From 2089bb3eae04cbfd9a1d2db78e3047dc7b9f4eaf Mon Sep 17 00:00:00 2001
From: GitHub Actions <actions@github.com>
Date: Fri, 26 Jan 2024 20:11:09 +0000
Subject: [PATCH] cli-14512-2315

---
 404.html                                      |   2 +-
 LICENSE-text.html                             |   2 +-
 apple-touch-icon-120x120.png                  | Bin 9676 -> 9676 bytes
 apple-touch-icon-152x152.png                  | Bin 13068 -> 13068 bytes
 apple-touch-icon-180x180.png                  | Bin 16269 -> 16269 bytes
 apple-touch-icon-60x60.png                    | Bin 4132 -> 4132 bytes
 apple-touch-icon-76x76.png                    | Bin 5395 -> 5395 bytes
 apple-touch-icon.png                          | Bin 16269 -> 16269 bytes
 articles/LC-MS-feature-grouping.html          | 111 +++----
 .../figure-html/example-1-eic-1.png           | Bin 96812 -> 102778 bytes
 .../figure-html/example-1-eic-norm-1.png      | Bin 100881 -> 114445 bytes
 .../figure-html/example-2-eic-1.png           | Bin 106836 -> 106326 bytes
 .../figure-html/example-2-eic-norm-1.png      | Bin 126272 -> 126225 bytes
 articles/index.html                           |   2 +-
 articles/xcms-direct-injection.html           | 117 +++----
 .../figure-html/calibrationresult-1.png       | Bin 23848 -> 28255 bytes
 articles/xcms-lcms-ms.html                    | 201 ++++++------
 articles/xcms.html                            | 307 +++++++++++-------
 .../xcms_files/figure-html/final-pca-1.png    | Bin 103834 -> 102372 bytes
 .../figure-html/unnamed-chunk-9-1.png         | Bin 0 -> 85614 bytes
 authors.html                                  |   2 +-
 favicon-16x16.png                             | Bin 1059 -> 1059 bytes
 favicon-32x32.png                             | Bin 2074 -> 2074 bytes
 index.html                                    |   2 +-
 pkgdown.yml                                   |   4 +-
 reference/AutoLockMass-methods.html           |   2 +-
 reference/GenericParam.html                   |   2 +-
 reference/PlainTextParam.html                 |   4 +-
 reference/ProcessHistory-class.html           |   2 +-
 reference/RDataParam.html                     |   2 +-
 reference/SSgauss.html                        |   2 +-
 reference/XCMSnExp-class.html                 |   6 +-
 reference/XCMSnExp-filter-methods.html        |   2 +-
 reference/XCMSnExp-inherited-methods.html     |   2 +-
 reference/XCMSnExp-peak-grouping-results.html |   2 +-
 reference/XChromatogram.html                  |   4 +-
 reference/XcmsExperiment.html                 |  12 +-
 reference/absent-methods.html                 |   2 +-
 reference/adjustRtime.html                    |   2 +-
 reference/applyAdjustedRtime.html             |   4 +-
 reference/binYonX.html                        |   2 +-
 reference/breaks_on_binSize.html              |   2 +-
 reference/breaks_on_nBins.html                |   2 +-
 reference/c.xcmsSet.html                      |   2 +-
 reference/calibrate-calibrant-mass.html       |   2 +-
 reference/calibrate.html                      |   2 +-
 reference/chromPeakChromatograms.html         |   2 +-
 reference/chromPeakSpectra.html               |  14 +-
 reference/chromatogram-method.html            |   2 +-
 reference/colMax.html                         |   2 +-
 reference/collect-methods.html                |   2 +-
 reference/correlate-Chromatogram.html         |  12 +-
 reference/descendZero.html                    |   2 +-
 reference/diffreport-methods.html             |   2 +-
 reference/dirname.html                        |   2 +-
 reference/do_adjustRtime_peakGroups.html      |   2 +-
 reference/do_findChromPeaks_centWave.html     |   4 +-
 ...indChromPeaks_centWaveWithPredIsoROIs.html |   4 +-
 reference/do_findChromPeaks_massifquant.html  |   4 +-
 .../do_findChromPeaks_matchedFilter.html      |   4 +-
 reference/do_findPeaks_MSW.html               |   4 +-
 reference/do_groupChromPeaks_density.html     |  43 ++-
 reference/do_groupChromPeaks_nearest.html     |   6 +-
 reference/do_groupPeaks_mzClust.html          |  16 +-
 reference/doubleMatrix.html                   |   2 +-
 reference/estimatePrecursorIntensity.html     |   2 +-
 reference/etg.html                            |   2 +-
 reference/exportMetaboAnalyst.html            |   2 +-
 reference/extractMsData-method.html           |   2 +-
 reference/feature-grouping.html               |   2 +-
 reference/featureChromatograms-2.png          | Bin 0 -> 54998 bytes
 reference/featureChromatograms.html           | 180 +++++++---
 reference/featureSpectra.html                 |   2 +-
 reference/featureSummary.html                 |   2 +-
 reference/fillChromPeaks.html                 |  56 ++--
 reference/fillPeaks-methods.html              |   2 +-
 reference/fillPeaks.MSW-methods.html          |   2 +-
 reference/fillPeaks.chrom-methods.html        |   2 +-
 reference/filter-MChromatograms.html          |   2 +-
 reference/filtfft.html                        |   2 +-
 ...ChromPeaks-Chromatogram-CentWaveParam.html |   2 +-
 ...ChromPeaks-Chromatogram-MatchedFilter.html |   2 +-
 reference/findChromPeaks-centWave.html        |   6 +-
 ...indChromPeaks-centWaveWithPredIsoROIs.html |   6 +-
 reference/findChromPeaks-massifquant.html     |   6 +-
 reference/findChromPeaks-matchedFilter.html   |   6 +-
 reference/findChromPeaks.html                 |   4 +-
 reference/findChromPeaksIsolationWindow.html  |   2 +-
 reference/findEqualGreater.html               |   2 +-
 reference/findMZ.html                         |   2 +-
 reference/findPeaks-MSW.html                  |   8 +-
 reference/findPeaks-methods.html              |   2 +-
 reference/findPeaks.MS1-methods.html          |   2 +-
 reference/findPeaks.MSW-xcmsRaw-method.html   |   2 +-
 ...s.addPredictedIsotopeFeatures-methods.html |   2 +-
 reference/findPeaks.centWave-methods.html     |   2 +-
 ...tWaveWithPredictedIsotopeROIs-methods.html |   2 +-
 reference/findPeaks.massifquant-methods.html  |   2 +-
 ...indPeaks.matchedFilter-xcmsRaw-method.html |   2 +-
 reference/findneutral.html                    |   2 +-
 reference/getEIC-methods.html                 |   2 +-
 reference/getPeaks-methods.html               |   2 +-
 reference/getScan-methods.html                |   2 +-
 reference/getSpec-methods.html                |   2 +-
 reference/getXcmsRaw-methods.html             |   2 +-
 reference/group-methods.html                  |   2 +-
 reference/group.density.html                  |   2 +-
 reference/group.mzClust.html                  |   2 +-
 reference/group.nearest.html                  |   2 +-
 reference/groupChromPeaks.html                |  60 +++-
 .../groupFeatures-abundance-correlation.html  |   2 +-
 reference/groupFeatures-eic-similarity.html   |  11 +-
 reference/groupFeatures-similar-rtime.html    |   2 +-
 reference/groupOverlaps.html                  |   2 +-
 reference/groupnames-XCMSnExp.html            |   2 +-
 reference/groupnames-methods.html             |   2 +-
 reference/groupval-methods.html               |   2 +-
 reference/highlightChromPeaks.html            |   2 +-
 reference/image.xcmsRaw.html                  |   2 +-
 reference/imputeLinInterpol.html              |   2 +-
 reference/imputeRowMin.html                   |   2 +-
 reference/imputeRowMinRand.html               |  50 +--
 reference/index.html                          |   4 +-
 ...ionWindowTargetMz-OnDiskMSnExp-method.html |   2 +-
 reference/levelplot.xcmsRaw.html              |   2 +-
 reference/loadRaw-methods.html                |   2 +-
 reference/loadXcmsData.html                   |   2 +-
 reference/manualChromPeaks.html               |   2 +-
 reference/medianFilter.html                   |   2 +-
 reference/msn2xcmsRaw.html                    |   2 +-
 reference/na.flatfill.html                    |   2 +-
 reference/overlappingFeatures.html            |   2 +-
 reference/panel.cor.html                      |   2 +-
 reference/peakPlots-methods.html              |   2 +-
 reference/peakTable-methods.html              |   2 +-
 reference/peaksWithCentWave.html              |   2 +-
 reference/peaksWithMatchedFilter.html         |   2 +-
 reference/phenoDataFromPaths.html             |   2 +-
 reference/plot.xcmsEIC.html                   |   2 +-
 reference/plotAdjustedRtime.html              |   2 +-
 reference/plotChrom-methods.html              |   2 +-
 reference/plotChromPeakDensity.html           |   2 +-
 reference/plotChromPeaks.html                 |   2 +-
 reference/plotChromatogramsOverlay.html       |   2 +-
 reference/plotEIC-methods.html                |   2 +-
 reference/plotFeatureGroups.html              |   2 +-
 reference/plotMsData.html                     |   2 +-
 reference/plotPeaks-methods.html              |   2 +-
 reference/plotQC.html                         |   2 +-
 reference/plotRaw-methods.html                |   2 +-
 reference/plotScan-methods.html               |   2 +-
 reference/plotSpec-methods.html               |   2 +-
 reference/plotSurf-methods.html               |   2 +-
 reference/plotTIC-methods.html                |   2 +-
 reference/plotrt-methods.html                 |   2 +-
 reference/profGenerate.html                   |   2 +-
 reference/profMat-xcmsSet.html                |   2 +-
 reference/profMedFilt-methods.html            |   2 +-
 reference/profMethod-methods.html             |   2 +-
 reference/profRange-methods.html              |   2 +-
 reference/profStep-methods.html               |   2 +-
 reference/pval.html                           |   2 +-
 reference/rawEIC-methods.html                 |   2 +-
 reference/rawMat-methods.html                 |   2 +-
 reference/reconstructChromPeakSpectra.html    |   2 +-
 reference/rectUnique.html                     |   2 +-
 reference/refineChromPeaks.html               |   9 +-
 reference/removeIntensity-Chromatogram.html   |   2 +-
 reference/retcor-methods.html                 |   2 +-
 reference/retcor.obiwarp-methods.html         |   2 +-
 reference/retcor.peakgroups-methods.html      |   2 +-
 reference/retexp.html                         |   2 +-
 reference/rla.html                            |   2 +-
 reference/sampnames-methods.html              |   2 +-
 reference/showError-xcmsSet-method.html       |   2 +-
 reference/specDist-methods.html               |   2 +-
 reference/specDist.cosine-methods.html        |   2 +-
 reference/specDist.meanMZmatch-methods.html   |   2 +-
 reference/specDist.peakCount-methods.html     |   2 +-
 reference/specNoise.html                      |   2 +-
 reference/specPeaks.html                      |   2 +-
 reference/split.xcmsRaw.html                  |   2 +-
 reference/split.xcmsSet.html                  |   2 +-
 reference/stitch-methods.html                 |   2 +-
 reference/storeResults.html                   |   2 +-
 ...gicalOrNumeric-missing-missing-method.html |   2 +-
 reference/updateObject-xcmsSet-method.html    |   2 +-
 reference/useOriginalCode.html                |   2 +-
 reference/verify.mzQuantML.html               |   2 +-
 reference/write.cdf-methods.html              |   2 +-
 reference/write.mzQuantML.html                |   2 +-
 reference/write.mzdata-methods.html           |   2 +-
 ...writeMSData-XCMSnExp-character-method.html |   2 +-
 reference/writeMzTab.html                     |   2 +-
 reference/xcms-deprecated.html                |   2 +-
 reference/xcmsEIC-class.html                  |   2 +-
 reference/xcmsFileSource-class.html           |   2 +-
 reference/xcmsFragments-class.html            |   2 +-
 reference/xcmsFragments.html                  |   2 +-
 reference/xcmsPeaks-class.html                |   2 +-
 reference/xcmsRaw-class.html                  |   2 +-
 reference/xcmsRaw.html                        |   2 +-
 reference/xcmsSet-class.html                  |   2 +-
 reference/xcmsSet.html                        |   2 +-
 reference/xcmsSource-class.html               |   2 +-
 reference/xcmsSource-methods.html             |   2 +-
 search.json                                   |   2 +-
 207 files changed, 912 insertions(+), 691 deletions(-)
 create mode 100644 articles/xcms_files/figure-html/unnamed-chunk-9-1.png
 create mode 100644 reference/featureChromatograms-2.png

diff --git a/404.html b/404.html
index 882ad8f43..97a455efa 100644
--- a/404.html
+++ b/404.html
@@ -31,7 +31,7 @@
     
     <a class="navbar-brand me-2" href="https://sneumann.github.io/xcms/index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/LICENSE-text.html b/LICENSE-text.html
index 8ac54bf24..2f6b12d7c 100644
--- a/LICENSE-text.html
+++ b/LICENSE-text.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/apple-touch-icon-120x120.png b/apple-touch-icon-120x120.png
index 2851b4f93731c1f58275906c0ce47ab4ee1e91ec..9c84456e9f91f1bab125e780e39968413467e044 100644
GIT binary patch
delta 101
zcmX@(ea3r21=kBkDG?T3zS;ISH#SXB@ix&lFw`|N3o$aVGBB|+G1WEzG7RimPckzw
kFsPQeMwFx^mZVzc=BH$)RWcYE7-84of1_~!<XF{w00t)=Y5)KL

delta 101
zcmX@(ea3r21=n*Pb`C8jfg>yCZETvL;%%&JXryak9%5i&WnyGyV5DteU}a#ipf10g
mfq_A_#5JNMC9x#cDmOnRGp&-r$iN7@hKtJA%_qmI<^upYfgRrf

diff --git a/apple-touch-icon-152x152.png b/apple-touch-icon-152x152.png
index 2cfdd374687f421bdfbc6b709bb50346d512f539..11cbd3bc199d49d4c336bfb8800b15dfbba5c0b7 100644
GIT binary patch
delta 101
zcmeB4>q(nX!S#YsN`ysk`lP(%jZGH}y-jot40VmnLW~To3{0#{%(M-F3<Lh$%kvl*
k7*tDKBT7;dOH!?J^HVa@DjAFnjIe8{Gssq&Jl7~603?|mH~;_u

delta 101
zcmeB4>q(nX!S$SnokNQ`{qPKqjZGH}y^VDZjdTsnLkujeOpL4yjI|96tPBj;I8*Ey
l7#LJbTq8<S5=&C8a`RI%(<&K^42-aAC=a>QJbA8BJ^;<I9HamM

diff --git a/apple-touch-icon-180x180.png b/apple-touch-icon-180x180.png
index 725ce5d174d1e15e8aef5b555a0f92751c759d90..5c3bcfc99977821dea27d5eff2d3bfb1b2b6f89e 100644
GIT binary patch
delta 83
zcmeCJ@2#Iu!S#YsN`%EAcWyAl#->0!DHB}-LtP`Y5F-OC0~0F~3vB}+!yrCr#o@_s
R?Zk0N>+uWJPnNUK2LK0n7SR9z

delta 83
zcmeCJ@2#Iu!S$SnokNS|=W^cF8=C^{q>Ob9jdTsnLkujeOpL4yOtcLQtPBheuZ!9^
T`K_He4(UywAMs9>v(E<rUyK>q

diff --git a/apple-touch-icon-60x60.png b/apple-touch-icon-60x60.png
index 167f7abf456a0e756905ac9e28d04d2351478d82..0af9c28ac949d743dc2b2b1a6f5c43659e0efc1a 100644
GIT binary patch
delta 101
zcmZ3YutZ@(1=kBkDG?T}3IA7BZESkM=WU{EV5n<k7Gh*zWnf}uVx(;VWElKb`rgaH
kz@S><8c~vxSdwa$o1c=IR>@#wV1!*mSG~XC<gNVq0Av^*bpQYW

delta 101
zcmZ3YutZ@(1=n*Pb`C9uf2U4v+Sv4h&)Znn&`8(7JjB4l%EZXZz(Cu;07#lYtXssu
lz@S><8c~vxSdwa$o1c=IR>@#wV1!+R+n!@`lehBc0|0lY9^?Q3

diff --git a/apple-touch-icon-76x76.png b/apple-touch-icon-76x76.png
index 2cc49599e422376d19394dbe5d372e420677a780..6d3ed54a9a7e840d470e0f19e221f16003559ac1 100644
GIT binary patch
delta 101
zcmbQNHCbyy1=kBkDG?T($~pZ08=I~Rdz<JQ80s3Cg%}xF8JJj^m}naS83r>x9oo*o
lz@S><8c~vxSdwa$o1c=IR>@#wV1!-6%k@><lb49(0{{tX9t!{f

delta 101
zcmbQNHCbyy1=n*Pb`CAZ&a*#NHa1-s_BPfvG}1LN4>7Q?GBL6;Fw{0Mure^1IIUES
mfq_A_#5JNMC9x#cDmOnRGp&-r$iN7@h6kt5uA00=Bp(12&K{=#

diff --git a/apple-touch-icon.png b/apple-touch-icon.png
index 072eef3dad9657f6f028f17c90286196da58e408..51d348ef96adcca7493a0b47a82e06fb4490ea07 100644
GIT binary patch
delta 83
zcmeCJ@2#Iu!S#YsN`%EwYF6^JjZJ}eQYN|vhPp;(Aw~vP1}0V}mf8kDhQUwO#pRRV
R+KJ<k?x>JHJz35^9{?!H7+L@T

delta 83
zcmeCJ@2#Iu!S$SnokNSYZ-K4t#->0!DPvtjBV7aY5CaP<6C*1FQ*8qSD+7bnn}Mv8
S-`a`ekQQlqxpT6deLesS0T=23

diff --git a/articles/LC-MS-feature-grouping.html b/articles/LC-MS-feature-grouping.html
index 8f41da717..18ecbb1c0 100644
--- a/articles/LC-MS-feature-grouping.html
+++ b/articles/LC-MS-feature-grouping.html
@@ -33,7 +33,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -100,7 +100,7 @@
     if (links[i].hostname != window.location.hostname)
       links[i].target = '_blank';
 });
-</script><p><strong>Package</strong>: <em><a href="https://bioconductor.org/packages/3.19/xcms" class="external-link">xcms</a></em><br><strong>Authors</strong>: Johannes Rainer<br><strong>Modified</strong>: 2023-12-07 07:39:24.923344<br><strong>Compiled</strong>: Thu Dec 7 08:44:39 2023</p>
+</script><p><strong>Package</strong>: <em><a href="https://bioconductor.org/packages/3.19/xcms" class="external-link">xcms</a></em><br><strong>Authors</strong>: Johannes Rainer<br><strong>Modified</strong>: 2024-01-26 18:14:18.95093<br><strong>Compiled</strong>: Fri Jan 26 20:08:00 2024</p>
 <div class="section level2">
 <h2 id="introduction">Introduction<a class="anchor" aria-label="anchor" href="#introduction"></a>
 </h2>
@@ -816,7 +816,7 @@ <h2 id="session-information">Session information<a class="anchor" aria-label="an
 </h2>
 <div class="sourceCode" id="cb32"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/utils/sessionInfo.html" class="external-link">sessionInfo</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
-<pre><code><span><span class="co">## R Under development (unstable) (2023-11-22 r85609)</span></span>
+<pre><code><span><span class="co">## R Under development (unstable) (2024-01-17 r85813)</span></span>
 <span><span class="co">## Platform: x86_64-pc-linux-gnu</span></span>
 <span><span class="co">## Running under: Ubuntu 22.04.3 LTS</span></span>
 <span><span class="co">## </span></span>
@@ -841,67 +841,68 @@ <h2 id="session-information">Session information<a class="anchor" aria-label="an
 <span><span class="co">## </span></span>
 <span><span class="co">## other attached packages:</span></span>
 <span><span class="co">##  [1] pheatmap_1.0.12     faahKO_1.43.0       MsFeatures_1.11.0  </span></span>
-<span><span class="co">##  [4] xcms_4.1.4          MSnbase_2.29.1      ProtGenerics_1.35.0</span></span>
-<span><span class="co">##  [7] S4Vectors_0.41.2    mzR_2.37.0          Rcpp_1.0.11        </span></span>
+<span><span class="co">##  [4] xcms_4.1.6          MSnbase_2.29.3      ProtGenerics_1.35.2</span></span>
+<span><span class="co">##  [7] S4Vectors_0.41.3    mzR_2.37.0          Rcpp_1.0.12        </span></span>
 <span><span class="co">## [10] Biobase_2.63.0      BiocGenerics_0.49.1 BiocParallel_1.37.0</span></span>
 <span><span class="co">## [13] BiocStyle_2.31.0   </span></span>
 <span><span class="co">## </span></span>
 <span><span class="co">## loaded via a namespace (and not attached):</span></span>
-<span><span class="co">##   [1] bitops_1.0-7                rlang_1.1.2                </span></span>
-<span><span class="co">##   [3] magrittr_2.0.3              clue_0.3-65                </span></span>
-<span><span class="co">##   [5] MassSpecWavelet_1.69.0      matrixStats_1.1.0          </span></span>
-<span><span class="co">##   [7] compiler_4.4.0              systemfonts_1.0.5          </span></span>
-<span><span class="co">##   [9] vctrs_0.6.5                 stringr_1.5.1              </span></span>
-<span><span class="co">##  [11] MetaboCoreUtils_1.11.0      pkgconfig_2.0.3            </span></span>
-<span><span class="co">##  [13] crayon_1.5.2                fastmap_1.1.1              </span></span>
-<span><span class="co">##  [15] XVector_0.43.0              utf8_1.2.4                 </span></span>
-<span><span class="co">##  [17] rmarkdown_2.25              preprocessCore_1.65.0      </span></span>
-<span><span class="co">##  [19] ragg_1.2.6                  purrr_1.0.2                </span></span>
-<span><span class="co">##  [21] MultiAssayExperiment_1.29.0 xfun_0.41                  </span></span>
-<span><span class="co">##  [23] zlibbioc_1.49.0             cachem_1.0.8               </span></span>
-<span><span class="co">##  [25] GenomeInfoDb_1.39.1         jsonlite_1.8.8             </span></span>
-<span><span class="co">##  [27] progress_1.2.3              highr_0.10                 </span></span>
-<span><span class="co">##  [29] DelayedArray_0.29.0         prettyunits_1.2.0          </span></span>
-<span><span class="co">##  [31] parallel_4.4.0              cluster_2.1.6              </span></span>
-<span><span class="co">##  [33] R6_2.5.1                    bslib_0.6.1                </span></span>
-<span><span class="co">##  [35] stringi_1.8.2               RColorBrewer_1.1-3         </span></span>
-<span><span class="co">##  [37] limma_3.59.1                GenomicRanges_1.55.1       </span></span>
-<span><span class="co">##  [39] jquerylib_0.1.4             bookdown_0.37              </span></span>
-<span><span class="co">##  [41] SummarizedExperiment_1.33.1 iterators_1.0.14           </span></span>
-<span><span class="co">##  [43] knitr_1.45                  IRanges_2.37.0             </span></span>
-<span><span class="co">##  [45] igraph_1.5.1                splines_4.4.0              </span></span>
-<span><span class="co">##  [47] Matrix_1.6-4                tidyselect_1.2.0           </span></span>
-<span><span class="co">##  [49] abind_1.4-5                 yaml_2.3.7                 </span></span>
-<span><span class="co">##  [51] doParallel_1.0.17           codetools_0.2-19           </span></span>
-<span><span class="co">##  [53] affy_1.81.0                 lattice_0.22-5             </span></span>
-<span><span class="co">##  [55] tibble_3.2.1                plyr_1.8.9                 </span></span>
-<span><span class="co">##  [57] evaluate_0.23               survival_3.5-7             </span></span>
-<span><span class="co">##  [59] desc_1.4.2                  Spectra_1.13.2             </span></span>
-<span><span class="co">##  [61] pillar_1.9.0                affyio_1.73.0              </span></span>
-<span><span class="co">##  [63] BiocManager_1.30.22         MatrixGenerics_1.15.0      </span></span>
-<span><span class="co">##  [65] foreach_1.5.2               MALDIquant_1.22.1          </span></span>
-<span><span class="co">##  [67] ncdf4_1.22                  generics_0.1.3             </span></span>
-<span><span class="co">##  [69] rprojroot_2.0.4             RCurl_1.98-1.13            </span></span>
+<span><span class="co">##   [1] DBI_1.2.1                   bitops_1.0-7               </span></span>
+<span><span class="co">##   [3] rlang_1.1.3                 magrittr_2.0.3             </span></span>
+<span><span class="co">##   [5] clue_0.3-65                 MassSpecWavelet_1.69.0     </span></span>
+<span><span class="co">##   [7] matrixStats_1.2.0           compiler_4.4.0             </span></span>
+<span><span class="co">##   [9] systemfonts_1.0.5           vctrs_0.6.5                </span></span>
+<span><span class="co">##  [11] stringr_1.5.1               MetaboCoreUtils_1.11.2     </span></span>
+<span><span class="co">##  [13] pkgconfig_2.0.3             crayon_1.5.2               </span></span>
+<span><span class="co">##  [15] fastmap_1.1.1               XVector_0.43.1             </span></span>
+<span><span class="co">##  [17] utf8_1.2.4                  rmarkdown_2.25             </span></span>
+<span><span class="co">##  [19] preprocessCore_1.65.0       ragg_1.2.7                 </span></span>
+<span><span class="co">##  [21] purrr_1.0.2                 xfun_0.41                  </span></span>
+<span><span class="co">##  [23] MultiAssayExperiment_1.29.0 zlibbioc_1.49.0            </span></span>
+<span><span class="co">##  [25] cachem_1.0.8                GenomeInfoDb_1.39.5        </span></span>
+<span><span class="co">##  [27] jsonlite_1.8.8              progress_1.2.3             </span></span>
+<span><span class="co">##  [29] highr_0.10                  DelayedArray_0.29.0        </span></span>
+<span><span class="co">##  [31] prettyunits_1.2.0           parallel_4.4.0             </span></span>
+<span><span class="co">##  [33] cluster_2.1.6               R6_2.5.1                   </span></span>
+<span><span class="co">##  [35] RColorBrewer_1.1-3          bslib_0.6.1                </span></span>
+<span><span class="co">##  [37] stringi_1.8.3               limma_3.59.1               </span></span>
+<span><span class="co">##  [39] GenomicRanges_1.55.2        jquerylib_0.1.4            </span></span>
+<span><span class="co">##  [41] bookdown_0.37               SummarizedExperiment_1.33.3</span></span>
+<span><span class="co">##  [43] iterators_1.0.14            knitr_1.45                 </span></span>
+<span><span class="co">##  [45] IRanges_2.37.1              splines_4.4.0              </span></span>
+<span><span class="co">##  [47] Matrix_1.6-5                igraph_1.6.0               </span></span>
+<span><span class="co">##  [49] tidyselect_1.2.0            abind_1.4-5                </span></span>
+<span><span class="co">##  [51] yaml_2.3.8                  doParallel_1.0.17          </span></span>
+<span><span class="co">##  [53] codetools_0.2-19            affy_1.81.0                </span></span>
+<span><span class="co">##  [55] lattice_0.22-5              tibble_3.2.1               </span></span>
+<span><span class="co">##  [57] plyr_1.8.9                  evaluate_0.23              </span></span>
+<span><span class="co">##  [59] survival_3.5-7              desc_1.4.3                 </span></span>
+<span><span class="co">##  [61] Spectra_1.13.3              pillar_1.9.0               </span></span>
+<span><span class="co">##  [63] affyio_1.73.0               BiocManager_1.30.22        </span></span>
+<span><span class="co">##  [65] MatrixGenerics_1.15.0       foreach_1.5.2              </span></span>
+<span><span class="co">##  [67] MALDIquant_1.22.2           ncdf4_1.22                 </span></span>
+<span><span class="co">##  [69] generics_0.1.3              RCurl_1.98-1.14            </span></span>
 <span><span class="co">##  [71] hms_1.1.3                   ggplot2_3.4.4              </span></span>
 <span><span class="co">##  [73] munsell_0.5.0               scales_1.3.0               </span></span>
-<span><span class="co">##  [75] MsExperiment_1.5.1          glue_1.6.2                 </span></span>
+<span><span class="co">##  [75] MsExperiment_1.5.4          glue_1.7.0                 </span></span>
 <span><span class="co">##  [77] lazyeval_0.2.2              tools_4.4.0                </span></span>
-<span><span class="co">##  [79] mzID_1.41.0                 robustbase_0.99-1          </span></span>
-<span><span class="co">##  [81] QFeatures_1.13.1            vsn_3.71.0                 </span></span>
+<span><span class="co">##  [79] robustbase_0.99-1           mzID_1.41.0                </span></span>
+<span><span class="co">##  [81] QFeatures_1.13.2            vsn_3.71.0                 </span></span>
 <span><span class="co">##  [83] RANN_2.6.1                  fs_1.6.3                   </span></span>
-<span><span class="co">##  [85] XML_3.99-0.16               grid_4.4.0                 </span></span>
-<span><span class="co">##  [87] impute_1.77.0               MsCoreUtils_1.15.1         </span></span>
+<span><span class="co">##  [85] XML_3.99-0.16.1             grid_4.4.0                 </span></span>
+<span><span class="co">##  [87] impute_1.77.0               MsCoreUtils_1.15.3         </span></span>
 <span><span class="co">##  [89] colorspace_2.1-0            GenomeInfoDbData_1.2.11    </span></span>
-<span><span class="co">##  [91] cli_3.6.1                   textshaping_0.3.7          </span></span>
-<span><span class="co">##  [93] fansi_1.0.5                 S4Arrays_1.3.1             </span></span>
-<span><span class="co">##  [95] dplyr_1.1.4                 AnnotationFilter_1.27.0    </span></span>
-<span><span class="co">##  [97] pcaMethods_1.95.0           gtable_0.3.4               </span></span>
-<span><span class="co">##  [99] DEoptimR_1.1-3              sass_0.4.8                 </span></span>
-<span><span class="co">## [101] digest_0.6.33               SparseArray_1.3.1          </span></span>
-<span><span class="co">## [103] farver_2.1.1                multtest_2.59.0            </span></span>
-<span><span class="co">## [105] memoise_2.0.1               htmltools_0.5.7            </span></span>
-<span><span class="co">## [107] pkgdown_2.0.7.9000          lifecycle_1.0.4            </span></span>
-<span><span class="co">## [109] statmod_1.5.0               MASS_7.3-60.1</span></span></code></pre>
+<span><span class="co">##  [91] PSMatch_1.7.1               cli_3.6.2                  </span></span>
+<span><span class="co">##  [93] textshaping_0.3.7           fansi_1.0.6                </span></span>
+<span><span class="co">##  [95] S4Arrays_1.3.2              dplyr_1.1.4                </span></span>
+<span><span class="co">##  [97] AnnotationFilter_1.27.0     DEoptimR_1.1-3             </span></span>
+<span><span class="co">##  [99] pcaMethods_1.95.0           gtable_0.3.4               </span></span>
+<span><span class="co">## [101] sass_0.4.8                  digest_0.6.34              </span></span>
+<span><span class="co">## [103] SparseArray_1.3.3           farver_2.1.1               </span></span>
+<span><span class="co">## [105] multtest_2.59.0             memoise_2.0.1              </span></span>
+<span><span class="co">## [107] htmltools_0.5.7             pkgdown_2.0.7.9000         </span></span>
+<span><span class="co">## [109] lifecycle_1.0.4             statmod_1.5.0              </span></span>
+<span><span class="co">## [111] MASS_7.3-60.2</span></span></code></pre>
 </div>
 <div class="section level2">
 <h2 id="references">References<a class="anchor" aria-label="anchor" href="#references"></a>
diff --git a/articles/LC-MS-feature-grouping_files/figure-html/example-1-eic-1.png b/articles/LC-MS-feature-grouping_files/figure-html/example-1-eic-1.png
index 4b3830af2c8c4d61e16d42c92716adb98fda8ba3..6e96a7c4459e333fe27afdb3d1c8937185aa24c6 100644
GIT binary patch
literal 102778
zcmeFYXH-+)*Dkt~Kq9><ARXz_M5=V@O^}XA6I7&wf^?FIC=sPMK|twMK&l8NDhi5#
zgeo;Zib(Grvd{MS{@?fBG48k@&p2NW1IFe9J8Lr6oX?!|dDg;RH`S-3=AZ@ufX>iB
z#{vM5(61DjGk+gfK0e8U9!OVXeI4iteI06lKtm5Ie}fwj0D$}H-yc}Q{N^11-~$YG
zu3Ck>TA2)O7PNjz0@usVl|QS}+>1p0I%mbGbK@){1;y<LFo&_VVRJq7%&ebxPVS-}
zy2^CfBthj_>@!Ry{L$-`Gmox6dd=rm$9n8Eet5e1yWzs@@3u6>?YnwX(|y}>mght+
zT%NAHsW{!Gw9twYG)(@#-~R_Zup9rY;i;e@2hVS`Ve*rNL+9i|{uImt23@7h=psqN
zBboqd1OvhyIA#urze7oyCCC$K2rPtO1TrC2ruP@C-OmoUU4<z#csp$b_h{>j1rO^F
z>j4-2mHfpv*N@mYa>lmfONR1hLw>{aD1G5h6u%J*u&1yA`v_A*j=R6n+jwYAXyNG-
zKXOJPf4xj1R?&*9Q>H=bYnTJo4{!gew}gt>SHZFY*y|4uE<5?n)#`Dsd#r!jUYY#5
zU0<HF-+Up7(5dWRN%PeP9Y6c)Mf3IZ%U3|7a5m3mx#vH?ze@<x@j~lNpGX^O|5=Y^
z=*LBgnia2>f|}021fc5Z2|6vdHehs%+<cL1VhxUce{vVlcTaSN)+h<;gU*>&0kEG-
z;jiEmcFMmMTq!hbR|Sf;?N6`7yIbptfZ?&`Jo5_ra083&|L<z{q~vFAN^wQJb8NLc
zuD7W-e3mr+#5GXkt?uEQ>;2qFWuM4CM%AGKRi_;3!<ybZ_a1&9dq$vY$Xs&^Os<-N
zzaH9eW-y+0qVTr%G9xUMnQeTb>HPbT9+rH%(#NDR!i37m-6jd=Stj2mD}vdekO%uq
zum}3L4VPr}YwAtMEjv()hJ?Bf)OH@WuVyFup^&TRaHLL7y%%Qx(nIF#YZ3nK)r~p#
z-rk*n=eJMY4t-g~J2RfPSEYaY^yF?#q%GT+;@$cLwmn{_v3>8VQ*ocK>aNCmBB$lj
zfPG}bLap4|mKDBi5Nw^Y868>b3s4Pa?LD-MKDA}t18cr5q!v24LzXpthTMPp4ccbE
zr-#nkU&Q!01TC?P4R~p@h)cZJNAc(VFPjIi+L7+Nejki=^h$`tO)b95@j^S9-^-pN
zHS2wkTdveHVdC60&&pnsr#$evbxbCbV~SQ6Xwdklr3|PS;i+*AAkW2<$k6;w9#X=!
zj?w1}*ln8e0c@}d@j+Am(0(SSiLDWBIfiFHkT(W-*Yr;CUJDCKxsudq{M(P@ZtSEN
z+HWskD>*C&3YY1NQ0VMRL!pDW{_9USeVM50Egha0>Rez>X_3MOrnaRQJo7OR_r%pk
zCWvH)rlVqGQwi0DhN8%uY<rWvM03Cr7nhrq2|9Rj)|&s*`WXvq?a1Lx2HTIr$EtBk
zX-DV;-vr>^u3GOK^|e(_uI-^u<UGameIXj1n_SKzKYVJy<(P#w7kHnzVcU<t&P#*x
z<tXEStgGFkB!o(=4n;~9q-=fbcp3($7h4VI-DQ@ks(HD1YyOL)@7*2m>~F1r9ZKnp
zQo}|>6GB{3>7n})+v6J!&t~5t7gptaHw6-&dPKtbO3D8QoXuk9E{EzqT5<&+^@?41
z>w))|K$ob0-h)xexwgsWcmHnZ_rKfO{XZ7j9l~EvAdWtNYIi@cCRyF=dE9AytWes3
z4?QnaX>xjTtxq8q;rjj8x0udex_FOajZ4Zm+QOf^1-tJ&+vsiq4nh_)4+$}Z=%T&w
zQr;_<n5><F`Xe4iiB;#ij}Oq5HDGUmYi3tpKz=b$!j`*#CCk~);QwcvzR+NF>Gs%4
zG#m$AKxgV>(AZ;|s$p-wGxW5qBd~S`bzs`_boyN|^X^=(c-!8Hac5gaJy=1w>AB#n
z9l3S0$lq!6)1ud)vr?LL%ha!#@O%4kmr&O^-Ga_n!o7aGYQ9!uBfXTGbsC+xEJ_E9
zkao)FPJsr=M=$?{G2WgNX^)@(CL!4W158FbZAOP(jXs6U+U8i{U5AsXm)Lwf=0FSZ
z<7v`eM4lkkfs5Faq_Jq(vOM^W-1|x~nYjw@z*ORPyeh~0eE+G0o6p-OX)rlP3uZg-
zp`4o+!z{}w?GMD9Dk_}swu8(VOFlBEHM7KDPZXqs|BH3beUKIGnSY@G4Ev3KQKm|1
zM7-Ax3x{iPEM)G((YB-NbO$ylH%~#tv`|>Menjcs!b8Gyt?(=6()`62;+38heXGfg
z<q(q+uKP?evno{8-pOXvS@r4Omln+af3160$d&(%25fnFEb5xcoi?vWeL{wxze7Qx
z2pSVNWFR=|0XdZBV+;jf?6r*LWxS|GbGrVM&J@g8Q4{>>+2H;Zs8UlC_`?s=G(v7S
zT<yE>8dmi2frj<L@y=L5zrnM2i!pfoM91doy~_Oq_@<xS{eT9AjZWhGFFgAyp6JG-
z2i^+9br=MVj|d!$tCjW_G%R=tcK!t%V<O*WFrg1JCZo_O$dj6sfh)6oQT((@N3b_(
z+lj))Px7R=QYX@HcaJXqeSHo1%6`1DXj{S0e~OLzSU;>$Q7UIYi+F{qxvB=NU_l#e
zHpBr9$2`(p?$dHJk8a%t$30i8$Xpphum#rMn-nWUbMKP}tABtWPc&ry!iC>&(1o1G
z{ue{un>qrj$cTcEo0eG4Qjl-D>YH5#Lj7il$?c_yJnHITPGh#*mo{-9*!yZ@BBjrc
z`MEC+){E)*ZO(im`<?9-AEsB0E{a={&^O_fXm<(-c)4q|9?(ita8Vi<3r%2*+kd`t
z)5L`BT1P3y@7!LS(V#0=|MxzouVC<Gb>I}PL5V}{6-?OU`rmk{x&#_qpk<-Wxp@on
zjrs_V1{35t9<HE729`9u*#>i)`U_Nr{CyWEw(?bHf7J!98Qd3ZQZtC0#0zk8)nSf0
z@^I#~4`>$BQp}6nZNulx--<Vfdw?P0eYT-6pbD=aI|_FEL62;`2kn&o`jDM++Yh+b
zryC-!Do+*kuVVyv1^x$AouLAau7J{Huv0kzwlOufw}R~CHQDBEbQa4Q@1+C?)6%oM
zYi=G1_xn531EwPMs+sdJVBPa=QxWD{<{FNFVL~5*7%*>Zwn8pfYwUYpKqY$Qe0PYH
zG$2R^VfcUG`@dJ>P_`_m*IWwd1n>f3U&6nk8O2-BIJ==pD)W+!2+}3j{!D$&SNn9X
zR-ZLsO^h^ssthO3nn!PB_g8edlXzp5gQh8XO53IogeIJBa=fq}^G?OX=~-EV4dEDD
z5sMxv!Nonv!Ig!){}>9z2j|NVE82pF*Dn8?4ass5|LGF`Z^!$;-vfi%7DKRm4zd{Q
za(b=gCl1Km(o?aZXsCpiyOV93UV=hpVThODD)7gVmy%4EX&uCuEXV^)QT?b;W~5X7
zo4WMd_?9nFt!r~5ZQ;3rHV@>6oG_H#LhkUVw%m`AS<r`tL5*|HT_~5PWUn8hjfP;Q
zZ>%pJ8e3G~Xj%f~HvR^8_z!SN?(gU2V!wj}Q2bpxU#1~sxRiUI@ZB2q=FQ)?*JQy8
z&%emgj_m~>0y|GdF!KM}{UaOWRY9+%4HXu{NyzY_zn>V1$bk$`M4CF$*l}jshntk0
zQu*^1qApUKBn$2~v~pcSWnN3I$fT2Ckf8WI!qcbqJuODmFgfgcE}gg2cr!Qg(ZzzK
zCoi}GqUJXh7)o+Bvm4`ea$m|YIV(jGn4<O`?f6K;uR*1XoS<DKo9YQe?^@@=y=hk{
zygw4LxDQv=5yaWF7-72eDnKDN$rirnHA85_-bj>gV9`t#d?8s)*-AI+4HvnF8m?4p
zd>Wy+;bo8z$fs$TJosRX*2UIE-=zqNszgSd**Z-0UG}_3Onisnt3+}kTj@j2H&ZL|
zh7^5&T}Ty%+AR8T{sX*@KM&;*fO3Oo#YmBAqlp9bs-Wtx;?1rmgso<l234OJWTHhb
zjLVFcg1s-0biHdC!KQ6LxcRe?U%`YMFoP+={iGXs=}%HiQ`BeN2oDH1F&cqF%4L$F
z4_qoNX7e!aSRqZM6T=UQm1!P3o~=H9$rYo_WWln|`Knl~jAf@1$%CXzD*BXH7YmcG
zVDf^?ZP~kcbt%F#lz^+rh1hx#x9Jl!kKLQWy)c*6u}EBr%K1#WA5Y2<F1J6v%kRO;
zV)QYyvGk9=bg-qaBk%&OmO+KdVRpInE6$!LW~Apw%o>LYgxW!n$D6*DZOe?WZQ>tI
zAwvtZ?(9o@J>W6ULY?9HgQyG2h++6cl!7RCBE}K(zF`b61&A6F30+<Yzl-ux6uu~5
zBoD<yfc`Z(YHo}Wkcg*T{Nk%~kuF=TTmAOy$aHq@HiSQwj4|m3$==0%CVI&-O;<$T
zJ=kUb(?g^byW_Nm8ayI_{k?6yy1(4qttG}(LBpV}FhbJv_Cru~!;OTRoj!Tmyk{V-
z)uIx=Am{q}+OOgc;BoC;8iVA*_T&;ciYez*AeXY8dhu2B6NB?aNkTC0*iTm+jv7I-
z7-Zf4JO$e@C7HbL|32G?D1qw-$zSl$+VUj$7JZ$5(e~3R>{;gImYqju7=bK&6PixV
zoe)Y-`n{+e&qUA=o%I+b?9x-cxM9=CD0)>(WfwKqdgu3NF@lCm8h);Ui+MC0wj2ft
z%p<7a^oQQa)>@X2hM#>$Re)l6a6%oZ?vk_i_`)?S(~iOSGX`x7*zQgZX*Gsd@KpWH
zRh=IZNrHyj0@c*5Y<6$!XD=8inx1?`-}s*BE24L-W<T~`*<W?;{c772LBn#>T;oBq
zbQT-p%-3Wmexv&t>&7dN;zEKR%yi)>0_F3!W;XTWm#U3x4OklJYRt=4nGLF?jt6(B
zH(2z{K7^y@=tKmRpp01Fde+ecUMb?X)oi(S-&wp(GnKLhU+;D<pT(u&TcjHFC!tqh
zC=2HS@D6F_PF{98;Et$1%TWS<dNub-sEiRuBR$3kTMev>_^vs_ThqSg03VokyG>W^
z11$!&<XomLW}R_gg0fPbkWtiU*+|)NE?DUN*9B<1J|?pGQ|vCEfjKZZ{CYWJQq8yw
zKY3s8M(-~;cRnZ75n7vH<!6YCYf9W<u{&nl=*0u<=~%o8ahlc=MYy7~+RuXP*B@bY
z)&!))xx+8|%gbCl%yxWcmKFN~>)JE%g%Be-xDfg3Pk?k*jPoZULyo|ku;uhIvSx&b
zU)@Sy2vLjVPVgz>u3##ngM!_;tIk99(&jKu&8!8{TsGK@m|L}{zh;qhYY9+-k76>@
zcQ{<aNiHP8{58A&aI7<IoXIky(5X(><CxXJtZ~n3(UkPeVm6EL!Qrf3a^cm~5+FQ{
zV;$DJywU?QdB|+g=mo>Q8A6}-;@brcvnc#1(xttnDSG`j?Pr)tn-=3L{MA#Q{>kZC
zjVs#7URRf>)c{U^-)J`*K~~GOu<NGJ*MR1_9+9@>=0&{qo7Dz_HX+DIy$OmwNfmen
zYhwmu99e47>5Q0mJ75>b{z`TR=mM!0acf^L7^YlgqRC&=dk`(vy<DSo#yFC%kK&<}
zJeid^P^IQ<f9BAH^jV_K_6kCYb!(qbZQQdk`AHq(_4%n2Tj`tGM{ymLgveF3JKq!0
zS0`vi_iEM=hx`T%jr%7MPASeUcDyLd*U-$GKD8slyEIVh6eiX3rYQQSCpq@-`3#p7
zJ7Lpavk@b1mKq^yH4e9qg-B2KaxbgVAG(rqV!ocknPS#QhGkL*AKkhLvyzJzU4Njo
z5s**&6Gt#7m|`kkRAl(v7;EPw=I`YxNCHvtR*QSDv?8eThREB-$%Q;MG*BQ(!lNv&
ziC({%@F%0_>k#%-{ES|*@Mo1V)oE<jQa~ujMzO^#UCLkp7)}Z<Y8(OdN02fGo*!$3
z(&sVKKpwm}{=oz@X_>GP>ts@vVe;@W@6nFj+yjgN%?ok(`jeB{wRB`+Vf;-xCE5_~
z+TVTFgu}t$$3_T2X;ka`!4O2@LV0X4iGiTk0pu{LXtgsCDcG1(3%{Q}WA_%ND_=eD
zm`3{gEp#c>s}A@L++=%LsZVQqQnoKth~q4#$Gp_(0ediBII+-eObIT=s2ElOWFzzu
zc>q1U@rx<2`@EHUS12qUK);+AF$mUbXB}<$*-TZVvN8TGQqpj#xg=Iv@IRI2=vN{y
zA{Q~GHIbN#?C~9x9^R++57MnYd*Y*pX~4z5XIg)JqEx8=w{F{Ss^#k}XM4O|`s9Vd
zi+cy(l6hEW9ZDUbX1)_{kP<R2A&HGpxI{BrQ{(nvX6*G)%#)%w@=#^V6Z9poGdRfg
z1}MTxSQN=<vsBbkL}x)tmL<<AsQGk>bjo4JW{3aTC^CoQKW#u%ix}oyMrAsBj&ms?
zLLQe6gbUPU+MS#ncht`|V84~(+7)DUTV8Sk;@3Vc8@hFg1AY?nP6H~tfy=`P_QUwg
z+)C72P5f=w9lev~_|G?2?%n&UFc=INw0VoW_}&Va@t>T-o7kT%IK4t|S5V_7;&P{J
zq2*G|VC&x$hACxi`jOds5T?TF-YdZMYSCa|V(4XtVH|%RRm9^W#a?D}>OVKtu3ywl
z9^1U(iX5UR%*O0_VRNwhxcgX<hC}6JI2%h|UX@*K&q&6zJ(oqT4}kQ~J=e$-Nj2o5
z7ReEcWUK%`=yHEp6>1l%zTx_dDEyWuPO&@Ga|5x*p<lk#Dk8}|Te^g*USwjrPbMqO
z3DF$>Xko%Q(yYMhO{;3n)v{V`Zs5A{jJvr8Wsriuf#jyK1wKr~UwBs$)K_JAMa?Bo
zRlR-H9)HMS-hMrcB2>t=DlUKeYw595iZAS~wESBdu0OjFznV67#;&3!2M3emPR&J`
zBAg^vGpV>tXqhu*m*fnCpzZ*oy>1QMwWTxDk)BQCjx)&0>?s9vBDsi`Ig#IXY)Ohm
zUteGp1NVZ2$_k1vlV4+wF+sxr1Ua-?wYi2bRrBY<ecy_v-#vNOk1rUv#H&N~Yw7LQ
zA5eu~t^Y9N+BSBbF*>lj5g^01hUxYU%g+}xYdQ<I|8by+$ULx^x;MX}GjZ?qVr|Vg
zo=NKK_D`hJ!1MD)uK(n`p!NyZCW{y96}StPjW}u!&hvJ^)e$2+jKtYlHUH8DpteuU
zTP5d43ZIC}2aDMt^R;T*BB2J0$GmVG$(2#gEtW*GgSktxAYEB2y?I#4pRBdEs4(PK
zna=y($Wq>k532qYb2qVZ`GJXr`A~2)Qz{`3nEIJ&D4KKF=l3tyaMkc^QhIp$fSUyC
za~&7c<~#_+MPH?YfI`|sE>c0qI#q{z=bJ?Qz_=o#Al1JmunC5SN%0pm<8l3Q>8nUC
zidRZh)gR6r=&klY_sM(}k1D%&&86YfUYERof^oGbnabSAYwkz@_EcM)m*_L}Mc9z|
zQ_L3D%s|whq~pIs3Qk<-{|;8#D!YWcAiD6FKy<KMO4#_NGytxY!>+p%Fhc{T6R-rP
zj*QdQ*e^d#DQr-tWx^&NsdS~)l&A2Gyvgvx9SQ$iWdYYnRg-a7lkM-*Z_p-<qcxvf
zDOWYFza!R`z-?LL0Cb)4reTf^&na55FykWb_9tQlX<@MM$xBmOtx>#!rdG0Gkd+g?
zXvXN`%pdUGj$i<O?ec3?zJNO<MB@Gv)rRpT!!}QL>8!jrQF2qX`2x48en!*0G1LsS
zIX5|UnzlBe+I5prz)F7wdDxf8^T$^rd9Yr<6KzUq0~;1L%JTX4r*C+lxuG*b=q}1A
zQp)U&x}Wxs$BP1sJY5;Uy6EaCr5$SPd;^vmDmJo>MP%|nvd{6kHf)7LgrUTBB)t%F
zlk6j6{^&tum&R%ymce3ycAOTY#|l)FdeFO-uS(yN(Z6DJ)RviW>@kWDp3e%E*-9XP
zvnXox2`z2&+A81GC9P_*vC5{jxN5t^eA<JTe|U0FyM;$7LA1iUo_>R=D?3mFnnj*`
zH7#1b*`mQ0AhBpp8h<XeOjV=q51+Gc;Y=9}Uja%LxifDDF<N}f>bO?MM1u{lpb}~`
zNgWWiKR;Ui$VJfOotkqHKolHrtUnbopb^Iv`Yhfhg?E&G&c8tj=f~4*t*mtl84~%r
zy<lGFI58SlXMeC0FBoJn&a38$uR5GN;G1jsz?(6s0<G1=wsCvts)0~;M&o+fUU&GB
zxPN3>z*GBBsNmj_jSc(rlw}!~qf<%ee^%z*w_he4e4)w(%e#9w4)fmaoY{~h_0c_C
zI}<^>#pzj;AZTcD{7}MYdf-FRKG7Kq(oEQn$#5#sZ{2ldFkE8#+>FEw`VC9_C>{~V
zeeg&Hn3Lr_6~?XT;gOWV=yXIo5b?~8#zn#yQ;o<&oJd&W$6p<pR<V_OBmcQ=B&`ia
z`}n?(&XrG0-Ar}N!H*)|c%6kBym-|dc2&#32j`OHq}yl@<w%Z$eUD$!b5R~X8mrXu
zC>@3URPzG-3g6#g6AxBm-F=}-Y1cG3ig3O=pw)#3)Vqk_zyr)FR#u$GI9V^-@Oju2
zZF9OP|7C`dEYUKB7j|4B4@mb&EQG5aY)BQ@VBhz7G}-RvB|j*S9VicBq?|zp15ERp
zI?t9hVZP0>1nS$`-<|7!uRf!D<SO~5{>;`dNXEcmnJh8U`OlJa01l~#Kw#$-L8#v8
zzQ>n4y!o(QjwSu|Hj}>8G9w1<cIMH#pUrf0DtwdX=GT+0_6%o!gJgp4ELSj#zjDC9
z&c06Q>%L%B<uZY-tq$-wUtJm;Flgc3Qp7sOE+^eTUp<Mn=F4MIL<kxZDN@<MwsXW8
zqYE!<QnB598u~xzC3@La8MiJNAVPcu<rP-A0TIh-PNjFET|-5)@g;5V#CeO`c3CS8
zRa?IofV?ZU=H9ljo7eH8KUMgr85X}hb4@PHXTdx;4~aGXsxwC<C8!dDV_8S{ZOz~b
zR%eZ$7JVTEC4y&GKL4?x90$ZT$sVbciGqgR=?ieqbPm5VzC?wh=^<qpz58N=Ih_7&
z5L_w5T>I82TUQHJbInGktrvKwtZkCg^ta?OW7(Tsqj3fs<}aB<R}$B2C2AMeeyT){
zJOM+;2BiN!Mc@0Thq;5*-xWXlE!V;*xD#`q1}bbI@^!BEBMDTAwu@nQ5U)_OHG^l+
z3d>J67)xKH=;a$e=8&NbQJm|kz(EDa<}To1B#{^_w())L$w1J&{ro!L5}DjBi0}Q@
z@yW}t_`PYyIh<~f9_;~}!i$1V6#yW*C87~4*7qi!i5C2ms_fyv<3+}EDKQ^Nm~W$R
zFw1z6zMCS2f=r!WHvi)&rxR^Xj7-GSa3LyfMr%1Xu6C7ZAN3h&r5st}e}m;d^eaXW
zrooX#e+eSRycmJL=&O45%e=Km5v=~~#@(=Ml<}pGoF@$}qlLX*MF<x*l4}0KiX2G;
z!&$a&XxSD+81YQF>;3t9q`9SAs6vhBn_E;H{<<F}y5)7mUb}wFKjoiBExz~EN|yU}
zBoCsK4nEzNG*4c);plu7o3-dn`4F-9#|A?b4EOOk<~K02W;VP-97x3L2a?aJXx~Mw
zN1vRs6Zp!gPP4#@9VyrJz=~IfRpma&%Il*}z=V^zM2hGneJF>&!J7aM|Fd(vT4AO6
zgu-~QB8s05GjW?99bH+=*b*!nsgTq9v+#}WTM!g2q)gL!XW~H>kaW|*#qSE~=hz>Z
z;c2z=HYlmns=~w#qrv?$T!3i^45QdR@Nss6D<Q5gSju?Q(7aCruN#J)KVT}D22Dup
zUE&Wlgv+FNy~x!>CANW2|9uuf#wXVgg(<g@0%fa%*L6*-TFy1U9#rC1YH3`<$|c&r
zmubpm*WHhDJ=+u2k^df)-e4o~xIUgNcT>B{aI@o2)*B1!qt_>9=2;9UErZSc@%d>;
zU5y2B0N1Qe?||~A;yS~1HI+P0-wab?vg1}N%DPnaaDI_+RP<HDmdVi1AP!T9>nqPc
zhm$^%{lJ?t7<_aEw4D!z%%XC!7M0fj0f;Vr{(|mI@r6rC$@jwJtE^oiVoY2J^UO&o
z%0o)H7`4jqiXx(7ubKHk@u<X!bcgV&JUY?ys1gChGF{DvXP+ILvJj^XbQWSVTzVY!
z1MCB)OVxn8j0pa8#(}J&t2Hq5zK0!_=Jf8FMYM+_9qT7jG2C9z=uI~C5iTgav)p;&
z98(S3_K2b*SYx7T#vcNuMmH}l**}6gsE9Z+gtG1;SQnTP6j4e_vOI-9eNlwivW|N#
zjyJ|En<+xLWxho5Z&2ksxm4eH$$r{Ix4}VD%HR23N7z31wdhTH$>lJaQuR{(jLUP*
z0J@XdTd$J}KN2p?D37~lf!cXM-p_Z6zBfy3q#1)XET_a_(_%R}1r!8^C0Z`UCk+-x
z^@5W%><3r>_CKZTKw`O-lkuNDSZ|Uo;|Ow`ZETnydFEsg2Iml59sHeZItV{u-GIyV
ztny-=;m6Da?lu>~+Nx1xhK=Eek5(~}xZFctcpg&yy4N&sNb(={#PS;9UGBg>mK=*n
zzBb?_f3W$2|E1f@H;hjG7`M>B0$n>22G<qDe8yoLiXK${tQ-Cwoy)lF%;c|%v4_qU
zIjvj;)k0)DUBStu%l#gqW*(it39u>rPd*PQOU+1D`8MPs0t&m0E7Zs?EeRKsQS+Da
zr>n7DozNbq3AU(WaL|&VIIv%GJEjcizb~~JCmnP%dQ0PUa5;DFR$$v?4|J-4@p<95
zT9N1^kn>mEpM#7eyIe%2xi_#@C7Q{T#>(jRNY!$zmSN6E{EWd8D*SRYWLKMHnKUTN
z8`7F@*CHs1>;-rvS3|h_--yvYwlM-upB-o_aULl6pX+|R-gKrpddgmDPLL$z5<KHt
z+<dQ}r9-#-6dGYE6AFFpo9ZVScERqqe5^33_KZkA4WLZ*$kbw;GZhM)`PlpSx@R60
zdUom+tv=;%b1)QClWYSH14$t>!4En<COSW!tE4K+J#^;a`D{fl;cKHl?9@Rm7{zdh
zU(ID$G&@Hrd@ajH8Q2pD+ZB0NYIb)AWjI8CchTU99xHrW#Pg-`<ZuEFkWu(0u}?7S
zsb0H_8D;$A6K;dIwfwsoSHFRi^)U1Mb3u~1>-BEe9?zq*(s`bq$dc^R;{#t&x1zSh
z62BJ!MYkZD`iYiPZvO~#6O@_q#O*{ggz4)j4B<_K7Xl?yUfo~;|8sSzghKw^<pO>F
z-K;Aep1)^9lS&<>g3O(8M@|d61c7su1I0VE=9bm>bV;IJvKP-7SS%xjz~BM4d#ArO
zMq#gu=O5DmX4hNTlu8DbSRfYU;yDVz-SVqlpe$w;^BJ4KG6a}X#KRwcIv>KUbC@@y
z_DRGufj>DHh|-hrRCiSl06&FCFL_l30w+nL*|NEaKc_KTZv{P><M-!}CJ8~4RQgBR
z_Yd=)dIc>WSwR%jo55y<{*<X+t-0hDuj^YHgl{_Bvq_!k!Le#7wJ}H3J@f1Z;~x98
zqpRXHi`~DXC)0m3&0l-=n=uEdgX~|@J$(gc2;3(d6RLD_*3l^PxdZB2;jVSBW4xZk
zZ3(&xeo4?WeT`B|RQ@d{ntf;?OA;0s^3S>@j&ni$d4TtYu%YnNC-b(LZfqam>{$cy
z;>^RJ2Fi-!u%e*scOim<OP$tuI$`QLc}B7<nAVkyYNiTltU1+lotu%yy`vO()>%J;
zXS#!VSn&QTn`Nx|GlBHyL9eYWph-eQP~~-2^4pkJ_(rkTY&oGn(X2|6OM))@uS2op
zvFp4e@M*+vm>Ek>;Q+=uwwGLhH3Lwvab|8fk?~s{d!%H5{<)-`CxNnaLl{Sz7hbTZ
zn(A&!Dtd$KS*w6jLH+5QF*nt`><}f#Aj%E8{C%*15ZSON%_6|wD9oy?-1LkjH;Al3
z@JW_cq5M^A^lo*&P0V||bNmdRBG-QP2MD;QH%?!TSOhZ+qjL1Iks|&l3LQOE9{H8v
zbgna#qGFR+eH9HC@~|X;TYxSNvitu<creMr8inp6AbfA1?Kf<@Bff;HP~};kr)}tt
zg~4u~9czS78PN)-9~sqg172+8D2*hOKAwdGN<vB%a!8j~q<HUyrbfSNb)X}o*bPSm
z3*Q^m4OQ;1bJ)3lF3F^Q=i=M|PMYBXwF*7vgtjmoB5H`ENO{m0A!tUaAR#{nyTmW1
ztq!Rzk=)xo*i$}`*dN7lIXV$E<3n!02;Rou<1v@|19f*lH$1MB-A9Xr68kQw8$XL?
z-^ZK)g&)MJO*{=D+rCg5n8Biqr=^UPzgK%EoY(2W+j=sTQF~cVZ<FN;*~+Q!ZWXzE
z<p7;baf_O*)B8H%E!FRS3d&#Zf`H+ymn9xEA4hOI_l4L=%x7Qi3m|00aH}BtfDtR^
zbgLJT*Ss!m{yN#WKo^*Cz~4_C(mezQ3n|@bT~*I+De|k?n!6AmH@pKD^#AlwMvXOK
znw$53<ICkR^XTuy&x9L;eQ#gFxDcTOp03G*kFWNCjB&f|cF>3w$B_yK)Q5QyD>`UN
zTWm*>e0~EY3J>=GYUPrPqT|zrm;@rLSL}2w{KlRgfEqJ02HO?X#ruWs*gQ)#=*qg`
z+f_ATnRc<Z60GbV;Thnlm*4(lz05?D%<4lG0XtClzfjO!V=#G%f*SxJa+hc^8wi2N
zO&i9<!m#Kpu=D=;-GZ>fC<~|hzAX(O!#-FEtUCQ67d$kY8lK$Zk+=o2VS{2RF~PT2
z!g+C``$6h8H&<0XGVRFjFNvoe%K5sS0Mt>lmkD(*0U@j*xIeqnO^^-p;|Mg8=XI3)
zdLIG}C*v~rj?a;9k<=#XuYZ;hasBu{AGQhr1|!Uhlh2Cb`bytT%O_sXgL*c|DOAKu
zQ9oewe5=+SCPS8V_sd6PAL%iTMV;6@cpS|OONgJUwz+0vVM}X4#zU;fC{^=$yE7BW
zSE!Tcq6~7xD7G42e^r0$b03X{y6Voi#V2wDFC2URcvY(LWuROQ)I4r@0TFK={-BF#
z>|<mjoZdCQ@KGAsj-0UbLd*7dqt9<J==glz#jG>4k((HEkQOCXIuh!NuCR>|-P%CI
z7LF~r3#c`F)>U~j(4ficY^K~wJApI@>g`cR8J-?ri9e7ynkL-ppo5o<_-LMDat|BI
zmOO8^ux*7B_+!4#V+;7voXK~z<zw3kF7r@77XdoC`cF#{8e>l5aQk7x{ukSt5^D!N
zvSnLN67VRxzmbn5#}~;yWicK>u)l8Ll#fjxf^A6CjkOQsG+Ri{u^jkpCRsr5L_;Tk
zI`7XZsu>X?-hY1M#!NA0&|^8ZkvCpY_78ELPnv?>b@7Wne2OxZ-oK&YdC^<W!V~94
zb$_b<k7$V+O8<vG)6$+k3&UgFz-y<G*JQk~S<AUCBf>Z}Bwir+^A>k;$e(<_L3fPb
zAP=Kg!ZlbrBjRo0EgLEPafwm>AI?R}E_bH7Sp7A}AN<Hm<a<hM)1}Um6$qCb3sQF%
zwb-+n2Y&MaQi8sX`^jzCb#wGKFVw1_B-#!&W~@mn9iGz`im4?Vpu3Q{)rCoTD3i?h
zw@S$Y(k@bSG<=f25pN5nmYYtTA2OR4y9B%bGi^<CoXe*kyoORowc?li4O!ey!QKzg
zM9()`{uLM^C34Iq=%S<gK!KK^0rWZ2uEd9x%g?1gUI@CH`|%H!BG`t>5N)`xMV1VZ
z4(QNpSnaNeHSp9E=LT@1FSoDEJee2hsDJZ;y7L5^>15kp=2Ubj=R{if_}Xq4`tm4Q
z!a^j?9iTNw0h(F1*;M{Yy93XIDQuoZAPX}paMzemv%zk9;o?KQ(pPc}VIQ~HX2H;1
zhZ#A`8JkVavJ%rK3)Lvp+`ZMQ!lyy8m;^z4%N@ookk++uo0VWskjNJxG0j9y^!|y#
zkJqBiT|>g+7E12$UTE-PSz4!3N2uDEOPu~T%AfW9tToQlunc!+tPqhO;B?&cCE;OB
z?{>PkK|aqkKi!DhDQqV5fUzk*D{L%(FzVuF?sr24n7(XSkzW~N>+Fnc{k<+_XtK;9
zcxb+VyQcT=^crQ7><Nu*89lIE`>!(6aPf!ytYzHG8I8ORQjsNvx`e(m-Ycr-lv*1Z
z>*Tti0Ye(~%H+q%hjU7ZgJ6;~wwlq4;-tM1^wQ}h+&)14<$R$QUj-sD*6QmW>xj&*
ztBLm~L2bTjhpU>z%MC;r)Xry|>=GVyd|Cg1?PD2;>80#NnPwKUi*(!K;!;BiA~%#>
zm*zg@3AE)6U_sqVwx6Z+`EtK(O2`*BDQEV*&J=~?ber>}R_*8@uIYZ6py6bH^&x)6
zh;;occ>mOw)Ux(iprLXCc3tX0nPFP)?VTBtz%neP&NR9bn*K~k3a(7EoUsQF`P*mx
z3{o$Ry<`^*VcJw*khzir1WO(#286QMaHp3r%Z5Ia0A=|XDP8?FYuCUn0#o9+Sd&zc
zjyNb5%oo<9PfqbO2o!X%R{Cv3LN}iKH5bk_8sDsW=()N{8(_XcTg$#7g86{E0k_1O
zF|F$^)g$N8Pu}mV^*lXsBpJB0eTx4@IOYdwiIYPe!gN(pJ>VT@{-p;|$ODV$Q}9e{
zyj_aU(jCs!#`S8!XK#|pXF>&LWL-NH2yGprt{#lgL<$N9)S=NcNYKvr!d<wx@d;0!
zPM~l7ESg#N$={YI;{nu|jBB5$U3Yw^9#9w7VXTz+nsPC!3y<6rlY2|oifDeFm09@J
zeh~=CJ+eRYD?Po@dcKLd?fsdbEIO`}&zoyQ&TQOG7oI=gA-x=oK_k}6`1I&X_^ZKb
zgD|Eo2As#2@v{$2Q1OpW;tnoIl^H`J!sZBzh|VN~=IO&G;AMZHM&4gR#zk|<JQP`A
zj5_+#rXnT0^t#&o-Z;ZnbRti+sBl>d6oYx1b6cCfko+M~4(HN8<_4;n?<8k6TwC2O
z@)gSf-&fcS>#+n&%G|3x@H-0YV$f`<%Fiq`w-<v2iUn*(2+Ks0QwZ`BRk$l@rBK&=
zn_`q74I`f)#l;3+$=`pVTtgk8O~?k9gjE0zBWQLm<-J9L&JKGOycT6b{K@h7L}O-_
zFpk5*!UxCHuQ@Pq(QrK}SRWi8enVC!myKBk207Sw{n-VIVNZvRE*eA^L&3x`7;GUo
z>$Ie@UkFQOp<4zk<i!vRm%?7uBKR{<UIuB<phmi#?{pYR;KkWv1cG)HMWm3!DDu8w
z$B1M#b6ou1SScl4jB4~j&c$~KlApzs9WT9vDMJD*aldBo?cllSbP%oE`t+sF0u+G4
zgoMrC_^`Zr^u@H_vSFTJ7UAZ_GkQz)Q0Qc^aI^JDea()&fk+32U*C+&Y>;*3-xzu3
zdY*CdoNLnNjM|r^@08+JEGbsiY4jKsOj~UB@0*Mbfzp+$;o7U4)~>qVhTV;<yK&~d
z)k2IzhKh`Lu6(y-nMSrwngV6BP;V#AHw@o;V0+p@i4o8674Ut+x=J6~rl)Tl`8TuJ
zaK9`@;3}8*+7fgEZ@)4dTmfBe^Ha(ptqmRQ^DuiXndi=;`)urAE%mO-g`#+mpgT+5
zxG(7My)YZ#)guXcze!?a0oa-98+~#aQr-(fr_PvQBl`WNb0vZ%_!hvMvHc)u6U0Y6
z*s{tW_D+%o$^-aiINfdsI>YI)C)W&UzN%kaQaQ~8{b2F1D2fRa+6yf?ncp@poA3Fl
zxv1nn?`I~3=w<MvAz#d+m4%`56GXZe^R)rz6`dDet!3?b2T2IYEm2It4|Z3A8vKjW
z9WfE1^=ncc7b1Eo5?i3L9rQZiR$4w7H(c1xazaU3g#3Gd@s6G}yE@xw$s?D-i#uUt
z4Q{(MyW<7rnXe1j4wwAs*lbV{5MjO(|5yDeySe4C)Xg9bgY=(E(MuM3V7BiyE2+eb
z15303=88`BVaV^M%@blx4Z`IU5s_%d&&rkUI9f}S$nWgE-ZCh2nhxT19*n41;<Dko
zV9?g*q%j+a@xolGHd^$B&fKW5aRfh!uw6+(>fz@>rvHbt$+(WsY?7meXxH`Xpu1kg
z!oMB&-LvP|&P{V~MHn>MoXvdV02Rgos1tjt5Kt))@R_=UE_U*s@9>v*Qk8O-wz!tq
zNH=Jqnsg~r7!nkp5{}E(DMGYQZi!(2rT%!Sy~pNivzZD2{rYj(B^D*z?``mvg0tD@
zHxA(Rmd4PBR6Csr-#4p{iQs2WaSi}t{5of@q3LA1!9d-xb6aK@26J@v3}08D_Wk}l
zL5`3$BFx{wV8VW>^Su3wNT!2(oD2J(V^Zz^D!8J8=n`loZ(`A0%7RkhCJPN~6m9ck
z_7~d`(X+XN-{T@(nVpaEuslfF-^Lr=|3rxizitm4dXp&9$y09uu6sSsbQtJOUD;U9
zyqk>k<hu|tm;eMbc_CWuZv<M{&Ru!#q}@f2i8t&UREDV&h~d?iI9;sS(?GvFZ#+c*
z)-xs&Lq|TVv9RJR?2zxZTYdnoR5w_^>X!rX@=5*HZ-p61(MnwVYab164DSvs!0?px
zuLIf!-0w|X`B~lrssAU{?}keIf~seJeB?>uc@%N3M*mHKmy5nD%%aR0g88SJH$X~~
zWyg!AS_xRYY2X-s@UW)P$lLI=VO}6Hx9Vc%KZ($5QX6YpQ2&3gOXEqPvwVODd&9dk
z8yp=oN~h^36@iy!-ql!7wQ@N;1WcaoC~au#IoGUfnbc*kQ1%tw8eBgFJzeK^d^+yo
z%?8GTe(FR-?-|i$g4U4g2D%JpStlGw%H%iS)99+OW_m7X2qAW#;C-@tY)q6HUpuEa
z(%T>vHE~`E*&Le*Q%X28wed*wwjLu}3YpX9$T2iG4=NF<uD8EUQ7%q957@%0U#@5(
zC#ncp9jTx1`YL!+zj7SiLlQK<Q>g&L=95{#7n*%;tVd!Ww!S%Qv=Dpg%ppkWO3{0Z
zkbvEX#&Cw<Kf7;*$vsYH4-V1VB$XJSZTxo*37~Q+iB(%Yg<VFiU>G7}X}5wUw@cNB
zst>`Gjvn$Uj`PIrJ(UvoDC>b*?ZdtW6`q_y>(Ee_0*hI0Yn{ulh`S9}amTD2?m(v&
znW+#3O`+O_5L~9j_Dh1i8xwO+WPh?UHBe^%!$$t+QpOQwp7WcwmlvLyVt9e|f+(3D
z9|+y*PPQZ)yrJRhWk>jxr_I=-H}vhp4+!(ucs^W1&Hi-phmmqKa-VNSw^saY7Oqh&
z&<Bi8UqcW|MR@-lx|ws2N9F<ecQPi4G3Z6Yi6`Zc&YLcd*^3aSV!kGsVg^<yu9q}i
z-aWO+AKp`SrI>8dQ6uUx7`k~~I-0iMT{TW%8A$DRzTOCrf2`VeEMthc^>YhkA!Os2
z9!-nrEgGpwRG+DIxM%lV78O3ki_u6Qn>zD;GUzIVUS4G9{}TLfs+e)e-OOm@K6S<g
z$`_!JJrq+k?^U(6v#kCyNWuB4Jx7E-ED9&x|AHs^Yoo8nxb$X<u$tk-&+wLdH_Hp}
zd!fo|G0oX=l7~}1T}B)O)8z_W9uD(c8Vo8GRklU64w@1YPJx%^uF#n#hTVXqif*ic
zM|-peQyuc{->Fz=%fuzcqF@&kPkBr6R#d(hh!1#;Q7%hk)^ul4n@t{rW`TI4K27x)
zQIr)|QoZ^He^<!}bkxqX7pfL0FNU&=8|0GeLR-GPYpE=T4PSo|_!H49Gd)WJ5b)s7
zOwj%n`ZK@%{If5`<V6{AZj^4c;3mw-@!ge64K=4{H|%t4&tP`O$FdT`Svg9W&e8Nf
znxmz=)&g;gzGZ5H4H9MoOCMQ(!7eQlA2i6rri@Pmr_!)2p5J!+(jE)%LiekzB-wy9
zl|@(=9yVovZ{qR6r|Fe3K8qnJeWu}Dh?CeqXY8~pzXkJdT&`su4<kJ6=;;?CY<A>#
zrrhG_ul;E<6tnfm+vBnilu9W@Ps6Vs(ali~O4C9;IoH;eMBgf5DY|mV9Ki-Zx~1@c
z@?v(htUuB=?r2r9kkklIVz_0o1$-cN^Hdz4p~d?5ZbSteXAR=Dvje5KNhGS5Cz1C_
zpu4V&;pdq^Qbvc%>wVw@FwVLsjr0Y2DNHGIUloo_@-&|fx%OXUiM&7`pK|^)F2^ZZ
zVnZSu9Celx_v?`?|LNO&f*i*F2EwJ|yN<^TX%-q^twA~GH<Pam`D>g!#`{6;G$&y_
z@T9#J5u)PHI9FC+nlcEH+)M4yag5G_j^|;{_F5AT5)3;^6kFvGvDr`#<%4WTn1-M5
zl6N0PdiCj*7ox}8^1|!CH3;tip+vvciR~`CsM4r9m`iwyPM&$CO}i#K#=E87U~$!A
z+<p}pWy@qZVLXuAz}g>^b!JpcSGx*7+Ps}z7Hy9=j>AzNO1TE!pJWQ*ox9=4V7w&#
z8(qf1nclcw!G~;+P&1RrwXf7rX9+R$2`nC5Mv8&bfK%D&FR_Hi(017r`5O}802`4v
zcKPv_BJK~c^|xt>>0`w2==@!*acovtC;TQO4FYzOKm5?QYI1ChD0<}<N>kSY>5DlL
z^jF%P5AQ%qIC34}j6S7=Wgj}1_d-1ParqgoRfGMJ@AT~jKbw{PT_jjn=+5s}{!8;p
z;@C&vP>llqQ=^n|)%#{0648nS6R*2p4pt*b3s~W+xq1?S!|lVo+_bgQG=%b_qZ^##
zzym(8(&m?@r{Uj{3X_I?KuE*ClvS7k;jn#35TBa%#Sj5|%1p~qC(Yvm9XWoemzbuE
zhyP}MSP73{-0GWUH>8Bz!whLQG{4+efIqg1j6#-;V)Z`8sgGLQlVc&dXW<0TwE<PC
z!D_rsBgKL8CXZ9r<B>?7gFh7U)5uIks%r6bJ|?HoeKQP7{F&}GLo_ymx)yV|@+qDU
zJIkMkT>NFEWjXy<*!$Q4zakx9i{{^mTw<$DthWl?0ks>BYL?1(C^j7RrgdIYaUl-r
zhR0KuGQlEGZ@aOmA51Q%7w?c%Nt?v@NPa&eiK2`R!m~W83BAEtJWL5|M-d`B^B~9)
zI@zl?u1gS!&E80-7<xV0Qtxw^*Ph5Dyk_}BdNqX3yw+uMt+VjzPa_n8i&S=pYx_ct
zD@~syUG#kvrfnn5P?{+YG3_t~TVYF>=yhEr^2vb4L!&kIk$&f{-g{R_N$4w;R=iB9
z#C2aK?uk(j@qen1^4K)Wmtsx$uEYdlg5ni@E6XUoT|(w2-|l<WG|^D&osx?e%*BZV
z9qu;ot#-Je(<@D@VFpX#RTHBv%u1=7w#8^ge@9~`!|1*vP)!IDUqn2!)Ae;${+?Pr
zDjg{N$Bmp(e(My(rsN8Gma<i=qzl~T>-CL@w|WV6AR>%jEDnQno7<Qa?5r|t(x4hA
z^YZoQAyn18M<!vo%t&jerdu4fmo}ViB!24DL*30eS6EKp&w~pjUhD|W*-JpmLv#KH
z!Sf4}^yb;4{^>&6MP-smM;5K_P5Amg&HLNrj27uW;o=Ce3d{b!ygg$LE+r}dmX>Rt
zfY{B1EbwA1I2D<3S{iXkIvS!cTKyoQGLb;MKw2OSL1#VaAx-7^+SkIUA;|OR2#OfA
zP#7bqUv=6|SZ#=OXK7naUaGPki#kiJpi8i#B<5k)VSQ;2D`{Q<p=*C06uyD3?zH%Q
zChiF&I+a~u4C)9XtQ-3*!`JUy<%JpiSAoIaRT=zOK#$iI4AOPt87&|Db73G-{Ek5e
zQ(j%1kxG-`nq?*Nw0Y3azRKq`fV2wT?B!`_G-I)}O`xQA3lyU>Oa~Vci!!eLIh{r-
zzu~-r{oHg{w(*0GP-z<mc}2w#cnY%pSS?+R`^e{{ufqw4U>rf2&_IZP9=R?6%VrU}
z8eEYhvRmhayDTG+`*ry#EYTMc)%Az+*9f3a8#)UGU({uvUr(Pe+BS63y0C5-t@3+#
zLzf?7cDUv&F#;IZ3pu)}h$FGW@N(R>@>dX!=F%CD?^`oQD_@j^$SIkvX1ldT$;7;v
zzz4cm!#mK-@Sl;0R)QZ2K2)nopi`y31X-n_5ey-fQ@f@K55yO(Y78pB8rmmDC*tY2
zC@XJ_)~aqeb!EUu4!3Ots(t{YT%yCDoB&nmKA`i<eu4Bt{)`fYR|W@f<ppMWfj0OF
zdvim*n=*7Xz6w&Lt;8UoScg(BjnLUYLZFaVVU$Wa{RKz#j~%3D`GRi#6}z$4xbw6l
zO!JrN1(AnbIuj>Qp8mJm{VjO}K73IDK$nWL2kC^d9;g?vp-c(w9q^BDX<nN{q{V?i
z?Px~`eJQtyM_B`|Uq~P~KI!+8qlXNr5gyDVeZz735ZtZ$u-D#C^YdrVlmuN@@tOSQ
z#8^&_{X5L^GB3YfmYFq?Gkq?T5ZZB@Sytmzs7_GWocTpJpXdeAe2h2zokE~Cg05{n
ze7a#t1s&18FkEKYUDeX?%6{tF9WUYkJ`2#-IjdUkJ5*Ng7rEZH@B!WKFYu*S^B5D1
z#h31SFdM%3j!qi3&p8Oq-lJQ(>#MQhclPUk5J9D((%E8oaUO(u=RFdy9iq`5_5Q^#
zapBXsQRLU~0i)Lj2qAY+gH8E)QAKlr_X85IC0Ep{YCZZ*>UEzpuY<p=?ZeiA-)UDy
zZV9aXQYH!!Uz{bE2wY*9fRRNvz8<g;ZhmBl{-BsskCV&!4jwr^50bKyU7R^sG)v-1
zes%4G*b5y4@%KF>JY6g5L-8ZW@M#*dIH|EXLt#;7(t(`dR#7p9OY0Y~8_iqqpW=U^
zQ~VFA##%CfT41E%6>1HM=l+mO*O}`;e<FOq`B5#$XsdA5N(I>H>FC2V#T%(RhTqp|
z$wC6kS>eEm=!OEd%GO($22Hbe?Oq{G*-3R@IJSCZGfeSRjQGVyBx)ELrS`E3g=X=0
zT={sFglN=A8qcx#RM-@#eTZiWVH|H&T5}xr%npep2V=Df73R%n7ahDrS18n>=|fDC
zc-hx&l#SG8CtBCVxWTir4{Lh*2TA-~&JDzNN}`z#l>D`0wmbWom6O=lYpxZ~$N@hC
zCfsf@k{r)EX*g5<K!IxhDMilW#3U>7$O3mzyAuLX>y`#)jGOB2O~2-u;h{^Rnc>c;
ztED58nI^EQ%))I2oZrFsXV)Bo#=MV;*j2*QTG%5Ae<@bW<Au@S>4BEp&*V%ltFF2&
z?W~^j<vl<XgvE2SL$>a_2p#x|Q3Ux_JOo_ASI8Q$U|}AXr0?nGN*o%GE~ayT_W?j@
z!f}oHZWkp%YhxM+miW7kub^$5(7nYTWA9;YzWz6)ykhq+s(qeBMJdG8jw~Q}k!Urg
zA%UWoEe$8&*~T|VF@(3Hv-|K2xC+J$$n%aQ>MZ@kX33?zktAnv!6RUn|2fHc^e$oZ
zRKyG()osJBTt&TTK12_6oZfBDF~0Tmu2^~Iv&`EEc|Fu@1K%cu<GysGYR`uzSre+G
zL0kV-1sbD5r;DK1eiK2#d-kK}9$@##(e}bck+6}aj=pRQgXqF_dha@-v1@`Ppw7I+
zF~*{4|0er8sEC*@uG!YO&wg-G;hR1JAbLodoU~Fpl;IyNdiEk1Fw@VjY1uP|-&fdB
z-$;NrJsQXrQM)UQ01$RRT0PSnycSzx%}>C$vyE8r@<vqUfYdv7PKtvkbZ27&>C(W4
zZM&|=Mz>gIdJFsk&9DpwfVGl;-S4BaV_%$_4CjIM28dic3?vGx6+*_SyST+HD8fV<
zg~cwsn2|lp0NmR{n?1Nm0R2{4BEx3B(O@k<dtw&s^ArrbO^^T?I2Jg%vT;Xq^Q~@i
z$pPs2ZI!NeZ%X?g@17fWucop;mTd`5YR%&fj1OW6pGFbS!07=fP3Q>h{sAkvdu}97
zLY1+I4^g0{>;!*4?x|z1I97z?AauLbmy*<qj4IyA9jIXpND3f*OC?_L`OIB6^r)3O
zgf3<z|IiNoMSB$a(Vp%n;w}gLk9j5OhrGg-3zJ>auWR@<a50Y)-;(qsSf%~m`2W`B
z-GFWeBbq!A!i(^Wpsu~Zm18tbGtSE5g3qUUZwpwb=mg7ah7gg1E2OAg)3HKS;Qd29
z>wy<30J;OEK!P}e`_3~H;+Q5^1kbKAWwTfW*`)s1h}uJ1+U%s2e!}jeSdzl5sz2Mo
z3&J?I2*)A?>!n0%Y+#V^mT|~Y=?E~oSNuemnEdJ#Pv<~$xA%aSb=xOE={M{tMv@@q
zWn$#kTq*g+O?soHBP9dY{SnPF(LY>rt?-r=Ij;U|R!5(7Ve&JvDBz}g6nzbI_SJM6
zeT4}gAQ!GsI`HPkD^wIM1a{q<A{eia(qCMU{ugQQ9oAIv{Ea3ALkAJ0L!^UJM5H$r
z6ch!iMnS5A^xgvj5dmoeN=H;cR9Yfcq=XuZfb<9mp?3(KKuF%h_gCI~@4q(>56P39
zJ!f|3Gqba^GqcVu5!as-rBRgZ`11=@_dIA8+rr3~34*R{c*;n4KUqqCi$uc&(i<Ud
zvquLKZ}Uhd`!7~Uy}eQ13{{&O%Xo|#!CYy*fjL#|GOFtZda(H=a9B^L+v5Gv&)69x
z$<EaQ*M$-4mIby8n)q=%gB3;<3~LMOo=Qid7(aOZaOY*bl6HwzDfz8I;f{5|3h8|H
zJ1zH6`V}tfK8)&VKc=PE^amAhZ3c+K$PtYHJ9<BGSz$AbAdH!iC|P<+FU7>hoLHEu
zxMH09gxsVyd0@NonGKambdeBvaeAJh?F##F#}_ejBG3qSdOJn<4h)bj8(HVVq=={5
z5fcE8lr}m*IY#s}PeV6n%!?joxd1|wN4-f+R#VJ|>vZohyC`E7Y(W?nsz<gxk59e_
zHiH!PJiP;ZxsLoW_Gy|$_m<*+`r!|$whFKA4I%Ktw$|agLwB?Fo&qenKE1*|+J6X9
zvpC!iqpSAbxK$zuF9zc$J>w&ge1YdD%Q^~;Fd7_C@YL;@SCJzaEC$JM$v|;0=<y3C
z=AAouHRmOD-4KZ2_JnBdCZT&@l^4`$fUaRCasAG5cQ2BR@#}4o0mag??}3?#!dbo4
zwSe)QeRUHrjaMork3My{1|o0~TyTrZ6uMfN1gE%j|L+vhL4`CUJ_^!gZfZ)+HE4zR
ztH<tFLHuV=J`jbAKwAshFG6$v<i}H3Mz1Dey*FvfAjr#v(5Sm_nezcA5mT6mMcLDD
z51B3e2;OhRrHAclv)_kF9&tv{ptthRD=^J`08>~nClVztGSr&%<U5{*XDqtb@Q|$Z
z1K8qJs@8&6_2r|pS@b<>Q(7L%sD&p;78WvVB50T_D-*Lpn~Z`;nN}#oLDV42WbfL_
z3wdtP5g)<2_0;2mPnq4=LT-3MnVCG~61TjS>e%DhygVAEsf?O~9@LxXg-z+gm_|>&
zj<l_^0?*fX2XDEqj1^<F=qB?YKa-zHjOvY;4nN((<^th-WGHV0=L#(#>RH`~5Jxcq
zL|trNdb1d7J-xDfRw|3F961llY`6b+nX6+kR3M6fz^k~$OXU8wKq}m~j<ANwzN<6%
zf4Fr@=SPi27&PoF9SyJrN)d9y$`%CX#VTf=Zt<JL;ISKbl{vFZxc)KQ88OU8(`dbB
zl1XfHXR#|5%%2NyVNgurZEfMw3V$$O%Xgxo@jzR=?4DXi7YM^t(8&E5`x9FGKJ!tG
zY4WON*HH>D&u(>?g5q59@D8aJ*I~Y<uXI%3n)dTUj6z3${@$`XbnUCAR@?3=cdgi_
z^`8Vw=343X8iF}}`Omm;mdGZy;u62{4(Gv;GkA`DWA7N(x@9tZ6dYZ8y*8`ct0Bp5
z=GuoJzNYJ@fpnxKbO$-;UAe{YcE!A3R5&Iw9saK8i8Gn!!u<!zN6$;P1p?$eo#9Lb
zyR{*)r)Mhs<mbzluELNPs|6AF>RZJwl)T~JbwRL2Kak-1LV8Apig<a7Il!SIlLJpo
zq`lnvYH3IfnDe#X#q|Y|z0W}SWKX^Qu-s$SZ3WPuWD>%40Xq!2{tSf4i|yUJN2WML
zgrNK8>5Cb2uy+Pos0{x?WpkjG;Y{#%buu|l(MY;lhQKln2Tou5L**lWj2asCFqv@_
z?89@k^>S_a`OqNmS+^ghwo>kI=<hvo9h&m4reXmZjwtr@1=&>alcV9ev&Qs}U$D|l
zPC?Q~Za4=lZ}`*i_QT!JBeo)^=`8hai~b1V$6BsY)LMQZlb?JgaXMWI6CnI(i_~&C
zeRZt~w5DAfKFjnxyB!?Y>g6sa9gRA#w{73F%2wt5VMlPPpcUxn4d+8H4hF;*-+|Tp
zUC&7t&HUfCq)(sVH`t?7#zB*c72LiuH0mzVunzjc9!Y*?{4;zc#+(oU?*IC*G;QJ>
zc`IVf-Yad+07?$!@@mpg@L^51nmduOQovy{5n(dsNafPsbSgId1W+xsMAh>zA#ANh
zEGpfOpL#)&LsO8i{M<q0bohvyqRTg=^rW`!$jpRf309@VJpg)SF`&b|!gM9$ufrnf
ziJb7px;OpF4)fp-4H42MMcbKMuHj8j#ftk_bn96hq}Fd%(pj`uniOvqraw>uop+;>
zi*U2zL6-0cD-?ZIJ3Ystwtw&Y1jGi+p``wqKG@OO^9Y8dBZ$3W_zR2pRn&NxSZjp0
zB{X3UWNR;2^yT>qalCvH!bJ-&fBj5hl9~V^@J6gjFed$pZnxEhkyyf+x=rpT|7VeK
zdxnCeWm*g0DMnp<v3pM=_JQe7wn=ga!YXBgX_AWUU*hg7;HkQ>))p83=#@Pc$)RMX
zkq-Ce!K<RV>X(#h%o<8PUX3}s6AeIp-7=Z#pDB4-Z@VDdxk|t2i{8t;bCxNAgb$19
z35_faR}Oh>{4dI8w6$<;Vs5d+V*e%D3^2^<vQxd_AU?d|gHmm7@%%-*HeY@~@rn7S
z2L;8yoPPBH6?iXovtaDL63FvB7buM8Xrj%2<YBXuScGg;KF`v?4^~SjW-aQRlt9l=
z{;dE}MW1^7O}v9AOFt>J$4dYXbHJCj*fD-Enr)i{ocU_a6m}+R(=H7LGLp(8U)h=y
zwAF6@jMQn)98;NdHY)dCQ~o|g+E3|LG+G<3^c{#t<W<X%a(J-^?16kFS3I99q}A-A
z)%Ea=1W)IqAS#mP&L>8ewoMpy;tPM+D}w}!=5nkUpU9kKIW!DHGJ(S{MO}KQr}Uok
z$SxHqEWi{_X7EIQm%Q+8x<_(K_G_$Q-Io>`EA16paIzG!`Is|eL!k`6i$Qyus=9&T
ziTm0**lVuW{`tN}$I`k0hDUkpj(c5bqE20lFEG_IEvcUmS!emSnWtZno_lA8(xvHY
z|ACuO{;_sfi)IqE*_GMgSe1jN-+q8M1eJXbd=7yNk_2pYal_^i@Q}AI2Fb+mb5EbW
z|F?Q8GexSprd4d=Tjd`|Bckk}(9{5dpcTfob|h`)&_qiDO(q^|<_m0c5#I8L2`H+0
zFewIoHa-z?K9ZH)MaZ;)HDg8^S{B3Czt(FR*LY{Qyo$7@lDX{FqJqi#L+1^vbt}D}
zTUppYsIfp1AwZ<qw@NYT5qV!SFC`xzox`=&*Pdi4rT#0&sFDp!JqOM4Mu5JQhJ)wD
zQ18#7J);^~cyr*%%9l&6cv`yK<H8&8FsJ}jY4qK*xXK8Bo=@rZ^Yr0-xsO*?ZOQx?
zBAR9s5-`CywdCsBI(<Z06(KzE>>o~T$=tnyiU_ZeL^AyaPbM%?zxYP31T3eVorQNS
zu_y>HExbtGEwcx$o-_#)U~}SlPtoYgqefgmz$g>a<&#3UIKi+M#ZLV?-lMxGeJO{L
z@W781sOE5DLa@D!tl}jUU0`F7VAA+|EieOvsXL4o)%@uMrSV~W&!a%plpK)$7h{BX
z9uENBSZFeBZsO&wO^63k^0#wGr))ntm4ERKIWukoufHO?XitIl(RFRBp7%NXTp3TS
zSi;pNe-GcCSY6$*&IZL@&*^{2H~Jh3*oyb?k0hjgu@T5~>9ZZC?lgo4gdU)mK9Xfz
zB^ZYZJ}Sz2#jLQ(s+DyScg9*KQa+~Y9dk5_Dv-vL=?ME`jXpglX7~~}c_~v>m9k`9
zw~J)AP{|mak8u9!g_l&TZ*9anmw=J<Q6WpidmXrS%p3}(9X`#XPNU9J|0gRm&dGZs
ze62a}Wz=nzPi=RLThJ^dmgW)GM&klEzE-GKQbF$d<jmb76C9C+Fem0qX0dnclh@aX
zjr#qw)x4rcektPVOyH$qNpqV>H%^d`e*Ftt27dpb7ar5a=Ys|xTnAkm3=}u1Po7>J
zpK6hXP8rTnUNC19^X!(8ITPm)YOhE^3WYROi;(iFrBhYF%V!?-LlV0|P3eboy}BaN
z0N$M>OkTv-*t|#i*+LvySyv>cjMPC_Th9<afmirTYy!$LE%gQl43wNUU|I#s7i~fo
zq!?=kPKh&N3cPH0IBa<(t*8PC77+=!kC@TTT=13^h;oRV{x^bS>?&x-gkUI<iu7qR
zUS{KT=QCBNlO9}%o1iCsZqBQqiqm}Pd~RkJ!!pDKgR?B5&B7;|DO?tAN7r0G3jZAc
zlJt#mBTQJ4fuyo;t5>u(0uH}f61gGeI`sB0?)T~j25Q4!P11cPma^^dVtHr75d!;I
z%NinIo3Jlh%021sdI>MnT~52)dP^ubX@iY$mGDZnRNqTrjTEH%Dk@O)K*`LRxX|rM
z^2T*;vEpx=8R&Ey6J<9XR7P3(Guu=@xy;j`Vj)*97?wa9<##Wh+h_}Ng>JEy%;i&S
z#V|po3OrLW?WWA3tDWiJ%E5P<wW3S$h%YC0RBB13KLUBysDI|na`hV>Q8R&JVhYyi
z>D3lA2&Wy)rYE}*H4p1&4(0H_Ft0bA?Cbal@%&-TimdpBcOP^-gOv|nB0+&AC#%$h
zSAyL@He3_9Wj52qZ|RUmT11vpfFNh=F7q1vO|dPpa*Jk)LF%r$7iTRriJ<hP`AvAb
zz5ZItMOsu}tdi{~*ALVJcZ#XHJ+mV`UR@MSw}(h*iyH7@c21~Dymx5LRgBheu>v;u
zxS6*K#ivS1Dio$@zct^+)^BHv41O`qrNq2Wr|WvDoyB%P$qcT0c-OR%gVCOir|;Yk
zyfp^8q*RH-MLtnZI?YvQ2vad1gU2oOf*sgfjCFK)u(B<)+gOd^+6NO2tt~ALKiX0)
z(6rPvyt|J#I{D9SQX^%nh49|=;to1#HpN%1STwV)SEAL#;CkiDh2U^OyqA>XNdqPw
zo#1}}VS`&qwWNU>U`L_&Qn)BQ7i3&o$lgC6QdXB9OjDmQ3nk5A40+yGb9nuV$Znoy
zwxqSZv-12EpxjTcNsX?nN^Nd2GeJduhGGtzDYO<pkb)cxKi`RB)*{AUA~`HB5?np}
zK(Xw4zexQya=&}r2>j{wWc^3Gr*LkrNA^dst?)I8sgPySl7TW8t@BVbD@N>7+d@ni
z&IaRzbB^#n?(7F4e7G0`sRHYMnP02R#UyZ?<h5fAJlQ10@97jo2hr3gw}k!5$M7=+
zQy3N3v>tPU18Hlk0#s}NQ4@Ai$Qc#f4NNS+p0;sXmk2K=sGj%=Fb+xf3HdG&FjM6%
zLA(o2Yp)d}9?^eY!*5VJWd><QTQTr```#DDFL6yDvQ;azClja+qYJ%@AnA6L$ig(d
zY=?tlJv}cYEidRtgM3p+MXeWJPK_kw+Chnee~m0)tmave!oR?EF0;DvNoh#*?ojyf
zy6>i3D|p`DO4D0XgJp1@1W_^EHj8tSC<78p8^82T+?!OczICbO0zh>{_)zOciMMPU
zl*!=S6g8RKH5?tsL*C@dB8dnMGWUHBl%IkrI1B|(C%GMDTgEEKMhS#flKYo+0DIIa
z+B)`ugL?X4F?22xY%f?c#Ri8xOy3k1$t_8Tf;8zTbrw31lWdr!2$p;>=$q9wlHq%%
zAf&3Ii<-L<A)zm?;Ss7H0ak3q3rXGakc^AC((0vry-dmG+&5MEFSVx_L+#E&6SVqM
z?Un8>KG*P-_xKuHB9ZZln_z<=(AA;e^}8^@6WNY{#;x!kdb%Y9BjpserQKl!*?$W1
z4HqYE@kVM0X!d@M7+)7EJ~^ypZl2FZ<i1rPU*O&6#lQ@K?Jo?Rn@&e{+3YtSUgA(C
z?6tA!Mj+iR13wuoRf8;AE|cO~<;s6|X#>+a1G+0<4W-!5F+x}HbZ-!Z@h*XU<dA(B
zbQh%;Z*8!B4Sg93B3;1evitAFgRHk2q4jDpeov;O%#>xAYZhh+eZtl_gK(rkpuoxW
zWN8cgEdh1M=)8npi3VD@`Qkb8S98ug5Gme>2Fy)2XOX7>JCPo_<zotEA);z&KA6yv
zrpdgQ8%1mZT(ynA?1qb)7s7k}k0#KBL>JU~1U5ZrKx$|IeIz~)B*1ba91HZ%#@Qq9
z=+#fY>TQo)KW(k}sQUY_bJb}P7cCq7#$@K1Q}Y4?t`VCl-jQG*)O$!enC4b4o^FIr
z#I4~v$djh}=V}p3bacoISJM0|D^p_S@*}aZFsI`O@YC}L0V;gR^GE(MCsT!%%-Ip-
z`(Iz>L|p`O@gUI7=t{8ig;|1TQtxYA5f~g0kH8hqp+`7eBuOW4;#g4BVbl{xTcFV(
zBYSn{<myxhA_a@10QG=tXfT(mRh$a2<xc;))!wWzCOvBE`e6OikFkyNMGS?~I^j*e
z6|2m$PLF+xPilpZj~NH2-UvT;zx3JO1p055p|j1Mf+8shIdip#;@p=!<t5$Dzjye$
z;mIm~i3&z-F7g2v=SL3Mc4$h#tDBtA5iQWu13zanOcrl5WYYVs2iVR;jpld7r@XY{
z0Uep(IdRZ%SnK7v!dY(>DRTlRit6OMx#&aHOl!uRbYMB)WHlFFV=vz`8%AB$oe=pe
zf>#ta7Dl^<*W!(9R)fz(`E+3?Ar{Ux+`Npq_zl<}(d|H&1$9&}Gez5VakdBb@a9nF
z#l*RYie}`+$=wI~-}QD2ACqBix9H5))Lnm~d}~Z3n3)tF<O<F@+R7iM$gO&SY8p1*
zbzPqnz%vuHz~*+5;3cYxo_hR|rzIpz*#9)>jv&+Rx~V+d?3%uJw?GF{NV;>W6`*W_
z`BFzy7%M?gYSMj@*>xX3a6HvC#?A!IdAYBjlMn^D{G6LG?zs+W5&($(tN|&w80-YX
zv*IC`eYoxo=84h}HV*gaT-kmBnu-t=8U%H%T2$q|C192RW?aVgb%<#Bm%t&s_*Y5l
z8#P|`aMC2UwL_=4tN&QG#_`58W@2Ci=<n6;Mkbtqd%*tG$M9?nx$FaNg=QZ{F*VCQ
zT=WN3vF{t-%kL?J?kN?Nu}3vGe{DggZhPH7o~U|KJdz_06I(Vfx&Jh7<R<i9{iVqF
z4UV4^sBQ<}{K<@Qb^?7ACz`QiPp4TB#f?8&WU+wdC-xzb&O!tC+&0>xccC+SzJc4t
zqrAPF<Q-ZTmpPuFl#gx>=L{}yy7i<G&{-fL*Wu+;(o|&o{1e^~O$K!kg7@h1K6dl{
z2mF$4c0=1~$xez4Qjw6=7C8~I<tzzrdtirNQF_^QE)E-dS+TD1&5Bz2NoRk3+_UB_
zP<2OQLqqD)ChaKqv7T%A0;8*9%PrTpq5>~hxXe7BNe_;ilJ&C*;>PtnYmR31eYO>M
zQ_Dl+zdvHp^Qo00qFzVhE@95$w!WK+TOsl>!uY&4F)i1v+~XL%u@fnOe!Xhe;zF03
z{o4$Z%v(x{8w|?E*po7<DP1o;oFh6SYCE&3hTSI>Gvt3HSYm+UC}w-du&)q}fKPc5
z`t95Z34(3qdfI#AAQ0SgO*RTr!X_8ldo<K;=rHR|ulC^39{sPIa3wrIL_nRMl%4h}
zeNMev-HXQ4Bja3<5V1M@y{S^RM(|Ta#i9{G_L_q8^pz00HX68)W@Inp<CujdMKjfc
zxUZyif6v>qOAqkXrrUMJ;rbVn+@OtPywA;xJm=;gUF*a0oO0SgHVS{AvtmsQMFUe{
zrBId7;Pq@2A=~GxOd~}FbBfOMd%7hV#yWa8FJ73Ed#sqa?`RiE-xbz`jmT~RRbNlw
z6^5+_$`Bu7iynqex8B@2p+7>U|G86=XBQXdXq;rl2huanw_e3NnoM45Dk5Fb(4v`$
zzt*U(9-YwWGL-xMr&RT`ZPAmzj;M=N;<nk_Ij`JW9woeBn|<Lk1HtFDwoKLkvEawc
z(D2bm>(!VLWu+IBT6w~}Mt+l<xN23m@nF~4@U++(=>Xa@FBV+6fdLh&nIgy~6o)3f
zEoWvT`(cLgl|pw5e9U+sJ>4D)Nz-DO;Jm7$O(~LR*;?Y~c(S<9Uas>CR{J28=1DI<
zSfIL<#kQC*_G-dgu$8acahMLL^HvTp*QMbik<#P8ZaNPXLqG#NlNBh!r3!Rn_y*26
z+YbYG{+4Rs%^fA*p9D%u(;PXp6(%sS8BS8Uo3IUzYFjd&Q2sHKCIxnVI-@~v==+?Z
z=QvWVU(y`qVi;xujpQ!eR?X^0-Zk-V`pA!qFQX^@#bmRwB-+V^j4ZFw_Q{*+Zfh0P
zB~axb_pCK<b+pix=-R&2(W!gYZ<F0Dc6g5J*VMF1>mfhY)a%mAcY$0;iv_*$NiGh#
z*)BMI>V<5pcPcezm9fb9Jo6w7vzqz~9V!t33Auh#=iW*cINVlBU|L+FX&&l&=YTW9
z&|VI8`yL?@)5FFx_2}4cjlcR+|MxaFmX{%4CA-#wHYM5x-4|0>z?Bzx+4cIW^_~bM
zKC<Z{9M}kNPPA+oTto@E87?#x{LYJw0)mO%n?8<U${#AfxEkw69%H?5W6%U;gg-<x
zQH|NjgL<B3r@fxLjK6fmgab0)z9VB|e1)_)1I0(TvnvH^Z5?v=$jYwe#f4qs;akPi
zpZG7Xha80h2xIb8tq%57WoSi<9l)6pmES2gvj?8aiNiJMO5e}GJi^stmzxWoWI%8S
zQpJsFMl7@ol?ai;@P`pKcJICYsPrT`Ar*F7NmL8;Bl;OxMw-vrJYD8eMHhYw!}C>K
zUDx=Dy-|j(&=;4Va`F>j2UdQ2r1-mU;@o=tKY`J8`sC`5JJ!aFoF?^${hIi1^yQBu
z+98SQca1tR={z|%?nY%}%=t?pS_~GfD2}!Bm#c!V;_h|yw8$}s37Z!M$qakx=oCl?
zMVGGEaWT-OTY}Rd?pQJQzJgtfE57c+mKqhw+LJfVdPt*wqe_lX?RAS>Oq6w^`^oy=
zH{nN%qmaLz_r*31?c8EwJ&}A|K7L#OX4E;R92q=AxFl4i{Y4IAB*eADY(Da_kR{GR
zsQ6+}`Is*8ndRL%jOBR?93}8yX9S$B8vuKm^1t>n{2a*^pNetJ`k=aUwGkFq%z1bN
z-uGVpkP)8{OLGXB9u|{vx-xDV2%6BwUw4fq(|0Si2#3zp&(ogL9=XjwqIW$PP#Y1;
znY;>UA$B5Iie!zbddz&}9(DRMkw)`|moPa5uXtTBIYm{8iqyia#lp*|%$uF@&>b>x
zQ>smyb{8LDd%~%aRLcBNS)Mg{nD`g>x{T8#?9QPregbcm(|po@t_<8KyFJg0=fxiY
zMTb`ya*r!eC}p1(IVHZ`3%Wnw2A8JnDtbwTtwRn#ArTsVP<v;5;@fE*99Ul>-U0P;
zRgqhkxH!MNb40)Nyw&+-B(5MT`ue3mKhNgge9Qw=cM?_uj1V}wx(a$rHD$XUxYRC?
z0}_z%+I$Re)e5r}plO`4$nNo{S85&}M9|9bnA=8zi$FcjffbMEb=M$BF5ZFZcsHAR
z?ViKiSq6YOw5j2x*_DnwWSb6?v=|sp<TC<i{XLVeD@VA=D&~7gOfuVTq&Ir<{)=vT
z95V2`p{q?UVq;g1{{E<zMLGARt*cgkUUXu<d{;E&>#)BC$C=DgYLd%6>I*c4b#$_N
zS?r&a>CpDu*JXpj*`KWEF3W78Zbprns1|J4V|qV@PE^;wU(;_vOZWq|dgM=^)%zz4
zPJgm8=DLJ?py`Z}u}QB`?Yf?Wg|Z#kG+DG141Zm>=J{0$R6I2gokxlnqi@EG^B$20
zZvkd1if;|{wVh$cPspoW7)BP|@$`yrw1(8j&1RG-N%(sdc?!e%I`$7LU>$cl4cjjJ
ziW~pyXh}XQ(G2hUb7T5e7~5O}j~S<v#Cu9gn!e<R<QnGM@*~qNRN{n-g!}pw%jD3}
zd?9wp@q=^=&-dM~{@J+A{*JUYCH+2&d^Mi)Tz4ls9JdqNHj`9&U0;9$7z2lywZfyM
z)ERi$zg8OAK(qO4FVs#7?1Z3?I}BXSmkpTc0n@yjC3Wk3f@&FO^n|l^Wq6duPgrfu
z0en5|oT8xD^H^~oc*YFP#J2}4rRvtNs|wSpuu=JG)xJg92@hnXaGZk^(=B268|w$>
z&9%5$YV0w!6<tgb*=?VhT(6LT?W2V?62QNtu?mNNg4%thkD_j8i?U776a#`j`&6Dk
zx|gTEI5RfIV9uJ^GBeOq>o!uH%;##KH=k0n6=^YDGVE9U$LM_T&RXk^Lfw``e5=wx
zJ$I-(`x+NQFs<${TzBRNZsXb;M~}3@q$oaBl>;2G$AYuDVG5>=F1GUzj}bYOxueu@
zYXkLKUWU_DiF^c<;bNdutWq7UwltS#yNXcFv#Gpuvef2yRQQj%%Z3S`#oMf@_3%8u
zJzChgL>C<)Zht7QcYX0&-PEjD#yGZ9%~h!7G45U~TSi#gvtgr)6fq!bB;u6aUvs@?
z&u?)$H%4y*Q90|X7$xYiXvhD>=&!<1q6y07^p4c<d_`3><UricHm3j6o&r*wL&j5|
z->D?KM)BC&HGDR(B>qYF@757hBH9FUJb(`~?MqCRMT(>AdrZA}_f^_X1dq|p@yq@o
zi$Zw+tEd+Xn7d|(F4uLn+qyz~Z`JN-CQ_Jf3Ti*hoODrg`yWguLFqUi*4E9_W}QwC
z43-7f4CdLP6}HZ4zLw6GR2W=E!JDl?WuP)=GpfRW#9ayZ->`FKvN@+){jG{8`W&9q
zkV8*-6Wrxb+F=~#SAT!<3<|5%(OUhQ#QC{Ul_3AzJ-_hO2yCH$EQX%$1xr~yEn&+I
z#eC}Mo$|p-2(ZUsm*dRoOFq_Z$?0u>ys_7q&+b0Rrj_`5j@F#uf}}h196kA+7uowe
zZVb|aQm&7!T@RC~cA8%4QUbe!4{de_|8k-Kl8khb^{q{VQu_uRBLN%=_y+GOC~IPh
z2S%57_x659BxZ*6TWcwE+&V_>R%goyjy}Q!HGNazGjo}-ShMba-1316er4RT_)SdZ
z%Yb>cXVGj}sh!A1pO%dhPaU_v4xP4apbt<1Z~&@0JtVsHI9N0`b50I$2SuC=)oPv~
zmi#Jr)~R`FUdmpE&0%LQ=*Ajsj?!J>n?>61qRHWNoh%q}VeQ{KFk;h$^MfPOm&O^-
z?K~w2_TFLA(arT=jNo768U^o6NA)3~BpcrEPm}zp=TXmb*{HbgtNsJ3u#=*0FbuqD
zAXqx+Pm;t;Nb)(sfPxE?jh>0E4qFX!ULUXyBtO&jsXrgBZ#1$ty=xJD)@zw&T^_dM
zBrUqO8wuWi?oZRYF>?&1z1OiE=Wr{BUCsIF4wsotx&iWg(k%{=q@(*WNAWS1%0JL2
zMxvL)e!(UK6IM9xrgkHpSOE)zjd<0=OX=kVn_3Nwfw+75jT(PPxJSeUF2}bozUGIY
z9I<2F3w=^J%c8VwU*xcTTmD!~GJXrj6o_@)RG}AVYZJey!>_S1$lU#QdJ=U3#zQ!{
zfDQ9J$?=oX;5W#PemeG<?wtHz2^5T>pFK$w!OHPumdrD)uk(yY^T~GZIx6p?+Ut9K
zk?4t=g9+Su?PWa^cG$#BzpUdqR3>@JB<Nv)1|f2shQmQ@y|*I1MbS+oaz>>@qcI&Z
zfp9j<Dt6y@U$xZaY5ru~AC0ik(|mG+UBia1_bE5n{7@22kGdMGsf8w7HHnm#(PXtT
zs?{2-dl)mQw_&{P_P!ik1A<XaK`fyq2^x9rir%M+1haJ^5shBQa36FgVw-h>Pcc=t
z&0;qWu$L;z>8C(0ds?IG@A^z#7p+WbrNyUbru%Y9`^)*>g3F;{uW#DjecJqk^77>U
z6$u&rMj<Z6_yYzBOP(P0s)WTy*YZ(NvkVZDO<9_fYVdi-I;&F+PlZpH2xV%bnH)m<
zjM<>c`U)8=RMwE13dy^rFZ;*nW-CTND>o#p!?{~|P6p^3=3N_w;q?<=elIgXc(xyZ
zuDDQRq{8EOt2b67{q+mG4l36)B-kUUnq@*?qI;|;{G@DA++J%z;cTmB8Q%TdLAdM|
z1SIsdg~hg_XX#;W@UO|s8%L{~FaBJlSF%ZhiMZ##l{{m|4y2-Ylh*B(R8;P8h&{GB
zvmbNFd?lp6LzN7aVSM=%lho{%@ps8AS?J^!{DX4ZwcW6`Q+UP%Kkj>s1}4lglK52G
zH$xY4Xd?JYyS6~0ZMX%u;d+n7SHUH`C)xU9AnE#2aGrH3d{ZLh6^@~4&ft#=1Z4Am
z23QZS!z3e5wVln2si$oDDdk`GT-ShDMNTNK3{BaWBTAP&FIjV7xM1{>?}4n)^bCT^
z{H;=ce(k{LzhMK6`%=Wm$nDI~1g^>$=xt~Er{{06w9`OSAhfXyr<)lSnVoAbYB~ug
zbLhUbsMvhg9_X3^s2<u*N$mqP6VHpw!xJlT?FJp3;t!nFg@7n-N|9{^IZo9hvJ^Ac
z>RS9|_dpB9zQ&l?uUGt>t##!Y+8|j2ZjV$Va);$R$&3XMFH$SGz2XI=1#)it@i(J6
z^*UG*E8mC-R5GV@yz@P+QEP*QPMcxvcxn<2f-7)^)@N+GrRp?bXO)aqX`-kY@Q%0t
z`1|;n!^SVsdnNe$&6SJEra)mirLUUL;5BjPISFs`&V?%bYmEypc-ptDAKjF?zkY9L
z#^<|Js#a1|6{3vwq_Ry6A_EjXzxthJcu3cinnAkaVm7Ge-{Yx!@mKgBe(zj*q{M6t
z&&$YFUVO-fZOUq(bGh$tc7?O(h$O(+m2XNhV2{OMH<#|^$lskIGjE@4V}Ce}Doxy#
zWd@0S8<3T2gk_)G=IKtVqnr<<B6Z<DfW+S^JM!_0=#?h3)5o;K9lc2}h8cJ1Zry)E
zuQU<sKKJR_XNG{ZmN<V!FDp_QVyM2$IXyFKOXEf(M&JLl>=$${uh*$*Vv&Ui`|E)V
zzverM`Z&f2`{)d<onMHq(Ch$a0s8~K9SHk?oVfqQIyg7Y*C6(KW;gO%1gs7EC*7v_
zwD03i%3&$y5aqICf#wcfa7awDL5IMcJIK<XEG_lI@{cdhknMPsQeATp%fJ4iCr*jA
z`~L)obw4+g>Z0D$cD+8~+o?Fs&;M}B&Sx&01=F~iE?l4OvE$yzttHFA*ELwP3u4e!
z@;+|c&B+Q$lPATMWT_{Y%7!NPyRxkCZd6Z~>LoqC8Q@PgG@evU5OWk2cW8g)m!#-0
zxBKnkoq;!K+x6oH+A9J1y{|ZpL%DNW#jO9)xQ9IW))g??m&DJ|Ip)M+LZPnXQ>;KT
zjIIFnG`<>)b<h*JA6R|;&U_(^qRmd;oIA`<K=O|0ni^a;s(5^Q_5CvqbqE5xA9!L2
z0YyPk$na1nah)g4x?VW*6j-SBP2%HGf$VP@z&=Pf)?*q-quRWooob<JG*Ge~cz<dE
z7w|2hGT;>nIGzR9>08PD^t=3FH=Wa)03G&ibJ~t8HbytO{#F05;rXcC;hs$7o4rE1
zY(4X+g5BII|IOOXTcU>`262HU#0Q&Fi~g+NLpd8(&&;OH4c2*W%@I&LU&e>^^GZ*q
z^k@CWJ#`}+f9d^tNSro7yq+aMBH7)iIQ-h~FdaCnTIVPHRkRzMXEFMz-UVg>YccV5
zkBPe8;4D;nGXm?Kb!l6dP3par)IHXKp3vTai!W2#ir%VZwSKH=Z5<QVv^~9li-%1$
zp9TI$*T4C~!H-uB_3Ce)ypYi=aN=82j_a>2_SMQ-Nw*w%EaOWXd(uTG@;>oJ$ED*5
z_H~|Gs4_$Lc?I5+3BW~(84t97Zv8L!z%Cc><lFgP{%+%trt@6egj$z|Z3kNoTNGqP
zliuW=U=#fnftM0lL;=;+P{hSCQ#DC#_n4ORdM3JkG7U2RjDIrBCRALTz88h>Pl>%@
zvK>2l)>(J-rZRB3Zuy1pi$Y<XPIT`KZd4L|9P&Vd^7$s}=|UevD(hkj?0k%vaGN+!
zN2{QJil(bjpTFyK_x?f9jZb4>txNwB>{VYSVNdltsN=Ep_XXbf#Z3pWW4Yt$%a$wR
zE?oR&y?UD3lYjYXEnigBicHClmsQkQybW)@##F{%h)M2=5m+;QHgWabb%)_Fsx|yH
zcjD6-<?#dnu@~DmW+X1qV_t}b?&z8Sn0;Sv{%U!}KFbWspn@@*Y&R?Hsrfp<?-6g<
zCIHJv*?y?cYgYxO6sz59C%#%KFWTPqx#&K3s2^%@jcFl`$$;vlZhPy=>gSy%8Iy{C
z&{^!&{76T%?Ivx-F_vppt!1*24tgxcwY(-gG1mVq536!%NLNw54gMvmg!h^HkZ#(|
z4tas}6-Vz&`qoXYhauNa@h25~`ws@Em%L6NA0MliAzy;i3uRM<=e(a?L!Zd#cGxoc
zrQ7H5Wqx+STu-{)Q<h&iW>i1B?62s5ibzAm@8A0Xafj6OaE=wt)^r7S4Xl5gzlZv2
z^@lbU&VMPZ3)ZeURRj$Swrnh`RwG)*2-d$Px%RbJI@oCs-y9u9gpJPMxeh^%F+%dr
zY2I824rV?QA~^9>))o07hOOdnCA2HuFf=&kF7UM9lg505tC&kBA`@rpd`um?+dQJV
z`GNIr95kad&~kLndrEboq1Es}PEJM7{c}=mm+c*Kn#|wK*RpsWgvR<adfu+>*8XH}
z>YRhl>LxaZv78Ewr3(*jc^PT61m*13#xLovI}XfXYi-t^K~_)Fw|7kppRLiI-sjlQ
z4azxG&J#{$IP_7u_LDEuGOf|IHaA#GRja<k;>$u#t6FqGWW1x{Vqs|c{M1vA*^>}+
zl)a3nsJS8km;L?WyYwft{#0sD6F%)0oJMug4t%(OF-lbz(QfCBy7V*XP-@&Bne@S9
z32=a~9lc;n*88co!xoVq?K8)}-gPq2zimKl#LaE@1=7@><I8b7Y=CeEuYFG5eu1~S
z{FvFZ&tVR%%zL|E<%@ffmDaxK++l-Xbw~`)YE-h_wWy%;gB$Yr9IWG|A(}GT=&hQQ
zy^zVbSjXe84S{Efii$rsZZo+!PDjt-c7Nv)iUR8M%K2M)P8Xo6$85U8Uk&|5ehVSD
zUH96rt&$QI3;snYtX8FloCjcCsBLEVDLN&aro%BMt!{d?mb?u{TUGV9YN;?-iwYe@
z!RL0K?K}_t7C~O-D2WtVJ(8_U<nKQqx&+O>l*%M-+bO3{%U-qO#LN)_X}HzhSA#+(
zRbZ>kDzr=UYZnIWw-39KP*XCWrEhix){+zhWfNNoKp)2NY=kLd?<{)H;}FdJybiwS
zYhSEZ;-b~+vwP9oRqE&y1(NSrp?a;V_GXP8F$bDdx;}Ap22Tkap;L|P*{XWz4cWbl
zw$&ukQE%J3uZPu)GJRGZ{|o<jxapfJEPahyJu%Al7w6G60=54?V2eYR!*<4cI;oTp
zb_JD|$W2Ws0*x7kJrAy$9V2n4>-#9H9znx=yFhH;m>Fg4C_9=zBDI!&EUg9QU0lzo
zdBdr{u?>O$-6Y6pyWiQ^7hTuzzb4DCny|#2i%q5JpU<>Ey&qg+EdNgolW@IzSZv3H
z4df4|F@?P?c(@dE(3xSrEG}8K<NtOi_4Qn6LRVwea*lWQ8*!KqaP;EoK4Q!LBRG5%
z$o9PGd2YmSu@AMg(cf|zG$h^CWt06!q0(m4@rQ?~MWU>N>eO8*S6)ufK=Q)6`t`Wv
zAD<o{Xa4@sC4Xk!PFVNmUWq`CjlVO0{)%+}kRP6}TalN2i6i($&Se^AJ2_%X{?Y)e
zbt5+ukMADkQ8t{>5=v6uc%?VyVnnpAq0n0z&$--*h?bNWCh1f+tOprr6i&UWiepsK
z!nDnU>{DCgAN*b-bgDBa9V92D7u_!2hH~err~8GJ;XmX8k^}i<%Tfv>@rzD(0fDV-
z>x1`~r_PHhuG?o_b+ol#4x|lVR;<gjB;sN#R>o~PK38^N(R~ybv{~4KIyHi6?Fh+r
zBtpnk)%s<&K#?^<E+JcU+7{cL0?4|H-Xc|h?r?kmd>ntDSzcA;_i~$S;sY7mt>tU}
z2_-hz&U@AIWzTw+EfXu9VeeUYJ}rJ`nqfI_eh9`6YhCJ{-QjgDy5P_oIW%2kq-*R#
zk|gl@)i;iJ@2`E85qMOIX!yi|Murvnkq?u5f%|Jl=i67PC~Bq)+e7ZZla9$Zpy&sA
z9v1)f@nB%sSc!kgl@!1-T06F=<|f6OT0URWP;efi%-!wGLA)r^H6o%No2Q0GPce>n
z!k>IHTR=9*WRTwpwc^$*OA9|tVm}iz6?3mI5~gNCU+46{?NRY=OtLIXaxKg2vV{G0
z)k^&tCRPy{Z^$gCsM3Zj$Zk~L*T;ppidrz|wJ=1Fp~wUkb-!mB0|2A3(2s}hi7(74
zwj(spRBdm(7>}Tm$Y{H&v+epTqK#F9qm*kk?w?`bo_?RELD{^tf(fdrZLaIE`xAF~
z!{V8DbCZ-5LUp80uw8|LUYHTJ9%CJ^Pik_&ww1TOnBxmu=SU1WKcq=q(`q8)sUl&#
z(^ATH{4^Z_D7#s=N#B9L>J9!J!jRMZgSw-K@+<Q5%ff5&^R)YP*XBv*?SI<d@p<qa
za(jhe(u?<>R+{o1_C~|t%+;nK{*lKM)OlFD@3#$TYqa!p4q08tmNTDNVO?D0j2D8~
z;Pp-K{lctnA~1ZJXka)Jb86sPQ<36qJma5c*0pWnJH+LxfjJv6D1&~Qdt^Se3pslp
z;`!YR_LAnl!X8aV=Yd=%ug$fI{%2*z_JG3aNJj<CD<AM#G$$sf;t+cmCy|$;3@CC^
zeN%F-m1LV)e=VM>E$&ZLk4yqT$Mi?<zTugH0Be4pg|7U(YrHl_!L9iD>1RJf!72Kr
zl$x%SuKv4IVo8CjKJfJonx2K!y;ts6C~N!d^1ReBV`{S@arf69i1|G@@{hmqjL$oR
z*Pb1kFg^RS%JTD}T#aX#;CxbzcrI4;g7+t>{)j`oH}?<C?^=7w?K3_z4PJrQo^|2v
zwU#V>r;~jOgT?Oo!P{WGHbQEhNAqXMpT{f<9V7t)fHOXbmvO}69=j@E%hQ%G71jxP
zFQTN9`!P^{gtf%VQ-n_!SOCyiX8S?!9^Sie_rgL`k@n+54h=^YMbN`I&KLLp{`<Z|
zad3@lHQovFf6$S*-Nrf%aZOoKnlsou_t3pu2GkVyJg2$hitIvj#c?O|Lgyavx74}!
z;Od|B!u|(?TOHgY`gy#Le-=9LzYn$T50`q&Z(~=Vw5|jDi@w<*4$a`&a`d+25N6Yk
zf6%vxJGwOYSFqzpYy2WKgJ4V0tVtKOu<8(|;YG_%)3|x$dAtkmOBZ+-KMwDzHou0r
z`$b1{Q9^1r-=2|cMB%>9>cAQV#A1YtJ)p<4<2mTn%A7Y}=pXsG`jhX^)td1}5<)P%
zosX0+UkqfitKj?0gq-Dd=}_vA7cf-37rL%UGtZ2AdxiM-u{g@@u~w?g=(5;qcv5fe
zm0+psvl{0UG7NKRh<5_n)5m6tX7^uTcWCi3vA(AH5KnmZI8;;H-8pR<;6UHZ^1M2;
zA@J4)ynW~#MbF{nZ+b1+wTW+vCrNCbd72_LGnGFtJj@3^&XF^Cb_XS7n&>`|zZ8u5
zSrDzu`@tS<2Q-(>zO<{+c*E`YB=Ma0{RG$*f3VEau=pH5$866CTF)v5X<5DQ@JM~`
z<ZVwd=4UU|yz)098Bsp}dg-wZLe!!InG57I(dF&4ZzVr;?(i1jp_LA}U-&fgNT_#)
zo^n`NGYPKeJYpZ|4fn@=kNCwy_$z%G*n-5v9FMpkY%dXnJPqr<;qG7IFsf@&Y>d&f
zFp-G4)f6wy>UyVf4_svOlo2D!?ru;O*Oz%)LTFH02CJA52ZXYwZUTM+9UY;|lwYkt
zZ0KrqcoI0tC>UDRdO&|fMG(VZ=BY5M>_)OC_YICLAj``^FAAklpSpcgqfRe*`q`%r
znHJf}OBTOznre5j=l!zrxyDg|i7F2D@ODG}-;b>hVE!+D6~PlD7hGfacb&wa@hpAG
z1+CCGn3DV36>*yPJ^EksQgPVrf3&jlDnwju#fjyHpw@1d^K{&@RljjEMq@?}h}I2f
z0gc&1z(Djpiq67KgE5CXC&_Pg3H1k=uuO`cx=d*vSK+i7tFv-Hk-?BZeEFwe*dyKC
zO%8rE<o|<_<3sV-1SJwajq7)!AeMnRSqLcCR<Qb&k&cc|a5T_6*5u3+9*(IkZA=L2
z=4usssRH+LxF(`}P?HS@q&LLqX#;lQ9}$iK4Y%>h+d2iO%j1(k7l0pB>J$U2A(-?Q
zup+O5c9>zd|Hmdd*5rdBv#~|z*@69(3G~96(}ea(bjNwt6@|m1T#yqFkkBGzoY+m-
zyv7_qsF_Ovb{=&K`2Ea?<qyK$g8o@V?4ph=VNi2BbT@NRMp;FK7QNGo$RnF=4Ply<
zUaoK-iQm!rdk*UJGoCtkS_fR9sHWIX>J7Px+7UhuPWfx#$yDPf&NC4-5&d>@_JwL+
zNggFbD0p`=tBq7)Wq}h1$BE`=#MN$^p<?1y-fTbqASf;kXz>Z~W&;_<$Z!U9u-QCq
z!_UJWD*6*#Fj`gw@;m7?2f|$VF75iaQ;(F>o7pV`n?T_5Cl}PqE-Z6+lC3<|RF-j_
z>R$u;UHPA%jLK^>TM&UhbzOIR^B!ifAhX=PjeVq}h@SdqUm4Ty+6L5hB@2AnY1$^I
zw2(-$D`Mgb)*r9J$d~@j7c9oHFSc$KX6Fo(_@jsf*x&;LXEF=$2@P0>LQMW@QIxrI
zLnW1OfAwuOrS1YB;J0lB>NCzB0yOZnKYPd$Fiiag**R0AJ%Zr|P<My`);~B*cm3y<
zjlPCERo_`QDzIsL_l%KeFTMWd1#q|_6*l}4*wEJNQ0Ny>bzmaGutw8KNJwZ<1juRv
zt_NB;?(Kjuewas+uxJ;mBVGIhY5ioN{q^iKD4o4lxEp{lVosb=uZphZB<8G<>i_#k
zPfm`*F*D_*N8u+=Qu%-;PT(_CoV^C+gM&c$@(2YC%(#vX9}pfF(d6{}^Vx&H{Xa-G
zR0B&M(-B1R{SinOSNbH1D97O;A0CSTa{YohaGM7JH_sjnK>UtuIS6#5flm+rhcTN$
zQy3W`FzP9~Bi|IUvkL+D?SEf*3%D?bbd@g><7#uA?~IoJp!`3BGq$#W5#%+80S_OQ
zkImptr97odUlpu-&Ms+xD-a6prpZTK6~_4q4i*zHZQNc!+Apg*zGJyVv^o2CY4G2k
zmisL6>Rpdf=_HY`(dm~)ni*ETG<OFsd1?}5q!@viPA$+BjU_c>VR~iap)MRp&4SQa
zs4Q4Xz)+{L7(yUt04bt6K_PH7IcWf+<w5FE3$4@~j`fdDuMl0@icAdR_5|KjdVQG9
zza7F~6tWEgsOf9vQ4_$?@#URC13XHHkKo7r@J&t4iVM$&Yq~-b^#SnwWY1=wk0`~=
z{ujFVM`t!7un~=Cm-WB(ShCFXelobnze6N!zZ*(P6U0B3V(Ex*IRm8u&4E1;#i=a%
zZ(dK98>(s8=y^HdSwvIwEh@c@GODuMSMHBe0a@~2a^JoyZQHSZK2CS{ucZ`h^z4i0
zsI^wJp<Ic<8<!~-_6q*FYa<3*osw|P5<N9RNwteT(N-Izq@33at%H?pDL>><h~<%+
z=abz7zyyEO=wO9ah84&Zqh+%HEn`KND~Ej#)C6VOvOoP#EP$jM9dL(SJ`qVSst?UP
zM~5I-vqOFkb4knPCE(l`8h7R;jcEF-fc7)^@;?wLpg~-48E!&%?~d8&`)Q6<vI8NI
zdKKSDoc{qf>k7D^N@?^cs~?=5{g&elw?MKp0F*PVR0K=P{Pni6Pbd!(tTzB4r3&oL
z{8zJN0Z(DnIj*le2pSS&ZG)5K1>mU;Km^c#H-%WeZU=>3os_LPZU#H1Ow$s~r0?dQ
z$yDV3y~TFer8tH43d|fPAi@Kr1a1O}+4m!rbDzBdNQC@f5`lwdrrXmFfiQ^E<qYag
zuqLzA-767i6ng%z2rt!Evi*TV(Vifc673rRt#x_-^PvBt`2YX^<g#njBrt*NDpfr#
z35nSrsxM=y<y8Lkx}OI$3bFp`*>?`Zgb~E6;7u@|OnGu9^1U=)Wn_WFr#l*l1o3z9
zIrK{bE}F`CpvDP?)kdDsPG7$A4&jN(4u{z|{i8v!)fTik4)6S{Q-EdVdI%X?Rg@ok
zU8aQmB9&LINB4C_Kj}d~X?moz7}*_`y!VaSChH&PAe~>hzY9g%aXxaKatBB+jOL;q
z1q=N0L^9V3bW+V>vgpej#OgPPc{H~eW)!TYc1gclcPGM|sK_2iL=_uzraY74;2rRn
z{y0%wK=<pzQ;T)G)rh!zIn$*?e7HxH&nN^5Mx5h<h($24Ed|`utR>WAFJDK-o+h$H
zVafKf{a+V;`MGFbSDV%;sCTg5N;M@Px)ZHk{<7qzvQ(wnSA~_k|E76=Dzsw4kGcL!
zwt#IFv@uGK?(3wfqMVhoVBVR)Y5h52OUh#vK$f(*IDkDs%*GbY$ROn=%NZ@$FBFOD
z7eQz0yw&a_v#`GVzeQjlkL!=P%`xq|Cy(qn2bR!rwguU{Gga@%%vf7aa_{`^=hPj+
zdqev;shCbzsZ#;?#gO{6t<hfMpWg_eEJ9XjLy`$`dSU@7{ank}TyxksapbrG7<SWx
zxB^2z(9nA0TZ6BVN|O_qsyGZddGbC0{A6YV!}{l^-Mu(1bHuX^|DAJzzNRLGe((XP
z#<-C{DJ58bThI!yi&}bkF;9W5Y7t_Z$654xiR=@*M1DET@U&(Tf>~<aElf)al~WI}
zJo=fsGs8QBjuP!_x~hJdXO}bruwDh0l?kx^F~EA(_Y|NIg5=4kik(ks-X)`3Bc;TT
zKdePUze6${4x)VT@qf%6r~FYNemrP0B2M4D<|KZ@`0xTjt4F;?cCv$F0A!*0DK|c8
zT_Oo9ww$<J><coi4J~N%p@b(~_+ZiMIQn@M5x%-BbZk}FN4$cUhD+7Y6aS^LfM{Aa
z>M!({($?6|R0Bj_u+buJaUR;$KIIQ7`Tq>#SJES*wLZY>_bgTK{=@6u|K)XsY=H^D
z>tTS`UH{E%nX52fV#wDP6<B^<GNmSJchsf=)SW<10vV<UCB?O_O{w)dDm+mjg5K}W
zf=gyiO5J;DI%W4mXf4RJe^}~e>o@h$K5CM%G&?x~{UFh_(djoD<~;#xi-;Evjj^4-
zudfEPT&9?jv!8T_Etep^7V;JEhXC;Wy3Xlj4Gi3!y(r&c{!0l5TD}IFUr#*%ljkkD
zlI@b)bIAi1UopZ~_y&e|4bd-(X#yi(xgB3C`W<wNGnkT13)19qQ%CphndT1=J>|^7
z83_q#6dn{;Ew2SR;L-Y|dmjx-^rIsv$$p#cJ+J$g!b`x;gcU8cZeA<Tg^=Vqer;OJ
zpr15aC@pE1F4dS=#rX2RKNDFeUrkes1MQ^+?WzTM>2$Hh?x#_o2}zhv-TPJ09lvnj
z3Cf?x|HEV;N6M>MlJ7l5?CKNji3(nAJ|LuAWSn~bMdp%t>xL))%C}Kxo^%!nE+Csm
z{o*E4ZL*Oi?&*Oe#<7L?FN-BP46rYrW2E6Ej3~wf1Z?icBwP;A*0hW;Lp-ALpgMSs
zbs@U{%VDYdFl`VdIOt}czXLNR>zWVYnxdV1!j=;e5@w0F<(!L8t$^EAeV8x+jDb?~
zd{xj_X3$DC<<su&%?%{53Mq)(gOp?Lj8vPD^PKAztSDgKb_?#1lF6N>2*S}}PyLwW
z3{fjbY$L(cP4fn^o=d1nuFu+pOSUwR^jgno{fB#RbSG!a?wE)G409Qhii6(F&0!G-
zX<-Hrg@9z@-6}+ldtdaEeh+N@`4z>4x~XCSi?By-e_<M<--r*t{Ws@_gWc5@BlT*G
z0qyfywqR%t<Wxz?$EwN!f~^G1AF8-)H1hjex4QL;9a6}gKnb4RzPC;3Ko!~+>il|%
zM0uxV+tne$cY+2u1z<&cZ?cD4fMb6S;7W^-`75^q-UV%N10A8)3U~tO_8N54i8dy>
zGs#l?zr@*Dqv`4Ht%8YRpN2<70B+e&fI+NBe69Y!<o)_@@`i!(Z_<9@<~0;R<o`2>
zP1+&>yJ=HD&L2^GT4gScMNODE5T+b$kk-iPi--ZVmO5o(X2}mdHW`=jGX40M`jtud
zjGIrs4E;uQ+Iej$CR=}DuMwX7z!y@&yDJ!sM8RVUT^VW|qbCe{qMOc<k@z0m1?(Dn
z8Y%fdqiJ|WR;Lb7%lSLg*X3c((kGvGcRr2(xB7Xw_Ve<|Z619;P`>JoldAuv)4Bj6
zg}Kv9F8e?1y?0O)U(-K2yGxR|NRpg$mZ)Tquu5E_WK=*hNDw9LuA-nwT0kVLtl$a=
zh~%^?iV`J><TU__gq57<4&Uc_-uL&rb?g3p-(TG-ikd2zIcK{2bocr6nf~;yJKkBC
zwV&TTU(#0Q)iaInnoS(j=8ZqPvZU6*R-{$dxj&6{JLQvosIKA|qN_jGx+C&%>&eN1
zLe$@(XOPCP(vaN5ix5t1MPS4~9x~-yjn%_g<9{z4m351A-2wBP?sJ%;Zg2t;<^R6M
zQ8|l1X>OeGXYPN6#fATbMcNwxZ?>g-wJ$6!VS!V^gWcTmS2o&jclOd!@s2~FpY_iI
zN?u3hw{U-ZtK(@oS?C`A9qbXVP}*Q$ai6K&1t{5+gUOR=EU@i_g{Q1@mfeYxHPO+d
zFJHWX;$OI~stOeUUi~ZnEo?gfzc|G!YK1nmSIhp}10_eKXV!dmxfr<HCwPf^-C(JB
zQJ%wn@)r!KKIJJO%c#p3%b9#wC*4|qaq$!FBVEP#w1bMo0XPOH&#*zF{U=l_(MF#w
zSZ9M$k6F(i!?@@;SIM96mMCB2$t@hIS1MYM{m#6R?Ao#Twl;N9bC}hKVBN#R&)mC8
zng3`v=s36#be{ijX~q7+ztaj@=LbRh>x}A8p>H-U@)f?gs6bx^yaeJC-qTLO54bCL
z7mECJ^&@YDiPsPB($(MIbFB6bnTrU}WdsDNSGw9KEy=}FZLb+7^aL&6wy#HYNuG~2
zk)N%cTzg`_D<LwcwtrKwY^|fA2ZYwgRWSCE-x-HL+l!^K*5ispx^}cmeJc#V6VB?l
zKt|X0Zx8dYP0s&>-#@{c>kHqm*?T|zFq?e_DRg2TsgFG8W310jMU{P#-2jQiMYa0N
zj<tWUbNO8#qOte7o@qMjhn$E^Ce~a83A0H@JrKIxYxKzI(aF=5`&}>AmL|&j&qVr9
z)RoW7u8)MHW1jEHND@Pq+rI2Q6hlF5tp5{#(5pLIo+qlW4dNia?3Yj%M7(NrFNj82
zNSy^x&(u#ma*aSjl$hncAjEVk0JTF!vn4a@qopgDTB)WG2hz2N9G=G;lc(55h~IJH
z5eSiBYeQWo2h3%T9}pEVCyc_=B<^}ZY}`<1!}HxI_5VvuXu%jZinkEZ0>jW)Qd^43
zZ?NI!vh_kPo>^cUQGma8n^Tj3)50|n$=E+Q4TscmzQ!cgXB1(wn2CaEz3JT7JET~0
zglCC^-g7b3p6q&u80vP{fg)-M*i(}JwCc9ogw)K<TGDeK-*WTx@HCk#=%@^=QU_9r
z+In~F6e;J(5dZ7jYv@-PjvWoM3aKso2l&KN+oNi!g<;2$oKG+@qS^}SqJ^|OkAdDI
z{rD|lc0qtP?NZicLzMSyM$@vW8ZeKZEXM#wqIBv#fyFef=8Mw<FVqd&^?5R^>rRHU
zRWm-C{@Mk-b&1ly#S4k@u|I<RBXTRH0Y<qy6;^Iv4_5aF6r1vJq(>qsQgxp?{FA!_
z_&rvQAhm@3M=e7@u3ZFxzu7+*6rIu)KHz8<-(McSbkd*g@QSA|dYUx7$fsAn7M||_
zhiL89{-ZNhU<pXi+BST%s@Dc}e<x4aq`H(a)R^H_rT+V0B-r_}Vsmz7i1^#8r#%T>
z%6HC5qVQ+m9LtjoO~B2Q7E5qnmkHJI`}X*zgXE}e!Ejv?r1P@e|LB|n8b(!MlXLEO
zf+8$GgOj^EkFkIuHuHO#%=!OitPx39gg#pIp<R5DnLm{)z1%PIej!t9qAC-`2|eU-
zLQ}wrD_XI-`u3_K3jfapMW8l>z6CHLi0>4eY1XzjD`N5P4}p#k)|#C4tAi?*z47E%
zxMtP6|MiCdnalCN*2e#T@qe}i(7qL}H=wos25NXJ=12L*Zz{o}VLM|_5wI`RM<*JY
zoT#*!3z%AOvoDxAz;vlySz4*InF?5zSj<m-D+t1hGiiVE|F%v=E+-L5^YX2hwa$kR
z*l_Q#9vMgO#Z>KXH>L+w_hb0_0-?SqeWV_e3dxuhgF{)gmx?i+UhmkQg_bttFGG$r
zzzajb1XU8LpSml|PCPY3U<_{ir=Wp>vM9AMecoT_C%^iKeS5)x21@SMr`CSme+Juq
zP>w&jW47>M#4~VfTQkoM#gv`lz8>!!@El?nJ`LuBec(UAWFBgW4mrFueEYes{by5_
zlj!|B?Np7dPU(|J(^EJ8z<aKBP0>%~OhzHz#T0e){<`;Kl!i1(40^;q2sQCju>^v<
z`PWT{=&(!mX+Qe|jt3j%E;?8hO<DUG5-x@nQ|tftTR@KLU&CVK84g~NPxW8^m3(me
zE{`Z1HpFJ%xVX}OC%bp6CwyNWR@AdJdv?1h$FbgHz?!M^W|~P^MJ0WK^}XK)cO}O~
zc!0k64-G2p+o#E!*~_pUk-8JQlcE2-sG@iO4ctad<yo=qyhPQVJiOR{g!%Db(?F-4
zDR^==Xdt8A%nHtlZ-2B3)9w1qwd_wIUSz>965Lo?)1caGH$kzV#jc2!r<}VX!Oo+~
zjZ=9;ygev*me0TAr$*1xzJ<@e1miZb_vqcVeNKmZOy&Ep<@KJ?LyFC?&r80oNlzNr
zZ4V{T`};Rtd^~D9x(27R&n#p{Zh7R7gYRfOZ1Fb>Atle!L*mcQ{Pz{jeOvtzD+zd}
zET+hE?~<-iEk=dRj-IY7!Ih(IOJ770wPy?#3;QQ)*205*CV!so8qM@@_2u-t?=;$V
zHSon46*;eWJ@MLp&U{hFquup5!d|F#hv^xY{nf$V;<C(}OR%F6{Rb)0tUJc@yH`AC
z)!U*}@(v%sqJ8IgtHSql5RS?ltD%lV9W4&=X(cz@<_9{;Ye9d{?e1dUAUiX3l?tT7
zeD9+i!rHM7{d09v3qKf>-L?LFdu~UN;OuXY8!2~R&G5MD+c6c;*S`#g)(W0o=<E$R
zQf~<9>223K#Y3%>G4hM}A=b*<Wi;UQ0zb`xJhAKsD{*VO%dbu1`#FD5w^;}Nczi(f
z_C#Ytd3f)b%Km_#QqdqZsWZCkkKBAgV(wU5G+&i0s;w8&Ysf<=`giyI=jy!NEYO(r
z#C^K*R8f|da2%1AJlzy0_}D?nLN@TqO{e&<92aig$<jh5>?ZCuPUUpfby}s`{u(WG
zi%{>HE1%<+Ga;)ssa3iHWxLea$2&CowZ&o6Q_A6sty^IKQioU8sSoQhfgi4rb9)0Y
ztrpNUjQ^h`)1v{b={(l!U2^$I@w+i@KdF~J*<j<nAAGq}DRaVpitEcg#86hP?ddfB
zKYYW_!S|Fak-zA=R{AL%-@5#PSefg+DiACo+O)iFqjk+w`+hK>y3eiAzwA-gAAmVs
z3%<0ubZTHVtW^T=^>*K;=J9Yc`?p=m9otpTv0eRh<x~%=uIN8n+9>N;T<XxM1w+F3
zEuRM#>vr#3iCJ*<#N#?@{!;Rzgos5`3{Bg*YS*V)uI+a)FWH;{Z~AA>R`lK&zdPJv
zI~m%o(aapUrt*3Mcx1_Q!)8^v1h<}l_@L_3a=<g#=U>od9N_V}H`>G*JtEBT-N)L|
z09kyGaz<8rA=2q}FceDH;uQCd+X($ffe7-`v-_q2K<~4t!G|}G@+TBm(o}h(^Qp)o
z*Ocu7SS+7K(O-%0{zf&l@p|!Sb-}%1@40XAe5;@acT71GR%5<ja5X?}pL^XF7^@0j
znf2b99r2VCJrw8)VDL9pmU|ra1Y$#{J@#_D9gvX>2`T;Kgz9=h9-DajEW)=^SPWc`
zXYSEuY?@C|@ri<&H;)$Yk*=Gywb={4IGEymgdZ=m8j&@#Wi3y=B*pE+W}`Fy{z|=|
zX=@u{d|(H?(s+?Er;=eu|8~uTjhH|?A$hS4U(eg!_2p)^7gq%q0#HMb|8M?w&obWY
zB<kI=_3n*1y8Q+T`6+ayq#wtYrBn9A%svP_oKM|Vqx5$dYg8WeN|wHlbl-34EtYf?
z*fc@^wCF#$VT0((88j>YD?T)9Qx{ezKq;GQ8W-5slGV#U)KTa8bAx1%8zc#z_54Yb
zD(0uT4*oh8J!;W(9vsX)_WX9wEiGUNGJcD8ZNI(BGfw($w&vjK@)$aV%WWoI+T02+
z$7hdY47?}Ro7UB<!e>6^9*!0*TU=YH%+-<vFuhCQ-j3c$>m?EW%J>f#OvW|tjd(s^
z!MJ$GK%TCKiu|_sMh~`)SKATa-F;B}2<$q}{O5d*ta`y8&_;uR|6_+BqsXt%c3Dvc
z6Sk6L$7n7W(!}bnpBflx((|2JO=FbTRw16eu3l84vhUs<*m3&_(T~ln?R}5RzEw{_
zP9=Y|x>53?xu^wiZ@WiSSnipo0Zr$3&p*=W%NDZ^eb*U{>p=}0BmC@0zjNcN@czmU
zYU?oxgP{9%PvP}T;;A=Te6grSF32!2Do`407yMsW*9xx!6&^R*eG7lEJl^VwLd`|~
z(OI*F`!4kNDR!&m>4YyhWL}7$6dn0;sP(j8`NDIHgACRCUQzR{v8V5kZ23D1?l#|Q
zS=dUeUyIFP0<IpZ7~Ve|U<70OiudnL-H7>erZi*r$5l+#F3s<&X*WSyG2#PAKcEkk
zs}o!Q>#FnYAfQ>g7CU#gOVIF#N4-Z+;pBM)K>l&Lb@n#nrj&2owFm1(UOO^>VLMgI
z{aVRcbhX_h$1&NVkFYgsb@VrK(MmXIep_>B_D`0wdk+nk5bq5wMvb0hW&Hc93a9eq
z7eElW&Q6IzS4AuCO86OXSEV~ddKkR0-C-cN5sWVvJrAGpBiJx`z=K8Qwbw;LsfjhU
z0`DB}rIA39QR2QJ`CIqC4r6TcYqOz!l<0}>7o!8vVXd2FC-PZnY&g2a%XG-D)oG#C
z2a~qhE3l75p>O}=44~8l+P40a?EIgdQT~79?Kab`BLJRJc29>v1G6D)GHE!w7_9-N
zx5VgqOeDmSfxkZn#SCWb%%(DV7Mp#a`5eCU9SKKjeTuM|RRJ%cSCznW7ihhVXzeM0
zld#YEr2vD{?gg~s87|F=UqT~e$A63I1s9>OaK$!@+Pguv`^@^0eQ$p%qS}0JKSfyo
z>oLk;`S3sgx+eo#L12U$!5MQNKcZuc%{8Wdf`cfZ2`egK`PE~u)!yC@MgdvxhDj5~
zw8ALFygVr+nb{E02buA-?~&u+@q^;x3c+RISJ4=vWyop#nY|WY@}5J%iC&mQ<`KqG
zkTsey8Zh*zyrO_YS85`@@;Je?skKjd+0-bbpgbR0mihA27*K=_#k_%`@ce+=S}^&5
zvww9`TFO616A{fzp)v88iV^Od#Odirzm$D~!jEu-iBNnzS%%nWmfE>cg%GoqhCy`G
z<RJ9Gc=j@cOM39$0PIWeaxtwd+$(I%D3d_11vLNo$T?6v^-yXg@%xn_)o0VEqH0TC
zj$#{1O=IaAVQ-7IPCXW5dS#Cb`MkqOy4SQY58i#>hmFIHnreDxcVD*kq)pPhO}X0e
zG*b+9DAiA;OI4xWZ7>P&8fpu^HJZM60q;K!`^>KaBy#X4>{(3Fd`*n7?cnBpX4*>g
zc53UrIO@gD=}=dO1&#efUH$KJo_e0^VFj=0L-{&A^mmMWzeA>qAQwJ}U+?f&Bi}}1
z5C=SO*DPtoQ8%K1h%ODY;r{C-HLTcL?6yylORd|Pzm){M-ljH#Hltj*_QOmbJ{N0@
z>TdyLj^U1lq)0XaR&SELpfTig$}>!jOyy}(SeT64H|w~qc1pM9l1M0BXCzLx6`3Q3
z`oX3lQMsuZNJ|kTR%Xc4vSX)l7Q*iEZ#AijQYen&(do&(-6gCpwv#X+@6Is}0afCE
z&H}(=M#pRD9x+UzK@*b#`aol6Nl(k*1>I~*Ebk<5xfs)HjBuPs)#T4zX5iGcE@Q{8
zAkGC_F5HS9S7m)<(T!>OnL5ku#PI!$QP%~$Sc-(DP|DH3P52H6MJmBj*iwXPqNhRU
z3qqIbuNBE2$r+*Tib4!914O&;{-h@(3L7naJ368A3C5zuyzbP&7HN&bp{jL%XqITh
zP9#iD@YjhkJ=elAVaGmyy?WG#OI8_J22R;bVeM=+{&<m3)2#<+B+ikBNgT)p1g))>
z=!zQYFS3K$3)N)^AzX)-wC=J{8mZoj8d<$Njh_6WNNPce)793~xU>8<AvX_J&50A$
zq5!jJtdg)q#BePyPDNN-seO!|#Yme$^ZrD>5Ur6bjhB1!oAY=)2LY{`@@V3d{+=o0
zh+yP4f=<>%Powwt4r3@)U4UGLL7q*h-{e~erHYfXq~7WeI<Q^t11m!HIzbv{s`}pL
z0<%2f>dW5ghFgChh|`PLCG!L58?6W@y@ut*zYyOVJ&Pf9dO7~CzbKS=BR%efT4jbo
zs~hE_+?qA5*ww!4!7*Ppc*I8&STJ70)~RJ14~_LAz|sE#QUURj?=Xu(AFAGIVK%n;
z=O^<c#z3bVcJeRecn(@VOlsooilNb@JA<%=*LYW9&aRCbQ#p$Q;hIKs(h@+T;emRB
zuX9Ws3O^xjDUvcW&|H2mH&oH+K$)UEv<KL)o-f&uqZUK?e~KdTzE8;ldUNjD4BS`Y
zb{__>Ba1?o4xrZ@#s#7NAS%?m-wfj#GP|kV$aPHwhp#Z9&DKVVQ0w3yaPk6j!dmd`
z6!qU&|1r)-^n9?tT1d`@BLTcv-TQo8ZtjAt8N-U2u-u(!;ZRqfA96#lk9zVRSm6op
zu(d9lnE9`X&xQ7_V+5DQKj=s)o+EFm{z)&CO&dImQ-`{ug(>vYq>qSc0fd?B$hzcB
z1w?=Fs))fiE2JaMR!^WSBATVM_;p!{0iNK{$xT{mO6f?X1g=+c3%rnPZwUAK(Vvah
zXHf&2AxJdJDS5#Pin);<Q`py`Az$o~?m;V?Z_#88WEHK0xV(&OT;RFo3G2*Lc}$_Y
z2@82KR;b%AGJaph)3c40)H4f_yovE)lZOI3M!}QwBzOL32>JUEpy;o7mkT!b`deg`
zNMA+IO|EkxBr|+oSll4&x#*hsjbb1;<j~tFp<1c0mnB@dfVzNY%6@^&gk>>wC2nqK
ztC~TW;buL5_yBZJXSvtq4&C2No)-RGyz#P~l0sACx!c)desW>*<dmkAvDd!pu1;ua
z>Ww$OAI1q2tsQ}fo8nL;Q1#TU=1pO0cIDR;S)Upo(~-1<yXQ<sdoOhVO~?^CdZ7Bl
z6Ek!ZurG(xmL2ww<cyqxWLm`G2(SMX)cT|rr7Ksis;bM_`*BH=>U%A+XlW`20@<Gy
zo^mdufjab;ZnecmBfndWM74Hsf9FRE1kLW7;b}<pLp(srOGLF=9Y3(BQ}ry0^ZNzD
zn;ZP>B1$aZ5u`VS@m+ftKnu{gu1<qGe0x$Ky?PT-R~P}&dH+6`yvt>lGCG%=Vr!NR
zx2nRVilGWmBHo!!yD@p)Xe)7{$f93X(v#Y=w;`u^JD_K3U8|+xR7muR9qx3&bc$tC
zK!hX>%G;L~JCgaG>MlgyN=#09kchhw&g%<Xt?ZOQp&Kd~D)2WT9H-?lgzX`~HK!g7
z1dbkUL%w~^y5Ybor{~ArInpjc-Foy2`SbZ$ysAvdPgI+qil@)~?rC!U%3Wlz*G|@3
z3zS-hIBN0Xf$?@6@rs*d!ajHVuNhr;S4c?Ji`n3f#(l4Zg;Tm_A#R>kD_tVQKnP{N
zIc(XqurAb!*5}KCrTwg5rV0B4zmy@yR2oaGPiq2>?Qjs|!$-}@ef;rOP#n#thlqfC
zmr49hK`n+;gd|U~>_ljZ58Mk#H<|D~giia)*fI^!xc4qcb9F}2(?QU<^>=@L_1Sae
znf`X)=!f5KT)&<qdwnZT9EA?~eP>PYqA;8bq4+bZb#il0Ead~_rMG_ufO>GRc;O5H
z=`JRJHp?wA>@-C;NtS~xQ_N34uqft8Dj}zZ(mn<xpB^at&K?7m6#7ax&g~XgR42U_
z(C-a&rH+nK{%a<KIJTgZW$?)bbxzeiJy~sfSN!C!rzE^;{Rxr;VXx-0j=G%G2lp?_
zYkC}BtVA}U0NSL0$b~~_b}V0Iwp%~0>2ZHEAn#3m&@ha(fW0x3frU(=+Pq!vrYT8<
z-uH57<Y<&`2xIeHh2P2`oc@Y`o!T8oc%^n(XN|w|NkGOHL6hU7$Hyol0x&zX+SI!a
zw3Z|vqwDm;0T2Z;v~@HOo-{y8av_rm0=~5R+nR9e%SpIz$>0o{_-<BK!FMOC%T0LR
z)Wi%=W+2^^FVG|N1A%b!UTsY3H?C%W_J;_^pX%+L-+iW&4e(+d1lk_(${lZ#aZ^ey
zPNzjMnc%KL`~}c-J|tCLJ(3_jo&`>^2fzY+imHHMX{m>-33JE+8ePK|Do&dUNmfSD
zFCLS`h+K6yYtN>}nR)c2HMf+#+HupEX40mEB0#-hC#|sD@-;cAlofOH^?{W`^eJe`
zy%gQyOv1=S@f)DpK5zrm-{|z!zHc@yY#T&BNjEd$aEJGoM%j?1Nvz1SV=ZnH6ydXy
z+1pK3xGvGAb%?}W$kKBMPMVejZCOPq7yL0?wYX_$_F|~T)FMSNGg3UgEkp^Ysvq*5
zuImDien0Rl)z?KEgM~;&v*g1EV(M_P1XCe*dcLf2O<qm*?-!{be9j`czeZ}t^L2$V
z_5QJz^R9(eqvv-o(dS1A&O4fndG~BA(}xKplqKJTEc=!kb{nn*Kov5qNBXrCDjkm_
zN!eEVzn92|xha2u5LmpexjFu33O_LYmxXXmx@G8VMY1%hpf@}O@;r@aiB#63^z_`n
zH3hzLCjCK#^af5nunP|j7KX}pd&cn_U>kMSy>K<ZQun%p>2pJdF|!TBF8G^6;e_{7
zv$>P;rzp1mRXutr9)mVoO5)3NW$uo?q3%`L;3xWR!YjSN!tqAQ&P3Z>A=K^ulls#&
zNX^s;<;jmCq<t>7Qu?eAPi?Ai*@hzA#TRg0*Cw|@apzFz<h9`yh81>}WL!@;uLjkO
zwZ0ObP}2uE^ao8A*mZB9pkiL6r-a|*nOO2RVI0PDRXLre-g}d>bkNcKK(*wV|H)AH
zPA2;Mrf8NLg#IGa?jGoU)f4dIrl^qTrc$@{3oh#G;|W8_wKz2uUIzdpE%m5E3-#yB
zl{WrMM>LsYM2)zjEZcg?%GEVKzv`I~RPu8LYF!tnh((S+gDT#1U81e#JdbnKI5bO7
zhkFbGwFrF~*8<eC$RKND`D4#Kq&v_rtX5iHib%BQw8D$|X|I)lf-fGB6rGYt;soAv
z_QSaA>}+Xbeg(`_{nSM-YFCZPjO33mO$hE;f`=H7kQ_*W>t@~zwJ!>RKci<1hOm_R
z0=w3OU!bW_tCN|S1p|vAL?;vD^J9;YFS2VxcbrgSzD*kid{0;qKFXVjZQ$>m6rS?1
zw#E}am|xl8_GZDq^m0}^u|t<q1#5KXov~eBpO{*Ae{@Ha`I~bRn_0lRy@hBb`~dlu
z?R}>>(G1G<-uGMw4w}-XoVQ)jIzN{__7w{0qPg!0p@vR<rxNlgSNDi2yQ%?9!UKHk
z`Qd*zx;J7V|JZygx<_jNI&30rdJTgG$KLKfVN!lB9p-)aIy}1q0bm2n3avu=0`r;q
z82;vvma}tH@H7eHSc2hud+eE}6!B$%j<peh>H}fOPPIP;5b@=VGbyWgJqM+h7h~pR
z*?cdnEz~{RZm++-PHZzZ2wMf~!UwN_C%-1|6#^}A+RSdC&VQeCpaJn%GatRb4W8Te
zSumQLfEaf3r&52DgbAqtSL;I>GZilB4mh)jg)n|3PyTt3r*=^{LzcZ0RRNP+PhbI@
zk9lyFkvO%d3|7Hn*>}`KWkOFKZtiwV>Ft4oGki}t1z_<xla0#NR@;{|2&pmwM;$=4
zfF|n>EL11d$9p!r>GFWkv{W2?r&zspd)ryNgLZv}^f3qgU3W3*2T~-I<K3ie`h=X|
z?Z&b{Td3Pla=~9(Fs=;F18ue)iBPsW2JCr%^NuX=6WzM7yGgV4b7wZFp=p|s9PhVt
zsJ6<AUGNOLV6`_$k8wx3?zXV*iB@0<cJM@YP`b*JR)3)Rq*Gc1t6LpIXP3~Fjky_q
zB+9!FOmvL5r1HsVuktYB!PXhCe7%g7#g^cVgQ8b^P_?bspVS=l&R6*qO~R~`PD#Zb
zf~`nlcLV32dqE>JhypWc-xH9;Wf0wn1;is!D*Qk#t`L<x!ABj7ji`y0p%|S9sih+?
zxZaz{rva{WBF!YH&O5Vsw)EDgV7K+PxeTj3l+1WuR_tPEs;@x_A7T9bI_`&=#_5L$
zmkg8T!=i{1a_-6<;{8}!lIWdtgXz`q?`H@gTe(kEE5r9;WRZ-N63V~q)jarOju3mq
zNHHb9J=;}rzS&^HG7Ue$lkzohf*IhrtmFBtcm4fFgciX@7~XLwnky6__S$^mfwnyu
z7FX(Tqrs+9#PvLB*z1DVUl(c!42y?m!*fSPsyd6h<aeaSp!+t9j#VXFAlVjKBro%5
zi$23n@QmJl&kLTj-M{Ejb3<0U;5s2A8KSU+>CwuG#kKR}o{@YRl7l_p|0X1V?g$11
z6r~=~+{r;{L_hRZw}j33{?<apJuO}X&CWahaO1DOxUw5}s1_=(ZJPH!T);2oS!*FS
z&8T<aR{997v)l6}`=O0`gjw6XX}Srbz;e;6<L*oamnQYshpLC5EeS=Kz?#BCs7g7x
z@@PIjY+5)xz1JZkM;#B>E}X1kt9IG`kuhQdsob6xz5w4ex!Pg&f;#q!qv%P}Uekh=
zp%dCn?8iv~YK5L4D8KDw4nf70R_xCH!KjJM!ao10b;(1DB-|ij@IK26l{9=zf00SH
z&Aj;pOaHM;f^4YBjVP~Q7s9El5kz+~nmkOni8IclDKO_5)d<V_;gDXc8V`$pw(jLe
z=_HaQno{DdK|MveThS7s@VYtHt-CuX6DH~D&CxMg;^2bCm5t<PTJCri=R(KQJ@7_D
z`cYr2aBUESr{j>>Mbx6C_b1RDEH?=AMxS_Gp}Pb;r%%Qv3-yaePW1KjSL-M-d{669
zK4moI<eUpr%8jrIKTA1DbYGFgUcg;X_b;n7;rShdpm>nnE2$=*5(#;jVnndmbovwo
zERR2W{k12R(0n&1gGX-tc_mD;uBUWF-LSai$ZSLx$%!xJQRrXRq`TD4b`HHN>vB}J
zaEUBmvNU^VgvR=M-zObK8(RG&0*F`-*V<AaLoKp-_dt^k&{n5@Rl8M$RRW$pd5a)n
zGP`5vxQb`!G~wTWzDyJ+F6ZUKswG3U{hDX5phI?VkQ#7*PQ_=g-t-LdJV6og49={n
zUsNGn>Ck&u=I1-jSU??U#M2jflJY*)w9Fv0*gg9C+`K_Cy_LbiiBLwPSuHl~JE3@n
zx3|7u${cyDBkg;Xm^IvC(##sFNL&GP36|K|xzuS6%|~`TkNN8*Bc)VN|GG4--sXGa
zJ&AIWOMDT)+U+SEJ$P+Bzp-dSZbMpCMI>E7k1wGhcFZx5CoOh*KZfDnQ>2%4D0Qb6
z)(ESnqNM?2wpz}wu1Ked)fq$Hz%kuKp_nB$c@L!*!cA<ykuj%h2`F6I_*tyRqDigE
z%nSIJpE@-IQv#SCP>`ML*mvAKZpwwKFUT#H@>w_CFRp{sr2dhPCm3-*gLf}f%=ojd
z0C4#3w(f)1`hm4H#P|cAkuMM2skB|J<>cuG9Is8=FdiN`h`4EEW|MtU;&xvSRyR9%
z*d>)d^8@zZl#^x=y86qHv--?#bNAz_TOWOe$cb~Av4>`DdzZxzE9e48J7emGCkQm8
zXl?5=R{mv?Snk|~3)n=4`_Fpb77O=N&paG&RQG$*8NL_8nP7MGzn$n}OZ#2BmgwZE
zny&QV2tJEu*z}8c&f1vtv%uOWC@33ZNrxe3{`o??C7xZ3C!CfJINILxNO4Y^7_v&1
zo|fy3I*}}NM7JYU?cmeJNs1W~O00gG{#mS!kgEm=cx_Ne06utymocA487*7_&k3dQ
zG}d2c^|PKc1!wg<Md4MVDRkWvC%?8~oy622m)hZfhQdf>{CtPxXjXFY%7J?3g!Xjl
zc5D0l^$Rx<t<t;9SX26YPvc%9Rz3yAuyrmc)0yC%t^XylY$)}Z^_>05T$S$gUN>{H
zh9gw}*n=?l^#VI=MD>-q625T2?<Tpsgl{+qq=Vf0)X#UMWy0Q_8>4G-9+IQ5_<#oZ
z2<a5*c!y-z3OE%`u&T!x#KH}a?%pHJK;!5MEW0l)s}4P_ZW;XgIC&JMA^U=+3Gw)+
z=&E5tv@Wslj8^Rr{iFeXsojN4RsHTfu<gjPWQ`w6(!}l;opisoV>INM8>r<&o0MkM
z_zX(>)rf=4*d_L@dxZC~4&2|q0<rg1hhFAPnY`%Q6Ga<&Qisw~g#7YARZr$>`QIup
zh2*0m++Z~?Lf&UXYwa@WEFd(xpQcBx+xa=72jHXsJBeH?VG`pg5nRK3VIN|jAItF1
ze|b*kow{IS{)j?!UCL_1t=BaAC4YcGiy&GOHp!M)=)+8LV$FE%=*0Lpa42bLk!}P<
zs0(RdhBj?Co%DWUsvzn*%@n&*#U5q=c$=JSiOtRB-#3b-Hyrm3IFEP+y~4qrKZ!DF
zgRx!*-y)_ilAaNGI#v-;5R9<o!8}D*a3ODP<mLMXA2^8OvHARwj$~3M{UTQVS$*j`
z8`51$=Fz1))Wj9NE~O3dk_=?3Ax2Klu7P-&J#o+qVM(Lg^UIsj0Vu=kzLPCvxxn$q
zHlMfHQe$A>AR7K%(M)6Dg%4m_dS~k`{uVrwFt~ByNO5t`W3EQIE-1ZhC?!PPOSrQZ
z+ab(Ez>PeN7wjB9Y{7SzK-1nljftm&aJ?g`F2yI;GZ)qg(z|8$MOMPov91R9TiDvo
zdE!1-7&uc^&oV{t7PnG>P6)UXb*MDXGl?}&Rz`LIN<7fZw%f+2pt-TD@O3OT35|BS
z%2|De5)Z``H0`9Pq<+K3#p8QTlL9i%8`xJG<w1=!-@H_jBbsg+9aYJEKb|UTWfuXZ
z3&U@1vro1s`l|N}LD{S)>5EF%`EqEYySaK8!A6OO%HHOT4d1%xc77mE*wQmP?hVAj
z3Jg8Z-z4&<)arC53Qaj!o8$5EAHibb>4RCp#=xu}!4fI#Q#6g#nkRNHb(OGf#L)tP
zW5$>0=s2IrKiD{|Ed4z*yvY}qU5wbwhNDCL-^G2%PcGci5_YFuDjIKuRSQy*FI~c`
z3iQdr4>Iaa2jiw+!VU%qirq~4kh|c7E&2XYv0D}JR6e|sdAY!_BFof?%Sp8M2eun-
zWW~w;h~Nx)9Z7)e^d1D1-XKX6Zk9oKwJRgzIVkiXF_tn6O~F203facE5P2v+;<<2f
z9V04{`9(eY<u84^ghu*mpD2Xihvdcl3;yq}(5kHfe^1b9aIj|*Xi`RDO3=wb>1w%#
zO4j5L{5HNE_J|z?bWju1)j%d}ACaHQhgKg0<*mEP!$;+B@!f{^1fNDd>3k;~&tRyf
z^;0aR5=v<%2+Y{O(T}ecoeklvEBsLfx8h>Ika$NNi81Vs9DlFMlF%rZO^+uSZrUD&
zbj}c9@cu?Y^W8x(ZF7L<Tp`jyUps9iHMelQMz1qxSPq72^RLJsWI$<<^w7R8qG|5v
zO4z{}^rs9qvt%I1r1=c!>}iUwJnAu1!!Fpw*i77~(>7eR5%@ExI4o=5vj>*YTfw*%
z)L%05xpE835M%bFK@TRrH`WW^o#4cj(4a06v;(xoy?i)*M!+42H*9WlP=i6Zr%vB>
zJpC7|T@Rk*O|VN|4jz9#NV-zrdF<*iq{5M;8Xus*_00T--ueOMy<fZT-b|hxQZEgc
zvp~9oXOBH^IyLftHzDz=@HW5&MCE6siCUJWm^BC;YQVP}=`v1PLQSW*%$5_X5H0YK
zWd|%G80ZFn-szH^!!i#ne`R}=-Tl4kaS_3=z#RMuiuBR6qx`K@(Xhw4L%seeMklZX
z^^MqVj@^ci(!^Z4DKG#f7(r)<u^-N>f@t5#j_V@n3e_~#_p>|>w!`QYxlW>1TNs^K
zoFr?PmyyNSou7=T9cF<cqJ1m-tVB^yYQM^2TGTW8Qdl@P2*sr+i~WK7L!T6hgc2LU
z^yCkrzH{Gqk^UhSL9`V#e|@uUqVOXY&;yb;yodS*ea`S!totC!y^9n@?E|fi-cBPG
zMd~iUC<ROn4sd^VP;F(|Me``qwY!VM9S>i_cixf28AlnHIKBW58FJuobu~Mb-Dfo0
zmh+<KSltsm-<iuejIZeta^s+8v*`*SuZ#3;lsjAUEJrJZ(og~$5Pargr{7IpL=3?I
z<h+3^9Um$vWjO`cfr~ytWh#1pk;A(eyio?Uco$VU?(XJKg>@meUsH6;>q8a3!ag%W
z2FIecAKdy@P{<Oo)fZU6%!l|JgLHLuSrkq&H|=2{>1A;Gm8X85E{HH%w}W~zO7K+G
zP0B;_HTZr`$b)5o8EbRL7gtM^A67mAZO4}O7R(l_kGCTWxM4G{BU>~62g9;@k{PAH
zL9LmC)zZl4LCFeOB~LZNyLY7P*aPFWMG~!$0E4oJ1Pb3wT-w$9<Ae4!-c+u7bvQJ&
zbw|E^Ajhw=z?^R%StsQI&N6S8Ch%eduJQT;T}D%l`+v=NP%0__GxZTp^;x^~+n>d_
zGOP}EEQ1Fo2((Y$JvI5Bha}U@ch}^V(H>?!+Mp)OtWj~)VBRe^MHXgibkI+p7Q@p!
zLeUs6ep&D-vy0;PL5<;4Dc5NAIki+ky>~g5$xof+Mj9oqycCPH*P8>w!w0shnM>n;
zmpDur<q2o3{u21aNy^S$fSi!s^d6`HOtpGsp7&oV7*|JVa(=fbZ6Xge<T)y!M9T_s
z9T6fjtzT^B&mPOySiZJH?Y9xDs0LKPDxUuQ1A_0VzCI7-?aU=4C%+gc<O05gYX<mc
z0+i0MX5#Em>C(Ft*2S;v0@qU$9ntPT05Ve~^H+E@gmB=Rkk0&tGN_crg@aiX>!Vq6
zP3>9kq`MarCx&8@{SK}_^McZfGrx-_qQ?12x+K0NuEzQdy9%H4KoCno*XLimDY8|W
zZ8q`p)kYUGB0pRtNRJl7<4v0%ou$((7+LR-C(=;vk(^1}$tP^%N-b%zAHIzdxV0{Z
zW8Vqaze>GF_$wWLB-k_Kv{y#g4?JgmXwZn<QQy44N;L$&voegO?aXfGDIyq*4(xi|
zmi9PWFreFfVf5Tf82@FeHwSS`^}?@r$g$;MbNs8k-#~PRYGK)nvwe~=w&7|wud@mT
z0OiTwWf(`h6I4(BY`=9&nux?Do7G0XQWx!`k}_(uAqr;cfaPVKek7i8Ce6K<BHA?H
z7|Dqv9_%+xRx?nflUj0V#Uq^{s&;S2Y_%aW<iu-^FfIuk$a%P)g#Gx(JYyNZMy*bt
z6{GY1HYz2)7~>Z<dl<9La6N?UfVKSeJ;EMFqd}B(>`~52ckHbG`0_Hu#EJQ@FX5GI
zfKOB)eXHPiE1@0C#l?#F9&lRej%s<RBM)yx&3WH<zomC=JgxS1ins)@J8nal@?$c6
z%6u$DEh=Z3zmFq}R-zK{G=|XRWQogD*0afMxRh(WmEf$yH}|A6M52w$N|*>yr}j)|
zu~GF#>4<S(1OTBjMm#C|eAVK_2h0S9EEO|`M{)H`e<QoukDM=LN1_;#6@*i?Tk>D4
zlgjTHm&M~(f$9UR;kP15dQ_>;>e<L@BpW1VsZW$xvL_5)p9HDfM<9sG4Ywim2h<Qd
zM-EQf#xE|eP*>z>U9}(^Au%v7_Xg3dk?lQsa6l{@oX*=gyunuut4>#)G&l@G3{6<s
zE5ZlZ*)KqCZH*3eDVCmHQ~hU1ro!&9sqcG9=<BM(uuw_j*h&n6_m>Akfw(l;w6=3L
zG*}BBSnEgnMELNj6FQR*E1-{sg}$!OW0&-#)z7~JRhO`;-B*`;Q&X@dvV}m@464v(
z{-F3eJ(1nhzK-)S<XRU$)E4~he&G1qrq8qd00eYK0S{y<fZ*FZlAe;D*U>HY{`8?z
zohjI!;hS*$2q)yWXJM}Am)o#Eu!lI1YUYBB&Woe$DiG75@mNb-4xxaH)*}upwL}fF
zll94U$bq_Pkx2JihO6!t&IgjK@S%;(oIIxIw@cv|u~8;ap5o>yPz^Yp-)nhf%=7-A
zj4cBOWo+0y+d?;V-6Na_xtjlNPj)UK_&6%=4iicR)*m5(9Z`my^yZW~&u_;~pP|5~
z7f^M-%3pP<{>Ue___eo<4sa26xq<nvy9v)AM-oE4g{bf2r$~4no<}{Vc4M(ZcPu6z
zbO=)}Kor+k(Y_ycv*Yxt0$ZwIwPp8XHtS;#F+E^}>yApNsoD0J`oQuOTvI!HpQ)!s
zFke+OuT6*_rD&DA&-i;DJR*6Q@Lg94<^25tlqaD`Tv`H!dxHT`(Xjjh*|z7cEL7jW
z{G8voS>B7u<UeNtgemRac|fX2dUILMI!0&L=Kelj;r!9FzSZ>rwCS9sYCR3fgoN=f
z+V-Q>cg=n6d1dq(`P1?8t(F=cm=i{u>epLzC{VnTeDy=UyitXAr3aXkex9={ggY3=
zfu)!{)t+y{Xb$Y<7J<TMAxq(Gp!*t({E|-bSwO^$zW;TrPfJq;s>y#{S98n?htJcJ
z82!rx#)dD{jaUk8{z{8l*9#CJo+E{Qg*~4CsRclB*1^lBx#`W|i26@;aW$d-8klh8
zx$!m3V9~R`^<7^I83~C$dq8c!vLF>hs$krXrr(pc@!Y~;+-2b-x?8ofVs%F2>!2`U
zlippx3q+%HOVXm`MItBfjxc|W=6uUwOsx_P-FRmQ2o;0&vI`GM!bn95kNe^S6O)oT
z3z~k>%`H8h4<$JjeLEo}UHzcKFrq=&*>|Q~Q;z4v4&~)&`l*UOE=TKxDs{%cES8US
zxwzhOGmQaguz$o7%dit3+@(rJAlH{!Q9Nn!pFkC}<Wp($-ggyWzdZ8VEiOWGrl0*$
z)VKVVQ}c5`G*JaYRzgK60#JeA?4CDwO{7{V73Btc^~eeC_k~nf$ap?3fX$yPB({5y
zy6_}t8EYyZZAHd-jSfYov&Zc#%0C&3=O%KOf3o8II}R8D)8`Fxa+Z<cIPF)p+RHMb
zjGcl;(Txz2knP&>6AVBV|5)X2>YRB(T5aacXb$=@wNA^BWT5{{>@znZU<cakg-|`b
z$)s540Rbd}up^+AT55Qt07DflE)PKL5>bm?f@T5Kv4s(^%`d^!&)Z-5tu)PH&dRhF
z{HFS-E)*pp->%(|UQIYce>s><l0}IkM#GfJj9BIqtlzT<U~!9YlsR?k%Lur(op~Vs
ze1*pvIA(Smu16dDtpCUgsuQoS_*gGK)J*w?KnlnSArI}HoGy;7YWn-mEScxoM1%`_
zq2i7~kvd$6b^A{YZzwxO6m3ePCo>P_96>ZVzt0f-Oxa}j(`a7?yo<3ClNWm}o09F8
zH+py5&g&9s52U7JkS(E0E!#4g5kM9vg!&rbP=`+;kXK9v1(Am{RiK)8z5-5WO`K|S
z`S!SMm4Te0eQ(k8V&{?^GmnZ_GRBqfjGUl|`ohh4${+=Y<A$*qYTXZ?`(~@)wCVoo
z#a7N5)`z~L{=UET3QeArJ9!!SNa)wc?N}zU7*wZ7)g438V;KrrJ-_52H1ZTVj$}!K
z=&pNkytr_pM`p!ENjwi{-zysDrV3OL%NClHy!?`Gz?%m-tfIw?+j-4IM-=z&2XCwj
z)zM1$)brCWu^dQJ+{sVj-+8E@ta06KsdCzoYTnQlwv;`t%pI-QxoQ;s>evXZp$rMy
zYaqb4pUuGz#0qVy#I$1dk{M<~b9-tJ)DhLTZMv(6Q}u+Q^XIISW^vYEUR_38`T4Mt
zGtKVRRwVz>wp!esP1%?`ogGmUDnm57;z*wLo9)A3`CYW3<8s$Nr5#~Ngi(uc(&!&?
zH++7ulasnftWbT@F-KJs$Q^Ntsi!Pjhy0=#p^<I^2F)CPrZJ4)YlCb^!WH3~J*k=U
zT7t9$hvV%m5Ak66OJL?)^HH&JGgq}V<#Oxp?0_rKSawGeci|fEm~dr~7?BA|M;iHe
zgp|@$4TBLwZ}T?=ysH!JL}O93vp;&F+KMn*;UOsP8co$G*ANytWDaPJ5auD=K<`BK
z9XruOOX%c!dl{|`w@cz`7M#$Bs0~A5<W%}G<yig#ZU~CBhwAL;UFe1Z7egMt#>QXz
zBYt7xXz6kYHg{z!+djzH(8=8NNy_J7eJ~ON!Soweq*=$K*#1*pTZ<|6(j-pOA8aHp
z+l&djOPVIs!Q6v>?7N-gK{Ya@L^fBj#GYB7V#W?7*_v%q*De84ALI$(g^-VdleZ#<
zS0@Zb@ZFVyYkE*E=rl$#tr{au(IJ<(C13UI#Hs-IL*<E;E#W?sazlB59=7^<`g5^t
zV`!ykOquW;f+0ci(<cZQ6H{Elkk9km4wbQopd@tA%T9?0m4s9juIxLbQu_m$1TN<>
zp#va6Y8J0q=0MOmW9-9;=8hY(SX)#HIFWs7rBa)x=J<$*Oj-)@u%S+9^qJ%KVI8bS
zy-uTG>z1aVAyhXMKeoz?D%hI#9FJ1WdHxOjI04UKpUk~~{mDb<P|G!RZgP-{))$)8
z3OUBs*T?p#e>~MdV;=UemM-{iCiR%h9S``W+vK49^Nq+YYtgrtk`^+)%)f{#sx)7u
z+6nV66h&EFEk%&)yGsH~@uloRGzXkJ3Vt^S;X_wDP=rD2e8eiPE)$=*v)I=eFEtrG
zoxVZv4t<Z%5#tV>--alGX7dTM!+YU@ssaR4+!(_z9zj+=P_4zj<2EM3!?(Mpmlh*D
zCJ%vv>8htBTH4>=C3zt+i<f%wCJ-tJa`3dwDzj=zH7sKqc4j;t*V!CK0k6kCKqZZ-
zC!@4)MUhpot?s~B`D(V%%MfTZH)c+&e=+WEM0gbD?ikHLbM9-XttR25h$PHoDLP^i
zp{x`;Xl9zrr689O%}jya<ryBvZW|mZBv`%#g^Am-OpQD{e81=QXC$%w*z4C-p2IQ}
zAtV5NN#7&8OF1=^0>0?OA%xrm6CV(C*97iX^{M0guxu&=F3Ui4{rmMph2VTmgp+Iw
z><^gpCX(>JsdhO#PH5(jkS_5Igz>nlWkwh+my@<-C)I(Qkp?wvSO)IHdCV7WGW*bo
zD3LV)e!OLVH%P4`q759#G=v3_g}A66ey9xG)eBJw<)Ks&q{*P`_>uh(kA7K3H@~aX
zxzM<m{ft1QqtkhwUu#!_yO<9!TQgebFaNX-$6DF@Ucs7CAx`O^6nj%3eg3IPw)DkM
zudZBt)pfbbTkm0D<+H&D4qsL#X1FI+0;;<5F7|Z460>EOw#_<+?mffC2+-XEk_vdQ
zJ`Nwt)oYKjX$ZEDbZqB{UtKRKyEgiJ(tcZOfuVzC=JU=jov?F~LoK*)NuA$3)t6dg
zl}pfTwfI-u=$d!M>}QEUJ<Z`H+D;DOIlt`XhU>Y#53UBg(zv9V#)L;9R!5KAlDeO(
z0u*&mJ&HW^aasuVb_ZK>R7b;N?r71oEXA%@xl_S^4b4rp_d$EVMRkre&;B;40Brfm
zN6E;_^Rg45;Q=cp%2SZCM{XauJZd0FtDB@gtM;J`>s_P80hm!~(8qFV^2Lx-$clak
zu?a|dedwkWy)zz(sKlv6zeASyG$c>l{tGYUj+g^3^yY~lcX?t<rFy%+28sWwV+=+z
z%otxh;X1R^L^rr+cM{N~-e-OJo@YK*Xzi;N*0iRgsZmAMZ-qwpzA@8<H(vIL?Xe(}
zr_<>NR<=L1qk<=RW5|_cNA-AJ!`FP3{$MghR>%EC^eXI&&?X}}=Hsph%fs{*8ZCN0
z)KvwzMgld=<Pu9OZ*2L&-8;yXC!@>Y?XR=doL{ljN-t~-k(I`+uxdC^`Ci3I+#;(1
zI_S}PCE=>KjB>08Det>GZEJ!zJYflHW3}{_XI7OBlbU^~!{}=Mw?k(n4AdY`kRuIy
zv8%L}snfgHPO7jDfBS$Fgb%(NVE+Pyxqlm@6NWpF9u0JAJ|sG(U8AtoJa~r^vVH61
zl6Xdyp{z4_HGL*_s4<Sz;W;~y{#n9dxD`z|=>Gclp&tDK4@KB~Z#8Ufe<-Q`-2gSG
z@ybV_N?euOhP5A|C+<l9oHn|cfK8A}I}ZmmF5Qgky00zZ|EsNh2sL`o7ds0mz=}T1
zH*sp^9w2+1=~icfR>Y3#m)LVjxtf`Qw=ZIKzQj7H3a$X9lBc{mp(dHn9OzCA-h2Jx
zP%+yVt5^9qCY~en)cMQmPqSE`6`p4Q0Rw_a`S5L?P_`y}#c5I(Nzu~B(lAM)q;Lte
zaGJfP+{;Xy^@yoG9Tj7mj+AhwzkaHGD&^J5@}C@rS!UwAJ>w*Ew}v#E{P{C=^ZQOz
zJ84QflyEH6_Ul6ggCZb+&`r#KfhnO((q$z6#5mJ$pAHRK>gRC!a=H2MI_p#MtKRz_
z^b1}Sq&h+&-;O<*Kh%H8k{ZaPX%J}CiCO?CVfI}ogs<xDcd$hBQ!dJdKB0QNC9VM0
zir+&e<xx~A{!8r*w~C986>Qur6Ww?CXk`{dXf#==@?zfdxq{r;_rzwiJgdH@WPID0
zdWYQshBRK&>l=u3`W>Pg9nMJ+pfsG%GfdKkR0`^4)z6?VqhK~igzkZ()1m627I&%!
z>~?qi-o|=b{@fDt2d1txN|FYIC)s&D1vsg8Xd1sINJu1@6*<&_!{3xKB>6{$4%W>&
z7r9&H55b4;Z&388k2HD{O+$Czcov?{o|;Oy$|Q3FIAHo>y**HE5m_(*<iJ;}F1e>%
z;nRfe2&u2Knr|)4K2H;<zN%;WHqZ3#17MOZFIkFv;mz!W>?4H~?NHW-?<0IghEG1d
z;b*@O!q#rTe(!Y09(LD`7x<dG$tF{Gk{nBJlF_T>zSjc=j+s<SH_>Kxwy0j4y=2b(
zYZ7mPe0mwW(}RBUmhc2~2au$Gzt9l;!`a1js{;{z{bzqBwgIMR1sTF?=IpG}+1tiF
z;7QphdmXp-K^4j+%GbU}rSU3Z$Q^Ioz8)3y0VC@UKbZw5NWJYgQaygpK|%psr*Dk<
zG!Ym+B}H+oI4#v(N3OE;G&EFIfmZi`xXb4D1@o1FU>vLeDbv+*j%TU>v#FK{Brwtq
zvM;ImD?-vM?wVMJ<T>A=7h4_?>gAUj8gTh$S0tW3)HZIPTFWL6SY0}8`1)kF|L_sy
z+RULcImeOqXWCf4p3)HY-3Q}N4z^jfAJ6WGkoFM0U9UEyU}={}VUUlE4HMoie)yIN
zHY?mwGd}so;}Iz#;i`}G;E|Q*dOLF?JQH+gW_(ZMq@)yI{9J$<aHh$D#hbUAzs)73
z80gUopIaEbe>|tZi&mSz0vHNt<KcX%Iuv)x+Vlx!rcu?h7Q~<C>EvvOO@uEnTN3#I
z<f;j5Nv^IS-Tr#-rx*EJuOFrCUZ3S_AYPzpO<5pX{K==J`(d=z;f}5|<T!Fw_B*2_
zKFA#PlFmFGgYznN-O1bd!mo$@^pURvkxkY(6MS+9_K=ayn<?~m;8P_3h4Kl^VAzhr
z8ZQu#HlyC?C6}QgBRP(UrHZvW_Es|EJ4MF95{H@(mAOWHimi@LG;rV0fch*nkS-fR
z6?Bh~Uk}cev*v)~G{IsSxk)l%TFQy05yBbpnfvk*dna=rBKwTAzJeuAB@pe+Nt(3F
zeI(<<BwuiqM&sc}^K;yd%H|@go=Z5{n)4SS@qA{6#KTzzpU#*xm0DV7l~0~DXkP&@
zk#)}7A)hK`QvvRz?q6j;+}^C`6{-%K-fHl4eq+dPw5AtyCx&BKDaBVAYDT57<DF8R
ziL@}%8VB>0+}4%NQ|E0zS(a6199Z>yPxbnw6n3Vf0o$R8J8?o7|D<rK;}_%iHN7gT
zA?2rS(e&j@DX(Z6C2&wc+K9;BX=O2fS%zZ)!tZO3dGX+EjRis^f)S+w8#kCJvA5?%
zoSA2t3ItkgCvM57#jqYwQQ|9JpSdDT(ISV?77dD^Bj}d>!8Z{PJZNKS0*+)C-w|sj
zUD$H(JV$EK9kr0kl*iN3w{Hp3GzdSrcS3mag4QHv@TfC$Lkt{#sL67q?#W#(wnlQ1
zFCNgz>TON9YI&XsIFO3kpQXNhw%zV1VzHlx98xXy&BS~gY;i)8GropbQA3<>FEiNK
z<=^}LJi5zVGW1scO&jA81`w2R6d*8YDY{Y}ap3~fk;~>SU%^Y!<W=bNiCUYen<;;F
zoZihzo&6x6IEgs&s5UA7*4zaXgo8n$d&TsF@vmDBxq_05S<dCZ@xeN><vn<=70h>#
zR=sYkYV)nqKrv}QyHPU1#vc$`1SaWYIWs>=eb5Yjd>n&Zl%|IanGah&j;bA~Q%-nl
zQ0jQ%j?Py+eat@oEYp7@_i1|pTg&0A2N$%y(YygwLJ?mma^2F8#6E|Us!05PI~xgC
z8J9i53;!2qU*Q(z7yP>cEDZwEA=2Gl(t?DD!XiqSfV4F1(hVXZje>wdETDwc(%ni(
zEm9)2N-eqf_4~WO`v=_T;Q^laJ!j6@J#*&Fnfc6&EsLvxC6(FBmVE`{eU}FrH@;&U
zka!7tMyf9_T99cGc>1QmZW~T5w`qYB>(3nCXT{GkI%KM*pGiE_a2$45N~LWa)v&MV
z6%n-*>M|IpB<b7}blSV`MySgAq&$Lfu4GesB!MMhJsi>m9rpXEtvMTD>!X=f!pmzI
zcK{hnS^6M}Yv?wFG)An%%1|7ns;p{Yo?xk<#!*U(znDl(oPTs)C#}B(Ouwr$5qKsi
zD$R4`3b`QI&x5)o?Bx|hmHv}!U79wwh>)9?r$7p0m2_BjcO)o)pZJzy;R7hh<T}ht
zfyMI3xc{1TxntSD_V7oV9(d_7M!|Q`GZ4sWSo!yu6eP{z{OaOLjmX`|MYLqUX$$QM
zTc#|y8G>R7gkpoS`4L;l7iOvaT9%W)PpqZcxABc^`+C@VROQ!~;ljjN5w8~7yd3J1
z@PX?j`py7by&{u$_!WsJ5llxE&tIJlWHphK1x*yoJqK3}X)v(!zD97^q#&J29<j}S
z@w<+C!H1tn9v4i#oi*B@H@C;o`Qgta%QZ^S3JgI+PYkC)KxXL9{V-24))02{V$h_H
zU*r{oOn1e=Yw3rCcS5|>vY!S#UvGDzwlS8RrqTFDGLIjUXVU}8<(UqQUb90SG89Q0
zM?cb>5j2B`fm|d8V-tIKojf$S@9}_<<F<WxZ_Ezb!O;0z(WQ(RaUffn2xv3i(y9t{
z$8<HM)+ybj$k>keHpwC{7O(<?mkG$}0v}L$rBig=45XU_iB2~Kk(&5A;I$QmxdLdz
z$>nn<o@g0*vtK^VDLu#=fI`uHHwmaq_Fi6z6$e1^X1zN%?|w*by2cB=3ftCNq{OZe
zgl8<aooiqkJ6E8}@7kJwNegCyg6V}yC#lM%%aM4d<OlM_iy)+ns9DCv0uw<4pEN@{
zu_#oT<eIRO3LUd9doRhyRh%*cK^{oDsAaz?pk<hT#jtVz&b4kn_M|fi8(?Sj@e>zS
zv^w(OBg81aLH(p7G`xOjRtR*T$@>a1i1!or07`n==(A1^UD08kFn-v^_i@F)xB)Fe
znDClK^18$qIL4J%l#M9V7%oyy2utOfcDwX%Q8~u37`kTfMX?aF*RQ^|DT`sS3Df#1
z4DtX~gAsBlGYeQrFTki96KEpzvhDW_KCM2D%E-Si4$=Nzug5yRpo82##zCtSmBm4!
zq?e;J@^IqGp99yg$20CinmTbo=$`{@Og5$(8JzI^kzVf4teb&GgtPrf&g0}yavmHw
zJVzjlSI}<fG*S(O7s8NKk!kbY*8^d|KE{<*U6MX55t&hWGYKFymTBMQS8p!U0_I9C
z-7%~-K}<DlME($#=kj_|4`UX{7H`U*j`tLB{y89>)4=LrTzw0g2pGyWGPOVbz`x6-
zdZP1;1}U_ol%=!EMnM9?@#Iimx%zCxI@vK^64|I=KYpzH2z=8WJ*>>;I_$rQRB7%K
zs(e#UpwWmY{kQZToel^3S)d}SRz4rgt&zX~v?9+ZUHJiXo(@3{nHa?1Qn*hS^-0)#
zgaFy@>!r$@9uOQ8&K!A~E=2MT+6rx=zU+ut$m$`FaiPn~%zjF+2rD3G{jlfvd#H5X
z!(iy`Bv0U0#?Her&^FPyI$e)-0#m}vB}{uZT6q&{4%)IK>wD})O-57+YjBw~A-j|m
zrbfm7>{vBKi$#GVT|}#i*i#q7GoO4&GAezvhj{e_wx}I6w#{EO#R@ZwwI;Z<T~GDK
z)VRKWx?$Xty7!y8b)JCt23DA!+?)um9SCcUS?oh|j)MRl$#u6N`aR=3{)q#AC+!6P
zXiY7Vrzhr;bP@XEdPEgt=WgBpCe~9qkr3Jd-a)qg$e#slo`GsH8Ar-y^9ZV@+E}A|
z(zMT?0}duoASF&@<jxg||D8l_vr}`~s;^i@Ock0cLIE96+e~IcNw|(+POy%ZSyU?i
zp%g=Jr(P!%5rrv+6cGkg_7zX7H!Yz>)QQ$*=4w;KHpfBXRF|dbHbKIhh6cdGa-WfB
zDdFF~kU^UFd3b(Gv0o}CZ@r=9Ig{ui{1TPtUX$(2lVL781`UW*BW_Jvlo{$phhpWP
z8FFgfyL};+$6Cz0^1)K~3eiE}du|<H#pP<q(J-R`wLIOW3LbBBsgo^{T%^Vw*km<(
zayDvkWlAZ-E~_xId=i?Rzx4q5Qv<<EG+oz7DL3ZIR%f8cB}Ce-LDsN+Lm!Qj?^v>k
z47-5#i8td`352yj%Tj`jdmO@uYGcmeUJ5bWTw?Lb+jaW%cHnxs2n5M=ZKz=`EbC_~
zBmjwEbyK9H;x3bCpZn7Rn%RaY@xJ@FsP6Co9#;jk$2)J%-qD}5rfkjlxl(0f!?77k
z8V7UI?Rz3DKWV<M(q$#$3Be0v<>eLXhq_b<uc@)NK54%1zl11lqFufpG*B1D?{>KO
z8}Z4l{^UosoN%md*{kmd*00(nL;6vPlpM{h=vNQ!83Mp)GJ@6{m+bDI*OlDlLY+Yv
z66{W`%t}x#-v~^VX?Nk^flrYOA0THI$v^u@SB^Q_l%y!A`G>e&we(%Y<k%{@<?$Ip
zYV4*|#9(8axyy2d!d(#3Pe@Y%R_eR<W4~q4X48!UzY|O(Jg9LJI9-&JmXN)1PLhJ5
zl>}eHxnz8K$@e`3&E-nYrQru=bp4=+)N^&d4+s4PZ6-73XGODi>f0K-kp0GS{@xF}
z(>bF`e2g)94W~A_Vdh8#)3E=NaK(!cP3E0=S}e=D{%s&nUr*$cTD^fXrxw{QfCUPZ
z>>I7Ct42>Kn?U!p_{(?HVy=w*?c6e+qE~+&VpH93*|5*ZVYqDwC)7qAvhCD?hz-8u
zHB9XAMTi!P=*%2DZi}C7s{DHsEr@L#z;phHF-n<Op$er;Dv;%x0=X-;La#wd?Qx4l
z)kZ6vD^Uu`)nzz*xLUdKiN<mIvy~(T_p$DCQiId55W?xc1(PLbf5m0nu?iNHwsnC3
z#v2pv#g$T!D{o}!21+b{N3q+{K|O6wF;8`=>B!=J0HkS~WeD^V%KldS;jdwPz?{y7
zWck5&2nwbIQ>OB$s~g{uz64=S{<3I5lWlG`yWND}F#Nar($8)18C_I1Qx7NCj>j=p
zpC&I(p$EKbrpoKYo94kEWhDauiI1zWx%exUqnh9V*|9%rT8|LyJq3Aotp~5EK1kae
zJTUV_roO4#L-5=L?cI5s@{W4CHuK3Af++9<z*KC!g;;-5rI^k<u*zLSDV9aX?6=MA
zm*cfg)~{`o(&U|gV0R=m-*0}oxvGq!yyV?2@S|u>YLRww7lj}=u+hx?q@0hDzV?km
znbL=p^)cqX>^TG;5Oc>zV1mOR9^ESm^KHW@_g49P5vO2JlHhgax8x_eUfr0E1Rkl@
z^e0;eYV`bH2i_el7WIuA6LV^@c&Rp@Y=9c9tHIag7#-3YvIc0H&CPh3#ZrZ|D+(CW
zITF;bFt-xKd`9`(1LMOI-qIEeTzKBIF3Mxcusf(PR^MPhQD}5C29e-Eu1zLo3c4k%
zl82UkGI^n|Zt<z)X_L7BMb?kkKY(J9Z8;HQJ#IOaPgClc5?5m*;-I;C)8hSpZ<nE4
zAm=CqvzwKW?p^IBZ%p-@OYGD2VaT!lBZxV{X!LiEJJOJOgbBF)^(9f;O@iSt(&2^{
z?h~{;Hq6UXFlUjx$sQes0a%&Q-1a-z+sRgaVKxGBBk<u(@99-ZG*o>GeKBxQ=EsZ;
zQNO>zE<-|p+Z)hvK6#t$p45Mt7djIXf0iSkfi8sig+WgoJQxFi#1<zhl!}$2=qB|#
zn>f9kItbiN#NRwzG{ahVGZR?f2=SuiwGY(0_5BmONy$cz0LY6mi9ehO7dIDHHt~T*
zv!n`pg)XaADi%8$5ma3R9EWscqQuMte?#rrI>zkV2(1Vaf@bsqu9tdqwHDB8Ta2yD
z&<DCks>?>yZffI8Yj7R)4i!r7Ju#d?-Yoh(tPWp7W0v01pseL+h+EcaKSU7V*qiol
zj#X(b(qWZx49JYyJ80Lx9n%7-rvrZ$P&BrMuakbBMBtrw1KSf84-&cyn6_Xid)|*)
zNzxesaxvI@5wxWNAU8K``)iZAvktN*1%(f60R&C7msfM_mx6$8>33f+{S#0iR;l)y
zAU}eQl-MAS1?G|V5ZV(0Eug$q#py=q>hE;$Yx#<sf-pjzzU)bhr_OiNjQr@$?I^&S
zsSjRgc2gJk9i`OpAUX>=SxHZMYUH*Wh#W5~z@EHnLUk~qgAQS00_Cb0Z65hFg`yez
zm6n}>@qIWXlz5Io_}VJ7Tkl&$4IrF`Dv187UAzBvM5KaOJSq*658)e2ZkR3mt@V5S
z1>ONZ;*>Covm&#DY>g&zGutMxZ)MLX?Q`+@ice7~$EJfdi0EJP?UBG{XZtyiuZdK8
zvSfLF4<*=Bpd>-E9Fos<XRl8@<eYIPA2`&Dt+0c0%VC5w_SD$8+GU-PsNq7i`lDQ3
z@BB_Y)Z)YPF9q?GGtX)Qyl^%_ka&b1sR@1JgdCqM<eN0k3X?V{64pk$s3{Y85CdnL
zPatGp1TRh%q=cr*)r66u-VbE-0T+?}hwSDQ;q>&kzGnX1+R=osRXBgL#wId8Un00G
z%_HKD50k@oWaRUl8*e!7UId{J(%pR1Kr4`dH$)lu0jB?#7od`6B}orrrRRMCa~&{;
zEQ6hGzq9wIZZKl$c<cS%nMlzB+h;QH@!U0Ej4`yh``wvFIC*uVei%Z6caUW%z!0nk
zYOMWx`kWvfOi$e$|Feyt!k%%b5JhnhIY;8*bN<?o^qN4}d-cf97`>j&PS_yya%$@H
zm#c$syl9Qgp|W+w+d2_^pAeQEj5rY=2H^?6dBPa^<<=9&U0mx8#(0$vaqdzj=17K!
zQ4A4_!t_i(8`OiWPRT0mIpxm?Bq97IU?{ilEil^YG?ex!*N9KuGVm$H>!NJ@6L56A
zF$`atnJ~JetCw-N*7m)z#x{t9N9)eDnjV_%A}7qU(lt+wAzKj?oy@EH7>ads^CLTp
z0CU1J98loEk2Qu`!=^53vh9ff(WC7PvS$`sYDgs@b?CKKB{QZH<MS-RQyf&lMvfG~
zDj};n=VqE81p!y$1Do&ko_y8~^Mq-IFjg?0hMsz^+y?2wO9!GZ;cq8f2-FFx4DQ}y
z02gzwoFNFN=@uU>UC5Ej#VBaZ*lk3co8St3c>D8-FEruS5u-V=KRE|yGb0yZm+})j
z?UaB94}-97#w$V@VI4Ot2|P1dGpI|z4~6iDM1L<vE<ax9;j=KxsL5FCWCfJiJCx4J
zkPSIuibpY->^6jWhK4E%K}qIkt#R#bkiG<`f?YAq*eUk?fRk}*In%tx2fh7Ez>feD
zqn?f7Iljams_B1D<3JXtRB0OZuah#^`oG|vLJM>@QSR`d$O3s77nIyM4wSJPFA0Jq
zvTw2<P=RJLRzKRXQqVBY*pB8X{R&KQ(gK4x6MZGbm3%@9o~VqFV#y@7@Vj%+9b|Ds
zxf6<oa36vvHoF0+bNNhBj-Rac8$SHy#$BOR4WhjG8ghP3>_+B#T|^XXLBMYOq3j2o
zJHok;4kpX@dm#T$ig*0PRdS-|=g~gG3(r`|?s}j<X8BzUu&l&5=4l6gNS-Z@RX~)w
z-)<ym1UMa~U)3V9;7=Iy9K%{^x1WLuu=hu(pv;Zji3OL#+2<OEYu6Z!GeH-lRyUG0
z)w7or>By-y4AOQ1^MiAG$9`Z#*Xd($Yw^G18E5XrA{a96)Wt?<MNHJ0w?|MZX+v@n
z0#{#U*H!S0)?+A+ubS71`hA#!9*$r;06oFb{x!A>*>e-Ha+JtlyZM$Jrt;FNEBMj-
zJQHbFLQph+JQYieRuIji*UbY34&UqX6csFiSICa%lO+jEIF`yb_l-G(ZxmZ1YGWa;
zp^Q^;IEKl=$QZm?1E`yBUb5u~PxR95^MJG4ZhYvv1r6t*cb~tRtkP+x6YpO}<>rp7
zFsS*E55A_v(Uivs>aY3v$ffGm`U2te7CA@8ersWLI6b#{Z8>sf(6z0N%;p=_vVjcE
zmH-tm>Puc5CI1o1FEcAo-4?~Yti4#p(1H?4$HZ`C)%WUN{ry?{PUP(8T6ztFf$K&-
zw}DZ+6<v$ED9&(EtclN+9TSGv>(Ky-jFbn>99uG|xGNIO2V0{fb#Bz%?moHPvb#7H
z$)T`<dAy&g_fTD=L(yLz!U?Z=a)ckfb-j;zv)0?mNHHTURs!s$TcjXjlbpjrnP5{i
zS5`+MaOmRKD1wU0MGEjHi*z$Rzk{%RcClDLVaEIVQbM{j3O))AZvrWjvOB&pktIIx
z=7NK^NwD{7AI!<Do9Afh*;<g^t0g3!lTeKpA+UxPnPnkexPm2`Up|8CoXytye&ejP
zaAVlKt4xMtoWWhS&3Wn>0S@|Me5F{`$h;CJY~5)j#DKC!8`|NKxR&c5&g8|kIj}GA
zH!`NPNx4adJY0#sU@1vKhSG0&bHWg09D!i=)p>%f8a+>JwGo{kLe}1T|5p;LzZ3pC
zn)=c0i<Rs^U4JUtzwT!j&sLz9if4~P)CkRWZrUqhBpsq4#&coe?DWL)GK@(TA<)Vd
z^fR<lr)T$!hFC)fg@4gn-Q9MUdta3pKFJC;oB;J`Q9<0qM6J<0;_qqnzEKc@nB%%x
zJ<5JniF`PG((^emZt#;<P`OWkJqNk~Pi9A@3|pMB8pF$b@MS*>x3-}+?u-Nrp$>vt
zgS<NBI9sqyobdat@!QJ&cR$z)G^qfW=a0yn@Z3Y!L}#PTR{6l?+m)H#GT71c5!3kC
zLkssqyb)$=TKPZ^b1~P47XSk5U+<{4WcODL$|jQuyjTmBIODmbiVXSk`<V=(ha4}J
zAHybEm8w4K54)%d5z3xMMbs2SWdn|2>zw<}6!Cbg7JcM~K}1|-i=k7EYu`m$)mbDE
z<n<~^^y^T=O}^(-vmtj8AwD$CuY{O{CrpW!ZS(NM_EgFQ)fxhC{kz#V10Qe@v=JEb
zSz4$rsj60TH*vl!2?Kv3cTsO%^YNl(t%<}wg*&2M-3u43*R`)4grGn5iItFC>fd9j
zA4x3{X?;trM*@M0cX$Hz$ynp_Fy>H>gvE65b$eZTm?}jm#I2QfgHX7;wwt9BX91V(
zv~vghOHMm8Xbsv-o@^m0B$Iowd^THe?$>+UCWu{<Dco(!dqVM7H|j#mt@Z{P^D?~8
zt(`QKREbaK)uBTBBxG0FTz}(XBu+|Pi9wS5@_6KAbQ~65{AByhbF|!o(!$bsmu%#x
z+7!Ps{0jLz5^Yx5yqNxY1SNxQ%dz)-;bkE_T#ZSpCCAvK`Ne6PS^FiU`nhIxBey6Q
z!uAT9I3W)B1?W7!p<4+PfO4pK`MNW#W8}1%Jy3IZi~>#dMNN#bxa(y9Et_b01cxK(
zT;$%0oZa^nHap`m=0jK&rOsP+-DwF@a$B0*)nEs1SxPW$A%tRy`p1v!PmnhfmwbW&
z*HI7|ik?%5XVTJHXv`P_)|wx>g;z9KfG3d+W7J|rh#;4s?D-pJWOUOF#_6N4SSgsn
z?KTHx_W8YguoGv`i*WZ&|JxQH%Cy&uoMng4#D2t=p{pF*outa1J4I)S>CWGeJXFiP
zPEf@2i7(7;vM>aYXW&B=;l2fl^Q~T)zXyO4_i@&)0cTM;DoYS|*8@QXIUO_C4=k3>
z#Zd}ePCwvEbyN2?BuT;<7edosvD{(oZtwpik)WW;%Qf2#>du)F*7%mfy@w#h$<LI&
z%hYNQENkK)(Oy@?zje>GLZ||#rV7784f&H;ckQ2T7>4v09{qSV&SxWL#o$&pc>+~(
zBKKzpZ58<t*?DZEy<5j2O2793-Hr1|3DXh>UN~9o1%7D`?}GA;6tPpJ1G$iYcXHV7
z{~^Duy$>0IQkmhbW?$AAH{HZ}Unt3c?4+>&+9tN}yaVYjfT$(Z=9mfGqQNlYUcmdt
z%|zCJHr>E{w*eG>gICW4HVfdddq$$)(+1d6EVsuitT}aoJ17>d$MwFz!b!5fFqYAs
z6%R`k&hZBjQH&`O%js0N-LuVaxzFqA?Zq_9C#v)7TYGe>>%bN-@>sUD8?yLmz^vo@
zD^WoFEdFRKEAEMe*{?evQ9z^6tGg4-58`R`rmhk6xE(>O8UnO(C<iusu23vu;YnXe
z<*km}>x@{jDXSWy0MKEsGULO*qBuQQk^D_<GSo3<Bj?*HeX84beMcDEjA#@gvmC4w
zIgLUO8idj7a}pL%C-Zk>0_-}GDsx}$s2QKFtO=981>Ld78N*}ucKAi0Z-4@a9pRZ2
zoz}?hA)O%B#dgwVH<>fC%qY0U+jp)*U2G8cy45ZFv2rp2k6X^Rh74lekzf|ot1)vW
zO|fR0MR}A8qsC_zM5^NxHS$YQUKTTc@n7N<t1EU)AvHCf&Ptf8%px#ce>~OcXRuEp
zT}W-pf;3JWZD}9SKJ=&Dw=j9whNWJm`Sc=uWV0}y5N0OsdlLMab44k#uS*Q%wqO}G
z8{|_@<D72!hu*ABhW2SEzcvjzisN=m5J%6@HC0BHK5-<8K9ZWv_|*x1NYJ_lbVt>b
z`&3y0yl8SheD$Pq+w&VtHfC=e_Nll?eDT(N+XV^y7ACa3vF*=qWR$wbilU9~h2n>&
zv1?giY7-54CR0r~@ksgkVJB#QnZl$GO=YL|@w5@q?ZnwVRd5%ITE!6pcLgC`6a%h!
zLuvQQ2Qz&Ka6+2^hC~H=6Pab(d>mP=wzfe2Cdo)sHB7dlQ9APdjptS1IonWj9OY*8
zxXn!sOBDHns{7k8u#=lqPW!H$E4CQ{x<ou`ZBGaRozY6f>1T2vE!9Td9;~|wrGfD(
zF0y?3eN;B7d!|vfkyA5?wpp`MIg6J09v>^ZPa%i3E8%MMQ|ICCdo#4nyPdLQ(7DH@
zS+6Sk2C~>l#Y(e?-bPP~TZFB0V&HKwYCwp^z1ziKMC5-845MAZb7SwJ+JcXA7NPs?
zG}bFoRcb#C5uPT~D>@j>?Y2jqaJ}v8Dz3S&8$6gVW0s9Lx(3zUC_<<cVPp64J>1;w
zkDrV}Nfy&S=DS%=%$#&&(bodLm12nMo$5mbP16U+79QgMmg&W{Gm>KvDC1E8^PdMC
z4Q-II?jyuSV(WXQz)M!)C=b4Jg{B*G^Fc`@oQjDD&_?@FS2f6!2tq2&a%4X;AL$8&
zxgIrpr_B32+tf<g0iZ{A-_54IZ=R*xL~w2)c$#eM%Kj3JrDUKCk1goLS`NC}FWkhh
zmYi!pB1wbUz6g3Gw+bdbz*1V?i_Sl2hm;Bh8u`5&bULfxIz6=m#|B@Gg$zDj$d00_
zgK&UP11Yw+wMzAduJHsJ$NXO)UW8bLmWP|Q59<Lk;8h4KZ1Ba@=+QaLV8FE+_ZFhm
zi~Zs`8;BiHzl3US(76UHP<v4`6JH)L#!&buJc7+r`w3CT2SLuD$<tR^pvM{wKA$tM
zAN<yw0!ttE)n%9v-qHLsn{n{2x$Tl7tvJq+ek4r=>xn7)7#Lha9tu}wB!cQM;n=w>
z!yENBJa50Lk^&K^I7mTuz6b8kN^EJCNk1ccqsBC?m-_dq*Yp4$mm$u=^mM#6d3%3o
z(!}>tqwdOAf3g!n0Z(#7Qq1}~-5&b$MuTeJp#4`4bfkMG`;fnlLrdhWf6V(ctL{lh
z+q38^F@`;S($8_IDqV|jY+`r8Y)3+<MFj<*2J@LK%8)GE&hI!suzG>Ea$l;WfAU2|
zZh^75EA@cXU8M<RX*$9u_w3lo%Z){?qr9|guR{=|6}M*0gHXpxI=bDsXhwUcA6z1|
z#_&2n%jK8vPgc|t_9&p|4XkP8$b|3|!FO;=aB+;IO+I~`#%G?Cxg$K8nd3uOB8(5_
zFnwY4HzUGy_r#5LR!k(iILgsVN6wE6j%>BS_tE#WaW`_VFjp#g)sokkM*gA1*4V5p
zn7>PU;#tr?roEPPIYy2{k$C$-If!8AsGc}%omu891%54|Nns;|1Xnx5baYx>Laq>H
z_iPO(qNW4p>+gz~rIQEpv-g#%*Ka+76OIb{6n-<wwozeNArGg;3Zty9=+~OaQDThg
zi3|tgLEX)0TONcT5h*C1vadKU3#ubzn+j>n-g*f@j*MS*d<bW6s{g8xIOJ{0I{xdl
z(LVJ{y~MNVKoXY%lnLewni%)GHfFQ@j^Z8S$-JYN6r`g4^zCao@8*aMSej-!{_sVa
zkZxtLfLt^}=istKOTByL2M|?*@Xf!^F|n?yDkdN~V&I0NE-W5S(oie<nd+{6M_T<}
zIRMQde-3GIi^`Gqppj=xeufp;4V~4lzP!&s&l>+57N<hFsimWq9dXHuYTHD_Qm62d
zT!?+_It6p^$t4IuFIzvieYH+Gwu|>6|D^m4|Mi)7X%5Fm5O!(Z?396&NWPB3Ll>Ix
zrlNDteInYDZcD)8@!ij4|GtJEC4(0EA_~w9@%!YAH^nKyh2-g;H9`1PbuQtuZ2B{s
zE*Z~QSAs2xu-}Df74->TVJYRPZK;j(TBmLXh$*J=<@cw{HM@^EA4jTUgwVy3VHVy*
zI3>(IxT!&URKU73-V@U3n8(l-6}p6vL$|?)a}yqTvCp}8pW1^f$OiCuEepD2A>ccR
zo}Xgt+)%ZFH=yRdm?}FxV{oGAi2V28lcVnv&*QUuPVK{3e$x1-WzU$8*Yr6=m3)6~
znzNfGxbT=r4qVx|+I?%85Qx;&$odk7c9gvo#Ry=62t%XA0RskG>_M+2cuW9`X`9|*
z*lL@-jVXm^WbyB~{HfFMOg{>O$BszWeXX}WJ&PFd<%>udU`$H>gK2-5z8ZOh8GECW
zbPetmL?=(q>T{wh17Lr~=djNzy?-*n@7pN!UX(_w5RIXSOsHTyarw36&FTv3(JqR?
z5-Agpa?00cYez~n=Yq<}K4BEe+<sp7qeCPH=@h^|7TWFgxT2rNr?0`U`}P+JE<cvn
z)7-@9h&Olhp~cYc2Mn)@C#ap(#mABs<!By~?)EcZ?t%R?U*G4<Bh%xLZ@6>P!SknH
zd<)T<dqs&^rm;eZV%2@vde}exMO5*c(ve{L6spe-obC0<86LFOAr!D4m_7AJXfky6
z9iBV!seplFQXNMtF<8t~3oAz!M}&OsIrWC0T~e&Se)}Bc;mXrZ<8UYj$+jb_+UWcx
z_D7^^>hFVM>l^H`s8^?&@cS6X26;95X4Rj&cncyA(mCejmQE@g%Iim+Ir|5BabXOi
zBpy^zR8m~0ps3C#S!@DYKB4p%A|152GwJaV8n?L?qG85!2zKM^c(o(;9umM`EP|@z
zmE)<7)_vvU=LfPWUN-XWQ+ui5V5k>&;Q9#3vg7Js{g7?GjpaSqpc$Ea<IWI2gN8IY
zz#-k^MBy2iSLd>rmI&Iwud~Kmm*!E^4V`SgdG)HK;Y{?zO`x3}h6<J@dB$$cEEK4j
zVmau{r6t0jy&?E&BUD(4lUwsy;e<B3Y6<`=Ops|z@+B7{6qaB?k9SFGj)`J1eVDd9
zS+}7D-Y_N$XZ!;jiM^ZTtbhh)>xH47M|Y~*Ojvx^Ai+s*f)rRT*l)TO$<xvT%8j_$
zcX2u<KtNtBZQrl%lkC%Ar`V^h>eVsG-s_XJI(&pNH4saAK$Z9H`l1(JdV%Zv?S}DZ
z2)w3ohpH+0Y{_GdF()(5sU;^ktICybra4brk}^VqM2VQ~=N<IP&p*;o8zurb$6FUG
zl7VrAZXO`|KK4aOXa~U-J_ua{SiAA=tFxc^mJ62aCG-jcqi}R5ONr(ix}OdU+bGyC
zL}Oo@uA2b>wu|3`m|Col;K>g!>XswGB3u`2VGGpu)O>BHM=o`ys9Q@AbAQT<uZipX
zsQPCx;O1Qz+~(6l8rvG})gy-opR>|AGGL)zarZ)30LH}cLa5rK2d|}5pmJU(HE7Ad
zW;SEjcKs!hjn0Xi%5g_Tr8)JgYliiK<_^35n+1uh<tXhfdTfAA7LBdby1qXCqDA7v
z`=%!k$vO0IQKVO~$6HYfn5xIuS&%M1-gJ|N)zIQ-QC4!^TljA60=$q!P;2`fRBfAJ
zt3$3uoJz^+t;%l_(dqT!O+%afiFBJ}#RYLWut)ORj}~g_HOs^*0izxo5}*RP2szKG
z5u}=A`4G6tHLEl&R*>G{)Jl1I<r&=CeX{F$%ld%BKo$#|HcvZ~#?t>S^n1R2N%bz+
z+1@S1&i>6W7HPiats&yHm-!QN$$K%Uwo_=r84`~~_YPzbybsSqcZxVl&pM%~@Xowt
zE6x}fIm{*du$TXFzLVcFi3w3VOzDFdS~io<o@IJF?`z~U^wQ%WdW*%8G8C#+=%_Wc
z7}EJwW*`YY<WQpGNMO!8r*mGJ%O349qTZgi=zeTJ#=1?Zl;#KH^7GMRV}F{HE-GD;
z3#PQJ9O<H$S%p(yKL7F&u5U|rQHIv;3|r3k|0604y+ux{#Z07-rsAw-<}sJTwlh7b
zs2TjRz!D)&&k(d2rPlXHn^-ev%|%(3lx;;^$^Nzvld|<+t&*x{7T>#w^r_$kpc-Jr
z#g4*<8MIhe0sasAlonk`ScFiWapo)f5~$Yn;zf?l?LehjvBYG;bcr~@y{z$*36lEl
zlKB_jN{ZU;6pVZex_rAIq#lR)7v`f`k8jNLah#Apj;>Oow?FX`D08fd?HB*tZ;v^(
zlMD{2@77DRBsLMP!y6_X^o?B%qXe()nD==e&R(45tdHycfgvr?D&~9cEPANM{r-=)
z=<UK&6$o`&9>vMmeJ7QY*r-~W2zReS!4550$Etl7Z828ZQ~QwPLzL@vE_ae9UXyc!
zIMSi*m#VUA<crrdCtAKG0D9+kXkI1M@2doosl9n*_gC7XpjltNw^K!h8et%plQd~9
z+wi#>e3SJf?<SV2zg2MAC;kcaE3LeIM*Xco<q?}o7WLxDvpEeH@Vrh>{cWzx@P|JG
zy>q7RB+5C6>FMMRX0|zG<&+6ny*G*H>Bu98GPE6{z+?kXwwV_yg&sw#UGMH^8}K+p
z_KpzO3V(yl>T=n`EvQShalV*ml?8_Z5nKEPwrttOco8oV?75C1O5x9NZF#&eU=$-+
zEOOIq>y|uNHp8S8+5FJy98DOPje8P3U-Z6PE!@D>knwH}kEv>lx(ZE;GCO9RBJ_3S
zp@Y<htmDDy(;VL|tvV9svVmX6MB^!go;Z`b+`!2j9zEaPbct^ON+~A}#jpsPzk(}s
zR>Br&Mf%)34KBVoM=UsbT^C0UD)kojF$quBd?>_sUi%Bo6ZFxdj&%f#utvNodNiWo
zv6Mx%-=k@@acdy~jU039n48!}Tav;p$fpmL2wU7$(Odn2w3mixdFB@ORL&M_H-<@h
z;6#_{3b3X8@t#Ze2|=_4W~uhZQ|hDp9S3U!4VVz}#rEyu-z^etr}qR3{!lf%Z(7l@
zANmH3!{5eyv#ES1sZcC1ax1paTMpHqYvTGivacN$X+q~Ye--A?f$$x4x-WbyU(e)c
zzoZ(YbD!E!emCd0V-C{18x(|%0R*%C1-<vrQQDqw$y4Xkhh0Bs3Gryl#y_Am#T0ik
zyST01&ddFWv~Q16fJo->V(EoZ-{YRQi=t(3GGV6CP?8ZD+tzOP6`A$oQ*^5JuG~Qj
zveC{W)O#Po?z-t4-H6B2)JZ<Kr4JDi8qbRmf2oT2%6E48I|3KKtY-KqBA?q*1RIFm
z_XRL~UczO5Ewgcs3G~Bsg#?Iq1Fdm`yh6#k)&0jQnnB&DI=hd#9W5I1OzBr*BpmuP
zWwlp<flT|er6H$YD>O^}4=urLj%;4mwh5hUhfj&0#Oby6D3C7_cvw<wJ)~`ATQfz4
zUFrcEe2Wps>&{(U=?J0B+1z)5j^TdMZJ-hr$kh6U-`9bwWxw<V<6FUnbB>8W&LYkS
zcy)A`>Jz~X`scgMr2$I_jM(BVuFayGT(}z|KI|$@HE`bT^uS?&piXm@F*U`cPUU7v
z=|rutpT&DuY1K$U)2aZ)2>%~*pJ@%}2nH&|`?r<d59QM;HNWuk9LP8^F8H;RD#JY9
z;Y{1QlmEoPgXa7!0%{-~kj7Fcsg&`))U_w>CMe}5<@CQ_E|tv`IXut-;0uUza6iqn
zXjb-x4Y*esd$K**Bv<+{j!_DhNX7HA>v#5q3GLPBXi0N8fV(C-b)Cw1PYd`a-$Qe$
zLLoR}FT&>hY5|M{h8MFiLhWA3MKTYhzj}tX()5PyNTH4ESqKiq^Ur{@VaODIvh*W(
zq=DvkmhbRwdjTo=9-)i(yJu|m(A5p@7~0!iDR%JdW%sQ!JA?EpZercJ`b}zacvFW_
z{6Xg~b}-#UPsd7yL~i5{Z(MHykF92}U;54}F_h|0${F|Bl=0-7UV>&r`ckfQ)m~P`
zNuAU`H+BpsC&xhZ_!Etjj$7?8oV@mi5%bW!oHc~;p9<UtJf<&*t|5a3^+nh8ZeP&C
zY}KuuOx5%s^QSlJ5?of<eZ_q)Tn>+D^K#!X-xlOizSHmn0V0!*(PNG|moWP?$x1{?
zAtjDuG&fotXW>2ApvdKGQht$a;fp%d<?qeZs<SGMAAoLEDwWK_BAo9Xyq$jqI7~gW
zD0C{)GodyDW;8cPf2Qg0ORtfLErwFsj_p71l$XEi>mBEEOE$K5f}x!-R$i8$3Ox6P
zN6;b($z`36JSic(wH;yiyqZfdA6$=d<J6*aV<z9+;TlU{TYpW=&Xr$wLouz--@BwE
z_w-by>w8<iyuFRw@OfNLhU(Dn5%<$2(|6sw1rwdrqhE;&jZC&o;f1CYc=G)(f-9vK
z&vpor^4FGw;#C=SZ>7=K7j$NN^eS5-4(EqY7tauqSpUB=0i9#!Y6g<xjAZY0b*l=7
znMOL<%*xrwJvh3gT`JOrbBf-)s(QV9rdVXc!12^UNT7Q%?)l47l^3qLZ!IWw2ifyG
z2aYv!*6;VId$$(f34Qv|sMOX(>1Su$pJ$?t4_>u84Ffvcvfulww6z+7Ws{Y0{vpoF
z{zWUy+0Rsj8gg$|?5&sM?YY{ow1LtIzV}g^IWhB}yxM9DhG@xo2dX8A1m4J~g?Y8L
z{h1#*_e!DWeeo^0aBc9KgED184M%LAw%nCss`j%xS4DV_ucJ<l9R9h!_!xvun;!Q4
zBK=Q)J4>1$;;se3Q=f71o)uypWcr<%bp`YezXzm$T54e&_Tkr#qaVo)H!e#>6hC*N
zb?wxZiT;#g6Sr|}8Cl!LToE<Un}6uY*<44RG~o18j^(YN`seurt=-^dMu%`Gsfw}&
zAJ<@^imxidh6gs<u4;Vms8H|SUY>2MZw63EO{k3AIi=gS+Eps-%(Gruz;=B0{(KC%
zBhB-o0pHF6Shk)C#_XXr(ZM4EVMHh*w|8dRS5B_~n-?IS6B9E1dVe<it`q+6-22|3
zb7BiV#IIj!AXe8eu3v9WuzB3y(h)87p{##aS-O-V;+r;ftR*$k*SDP;8t>M^v6*MF
zqu^`<6LMm+ncRg6|K_kW!t;mK_idXXZv};e_b1=+nQ#H{nNmGFJmN<ATWb-fx{xE=
zx;m*z(xzI>byyDTJus}{plf&PL0v+yluMN__0C{pSnSp#!=6cweX`Qd-}<jk(KCh%
zwd-w9=a}tX%2h*ezs4ZyL1YA4r+a+&FN*XzOV_v0eBGn3Yn_|s)OySIEkyNeHX^^#
zoB2~nc6h$3cCqVY&8zuc`|z56F&&r^AK;(}_-)q=TWBpxc}@cp@BU0C#1@i1He8@u
z_`3jaJ9t5D(aj+W>EquGRc>!y3w#zTU=nhGq770L7xQw7endJs`kde5;r_}CI!mOA
z`AEVIX;`wEUNp<r`=XFvM>SauEAkEUbBPL<0Yj_SY6W4s-F*WJ1f3Ds3z@V_{M9bz
zk4}+&w=lj=)mJs@$%qv2Q`YW~=svS@HWwtFN62_(7&qH}N6Rm|<2d&-SvOjz_C}Tl
zaT&!#mqn(Y+{wlBK+{vA`q<n;Ji8^CO6ojT3TbTDd1#k~WU8D$2s;*Rqmo?PvSInk
zQ)^;4V@|Y_QlAE&KJ_0$`Obf=yXq~X)q70*3D#w@>;Glm|E-iGefArZm+XDbMIlY$
zc~(=pIr%?2N&V}S+l2pwSahk<*QfP*ygXxd^xq`*y^?S`<J=V2=-{Z}lhR-LCa;n!
z5iG^@z$9hOD}G9WZGoR`Qr|D}O035>heCx#=~R36+ay7W3(tM{7uM&u)O4x|9>WXi
zU$^Y^vVI0aLwF9H(zBFA_#`~Uu(z<coAL<Px!h9D(r@rai<F!bwOs&5soPxpHR?q8
zJ1uK>R8M$+=T|3LdRF58%YkG4V8Phvl~vY4e7H_iY+N&Me;oOor{GZ<1r$b<t9C8v
z{72IH-%fI67+&(oE<3H`4tuHnR5vGsxuQR1s-IxVBiX9_Zk%LfrWaEUvthr^=c&&R
zt-8ZcO9;0=gYIx(Z(@vd!%d3+bE2QDxpSt<t)<Aykcq90nzLF&DMu87Lij5!e~YiN
z7omO@j>Zkja_fP&rA>X!YyW&J+s)m=E>RW}eZ4(vN?Gg=WE(F-W^RcdMBPI6pOFic
z->u;N<!}1VG{dh<veMA;?u{_^JFyRo{urctk3op!Tgg1*W=$O)2uO;KK6JXt6408}
z(h#1*vA=B_3UU*r)BhMv@v;l8*~WYlZD0fb@Dx3U@j(bZJNqJ~?q&0%yr(mg%Uz3=
zh-{OdvC5y_D1#zYMnu*8Kz@i0MS{~)QaIK6-pYhi5V2<3E~S|LNp>w-M>*@|%l>@P
zpPX#{pj+AvQQt%LlxmTr+HPt6&V|gkz?9Sp0cwxAk1qyd2iaS=?299}V~4AH7UiZ7
z+p1E2%8KmX_;Rahnfh#?nRRevgZPt$xlLYwLUk19!iYHc#}2h@&JDg_W62W-ecRq#
zM=d7j5A2U5^<EFZB%S>LQq@fK;FWx8;1tX_?kADEZ8vo)Df_*5{UJj}F7RT>z`!v5
zgw{{Wh>F(l!aJC;EX|haK{t7m$+`T{?l*C$M^1#F7s55O{fIxzqkvt`C@ufZg5ks-
z@3y|-#3<b}Ic>!ay)s_+pjG9upaQZQsHmpk^Ksv8Oz<qh(g(e-Q>O1)fVOcvzim|Z
zD_RJeGAk%oT<dI1U|Mg1f+zT4gRvT1kizo#tvr5l5{LEsw2l)*CoAK9k{hKy#jD3n
z8DimWWYc%jRK>p8P+fs0@9DN1bH>pz&5qv<2uzKq99XG`h5RkB^>h!xVP)^AaV?B(
z#prtejLN6?8Zd48A@=(hd5!MzM)!_M%9#tY#jF$X?}mrwM^qW5PRG*BMT*0-_$*ks
zk7Lw}2-iQ#jC}e7sFJ*&AldN&glWEnX|E4eC)qV(&PlWnl6Xf`(ifzD;Wp$|$R*Xn
zqMNt7M|Kpu7OU+?^f~PPPx;%HxQ;4xw|5>7bp8qOU6_nvZB75V73o4_9jNe>s>CJI
z=E(oD*?YwaC)cM|dSBdulyCt#<E#8gQw&89kIC{2WgGsh8m(~n6~%F7f$m4D5iKV<
z4u82zjjkbDHr`;eFonPDvG3>3YgW>!K2$LDuQz|06twwydM?y8#`lAefMo8?U3nkg
z8dj_B!#L5^?QST!^T6*DnM88>56e*^)Lq_p$F661#l_#g>#WuoPjIl*83uO~x-`kB
zSkYU~#gRQp=h^>7q6hs{i~a$nC`_k*(Opc?8kc)-gR^Ba_f~#eKxXB;(JmS_KRz#5
zlL(=+L&K4>Fq|<!J~|&O9`cc0{g`B=vx6&V$4K~7#~K6v%WdM4*qgLL1i6`$$UD+I
z9^kBdyx)=v1i#WT1v8a4sw7308XYQ2zGHN{rJ{jwjlT9f3Z{;}n>O{Dy;M#`J;3!Q
zOI-DO?Qp~}%egsS=av$E*YDmGz%*f>bkb>^JdZD*gH_tCy*dn9trX>T+i?DSZ{GLF
zz>29gJMQFZU&cLg4#Rqhi1{uT$O=0TqqLm*gB4X$ZWRJO%D0(Ek#jEaNKG)SO!b3Z
z+X9|lN0N|xLFXh=YGnfR2dnvo?9n^#Kxk(GbEQ<vkB}%LXirnPE{3#&LYs%ii^wiz
zX1=s%(rT9}I(n9T_1T@Uu%I=v3*lnh*3Rvo_(ihIVKG#+>bT2pHJEMB@Mkw2y}Clo
ziyBGfayX9amY8!@>?-l^sS`;WlFW=8AN?!x4I5BCX^N3l>^gN_CI7&hXIIMoqey2S
zAMhaWkYj|iy|in5K2Bl6il~7;WACs*0Vkbe>6~gecKtNn`;Jk&gZI4CU>vCm`-jVT
zC!3QDw~#JSU``ds=;IV8KmF>8*mncs`K6}y3W_mR_LeCV3N$MWBE68}G?`GHl4+OB
zE5y@;gB9wrjWVf^Ue^~2<SFAD1HbeeSQXM3U}9Sd-G}mv21Z8{lL+MEdYgKc!jGg|
zgPfK0qnvXArx{|<MGh8d`cC0`$a5VdpltoVQSfcODLw116(SGE;4MVvRzycv-o_&q
z5MVP{dGzP}0h>pEj*jM+Nmj>`NkUe^FY_K2N8+mU;{y&HrL<6T;}a=Kh1XnbWb)Dt
zFQe>}<i*=_!t90SB<S6(Tn#_Zo6Kd(tESqd;%_*zCI+w7IeNWVcz@wu9N{wY29CBV
zABq`RVDAv-CZQhn;g<#uj<vFQ?Q3#bu0p#>MrKF+&W)VI9160+YnINNe(Ur$3da`q
zl=frmVe1~P-^FVPNoa0$O<jJ(HfQntp$N2&*a=&QwJ-3XManB!3%xl7%S=x2x3b?B
zaQBy~Oc|t{NWIceCn=jK|0i&zB))g!9;$I=)XU2%pt0+q?hIeXJb2e~?Zz&%N_wHa
zO0ry(2nP{SlsA8qGE(9C%QE$9yOKfC=ney8zjImRntBB8ahaTKSD2slps`!l(6bwk
zug@p-z{oN<=nbo^X|T?cLGcoCc;YsrDAy-2Yg@9LUCaZuw<E@Z3hXVOk&8t6CSek{
z4|I?IgbBA_J-ysrq@7a}({@tUcB)Gn92Di{n=&x8O^OzXE7q5i$-13Ea(o6IRh#YQ
zzDgF4<6Ie!=dhU@w;h0O%%yI#g|DtWxhp1Yd{K!GTV3Ihm7=Ig1FZqEUDIC8Tgm0U
z<J83Wzo(!fmo-U?ltMG?$tnla@P?2w>7nbDR6UUJTNJy?+wysP_~xa{`Z@J{@?Q5N
z>6wMZ=ZU1x4(V@K#tPCIP;L40yEq7Gv5N{FKQQ5O<3_Vlh%-#JrtU12lP+@P|M-(G
z=%w=Gll|9)GUdkmH8pd^xxOmZ8A{Wv&Uq_-T<5F<g$o8_$-k7z+5(OcG8w{Mw`NdJ
z{6zPh@QB3YDocawBGtuiWBR7#G#oQ$btjvZ*qbkfRYaGy$JD_>+0R+_IH*%!{yJ!X
z86I@$OS;#RRH#Hcs+1|0nrc#Kw7Ay&K{w~UL?^rAkpNruW?Sg?PGr;5<ZAE4<CAy;
zh0CT$)gBF!(3s#$y2oU6ZW^psR6MC<$A9@h%L*JR^oVw|BiTIhN&t9W4r_u*%G_-A
zfqzF|$>4T@$gjhDf}vpv>4E{&xLH!Q(oJPpX!c*lJRol1OVdvlLQah$?f`K3O)uun
zyJ_J8A0NM)mcQtrz#YsGG@*ozxxBr4y44`gHg2O$zMPAJaj!`~CTTs(0yX9^6<aDW
zz7Rwb(!8kNcK<PYO=aurNm`&s9xk;VL_Aro_3=7}Ki`NHbg6)qo&n(kB=CN3Ia0fW
z2`@#;=TFa>Lr)Ij{++p*<&%x^i<_A;(;EyL#GC|jy&fw_nXHQ>W`t4H9^&s9p4xf-
zVw^{X3P8l(kjHgbIz6IDkbmxq2N1#W+uMLFl|p-hRyrYICXBWXnVJ7G5QO#w>6qYv
zn(Ba&wwm)8fLZiIJz5=5u%>pMLl<r+yfnpDu_aT~ow&YuY)nxRe8JAD7kry!ibA+&
zU0Cr-P9jm}vFGMWPmrGRozH}Y4tR50w+~Tz#@x27PKO#6rA9jbV#A;relM}n=NHVK
zp}X>}nMYcQ0KDB*oo^l?&N1|TZbju4_Q|VvR5@{bS4tNCexfHT%f&}pG)_0@UsgAK
ze4nKUB}0)`c1CYW{);OE)(?aYtJK=3MeB-h9pw$InWF9CPUyy3C$Uj<sA8+-7itaS
z$%uh#GR|n~!noZI;N#~_Jp+SK@zX0XsIXvem*jOmkCa}X70@p9*WzOP!$sxKeZQ1=
zqp3&Z#zyy&{_A7+`FnFSoyu2#=Ylujr&F{ED2_mxZ9QVn=z&$>Qh4ik-_soFM^pdC
z(K_V3SZV)nM1CF1dBp1@0xc6WC>|^?u$}O5V8|vqnW%I+C+b;I4EbcxW2ArqRGeGW
z`2_r{H&{LTmz=F~%#~;!6O}R;z?aQPKbPCIPr;AD=%K}3?)$c{u8bN!ULjPP6L?0e
z|5%7k%>{`%yc*ksQa6U!`|T_E(a%tEt}}|KPds_$#NHe-0draYQVY0)ZeLi3WP`SJ
zJkspLKuRg$BPTI_u3|mS>(wM_6Hd9`cy4jTPevWzA)TkFG<Ff)0G36IDU|7AGJ5UP
zpb?+M41lQ+z3Z=-fJ$7nM@h_YbN4X}s!$?wcl~I%+%H*D?`0IG{<^;$tKd|+qK#bX
zm32KZbg`-$;cxD*7yH*dYn?r=|5c$;Nodic{U7Up|1Ik5*9HdEP3vWH`lR(1Y!{3E
zi+HV};@~{ZxfcJ|34iNU{_?`1B7>0y80{0#>2`s^zQkNMFzi|eK+%&&sc=k(<_@q8
zqJME4fx{CBT>wM`b<i#!{KHli+9IVm58wgL-_idqWU4j*<c_A+@T0Lt>(|vlt-AaI
zG*7^uME+oF*UN9f8)6eahkx6|PZ&Vx0!>RYnr`4^%R8UOlVFM?0%Y8E{tfkBt_~0U
zRip<H&;M9VZ-j_)3|nBoF1t?yJlAsLx)1~U+M9sx{3BA>1LP8^0TP>Z5GhaPZ2w0W
z{2!0}f&&)pe?9pA$e|U`#U^tkSSBEYT8FI9gZ~B+`k#Tas{uCqZ>T%_;A?i=J<Mr@
zNaq!vY5Gcqp^2$UCSj1|-)u$yFV%<Of34$<*3yuT&<o@&>@<x17gUuplo~r|VEAtW
zFEmc9fpK+9NdM;>xg9`I3S$K6L;>$K;!m{yd)xmOgnZASS-YT1ldJvrm;5hAAd-Yp
zk+Ni-kt+6mD(4%yCRK<5ul#H7Nd?iQRBd{;Bw<bTF6Qf4$8Ei%t50l-{Y;9KI+yZ2
zo&i8ap=$|hCBB_Zr-B{E`6F){88xSOK<%Xia6s}(XKDm&-zZR2&qrx~4GSX~=MgXh
zpV;@?P|GX*V+Fp!a0d1_UbM2DzqP5nz(DdFA^?TEg|vo=a_*q%HxA7m5DV7#00KXe
zx~R36;BYRW+YqUNQ#UkDB5coSgAGmKFFXTY0{GeF+PH2jv4&5XVNzk>diFKgM(U)6
z9lD(J%l*}wj}4;|r4lmcAK$BLo+#t%qqkmh%VhxU{!cvFiJ~Z!h`7))gGRc<CSpbq
zRZu|zAfz@)F=2CE^`_g9W$2T_e-E^F<-GN$X!UZ2Cp|T!vsFjbRrLHhty~Pi_?rs@
ziGRJ&`4Z1JjHq^xA##w{yejFuHVRIsX1vO2PBs+&!cOByJ+l4aS*YBbaKDo4for+c
zpg9GP7yl?en)8<D+)m&gk(s;h$athXg>H$M<xxqGwurytRg?ig__xFw(nu?eqk9yd
z8v9I`_yw<*pxIl0CaeUe0TYz}lJ*~Kx;XyF?f<oor^JxWm0z>*s3IuJNiM&C9liVS
z`LWJFR`|zzz!n3^LK-vA@@njY0rL&e-SEHe@t|!{QlLpmp#^E{ga2{H{$px4Ah`B_
zIszLC)CgLL*rzEzK*Azge#>(GBl(SG9`S!kQ<p%S7^gwux0|dg&nLmo>CXuQpK3bs
z0j=VL0de&I=n!*2%TXy;0K6V2fYrp9VV>DY{>4?i$}RslMOPrE@xQ48za}|HVAs?_
z2ORyf7ppl0>=0ElVSU%ge-nSN@IU*C9gxreQKoqV%YY8pP=fzW=d;>>Z~u45|NZp;
zOZR_g{EyK8-!rZM!`@p)MHznm!V{#_01_gdf~2Ihgff7jfP^$iry^a0pn!lNNJ$Bz
z2!f<E%!mlmT~Y%`#|+&w?>+vX=d5$iyWWrQTIa)A3w15^zOH@ky??u|*!Ko0LZ}0%
znZdOXyjVqZ)WqkA=m`3rlV(rg5_+8`I}%oZO%CD?X(c;I-NmIEtUtYR$V>^cZo_@%
zL5ZLmMif<ImmXj+FQA@8JK#5vX#_HBLAuPFl@%+BUj3m+15ts9X01qepEf;SkbSr#
z@#R>8UhQhp+>vOvt_;WHC03@Q9CsC|yi5JP5NnLIYD2jBS5}HLDDb9G&9Ur8DC?hD
zoEyg4`uq?%*|hI{yF`9rX}-I)g4Z@hi94b~h`XMM=#CNWs-_H>!TUOs#VSMRwO_l3
zk5(VD+KvuAg<gGoA52S2Omyb~Bi|NTl!NLM@`$LUq^uyk+}O;<Mr&yB*Kgl;Fsp%Q
z4R!jDeDNDI)0baMd^WQ0ltwovrt>D)U-71Z(YK=TY;si0U}MOB=M>JjtJb^*n=8N%
z^dfVb;_9l325w%j@(y4~KG$K{%Qgzx-xo*vlATK>H}A-jTbD>+hvhe!9?}Hj#WYZ6
zUGjBMYp@>J81xm9K05W8@({+?W8ZeCFC2zz)#r2IDBfo;+cLSPOkXyBXU>=3#XWGf
zsr2q`j_zGHlYtL^!`6@&)_;#f=Q=@UJLYB^ZA+kHR>fnQy>HRBzwx=}tLaFRO7=Zv
ztK`ur8akkB;=sEBoo>8*hu=s<D1xuX4~Ff2av4MlcB5^z@#G8vKiWzu#gFmNe`MPp
zRln(rB~>fXsXOyhJGs>ojeTkQ&`9a;C9;C)aLGCy`wI_RE{)b!ETRokGwS7RXNP7L
zb(>F>SB3CDE{jy$)HSdNsI4JTyU&&`<J`b;b`#HZ8Jf2$q5AdQ_;VvxPF09>;4-p#
z(>7@zxr#eoDZf(|tq8DMdYz;#-CqgG8-l75aYs5xy)|L)V<qY(Vuyy^zl=m23oIXc
zg3pw=V)Z;0L;no9D%BW}n)CMH8~UrtADl0n{-YDS{3Ni&{I{NVT3MFRO&z5RVZp3N
zPM7JtJ<4x|+#HbIdzsGPsrO1H;R(x`nBk6|(+9bj`qy`QF9)qpk!zT9#YFJq*}7X}
zFFU~ow*z)K{+iK5X5*EP2UNnTT$TO?HWxnCy&(@eOgNA5O03+9{p)muU&e#V;p|1V
zSSuCInxj^?_~mGR6|d6jt>CD%w(Z9ONaLEvG-acA9FQAj$B)pUwv2$O!(dB-LMUcl
zSXB2{G4ZOH6n1wT(1<}9B^$$Cj`>tpKX*U=Um3g7RI~d``Q56*H#?Y*%7F<^?Ce~5
zz#d_IT&C;0@DyQ0oqJ9iIR2A!HGJT5f51BOF-;)57<^;!aTM@SW``N3MQVZq-PaYg
zPu>FFIE0u|&&YHfDF{rn%JiQ<H<vsVw_>x<kvnk#PNS>W;h2Q#&MoiN6Vutyc?VR@
zc4Z#+7|?9eb_ABLK8li?&m*w&tc?0Ym{QGSVA{;Wqvt&HkEBxe^Ho%)XS)L(1}-1y
z;xL^-h#gjFO{14X2}L5b@Y(*1%GzBeN*&zs4ANM3Z{owadM?lzv9l29+im?g5qsy)
zl>kz#c!d{}l|LKM63?299@Se~-Ot+=KX*K4M)I-j&2@jS+}-C*?0)>LWWAg<?~xQ4
zPz!iH754RL<D~&j;DwmtU1s$i{<e(puh%>@aokw1V&5Jkn#4Di+q&p`QM;S{+63l?
z6+@Yk?2n5hjVCZgK}jU{+2;@Y{)ViT5BRYkKIoXC`L%)%D<NIm`KP?K|CE<w;1UYx
zujfSMDsH*&a`s!a@i-=3p819h=u2*cQC)8yx#IMK+g;SawVcyX;nY~ww1mo#M|(bx
zqj#n3MQ6W8cSVJo)7ML%UT*zeU5JwC1?opKW=QJ|ThUp(bp}xgxhcm9gDvF`U*OTX
zPB7A@#-U;Buma(YZXySs64@<mPQBbnmq#3J*p|vW`(6SBx(xPyY63`KA&`D#(P&QN
z(0BR39S1B5vBM5E>L~H>n<vQ^RisWjo3>Cq1^{?-Wq1}3{AoT<z@PEB*I@wMAetXs
z&RO$KNq326BA_lT6fyX|4qj3^=YdE`siNo%$YJ8|QIXc$njLjlk}8QTj@L`tmoDkN
zpVhursFf<wdsfEYc#8D4tGe*b-0O;hzIYU`M`tu1G)Yz;1DggO<IP+KEE{qREM-Eo
zG<N6f=CQq!R^;dFr_}EUP6zYywmp}dt5(Bq@Z*7hQ1yTF4~Y{5BsKs@eEyOBl^{NN
zXa4d+`pe;9_t~W}>G0rW9(F;#;)i@f>xnEsE2vq)P<i|pf+5PheWSRSGVIAxl8=NH
zf!f%vs?<^B{)%STp4jK^vr2aPOY1MJl$P=xwfo7-_++7*hDsVJQgS>-AAC>Z&x^dV
z?yy69Hr2}@=TFp+Es@rncUYuy{x%u=C?8;)nnto^Z!~ZO<>Al2vwztC2SWFMAoK)4
zm^1oQe3<~kE&>S4O6L=OAGl6k>yyrQ1m_Xea@e1WbPX`oo}TTK;oauBd4F`<Kdrim
z?Aq@>x47?2Rep?b@ZH;gBV+HvI}pUo6(3#~Gc=t#jLW^45q@N>K`_YTH%|k%nR__2
z!#q<6q=(R-h@sD;rdjoQF19ya084j}`5(f+5Hve?{;+H%<i>=*#m(;I^L+eX_}ls8
zK#qU7{7S$jZlxg{-4LGSq!h5DQbO5)WFoM5B}9<ajsEdKt8~Yni9ipHCzd7QahSD6
z{h`SDN#2vJrsUO5fz9w6g7dZgZaF$9&obsX)BJoN(=5&6{OGujiha{OG`B@`HAmIw
z@D!<5Ad^uu5^)tXT&9tk*@o?W;nqo``KasvY?tkOB<H`s-_KW^e|ku%wiAu(!E1)i
zuZB_%ZnUNr|I^{TRotDqLvi8Lc^~}BJFes70`@&cz~2?Rk(Z53iS4YT61u9)rV<P~
z6u{q{N~b}qFy&q({|viPrAHM`);u85*w+Biz#kTOm-OrG^!RvxWo(%B#+5`j<C*l}
zX~^4#$_({FW@(;}M0J#AbeZYPrxd>azpQZ(3l-QX4O1UP!gs75zKkzGV)hkDoPFoJ
zjhNU;wU+<G&Rz$(J;2fR3F_<&H2oKyePl~s2{6wer&qadi*be2Q!_iAds<V<37PpZ
zcuRRhY$+==yZi#z|3wl@%-g5l5G+I6c7%ktM$Tc}7Xq1%kUG9i3z86)6ZwVOLAfdA
zuk)L>cMWQp<&Zyk6s)Xno{wSjVvRRHF#S&<#2H?}Q#kT}m;V=n{P(akD!-f9UvU8l
z%d^I@)C8^l+D$Ix@8i<d@Da#&z<|}hK9*c+S$-;E1NJ7jjYpA2hz(}Z1U4NGy0dNF
z&DajPZOeaXNLGhT_!JZL`#jwCmbb7sl`RtJkg0V!-6V9@z%IkLwI7hEKOUH9eDpO>
zg%IJCZ|DwB?)Ce-EG>tuVKDWTlm`E+%mGIkz{;5#_aB+B5o8{+KfyVsUj7DXT@5sw
z{y*y{7j5+Vzs~~vpGMpN2R$rz@sy=Rj*M(EK-c1H`$%5+&&~_6)X}H{Xp)g@3ZhGs
z6zByU)O{faAI3`Hb@-$!*Fe0fZ6qiIv>tlJ=u{G=hJI}03w{buF1*OV^TAn^SHfLT
zx<c|G_$NR0y(D17pa0|QWti&!5ZN#&_!^d&6XFN;gz7=`_;nBHp*&rfY5wVum<>3>
zD!aKu!zLUpZ*$w50XzI8rwp-AUUzkllyzY8=iBYKrA<Tu%qP?pfkg^J@Z?-4Tsp7I
z$v*<rFnqkHEh%|A2zs(E+$DVKI6@ZW#ycItD_G77Vog=V-?D8+B|@#aSb5`N$(L8q
zUx7WgrsP9>nVzBme~%|=D^W<C_aF><?rR^-2Twk=PA=qnLYRQ;5j7|m<ku%W&1vQr
z@HwOuis<>`ITp_IQHgBjQ+E-#hpg<}B{ec>02ah+)C<B<mO@sf>db-8)8PMRE_@hP
z!U_uc&|hUt1M=%Gt@ASo)6SVtjvl<pxy^#UQ<4(?>*DYCH^p`Iz-N3b;qc^Fi9k0h
z?v&Tmss{6?zD!n2`MvvNT>~$D$1UN$1;9&F+2WpPPv7ypoQs+K6cEq|oZ-J`qXJ(C
z@j56y!k+8vhlzr|IakazD#d6f7uwqX7sozviRS|{75w;20Ub^kL=t}}SCX{G3pZfz
z#V7MpaqJ^Up)w$?4cVIr1u}Zlv1)4D=C=F6jBk12-B|?IEZF1HgHR+Oc-a?OL49^*
z(q10jTkmA9$z4BT{54;413p|Kjt3gLhxFoy5#H1wc%NzJ#f97gC)xzH$U2YFMVCcQ
zgXZB{XOake1XV;GSm#*?u<CytXU@cb0xz+FlHXS{%jFIV;Vs)d0s>!0{6xso(_OA2
zj-g2673k{U3tTd97eUX&Zvk`te`x#v{rmsRf&T+Iux#)6KkOnv&M+blQG^g<U?Z#G
z!eUSfsAp&{v@Y5=O0Kht-<#E&q2g-5RXL_fnhfy{83oYItbo$_A2<zEI7k3&?u(V*
z-gwR-03n5@!!{0S(c#Gx9)+_=dmi8x8FT{N1Kf?=2OyR5C*(ey#c?Hk9|>R<z{e{T
zqbg-{Gg|JA)*|!fQ)2P&{eI;TW2wm<k~ZQOrIDS1$Zq>c>ZA4V#S?)Zq~#;u^p2j2
zw2@z!PRINH_(GV0f{aP1Bd9NUZA&eImx>1%c^*I@Ft|GUGAiYZv3_@>%e@l)HDkyG
zBu;X9e!#tst3s$6xC%d;S>^U+JMf5@CGU%05ZH#JL`?VUQ}Bn`&}?Aqsk;Au6gy(Q
z>e!*y?HDupTON2(bN>F?a;8YNzz1IGfeSC3>_3nX!|GY;>(R{pwg(zJ+;_$9>VjHz
z4r8r?N*BuuyTfzLYKgHLuBS(t#}~c^VGJMbIVXapf_u&$qblLLabJ|WL>D2dP3QGA
zAD@;(YT?~H=xU=pbdebA_j7;uzdpqtx!f0)H>rD@S0e7Weg^Eze@|?%KyCXxdt87_
zz@QVVLnh$p56QDT&~Wo;E1QZVTPn07X6>F7sFWlZ(8Flu-jkAb+|q74X`HRf9!jZa
zlkAcCbjIst<vPV-ux$Qdvwn0hQnDwFciDPV{OI-RXzl@1D1jG>c|-Npb!Ym8XPECM
zu4YZ)op(L;J@yymd&;#$R6YSu+7yO40j%(EBe1<NVnyTw3om^{R~Kf4T)@#wS0s-#
zBvX86<1Ok_l=<hIR;h|s_ttXKzU%DmR~b$HmERmo$+7T~Szz3K9bV92-XZ`(>MJ3|
zsxKOP_lpdX^Y@uK7&ne@=p<gA+iV=KN3jQkWMygXC38tcwUhv_ldn@g`^N8B0%DaH
z6c*@sN`_Uqi>G0N;V|;23Nlr`8^<c+W?_OA!F$2gOboom^O^WQ{PutBvq3DY-jbC=
znc_wcYiOi+nyaS9+9MOp+Z<=F`ChlaF5z&mCWV~c-BC9yK#U%)QzG$4kZi5`wTH2%
z{e`8AFN?RSNtD&rTxzq$kG4tILUjMe8cp0Q4+hmXCAvLeJC55d^|5n3HM`$2vIgjo
zL~&=fe2>0%%Z`|NfF*px2|z-46T!p5la08iC_t-{f4;gYxaxN!b>)iV5Oq(|S|VL(
zZA~K_J!uB?cKo2<{imPrLr@*4EQFz9xj%*Y*Z#()2TEKsTKsS3$EQ3oB%3DeG($Kd
z#DfcBgx~SgpS|)~<AK}yg=y9!oUe>=Z?-ci71M&YX`Zk9@)bM(?*vS-KPW0p6UjfS
zy|`!4lh`9XttdF}k<IJ~j+4X~mJ5tOZzGQBwojMel-i6O%`Ih@>S$~-3<YzZfAI&S
z>wD!B!AuaEC*7FOjjJ{{U$=q=ca;14gv0OTOyIm-Pq(S;oR=|^CFH2};)&4LD*}xF
zr9E{${{byuax18!q<`;Vdw^-oEX>69^xmzWH}|`XWCQTAQPaOh1iZyOlrL(x^_lgL
zihG0e-kD+gGM5D)!MxMfnulGOtx>;q3IXe)vo`oqK@ZMjN4<y%^qD@$YCO1fsQqYr
zXNf|xN4^0xxHVt-ND>ieSYDU>2TsaY+}n#MFhhh8vjA$GqWsX@%qj2;f;#n0lWhcf
zUmHuQ9ozRP(Vv-{a7SCxfieB<53U6V2#u+8K8a?{P_0*Lsk(^HoB*_`LHPTu{-5}k
z5ekr4pZolV+xMo#DK3!Ax9V+`qBML%mubhDLanRvSPL_gq0++-yi|~XcFVf&ABI*P
zydu(!+*p9V+>axD-$3T_f^ha|h1(rQ`jjDwgVSFZmAVGlA_R-~wqnm*+a3@x+!GJF
zm90!$CZLo#qw_eXF0ZC3QAcvK@3V1`1v)q@-k}s}U(Ma7@uCi##p{6Np>%k9FRq*q
z{<HhPtx54g6@3c4<h#$_ex1GWe6ILY^Iv5siQB)b7{^+UTfvIyW~<N80E5M}!yNT>
z^UC)B=og0NWh-tR7WTf+J_`+`-%KAeJ+FWF;9hFn&Ff%Ur`r3+^-a>JQ#<pOQ>Fu8
zYOT~z))W1XRumV!!~y5k3rtN&pbB%l-N+Dl&KY@2%n|lGxj=vFb{crIjw#G={BvX7
z9|DHpuBT$jRcyDWJsvqV(Hr#KRGO~+?Otc)m_=4E`2C`N+@Ji0A00~Xc(;wDho4rI
z8Ayv$-_@P_v+5INr~mO}sPIpj);2K91p1?gyZsm7ZUtI`0*K@q`p@Vi+=`Xonh!oH
z9YN?9v6s!rhnsI)dfGi-Y=FaO)&t(?ikQC_LcSa_9}$2B#U&J`<z@d7Rr14T#_N^H
zCWyPMo?Q;7^mx3V>b%Qz_jNO9uuEAjqq)4wJxbg{Lt~BMBxkgWA*wRt=ATmqwWXk0
z+3MNO6uAwhW~lD)2XTBBewq*n1H4G?S_07BwXd{UioaW;Uu(=?ZjO1&MrWhvIK5RE
zlCYIIRr~B&rbsYi<7qg}*Xn-l68#H%HL;G{#oJW^A0@_G{s<**)weY{r8((mj{u~A
zwlO?WlU^Tdt*WQPZdA|a`i{G-U?#s<RWlW`C~bt2@D|4ds*m@}1o9>z*4Nmp2YDOq
zg*fIM(iEPil$ATwZvTaVP^2A!NX_`88#?D$xoM@lhZ)~tLDiu?+1q*o*<?kj>jWQW
zDiODV&)Ztk*r4GW>YsmF8kBRZ-z^ZVI{e`lVUZ4mMGm8~9y|vb<S9T-=<`i1R89$P
zmKo60Hqy4)ixh}DTK<Id;GAte{sY$wR&O5L#fGAYd6=l5`@3eV$CM#S;(FWR&|?9N
zzY5l}jW^ay;DA^7D<FBe&2@ARDm5y|Qn%i@f?~n=qQ2j68^^YpDTxyT3JC<ePsn2c
zHviwB|KB-Ktom>W{3TxywN((ARLI2yP2^YkWCf&P2D=3Kxo}Q`29Y@O6Cv@BYpT5P
z8Q{?ZVPSa~l_Fq&4`~QSSV-wQ=xpSwjYRH+j)BVU>vHvLoKl0p$D`qd5sF9o8Q4%s
z8PJgHrXXEsyG!zbJu&#DV@d&Z5S5hx#FuA;-*{8sQ-NA`iP1^a)=M0-WE#n%HG-(p
zg2=%KU*lCo;2Ug&xmbYe`9a5H%YX!`Ml4Hm;Ytf#eIqd4DElwKBDXja^cMOlbHRDE
zK@4>Kydrd_OqDlAKuwWmF^6joVuvIJDi;d7lrh$6isPd4ECdx>{>uhxc`8uBm(OG^
z$z;9?Md~QkfmPXF3xKkq)6i+ih#^H@#EBoUb!!!K_X^pAWL|LBFPsTP0h(gCOarL{
z)!jLG!j(@LONi|{C`f<=H}|*EXOB|jCvPs%7IpJgQ=j-$N8TRmD~c4mhbRYF*~b0&
zWM1;c!w6y&^(aQX3ceQ*I?Rl{9<zh7MG(|p`!@;5#pahJ;wPf-@uV}$!s5?YusrZk
zw}n~?FFaNba>Ok)DeFd$&Vl7YhssXd2{yq6CU;aPm9%&pRT&?;8-CmS-d=UC2*O`g
zF*JS|kc5I&$0QRzkA$7oeo6z47B5O9sMpvk8)z66D+`{}WBw!H!RM_|0jn2?KjhFS
zTA^DZp5(9Y<3|m_@<@h88lpwu!-KK1q(^1dz;!ZL^k;Ei>t-1)eJ>r~r7$AaB&kVj
z;<x*E39KlwgCe9JEP5a!zg)VDLBdELe(nsHg3Pe7?_v-<5eWu2OKDhBq#%lK4F`eB
zL4JS-8die&ZApIdLVU<;!!E3bsGeVOkqPS6?|%H)hNL}xl8h4=Sg?-^Q5Zps(W*~M
zYB>p~$1eSe!<!9Bq6DKlUeruQ-9`8|(tQTf3b9<>j3OsT#~%&0y7;?HyC0Dd6{xWT
zWAl|sl}y8br6tgHP$8Js3)CDdepU4k18U$W7TgzO3{v6hrAewK@s?{hRERSK${$Pu
zKaVPG9e*stX6I(yH2gs-B*sq|w<x!!9tIG^Me>>oWU&6ruz33v$s$KnpkvN)g)@j+
zF2#FQ^-m+pe9P<x+~A%Scq>D&33?LhCPps;2_q8j5uSuD=KP_!4~#MVp%Sbr^j(ho
z;9`J!z(-j_wie$wM27Woy$)W1mV4rTfpVGSt~^RPv+W+aI=)82FZ@6pI8l(<`vS^2
zSCvCXPys4NU`fxWi6=!3{h%7*>ldFsGS`Cw<f?7-W9l|84uLV`gCB%mTnc;Ny+vf5
zRnUpZGf)yh5usr*)`PfBZGaLMBA|lviNTAqS=sQ!A6>B1a_@lEn`BIhf(OkhK%B-~
z%&^;tqRwoi;0kKqpLoL#>L(udr_}|&K?H2~TLV~z8<<UjI8zN2+2oirGK4I^*{K#a
zm;z^bfTeNKl_E&=ZNUm&`2Fd~!5FJu8uZFeOUUz^-cLV!-&f;7)$Fto^s?|4fgeRs
zgjr_Qe&a}7MedpvS~j!o5E03hYw84rCm%EiDA}CrE-Y^+oX=}Lr}Do4**mn&TXn?^
zV1x@9fdV~6j!7BAfcwMyV)_#X3WC@)R7BU4b^r^lCWh1J_}c8bO<7;7U|_fNKVJcT
zCplBK6&WVQ3{PO*a2)3Hq7X&nz_<l!{t<Tt0L#eNIXuIVYg-72#*uN0Y%6sjn833Q
z6QJN`6Vps*s_K&<di^hJT(_H;P1)%FSX6%a-Jt!WAs-l?NRBZh<uUe|%{6aLuf?Ro
z;TW>{^uE900wx@~Q<ZoQF610OcI|ddxVA^@9NVG*SEKU$F(KR-6>|rzCGbRy$WPAY
zjzJa^A}l3e0<Yx0o$H_#i9C^N(|jpyKymaUAQQ-()&tEUNTOO#pYwIgc+HnF6Qs30
zS8>X~*e*l2V!|n#uV=h{T%ppgum0Q#C!qxpFZ5m=bl8z4KQl*y$Y;{$)V;{US({Q?
zSZT>=#E~aF9spKK^VBa40%tMK`*azHe=z&I<XQKd%lSycp!BG9l+U^v+9pPJ!;%#k
zFZs0{RMpF3eBnrGjqkJa@f9?2W);Ox5u9c$P~*Q+s?7xPimXqeuu7J?k;cG=czv~8
z#{+}!2_89D5eb6HMQ<#1b)gnu<+IbKz-YXO|L6;c(tw(VEI#%Gp_&W^JV4B@yTWMc
z%(j=X%d!~f63yc~A&lNPKH{h85$c@EYz0kmarcw29D79thQ7~%I@5{YJe;|(_Ea#O
zJ;!5E+rOgdW~p~IQDMbx(h+s)7tk2vtMvb3yXGm-;B^UGU`JV|0e5613v#aCzPDCv
zvMcQU^iJg{?JRgV5ET(`wU(i^W_Ds9!^pt+o|j=$0$nCjid3o)FESl9+|{t@)Lj^B
z!CdAsUXm|Pp>7(!ybjx$ITzIUN^wP(0FgJW5UmZj-42v-X4|}KO-+n*2jgYxAChnI
zsN9!Xj4+@WuKDLmMrIIBSc&aCTsQqeFjJc%nu@rc8{kPXsXGYRq}|C}r2)w@(*8av
z*o=s~_0kZ*lk$oZDA@ec%Uq?1%c(2keoU23q4>_M;%2*;7$?Dk!f#iDRX5Cc&rlZt
zcz#iHi07_7CP1qg=(9L*gYbWXgsoBrzbcm{vS2=@qVyw*Zc)sEcE^c{2;vaD2iS*8
zx4Pc_XhiU9vO?*hG1R9_fQy%0Z#LXH;lrGP5@RWNWr(8117Z7^T#_(IsV)=U3KaWI
z3HaOv0buO#Q-F-^t=}~_&zp|JA3rDd_WkT_-lj<lm{KFwhYA-137473kM|Ig5zpOI
z$oG$XMfv#&q|Z7+5u<sG4=ztb-u;1}pUq;vu-OM#iwS8ZlT3a_$rrEkpzag@8jf1?
zLcPoc8duT6^#5zsmV{6J7p050l}@5@TS}8!Qfe8_;!5iaMraY(`795dMPuqBlvHEj
zDjj^JU=yM8<jMpwb##CM7)+RYgIlZ(ez}^8Y9HMfh{&RN{YibuepK&D^R)$PQh;tP
zMS|6xyqAKPIZ&{%{p>D@R$avah419<J`zxIdAlSc-Qh~PcF;)rm@mpQvuZ-M`62BO
zt>kwp+X%tpReh*G(^Z|g0%kcHQ$Sow754^y622alCa~MVq(qEe_>*D_X(#qV^W_1I
zj2=;^rdWxS0YWD|zmq(qYE5I3)*reuNZ8#B{~I1OF?Ih$$qTo%0TwR1*oEqmd}$&e
z@RyHCgt&j;>Nkol3sv!CUSG3jYTMu0HOQ;9aHoQ0ls>`%0P}|@ZwP7z=Nr+3qWDl?
zRE=mSH&s(v#Cq{+PsAq0Yd+GU?NKYrDd=LcrUp;3Lu3-Z2&4d17CXV6P!3XRQY-C@
zFBYzyagqa=$hBQH?sd}xUT<nx(G@2E{geIbz~Ba5H9j!p&3pmOE%a5jL<fD6!v5`y
zKpYc*MLKv`7^z}<kTQM1qdS#auJ<$d@lxoXsQll!E7{~B&=WeWH2S$(ByfR!BI4Sw
zsN!d@d2FI}ZOMc2MViN;bf+t_q*f#6{|tz4K@R=#Z4?CEiK0bGqwIi_m}#>V_gVON
zbzN0dWEp7Ttp!C9UnO6bM+GEs7rY%!23k57M`1?v!4r^9Iu^+lun&19Ms2bv4v`3<
zsH9cxn<mQ032MDk|NVn$<GQx1fCON$12u_w0G<^qZA*dTVWQ}kl)+$sYcvfC3Gt`Z
zyE;cvI`WB`wChP57P_Rw&%p5EN;TV9Ugy0CBp~>?Z=ZwQ7X(qczoNK{iGj|j96T0V
z^NE9qT_}10HIFzadun~Jp28-)ZPql4?iFeB`?oETnEoI2kz)T4o@0hJj+qes55hgx
z4#qDTP~{P_5I1l%SRXtO22MxmpLSC0yy_!sp*Er0aBuER$lU&zt65U;M2vlg=%Iyw
z-*B8Uto8b^)s@{wf=opmy78TmRob9qS$i(q`VSv{;<TW=xmwQc`qw`woD?{qrcfJ3
z3(wIaU4!=kaVDQx)OGCSko5}0FIzT<1V~p%+BTz$5I_y9rF(5ZYJud>GWob_+$!u9
zjh4Bo2Cvd{C9KBADUl84h27iH)?0t^RAeoe=1;w<a-?#Z9x7RkR@dP0<EsVd2p!3n
zKIH07|A6R|BL&jspV7v%*$83<(CVmi#00J0t8o%Po_P9;`>*^LD;{Z*erQeGz6|Hk
z&}XEB4`}d2kw=WwMg)xfN4(2;9diF0x}MK2W4tF&3=Nuvcan21ju~S}cIz=huB*^V
z(#5Gm<1_;PRQ|YtCh6r;#a|tBApD_BFugM;NHi!N_KK95D_h(s=&1F3wYSb@byQr=
zO~mT^kixRNP@y+p04w$YcsF!jcS@OBkn^xiNze(0SFDh*#l6U3KA#kT)N*f)ym+Ee
z91(D3fjXlG+!87Fn%w`ea8)LQvU#dm1X%20ogaVJCDlS2(Bv8eF`@JH_k6;zU}x>;
zer~%fj{U<pKA#(*6k*iSfyy4;r)_g`Ak&a&PX}SN89eW<-M=lIz-f^AP8vT?#tP0R
znNf8%LK2%22@CWZRtV-x_yH$ynA>o^u6BEmDV)EJ{3;b~E(&r|9`jEV=F$~b`cS|J
zv!-kod?c4HeswVK3ORxO!}SDyqz``eYEHq`Wc(o?<{dl2A?XX2Uuqb{IMcTK(*?~M
zH?uDy?;PmfxdSgL?>)soZ-Z&x14q-XEW>GLZ$T_{GPVwk(7U277y(52^NJRwujl4~
z5z9)+-HY71Yh!NHV>G1<p9`~#_W(q|%kU+K3&GBL;OzE<5!)tXWzNFAtI;;g?(%bT
z|I)5xwJU>s%kY74a)<HJ1)kUYmaXRtL6nhA(4G%nemZVqeePE{BHg#Xg$fjUGZ&r#
zlJg*Y7I2mRRoAl{1dmwE3oNPr+GJ?9?_5TyfQ!2rw{Dsl;O|ZDR9?Lil>Mhm#{WgV
zwHUYdPuIacyms61PB-iP(kOKu$HClRhg!C=&LKQv?QOw}^)hGj{9R=apSB$7e~8en
zRtK>GWjAiyTS)0nQ(4Fxo*BLQa-o!34=Q)9da|ZuabrARV<G8(p9Nrxs*Es+Z}>jS
zHnkA$8R7hBR+@DFQB)JxxPIz3&2C+Ey<TCM4Gz;Q79rvsMvb_S5c`}v4J7GRw?igZ
zJWFJ+4$i^mbERN}$ZmtaPX?5o8HmgMJM`P>F^z!#2!bA=KODduIfjBqg?h6-TjtEw
zt_P*c5UG>K^RYynK+uy$3-YM15jt_PaJm(|XT5Bolzni?b<k+~;_ax8_cdH@pqNEo
zuW*ETJYy6|ZD@q#6$-*GszHL|b&!nqWf%u=W_yNYv%Ij$e$DTvQ5*4iAZiSCOfKLS
z7qLp_E0HCBZGAc-5#q(Ii*@8dG<vYx_RhTKLtO%nTcCiHv-Qn}M74k9iDL}zp^?#W
zk|bW|d)5B&49!pqpkiN~$=i{9l$VbrI2KP~vsOk%1V)!gPlK@D8^;_F<iNYsADv_Z
z_;WCmv?)d&{k$y}e~8*eIr${&{dCOg-OoyiJ$yCmMecvs=oND&h!wO;!}ZwG!gP2L
z*lze<+oQ;_sBltG9Uw90udzkC+XB0t4J4GwqojW_iB3Y_IlVyE*hlO;hwSCm!cJ~^
zKlh|{(fXA_#`JM`OORCcK=^y!(EW%=PERTK=)w!2T({_ppc*Ddc@PtikQOl4Bg8|D
ze?3N7{c0z83qiML62y9!y+KZuBm3nODB4TJ(?@JfAkaXz8K|Lo{xU%Scwx7G52&{-
z=}uDzWF>$M1~PyeS&7$l@4hBX<R*6#j#f#H+3Kc;NsO#0_rIdXSce<PRE-&b6zbdk
zKJXw#C*tQ9cK?>QF@tW{*E8J172VW$pQfMqLQM&*PTR?h%QT2|nvC->3fPq^KdYhw
z)iKg(;!l;v0u{I^7bf~=I|xi}bZ)*4($gAmka=_@_#vJNtxPkcx|V-ng)TRm|7Jl;
z`^xDyttqIv(FE+u+)htsKT~x3LO)^6B_^DxIZphx)2|TpgP4IR=SQ?};p&S-gi^jk
zdvyfW2r=a$R>cKKlB~&CGlfFH*ek9WY9Inm9#vnG23;WK9jpg<lGy8M<sfni-dr})
z-V_yNrqsGT@zSk43y@A%G&8yzVmF5VLj*+3tuMq!qg*v~+WsP&DS(u#g29`)eO0xl
zIcc!)2VoQJt8wiy{t_l4!kgTB^^$7wWmIg<&7&75qA~0uLcj?TUYE&)36S_)El<>(
zKFAz$4mpN0K`tj1j-LbP5FVoPQQwU63gclx3b|C-R}|+qLDJZ&pZFQ_AuX=A@6-;A
zN@#6zMqh|T9IIdW00{zZrc`?`p2e6NFWDML?mw)nldBJ*esK~!*C0SJZy$3=5Ha2g
zmri6YnH>BFzwhh+Ib~|=I`qjQ$%qt)NJ$}$>M{jN+81TQLN}-)<+v$rQ&zAlOF?0B
zAN)a^pBPK2n&}r+K<U#-+{^tJ<y?#dkgAkF@&NOl1>2dxUT}t1Pym%T9O#D*2=ZrH
zPHY1;1eYNGYNrZ#y%&P#Jn&S81&}Y;Szaq8%KeoRzU&oR(phayd{OqnRGUADMee2d
zk4jf?g@U(m`{t3NuuFXgUb?r+AFBl0F)(D41@;lTt%!Fi;>hj-6R`Mg&w#NH)x41D
zv_XCMc`dG<fbwPcw^78XO4M-^z2n4vu7mcu_u`q;lx@U_X!3JE3S0Lh&rDV3XExwI
z>Bin`;S8w{juw9?!F>1e?kG(iE}U`&6SkzcEdg<?b|D$FvlHq*{6HM<mf_%2{}UfC
z_&S9=LY8l)j5PlTP6QS{eHsj?PvM+v5g5&fk_GGm$cLQi`$)>NB{-z=<L^BQDxm8y
z??T%zQ)8$A?W@}XpD8fGtU32{nSHrP7zP%?<r#rk8oE1A){-!xEs_mkrQW?}%dG~t
zbi_?0y`F%`^j!S|8hlWr*gNu5;hNGLONvR3#cFk)2*SKp&cM7bHVLEq=>jSvOf%!3
z@%-Ry?RvgIc5mr?|3x($3ER>|c+`SfEo4q_ZT*D5jW;64is<)|v1ck3=o8Zbe|*Gw
z2zYGC@hjhu5)3a}p)!_oOV6%`7X8i6E`=oR)cU8;Gy;!=i3sv08^ki+31O%(FY#7D
zi?)Su5zrd+A?G&GHk+C!`@ucn%+gn8Hg$h)+l%?lMp_%D(0K1XB!}&dTn<n;zeRmL
zO??(@e?2PhwE~@|rE)6;Ce$%{dRnLCqKb)TC)(L;R(h;(?HXDbz1aIA7RVF5X3sMU
zBLGfR{E^w8nZx%r#s4`Co+H~O`A-l`_0ew!1Q^Gw6=VS~ybYeKR=M9q8?|{1=WVHl
z%sv<PX7_%WvMI*9wK;@xQV79RlQUff(sUbjw6!6S+}}`%#&>C6g@#83M0r0gx8wFE
zYj?Wy#8PM07B>NVJpqy#ATYninP|}maSTpLY3qt=I&967S|NC%?uQD11%m|B2Mu~q
zg;3Rtn;q@_eagL1;IA@;@CJWiIOd=B{n_?+D1N)x6ZD|?mt--U!bvSR_ce>5JWi|o
zYc1vvU5HMuX;9bk`T}sWmCD=5XXbe2{Pf|EcZLVd?M$?<#!E6Ze-9PCybB>9D$u0}
z3_=^U{S7w*I^Pkc)n0$_MwE}z66?~Ifn16uYauEBEqR0LWm$1+0M<VyS6C?H2ZLq0
z6oeJ!9I6kRBiSv6DLt@oA4UNhJt=pgZ=xZq+RRm<Q{pX(CAoW@j!@iNCJ2-z@I6lP
zl=OrUNJ3CK%)#?KZ4XG8V6V1LL-Y$qIHyGk;%JEV1oamMO_p{)fNnC<r$}7|%#Z5s
zbJy$5y$jpt(zaxkHI|%VX5dOfn4tb~RsjSW2YpVc?zd-a4y&8}Pq+vwCGq6Sju95v
z>iHa1c?KMJO9X!t<}{{?4<Ny|>E^eP`Y_Cr?GoMX<R{Glvy$%~b8bw;y)!V&z?(~E
zipT3p=7`pA4F{JftUSE*50;{I`go+l1Ir_CP8TP7@GA_aa%Cu#n3#~+Q`PvffcSyF
z(Ow>Qv(6Wff;Iq48wIDNA@(Eo(-PqTf8&d{4iUlzltV{RT~-IDq#B8g487QxZ_xoF
zHi<?H!e}zU%M87?TTIsOyO?iG2O*&=H4UN2v89U4yuh8pJ+`>qh?Aly+uF~-LuLqd
zE?j;mVetmX-$#IM>up{5JOl>%wUOl^E|*A8j~%9`%>UmOw{#1ua)iU=^hA6oQx7w8
z*UET~eJ;TE?y)1zYqrQ4zo;?^dhj9O)_4CWz<vqC5DyHWJ;-b#0SzE+SMg4E02Wx>
z1jLvD99SlFYE;Tf)qVl;nSuXFp84yTCvut+?MPlqlUrin(VD0`usnMn#3<c<_$K>?
z_j$m@qjxDH%-i|Y*sD2srNibmnHVvsRf>>B${-n5JtkM^hBz?13%X_mQ~}23y*vLK
zer2I{r;J?bg%eDaNRe{pPQBJ9jBCRP2(a(9d~ca3pp6-T%)v8=YV6Eyo-xPC1%1ny
zEF6}=*X{Up1@GMT@d4@b=xSKlLRFs<yc#V_+L)#ZO!TZs0G!+B%!8`3YXcaUOrit8
z%mlLOtB}N=Ns<i;NfsZPCs<Q(9yqs4MVRV~hvdNarFD$61M^<AeKGrMdv;qrM4r3=
zV)QDoyfg@hr|so-#7oNnYhT`6RcBsppOd_Aadt6cyi&>A8ILPOy?NV<R1e%GDuE@!
zBR^)&Ih%DtJ83({MN^7V_oH?L%K0oPXHvasy+v)aCoB;>!11U=I=wchtFW)yV!3+E
z%A9?>cnW$Es9wIG-9STB?yHet_GBaA+$mMu{bJm?3Wcx`|GPU(zffXdJZz?t2DMZ^
z`CTW>R_44)jX**$+}W(*kos5Nc18pC+b>h&5^430j(M`tiOWD}X+w$*Y9me@SYgj0
zx!<x74g{>?keIY6E@HXY;_u+UI#5lV%2WlPUPke!MgkU_8%PRs`md!FIZL-YL8Iwc
z){7^v1}L<vJgrwIC`l;6Z(>jlC`XhT3X1w@uXxr~i3*fbKGioT3%MpIUl$|w8;DTr
zj>D~E(tulYIN53&DB3ipN4`0Blmb{)gj_)_7_|`oj0x~wBr|zt;!JXsi+=KiY!Y73
zq5u@@ceedFl`6ms5Y?MIpBjrQMxz>k;^lDR^D$}aAwu~@3p1lH2pcJRcZ@nPa)MsK
zsF8={ryuMXy#4P~Qf98MZUvH7JaqjL2h`$~M-(Io4sRSlk+=QcjVvJ41P+t+&D>?B
z4WhDo>-2y{3&>>HKjMShsT-U%nGF!?Wa^Pdy3jW-P9QPeCrSd#U4V&A0+RJa`d`Ne
z2g2>Z6E<*Po|$pf9osJ-oPGZz{pHbqn5x5JV4LZ@2~<JisP!Jo7>Fkb2{Z#CRf6+3
z7+S~XLi<6UzlAeE&DR!NQt~-@Q-NcrW!@)wZJGn1B7^BbJ9D&ZO!WGNCcusjQ-LLy
ze#jh=J5A|5=9~WXHssjntNFgO2|%65C5j^KEBVhVVMrQ%5YF?;7MZSY8VA&3>eF$a
z7Bl1L%#^eS%OL$ji`j*6Ry@*6lPAAUB@uf5gmH(uH{ygzPx;0p3A~%?uWAGk-=f%v
zi6#FQICOc3-h=uWQ1}t0sb9D$m=G*bpGzGW?TacW6X<5R2R#SliJPh9kky0Fp2~CW
zd_<^6SVpiJ^xQ+ZMO2e3C9-3N&pl}@Ak828wzpqeuphp{pWeGu`9<&{r3LY>9lFdY
zT+EZS_r*z37g5`Ag(TiB$sri80u$Bv$~-gz%TqgflbH(aLIAIfm#F>2vG(U2b7xto
zD2XSx1#7UI>(oGm8(GS>$Jo>7rvY?w*CwelAiBh#1ylrRJ<sv-vr4CUra#CN8X3^L
zE}*>U^*#ErnMNI;j8#m*G9+`*O8OVBlbLwwHC9{^MneZ-_@$-EnIpzu_+N3V_&Hz*
z<=6vWi7R%LMX~oYBtn9rG|;?txyCtDa>cBRUY#N(R13o5%C5U>v{z#>LUpuu?G8G4
z^o>HZ3=n?qyVkvZ)xo)fmnQppc|qe%=Phol#YvSDKxMP9n-8S;T#qR)n1AbagmAWt
z3H@hIqd(UmYpQMAyY{nmm#kPQ`f{6O3~ET$(p<|Qe<&a6&4spYGcB)s5S%;|J3uFT
zOZY>>-E=`CYpq)F2iB?<JO^T}veB(1Q8Nw*L~U=Rh!|3Q26;uF(4iCOrt-Or#xL+h
zSSR#h=3XQ@eMHgzUNpIS+!pxalfY21wY&NuTY7l(T5Q3KhsFm$Bm=s4OP*+o2uOy)
z$%cOZ`bF@7K&giXG|ri=D_ccr`ZZyqe_9GngibL^w9fwievlq388!DNC+JFmKgFgO
zjqq06d5O~^SpfX=VBEb=3`!Kc>!|Bw_AgbC2yv1#i<ki)cW0_hpo}Sv$jq!NLr9WS
z;qTG2kliIyzkUe<o}Q*1^S|v};gKE~ro;8{OT?O0OgIid-Gq^w>fb}{BS&UsS3hKb
z<5tt+OoEX;si^3x`T=&2-$k?>Rdj4h9WiII(O4x;90?x<*;YC^CO{|IBP-w1Lxc;W
zAY{v|UP;dxyx!!@T`t{T9bX?An|(B!dVYKv(qGiLkzEqBxefekUI>f<pS2mFXC=Jz
zws#DE5X?xfOdd>io^uh`&zO#jMqFQgRB&W=H_HB0ntNzopO!6nm9Bbw0Bl9we7&CJ
zM6-ZC%z>%91m{CyNI#ljzj`OHho9?QaX0iq&UQFEHcZROuhqg};=t|PdZ3&>@{)-@
zkjf7n9C1Bn9%A4A1T1pJ7jSUX3CYf)Q?u<(weHf`+zb2-f+@CBhJwOXqXtf;H~b;M
z2^aFWTIJ~%bVE~kM((Qvd9;!^c9L3fB5+SU9BM(_%q*7>6DZvGN&Xz)nxXj(mp-c-
zB4&iob3^1I4&=4=PeoJD?jvjQhZe*kX=k^Cj_C8>%#97v6=!X{#7kUzKG5ySOS0K5
z15u&wVr2)cg2As4`7)p*73GPg%}uQJ-Zbz?_|mC83TQb#`C{+;us*Fr<L4hs7)V$m
z85L(Aw+ON5ml@}iyoh8aq6=A@d%mPt7IbLRxPDWXl<C>`%W|iT43KzA8A>om@O2Pk
zA^dCrDWPPs5{pVyd#@j32&~z~rq6$TV}QQvAeT=b2?|S%sl+d7xc6J>Bn3jM^R_1T
zO~X`NWSC$Q&J^b{$8SiqYe`Qi(DI+dkvVsxh)Sr5;E|s>rtv{I;`)%Z%knl0#W;;w
ztO_;gpjjAfB6(SN$%u}&%p~NzW5RzscheC%v6E-l5ez^j*?J^P0B;QryJ|X&-n~}o
zioWh~^DT*{^u+)ht%_BoKo(-FM<XEVG8Krz{M8z-7<`<UG*H>7JK{idb4Lv!dE}3g
zWESwtI|vi#ynaZ(^hwD=xSyJgNtN4Sl=};<6-&z;e`Rlm!l}Jfg4K=d2q6LWLyc3v
z0R||8g(FyL(5*`%JkGQ~9Evq?4BNi8%Yx5bBRVO2^G+{R$b-`{D|L40<%=&Lt2$$n
zRB6Vh0tK-nNpCRjnT55R6GwR0f^4L2>EbCr*5tI+?Pg~fT|Ili>yvml?4zIfjElPR
zCm%I(b6u4a)4^VT|L(TDp9i64QT}~Td(HQkg1S6VO|kSZ%p)X#+cJ3weR_$wxM>d1
z9OVTs?T-5UE*+r>f7K%k6YT=(9%{NAYHK2XA%#RzoKoRUOR5jx;qq;~LM%*-2lcf5
z_-HzUhORogXHLD;J3zv_KGMO+gc;`~{v<l~PK#11Tl#`F2oOT--IOOS9}kYjv5~0v
z*WUESF#i~%uC&6+qE=(+FF3gLC`E;~^FW>?$~wJ}kk%1mBNh0y5mJyYuIbAY@_O1~
z8SbOgj^Tb=pvOoYmnC#6a`r}kjcVjDPW|0xLr7D~S=Z}{TT0CnYQoU49MgxBzYF6Y
z`EV34-ntk-{XV*Lc2y#<=;C%G5%a49{Z~%!fa<c{*jjwXY5Tz?3>)6R6qZx_PEIh7
zx9!@&3yRG5A>PSr0T1=z=7{4oPi9hO$kf}$XgAa+2b6h+@(rVUjuW+9UDL~jXWZ@<
z3;d4ytO_wC4VRcGyxM4I2p?lwIlNjM<KW7S6jNX69%xN_Bk(NLh;(EJqP)t9V0m&5
zoQx*X{&*9lCwkdJb;5_{aJ(J%4wtE<a>bqB5o4t$e2RSY8*Fp$o8x{}_u={0g>`S$
zN7O2(5>Yd`spV~S2--sH7$mEHq*0`EqwgnY9X`i<2ZHfhORkl;7_jNijW}3s=Tr^+
zKCIcUjX(~hkP_0>xVs;rAO)Rp{B7)Ex5uRo6V?NrkUBjNz7R2Vn!NNvDRYNJb6FbQ
zd)#->9?@z-^u%91h`BzQOwxD<b)DZX4#_;cr185lcsu?9&Go1wWR@~Xa{?K$6vw_<
zM1BV#cN>R+R1)xbs-Fd@^u>YM@5k3R2orC(Bv6CiU7lY&ldcGPV;?MdZNv4m7^!li
z7O0*+fI~u6{Lb?&+%>F--I2Xm3-#fvbO~wU5mEGzXsV`?fC9t<x-P>lxZ&bUEiXOa
zN;5oy8bKMJE{`M~YHQsp9xUGum$lUAB}PVdsGYilaW&6p!HA<`>P3a~@DER#_U<5#
zA)7i1{h+x#8i?X}Y5|~;a5kAdFoLf={*=QVJ!xg#h}+Z%xr~c&(yp5aHD|BIpEIB%
zbHv^UnXr(Emd%`$0gtyEB`XoH55!<HZX)jcQa{$FguNpr=jYnyRr}@+tyTA{pZ%1&
z0#JjG?F}cv)iNsLmC;|R*n@2;a`k<*uV=8FbD-x$uWERtZVQ$XO`86xymCUWWlj8T
z|2T_~=F9?yA}mwx6winy=iYTeHeNXqHdMe<oJZ(GFl}Ja{X>Y2$NCxbEBsJ=710dI
zx*wVq&2YNx{t98ZxBW=9+k|{ZKZ>&CIEU}MIp6|J=<}LR*FHs1e%F(Z9-x1Pzh*RL
zrIp9o+SD(K0a=k|ldStBv2AGY5*#9Rh8Nj~>4=`#sr^fL=74~-OY>N2x#BHA{}gm_
zUmG0A18&sTWCn0p5|ujt?Qa)ep^w14IuMpHKeYOt>!~03$ptX}NJxMZ)`eSe%^EE0
z`(0zBK#H=t=Gl^e<3l=MZ9mwsHME>-FF2L-sGnFFw9bNNH-Q2+(goJ%1OkGP#I^Di
zeC7qk&|8TV6R`hxbYL1?ew5XHdkKU|9s2ed<2upjXsQ!qrYi-lX-cFa${24`osjZe
z)5cOJp)Xr<1@ueblX5xdB9B$a#U!i7@&JlBNWX#Sz#$Q)E!gj#j9Qky?1te^Y{WY|
z24<rYQ@mJdEd>Mz#5m3G8zhXm9yW}8&uI#%N2DX*mM3U)^e@-A(KHB!*vxusUPp)(
z^^EE5BVdz(o{NTYLmlNMipIU-Tw@!X!kbfd7M%9bRmje>m(J!ILyPDm?>oWR$|Y|T
zpMcQ`Nv+TsIxd>Rw9VBcj)IF}zSK?s%;%vJUF;DD^Rd+U5g~;E(puFsZtteAU_c;J
zH&!G;*gNP=6ke0Jc9Thy5*@^LHbGY%ekG4+PMsHw^AZWzVG_mS0IwHS0||J#eAfy8
zm8uHLO0f$~GNx@15C!!Y;Is=;fv4R~RD&o3IBVVYNw1Xt%D0}zw{h9Q(5iJ51)lKD
zOcql1AfSfJ!rk^rt+Y9_#;}p0swD~cB4ca+v8yj-1#7@{Z+I^6$-_+xfHjw%0Big~
z7(TV73sUYkJG#3Ad0Ez#e-?92m&Gf<a@XWiGPZrfmM4!v){rn(xtSNMI5`P*f8|3e
zbnHE^dS+9D#@0lw?mzYhci;ICKjm(MnBHy1zxI%u=Z`%$nX5z*4c$%~&>rZJw+tFf
zVI%TP5Gmtqtrq1aftEm{%)-viP+43`nv^7p5Ji^^P7zf=*B#2&{lKG-#OycxS5lI;
zw?ZU_Vc)&It_w}yDhOrI-yupes+szTVlQZ!OYv$#cqFlT9N%0EpCZCiX&uwD;fjo@
z<x=H1CP~(BCV6Z=CkN_ag~pi!y?gT{xMLaRFfmhTjXZ9ZggZo+=QgMxge!a=np|9+
zGWt{y8kT#a!q`e_#-y%Y_+5jv?QQ|Rt$l0RCtpl@q{eF72&y(@dda=s>a7e!On~mO
zs27@=9;IiaPyr2?G3~5wWUbGCe#!4Ra9whZQT=@0gXxy^ang2pPoXWEDRw^KW%#So
z$eW;HS3B4uosDHy?h{DB)%rwJV!w*}1Cpx4uuAO{U)`LwhetQC_bq5EOeb+Vz>S4l
zdDqtQAjR>G7Y7-B&ttavpHWLKw#g_W8CO_8`{DFql2r;f95jSa?8drd*#d?io%-Dx
z>Y}V*8kSiMM&D6jY80F%4w0ZS^rn?y7N0)1%g7ireZQV@1lIpMz0@9Wx@7w`_tzlK
z%{!-Ov>*RaKKVk{q5buUPHb({+gMHt3Ek##p9mrs1&J@uWuHivO`83!f}`(h(SW>o
z)pi~++}!kAJ@xY(_<2Y$$4Xv+Qq?VI)o2w`))zcmn$6if(c46r&^7Rm`>)(~`u^tk
zORSx14#hF}__|lCdDTPiH;4Y59^CR`{Q`8X{nGTo9<2-w+WbPdiOwX(oBsSfvc4>R
ze{tM@6x|K)q+f8*%fvuo^we|`-h*4NN%5wSf=ICZTF1~j`szu6RkP3Fv~37ZpwJL;
zGq#K4(RBod*eZQu)lwsr#qS@1ZmkmQpML$dSp3^i3hfFX(Y8Y`+<K2MeM5hG^UXWa
zPMv9uRtJqDwaNro&X3L5%w(TFqf$T7EnY{XIX3vwXyDX-d)49fy}v(z)PhmJs1jUg
zTQIW3iY`ma<;4}ru1civ6%Q01>SB65?AP+7G$?0~Z!)JbPq|Bb_uGYrW;H1B>Kf;Y
z)P=<hjs|zjX070qzsm>es8B$jV57v-er1Fr;xcgS{2rpr&&tF>9~4f!{@jie-78N@
z1!np?{iOBQUVcb$zxW-1-y3{BzCz!Ai*?4(zuoglaMpxa7RL?4Bu2h&<zM`*umnn2
zF`B`|)t$cj8k1>GF?x}!`#rN|<O|8KVtyd*Ypm{G(^$)xYL(UNuA4}2qn>fg_~6+^
zO;@etmAL6bRVnmgQqOMZ+9!e4>0Q*AhtQXx#BC%=m(IHB-h2&hfWnaH_`a|=MI|J^
zYXF`WQyWe<neL^aJ&;2?KPL9%=s|>AOc!zbFZ)lOu|E)<;$|W6TDFs(5Kgo~K{ayD
z>jRfv5?U8crIGRBJT~?Dw511dBfQWv%MD~B_Lg#U6_IG->82sZ2Uest{eP<a&Zwrk
zZPA2aKtQSrh=6pJF1;htL8{Vw6{Je<iGqY8AkwP}2uP9MOH`UvX;LC2(t9XUVt~Aj
z=R0S-d&m2K?_*@J_eip{=2~mkHJkVd-s7p`{Rv|$?HnLJY<Kxc&awp*NH@A?r7LW|
zD?iH;p^!~{w1naa!l@#LF|`>oRG-&zM=PNpe(`r{KXC5$d)UC*B!TXGW0$8GSu7dF
zurwp|HFt9eHi5ZOcw`-Cc7u%;n^tdo6w!|mdf-`jzPbdq^ZRjhRR6-1C~vs&)A9=%
zdN94nK>(h*{^*aUjVvqmbegBkEvBcsL3ou0=4r}x@-FNbE%;FFPg}@qp`&LQp+b@v
z|6?{12p|h4uJ~L5g-R-{tc<yLsvZcdhkTywD&N3WaoFo?Y9W^KjsZ$XkC^hFMrMfl
z&rpzTK{`|;f2`X*|KWDCPd5^Gqwq?XB9c5)YW|dZ_kC!{^)Y@VcW#Eke_{c+IkqkV
ztk9t&Oa_ab)#i^z-kCTrW>!xtW!bld7N1Qm_%pieo@c*=mmyN2kE7ucQRU?5cQXQ*
z&7_X|zPr}ib=_Ld&EuprnoqRC!zji$kY`<8FCMt_dW;NEf+g;^WS$HiD6tRDyS)JO
z!qS&e^6HI5jak+g;XJu@8PW}EuqjABw+=CZtf}-fK0I{if!T%<yRW7#t&C;Mt@f<I
zk*nOW#CrKrcQdNC7k@N}5<yd7^u`%upv#rfYA-gLyHCXJk<4A2!5`+h#zTuHm5WDj
z@Ak1<w0lbl+c$a{hG$O*G!XbxjX<5q39n#@2mTMq^X<8IIkZQv5d-z;M{pZuEU(jO
zd0U2whMkUvH$Q6F5h2^VA|q8~8T8V`X(8ZMT0fa7XK@<rD|1*hy)9~%$_>bo8B;XX
z)`0Z_2%GRj5R`w*>5-OS;WcEyV~<lKCTYS5BpkM0UC)Df^J+6Yk>)4=O!)gjiDh1h
z0H;&i<fes*65qSdt7G<Ee2XNOH_%Tn<TpL1M%b#D_V%d)S!lZ-Rek~_JbwL`Ujl8X
zm?yS<c+M?_RPDE-h3+N-4ikyFAD^j2tyI2-O7`>|T>OcK%C<D2ux|8wGAmB%1BgGN
zU@w+KpJE5xIXyLw$OG)*qj+&Bxm}Y11(8kpwSqRtCyCp-wNI~WFw$6+NcVwgd<fXt
z>?cigHFHS6q)Hy?t5!OEsjMG;yk1LsQWeS|PNBNTR>d)#r4fZ!w5iGhEH|Z1$ECUo
zlU>7s%JhWDt(hOPx~+~Oa9O^G{jfvj)-Y(brp+8hz|9ZW=RH{qQ83R{NN=2pe~Nvx
zzrkzkU+nAwitRtCnGlm7jFnolO|I=mc3${5Cp6={K^W@M*R6fJdZM*;3F|oaN>AJ3
z&qBLrU0Wv~NnmAo=N|-%8e%s<NnoQDwo*D{e7OmqT0se@tJKWq=8<Y8XZ@DbzS3Qj
zUMj_Z{gx$JTk85d(MF>Db#&%WQzHLW)HV)=j8WFfIEX3SJ1pm#-7&uZhjMplo-Aa7
ze}=Ojtw>O%Ahysv3%<$rOw@v6E6SHun9Z(t!OHT2=@^;BLqOD8Uy7s+plvaw`LE9E
zD#w9<F&#`!e}V?a4Ab*@du#ZJ?i=YR9%)*u`-Aai&K%Wbi)40AzaAhTC}+S0xbLTN
z9(MSW(Y&nO;w2v`64~5nH;-}Xq$3YWuuA5#aqb8cOY|eKv(N1HNC%h@wI!2c19*k0
zs~Qc9w%uoFFsPuWe8U(xj9#B$;l`|=_=%WODVRDi=R`pkG85?B&j-MHzuJTZRG4X@
z5&7{B-t{T%?uW<9<rvo*^b2G$8>(Y>b&X_(*h|-Kos(VptIFVMSY4WlnAc??unVu3
zf>-`KDCZ&kRoAgaJN1*thqT{5-nj#r*K6Lpg8@|e9id%u_LPAXW@I`n8yW!WOxk{b
zeI9!ueN?om+(neyRt4d{^j;8$5#PEJIbJPf!t@w@jC80}HCavGW1O(y&?jA`U}#{n
zdS6@pA+Gd70e<pPD}9nlfOzMwWxm74J^y-@7KXr(<(^^~p?Y_D$_28!i2Tr>V2QL|
z=9!E$F}u(CwH2nGH`Rxiqv7d1rhNg%&l=0~8>B-dQYz6YXl}8MR(OQ{Ih_FQt1CCx
zGJNHC3sRXBF3%K@2M#VH6RB~o{XMYG=&f^{f`6yh0HR9F81&RhYS$K%tsH#_qe>Cw
zy+YF&W~EwAU6IafBOW4k-4zj?b90oo;Ub0UZ4;3q(Jgmy$ERnD_c5O4-*FM_qqxqN
z8hf7Isl1Z=c?X!I>9J=R3p*j32D)GIX79g9JP4)@yc$R`J)iZMDY@>J!oZjCr&9U_
zcU&&-wx=?&n%S%p%-&y7=%`t|^?K`g52tW+t1p2_VzruJtw}*Vp>;D<r(-+PSHvrE
z+m|hy^UF>7l;~L5Y?D{PELClA5)-wh)ZoiWx8yXnIa?H{<bY)@R2c??V(MZtW{mUv
z<aKhU1-(rDq<JH@98;MSX>B7C1h)1_^!+0xz1zvhNUNe4s3obUg$9^)hZ}hjdiP}z
zGhQ+<9@)!j6PJ$pch84YJWTo#@YA7XrkHCC2VsMYh+%?jbVJ$khi^SG49AVM%%jG7
z3c5&*F4qAGrIssA-NAeVsxjDC`28wFbHw-8Jj*EUch3{#K*BIDaiP3by%zdF$$Ym9
z(o*RI!-i)RIZs_3?hw+BuB(W-aM7eks5thnUY9Quv{cSnGWt66Ggsjp#6{+o?v_K!
zv@297NX?cq%wt57BRjwx_hyA3cSyW$${sT>j-fHOQs01~!IX5l?_Sjrp6=U|Q-VZy
zo_*+D?Q}n{ULALtAcR%D$hhV8Bs@`Ot>!YuNWGEn-I>)x&h`p3aNFI6-=x)KyYM!M
zg`!n@;*z;Q1t*k9hsQ`UuV>JfP7RStEXWy7vVF_whO9a#-4Pz(vo5c_T!Nh2piOBj
z_QBWV!9Snf;gC*7cm2)`vwpeNa!LNYEF&+Hn*HS(z1+HMR2^Khv>o;*m)vLOTlMEl
zECGtDehd(kTqnx6K^m(dtkVW|))}sli`H!QW!a_tK=5eIN8miyvOW}!BZkipGS7b6
z=d9&)@Ca0_HW$$>NPNG(ATfIR%KYA~*Xk!4n3yJCJLQ<|zd3G|ptX1^A;NHk*V_->
zV^9pgvPtGg6nv9@cbai@&l=15ODkzkZ`FTDt07xbsueuuy0CQ2$F8yH_U<`Z_p?7(
zq6EcYhCopIH!TBJ4JJ6N(}rgehcab8&iCA!B2;0V$10TMWGZU)l39EWGqU)AFj=Hr
z3ESu=LmVWN@4Jg8Iub}n1+Y7hT<}`rI{e_>sawpt6r5IFTb)rxG@UQBI-|yEFx1tq
zrA?tkO>;@9fY>@40&pVG27VmS5U5qSjO0a7qJyz#6-b4V2>)<1vQv;Sq|A-zM6GWp
z2fgU(F#~3=8=Lc%bH4>8*5OL)W$oR9xD3S5t&qbt)KM{}2o$W?g87z4JJ4KyzI#Hs
z^zsRx!-xdVgpsE5LzXfF<|%T@vb|9c&~>zuI>_1(3=|nIm(T`w-AJ6b*fu*9!UFME
zZ0MXT+SZ*BK^pH%%SV-T9%STowd5^p>q?*6u`eyGnwe?>MucA)1eCicgVeu~2yV+V
zxof!_=`%m3$qwDV)U~x6{O&3@oC*GEsIx$#f=!z3qxZTR0AJ!T)>}+r#=}O!%iKa2
zfc%#TqAQR`<N$`GHasASzaLCHt+PZOcw_phS`w$9p&8LBwJ_4$pWlHBn8gERHay)p
z0F)oSg*xDK(ADy0sbWW`ElcS3RNANG;a^R)BI2CvyJ81WY`MzrppF~K)o%73%Y=fu
z=)T^n7YyYRAI`3tvWM5oQd*F^$kj6voX39lFO8d7UpSKel7Sg{r!Wvdc_ap{x{-G8
zy$FWSu4nwZ6RYoxD&Km(bFw1Lm;Lo^FL%dYFy<2*N{Y~H^wdGEglh~>R`?iDOQmoP
z&(dTB*?>~fx97y6>r(7JzkQG*T}UL0%q`B+xiKep)d;b9oTc_F)LfrEKNpnNN^eH^
zeiyzoAlz}Q?J5K{5aTe`-H=6kY%pi^{Uiwans@}#y1#9TxnpB|9de00X`9$zsU?k}
zq%$d_PXFxT<9d0tK>sJ!=f4@Ml`z$}i?_WW0v-l`Hy6@r?TO=WIDdALRi@@v4i(^}
z*Z?I7dJ6^)RG815S5aBe9bf7bTCdeFI@IGq%3HsRQkiuj5s;aAoI=Aaj|}NgMj&Wd
z=ef1w;=oe6-T<4PIN7@C0>PSsQZ|kkNUl~Hi-VVC%<t?FpO-dcXX|1akP1aL1$}H)
ze)HXKwQaQ#5Mh~C(tyU8sO6k2jq2D+GhO>TP-H&4K>kQ$`w$gHigS6fAT9G)H@0Ow
zQ-<+KlfC?V-2P?=+cgLwF$-19m(6)@J)&oRM@soL-+0Bm`kn)Qb-xloJ7I%h(BV5#
zr5uX9hrXxAGPceKvR+hPi(ho0hN?@nP08!;99Y;FsIrAB1RdT?d?B^SUZs*u3{1?`
zv4?0z6%BsZNdCkyeQB%I4IlbReavEvmR);fU}GyZZ>@^;>B62k7|Z>oKV_TC2+MEd
zXcJPnmL!+RP$mB&;~V;+Rtb9LVIk?)#%Qy9Gwjt{Y<9G&)y#SFCP`d&iq1S<VI7+G
zS{=EngC_Y;6>XUJtb!N;$#tIs>UN36bEmzA0RZV-0hq=OAxcn5Z&AF-f+e4!fR1Q(
z5LO=YljNtCmR;9}R3sJcJNu6w(2k6J^-@8{))>O#5DxFjJo?K^n#N-RI3)bQC{n@D
zSEO&lO7k=*;yfTI$>vk;wdCu?b+k<xh_jA&!{7igVeF@(D_|xv)i)hS6~_z{!Vd1b
zG8K|W8FGG|Q?hR^ydkDh;dA~(mia+f*GsndZ^Gn;@&zgA*qp>AyxcTt?p2c(9Y5-j
zGiGAwQ-zZpyb5LS(@;i*LiGhF367?!<;jxQ0{Mj~*S!ICz+5(;_af&>wCN^Dj|x!Q
z=?4A$8iod0^4s%-7YcW1Ft&YpPfhhTiixle!EUp}7Lb`nGe8%+LnTuczea4GN3gV@
z>}E4;5MEe12YCs_ZHBRHO-TPbXuElb!^zTv8wyF_fI<_Bad7w-gOx2l4U*Ey#p*wW
zKY|SB6C`bJl|~fBGi*gj(To+%V<){aH!weXNRGyZcH5E5e*W@gp)K@A4n-$tKH$As
zUHAPR`w8fqxQZtVv8Kn}+SuymWc-TzxtXg*pah-!e2gdoU!E6l*73Ynt1*!$@cD^c
z<F&xHBRm^Zal<q0OWDvN4drJjz;uDn4124VRNU?>lIuc#)v>!}7vMItMr3ty9zH<_
zEJk!88@u>+46fm|8Brt7KZ_;AYoV|LaoED4Am#Qh1^fv#kNiCo5RFiVC)OQF$q06r
zw@)gsew`g__)s&12T0g$`5B3FHS;&BZzWNW?>ezv7omyEjG<FjLo$qB7FFdhR~-H3
z?OU1f3ipk+p7Nxk$T*G0&zJ&w6+)bCdJ=NeHHN87L(n;?CsX>iUO8Wh2Z=>Bdnta*
z?EQm=jL~+Nv={j#@{pPFhw|q#2-C&K0n|7M_l=0QAKzqV5;=^ncOcf;{G9G-`X<xo
z{?esp^E3N#w47KE921wN1c~is?-0MQK&hXi&1kZoL<S1x+#j6HSFfJ}`WezrGcONt
z$Ake7y@=I&y-NLp>{VTMx98YF@q7-YK{lyZ)v!Nk8vC{tSgqhp2*~xdsB+G3s!Uwm
z@ZutZa@42;ee{g4nkiA~M^8r4y7v3UPOFnNW+eL1pb6GoNUCNyXko1z`RKL);5;k7
zU&cSuI3pckW-B|XGdv)&LiJsbV!_EVX4Y*W`e~!Csj!}}P<esT>eCFFw7t-HJv`sA
z6Rm-t*INr`e9RwdviR&lld&&Fk0+!PbUCG$H=~YAvp_)Mx``9*vlOr(CoPGwbGdGp
z{a%GmL}SsKtU{cbe1Yo0kgQguSSV<^dAH$d_&yx|eki1*O;ClY^L5~><;WV9=QEFs
zDn`Nkwyw=VWNHTPJ?FLEX)2QWr^jrHW)CLzGP^Kif4Hq&^@pXPj$`_+ht=6;N=vJY
z!DT!jE*!A0d_amm>Wnmbg=^EFhDim_tsdf~6Vzfdt#nfajvELLa?EOxWPzV4NPd!9
zpeh(&yi)fb3%*yMkY<-=)hdGv`Ak}`XW)~mUC+25qLn^E!-EZ3RVaAY9}AZ@Q1ykS
z7@L$Ahzr%`9Z?Qaj0VaTwY^!qUA<w)?qPoqO?ly^n;xWMbMaPJIX!te;q=mzotwDa
zWB>CJ;L>lr?_U*j&kby)dp3+}?*g4KK1_48+CK`^a*?TwX;ski{e#X|$(|UmF&RB}
z;OJf9rIOGyXESqq)uxtaNRBrQA3Rgmfrl0;HqbSR#HF$*n(1y6ylD42O`do}IpP|c
z&GK#=%;u)A&z$ZvVuDDZ7ld^g;GdlCV1dFP?>g^xx*g*ko=eh<yvfx;`3DFnYxE;@
z_R}GPMdz1)>0?~`b~d#$Zy0WDDP&5%lgnIhCNnD{ZT8_rC#<B?O(>mQ{S}f%gDqH)
zo9rikSz)k4{E42$j!&gnOZb+ag?xVZQ)?KXKKTpL2a<83+S>;v=Wf?)c{bg*OIG7&
z$r3ViF0B_N4GCe`01mVu6Xuxf?(S&5)&<kB6Ps*=t7YMtlXQqXNF)PZLV!y3Q^S)l
za1;dgoKeOuq?(97=$Po(Y0v!0EyumqUw|XHwMniUqpu%f{=_Xx{hgH#+^!x;mn5eV
zDbm6LJsv0BBS^hIIjf7A1%>j?j>Vf%M!6#1*>x*cZ#&NRt9<<ps5CCN0NWABrEF^)
z(tQR)OcF$e4_qWh*E>7$wkl(07<LTTV0dyk)vjmK_PToYhI)NuP_N<UnnGrelb6K8
zt$NelZtW9SrO%_dLBz;_)u+WLT%)BauF)*5Ih@j@v8F3A8NyA4b^JrD1#4HWQ_=Hx
zwMvFG%R4?SEVVOq-G$b~cEv@TQCmHEK1(P7-x(ONpn#U!ZII>d|A^&GaO@@kuE0{y
z!tod(se#Nt?Jbd|G4tsS)zoxd<$D!KarS&&I!L(-5LF<+DILCfMP8f88dCXm8Jp1N
zZmt$td}rauNb*E`rSL+%0?|*2zTERyeXiT<%wBxL`z{mQQaN2)FGe398j;`B29zU{
zlUsI3t`Uutmk9qFlNe)4%SWdxx4DM~1Kz%6D3=U-d3M=Ac+{@WS4xwg%HTD5X&bv(
z*{P*=|Lt=f?X!y1{a#sv8A`fs+^-#i5s>yg7AggLtJH2fG9<0cAK!mIh7N%hhd8Tx
z-F33dRrX!W)66L!5jXTEq45b+!w(q0d0)F6==v)pmbjMu;oa732E>dC5F#N#v+=<_
zW^4cc3-+CG)KadnR;KjR4p~XPxvX{Nk}TfgaLxK0NOWmMnxxsE))r=>BS&uo1u9c-
zc7^_&Ty>Yr^`BvW+7+MP+ZU{zJ}n$-%3o7{JwZR_g2b=6J!$e@2zj4zb@!^UNb9#F
z9)1qHR%_y%-YvBLG1&T>Ob9(EmAMbm9{wFd3%wQFs!S1h6&KHvPY5{247iplVPD#B
zpF&;PLSx<OUad}=s+7}a));V(u+~x4Y%E1vBK`Mga5L4E$B(o<Odi%1EGHRfn@khG
zV%NqfMI7@_DCbc(n`G)yrZ>?~r%qkuiU%h_KzC({+tU8j;(Ib@tF7Ay8?6xbSuZoQ
zY%15f_&j^*ax-5|U$P+oIW?TAdq+MlyCfHRo%xK3{^UEI@@l`1D>J{Ypsd?N&jFGf
zv6M%aKFs+%sshr;xPdEO6hK8TH`rTsqPCASG9MU(Zu6|jsk;mPqOJ!=y|0ZB&6b9E
zRrZeTe<$j8jbbO(uO3^NCX1KV{vuQ_*$d$AZN7%D%G&nMmo?88eP^d|5im3M2}^B_
z_*u8}98u-EH^nO`7v&Hs<C#VpT=0q*ET5{&lY4gA5W%>{^F4I`=>6A-rycIBzZ;rz
z_x;1^Q^n;~%z+aMjJ_5B_cg41N&O7ZkiXu%Dr^4(jpYc|r|AK7ovo^kK7W>+16OA~
zy=Pjsb2`pV6Hrz8Voa2^c9XZYczPK#a~w*hP(!w^Z!T?UBz0VwS(P!l2!Ipo1A~@p
zoNt{8kJ1R~_(HGOhN`xr-lfN8{#@2M3r<&&{T-FX>9jO=`Dg2Q&Q2U(J-gF0vD1F(
z_x83N<XG69fFp%W&k@mawl<UHGM3nc(mR^Aaj~CacB*J6yCfiZ{z|NA1dnKN^ds`r
zN4^X65kL&jjqs786BbGF@5l*w^gepoi?|z2cg0HQ^06qPYalsmS7>)m^tyP3iKTDJ
zu+bIPsiiKMOS8_8dt&fiXx64L`f}){z`zEy5YFw~t_C@K`V1o6RF#Pvna>*>s7hss
zoVFt8FYy2NVCUMe(4FV@+b61aif<XZq++6?$@pg19@aAZN~3Vs<nL`C>8B+q43`eD
zu1+k=NV&}D=$P%xX1_eH3J;{(j6SaR4^(*il7}IW!P&ts`Q0~f)eom~J;^bbUtX+f
zhE3r9?SY<NlY8DNj*A6+xoa4t64WA%il~VUs900l3$&7f6)DqJ(o^$*cw^{hbCD3`
zoMV-HyujHR_65yC;ria#;p46tY_jbA+2nq4gV~n{1J8Qx_Ks6{_|7a7RLpK#hi|bT
zdGXpek2tN8QKApYJUWicEP^BjM>H4eubI=(H@96Ad7qx%uu$m5wXM-7t+}Kh__8$j
zOs<N|zUtAJ>+ZbR0hxp!bop-%i`<#(6Uw8?SuzrX^IMH>&tsY2yx-xui{h*YzZ7=$
zPU_8&-_<*py^0p0Ll!W~j%FIaW3k68JUsH12fszUk~f-(1$Wq0c_2!+?flHYRNoXc
z43~N2Lsov<@_MsP0;6pxsSV<rJtXykw$mGmqD&`kD&c15%l5}PBWK6+!`=9UUfDOZ
z{YclqS}|OQQ8|o(r^C46&8m>s%Lvat)Sg{ZC&N_+JJ+?h&q&|-Q*N==ri}VtPh^pg
zQ==O39lWho+g{Ej)%Woiq?|=+F!tgi4_t)PC|=e02F7z~CDgbD=Dgl_z6A4D?SWM*
zi;I8ze!}^{KE_S<b&<x)+nxn{mcfzB1HkuY^waodzU(_jn^tcL@unr#_;PYMGTza0
zxYJ2+z+p9B_7EZohy2dHiZ($8iuGXv``1!X8!^%m^xNUjlwbh&{=(M)hAktb;nucC
zrVekAYH%u=+SxzK(lmkO!se2X#WUum6tuJL&pjNb`VY5En<J}#(ey+|XENNd4Glzm
zduPl2vOm2$=8_@~%f=o3GGj0n)!%EhBY}FF`D~#ohhvdvZvs9(9gg8+k@KPnV3v$i
z19ZvA`R|o~GNKm>>(z5Y`QSA?59ZQ8RhG0qdz(Z>o8ogNg(UtXM6KJ%SKIF%cYt7j
z$(4HzW1F<0cjKB+W;da2XC&&`h@;LY=s59+NV8S^*ckc{dRUf=<0$GiODc+|%pW=9
zakn|1ZQEuQIHP13M*UehzW+evjLvkn>wsir@$(#dXezHg1I1{!KZz`(H=BoUaX}OO
zsdLX|m?N>QI8AtQX;cTq$%4kfZTWuTW8b&0toolwCph;eEY376JKw#tPyNsbk@{f7
z>vU{&y6v}=xL`=a&_JH^c+H%fNk{gT0aXTF^w;!!>1?&(@&H1VrXde_0%edp;F`R6
z*{*y1@cbo^CLT9rcLQXi8tE%V-vmA(7AAipAx!5OX|f;2vTKhNBtq+QrfsI{z{(I!
zO>)0zJ8GBtSmwrl6+h&al<i1<P*5f%)bR4GU~@Xe7t*RGQG;LO&SR2`;z`KqUsQ{o
z(dU}mahua}p9y|_b8p6<d-*Yb`313xdi=}fS~_EWa{Bt%S-u|&&!onU3q?`-5?ot4
z;pRbpI-{AA>h~~iH$f(R*YD%=tuD8fqKYzxihe#jx+Ckouf$@{6|Jgkw+a^tazo>O
zb{=#6td3v~LCvjOb!9~)7$DmDlsH|HOYZ)!r~Oq+8#L=o)=fLg8wR@Hng=br1=VjF
zv=Eqljd-O*o1-e!8mj8eZOdcte~XfS(Kqg;Ma}|*nU#9-eE`F(*>%p*L>3|g-ax?Y
zgPfBHl9OAz+2o-+^++`YqsuGbVX-$Q<>kfa`Ng&ETrX%IXzNriV5VIu<AKZCjibbh
z735m*QGBI6c3Rc{6@bTU-m)?+@VP~f_~V80JmL2orRsSW^Zs<qX9aV-dBG3mWvbO{
z=DeD+9+}S=zB;rXCh>CXvEZo181Ty)i~W9#tC{|o%QPuuyF9SrKAAbN{aZbo)w<nY
zxQ-y0zj`L^#2hj;=&f${?Ok<SF019&^U=kim0b;4zZ200*7+|#zPfL034Uq2hHg|r
zomzBkKXdB*xG&z)ms)?T^oP^Pj~=53F4wlVllwdNHNWD*^C?;m=L7&^H7GuO6z!dT
zc6B##Fnr(IpMEz4i8l!(L3)A}+F^JJY2LLsB+5m=*7AKg->@tDh^sbc!+V)C?~P)x
zM1)*JLOaioOqUVP2NWW&UuJM}6tQtLI<fPw_H~g_>IyjeKD!jnW902+Cq}}(A*Y&s
zRVy;ibhO(D;NUueW^QVCQTp${5`?_^h9Q2MGRg9*%*<-*HnN+K%;9<zS5p=frs8<5
z@*bL6Q0j*D+2>=`2EkP&d@qj(E-UKFEsEXF#XqkP*F8>Tywsz!RJtm#+M|``;{c4-
zW9(xS%8&2)Q5_J<9MUVkcRFpLIsi#lF}IM7kjnYxW*ZZWzqaiqroR{de!QVc_Im>M
zVX409_JN-y+Khb?dTk@>^ltPp-?;$ch+n?<@XJPuF-e+F#KgcGF+-H>m#sI%{l9w6
zq}VXGhA%3;#eD~8P7=NbdeUQ$J2Tt+$($qyT~9<XDl9Lx=;3XxXKES_%L#g%SHrog
zUYEv5#}(R47FsV$<}JT@<r*E@-Wg}_Y&fAJC831ta=KUG!*M0qDed7|{0*oUc{S;5
z#Cc{Pl@HPQq62;I?sA26i8aggr>RoFGxgkP>ddt6bMLZaxjeRRf3RM<^q#s&-LLj0
zBE5~+Q;F?(?=!qgmOP`)hl-mmRr8=;;@vt(IAHX?*miyYAXwu?uQT!xSr0^t#UW;b
zCcD~RdGuta77n-Ol%l&O{2L{+cu)}<k5|pUgFw`=8p`({8HoJbsm%)tCvQqs2c3u@
znJ4nMb7`ud6Os@!_^j))$Ck)%NGZ{Vct1KX%k@8>uczEkF6{0*x~R#OT=#?@phFZ;
z!Xlqtnur@70;^DN8(K!XAmd8;c?cRG@(_Yu#Jh?wdW{}H`7Twu=}SrTL#&|XHr^u+
zH=*K?koAM2Y4W<~>an<+aUqwFq!;kVI1l>50dbmdfsOvnH${T<8_D0MRh?leQ3M-2
zs~Z96?^y*jRnpjatNLc(O7bBx1jY&$uy<95UB)<JYLPta^N9y^u;P~AKa&gpc(8vV
zDp8mL!lYJ!Fd-l%qP|-aq*tF}lBS@+*|T}6m*(3kr$+lClwC9)`T}s4l+s*y_)0&C
z(I)zTe?_ZR6SBO-Y8UjoU``;C(<L^$#S&DX@_&7=ndogmJN3yG%!CKM1K0?Xo>r*~
z!nZz3^DtBG>fpc0wV)#?&Ktwn81>+{QLlM%x#I?AQ+|VoliA^cvfNeGm>IwJyI?d_
zesJM*>tn4Ofh783f>aHJ3pjw1nL&5<zBW&LY<hUNtU$CYq7wreB`psmfIHTU_ndeQ
zH*K{f$j7;gh3YBu);zp9>oflO1zX;Ke#O#;^lzD_i#>w^2FAtp;&tT~3Fx0&jT?%T
zpo=q<i<c<uCg>tn{^I2Vk^=qv;W5xN(7!t<6oQvP0PFeJ6A(br{^$1Z0ItPnshju~
zgV>!~mgk4m9}m6W@)oUCzEHetX<d%J<f4ZQV=a%ikt0S$07wZ8f^k4j#jyUl9B3U7
zH!ju-mWh#;1hv(*cCTkmfz%Me(`4lvt}ZRB=_O&@l{?Q&Th_59lV2mYo|$sir~NbC
z&qoRu@o+%BMyETw;+UEXJObh=s3lubm)2_zVuDT3+s}Px%HU<NU%+Q^FX4?u=*jKp
z>aOJf!L7~<f7!q0^gsUp2E_o(8oC3bgaAp|B$B0YJS9W|#3Qj6ss+Qt6>^p{(m@2!
z76hn(#%wa*<uj<pvti4%%eEl44XK0*Wc-@;m;l!`vfRULBaPt@BqgSx_9oStJ{9Ax
z5hklGrxi8=N*~@o8S1DEKhYL7r^+pn&=9G;z`egqiTc-KI^OZ3N<z>z5+Em6g~0Jd
zt0Ue7iKSz?vY6G=ucz{?xwZtEgEL(sRxnT!?wqvheaxzhDgX)HUKaO$@42M6K`Jc$
zal*DLvtr058{}L79igy`M_Lo9!M!mH(#Px3!F~H-F>BvXK6MfNt;11*6G5j=*2W5<
z+*uXTIuqH2nKS1k|Ll<aMXB0C_yKr<<agp@4+V+NvjoPSMj#fsDt$#7BruYgUkb6|
z_kQlp>rG@37#5zND0(70w4p3#p_{Xo=6sr~nWHVInL~a5cmp(RPW7oE94A4?6DCtK
z3I44scXb(eRTxrJm3%vd2=GzOZ*6X*k0{q3e%}dxBP#RXute0$gG*<214q9Q?dRt-
zR6n7!5LfY<680GBgQU-B1~S5M9l=_#PMsV@!Kaw!tjV2VNnSLe%<t$--cZ-S<uS?k
z`QVfGQ0w;PMb@EXcY(Pdf`_~7r+@xP_gdv(@l-{qtT4+nplw_O%dDkmvXk?F_+R4%
z2fRW00FeH{{r(tTr|CxC&hifDa}F=Zj<A}T)juM3AErW_rB>AT2T&VZctblNt1ByT
zShc>DC3fmDwkZ*WR_p5GXGN_8{k(pB%{zVeFgJ{6C0$Lc@lZ?7f^6%<(VzZ?<#cn+
zMlpQLpJkCzH@UjjdCsBX3o84&9iA7-4V?Tpp2o~5G!A<I9Qi&?foxpv&D9+w&T)Za
zY%v{-EQ`sf;R@2dKJ^A=8^4_19~98E8<Bj^3cj882dxw-4`Kk%r=nNwQy1|?ED0P7
zGE)b4-{)Wj=Y9Z7Umu*Y-v0fcaTEVT<>(7WUivBC7l>yv?WIw83j|&fO<!(nzLeN{
z`NSL}fmre^bCUERHqmiE(tRwvxt4N0#YKF(Bw7oh1553SZ$2hsf^Q<ILFm53%~WaF
zkEyRd%BRyZF@c?LCY+MaD0lNUdyfp|WM#gd8G$42cU7FVo;Cm|`Dx+-2R6!S<K4_+
zND)ineaHX0AsgnDpeNks8PdF15#;Mj@=nmu1j#+ZnN$KlEu1;WmJk@;913^L8n|v|
zPNjvIYJfIL5b(9Q<yOQ=QRkIY?sh|x?|*_Bz~S_gEGM!wGJ$lv<H!++v0?V(AK1fP
z*Z1R_k1!R;1Kc)1vsR_i{DatbAwxOQkkC&PFGNckZ14c>M)6Odzu5jaD||{E$R%9n
z#8XQs$$PqvAE6Zq(&3*IHzb8|;kF6VU+r@<v4ncl{&CHUkI{}3$%|)T!^IB?dRnn3
z?7^QX-})WyM&mjp9(h<R`9k9jwWy~*9eUmKx}x9Uj@07CXNOs~8&SRLpkMhI?UabE
z2O55*{@0DtxM5;Uqwo<eKOkJ2yPK<&b4n0QMLEZTLsLxHtwy?WXW>A1@w%@{o3@f@
z3C*)>lFqC62%gUQI6n}1S@e6lC0>!A*q@;U&lHc%;^00Q6*cxa3>i`~WLD;S<acN%
zH~8D?$p@A!xej`uPuin@m>0>vBuin3u;?+H8Rrj3Cy|BZ*YlA50g+YQ$to?jQGr`0
zJkC9!g8;!DhG-r)!!s@$W5fPDKi!81r8#qNS4LlEKi@WT)k+=6>lf4=5p31JAXt-p
z>v`F)=R5}k^l29Wok;Eyi1KeK-3*OA>6{t|(6)F)S>*8h*4|vOg8GZT|3)rXa{Ri=
zGZpe|ftZvtY{s&%Q#Q_Dyn3m<<tzdh#?t>OqRKZvf8f|sZ*}h+E-7UqtQE9UFn0(N
zMtMI6AOw;c(ggzwDBrdp>J5TUd+I_W4-U>)@BZZ-E2n=*oFzLx&P+`T=Epc~zMDJb
zc5(WzD*_Z?p3@%GEX%8^0C@}?SA3c;b*Q962Ro62t6rb55iZ#UdHs40@O96HpsdlI
z;Ay3b(`82O0mPdH<w$MV&g6wiIlqTKZs^}0fO)Rq`$hWy6tV1ZFZ>d3Y5A{1`~UZ$
zF*!P2^EtQD(C0V2?K;v=^v}Y4es)eWgQ&d~>AVe2<S}2BPe&?nVW@X>hr0(8PV(Ul
zCt>WOibV64QVmRgpzVt2$H*|;`6r@I%2TwmIwA$#_;G5})-gOxVpq4;ZPKLqf;0$!
z#o5RImpF6bcn#aRZ)Y?@X(#@WquL5J3I^@EV$eXMS|X(LAwB{V@Sdd-R0*bpv?5Hs
zS4X_uB>svDEm%I02fNkOI5OWb!ZhGr`sOdug~Q?cGtJyyPXH13o!`#{uQ{}9WM%$0
z3iuz_B>sN|pwQ~540*fiBkB@RwvyTYue>>odFMf|Xkg6*PV;T01_VLDg%y)rC@=hT
z8z_HD&I2O3=&!8R$V#+B<N@Vw)e`{xVDf*pX@h(wAa?{X0?ZG90?GU=|GX(g{BLC!
zP<@dCm%^F)V7xFqP?(XGgZ7NehMj;12tuKEf)ghAjn$r^iLB|8Bt<A)plSGE5K@7L
z?NS4n)nUcU;nPs!9QhXUxr|qjl648EhIG%r?4&{!j5FfRDfXM-F%c`b&Jj8veRLV~
z5V>(G-<gEA@&&c=be@MOw_&WI;rVNv8MkzffU&{hKq8jHxx(a~5!My_4oNuui9Jw`
z1?=h4O%9w7{RJMF1yqFuDS)Zz@1WZc!+!iJZ%BeP&Xg|0R!lfi=b5*JNqHq@%o!W`
zDup64UYnR)4rO8>5f?wI0gMa2ze~rdVMZKF3Cpjj{WL=fOE*d|Pe-4p`?27$F{W@w
z32>sw;Zmnp__>LKDON>2x$wh3TDDsu>Qd5#PsjJ?y@55)!tQmpl?_)HXv>XS6SX}@
zFomk*E=nw=)AJX?00p1yUwZz9F;VJjot__|5{?5Ba%*okHbWd=bGlUjHR_U~-}M^F
zZ)sdpk%TF*J3)u*+KYLMfcoUyHh(w29u62U!9)5OFuS*nTs9)GyV$Q>89;)$e<6Xm
z7f8`SWP@w@X*7W%M~jn@&zvaq`iXrD>qkYvZ2j_gJ2nhgKuHsb(bN6(7#$c8KZH67
z%?F1kkwSK{h&Sgzby^}GumZ)F`{Ds$P5gzs6x&RsXu#nQVdZpMKBKZ@442`;sKFa)
zc>WL=0~NgZz2tX~-NVCyu3>-S{NG|i{4Xc`vuy&&_M}E8&&?cT5b&p=qN7}?Wb^$0
E0KMX3kN^Mx

literal 96812
zcmeEtXH-*f_vK9@K|s3n-jyoSL3#~UKm-M;BGLqu-jawY2uLqdl`2g@iYO&fQTTy?
zK%@#pK#(SpUUTR2zW?_>vu4fr`7pzRb@`B$=R9riefBvw{;IhlEj1@K006WoBRwks
zKtlhX!JIvPsC;;w3q6poCWd;@6Z$#OZ9_v3Du1IJ4*-DY$>|>~`PW}}0N@8udfL}Q
z^Vg=sa&Fr^CxRQ5#Z@oH*(S1e+^t!)wJEgE>%sJcZ#2WQTiSv$hKlx;-B!0vl@7gw
zZr(XuqLz-MjyrpSKktpuSskjU56>smeC|9?2#!X#r!FStr9Mw(<Opy(h%OW>tcv*4
z!RUAG`j?3L!Ur>mG-1^L|NXzP0^!}g%}<0;oV<%@RNCX@1E;i7fpm-q23@Vn;EW)m
zfww>ch7v1>jfzw5ZWJe35;=(~L|G!Yv!?SG<s~0V$$N7R_qaC1U`}T`seLBCY&BI)
z?<q4YpKE=z9};zD|IT`t*zn3=fS@<52zd`)NXZIcf<1vtH0|f7$$$L{E~2+^!nAx_
zxP#ea8I<ntUs^~<rvPrlu{rjQ0Fg@-AHv$;GG_iW{aq9~6E5lPGd*q{iZdfKu}c$e
zNlTtKY8wFxg4pDcfl#)@S~9oT8?JztmuDW3V|S$q!k+BU!G3566sN}NIMD1Kf173I
zE4DtlT4+9H4PralTVbJZ2;}wQFfPEvr2EIYAGf^P3LDf1!H-?>Rns(>x0YxZtIrf#
z)f)jwf4AkFsW?X9a|?aufcMUla%wY2-tmU0L0HBVHx)msUk=QJ#!Hir{Lj2zdSeqL
zp8XU{j81Xt`Te=)9FdEF{>J9b68-X9w58{4A6AU|=&e-8qJHzPH+wC1{P)mWM)P+J
zvzM%tvrk&>9Q?)iy?sT4c}Hp=Yi|p@N2LYnGfnq(^fTszrH?9u44Pw%lv#yac+n#3
zwA*AQXnc|Xp7e_B6S<oe>r`_w+rmW8xtIeBd<FZ-i(3Ubxy_hY>A|!l-Y*ue63l%0
z8@0Y-xz!wRb6(Xc-C~tJ6zGidoE0UYd;BcNLxNtO{7T#z`bhp{d{p?l8C}^wZs^lL
zUh1@Sx+;IK7|css?Bt<N23KoiqfWPAv-_h8pJzST`;ok%c#xk)ScR4Y49)>XDm72{
zWBn57{YsfaFXp!BrRv?i%sBeR6Ply9fVt?(YaWDyx#@Gx2JiI-`XRMjEf3ZQC8&KH
zIl0G-!U75u;xD=iX_!8+?zjWsq<v=XCvl}sCq6$nLYQ}cz6eOZ->^H2h{e_EZI2_4
zJLB2&{!02syF{>(o_U7<x!!ch%S_t+xLap@lnu+4Jp}!g=D=HOn%(g}Hv30Gw7X7p
zYmP?|jchFc74r4xP{`=hkZUi4CRTiEY4H)66`Oarf9~_?Ni{RZ|F-elz97@o=(qfC
zRD2zDuJ(Js(Dm_wqy6Mw)yK{$4jzh2>8-yx;twpETNv}$2Gq{&%jj8b2^Q~8YkrG1
z>v~QABLXANoE|QgmHC;MA7q`R*EFmD2^Bq5F<;pK=vIq7!rcs=p(!V|rxE&2m+^m(
zNx})p@4j32r<x-#!$MSp!0dn3b2g$fJI*t<-;`cL=FfdOH;*aPB?`O&9){ne9t=^d
zH;8cyDUa69aW7Mr_Vu4}Waie~3{4qj10Jr(05_wOxiX$jlzw;sj(sGrip9BzYj$7%
z3TI_4JmM=~4il;Lg6<N!`YV`5!2Neb$(b@hulkEwskMmQhLO#u)h+0zCcnm;M|q<0
zF0?DQ0qFPI+zi$O=F(S!@A;cHl2aT0SxE_aHeJvl@zEj}Mj9HHrNI^<_Od58%A*(X
z*RDmuzO6*$+fL%5W0uH9E+F5(BdN@V<y(f%c-)P45kRd3{hvZ%kpZ$4$DQ)Yfi5v8
zXz4XjrYuT~Fln1h-0@%<`TUfRsHVUd-8X|Gq(_u*PdZ8Q?EM)g%OLz-7ocf({_*So
z81|W_qV6jC=e5q?;T{PaFCiSrj7w!p;;ohsh=w+6Yq&F58pu6SG5_vBocp*Xb!V+u
zH;pj(uWJw5L4Xr8aO?+N$7FYMwmj)6A0dHjO~w((A9Duv*Z|wBxVt7oAm0w(4?ZtN
zl4lxbFWZ&9QeFAU;l-3+7oW(*Eqe+bSxBgF^pvDZP%oXl_&h$<0gP7$lCKq2f8Vn!
zgtuD8JkY!`j&@6LNZfmnUsYSl_n9QTLqmZY=D+kW6xx9{=rI;TRX=i_2YIZu*ZPHU
zQnO&a80m!SnF{Ov(Z}6o{FoZMwGhWX>(G?i9HLqCE#)1y$-ek-6<L;muAxNHPp*Pq
zrL8$%y~`Q=WIT7&HEdPynSBu-rD(7C$L`CKilPJ~4b=&3ymd5~LpF0~;G#yhFs;OY
zVfm--e=UJGF@sGsResdl8||n%K=zG~wKw>oUGs5<s@ifk0_4M|4a?%-Gmb|`5?<&m
z=8**2!lU?)vmW1hlMJjf`o)i<ESbCaO8(BwKj^{@^?d&R7{6h%#2WEq<JHN;%3#BU
z9U9(V{ViCKgyR6=hH3Iy2WZ!Svx?vr*mb%ksH$rf<v?_5j_UWj;o~cMbKEK$Vb0T|
zAN_h)2z%Ezz7cawsE-C~n|*Cn{rZqgQLpRKr5;DW%C;P@tBS$ivh1S6R{W@51v9<~
zo}C{Q8-EKA_;BeBSlhe{A#pkiKS;eKMX~F(>3~|XT?C!UMVt_+?I>-bFM$?G5bGQ|
zR7eB7Fmq4;V^bn>MYAWw`|>CY)`E@K`nl@flaH&@Uw(g-lOW7Ve|cc)GW#HQcZ#DR
zS25<dl;ibrwDHe&*LA!c$E|A~`mU4Ce|MfH^!ZqYCo|t-w+VxN8;E#h*Y0t9_Ok2y
zyMhK3ksb9LBE_upao3}tUHXe#?JV0eulYRm`{%h!sF;t?miWb<MwADgj9Kc_FyEOw
z06y2IW*R^IJ+cN}+FEzplX!n!#Cm5ANXhidrK+1OJrmBpKXpPFJ5ekCAe&HWar~V9
zbN%(7?e{PD685~3uk;2{gvhKH-TieSE(5&S**nX4#K(TLaZ_Wj$araef|zqB`Wf>{
zTb7KGSh)NlPfBU@DObJig&6;&?=+;$(~#`Mq=9F(zI8u{`v_0+r(Kh>2UG6=VE+lq
z(GOOl-mA~|e4#xy@s@}N3=B!z8sKMgvDCZ;K(KCDqkRGvZO-?}7wSy*4hFQN$0pLR
zWU~Qj$jsAuil5F?20E8Gn?*Ej0f!AG|4wTOc9x-Iu46$no@E-RRkPBW|H?X4!~a6V
z-LIc$2h?KOzmo>>Qe7YA2H`ki>0rg%YvO-7_6>{8r^hfyUUeaSBwE+h(vrB;GbfPb
z**{-TOVS{U&&}?j=89cBT*V#l`8!r?b{mZ59y6e<(*K1G|BKZ5zaaMhpVQboDc6F4
z+;2Y4{Z%n3r2^U5ZZEX)v@<!obmckt-S-q3X0#U3YC!nd6D*o?@j6eOaQqP_1in>G
zr-vAXj%UZG<GOpY5czMgw;uaZYWd~TUWxz5DJ}e;3A~F9<xA`x=*wo@$hH-yT4C?p
zHFA+<5a4y!%#@r)@K$`S3ea9TJ$$@dIMx2s@AG)E2~yioz9>&56r~9V4f;4IRmFU}
zDhCa*_Z+lQfIb{FtF;8hboj{)A8tb_fYMz|!Z$->y*wT3j~rMH5in|{6FUF~14alj
zKfveg*BbRZ1o9a>V7c=tJY5(SB~^E}23}*DQEkUzM`Z_>tWhs~oFPGDnI^pRqWyBc
zyYGCkQ9^a10jJ{WcZh79;lA{bY&Lvs|Cd9Wu!5K$vYjGltm}gbzb4%k_>XhTGv)@O
zsI+jKLRdTZIMw*mA#8k`wg&TXNoK4lEr%LVicPhp-1VKqcVOS8$TTo(r3oj=_yO%K
z<F3?V{qWH`^<YdSn?FbC6IF&6!l)IMFvWU_dZGI87{n+{12~8@dK;u5?`bpvV;F_e
zQ8>~_!e7G&C>?7qU+{c7qaC??@5^g^Z^~dyQ%#I4Dy_6LeN6P;F)p05a4qw1&s&78
z)=g#F!HZCcR)w_O7U$07^hgq&d-C9H+BYZvek&DFH01#-042DeOfw(yIPO)t#+(pd
z4xfje4H8kUlnOI(&a;~5!&G2T6A;R;7uRNCc2t`;QnCU?Oi^is`Ax+xe9rTEZjvzR
z3Q6bZ;JYqNZcv;8j5ADuB;KP0XHx^T(@L>TV)L`dXjZ#s!mg_Gr^%>zF*S}H`LB;F
z5ze<iyeXD~GN~t+`8c|V?p<sd)(`?ERH|B`)FlVuyAldy7UjmyE(@}{|G6C;r9He`
zhb&O4&qH?&+O?)*>Mythw2Wo%QYY1bg`Ni<iT{JYQ-)xwr>VbQyIV&C7^lf!+I)=X
zGD$MQOkwk|N7ymo$QrGAUze!d<Md=ZQb{oLaev}J^c;7KL=6ma$kv=35t6*WRK46^
z?Vb4E;`H1`NF2JTA)r)hqlg#YX_ojyV)DE)_2#9Zv&Lzqs$tB;vtMS8(d_#$QbD{Q
zS=S+=bV*>ErTc-Z?6>l@%r4;3-#au$(DtNxz$qCDPJ{}Po#&Q+v_3ZC7?#4njXUzw
zuZB^MBAJbH?|hnpk<5ucFNS*O2jR1bXKP-1!x;-fm}%oA90&Sk$MzZ7wqL2k&&)6<
zrY&j0v?cAl{_<{POou9@AR{!dTD|k-i({(rPFOz06&qD&X4fXCcjqzC+0yj8?Xz4&
z!)eQDM9(}%iI!X+hllElJxqT*T^|B`J@Y5m8H78R{COm3W@Fw(csFO%;eqWF(~wc8
z$wy`w{$1Dm8ks7L(iN(s<)*2zn+y;r<*JCD$4R;yy=5*r;gw!@U{mJZyt&@-SQu4h
z?wIiE#7SNg$&e`4VU)2TJ)1q=cHP)$Se|p24nxCkUsQ9~);T)cHMQK!-g(Pv>T`-?
z?lECokfi&?ye)!qA#8*(ABs9COkr*j6ffhk)q3v|?j)gZm_hOWZIhc>F|%{Qa+ErB
z2*Xrh8^xXFpdhjMpFy)e>;nuNO_Cf|S77jYC%Xu(JtaOEpN-jLRN9k4&gjgWW2WN{
z-Dg*C23n15QsXtszTg-VR^=~B!K2TdXCvmpxaq=@zAQo8^&y4XUwLQsEX;x4VdD9i
zsSML9{J65ojfHr&r%`R|eVTBkuA}ZeHr`BnF|QClAffP@Vr?q;b{hFkjw&!9;+0%z
zl&JCk@je{x@3}m_0kdTu$-s@4?Uz=Ds$8vk-ga(3jAkc{PbB<D!s1-EKPnA5t%(0p
z)-ucJG&v=CB$wKBW?$Cd-m_a(&r=tgDcfeRJa10uDxpwp8usIwS0wGix;>qlRe{S2
zpv*m<n3J`A*_o24MHDXnV>i@*m%*;bSDIMWO;j~S(Uf-EM^tSa5}#VlAL83?owrLX
z)z0t$A~HEQkOQl0{UD<!2g&+NFnkR-P$)o0q0;1OXMmb(<BstB*}3Mf&-jwEmJl7s
zx8I4vNjF_Kb&&&?on6)gxJZr2?ivHha`BxP`Rr3LQnzB!w((p!s2JN~i=YR_nh$<J
zEA@8mY|X84M{~);w)O~t(sKuJXN0T&K-`G{$TI9UouHpqZX?L5H=VSIyp=WP62Opk
zs8xtqbMQJ=C>67h=xjQ(If19-=&)7cQbumR6;Pk{DouObh<L#<b8IU!m2#K>Mfh;H
z@UnZ4rpP_y`H1mZ<N?fR@9o|(G`iBljjp81;#uvR<0p2*u00+wKH$!GP7l}<ZStL3
zhZNzRym^e@+knp1hDyG6I=hDY6S%)_gvE2-*hogwfX=*^o*2HeIp~~7vvZCyz|t?;
zd&G^5%(#FE6!aIaXC`{huejT1w`@FYYC5SfOThBMWmq{q7@R&XwScDqP_F5msOp$t
zootw?38O-X`jr8<oN?3-Q%jdg&lCqkUX+JBIKo2RD*rzATxkWRNO@LMeV7uj7<S=i
z=W(XKr}!m|1t2G;d{2dBLzL5#DJPhew;fw(hcj)y!!yQy`Hs=$2Vb<6!GtU4|HTz1
zH3tt^{55uXUJe8QNWlHX6=s0t=i<c|_j&47WVJlHo~5v9<<99mOuP2Zi7AG*OuHH3
zR?YN7S>2-GzAr0Y9A8bXrYL|EBljFJ4HIasDe>IJiy$k$xhtdn0H5BaBO(EK!|maQ
zFg^qw(DKO~*hy>`-4O}TLZAz0#*EO9BBkoRU;br*5V@{muNexgkiRpfy@L?VP7+%2
z(2YJ5iVT$t(6!=}Z?nYN%_iW#vwXH9z1;y6g;f^~+L<>?Qh2K_Ttaz$ejCqJ>sS1)
zibe)O?yR4V{XAZF{TD~qj^2=;fZ2RhV-xqL3f{rl(WN^J-^O1wHmCDx_#0<A9VVGs
z3EV?S=M!7=FmiERAWozg5!Et3gzCSTOvcse29cx@ibzGNrgf@=Xe0D8eida^0civW
zuPG*F=FqrTBN0ONiFU`w*IXz6KE?+U^k!3X1gF2!V!l`W+$jxMu)!*2aMmpsguf43
zg(XHBKBwTOI^aqvtd>?p<))o;xEUeuX1@YS0ml~F`s*$c9RAbWIMbV)B_}F$c6Ie`
zhOP_?JR@N4F!3~8W=xUn-+r+80RK<dlO!Hsh2<9=Mon$&vM0wvBbHt^d|r~V4AVJ-
zD1!@E@)U{Qx!QpYZTc1skrnBZr8KEj(aD34C^)y($Wq;WaA-Ben<(8xL*n+88C9Ad
z7D_?WwYPn1<@7(ZuwgowYX}`o1r^zr45DvAn`VmRqe#^N&AL|Q-79NAO8M;^R+uAw
ztM}8TAG%Q%CD&fIGu7XyY|!NauA0ngzKo_`>BVa|cR|U@N&$X?Fni4RSCXVo?C}gO
z*c&<QN5^f(4$JpcGkf1Uf1+wIy_XwE;~!NS;O5$k8{<n8w()r29P=?)6!|UiYB;NM
z0edSTJMYXnrk(Nl)1@H5!Cca3a;@rT{BQU6H*Sn5>qd$lTw}6McyXi5b29w6iqNV;
zIvnVHEgVEJ73Do&4~*aCbnO0Y@?9eM1(|b>in#dH)$@xC$C285#Lg-+q>lcGCzC;;
z(r#7oji-P=wXD`(t}KzP@_W#Qjw;+Qejr8Co-+w6$^66|P_on4^-(k;P(sU->6SPs
z_a(FId|N+_Z-g$HrqrOsL!@%-mKNa`-{5P#i}>IutlhQN34Oqpa_0JYaDbdyOg?7a
zBxy|l9nLW`Ys{E28nA1tck}B>);_e>$3n<jx@my#gFX?su`^;~I=J&rJ_8a%L{{Qm
z;*p*q{ys6-HDgxLS>Ks?j%N7lBzYcNEgokh{*MGuC#euziQsqdS%UM4MqBV}sL?$>
z?w{$z@}mGB%1El#Bs=*vS6A`BTpVq8C+<P+R{RL*y-EC#afN)?am)T^gAu<d=Yo&B
zJqrHGCNhwgvNZNyI23}AP=jyS_xop2)bO{9wlGLEW5waA@<E>nFu)&4q9NK5)$!6@
zW?06)xX<4PL{~tKnRTEnz{em;>h}@`7rgOnWTaXDfbOFFYx;JI)~1L6qv4DKlSmiO
zKwZ$vfKByD&q+5^0)VdFdWpJa$$NrkD`gXpzuh~m1L1M^rT2wjBwZOke!<J!Vu#4d
zyiWpNC$112unk^Y(ECxHgZROP;5_AX0B=o5{*q^A+UG>KP1L!|1e_x8TC||amVlpr
z;#Hmot_4l)_jPRzMph$qbtfv!cFf}`7d%@O+%)ZzAzJ3s{YsZ{=uoH~HS;)!YQ%7O
zIt7vjmXi%M^^6O0ZZ94}!Xst<6%M_Bp1H2Vrc4R&t2;Sy?9u#Fgr&DyIyWMSpTBXg
zJx&Sy-gv+G{UPYCDYkMW*{JJ6yUIK_l+L(v)Tgu_9LN=GfB|K0*M&MS1-i(}RZ1+v
zs##mutfv+9xQ2J|VR|=ei<2<)+R^%ZZfU}vch{~>c9uOc9NoCOe{=^=l`>W%&R%Id
z4!i2dT7Uh!wvosxvr?IXB<xS6Xhu`qg$m0iWW6872czM@h?W(g)U)N5=MjBM)-@*7
z)NnukZM@|Z{Rdb#0UXz6uiGG$|J3$z0bJ*~Ftax2P33ya(#DW8^kJXGAEt(XzLW*z
z_Bi5W6Gz!I!#`w|t+!R5Y!;H?2jQ+>({W>3WJIY3x6K95xUZtUWFP&=)aNFNanW(?
zEKzDCEB7JPi1?tZPsfag1ZbVb`|vo3$p6KWZ~d$f;((fHVAR*fTPTc*@#r4jomO`Y
z-(Z5|LA%aWFFv684Ho@cSiI5|O1L`wMuvBb#N&%+-0>2aY?jFxii7CYRn&^M1iT2L
zp<P39sqh!x7CQ(;XI0Mnq9id&b1LF_+R6pPf<2mNX3n%{9%9ubm>QWv9v(j!y8xjZ
zPZ(=`ROR1AaY0YVaK_DLd|PD^l?shA(<Vjpjo9I<4Tu*Jqe2FzO4h-iNp#)Cov9uG
zS2kw^vCrow{GbAV#Brk8L0!-1AMEae8n$~7XE|sOj4_GmEg~$KM803=XQiBBQl{Q#
z_XqyQet;OT@%};4Ed9W%{S0h1|14#D`XP*pVQjivx0AZH{mJTi6UeY`5l|g7ykg;0
zBa)G&qd^&(%YNySE7D)<n(P~9txuonFn`9mT60Fn9PE5Ib>&|;8l3#@(8te1+?dAu
zC{m=}<7I$-Pl(q0lHwP_m22}|{oqO5W%T?zD&?3~&qI5nFtL#**?c9go#`*w`D(Ba
zlu!E;d#vEhHWz-|d>J#1RS+oRQbGu$hR<iPfgLQvZ;S<>zsbOw2xvkVLvnyUl5L?Z
z1xcDT^70v11I)`imHcm8rc|4B^H@Z<+iDm^&`4+FzTOSZFa&33-Acw^yEbL*#)ar~
zvJG#sd=^IOivgN$b9tctR{Uzrd_6+zWYva{-4>p2x9SR%x#NRU(Ddy|4Q13V6xU;q
zOD^z17K~{2SaIhw4ZIY~nc?#7yqjWpmqbS{5hDJk7$$~I;r!&gi|S|Z!J{2Rzt%N_
z#e|tZaIq~hO+Bx>>BGAxnR>FXo6ae-U3do1F$sjX=CqVL*7N$0ogaKx5nM>9>s9*T
zx7q+Z!;?}N8%RguDb~{-`qvFx3%7&P$coK5+@5ZDO`6eU7Wr&Gs%B3C+K{sgQq}RO
zmDktr01i5mDIrv(x$VazLBDSP+TdRylUL<&1OIe=^oDeadG_}>{eFXU`)rD4FS^wL
z;FM}9)@<(nczmJcXNIb$|F$<7$E^(Cm$%$PUyjdF@OKxBNmCW8)BZt{<}Chj!4tON
z2Qx&ho)4};wDWCpq>yi-?pmQ&#*UqurLea={~Un^JkRuVi?1L;%_<uPy%#Ig1i&-O
z8AJEA<glN-$B^hsX~oK|o6EpBg>v$56*ZoH3!=eF#PBsgDpKZT3CGmsYfc1$og;@`
z+qZh|(E4I$(Fe4a)UWK`+|}%%;a9gdGfFGJfm;LznVsxjix_WC8_(x3TgsjE;qT#j
zr>U>z-l>!J)dZH~jRh-BA=&vhegNm{(bJ2M5GdkAunQ}^blO5^&X|2oyO$|fY11r)
zv7wntz)nd#b>}Rb5~u^XuibTkL02zCK>ezvvsHJq75uK$T*BYwuAbagtQZyzb5x0X
ziPEie06`a{1c~_DR-P&WsrC-ee)>e><Zl@2q|Rvzl+tTgW8{IE#7&@<V49+%l<c0k
zbUWd-P}d~pO1?}q-;=<rsGDaZiyAi2bpG-S-{1446S5B#K-=Ztii2+vS~e2d;PctG
zJ;8GEX=;b7UilQ<Huig3oJO@ja_kKfizE1CpWbnD9r^(xk2zZN*j?W4hFX046t<-{
z{hwZdtjjHTUR}%7xByybu)?!5IhTH3xaoOy`vF-qCT_I^AA}#pw_!fT52L3LN`M%A
z6<%S(Tuyj>g=V~ouwNlAxJ>JL(Vy<`4Vxy;zxrr%2>%bwHJ8Au<MRG3DOe5lFP3-?
zKtHN=e?xEB46k4I<sDwO+4$Tr-m~n?)ghBEnSTAQ?H4DQbWGhCuPq!j2MILo=k3Z?
z_)Bv&Rsq0O@0$eH_Wb0Gg9Js>i?lf5ARWy`umWT4MUR>2(u(`l)pW;`!N-8m$n#e(
z;}gf7i@KeO`vtRQ^=?tbt38@E{kJpZ-$j1~ZT6|_Qx?Yn2WI8^dS6ORRTGQbYY}#Q
z^Mm1?lqD-o=Kv{J3_gU?<4er>$#sBTHl5w<twf1>Y#(-v$$Bjt;1gP**v(al8$wUP
zWf0GVjO(t@Ua@y^;PeU;bz_TJi3tz9s7t;`+Di;B7b!mfZH&L;>Mfr%)fYuuIJO;5
zJsBC_wii5C+ce&twJT_CkaeYT?Jq9smQBkg3?AUPvS727>bVV4J3N3@X0JEZv<6;f
zlX`B_`H9YyK^K7KQ_<3jDJ_`%(HM6V<3w2~RQ~RSnMC8v=X&^cMXzpe2M<!r44c>U
z)YA{ifwze1_}TB1dk9`JRb;7njYSmYbfaFyH9n(3n5LA%Yb5vCiSR5dkn7R0lVSq<
zU!k2g+l-cVU_Y;kd!LHr?Ag5<mzR?!su*|RWL3~E=vF5Z?W8X}oqBo5Gc3dt%1+2L
zxNrQPWe8^rVuW7paW!@ivBZo*yB8P!5$W2-yW<t`j?E9~@5%h|Y`*9(+rU7qu1sF9
z=+f^SjB|>toZ`#?mn$6i0+rlW*Y8oacRq9$HcM9LNIahHan$R^5xO_U@!_A5#KNZ!
zmn}VI8$^4o%Xs`^+HX-n*T>lcsRWu-y;a!xg=U2(Uq5e>UFwd1j{j7~QzxP-Uxxi=
zO!dNviSO(ZFblkP-8_so!K(ePjny@WI|9LC(s!OFu&Nt4H2t87d~8QDS}9L5nL4=&
z;?}5L8HsZdVO$GG{F#QcsFkRgG_bl}(2~ygvb0fEu#e+pG{bPQvEZ`jvPzE!(9Mj?
z6QX-`d_~y9%m^yU4U^c*H#I4iMS3*(Rh4aJ*2ZZzb&SxjE?n}QY#N5u%M4X(X6i>>
zcGHBujK4&PgYyPF-<WVY`+Y4R8z7D9<1aEhrb~+aP)NlMJIG6MaMN9pfkJ5Wlq+87
z@wUG*>)D-44H2P~I`u&}UN-SrU_Mnl<)&v!ZwXN50y)T!bsXax#+csPGNKfQ8&fPO
z&0j=g@E@AJ5tOo3br4ehj$WxaqwoglW+<?eYZLeVtJvlh%QlCV;&#=YjsX+w6LIQt
zv*nA$<&HyM_|y~+TZWX<QCeg2M)}|nlng;6Xx2x?1{4+t()-`5450Hrx1b<#)@a2h
zCI$3BOd+1Y_(ZB;`DeMQ)*i``gyS|fKSw-D{?4l_td2a+u_HuwNzRdD_mJPmAc)^r
zw^Z~HJQ_KoiFH5`g!6y*?Fit5rl|}M^C0~}=^eCuc<nu+^(CQIU?_cNKxZNCz4z5k
zP5eK4JoBmD=+VhKY4yolly^PzmQ4C@<{fHFnEdXWh?&k>Wd3FL^tym5Z1Fln3L)W8
z1*|h#IA+7x$E(hQuwyQgh^tx3?C35!)~?5n^SbU$a)snH0>V%S7HFLY73|@^>!B(o
zlhS}K&(gTwcGcb?ZJ!J*8&L%B{N@a5GM4Ppt=zfFgs>!gBqm^+F;djN_J|0%-GsNL
zGqQ09d_+;=^2c4SxYPt3e=lp|1>TXTpm_#t^3)dk%5QM-nVrTA)(RE24R)lgVZ4mr
zvu;-5xoDES5r*1iN8g{$p5y0q$VG@B29aPZ>%QV8zR<Ma!na^{>A%-B|E1IFd4>e|
zX;ZIcL))!;?KJ*#VLZ-bI`c@pcFG|4z%$03wmWLTfTS<_#cocIM>(&#?JZco&Ndm%
zZg<ri6qJ4V)22(uP_tT4nYJ0W1{n2uufn+LYi!3GBo?{_PFjlb!e%}UTblTZvh@%K
zr$aC(*J;Vcx#%J{PAcGmPOkVW_;x9~tH0lu#+M86M2BqY&~2pa&aeq~$LE1ZFgA1l
zA7)*qnDw{`Knzy29vW!nH}UxJgLv|2@6U6MCh;(sJ<DVZY)0ptXx5=|BM-pEN{*IE
zH646T-2aT9uSI^>=MyR4J(j39tX)s<$}V?{qk%>267<7ZhZ-HWwOCS3aobD_a&a>B
z)#roP=npO$h1j0YDg8_9Mb1VV6;gebP?&^W61k|W!DjdTP<7EBexyT=Ejf|DjT8Wj
zw9YN&{a!>5MHV}J3mAvJ_&M_8c(C52PXK(hFL{y9^#@`zD&-Gk+*FG62LtEX_|<9m
z6{6ZMEVJXFf()|)=vbp(D(+(xHF%T`871O5%h`pmgW%kTlI4iM(X-6!p4#p&QC_!G
zws7(n6icIFeM?wF@uy|UWo5FHMe~>)8ns|#A^dfvuw*4CbbZF$RvHgBUXl%p<j>rH
zx5c#vyh0^(b99AXTBy?wFE7TIUi5Z{4bTte(Hu8<%LW8&^rLM&fj%&!j1pUPRkQ7m
zA|w-RuQR8Yik|Z#Ke4OXJAf)<HwiC}lS{tar#lMX0kU!5aI~8(wB{*~wIPkvhZSPy
z36y`Xm$=<=>P~cca?OEYoFTtdjd;#XdLeS4m`wicKxF8t{b8LcE<1r9&GbrU;cQaa
z`fZcPK$FJmOB+e2%%Ysy;BHrXkNqJXfPuW(G^=?@qEV0iDR_j(<bvG~F7A|smquGT
zHVtlS`l1Gb3WQ8%F!!0TMrwH4d(V_ja0#0fM}Y~seH_6Te_=03quze~if4`;*~4nt
z(UoVYxF4fr%aBM&`$rWhL^|?^pY-|mUkUH0!xv$WUODw2HRAa&XwmYMXsS?#P>Z(+
z!lk}=JO9^rwN6!RbC=oM;^+9E&)%eqE8RW7r7B?d;7-E5Tw0OHx|=tk%ET{~#E#&4
z#0%eLOQIqEjn2F&WR9<1V@@Zq#AJYeuxH3%_&sD6@D_He)p`+0O|w*MG~DzKHv08e
zvAr=)>%iSLuP=^n>#%oM7<5-q>9sJ@yGdE4^OG9-4!~YsZTuPBxqWJXx7rgkJsGeJ
zugTO*s|4%Rci5lR3FI^?Xi;t0{}Bhd(=4BdMV7sK60j%KNl>pyU?cyum8J7={SkLe
zI><FaxlggPv7|iL_!|VCFe#lab$rXQg9v9GEw8HfUFPjsf>E+?+QOebJf=0uqW#W)
z$Q_dj4)SlFG;|R=Z(-J{kFpszc6^8t+cP>}tdy>NxhwLpjsGCx+tr(N2i(LBjslf@
z_FMGMn9+L#e&ck$%a;aVyOb^bcYs9Y<Q&?1*wB%tE8P$WNp7xOOUak!tBj9K1zP4k
zq3PI*Sg-R5ZZumdnsu*F4NiG!S8n+-(aW2oTF_6pJSn-gbF`e?a?xLA=mYv(Jxf4R
z3MQM;J7dk!dt(;>ws7SaHfY*Azmd3%aIeI3KV<9yGdI23d6cJC)3NSkKA*<;XCuG5
zFx8IEWZT%y6+Bm*Lc{AP3-9fid-IqZ;}Z6y|G+x=3eK#!AqqgTJlLg)cYW-)&dz6c
zHN88+6fyJhcHH(OPC*tQ<v8;`-Hs9{uo4}eZnti~@_6$+8F!sYUu0M+C7Y>a?WqvQ
z9(?p+B&^guX)u%Eev$?DnG5!wUo)3HZVPsqb?aIh=)ikm_JY(^J-uqc*)Us1kx#+m
zO&y$e_}_`Hp6uHZWWJ%=JhC^u&TaEeAF!p$^;AA}Ti!q5yXAIrApL?9^B$9-kK<Hl
z`_s76JCHm`iAz%{61Zolp(mO9z~g7mEaS*(uhRGGrj#!<$Ljy=vq>bplKDq4XxN}^
zU0?+OjJ5=11ny{`H*AM$60iePqlBt>Gp3U~u#Zl-{2*EBy(g8wxNL{*ic6TZtoJWJ
z+AcFsv+4{15I)XehpV6GN#Vn6Uxj?U>C0yi(2AB!p%Si(phK|`QD83W$G~_jZmNOO
z(lzvY!V<l88M2b^7Ao*ZzeLlzWt(*Uqt<XMxsD<CCjiLPnmHsh2&TWX>7@Pp`B0%N
zBggSRXT#ZV*UNS{7&j{OZ!`O!ce6drQsp0XUYk?X9fbiYP<l$2vAS}PE*}W%|I;m8
zSq^1g6XJO0bndMy^*7mEQzsidJ&#;gheM2SQs?WR^!$8|cWI8zDbpSuJV5IcZ@Vte
z)<T*O4S>kVcsn8D+vXD|0aaxAC(Yr13zd}sPnCvXqj!Hm?Ub=s{QRC)wShon;%^Z&
z59wUM--xrM8^mP2{^KnybNR(D%bV_Vg6<lm00TqQ0*Czvw+*@^iD>3kieqm!Hv;Kq
zHdL8eikqzy$oM0``ZM}XlN%55B&DwmD)SpcNcu(s8}oesKl;V%#yi9dhefLqarJ@m
z&mYF%GZYhfL7cZwa=|6wEn@=vtX9`sn!h|~UdxYnJzP_Wv0X>`gWG}dA1Vg|%=e=h
zE0@Za;y%DWzXv5?Pjv59{CH1btUa!Xq#g3s*er_V9ZL1L`I#eu1;1avJp#2mMa;o}
z^)+`}cj<qy#)G=TCv9;z;Br_Ch7J9dCgd;l<1m+ZO$?iFF+vzKYPIXg8Onohh}%RJ
zJ@4*#lVy&WEQ_|R@ob!Hpct{aD_7V`W~!zs%1^}dT;}%C<Q%$AO8%aSBzbSj%Qa2)
zHI%(V6n}EWsI=exfLWH=S#54_U%2=|2@jZ%BiqE)Py611V!XLty=xXx77e*JksPEJ
zo%zRjH4IHvJj$yklMdriSyu)gP8NO63#cB+O90djgh!>~xDqH~ey;Kt2SuZA;HZw(
z{6i>Yy_9LgPC$d<trVPzsl-f6y&J2)@IuiSwgA+vVJLU{O-(%SdI&z@sC`b>cSaL!
z@G4yxSf$~wi9{)$-{&xYTH{{DbUMEDb?ONjH9#20Kt~Ov(vvRwH~EX>1Da-GBqmGI
z${SOoj!stKyDavm)}WW`K`9$Cf%f{cD7Pj^I5W*H9&o!J9(QA*Ufy_>^jBcWNZ^@q
z0W^|KF&-ZmqVHNYAn9y2_ZS2Dsb;}|d`1==Dla@@Joa4)n0a?3i_-Omv{76#+)oUN
zT(k5(5)OcR0sJpK>$LWpiaT807i&NGA>~QsHiodnHppEr)gG!lK&!8)UCa%j-zkVl
zlaJ59KHzBkGTz3YP3div38i1e+r^FtaN?ex6h2k;(4L76kp~AoIOkTgvg7m(BYX)D
zAL|sF9h<XEz(`6Lb>Am8l!P$VYc8nQbVvk!e(?)*V%l(_8#=%-k*JB3p<7NJ5J6`e
z)5Pxp1{xBqh&@}jWP&AhY-Pk$rE_m3Dls-9C1H}DPlpq_1m{0TJQd2Mxn3(OTIndk
z1B3t(j7t4)1-+KZ?JXNS&E47YEDa)N_Yf;=B@T8b{8j1q_<Qe5%B+V<kb>1<rO&ts
z$G99jnx4U1JIrjrYllO1lCLY~<CuWDb%2&t1!OjXS7hy1ZaN!&!R@8SJxkn*&pj#H
zFb+&Znug~L4Sr#)W-I{ugY}GgMpk+W*~|%GMgdI?wKByGubD4pzJ_43#56LD5h{O|
zgv|h7F_Az0WeO#Oe(*1W_HjcX=orL5+_h<1JnEAw2Z*Qg0=%zOsViv?Wx-iEH*F+e
z>vML!KsDKP7e>onKW68M;>4l*#LzEUyqTq<W%m4aH0=%FS_S+W78=`aUy3(FVzbbr
z;^3gv#rXuhh(bR;jv`)|`xw<F{r5=$U_}9`PF!w7LX~mH-KZ6{4@HQ=nhUI`PqkxP
zAO~Is<A&X1m}P5`SaU-D(E94Q&r1q@i^!1omxhgs-eFA&WwCrmggnl)yLAX>mKDEF
zg*Wv{OWBWOc>wFIKQWP<4aD)@Se`b3V^cpEZp1U<cL37Jhjhp>huT)I<IK^@R#Uw@
z;D+21YUN>cA1G7^H?LF&?lAwtWLe{K-U^$=K=nt9I1c^K`kzCdv@;CXvQU%@j2Tnn
zB_UDt!HnD6>`%RT0K!!%*FMDM4UGhCVmozx_)8ZKl9B$0?dRKP7dGG<C7sN~ozwXa
z%dd_dqc4$qMU$uD$_@3?_n^`%Q8;%HtZq68=cM~}2xft)1Qc|-y64T%D4s0GhsU}^
z1Yx#X8Q5%0Aw#l<(-p$(d?R6SmVw4es$kzAYDtZ1P1-v8sbgFVl|IqzX*g>$ljKP%
z{vb05hHV*XSfkIh=CCJ?yt@5iUax?dY&P!=h7jd^^Zm%SDDzgJzVy$j9VyQ>o@>_?
zig1=5Vk9y6i!~_XO)~YxJb|(buL1zw{kvoAc~>#sL+gFTn{V<Nr`ak4NzK(V;nxV)
z0N(3udA$;V%>IR*+<j|~!3jw(8n*Kqy{6(f%Zu&@O)$xD59n109CDYEOdyU|<jjPB
zv2P0NJaBSfy;`*D`0+Vpn^rq*X?A@*9E26|-)m3ufGXm}BjnhDCnBE8rIh^OIpxxC
z$}~`a4n4orD^Y3Mk~`%k*|n7U2A~KiUSJwQLmjqt-+uyjNou$RAHoswp_4|o|8?e|
zDr)CY9dt5F%<oKMU*ALBY@^G4YJInkfPqH;H3aS%*4{v^<UnCQh|32ry1txcj@v-m
z16y&M7&U|;f{jTIShAk|M;zsAyoF6LP7*2reyyrjy4`#Z^JgKwS*E*<pQf&{Fjetc
zyz@x{`+$0C24FW*aiPS^SJ^Ya2>sb9R1aTxARtu@@e4nc$D*IHV;@`~A)v||3?IW2
z-ccQoZX6qOAQw?aHrD%+ZEwU-U_>fQAafI)-#UTg3y48qQG$5@Q!t#!=$`|h?viwQ
zz2Ti^?f9efoF0g7EwV=Gkp#7=ZYjhh>2b?jpbY8j_G7tkSxn6oc|Y06KkHd{Xf$v}
z5pw$H3xHYJjLC^>?i2(TngK}^mF|$H9~7PGd)GxvpbsQ97s90dHphL{*q<O{XaJLy
z7HDn$r}zK-sN01SQ|g)hx;6;DcIz(p@)_l6b0FxZg!QbSh?no6n87z+Q-YB1c8Yu_
zvIMNPneE)ZAZzySn8Kaww8sWAZsi$}QlKWXJ)T`4R9u$jRFf$<C+U2;oUd}BatoSw
z9pw*&&kz1o7KovEQeNL#!@p3$H>4=l@k2R2WZWOq6LH_k(WAoM??VgZJ0ZU+I=)D8
zIiRP4=@({HaxGLO)d78O`M}FH{cx0j5i}(%Yi3^<@laved2?!|6nqzjU`+vZqT_}|
za@q}d{r!)5FTXaN*?j+--$nTW<g?e)75fkLv7Y$mVVFPNkeqa-`Y5>ZYWw`cNZfs`
z&y1)%I3Fg`z_$fcx_ItB8Hd|V34i_zEo=jM=1>ZkhxNwNGs$sqO5M(nYle58TSKl5
z)dC{4)!VC3C8RL+3rpNn3QcUFp+bOX%IO#;>jIVTVHB<|zXjER@>}8MEu#1ag2mD;
zg!0QQ2k;tHJOH5v&F}YN&M2GT5S-BX%afvtg`8CyL}%g7j18<)qWE3>X=yoLKe^%)
zwO+T5oO`yDtXbW~O21MkZAa6Sk9erEz+XhOJnA(!{KBgYp=EV6Dx~ql@CxnnyQ+sA
zKFjZ*;36<#u{;kjLINOD>&y<DFsehM?o5rn51tM{<$i_WIqrkj1}qr%Z+HgsX(Q=F
zBXb=?D2Yr&cd5czld=rqAXCW4VR8mtg&;~$Nr|?G08DE=5U7E113o6gE%R}?ix!B%
z7QS5hZ;BUv)ednbP|0tWau?xf{{z;}W5>2hk&<P@R7i083R>-D8iJbC%b$$}Q{`)*
zz0=gM*?VB@8=)vGj#jB})vY_YcoTO(4^<p>g{SRS&x9)N;Es;89eG*_AD-sY^_v=c
z-{x}`ApOS-77FjDzpHlHvp#cmU-fC-{P1NWoT%fvaC_((RQniQ54W>IpRRR|?WE&X
z?Q?05$M}DNpYaKSe-<_IpL`29`ZYXXmCiCI)MQf%|3PwFNeD<amt}%_P#-3<!p<{T
ziy$i!zVBJP?$I3V?QP<=@Q4RsL2<OgHs|p!dK@np^JL#EliV^NiR~zE;jR_*n@rJE
zoz)q>%K7M66N2NofKxPv{RfRl_QSz*HD}u6A<t~!Wi$@}b)URU|MRN<j52qQ&3J=U
zJ@o>gp*x~d;gloJv;%8&SLmVUtUdO6Hh&os2Qg|Ld4cwY0@j9lD!~(-5mzoeO>tn4
zPsQJi<?)7ZJc+!{EmVm=ma!2WEQ=O6O5_33-d7ZDf&Xe!x{4#;oVZV{?urO?;dI=)
zLHuN`9BuDRmT!Mu+L=W@cOdTy+sdR-TZDD%YVZt?WIihI9D{6UL_Ed)LOu_WiFj>=
ztszCe>dGMqGwOq?K(ETWF@CbUqP^V2yFGPv{i7-8p<Py3%xvKcIFnN<4Y86YWoy!_
zZZyIl-f))0+O@PIfh~yV@@0eb3Elf>&veZyb~vUh>S=5lkBpEwtIv;<9Qm^lGucsu
zx~91mQeOA|`L4|6>K~vLmv-SyDH~=3p_Cm09DNab`~Lb<`Mdu$+(xC~S(GlsKL4fS
zgPJTT?w$D~ct`RF?T<70uy*Eg-tk8{fB9)K?H3n11aR#uM)Eexs5!Z62k~$g;s(Fp
z*iH&z@8-LPAQAsd%yp*dv;MbC7*R2UhoHKM6k<TkD=%x+_YWNEK^)B)DgEh2&aQGf
zL8EM-IileWiXURlcs2QyIh0>Rrqiu-C4S@k9aN!yfyh_5Qnt7W+mMXl8ZF6l?b$3e
z@+_xWTfCjtlC^y2*?lNI!P`#8nQ&eWldd;!fQ7y&cHTZ!wsiXf4Whb3qxd8Kx+}X^
zJYVI2Wv}}rn5{d(QLRLW%+h<yW5+mN*-5ncpO?}7ER{~{#wYLDxNKPBCR4NE9xz*_
z@poz@-k!!buJl72NY2^<6#!_EsC8J>yek8LJGNV(yWIj+9Z*nOY9wRB6whgdhIi6l
zv7T&@^@lGse<>8ArQO02qPC2u=(WQeo@xox)<R5@jW7iI2Nuk_YOoS&tf>T+Fp<-C
zWFZ^9;^mafr6%!%6J=!u_RM=1N7teqR+MT-TgVqk!9-c-<eO7sSQe}sMwi3hSU2q#
z`cNzUkzHU#WNH&DDVlA1C<W>)pSF<(-A>o`clR+E=1SOV#y?9$ZwSHcnMAZhstd$+
z8rS14%L)~Ctv-RJ+(UHsjZ;jFAvAv9q|g0I4H-PIokn>ZvG4E+%}+}IPQbpg4a=oi
zAm}Kc#s^AU()~6}`SV&-vGV>Rx-tSft8CX8VuZE9U!Xmv1Ph%nL|;4y4Gl(LZ|SpW
z*BnmH`zV?6brt`0Pt|q1N%E~kL0iydPnV2@+6$vyx{V=HlYzb7TB<dG)I`0=)tcB2
z_Ryz?3pl7EA&JtJ9)v4OD$GE&CNG0<#(kK-w`aE`@j_4CO0=m}8FYSaf7w(zDcx0}
z<4z$&o#caq0HlpCIT&X2g|f-^ECEoLEiFDC{~qu8{n16U>~>hab^~nxmi~uVLc=5Y
z%EVjT%CI8_nlo+Em@jvs{AsVtm6>@07o1;e$z_FO7)+NPcIc3Rj<CCr)Hld+NC`O|
z*&P}=q+h~U^Y?r0H1Tg62as$kcnBp!Y>Nyf?n?|TjKM;zm!jLNfHKc(P)AV6`r`(q
z@l8`nx=0SUzB=E8RkkYZ9TR}(`u>qA(b@TXz%w&F`lC0Bb7{y_Xj`p%@xM}vD0ij)
z0Cbpk!BX$)c(WDdE%cG2$q$FScfNf_rt=jGSn-;*y?Dwm4D~n;fC=#vHcyyT!lBG$
zDKfmRG*v$SN<z+etH1{=sjs-Kkgp#eGyDx}adlY=$3Hn`Kl;o^6@?={l=$am>+}R(
z=7X<<PpQaU9w16;8aU7=P7`K(rYJX!@xMkVM}W(aw||!S#5cdFT!LYzEUe6zjg(H1
zf^6y{;GQt6jI`>xea4-!Ek0NwY$oxS6#>^$P8KE6;0stj7r#b@d>^Gyet5xM(UlIm
zQe&{}(JHRHaSBtgt2Oq~GNb@vKGid)MC;$B5A);n%b%_RztxsUPn#tO=;%LmHAv`{
z0h7};IJ<?8pu4vfsRlg0?#Oe&eA@BxDw4LLT@-r1;Cw4`_yG%L=?5jqzG7J!lg8K$
z5>1H$MCIZbXAY8Aarv%a++JsWgp2V=xkKUZb-zQpgzYm6evC0JPPZlx44mD1JZ?M;
zZ-v4lUH!RH0mDf*qjJaJ6ZXEye5C2q$ovC!68Z}dRcp9XI@9)OKHk<*Xh4k&9M2X?
zRLYcGvKbKPbc8NDy7qg-cmG3hqHW{SKd(3O;Ota-%pC(c>XLj!FIHHOPSDUTEy)*u
zr^DRauKmlMB-!X)+1<e}j&>0RB+g6D4`rQG-5+XE2|v0&F#Ppy>GPol!t~N!cf<yv
zcBSf1)t}y-lPiN!Rc-fTtv$+D)%PClwNog|g9qn~iXMFpE>-okn+G=-xvGjNeI%#R
zBBb-q2of)`V_GIxp3PIi+M3dFHX-em?VG5SPAyaUjz=peV#%;*mf`H9xgwqa^a7M<
z>Xy8kfMMQ%X|vK5?{yn_xyr3xnSom=%dW^9!J4JX2u*rPsRCuXeRrdxhc&Zzv{4V-
zkNu{d5UX)Xlz9^fXWG9VR+r2ZU(b6-<8YMjSDlR9uGJxLP_?W3d0!bW(eZ3LXMK#O
zAW>AQfizQ)JCz>E4Sai{Qn$GGM4);PbkLPD!7HTj3Wy5SN)|X!w{r|zVEwcf3&n))
zC?WuwxcsnwN$-&5_YiSv2EGYO(J=VhYj3UEkjs8%-{;(UBi&E?v+wmvO>?AubtZ0H
zqX+H=zJt}@pAatPV>H>Pf%=voZP%qtPnA6M41_`{qWq!We{h5?d2fBb^42iBH6@UL
zqB;ELiNFcZ-;SvJ{r_xXn76(Idx1EgmCT8AKzrW4#kLU9O;6ewH6jOsS9=wIz1y8J
z+VSZZ!P^VmqJEwvLrXJ{Bd~a~)K>*QIRJNjodx_|mOEqdgW89Jx8H7;+i@~}IuE_I
z8h)vER1T9jxgSkA^n)57Gr3>LsV)YSFve{l?)|a6K~Q>gtKnvf2W+tHu<b6zkW+wE
zgLZv2%mD9W4V+}0%}-yKdM2MZY+NO?&>f$_z)3yvr{{B@L}g)^EPbd#sSw8RJuLk(
z=C9Mrv&jjNUbh7&CSqCCBkE=aYm1{f_;`i5lai&qTQZYjCghwl4^H5npjT)@V6$h)
zvHOy?->>Yt5^cIL-3|pnqzW3G_`R2TenrNj`rw4MXLq+$ct!38T>43%66vP+5fw|}
zxqPZ+FCr%;{PC0z5}>0bKm9;MwyGlxkew)Ru-zFbl@(l?S~*reJC3|^&L=Se)oiH9
zX!3jfO#6$)1#9dlfhf8+1=BKl1jR4tN;UJkgOkp}j5$mb#!e#|=`SxE<8xV4baP4!
zV~<r^3;dlS2?MD-#YkzJR7Cd_I+6Q_sz0^X6<_@YP4v3&cF>*lVcg4RHDT?U<7f8$
zrbsh<&btL55y0eTJ#+SYnq9PAEWF||@&1a)w+Hbb1~x;!G7-b@a+_F8wOd-r<9=uk
z(RhNT5Gsksb}NE%1T6jD^|js&8{^r&5U5$YgBJ=5`x6}e)4`o;A6D9r_l}FKjGf1T
zX-g|XJ5P|A(6YBS#o6^b1xL{LHZUm1{@&J%HByo*Xzs#)jy4Kms-*n0GVth$W%?~z
zPbHF9)l`@TjILjddu_#RLcSEKK*kB+3Je|Ccu-%zY8F%?YW*lW_18X)JbPC6uG7Fv
zmqVVj&}Qo>rt6VrrGX-y19F+J80Tu%k6?uj!lWN=oAtb24W{~jqmcs$l_5nDL7_X$
zCSDiZyJ8>QphN=0I>d>t`=eoIS6pDSM*cqJ{En1+uzPWv?ux1&Qf(B&qC~T6711|g
z?dr2^#EOoKdGcb=#%!6FzmMvq2yKe+D;i~x7#k3hK4aeuD9dT?>bz&%m%Yt3Wow`v
ztgCUB)OgK8S(!IftaMOo^fRc$c|V316vK3H%^L_RSk$uMcNIIA_!#o2G}9M%0d_=I
zcJ$s&Q!}F{aEq#UTa=D{-r*CvDdep2tp!2n0BBn9x<}h7>u-Nv&VymNFiwJs-3J@^
z?^-2SERzRc+YNtHPOLltH~eP%DiZA+4+R%n!U%%IhsUAtVOso7in{FH<!8k{h*F%G
z^w++3`Adj8Y@i^?W<>Zdo*gV3sP0I?)o!tEUKF~{b0w)VopsTbQC8yQVhS%1Z5qjI
zO=!{&7Ar-cnK2Erf51YbGje*kHVR!G-8<*!&}wQRj8|xv(tGqqQ^=M}(qj*+f?juo
ze@JrgWy*`-IXZ_xix$59=)%pV$}VSoS}vtj-ljZl69g%d4Esd~gGEI@8x~(zE2pvF
z^1g%A@*N1JeLdn8jbnQgwH~${$Uan`{|Bn;I6sK?5zE@(HQPCtezQ8R%xUmppqzhf
zW6vF-sZ2k*W5GE;+!{BL-N-4CG4y}1_m)vnMen2V3`3|iC?H*eN{MuXA|V}8(nxnV
zGl&WZNQb17B1m`5sFb91Gb0E?Gjt8iJ^b>0-}~*q?x(xnwdTWFYtGsG*?T{|pKxRa
z*7X5!;pjQ<w0(=%6;o1Y51{erB_Ur_`g;l|oqk?tA7dk^vs3$JCyHN+U(WAuGKm`=
z^N}^QqDJ}PCXqX<9gdc!J9J$V^-@#(H{)GgQAe_Ypl@EZ?8nQz$N4fI!qwb=ysT@i
zJzEocO|)gH3fntTOFN_62G%04Jf*FIzWenZQ#r;~vT?bZDq^J;lwNXId}SH11EFFL
zLxgI=y>DVj?U1ZBY$8og2WpEWx#=1`)IuE_e8Wgh7Z=~)OX7|c%6`*$cK+_+{HwJc
zx)GgF&D`OW$OK;x+-s{84Lsz)$;*H{6Ub&Fh<n6gXnwV*14EhP;q{41CT^HsNUf<b
zBU31ONOOA*1Ka(n)*rjYd8yBpVpt1toczbb`JQvu64ht(T?m^5H$1m@F5dktPn#js
zWadvy+tST$WM@<o27J1(9ZPt0rFMz7{>q&$tEtRQs?KfOsrY*NdCz+q3r-HM+q}hl
zukfP6U-Kc(N2|d#Ip5MuY-eOqw^R)AVpyH)`qUsUfCQUDM`oV%4}7CWf??q)1{6GU
zWb+Cy0b(reRIi+cBX6Rf=L-d9z92MaR~l`^YwRWz$i~lBBQYAkFc%m3aLDQ*;h_Z`
z;HSGi6PF+gs^lJha;H0rEF+mUl%PhzO!V7`B{p5M?B_a^Y)kDh#CmK|i5Xl0c5`)2
zGfRU-E~y~dXK!nL4zdfH`5b7ROYX>2J+sGzP@1Fr*2e{_d=hj!;O<1u6CwyaMc@FZ
z*|BBe^cG-bJnO6|Y^D@+`tqi&<Gqnc4}Lwdl<gD}qZ}6dw7!}+m2iF8hUr{~7ikXJ
zOBxV2V5LqAq>B8QnSC|qM-%+ZQ2ZJu@8DlPq8O9n;-nVH>FSo<Z`)*qnP*{_B5Qlp
z`8wzWK^SuY{^$xKhj7BqmVxaOf_cI|>@`!3fDS6p3!E&OCMjuT_Kd%XaLxXeo<_YL
znM^NjN!3KqL|7wQF4@bqKh=5<S%`U{NTg5@_cneY9u0ZPx}Z?!%IsAN<*9K#coA7y
z)WVcb>DsEt8Vcd|au6=J^~?GQX|h(j++zQFvEMLSFT*+d#QMjNz)lbpu5MtK$g(1v
z(SF9PctTy4!R$NHt~8?#0d;XvTtpA5^`D5?`1jwV*MvU*s~O5QxDBpK>SZvm3YX28
z&A#H6xf>v%7qQjI)@tomY1VrTno>BwSq=Rz6upA2|G=(G@okt0&+X(%&E1(W)MmmC
zq4`tapf`bv%&Xq4p-fFbC0~#O1h*skVFO;hQ1=b;kV&~*Uh2!Ds&#(f3ESLg?nzmy
z3}FcAsIB7M8ZYV<ePc8^VQwzV1)GebX%mvX+?)E($5-+4=TayFMW!^SC+dHjSVkL{
zt94VY-s3E+Hwv@tj?_wu^*Q*!xu)LeJ2~-jj?9S$JI#nuuG~kOg@W_)h4&YAS{5{3
z3EXFScI(qUeO)jg$1Hx?{#-<~=9L{zme!Ls84L<$UC>gmYMM~>rxy_-1-TH=fZ$-P
z?_lWQBae5UsDiVR*xI$(8x@iqe0Z8)nh%a8+7^YPKQ6!U$4h_Wx3Fc=n8*umL<?a)
zaCnjA>H<+RAa{J`w}&!}Hym{RXbW+~ZJDJ<EmQ&2)$Gf2)n4)&l7Nb3R=*D2$++kW
zd1x03PdLxn7K0!+&Ci=%SgK19aS4D)Pr*RPEH;=-LlMl8oq_9M8MS*@N|JZ~63AQr
zvnAUrCDH+2_p|<?;3~Va9BvWr!;?tUd;879YaNwKuym)k6SM{g8J^TBj?T-(m0l3|
zK6*A4ei)vvE%#L3>UE#xB?NI!V~DwDX&9z{QbmCHqbR57$*81Wob}7vn*GN*qMvo)
z$@qf$1*7Uyt70&W{BL}7u@7>0be<($xKA_xeD{@5wcs#sVTn-v0hH>L{gv48j!0JO
zJ@zOh`ML^fM%`=%MEM4J!=+cO-j~%<X@*QCk4u-oJ)=}eNGqCS;n~)<2i*57|2YVd
z%xbzybAXo8rP$-%o^c7n`1XnpY|E^&&k>k1C`(3V<oHVp23oglCNKTKVF=OO2hxa_
z(G7+j608ixNESyU+Y>av^+-&$(SYN<1f9cbsL)W6RitNe4*E%r>j|%<E)9T#Lr6}|
z1&BEZHB4012Y6b)dJD~rndC3yIvtr0Z(#ILcD*fCc=zYD%6ab>@u(j}eUGje>G3H`
zr=j%R!?UC=+=Oki&dnuStA&2bQ(AkW(~4)A^xefLuIf+{l|9O2sBk=J+%}XPMGEbQ
zmMtUDtgYVs_p;mT(&%mkP{qNYih|@=3Gqz7$`t_0mFAf+eTm|<JPH21b&lWQGsPU*
zilN<P$*xm5pza7l>bd_G<jUo^d3WqPuvXH2r%TYO8*V5`i#YmFzJxLY%rkrR1Hol2
zj|xLA4i!mn7_O_Z^Jy1^va%9JR<IujY>eSJ%&f31{A%Gjh=VmG7_<z}OjwjBnWyoC
zKqsy=H&ZA}9rKhgH!njG@!@%^j#uZ}eXys|4V`;uf_jp*?Vacr!kX<I-kkGE?H;HY
zIBk|cPEh1qKkq26upm6|v*g+iM3%sqT&%cH6qFa?Lnlb#!KKGQ=o~#GI>&0J7F)c#
z<t9O-+CY2CRhWk-P!J~hrXH~gQ)69_w-9UMROJIqG;cJcBraVKegGw}_Nr1&Cdy3R
zhPcIX*mk1DbDLmt;T!GVCP7x#ekeA@SvFQz@D6x~U}tk^pv$~(5}fzO2QQIkrB{+%
zl3z~r={e0SZxxFNM{#DIDD($&TE)zEpT`q9I1p|IC`!=PDS|FN(xiIe-9cS8b77nY
zi#zkwSS&VDBWcJX=B+#^e5SZj6|=v&&7+Z6VzmT<jfeF-_T_*_ngmDJ(FYJuKns$A
zx-HCpSLwD4;R1gjKy#m7hQsb44wge&2u>+6QKc^QzX0kVR-}IuQwj$k>|^J<5}3K@
zts4I_h^r+LKJNq4k{|US$^l`)-<@62I7!q&gxInSkKMLAUPb?=)7@&rp|3!WSm3iO
zZy^S18Zs_X-H^u#tTK4>8YH)*AGZmiYA;wXA$z!;z)q51CS~DbPtKLaZcQcx_*D|K
z4w`3bsZ;R;<DaT#A?^7RLq9Fv`Vk!Nek2--GJO?oM9{=M3Z*I|9p)yKbTXV9Se6g^
z<Rd0XjAg!fB)q2N3$kPz`R+$w?bxepZFdxeB16$wpT!F^)SvmiG*!w0q5^uxW2V*E
zT7%u5ZNAwOv}Q$l(O2HcP%c!IjGX31Nus!<zHR=Z8aeZRnM<>Dw-Z+y)GrV}6rcW=
z;mnL&+HzUZsrR%p%v*ef(u|H<^7G3S8VYKn$;s>{g42_k%KK~F#_V>O9ac%%4He?(
z@30s8?o>^g>w0;1`qmrX;W7z3Yz7{S5+M@dwTk66O>t}Z%~7QixiN<~$v`T)?q=$~
zB_Vhnvy0&yK#s}tVz<+FVjyy@2V-humyF2qECe(kScwjl>ie&=>YJk?2-2TZT))52
z)q`POo!v^iR_ED4$p~fGiQqE-UtQzAs1HRoG-+BTXNWAAhzK`-WceTmcH5wy9p%_L
z+A>RsuZZsKkJc9zgG~MT3L~tEgblCqJmF_2)TWsTMB(Lk4Bw;Yo}!MU_NIOQ9a#4Z
zsS8~0qyX=jv~~AlwFvhP`{-5d?+!sg-P*(6CoKdp@M2j!zu<zd7{_m<uI?aCodNS<
z5fq0i<4R;4o~1?MxcwC*vmKnXHO2?22OhpH;w%jv{jMl`RpUI@=os7=Rp&AM#e1KU
z9rAthxXn<m&kqk7>2t4|9mW2Y+wS%xMcE_s8>Jb-J|`aPRlv%Nz_4eCI7PQ(p73}|
zhHV${F}1JfqiXy$K7>1@lAZp$AKn+^$j821NSQ$y5nDvr#~?721G6BS@=dL#2d2hm
zMwm?`oqFDXgO)<!ID~y_i%)S8NlHdI-X{vNczfuYU%MvFPU|J(45P&xj`X3eeu%0V
z*#ISdJ(RqPr<53AaK)GzTwGTEG0X}p>N=Er05ShL-X6)!$v=C!L$t4uO0>VcA5w&<
z28<8YPN#6CRh^IE)Au2dfH>ubKPJC2WNw8#@WR#72i&f?l;Ka|jH^lk3ly~=-MBG1
zqU~37f}1!jkmf_sAt=c-59lk(XIF3SDIj2F;Dnwm(f@I=%!>fia~24LF#JnPYqp4~
zlEC?TL`k&V+9nWdxw{*%F3jsRhMP|xR&wqZ60Z_8J39*rd?eLbD}$Ig(``-;FQI@&
z9HpTux$H!xW9UeF#XzqvEyIpAe;tFxIupsb?67&tg_2)Hl**MPNT)x#zw&8kAace(
zdVG}!4#_&tauu~{Nx4T)7Oy=s?#}ZKjf1}_b&VuNlx(-<Xg0(fu_aVesSr^@z})CY
zHF}vuI$?3j0Et4WXmB@M3uf5w&tGoU4Wbf3U!g_!M*FdI;k@asMn9WBp6o{$E%X(0
z#QiLMw+|dl8tQh^$>(K+{Sn~cgSsFw%24^IE%6D2GxyDm7N5lqv=_fchfoACRI4{-
z4Y7BK-%3`8s~d-MT;7`C=mPW|SL$J@kG&TbS6P58o%R^ySL2DkOGry05oBMJ(>v@>
zM5(epkimW@hmw{8qAe6^SkAF=plRwVm>d&zclmAE!G*9<(B13Xz=s(zI!<MApM;f2
zR+-fc-R8!Bcnfga?WDRAUO6_6j{Cf%=a}^>K`XnObO-rSQ`5TPGYdD6NiV}&1~;J-
zD*fU9x}k$uRt#>H*=F<!yD#dO2)7?zo|6R-(~(Gq$3GAsks0tqb7^*&3uQHj53&8S
zjJy90sL^oh6*Jr%zn+e-c5mIIn4h|}AR7G*Y_8?YJ?w@O`Z;k%9(aY=U11>1Q^y}k
zcAQ7>5zSpD05a#^l_-l)#uIa*lg`96stbHm5aT53rt)gf0c`thtE)9!17RR0GD@)l
zI2Jz(HS)uk-<gZ57CJfWNQO^qmAFV@RHbm}hgtZ{?>Wqr3LPbLRKEKwMsV|e9yOo2
zwa$g=4uJw$cqSf}?s;Bl9B1|3uBjA`mO5F_X-w0NKz3wS@r_QREBE=M4z2OIyJsT~
zk1X$2d!b4Z)PY$jt~vaTdY#M{$M>xLcd4VY;X=P)ETG`1klBU5B!Iyd_HM)Wl>EuD
zmXNePC8l2=4bnuw<tP6xf+P6(xYEROzaK_`>-{Kld`_rp*+ivSwN>tpsCvroLA^YQ
zwH7s8uqXH<fg?_m@2wt>3xSkW1eS5nQI)V5M>lfVTm4GOM^|MZ>fSj>rHqYlaob&0
z?rMBgdM5dX&Or`9aQsOVt-_twnedO{*4QaiCPJ*xU=L2v(sA8?8K<IF$w`>fA9iEL
z);nZj1u+b9cGh%Zm&8Q|)FS=V?h4X**oE;rFR3y_A3kYZe$^c&GQI~*Ap{K(g+2Lu
z8p=IFSDBv9!{J2$&B!o5a>?iTl*I_56L%EBWa;BHLJ4ae4M2q?0UUZKq^C5m3?~DB
z+)#;$`VQl!ln9`0!2HeWjBO`6m^=EJ9PiIRinoNDK2!Yq$siQ3CV1{;ak$_(DU1>T
zn?hABY6v%5C!EK$2cF|sy9;>s=sdV4nEa2_smiw9Ks&-kR!E^M=ZdQj4&DO>`OV=F
z(wxIrs5>J5i)!i=*-y>Sd>&y8tasaMNIN&a=hQ;Cb<(?|qLu8aFX>`314iutdFscm
zkq}IH82R}Ft{??(Er`A6oICmhOqE?=>asI-s;jfS)jKL?TNSnKgIltQShofo(qdE+
zlO+d?#eP>veA9_;FVYMbfke5J?hu>&<x-bdQe(N`{x5$w2{K@HiM@evS@?o@Pz2sg
z*b)%!cUB&+7d2Rz$z*A)CM;Q`V2hnfP&t6***c+dwtiRsFVnsh-aIFy)EnlSWhLH_
zwu+5vFIiFwAHDBT4DpaG5Z2KnDqM!8Ps}$HoMIWJaOB*8LN}EiKyiAADPr1ba7lXp
zd{HfWbGtS=1NmjxR|FG_oxWL0N7Ceec+NA<2QXd`#u4S*nT}+L&Nf|1;iCpJuxMdg
zk51;>>2Ct|dmCWCi(e*H0M$a9Mf@v>I+k>+o``A=q_r}PPDO>xb5j*;_6T~TkEGd2
zH6<)|@e4^297Q%`YvX|p5o;vv9l{8&Th;1;cLaNNb6wkFijS)k&U5!HApEunwT~@H
zn}n$MJL$@IX7B~qmxn@Zy2hg8a<mG=xEI;;7Xp~8Aftnwe<P0y#+Vb|EfPiThR-^8
z+MSs_@)~`}z6%Xp>5*kGL+nVj?RX-xK4wxK^IKb5A-61E>e((MTs;Aqo`rJB*ZA%5
zV)*gHiyPY!KD3^sVPdaZB`xgYZYie~P<zH^>`e(+gcXD6?ypl>|81xuICWUoIj^G$
zAes>B^uvR?FSgqi1$Msc@3KUU)TQx%aUsWmh?KUK0|ml`l?F(RpN*qiDBXQXj;Q4C
z;zU$Ny`u)_`kC=$chsxHQVl3*FVR$xZK0Wg@H@dX%Q?Is?OtIGTX_qb^=0iTaKpE{
zafwM}T$1a~HeHNk7e#u}3U6nl;bM3eLqt^*cN`^B^?SH~&m;ZWxuO>}4)vtB;Kn_q
zzF3y}JD{V({%3GPU6bc7{KkG=!a2Bpn3&GOnVf2D$J`mR&gHdH&8a0ew>#<%UJpgq
z9^+G&j+t0%>ZP%LWFE|iJn?w71JDPCxx!t~*hr<8;e?7U38gtWEfCHL4VCL}00VW0
zl2l6k`2~9ovtEC$%2XjgV^rCU*0*g`Sbfjw=pmA2xPaZ{;K?CuODh#Rgo|Aa;Zi86
zZ(H<^tHKQ-NXCVm?(Iq2po$m{K6ZmYdAUf@=XcY<0OziazWq8LL{9yYgxPPGu-;mq
z2=soXxrB0EA&rU&ycu02+6W2`HjpGuD$$iW2qD6Zos(-EL{Xl;5aI41oVjGWKX>C7
zS2P@Xgot12sEd7m>l3?&sL=hY_kC2-tTMOe(^<$fICSV8x}l^Qhn&T3giVpXGn>zl
z^1bixhHSEh#;l6o;++)`cymaXsEP7Zrv|D!vB!ajiE@q6f431Q_Gc|IFhlCH=T4zS
zr{SK3Q#I{~K~F}N{+C;}(+)`>#jHPyUq=;+xufE;SvZbAhmZu!-(>%4KTZgup*}lZ
z{CTM6#uXk9+U4-FRP!x-ix_j-Y5RJW?%Eec8v3D3V$Y_xVq0ZT$-E&M5`5$Ks298M
zZym+iPaGd5=kh22obCCrDg<TId0M0+Wu0KD{~((COypRSiQ1Fy=pkI3%!8!WFHQxB
zN-75YT#W&c%i)5LLzpzf^sRL=Ys5ZyjwMMrYSasL1eo^U=E4PrDw1WHo^1L0I&9xW
zl^shR7JLmQhd)2m%O=ZKFK5pF^%X1L?t|nkw0?n`%$(aktG01tP;;h;8wDZQ*{5b6
zC~B6(Qw9R(auMAkp5YJ$CpdLQy-+Ne*91*?vsD!yUNC>s(GGn~WS7F?A%}@Ord5lx
zm5Pd44Y8R^(U~zl<+O&`wd{7q_Pq_LE}c@8m2&66IIR1%*76i~SYoF@&z++~5Gxt_
zg(M+WVz#=$winSalpEzuo|NAZ&sw8!s?(Z#Zhf?_7afPU%QY)}^4FD7RoCMA-Rhdn
zC+tk7+qbGzQID;k{Be=Q<>9v#4{h!cUXj+cmdM)I6-HI|Pz(A!-jS@Loi;NzWf)lX
zrNz6bm2h9oxyRl<{1ah&CO@KDWMiBxAgp9-`6&_Kk^2P%Z|nZ0ISK1lot8~NEU*u)
zQ8Bo5utnag=M-KZ7d>hUyi2f%+#7X76rSn3V~IU9I<m`-t4lxv-<rLNF%n{)>-D^p
zYOA6k!zU*4TWlUw)rwF)A@?*{0jTNrjY~I`%`fG*1Mq9MU3uFISLtvfmZOx2VZkHz
zD>>HI0wMz49^2ONJV23UCxqlG0#->~_}UZpuBW-HY_6Qg0xjuAOqP;DsvFI9x5bm!
zPThqIf~7D7z04V1n*v1q;Gt4)FGzJFrd+ZinC}_(<&(lL<1`lH@W|Q>kEuTA9ECy~
zw^z{D@WD^(q5it+oFPIZBeEZ@m$c8ONu)K~R)Inv-F2*U;CTw=KA4EUy^xWbcUJ6F
zsAs-;IyFQuPW^A$%?7GunhIZeOOI<rk%9>bU3NidG!T)nMj-z2Yhi3}Pq6m-&?Q2|
zmpPc8(xJ~{-~u)rJGojD!9F(}UUZf)+$;5gjzk<RUjCN$<rCXF+4Kf#j&pKK20C|-
zvyO33fPxcz=o63atp=m}Sz9}Z2sTQwOHe^Ux|)nH>1Um&3+(8y_Qyd(MuwUu4D+KI
zKWn)HmK7%D5?xiIM+^0HRsb<spW!4QZpRTV2t7%X<m2O#w2kLwDzgt54ecbkq%rg?
zilx&JU^~*AT?CnCS|E?Vg&bouvEofW<%9;>(8r6fUPFIbBU+8Q2R*_Rt}=>=KrmYE
zkG673MNiu+LSQJthsF?5Jcim{PQeIXO4rPuXz0s*)zW8Ma&<mZ3j;WV!Ge8vFKNLj
zPj3bb3p3&85KtsF6X;u$05n(s81Lb_#jPog?cZ7@5V`*G><SBE|J_f$e$;_<?(xVm
zYwI`I`}qp~w+CH$P45qHK<I`AezO$UMDKAaPZTvgzH$Z?#;!HBzbU0{HG~-Yw5u?s
zJQNxx<Kovhq!8TyG&-ynO}yXSHY_I8<dk_pOD!ZMw3P3swca2d<L<wXmN6y~^-*Xb
zK6PYt1Si&DgsrtLTjKC~qGrN@z}_x1N9vKqfRv0o=U*c5lE%N*Wjv2`81ZbEgu9Ar
zGN$AH9SiV`Sg|FwES%@%a98dm6*X0Sg|}CdiV$`=tfg3;a+sW>qKKS;>{q-&MG>)$
zBCm@r2*qDq{E9z|m2mLo$OfuU*V@Z!q9ftKm4FL4_x^%xb{FWoaE&MX9HM38Vhk_e
z-0*K>*YJwp<S0$Qd4{9BTBr-4i*`WTZl20yUOXL)GLA_1uhFI#UUx(YIq%H>NWpQh
zlUUp2&E7M%c)NA1nd)Y=!bH%;3lCImr;)SoERMQ}vDrC0q5pME!4z?l`n)k^?Y?xm
zPcWUHz5Aq4o{EHz%X&v>f+$4<_!)#%u7=BwNad>))-X@aM)516WLyF$^*9*T66OBv
z{e(xP_KaMK^v_C$DK^?sOAu?9I-GWJ^_iaSr%&R?X&*w*w!`v6ADf>CLf1p~rn%<$
zX2Qs}Y*HQrUSCJR=a%$QkA{Zn#fnx<4jskx6JPYtwEIZQD|m9@jOty=yuO^G{sUyp
zw8-2Cy{O&u)@c1?|N2}qai(X5Y_bvO$g_B=bv%~br+3l3T_WdyIze5Oim{R@GbLrE
z=qm%}?C;w*uM+80So-66;Y>t3154+vcaU5aXYuxK)HLr?Aw4O(>X=KCj=A%G+l$h2
z8VA<FVPx&8ocUEV9jB4%(MA%J_yzRSHI=haHj7l^FRJopm5PmKQZ~PqQy;h$<kxuX
zOIZ34p1H0!S`(E#QHvWhMr>x#qplz|o07&q+<r;@O{WOAx;xQ-5OeSRnR2K^Qw34Q
zVDb7JHP7m+6s#|r`g>i`?$_H2A>5v9t*Ots&)>Q9-8eI~ZOxFpuRSzkSbiH|KzTN7
zdPD^8rxu<bamlRM*oSR_{F>o!e1MpRP+mT6(QT>bV|<KAZ`l0h<N9LZEcn`*fd8qM
zm|8nSaX;mJpZcK5vtTqzqD{%_aTn_k&4*<q@8izi`wbXa2Ga=L{w1|~aQXIBGq(m4
zkZ-e$zl8jB_Zg9)%yPo}<7b}1<Q^|KYhT<eWzbbJl}f8ri`g9KL$q{5vK$35#GUmo
zr8KA<E!-d#kuJ{8T3UCjH#YB%4=@VM5QUI8wa_Am10!+3Wz+G9BGZr~frTn1OaD}h
z|L@r?D5$27u+=s3S?twqMx#GfZ|GP}o*yQcvn_cFf-JJ`T61aH8$dvVz1^jUiRv7D
zoU;AwmNxK`l+i;aM-VE=`f0rupgARXa?zYnTRe=37AW_Qk{bYw6$E*Zw!u-34<5$k
ztMcp#IPbNK_f-bKFTxxLm#HwR3jGUN7QZVCy(ppgC)9fFsK<xd_|xE%D48j|%5055
z=YbO4$=Q=8S!Wq0m6(myH6FVsjCztLX8T6M*5o;7pG#GYwS#X9k@fe4ElIg2t(o2r
z<EQUknd3{?cCMb+6}XF=h#U=B`mtlIzOdylfC=o~*$JvHPGoPWDE1NzR#`^}okh+a
zHW1nM@BVV{i8c~3Kfn7hnY0EQa67{dv|<3b^OoDvyT7=&iRYwPGE|pZo=xM;tnx~x
zyY;gKQS7xFsNc^RR0=<cK2%>Z{Xt7ndauH5ub$J9Y*a^)`;>~^E-n6T(cD{gIQ_<R
z)JxJKGsL_Ujyr01#ka`j=a!_#7+a@0&O0Ma`@|2m`Hh^nPI*!E^p0Jgq<0cs_DpAZ
zhJkUa=tAfBu%s6?lrCldvdA%qJE<<<XU8xR;Sa+AgOHr-k>4pP3<Ayqgc?)qS$-KV
zhNnd^#GjMLD(ZEKhq-Ao2U2|cAda%x%0l5NYMT>d+`B*X7qFBc7Mm+_q0|x<ey=Ar
zJe3W?38Q%Plbg*rPHe}vAHxeFyM~)sX>O17vGrHf`t+~Q-i<ikn^5D3{}k%0Js|7Q
zc$v^a@)4--pc0GQpnkw1>eK$_ytbziFYld8%g0ToRO`trEB#>;p8dn26uaU(!O=07
z)`YXinKndKm3oesHFVOa=-ae^`=d@<n|xk&j#w*MoC&sTs}C5d4(wMZ<qwHc2j9GN
z6vX2(ad5h`xVtxB-zhKe$@qM;mXgixlO%|h^le?#nvsRl(Z+Tlsat<pYm9>7mqnNi
zCDZ4|dV5x>91wPC-hv3`?I;bZD*Jv}&-G=ea8X~fVJM6+tS@T%f)lt5s=Aw7MR}^#
zyOXhSQ9r0Ej68$C?Y&VIZggi)yy{+j&r?36)EBxYxdbZ7U8u2Y-Lv<hi3XIepLkj=
zbV+M<7(g%w5~#2ot5S4Nl$?rv{loOfz^D@Ad9IiDdUt*i-QdT|?HPxcD7u^m>nX>~
znxiRmsB(lCo<|WyeQdU$lrS7P=CS*trTy_pEbn>Cr;mhu4U{=#Oo`SHE)SxOlUkiV
zWUgG!RPf5~@}1Sj#9VMh4W+#h>2cv7pK}FcOb2i!-)*pl3`{CVeqLd`Y)f;#M`Of=
zXPwXX$4bWnvw*7sl2=la+Njcd@8RK3qlZxYv31Rb-K8Q5KJTS9O}FHfY~LYshqh<N
z+R97sVeXvqc31<VqpwywCNJjpyRPk!heG=O@jw{0QR0yO^Y>A^`)!IO6fXVlEc4&@
z=iuG_tI$0306fuq75P&&M$~J1^ycavCd#-zU+1hop3;(0JnL1z+6O;Pe8Gj#cU$2S
z9Yd<y_to86^sB@ud_5lv!gPR=iMlne(Rp);cHDfNqn*qe_k1;lk{fsKNxs`%PYyK^
zrL1oAdyOEo)>}}H3bM%S>U){`D)j|NR<K!`rpg}l+_w|Ad-H28#TK4h-<IK8rVb<M
z%gQtO^U)A%noa-#Hm>+kTq-?O!Fb~@A2!9)anjYr6dL&XBDHdp(=s$Hcn1aC`b%HE
zw=okPq^b#dg*~|0@jOKC4#TR(t0Qket;7<36M5U^OfRByH<U8djz_?qL`e062eKBq
zV~=>yV)HkpRR$=g>*MaULX+d|EKo<p{?$HrQwM|I+ZeAmPz&MuTSg`B`DW}{@BtSc
zAB~-(Cc;V{`4ZxE6rD<?hl&LAYjL~)4HSG_ZjGLApYpTDrAhNw<k#q2&SKFMb?!>g
zYQ;7{sJJ!i?@y9Pw#n#{tPUIJ%33+D-ya2kE0<WU$$O$y7h`bSpVywq9p&8?KAg%T
z2HwiBTrAM4avEO9Zo`>tX1B%YjY1g3!fwY)pDHZEWd)GARu_$_N&_a6Q)lFR?cVr;
z-<|n)*pBU-M9t&$Ld{GV<MF0M!%Wg({rfg#x=JCHv^?E3)JP-|wP)kBdcd7QXt8R3
z^5Mf?lZ7u01&;%ypD|;Fq8kuT6>~4YT|Vlq%njzBxG<g2d@RA&#jNaonX63=98Uu*
zLqaH}!2N6LdubPCy}O|nYRkrx(HW)fc&D!PXm>b6)9spK`{biU2G~KV59=;l8ZAxj
zaf9|uY=w=`NsWZ5KHB!7zoO6F*N-f^wqQN)EzRoJt|8eLf0waK_^BTv7tXdNQ7wJt
z?}g(xK1Q$DSmoG&_0q94{lTTpp|?*pWAb!1{I5MB-CsgZ?li`HS#0~Z7RARuowXh^
zDE3{ggfbcO!Tx;aFj>*Q^JJfN1<i?sRor(^EQ4oEz&$D|S=_&iKHT%wTldSNJ+DnJ
zk)2kGX#4(F-%;Q;u44td`Jyg&8J-)6=nGUYH~q;qSruJ!$C&tMvxx1ia^0O1y%{}+
zmb{qM?D{fE=dsAbjYf%2es>%SY@oH&%EFInsMe=6IV6kHsc>Rb)C}(zpAZ?|I4RZW
zd{f|8asF^lZG4;2_C(>Fh`m@-cZRk&WUJr(#E`tOgk58Vj>y2O=;@aamn)<Z{bfyk
zQ>w(Jn&J8aOuC-DpJd7!xT94GfS?_JXJ3@ds}bCY;RQ{pyYQH(aRdZp_6H<iPYbdr
zX+P75)PIP+EE~LOG<3%(SpB4FfB3E*#CdN;r|l%=NorWd$$89|j;cKVsUW4J{hc^P
z<xS%;X?wA1!P&TNlMm^`vpkZVX?ioT#wIf*(-3F@liJ4P=?oxQoi<epmSg_Aamx!e
z;j?q$%baR^^;RfKp?tWzat42IcWUpQ_4`9gZtWb%?fj8*OBx8FE?nqp_*eLh8*#CO
zliug??vvRTfetlh$PTQW^vr6E9#wyESFT}kMzy=RvvoA|Zt8L8lZSxjy{&8gY(6H7
zcR&Uq#+Efk=>A|9Z<t84P`EYtLw7036)h7s3pN;0a~JujZ?hvKLFe8uTig4lD1Jf$
zr@5FMZJ&grtoEA~L#Wnv{-_`~K@1orqmFzabu$D%fW_5hVN9m9$((m-Q@XgAiHGGt
zt>>pA)=2R(yveJzLsW^zGq`+Jv><8nZ)$>*;TzhBfG*dwqxS_-jR%{IE8YiAxnc%m
zT7nMSEuo6aYu^;->=9Q2&-H-zLp3rj-_N)m`sF`o{sAF=s~&vm0J(^+O>F+<dyVeu
zjLqxP_b*S6;1hvQWl2ac495*kTreKl5r||y(7%|U)rfho88h2Gx`drqLWK)*^%&Op
z_PzC`AtfREvnPd1kUJlGd3wmdL|ic<uj~pb&d|}GORReH{9*b>C5`wcgxyc?^scRF
z((Qn6Iw{{Z9!UC;*>9+i2FW$@R6^VJ^_f!HhcN<zgD2{*hizAh)6Pvzo{_v}RPzlf
zX9D;Xo21Hny4kg0mz#nC3-@!UE}G4=P;M%QqXCl(MmcDWz`=-_^B)C)n04ZrTDoo(
zPe~p(?&Xg15cS7t8jnRqhp-S^jWbppU9#hYRzCszo{BpzbM+ZA#L=SE^3Pv63Y9gQ
zc0Kd|AOuaya17+}pvGutaS^v=oL{U8b5QbJ0OnIY!+}jqZ3&T@!Ho0g9m3VX%+Gi9
zxhPWNSBmmzeGA&ms%`{mzJ8e#^=Om^EJpQUt^!{csAA;CMmWAoot&#lu=u1L6L!Z#
z6e4sZK@$X29#35#d<aF0Wx7ngKkIx>tg-89O|fdl?mJ|&J!2uiL2g{`hV2pdB`<ud
zDCJ`&ikBsSLQDi|u_3^qJD*tzsae|1oNAU(lET*Q8XBGMB~sa?+?K#&!SiGKett)5
zIWvZ)3hwl7ETo%Tb7g37c)4{m@8?N8Vc2Kdw}XTC1j=W5wXA%Q>hb-C0lfK1yN$>9
zOA782lG2B|BgN6L8+U~5I^@Nvi9o_FEw^##EhwIr+nz))Y|}xS1AAjhDxc}YXHsA9
z8w>*TgGnN-_9N=_Hjnu%ai@649|Y`nukN@YLn-D?25j=@qvdR=&us!R2?YUvyP+!@
zc|a5f;_an^kR1M(I#<!vMg~~i$6%}2H`t}1ZVkc%;zMisPa!X{rhn`B-7Z4?zA-r#
zY?~>fas5#*!s1McLo(M+;bcH~@0CP{f8!D^=3_7#KI{J4C<o3BU(guvSyU&Hca260
z*;FY#F_<}Rbk{SE!vpqiIUZ*5@#jI7t0^3I?05C_!jAi-?6Nccx2W~{m8cW{yJ6Ji
za?B7ee+>5kds>agTx#^sjF4yLZR3ai-hA@KH!!Hk+U9FZ?XlB}1k72_*p=-Vyw*Sj
zvmd}@5Vh)1lxq3%Tn$e&3(38D-=*kwrc9sh4k$DYTZ0}q;@zje*F>CDy|PkzZlhi-
z$i*$hH73o)jTRJ?$lLoBsIER>AL1puOier^))Cdd95XGIJD5!=@Xczhq98Ef;^utg
zrH>mjjIeRXXdu`5Dpd?Q_F-7Nha4tGxm~ZgHB#H`bG!DuF>6#y=$F86d&c{)uc{<~
zB+@Eq;$!IwJsR~N@Bh5uP^8^{yI|$w{nY%^LQrNBzbeNH6C(X%#{qnc3-uH+mvI4-
zZv;}B?ClLS#q_sSbvyXn=}hN$$_}PIzmCz}+t{^w%&Z)&i>IdB8hZ``9A)-tVFJ1i
z*s4`eccPu_pBId>k?02NAqmvxwE6&*RBUN2E&a;Pa$1T_f}S9>GmGJc?8-?6H&0Gw
z=%s40o1&<h<>E~j$(%V_i(cPIrk_ux@`fg~4nK`S-Oba?07n4e&Q%`#B_+8J-L&Rd
zHBVppfS-r2ng7_xsKYWaVOn2fL+mFF4!vPuCw4>I%|BKzy7D4=WlLEnXS4D6N3(fb
z<@Vw9vn){`c&d;a^p>0^&vZ+T`SA0^$;y#X;%T|+gwSQR8Ig|n+{cUqX)?~!btl@)
zNN;h5<J@j0DTAtS4RKXuO<$Q-D!C(CsjxQ%rGi`gzfACnB-0g!`G^aGe!stP(AwHD
zz`|3E{&?I^jjQib>*>fW2w{>Q8p!gTiW#e)+D<O=wte=#W45lTzLX#aJ;{8r8Il{F
zIj6M!9xE&L=U(Zf4IoxlcOm-N$6BbrQiln#Iy-XA-MNMRX?KRT^QO2W!mvLi{a^qw
zZznyaN))OVH#F@n8Fg*f!K_PSOEyp4$4|fT*$Z9Vyq)(Mdw@St3*w3*JKq0-Gq;p_
z)e64SDl`uioM1>F?{I4$HmMvdX!tI&h`RJo_CF66tPD%*3M0zNoO$TmSJj1UK4-<~
zJ`C;~XI>c!w|;bKfZ}FsqNr@|)g%rt(KWuQ-D<B@5PlWp`Z@@5Lr8-$W8eIx@kjpU
zx7ti4{gp3#9;U>HDj#4vELLqUa0%qMA-9DFjri5=az1A3+;iu4SwFl`8HJsZRX^D{
zad4~AS1PA(Y9rlUC_2q^1zkC&?Xs!Tvt8ykR?nF~;icSCY#)GV7B@4(oWE$U*Cc-b
z`0T#w&n0$iP@RWwSl+w%J&Pp$M_1&L2hrgh&f)%eV%kuPsMu*9)K5fzxcyk};n!}U
zWg_oWCY*98fxMUo$Y1zude#z75nvu1Gp55$CG;Sx&r!BBjY@4_?D1@%KP?^+HK02M
z78Jy05Y*-D3~p>2I!E1`JUMP*6%wzVBKWOMS9A)`&NzbSjA&{!-Y|Mr7oik*C+~=Y
zhC0hfRx^Yp_G?$CENJRCZd+j)=h4J`D#32Hq6J(8zl2>xPZ-BUYpf2lNp5mSk3+X-
zU%I9&$#wb;HFx?i+_$)>;JlKnx5edy8{S1-%4ofudupQ?RG|aEvQRmUyNtT3yuRfn
z1ChLY(MhEQ)FhXT#U*hLE|x?ehPe|gs*vdBnMukikUS3spW<H%g%6$zs5<cwfxgC3
z2TlclTO3tX)pb8xkZuD$D^$X|F}rAiSO&?XX1>_yAJ?rDU9lt=UJ2<V2M7NaFx*S0
z?4JtW_B@vpdB+)-Sr%*t@{I*Hd^~Co+O!2Z{w<e-10t{s!-VzskQy+Xz=bkcZ4}nj
z9rA(8E7f(^syP=Q8)lxS;3b%dtgT8N82~rMZUjyLa#($8p3U-B=1kCuh5`q)ja-kP
z>0U^q3=|~32<HjXPcz^|^&p6%va9LIx^c&n(Y-~c)1+J5I>tecHxn;9EM@RF*<zh!
zc<<Sn(eB_+9)02c1_Tpi0cj+VY~?zj1p?h*yM=-7LXYVCP&OOH`_bKGL<4hbQF_}I
zdHtor30fIY1j#?EMIlmvdLL5Jcfpk#I?aSL1^QF=+wYVLmUO>v{<+i6r#Xyen>fNK
zUq7`0pr4w!$BYR@>Y9<qA2o>6TLHjb2z}i=fF3r8%N$q%A1z$J8uA8s_1hv%iT%|(
z;Hmx}Pq`aG0ru`O9HoQ&1?TSAx7RPa<hL+SUVrP!IN^(tzaG5ua|?6F_}tRr08wd0
z<jF~cc!6t*9*BuJgt{qldi~6|$Q!50dvyBP!|5aj=wl?=2^=xdfQ$ZafiwtXZb^&b
zMH#T~n*R|*5g>w4p+H(3P#B$J8GhX{%kfq0J_z9q)A}t=J{5+_gQfd_Qv@n*k6mLC
zP;5U@unz+24<8hgw&R{)$hdB&Vuww?{xU9BhZ)qA-FuOxagdP*uGrA@<W$wArNiC3
zo}9xbg$Q(oC(d+gw&9P5CG=p!q%ZKM02AN;_rwFczTlv>R)li69QcaI0~&_QqYQ)&
zgF0@6G}FEX>YOuK{bwd~FbMM!5fk16vqzs?l2y^22BMDSH#A0oDI@<qrB#X{=ypdc
za4S@V$i*gbs{ZW%*G>c#bl`n*VZq>=V}~b>M~A5q73Oyedm5&&9pAW%)?_x>jL5CT
z_S&`SCW5q~f>(`^Nbm1KHzuD}F~2@yB^ASUKW%32zB|~k@*xv>38yHJeDw9`6SK*U
zYzUS&kr366Xbl&G(1K!u4C}&AE)-|q;D;3V*sd3)<6T4tHIW+qzCONagk3(fD1;k8
z6q#C3Q|gz<OU2whouda2k(9fep2osKvA7&>!jUgpicUoZK`*m&C1RSURS^XU%4^if
z1}b86T0%UGu91QkBo!TGM8r6HJ+NMss>00xe4~ZAB^9UIDGH{=n@yA%Q`di8l$xSL
zYOX<nmgti2Zg?68*;0DY>JVMP)c!HU-RUReKe<qE7HO0VUI}8@9)2$!5d>oaEmHnl
zuqfuSsg;)~M$$5{l6cfcnfl1X=lTKYqA1=vMX+8ixODMnMo326iDMb+&!`nL9OqQx
z{|7&xc@kibGJfnj<OQ>57UI!gBA*7EB`{09{Sv9egrE3sp&8*^O?t$*Mf{Pd4X>%`
zdaQ`3CV#u@(R(P49(_%Y7>C5bq@szjJg`RUS{+To4;=JpEy6`S(@eP?W)e%FY0~3=
zqLL~Hs=f2srCL@9cBWG|hup9H|AE1raj;gHK>cYpIUc$1Ctd)xrT&fDBF_VVkrC|x
z%IwL?bzS|4U;jjx(Ho~!m`7Ege)ZTR$iw?34cLLEJ&u6XxTXjt#F-))iUlnvW^$PT
zF43I<%BYz=$N>-=|KHFo@geX%t>nP-H$n(bc`>#qZbZO3b}mToy4RM{zsKo(h!Trl
zxx9#qxm=<nssJq#l&`-4-&56{yZbK^;@<QHpCf<^VFV{o>lEl<oCmPlYts84{R%(V
z+z1f!fK)LAfKHJJN+D@?W{-OwuZVZ+K}}4j7~yyc+Eex_(k7x)8B0&%UBQ>N--sM4
z_|x!qIb_E~QEG_0W71(pn`{tsNC@pTSQ<a%RVJF`xy&m-u2xdpE|7$f1?&s&j`Z0x
zOa`s_v;%d{#*inr-MEntxNqHIblE`;A|L9{o8MJ4$6o60&Jf<n*&)9BMpPbdhWlwb
z9x;Xcekwp3^%lV=&)(4_cO!%<19S{7(K?t{+ehb2|F{ShhYkUHH1ug$7!wAKfUm%j
z5MIYWp(Y~fL9maT)}(tMfh?Fy=cRsOFzt9CwrcVzVTderU>lK#$3k?Bw~PoBiDyX(
z>fJwxcKD6GyXW4<2j7KH1jZZ4&H*o%Y-;G*<Xm-B+G1H2PLWPqk>5)r4VI0{@|@{5
zzfTJs<+{RtuXiLUI6w~Pj^s#`WCAW(VsWFOGv6ihH9nD(yd)&_X%3nfY>w|zRF8n_
zT%9U?hZ5{{?Ncwslwk6)641rJ7X#fc8{o{!7#1pUh4d^`c<)}2z@DuU-D*&P?2+70
zwoY2yQ~eF=gpJSl$GD@7Dc%+-o>6<4u3rIAKAL*Uy{R#JD0vxbVTWbeK1DtUp!^zX
zuyhTj(f=39F#t+U07?rpu4T9|DC|jy<COP<y~r2L#;vi>K>{D_@Nkb&o+t)kvMYFI
zURfg+eW8dk9q$YtR4O|H4Ni4MpEV;(y?3-eo<9>SKLTMsV?)|kY*>B`GbQ0<RgN$-
z*p3&z%uw^N9eDbV^GYqLw~SBCBBw?raP$Pr@I83yG<n1e@^78h269tXlM&Sx8!~gC
z&&i=|5u=1_C7VzHEimVDbU313qe|jWxxo}bX5NtNkb6H)3(#`gu&0W|y9ZjGH+BU8
zF(bCxI5j1465o0M&quyCdma-2Lp%8255L$I1X|)G^}aQ{IJGJxf^h>GIvd$9vyeN3
z?nI@`xg;*1_s{y6d)?y&7N!9FTn$=#)vlUpUIO1G?uITyd8%0m%a~1~ZN7z7SbZb1
zfsY3_fS+(*JRvM1{jD{HdnA5@I)VZja8E4>1o!>7h2sOc2jj4CN$C7~CofE?0M{Bm
zN8i@{_UvF$_-AnaOQ9ps*i@H@;?EM}pC;tm#vA^8)mTbPS<EUnkcCgiFJa@4HhYha
z%rI-;d0Ekg$(hSfQ}Ud0G1c0!irwHxR}8qP|2s@@7swmG!?z?s44-KK3ln+5Svx=M
zyzg@-%kNTz0BAEjeP~Z)3WU3bm91XyuOW=j(%SlaDIamr1%!in^J)prnvv5|k-~~P
z9nJ*6W#KFo{WyLB8NZB8u=UNg2S6Zq2VC_bM=y@}Gy;oOD2wc&m-fGl8CKkB8AhMf
z@6WtzW;O{8k(+oc>=FymEDJE+AVA-3K@Ot-pznI%Zx^|V2;p)U2>u+Xz5NP6QKCv*
z#2yqT4*FEHT*{n^W2~|#t4s&uG;JGC)^BIQwp={Iq~T`Lk^91%0kekW#8Jb36dul3
zKuYSI7<F_16<q-I>eT96<c|=P7VIailJX2P1%l&YX@FwZ`hj#Cf2{NW1Pf!lEcL>m
zJbA7eo1Z6oPV~%Fey#?@0!pa>qgF6ZPMG1sN_^w9IU|qPxadXmyZ?gvp+^)nO$b{w
z$NfZ2S-``p{v8YOFTRv;h8T<1sXr3FD#&zfo~ooQ2v7zvz7FfI5QZ~kXq&uwc81LP
zytO^<A|t=tVx(wmlMVkQX)zn}kb(V9b3ZWEEta8S3tev<UcTTXO&1E`-XjlAjBkDj
zIS4(K?fwtisXIRnD<^iGuG_p=JaxwkIEx<Bo{&1psMGeCT$%!?^i8&bWmAFP06<Tv
z^CyJuq`zY}aZ4Qv<^R}U3wp~+)~he(V<@bT#j!GFI4<3JCyyV8G`Ds%lIf~I$;pK#
zN?8{N!Mc3~1>jJYBg6UZlsue8(FMSjKaEnHYet}Rlk#wv*c0I3Nlty2aF?Q}Z~96V
z&i^@h_#|-ftNWquSn_V%?`tku7HRhnE@^bV=f}=6DnI)~<ulYQdaLTdEuoj!48R;m
zw2Xs&Sr~j-V0#BB&QMN>cK_wBmC@y#_XLYOXJE(N=}E(4E75SIZqQI_=ly_oTzbh$
zqyIoYuh9A{8Fh-l-JOye^-`CoVZlHogH?It(h-{%+hi?2JXAMK%Haqzg}{Mjm3Td|
zoO`<#TK_?Tul$8pHva(i7a)L+f1)Oy)_Rst42zT>{ob&NyVMN^BvAwSeE0mC2LR?k
zH`EbS&Z#tbRs8+T9*u@Jd_P-P3f5vzYn@;Q9GvJ1?8ynxm7d|OS!knekv3Z3-91vd
z!L7l1X3O)I(;GU+&wPh+`o@dYfaB^Sf%R`@L&Rm|a2>#wevM3DJ!x;;5h`~2533ZJ
z{Iwgr41K!?Kl<NTCEq_;<s&**SXe7SMWz2bW!5Q*g3)@5X*Jm9fp?ixUP{)3wF#kx
zefUHt%br+Nk-}S{eKgjWcz5h;^EgD^g0@HR0W|Q?i;|LrdZIO*+*t+`A}xp8$h@Eh
z8~#PVr(6L*%Zd&O7_9jEU&#K!#Oq^mcM525FQYFX86Ew8&ggvPP3kN|cTu!hOG+U*
zL+d3SG4lUn<V9dF-o#+%*G6Fnt%{b5&#^k~UYY2IU=WC%$cX**_c<6jMKgS7IB)KO
zcPUXuGk;i2UwI@s2>P6tSRFAuaV@uhK1MZQ1(5_{<fm%Z#twaX7NR$fs)XV0#F$ja
z53{C$-$EJy3~(FID(t=%h5e^;Eqm|@VE=T;YyZi`%dWXN3&1foPu`L|Vfcisz<tJT
ze&4*dPYHiJSwk|VqAA-WqekjJdbRsn19#ols5!+jL}a*Ju@rRvpSn-3>XuJO*0jRq
za%WAW5<tQ<pj}V}m+V{t0#|S~%D;vS5`T9(AjGs=dcb@QQ(2_j#J^zrr+5a)1F+C%
zy069Gqls(r#{!UU2_RWMlm2#}!u>}6OS1eAwYvYAF8F`bSL1SFMAJmPo4yKwEjV>;
z3)0H~BrG3y)B<$Ko(-{7s3c}nO={@_l>5;&!7DY0lR1#(=E%xOwguY3cjI4x6F1##
zBFxxQXa+_=|7V2WYXg;&v@Y*&Bw$Q4UF+>ifOm;ryBPG}!+JNJ94<%r81)h13Z#6+
zQ!!j}TC(HLVzCl%g(EI`^lBwgjD!H_Ld(#E*Tf1PC`A}X-Fr(8+AZ>wtU>>c2MS3N
zU6F~qRj>ekaG3zy<h)Ma2iBml%lR5hyg&C1@OoI{%FaSrAm(_y!+8hD{fihv#LEMu
zoZ?^N5HU=NB8S!M8^^)nazu*+^Qm%I<QpmIpJ&Z;*JkxQ2VqW>?e-}miSydH8NKoN
zN-atF^?O2u9ja6WO*TB--n@FHL<-y0mHCSEdMI>5^T7g~Zdx^VC;d)b!U?4i)h)Gf
z7RWMUV^ubdQ#JD)V1g8m*j&Sje}@!iQRZDSL5MgLiGTSKiQAK`ymap@ht-sYN0yRZ
zy&XH43JPu~2X^1Y`d=$6qI?XWB1jnb?-+SEWn_48@hU7>FQ;dxK=QW$C|Bt(X^34;
z4@)*OS0)_~Xv+KWKf4YFa`V2{f9u(WbHdx{nnE&rm`2~$s3ZwLyf)|Jm;wD3uRF*L
zr218dmk7&aZ6{x_{7vS+hFL++seo@d1eA0>XpSPYUj7s`NpW^7e%&M*IP{Whlj|C7
zN{Jak3FGgBvi5y`u;7(YWRaxufjU4~(sqll_f<!~04iY8wV6Je;6V;j9h^=_c4Rr=
zhmcN>em}Wj#7J7-MF~(TKpz0*%Wj~K5R+sbfS2VTd;b3VBToWYp`8m*jWf0nCsR#Q
znM{S@LQN=s!aYS9hWEq9!l9d(3$-N9YfG}A_n+42e{t>q0M%aNwiW`LUcCrh!|mo1
zWj43k=GsS*klncd6kkDJUhy*}fj+BeOWzBdilD(^hXDbdys}@ZoJw)j4|VWg(IhCy
z>)U-M0DnZ^%`E*~5{e5b+cLWtmnhNHX!V;oIgmc`*{%0~(eA71-yXbMKiT?!0OJ3V
zt^5Da`+wsOY{b2ABER-<5B!7TfoAfm99-aGJe;W?D<TLX2NDAT0pTz`*kzP>Pc6L%
zp2t0)ow>M&qz7x&5Aop{u+jJrn{6^#ExZoA5T@yth$%<Ik~~lPN93pX_Pl}M4Zm$)
z?$MoHPj>I{VbBzw0MJt%Xe|sGn+wrUrMF#$9>C8m02y$G#JU5`>M83a62Pr#j6FK_
z;@!cfWaZO;16#oYT9~%YXzDllfyr$am(`j?d5*~W&8@lbn{{Q$!%dQjh@Rk~0PfiB
zD`xH*DbM`Ji_VMB?75wigx?R~uBQNa888nLE1=4;7SWl$rVmePbyTK@{{~bwuomqs
zTrXfT2$A$da0=p2GaJ@3or@ND?3urDh>sc+&do^i#QVKtZwhWI_88FUya5ZDBf2&0
zd~zb5X!uC<45=Esf4@P3N<PWN{z?p&g}di?n1LGvCI%!99L#@T;@z0h)w3(&w7l4I
zP)JO-z<J2G9|GvvVS%kV#(;;7(2F0zxfagwq!kg0EE$uWng&Ua5WeDIx~y0I;$Vjp
zp^lF$o&w;j{K;I{@JI&gfcwMsO7~Bi(b8&I^G!0ow$;yk9mSl#fIS47SdHF)3_vl6
zOA@yHU>TUxP!c6Dss$mA5?BT5qg7*L{79{xCS??J>UhHaABR!Be0#bSqU9<zVlvqD
zOfj_G4G29H*kdBT5Y69H3oD!Nut3j>*_Ts$U7+tA^L?p@*8lup?EO_hRBzNa4o|=!
z-5}i{QX;9eASoa{bSlyy2uRH!iXhS{sWeD;$|y*PfCADnqS7&f^f3F|{NCqz55Dg^
z{2%_$_Q}kB?>nxw*0t7N*UVPS1J7}ZbE3D4W5G<rzr+m=b2NYLZRxj%U(W2Y$L=jI
zuLfX$fxy*tlAR*G4sHWrkOc@-WdF~TjBzfk(D5c^meKjSZXff3TKk1H)Sh><hJ(`P
z&b>F4^VLT)iR*ddbL~fKCG9ON`UxJTZEd6MPOp%QFCTeTPLp9CwC{wHpEEPJ6bxTn
z7!)eAOYa$tP7^}8kB`%1I787YO*GCT`S8(EuQ3SK`nX4H(c)MZ`oe#{=lZrina$UI
z)XPqKVE6W^l7UOk!ZK>!1rWgGH}|xL81v{vje=|GlL?;9%HjWgy1s3DZ{fXSx>lpa
z0H8N%{9aF^Hl$I@?_FLigPn)BBXB-{@r}4(uzS^q?A-3ne2~<z7W5vULCB`$vYp5K
zdC}vprx<`gB%7J8c~AM?nhb5(_mhM5<16Sum0y+8{TG#j17TvVfU8)))OYa44E*8r
z&)Eh-cAt#B`Lv}w&MgIQ>-+=onrvQf@PAKaBKeTq&--?ZJ6ChkdNjf+gs?j=!uQ@8
z@s={S+*GP+Ya2d%o#!<%LXW0ItD{pN*9J-5=p3q7m}ukvI(~<pS?1}IGCjLJN>SsS
z9HMqA1lft`KO`IQoc!Gk2(|)2^nEwdmz@<JLJWV~o%h;?*y>!*0dLZ>IC!hfGXL)r
z@-NZb8TeV$!Uy8lc^R4Z%@@x<eb9H`3B5<x>&a^2Xa9|#Dn{7V{)<we_v3r-2_(*`
zKOBFtcM3UMJTKTL!an~bX$V<^9W&4Efy&Vs)3or*E4R>8hguY@z4*RBuojAOyxucV
zVFi`Hb=)yFe)j3k|D3D9H{iMIxI9<?J$>32g6L&$j~SxQ?xtR#zU%-6FTw|==Pa_Z
z7iF{)pZD#twO&g27LI$9si7^nC3^|><Caf^TBuKz5}?bPx}3XkurFi)-1!x2^0#;k
zE32YjCIk_o<&(%(GWYrkmg)D8ke%PY1$}gQEN><PXZ~&HL+i}U0XkNf`!fPKgY%(O
z_z!PXGEVKszYU{|B-|TRP%vNf)C>ANh>h;ALa~vssD6oEYOsApCO@{8OmfnLx<>mK
z!{AC<FJsJx+u_BsNSF7;_pc@GYZq-JGYU%3U~q7g@ajMnXFoD5LO=h!wf$nc>FRJW
zxGHlQ{YDgq6K(gC_$lH44ZS>wmvCeF4~GEFRaE?Yd1BphWFxv8&1K;GRT4wuKd%D&
zFx#7zJ%k2$LqfL*9Y<>kn6EaWl137IrMy!dpZ8s|3^hi6Gp9a1WLV+g2VZ=Rd|<?}
zD$EwsY5n>^*Ru0_y85JVT#=L>Ud;3+mOeQv9-u$>zu*({f3P>&z|WCR$9c!}{Gib9
z$j4B0!8nM#o-9<N%D{Q=0w88$PB>+(L+$KYPZC50OEa^(j!fEn#>$nl<BP9lC!e2x
zVoInj{Hy%ypl_@^;&HPw?&T-{1Et;8NrLC`@r-*_8PsQ_IUIIk(Xa9`nLqvkp&;u&
zsP7%bt4RK5UUYxF7e?Fjlp4t7aJC?l<+sK{p~%i|oOJ^tzWYb5`{S<?{^w)m%{%vl
zm!|adk53l}A!A3YFV?Eff(^hz)g54NQibmRF0X1{M8Ch#j9i|)WBDG+q`G@?`Iifw
z@59}vdgySguo`}|^i!2ApsXLXG(R2^s8@jsu=};~=fh1RtP!?PaKI2=Q}{P7?3l?p
zk2p;5IP1$~yUOa)o?wQAqEt6ikaB*J|9Spc^YYjI3O5m&?$3Ju=Z0UEB<vM%3>gxY
zQgK<bS*RZT%1t&#<zE^?nh)|J0sAaGOnVi3*1GqY_4~@vKC4IF_a>dHJl2k{&;vHh
z#=%QM?tapt^6-1(bvG_y!x01!_y3^e^6AyqXHpEJd8a#@9Yqsw-xB9uJHKs|ODEro
zTUaSnq@R6{TL^h9+_Grsyx7B+!`>Lv-&a6ebaiz#-{PJoOMLg-p;mnjHvh#1eYK+f
zDk-E1VBv8B_?%HSpVuOaab4gaPkh<)D=hH%`#hjfs}X>wnBE1AGO81FIZSQ;?~`LU
z_>os>OSf+7FY1y*iTPTl{P`hLumVQtX-_C}sVMjOyH5Twk+VgtA*RQ0zbiFXBg>+M
z9>h-PMNd}Xf`QAv=#*hRFe35Sg^jDhzEwU%o7MdP{{8<B#Cl64jRP%>O@P%1hI}R&
zLeYqHpxz*VgtHix;XJXvPT#Y%vK99<oJ@O{;c_(OV~Qj;+PT|RV>4zeQdkttlT8AB
zTkTQAF?5-iNhd^3GUaXM?TgoA5EQhFx$W6WyRv?+^!B}bpNF#*YZsTDrTjr9-jqq+
z|M5SH2L6wY{(m$A`G5WS|Nkoh=TGNA3rf>xnFP|LNUz+){+%l@2JfdpMc69pFb6d5
zU*ytGm#NSn4|%WvL?n|VhClWk9Tj={fTZx9D|kCTIua~;z#q8zQrHq&;SM)Oi1Pb?
zOXx)XHg&L@ReMwpV!azK3;5x2>Fp`1rvnz(6d*D(B&j|8l-ZQSRI2J@AQnUo;A`!`
zzS`qhmP))z_cmlQCFDX_#~AoZ^e~g-xn?#vm9sH?ZNsH01EOyOU{hntMEF)3$8-Mp
zEICMzu)>viY0%;x7Jzyi;2#+Sluy9<yj7v|y;7%8oo;*G0PwFyG;`}=1Tk|ME*>R|
z$Eqp(>D<<gGn`0-B0`bO&9Pk)EI#<2R88bbR7I)?+o(x_r|>I4&VH+7dV+BHTZ?2G
zz1OX&XSimXrt0Ke?2n^g^1B7NOrWRXpw=30^;k(O0I@)A1K)TNp(O<7PK{*gXA;^r
zdT8s+9f2^cjkd(0lC(}Gs2r|xw23>N=%9TbZPaobjK|HSM;R<hXCQ4wugm#$AXQX&
zk?Jvs=;}F(ZQ`Rx53HdiklL$j=pXKhV(2{XbKzo=Ge$7~%c?p11w#_+|Kaa8#FJun
zBWZWX&?aSakP1zZhi6#?(y$87l40^OoYFmN|JZT9_UbA8jN!EMEm3Q=->X%lEo4c8
z$?Fao@|#$FMO|Fg&%%9Hz?;YCz$OUFrnDk(%7YP<c(U<vJA{DF-9X5{fuTO_ZKX$E
zMA#0@^6!>LJE9U@g{#3k^5?sd2*;;zo$HI)9LhcE{?W7<)lg8`a?WK<50lro=b+y|
zT9bufw-R|5^o|)Eq=GV$?$~3$@o%HmxisrARKSU5nJFoceB<MXyTWe2!Jg<U7So*7
zGx_~g5MiNE$`K!eisnLJ-F6~BWi$$QgqZR;IHiJQ&3t&>eOdqo+g+=M)K5uMW|4Ox
zW~zPhHJG`i3(wQP;OkUtQ>z_0I|;7D^;xM?tA?;NY-qGEt+WrEVeHk9NU%=YhQ~54
zI($P^kP735Cy`NP@e;61$h5lE6p+QQ^Q(<zhGypVl;(jCWp3(x3*5nGHd;e}vFkjB
z#rFkWu$wR#@A0@$!(V<wD*H%b<p)0yyBUq$0jVqXYl~_V&Bx<(D+|A<usa&0pSpG;
zd(hk0!6udDq3q?wv=2$prYy6f*gR=Mk1YirGG6qj52U>MZWc1=i<oae;O^zq!w_Ed
zYoYfHmuT}>)sKjL`QtrAt8~6zldtt~)J@?}-hH;sropmo!4zn~_wdTL4O?*%O3^X)
zj?SVJ`0u2aFe>6zhQQokBPgBmm|3cs@^67^TaDQ9UQiPADW<fYqf<4Iyg0OjrU;~^
z^h0Rni5o^DFJhi>9Xy1vO>;gW_lQFPQ*Q_<w;zE2N7)ITXPT>d5#Y@7Q9H#n!SlAH
zvC5vgc97xDlX3Fq>~1dJDU$L~>(}Fr3`zVIe6_H^vSvE;=kXh?RvY4jA4s>gtElVx
ziAZXRy53x)4?q+PozcE4{^>%3mc}*cqr=JKsiA9`G_Uz}9)u{}P-yL@IC&5%k<O$Z
zDCp~1<42Y3kNW;F46HYU8*E@t8$t8&BmE|}KkLb0P4j?8o+=xR>-n~EE+uybs0gMF
z7t`x2>^V*5Babw3k0S#^dEN%ING0-k3-yqOi|KpEQ%uYAGSh7y)ePUdcY130-JXwZ
z;ZCIl!N1L9g7-vk>pDDf=w5K9XPu7D9^IdXl&ObQE`QgpY%wn_(z!lXikW-SQlK7E
z`W(rxbA4h-+KxhkNPO$lmK}wn?W0gOmnC#EfAg;|4e<Nv#T4T(?bbUT4C6y!!~AM}
zswJ9WE*Ftw84aEEH42DZ-H!3h2jcOa5)ED*W4DIp$Y~~4iKwQ!gK2{}_Y(9{rbn*c
z<ZAcZgXv5V_vj;EMuK<o+aD1$pZ~n2#53dc;x(Rg;W{N>?s`tu#Z3%Py~?T2U1Pm%
zND{)l1>ww0=&pRN4Nh}bT=k2W3aQORcc$U7Y9M8uf6d>+q(zF2$ldNlkY*L-n}R%g
z!hzm#1Akh+xKKbnyNKbzOtt6D!xW&?8MJSTH|vR#q}%MYevKq(MV{-vx|>p*;zBx8
z`Yo8WxWqQ1$R2h|7fLGs_J{ypK>_VCxC-$rINi}p&k;IUimBS=)(Gbq{99kM{l%aH
zvcy)@NU_%LD{a;1KB-JlS4LJjqLwW}C}_EwuZs~I2#rQ=K7_pdvCE-7{`~TNCIVF>
z91AkhIsXlPlTiy;CUw*CS-QUXdFJl%(cy{}i0>jw#JLuaY^Y~CpF8lIfodb;4%fvA
z=jU^S8X&W-4<a#15x|B|DPOjcV+#&XTs8sA({9Sa<e29IUwR(ge3op=aXC|=dC1J$
zZ!@1?PzmT%zVf7WSPX;X%9cL3hS0Tjd}|YOdH?Sy94tZ;Ou30=$IgCAXjN}9CWY!$
zUI7JEJ~sM0o+r$WTmgq1nR>B9!rlGt<#9I!GHc$BwroZ(rFgvLMPCTy^bl$RU$QY!
zOsT|#l#Z(onL}G7q?xpMbW<vwi5*O=*aGVPeAFw7BQk*s9de#d3VH77HhLHEBv+|O
zYNznar0zz7$2HOBmj_P{^~P?>vlMB}<ysAUJi?<+rG2$@>a|mDALth5<IlX5Rh~#@
zkbN;Hk`<9Ys(?dc_Cq@%{vPwBGt?(WT+U!EoBOH-rUjGT-qL#Kh*CRlm{e!{Z3Fx*
z%A6#yu$dN}b0o_^St<&x6_K`syID3C3FyH08k<lDT$&%?=#fab_MI}c(An{ko3INr
zOly_3^|_~rwvydT(DJ(Rx{<oLLI?h^uE9cS;rIDB#y~-WL?v`w<Qp$#)T9sO^(#vB
zZ2OJ?hEm`&ug<1TncVxvA!-))&-*H^(@Z^EW%KCJN9?(OnFe?L0zg}M-J*lK!Q+vQ
z7LGFM0y^IN;4bKS%s{_!ag~7joDL+I@{_nvM_$du5&Swo{@rpiY60&K0(;M)XyNEe
z5X42~Mj#X$kpZqOiwiSk&-L_l#s*YcPazEOfhAN}UZ-z;o6S3n=vK9(ove{-I-oVw
zE1T%kIzeE{2kDGNZto>N0I%(4HnFt9aQ@dL44(6|u?!nbWUa*>wksl(_RrTVU{Soi
zf6;?{N6F6MUP>`l6)C$x+Pdu(8)&ZnZ&uR5hfp5@LN%U>$g({Ae`^88A!A<+CDa3{
zaJx3YIxt`t?o3c^Sk0M46%|H{ytl;%v494Sc+Ep^;HRLgzgsfDJsn8q*D*6=zW~Zo
zwCZi|e4VteZc&OyKH3WvNE1lm*WVR(l03d3-vPTNeG9fF7-223@tD1Hx;(_9GA6LO
zhYL1qo5YmH@~v-t$LQ$gj(>cpTA+OzX^Qmb<tAlDO5FPytOZHPUs(qfiiE)IOtNVe
zV1MVdpz9x<%No60!lo?34sK(A`I4knx)fEz^Q`j3!B^!o*13_o3+or+zA|;n;kLeZ
zQ9^p6prDcw%$G^B(Ld<|HUmRev?y`NlQnSN33=O9$4~jpT(d}Tv)kzcH4JE{Xqw70
zBSog`DB6mO9rnE)x3|0G)Xlp-g3i#r0-ncT-ba#yGMwb-WtzBhafGs^RRAl)mQ)Lc
zBe5=_gc*H67B2!DP;F5IFDENEc(qH0r(Cw<6H>Hp>!z4ygD({7OcXI|jL6kn;I6bL
zovF<Y=GQR^xZ!&phfED0g9<_SHa#?0HLMw1_MSIYrD>OnjtG5WTK%B_oss}6=HQ>`
zRE<wx{Xy+UH`S$GVy*q(Poe1BpC_b}F35sGoHa{YyaM``&p=>%$VxwUt&?QEUwdN%
zLxNs(1HqitHrVjVSjLMTwX7HN#2b$X%^^o$WJY;!Ln9y?$&Xi3Oa(zq*|#KUrSQKE
z{Mh06t{xd!=Gg-eB$^s(<{+Du^O)EZ(23e$$5PU7FURjzeLn6%13gh|W#WBrTU<n>
zTIkE2I3N%k%>d<#2Q^*-*zP1-$O1BF_L`dp5~h9KOqK*Gu?$(TGQI&qsF#<+L+^9q
zLd^q+j$Ukz_9^+=kI1LZR1`Yn4l~e`>o~iWRRS?)NqUcHhC}}OHTYzOMbHVT9`dD8
zAYKK~;#RUSzp(>09TUcFeM(3N8?`2&WzIP>ipQJ$llS$o)qM=SI(fnggdXbggG<9l
z&(@Fc5yYus=P}cfPbsF?-WP(~LW1yrZq>nc;`PqY#9^5*z^kKV_5+VX;3#mU*mF!e
zCei&|{5pFmonO&ETC7SZZfSW3cM8us*3_)WkG&nallv?U$}DvhM!b^yM?GnW_)WT6
zG3~ce)IVFWT>OPDiU?2QtM6q!K8Yk~8Jn42YJc^$quzyc;^<jP)WaBpnS%Ln=N~l6
z3I#bJ@GV}4@D<(;Z@EFv=K=n=c)&EUnGC>Xbnef3<K@YRC5oArqd^5k8;v~cdcT$D
z8k~)`k;;lCpq&<%zR<z}DH)_ouLbPWZVrm`Gqp$1#~(w=Xt#(wsb?*dc13tpb;8FD
zNH8`I7EVSvMd)K1=3#r12-QD5+$9l1=6nW6`yi}Kmse&37R2VlO<t6_FhvS>y@f#}
znWxt4xz6l1Bb)~C$rQA3Sb}lkBONQm>{tvy|9&m7i?U&WN6FmSA!SnEYy^0s7aK+(
z@}>8EnCfgU<ed~a0q<+q6lp1|hpYZUav^WEUeLkAImav0pgob>=2&irxKB_I*nnD~
zjIU&I<~hf^;w6tVqHo4nISr0(GjuD>#2-Qe*tlB=7lCE8ePzUI{`osVeaGs)E`N%2
zyW<z2fJamOLv}d{o}ms6o)dUnq+tfr;@}r^UhzA?0z}c&_=cFS1dUv9#wx2TZ0~uM
z`=6Wq6>q>_{;F_@4?4t{zt~YIYD6K6%VI1Q%V<4B8Fye|R~uGDL4~ci6ED8l36{2U
zX#C6O(s_+I5&Yg8!4hRD;VW3jCwT!7f$asnFhS5-VslXm>3FaVO}nzPauw)^iZqFQ
z$k$ZB=ML?8UrS5s5cTP(D2XYZZYZso1Z~g2Y7Yg27a?w)-onb$<t<cm;SBV~dZYDR
z(oOAyNEE!r=P`yA^F0Z?fL{syf)$w^%`*eig$YaEqQJNfVq0U9+=e3TbvZM#)Dp86
zYZN{qDDTZ8S@O3L6W2|xS0rTpra$}Rjhh^h*+4Kx6-(L`zI9b6uR@m~k=@_N{N(t?
z(Zi;YTHQk+u7-~=(j_?EB37p$X>6Vntz)*|GYz;KwU8OXuK9o-$0jH=Av(B9pBLy)
zPSVAi1)oiPFL_mc+~xKg=#8kh@>4;aqLIWgALc10LSv^%2Cal%h5}UEi#l~2R>|e3
z$HLg_*kGqu6Xt#KGh(rBfEI00n=;~SSv?|mQp_55Y807xC#0du;Y?^g`|8t1a@+?i
zS9Gy^g%!bP^?zVTp5#7?ar$$`eG=6#F#w#JIdB_3yDC7$ud`_0jNd!#DRm-etsUcv
zl!leO8(+l8#oV<2{JBgHn(*9HbGbKV^2yFu$PBrBLe{Qa)Vi1o<%z5>as7SWrF4uk
zh8N3$De>x`t4O3DBB=HoOpJCi$+OHvUL84f!}=>F8WfevcjJzX3)g#|kk7xobHrb)
zr5md`Zn}L}=0HWm0J!eQhc(n#1h_zX${sQHP6gj9Myzky?DPY!HQiiy>K5;IC4LZF
z-za{45){s6Qa>_rdiOPbGamCZDSYBd1i#s&ov9QMdQn?(ZK@kE5jINW6+?(W-+qX_
z?@+xkIugmZf)^AM7Esi&{SMya7k&TQCn9~mD{e50LA{~gkA7Wr{0YY16Eq+^J&2Tm
z4y=%(JFZ(tepmX~MjXsHT!tOx)?^GLPnqy4`DlI#b9EvkKI`@xqK}Y4biNe0E4TvN
z;Jv4Pu=*X=lC!@a<4CCxp8Fr)_-(KZ8t`eRO`%XC^vq++2P>Aw#2_$%Gi8eZclgvG
zgr))1i+$&J|BHE_Lv(?fm6d^K%IzQE9Ia#QE{0Mf>=0?To6a07L=M+?Ym^K(ODU1c
zH)&P`wLWX!Y7tR_&OX)1``9syQ5=l8;%2E6sgrGEb^y>etCT_gW6o-FXq)3&#n?kB
z{EB`(TIp;zIk+80xb_6Xumz|fMc$jJZBPx06ECHCEI{PZ*Hi>_%#43ro(8LC<@k~U
zJ13W#vAFS1ZQ^U;n7MY2dDvcz8kJq6p4sFmE96+}FYGpr+&@{5U(KN#bQy-$o_BdQ
zDQ%V-W27rDEiQXCe|8PLMp)WgbCN1ux($TKtqhnU%>(o((T97l_lK=HX1Id8^!A=U
zcsW+PPKj$_P=M@82)x>De^&qJ!j5G2{qPmdl-u|J&y+w9O2uh<?e4~1C-ZIk7aa3Y
zl0y|voyBjX6lfKiX1}olY*tjvM&$c@zV~{<p>6rdh2IxoDeJ|jLXCyD7pz7kClcq3
zZ$SeDcOWWbeyl86=#XkNzr_<uG`^?zN@u(eoM2hud|Y>)wkTOANi}6hFtJ}_p?8`B
z<ht^{_=BW8gIxd7ksbM~4MN;z_yh2pG!U`vfcIvfsn80#u}*I<<2|S(QV_-2Nbf@Z
z197NRnPjd2**BRfqm4VQk{zulm5;pYCa3Mqy;)C4lBY!?7AO|f?;R8ndlGGHhxI?`
zbJDU%qj?`~-ohGx`0u9XJ`tbTi*1G#m$3MSgeaa+#Z78mR!DCWaPW}MB&(I@L!5;m
zlW^G8ws?7o;P$w^*4Ie*+aFILmj=T@!$o9VeuUlO%D!qyMaEWw>*<xrjlMa#<v=3`
z9k}P!|8Po8HtyO)rrfjXwUOer$9a%Oe+4SACAYR&LA7vv8aVq`%rz+6(0jidO{mM(
z3=pGy?)VS%GsH7Itteq%M&Ij_e=4lw#;Qs1Syzp|8O#b$Ca$bMd&%7mT9^_&3DqN$
z6^%}Ut#lre8k=K;x^FxbEdY$cl&Ql~Vu!Da!ao><A(#dC3Me1h>|y`5JyoU{=S#}d
zUr>B+%@}MU+LXzZ&XC2g?}FS%Sxy6$3Hok|Bu_=X!$gXSlkzC=jFL%+6`QlXW{*IW
zx$)hmIcQzP?%pw2?Qc+kJ-JKay&o2xm4b5e3KjG-iD01(k_e-n>5N*Nce|k|3QTY-
z5zT@cckTVNn0HAR#U13Ty6VnX_kmX+8db;Kge*f>;fC<ISpRpoFjL?sK8ds?RDl=+
zbOcMozRKMV!NDm(U9}(8<m^lwKZ|T!k>hCAXuQ2e|6xz1^TDU1#S#t4Z{yBMc~JIb
z51aILq7&s}fj@l$Zr81LAi?s-$9vFB#<|KWE*(_E-BfA11~&}Ti+<WY2t@mUXP#yT
zWI`x$h?R$V#rqWM1VG;J08qs;)3XPONgDf-fPK;@Fl!;4c@FCe5J+1V^4Cv^$6oW0
za~N|59hh740beAzYooSnW#vauI^AcCUaM*#ah+?8pE_0yGYe8%ZrA55iB90QiG{$K
zE&Rm4k9fHzl|!CNOO5(vd`xor6A1`)I>YeW_9Xep!D;&i<^~e(MGND5DXf?76?N`V
zpB!+o3I{Xcd2`Tj{+3Ssw;{%$Zxd_jecFr;d@n{877l{axC?l-tR<t6^Vo_8xDiXk
z4|MZBW#4ONnIK_5s#FP&E_zU=qocGqi@*6%0~I>m9VL#)V3@O#d?eE~TnP<R!G)}>
z<q6(ma!_#kUU(bS!c$M(SIKB21!GwiKJgoy%aS6Hm(%X#3G25>joqS!fm-zd$i2^t
z&1)6Mt2A)QO*;gqgm4C@UL5ziPhmn}gycvRk2PJ%GyV$G-Za3vo?24%ZMsJQWx=I$
z`_h%N&`x`pTOnRP@lDRWAGeD{pD0u(%DrK3pFD;Q%k^k{>Wg#Iqz44`3c0=(>_q#v
zI_8Ca3sE2<f!6@0toa&d(>~<U>_O^xQ)af8)xCG<44~H?gQ!SI*HCrAZ%%(qI`hcd
z2Q&D|eb;)-`B&I~?&sev?)q==PQi!}rW2o)uv%|o-sv5)99kZo2P~y{3H|;vz$AkS
zrt+>n&75~i=6Z%ghP0|+4itAfCu#9#PWi5k$DB1BDe&bX++&S3!!i7$;P_6GE$Mc?
z#{w*v-q+}sqz>Q40fiv}fcYxbmkypsvc>LU3pH|u!s1V;&W0n1<ySP{;th>L%FU_P
zV?HHxqi-eU?K}%f(cOO<Ed*q`iQ~BXPBlc}Z{~zYsuEmCw}4h*CG#9ooIwU~p3gKt
ztCOUOwK1s7AEFykk#RfEv?CfPnSz+VX%Y2Yw%<~kMl#dOJ(nG7-hs|eB-Mxr3%X5G
zm=Jm*vlfrlK~-ax`l&G`jf!WMeXL>%bj%XN=wE0N-Iyl~UZyjg$#MMwdOItp_gQ_a
zy2+66hJuiM?g}Ds1Bd_JU0Hq9nM*kp>K`fdcwR4Ne8gR+gR-G9<{+!%4aO)boa>Uz
z6>w9``zLKklF`tme38*dhCPO^-3)%<Sp}`SX-O!1fT|HYfjXpvSj!#Cl%k?gVPiH7
z&`6z+!V`GJ2ZMMXp8XIL&{W7L%Oz5r-GgYE$VSz#S@Mm8Zpo^83nZ91UZzEkm_XLx
z&vT2_X#8!t6G7sPu1?E4^kS}{N1Q!~-8YnEuQ7yVj2KL1ypulzlwFtg&CZlu8qx5_
z@gmRx)gxPN@ftD8q`*ygAuLzS-MOjiW-tAPZ}{832w%$T>A4+d56FA|*?|5j=6`}m
z&&M14C_A0ve?LvCp2y!x38|_)?b&6)YOH)h3D(Pg(*=D2`iXx4u@B*RqC+}APMOlJ
z&t1Fh91Xx;F%Q*3mf*9YJR%ujvL&K2gfP~bv0hcKvN}CCCekFajtl>w?fS>1VTju3
z<*Vtnz_wO_3#zuNZ*A9o*<EsuK>Pd4%@u)h!~k0NDp2pHjuJ7>7`8~Sm{jyJWO?Jx
z>qRF@hx7;ZtcEcK0)(DNcQ!uSczry(mUY(j33xWS5!+r|QQ=PD8Y=`xNUXM>iKt7v
zrkk1nx~o4K)Bhqbhqe*wO)MVU(U~%70q(P*m?sy6{m#L-R$g0P7LU>PI_Fr6bWPAh
z5>-Q)YG4nB**Rx4GWMUgnXd#Tojb3#(WL;TB`M7lP@E)I7W<@+@}FCu&c~OL)*gwC
zvEL7(0$>i{3`>vzKt=EHe!My<&+wi^jWs9<aUx!vsY`Z`^)6`TtFzR?TIJ2!dt|S9
zs+j6Bz<Ii_lCSyE;IaitJjp%bv+<acz-M}!MUJCDcN&zsHF!JM4pAsMRZ{<NNw0~}
zIeu06y=~(d@6^q+11bl52fo2kUoXB2o1HXBh>S1mu-2H5RcIbQet$SJ@Aj)IIRn8`
z(*g)Sqb&<;RYvsvpoh;0enJl>%Z|U0ny8n~KpjpCgmo;SJtERI`T?*i<Q>)V^2*1A
zg0ITRvoeZN$fKkT)!XzZvLkx5{42muEuMMmhV*mR{9+@Suc!7!6-dHbKJujjNCnid
zKCEY_=@Wf-tlnaWa!yH`2K`U}zJUA=Pys;Uv6sh<wm66-92uTg4XsM|Kt@lH&-?M<
z2JXyjYJ~r`6zJoNFRWZcQKoq^)jj+G)mr{hOu~~#gh#K&@fwl@!VS=#iE~b0QYJ-u
zbAimduBv2Bt$KXoA<)Sy78M|jb1wX%xx`*r{4+nP*`LeU_~o21^q*d#<_5sKxfa3R
zmhz`Lpp5nqsM*m(hl87SXzce7fM-;@X@Oz%1|~ET?-avElLt`=(0w=Q-wWc9qoCw_
z-=Ir^*2<@s<=ST>heY-L+$yyWwuU{LIImsxuMxvJckflB7+uO#L%Nn}by-Qlxk(h(
z`>U%<<Y-oX&s>9a*XW|2ZRcPVS%;Mj_(ok@qPN$QM7xC#jTAT__OJvV_caybu9Ry<
zaKFgMboJ<2p-vk6+ONOInoTXKk#FQ9pJtW&1WVx_yM1@d@-&+HP_$v>u4Y9sn;6rq
zDEnmxS%7p{FD`#4TA>xq102r-2*L-D6&j+h5CuV%{{hA;=_0oLeWzSq$j`UUXfXBR
zseN!6;J=geJVnVG5rRTj&JU1I=<hJ6mZ<3Y=)O6Sse&4&F$~k_jj0|fIiMWYK8?ns
zsqX~x_~vc#>0Vq0KVllSi$Sq$>-NN>)jRXeM<=Yl?sBm0hRZNGHPwxty_MqRHqH{m
z_R8HvmQn%IA=;g`MKmMd>TH`41e1@q*o?ov1{6xrmO!>Po7Yjc^1JfbBdVpfbPh%4
z=uY@sBmkzUNlPAJE{R{zPfV4st<E83N6ZRhPV4j0{7*jL_yb&^4<2<f`-=K<<9skb
zG2kI8AqY<aJ;Q9@x>GVBYj*;f!Bw;Gd?e_5qfWr3EM$rI&<<R8NG44jz>ueQ&mQ$*
z5gqdAE>_S%1S^LLLW3V?VVqW6C0}0sjPu4i_YI;s>gZK>X3-}>7RS6(a=Xv|J+7w5
zE=osGSF5MN&p!M?7w1<vfg!S$Q?dQxR%}S!8g29X*8V`#FBNp_1mSe*hJ6x~E9CVF
z3{$Ne7b(Ki$qA-soqwXIp=^`7Fr~#~96#lI$3cJ%{O(&0qD7y_QZ{{b(KQwLu(~^S
z6h+zcA4&Sp=8VtfnErUWMSOdKNgf9FQ?r@_W{SUm-zDA%Mm+tpFc6yqkidGfux%cZ
zsj+OpG_(MrJbkb({rc4*u;&|~)>?cQ^mV)9`cvDUpo`q4pj855hhyApz{6KX4Ak7>
zx`c}1y#o=a=yCC;76iyw^sKeKly-G_PeZObB}U9CCY<$3m%J!K{21)2dFp?5h(4Zv
z{0>gde`uV(50u41Eko(Zfe*Z`VBevP-fj&>^zo6>?L#%$)c(WY7-IA}zY_H<3SP@I
z+UdF)=&k?NNwHb>myt1bz&Nz-db~+u4ok6>`vASt8T}%DEXOeIiae77WHR501d|7P
zz)ge&QZP+gk|F!LW!t5UPFKmO8E@Yr<aV*(d~S!hsR~AmoQLNf9K!ILsBIpncirvn
zc_OqWe<Q98%`&ERyxTkl?mKlq`}^gEIH(;TUgtwwu`FBar`n&IlrJWd^*!|_He_&D
z5V2pKH-D;&y+1e#ZvAFF$W$|qKn(ao&bbPHOKML!ZXQnWfl+jJp(mpgkzGCq#1_mc
zs_<xT{q~|_x3hk!W*h`EjUKo*h_6M-%Saz6N_0ExZd$)cLSR~P#nlXa&-MD4zZ!N9
z+jq8DIWqv3D;sWwQoBQ+5;!8Gc-U5D>OdM*3XY}6D4^fR-aa#9seeK$X?!B3u}5&B
zlk~vu+aXHcHQ$PllT&7&SeguX^Ygm>Gt_N6V3sM^I!(LoU&<u$rL{qRI^l!;2>;9i
zNw~j<3=!TWG)mNkdJnw9NG4re9S{H0Cs?rp#!>bLVF^9Xc)HM+(4HM@jm=hl^7-R5
znkclD&fC<OS8QM~XGbm?y&YEjvt-x34333k_9EpVX4%;X#9*6MDE1gGH?By~sA9y)
ztiU!e<Kt-E)H7f+@1JqnEi4jbPrz4u(PS5)K7MlP!~4-n-n_4rDUb{O#wQ%L&AS4u
z<+pQ)z_g}sLtja)g4w#>6re|!f(*}q`y+U+0^26)>Ml^Luy3$@&4KncB#m4j8;1=>
zsM_?dxh`$%e!bSwl2%vDb4hvJ#`G9>|3<-m8F2r3g1%CJOL~&B1iNYddVQ_D06z>P
z$g;3t24bjQn12zko_cX?>eld_10tX;F{Me%PV#kVXNIvsi+l0DvBP*>`K4Ro>GX4X
zCpMwiIaWr4>3;QH%J&bSfbQLq6Qajc<3%w=gI)|!=JRzYOD6z^|MmuwQu+wF50rrE
zU==u*)9J`^Erf(tX3$V*igFS+`@Em!EKxxnMN?enF9qFSdwH%ND8WBQTO#fTt}Pj>
z){Obk5J=;J!fTPXXpYC{nJ^1z*?5<rWhX%ncS@D-L)qkRnc4sQ@Fq@Gr~Y-;KbsnQ
z?&ah_8kfX7kJBn9KU_lVMa)4Y{x;75j06TC*r{^O?${K)1wBPa=FqK{G-&3X7K#~#
zR%Zmxd+-Cv#`nigsita|t`TGuN{Ia<bpbH7k;`Yd**nZ8|7>z|oIeBLd`d>0(JJ{@
z>mYukkFUA_#1)kf28tucM|m*?f^@EC%mYuO*H4$DP99s*A}fy80-VwaJ>R->0Lc84
z%j7|22{^vDww_+B3x7#jB-cdS)Ib0>ule&Ef#$N!(;jbUEDS5itJ1^Qy0?bei}{YS
zK2~W;zMmW6GvGN^JdBPbk8g~Tfv#rIBo(`}X5j6EcyPB@&QMf-5)R}mXCHHb-eF=i
ziB$-LeCxM1@%onJV(~}dnrdRhF=0;67osuQzZ5GSXmQ3nV|J&8PJwEjinJ9OVr26f
zz((i4E6!SbHBfdcONHgo6fjOfu?8oXCr6OuG`yF8fed9Y#`T<QPW##@#21Rq*R`N|
zAhB;Qa2qOmOSpm&Jh5?1v{7*T>k2f7RxLM;zHE+BqCHk9@^fPL-gHf?x0r#o6}_c-
zbgBRZZ@aUnlJk2;jiX(eeuilVfxNW!&|r{(L9${0h<{iXXg;1zpSo=;3;mD}mZ!YE
z*}xe4z2s>iB#n3Er6(MF*xo-9V?@N9L7!^s1~NHNDuFNf?i-_o&zPV{>Lanj<Ye%A
z-H#rQ^<T90_womq?E|%&qp%T}>UJCgeziwly#P=h_zn@8=@->DfX~dIWAh@X3qZ^H
zLY^`sO)XHoUUqGJAwiwzJlaE6JMZ#r22?88x-%KY0jbhc7h+bL3ur)8Qn!wcQ)m-3
zQ|IE{WoJH8tcLsf=zsStKgQV8nUKhhH18lT^@r~pXd6E;y9KOOlPSs!NlM&METKXS
z*%bc1v#58E(0P9VjQCGA-o$bED@;L`%&e~l`yH(5j(__N{PonLnA$3}p!<WE@2<E1
zoRFqwE(-V?vZ$3aZU}(X=+tWEvW4Xf(AW0!%Hz}Vy4>>UxgvLS{!#`nVm#u6O0s+1
z&8cZQz)5=YKmG;!k|vI5UZKjI(L;0DQkS7pM~_E=F<2#D?qCTq>zu?XB7CFb6$n*v
z;zFAQfaBL#odU&)*D^0FSRt%J{gh`~0>rI*PB-omY#ul}aJs1YNj}v11%d@|vwCqa
z3(?j5?MRN}!?Ejzr)dudWU<-JF*Hk|DO$CIt6>=3?H_Lq>x5XOgPRw=696&&@&9AI
zX`7#^TpDkp1`RyhaA)^x?X=ANm)i_Z3~J%4V-#5J!4Hf@6#Vg^7+1-u^4$gYYiC#L
z=y6982pL|`5LN-p7MH9BJmFov&Ce-}T2h|#%h$y``&28Xjv_5E<>dL@COCG8s$eh3
zZ`$QO%wvtMj(}pcQLC!XQn8<9ZctHsd{jOeZNSu<OW8$oLWb+Mnvcf36~7j!=qpIS
z3>s+au-G=YYt2r!U?Hl+r_?xTGACqIM0%8X$!M!Bd9tJO4^XHVNgs(4EJXv()fXat
ztg1qEf2c-9`Pz7<T@dIq5ivnThoaR?Jw2`;OvzW=smuWEb$$oLJXU0*k$M)Fcs8eU
z+F}r%fb>~6aB4brNpyPI2CinKo8vhsrC_FJrJ0+!Hock)WnVe$oXihYfE>Z|ArTO~
z7?`~{kJRuv=Hbwq3g5bM_!+rZ>en3X#tSB{g4C>xb@|<SZAonGA1$z+4AVVJuB?1@
z+Vh4UeGz;3baNSs(o562#VXG(@u5@HI&%HmSR?!`!-fD>3-b)5Gq^*&fM3a`P_wT<
zTx?j6kkRemHw&i#{}&_}DgC+Ie<^TwPKZydVv1DjpF%6!L30Y0AiVxZ@%?FLt8kq^
z;B1u&Mf`6qKyqbm9u+!17I-d4OTOCmKiFsF|JiE&K_`)X%DN25h(G*4jJW43U{wr|
z3LbUdzLL|;&DOLh;zk3KXo+igz@U^>%|C|NxD^udS`G+x^kN<Rm`{=~Cd`za&vzTj
z#1zQxX^>`m{}2y&tVk^oEocTf1Q?U5{V=wH%uu1-+av3@uJ4l1=7I$>l~s9t$8JtW
z3rdmJG0Bi8iSTn!nTC9sB1M8`f)D}kivFt-%4=XZiU_O)Jt2)2GG{$De8p>R>EtFN
zuqc3MSvw7uZEbMR_yOn&89t<yG_GZy_&6=3$u17fXvVcwkVt|p=E>)&t$aRsZunt)
zVo?Q8L}=n*K@Gtr#XoOSoB~g=Hyk*P&i35<YxIiVrB+jQ@m^L4)D}BTW~4z~zaz;W
z3F2h+pgJZKpxJ+?q14ITKWkt6F6e%z+TkE2#t7pPdy_jZmVBoXT|igFw7Js?^?@ev
zF>xgeHVXLhweS9%Tfa@2+XRtM@2CCHhz`2b;sjD#>0qW{xsnZS7<eiJL|&7Wjvu70
zR@{$vd+t;Z$B^S9lXjgw4RWDfYz@s{Vs*yRAgPABUhn;JScCI%chAoj@NlYW$9=h?
zk*7@_^QQA@4xQ9ff#&i&@?o>cA@jwE$X(I=2`?dfC5#ZHhq#4!iHu!sQfz^?jqHR-
z((EF1#9ouNcV8tvWfu`)`N)Puo~~g7%(0oNr5~U~TKL@IK@&F?F?*^RXnnHyKnTU_
z01+cz37-epM}t$>z-oo42Q&M?FAOJUAT}Y+T#x8|$YmZynD5Gy&xaV>7cFw?vYoFT
zim$t`X-@yqSnL5^Ki2k_v~n{f-=#3I%_M#Q-hp3ZI2<9~Q_e~|Ayo7Kpb*PGu?_n3
zf97QQzCiQH+$?W7>cHPuo&v*7fghyeuQDVZU!vU_+6%on7e=THqaHz+4|_>`1g-Hs
zP`LF6^20S*46B9^=9s%138tyPSp2}NB)@Ls@<k+<?WA!U<S?~V1NN36s)}NJ`;-*j
zQJi)PhL%=#p2jjwAnju2q_${C%lzXm7c>yIw4`Qp(R2k#FP|F|%42l2<{4r_jqG*h
zbD|yUV7}V<tqNiAg<B!DO)}lS=Vv;`np<J)Y|=Oi%n1kGXh{uJw5Dx@Y~2RQb2QSh
z{Kp%}JX|jDBtNPgq%)urlZf#W%=JFtS4VTqV^twbSv6M6OLw}h=jhwFp;gG~3#ZJL
zy^fV}sU+Jg?71EbG+=SnUGU`m_&z%MPcL?9A4I(S7ng{K0YB0xqn9}(88X2T>(8x0
z6Jg5`0@=y6dhPWh%=>euyvizYOe!CCi@R1t_^kfVUm<~rr5PZ&I1)?|0=gMvXPxME
zoOSn5kzKF+@|0M$D8vRu!M6oi-1&m`k5pj_^t{4ul=JtR+rP`8qshx`1{7#sG3GUr
zvmKamC-1z|5)S$uX5-#obC>9L_Vlpg6G~6C_^GsHcEmStbdBz7_b>C=f8@@oxu;07
z^j}YfUTw?H;^5XV4u0%-0qE*Tz*LCx#kOb2Z!$B1l)WS%A-J*x=5E?vmmdN)KcPl~
zbu*=-v(Wo2ZY;j6&cpnkr0iDKV`}5;U|>A`s8M6@zAg*S-$BiBfG6qco<X*%P~3#A
zc2!3=uVk~kfCIUcw3k^5;2QHiuvSo$Sov;PGCuk$wCn1Li`(}KSi`Npe<vLeFnI#R
zMJ*-7n8X-5ho>Y@d3oZEz!maX(fp^6@#>2wXWrOp_k6&5;1X?uBeAi9;h$fnNNt7t
zqVpi*gO%=t>A|5=DX<(&;Y*B0caj}HVt;j5XOw7(boT4&Sn}lPi@ho7gkpP**IC2U
z!v!NV5({eo-xWfY*-obg=iKdq8UiH)=+xMR9u+xQP9}Y#v@b}$s2GZx$|qiP1k0cz
zDrhNtKpI|)yW2*WKl|Cgo8Kh~rj7$OefKPNion!RkLIu-=Jmo+qjb+&(HC9nn=QM5
z6~>yQoLlCSH{*(N`$vDPJaFM{PJf@$$K!8p^5v#3rv>j)c1a=-q7^b-UuSM!882tO
zgdU@5{DVUFk@`Q;Qj?;9iRA_{=$F*9Pkl)0E`5&XPj}7&hf@c+GLL_hWm_?Cx}PdQ
zI@4+5q;aLK!3$jr%Vp3#h&kKZjK<^@o7<xVwZ~MrXwsuS#MeC&(aKy;L1(QFq;=gs
zm;|^`_}ke{ZR{86;Fd<=L3mm2+(m+;b^2?6*f02ECsZ%?j+ZV-8u4M}3%9ivd6K%X
zY_z1nQYWoJUQ7o6D(asoAe9C>dbcngm#JW%gi4o<FJSo-!sI;%q>P~j=s3z6bS>jF
z7~6q_rLPO%>5Yy7Ef6+*$Om?**kNIX1<*=kSs{a%t$ak6r6R-lMBknX>l4uV9!P}>
z!CWEp*!6$n!w>1m#n(=yYI6Twa{=^4|5^3~YJ=HrgJ-|@Oj(tC^8uNrhFR4J%GL)M
zTku%rg9#ml7bg*4s7-_Yy~JXM^DcN^T&7$P{q~5JsdAM+{*6Nm%~MutjdE{h$acO%
z(Z)ZaG=wEUIiZ9`y}cf#?-k#4X9c!sEqYsC?9A(nr$m3f@f$E-NF<*)`h<|qKUO5(
zIBj@lBs8v+l!stfTfo0;O9+;8Q9+y9+&~&P$LkWNal#ou`?Abgv~37T@6c@ptxiYG
zuw_xr3IK`S=LwlxHG5!(o29}~vnubQ#fJC=v14zztV0Oe^Nt=m=~DxF4kiDr>WSB#
z6fIeo%%tT+QT|inl>2ucdEB^a;<@0p;5F33&up0@pl7~!Y`1u{sG_`Fbj!bVTFQ0x
zB201AIly)FYuMWRHwQZfwP$V`m0XY*i%o#IO~lWhsY`0%2u@z7J0gq}@^0We4#$ku
zy)h`#58ku?xbkXRFDEO2UFAw{dYj2GHZ;Nhip?uqA#)L%q=0wc*8l;fSHQB_0xM1h
zwnW$I4|ZTelD_qb(muX_Dq)#U0Tj`6D}C&S5SBrSTPY`Mn>b1s{$3Phe5;ZJ5_6vT
zGmy%BL!yjcp|`XyV5jd6bkX#8Zaaf_@Kj8YxpIjpB_tq!V#wYpkY--mDkPc<62X?m
zxZB2wf9N29V`zIw!TBm6r79B`j{v9fWPuOs>fE@=D^;(Cu0@C5DtR6YYL4MEU*q5m
zyWh)B0Ht0p2q5Fh>(f=cs_W3A@8tW^L?>IDq}&zaKyxD4LVI7pa;MGv;hGBO;<rz+
zsYtwOVEzFeN`=__c6yq&)oSt<J~5XjK@N>V1YemcpGAxdguKINmi`T&IlN&Y0_o^o
zoZ0Sp?|l531lm%PDoX1{n+_b_Vr+Z0c?TjitXDCn^Lvb-3@N|2DeRqEt@}ItH#M>b
zVo>KBy;~A6E|^jif`BL|FcPSt_6Wh8?84oDPph!)(VN#H<U{r{Z%2C*R$PtSns7tX
z#=I4ClvQQE8KlVa`o`XRM!W>N?4y3OWH|3g>At<YAH``rPJQNy$<48+xK>AwaEs5J
zS;S$gQ4n5Rtv{%pB^bUga?OLgM0w~081vvYmv}4@30Pn+ejCnMU%@k-D4@@!BTYi|
zyLr(0n2hi1Bqn)o*yUcMcu?2EGpT9R5=r9YB3CC+(xmO4O9DLcl>K;$X`mD0jFZW>
zBO&l$dBl{`UK^iEG+(15XvO1pe=43>*BexdI8$pxLF`G&I+C@6?w{Is9?PACQ9>q#
zZs{2QM1^F8vfz6hlI1h{IM=~Gg1KabbL>$`iefSYs0UleaDp=WDbX#$LU1)(3vfTV
zJ5x36D7tGKpb#S#xe!H(+=fyG8gw*%(T66^|6EIDqd=Pe{#{h}rJoVI2jl$l3Nmyx
z9WaiQ^@eP7FlBi8Z6jl<-zr{JD17Yu{$aRG)Im!HCsiBo5k?dnh4pEQkGjkM%aky^
zOUnRj8^J4L*Jn5;Fdb=h>g73KKhOQ7+6vw!2b3w;A#(Y4XCaaEP(lHk4XsB{oev8m
z!jDD1G?PfPyFUiODE+oaHm+RVmxEQjkgZO|aNYfRZ?DW)TN`_h{uL<|DOG2X=&p}m
zem3$tm};gJ8mZaVjakI*eSG-AF7l#)-xc>2u|#CU`Or|gC@k~!p!;b-%PB^{lo{r0
z>5)1vzVtl<3h5z=Q%5nK{*1oPLVDg*MQjI=`;KoqBwPCkFo<l4Co0{z(k1eoti$k6
zu6Ah!9{i%uA+@HG5~|1*?6K`*E9H?%0W`8KAVZht6t?9adv2E(0z-+8sq|D1%bA(z
z5sX24&d7=&nky2;8O(Od6ZsJ0Lo`YPsQX$^Xus3$oNR2`hGE6A8V)G`)Cptiq~#kG
zcRMq&d3q7|PTls^N44GY>6k6&AW!HqP?nVN8A)<riG2B3iXDT1<=0!sKa(4k=pz(}
zzYSQMjKdf*Tz|)B8|$^g`4o-6&)hBe{tqg<hS<BPWpqQ{@CgSAN52G&ptqgihm{%;
zvB3Kwm-2r+XU@SOJ{71wu!ncS?eqx!nB;-ToaRman;L&m%IQFL*%7K2-&wEYms}9l
zS@wDF7KP0f8-(bi74k&KC@84FjtQi^A|0hd7EIEj7W6R+6NTX*o1UNO=)l*y5@U3+
za&@rH<9Z5VnTC$W4=&ApGMFv2kK;S}4+$l<#Y{su6|W`ybOAa%Gj2rGXVM6B_vZ{X
zK@uzJTe&_xvV-s@WuvrB@XE;AzM@Ea!huV4VEiFFVoF<KX$}e-g{Pe<06d+N=u82^
z(qwMJ6BIAb9oy7zzZIkA$^BE*f%=3bQYb*ICj9_YjSYhP2%@l?cN~;Y_MR!E>F#S3
z(_RNkdQT(B6rNaL1(qc~-o6+c*rXkief{H!LGPxL!bcTiMK4FMV&z@PNv)%Zzr7xT
zEdE;V&o@{fSgUe8!>cA^B{&2{g;)8xyj<byEhJ+j6v^dWe-Z%;V`-BbA<~&T`Hb=V
ztAH+>$9<r(9uKB`k#k*8-<jwAkyAbKz^5GsrYK)f+u0FnUVrU@xcO|bz=><Ue8!ga
z!CO0ThBkV95%Pnb2I26#qaW!Q)!#IQNA{%2d<|BFvW+*0f`T7gs*fK=(BBb0)2qk1
zm3**v#!yuq2;Ww`Ar$!aGeSr=9Yv0ogG@)EVDnA3d6qJMiO-hdU8(r61gD*)XHOhl
z4m-uh<4g&9WeVwdQf89K%p5>ALRoe&Z^5%IO%h8prFmRGZ#nb3)7|2!c>i2dEux3g
zBS+;OP4uBJXYf_LRn=&a!OHVf3<*YsXrVNmwKxX^A78$tKi}(^B;cvyf!-^7l+^_s
zdzlV?8wSGu<Bt34$EK1Ve+Q#uYcbbvlo^K|-Qj9U+rQ^Pelq`ERTaJf#lCcm-q^ZU
zr8wH3Az=3?kKa4A)8v}5<`lT;cmXCt>D*O}|DgTD?2jJR1v-vznp-|jcd>RpWNKt=
zn_nu*ZWuatj@mGqm2Dl=k*a$Wp7}xzM}j`QeqQ;<4gv^Zrw6HHZ;+obVr_6LWDg2s
z0>HCAz9D`Of8psr?74?#oY8o0>~~RCw>-pD<=wzIV{w~fDVgm`*2^6>Usy}e^4FWO
zmai!9*f8I&;DX5+&#xulCpo1$NuL3>@O`A2(BY3LuG}M{^qOz)`ThmLxdt>16q21F
zf$V#A_o?EXf(VcWXR(kT9?vx3v2GBn7x6B9N}AJRn%aSbcg?{u;fvUVg)0j+&+1t4
zB6VZ~pPZ;+iB5i37;o}$*mA({-!^7W0|;TB+)@qnCwIar{r1pXXbYyNzI?J3MUiT5
zdv(!A0?1hl@!D3$X|&Ks{i)-K?*1{@lK#)=Z;+yyS2oqbsRqUS_H)(bUx20)OXQ=<
z;D`&UHs+jvXHt=^_vK9#8@RSCLYi)Q$?ny){M5nBBCo!EoT@EY7yg>lIQVVIDGTNh
zxvbZaI0>hAyx^K+dwc%m$*KJq8QS}P6_2p-X#MXXf}tbZzVvMJEOBk`!)Hv5RA4Lp
z6;BO07P}zZD{B6nQ#=h#gF`}aE}|YPYo^xbTW;%jcc0?y`yP_ms}p2q{fbS#ER!p~
zeUQ(a0(BK4*iERe0bBLKReq_MppP<bm5@6fvhzQEKG(EoIyC-G$$<I5?btrGD^;;<
zr?=7KfLC4=c=vq;*N0B<yUb{Sw9RYBuplNTAku$G!amS3XH7Ozg)&&-^A>SVq`iX(
zO-|cT_rZ+o46!07LVmPQc}<uh@IJNEI~NJ!p4t4~iMio&CU5o=(i0h^t%UNeF-JJu
zH3|?WbH2u6?grtu3XzVKX<Ioqb8cE?hz%%pdD-vJQxi&mdJ9SXi&4Pfk;%(Ww%(N$
zrDtO5U99f|_+P}w;&nYW!t6aJepD~;x2RyQd&;Qp1E?#(l)gGwKK^s@m!0_ku=ieJ
zO+`<?a6*+D5TrMe-jUu_iXc@%x>Nyy(506I1wjxH5Kw8NASx{p=_R2{7Z43SNbgNb
zAS7qw|2yw_zjJ*q&cz}7$&--1*P8v?vu4e#nKi>KYPf{@*#@~=e{*MD)u9xU>l1qS
zPb`pU7OVI(y`FmdDz;Jv8=<JU>mQZQ<l)L>Pzff=D=MGeI~W_2oR;@DO{_UnEO%fG
zrQV&F`SkvZAOD&;pmGM?{IetccI@VcbjPH-K_<Ea^WseLF<BEMR#pvfUoLlgIVpi(
z^(vR&rTmF9B%+Ih_EXZ-B?2#k?uoRadKfcH_wIX!FIJH{2Fn&4AWTAvNDTAItv{_^
zf4vf~u4(qtkrqlsSkB-B%MV~7>6FLL{flKU{H7&_T&sm!sR+tY8+&Bi^MlwRv=(T(
zj9RZ{+IN>okq4f>;cWdb90TpOdV->)$$b|K^1K4*A!$pCza@{&=7ge@%rH1DYCLEC
zVDQI^JHFImm+|K4=a?I48|IkOGsJk{O6**mB%{NJjX+Uf$`e@k{y{`iRhSwp{(*Jr
z9VQR+Bz-zS;ypVl`6rt18;8vpip=6rSL%W4c51#_{^BQ1vB}8mH5doIr`8s+SKyQ=
z@~J{6jopmNjivekY&OOVzBmgiB_WFbjmDEqw}q6N6+2`B>L~Y6xJ<|#Z&BFe(MX<l
zj#F*SvnFF|;sPbYUn_?--2XDoVUXWxk>fNAb^~kL&avHpS(w)8DpZO(-Z|5S?}%e1
zwgy|+ksa8Y)NW&yjA(mr72h`}7W_(|CW-?9p-Cu?xaVQUcC^B<JAU-+c>qt?2)!1s
zc&YUJlNU`)9Y|^D_hah=E2<DuCAw({fxX;Su)+2r8_n5cw{oeXuqOVkLr>t|Lj*jC
zs2?X^9>W-GsbzOeunyRvp+}(&9oFHee6>n#@tCB>_rk8?9BDn18}PY=EuU{1)4t#k
z@o&a?b`Sy$=F4!|cNE2u>)p#zlteOrl<_jt8kD1G$7qu5(>2^pXk3dy{0A)#J3UK@
zA%Y(E<B)`Ya|g$k63FmQp8WIZ1YR<$;Z0L;`$zMt&TPKyMv6^JOOmwE<zvCI#t0_9
zHGOs|;yK^$-1kDk(~?Q|&9iWHLND*hG$e|Ax>88s>&=);c84J<xHA$#Pwm#EW>S-$
z^~)vu`!c&CEbgALrh7t9`e2?Y+8m<i-Z{Fb*YPG1hE_n_n{&K`zl=M7XpUBD)373L
z)IN2D(<R!&E4Zr0FV&vHm=wV-hL`S(pLAh5<IN58;qi76AWV7eE2E9|eUz0FnC0`+
zT(shXCkcxAgpS%H)RZ*TrJA57{{nu1KEOJn#n^t|t(P4IDKgF9+U10oA_WtgEckx4
zr;upT$=#{y+HaB$7@gk5)L@ldIJGM1`&o5$CYP=N!g7bwD=}r~4DKj|vd^Pvong->
zGPqA`kP)0W4;*gJON4x%d+GO7Bf&$}^_NpAF!@s84>0h)G7<X+vT=irP5csFj#TC5
znn{V5SVClGaat7R6~rYx?qTf#FM12S4T1^Ra$x`Z+d;oN%T^=IzInu<O3|n|W25Oy
z(Dj-7;l9Fvl2az>VCp=VK$DhFx=D%R;AFrQw?ftgI6w@ZYB2d-Ftr--*rl39u|~`I
z(0>f$u4C%_;k^~9TvA^B3;XZy&aod(p4uV6V1-5Tz(_c6+)FKBN=Mq<6i!%B-F5su
zV9x0IL9c_@Ox+aIb>4g^OWc@OXQAJ1F#G73T0twqaxKzEnLaMX*Xo&fBdCbpUkjL2
zlj@R126{9D={C{<7k+$~tYwF-PO*|;2T}b#3}X+YXK_)Oz!A5gj|n|0kY)yY^l;ny
z)l!#vYSz2TMdhz!u*LrN$y<VmD`A>kxoV;9F5=1%b#j6Y^!-8>^EL#esIrsf&qd3`
z_RA?(_Sh`OKzI}c*SEvzTCWhxJ4sDRwEC8xsrz!(i+A!hU!t)xFP9+`XBdW1qUBE>
zpi}*!ZF+sTZF85)10!R3pBcley7!Tocm3+++Uv`TLM8$<?jR4B>fF?>%B}^>N8e$m
zvaE>5A?DZcAjL5W-(b37E%FU~irsC(<!Sn*5RlSS6!+B2o$J&d&ltBOYq$w+<+AlG
zm`W4qr8A0iWhhCMXm(n4a-OoddI&p0c;KtuQwma@V)ZQIgRbF^h}k`z{HkE}UHW}B
zVrlFSJo01f?@fVizFiCDOTo2Wm_E25W@kPuX(!;-@h!}={vPMxS|kG_Rs}mTJ3$;y
zA%#4Fk>$SjS17Tv8JPc`WTw~7qhE$>5tFC7H)%o&;`do#b;&0NU52QG@bYJLR!YwX
z|3reyUOu=v@7`t*jUJg#+*4EeisRqloE!K&1sG>(=(#(-3#FNtg%9-wjf%xOFx}5!
zH*l{TfuQSHQ)i@EiE3%aoWFWv*UO40g(>lb=Ss~GlDU4%hE+XyBzg{#7awdUT-BfM
z+8)4-;KZvs$z^*)em!i+9ZYe`q~4~qbh-+^v&D!2d51ZsRLIyTS0FxJ{rf8Mu}T(k
zW=(zH1G)3=BjPEe7HgOi?%ZYULVsW|`IOmUfL-z@YlU#qD$8;S-;KEC_>j%+JY2C>
z)rnM*4@EH@>t5P;P%r9q<wO|i|B}_&`EybQ1v%$q+N&AtG{M=WBTn#d=@cNH$@bwM
zRpN>ePuT0qReV~|7^b|t2AktVKL6QT_q|2z>u<_jtP_O<!E(4gw<?PKk{kn1U59jg
zc6$MQ>;7J{yt&Zm&1VSztNlp!CC6^d2U~55-rQ)_%chjGG<Y5s(j)bsRAv<qPYP?s
zC4-LW0h3a0S;OD;H8T6AZ_+inExasR(E4@=a@-a~(-Gi#5=n-4hgz<Ol$(E`>{x1`
z!rt9YbIxZJM;KpEFjRMzQd|ubJ>iXwEY?_KPN)C4j}i}02!7wKaW^t@au_tLJ^!VJ
z7hdLk#Z}3FY>|STr}hl5UVhFK|5~H5GqX)7An$Uj5-`t>#}^FwMpxR75E9iPF$`1J
z)}hdvyPq59WE$vlZ+AGMy|7)gGaff5Z<2zx77!Fh3VXj&y#s8EEAUY4xwUn6eGlBO
zvFt#r*QZBSW}Bv;F0F@xG_}W_U4T;Oj*2FiQYTR2elul)d_1rTURat>olgH(wtSB$
zgza$~0&1sH{Mv;vpOEbeOPW4d#>E73cI$reQFhdyzf6d1Yx&{-RrN_EI}PM9=dajk
z`d_a@+zRAr`ZL2?jpF^or;zNa(qOxdwBS6!p`eGl?}$`Kb`G&irpSqAdYZ0cFo^Lk
zkcZraPiwJHUh#)H=?>yGu}6rf9%7$g6xWzCl|c7e!xMI<Vy#BFGOzzk9&j~XZoaxn
zJZ7%F?({95Y8&MH<XRaY9#nfBD^XsMH%*kio;lqIg2Zsv^8eK>Z>`fRpJecSO#zx*
zkieE=bP+sD97&@ic+jpCG{kZG4mM$C23NQ!6q_=C2`{5K2BTu3pBE^<WwouwOGnWx
z)yZxDd{B|w&)}=Bz(x#B{qy}4V!besaFJJs8#HdM7xTfL3}c4#jM3aZ?aa@GtL0{V
zN&V9ePuSe8Mn1#ZSZg=h>nG`JS(-?-xwk#tDq^C$1npGJnVMUS3uRL>sC3qd3*OAX
z;=542S^JAH850neuL*VE4O13^-<2qzO`Q2{7_5|TWR9B)ZYG_5S$}|26^d_p+diH=
z{I1u+t^O(IM*kOv>#Np*Oa7nPPUy=v@~E3t;;15crj(qlLUT%eFQ1ucFbVZbbnLH7
z05XXz+wtz}?}kC4XR=rj3%}f#v=>6D8tCm=J6W-sAD|nNQAvQ+4{#Iv68EkpwJS)A
zeER1MxkJq~8_RK6(6**l+~B4%Bd3D&;Z(r<-WWV!*BjA%pG$-Bxm{8)=%<TD38!4f
z&Xdl$VGDX<kicKls5v;5#5OF)5jAraN{)3cYF1+xKY0J)XW(A3W?)KZ@|F~elQU`p
zRKBrIhnsc|EXt-pHllm2Kh$?KgvBuKUj5R4X0Us0KDxVImctj~I=8SWOJx*4wK_Kf
z>X=_|cYChSJt{#2ruLpAk<UUT=qEzW9@6!o+)FH08wx~w?EU9zb<XpY!5WoGxk7dP
zdqZZadihhm#6o?ohT5}@SPkrlHaOjk#+zYd@W$qDozr1f5aQ;iKygw2j(HK*XH?L@
z@4ikR$`0Lxw4*OO1~-7Q_O-NaWuxb<@o&(I@U~rAu>YsPyBTT&=_DCy9m_@|6xt<<
z7J0Ethq%Q*s~RajeA7$$2K@KhwlEZag`KJkVgZ=)wkEqbpSH0Ja--MA<A@E+$xNJ~
z;`KAe2A1yi9q8(LzJX6_Z|z?<f9UYJt{EUY5<|i|V$QkWK$W_wVUd&hDF*8XqH(nL
zxzr(cVgTL6%EZf^>cLykt<yyjDf(FeBw7?vwZO_MiIc=j#H;@q3jo%lZB`6e@$%pB
z4og`;{?H0YfWxdd5B2IFpj$#taMmkF;TVDVdn`_eZH7p*1Wx74hK2bvQ_fo-k(xz^
z&N5hLD7JW1lV*JNT3wFI^raI?j1wBLfRL<gTeDHwF7lusJVfqfVhr8*An5hY7<JWi
z5xJg1(q^H66)EhLSN(Y379|g&W4KF<SzP4B1l#kT5l(WVL@j<Bw_RYv$1jLyS6n!a
zfoAXAk#K-!bYx$L(dQzM`-Gb7NM~;z%7IuF<~63)RQJ_HYhCH;uH*T!!OztkvEM%i
zwG0@$(pCH5j?dm$f7n$Hbl%M<BByb9G5svckCu~0i@bcDz}BoVl^9G<|M4$@bYA4j
zWxk=OS~n{9c(|L8gQN;=Is(Fj`^0^U!YVNQR+SJD$ViAUKP^3Lz<fmQX(B|icGx4u
z0(DtmY5zj*7+FOIf2pO7`rdJHQ^z_%0c|)lWw7<J*BDrHl?Y5{Flk?p2zrE%kV&4N
zr3*-C+v!@r85&A_Hu(NYiClhv`k*DK?2`r4HA~eeT0F_z>{a$JNF1p@EQ~V8P$@3J
z$j1j$^Ey@E6o3o1wkKYW3Rt<MbblwE*X%7gCUNXgoi9aRxy6a_K9I2-afr>T8AJP#
z5=Y~X&oAwwIKx^?8)Kg`pDZBgv+Dhxh|2*Xm(EMB>fffEvB}QWDxcpL#?xKO<4Rrl
zTv^H1H3H+Z6Z~D0vTlph#oXz5gVD)r^H(GFRvvCwEl>^L|4EN6l%FUPwIGD1ed{BJ
zf|L|@@$^!}F>DTqS4hir8x(4lHCz}wu9d;`y|v01Hr1w9E*BZDoP9T%_$#J-_A3YH
z>NYYWd8cDt=r*vxS=})whxE4oi{a<6%dh8{uH!WHwps-YflP5EKl>u<y$JVMRbp<R
z52Y_TF{u1<0DWy#r3F_D7*Ml81UHFgY&B+pWbiH1n!_B(7d)HbGZ4eBcun*=x%#Cj
z))TQ>mooMYS$FU8UM7V~LwpN4r<3FgHT7{Y?zt+U80&qP%o2X=n+MNIC^e6mI0pgV
zDo8TwjI37d-H3k{`$`D-_?1?V8q|r;D+<D@LeZ|W_n~lFf=IxeYhys#)iEo#T@T-7
z8AoWO`@Xcq5j8fCt+%Mr3$a=$=QFPPc}roG&Cl^s(}pp^4@y>0`Rzl*v!A#*sUcTX
z#RJS20=spzn~I`%%QO#L5b}yb4(S>Cb{TtJR=5dkz?k*xFJec*``U}QK$2JM0>8bL
zNtjG1rhFXO@ioVj{Du&Y9PQ^cU+~A5<T&hXY%Dp7mt~uegZ8z}64UKoNt7LHJN@&=
zhqNGPiU#)Binp?7Jmm*P4&17wKME7O3{uZj>#xtl@Kl)1rm*N`8;R{oRn@mc(gJm`
zp@949^SVw~6iK1QS)!Ts*Pc?#eIHy*=DqEF*oT|19Zl+iQ5?JUXH2O9oKP3?idb{+
z3?%@g-z|HtCSI;Q5J@;>eij@-k^8Qc%IWg~eQpbZ+RKg`!Ch8Yes8LUYc64FXU{s?
zA8c9r+B9ahgR`x|3wEk%VwFR*1@mbP@zU5Q-zVAm1QVDTCe^&*yh4m#$NQCracRKh
zSk!imu(W+?Uc;SYvp_jnlt%MVHRhi7))e1p^yJ7YF%(*Iw?+<>Moy493M~zJ;aGTb
zrkmO+E!oW*KiK?2X;m+l_saq5{<QDD(xMt<imTqjChVT_>?K0@X5?A(AO5<x2bh{i
zT%Y)lD$^9NrYd1S=;1kn3CR{E=Bno<{918cacOq9m#!p?91MRz9Kbk#O?)0Aref6*
za$JzO!GavuOkB2k1gt<1NNH3B1}eUiSkw4^J@g#9tdtUeX*zc4E|!}`#aRkrnyD4>
zlmb+ofIEQKrJr$2e5|w{jFI|<)6&`l*7j>p$AnV=K6o)qhlM5GzdVG#b&bmucg4)G
zJwQ-~f}j$}NtIS3Gr~V=EBBp_nkV^wR8MRL#t-i)ArvjAaXHm1&yWDRLAgKD{WYL2
zXZ!mxDOyc5uOX`X$i<Vjj<uc=-+~!vOh!BlMuyzbIyzJCApC7Y$Y^iC2fg?{4jQ{~
z61Gj$1o_!B^>EU3dmZSiOy1v7947(6nUr`Nj00^z0wOEn6t!*bOKV4WAaHUM>WU3Q
z%=sbPK(~FLP9YiiS7rU}CSPG+1QvYJrZK4d0fPVKn@bpUXcpQ-Z>l-~YtRQU=RdK+
zz%2B30>hZs;~ohm-KNhK*6XKryoocr*|GRX(Ddn?zSQ=9$T@HF!*-cnPb@?PRUXE1
zqP{D7gFBzR`GfxvQPrQbPkk+gx%(UinI0Q7dSlTM?$pAqX^X3U(oQU_*PXAQCj5;=
zobxzP*Qvo7v*P(ji!2Bp8HLZcre4xIAA{=i@2e9fB%o$dhi>1b3oUEip5=mmP0oL~
z*?a~4nc2}sjn>{JVUAjzTKzMam+=D_v?d#K=>)<xaY=EQTcP!r3wFl3LwVa18)<84
zt>O3KbY)UYB{4;Zm$paKDb0JmoJ*-a{dH_Bq|CmRB-h!Wq{NC+Cz3Xgq?uls;g(=|
zkE(V^8Iw-&0<AOZc*)wRN80u2w@5fuxzsn1IO~O#vTKNWSEvf!mn=n8*Bec&NrR<K
zrW@!_!p+SEgZ)3CJ3>y#F~{do)En{!H0=Q<o2?^=drrc&ZF7oPLI1b+h4?-1;R+36
z6o26grw0gv*deh3nBuPZVkko-G_at(*e#n=_;*a<(?jk&)A{A0bl2EAP!*(%q<}<G
z<XhF>>vSM}$X*YrA7!eEx=ei6)tL9YD!Pt6&fFYTrT*-bw%1wAe@sDNcUcR1YbxgY
zrFfffdiu?W#l%zRW<+u_CuX<o2-iK7RMmsurBvuX4;n)+`%kmG=jA<7pfKslzLXQl
z?C-3M&gRyWyH*E|M1o&D<zQnl80<lG6LaCbIVwAh(2m&y1iN-2h)}%rxt|tAMw82W
zOj|!FI0V^SJlLdY%`Lh9RjFPBvtA#GXuR?56+!3ARFjL4Vpz;A!c2^G!#GLq)|~Hq
z&t4WrQPU;Z@`(!i3~5##kPEL$Fu?{6OaI1I(U~+oQ%m$2It87$B4XiJnA28e^ebC5
zNn#`s9M)BBpfv>!&`YV6v)F26$%Zhq$~@oh;3zj6d6g1nF8e(wd0RC)RJDO`&r(FH
zi{r$wBS1dh(8UkT<7NJp#jNN+>os?3U=j5$%Wx1gCxde0;QoZR@2*o-wT+fw{oV3=
z2-R|W<yd?5#jY!($}c1h)D;4V%qP)+nicEM_2OI&|KRgIasJ>FNvu$x5@Df$Op`Q<
zAs9qPcqB3X;H&d}mm`$pldki$9LWXiaJ9)NWOvM!)4DolleXAp;^-0!pJlZee)7G+
zciG1*GUOyuOrjSi%a!}>&J7(~{Uw-tSOVRz=GgWhjht{lixC`*zIGI4y#mW(I*&*m
z&Ki+JMg<`|5?ob7)Z%MCEc857{^(L?{(MD`^0{5wI7E28FlyC-!hr(DdBokwmfLAo
z4s&Lyg^cQAI$LtS$IDNX_`;$+8G0!Ic4FO+UQ~=~iEB@xyk>_ON)(g+OQ0=3Kw-WI
zG9Q(?suEkS`1iP`94#PBh${Cr&zUEcK)bYv{Ps6Ihi@5E!v%ScXWN;fU9F^cu{p^J
zY=L8os$#z6sF%|N-`(b;zl@~ach9^s`EZmz6Ig4z?MvP^>(6zG6O^QXm<{AIs<GxR
z5N6q~1Qw#si$=24VeBkCe#(33W3-RoO%k=Fy2e9w3DmQ8*=y5Bt|<!-7s6>d>v_n#
z^AujI!2xX_)5M^-V3#y5`97kH|BIjL-%LAE|NCKs4=@8mgYzK?mWhe_nV;}GVYALm
z8xFe8%`gNac1{8nF>p*;MP6y730EZ=QZtdla@Wr&ZF?iVlha%?Pb|TxPd;X7<Mb(M
zCmzKGBLBWpdManPka8Z2VfW4UxHkVys=BkSA7XmFiU5Y)xT~A*TS4At1&Inq;=u!z
z?d=4M0LJ5fxRon+tHA;Wip%nR7PhC@-U2dQFtFCNNLsKPS{D*16QdVW$CDlrl^|tt
z1N3#boeRA+YX()zceA;;Q?JaodO{#1TH=l){!Pf(^aM1-2^S^SF!b_af>p))q4(Ft
z>F`&}F9-5Xr#jDd1e9O%%){HTjtnF}3hCC_4B=8P-_;pi^6gPA)-sqF*Sf6?{e_*W
zFtSiQ0T%vXUjt5L{Jql;hRJ7d&VTvZ)K=knlf-<!A{8kTB9X{_`|iw_+Q2Gag(n{-
z9`8*Ak3DxyvS(+|dLRAxy+&|12V`qS=zIJ?^1Thj*GNXo5bl($1Vd5Oyt_bO2Tl82
zLYrSt`>M|{x5-|;Zn32pQ7wbFv+}IRsqC*|mp_bt>Ra8wqOcY(4KCAuZ6|i;solgQ
zuO`f8)ei@}n{`TDsSTJ=S60Qlh=QD{g!rQtPVch9r#^h@)$L3xaSO$NAZK?|X(<*f
z7SpPBwSF38-qQUe58~tQggCJ--RY4ikJZRo_i(yh(DSFOqHKC?&9qyUU@98WE+jEa
zEZ`XXfzU~SJ$}V8Q`sF;jy%7P_%(5_zLu(*+t9afvEl&6qxduiY$-F$&&u*B<V4N3
z)w)a*`s=&bCb8zNg{ez*+^+-+)$;3+_R1`A^24Qjv9a~52iadFetU65$)Cylv3RA!
z(z;R^<0Ud)oXBBip`>UCSX@-IL)A8hpn~<nW^LVZ^4RZj$>kTa8qpw`&TxDdk?=s<
zPh|fV#608!c4#puS6iyRuOKV|_Bktl@G>2R(%7qE1+!ye-?X?A@m7vXu9e=wS9Uo~
za}X!GlZEfbKdjBVXz1gEnC1z1wuU!hd)K@Y(;ms#>0GuT)ODpRM*U>THJGQae-wN)
zV!=u{q{1#!=Mz0#&dvm<Y{oDH#MEs(<ZM?`kkw^}&)*McR%2X4C$7c0rv0eoEroFN
z_@Go<t6vj59~(ll!KDr%xR*PwK(P{H5DI3dXq0&Q<^2zfLQY&pcr{ge%=wol{j*1)
z8*hO}!kxdDidhZ0bXG|xlo}TAL@_?dT-Kv(`8-U7nZylHTOOadBn_WXmRU?wO;1(2
ztrvB`Wc>p-Kb0a1+TF)=^UtT?K111)!R0bH7`pbQMHj~yj%IS>m+s0xVu;6zUg5YT
z0b4gN>A6L-)oJ;<(_E2xHdM+|<vDMVOI^h&-K(+NQYbH3a?LM!Wq<hdHv_Q0;Z@Gu
z;4y1VEgc5^OXk;!ZR|NkY3v-?w!zg3(F1yKt_Sh)s)iA?CwJc;z(8T#ipkPzir7|@
zR=uz@qT`b+<Vb+LWvS;QAy~qiFe{K0*|9!?FT8U{y7~*+IKxho;THAj2ZT-V2=diB
z<yHbRcIV#HxlHVFj{fnb27ki^BusdLGcvGD;uh_~(D`<fzOEQCw;SS`D<EBr;5Ofg
zYYzMc3cUi64HJ;0iS<S@NE@8Ncb=35-m;2}XK1l<^1&@2G}G(J(J5Q{$xj&}j_vJ2
zkWAjPxy&h=chVtKrp|>!H<V9ZiQD4)Q8|$t*Xp7%$HEiR4Ohaxbf^8@VBUGHwdu@S
zFH$`GtAn-z=SoO8qtsN7jI#m8C@xJjBQ-*olsxMQl1J~H@ZMD*;6?hB`PCTa{(hIb
zF3S|5G(?}pVZOr;W4c`5K&e`)yAR(!R;6%_(z5(MEUZ+f(z@6(rk4tFmTPbXS;lgT
zM}G;m*8WQZ)v2E@Zwf~HU~3g6ORj{KapA9L(Y)k0b>qy2j4UfQ*=y%dPq3;+re6J>
z?$2e30V?B;hb6`}nlB^n%yVrXGY9`1NS+xgI;fm-BtO12`TSA3IMT*7ol@k5wAh3!
zZy4{c6!!W)=L2(VHpC8zW0yO9$<>nmIf%1tSk?Q-lp70ZNFW{A4nlCessv#kPMsT_
z!$xat%Enia_JH`oRrHE#=SF1Nwd&CiW||{g0$sv7I&@u+aFL}CRt(?p*Jbdj9yyCs
z9p8^>ER4-+6Yfrpa9^ys_tVM5H}Lf|MrK`N?b>gHdz*73jF2&<{IJ4@tEDIJy2x<3
z65VMh8=cx9p|pGXpyw^35sh*;mSCa43M_*b2=5`b?e^>CxOES)_WGJKi(OFweOG_4
z2cq%zZ4)Ss_@veyT-5Y3c4z8UaVn#wwoaxt_-K4o==dm?U=i&%sOdC(srpFs`w~;=
z!IxgO@bSML_LHHB6L*PL^3SVmu0B`0D{sKhW2>wZ;3B=Na|!vg0tjOIO$~a!*W=V*
z<eteJ=%Hww=sCroxtLw$*Q45`1di-evd2e3pAmbl?dt=p+uJBEZ|i$1w$WMq#%aYy
zAJJJW=SvPrVJ<H~1){Z;RTB>u-Nmy%;ZI`%Ny$bAa@loLPmEGr<P_i2e|ki2a_yTa
z@?n<)cBdWOFxMxq5nS;%+IoVOm+XN)?z-j#4v21wpmyDU5*z{(-->NRFyqMrohMR1
zgy;8<eEIfZ9%JmAHL3X<9W48pfi}I#QrP3CU>caZAMU(=Jmz^sJE;HpK6%yguAm1K
z<j#IQRgM)8?}Vw7!KU>cA5J<YyH~@#HgAFb7?ix-{nNf=1si)eN!RSfh@91N%Yb_l
z9OATfENmExKeMsUJ?BrVG9o*JHD3vP*HLxzK(sKe;F|bygyHLVZ1DvZD>$ZEOKjDr
zu)sF+3Ld#+#>xeH;2`|uIapkeu)F=H;h5E4<tTbq^U!Bjr6r|APSk#aSGQxeGp72^
z`<N#aSBC7sgq{9`l?;;-Fo+6|w1#%dOIwaDO5w9opJI%vN2s~aNv&@-v3cL%Nf|t<
z3hADUC|^K5j5-aB^3~1_x9;f(@!)P&Xf?VE<W%OorCKLHy2fXN{fMxkrb}r|ppzIT
zXW@A`^phrUT`|CAF2ZDOt`8Y)r&0?OPCg6ue@x_I`McvrK#RyxLS!TTlL%Kg)imhT
z9d`V*i%I`FeV7|2(@a!YE<>`{L#%0Nfj|-(u)q8B0|<y{8M79%5gv9WL|)b0TdTo@
z@aY4L6;XP>Y|TFA!6(mVIx!u(pAE<<mBoa>M|xB5AmXk`^q{cZHDZWU(XeEY7MT?F
zZ>B-bDQYN}NB3$IZn<`6FLh#(g6n-&Kv2iXrUsLM(ym2HS!|(`+>(y!14uTJZX&)$
z<Bsv`<Q%(ei$`?HRd6+~?K$TvD=4q57W~0*PR7(i<*oAI5ZB0kpGtoRVa(Gk@4c*S
zN<`x?A*_lA?%E-ar*^n!Z}BvS#T<5yPN%s#V5HYjFa;bqbrj7JcUsTE9GD9z-8A3R
z@I$MwYnw`H)v+t7l}>^a(~qAwp!mZim-_y|Qp3`fQSm!cYb#EFD<T-=MDZ(3gR}`W
z%>VmG7)3Ij^@7l&H=$xF!GxZi<{la%qlC0@64zZ)=`U;+?B3%&3(eA(xH<NIi0)Br
zH>+y?z6m7@cl$hX$vFsp^071MJL>f?)s99Q-o}-Qr*-|y>?0O<A<53As)j4MZZ(Wy
zpMQ7Y^k-(DNfx@%Q`)qNSqjuiibzs3$)pZ`RnBY|e7ZMx$X1IR-FA-wr3-O(R?f^2
zG&Ius@`*fUG#YVP^QB;3G{vxC@F!E(-1FJ2qtQser9jDi;uvODQtkWJ!dE7mH@_9~
zb)19dIxrsMVv3$uZ8aFUO9)bRU*6ektQcF9W?;N|9;`32zX-a9iefl<UwtLD>z<a5
z?+w?GKWsvE#}<#(Zb%6QsG9_QbCFUljwpQ(2K5#=h!Q!m^iPSb<wIIGpbIDdk+1RI
zkG`o&vTyrdqCth;!I(R-HR;`+H}RhM8?5Fpz5Dnz-}oL^7=(DoSFpqIC)>K-9U5G{
z6c;)@VvuWJ+p0&~M(J|5>M8w-uGi~c5vu?#t0moZF!a6qPc21~&B=cJo`*`lhf0gz
zHQ{5QCF|qsJ%9Z?42Zrgbj_Us=8<o-bYohI3#UUrj`{?SKiUueBW_%F$bZz7L%UuB
zhC2f}vv|!~?`hFR@g!gSl`gw@N9`-SU!ZyQDnG*<TW7lXOJM|!(>Hd%>+YXBsE<!R
z3hd1c+-^VXl9sy5b#SsFxbRL$@tLRyk*k5m9M3<#nMIw6q3y|<Th8Z%W$oBsYR;Gr
zJZ;0EJnj+i0eMrSyx2hgwKjgkT_mrvRZC)5^iW;%90dy+__FF)@pO~F9N3<RLA|~&
zOhX=a_b-0x2vTh54v$)>)yD7<=%GhH&&s04YJgbK(WKlw=B`QUUn<<jM`0=<n?Kf7
zn8ch%B!-fz89KbaHavG_!>mFXJ`b3CD^ljxAB62C%s5L78s}-OeVDs<o!Q(XCm7!J
zBH<bFl~z-r)LB}7TRpE+U<Ad&7^#s7L}o2PSzQ6W_!E>l_SDD-a$QzI7ggx=GQeG?
z?H#WNg)6OGdwDCPdxPvVmG5~M9~j*a6VA@M*|vFIF)4B#z+o)2D#Z3UllHH%%y;H<
z-z0*jspDLn!a-h-N?One3an)8h{|6%?Lbiqv2Er)d<RP+#2QeXmaZLucvgV-C10@A
z@<_Cj4W=TDQU@7*iI1ozvt54et^nNwGs2k)#t8RmESm3V=b^CU{#@_bpSoVn*iUp0
zQ0?C#6N$liEl21dCy^ZY^toPkA9*$AChm6kMXil;-eSpiC;Zq}1j>PRUEyy1>Jut5
z=N-6+{1^!|@YZ7sYQ_g;;6SGP;Mc<rQVUn5#JO>v9QDI(!^s+K6w1EnBjgLk>uEc(
z%yTO%GPp<z?gnlTipV!}sKSQxv)4+9!UocW*E6t0m8Y(>bg9>?VK3%m<2QL=G)lbo
z-?!_`Pi+}yy8F)pBN@7m)sqOb&vfg4V68I`M?q)?#j;41+Rq(1$dr&HC)RWD=53hj
zt^@WM?%3|$@Ai-ZT^K*eT%|O$!QAkJClfO;v)U7;Jk86_r_`a5{PBDI@PrdFgb9T5
z;^c4`Yd`)qs%bFAf&HV{>iEs(6X4~yZ7&2hXZ0B_UH$#z4WClv4qw9+xAAV=3<;ig
z`q|rzNf-K$>UcH9#JRK*In|KQvLouw@jBmJ-m?IkK5nhil_tOWxJXd0_@mxH;bj1s
z^xOGcEATnIFQ(S>l+D?aC?a>J_hy2Z7|Is6%IjKBl+iwCqCJvaX&}uxz?Ps(f`!~k
zcMtU_7uu<3LY?HG+_-$|1#}P^@K73G%Wx*#+T{3Ik((I-MZ}OI8mD8Tcq)6^)5pI?
zcwg~%y+-wFllrNat}rC7tvDW7t9jBXn|6d>K81U0sIMjsH)<saLCw^j24o}h1)){T
zf4z=ALZ3y7$nIV}+$aIVf$#yMU%MX8F#N*nzDE=R@)MhKMC~-7J&+`y-_q!|1oTyw
zB8_L?)9E)2YPJUw{F}JLMAvtv-)B>JDhPe=@^Pe5|F!s=_4BwSy4qUg<)e7R$INzz
zm$gb~-)MS5-$`cFhGY@_-Zu`x5P~bx<2X$<1pPOU*p&RniGu3rfM2Eg^rAx($2+_&
z!J`BGb)K_OXqH>^CVRW8_iu|y3uWE$ZSgGjXd}6Vr{+P*Z1LybHEtN%?k@|B+5Vze
zJDp!bStYwDG<l)B(K=$DRyde(R#hSS2y`oYQE1(VOv$o~X~mh8SYzwZ6I4tYXFxyV
zHKINn6C@`}3)YYe7)UyAnr;0p5)~qw^7lNrkp}Qbk6kdHA6V&b>w(s`0agy<9-cxE
z!n5zH(W#>*>Fcy#JujMcd!G{b{@P@mn^uFH!&z*!o9d}yI7xE2y-8rlpQnp{Bupi4
zyfPoz6Gjc%XBmghG<CFN5d(<AcI&=g7^I6ZlNJwT{44*P{7}ECk7y~;c5Kbg5`3yf
z>1#?edpp8HB>BCnUG@wl&&aQ)%(jSKys~WVQib49y?)E&eZr1VUFjX$KSdgUif$)u
z5tZ~Rw|LIg-ip+rV#5d{cM5HfeZyOajx%{!EYo!}J;W<b{x<xW>b)6zA{(c>z?Q^)
z&HB>2duN%I?G(eN*T$R&lD>p~&3H9pcxAi<1G*vk%l!OB<DH!udM)%HHXiL5==rK5
zd*oOqApTwJ7M`C8AxIQ>OrdY_CR?j>V}hwJd)<Fz`!bL>7v5VtA?<#DEoq`J_^a5z
z$rgxXT<TS}yvuG?+WPnHZhl(!SlY)tu5`VqXLvKep7^Oqrj>Ic#i<sDEis~B+bxhp
z5u!Xbr!5l+;}!Z1N+U#rXj1AaId2Sic1Q4g1|o(!i0v`3UW%~5%=txB@>cTw%^j~H
z<L*Dk9|djvOuHAVUUlDD6A$eij>Ug72AMW4Kkn3cLDwx&x_dynNw=cEvtGMTq(s+X
z1kew2HN84aSsILadR5y6x~6t@kN)bz6y816p1p;$PpJ?x{1{16w$Yj9V{L7bo=xqu
z`pMMOk;yUi@=0S|>{D=`->Kc;EDth=B57yNeA6#!TCXazoaWVJoN~-&=X<37KN+?K
zq{cM8H*?Q2p@O5niE8KE%DM-ip<3v;^n$I9^TZ^7=c?>bJ9%*5vuFJLqxNYp9l<D)
zDXue?eeuc^%h9R(1V572N4eP_=NMSm6r4i}Zq1ZP#6eG<NpG4Zx|8pWkQEgC_E{o=
zCAKt%Czcy0ovb2Ks}0oouUs&4p)>OL4y7hpdGQH_8GWw8z6zAAp};!O09ezZakv?5
zXwK@K>!8ypD}@494sBa=PtNGj(qQ6;-kq`KeaCzvKfe^LBvao$CmZ<4zjV>&0=f$9
z*g-+^5y|+6&RQhd8QAkeu4wp>!|pWHH2wFN+sr{t9KC{1<=`7{U!{h;<zQDx`Jb@>
z<o3@7o!CrOBt4T9<0cIJhE;7}L_>~B^ItO4k<cJ_Ia&z0Y#rHvAJXqnUSDaV#ZSW}
zVqm#cjoMaR@qn|N!?)XJ(&i00fM;36UoHQ^QB6mb1g|A<J+SCwDEIds4)QA@Q1GZB
zQpI9brdD*%L_#`Z+<r5%Sq^9b%E<Di)Uyb7{S&V>yaGL$8)z%G2DpE7fV>5*v6AVL
zfGu4fblQwrFh@Vd48W^KRjxWMyIy^vqoY$C`U%g<$hf8<M%2|Q{v6>00zo_10NYMc
zY$9RsWarfz1Ia0@rH;<}NAIBHx5aYVQyXeR{G;KIeYPF`UGu84QPGMit#T^GwMay@
z<bW6Y<|mD(@GTvkef5_iB*kvN9k<L&jF(V}`YV8u++smUPSh8DaG%ihjfqb%pByji
ze`}KJ^VnB^o$l4anclQ^mjYe-ACXAV8G;R-?imrp3%?4LPE3qpTE4p(I?}7Y3plSL
z)INIFy-D-Ihf4X&@7^pQn7Q3J*IS-VtP+5^7?k#$K(k^tRaxH@haQ9Vvm+DMX1rIS
zP_4M{SDpAvz7_lM;Cy~Wu1+^1`-t1c`zTyC$fJkA({lvgRFFe{l^!jNygDR^GN~_=
z$~AaEei-|j9mAuao)cRmT2jIS+_%@RFeXh%m7qSeSnPfUhePlKA6!&$;k)*Yy#|BA
zy~7EpJ4z~_pDPklQj^DwQ~!SAM5Iofgfe2at+lsu?%lD<PDw8;>{t^LR>;XLgg`##
z=2@Iy#f9&oMvYCr_zG!B&t~`r-V27~o0Y%d*Z7S~_E8G6x-kVY;y_o+yg(1c*T4w}
zTtM#^Wc<<DKX5X?=$-QWa6(maZr<!ih8qBTj~beg#3a%EC?LfLZX$P_4K@5oO#EP@
z7mr`}-{Tkt0!b4S6aQwIIvIWiQnR}TgkDpF{|4E`n$~D}UNph~?^%0zf^Gm;Mf6U^
zR=LR5C%*v^*u#RR?g;>=Q~-MvfuFJg{R3z>X8;^xrM$v`;|zTc48hvZ0GFi~07fz+
zz-)q{t^u$IZ-NM<yU9law}JB~Zvgy}_!rn73Xq{MjpM``Vberrsdn#TU9f&rZ$yC0
zJVmpB!8%$^F=}!_tUl)q7|*?Hg@%zR#+?y`S&omPK=z=I@|T4HY9x#PLyhIX2CHxn
zfXdwT*sS?hEwX2^Kq`Oo-8BJYpkrH|e;o@{fEXAV(;fa+`wxKJ!LIWc-?mHswUG+=
zmhaOG5M~SSh>ey;nephng6QG+^4-<BMqiVn%}$O*WZt;gjuv~{r|$FL4o?pPy=Ed*
zY*ZB+%%PS?6oy22jjr=%z>c42W4zu!LNP|O8dWVn)wfb3$D};?dju*i$<6zTSInm5
zsn=NgR=ma%h}%tnHI|W`3D39Y*(4v*2N($lYS?MlV4jOJF23&$5{El>wS%HaPLy|f
zu>67Cn0rmbFVB34+I^k|nWQnXOC9pN-T-Ue?{RfjR7XIPt7gh(FBv(7BPrO7fYwfF
z0)Hn-m*%moaf=b{yjc)DXAgKO9P+)qMuS9Uo-B13qT3&5X1-0!Xc{7lTo9!P5OpS6
zDkvEgYi!N)g2bvMmy6LeiwW*XVPd3qRkVpVjMH*&^FArsE>?WZM87Y!BZf)1B>&~B
z2T3tZqYfR`l!pQhNnuD=>Jz*2tKu~QXm13juSFA~MApn*ue=ra819M;^t0((TP0g@
z4wo$*eg0qws4M1o<7~uW<1^l`FdU<>UH*+J<Li>Ori95wpVZ#O<G`QkF+g%-1y5lC
zHyY%ePL&Cq8Q_+5MK+QGz-f34Mm?vy)k31oZdVjS1!&pJ{<Eq-srdN$Gu$qC`rls!
zS{R{s&RM2TRF@FiEv%E!z2JGT|Mq(j`q(>asL*v!47fxf_hQr}k|i~X0O!p>MR%i8
zhHU^B81QWZ*st`TMvEz<crmoJmf@|W+vL99Hc7z0i^u<639xVCUpNht4Gl?v<=CFD
z!pC}j$ijT=?p-ud@$SExxZTY9x}D?594&^WMAyM{;XUwP%mS~G@S6()b^aHDol#hn
zQ~y~HvODIn5y>%_mz1R>>&C^XQT;b+4(unX0G?(vXrq<jLd*ZU3IM=AV|u~-|EK=+
zP1#5^>C4IVVpGsWN-n#uptIPs*afdJ)J-SJ-G<9)g^=k>o0Px6g=EjX!_Fhdb0Oo)
zZ4y{>jK?G-idVU50Oz@=U528e4^#7}^k=}3MYXmEeSHa`2XbmIyd0WA*Tow4HQ$yj
z6sJKww>@|_9xlXRQq$O+TxPMXtymdgFk^9~y^=-rbl9<bK>6vg+D%Z}0s=TlYtD_!
z4vMDwVKwJ=b}_-On%^2*1^BNFfC?zoL%_r!tT&ojhCa_`9;XzyQefK=86%~>a-ho-
z=@AB{*|otE<50iBa^7KSwh9@L@x?a&O*!NE-{&MDcMs9Jx<jVdcx}%g_@hJHRlH+S
z6s;s&6hg5yN$SEhIBfd@PuWkr=_WB0R=<N{2~V@tv|@>2s#(5Sc3`A)$2&MY$D(o|
z0cyXHw3hvqSG5XndJr+Uu6m=QP*Oa+C<3mp_-V_}HXqVrMkw2wYuqp6JSZAJ6lDC|
zejC9%8p(7l4LPoo*lT_RY*~%|X-QRoJ<qZHvJ7hzTK(ArIDZw*3aqy8_|t6-oS7ld
zbn=zQdv}RUh~zaI47qHErRKNOCV!2tF%(<Vt{;Ccj-h158)KCCT|#1uM_DBV24F4z
z59=%d>!k~<#jTdLLn;FTW-O1iJpim7ckFEbVJ-UKvCjRUSbJ-J`wC!vKZ3R?j5SE^
zh}19@F{4+>mzL@!2Ck0HN1KW`P@VYKqcE4>H6d<2{*-k5`F6zxt$7?g#18(nbA@=;
zo^Sg_<8$%DaXrbBSI?rtpaGk$<%=D4?a>Fxd9a&do5!xl<NFO?_j5lfKy2T5@ZWO~
z?{BKV<EHs=g@bji8^E;;Z+aFn*Y^)s!~ccr8_10P1+J(2Wp)QeS%-q3-$`$g>Gy;&
zo}67y+oG@Q-RTEvi_r8~d52ZOy7eMqA37IysO!>L!#52v`tvFZ&5n{@g9*0#e5dwZ
zLuV4ZmJ&h^`}mNR3r3FY*t)CBqb4-XlRA@=6;)4$(cV7j@}aHkclkd>FQexVdnnm%
zI`v9);<`JySu!tFNRjg$Mgo4<$@uw-BUiz}HucUJbbL{}L_Lh53~y03wqIm37yqbX
z?i%K@J{*kIm=A~LC~Jx1{N-6zkgv-Sk7JZW93=~Z6G2HU+7;nH`qY`yPvt)}CtH{=
zYkLrCbMG&^wQMFl?jxEEP(Cu@e)nrQyQJAn<A%W0&m*P5?+L&f;rNmC&lmrNsuLf8
z>TOkyv#a^^&Ab@uW+YWr)1?!k-5v~u46sK;4!*V8g8c6Kk~3Jnm1^yC@dVCa){^mc
ze9BuQ1%jcT(vBi5baUFIH^DNBHN-`B(b8F#_^~&IpD4IH=epu9${X@3Y#WMoM4klC
zQC)zmDhPQOsQEL4vZi3p0LM4#V;fLW^>H9!=+vi<pZBZvk#^$}On8Y@zK-5(!=vBU
z;S&#EhKAP}Cm<}U4{rJpH?z+NhZ1auc3!EkiUDCNF4^f?*NHw(<!!$oqmRFD;T`UL
zWBgupInDIh)I7zQd5W97G%p2mID{FYia%?bKhm(Ktww<e0(o<yIK+!(7%K%FMUoAw
zbCM-v1*1suQoGTC*ZOUS6UvZDJ5=c0*a!0jKsuBzf`8^fO7`FVI#3hk)oC2O{x-6r
zs`ek{(jOXdFUP;~j!F`N-`+I@!$mB8Y6>ej_ir77(uZnkYt8}6zh8fQ7Hu|#IvO4a
zMp8en=2wnR@l8_n70l|SWgoF#O!EA681rbB3VD_!Ny<fgeEz9(BkwW~@dLG|-$$>5
zB6}X%^eMg-zwc?eiZ^`Ci$$K<lrCw@p=b(e(&k>)!x+jKj54MKM9OgfT)VCkKAf>M
zOliLX%q1c38lyhFy1@MT1?F$>0&&J}nva+Lnrz%0e)Y%bm*RIiqu|Rw!CvQ3n<1KU
zQk}25KCrPDQLBKm<wIP~ah!PzsmYu34Cfn_n|*R#KkKlZ7syIjKvJXqgtw)~+5Oq<
z_$`gT$(>Z?m<AeJ^d$s_|KR1}EIsMV9cy2Ju{TvQ9Tbi1+_GSAZ0QWjN$4R0+?>7N
z2Ig&R-8qXrPUmbh{jwAk?80+nFY1#Z#7hHZesHsRO8(yLxb3p`3Za%$@}RL~GvO$Q
znEw{?BkUi0*ZdcIrz~nqXA!;5>OSJve?@&-!kY{{H8R|OB-P@nxl4&|5afZWP8Zx;
z*^usO^>*x}Zd=7L@euV-)FKMckR!V;bHj4&TXMl4OY$t2JD6FBzN8IHcLJmnrZQcQ
zOVS|~1gQZ?9V9y}%CKHT{DSg%8Z?oj^@U^*p5qASXatC4LqH^7@W)smWxDTJpLcWx
zAn7(Ztdh%Lo&N|2GAjH&v;6%J;r<JRRdXi^dH~3MQD#$G!ZdC$wf&;(gZs<L-_YvN
zznjSb#cpAEtNv!ahV;2HM2TPTLScJQ0SyespGO?=Cr;=(Z0K$Nk*PMaT_1m>X<<t^
z3YXZ@1|(a4yP(m3=Ijv(cvG@Eg}-|k-W+-MH3Q;DdmR0KRgU`qFPCirxoq1X^?gX8
z^0lwXS<yisQNHT9qLTFI>BQGptY0XBP454ht*!ci*}5z<{ws<b_oMGLBhg|F2ry6$
zpD&<N=i$jMjJju3Lg6o4k`0#Xj2X~s%cj&FqF;d(Jv;t%{~FFr6q2282#aD4x+Ko5
zfopnQnstF}l%P7Hw)0#Y9X*dc`4nimKQ9W1>l*PgFPyy%&iLnB%8CIwM$xLQr3bQ0
zw?J-799Bkk0H!ueut@;ZU;i-e`^U96{{d>P=l2TW1uJ23HyEH|022t~=TQS|e1#~K
zYKYf4{y=lH=kOQxuwRG=`mEg*(96g)Q@TJ?3<bZ%07g;>-)&lS*93w=y`kanC}S$V
z|9>h_b*#dCo*LYg1ePpGb2yJP$S;6Yw?_k86<?`j?Ck^O<`XU^yUBE5va>DgZz>K1
zh8aPj@|CaX|Ay+z3sfH_OcFkh<9^UAp(Y6x;;2SR-pRX;m(#^0I~CuDvv5vK#ie|?
z&$Lr<B_F7PS?3~4qC^V285y&9xU5|h?oU@cg%hdGJhehB`IQv^V^UfY6uCM)y3z-Y
zGK&hgD$YEZ*V&Y_QYF0m9{Cdx1vsU<-aOyLhjog-S>}WWWc5X@ZhG5`{_iSUaKt~s
zE9rHXYq_Fbmvb?&YVqqIqaImB8~jI){(tZP_aactdVW=gegH7WF6<G-fA&a3S|2R^
z#k!#ym_Ngl6zw$m{KA;71@Z;~r$hqw&qE6sV2q?2SyOY76_6Js&1jLRQ>kAvDt#$+
zVX$yt7%coPx0g`*IqdrXT1ku71~7<7=z{?*&3@Q-6!^eQqW|AvDNcSsf+xqqy@6ye
zE3v2!#nU~s#mVu+yb7-%jOpD$hxcDnAaE3{7kdUzZdWNK)k0@8`;uwiT=RceMZL03
zT1!H_lsCOfc?Ia<+CL-hFV8<hG*-2(uM*$-H$hJ1Eio|V>0>u69Xf9}kI1o|CI;F8
z#3V&B>q!8%3VElU*%HG`p-B@(8zz-UoJ%fV37SXNpdG=Eonmx6Vdiq02*BF<&x+g4
z|A1v?0mjhTXNN&O!!|XZpBdf<ZS%n_h=Ysn3=(NGTC`m4ddCaz?|d&suv4Ym5OaX+
zVqeR@Fli7p9^^L_tqHB<aSF&`_!9YV>IhtW7sr06(#Lct8K7bs;6BUgF>CIzILPK6
z$H>44rs5~bcnV(>>hX0-lr&b7H%?eTCtyl9gG^_>GzR-BD@AE}=id)L)?AsLtM~?N
z;F?17#x9{>OwIz8v^kz_p#%(CdX9i%qdfAv5cEpf&y*uxcz)1|v0*>Va;ZUDaESVQ
zBTd-zzHS&=aV;p{cFR~lqxdfJJV)U3F@|8M@en!z*7s6Fd831x^CFx1&y}0{8Vo-}
zS9`#bmk%hmA1vY5)_S<tsmwA2?j!!|GBg-u5Kdzlb8fKKV}q{f$g9(%aY+~WYz4HL
z@iP)iEa8#)xCs&v^>9+bYB70G&<5vO+<6R?IA@0v1r63UYKiw~ltFx-XC%M{0=`@p
zgX^79@tn01{6Jk<Xe}{A;E-nu_L-ausOZQsuDIm+%nxjLS&6-mgVLS^&$%Rh|J8r*
zOlqD7l;MSxKR<E%qXkBhFDvaGN>QZpNs;2^kt2n)dDP`JT;xFVp$)?ffhm|lMnw7+
z>ZhAMnC~;oL9E>K5BpkIr8>Mng19K7@-RPOG$&lUkxrc0Q6W^TfWg;`+h&~r?o0rv
zy!`);|A)Z;od_uOy!s1NsBY;11OmXO#ZoySnpl{K8iUgVtY0i8ybIlr?n2*A44`+r
zf_;po$8uxou-CB+P6G_TH$f5=^VVgl$C)4^86JfBlDZ=LW%d|D$wg;xif?SCz(<}o
zcZi-wa}CjAH_$C;ekUG(L4RhMT5Qmrd(zeCb-*j<qE{XSI|E$9Fm)kLKD8*-E|%(k
zWuzA3;p>fU;Sh#JZjJU0Ixg?udj*<#;RTT2d+NI52Rv4SvuH^0pc%KGE-J0#jqc_r
zwsn=l53Q*j2J5=D#EaL(9sMNLyxshqW6Rj)+Lrzj-ON3#uTxAVzO-TE7kT!|Py=t6
zqE!g%Q;&jk?vA-nzox+cpe$}Cx5psn=rmWj*78vJYh_XyyQ7Ke;ER52^4pY%!<Rk6
zm;3XmL=0;F`%U-@`}_3D+V}NwHh5c4u**`MP-Q7aOxSl3UyBkSRQk$c2aV1Lro9PL
zlBlOyDlf~pYou%(uP95Vc+x-h``W(>5;^68IJr&t$NG%zlbr}<DV+cTckkzV=A*=e
z25<d24o1&6tf)74ue=hNAWQ+T3-NMqE!Ekr;Fr>KrF$MEG>J;?reH_crIcUte?~V|
zI@Ldif_kq1H3mu%ri_)2_E6O<y?AhEplV7a*(RK+&b884Tg-32cHrksf#}?eHk09>
zy!mBzFSB8FLl{bE2R4`R6{sZG+9Phm9vbr*#+x-)buRZ0ig5m*LDsULb8kNrZBm9s
z4Yr1k=8jN2|6pJDk^<jXS|jZ)DKz~b+<o<1)L-=H1kBJK(hbre-AFeIC?O>&JxEA1
zC@4syl$0VN(jX}@>L^Gm-3%ZgJ%G|M`yM~v-TfE#*@xeTnH%TaQ?GOGJ=_a9k<#Yf
zd{{Q8g>hP-Z<%2kyhBYzRU^M3t0nwwzszX5RcQUri$z;MMgN2oBu>Z2u(Jo$;=(>W
zI4<6=5~Xx_+K9fWyA`QpEqlCYm^&;$?Y<3GHq%UM3Sns#It*b&+<H>@F}`?D5==3m
z+@)Z-NI8e<EBa!MF8UbOmHK5`@$i)6{n#SW_KPTcuky#dO-i%Tv$$fsSpm$;6V_pD
zW1mid#YcrRMkldxFtPt#yC$CQp9`Pb)?cMp;*5UE6&$V(1T5rM&Wf2IUl(=<A5kl*
zz?`ClL|N)Q*4p33;7sy8nsai_j}FI*XxmczYIQ5!K@5?p<`qfjk#~kp?aQt#Wxh$x
z9sA3dyQMJQ*^EvLU06(74&|#Tz_XIk^;{gUA=r2q!3aKK_aL8POy0pBmER@}hPJKU
z^tK>Z8T~ug(2%HrmUE_#nX+^?&=&L=c-^&d@3v!;(E6^)>QlsJn|zeRYC9!)z!hr$
z##cE|?7=_g1qG~oTdlTrXH;~1<y_M_j1+NVl@LFiEAsnsXka$&jq|}J;w=^+iWC`p
zpx$BRnRS^j_(}F^cUFa(ie-zxP1iAzWEE^+xYML5*Xl()8x^!?_s&qoxP#P*+uvTU
zq4{Th6v&I))lkfzIfX-HWej7$>adwY07ANN7k{qhzK_?M153!Mh<5t71lDRvR9x=5
zo!$-$se#!?yda79{ilT#^_roMp_ymf2kTz%$J7ccaC>d-qaz3UUpk^V{M$l~V7}Qi
zUw{m=lj?L_+vL^F-Hz1t&2IR1o_#btQ?7edYSLB7q|4N`N55-y^~dAAHwzR<eROX2
zcuIwgt-~J`-&d8D<<Qp?dn*r&VKsxwi)JuC#q0V71HcJhJc^dUn|oe;JAv@G%^wUR
zo1vIKp}${(KyzBFWj^B<X?-j$!xPLQPk2!dYfjgCRg`1;=!i~VLY?FLN8y{pgDQKd
z+k?irTT%8BM`22!;O};7GwVWU?1zG&iQ0!dxZW{b{Wq(pXR0X^vKB*5`KGHnF&Sgz
zcsBUpg#S4W8^Z^NuldMjsi!FAqjLg))`Z!i5n${J|2GR`(A5+9Sud1BDaRv%Z|;4E
za+|{O6-2OnRA1iGcJc$6>A_#Y%-S70CBMTX1RgB}!ur<;qq_c0I3dL4%X=bUche>q
zDw+B9#oGum{mnc7M_X=&oF;$gC2#gR+dtME8|L+$lf_74#`wQB2xT@NUlCemDXqI<
zP=5{wM^jH(MvvbsALr!`_mt{3<-QDExltruvvgr1Jr*AE>_mq7x2-j02=61G;BWKJ
z8$Ey)vvKaEJFzO;?nkLf1;U^4dPvOTmhgxFpB6Ke5hg^bQD>W2-eXry4p~t~%yT<d
zJ1tGRFlRQ7FIcYTZ%O^lE_YS>RWtj(FUZ|>6YkdRll9Sea88C8izvGae;aM6-UeJ!
zJZG<EFY^_t|Kj*M3iqMo_pv`OF+zK-eHHc0v*T`tV&+b?%ek83|JaB7ug1YxW<+r2
z;W<($KGNua*8+U3=t$4I<6K1>Q|q4cQ=}kT`7n&K@!?8ay~o^ih{?cM=L^B;mzjp6
zJyh<Af895ShitP4*&-x`<;L^$j9AhlJGa*!vaBtvAlP_ZgF$15>|DMJa<?W;jTqCZ
z+zGGb&0At~5<}I~_jt%~K`Qc(=CeoT-?P<#UW|$U6G*)0j2Fn!Bnsr;(Vlalk;hw$
z1#eN~qL!I-Q(_68M?O*JT}#{9AA>P#Hyn<*Gee_OD<2M-3He<>{TWu027*rWeyXol
zHOKH2S%wtZMqYhWs2rdGUB6=I`gSG%q`T(St2^zZZ@19;*g+)g?$)clH0IrRM@pRf
z1-*D(e*Hf=*yJ4i4U!j(R4(m1eP(_>IPoZ8a$*z=!k$vw`%wtHzjTfCkKJ0ko!iK;
z?A;vAw$x6VAotn2U<`cgj|zoVa$ix=#lFh>^0e!~ESp=3QL*a{P<6AGhW{(_c78@c
zOyRRjfMH**Y7IW{Lw!T3K};dS@uAOaAt~Ix%*URYTSNRqo)-*QkpD_sOUzb$0ozvT
za5g=%n}4n1k{YXV7A6i-<MPco(O)+DY;tEWJc4wJ;Y1q2vcIKRH04pJ<S!WY?xEe!
zJ!csy9aZfN3m(7<0Du2)dS?Ist*Syxe;WM2Y9(gF70#uvA_np+;V8We9P^R`yv{4B
z!GP4y&-_q0LjG*Fc-{m?q-M;tE4;YI9k7i%|2)(nWjJVNG{jML;lj{Nsf|FseVcP^
z9I)dRm-r~MBt^D)8x)2n!#qQYI2w_!r+&(d*7uZSdC7)sUkJp1Z~njUt%20NF4z!8
z1;hJcJdkwWac|{1o9UbH%E^BfW$`wp;G^VCAqvokZ|})NQkqz>tQjRUJA`xA!Bm?b
zevZI`s@?DwyWK!^<AFtYH?**F<?P2hbzY)Ks^-`_;n+dWlSk?qCC#+_np2`GWtnWr
zcwrv;MS`3NA(?tCpIDnUG~t6}xsl2Sx>fM`nxnHLe;u2Ip!um-tN@T6Ugz2n=Tg_t
zASfxg3<rONtlU@?2U-~b?GWz@zQLc3O?hB-CLsIj{W+#VGObSDB*G(bF!(Oha>;j(
zOEFY9)h^Wcu}i3z(@S1`stJf2Sm{>e=U&)`zoX(c1RJmT2MJ(!z~krzgDSk@zN7q6
zI&WJaaMvH+w<BV_+3*2BO^Jmb(7*tC)HDM+k&ZRw3awJ`|2H{o?JBF?EC-zb?MDGF
zF-BlCb=(^kLGrnTYm-mk%y}FCu^&{N{PrZAmF@UXI)3w;K<pe;PzT29J9!pWm7-`-
z-k7$WEww-3JgQ^DDcMJZFq9zG(6e7VWzR(Q&4!6tpm58I9QYiuGOPJPoCi8MOi2iS
zOp9^eKwh$?wfX@Y%P1kC0O{HBd6Tk!$1LtBpv1Yp51yEw#kj9f`+`geCi?U17!aO{
zL|C^MkF$6y;Bg)(NsF3ZI}%%yx!-c5xkVzN-RHW%HKqNzAPyM6X6wRD5PpZ&WuT09
zovHMx^t8Q;=xeQxi1ldu12L@Bvg6H(z-3O<CONTrMoA&#H|pb!+>FXrk+nY@HMDA>
zL??W8%9i_4m>+?rVEjlB0FY?cvSN`>$|2$0x{H~Vj8qrR;6n{3oI_oV2~(a)yy=n$
z@QolY<Jm9mZO`+#LW`T%^$H2ZKv}<pf(ef*g_ZH+*nI^UwGu~6R(RSCX@z94=oXQ1
zTElFOY!%X)U|3>z27?TX>~2u-K&A@u+>Q2f{WTDdSfuaHHtGF!r~l53EI=$aQC@T|
z?(~ncrN)8!{<`s5jDu7N6@XkozZO|$l-M)eg;Vs>cil9)H0OL2eL#W`!}``NJ5x1Z
zr+I@X`0g?YIW`>(ZywH^r~LlpuUZfdvJz?ekpBE8@QBAcfP1WO17&fGdtk;=!4ndL
zHE{Rqho(qLlqA)J;8h2(lZ^HlD9EgbWD5#kBXJ<C@tE`C1v-tr%eT$}>9z~j@#?>&
zo&W{+(%X}NFWQXW{iTh$!TIsIqcQ>#>+x3&5Y3DdlT~09>7RKq4_=F#w-TJ(nekXu
z|Ni)-qpFwA1e6kB?x>&P*(~l)6^YtH@g#?z{gTAY*e(g7Zz2<t!`bizEkFYVm5L;@
z615E0c6Ot_`$C!0He1-1R-ReIA(NC>f21f*CMLJ-b6(==C5OpFMucSzLexo-bJh#^
zc?Dbn#2J)dM_y-FgRKc%bC8-J#25oF1W=gD@c>BMcqFoK8dnQWmD^~lsXUm=19P-=
z0-tAur2?DRC8kAcK8zEGPRik$a2<l!{<i0A$S2ztX}lG$-Z$d>JTA6z;O`{MQ6TYE
zp<sd|zacLJV9lPE&ixdxpKTKk6gS_@Zpvd#RGhp*z0d6dT)^nFdxr<#ZxE@$O`vWr
zLPrD);Fe~73jNJ%#QS@$qC4Grm5<6Jc5!Y)SEhXZoEMKj?fg}xT6p}cYpn}_tTIZ1
zmjJDpPn7S_^LMC<focMN)JuN_&X}}e(QP35e2O-StH^zR$tBWYelz3Bn-u8YIYC%k
zG={RWcwWa)7g>W{TN}I@Hek5f3OX@#d|0-8i^xy&z~ayuKo~MCWX(ZM-|h$5?}5~Z
zNuA?aaBk5IYf!mt8j9}`%daQG+n#H3{W=y%E3yA;{QS|C<|=2UqIuIC&h*3X>I=CT
z(iqoi4a@_gttK%FJojAvL9b<e-yfba4A7Y8os&8i*iMi8Cq!FkOd=gX{f%*`+f<ty
z*o+e2`ghwPi!zCV5jUJWpEx+%>=gG)9uq<|YQvI<-}YT8!dqS!L?P5uc{4D(2?CQL
z<zUe3^0;v%<r;~*=7AB@PYZ)JjT3r=h;Vds52#1Il`6x%KhZeD;8-o;{A{cp{=>?#
zX~20FH`L6K*@$x}Ep5a#W(aWFLRm8eZ3$0!V*@y?%}i|ZXaCh87&$?NW<_;Z>|f+s
zP=MaGuxaF1$YE44GZ{`5+@Um|=}}8n+rNMF42g7rG89VV`1My`!C57rAG$*t^XL+7
z{3x&#l9{6qF|Q_u0ebK1Tk>sZ3fktiKHOtWtGX^Ifv1nL3khdMb==>W>W6WYHxEph
zev*lX#(#p^Iwwex%839?i!uzdM8H=Z2x9EpR?nKiNvRO}&snIOd}shr<^6}F&djQ|
z^@u|#UsCt>W}G9v9>%T2(2mBC0f*S=fSos0{YLG!Ua<=yf(iSu#)*<B9Ox2=rzn2_
zNZs?^eNo3e+oi$PGRkZE59JVcH3$fTHTFBuS4lrZh}SU20Yn-=55GCp4s=WbW|T<>
z<l=lr>G;GI({=;GRPf6L^9_(#)ElY{Aw~d#&N5hlMNx<hGSD1D5#xoxZ^pC%-8y5I
zV=_l}W2^b7@Ukd#j$|%Zc#a~(f%p^flbgU{NQ~LWF|5R&9~+4a@Z%|yyauR_Lbuze
z=clkIaywp+<Rs*Ku%Cj3^fJc8GTf1s0deL?T23P&JdJqu$?;zg{w#zVYWqB^<bH%h
zmK3%A#55Q!5nyPkE@pZ3&F<*|m0`p5KqEuS``*O=w7w^&(o)G$Bjb<kP{3u?)2|;B
z%<Td6gB*7a0-({(63j8U#+wI&INrk>N5wo~^Y@5f#LPd8dsNc6p)}r!8f|N!)Uc{e
zas)6Qdm5xLQlag6E~-nEQWsd2*%tr`Z^~9d?&tRGfhb?n89upaGhdoFTDpcIB8nWZ
z1xBsT6oE(S&po>C<v^5|`5$Xw2!J&#_gU-QYp_Fqo_2iU+XGpykxYVuMY*m4QhKN1
z+KEEl^?O9Dxx2UVY^nG{#{_C+Ke;AR|EXTD)7Q(<Y;E5O)0N@m@%R@p=9BGG7Zv|Y
z>g1B!TThRomFsvA(g;NaKh_l!;>ftHvuc)VxBw9&)F?xcAm|7LDDfm^E9+I;52tcJ
z5Q9YN<kxQxB7u&50jV6ZF8YG$(!=DahJh79)Vbkv?UCPs56{)8fp&uQ)0zMh*$u*h
zA{blK<S}PJ-G$>rh5H6O+@Tb$3YM*s*iREZ|Bf!Zw(p!fs1Zquq(y#2o}w1P_oT(7
zE3#t+$*wL#Ta)x&35T+@N?Eot##Xp$@pxD$HeD8}18AiT#up>K?D#DU#gCqPB+ymv
zg%k%j`ctX5lLIheGi9u3Kamce_>$T%&{+aoND=?gg37g4wi6bN$8z##cjPz=owFR1
zu92F=h-7mW(64o;;vzBv8q!o^^=ijND1bB-ek3eFr~mdx#SV@MJ#?iR552YyiY6!o
zuhFb@Ww(-GZY_IOktb}#l(1)2fyAH-Lw8ug_ig?=RYI>9O3w|Ei(n4IyM!rF28bGL
z4czkM^z<>yeV*m2FQw?(X}un3+t!OL?e`78Msdny`tQQrdJMCuaI!8tko(CuiqkFh
z1e$Q=U`s0Z<4Gr3NxL6{#)*2H*S>ROXdJc65Hwg~qN`{jfKNe4oz>kPXfxN^0lO7L
z2V^QUM?8cJdkX8kD{v+&Ij??u!7->w0y@QMDgQA^&*rx(W^iCd67|gBxwe^q!H=t2
z<NO^p=Hc>5eePV+xg6oS*c`=w!G0cqjybaW3tpziJVJ&e#et@5su}|YO5{^{IRYmR
z8}rFuva;EbGd592I9t61A-;?({j&rXh`SHz{V%{4q=gD5t`_xd-$>MaJEUEHA9F8{
zA$=@#*AW{9f~)7)>qr1x%J8&w0XeE+l$|K{W@EYuYPc8vn1WNlH3DVu-vSE~WCPnO
z#a##`06HxwFhsh_uv-h?$|fPcGm85Y+k?dQWwJRMfGd1<blsbAi?9iSU57d;P2AoK
z&^_wK(W*a+`s%N;um;O}$4Gxvk53B#CYBuCb!nX5iZ{Fxheko>U>$OwW%?5eB2VXz
zI+y7&U*5_`^UoJSbbg+pk_J|ymiG)O^&qiZF)yUzSinj@l%8;!32clV;lLOIOavxs
z`Ssab6sx18f6-&|geUcp9+ws9W6RpEKW%QeXG;>tnX#|TUhtA~+LE$n2?E3=`CVwH
zf#11wF*g(hr*;3E95`A;kK7^c$Qs)@Xv7sgcJUWXf61K-T@qZbngzo(LSPw>3`q>k
zf(quY9Q2HC2ht!!{2dC}Zv4GO&7%iRhM`wuuO}}_u&F_|djp<Ld2r+oQuKx^ykzGn
zPt%qMuOYOw%|M-c`9Edz$C1XQ$}MA>8XW#jx_8BgBD%BsB5#y|;4kSCOWp*9o#Pk~
zvi^RL_fZ6nW@r8){#jSY9=1SNA78v*fSfU+iT7wQBE;*Rpm<;lsi5;D#E|ajAuqX4
zz_ME|bOBl@CYVLT7wv>~%Tv;PFlnMyVHtGqy!++JqemXJ4l5mswUp!u8lLZ5!2VcO
zFX!bxW?(RVpRx%$bzOJWJs#h+mOiR=e$)nf_0Nu{GT^_-)QQ#81`A4B;$EQVP=-%C
zY&?@9&Wd!otnL2}>^_3~%b;y?8#K8t>YjD;n<>>@)(-n22j4|fBFT}s%+PIM-FKQS
z)HxnwwG`FyCZoc5Oc8|e-`#g8?Ll-&o*`cOaw8b9cb=L$*FjFe`zkrlY>lb-sgzzf
zgh^hS3T-??&?9Wh5J4CyDtQ&nOFWCy2~kz?lzh<#c&NsFTs|#orbcfG{gFG4nBtMo
z|2V8y4Fzqvsuq#vgKcF8P7OW0Ddu%@)VCeyX<BJd1SME<C`EXpM9XO9crJOdd>Dyk
z{um0#yPwUxYXfX`ppZ8uRx?@$+_B*m1G$}YvAUe0omKBj*nwEvgXlYis9T`yin)$F
z)`v5+xau9AA$Q&ffiRf^>+6t~Uww+7n9zARO{CXTg*|(W#&R+)5|xXR%VRGH8*Fat
zNiAX7`rE!pYpFw4a*YWZLH4-~a7@HSD%^ocpLwVgbsoP_31fvUh3XlV3j%{T#c=0=
zqO*pkG2&a&7?P*mr^BzvBjEC-9@G`OarNk-cOve6;Ixh&%;Hb-7#=RN>}E}~LtwBz
zeF%bHGI9cWC+4z0;h$w((1U_rj5vE|HrS*DH^^cKuGfaDcA9vW7$<~2gm*wYEX+i9
ze)})G<y+4}brpy>o04|Ace*X!{*rMDtFBVvta8wi#?x3RSsGM_%)uQs-s$sHnE9TJ
z^0!#C<1o?4mkIG2<u<GBE60U?{-b8C4~o(p7%pfMMSOSyGBcdQW14UMNi%nXHH?-l
zU7G5F!AfAv9+b$&k$~NymyzVZ(G;zI&|_Nn>yfaK_`cOXD2-EC)*iSKN%=uY{qJLm
zr#UOC{eRS+@cL<<m||R(*Ycsw;FLIoByWFzywN7L`~7pjbMbKzYuJ(}25q#Yh;on!
z3Bo&4B@E0SAWXES;CYl3!~@9T(IO8M*ucU_-53RwlOwm}<VgyeA_aKG4dug!e!trU
zmf$-KA(kAEh3PY(2cs33a?&fNNaQI#m!^q`@kv*l|HEkCzEp<Zm-Xk_0iSDzFlWg1
zJ;%o*n?XF}prk+_HHox#7gCHPh7jZVn$3OP9%xi0Pb|+c^`tNFWz@2Fk0J69sn(>Z
z9@GhlRypT}IA2Cen#cedq^huD3Ub04>!5j~JgK*Y9e`Clu20ffju;tnq!OmYVmWH&
zUI5$byX^wv<Ps~#!Te<ph0#oH)5c01ybHGCm2xQl5K?(16;9u8icjTN{5-60RxrlK
z=6vP{2*`65p!nJp?Bym@1}=JvP(n091<gkBr|{+hMXVRn)P-actV(3-U}6Mp$IMT-
zL2@!3h2vfOVfQlYQU$B%AAGokGC>cuo7B;TQ2JesM|j=N9gb4K8tK;u=KS)(xT7h0
zo|8CoV5Et>J}8S+v;YXx`>$V6JZ%lfno}R}?i&zQTe=XUvf~>sgL#HtZSe$?L=?*B
zof2b&Fw&UvWa7r9Z?x`Ce&ki8M*P8`4vh%ERGNTBxxQ;oeKFQ!vcGRQ|1uwoIoOCU
zd{UCG=58kz3jb^xa~X5IE;I>Mg)U{MDbJowl@tQNR0T|5PmTb18`*|(8F?wddGX^i
z9KV7Ip&KBeTd_DgFB)kaUobO0c;HKTWoef#)1!qE!BnDC65mvTqSTE7u)a|H<G{^X
zWl0ZpGNGc4=*=t<c+C?mnSb;C``o^L)IOEia~5g?Pkf$4#(<$GGVDZJ7?)~CFHiTG
z&p%BaPe@_FYYbt@c@^5>ekM7)B%6FROMrm*>mOvH3eeLDjWMTa+po?E8`eN$Wc~b&
zkZVAfl`tjnL1U(FKNy+9SVkLwxoSi9bxbp7{!$=PgHRAQ_gf^nu5zMASJC`IqQKm5
z(*|M?@VZB3h;YFDisHh{tj+6QcF-iWhuZHQErtw3ad+*<4$m5{vxOQg`{SR^imwO(
z`mO72x8{&48z!A<IF&D_piBGRWid_SHHTsD`axYl?9Iq&ZSmJ={CX%O5cm1i6jK87
zFj1YVzv&6&4Rp5nVS9^*w~4&6sgr<7#}c0ph%Bb$@EqU44F1{gK`v6c4;Sd>(}$4N
z2h`Xp8z-G(Z`&TWpajt=$gmytop4YU>4ovyzIu&oP!)ot&c6&}qjz=duJpXX_Ra0N
z{!QR!Q$j_S#h5*sX$87#&>sdQEe;2HYfP>=p4W;|O@LHMlltkp2rqcq2C`?8|AQ+5
zGpquOh?jrj>xKD<8uyql+}KyaWatn~se3^!lMY_3k8UyFm(7&>`z2Q&z)uuyav<I>
zg(EtW^x7>SQpVDwr$0i^kj|X>q~8<y2>F9(0U{If9&*}h8db|U$~dt^s`6d}{{R&B
z-&nvrgXC!BaQz=ytq4QPAE+A#@<;}KnoLlL*3AHwQ_D%e+^m<$tn{&u`&HfWc*FjC
z=LRRL$LjPiyXN-GKp+eu`#1ctrWNCqig_zBf=U=pgU`sy+)P&iJw24&2|c95`Zz`|
zKO8DxI_nfbTfi)-&5t5J^8KvH^C=47-o};b@4{<J9xjr>@0F}IUS>;YEP?|bx_;E0
zs*bh7AN#`FfrKEGh2t{O`atE%Da+3PxK>W~^STLg+=&T$uA0khzyo}j|Ea?oYJO<m
z8VQFmPsT#JGUz_I4vrPTtUX*wZ%^65{Z8WXC2;63Nfc%N^U93H;Ec<bEaGVn<N=C{
zY60ous~-a_U*E2}7le|>aAV?`heR8`Bc$y9(}r17H(NE!x;R?%a=K5Hoz~x2q(+nv
zZ>4qIzAMqm6G{<Mvwgip%=ugm5Cl1p6p8(!uY+!d$@h;`Q8$emWmH93p{#kk(<?YV
zyo5cWX9Atv*k8l3W6DtnPj$v3{J`$zaCBa(B$_GOu%E=hPPpi*XBvmkc;S9WCNCG@
zFd$^R*#UE`11)?7W&~3uTw>K*%cjf0pW)2sWS|I!5UQ!fkkPq~?LU?;ER1o7517*R
z%%D?OsSEu<_8dUkWK-+=3Y9NJHAm1kPM!!M#!G5Qf>)iJia<;>0lUf_>N@2xH5zOH
z+$#TxpJo_(jJ}h=PR2;Fc}p)~FTa%TWGLp~7bF}x*jeF2iA+Ez$HV18uYH4s>qpf-
z5@Q4JQb0t61VjK@R(!-qpX64JKVv@h{->vsubwslUKExbGr#t%RIJHvx4sd_7j$D5
z_cI)RvV480h<LykwPnD@NlaD4Pm!^cQ(*;_r~W+58Csh^g+CL!fGGhkCNf}fpn?y|
zk04lyIzEJeWc0qbqiS8!NmGtVe(142#t?NMZ46-}_;Ec>ALZ7jafmbAbOEHFS%jEM
z%zbq;AG`$-P?zVR84bdl!l3InKv*gSz>!m7lpX3;W560cX{F#0=fVXC#8zq`^cccO
z<9D&g6C4uJxbd@M*1`~X7-s$HA9)QuAG;978UhUm2c_s)AVsm*0KKeC+B`gCPCdWt
z<{h`7Tv0lxS|nqP+*A3P?;a?6vgpGi8I}_YgaP1-$jz!APvLo8u+v?_ub&0&Ca~|l
z;|*bjzv(T(3za_IJuraNM7)K1#55(af!^*%gcV4W0KrZU&cUo4kexBD?`A}dkRr%L
zzlApiEBTU>-dTTwF`l=!0B_m#QcVJQfSUqAPD6|7ylCquEwWPAgGvE=9{OQ1$x5FY
zuak|lI^atX%%}y$XCHD_CUZhQ;lBUIFap$5(0f1V7Ofm;F89r1xW*h2R><)#unOkW
zKWeGpjCXnU`G9ngqQQX}g8hkrsW+UJ;!XCh1D<E?<1b7DWH1symV{r@au<ezV0O+l
zy;}ddH^BWgdh`d80H9n+YuU|N3G$yflZRcZI@#pd_}a*N{P23e`o0V&iR&)$bpov9
z_Y7^SA)c$qT)@|%wjx`4WKx5Nt^nJNddTTQfgA=ICa^(wJY_J6s2#Lb<KBG*;4&Mm
zE)41rz>hW9ZPE#79cy!rE#S(}ROkY62`XII|GKTcy11<NlEvg+-KAemZd>cXqZ)R`
zvYRJnyU~@6oC88~VnGT3kut^(Wx20-0@eHlcLl7b!N8W>eL2P6?h&jcG~&L=?aPRj
zga8-5o@o1s=jVUCIo~}O!aM$n0%_2U8(aq=rS)<smW1!Xz`-KGh?4(Cyz~6m4`tdA
z#;IpHGcNXJyW|*6^z`oHJEissE0T_E;Q9E;t&Ro2v{IEx$w!WU;UnPIAN-NLgR^jI
zFN%gofmn@M>C(nzGfq*2s0`b$OH{Jr{{XOxw{+7`3}X44ZQ|2+4zPS^M9qU+q7v*=
zuqkFM#vG$<q7cH6j2ssE=ohzB{b-b!fY|hPe~6G$1{>%%*&a~9L!A+NXV~Hy0C_)&
zueiw$*_Au43%ga`jPrv;6Hp9fa**3<xqiIRcI2uP$2uAozp!}nONNLMEI@O-i9Goj
zE)pBS6<XTyPj-O-8{{ao?-nMXh`d9p{HW2GU<lIH*O&Lq^DdP;Bnl{vsJADL>1O?U
zDit$L9%9#Iq=zhc7#GYQlo}aRFgh#;X?}{(o-4Vv$y6|NWlC7~tFIX%#!qokq`}Jh
z8{RBH;WP)W(p|-}F2w%tS^y!qeyKs>GprZpozYSVhJZ-T*pdXVg!=?DAiwWoh#Xs!
zjZ;f5AM%-SmN}m?Bjgawb?`bq|K>{u)Gmq`B?I_=1=tsc@plT`otI%Rpjot*8Sj*S
z;V8TbBPc$Xxf1YXmYC1tJiAO-w-<?9v*hdQL9H)Nx=}ZlUBLQ2e4()?N_jH$^2DFV
zXduAE88i0jVB{Xn=z2u*W&y|n6inFVpldsbfC0*o%1Nj&AYSSUz?2}vkYgITZ56sH
z>Z!DxLn~%|08v><>##K7^?W*;5FY1J4T%u){lohFLNrs+7~sh`D~YYfUzM;0{xU_y
zu2O3K_qPw=%Z_#vWEd@?-6bwSm8{Zjd}t?@XE}j8QeclgPSvR0m5a!2VYWY!C(Ikt
zPDYbsAy?iQCG;3*MB;OZ<44;!8`13sHz-f2u})o(6A*$Vb{uBMACzt4gR9)&F#%oc
zV|Bcoh~=(8ZZYMH1+Zb{iL}k~@mkfc@^{tvXhPmb=gsX4$O#9U^<01^AfR~`mg^;T
zF&HI-N0+`#r{p{dMrkY$r?eX!Fkp*NK4e$t&0Lavx1u=pHf0N@t}G8(9Bc1yE~bX>
z>-mTfqgKhT7Nck}*DcrZx>0!iaCkom<%^y>8;#LNt)YfJhVyVNF)K)anZhC}OY*sy
z5AkE7fCJ9{*hO0~{oVJBC_P`!GI*$xr2@4)?DHEc&r2nqpkfm*95|S>OyW$sJ^*Kf
zd1ApqkVD9-bD@$w)|?Tw?6z<^9y$wprg6=;n+9**+yV18MUs;d+CdkR*mp5i3EpW|
zSZhh>Kiwk;cKdx?NVT|f2W^7%YvEN4dM8mJKP$5Jh6ZxmkZ+jQ9HL@Z`cMv2%ULx(
zlT%W*1pr46o$?*{YnI!ft(?{8k-Kw;@!?S&1{ZAEV-p#r=*y1>=b_sW)@@FFplIO@
z!caRFQ4gIy2(woT1<A*RXd=|=$U?5Q(wy)~y6>x>zCe9KRiFboMs6)+hhI3o61TS#
zx7OMYY-RX8e5ZymLUx{WY8<r507w!C)SE1caDw-0UX@k;g?zaD7Vk32Gq#woqDi;S
z?tgb$e!Xz7?YRnq4?ne5b)E}b!D$DjJjBX1K7iIK1rzbiN~x9VM06j9vszC6M3H+I
z1)#As%Oa?MVPi&`L<k5q7#(gvgH%(`F&v0g<uRYu9Z2tj#(HsvzE2`5RSYsbr<AM-
z-Po5HmlL+n{It=zDG5GN^w+*G`a-NMDdKs_r;IqNApw(x?^o<QB!6zUj{HdixW9R)
zX((PWWAvB>g%d;ARVE88ysYIclX?J`s+fyF5Z13~`|x=~^793)C@viI7AnKq+ii0|
zte-JA-+)a)-G*|scipEFb7Xx{c4XcD-g293N8R*Sg(rWn3*B0bGB!&k_M$#M>!mCk
zghlOeB0XLHD;=m722J!ws4~h!TfK7*vquyf*RvsBRPKAb2DS&$EdG=rrH3;4b`MiI
zS`NFj`3#W`K;$h?l=DjIK8qtQky_lZ<h!~8^kaiN(Zb%=>wn9{(slSk**>=cNzQH0
zQnq+uu$ZwN<-eq(4geR`x~Qr#w(?m{5HAp15wYe7t1-96_b~-K5HLW_oeUW7(3e4G
z*GVp|sne7pOH7z2XldtZijD-dac`|K&y5PC?*m+ape)wkN$55A^D0zYnC%U5DIy6`
z^LjQ3!q0qZ(l^HOHA`a?o@}7rGw(Bhp?WDV`JE3;n-(x}_`>s6Fh&@NfxwTWoO*T@
zP%7x@)SW-7W9-3iZitTNp#u267B|1>fZIZsz{_RtSlooxvuR&gv;bP7KTyfv9BWr0
znmNpmigsaF0&&u3bmUS%^2w^QDyFWT*a9j3i2o`hMDZqq)&JWCwM_o(uHj_2$v|y4
zEjFy|!o=|r<_@M3m0L||yd(Sv{Bf0JPhT!J^G1dxSH7l^N*%p_-G0|Y=L`}+mZI67
z52K1^k&_NYa~3OYQY_f83}oy2P?05Mpx43(3^>pH7+G9208W^-2m`*K>+&E@1!bDR
zQWX4NGsf`U_5}x47Qv2C26B8O6Zj-;zr<zg)0YI8r9Wwk)_|_fhK4hAG#9{b@Fv?d
zaPD?~p+>DulQ>2@Y7%un{WWj)84rBeynaCP8XnUb-?*-1&4C?|_#1%ydyn?HVlR7J
z4`thFt&MmRltwQri!C8c2N*OQ3I}!b`jUi!5tbFy`krrGsAG$u9Gts*nQT+DG$&p}
zz|%$T%=jpM{v8mg%%Idpt>ti<r{wiK&e!+gZgQBQZ?xvbgT>9PD03F<E6?Y3vXZ(1
z6DoWJ#f1#ApcH7C#)<gyFy?JqU|*7VP#M6VZ0ICN7uwBJW8c~~faGnmX@ImceNIHB
zqkb8!(qXPU^2b)rb6Vl&ECc5BLA^?*DA5iEbr~9c-Iqp)`txJ#Q5+A~haeO@Wy<d+
z<&b_8{+)tEGrEAiV%2+JnwIg(gFzThG6`1&vl1<-Vis}^)ID2Gd~sag#4F1AhBZzk
z>%T;Qv0ar#@l(|hvIolG6-~RgXyebrAp(=~8}KlJR@RWjsW-SH6dQUd8o9VWq(a$R
z&`OL=Wya3TPK0(sK|WkYkl+D&#d8POkaB}*7Xa2Vq&j+rStOUS0cZ<d;G=+8oL{ET
z+`2|P+8O6hb$0P_;re&TawjTTJ-X&0H`X7(f%48}068#|yWRl7aA91WJo9U*4u4Z*
z;v($p>s~GA{Bh=xV6&o}dMQ8ao;TMf06dQ&07+!Q&fpC4-`xAPMUEjWIjW<T5IY3Y
zEn52ZWIViRc>!ua7XdJByjnC=*CDKm^D3fRFEz?(ed=)tZ@&#&r{_VqQ{Q7BCS~;;
zszX*a9ai*9=2F0?sG@wyhNM$EvW8m~oTzym<*0iX^_T@g^sYT<%%%}&7SU7dBJ3a@
zh;z|zT)eq%couVyOW9-_yU6+IGYowm(?;bR>*8C<gRiz>F4cSXXAu(pBI>G=ddR28
zTIv#=k`RivP`9*DaKH09{Z!2lJi%R-TH!kW^&5&(3aD<nZl{Nw4~UC43h2w?tuokF
z&$+p+m6Oh1Nuxx`yw98%UqD(54j5e#&HkD#Y7Ma^yARu6f9SsKXnHrD-hKq3*m;?4
zz)%MH2>J*aB~jJa@Lmxp?P+g2iy|5Zr$SA?kt1Y2{30h>2Zck<%@uw^mvO(!OB_a(
zkn0y5cqgp@f8tzGhro#&ly*L>Jw>a)JtZ57YirNEZ*JRi{Ggm$oT1Xtn2QZ?9Pd*U
z4f!d!QO{ogg2@pa+lKCjnoa}#B7YfUY>4EjkL9Q1<e&Ue)w5^6JdE>KM@Zc#f5SKv
zUGYgn!`8o@rGCfv2;<FB=Nd*pBu-_yqYU|eOE@^p7Km7(u>vvBnCi;>57uD=rpL3P
zR5YUrcprcV@Bu<NRZ8qLfJ*=LiJ*vQ9w>(r$*PH0ldM1$zdeysLqoulKZY1cgvm4%
zW{ME*8m2z}{4744=#ORyd&mv;@1J$F-qir$U$TK!VvfDrWwEJxK7uM93V&-ipH1an
z=SRj$utbcxxh!%-V=2*G3G9XusP9nK%fCpMV>^+ki$43CZ+sE<Dj4q#`BXkI39Fd=
zl^x#Vc(^FR1SI7YJwJ3;_9vWMc4VKnDhK>2*VWjzxF-LN#+;}cR`t18ivAE_BPbQk
zV=d8_bLW*r+({#>{)tE8pOfpc*$IoIA4l`QXO!HJlSIdyUoy2{+EXaQFA`b}0#zlS
z;Xa~|LKXbA@2=re)9M=L3yoOQ9Dt%ZXx^<h8%Zw_|F{h!%#QT&Wh07JsrE1f2N6To
zQTuZ`jtU7RgPRRg={Nd%%SW9HJs#t&kBQEvMgGeuTb&N-?^)x;)GWUZb}$T1g_q+o
zN+Wq+_=#8I+a4UdA?%Nz)5hQ+_5#7o7lN1aY6U<!E30$-6L2Mzg<9!U&hez1Wvi?Q
zocjTqn#=!gN0jfXfWx8GxATt`X=_L3%?b<PMjF?B7(%{&zG{se=xTdkPv<l|gTQIF
z8)B8iF79Tp#C27@khNq^Sfi0|Tc;aQgQfA8z=UF4M^WT@_MoUN3W9B&?d{qlwIA1(
zIX{&FH*Gu~7jo8=gnt@m05~ButC#mjl+X9+UqXySfNOHd)m<BS&MoXrTQXS*Oe%q`
z9D4Uhow42c!slSrO?1k!C(MmZ)kN_0_=}a3JJxNii}DkRQ^zrlJF}v2jcD9p43KYD
z9T!3VIu|8-$F@y4mm3;kX){6EUU9j{+l*rq=0y)iQkI<UeSd-O77+hX!$u|*fBt1#
z%a-d_;ws7zjf-4|h-)-dDeMQmlE}@uLdZELGmG1hDHAHgly@A}dw#Fti}-`yq~8VQ
zXBGxLD%<u7TpRqoinboto=ShiPwJe|`0eQ0a%>!G5Iv?yo-^-Y(ZCwQrQ-V(aBr#V
zMC-W5qb<LIX3Kk6sCEh2A4+`=exXddgrioHY37e|m~>wS>IoaT%>{>w!vVqjuEBx0
zUKfzr_Z2mMfjin&F=}WO!=4=f`4b2{Zd~N@uxO57Q;#=+m<UWqXQ6&mn~KVE&W_`5
zL|{<oYh|HI%G3tRL=w{7_I9!2Dgr7k!dRk*q{PH+X-c_9{JHvW8TxJ`0Y!xyQOBu^
z3g+YP@84IpZ!IhiALf_0Z{@!a+uC`5Rd8q@C2mc5^~^0iMqKlZM3QI8fSY0R)y|!3
zd}hQxebH(Z8(f$uC(Ng%6#AAu*Jm+#obyKchsudo-fg$tQDzv0(`JqlACyl`Es#+y
z@wL$-pM8t!N2b!47vY{>8~m9>IS{3HPXU6ssSmuvTyn%6qZ?BTS=`1^hipg}-(Pyj
zediWWPmEH+@n4s6!?HF5;dq_5`j|I_^5I{ok()r*43vI9a+22Lpsx#h!Q%pNzgMO+
zaxMM+ng)jZ6~}jq{m_V)*9N1GKkt&g%bK8J17}kODa}5Tr7lXo7gJ3t(M$nZ0SfP!
zg-9!)udz>1>+#y1SJf@r75y+IA&(C%D#<*+%aodt#lcHO7on!-6+gTmkkjeEbl>Ua
z{-b>6T2*H#vVHJG?bZkklbrv`e*Kz%O{gwTZnx}?mM)4=OF*ZSF4ol?U4@*D@%gr=
z!7Q+OXyz!@SX<DL`kSbAo+7i$=>o8-ny$@s>R*0Wc8Vciu9HLwVvCuTiomK9({WFg
zL}%d<^dpn;usE-TiUS@Q3NePEaM(xzIt*<3C;Dvyug8)BBW>ex-<IJH;T4BVc)LSc
zTQ5ISS=6`>Ij%LQT9m_A_kIY<cw2w3g=1s~C7Q;z9g%x(9h{WI<u6`B?ZhxokDdWH
zO+2;I9Jm!lT=Fqt2-%|y2Sn9ONfb?i{Gug*f^@6j-Yd~YrP91z>1PM2%H+Jg^7>9w
zr_vnAjF73<bE>9(S{bu9*wUGHK>8p!luJZ^lVgJ!lbNshB3KD=iAni2L?g)^$$`*a
z;HpcvqIDxh=oG+fc%6<(IU`9=!lFx6HZ#3n^(p*dj88#&y|YVwi)JRbgwUq#^1!sV
z5y5eQ8eH?@Dg!YSl!30<c#ws#A^bghq!CbT@{-@SNn-rd*mY{26}Lu?5Q)X+zroZl
zq7r4GTP;kubDJ;I;VsmOmnK0iM)NNxXn3w_ArrFlSc2Oe4#V!(ewVzY-r}{>99#8%
zQcpp>MYv563^vsWE`F;`cMZ5tHIh~jeXTYPl6X1tklPXwTED1?(&Y;OaPGR<6v&UY
zb+WLPlFBHH{~34;J<j!1XLExhXYPycN7vjX`Ot6}LwWbo_hOI5gBt6}o!Otn2HXPA
zd)f`{PP<VI#R(Ntf4u9-5}y;vh~RNC4X!{yc}00Pu{KT&E>dMN0x>-agt;x6M@G8O
zTk4qbtGV@WHDHOr5oSzzOi(FXcuaSUx8HI9nCXa_)2EL9_^*erPCEpamyhj>r5vNz
z`^pLXV4%UtW3Yqct{@}rU=S;?WiSg;GmXu2G(Ze&Z(3z(z9Ww%?%nkp-XbfOHYuqN
z<wDpzfwn*#1emKB^>p<cM~M7)17svMKiEGrP@&B?l0n-IQ9U62PlAF$e*~03PYq`D
zq_1kP?@SS1RKK0-AECj}6idC%Bwgx>@j-n-y-j1=Gdlf%oC3W*Z7;OC9;~Wt>1AVM
zBt7Op#tK!1O!FK@3}tGOsdau>p{k$S8#-0N)#<%OKR9AEi(Yb!4=Bou2~QerDv>Of
z!dRo+9Q6}Tv2<9q9fBOW|B+;WYR8JikvJ`9bcjcfEcecE{cd)}qN{EKMcu~(Se&ue
zGvH^&PvC`_e&PgWoI9L4Uv@z3{Xms)`$ku&>O%U6u^|(d$6{iNN#LMr_$`#vhm#rC
zPd+oWprjoPX(a?ST0p#&tyck3DrOR-Mzj%(xw}<wBddS0*eY(S)C6=Y4Rz}p%o2K_
zl5Vj5hJwKi5G3d~3goV}rfs~eRA#ifdt&jfo+TAy+v+sSBWMxpZX9HmPw8d}`4q`U
zfEG!qD0&;&$gEt~`e5NDu8?zwi{WQDK}c^dV@5m14n<uo`62ft+9t{y08iIcip&oW
z)nty#4cBS5jzmOvrfw!(s3{+>v%5{xAy`V3;>V&UsS_8nL<#Mw?u8MxLPI$HK#XaV
ztC|B322A?_VLOa}dKHips`4n`iU%J-DSr1@od?R{U^c@W))H_mN?#mgZ|~e=jOxz8
zHM5N{J3YT}9rLx#85A=K7ONYjU>bKM0R`m=g4TKM$T|{o*PFtT-8r*HA3x;k&0DDY
zeJtud_1k=vRT7Z!?MhU->qs6P0dE!XtzMrly}iu(JHjzZuli4jU%xb65^%xQ<DES8
zwbH`3ITqZflvm)48^ufA6WX8Gxjxo^yL|P+o5SBfn5Tpu{QyUrpv}xuN+Luzy5T^;
z8ux)d1OT3<(^KfIGsfIJM6ot*g~DuvSH3C9I7^@dc?DD)$Sz8`j0Om=Yh>+!RA1q$
zYq%yzt`AosTgBQ$7Wuh+cyzk`2A#W1(hFWUfa(C9WATCwG{28h=Gc3%ACvYC$-$#j
zYMBSQ0yp?Q>NX<P4o=gn4t#V`I6=h*SqH5{SRTx!H^*KJ4OfsL-?;5Wx6Fg`b%*Pj
zsTr)GSaRBKc>)j7O^=GITI!Tkp2jdM;mA)K8jsh6XGtU2N3xObv<2Y5ZrEgMDOfYQ
z)<gU!6OhS)Hb4u+s)5Bwi)@2_r#u4pJWz#WNu1sXej**}$~NI)=IRUG5J<QkKMQF8
zIFVWSHq<wDag@f1y7Mq5F}~ZplOl%Nt%d`^I<kTS6f49OL@P_6Zzf*)6=h1r)-mpj
z^!7>|mVsPIAbdjMSf*azTxYIxo9A1|(o@;OMHRM>CJXGK6c-;A#=#U?1>7DokUx{+
z<iM$ad}8h1m7Kva)t2Nor{Y$Tv>3h`u4x+lgRLjt1nHI;0IIAlO-5_P2pcL<K%z;f
zYj(5srCfC@L`sge_4*T1kco#r)=%K#{E8Gqwe5TICFSv~qo&lPyDDI1XJ{;Mkd2M$
zuit8KKdgwkt?l@(wS9|Ei8VA-DvOW69S>C}&1V-ohZEQ51lCuBvL>gpMlOO-KM72s
z;fFqybu3eP+Pa^VT|Of1(tJ4;H^l|aK$*i8eRtPq=}-5baj6&$P=(0%c_`x%B@_jS
zJnln&?`aohgQi{7&Y;%rQvFDZ05Vlw_UHV;JB4X3%xrxrb@9(7z>FipXie*K`j{c(
z{9HR+S6ABWDE53wPJN$@F47SC@^ABYsPcQ~4tZI$Y$xv7!E|0k-bDeDN+N#S4w@xx
z!Xxw5&QV?_<K_d!!jq%>bZnT93A5#gKE;1iVnnDLw}Um#7#(?j-j1e5us4c7?M%`W
z{<D2?OsuEfgOi;+0%~DLG~D1w*ezy0F_<_vL!>D4eM-$UDi_nhH)KrpA=`%XGMNKU
zw62yEL{#7H_Lo(ZP`7e9s5wgQGbyEByE$DYxhb9TOA2-b8GhM(8FiP}L)Jqwy=;9G
za6b5%w^LiN5nD((Y7f`<##D{Rh57dslmx!MyBySd_V*O6cV%y5w38#3$xXf}g}RT?
z>tY*f(Y!O8ZmvYypcVLg8`70EJTmptZ=J^D!flJnN;~3i4UDFwCw>UIgBqGp6$Vbp
zh+`vEJyOgSO}E#WK-#se(KzA$h)I&Z;|m+eg9*E$wVBh97Zd%RB~vD?Arif~zjkrh
z^85ipK2=YyKITEJhju3uv`w0LTi~H6M{jh^E^kg<y~fDrmswy-nzqHsF?oy_rpRVX
z&o~=gTb6O$g(diT(j9H0y@RByKs^77ko!4jV>bAzY(zd<#6E@E$?*%(&)%+A6Sa&C
z;H^U+SwZbSG<W90ZbbO|5y>=T$l`aogBVVCn)=g-V*9@6eorPz!(~=Xcnob?FiinC
z6}_dPiC3Xt`YAwvW28-N9c$j#SC{)o9W|K9T><@(Q_KTK`7w=Yk&TVDYVrqv2+C=r
zXEo7;S8vE*@3@o<Etr=MgA~kklG<%e17Y5riZGLM;5>Z3)+n1*QcO|VqG`eG_rB$A
z@HIVVTR|Sh^IAT$_Ax~V6>tPYYxU^|HNeS$q7J>b=U@>ZINV%H<#BCEA$#i|POvK$
ziZ2p(T*!u6oOrMHrx}h1olgyf60tP`P769VN@|q?ZyW4CY2P_6(!*Z+sn^XmWvvk1
zYjP{M?U8d~-<}cXafDLQw@LZ87H)LEMx?ZQzKAW`K6LP&{-wO|FQ?164Z4+GLhWWR
zT~peCoV8^!uH1q667Kqb2^deifeA(RD+K+Szwh)0^-S?_2D+t<GVFg;)?3{HKQUt<
z7c(}Lobe;wmy<8_6F8XVu2%oq>+jwCG`a0NO(yJQ=T0f$*DxdwA&Z|ls_dL5F_Ow<
zF?Lq*6gaw-t4lV%?qCc2_Pk`lP&ffL?WA?w4uh+a+b-Jlb$`gPl0TVaR@&p8pv>{S
zI-L>zs&FcU=58I7Ju{_A{-y1F;;dy*(EPRQqK*GxXX*U`gQOo1b@l}9dhpf*Q_tet
z%Z&(CNzZ7ZJ*pEJ0yl7+V@!L_&SM6nuiyrh7X7P)ysruyS-FqOH7#He*7x;n1Qb&n
zVnECj1<0r3by0W?kJvBM^3I|bViU3LZ?e4<E&FYTBZOLCqO1RiDj`$<nztjTOZ0Tj
zlWZM~ODwMY_9TQ|x-gJ$9s1uXIM1dcm#8(y_6!aY>HX}*HVzaXw;;Ihhz(Wh4seuI
zu%fH`Xfl6h-K#XPS8;y*z|l&(bjZ|eNu|l@W#79#oieUWCHVWTZP0Cp9we=pn~2BN
z$U`CW9+rfhTf&{DfR_Nrp04R`mXn1r)FTj~=rapt#>MR?I3GW5o!VhHN(2WMROfY7
z@8x>{JK<havFa4p&UY{jq!Qs#{FJ?>CjNHhO=td^QZM6e>%8^a_%W5g29F-USbqir
z@@9kYNaF)W#*Mr9@2rZ{2HeU>xm}SOX~Ccsr5H857H)!1;}7Ptfi_DfN9Mk6d-DVS
zZo1vq9j$foiBTB$9iU?a1weI8aa}T5p05u8=k{rq(@Xo%rDe}yex1{K#o_D);}uwe
z>jznbeQz7D>l~#S2A;B`6*n}ZiZ_ocB4*W&sWYYA$i@I~8~p3TdmsADl-zNxvpAjT
z#?xzWPQwgBd3q#^SKNelofQ`;Aaa^Vh|BV{4-zHZ(HDwBJYOcMp^&>DPF#kmBsYS4
zDVlCQHAeNnn-5XC8Hm~Qt{3si7T=xy9pNGH^LKPK19@MhXku<us$78g^VY7Xzlq*2
zFh}Z`l;4M=1o;%z2M$`LqQ}QA%s>VDWbrwBGbAgg&4uQ;EaDMWr-COcr#mlDqrqeV
z#8<%rh|{x@q;}H6LZ|1gRKtu)Dz3AWopgMHJj~x~`E=?g+c}ki`-u(b?dD}|hW`HN
zQXiu3&!&7^i{?G}U1)6ETg$2QGnp#-P`pXL-%<GXaWcbMUgf3^)@Fk%rTjtQH2GZ4
z*SPRJmjnaVHN_2)5$@vM_gnR?b~f%_>l*0-q=re`wj-DvQm#(Q21PdZ445S|*A-m6
zvc6ixM>xkpB$xkcZxNzAHWR8oLlmKg&@X_p6R1v4Pd^##$LKwW9X7t-T>#IK&gFe=
z`*EvpIpXziZ?)WSlG^LH(<wY@Gr@$zRE->Khwq1}r1W=t9eiH?wwTh(|5)>O1iYAH
z_3E~@=9_${cR7-`ylL}q1pJha1XSxQ4kgeSm-XX6qTGbHmN<VR&j@}Z+f!T9lzyC7
z9DDE~2(hfo+;xRO1^zF_RF!lg(lx+?M{{^dC|*9a`50oX_2kRWp%8DH_LHOT%tB=z
z=I1F*ubi$wNg&vR%hHn1z132D`xVXI-v3Vex%lv&U-nlE9X?+ZY`OJVWVVaB#mJ#F
zm#bDTpovz|?qzf8Eg455%U0DV_ol#{<Vk_->HO}*Qn*?%e&&npUoFmTB51mwL|p~f
z<~!du5r&S+@3V^u9CS@?M&$4>k64CBFZe&F1k6>v|CWT@nH*2$ula6T6S1TYNQh$5
z{U1Y?N;e~p3>q0mh^Pe;b3?`}5}ya8j2_Ik1=G<<;f5+3)ov;p<Odktoi{PFhF4gT
zR)pzz-aS+3@3*?hf0wVspk_)Fbgz?*Yxy)?^<oHrvh;&N?_T^c9T5hBnDq3uZd&+q
z>I8^_kNynDV=R);iZ34gP<eLEb>czEG6nZKHf1QgmRsAof5T$rn2vPd?)pF{`AJ|w
zc}T34MV&XtPwK{O1Kj~t$KvVyi-7b;79Y~OmR|M%b*z4xp}JqLN$Cz+KB9QpqQf)O
z|2(}W+(s}w`G%N)4n%BqjCA*l*327)GFb0((o=mmA1#B#ia;10xyaYM(%y8*(Szuu
z^IZ3Nz0I;fF)V{(wp&#eEgF~LbPMH+sh?mpGhWg1Jr<51TGGM2-Gz1$Svy=s_bFxi
z?FK2F-MO1fKO=cs^@M|O<b<`)9Dc_yJVy7s?z+1G%bdAn>wQ9M-AZ%&|JB}eMm4pp
zn-BpJsUlU1fPz#h(u;=!C`}OQy-8INkrsgjQ2~)dQJN5tqDax8NRyI?AP52ihJZPe
zND&Avh(I9i?WpIxckg}czO~+;_vh_3Yp>aB?Y-xlZ)Uze-#o2s6S%%55W-UIsaSsD
z@i*3-j3N%q62Bb0Gd_`N%I28@Kja?ZZ>VcW7740FCAU@ZHA3|fU&s4bj_IkdEzk8=
z3*dXB9RwGYMfb2;8zjX;)F`c}IU$h9pLrGK^GY)&m$gX&v1gF6-qWLYbG3MV+{SKq
zzZoPhvi)}OK9`aA+&5N-3>K7-E&Jm%M!}fH);A6l$0u16G{R>4_ZH>fJ>JCXm>y#t
z)y(5d9!EQT3~@6(HnQQTAafvNb$`zI<oO$81f8tmQ*_Cy7raETK&88%h<myrH`DGd
z%;DNhJ4&9eWbrO<U2B4PLzWJkrt<f+oqpqxf5LD2i$R_PU4uROx_WC3vA?$`XzTk-
zOo_jkP*lRItpB*2FrqZD3P1Z*S|Nos)FL}Pc}2rhiRr<65CDjtrxnbvI%^ZLuav))
zk6<-Pyrj#7=@{j2#WNKl${ORydTG%@tojaH;b+sLumQi1{&=MG2dmxs)C%9w+qEyv
zmeZlrj&+TX%oGbBZ5TVxO&F#0U+g^b+IuMRx0!4L1->xay>slfY(Q`QjjN(Jh(O+7
zJ^^Ffyl}Z>xfg6@?iQYndE~MW)}LPb#f*Lt#T`|sys>j=iRM<=du_l?PO+#BP7-x<
z-=8EzGV`fJ<c8YU)j%?Gc9$)8P*ShUf+Ca5jT+|z?v(FSWmPE&Sm`-8?E2zfRQ}4$
z`wh>MHgd>>@~6HBIUG)z=Ndxiq!i70u6!zIZ~eXjDVZ2a(N-t4#mk-c9)1DH;%@zd
zq3mek9O$gQ-?b6-nHw=e&l@HU5AN<?x}#~5z%6<?qo^n%F%py5+boY#HfM3pI#{}K
z!_nAzL0b6_WYU`S&K3QrHtiP~{im+S>W?3u=ju1+@R)eEAZvuQ<*!K(FNsKLAl6yB
zYc6~=J0-B}t2xeLw^~2N6Qa}`m4H7Yp=W-56J{$*&QJ=JJp0*?YHYT>uZo97XfF1M
z>SZ%_@6mK^lO8G*=jT_oS+qM-yEP%;;uh_GEoK;NB)vuO%tm<B8<s_HDOG4R>2*#W
zs2HEo5HXmzjm7#<oC$JIlynT7WlTfgeV!LTVHBmxChV{7+21T9^kVgFr`&@8rD7k<
zqSu7c^IH_8T-TXi=#lxTEh6^|;%4`~#yzjWz7e`a_^WGbeSMEb%Av&pmsaGh-e##8
zCEeZFs8kX$?jGBaQSwxOHyLa`pl+YvZ!c{%e%IQWVKCI6Zaiz;Ia5JCXT!f>vlH+f
z-3UxTXmEKnNC`a6K^4tBlVbx}hfM!=H&l!_L`E%f_n^X(eyz~1g7&a#MY}GH@{Vt&
z`;v2)A=^~P*elH|N24OBwmnb2`+b?a$0?6nG;UQdw0usOoj3jcP41YJrd=MYEcW0i
zKJhT=^L@cP`Fa8Uv&N_0&h#*(yM)wW_bq2BE|`%fy{*B_-Acz1400QUgSV%-7|p^0
zra2fFdX#G2oIpsh(T8^?wTz1m8m}ouMOs%cVn#3;4*3@tN1*(1t;1b439t*|M1SkY
zTr0X=$%cwY^&kCSe)vS-c*lLR7xqE2bN%nv?z~YFi_abJEA^vIUZ|IZ-PxM8Z46J4
z49j2fA0o@kjqPkA+f`@EIW2iQRbH1}9MzQr1WyUqv*MsxQ_BJ?xL7huRz+}!Pv%ea
z^xLgNXt895wbH&A)|7D|Qc+rU%_KkVEzPfI`?o+9LJZu66jPCL0{nnRwdUm#>1`I9
z9DipPkI{EcI2TdVVr-UzyQ+}M!~SlZ{rzUNq%Fyj;8t<|+JoTi!pCD$c82=Z!PYA`
zY2r|?nolfgj3po`EHV1TD+2Fm2<l!Tct<$kl$cW8U3=69YEL-b;MC`OPVNOYsE9^T
zgVM1zm`+`e{Ye;IhJ!8<$2ZF7b?yC-mej(w;Iz4+V$p0+qDX4B_H)|e@#c~3L`5;k
z@LO&W(AiSAvqA<a6Aus&hIJ`rHXPV?=Rt;X#_5yVHxpL#(AVM1!Qld)jhr9zxt8%!
zcP#zsM+3Fi3L>)RG9GMa9*j9gmLm*MpV1y&mK^<HJXI-O%QK(JYgX;pUPkf0F!%Ox
zkqB!|ZfRJ3sF><!L3iC~7hM~d(V~XDZO!1bjjsDQ7@u^Fo|VZeWIPeXc`0PDuV1cs
z&JYkeVms<<7Xj}9i^5c@F3hRDB9}DpO;XaibabsQ-rQ5$77497^dLnXu*h`dKVQ8`
zrZf3Xs-XQ|0ekwNUC59xYABu#R-M>VB(1L}t-H-q6LTe7MQR#fD_BO+$?Me}#BJ2#
zh))wdHpqSgk+rX$V$g+o+d_TL#*D~wDoR9_q}*3)yFF<e-8Q?4qDq`c>8L-ppJ`mE
zJA%oG*ZwrnXRWT$3@AZ2&TQiR^bbxY2)Q7g$rnurnl3s7vXyp3|8`YbAU^1gxnHwM
zq`1p_vSQ#qCs<Y(V%&LzyWWdZX<+9gI;r<?3HH-)cW^mv!)+KuFYoO$C^3E}j{M9z
z^(LEsQ&`!dvsK06+=`VuEHfWSr!XXVnz|sFzU4dY6KIL7_$}ha%RK$9I{odiQ$$;y
zt7j4fH8{Jbmam@e%pcjqVSoxC`XRoZ@H;@9Ii58FM}oCqerZBV2}UTZzQ?UheEaTa
zG1g&NR>a6ajpAm4{{Y~ra{6Jsw{tzK0ir8j5mQ5KE+p&M7|!d<TxCTS^iNS{v^usu
z(XQxT%>qcA`17+tU+lyNqyW200MjdzE2|7hQxHN7!3Le0ej6SlEk}xx-dW+7#;(Wm
zwJB96m-&CaMNY5=^JvCj(!GclCgp17RpQhMq0-nz6Nz`N`(1dfdE>K$Y9IAf6x#CP
z({F_y-eBx3c?nhZO@Mj$7^#%e#_OJ-rdt~*?YoTEWXWGQptGJYvjxE_^=#7zJ0#*%
zG%m1MgS8=&?DPlIh>|k!%Gzj{Rc>e$ANqA50PE1L4ZuOk_43vv*iOsbda&%c;M-+(
zv9w;QtOae>T>8}A+bO^s9}oVpk_QjXRf@;l>$$&@)J9G%l<0hly_GFHbman)^I@|(
zkaOmNFSGxUArTO5*%ciDS@`}+YO{f*^)Yeh1mSenz?jvn=^CzEO?v&iN#kANNOv7;
z*r|@_LCe6HuC?mz8&z*E-F%YTEvxqGnfEauCLN*le*M|2L9(&Iul~JdCMXJ7^e`*Y
z5yGu@DxVwa$Sm3-B~J@h2N`w0hsOCwWQ=Rzid=b$p{|7zy2VA&B$tQ`nVO+rKk(Im
z3uE@ho~Mw7IW?H2>L?(S{Hy{lwSFdT`1y1@TO9POD$8|LAN1Edq9!q*zdqKyxdLJa
zOEXKz{weT}ZSug=81D-R)1t#HKN`?mI=!?*Gdy9o?#lBTx_=(3GkMScJ8)=;BdyEM
zm~*V*8zZqn;>H&5UB<6o2A~Vl57G8@JZkA(D&j(`RB&^%IBzxU7;gFz<j?cqlIIPT
zZPuv=^45E*p&t2Od%}58a;8Lqnf-MUM#_e<@Xs4_EACMZwhqy_a3trLzZhJSU4<C^
z8}t(Ad7irKp4k%Z+;N8+o?uy*%FWP<$wFJNq;mDT|DIAmfcXm}zdHJV^r(5{ZL&A(
z+P*)4y|AzYB~v~ALSSu{pe7PD1BiOx(bD;<i^fAJ$*FKWOa4oeXw+yX)P=1hCC-+T
zYzr>H=r#*QgAh+s55{o;n?R)+AaUL)v%V*ZM<HGu30kTPUIZxN0ow?kND;lcZd$kX
zArpGfq3mx&W-^m#j@gMo*1-IcU$!ae)Z`OB{FCZ~v)-k~+ued*o1MG<)6PMR<!`hs
z$U4|4he79UczFvOVx~>>72ha*{B3~_iaehOue0jguHb&yxRni-9NfY4bcm68uN>3?
zd$`W3rP%SfcB%4li83<DxV4dW%Dm|R0MUndu}Bu$+c8?f+z#G#A<`<j0DS1>1c*eX
zlA&B6#dOZQxHQh}?!j+;9o6r!A&%O_FJV2m!ne{;QE3@9mwQ<GzCU9^L$VJofEh-P
zwxn2!qX{2>KVQmoNJfftFztmLfm%y(PxMath6rww(DRJQ71f}#&%#auHx!eEScvMi
z@CjXRq8^I$^OSIng5+h8{lXrDeRKgon=Vls)Nu;$9Cm<Ikj}ZdaonrU5%DaQbCZVh
z^2fIq#Xldf(}NFBypf!zgxW~Lyt^V2F^MVmJd#5&Z_n9D#mz*<BVv~m9{Vss&h2UH
zIc#y&+rP<P1e4~79*_Z?;Clm==rYVyi#AFb7MKKyv*@Blb0W83X*piTI6@(Ip``^o
zxg*qbwmhhX^k9g1%}`*yZX$9WoOH84-97PA%Q8sMoo8gOr5ZEYqJ}bbipD?u_~sPa
zNMRlMNoIHTaV^E|-yy;6q!WK7EF3Z4T2|f02S*NP;Y{xCtJh{B?Z4uUSRk)aBaKfD
zhBaMiCw>X6*r*htIkA5zg~~tP!&_DvfrMG7vyH#f9?h~plF;JaMTEtJ&kwgaeZ~?g
z#1`n29oJvD|N8bHR%cbR`ln@<ClQyW&OOwI7(-TAl3KY&+PTU4gqHMPd_Gn;T)@TT
z;&)qDHR9rm*8Q5Cb06Qt9?Ap_rtw7?bJnwVSyq9_mPsC{kKmUmiBjoreM!arJNh;k
z_f56K@ilre*+gGkM4vJ{T{Wj=+kZFlX%{l7CN<EA^XbRm>-u)A49~t9_$QDqvcri}
zJ@3!QV<Bf2c1uJ^7Bk(9&y7di0U7eiF#XW(YHoafh*oiMgrU;$X<P`pb9Y2?Bm#eW
zPL&2B#4eTT!MCw*<VmUMC3H~QpRbqUjQv!8(g}si0k%$wXELIl!(cBC0|Xvet;7Q8
zWY3>QoLj_l%q-=RM3rw7$1FhpIEzoxQ#JtJ$XvG{L|MC1g&i9+-pXMbZ4w!*ezGIF
zjyo7DaDxttQMPP=W!<IYFWHsGggP7Ez2JIe9i^cWL6pxk<14daV8Abt64S-&*W;Jy
zy{7(X4h7DA+@XdgtO@z@I2p9#BN`JYb*X{IoQ)VJPTc<mC$b^Iop*F<l5A4^MLkIf
z8xJZYb&QsI6(lM1I1PLCBE7XN+^+OoA%Kv99UJ@WY{O=|^w@dSj=wmXN-PiFx9f-K
z^UPgmtkzQ4Y9!PNmwOfxjY)wDl@kNNtfaZ{Z2dJL{{PZ=RkE5o<*1vK69mQP;T6OH
z42G;x33RKp5gdDZ2Hp=>unu#lQw$xm+H!<BoR_1ZAxRYUk--bL)yuwj`eE&e0f+g}
z_D$U8H=Nf;@U><Mr-c@6>=Yv|iBkUq9wQKyhoG7+z%F^5zSi_|uZwL-f7H_OB{TtP
z#4~nM{h!H2UD|mK$@)e{?=HRU66r{$d6%cDb-Zc>Lm&1&Wp_OS34O2nm7-F&kpQ#x
zIHGjfuC#6On&IiYF(uOG-FcS-zkZ+p3CJBjxVP>+pRZLHL)>wP%hHf~g-*dOAMu_5
z==l8~bbRT*qvNlNg!7!beB2NGSqhx3m1LJUrIHX9a-YRVIZ~RT>j%B^<potuGWbih
z+vpN>()1A(BUAV`J!8hS${Ie6VWPn3{`&RkC3+-bdAEzmcM%|1v4?iA6eBz_|Fz|h
zh5y7sKImrkcj}s1qi8lW-3d2zYUnwx68JSXP@2KR^j7%<PSQDP#=<uBIRaon<YNjl
zl6a1e;<H@YaP7Q5R?Wmu2Q32|u)Mn%R>0}EFzc$DXLtPv_c|r==TYkrqt8By3j>$8
z61OO^6&UEEQ>R7k$Um%2*%0}s#{?fzxYcTd*wr39F7iMkf<d5GZZb=bOq?8v{Th8A
zN^FZ{LTE(u`KYQXgj*r}ub*fCFa>ovNWD%tK9gt&rXFspjuzT$lNdfWy?zOY!~oHF
zcxPi6lRSq{6j$y2BY9>@LH}R=HyEfXqv@UGP+E`v4G>zfOvNO(3amD^%yL#VxDq|%
zB?RO+iO&1Uf_XEbo*)azmxBZ;k~m?lmcj#x1%&@`LwSHh5|1$D-ds(BIiNULKvtU*
z0COB?>aQ<>41a$^cYE>_S`s7zOaLQ7BASwZi_U)%2-kL{7kUSLm4=VJEA!_gDG-3I
zUZiLOky<!}u1>$tk_C#WPSSg^K9q$30SC}=>cF}p9CBq5zf__SFgboknnBsbaS#wa
zy>$0Y3Z0*P)W;0?>+rg&C}_HbK|f>KMet^ww$bFb<`bgy!FNC?7K1`bhW@F6&FKU@
z4K2wcDJ#mwPr4J+;cJom*RPMti-G`WwDPLH41VYK7;Q@w)CI7fL}Zov4N`6`r2v$g
zHkydJW1bevE5pw@2zQ4cLVx<oJOd&WP5zN<dW^+R6GS!1Onj;Z@^I7(cT8fWh8pb1
zrJYa4gtNq`Z_c0^j=`Rl)omRI#=+_*OW_q;jRwnZAx4_M=t`*4j1c%~$Jt4XK(9qr
zw$d|mTuHu}v)H*7=l9icKsjLhiGQZP;LwcfL!vZMo!n&0l-Y^cawR}#6!<Cj@6Y`H
z3Y6<$Uj-o_@(b|kTpcnzfdGbY-;d8ff&tZosNDx=PrC&Pb3HSsXw(vR3>Z5EQ=J2H
zX91xI7Ehlv$$QIDC9ZHh@AqlyMfWB1j8Pm||FezLZEWzWK<lq9C`(P21hz@+Yizp&
zIZ~PnjgnE>D*zj_mvD3l7A_K)=RN!cn^XcE_8>xmw;ZbcMlUJ6jkn++pCDj*Zqe7l
v5v<{$W>DE8>F?M7o=g8<`F9xLYG&}1dHWO|ZrlD00xk=)bEeNsd~W;)Tw#9a

diff --git a/articles/LC-MS-feature-grouping_files/figure-html/example-1-eic-norm-1.png b/articles/LC-MS-feature-grouping_files/figure-html/example-1-eic-norm-1.png
index aa0b2830fc9602be6f7e0430aebfb20fc96bb49a..cb3ee7c96a2c18646150e2d740c1649232a1f3e8 100644
GIT binary patch
literal 114445
zcmeFYby(By`!>EYLZm^Zy95NJy9DVFlukjqMLIUbHz}Qxilj)3G@EZy5Re?bA>GXe
zZ0wnzZ~XK8{rmm(>|n>egJZ|;xX$amuKT>MSHd$xEsDELcL4wZg^sp{2>?Kh|40bA
z^LLQ@5u1-6h#mE`H1H$-b*0`3#t)=^+Li$T0BhXeH!yYM#2EnK0O)9_y$C7XUk)uS
zGY#E7L45ef+adNP>nvkzq&;CW*W38F1WlcsK0N+9`aZuL4-I^zGi@?$q$ZXaP(*4^
zRf+DasXo1ow14*!5$m1zlELA704`*ubS12&qq4K(`{8Y*O3C%F^P5@gz_756;<Mtj
z!Yv4oPR9TH`~Uw0%6^D_4$;mi``L>Y)tOv(!sPO}XXwCak+}0tPS^vO3P6THlpq!8
zO5n@6AeTh}Q6Uf^$ssTTZUKBj73BV-#jsNvMQP?<R3-#Wq)fFh84GuZTR;b^VLC9;
zcrg4Qn0X?c)v!}#aMg<tV}z7P5+U7@Hb@NztAcs~oy-p>2(&LIx9sO2evrA<o8*yX
zAA)&aFQsHA1(9F0-LvTI53C2-v2AuR$5TF;(eVs{E;yZuMHt0p+if+7bs2UJoGOj+
z5TKvI$!A`Ap|)fTK?y0t&+0?jo%X79CSSe6|12|(*ZousKP2NaBuOLRp{ZF9;$HI~
zJY6HXevR~SFy27J1>xSnuU{3nh#k~6=1e8W6~T$CpcxAVfNw>k(_|{8YA0gt9&%wg
zz=u;DFg&Pi(_Xh+vgRjUA%_;_Xj1$#_=_;DEM4NiC%Iz4&ouzn$$wV|5ZMl39;>)j
zi&DNA8@h)ak&ODC>v{iHZwEatw6zHu&lx*14jnI{9~7;$12xV#mFdm85L7<!4ufoR
zEdJNT1=X$LybYfIia@Zs6GSwaWf3b!x#g<|f5aJ8CpBnhcQ!}UQ^E7|zYp<$Hx28G
z2`l>Unm{!PC44}~{tHT^tb>gSdeA7j2P`#Q*nlZ8%Eym#D9a+uBZ?R-N82_uBD)Pk
z)34*ucG&@oztNmfr=2C|^0hPfORe%Ve$z2wkrc4Yd++-Be>OiNj{b8h?bD04w!1Rf
z110cJ<pBq`EO7zmJNc=tvS9ZeOsgh3KSOf+J#GNk_!w_r^j~|V{&QlVnW(2KnXXS%
zK@YaiMLQ@ry@}F7PH8N+pa^oQ^(s8<%oSIj{4z9WeTKOjnQYt<E{5v{1{BTr(KhC%
ziZ_+pJ%`xm_Cz(jj(WPWx1|TRSDlWFnUxscSOBMH-S&uxIZ^p7-dT{P9S3b?_Y_mI
zOnid-Xmui-2q;A(u{Fb2rm+5L96Osf2Fg4#?k}s5dmH7Pbe0PsxesP{uY8h0T!hCa
zLSzlflWac#UITZ&$Ac%OyI|EdTkFj2VinD<A>5L$@#{-#*bAA3P|uAP=`V#iY!GxO
z(J0IAEOEGfS$E5)vY6;`_V)So0%q}##b{9Zo*}_)UY>hYdQ~J!Y`~7rUoaWR);$Jr
zLxtOv3<Mo5|B)*ZF2en|vna8IUlqHclC+2a?O8b-`}3ilc?R)chxxx2WvT$!GAWr^
zF53KvasQ*P1ROLS>rT~7urQO&*myZJYU}ecw8em50bmXECF!vZ>t2keuvSC?m}K5r
znz|1@6|M`PdOTjQ#P%6W0r`kyg#<wdg$2jwEBZZgcpIAbetRjYG8m*!QY_MaIVgYL
zkumN2zef_qh==-tINr9hW(R;5uBuQ|`PRHm=g<w?MO@WuA0L0rg{NC@tmNUJp0=k7
z965!N9p|zi`vWw0soW^d_{mR}V%%0}q6}CeDqj@0+>w>A-t09pw_hv_$x*?81uWZ^
z>DajY$pX7*5!5qJ`u*mXtkrXfgl>K3yL8`2JUTIBc-Z^qp4ev$h^|3Tptg9wV;UT0
zg=7nl<>smsA^u>L%Fn<ya={+{5z8-Ez=Dl7UopTQ#83g^`{($u*Ls3S*sI~Ql<Gp<
zBR@ZX?1m&UbiRCVn6+|zV$T0dP)CZ#SvZ3yiGwZIS?P=umc|Bgh-$OlShA1@aAW0C
zX2W1a<#21mvkLVL$>P7qF@STyFM{qbg78ao!Q@J=W~)@?w#wQc0ty%ZVEkAi_LFVy
zdP4SPCowm@+}<H^Jjh@n@2VL#n}Pw}v>D#G=kXivEoX?TE}w&mVNpGJ(TdCB-~w>2
z{x#OkzGnHTv4haB<^lXRbO+PI?KSdDWpHBQpMgiQPbGcs@#xS+|JT&7@KZDXhl_BU
z17M2k4nkd%=0J!#WGXgvOaP{(YY!$to6s*#>k{|q{<@sS*qv<I!K}*HW>mZLEfOsP
z#4j8yC^c%cWHRSn?;Wu697Lt+35^#RG{3fm-Du6_e2Plp&W=xhcS!hMB*j+$k#l!=
zRz0syX7+#P|6gMF{I}XipR0PlCQ(I{=?{jjZ$MZon7XqexL2Ra^#}*DFb*B7hx8^v
zeeE%p6#AkqAqTE{;1WAUD`%4z2BKJOKaC<-k!d&FMi;|8BX4d^&}v#IJl=*d#>^j%
z`<HnuG0S$m*{6d+CEk!|WlaA<t0>~JV4uOqwT3&!vkzEjx5h^ph%tBEc~P_RvN;}R
z*?CCh(0rM%!qx0nSi92M9}s0MkNf;*t~h%y$isgbkv#@?`MzadxA^U~0fCP|5pI7Q
z%KH&o07$+F$Lu`bs-R(wUAR(DTEkNwoCl~PV%tg^GIgguyX<5b&Hf(u86Qr|PDcOt
zqko+>E94&-y_F(Z&`&CveVrtwp>_{|Yd&WfoorXoI|?SdNOrBbe<0Gj)m<K-r#I+Z
z*(2$FSIoNW%bQy=JRn;@<7GY7WAc4XKTdD5Wpd{$PL(L(V!N&+4v!GreVa-x0X6X(
zr3piK(Iz`fBO1a3`9V=Bk1^0aYaw`Kshd_!oDhu1;ND-~RrpJ1J^%6Dzq~99Po{*K
z^~2-7h1{k~M$L(c+O^Jn?%?tJJlry45E>hxy|G~BSIrilH_!n-@;E8PJISs7D>1JZ
z9d%6a#gm&CTC`f~kZRdyy$6bjaZ129oYIqTSi^jrZ4LbP<wl3Pm7*&TJ`K3Ftzn&9
zWvh)M;c-_K$A`;B=fB4^^1nXZ=<^0(D9}?p@#kt#{FG{O{6!tB9ti;Hgb)2BRpSqH
zBO@r_N(N^lC!}bwXWJn;7j;f@<=;r;G#V>;i>p>v*4yJPJHW8ST?gBGQq((>s7`7)
z<HJtt2tLbM%f%z}FN5+G31EF7l1M}1I!>2<lAmB2g$^89W6h06so`oUETOML%;;O&
zA(@6*FM>dZmU&!}%0~N5(D$;?Tr@cbXk2*+jgCu1h2KGoH|k7=$Lqg6@%V~Z3xL(Z
zZlV#?xVofx!(L#YN!QY!jv#PKt=G5BO6?5Iq5qE2_+;xpiOK&<eE+{cfihsm^)i6M
zQ`jL0%FL;gX+vBrc$*dW0}ltRE5^K@5yxAN<9me(CUEeJyoZLu!mvK^JPC1-&k*aI
zWom<vzY&FH0DO^%XB6VWcS$%y0C`0V<S7G&*FCL}sBrp7)9r}_-i?rAw>67@Ju8Io
z1*T;3h`6eJ{_mH5do3RkyZ!rre`OZ`Zw-%&N*<l=<_8Tgq0FCjM)Aw)#xD!{cUdUD
zH-DEE5ps8~3z3{wrnA`syN~sG$0vnPOfHH~?6LUhQnpg?`vgG>S<!|SKtPO^^W}`f
zb-9`keSu!2`oe6_A*1nz>&a<#hfZtqe1ltX%D#;Ga9YLXEj}HS>=S!B?G?f~I*VL1
zt}x!BFrEQpakAy8-M&|wfLf-VdtgDTXdcdEF>!#(Eqlp;E9VO(ZXi3sVb!=6!-$P<
z1!=JHS{#9jv0G@RCAPVH7MlC*CoIPiZ)b|4RNSfMfWMj933}vX_xSf|?*&?|P=$o=
zz#ch>G#H;;baLSCp{cY-!k0T51qo0Q4pa{24hMgx3j%&k-!2qH>45ss^`u<vcOnSp
zLgM+q5+4#jAQ0kEd5*ynYf}Y%<9$pwM>bc`mhO8|SPsjCCBupY`0nsjo4?1=0z&Vm
ze+wo|x9w(3FKZ)jV-5NA`iJ09DlHu$RbNT<0YmXx_a%Xy(Vp_JFXTzp{4|e5u<eh3
zBpvj$O1c#Swv&4YL%LP?lU0j}x?0eV(qhMmFJ~b<4HWC(cjvF_tz;g^@rhIN5Z_vR
zr%C;DXSdTCt@HLKY+hN1>FtUAn#~G<>;u{8Rh)_;XVbMEpHJb|z0___1ep!_&dGOc
znh((mKoV-k2+!dZTWH8H61yQkCBMi-<}jUIX*@Z#cRldCchcFwntovaovnfW*qZC*
zM;F1yCSPkLhl9ZSNc5g&vp7fZ`JW4Wt@pg-&VuQW+KrO}<a*a&Z~vUclM*qFlRxS{
z2|$)b$vmN3q3LVFC|Lx|4%xxBIB*Iy`kg&5psY`k$_*j<>_2#!fkz|hC&8a2E;Fuw
zq3itxeaTsLmlTf%hYlHOCrH|(hxTjcIoYm)khy7^rcL$RWGtln68U?&E&yKu<**<9
zam}>s9EVPk3kJ@U0|zOS?03cT8}gc6j1wG|2MQElB<Q>YX^GRM?M0?@yNT3FS+$tN
zSkq?G2#GIIEfFpiIm|HF+wTf^j$CLGP;!cAX_<ihB#=(!_2zW!DU;NvhOqGXXzarN
z#d8flI}x^m=b|5;pOGJfkXmVWy9U<quXmO@X@5kuG2YO}BT^}>3a#%(Omzb#iDwKT
zb%H56+|C%dY*sLlA<s4W&13{9Qsn|&kYE}&mEhHL5x?KY>}H!G3W5{Rt6Hrk2c&FV
zA31D~J~juTsOfH*AEa86gHxODq8a_v6EC$*2f)dAzsY=I$Bfo;?6(sVrGwIm(qZ3i
z^q1&C;@sKltHOYI$T6ApzEey2`Sx@C{e)UPczL2!WO%Mc_v%)$u}xXp@?D($Q3Ux6
z|K;M34}eQn*9(-8*z`&eSTzeX-zwYouI(!JVO)TKf94HU|Jm>Z&=$y^@#koW%zdfb
zEdmnIS!~FHo(8ffyNXP<XztY>AYWyQji_grqGv(!lLCNA8&6u6E5+Im^9*_JXPMtq
zCc)f6o2?E1z=R<iCeJG9uxXXi-mxK{!=okQ7E>Q;alsnDv5|+o9fEaAneLL%Uqs1p
znu+X7^j+M8<EFb~rt#SB#v7S14bDHnoAU9bC*pqNkE8nNq=aZx>qJ+FAL+5JBe>)s
z9+C9v3qbKsA;{f7U9m0vB)@i;2u@2UQlCTcYeMTGKlSu&kC-|?`wL-!h~bHnN5o$a
zqg9<m2UF9Od4O&{gb8<_aVVL9J#OicqS5BYM=xeCDMQF*Tls}W1adCN`CWE{D*}JV
zw*7z-wzZptA2}mk<>W`T=olrXtnb%%f1EnV;c<8Txd=GoI;09JS4Km#wE5cJGJLBy
z{9<po^bnpwZAQ%)I8IA?Cyv1SMcq75mgtl*jcNGeJ_sE*aUHwoj*JuFLt6{xtcwzk
zjf?$mpg;Ctx*?&?1O<o}ekiTY;Hef4<wj!r<}^Z&ag4~4Xlvc_@WVI^L*u{oHsN5t
z30GOrq;Q%j*EK<XMJQEtJ4`ER?UL*ziZy!KO45T@3=(dd*rT6$&xw@G-I%Gt!nlF8
z5%v4>Tb;bBzmHEWJ8zqJ{FZj4OvZ*=2C>=7yQ)-`%&_B#tUG>NGL*NVGS_IMez{`N
zqCBE5=F=!!pT_{k3g<Z~t>3t$5Z}x43nb11sf#2NkUsWoKQT<?jj`uyx;wJ+yFEuQ
zLy~4=G&(au=c--fl2zl$k3fy@nE9EFOHOAbf8Up`rtXAf2vk3dIil`$6f-g=&A$DU
zNE15c6aGiNVYzY47+(*@M@AF3qMA1aP1r2}v?NZ1o<k=iHoFyINOZ|VJnESP08{N;
zZ+2jJNK3(f<{SM%gi)0tV`OO}1h=ec3S*J&`+JREn-PaMmAH`-*5>yYUz`e30k|S$
z=_g_woae{nJs}D(VMGptXIoC5WK}+KVV)C|Q>fOBLk`H1Y%NDq@W-IsPGrluRQd>%
z#^x4j8^(q<LRL!pWONc8zZ|j-EN>Be(qJ3g3IY>%W#(10`vf3}VDxl0X{G1m4T)c&
z_GXmhqodBLJUX<`ywKry8z|G4&5}N#sB1ta1U(8%72!`we@G<gLz~p2y*ZaDBeEYZ
zKmX2n{D~`onv2n)emKBG>So+9y8xH2JK^4u&EQpaNb!yGxQ?78Y)V9D(rrCC;TO)p
z?a}HDxnJ0;Q!3tWjdNDEFNLz?r#a<Y!ZB~ejzbe)L0c@8FDPs}wmQ1nR;0SZYE-rm
zc7Y_Fqa4@dP(aVh&u&0cRNtDwz%ym_7lv-xX~%rXJTFpF3J!ii*@-8WqmGr_4H!ss
z9wmh--{D1C_cdW(FXM_JsZZ*B(C{2`35orp&#^}()I~Aa_=1B^U4%>I5hQ5C#OOFY
z34>~Q!IVs+p4u6ba_b-Zkhtr<$RBlm9BAyc`Ms!0tBT5&Z{)Hu1M6=IS*Su;we_NX
zX#>MYTtMvCp2{r}#u*J>q{qpG@HdVsC8b6u_mP(mDiP#e!r@>Zoyj}v)iBb&Zmi+-
z+7?~kZfD8YE2&8jBnwj}wt|5F0rTD5lFoawk+yan+YV9PBgBd-%|V+2(nTlqn7x$x
zt)@M?tHgVzgdL?8wjt_?w|!LZEJo1}#n3BDx5T|B=5s%?2g(&GLiA4hyZ($ng=LZ#
zVC{44c*IwS19xeUnUL70s_c;?4oT!%u4vg~Qlu=`oiUcpCSI5ntTh^}XN*9th%N2g
zTU~|4vD^T)Gk88)=%6WgTW+`L%F%3}l(oSkW&Oe;Y)`_7WoushJ@9+cWgnK1hj)W3
z(kTnMGvFq+*JM_d|ISxHPf@87>+|UwB{H<!D2R{*aP(`mC6-sm<g*NcHSNZ^3;iG)
z-|%%IJ8y(0my*_*-8<%q!h5QGjqd|xZe?yUpXMKh@vb`1G}GI{R$FbxqU-rC8JK+b
zKRKE3h!kr&oq{w<H%XEVV#}kEgb+#PQU07AcoWY2IrPKr%#b~wlFoF0WZN4*`D1A-
z!X>GnV!v@WeSE#l^{YE0@oKnwV2R#wGa;&clkwbGPt4DF^zobfc=bK7lfarDjA%gL
z&`qNz+kun5*0!=;-yn0Hm0nb((vP1Cd^!TvXA#LlJX^sZT4Zokw;_X*QYw4vr`XsB
z173ZmeFotzIY(o#n*b+PC!LJ*#4grG`qej%bT;v(L1d7d_~;X3?O^H};^E6^8(}iE
zrURn8{pt$5HeSD2P*4&Q&8R23LD5?K%F5dWtb8ms6#l-B;y|a7WD*W>6{?g5QSLlt
zE=%_#P-Q9!61goj$6OJL?O|X;O}LRsTNpgyKB3ee=qRSkY0oo%V{ZS5A-2Ax^KI3S
zaQWj8j=yoMhL84qKXu%7$R#fbUXA3@VRAjaS86ESHxSrKEmIdvKUfSBVd!PHA%FCh
zMS=EpI%l34&B&(jhw;b)uTgGm;L@*^fT$t=fcTz&4V*%7G&;t++n#;8eNDBH=V$qu
zj66E|uTDGq1aTTXRQmjUQ%8J<sPUu&oaO;RX{p-0aheQhs8RQbqJ>VvvT3&-U0L&>
z>8diP*(|W-hr9;kC+;L=BlG*DWcKBuDlzc&ygO;vW26B0rJo6TfQRtgdgmeRmr;7`
zeSdDdy6#p+u>j>IzV3<AxrQT)d=i{FT@9HC?<ETwey;$%Je8o2lWI%dOb)dn;?Obi
z`fkm<v{GzQ8YL>mSrX0u<crHd>Jux{LW-}CSZ$bP>!P@ePeZ$~lj`lS_D-`a{fB3L
z*K^H-1>`=qmy6B)_K5j2-=_;n&acnlaTgyZmCa{J%G1@)b440*ZQ__FOFz}K8(OJl
zZ404TX9-rq1xQ%+UkE(3CN2z8y0NFg(2G^u2Lk-OQu%;6G(uz}<#d$*-_c%v9bKhW
zJcw_-bY5+b3-#DNeYLA3d%cg#&x$^x0bOh`A4bL}N5*D3>41enA6=x5osd(cJvqn>
zLR96@o$Uli<VU<gQRD^ZUnnwoD2f$}t~1V)qlM#t969$l3UQi_Y`1VKk=j_uwF&pX
ziimx4Z;SP!GGepX{j1IWL6e>g_zG5tGW159mRg?oegVaEQSb|i%sj%mhATmx48Ees
z=;cJHug+Jlh&+Z=3t}6k@3}@swBhGnj#NI27vw8v^htTPhxA_oW7^01{>!wEDnH2F
z3?-MyWUJ@28fy|yQcBDzeM$CL4p2F>eRyMH=Kq;9?#W5Nd?bdvt!eD`vNGH!Iaa;m
zI0g;{Slgy^r?DEHA#s+ZKKcz2UfxX2h)IjL9Kiac=hk?@65_dlCjaifM&AKRVt^4a
z_swy49noPmivq)`V(A-Xmu=hLKXh_?Y>@D|lOF!OGsVWcjrlup)Ba5uqP*Sb<dGxo
zM}FqJx*3vrrvuX{(n10+kfM>&+3hY@68LK!|K;6RzCJ51=7&osjMBv?RG87?5ewhD
z6TVk!Z?%|2^%y6!VY%)i2dzTKb!@!!OHcRzd3Kq@Qo~r*^Wi-mAb~-)$XKpN?pUYv
z!L>atqwXZf!iRxJmz4>YD3XEONq3rKIVPDtYZ5dftum+NgVS#$As)KphJJ%{i{e4*
zUkT<$3PRjX@LKYEV3pB3U%jl2(0B1SHPHcyr%J3D;*DGlHU+`MR*%5rO8X&i4q%O#
z?S$Hp^PGbDo7Ks)%YV9pw-%aQo`(L*KBY+dOo-&0n&={THTQd*i7Sil^i3QZ`$2aJ
z<;rz(v@q6XFbo?{-74&&gVuUK{`J{QYLESD&$h%hM&dcoHANuyiKD4~%e@Io*v~Zh
zp}s;~2nvmoEg8tZBO!2L&vDWFssJ#Tlvfl%Y;iXtXPbIokNjg46Pn`~3?Yerelw--
z{xDQJ{l0ktV5kqbWxcrMR{75xvg4THAFL3n?#aYE7_BKZV@7lU?b6DJpf|sYrcv4u
zw=-a*M^f1>UXoi?b-cg2ps~54yURRl3*0#wgF0L3l7E3Ziu%l7HAnnuNy<~AaJXAz
z&em^qn=lYU(RhV;P~<!H+ob#bs(1EtX3){dXiR(tF<u(qjqeh>m{du~d)!GM(kL6G
zROVb;$8w>dHe42Cw6$Ix5PICca5AHB87a@NBAv;ra6g=eoLiWS5BLR}7zzvTw56<?
zyN4bgUWHqwSYRbT@BzL9>bRp;Zwt2PaHPkr8ar7d3C?N~(Ms_P&i2A7Ix&u}jyo2@
zF_B{aXHmg%h^+!)yEE1_)WcglA@*m_nKAV%k5IP$++fp<<C@n>;cAn8cqIcc#0-Ew
zj}R+qqY8=N)_Oz~HPwwWgqA-PmQ3p3`?c}IyZGhWh4QHVVNYQTt>4B=o31o%)ra+}
zd6fA4_j=jWI%!rDu^OjAt2SQ&Be5anNG2NqR|im*^_+s~xH`LYBpVVpe;hW>k;sOy
z+F4=Ym^JT$)9Fc5ecM2(`t}XZ@3w^dB!vW_v&W*D9yd3Sz822KT8#eo&YiSKbQAQs
z^Hn)tpAG=F#K^IuANoD=yE_52pky(e-VG}ra!`JUcE&c_$-H117}FrO5_VIhP$YIg
z`JMZXyOvP-X{09mA3ltz;p$?+gpC$1k+3!BI|XYX*|Eva-bMUq60bqT09c?aI(|bU
zq5NhfS}Q3aS2ZlOicDc$1}1#tH-!=ST1?=lvN(!Q6(aUX@`6JZd4&n1tlE9}j9Xjx
z3Q*biHy~rBa-5FOk?L~81UgJM_}z1#>N_He^S!I<*$CUp8NAqU6n?}`qZl0?B$?md
z$GurL*bkWJ_QKliRG&qss;#)9XRIn>3;w4U0I);E)qXNDoB)5aQb*^_c3QL+u=#d)
z^&mlvwR($hiV$`JLqYiU27Bx=SI+q|LXipRkFsC(gflgH>dAhU`_IowXbiTHO>%2I
z&~N|t`#Y+!7g6sr+)?f%iodIYYVP#F%_*S+^TF;Wc-*J+qQ3k|Vog+G=OAt`0r>1j
zF`Goid;6~BGBbM|Ma7}&=2C9-Kx8|b`R0e-6fE7FytpD$eF6;RVZbx#0z7+v%{@q$
z6vS^_!cOm1y?P(i9rnB*&e&E;=WDW5z^=aL(k}FUy|I2}5F`4#t%O=buGcAnC;Fw?
z-3t2XsrX&edKSVIp#h7I9m?GT9(b34QC*`2Mf}h65+CUcUR^1nc-|f(X%*IZX0GON
z$%jEa;wTRszdeWed8Z^0(07Vr_ORD?p2>2ZKmP6OuN#{ARfzUY*@}Ixt=`(Jfz7^m
ziDe0Rk?-l|$~~^jQJ!n67DN&mN9|z{NW9ED1l@q^S3yW8^@{=@Ib*ku9CP_0+#I<Q
ze?-bce@FTv1DAbVO%pRT5TDzjEx_X+<klA*1{s=tXwhU#yVNiOpvEPxclxO2=<WH7
zDd&Swsb{0Z(8oWx(~60Eo}4+zYiMHGRaa{J+n<CbFf2YyCHX~|vhVU*&0QUNdgGHg
zb7z_))13erhCIl2vZlci`?CGXmjTHHH5*Sp+$Z<az@jeQk(Rs7N49BZpc2pnpjTgW
zv)SXw1K-n6-T2iq_QAkXdcGXIAt{>$DERqj;f%0hbiIxUza#fAfxe8;Dnu88sCs7f
zjm$pCt5LwH7X4f~V9)B2eJEM@4Dl76pGl-#l$AjiBDCU^DNb#rTI(S4sTk=osI<&s
zzN2pdb{4kGr{jq)Z4DgGOiP@~09>f)oTXT+3DZ9RI(eDhi=w#+xi=Yr&+SWSGDDsB
zXu}_{k%Sh_9%ZoM>!AO#y$OA+-2ZNEG`lx${rbz@YJ9Rr2-}Lbg&5Jz8!dvisAH<M
zCc8+7(qYUAmYUV$r;?68qgpWI%W1FoJmq<n2qJu$gwFJHei4Sem1|b%EjheWWVR_T
zCHx)YAGon+n<yWX!S}bq&?%>Zd&%tB9>JGEVn*}f(l&Q367k|DjQ%ie^`J2P@Uxd`
z-IhY~+;MOLTD+SuB6mP3!h?<|Jg|G>{!_V+?6g96M`no)fBL2AU6)5Y;y27w|J3Q<
zrCALEHSZZDsNdqp+zPMbR-;BsOnz=Od*#5dP{)04TXnB}tY1>@A~X8V`X{q%T*d$?
zdAuoLnT-xDHIZ8NfntAg1C_<3Utzc_2TP4MqJ<u#2s-H)jVpolAQh5ZU>hNtv~dUC
zdOcxL?I_YfDJDT_-29q!Ss3>`Xr*|mgXjLpnwf<2a0|<0S5LP2k6K3v0V)DAJ$rtx
z8a$EXrM6eC^#0Fng%GGm>2kqRWv=1Vv2eh~b(aw0#|H#o6k>Rmz%aJ!X~c(OKJlXw
z5MU~LjT2f+#SC)gIy;uhe*QdxG=*x@qE?_9_Z~Mn*OfT*H7>8LghJ6w9#I_$u-D@e
z$SNvf^1L)JbA|W<=qF_H_jV)-I25N=_iDZ}8#doJ+>_sufCY3&E9aM%{!5YFOR_bf
zLseaax3(&A3W$<KnG8eGNPg{)REL2dY~9L;oq>rPR7UlZ%A)4}x$~-pdATy-RH4?T
zTR|jq_KF90ihlF}+H^7&^UJp7-G?~!m7q3H5*`0FJ7LRj_C8B)+LL{bf920g>XF5*
zXZ9XFY1`^H3z-l4M!j3sSMD3<i`r7zJd<ms^t-$67_<BhUjyghJ65sq@>`M+@DQ{m
z$CUQ(ctcWM=Ad)Go|)kATKo46_dZy}s7v3Gp?3pTE+4hai0$m#Gc#XXG6?ut_+ZjQ
zJC~*==*vwQ5Ap(blN3olTv=Qy7dy@yh3nigVd7_#22gIFhU?bn;iDFH5riySg-H1+
zR^5^J*xxZIpytnbW^{R^+vgYQbjCQS=~YRc?}EAPI^mcdTCgtCPhQBZ&fz+7YwkGI
zfgh<c<cX?cE%IYdE_hI{{)#{+zwJFP|LZ1=Ox{yXrJ8u2w<C$P3f&D<#u9Ct&P*Q1
z6+#KTSI&DhnSiI#ksWtsa|f7R;}MHg>MK2&6l@9iKBOWl8;yWstN!<nSC$GV9#*&F
zbG7hGF6p1O^W+b6Z`%F8md7V0oPKuu`VWrF&n{r6WVMVRA__A9F552bC%C@5xdQ*K
z7=1FZ`Sb0V&avVsZ-TvjmQGfOPm=-Q>rcJDj0(GDY8`H|xH~;9q!80S+NR<xoIp;~
zkz8{26$-szyLfy&g-_#a<`qD56d|6X3zMV*O<{HQEK$iH!Y0e`TBk+}|3qBX@^9s*
zeLrYk%_iF#!%>0Rz3}o)M2nug2yX-Mm~+T_Aa{ot>8UaM($c#L$uK0b{fRJ`0QdBj
zkeN4CVg6cU!3{_o^b;-GVm794Uy{dCQ=9S=X22Sfs#KNB|Mg18!;?#-_T3v`qt0>0
z@XzFEL@kQ{V_uhR?!#J}W)Z)kZAADBkxGB6{tF0V2X9f;?!x#E&%1sEejMC;7~ihV
zN;lg1gf`N;QF2kPw*!$%VYqGq;z86_2k%Y5@+kdj4r3pmD)sz7^|2BBWo9z4m~AU2
zFuM$I4x#9_brO$(23{*RE;iJk^fi`mOS4czKf$w|k4&kH->6`|-U*G&pDV6X9{a%G
zt1)TgiOQN`ET_5gO>7FLcCYLm%!i<abq6zoKMD0DBI-kMc#U7E&EUi$i~H=TW4bE?
z&kA_okj;_#)OhqbZeX&PjDzBN37GJRy73fG#83mj8w)%7$;dS5ovuuV%Bl})=lo;u
z-WZ)M@uq>*_g9}ctE}&FnCjgP)OhPGXC}yEB9G}HhgWRg|Hnj24*2*RMZCDs<I8I#
z7!VOf4lnqi*zT<ta8(k{8Q87toPHnijBP7qOzxoM%07--=3F9vJpBqbk;Wb?c^obD
zBZEAc+LA=d_m>}HJBL{}AA*PV!yqp4kwmNONBm7&zWe<J^pWu(p*UZW99V0b_<9Kq
z@0&}3B7E9Oc6yao7k$pPeni_6NmYJ=dd2E!`X?$4e0S&5ScFWk&^4vZt2TDooGM-$
zse(L?IhEd26DI+pzop0Ae-F!{Mx0UX!<Kt9qh5<AVN81DwWSwPD&E?nPi`sw$Qx;!
zT;z{^hExR1?mr<PIMPzGe0kR|#75;ityXVa#@t}lcliKsBt>7vx(U4PUSP=7&ab*9
zlH`*x1wf`*@sn2zx%_|!Q=d;&GzP5X0(>`W+HU2MrKJ=tA7x4AXhW=17iLIjn4+zx
z3Jo5*zZSps<eJO0ST(Duz2Rwm)MH+=yg8MpmgioDMd%L0Qr$&JvLLN$%8Hkox{Q;)
z5MNs7<S)|v!zA6+b9&YMCSNBpe11&ozA0^_#t*g|yu@=P-S8gml(AHy;Z+kFJ?ng=
z`DhURq(u>FTCEI(1KoN3DE%xq`tBhQ(ojN5NB<57e6LJeBM{(w{^cOdAoif+qL@||
z__(mVFK)SlHKewAAf%h|XC}M?7CS97ZmmYR93FOlJ2UEkB(i&Hr%ukHXd&MUY9Lu(
z4x9cirgQvov?y17nM^hR9B)&=X$LGtG8flgZC>v0Hg*dY>L>2=OcO7&2dI3y3P!fT
zNH~t}h6<LEMC16Hi0v!p)nKG);nKg09S2^jy*;A70WGP%k~S+I*o&sxnuEMB!QqRv
zr!JDWl)pZS1tG|*nWi-Nge46jqP&*`yA4^KwuG4Xn~3VqobUnpLoT!9Y}S>Hp57J>
zM^^)al#xcN=;`||MdZQ~r3G0ua+wSbtnL&rVF>i=x2Ks?vy!v#YzP$RlEf9Bnk^h=
z_qO)2y9Yk%cReeOWTw_bZvwQ`q8j2Z{u%s3>lg%;-iFGhpZIwqtaz#s_|4u4;&wdy
z`e)BsW2eY28#kX}0eogQx@#ZvhIPFkrX!iNmS?S)^P~sZ2As@J^X1+pHP58K1V}?L
z-4DrrD+rj`78ZPl`ad*w`RMiBQ3_TML0H5-+Z9@aPUm5vA}8QRo3IdlY&_RUBz$-^
z$z)FlL#YE-wR>{T?@_}==l5tmJM9pH2|#jB4-!dp4WEB~x#3tK{PsR$6k%8Wd9g$^
zw?v^gSalgmsCu+CG^Fwf^|2B3@n?<I+sfPxa8Q{)aTp`T$ov#G?TEtz-krF;G{&Fv
zhuzexLE<;wY5!%3!EXF`E926}E#<nWV}kE!*^TRt*SX672L?(jKA5;MZ^p5E)o(_2
zketdu@P#}k=RG>90IIhUqJFIY5|JSPOqI{)OYN}gG^4n0@oakO%OO6ExZjKJU&9su
z$z#cTqED!u&+s0{<odc)4CqHGmTr6YxXjEMrFpqav$bL+hk581&{;3{dZ&(|1X(oA
zzmYiFai*&c{U<`H0lZTh!-*hpcQ>~pL~Z3cJVbrC1VoJpCFGiHeUktllR2sv*i)CQ
z4t#Eb*g}dD<}WYelPB2tb>A)-n!9q8(-i1s0lrugEWtmcRyt2FyoHJTMS1w{vT@)G
zBU}qziB(F9Yy|N7cvb7YJ6?o%`GH6K3%<c%?a2ytp4J*6Ec0CC2F&e9^u)4d_TXXJ
zASd{~Yzn$3zgYGBX%LT@+!6o-kWCy=%8~{~(c|l0Y~E0%13f?AlqS_Uy2#`vb@8q&
z<QHX01E!P{`+SzoXZ6Zy@iQkI;eQ%^ZaY;PEH~z^x<`VHg4LCoWz)lEu`iIU6Eeu$
zzBZVAe|t-GZ~XdOpDW(RE!Fn%pjWYRw2JV4)C|w`7$FR==tR#<ds^FBM<{L20&gkN
z61#d}D@%hnuWu8F!g8)T#yN<mJaOg{$Aq;qoqPKVT9qM#!>bQsuH0$Xh2lq>q&dYG
zzVpdX`Nm)XWxTfD`wYAx(ps(+FTbnwj#m4(f0mqQ#Fd}C40fdMF_U&B^`{GUO<lnS
zN*-1=r+t#?WBv4v0nN8A9B&#IK4I~o$_0dUaCC;n^c{tV@@|kL@qviaeRXbOy4X>)
zoSnV~X!ue2H^5IrjMVSB$^st+zIVica^qBmnWhGA_P}10pKo{}!>J3299}J0v}Tok
zuya}Z9p_&jGX)6-A-FO?sSk@5Nlugdi`9#KDZc*llRkdAyVmSUcLR(NN=_ZOOr|^V
zBIu%N)Ywpf{nW@A=0ZO2<NFxn;q3@G%Y@Cqsx@3)w73e~NRWU(;qvVG?o;=)i|gNJ
zq>dcjg;?l4i{GD{<U6P4bfRJ{is8e~(|UBZe9GoQ9P&$p>TFa6$@aS40U$AmpK8eh
z7lq&&R8o0g^=uQ`s%EuvATws1<61JjI--!foXZw>PKT7}<NIMmexdOP6|U}R>KOXO
z(Hr?{NW$6%E+McRQ(_p<^h2z@UAbx6XaMZ~ucsicnJ9S!?8EA$;^@nycP30C!qmH$
ztk&^iTRqEwf0#n#<H=3zLtn)u!as9Xbfm@5DwblhF5$tgylI*P@?W)FyDz#+&_q!}
zVgNK;Ln{b)d?V>&BY)@i-JMHcZu!Ud{NnV7yyEX3P>&RhapaI6J)O31eNb}k%*^yR
zzI7=?w!V=0-RZntfcjQ=$6^GZQ!V7RS>|cITBZ1QZ$91JoRU3@x%ss0=NMX)qD!%e
zVOkK!rmUVii~~T`*Y2+Mzvp!FbKBXN<H-{R+H;znwlL*sYdq<F<oLSOh3}YcD6+fJ
z@c8ku5|iE|K4c#YFR)C_0h7zuQ(Kj)q%U-PtG^Smh>b73#O(>j)f*UpuQ!?ugpCW4
zz`4x`Y)s^O1V#^zzInRY3Ad{ppOD|I4`>iq3p<Veiu`~sm&@4Jh9)s+-_RaQ9(y9)
zh@GepULXhi_HVV9?nW@79pF|x6uu(<y;jZds#*}HPY+hS1>@i^MsU(dPDfIE@0cH~
z@v4E3*F%fEoN{dpoestW-B<da{nb-q+(jCAFPQu36VOW+`KbP~(7=nO`@)x0yC1VY
z7lY+<ufISBtrbt?N^?#WcX84I5W1Ug6S01dbxTP;D}ADsY-Uday%-oC@cMrNPgfJ+
zadKlhFq=B}fx)TD@@c@SEqhr_ye`0*8Avjrc<cXoJ$XBWYyNalGMB@eUG@RGdr47~
zJ6X3<zUz5{L7<&hw-4H*RCB9cqY~-F8de|VJ+{4<x?*p`e_>IFHO$V<SJGU26w&$u
zPv5!|KnBfWjV*b)a!3xH#P%Ys%mN<I+Z*2m-X7e?-|05dY&<P^<&bzJg2}gu9w>8B
z6@k2!6fw0Qn3ExA`Dft^<eb-$4oNR))Rx+ozLIOPOI81hWUI$zwtTK+buhbkD%vSv
z>kwXMh}WIzG!0=s<)DHI3*xQIpc|T}n_*JbhqI1Ew;X6*zkrQ7vp&MDq;jXJ?bqN>
z7Bv3EwK(tZgodNXDN9}HO_E$rJI@UHqtc-o(|k~3v^vriW*z?nY6;s*kdMYug{)aH
zDQM<WGv1wopZ7iCY9jt3@zzB27pL@0Z75?~&7#clQ%Id5;05Rp_2gRLBy3|^hN)0%
zda8t!W$}o4pQbHw02m|CCyrrMSG3Q6kG~ls@e2D)Ow#s`O=$mSRDCT=kDwI1^(g&~
zLZk7HcX`YQbgOnBkE*bGD7XXq4V;{q46fO@vq|lo{a^5{@J_f1jE_G2Aj!wJGW(`~
z$Xk8i+rrGCWAAfTg3bt0Hr*K{S8y=sqkqY}u~t4oK@9(B_`TVE;t0t>hW?6WuOz<W
zAU*&ku*ZNr!<HB$;>SjF?g}{K*o}tjKRRe<33={+aN^{7n1@jk%M2!(*kRudH-Z`V
zq24(77fpN4T;uer;P9i0T6eqR;K7G=(0H_$g+>;OwnotJdw)>7Qbq1A!DZ;BFD}SY
z!j$I2V!ZA{Y;9LE)K`mTABSPnW*ZXgxi3G320%T-k9&HxMOEd=(8^*Kq$l%lgid_F
ziC#!U8BdG5eVsnvi>YWeqlAj}?niyo5@f-*vg8j9$1D_3^McLj?NtoUD%w69kKRK0
zlMgOBIBE&T2#c!<^C-TW$^{Xw*X79SH7NEIZh|Mn*vmDtysj-`*8<ki$&zL6o>97)
zWC^_BEr307l=|5(iy2)<>PB<sw3gT0e&T*ijVz&3CnM*8#&52yj=uv%|J+zpM8u<S
z%j{}qaYr-6u$g^a=kvA6Rw?ZfjuJhu;s6z~1Tng~lsr?g#CjZu((H%LAiKAS%{G65
zV6z~OrVkj!->?3nelbXLg;3cjYh&*Vr!=b5=M6T_M>qIM6t!&n!}M>346hKaAxoP3
zH27?cp5uaR)lrt78k-WBR9^|wa*<V%I;7+H+trxgkUxy`(p|HYcM`xglQxPs?w<zP
z)!a;uq#5z^$&Rc&o{B$>Zi%FXldS0LGX%9hoshXwOL>>r8!en)@~<7Nj&aOTY5UAQ
z!qb%iS=%RikpDKyvSLKfS#z7*G2xE@Od;AFwBc2<;KJidV@cUv)hg1`M3QVNm`V%$
z=$XijUdHp;w<}+=9iJf?TA7((YQ4Qy&Zv@N&<fF&P+Bqj^O(*(+S1XDYh0X5i0p*}
zw~M<fKEnU_u9W=YPm-J6<O)Tn`AYD8KWD0y;CqweL#}{9nuEw_1~j53WJ&FnTo2u;
z42NkmARKODk?#b%$eUbcjqiCG7zmlhmk8YdEMzcD$A(>%yq*oMAwN(5!bhS=tSB1`
z-)lNoTtA{Nhk01uj>H?Jn^Py;2h_jCJT5B;n-mP;7+l{Q)39vWRpy<F4t8n7w+=jP
znKmT-80Wr}_br<u;Zd2|Im<B-S3$6_V{=nrl1e3wCA6d$S>N-s*==bxQpzuYO8U4!
zZ|xMD%ZMGYlv3f3g4+=siliJo>3)9VGc>dD2E^Sz3bWLJiL!3?*&K1`WG0{3Gg;KX
zI?i50b9HouH9Sm3UE?zk6D(gSq%2#*3)&%j%#toF^AAH+p?U;c7yWCf-r<r)z`s9e
z<UGFW;XAl;`RZ78!H9S0$m(od=$DZ9Rw<DHR2@rC#Ibo(d1d@$bd4SfJFgjaovnP(
zz4h9#*)QAM%S0F%<GLy<O#vG0JMPoD%di>3*)j0@MM%u-aUnP3q9$iI9&}|kD@U=~
z|J1twDhn@Ona0(->?~zl`6@~f>^LK}G$7!_xnC}<yR7ep@v0w{m!EFq(Z6!py?Y>j
zh8I#gf34agmRvK7ufJbvWi~FRu)6=!7$M(3je3|N5M{V;N$3yihFljvwWevj?M}Oq
zp0guQE7WpQ@gg=3u-(OX#!C7hSX|$@OzSPf7>04wHQvK?aY+z`#fVxqJrqbqfYrj_
z{C?2qTO|W-Gm5c~aft~K1&AnZO@Wzt@u>n#;dDqw1<oTrNvq1AeXE>3_scbgv$z1F
z!sVmLle`inc^ehjeG3-**zLJ-Y>{&V+yM@m)9T?KP=AmPi08dDqLW#OqZwHuF?kyv
zFbxz&oO_Y5WUqGp7<Wh;_=Okhsm|Q>*(}JHreaXFE;vLw6A2aZvtgLC@lna=g#FEy
z_?cV<6R1iU)w-|x?g+=rx31zgcUi3DyNqHBp8H<@Gr9iPfL&c|VkZ=04vGcWyOHtH
zn-frvzG)CX`_LbanE^do287b>I=|1M<$WLYN!d`GPL}l4y<hX+Zcp3y>Nc59jvG8o
zGD0gvXw}Bx-%UB9A9dWz=m4%}J-6L|#0sHIkpVn3Ep8)>o2jZ+OB*Qs0nbBnxAm8i
zdhuUzqw$n|JsJlcTM4x4^)VRovOR@kyX8|o$7c?9J9ANer$c%-6CBlX_z;C|0F7SY
zIy=H1YuPM)e~oVP>2Y-}XXb=zFUg8><Xrk{>Yu~KF^2CjO4?YaZ*7HDSg7UBTsq^e
zX9Pu=8_VCYjEHa~pen_Tll3Xd!!pypIJ;<i80|`(b4o8C>U3i$JL4^0S}qQSa5n1t
zDeTgnoW;+CZlOB-0i=gNj{_Zh&rW#`%<wG?QyH1fg73e*@6Rb8yo;m~8vK%K00AGq
zaVk#`H6jW6!m|Ig!OQfv5){o<OBkiSR943&T|d(25o6`LraiQnitX<B-1b8`Y11H&
zYP7+KJ)h?y1Mi9<p@L%CHe<I|+S6YTDq@7}5e3K-5vXfid1t)A6Lad8K>GXq-+9eI
z`-ELBb55U7wMjiwp^|m7OYgL+(mL>}ph1M^L!3ld|BC93Ii-E>fb-AVZ%ncU+s|i~
zNPZDW3$lh&tnWsVC}ko^ku{Rv=^S7Gi7%GX26^I5mYU0k07Ti!nMJlmr%V?MyOZ;Y
zPHtF*E(_qE=dh88i1dVfz?a1!VYA44rBN$gW8L`kw>U+y{@F6Vyd(!HgmMN#@Vkhx
zrsci#$N{wm1;~Y^FxV|02Nc#N>XAXF;u*lqG-sSmTR~jI3|(~>Ez7y38pcwL>uF1}
ztK5P4_yF?zb!m?Z;*BHsk(d#Hx-lT>lo=S)wE(7=e9dM2aaLxi4>}qo>2o(ia$W9R
zCvh^>o&?_mC(qGabN!o#;$x6>6Xx-nYYe-4*9W7u$y#0*$Tn*H0CK7sEIN7B1T=pp
zDsZkSRxOsArxe$gpu~7eIhsoud++7}zG^~oRe|rbN(PDkea<77c<S=6#M3%*WQmXE
zDpH6yDcn1pNNWxD4YOTd>hC<)I=-=8S|v;*#wiU;xH-t&w$UyTo-|H=eX=76dG6S6
zS-XGyMBAwKsskUF!%fa~ul@OiiKA2_rB?1b*1s#NHkZoN+$U)WI=kpjFFiVq%*S;j
zatidMJ!q})rKFC2C#xhU?oe%}qR1)|8VEjqBQ{X`CRZg*SqUbpqW4IO`S=)v%5Sg4
zSwm#jZRKxPik_4E6soV8U|JxV0g>YB`FH+0P9t}+m>W&e{9t7{x%E`n^#Oc247#F|
z3PkW+KWu%*?pLC>PxOXmy*<+1lINX`sa%AEk3KR@{(ml~1+*~(E#XT+Be<`6Sv-$A
z>x$q^iaTnK;yo?`$2=2SRr3t8752^`pvn8zh}68Yst>w6!t!=vKWp6y9K;$u&*Q9F
zn18BI#%~bro5_XfnIpvBs?m?&vdYWxIpt_C%&AZ0Lh5;R@VVTGkA;A=M9;_9As}S;
zJTtoi{&5L+n;}m+x~mOkEOy#+KDd`z&|W3418UW%HJOPo&Q-?0-&)Y@T%@C8KYH7x
zQ^FL?m|T7bUUB8h`M0||pXJL2c-XmXft6}5Zs2F{*7j{;pIyiM0bNof>YtDe)4Wb#
z{h{;cFNsRPC<{47P>6X-Nh@EqyQeHJxIdsyg!P6;Oq;*D*=52)*X8~)0uZ9WL{e*C
z*7G@%zAT*6UadRlP+g-4CY<mS5U=*0&QX2+jT35NvWL+8zQlIDFzZUzoiCN;1J2iz
zA?LyPV$6wyNV>mSGkM!*i%__~+@A}>>|Q#K8|Wh?qALX$Nj&dui$+2bKCNRLvX4DC
zF|sBSbw*Ph=S;4G#BGW?6jnd~+&IGD$mneBaxLbGC*riH_i&=`BzKjRV&^cZ>lY>3
zC4fnfp)CMAa8{IZMUTg^OEvQIS}&){-JNu0N!s|beim%=>z@O{5GV1%A|@dww+stB
z;8a>=FWY@2HE0)&sd%UZ-4tw-pkJ(Rx5mbDaqh3hjB((RKYLoyqw&BCtK}aWk4miB
z+rC<Z7@gj5PHdijNp6SvGmeQ@c@+MYk9sd*n&tJKR=07R@6Y0PJpRxt`-J~bFMukv
zY>)#?FUx2VxBZ#dQQYBv<2=GLa=XL^)D?Rzep?Ou1e;g#iThjrlsp-&v<nKMb3E?&
zjv6iU?%52I<m35DKrH^j*K5*b$Q5+NeM@0ZmkCgy%_a~}{~bcRE~>V>HC$y6ESR$J
ze@W%Qnf~mfPDPOc#Is_;q9%d_xds#a8tSNK(%kKiDGDQ;BvV%k{hD~HG81-VD~M-1
zGoY(+31?hYNv@m7r!zjhs7xZVSH*x-Aqi=C?Gen>d{&k)fENH#maBaKkfi^BX9KSs
zdv83Tt8lvDe-)OEyslAz;}CzpyEI$&1_2dLDC>Jd)FRa^D-v2}CDYz`P&-y-<M4h$
zqi8K3Y+g`rnVb+@QI)9zEWdHL!J@gus+8;Bcs=loC_#a$4tmkAHl_<+2Y?MTRTZM;
zfnX}kePgmSe@C7D<?^j1jW_SjC2f`D42TSgC||wG6k!f(m1BB_zZ4ye^`uTmFp(_f
zPBcI2Ie)Ys5KX5?Qo`-AZScAA-B=GSHf{Xfi@M0gU{bTqxmR228l)fd;Vu@1`7-;N
z0}^YvZARS8(@-TQnUfKXs}i67VUJ7P=QqJEKLFan>bEq01dWLC>%OB4Hds;`d-OTl
zul_etD>d((5od0M{=O|oV9W^^FPSqvlQS*><QZUaUkw0?TI8=%5ASm;n~`Mk4{bu#
zJF*y#OUxs%l49FXu?y_IAONLc@ghB`-Po}$66MZvBd6SJ{`_nm!q<SmBpL#4Wf}`B
zbBZvjec-~F=%Q$L8a>CJ*=T7)LNIYe7&#o}J^QRXqmmkQ>bzWsLwvy)QF)%1FLL;5
zJhH8~G{g>k>QWmG`j1{K`7F?ft}Ac#O=%`nZD)FwvDiq-(v<V!-G7$7TjA}#6yhUr
z;Ce_Qq*dMHQTi|8*sZ9Z4E;q0oKKvkl2&FS9C-8&^Hf`u3U!y&n_``u9(W|oT3qrn
zYBy>}+rV%DHz9HB`OMo8TO3GN`ogG|{K(cj=lf#nKtf%mn;oypkFpu#yA3Z@AU9P<
zB+nN4x0BP@Ma(G9pT2#{I7Vpr={CTMEF?I#o=3*7L~=m@c&-vZe~XRg)!;`HB7Nfv
zQrJOTG%+vl)-$bL1OfPZgx$Il><#J70q6A?qf7)q=0%DDzsqPz4GV%0Z98O*Dt-Am
zy|lJ+%3}5*aB$PDRg#7Qr!|hQ_sQ3DLcA1P9vM<8JH1$gDcOeAtG$u(V@zfoXQpKw
z*FwEaz!Bt|d-n+3eE4+u_%;>)+*r()9BfIv*TvK6On_FAMw*j;<24T3p_9SUkoEAs
z#SLmi1L3B%>r_iB3p^zrg`*4;m3yzUdz+&(n+m#&m=5#-`|4hjrZ+~)^&v&4e6+dZ
zmmg))BRt>3jowm?>-dT?sf3j9y!CJO()02#Z+>6pzvv_YQu!20{gU#S;aGbEO^(E-
z1v|iaLI}{x$u$ov=+pwk_G-%f`_U?41K_|Z9}QyjQ3(yp;*{ZZVOgDKoTmwK6tRuo
zxsk_ro=fa}ePwL|_$Em(uJzyio?XiSU|y=qqa5T)gQ{vdRT+WAEktyGz&}ES0bi~t
ztkgA)p};DY6~T8%%joWK{vVpYI;`pUdw)T9cSv`43(_glAswQmbZ#T0MQNlIF+n7J
z14?X@ZV)A;MuXJIk=t*4zSr;nU9anTPTc35`#I-~^i9fh20gv)Dri-fsw(FBG%M#p
z*4t3d(@XYE^N<X=LRK|0UNeTC#JhAPoDZM`zhVt_Vl?(HZVn%2c1->Ku-h1+c6FT4
zF||{6v)eHp)7Uh2sX5jX*NMnY^tB6ZJw<Tyf(>N2d%yM1(1?@VH>IAEk9zL!_a0vi
z7u>zi&A<!%R%x0}j!;34*}QC6?pA2uDR*^0lHAO58eb<s9q{J-2iU?5o2w^$K^(O?
ztB6E*91Y@hLmr|IhyuG5fAPq{^z0*UYWt()g5}Py%=ACT1lRwaTe`A&x!i!5Yq*<A
z^Iz631ly0zq+529p3N4Pg6)VSdnAq27md~FJI#a3Q3pTS1gZE!waTdG>GIEu%uJ?{
zK0$w~KIlG^Iz~TUp9-t0<h`oYYQ$BM9fF~x$fSB>4GKW}<Ex`~BSqZ94cQ~5)1SP%
zp_EDJwuaahdsz)eL2wZrHB)}UT^%$}ZRB!wzm>1MY;KN?hf7r)10CR8BbXjtR6O7F
z0V>pQ=qm4V3AyRmHt4Jsz!23mrAYyFaUV*};*3fvqOKj&zfKd^QC@Q*4APa5s@zcw
zL()HlWx4XXu3)GNi6!;W6~f`yqv=`Y>+a=ZH@F&_ce-bAX)m0fh;l{i^_aUcDj#F}
zGSb-Nyy0>Xo3la%!^xb@R;DgsIDC$S2ytHp$bO!5^A&gT49}|(nTURRyK7=e<b@-@
zcjju%ZZ$wq-RtInd(+fOb5Tj_#h;?{53Pw(Kd<!6jxZoh+8H<jE7N2>BuD_aW*7;p
zZc$tUuVq5uLjqh>YI*7mH~Bio_pw}02%4!veaiLTBS(y8JluHE+#<skX{v~p-1P2L
z{_w-ewpV^Pdg0q`!sR}}T&-Y4j*O6gzDZI;wNE*o{=lw~JvuluL@``23ii;fi&i?&
zkXUB+W63QwOa(hb=I;4bU*D=!rzej{G&`2l>8q|*st-yQ^~Jk9UBS`$6-%deOee3m
z7cOOsJJ|?&PeuhLeFZQ7`KrHC+ga78CUa=*5A8n5hCQ{|8}(*$<@8qP)l-zgeeE(r
zf&Dvk2S?`I79I>YSvnxVS3r^HUt1wCE=#2iF$f$3b`s6!^%A6f1n07KUJ08*r0jBM
zaXaNV8g{@yp<uPy<L@AH>v@9+hbwQyT#YzMFhd)$eiOh6hB^N!`Zf6@Ue+lexqRFr
z1iY9<f_`XZ@i|#_K5xRkZDmB1+mF3}fkgBnEdF6N1Xg8McCu*=GapuG?!svlaw5I1
z0q(MH#r4f(&ap1OO?N{Zab}!8ID?S6{&{xf=W1I|M$I<|;dcbtXN(C$niIZ*$+5cM
zr+jOy(h_vum<e+sWpiFZ$6m~mTyta!VhR6pkh~NNrO%mW_y{i&bGXC_wA%t(CG343
zI<2Hfx?^h91{AF~K?k*scM!Iymh|Prd=<6(Y56a2n5$OK_3_JN28U3;rj|PS52NJn
zH#DA#dhtXYWC}`T5n*MwChjwp`B7?BvUJY$<(VW4GtEMH9+$56pB%9y5kAKd<s28{
zka1)ai+2JH_nCsFKo<H$XCSroArUU}dHe162r5O8mOzE>IADHjRRg4}#j__hpCEf_
ztCKcgToUD+1{@^~noS%ijV>Sr0Os|FCo02yoRjoMf`r<>kE?$cCjqEuDr{vwts5Kt
z$WS$Pd+In21Lwi7+5S#e42z(t0SakvTp!*5*~tlXRT$5sN`#uAQBS*R{2%h`e!Bg=
zB1sdD*n{y&Tf^X`B-t@Be<@D#3zrQn<2ZJu((mKGw18{`FZy}o$VcKb8{v6_xcF&;
z`IRQ`&wANbrT{mR=8LU88l97oZ{4d86a!23ote)li@w;n3bg*`Wl+tDAAkj+<4xg`
zgX$>GKXe2n90JsCe}^sW4yD^cR&>0Sx3(glm$rjD5Wwi?eSd1SawzC#GdR6P+!}#e
z>c{%mfgzT&KJHXAUaOtt9V#)&hUB+mE~;3AOeL8-Mq)BjUwU0h=N3E)&)Na)6T0M3
zNcL&AG`wtxMvaMV4*%#v4k41#mu*qDHlPzJGE#uNr|ZUz{pzPhDjwah=c{Q-&-MtI
ziqMHRlDNF?FFzZ1a&rys+<qBp@v4yoQ~hBiP)_u#(#KJPVLS1cHugICboc{&PGYq`
z?tJmESlvQ2O1Wb9G_esEi*MrZ)QQvh1vLU@@8RuxnRMmPOR)4p1W$(EnQPjsUh1E8
zTX*OIB<uj@D$N(Wa!o2P;EX{l&pk4TmjAvc&og|LMF`Sa1ZvK{ZP-;NN2!;w1Ohct
zs|Dm$&qS(O*FQ8+%=<zQawR5slKac659jlP2sG{5y~l*IO$vws2)4b8*oG_fZ!yd2
zrlBW7h^vMNy)W+M?nCJFx9k{*OVzJ|QBGb_bZ92j#Ne=TTKoSDJ&JhPoI-M}CEAsn
zUv65-Ltv%KU|tpN3dRyS1;M^pVx`V4peAPu#t@?587G=M|4y`qx`rfg$Y?EHZ!v4G
zEN%8ZlN*bz0@L8>8Y6(@6|SO!r4IiRsj++ph10iE9(O#8r!((*RdayEx&{*99tMBK
zguQ=?LLH)Tm|y+Da_9u=NYyWdPb%<$#N8*GKgOpfW=HapyNNExWY#%IJw)ijQZ^>c
z=6NjZwL2=_Ud6Fre0S-@3gzQ=f9-pyG-E~TXC4xO_|K|MA^l#-vwCy(P)oks!$tSH
zX#jLoq27Mw>cTrkXAtdHEoddVGb5`y>A5QthwzZ&nn<nA(=e~y$J9$<8&Nb>G`B_>
zP<5pBJnMSQHaNBoVv;j_XYK>~X~^WxSgRR|8xTlBMCGAfG&^${K+RPAo$wlxRP)G!
zk@&dd(}Q*eS^V_#?y{tvBcwoBCZ*pZtLIyb_jQVJ#hT#NJ31FJhf(-o&*^CmM+;FK
zhCh3S!N2JS5G;H+ua0_*W-R(kS|)^{uY`A1Ly01Bl02Wu@CGM3rn_>9yE0Tt;~*GY
z@?x(S>^a6nVzYNg4hxFI+y_FYNX~&Pz|xo9N=G_t|HISFpgVcc!hD>A3xYIn1X3i7
zm}d@TMVIy?B>e*vN=;o6%ZL$F>5%$&aSnB=YegCSvDcRK$xMx^b$QGiQV=qJ7P@Jd
zUDS8lnl8x&+TCOU_md`^S}4Ud%oQaynOdpZv;>j>ALc?kUy`kguRypDvoG$W9JwP~
zNi4JfCww28pxqzrPsW|(7e#k-y>Bc;uC%zFiP41O()7RZD*)me-97w#-2Rl0@ju)A
z^s-VI^gribZ3w?d&qNa_<oVWRU{iPT^H6RK`Z2bYeqRI?`4To_pWDM>T1dK!`}gDV
zmiHFQ2=%5rm@oLP5-Fc$(rrQ_jy95Y{e|t#MEHtokXkaAW%zwg+3g97u<Dyz+G#z%
zFM;2pHy<cY@**sX0wKHW!q+B)_D5-;$X{xhF#1kl`{?nHh@oQAoIFR`c-I}26FNNq
z!*II3Nmn$hX6PqIe<czqhL-9M|CI`=1Z%ZlVX8;D>rwafpTxy2i0Ibc-oA|=-}2(w
z3FSmZY(&>Zr}wdQT3fetHV0>>DgAwzasjQ<hL#|_y({cnl%v?9q-Ev-&YMke@+u->
zen36J6ORUpDC#M*I8$l^iKn6=2A^o&c?7N;zM}irC0L3JjF>e)V*AMnN-%6*eBGrv
z-Ep_E<VXZz_UFDWvATXl{pQo40ty{#u|}MZHG@`G)Vdq<S){{tNq|MLxyQM~2cOE6
zBNlXK_l!I$Z{0qX;3)+)|1$$i0dqpj(ElAjIRRRqAyivErgQ~i3HI@@pT$M5auIWp
zazzbShA<3$rWW9mB8wrj;qaxbfS5-pY!z@x=4NOISQ;ws7Qx`83GlX@Y1_%7W!%d-
zc+1PEV@W975TdFhEqB;TL)qzUG>B`r{rRMTbeamwRU5Y-(0!n^+5ouAe0X;=9q|Qq
z0J3WUq$QbCV?Yo*JVO^M>oDl0|NfL}|E9gYRiRP$DGmd!u~Nl^4Qq4!N)Ncy)mEm)
z!t2MvBi7P0p?^>Z;^9_q#5?Nh<Yk<!n$ulkb-Iins+M@Kq;$<zQazAg^i8pSR;^fE
zN;j_JtV$FK3LyuTRGKuLPRXJuYUxk{ySOb;bS&aa&5S&lL;S4JJ{WxVknB4L3u@VY
zKyQjDg#fV6x8UQsz7iulVsYuAir((`lmO@s;^!eBdjy9V)p<cS2cDcJ;g>g-Zt9_0
zJXnn;e13o%9$<J7T(bIpyy!mE0idO$tSNiNEvF^-2S)c<4p<Utj7j;Z_`mol1We-1
ziOj3uQxdt?GTc)-mDwI*fWXnCExcr<oTS^}w39Yc3e=vM|D=Q0pzP{#bKZ8wWE1pw
z8T}S?Kf^r|7YXE^0=qFs8)8mg+`#YKB=(>-h_#_6P`=_-h@><*`RFd<6$nQ&UzlMg
zhd_#H)1czJ*9oKmj_TKr+R%R=UH}l8NXsX~1X$K{S9^mUYR?iYqeVa<F}6L<iaxxb
z)JwH9X~ERQ1}I%}ZWhCRV_q}%g;I8JF;iiT)UU7lxp}uJ;ZGtYA2bIFXLjyM<e_po
zni>zW^*Yy~7E2OL(6PQfu63c8%DHH;L8iv17F1~CrzNYN3OhjSDGHvg!qkNt5%#A%
ziup8shw6C@{_*~aR}?|;m(<F~O_EP_@m~73TIiy3J=*roLqqlBSgHS_Ek4T5d!pv*
zHJ@EcL&iVNbUY#4X!_20O^vXbhpf64q;i8>XbxDz6~}j%nMWB|bPFCBQbQfzaXxVg
zqn)bv(6pucPb&oKO)k`zY?$}G8;5O96#EkypuciBi2c4C8VftflW-V+<}G2?1(M?V
zVsCxZe&^CfBq4?MnkH3Y%5QJ^R448F#SQZfK<h=6LZwj0AhP-lr%F8cK($5s%Tj8Z
zudz81mLP5*7N?w-9wR6dCmZbY>8pfp3O+Q450_?uv%UDSIv15?r??IQ3Be{Dvtgs6
z#60gXH=ytoPIOFMmMEY|MW9L2DH>C&Y)^d+HTCfBcQzeumDw??dMqpmkOtu=*r<yq
zDTlu+ew#5WDDIOXeen)agTTG-1o2d<7UxA>KNgXsllN9pk!i1DnEp)VOx#KgQ%f^p
z4s-JkPpT(UlUQblwt(%3Gpifv^HD<Vr`g@Iz~qnmc=Ht2GP=PfRo~g%G*gxOE@*iM
z`P?e$-*hd0NRXw1FhElO+kYr>tX4}WJLuH0K%dgjcnk5V>GnFa!ownS&j?%2Ko^FC
z{7pE_ectVr%N`Ar_Y?a>vc`bPhC8Ef;v~|Ivd5XLZel**T;wkYL~Ojfv0N1e!{hTh
zKm<sun_?>fd0t@eF!4ukto9oRg3W%OB#v+>gVl9{HC^?mT4U2uQQ?-yYMRtwm;ZUf
z^H|DrqQhOH92JuYkS}deM3kzJ^-u%p7;pb*tg1}h`!T9T5=*~ZzoU>#!+f8;r}`ko
zmv(99;Qi7G>1V<D1pFMn)|{ZJ8m91p5D$UsNTjD~7Y`*OHtsrBK-c0HIi1Srt|{dD
z_(mY1c{tKW+?XMS*cF_=wWqjzzxy*lk<h}ln;)hR;0E0Wjv{8?KV(aNAY)5&qQJ8d
z(#Mtp_z+f{=C>u$bMmP!M?F%jP~ikH^|AKvcj6&O<&1zT?8nAQ+VGzZUSa15MelbR
zRYG9Onw<|0IcmwQ6IW?LQA(aRG;%TyukT}h>83c8FCkXuT)m36mUU-;Ds2S*5#J^h
z22a7Azg`V9ncsgOAK`KCx7|`m##}gJASXo(4g%MS|HtvB>Gi}-`m%fRuA9I+8og7#
zThv@;ukQ#1PXSnDuZGLkA<`GMUT3)6st-ajhz`&-L!OjViud27i<}KNWZ4`d=S%+y
zoie#-e5Glyh!<OMX%42H!R9Ux5IW`&W5MjMl=HFcF>$cUnchv@M3We`T7Bbt3*ged
zm!g(2<^p`$;6eqf&!hX{{J2TZ(Ior=7<zl=Xb)yI$VYb!@$D2*9jQr%?S48+hbVai
zuWViQmJUv`D{1^lb*n&<1ey~41UJTju@Av3tcVx=tBSvWfBmhVQgCsQo|R2lhm;>Q
z$&kCq6Qb)9^1y`i#OU0jhzGCFl7Yt`AI=#p22d|kL{Bg}H%Z?$h?P<a6xgt)e#h0d
z0q3j7<*eettkR+6qK7n>yfbIP?kxv(+<aI5(U|cGK7o4o1jg;|+Q0v(k7B!xtlXLq
zjhJp=U!Q{p*+r;URp!Z8*MQ<@tPg$AeZ8snC1>G}!7a>fK^n&T5DX0y!B#gLd&=Lz
zFNiB-rPclhV)QfphR`!1a8Z6Jm(B?kUoaQAlle^EqXG?CDF?}3K^mHB?JZ@?`apmj
zcMYh{d-n%-*A4GRCHr4|H2Kenx#JFAdjT?H=d8BP2i>{DPYZECNkJ#x`?EN2WB{-;
zN#&2-i;z0%-fJdLC?*kJJCDuS51t}N5hI$rt&*VBF|BtCp8e6rtB%L3D2iVttssAj
zja`P7%C9nmZkf~eIrBT~OWeRNj#9GCDVB5j2V$fz%j%=e^m7TX2t1Csr&og){ZgeO
z&u(-~F;as)niV%9x&nA`tdboD5$SRrR2$wvg4Wma=)1-kL(N}`)ls%`n&^;tvvS(<
z#5jF=N0PKV5}Wpwl+!jux%w(HUDW`uu2)DUw@cjf;YG!;JDd6HsJ-|>>{sDHB_{ia
zIW{gS8MG1zXMfH}nPB?!sSHz@L^*aexbff}6?x<0mc+yZ>lZpd<fIT`IP5c1!GK(L
zs1SL%wN^FZx%||Y7_)DcJp4+ktD(1@=V5-`eCOyoB<ANAVw^<nd;(DfcPyiv;|Gt*
z$um^XG1MJ(KsfDEIICUM(n-iIlpxhQ+ST$f;w>l7V9NhdcJhY*a8edX!SaHr_EB&D
zN@10KO0?Eb?1rx=`<mb`Ze0<*1emsJ4nMR@jpy1Y)VwPw7DU2C(E2d`jxJ~xca4t;
z1s<}6BATGQmVbyw3^i?0_g?l;IwUC4@<+Aa1mFJsdfQg6HVhXc7u3Z?os`6!KyfXV
z=yH#-3rR~I=>bbON%OrKBQQt5kEpjtRuA4c1fVJ#eoeqZSXZSn2uRH2dtF+z4fvi+
zKt$Tkf6PBOp>T<n8_}X|t+K2Fos`|Ksvn<L%q9~ac%1+F_SLGbwF^%;o@hx?6Dhf`
zpMP*$f9^h6&WcDyEIAC0MEOSqBb*`@1H7<UV&K6c8q8{H$>W7P#w(5?otwfmn+h<-
z6s|vnP?eBz{;i_N(`tvrF*F^#IT-_(`o>eAA0+K;9q#qGaN5S!sJ1K1suM<$3zi7$
zrv0BpjYA%zk6Tf`!notdl8s%9ue^j9$+53*eBT9gq7HiFn3!{o9Q~2k8c7X<&$U1}
z5IXia-1<>-vw8W!7msD46~@{wXpOYd>po~ve(-%4vX{+K43k4wqD<P?#?n}7#1JwR
z1Ej8&h8s_ss+z$v0><AE-o^1a7`G5Pi=0Gs^#}JyOh;}9mM^QfrXZ5M8cP-tqOyWu
zdR{3_;FD&#&;CtKh;eSZF8Y1~@nm78M%)D+MKiobLvC<j2G<Ft(kRvlwoQhVLw+>5
z?7JH`_O%wEU4ix-M#e3`W7i`{gEX1s<;@uPEW{ER$9J~yg!e6u+ZH3SbG&fY5&Ogg
zFAi-!(PzgfuHOBoAi$XCk1MwufO#Lso-Y7GmSFoX7OcFZapCHPAq23b*eG15r#7i)
zJ;;F`-T&BpDg<lsbo6ka>c{nVhH(QH?~Sd){`(2seP8m7YEwx>(!+75@?F2Q*x~s)
z-9sS}+}m9oSIC%quY8s!cXwEHAKlmw4#GHwYf|sBd3rCUK+Pqmf<&5vyJuQjc*eG&
zOuHQqt3Q@Z*LCZW#)`Ojmy-B`pV25g#DR4BvlpH+pMIJc$Amjw&1WuPc)Oer1dMoB
z`mvj-yYP<$R?-5{hT^YZIh!Qf2B^*=`qI5H#FrhD7<@yB35?q#y<95Sh>IoB#yy)N
z0YZ%iqOC+(Nl57K!^y>YCy#{(8xU<WS4Sd2xqTmESvKA46G7WnX$1HC4-(%Jiapez
z<G3NrDKL45d^=+n#r2)=p&$@fQk?lQ`r2*9VVx^f{zEL|RhzkDmJQ&w-MlP10Fgbj
z8OdOuRWMzi7@zQPrept}ZR3G$QJgJGV?O5;%)%A@&oRh`Gw1|u%VV6r3)#oh^d2Rp
zQSIVIbXb>SXuQz9AN8iPmNL%1cL?P{t%+IIYhJbAm)wIb3U>bqszphmwPj29kj1+w
zYMTlvUc4lp-%4U(02k#JLKD1C8?-cukG7&kPN=*2lr6N^5TKIh*+n>te>iXqIOR4n
z@`11~0zC#WLH<_|-^`SJS)Tf#9(BkX2C?9HHejVMAllY$qq^IdXwRPLnK~tyVu|;E
zemA*70rGJs4oiKv384jX*><3M{_2xFhm|UbbU~EjAK+^S^R<EQa50Q0q^Zhoz6J$J
zWSwLZH|}7ld0z}}XUI|jw-98*OkAk8Q7=;4VAOdsh4{dw9@po+1W6HQY{aGTTuG62
zQ};ApEOG!UMDR`9yEwb@Q$9Mz9!?<cp7;!ehm|xpuU@>f3ssoE`>$Sce6>SfR-H7I
z=%&tkno%KX7vhc=r|b%QL-&6?UU5`&(Q<%sp{8^pJPTU(_e3|NwF3nwcFr`?)W{Kt
zGW}VGz1Kj{YcA_>hJO^`MT1&7y(=5OC~%*=4-O{XaNQQ`f&-nHDf{?Ke2G^L+6h&r
zz$X7aA1TK(P*OlUFnoJm0S{~yM3|#j#=jC3&eU01QoW_WxtnRZ_INoMzZsJJS2~$)
zpr({0)-&=LUlBYJof~3oVY;S<VmvWYIorGzNGz_<y~I2lk&S|oN}h^7uXR8Q24UvK
z_`3Q)cCoE?ICz!mZNljC-jmgm=SoDUW@&E8`$3W82&9InMQvuBy4U!Ey<z0g6on5B
zM0SHNl;p5<e9=3Gsh=${Z+!5Z=(cFmvXU{X3qh_%xKY>t<>=L*7bbBoWO3QZc)`K&
zuJ=E}5tvjr_F&4BE<ngVAyIoKUc#S59iTi$1t-m;VQiI?`~STFF`smAgkvlI^FCTS
z2pEx}H*qO9Oz7MO&ag$R27gkr$rdV$U(%CWyX0A7x#*yMGBchhS=}XH_Sw=T41m@{
z$<4su3jKgWh&8D>L~p+7mP@N0DxZ#6zs@WCjkEKFw+A~RAjFe$oSYPOJ&L5hZZ<&T
z`mxK$XfQ%Tw5z+Khj%HI*Ru&~W^YuEpQFJ?_rVaw!){pi2ESynTp9$oba(cL&9gZc
z0-$3<xaYm#VW1`jwzRPJexjJur<-_uqAHILGrOZ^f0m_fX?dFiO8B^+@eZ#nY47KD
z`O+7a?h$Yqg-28oC{v3Qz!@iRh-=PB<Tl0-pfV%BCi=uxPLwOK5U`VxUp!3i8|nxU
zCjs5>^E4`=uUF{5zl&X9&Yr<wOoMRGQ?w6&6O`YH011Ur*L}zd_R2Acs=zzO(0Q}I
zSNG$Xhc*_n^$beu8+-DgPEf5rSx%p~dQf4uCFAxt;`YB40(xm(%Ua%y!~<IwP;m13
zWW8#naiv<lF;v0JtDp{`MH5vm70^@V<8=ez>ShX`I2smH_YY?}W#$dIy%@&_0j|vl
zKbIt6%?A$Y%bV8Pw%M(WD0t4o6JL+fKl=VRhnDzs!j8_*fS9;r0fS4+YWhZ)$9Wy}
zfnxiAi7q~Ka%YtE>_0zm_FP(Zi3P^wQT$Mrxh~68de3Z=?ZkLO*h)HtU1I0>f`}S^
z*;%ry;3~(ua1)Oz^5_6H7d_bkjK+!Y7@f<tj|2%bjkDxjcnf-D*X7B(?NonX+&|7y
zyJDxIeM$+)%DL|3I%g>4`9&XK?HW30;ogfgj+tJGrD_GsU5|k)2|Mo}&8A~VbyqA;
zHM2utc;?KuZC|Y}Q!^wLfg@>sZhG(Cbi>%Ob*-u!VFbL5goVivRIqy-F}|@ZiSqQ~
zU7<u|b9L^Sa_OYW-;&zA=gVH8aktq|JzKa+-8jx)=CnXU%g6H`Xf2{ZKC9W;EW!{h
z%XJH)K=&2T3(h!_NeU`fu=CEcFB#h(%3I=uYEs%aNEkQyOa)qG|MtQqS}i2?pQX5*
z({F}AyVoAUl&niVmg216fK3Z?{qkw%vJ`prKcqhjN6}tkt;NyY@-&N#98X|Epq@;b
znMka2EzWc$U>LX&b^J#>>|g4IAl`JE^%HQK(poaM<Z%R1(0TSlaWIx^2s#0}%P_>T
z0tuQMkn9o7K>0*H_OL(F(py$VEF~!vsz|ZKfrkWNr4*EYA?Y;->qFr~iNpf0Xm^i-
zwWey7y-0#HQ@OuqjaJh5-6s9s#pw%k_s)P~naWEl>3vcIF4k)MM(Cva;kElkaK<^b
zDjH|`Q$JzM^9T^i=lKD!=}CF5i@Qy(|1Qf9r)$YZaSY;4J}aY^bsN1im|IP|f-J3C
zn-P%>z}7@l0axKYIf`?}=FTz`bs7G_R#Yp;H64mI_|s(_nSi0r3~(1M_)?wSq?co1
zWXKmF*ZxUK{a<ZJ@Ixy@dQ=XvR_b_*pp{LW5vbRL92#eD5+M}EL&oFK3$G%0Dnp;X
zbgl}Ue$c7UxuTAqyF-VS^gr$(o-8g+Vl9^_b}vaT*u)Vy4>Z2!X#Xc6$LCn5LUsN+
z5m1$U3#^)>=ISn1kAi2NRL=9e(UXKD#`&cN(jP{%XV}2I*jW4k7}nLKVUfg(7?QF;
z+^1I&Q<JJxMrZnu5?>;JX42va4bzuvYXO(W<UKtUJUH9$Mk}HQet!@iOglzQqLe(J
zq<+kU&}@v^Q^c`19fa0b=WPU<Pmv8bb$tbHB-OCZy8B}5dKPo#EPCSj`ofNXeM0b4
zx2W0M^@`YXQpUsiGks7lrjI=gxD6BzN#3XbSzvBc<0w>keI#Vb-$^Q_mHi;2@k)*^
z$F85YOfXOcZ6;nt6+s;b%nN19VeM-`2;&)&@7?q<6!rxy<Vk>q?G;i>al8TN=iqhS
zb3$qCgZmoUJiu&|bBy@fz;(Bj*FxS?8)}inJ-eWm+1XgIh|4;6{N;WF7!~hP(ED>s
zjh;SLEZpWK0#JDzX~lnGm*kPGT!Fc^nl`t7^qUtP9;iHFhI-?0M(7_`Fi1B8k?myD
z@RBv^o;NyVEjvllm(h6%v6kBscHF3MO4U>+QVl{);@k77wj)1r#ve+#kN(grPB^3c
z0spQt>~N)k0yVN%k-U^z9mrn=asUfIL<55+W=qnSAIP)Th0b%KwGjK1MINOq%+e~a
zY3-x0l}+C-+oGux#RhOmNp-I+m#zl92optty4$zar1<&E;%j3=X0)}oZzD8Vw)pif
z?$yt57@T?XD4shw4z0^4XmM_bf|CUX{j+<vu*d;FXh~g`wu^o;tmLWz{CE2Z@tJNZ
zi_pu!0}H5k3t>k;$T?`dS&2?x{_qg<+{&`4_1=3^gq}S@=oLVd;@D1|Se+t}w4UHv
zTD;*zeC0tt!9k^QfzJf-&dYV!8P0llNkio*_rHQ-I|a*+1mhmN&n!(QkjxD|J9{%a
zugRlWua3BS2U&&ON~?QO7t~;KF=@LAc(GBtbA-~i5^sr~dYiYDpb_7LzHlIUl5Fda
z%<MY>`@GK8`nw2+nU$Rw^eS{|V;m}iE3vmiRk~dw^7t%7NbVTJ@I1k~yX0Yb7Z=&m
zw=(Us%ezMs3aL9FoJs9JG@qAOG+RAcjuaHvMmGTba%2lb2g+&#q<w!%Q&Wq2ep|Yw
ze8Odk$xl*^z<Y=t#E^zU(5a779dClxQl`ytT0-hE0)(!C?(`S-tB@-*UR&Y4;tZGr
zb4?Ra8ttlnkKBK0BPGL00I~(wble1(5;XsAyQXGf0<e%@i~Z+)eQzHz_$a7H_R$7y
zP|UA%Z@mTZHOZ4*5#0v<chR<vp|iOzo(+ng?|K7>a1yAWD0^7I@COg_&!qOLmq>&?
zS_@nhbMM_9l3?NmVxT4i0nm6V-~Kl~<J8GdC8y7^V`vN{QM@>&>l%9;f4@eVTAcz*
zgZA|b8{{kmUgn^1OwRwz_BqrK%xqUQx-mJvfO{JAd*Z9sCl<FsR<*qggt#Y7>R7M8
z@X#JnBeda8MrfJvE;b6%7XZWsh(|$yhk)_B_!j!>J2Ys9)-HR-0TWbC7aeEI$&_6c
zTS#I4_fKQYQ>6H30;XT_o+*y*)aX+b;7P(S=z?(gB)!P0Q$xRBzV#_CX(ORe=$Xlc
z=UuPmzc2%&6yYK(1y?VdjDkn+Nm#9TD8sqh$oYHqOj8<^&q_8|4j3y;0@}5>`cbui
zYxZmBwv9=bDT-@qB~iyM5iyZuX|E6xAPCF3_@U1`D;NZb(+k!x)?$dWv0q+Jwg0L8
zZF)QmWuc)?(avx5Iw2izDu4RChN)?ckuj#;ZeR^%^^QGCDGLht?&CW(PSvm-1Rbdh
zVO>u?Mb$S@5O5TXa|#|KKA3wJV{l4N<_EfVK;;$tSVLY@PI?s`GQUIVu8?osJE`ik
zyJKX4Wb>81SLNTTBrQ_WR<Exn<$+<W-@0TlwK%arj!6tq>~cj2&C1gDxoJabXAkIj
zrSJ+6)^HzJUXw;g!8Y;b<HOGx_1MH%+H>An^9mVz0mM6j@1p+Q@+b^{1sl~oAnvzs
zHAokC`(XEPmBWoDVu^Sr{WQaY#X748Z7UKi7vg~6L);Un`0!D6ULWOzc<U9%Wa$X4
zL2x@w`K9Mt83No$gW0FNFEonZ|0?F9;`vwJl)iHwD5SS16XdWlu#xZ>>qJ{d+`@5g
zVS4W{dIeUzJfpL>iuy|3s6JT^3-H=U$RAW?{Y(T52*YVEZlmvZ^plmtr@_(0r%B@Q
zfpmO=QJ()%fBq>-4k4B<L<E9>v4wYec^7J4<!!4K|JKc2g;B>wRgM`xNc(#mf<OPv
z>Hd5znEUeSY_hlC*TI<oL?H9J?V>a&B`T`?;Hs*}A$zC|Fb;GL+rJ{Bk~}EW{Uo9b
z+8RPE5x|!g`kq+1hLKsSwSO97Ue^ou^C>OM(!<`P0uNcK!~5B%iLO;2eG8R%FQ>{A
z)kvC`%tmDjf>``@?4bJFVFQ{^i`UY5h7T8W<qD~Eysoz@5$Pm-(#y2o0J11_hSdJ)
zfNQNZ!*CSnX@uTOAvRc!d-hEynWPPIP7;`_Gsll;jJ%F{odM+)G(dB_?)EwQ)~_Tt
z+>wj-i2}8z`jy14@I}v>f)^WMs`?&3yLhd`i4NMrbD(C^{SWVu<>Y*V0iJWb>Mty&
z(eaY2XnnH>V0ssqU41<!>pM-$*w;aXWuHmh(HUmtwtbl0r#Rw?z)?BXzMHZi(N0aR
z#BZc%;<caz1|Fly-BQA0^eS!>1IJN*$>|6Nv9VF|pmbmpeU}BZjFjacJgi@h^)>A7
zK1#t5A|jCx!udw)ua9j~;iqVE+wZb+hq*+<CDPMQ+?jfHv7dmJH@6>CIA=1I3<i}z
z=X`2aYua=Xnf)&GwV(v2&#W1a-TwhCpsobp67xFtQ}M)p6s=Jf$xHH3*qfJqZH6*x
zOHTTjZ{77+#Tv3m0KP5Gc$ify)l_6b+TT;1rI-lIjEZ<4;k|Xz?Ea*WDV+xl!vMuJ
z)Yo=nMRX<qcaui8W0+Vq>`Hy*$+gBuq^nk6IA{p~XFVqf$=+;0Oe1gv=jjDEAELaS
z46en;u@E&uFvC06oushB7Tscj(MJ62@vWhr2D3Hia4@R}c6mMHwh+vZCLeYwjKWc&
zxO~m(Qw&k+WK$R?eE$py&(qb!?$vdXBDXOC7oWC2;a~eiMKg{OawsK0i<=F*ySE{E
za~k_&Pb{gMwqt21@k8xq2>a|)nn{=l7Zb?h<5OJ{9?|q5-Yd-;`$gY|*GCt4xA->k
zm3bbFn)Y}qF|iRpf=Go<CjRTTSx`t9|2J5VbP9qx6nFV~tq9=z6=6di17zah!nXbJ
z)D*l`ESJh~cM-_`adB$s3~dmVqLzd1&<iP3%zt_;&hsy?69fP!RqlC}-2=x)1w=m5
z7W&7aMIrYihzgMNAtEC5dd!SMOzy+ykU0c|zJ@@V`6}U?8?I=wKKtmd8`XI#)NXR7
zObyMI0OlaT=R%a@$wsszhk|m48D+Y>w^i!VQ^CWHxB!MbxFm1K7F{lGkc%+BH*zD`
z1c`@V_4n&n8}#;iC@l~K0O4*O`Q$P?)i%_wAgD<nMLugrz$q=DyvRrZKJ(bg2P4Ol
zyml?3zJXKuUG1OVI*ZF_->ZNeYOYvQ>}L9FI`I?wvK&I^w*!d}5f49OnW_@D#Q^op
zLo62r=*sj4=lk?`EWef+JNHF$Z!6B?Q-`L{NxHvXt@-w)&(#`I{xaIrxP_tDgiYxl
zF>Vkzl43EaX6ASsCk{z;Jt&G<TI4WyqBs!E6{I>Lr94@|bb8xNvcCQWUpcIc;Z_44
zzVPR5zgv=Unx+Q8h873lq(g}JNyJ%~P0?_|gv$2>JjE0|=&V!$`(Isz-C!%8Y_B1Q
zvtFK{o$3O_V_%@;OUYZ(LLh<Q{U!ud(z{bClZwp*EcPzxMO5B}5|_XFL1sC9!U#C`
zK)-slS`X(9?V{BQ8zbzN`!emG9i1U4y;~&s%+;1O{0mB5n){jwMPw=$$u^#`5QLu)
zCCK@Z<eKicQ76{Si$F$DYawE2owXxu7fDRv7`h<Vw-#!X&zVm9m~YEO)AM?V$N7C^
z$`jt*IDHIYj8R#{@>rRAo)Rm7YU)nHQ6juhwr*VKeJ~+pfKfP6p{exJ90o04OS|5>
zxb#OXsa|22qxNSAJrjlfzUj0h!`T16%SMU`q)+S%D9%|n3f@};bsY43uBi$G!XrKt
zF;wm26^4#n$ht8R5#-gyA(iNZevtVTPl`Hdat<+3Z-X{IgaukE&KskbKy|=HgyM>j
zTZz?M8Q<}kT}=(n8KMN*DTc--RE?tof-Z=XN2;q)FK^cIOR+!V`=a917_f$)+S=~_
z>K|Sgg&C@Ao=W<t(9fgh6k4CLlA!sg=*sT0n<84*dsIKC3g}^(2&tZ2Di;&QJn_dw
zg4<0b_G@OH9!zKPoC`bbSLzaYUPLx1U2#a};!Qv<nJCL`eoUoB_+%@Vx5RCpt@h1j
z=weJ~d(Lj`KU!2Eam|o|6_4&|l^{Z=ddzwyd-^%Z;$4VG(0WUd5e58?1l+M_4~XL%
zp2aD;v2W~h!xkz(xWixpGlDnDC^EDOv<F2t8bL0qpRm5nTC0weZwyETo0C1);Hn^Y
z341>wmQ2ITnTziRM{o`wa})`2!Jk(HdrI*HAdY@V>@Z>YLR0=IaRM2_D@|<=#44d|
zt2OCEe`{?>M6cV6-;p60^5>-E05>QYov;{UWOaBvrK7Vlp&g4;=CSn~#mf!%U_c(-
z0Q{SGqz7iuGZA@%X!iNb?f2GOFj!z#_F)@L&6-mnnZy$CTAiYMP$aKWT0`bzbrbEo
z4bTcvlvq8_yU<FrxUMOnxZ(DE6)!&P1Sc>COB4zy5e@aSKl6^};O1z9N+8b(QA+Nw
za-@>P3|Cd`8L+>)_+~IW{*CWoh(x*~Plt%odehuL2p=0Fk_kS1!!~ss-$rouk-dv=
zeRmQ)eHZo4qWFq3bvp%i$o>~nt)4&W;|O=65v@7lC;%6Zx{$8{E|z7hthiD?*F21%
zE{6&jn-ce)9UU!qmp)GQ#{{>@ZI;|NSi4sD%-Yk<ob=BOYIDTIEAa{CKD;5`h;0@7
z*h!~Dau*1Fdf0++Wjaj2KgY7RgGY@hG)sH&BU}LS8j(S{Q8tRqSABaR9C(;EOr?-V
ziDGW|IdUKTHt{VzIZ;-R;=us`M}~J1*R*--`nz$%U8}{-`|AD;vm+);;@{x}$}47&
zr`)!CPx;ST6=}_V&<bv_nzaV<q2WB|lr8+DdS<6`lK(}e!PEGrmIK4aL^sYUVVktT
zU<0UYqHC6S!RSx_%l`gM$9mo596%8iLK($qPg&Q|eyorPow`It&CIcF^P?C{k5!q4
zb{b|b1*Os@c*yYBT@B^VI>naWeP43b>^Zx5VN%ETN5?4=Hr`|3<+}GNnMQ^IAWWT?
z83$(q{MTjg-^K9DD2DZN^QFwS8`V^wm1@gJGG}Hq!?`!2#J;yra~F3*uzU3<*@lpV
zo`Xy5uq-R!aSIeeYY&n<plESxI8SVG`*~Qbcq$&#%nYvgN6K#<*}kl26kAv#^AFje
zp|4^pabCOf#$9P07^=GvNphF`SAsb2p6SX}p^aZvOvD8ZH-vHd90<9Y9}$VahSS&3
zC~GL6!Qk}sQ-i}Q_g(;wIo?AwUYJUwbYqrQMfq&=lLquUUC&&h0x8*dCc8U4iNvTN
z#Rr^9Ilj4=#t4+%C<vfK@-hVkSYLjEXxnzN<zxPqUc<RT8~F<k^EPEnX*n3hm%e&b
zRE)8Kmf5XI_>#}Ut#me2dIYOruYd^gl0B^|08TbEp2<NuwIi0u5rc8;#Kfa-%@Bgr
zaKa5DgCGG+&o7!Xt(a`dAQ&_o*}$s&x|Zp-MKjX{jd#B^;p5PR78U$z5|)IX*$$^P
zj$p35j2QqnawcY#!&Gw<Ff022M!<u~4=s$RJyCE3i67c_1XDDdv(a92q<<Nk_Uy<>
z7FNJJiQ%>1OhusT__(2Om#K@NdJ^+<obZy-Vqdf8NWwdeRPq)rAz-jeZbs)s{;8Nn
zIoUi{bS>fCy%+QSQiRimKcl617teR^h$t{I{mI%PaB1wy>8p-5N1i+-c7s7z6PNBi
zBkE%qc88RAENTJOL6SmW7G}i#J#YBYUiY7J|1-0Ack=2TrB}?IGIpvMa)7QtDwPK^
zXRCoZk8x1!oJThUf9b{6g7cdzfQoTm)FZ+`!364kUh0cGJS(jWh#-VkhWBJ;mP&;q
zy;SW#iWuSm`$oiu@tEFk{Pk`Kg4nKnJup8<yBnBI#vR#ri+=t1cZjgDUzy(0=3tYC
zWp`P4^qf73avL#>Fra_qLXhw!!;eSTRTywzlPa&R%l6g2LOOSfitk%-@ypu6l<<h7
zcFDI5lK((0gn5_?32CaUy7EK`3Iiff&$J-fpW>~a(yWy^iAyBn#D(sEMCM@N?`M=G
z#{10ROz_@zNU<{l;O{=qcCnE6<@7*?^liP+UB1EsrNVvovs2x3Y^Q9^-wp-yMFy_N
zSfRnJUHBG})-(S-tm`H(j($G&D_<M$lTNt-CY}rgXU`PLP@>?Uym~i8l<8A+!{CTn
z+AEzic<*6z>W_*=fs#>BT>X+gCkOryCRUE`!AhapRFGdR5zjdW_T8-<Is)j}G0t7!
zxkP5J9)3t6Qv@fEZpRw{lxF*oN?c9Vi(3%>KpH8CdpvB!*2p}m&KRD4JOc_PN2emv
z<-v1h7D!Q0??i`ZJMAQYd6oiJEn~nsfEZ4G707uf&t@rI!bPLd;h#(Z_8sbPVoHE=
zQ5C6tT-WDkA@)m2mU<P<tf4N+e@mjAikHMlC3tiRzo(qXU)~A!eSXl9wYw<x^@YnA
z4-d#$sBq!O`(JXjI%#Cm#q*ak%Cd|@%-}W+b?Wx5qtCC7%n)W~VOcCab>ze$kK*V#
z&H9?6`=rmfFG!Z+zaaBx72RV?7NW7S$u!|vc;Br~<WAdG*N?7yAy`W~O88!&cf$nd
z>=mrm;4_|+#GP2MH-;Fr5z6)BnCJXrukZQ|=B362P);<W3@<W^0iloOeYgdL%OA3H
z0HRpvoP%_Nvyh?`PenfLr~BF4n2kEl;o*S`%hK3{gD@M`pJ2_~T3qv47Rj!ivmyBW
z`#0ZWZkc0!hnZZxm;mWGrG3xIKy+sizt5)BscrO)di|m|#wH!LaF@zet-N4Uej=n-
z^I0~CrE9dqb0!+i%M@5PAVys8h65Ruk?|bq0RcR}mSx2~Rn~;{6b+9H3CqK1YP;6#
zlDCK=0X3@8z_^+Fk!(|UFHLvV{xFXqWf;U_AL>Lz40M0<nP(mqrL!d<M*Dc1%CUZb
z1GIlOdcT%TXg4X?AKj$(?GclXbq0p+D~meG^|!A|(d5ru@>nzh*Ye{U%sM1Imja7%
zc71MFB@M_iX&K?|Y1Mzr5xylyg!Ib02Paea<7k*3k>oImtN5|a=Em`47gOwrzaDW4
z%DrD0#$LE3PsVa%qZx^n8*cMb;xh=^IJX4cn}wVcDDGAK{JX%tIhWZ`i#zSPE={I6
z&BUvYAv*{Zc*ljSl_K+uIYK`)aPhA5<t9lm<e&tsliVsPE*2(c%ZZZSoN%z0Ti!0-
zNN6qIOXu+-2Dq5V|A?mX`rb@vkYl5${o0M??0_nVcBblT_Scv-2~X~MgBLM|d+&>O
zTg%mnQ~lt(N2gXo8`n#vm-2X>rIFFquI6-oF8$!iIP~2X&;nx4y%(&(%(07n&Ofv~
zlCl12#@XV{Rfi3^;IyiM>Wm(EX=ByOW$nbsfMxtg%Hi8x?%;3nvO5tJo=Y&TbuIX`
zZr?%q1vr_e>J9^drYdT*NB++}Rao28fX{!yL0z&ORo4$n<1Cd&!R$$^*gL@rtWF=F
z3r(7~rkExzye<Acrs<E|tACGVA5y<aBmoh|CcUlFCg;(`5|rX#h)3f0B-M%BpL!cK
z+(k{GW|9U9J1l!x52x|--7mn1hYKCjhD6H5dDnh>8v<})#WqV4p5%{AAcISOl<x4h
zR(0AWqgYR=oOaLA12;dKZ<U!gt~vXA@x&~@&Mo46dc}s-H-x{KJt>K+?fb5;+9ueL
zuoLn8;eu{okx+wsK+q=MS)K0#?xt0tsiL?K{gVrdM_t{0zx9l^ccVRoh~FDBP6KPn
zo&%Pk65kB)fkFcHTiA}ol4j9=^$yC-_3{MP1mS0T!NJVYM|dE$_V)^paa7{L-#KS2
zXLnV!sPZVXjtE+nl0G9hX$D(Hx<vo`;rT``U6BVb(8p*W_<K*-#k9<)(uIleyKXx~
zZV$@)JddjK>hC~cMe%a1C}F|8V?)@cC>e1oRFA8oJNL_IcO#bAz_lgi@QdsESKp4&
znKuTH0vx-(d8+w6CQvQivRABZX@B|x*^-oZ>`L;Ij0`|x$<YKYdD+J`gq1RUFVr8?
zswmUW9`&()oAf1h9eL=hy_yr#WSal7TSoRr?4=tfOC{z~Zsv2m&#1FGx)P_~_tc|`
zUlmVlqA0~_#+g+9@@dV%o})IkcQuGJ@=(WALs)Kl)_JX$l+?8?cl$_@&O(#1<DS%L
zX=5@P9TH3LDDp*krT?c2#UNNz{mniN*|FXp554sa({%#i(Sg{+>v@7>2l84<5Eoy4
z1yD$Q<=SCs4N}&TN*gHF$GOno?QY<=Fm1HD{PtVtxcPA)N+y`lgbZ&*bcBFp;pe64
zPp$(9n3$gO2Cd&5@Ktb*6V7jn=V{;C3Zo>?>UdB$$C77ET=)eE7<}<-ePp2iYw9Y1
zNq&psc^!&{gd$o9{@p_+^y}ld-nw#oW+UH}Dtkf=_Sc_bHll9$5FivQo|D4k$mi_U
zEtd~)Lh`Up6ZU$xntIpmLE)V|L2YTb$QKmSluqpGsRAaZ{7uiJNSiQX(2IhSFEmx~
zDtxFSTorPoyr`OIHmq5T6cmG917^wJUexKIOx?OINvf5qg_^Aj4OfsG71U?=%{bdN
zN3s!d@{Q@!0ES{QN(L77)*|-a@L$T2B*eYXJd>2e*51YincZ8`BP>VAAwWfozM{!A
zI{ArqG6oTE{@y%9Fu8#Rt$w_L5b!+-C;R`s03A1mC<z3Izs5x3Cz31@mwNgq!t?1W
zxC61q%GY?#<u<CO948C|X&aM3)iLeffccDVE&~S(*dGGC0CZC>D4SoFXUu+gRtEWJ
zb2vdCUKF0j)+m!C0sHiyPgHZWY!&F0X{{K<1#b`ua?w9mn3zyW?I~Rxy{8rvHAm~l
zDacv>(}6Ga)&V|;1owAM!>sBbL(webBSUlbZpgslWjJtV*T22@(B1Xzz>Ml6!DcK=
z#sEL3NiTB?unALM+`W|d6id!ytldEbaD=iV2Jq{kX?z|vdsyxiy<E%KV$>qaQ%XC!
z)7svtwAdNj;<3~vTnbgooMdOhN4&dda8u>U?GE<E+NbO(0nZHFA4I0sn__Q@!ERc?
z!jzP*zNK57$iw#ubuHlskNm-lzeXlFILy36DSl!JqLuhu22QkbNFkrE?*C@Y+1LSW
zTSt@jH<!3ICn<%AWf?U%agD!yjta1O5PwMB&;8h<^t(SILe7=P;oQ#%*VRaO69SCp
zFQuPA0JhOASc++i<q3+c9yYdA#-tI1jD4f)H`$dtKF5(Bl>-{6siI3)-sF$2|0Dp?
zfOB3n>&lNkqyIHy<OJMLNkb8XaNLKc&k&C`LzqILaWqc+y7)QC)@CL>yyN}p$J){O
zvhF!0zRf+}La-5aRvg531fWtuT>}J!Reg7M+y0Qq?|ThHA!}cKzPR-k$5Xt`>+LF+
zk9h7e{pwTg#d@^53U}z`0VxPCJDy(#XYii{u$zwGx(cO1-E@RM2a%qMmEVoY72VCT
zZ(i~_5Jr>K9L;IJw?ucKs2H>$Jjq)nF=T*mAQRX&f)~|>kgZ*-<-8Kt2MGaN01zr?
z%3=LXz@cC#t&?s=S=34-ifSW_$Vq*eNV1EA)N%@}waxu52Y(ete0p6Ka_B-3k4gm|
z^fcZOxi2X|73P%p-`EvLk4;}x4_>9vRDHw$OcR~dz!gWI=&2r(zO?`7C&&bx<avJ$
zQ|Pi7Niw$+)W?A%hJJwRp{!4Zl7&@H835-DKG$EXBg@Y=p=#1yICZoGid>+w{3Ea(
zj0GtR9d@gtuHrlA2DvQ=7L6mOwCqoC<|?=p8eSWGKOF0!9IQrK4dStu9)KW6NLBCQ
znGA33Gx+`?Rp?fW2eQIbTBOP|k?&R3W||40nc#pN^2{992_oPrM`)UE1X-YCnWyH@
zqum6QZ2)9EgZdhT08Nnk-F-91qT59jw2lChRYg>JdIKlI_0+bJ+3VT@AlbtP4DWl4
z`k?*n4IV)710c-ah!&40$={2umidQE0L_Q4MA!05F4}m#k?Eho*P>GQ!r7%unKg+X
znT#R#oMgUt+;{Oc45sMiIC-<G8FogUp}tfWr5UZOa2@*=9Quzt>v}O=<@Buqp3PZG
z^}dXJxovd9+sgJrhfe3j;GYOUn#l>U*5&wP>^D<*Dnx0jc3O`J1BOsanwZQuMIhwG
zcf3?Jl&E(JDu}p9zop}W?sg<tSSEoknDafoIwv_GdAIkK$s%!}FQK3iK6lF4g|wXe
z43}u)_qowV=BAFK>%W@aTf+#-&K^o2cVZ`cF?;shuhGi$BNL3@eZ-n`xSshP<{)J;
z#6lYd*Q^><SEo9Uy4Iqp^q=2boBH~JK96jkPNf-Um0L+uwes5vfTBeSqP`AD4E=R?
z0FxW5e8N6c`helzmD6P4uzXu0-=E#hOtU<;wexdnZa|!Yicl+uAUox#<Sn1eFJ{Td
zY`g|)!D|N>wK$M#XY)cB&)dvWJc$JVtv{rXQuc?*iYoU4`X(dazJ7;zEBTgE9M5UR
z<86c#ErAe6nGSDor+$Iy_(3IyUw)&F0?i*84?iH``B$vkf=jql;?wMzROy)}^ghKb
z$f@58o$7Gip&@wKcChSwgn1;*!)pJiDrz&nGuiR_3O^}$Rex76D9qyugO1{Rx?4~I
zr>S!M^Z|e<lHU8Hhv<U}%xs9KcQv+j5;+5$o$E-aRPpoMnYE2BQ>8Ddz@XHIK1#15
zOv>N*FDDJH%H1!XeMNt_f{YZ%2{k0RF;hO&9N!J7&a*&s=irxTp9P{82#2E;(Dg}Q
z2;1%Op&$Y8YWe7FgtF;|FO@Wi{8}hHncCwWHde;qfmVnhDUYt1CGpvkn)pQ^Z=nLW
zM$G5C)P9J>nfQ*nsqLSldvgQ)g3q+N$O!N$(}deDz;+L=b#%pr$A4Gz1Rh<92Npmm
zGYkW6@g#>y*7C=4M&xlrx4(@ctHvK9Qei2K#ZSJ-eg{p5*_WUY^5)I30dQwm>IaSf
zHj9O^aV6$^lwk@p|Bt=5j*7Ak_l1X{8>Aa#0BMkJrH52NKtVdBK{{s;>6B2EQa}*t
zZWt9oT0%NTKpKW_X3m4}e$UzKeEa<Ot#!U{t-ThD7@j+xtFP;h-}ko8SFM(3VrVzF
z++xKSGUz=2D(=K)BH*Mg;WX?OzM{DaN%v7ikInQokJVxthaEF!zFa}l+AoZA57e|{
zpIHYem+K~Qz*XxvRy;@H^3`c9CxlccH$`YbW|5Y8WU)Z@D#wob`a{An$K3;!=$KM!
zRVspH3dj{e09auz7zFYMb#~XyE#ePgJ80yRA5WpM!d_;0#m)udN=pq=`HkG8BdtiJ
z#s8@q?UDaY$|tMs)$BQ$=qE37&UXt|KK=Xy%qPr_EUl><1yW*RD`8P&2jV`h51a~X
z51(EIO=JFG1p4=%&xU7wsUG=L#B}E$k#k{0#h4sKJ`absFGaB?b$1Or3~Y$T=-m3k
zE)WmX^vtOW5((+}m<-(oOM>J?r(=uLIe?}##htU5>6;BKLIMS$Ox%PxuQ~Eu9cI<K
zOBQL)r{78kdB4;66Z)0Ysq^W|)s9`w-l<i?aU$|m<%_q5Xmlt&P7K;rzjD6{I6f{6
z_$-ivd<QFY$_SiXt|4A=|3g&RrvfDqJ@Wmr4XuE54qdU-h(y=#Xz7-yfH`*Zh}maO
z&c*fBIjh~0?-+qU<h5MBJ$y1xiPuz6H>$3!!X#J`9~C9U$SCXdjNlf&Irh`UL?21X
z(UEZ-4O=P};Exu(mbPz~c(S3K9sLtzNkZ9&(vxZhhsc>{PO@L&tA6j^$v*Sj5pVre
zr@vP_xEzB0l=SzAs$6k(!1dbsCUJ4K9c<~_IzsEppDSr+s2i|ZM))<DcHmTD;z{cL
zy<Ksb9DES@V)c08)%1Xd{i$B}(SE`5@2!?GP?9s2MiyzqE!3~&_37!wGz+f-%?L@I
zArJo51XFBz0(TRwH6R;(D1SYTuIzF3_xG{BzX5E&_%giH<+FLMU`k;PdaHA>GQ5K7
zMcmW$_oJ(sBzMa!=u8Yy$BYcL+{9lq<F+7nXgSk0K^Vl7!MI{!Qr#a%Btz#ssOGKy
zb1@e8pWk!?8?)AM%?DnL_HqZgj!>A<i;qrQg@+GC;k7U_Zn%Qh3#<E=Ljrt8r+%ra
znD8VvJFNJElasB6Ps6%(<*H!88X^l=KERO{Q((YfKa;m%)o1=TgS+_n*a|X(1uicn
z_lB*rFx~EYvbbYhyB^G6?($gT`&sE>G7x;8a{Mx)KyW8BlDCbCKsdO1bh5eFFtOra
zRH%B(g==WbsR94}1qaHK*7SS&riLG)m@O8cAbl&oW?zJkz@f)SkA^I$d`2^q4LgNf
zsz|T<5#(RW7gd5vUH}BxjC($fu{oz#aHKd<<u8iwY>r+#RIrqwlq7cj#*D@naqOej
zxogccG3m(CgkI4E%G|HW>}xO;*Z~byr-0gj;-Lnuzx`k*4vP0phY_AF-(JWo#RNV}
zs&pxA$i|FBDLbX_`dc$yl9ingb0Z%OY;=-j7Xmez{IC+WOG|-y`@OZVA1rcbBW3Q5
z=Ev*MSbdV4%drr(lY(q($?rl6IG9<ZMK?MWOlj?>AfCcivte2$03k0#kDmb%gYq~b
z5$H(ITWgIWYc%j;Nfi0yBG+GT^+@qUjEfo*=H<|lh;4;}=)@~}VweXpYjk38zHW%@
z4|k#TfdldgEGX%3yt_%^Nk32;RciB!u7)sE(hDH&_faXop$ms6C&Oa|#o{39gTg|%
zOt(s^3w$zsQsnW6pXf$43p(GW=uUq~m>(WWCZU`>f4E+Q)t`=jT~HHAc!2@ZB?`*T
zK_F^<`HAX=O4tZd-khyc#2-X8ZbxS$Nh`sw{az(j!H=P^E%yRBYI<eAC}ul3<Tu@b
z_4laJ)yIImC>7%Ez^a>qzYO=-1%Iu-S2togTFpAxO+G>V1h~uR<tL}jpMOW714dv-
zc4{m7a@R`U^LDcnC<%{Not16i+CXTar9bj%a~|MjUVb8W9!Jx~&q9MR?l(EZ#3#c`
zVc;k01JL^r9~;@j<`e-2S1ci2e_1YjK9R_TD4X`X1f?}odZ#q>$I8U7AGI)*vX+=I
zzs99@)}9nA2(L%H)6y>_*DF85QSE%6yN#i+V)e^&soXQ?gYt`j&G3DV0)#SD^CCn+
z^hq+bTg}N`ARUW8PM={4cP@oTffh<rifiKA7AC{Ni@!4kyxMuplDb`C3a2Knh|h7G
zp_s)N=iRv<E|j9f+%Q`bpr7CYkRD#QwPm-Y75`g@Dh+<gR#$aUvQZEqogAu(z-}yn
zgX3(#`Y9mb?||Ym38;f01TVhtM3rt2g;*=md8w7?1VpzNKiixAfE7-_h}>zBn!H(Y
z1%kVemh)L#yr7>%0=Pxk@o<#QOh{|>-D7T@q7?{|7C9VV8a|y4qpYGz>&|?#x&<h~
z`rBeY1+0e@f7{0ap{nFRoswMlp<A`|-zS970_SzP_8yQ`0!U8eR|<QPw}wi2^sp51
zvmb=J(0#+r>~Q>Hv1y3yDY8HOoBg;XTAp26ISEIjeIy~2M|q}Xc-4Yz>^q4Qe(tBA
z<_BM}hiw?_qWR@$9)fBxf2<irI1bK4esL8)5W=(3E2an~i&DfduuzJ=pNXUW!v*o0
zXpoD|4&%Z!jX4mdOA{oLxfgm8#-KOXW1QTPf@fvA;zd_gsq9?>nZYnYVLw}tpZtKw
zvojFGhatmn!V|eNwPCBi+sRz=sicXUk@&GNGE~L34`27<Eevyq(e#E<0zdFs=ikpn
zyOg3?GIEkWQQ2YozbR2*F46scdc`&&iDzqj8#ljQ$pLSfmXV|P9XMq={2ONMCrh*3
z83Q9@E8X@{&0)W^RAl7_p#COq9&WpT<xT$-7~&<gODW>`rd8AHv$Ygd`=`!5`=vY;
zt{0`Z-+vgTy2EK3{wT1+QSf*Rag$wLVE^TZ>NRkx0uTtC{|Hx?Y6RMU!;?Y8T9TZA
z1}LX90t>Z(2B3kEA7}*z0`WaAWI!3#&yAxi4}Vk6_m(z6C&zq`Zvzh>;RAaOfDi_g
z*M3R%rm&1k0bNe0^TOKU2gwgmY8vEz4f}0v>hEfz#gBCSpJKOKeET);(j>6;qNel{
z|NC8{ObQ(A_qTBFQ@`ZkWPpkdP(;Q0V$L3o{qnuY-q;<R|MmMaFgNVNCvvxO!CMcr
z5VX7}iq=j|e{19iZdp%8#3E!6q+K?UMhfBw0vxnOQ`A!onVCB<IZri0rwj;m$NK6G
z#OaJ4!v*alF~kNTPGi1HE#h1W?ipAW&!ICA^@wu-BJ8#hfZ*N<K)65|6}w^3RQe7)
z5Rk%+)#4)<yF3w(5H&ToxS*-BeP_bZ(Z4f@S`d*0Q8OI<+afA?-!fI)C(4P^3YrS}
z`xO;M3i^*<A*Ad~pmBHowJ-}pc@tx0trMT$$4dqV4;2qNGRdt8_uC(cmYuVc-tJs`
zaZWK1ZKa5Fc%RV~og5vaSO&@!=I7uFD%zXrj0BJ+tq1X)*srNl9Fj3YGJ^K9%%*xP
z-^vjy&(jRw3Bqwz6$L;I0#Y&`-;xaQ99YYEsy6rX{_-yg(+J65snc}M+lf^UC6Xr_
z5Coo?e$_M$4c7=y?9<hA^5fvcGC)=$F8**Ai??D8-#XtDsd5Ke59X>Czo!l$$TW0+
z4ym86R*m#0n#0k#8RVS$*JZZ1mY(+8FjWaQbjM0vx;>F6&nx_XxXJ6R?9n!QQ_Yoc
zX^7Q7+!_Y0ETAjWB#~)CX)@E=?U(7URJYkAdF42`%+h{D=k)1m_Ht7!lH;f6P(a<J
zlFxFPCbUp-gRvn~RW}b4-)4)dKj^g4zRfnl4>X#M`w?yJhVu$<UI@Su*|6s+w8OV~
z`a>0|X|-5GWl&Qe)6}RkO~^Z4q^d~bvIgMM|HGF5yWhnFia>05?l`YOr&t&=*SdFy
z++Y)ICaL|$Y2mYQ!yah|I%G7$2I-Du3op?PQ7?Tmi2qDLxClFsy~0U!&(`WvG>L6I
zL=$cs-XHD=$A^z~OBY)na9|~Ygt$jrK+<$mwh6<}R%h@e*kp4&Qq?CnP?z${7|$Mz
zgRRn$zM?o&&%5@D`vds4okoiNpbuld9M=!<*6S}}>&YQ|XeHo&Q^jgLPvVqYmUDTy
zLs-V(ez;qY^l8R)pqco7V6nqmwqzfA(*BYSPY)|BrM-hGbE0ca+(rHzR5~2*8{_ii
zZ9B>&R(Onhs(ffa;vG4mr-k2u8-gyp7XA@Igyf1`zT{#{b%kkBKkHDvN}_X|RWw=?
zd)M4>e=u5aEvs<E_u{hS$C*;we&_TQ`E?rkV0X1cr3=O3mnq?l;k)<NeV%G6rDFx{
zpxG`s+1fDV0KjK*Tk;uxkc#!C@$?DPR<;<aG_|R!cR0p=dcOb{^R|nwpR|ik|NDn&
znPTxXnet~`Z14Bmno<co&nM8jF~TFkF~Z;C?hi#rl)y;0ZdqQDKNyS;>x<TUH^dQi
z-$;jNG!LBhr$NfwEh3vdXFv*SyLEILn!68cT84>Lvt97By^FOSw3vLq8c}V@muC<E
zxzUmmb_e@t^!kS60T|0HZU8L_-<`dqZmT)vtf#NB>^jaFR4LC$cqInsfosz%oOSZd
zns}TId&D4RK86oFtJR)^64$=dPIn|vOrNT$hmDXrD}q8eW-UF=rktibd}}AK)5#Zy
z8wNjnUuoBbVp3)&2I@(lEBf4Bgd~=aEuPYJ!~Rdx>jkFA`@g2gmVgPhQ=TS#u(6Ae
zwoMNuu35WU>;+D@Iwsd<on?Da3buK$EN6BO=3-187o}p1zp}hL8*mWMBL5j+VN_AN
zAX*I4p!EJtlgw5Z0t(;@wkRFyAgx{dIPI<U=Nl_R`Sau(f6Ehj;l}c}E=(&S>vkPb
zMzi8h6n_Gy8{=T3bY!9^PW4wV2mh7JG(av5@YX*9a+&uZa)~99_UybhEWvg`*H#eI
zW7p%ReFUa2KZX%{Zrwpu2m37$mY$dSxj@*gw%eLs-#HU^esGBOS>?KdCrSyd1GdP^
z@ekUa0vLN7Bg48i9j&<){CxIC@`j@IR<jBx>b5RfZ)WsUXFD3*7wxcQT01H6>D+4j
zP0$;FsYw9Ugt61TFgH@UfWcpevEBHgal;8tba(^a0*O4BS#jvp9RG|BnHQsG>5%Yz
z4r0TFx>TRGbaKm<8#)rtpFgjXdkD(Zt9#u^lZkD9p9FL1wmWry2tS77y&0Ai<>7UC
z{>+OjVFXWY-lZTLE8Cwr21H|Q^VU#*gT!C?8)RM74&)$ZT!x9Dqy5%Y3IG8VW@AkJ
zM*vl7drk}#H6!}&-m4;cunWka2_Szr4DP}M{TVE1c!V}bo_n=fg0EsWol&+W0^oz~
zJC`PQ!-hSta@7Y5LeE9sHZ|14&!>mi9EU4;R)sE^_e!?U0PT!}0_dp60GR3_eg$?L
zF>pbMZT_P`iUVuQd}J%3C(fQ-L;GL;K07(H=p%OeIq>a_yc-i2Sa*v0Ffv6R@rwL-
z$<#h+M!nMI!QqE#+hAU{!ut~B{U9p_P+vPH8V&3V$3y@q)tqVY_E_V34f^XmXWwt_
zdbgkKOZK`kxt%SpxL&Fiv`wx~qmwJScXq~~DI(SbMI$%AN@KKkLY&e*3B?$_ZFL2@
zUmADC@DtDSP(_B*bb;?*X9J+3bW|Iy<Wu!ko&R&11Hd!^wlAD{Tn)R3pO`cpyC^2m
z{3D9d4#_*;LsIZE{xk0_rzeunk=Fq6wL`Zxm-05J9-bs0+bg{xxIziIvbmr;KX(1B
z3gRn$(do4f@$cF^z)~NM2T;B2$JxAU?He$b&u8&O+1@{)XmL&VZtOe37@o8&0tS3R
z<7!?QZ$Y*GPs^wOIlVz(dI@X5^Z*WixG{%s$P@pX!++=}0npE$fjyyE$nHTOXubm=
zVc%a8cIs1`98P$?#SOuk@=CvMv>VQ9+*>fCgjqULgjXJz0wI#+$;InA8562<<S|;P
zu{YPlgF~=%tT)6IIuPtH{L5M^5H^@=nLaBSbN(O@7Sl|z6(}9NAQZgY7;?t9M}y4j
zqUiCCS%fJJn$NPY0dn^ygKqS_C>l_UbvwY#Tz<eDd>E0Q*!Besmw0elwWYIgKj_Qd
zd#HJg3hYnBDF(Qn{5ZTm*0@$oeHwPJ0dodW^4Cl&m*`yinY5VP+R*lKZ?pG8-p*{t
z(B*Z?nN6nT=C3;UwsVl}+IJ__nd0&3MhVVnA)tSwAkbCAkRsfz_B!u}#i^kpbFQ0Q
z^DYDDKRR<`vkSIXPTTi8`_2|a{&(%}18C?|BZPBNK=RY&6!TN*Uk0x{$$!*x7O6mb
zX4~Llyk9u5`NBfMmK2kvaz$mSOhWp3N>OgDR5wJgwmXen#Lr@&;2nU+{fGKCebIP@
zh2p1kkf(72CiCkFFXGX3hgSS^z#I;Cg#;ufeRc+1v<#c<TaVQ43BX0)UWWau`u=~m
zg@^0Ockc-Ul*6q22jyPTQF?onL!uT#vIz#_V_=^Px}_WENkJ6e@Q<n@0Ze-gHoKL<
zwcc{}|F~vygPr3&24s9!6@ps2`mbvigPYab3gM$E&!lWjpS{b;n<PIMQJ8I@XtGMs
z@TzEYsBoh={J?b1?c|e#?hw9<Cil4VkUUX=t*G#Ol}&Fk)`f_iXtx7A{ZCFgNa;VE
zGWj3}?fd44XCuMzXS!zd<{mlEK%gnQGcHjjWCUVueMrf_6!39>zK3;bt!c;5j1S+A
zYEN`{aA<vZj||DUG{cIEj=xZ8eSdmrZMgK#<n&B4=Wk>q$kvka|A=hD6zdrOs%;<O
zLB{0-uwUv;Mjoj!yr0~~d}uuUa)|*@y8iFRe*Zrny~IC++7xf#jvNP&D7hqHQX!;l
zKtxl_ot^`NhPvGnY2H@VOf9ScFgjAjLja`DymzL4V90SKDclh*2hZ}J&BI2Mm=l-r
zaXbZj@cy?4aVzeOn;v$ewmG1wQPDsWRruTX>di&EA?SYv?AWg4Nbh^#ZTvZ+Y`kne
zful4qe{Ejl@KB&qeLQF<wY>+3`jJ7vF&vsIGrxfe1a4Q#*GqUBL#S*GcC9TPRx-#E
zTapPCsg+ttcasbLljTMIY>j1yX`ZwZoRLs&FQ&O&a_{k34Q^+X+t2dj!lmphJMRt9
zomnH}RXMD=D7W8$i$O5*xwkBsUfB&9iCKT9y+9;M{?8OJY%GcRH3WK-I!2}3yaR_3
zFGv7n@nov;ego<_a;Ja>Bvjn+2t)8$fuNz!EsscEyKZ;w5fMiodl1A(2k@uHAaBN2
z0cff?-f|}d`mN_C)1>%J4buBuTR9CzR75UM3^n9ET5E^hJ0kkgiWiVFgS4!6JF)Pn
z^rr+!%KLY4APZ%@oz|xZFU<9ZGA%6s{<`&O`Y>;cxiw+to)W`Ue5RCVr)<ygP_93t
zK9~z?K>F`!q*zQK2#ui7`C+mB@FlpDWx2_>HH8*<p~!!k5`}nKL&Xlo&G-$s<PO41
zI0-Q1ElGnwb8a;2CdvLN)W*L5_TRk#Rao~7DjyTg=?O3!4a;_gL}vL3ssS?{`<MCS
zxRM}EXZ3cqK-u482;;8XDuHR?hv8y(V4v~<5`6tP*~b1V0fI=_JU+Ry&ZpQDv$Dol
zLYX6_ZRn9^G4GXF0_3VF_nufi#8<LRh5XIo)q(5z5z^p$2As5D9;)&KOP;9Tz6M0u
zAZ8Vu3w>;0fdto;`0D+)JYi_+-^p;^P~ra*`Qd+0<G(5JUs1roao`7QD#R(^KbbEb
zumGN-rJpi95k3rt0VM@;>@{#DWK`x|J)9hDf;9|E##X{X<7JBU$;e^#GX|6?C2c|B
z!dS&vX35K;Pb3yxvI&^UuTp<ZYJ`i`p0a3{cJ@>cV4*3SSOaJSz)he4L6MU0r|E^(
z4p&lWCJfh&k|$ySOT(ya2I8$U>D`bP)$Vzgya+bj%x)yF@<}*!XD1?1>#1Bl<Vz;)
z_R_8V(LT)Iy{}l#I4m&)0iQivOE9)_JwBCw9UA+h?-?qPe3}0fbX4Rn(0wB2ukQfT
z)=miNhf@m6{J}8Z#<8LCun)r66}Yst4X|3z1$9K+S5YaDKV8GqZ$DnI4U77{XY{~A
zK5!=Hwx<R?Y40XP_(s~g-wpz;G+sab-DV6+wtXD^>I=p_|BZ%MmI4-}#Or1Vz^Va7
zyG0=Ie^bceBe8dwI<|-X*>Z)>P%uFU)S<US7aCUaRd}_*(qO6)w~D<H*7McTBGs)k
zZj(9Ym^G2?aTZ}4r}kKT0!ht31FMu21>;sdZBW2cRC9w5(=nS}D<u6HYxbt4>YR+T
zB}N6hMxc11|Gp{!+xYiYk25D9<`lbO?W+adt%^Cb>A*gd{v|w9AN%3)>8-=Ur?^*w
z1^CqrU36<%wUdO2gA9!cnU<ALUq86Mwb)$cVd&hs(%84tdbcr@SpCAv9kU-98eg-6
zksYctaqa){-qH{2{V$g`xN}Rq$~Cva<)y52W96^KWmEV5oWpS0OIX${%=l(r?Lz=9
z*vJ9mAM*g@fyo1?AADq3BkXSpbcVum)k8$tfE!t9_Hb<$>ZQPMC{>F1PI_4@Ay10q
zPX=;@huyNw*P4Rj;<pC4qhcPlz3Gk$(glU31&3N%V-CJ<6?-c!>JE_OADg8gdSeta
zF9R7zY{)&mv7KDI#t{16l|0y$&SSxi`M+Fr;NPBC!k$ib-fBwUT>Rb9wKyLTmTWuC
zeR_!6-0mcqQo3v&QI3Qh%L4Q4d<4kjrsYksqZBaa<2@F`Q{{bw@|&SDXgSe0Pt=s_
zK2HWy)6K}V?OAveT`taVs?mi=pXQp-ow()A`V;@M08zy%>vIkI+s*Px4(IsG?{VaR
z3?vjiorfmC^txilS2Y@gL3KfiE(Te<PdAypo!v2kq(zQp01+2*g9l1#F?Dm77E@i9
zS_Z;mfFPRz=u?Ggdq8IzZz{&2Wu~AexO3?)SprF)C40v>74lbPs7H~pG6}z{U4W^V
z*Bd5!nVF&-TzE5L1To!w+KQX`Xl+WN<SB)NJqJtYXYU0LTV~5Zd7Uj{(EW2<xN)Ud
z!B^3&i9i(W$4;k1g1Go$u<Lg_Wb9mU-gwbzMqv#DpiSjKG-iOg5x2h+QZyC-QJdWx
z?%Ofz`e7Y-{%BUobgJ=jLulu(LTlRk(k9{0X=E`Rj+Q?ZPRhJ>3Kn{PSRKE6m^7dF
zivMe61!(WfR{+wB4vU<+VNus@INTxqDd^cT;~LG0hv{0oQpzlBmfq>v>giM&A>#Bo
zsx`BljyAQ>!tq}+T=xU=<&gpC-Gylfs49O5eYIB{zqQ>d)`(Gw3`PBD&}Ujc1a4LQ
zeqf1gmB_t%%>o_ZX>8h?E~ZJ6E2_d6AraeGrJqQ{YQ5L~5bI&o-M*BVuFKYZccJ&d
zvSJ)L%wMn_#2fNBW|R`%X41!cKeA|BTZFAN+`f2MT}T!Jy&eE&Acal_phLyi|7ssb
z6A&J3VT@XP*$_Ge3;9C!vWE|3uD6pfczlIlZ#f5>eD-w~g0=8iQ^kJw3-35txB!_<
z_B|5)(rrxT*Gps8WAj`&RI6Z7Xd5nko@h<#(xTkL){6i+C%YCV-iT?z8@oG$s(w6c
zFdJ{pQBRBu9c35!YwPL(VT}s^Yk(^4uO>gmsl10Iiql9FNIL=*E}S-=(+fGoq-<tU
zkO7iO)y3m`Yt33txWx0>`0yYaQR?a?)z4o<*kU$&g}Dz?xkqQU0n_J5)HCQ%c-01n
zIf)1sT(dG|P!JQ&6S=0kwWibe1qG)8ccp#n`%47vn{|vEo69KuJrW8@tP4K5Dt&l@
z(iE685Y01<Y^QQCm<__95C>mHJMqZ+!xrua3*i;5ZMB`gm>W*3)o9P9R!j3%5_UqK
zg^ABdD5|VH{ODOP>goM^E+qc)YP<Q@qHTfGHI}E5`)qW$uvie5Myhz~KFswW`+j>F
zGj+oPDBq*vC{S1|z3d~8fT<Z(WLfs9KvkGaug8y>@djg*!CI!}>Ip`Aw!!dsTlrS@
z>DxK9R{sfY<lzK=b{a_P&q?0Q6&KbYA+Y#@OWU)A(6kt^iaqAb^a)`2OVdC~y*Z6w
z>B2AGTqTwaKyj&4=EhujbOY(22Y_dudSPBn1WE=MPs)Vd_7tqQT)C9YoG3vkt+Ec1
z<g`kh21-4A8sK2?dN(9!A#?Ur$Nf^!D7r|O{YkvZ>PcR%oUoEVve}64*IO+TrU95m
znAclf;z?=aFP{d$h=2-8oBiCrj*KqPk(cW<d-Zc2AG^#GXv(ixt+zuI+B_GcYkkIp
zrNKI*?{9>{3|OryMBn(@#+%?bw1Nt-2-8~;o6Inrddr6Xn$z;rJ7k2}hSH5qcZN;L
z__ECig}%FyKafK$68a`RR(C_{$6AnhKVPGXc&O!8>i6eiwuvX^E4tb^SGX1gD!1fR
zo$;`@>!2MDh5c>=`W^d$@+u1vq}A*R8oy*qDGamwue6ni2;|XcG5pE<s#+Xve_0vn
z-D3y>Njpj80fi#kKW*e?t8F~=&04_K?7k9fp~(a<1Fwy4Yy-c;!^=~{kZi4MjzZw2
zHTowdj1MhnL695Z@&C<oGs;o{dCT~+9!{8kUEFD^SPad4r)Jq^M_gv{UNOQYR?Qau
zWUXy1)`Bz-53MO6e<;33oL^Nd%4rg-AAU-CIDXCJ2|Ecq#kS<3g$7}dq`kLHJY~!S
zH()P-&F`;lDd+#J;t{15F88(o;+&w>GBgF$AmTrq@J%`k=pv91{8!P}e}DM@Dg};H
z3b%2d>9~Mu`}JNzNSmnx<bz+g0OdgFA2D?4EMORoI>Hk$lAxoCQy}La>(`5c#9oGT
zhAX|vp&*sNiBQOvJ6pa34ccvyOP!>kb|+FR4m?1yxz@0I-sBLFir+lMy4)!{c=M3s
z`y~vJyS{I6u-$u8Xi*EDCj5y2K<*4sqUh3_d~`e*j}NWSCx_*%X^-ZG8l0A44e-fP
zGCJQF3&rKmJ^*L6aF=QQ%me!#EP#3HJy7l#O$lW1^qr3V`QWCfQwGc<NnlXl#d|$4
zCQu6}78%yb$7-iH4&ySG&Hxtm6m@2{!YV~%j|6bD+v5EiXY)J?*f%*4Re0;I00(pE
z>%uSgSHfy79G*B;)J<4t#1L&QZQUzYDeMWJrPUWTqqm`(M9Y{5+QpP!Ff-(1<in+~
zw&%$m6iwKvA3t)5re!=V0=r;|)vigz-W^XvP$8Be>K`>XZ6!fIJ+t8+3CmaEEHCo?
zpp!Uccqcr3SCf6L-!c0d9PRW?6sbJ#(CK%ofJ$P&WVw#Dq9@B992Xn<+4>6g*#{3O
z$7wF9Gu;*vd~u|h&v8YDl-11L(TcMo@}#b^ZE$)-1OtWTsD6Y)oc8E3(OR;JpSBmp
zgWKq*qBdK|lEMC9{^va9@FPj&Fyd)S*|wGzr*76NE7L<ku|$VJ#vD3Ua0L`uJyc|1
zz%vE<$r7D-7y{F0RFi#6UsT6dCG&8KzO?5D9>Oeq@_rDDCo4os>Ik~;9{~Q<2a9<K
zlCw|N7<)$0BpUu_@M$!J6tWC6DNKV;f;>PILfbkqap{i{Lcs2D{ghn7-Ut?58ds9G
z*Yt@dqF7T@Q!<hKu}n_3MG#WjWejlo6AN`)-Ajjkw^!=JNp1*se#mqFhh*VrET~L{
z+=!!LgooV7eGaBCo#a*gReHo8qHmj`Na6FN>ksysn81Z-0GC8OagocO=U&8}dBOmt
zkwc6c!dP6<+(*`v-IJ)Wr~D)>KOeZ6vtN{K7QXI#9!~8C&0=nOcM<xRcydDXRA<7-
z_Jp#lN=LW+kt>yyjzUh}7m&p6bN>fs=Xp|*nXF6LPU896sd&-*fGnyTp(}B>^^QJQ
z`m@Q?1Q3qgdoZWOGDRFfUYJ!c=H;Zbc1Vk)sWU#e9413Rd71ONkD^j9;DY`0`<4jH
z7^We0KAA~;y}ow8Cp0vL_!Kl4&=#c%34bnU5N+A*)!ks`Om*J{bXy4`nByr{FRrf=
zFWMoQKLiiaCyT-t^+%S5w;P@Xf=?Y#X`tHJs29GmTA6*e+#(W*wy+e$n5}sm<d}FF
zrtwh@jL|VFqEpZ&d9;|>R6VqYQC;F@)|3K=jZ}O%b?W-H3@_i>uAq~01Xn_>jlm{$
zoTRCRqYRj6Wm$3n104lp{S|keekU4&xNB3msP*{|+$|iRA#k{rcF$N7ND&jzILn$(
z3%3IDDXj$hYwHAtScfSR8Mo|`oeHuWv8z;wPO+uHhg5gm2DzZN-d=AP>NrIKTmimK
zlRD=+ABDMBf9uEOM#|YM3G?v%WWx~x?_$v!6O(OlBevk!^fHwz&yoEp=0=Cm{f?eJ
zWwl;nR{Wvce21qHMtt>4#pdW|u5<8<3~@Dmup`&Qvy{H0yU5MCEteWMgJOCN(`!8(
zbQ=Bj>qhX|qiyZn7d~WO5iR&tOe12IHoSX0HiB7)?1AocC2=JO4du04EdozlZ*lHz
z&g)n@bHc}|5-^*R%dkZ#4}~FSTd#fu!+>;OQuM>}<QQ3o<;i+fN$$x%q%Bz!LPnRT
z>(&IHwTG)POoQs<Ow*9z+xzT$<Ih~q?NJgr&vc|gQ=SRQ7KC$Tb2>WYbw8ZQemUaF
zqUFH%Hu*PZy5xH@WD@-$FODM+p{nE&hwRP1E6u$T0N!F)+A{*_aOW@2@9v$Z2t}{Z
zHK&o;+x@|Sk=qa=>0yw6@k97De=s$1shN1-t1Xa+b^#@_gnKvxLK<Db@c0~ty+@<D
zTjH)oOOdf@axaL0Ql*i@<`=ciQ}`pqEXTDhr2mQT1IDp^8h%V3zK~}AD=g$Y@u1!v
zjk)YK+f@$ULTDK&u$3cdQb_erSXkVgTudKE39J;o5{T5(1XTA&Wrkcs4edRxbef5L
zzgI{)DnoE6TWU0;u2F;gVrlFUx!Ng6X<&DmDb+|rY(U?ZG{OxY@oAAlgmOwT*l6@n
zz)Qa~vuVh_xu+JvGFYf#u;mI$xlosl{nV7FPO@Bjbk+Ha;4|;u@&y2w-g{fK^Zt=!
z0P(NBXedI1<0)b9$LWx;?+Zlnz`Wl5j)S^l)EWr(_&-@9pLM%_P6QEmpIf9BhNlkU
zEf`G&@b6J-8Yg-!*{goUc&lPxjBi(qjS%b+Asq!j-dWn`h#XADTI82>aQwCmd#f(o
zECqI1@$0;uXzPlT0x9A7Xc%+%TZ0kmBO7)uXefFmX=XVsnEjgEroC;4cN?;cDyAx+
zeshavN#F5uf&O&2H;t5gmtu1<-V)0^S2eJh>#;4u?eV!ila-~jhVg_LXYc@vJ*I{_
z4C~6e?jb?y%r`tx39K!ZAe$<t8IR{y`B_F>WITIwNp+)^f#xs=5sk+OS#i(1o_7m}
zRSaBz!QTBk>eIL4@U&o5`ACK>$;*5oFS@e#Fy977Il_c=LOvR&jhQH#A=p#3odl}z
zU{s->*Zd#EMrp|vPkbmHRR&r+mTBd2SK@{qLQ8W^<+%s3(@ZKoYo7mzv(4%?Z%Fh*
zMl5(~8^dcz$=bHI5!|Hc*Hg8*H<%E+KQOfeW>29)Gm6vW@4-9YyXur?4cntF^ad(1
z>YBsm^FH^SjJi=zr>^Ap3LwDofc#O4d*XNl9I^)6I&RZt4_zooIcGnL+B_fG)Jqu$
z4qg>@r0n9fi3eBnK770ndB_eoMeJRyE{H)n6tX^D!Nh*aT=D2zZ}527y~*<b-F@ZR
zmM1a8q!f;CaK<g_<JB%u4#TBQ7VSiycYTeOTa~O-69<Rz$DXD@Nb7xq)lcd=Nm9>`
zlURA79M_^=-<nfRdmI?6=}`^{(Gl&+{StT$u*ZN-axtT8JjAw`n-Qx+yKKjBi~zKq
zomkc?8vEe@D5e{22;5d*D{c;jXP%;?e&yxOVQ);v5FJLu>Uq#))W<eJ7_qn=l`8Se
zUL28V8c*mx8sA}2*=Y%hU*C(l*lvgAeD$aiM`@N_OU2(Ylc*t(?DEY`|H_5P+Xive
z=aGU8hP5ZLllm--XpxVVb^Jkl8k!Xq{;Qi@yoHP0;6tWfB9htw=Usdmw{~Lbi>3O<
zbuZ{g1I!Fj%=a&cwnu*Kml~XLh@M)9&4)7}Tnx|DZ)HfN7Jf+A!&^`}LRJP4N!K=?
zLXGbSDemFre+Zm(?TL@+UR8m|X9aw73r{4T6>^c5@Lar;2-?Edpx^|%<5b|Yo)(z`
zj)7Ur=(!1B!yC%eR=TcsU(QtOCQ)^kj;bUJ`%ri~J)ubK=AH1jU2SMF_G?&yIGCBB
z%xvl;&0V)RD?^u$m>4@noplhxh&=)6lg{Yzv66D`S1^oGPjDMGEkz8G8!%0S*pE93
zKH+tj*RGkx#2q-s${coc0PLb2xkRo*%QBm|hs(}In0QcY{K0wO%__1HpZQxK{!Wnq
zU%mXazNpaX!ifYS5A=%fr;gW>75IH_qCQiE2<lM1DT5!$=bOw2saDp%{4@aL&LVnN
zuW(kIHEV;yoD<5=894*S!UGa?^+%UO9jk*nKr(Rzc1k5RojfHi1T~N<%mPJIWz5h3
z#w^m#>k#!u0Kekp(dTUisXm3*R4lA~V<fulMFTmc8+06jA8eEo^$0F$7p|fpqsH%C
z+rx|L?Pa@MXOOlr-_s8paP>58Sw7Nm4yTXDsYfqs@~O3mj&M@u<0Ay}qwm{0=W4S*
zzKb4)8X)-*JKm;Me#-Aa<~_M)A+LuAxT3`U)Fys<a{Qtsqq!_D5N3ry8J}&mAPC3(
zESmApytYtp1vnUUFv<yK5$mrSa7AezRNwB%&56N#Nq0*tKJnLd`hCY+`yO@2Bn}i?
zAQttq$h~T6i(F$mj&xvc$?J}Y@aYAE&1M$f<ZmY3ROK3i)LHgs=2R50f9@*A`Ml6Z
z*k-U}1N8VxDQc#r{d+=dn$?>979>YEp~c){svUPvwqMHNath#OgF@|E-xZ8FT%*R#
ze`F%=+-~}nc|c#3{DtzLJ}qv&Gd|S3EXVwZ(7dsGTKV8V#YRlB>>ZUBIZ4|Lyii8K
z`pgovQJ<ZhF?~sEg)bVmN-uDCB|26+!V<TOj^2M=Wugjb`!kNPB@AIZySr!Ryle#s
zlr;vWWjpi~nmX>c0kPS|gs{vAs-iGt3F-Q|9DF?(;Rohd{{Bdu{IP{Fpe9VmU47;;
zti5!*0vAmcP*(+9*Ux9QCHPTtj}5x{c*7^(k)z9|B`qAkcCtS_e&*@a8L!4*u(0D*
z*e!lzaGBL;9FM6Kt|E?;2|sP>optjEQ*A<=bg*;?X<c2-5BD(ZRwg*@WGBOKx~@C$
z=;yNyxF`Klm&pz$R(`w3TKPQ4Cax;!SYI48sWe+tR`uG~d5Qpk#)3Zri`uE<7IIXE
zheiU}9a)dY{WVjSpX2UPI?K`N;3Ngeyy;BM<d#Ae;q0!Pzmw&RmB<{rJ;6FKo3xJe
zY+9X>2NPSkP|q*=>kD)KBTV4u__Hg<)X1HyD(WJv;Et@y#b8ipGpF^V3+h^0`_@Jr
zy%#=D<dnR9YrgiAMh*rSbFL8r?@-aKT6q<hfq*5~5LghWe(W;3=Wn<pDwPovTG_|Q
z?p#QjM=s#n!4dl^$Sn=ghvTk?WJVGrCIM^NDDwOXJXV$SeAdvkIA<dT;`D<>&+U48
z9+B62IMN$}q_%O+i}QV5-QBZmS6cskHWycDL+{;;@D?aQa)dsmk;>EP$t<O+L0VKb
z7CQ%@kt2132PlO3jMnq+ufWNVw+g9ZHek2%!QgyH8=|WWQ<38N6YxYPf)D*xf1!%u
zRH-SZo?n46RY0M#e{@1u$#LcD{hp$SGMv!FyXe!(H<X6+M)G(A9w-gSDcIzp9L>*e
z)#T2gG2{+-#iPZ9!tlgL2V8n9KswxoN}60c@gjpf3xukpyq5}E?(7zP8gVCn<+?`9
z5R<TZ+7<IX0|cn*Nm`j0idwO4`B0~PF3&POAzSEEB}DZ2_v3`^2i5#l8mER3(j;pa
z(9~P{rEAFk_vd^M*{5g%;Omc%C6R|+%C4w8Ja2EAQ4*sow5g3vC}}X$b7wo7$MmD0
zOs7i^f)rRS(XvAd3S7PV(IPBUdA|l?=iH$KhgSuftHcv{dR3J7Zi+0|Zi`>HnAjhW
zgbz29G*JcIZggX^p+*pIE!%zjdCHp|917O=XV)#zvVww@KtuK`q!@9Vg*hl-lL|jr
zUzx};gUj9qx4&F3oKS;ml~MZ4G74nLXGPW;i#RRiO}UMU1y0Iag^ivJQU1}iDg5?$
zJShO4x%%^Y;&{HbGiZvSDfiQZYkiwfr;FHmcwxFn7c*p;gxi_|st1X#S>tZPgSl(=
z!^yV}(Tl2XA46KVb&Y6^<S{}+SA@Vt!ufM2gYXFU`D$z@`bW2L4;E=8hdGO6GN%XY
zG&kkCF&(dmI3~UXT4S?pIpqoPFlM8QX4d5+O<g@ubKyYgw>6B0)7EIxPU^}L)JWs!
zhQmB_O${~B7eSXkCF5`FAWL;@%}ZF>hnAgvc|9k=)my}4j>xVTLaq;<{`nmK>8euG
z1`l2vBbyTUwu?XqDo9_12T!YA6NI0Fd%zyIi9r1^!pE~Q3ucWVd`iJ2whcdJiS1F}
zSSJ`?JLZGz^)QVartZ_3@QGa;_f{Z`cTZ!<F(xu*4Xfu~i*@T*4dWZi0``Lmo+263
z&VHv?<4QLi^LGxOSHPW{Yfjc?1&2X*ULcsTdX!~3vtSL<w|^HEHdME5)o>gx@}Q#*
zPqfMhQpK|qFwPM^q50u=Q2BQ$#&{6_?gcOa@6m4w!x$#zMc@7rPI!5R9b9NqJN$dz
zqOx+VCNdlYSgA^4uy#2^5iEN-W?|s`(f7_g1z64<oD~J~5GLuGW!^OBowo|Z{ke3M
zQYTU>2deX8-Hw*dYsMB|i4_>q9K$UsBbOdcqno>^<zd7dM`VS?6_8aE#Vq=YEU7dg
z8mKe_Ud-g?#V}vD&Y!OPp1R~uKI@pA)>E|~k-ts0`t;ZDBXLNe^*aRR;VNs|*X>(C
z)yw`ax|AiU&~N5#Pxw}!rY6zWfWz-_)aq7aN6clOrsmjMCVEcUq5Sp)KV#QJm`PfB
zaLQYi>-x8l@@^3c&;TAQudA7!k+Y)YfhT(qQ}DM!A+2JNFh>#Xr=x&vI}}Eucknu|
z`EJR{S+uF=S(uH5UCctV83l%^a%VLkF`cjODE9naL<u5mG;_@`^#It($)7XyBde{U
z=j*Ff8sPXsdvl^8eA1#PH=2&DLm{>go|~eIJ8?d~AfV4Rft_ip_Zd<`-3?1Llw}AA
zv|jdLfBXowo>T<c+)cvb-uYf1PqtbU?65!nqQ?{fVoIUI!dGqyY#f__P6vUcXeCj3
zi9gcEUjj;WhVAKNJ!^iOT-}9NAw3hVy7@XMRx2uSddpt*`_~Xnb^cFDJ?O4FMVeue
zd<~VF7nx+gDL)(DF4Lw?YGkKX+k6Kh+FiR_v-VvT*+fm&n5trswBHHa(R5QA5KCYt
zTHxr7P*uA~?R#A!uY@AelP&VJ(mz5EYu`4;e69)<dYUwuYIGMcbpTcYBO>k(f{W{;
zy>Nai8{FS(Rbg1{xRF2e$fv;xJmR#}RQB5@#T@GE#tqoyMj>87gGEL_XU1aGW;;Nx
z_XeubDo8>k0TTIjUxT8&-O*1pk<r;nSpY4EZ^}P=Cc^S$PbASP<m8TP$881~W?Gx@
zFZl4*Pea~qT*3p9*?zwC7Pl1q!@7SEk;Qx+$r_uHDc~X%UWQ?2l4oi*En^nV8qKwI
zaB&a<x$go6gWb->l1znXePFjZSAs?)5G;{ymxcAjeA=VzVrZ8(bkxiS3=N(e+fgkX
z@jQ;ZYM2(thRMxl6VBz%0!1ALP7T4MdK_Aj=hwy9%((0lj+u<0FoG+bGilK}lu&-8
zk0X+Ki5qnbxPH>6fbN!Mkz>XyQ__#e;m4m&(7>9&Y2*EaXZVUJkaKXKu+4aG22|?`
zR@fB;(vBrM=~G7Z*bzf~b#DoBo7}?8@)Na6VWWj2bz9$ZLlt_@7T=8aBHSYrC++q>
zX{XY3$&V28)P{_b<mOao_oj7pX`F^7ZBVU$S-{W$BBR6|HMGW(Gf{EL361DGtMt-b
zyZd~9QM*;)TAyePzFgAo0b_0QWH4OC?tCw+#b#58-cMQ85F&ZOPh7bM#+I0WHSY9n
z9FzjC<fiqMqdB7%4ejMb%!XImT`EHBGQ*PBS=!qcvY;$gEI=?Zuzd{9&%c)1LBAJQ
zXV`u+b(Ljjrxv7lK;cOxs{Fp@&fco3wH+xEN3*y=uDmo&^<9UWubuMuCkQjcsgJWx
zspE!#NpRB6?{WBUU}5^cBP$gjj(I!y;w>~`e3@gl;?S$3iGeeC>NMSTKKrnQMN%p4
zi#g;$Y8<dUo~#h-d(KA-n)DOHdQ5(5f?Z40Dx~?D^s1AcujkO^terl9sSsOC&>fs?
zAUA_$BWnvK1;h{(ql^%S!$67}zWaGJ2>h6dU(M5pv0di1292)zN9L6Q%s^3hY3b39
zCq+F5LrmY?q^I!a7DhNeRxNyz9qq`@NMn(vYd`)7s*`afp-$GDt$E}(=ehDYH7aJm
z6SJ+o!zhsqI?Z^M!nVJe=K*Y#KbY<4j}_L_l;Y)782nLmr=_JTFnQd<J2eW}Z!M@v
zW#!>qjA>5W)xT^c=Ui*)l#Ja76zkPOcE;*?3A^Hu*&&L+;EFqK3-Xb<!tu~{n<+++
z@mgU1?fvr_s(kCT3Y8<xlDgeuM4C_0DJBS^7diNbr|S8!BTsxxU?c~hEh`5t=V$7q
zhTDvEEkauxXDsSu<k<vmvgLvZs6p<2XK`B<x88#9w66OkOMkNerBcYjOs{|?R@DnA
zKkeOMnZ5nHjO}K7F8MmI{CyY`&r-z)5-^_~u|JO;D9`a&P7iLtMw@yR3#-mK^6!_j
z2gaGUE>+KU^CE)TSu*i3IINHf>mRd?33Zy+KZvS#GeK=pXV0ArN9=QcvE^gcDx~BU
zqj>XAF)9VjO02@geqXYCVt7LteQ8AH{rE<xZ15i5D>>>=q8P>JXs#u7a?@&%-m4rn
zo!gq4t00YuiVcd#)4r+lnbi{3E3_Dk)J&R~*d6}qJi1q5>PfHE`A0cv&Ysc*Kn_4@
zk_=+38Jt{<!2M&H_?K=*%NW;uo8L>Lx45AZ{b!4d4f!YjH56ywjf(P#ms^ttmyNYJ
z(jG^8!M!XgxuhX6n*@En;lj#6cOg$L+nQ9uw0G`wyBTJ_O)x*yGzcdJ6_~|4dz9>-
zqy({f&PWfTq8IELf$K(cidMalEq*Xf#zc}?I!VjiN7+($9)mYB5d1NR?i>?T0V$?z
zAf&Wr=j&tj^%fj-x~@k`9tO5Zy{p-x<YQ<66GgxAAYWQ-Tq5kt@__m0f60RVrjoMw
zw&0C&=y&wKT*TU-Ns+ezAcoTr*G>aZ+x*=REqDj?wcb4eYIX_xmNdZza-_d4H+B_P
z9vV&JmYqGkm4nlBAhy&Yuk_q16luwEZOXuMB#X4j$6mge-g<eb^r?mMi^8zq)c0(&
zbav=qrQNO4vAFQ?*FfR*Gk^S1Vb@@Q*Q+c)>88C4Yh(lM!rGJ1-gZKR3(IQF*MPa<
z5UhQ*Z{r*ch5)I8ULd=ZZvy5|(Bt37c^n%TM<mnZUc!a;vcu&;*lXFZQ&;F>vav$S
zW}RF{>{W%k+iXWquF3%;4WPQU`|yc=6-HgS>tfP;ZDt#)&ie{KBvJJ8A^Q`|XFViG
z2&$gSjhX^}RV$8#%U`?Hmp4`28XK{m%dx6uSxJiWMN1Fs&!*rVSL}o*dRNk6ce<RK
z_wc%EkvI;t3OH?`_#f--T;S+adOPa%yJZo#)tsNYCqqUL)4r}wr)P3QL%*C^e)b_n
zeveEv$#``u-z%(qZp)qxnQ$1+Mw2v4HEQB3{J0!ylWrj@G-PeJ27)pIJ0b6v^*6fj
z9(yO0JwU~XGlr1~x`w4ZLwGzm5Z}{B97nQ<g&VTZ+poxku5o`BBfppatAY2%U3wC|
zANkmoa_A@hCX>H?L<S@U))OjJi`cRsHx>bqzdz)#rO<H4#S_ev&nYA|wqs32e=>@`
zlU=4epl`ZwpveX=bn)h1<?{3=>MZsX?Nu;nQi#Psp?r7IZ(M{ZwfS-j;tzL?3->Si
zgal5bUB8@*#hyt?HKIaO7C^nWM^!kxx&>=<xy^&Yj$Zo<si|DhcEp)*%mU^-_>d>O
zwdH79?!l}m1u}KDXbV>}H}Z%IQ5<`q^_>K%or0FyW6?BD;4J?LwQ^!i1O=~1qOb;p
zft5YWDy;Xx=u8F1YR-CG5NVl@pfffdLLx=L`4HcYY=Ga@rTP@6R&#5#KYaN)Qyv$s
z%K)sA8F#}{JDctN4dV5<(29=_Q87=w0zRY!sL2mbznZb0O9n1`(d;=<MV$e;DQc<5
zj-e1kE}h~0edr0ok^TB!tHz!$qB4duytZaCgZBW<$)vyoWmLK;=Tqf}%PZlIu{^H4
z7z%ry9va7TvdS~kTY=kaps9p=<$-)Zm&6M!eMN9wOKCD*Gwn!BJ~T_NQ6t*o-f`(l
z!_(s@1$9Db>(05ZgVU<;L@Q?`Rb4K9@^9}^+AJKK#RLw1)SHdm1g_!Gr{bS&)SbBO
zly1Z9N**6wijCCTM{FMx<>xz){X)HwfKT4P7`(gcsb>|)I402+;cQ(oy7%0-ag6g=
z6j*QlAdTA1N%5)Yv7?nb<#+86EqQFheNI(K2i90inr02)*{jzp_L8s3^;W`LP?iv?
zAZ?4A5`qSH!${+gOP)@;b5)-_mD;}VYp0-EiP7#kecel>sib{XPFMUTl(Jd~VUO5g
zS%*(OB$R*m6}wBkhb9>GCOsV}_12Gq_6(5VCE^V^1cU09nF*~LLvabh5uPk^a+{+x
zMtO)Q=EH(PF|*dr;!=AZEavwFf;-YpcYss+HDr*8v~t+ZJnxPwDrL68f;lbMu=~L`
z99Erz8YQKiZ;HHEim~d%&{D@M6Da`4xcTEtaaTgu%-|e@&M#-;Ry;Q8Prtj=mcZkP
z$B)xp`QE#Hdg^)3#XI*J;b#9)+d%&=+QrmWcW1q`I0i32NvpsF9UeId)3mwx;H71J
z{~{1+y_6fO<-Uf4r0II1UaWkO6@>cvt)id-3(3W6{tj_aa|?1@V@!*ct74QNiZA`D
zDbSQ4B<WRr!*MEam%KF_&l9i3`_v8%w~N#RMMdYfRFO7F7-f4_@tYK5(nue?s$lEo
zVNT7)4XS*{#eU)}4}IhZzfe3+{E>^?MqEx}?JTUG`%fm8mQ5a-(VJ3D)fe74!Lv#-
zZASyM{f@Y^dEZ%d<YeOnn?Y_FJKyUAmqi^sG}b7oI_sjh5N@0cEpv`~NO~dBa%a?c
zR#oQ8fS0dMn{%p!>tX)=a7p8C9$v^IHyR&pv<9z_bs|y1+ZaCyb?+H0ZusKP$ElI3
z)mLXg*_gVHdrWIX`RZFZe{A!2v{EpN#lz&$su#?}JqS#gLZPV)e+i+u#s1;WqjxdX
z(YN`CY#2ruG&C3-$Gx&2E(W?CCMDT-GYH(tA1&0R#KBToPoYp&RwjgMawu@L&$TWn
zg$1u|w-1%CZHI0iUWc5lEwm5G1n;cmdgs>X9t2z5*Hx?JE^mBF9Ky<QdH9oVTlhiO
z-Xkl~7L;T7BA5=O@!?s%(>_DsKG|s2_Mn~fI+pZ{Ix*b4{pe&b)+VM<sV<UxU!py~
zh!aXnjhd9*dAQu|81XwTbZz=xlr^=;rN_#k^e7Wj*T`YH9lx^k`pcKSge1$~wAhoW
zLoXFTXXfP|`2|Exv(UxMdr0NH2_;FAf@aZ64{@vxw$Gc|2d=Ms19yh_oMxyZe)r^r
zU^6Vg&L95XcxShA<)`DHCwm%Em1S!AzB*Ii*taX6(W|RAzipCB>7F=8EFkbx6gM#+
zbV@owborX*_2<%_JaN1J(K3j}@Z?oL4fR^8%{_%Im52|G_-n$lrf>I|;bu^;JsYqH
zlZ2<vbMW_>zKkWW(eStr)WSz3-_{vxG3u<$G}*!rzs<+pa{p@S&6kBrkmG$#AWH^a
z{8h&}&)d5H$|~TtF^40)`eo*|6k|p;qN?lMxbDqh))>@;#T6^}hllLimwECK`q0<k
z6t!*P7!byKFZ8fAc_2VYn(dFZ-J{(u5$Ei(Lg^<D40F=%JZo&=2uxFYul8lpLENMy
z;r-goW~JdXcV~KrrCD$J#jFhyXHN-k`k}fno(Ph-kC8*lq_~TGtfw^oLelnWQIEmJ
zXWJdbqy^IeGwgZ`k!|@;jx6r~=6vzHrzPy83wGGO8N8GGWHrvXZNXnjteiwn+%3)T
zKckPLv^`mTCeB^pdX?#Gx0odyX`-gv#1q!_sF0(6mSlGp(TQb^jST;^v_@RdisQM1
z$w{j0Zl_l-H%uBUCFlEisbrLUB);eWBQ}uiVwC=cq@q3;qgzcI$Q4Lnf9GgpGnSIf
zo^)kHMB?2W|CHCckF~4ib(`A+r7hRv>chkUa7ss5&Hte3JN&7B-~Z2XY=<NB*vcjx
zo2-O`Bos+@_C6$=a}XkXWY0rMg{Y{EBYV$=Q4X@#DdJe?{9e63zwaM#UXOF%_cfo_
z^SZ7VyPyOAbH)bE(9U>#-)e|9-ygW=hd3t>h%+2{^iRGpd-08XSr+%mWXj1&6{$>x
z8lILfhV<gr11eCWS+0*=b3kKKbLMXL6isMH9-j)d3k1Al*AickH-<{hVl9zZXI8A4
zDZgO980>c?9`LhX7><Jvq~?1t3bSnkwZW}V{fHFGYD&;enmU7OBLz$c_3kwXrZ^0+
z&$hhARVwg(347aN)*rv-%k4}l*F<J-Zx9!DGlB-}JYx3$CeN2l_&gdtzVVj|>l51m
z9=&aC*Wo<xl^g-Ran&`nLx_)eoywLvy6kN4CFb^2qgp57#j_9}xLx2CH*)G394x%k
zc~kN(7+)cakcLN3vgh@N^kUZ)GJ>koX|$m2&r~YSsefp860~sDcSPLbCSx1_1g~hZ
zN@~sKMgPwa9vxak)hk|7k8B{3V4~7Vv6vbc(J%E#DV%HPxAVR6D4vAO@({ZEW7d<a
zKN93i_^VZ$R9sXx$N8@NTVz?|636pqZ{+MKa^9u%(Y+E+bMbuU>H6>V3Ury1%i}%$
zH)p0Q!qnm2ij!q~e`}1aL_%3yr^PtU_o)3SqSKY>-ZVvlfQ_=+^$nx@&;#rAz2&QT
zgya$01^)N(3|h!EP1v{jtDZr%YQkH$zju=3dhGDCI5M@;h0+9{8^#9_K3l;d!SP;8
z_%ImONM&3dYWYXgw&asM6iW@fzTy?um?+jYo_;mD&X<Y3fOVPISli}p<~e%^j1YaY
zPP(54js}sQ;FT>_4zG0T6n$Ge)J$4*)*MFl-0Du4c$)+4kx>b}6YP4B9s2goOXUuI
zkz^~1Jp)&JBLg4ajW%{ywO6kX7`aV4&8z-Wg=9D-#C)Sro^Fent#<Vf?0y&10U`3x
zi!?_~P@FOS(o<FV9eV+%j|bNJRnpI2e04}*@od|e(rBIS@-gl55FIA*gt1Wfa^4vf
z)nj*ba{Z|wWB2Z>$^5K1ZG$YzL$O80OQsdt^65bRHSN*G9r8rdn~>1EyT8~FQ0(!4
z=k3dUnjD_J^i$RPf$#?XfvdTaJ$nFuDD*53<6rsHP>$7{89#)7Y6TqEUHSUm$5JxO
zl9*j(uu4k8^PH@oDu&*9&3$I{MWy$gAq1f+lkc4CiH#Yj6`1}3+>zX}|F6CmoyQa?
zYvmKQ;-KzmPWG|(fGx~vT&H~vw1rV^A^S!pku2i>dAl@DeDpK$w=n-+QIv8mcQ}>@
zww<TyU)|cS(PY-^^QSEcff6M;54QO2>gl8xNjwF$bU!DjL%TfqfqP&~*)w^=X8P<;
zD<AXd@%_>LzaSMD5!!POfl;^9MZ2u*y?AhGqY8yazcLlG=SJSXeI10?{}?=7Vrmi<
z^KkHH&WM@O@dfM$I;DiNQR0ac$MSxiJC{1wVrjK1EeNHYb%&Od%^d^dyZT|#k*X(4
zJD2Gi9I5GaxX@3x0Hbm`f|lQ?yZ-wSf2N=3w>qa=OhW=`#JPxGwT?C@Pi~&zYis0q
zJh<X7YNz%crbJcu7l}!5a!H>-V>7vu!a?2Jf@?~Tt<p=(J5lT_=-I9|a3uF#77Y|F
zVTIdnNCW0`qqq;O)rDAZryO@pTCg$W@Hfa*q=|f5t0*g%a<6aCIN<5AP=TEtL%v)m
z>MJ;bA72XUplitwS{;95NMDh9O>h#N7fbe>eq6(%ku0z6+#3d83Gz>H0u)@btD+{;
zW%!PxlZfbHRs|0a)Ln3B?LDT09iC8V`o276s>@s(Qx-e-eTKa@DXz!f61paua705I
z?0PMl=MHE5cMfaSl!J%}$njA;Llc?v+p{}uHjJdb7=IzU@_ZpcvdP-`&mf&IsD`<h
z-xh$?Ld%u*prV%#9TLo?f>c1L2Pjg~`g%(J!C5dAyQr2Lv@$}EvfgAUWiIYmb$poc
zO%{A~i71VuF{5@G7I#EsvE5BpW<19n>8EAdeOTbS{JSAP$ZVBuy5Tzl3fN><)xx(-
zySGrewYCf|S2@)C4d=eL(UUH+&8uGG)(|{74<vMetcJL1OfTA$Ob=<XPz?yw5Aty(
zm`eciH1hdP)i=TVI4LZBgUxiqndxI27|Frh6caGvC|l)42vA}Qep34|WCy`udATR6
zzbz))>{gtUq7u*_EUeVd0V$*L+xm5^F2|_Z<&kdYhChBKv%TU-TO{uG)wV@rek(8t
zMYREaGGeH3K$)}ifyEVGn$Asa6>DHb14<(xT2EqFyK3lwf0FBX3q=YP5l!7*HSZ*3
zbD8dg;^`AxuKDv$&UUb!nXr#@$<WdK?6T{ZNjKCuxCxWuZMCnqQGg1efnS?o_Vfdq
zHir+_c1FcNR8*PxLRMGQ+>cvwfvwDELRlaRBZ`|;!Eqfrr{|ljmSBy9!n{7-y+t{z
z`GtN&Ve;-F=x(NJ9fZV^5}(M0?h^mpY{t8TzT8EZCL-O`ly)}68G*5(J4;y_9;~jJ
z{zLor@j8qb673c5x2<&0misG*e#964E)7s?_g3ANBcoLdk3Y^vVqHEU4j1+KIjVX3
zVKKhDrI;YkvC-gmpMb#4A9^42G9WVhh^i|gu5cvPKSKo(GTb8XU0MTydPtaWTU~NH
zs}voge8AN7YnIC~mLFaphNT6`KG&+D#~}&}Zxw>jS%os@@+pNY9>QU)O>rp)$8E~X
zeUx3M-%647bB|*|Xu-DeFfLCmm=04x1@0Jcz-D)8OMQ|&f$b#Z)Tm`U!Kr}j3q!EE
z=J&n56$85JqVk-o%k)Fg@>~i^DrTW_%(V65!<s*fk%YylX||hmR2^b%&)~GJV<8=d
z-w`P%J=+}<HM}|>X(p&0&F!=Gz0<E`ktLR#@MGdJGH$}P^eq&TCfj8twU@i}hFDv_
z76bkxh#drH=4+ed+8T9&BV!OR&#IyC(0!m&(NUY287@0@XIq|@0vs&U$uBkS!^D`D
z_}xHpLNww0%Ewy?t(3w%u%pk7lqsvd+a2Z-q7{4R`F{<WWc-;#+b+zT)LEnbE(}(o
z!6G1x&xi{?4zDXwS@`)yB>{*s*--W?d%cJh-ha_DYajJm{px(&I%t|K^Nf8*cZ>2{
z0t<y(HWNk4b7o5T-FP1F^ff~j(hj8fc6EeYN@4j+5DuA>dpZ!@`|$xh!CcVR(dEUM
zfE#%%q(QoCYYj68-UZ`3GM$|msD8L7DrT92Ws-z1{rSY|sx(TQ@cbmGgW`V<E%(y4
zT~ougLZSi98hb;LpB9Q?@X$Mz+NLt%Sd}=n|31W3zGR66or(;A(R3j;IRRDA#`E=@
zkedQa6^BXaFV!N4o1kXK7GRyo)9$#M!X-q`^my}im1Y_-wdaYki$?stcb21}SdR5t
zpCj*y2U%%@vRPlbRG?VnGZg&@`6|;|hfAQV68?q9ex*;q7LLrJ%B~7=b?N1fJ;nBR
zdYcLLXcVxH2zYZJAx(<*OTQt!GV1|HPy<b*S^MDr`JtnVl!ek-DUCMh@28}_3YZsC
zgYrA)nE-C5ZumT?tVV+1i~Cabl4)UQQOT;t2XK4Q$J{`Cw(1BU>`(X9w_g4*4*H`_
zH!n%(pwR*y=(e#f)%M=Ox8eO?nO!q*AHHUkk7aS*J0x_}WjuC*o9J4BG5+rr4$n3G
zx)I#PQyCNK)?gW*1RF^ML3iY^=%h&4L(D!dh#-YIsu(_Pari8V-|MUg=h@szh`S?e
zR9RKr`J3NuI$9~-$pz3=ECrE6_~_}cc={WWs-B-d6eQ1@9O+qH+N+G%U_XzhMTsDI
z&;FV}wKG|C{FIk7AtqtLvPV@+clO3h#SM<U1GtpJcnNX1Kois5;^j!JxNu>N{50n;
zRfo|QD*T?FcT9T?+MeMA?MtN`)1f}QX0^tY+k%oMBy|%|cuIWI=6tS`MLM9zeFvgD
z;~G#RlwfsN7tuu<W`ioN7E;e47nGQ?c3sEYgQH`E60&E~<6`PD+QMr*`T8H#0vzys
zYGa=8zX^cZkwQ{Lh>eC&>`Nj1JyaZOVt+)^s!f&|+{ENZVT@yBW9tXxA3o=21W+HI
zFF*GgyFB24;-~6~kDUgNehFcX+qWwIs`)S_f}e`)P*vV6aI~);e_z0`C!ita{=(3P
z=wEL|p=!7nO(Y+Y9>IMVd_$10Es|gVOE7!eM>S8d7E|$!qga`*C9M^NHi@U)v?`<3
zcGi2jh=@jH5jmK%nduB>zKV+w_fbp$FM!BlDDMXaT)|<kkm3?}bW!3OUu!{Mae}!)
z{`sDo8(DqLaB{RtrPFs~eZU%@8^t+0BH6$C!H|BuBc_qGaI+%DjW@ECmUIEnk201*
zr<cg{<(}-ZmZ;;e!M=U<XNh%!^En=NtmU8AM_p4ZZT(Bu9q76)tYHB<auN3ZPH{um
zl07Qs#mUAuq*3&jUpiHV%vO;Hx~J!JFM|yd%&D%^fo$m`bAz63cR*f#Z?axw!gt}u
z6k7U_012ddC^Sl1<&nb2RS#ij)+WAC(JpT>>rKap$ZvI*{KSX?LVce-lW2ND4m4y_
z{2wpHLpBs|e0&RJ^oa1J#=5~tQOb=Rkv2mfr7P;Yb1h*_e3D#0RPwXS^0PE0SrZ?K
zkEH(tV*1<l0jdf<)5Sq1f4tT*$DR4HwLVq@xt>~Ew>c?t{O=Qmy?b%B%+9m@erS7*
zqLWo9Z1v^)+Zx9FI-`GoiqnqsP=V9H?I6<2FSZ@3-<KAh@IR6kv6>oN_*AbSc}^CE
zw}dfTiIwMlVE+okg=$<!rTA(I&Q5&E_i;n(tLNV-%G=fCRLf}Uwzu<ZGR?%D%?w7+
zpFZzUw8sG>#|og4bdhDYZkDUJYMdjP2k#?N+<LdsV{)epYKc^9AG{Wk!3xmgZ&$w*
z&rzO1i5CW6nN0G)X>7rf>}NgrL^FsgmCrSvs034QUY-PV^8X&+yXb{st*D#y@s(KJ
z<FcOAFV3T3c&!`|BK;a$gHO@_aRDZab;*ZhP=`aVpUf&(GOO0;{>nfl57N$ju;?6)
zKVln<fF;%Kdn~C<Ni3`@v2;l8|9$$<cHz;@lL9WgBS-q`Bf*g~PKA%bsrdKhS<51w
zkmnR-NxyO%%PfnE8!qpc@KdkKl{rn*{qL%ofa!WDC8M8=7=w#yAM)=c^L@Ac{4OSf
zndoj&+pZP*88m%&FAk@KA4}q<vno0@cXhtu@U}vEE!Rnr>3^S?U2VZfwr`{S9)U4e
z{77>bMzN0g9X6{Ve&`wV*?RKA9X(;wP=7?EU{BJ;go%cnC5^d9+=0E_shajCDU26?
z;F@MW(7+jr^#EPN#`uel`hv}xiEWvka|&03tA9H=I<*`7f99cVM!LpD!X7Ko0Gq5D
z*w4nAYHB<ZPtCr%7W2u2LdEWVha(m4|EC{&`4d$<4LgEpx^h(ZQF2`CuV2@@+n7i;
zxZk)Q!zSe!@i5hs?H$G&5N$}Pq?r155hmq`P*4j(PDAbrx8A~cL?M3@S+j2&?aO+&
zHe^E9zAmV)3iksj8+tl9iJh{#nz~u!U2|M>rvt-qzZ!m$^rQ1q{i6XFa-xFf@N06y
zaKeNVfE4h9fJHd0CmE6aN{Xr42x&v|-ppmchL}*zD<8XDCK*83XtlJUu*L9?WUY1P
zdgv!J^$#(>(AVbg6Ub=?$mMAIw#+UkPNKt0|EmcKpQ&N!ivZ)SHqpAS<x(f25Zpwr
zb~@~cN~0gmi<`^*=3=ib<U_Q=22Z`CFpWinRarwh1BH7I>$R^_+>MQ$GKlNovV9j0
z8kyEah{^>&P0V|J5zt&pH?ncpHrizA2no*9gw)X3+NX%wu{Te$TOBLGfw{BaK;$IJ
zg!@Q{6@(SHK<#J8?zgKmouPIZzZ*~SFx_I&9*V8ZxdcA&bwoQ}NG|0wsOTJ_d<%y|
z&Q#-06;i_2i1@SWl*%93X3T<atPO~`t*;0_j#`TES#Vy;Vg_4qyZ<LysdO=MzuL%{
zZ^%AGAtMkRJmtMmyyn{_D(D_Ie8;ODTsB9)>&JH1&fq%Tr^AvczxC{rM71fV>{jll
zoc&*ZlUy&mGF>vA;D7(i%~WF!?EM%UH2RJUuacf?&fmi~v5_MTAnqVwL9i-;Jr^3(
z01lOhYt1}=Db0wzC?Ntuz*HH0uZmXI{M4Lq_J5ML2X1x?h)Xc<AlkF}M>h?o(D6);
z`ej}Il!#dX6|p*Z#oRVyE_ZmEt+t`NzzW@iwF(V|L9sW1xYuQ%1+THk7$9DxB6gBi
z51GG?e{L9eDj}T9X6(D28q}N!$&2j}tS70K{H);Kif7C<ZN_}~?=4c1m$ohCWGCy#
zifc{?4UR2Fz##KoX}m)QOjyEFm0m>A9@N@FzxC9>NR0hZhdnQ0Vm#+WC*pCCgy}F}
z%_Fx5>gim_COH<hC(gl;vD^rAS-vwfg9fB4w6Qmg&|EH>wq`aiq?x0+piWA-yn?y(
z8aGq*4r{vK@VV32dc+z2Fc$E>5Xm`^>f{2W3cjN0V{gD~90c#6+XlWPXdQFGV>w)r
zAU`<F$8JPndG9{{LJpBx{rpIWjzN^+{{QaZoYpny!4%mbV!FXgd(112yZbdY*$&T&
zQ>52CM~ajlXc1KOy-UvaXoEvT;Z%b6(+&waQzwH6?yOsqpexR-ZfA@_Zrq-69rSmf
zc0TOdCX3vxeqs_kM|H-XyX%`wGjZSV(y_}0jjLCTLHPS+Oawsv0i<%B&RKkYMwi5h
zfr+)P+UfrABg|jJi{Wh99@%3gK@k<`wF+De?h79FYQ5yet7tdG8B^JAybB!Z{4Wp`
zI{V2cCa96Q{7C<gcAW%|#wW{~tltuE{gU|3u^B1fDJ^>P-=P~7+_IjYJjAJK{iyFm
za3=?9ImVfD__-%KDJoQu$gP3cS^KgXvtGfmSMl6EuW6zaB?W#aQ!i7evkruU7vH}5
zQ1Ev6mIFSp>}!s4iW?3H`egtNwb~q|BKvu@L-y+t^}H{YpB~o-sd3R@n*)3&E>j#A
z$WAohvA?AK>~@A2g)4R9)<qzSW(5?b?E!_S!1r21H`xc53<9$^$<2v2RJJs&^cuPs
zKlllspbIH&EnWy-6)rn_?npFaOF^V0ROP^|fC*_I>GX%Ms6(+vzaiqv{MI?R^yh*R
zkOo#)u=2%SCA^j}&jojW#%pP%M!1MZAIV>iawva9D~w=Y7NwNyG<^w(eP7Y?fYzR#
z@40d^F(x<+d$$iRYP4+Nq{a71YV>7oi{jRf%mgO_Rw~%)Bsx?QuGORaOfwwXRU;1k
z7<qA;Ja&8Zp)h@H-!`TFSU~#~+-vISBuP1Uhz*v8NzXFH0K#d}AV+w~R<ASmhNO?g
z-b%jFtArE5P2t9j+>hZ27kk0FtZO#?z2@4|s%6V>^F8kH(=c7-;<%kTx^I@FkB{NQ
z?+rD)nR?(6lVz)s1{r;(8izx0|Fbc79x-C}=P{<2>_*!f-?;40+Y)$Uh9<5!U(UKa
z%R|+@2C|AJGq#AZe|*trCL8;GLz_8NosdUA;txl@{|B(fn~}B0w8EVZ$=;({>Ux^M
z{2bf(SkUbll8324lhm?i<Y@@SPj47mg5JC5+)VINj-q19$LBd9oY(~CN>bGd0@@{x
zQrz%qw#=C!^hw1zMk3}P{n0imra_A#C|wuDIGN|LV!j3@(Sv0sPaZ{N9-?n){}G0H
z7fZ9H^aA-}Cb*qSttR_dK_=RTYOR*VPl8xhDF_k=MI&l&sq^O3*j}IlbF4+HG<-3>
zeSFYOFyb!_rh0J+5FQ%f1|ayO%DZIp0IwJn0@{C2{BjiavXjh01uiv+AV#yXoL*(u
zw-y%<iWX~_9D1I0<E_R>mygxDr6ZhCN1~BkC?hxzKo>kX_rzsnBjw`i4h?QsS?3$w
zUcAz`gS^*2WX_>udm(`Ko?a^?GJnVX#OEIEyk5o8)#3Q|7Ht4`VF9GzvML*um-BlF
z!+yRveX#&9RHeqJbPjYzaW`3*nnNG94dYEX`DzOWaPD6{tW_oF=a_6M#bzh?%zdCX
zL767jmS@lCr_awCMuj)1N(>cA|FFr~Og#X8eW@)`Ag1cl7?pPGHI;qN$x%{;xVRIX
zQC|X#G01#h;{@MP;C=Wu-Lrc5MwCNFi;?`x_9D1j(j5I6_1z~N%mE4mNgm)}^QR1t
zf#lDmYDkrjU}Yb}-Q}SN<Qmq%o@umD94O2Hc(EbAS3{n4+_)W3^b?=OOZ4};u;yQ#
z1%Pbb;O*(3%X^f;Acfe)=Hs{gjew7e0dlnqwADfT!~L*tSB!0S_I`ekvw0E=k{ajN
zvWM-aOR6p~w6lH5D*Zv_qj#&r=;nU4#4|W@<Q4#b8spqE-+KtjGZ3%6^Z#7q-792N
zn9rEAfJZ?mu}?hpk(0o;XwI0p2x{WS0~Z;8onv5_y@lUf5voPCvVI~Zj4PNY8oaq*
zG!o2p0V!@2^Z}iDr9CiV8s$+u)&6&SZmL?&+ckCY#Kz4NY$@pU4&JRaZCB_S4CO_&
zsst6X`Il;4Z4vF)zup|z-6nb$os(>n7z-N0<gvndX<Hfp&c38nFcdr5NwoyS8Iiz!
z&)^I);&$4^UwYi7G0+SYK53Loxuur;wzbo1FmIDCbu4n|k_YUF8pV^fmQ-?Gdfi{1
z+LEG$$$ox+CPyuhp+^VvBr0Gv(P>)!pUl8{ir2DY*4Q7z-bEL=Kz5ZuD&?2YMA#(3
zOa`tyapuf|-+UsUd$l(F-g}d&2xL!x<A6IKm6#+Si5r&C1i(u3f2=n-2R*h)j|J6(
zq=3IoVN0R49hZO3*N<jC<ZX}yVJpN}M$Q=V4wVc38%rMw)Mn-J>Op%8a}>7llylv~
zW@kmwcG;d8st1=!PA2PW9It)vZWsWP!-OIlQZ(Mq&<lSx=V7Q~;Iu5wGk0a4)?b!%
z1bccNOB&myiZ3sYE1eW6fV0SZHiHQaY9<`B03m@r%p^8~^(+;~Uc*uj3!F8%#xoyv
zkWUtA(csx1rMy{*hkELa#Q2W|-0VU$0$`Q-9`6&<sN7b6eV%<pQbgG740We32(5_#
zSu*s}<u1p(>_qd;uQEB5GAfKmRGS91TLqCUM)$dq`r;xW6i@t(%J3zYY0B5?DKvFP
z#d8$6b(zGBr<*g#V<!9nTMDz`l4xE+QfgcT#Tn>VhCoURFtC;qd`6EdP`V$880*@L
zF_vJW@Co|*YQ_T!$i<30Qxs>PkN)U}1}y|6yQ*l)xNJ{@=Ak$1NdC$a@Su&eazU1c
z?2Y<#rTr#X2<SCGKUn4_RA6OM?s+dd!D(6xxRda;@Aycqmf?sE{iKNOSSRE8EoDCw
zHcEJGq4KA7RSu#zLA!r>@4b8Q_w7ZloEGK>wTT6F<IdxV&8N30KutAcvTuxVXo@oM
zE{ztokJnn5cV_VBnq-JK5^`aaT(6r-y#ARJYM-@ccQ3HlqbxBdx_>di_Ml0dp2C)i
z)R(#HD5{#zbKWLXopC<R(=y*>ie0^yU_J)u%b<u)7KyrLc*j~;PC-(0`qh`aety#j
zM~cGD(>x-QyeVJBoH0Wd(i!Iqm+mv6<h?d&8Z+I2fgFlu|Fc;&A)l)-YYz(hSdIpJ
z+rT7u!&{f1rfyT&f?6MG4Tb%I(`$93`#Ssc2zvkCX*XvQQf@yJU&vWPUyM;Sg6TcM
z`~pi}fb~JK46VBN_8}~BFJr1*ZG$d*np`)H0;YZgpixwfi@(fC2L%4O9q9iRE@~_?
z=Xw@G-Stm3ral0LY-KmHcVbd!m|N~$@FEgzRJAkvN+Y)ToPVY(wMr>j47BDxBP3>&
z^Oj(Sh<s#9;|T=>4)8cSG$-URqmO{iYqE8Mc?Izt$rH4zSl7cLWngXjZv{>aJ?9s1
zzG-Wnpv_?g!RaI6L>ZKwqB6IU@3g#v+50I5peQHcS$Fkura+gGH%~M8!O-QG+DfVp
z<F?$rD@z&#8hpUy3};+S2*uvW?BZgMxx5gtQ*;3QB}=B(DJL1wV++YiQ2vs=hKV>i
z8z>PMd>rxcm8bDVrI$BUXmW%aq^Z2Y+*fOD3M?K&q-+D&=(Nz3q5Dcb&CmXDmU_u%
zL_>*HwGJ!3l!^pXGEd`zMoMo=<cg(()mhi#42LMVkao9f7XijI<cywcJY8>jt%Bv_
zQBIWmaU<72kS+C@dNNSoF;XmyrNgx+?f=DjSKALu-CoRaBu?C_|JJ%45UFwPo_{EU
za8Xz{Lka!t`<sWVs9>fCTu2e%-{d^@*Ga1sE>Hm-N^)rnYIYl2kKIPMm>BlWVMh)a
z#NmjfYp9bT!c<^^9S}#s`L`CwHCBPrzym<P{$3Y<{z*`b52ys9D@1m4M+9Ob;@;5V
z7FN|>+)l4;xc<up7O|#Q@AS(AFuQVwAzM@y_oIxK-ps{_Wm5J&YTT1S`{@@=@ZnhS
z*e`8-*wF&RE1(7rr5{qcw`2F?zutInVJn3ax~J!3q=BGqp^p6i>P73!uk{wjP=(NS
z7l&3=#gy<)!16G$8s(Y#S7;=)rscB~ckKQm%ApuCDXGPqcBouvvTqU@aBF|Kf@`}O
zc#_(Oo+~IYcGYP9JHd^+?D=7L+7mg1sfxQ@B8Gay0383+LWV*>Qp9D%x$2xcs=-AI
zu+}2Lh6T0@>g-AuHrb&*N|I}K;K=IjPcNY_GqHd7ak332kIh-4L4%dX?dYesD<23+
zUWC5O(zWKo8noEq7;!&<CSOvA^%mU~HLGajH_IE>hC&rL*?_=D?Gab#lP13_JOi)d
z!5B5`(FxFv=YmPi>;pon&&JD}Ilbt4?|8fD-=#u<2>jXo3Mqwv?3+I)UaWEr)95+b
zwMKxcrk&qsmkm2*F9w6j$XTuhib$Ehcl0;WT@C07^Y)g)FNn<qz#4Iazh#p^qBHT(
zjpOQ{7VM9Oo;YX75XqKD!6{dMAsjopkKL-s4z&-2hDi>{fUm_9&hs+(=QV1mtSAf_
zb{5@JFX61LRC30*fzg%4pWAB+L9wMko|&V&o?lkf#VcfFwe;@(A8l}i%WuG1^JMD!
zU!vmM)56hoy8AsZW)b?_an6?fO%pJ#0B(~Sz9L@kuoS6Td)8mr@HrxFYVC7BWiym0
zmnD-j10!CW{-q}CW^Oxyt`m`l|Gf@KQ77)BOled6$#h>hB$PSy!>r2vX+9v=xq{EV
zCG%?~g~^!RU9V9dIRh-*Mw{E3%drr(h)xBg&+sEVohUS3*M8x7g1IbUJg;}F12JyC
zfoCM#B~)z?FMq2=8fU^5JLPRQRyn}tQb<i<r8kpp{eW`=dLPr&8omOQB_e<yZssn>
z?&k@4ejg^#ms;IVo!Xu^2tbn9@U^Xl6zWvGyRZa<+pRx=@cJK*p&%`YOo~tA%gUoL
z%?ZEFXy}0F#=Csv3>r{K_H;?xi~ohrLN^B<kWDY0Ygwn}GP|YBF_N}$M@BR@y&y6e
zAPd}iN1kPIi~GQzPzlB6SKw$g&*ZZh<bj&)e|VrQvH~A`LyF5kt&N24!haYb1F-}>
zNd~{<;ixv4rNNS)s@$e)#*Dv$vQa4FdLLogcdmpG{5>whAHD17dglRGLgxR+;b!su
z_(;Gw@ETtciWJG}mY44dBWP3inrBS;>M#yERJVNAF0x*HSno9{#M0#+k?$W7iv0<A
zZ$g>)pF@4T`i=%EWZjss6wLop?C{!6TSJxCQt@cyN7w56BFJ|m9yyx)<dR@81MY9j
zhGWhJ{-@0M4$(3#cKD7*wO}S=4V3sM9iblIXxplp#_Kq6BPSDZhE2wKirh%)|AA_p
zeKc#vINzbrG+je0%boJ>>#eJuBh+XR>b@f3O05Sk?DFH@1R=HZtxnhX6Av4{c8%b2
z*@~T`Vum!&rXOBbx(`RL03a%0z^wQ>V4mk>WqiR6_Y7@m+vVZ6Z4ve@jCgeHNmlnJ
zLj=|(UIf3oqWHUP{6xG?@jDOI5&x*CR#$&K0PXT4!^0LRIKX87R5Yh&IEl7}u|*;7
z^rdoC|1*j4DtQNg2F$Fe!m+?xI#cRZ<#OsMq)f{QPTS1mF<gP4C`pWyFEV22jF6~<
zs_{;-C~iXj!}>Z@B~{0C^RzGbMGqW=eCmN1lEhZo-!d7^c1++-8)e>v(5s%w!TITJ
z7pJx)yAZWcX2_+^haD$3gzQ}3gYOpq<~7LlBYlE|YAvg!D2@qAbXIjQd*C-;RYy7(
zhZ{U%zbSSa^zfzAv<83-g0vV~8H}=vIa1GsCf~JFgv$OE_&cA{SlZ};Ib^liwU_)$
zLeH!NnCiP6X3*nCau+nV&n|3_4{JP+#FvRP0duih1hQvt3~GFcUvj$!_-borTn%KH
zbOe;p@83t2Zmg;f;uuP#Mafgrwmc}Q_*Yix6pE*ptXuQDFi$F@nPx`nQX}&-!vqNu
zu&H4gZsZ71U2q4g3tD2-Yx|bqQpU_E`nAwe30lWr3Hvqa6Su3R)}Mm}AP#<iZyIPM
zw+r15hwL4#STS%Rd&KoA#8*uxCODPo=WaSH-P>nV5tqtMjJeQ1@U5hLs1t{#Il4bB
z98EjR)>m+e-;*RQy?B35(D&+lN5)5z04Is`QXIQ6-VMo)K~s~-7@Ma9HyXOS{9{Zr
zzMg9EuP-mcL}^^6Ro%#+0CX=GFObg*Djf;U`fb>u<)$zwqw+C+j+%t58d!kMlyk;7
zkK12l$}k1LkL{41#kluG;i7ThyYD*SHF1=%kOJWhr)dE71U^NsU5dDE%SCkQ&E`d3
zxjUUuJc-<Gd5y29Vk*yYnT$P?c!FEV`h7*@XP5S)ucC4^AkMIs9+$F|@Nyu9r*!0s
zk0;(tE^>hCuOQwZOEx@Rl3z=sE&9&&&N%;^%rPjlo(cKZg{UeiaQQgEX~EJAJ?jt}
z84dU?n6;k71L?>dxIuWD$b)$~PA$0#FK<qqd;O=YMrW9mhisZ7FB`cBbkO<P^*(r1
zboPP6a7V&ViCBagGQXuOMWDzs^QbVJAu2nvB7~X>DwH;9P;@7yMuDoRjKqkHSJ}P|
zjMzW<=d!78&EzjGmuDUNwn3$%p=m(WP#TxYHRc4ED{7KEj?=7xQMfNiB1>}UWRLOs
zD6i?qiv(c{_P=?&s|qPM{<+{w$>9&bx}HYj?@fk#ywd|p0Qrv_C#fs-hv}yed}Qu?
zbPY=X%{D}S#J%@p@$%%%#jOD?-N|*9BW~heMLTHVMt3GI9f#(!6`lRUbSJ!pi3*&R
zDkX*$#Jy$;c9+X@a(M*AbZlu%qYk5gmhAtqR!n{Wg}x5dj)18sIu>*d4Ec;V^mQ;)
z{a~~*Q5rA*@>Wp8kFGs4+@kFy_xwK&o_DKas6zuK-T!c<@MRZzMjM?k$5KrnTk^WX
zd=R`|-8{$pmkAMX%hEa`kRUW;v|P(DX|=pGn%~cWRCXch-_Fy)m8s=Hggo{W7&FAu
z(OeEKX}c0yr^_AlF28M7fXJQ968yFQ6(jK?f!)!Y)C^GE+F|z$MYS0;nLwdRv+9@N
z(c|*Cm+?gmB=z-(`;kjA6}TfDnkT<h-?}313qZ>1!ew;pCPJ*jh}gK5l~HH7tK;E<
zc;0yX2tZy+%(fq^GbpG<8nyJ`(Nwlcg<^gNxuDybx1+TJ+X3Fk9TbCnk#%SMTC4n#
z4ASI03ks2;96d1K17%aor_2Kp!L)*tL))7C(?@gwmq*<u!UDY3xd7&k4bSN=`%PDq
zw~*Hv|K<A12(3EyQT$6It3M9gP9;N#aTmDt)Q{eK)sfcGR75wr`k^P$;zB+e1XaTR
zdZgxT{{hnV39H%a)Un~d;-va(@oXda`gaq@u{|~P(G-L*`q9)K93P(4JZClTPH+~d
zWp08J3vxeL@pQll5Aikl7+>5E+zwtoU^Ie?v+ml0&cz3q#NzgsDT9HR8ad@mZ?)q6
zg8<*~l)8x=4MK5x(wwv7uWw{ek)VMV>Bdl#kVy}v!Owf?u~Px@vt}-^5I}l9t{-i6
zN)JQ^Tk4l4A4IB8j{OAB@4Q;->IVuqZsZZtH|c2cz;UL>G?=lNAz^T*6!W3lB)W{J
z#0P86eulumWV0$E(Gl-_JqLkWW0?d*+mJJuGu5KGs7!O1;@k5*KPp^*_ui9CfWL6F
zw1d_4+BcRmgI==6m2XQC97^lO6dE*<bi{A-d!sc5OwaF<)^eFj=Vp|c=zIAQYiMcw
z*EllG!Lhc|ZyX3uffpp0g3;7RcdQjU6ZAxI4fsU~oUklk7uVAr4-5FeptRe>#aVu2
z^eFIVq6l#nI!f70e<o{&3a6c&9)+ao#b8`<-Z&adIs6@$E1!xT(7z?0YjZ`(bX~ow
zji0Q^`V;-KSTW_szkElk51=2#edV{l-c)>b_P8T0WrZ=dbm=RG3??v6eu-NHNxdr_
zXq6ybf*oDzxH@cx$MV-TM>8X(Cri=}1bthHElf%+cP9qSy?X`l%ea!DjtgVF_&J=x
zgHJ-(wBKzLqAkBbm~WpKw;e-cZf|@tX9Xhej*1j=m9rBlOIP&*5lU3R!93R}ev}z8
zR@Gyp=w3Yas6!J_9})+&UZ?xO!vM%#JP{xo0znimP1FOi{f{o@A2@#n&&d|6>@EK4
z0rH~Q5DMRghYhbVRlSH%Cph4o-_}n=QOu(9ToW0)4F93_RyXNK*IY#)%U_I_S{#W3
z;vT*h?9I%Udx=S5d5}EJUfqe&k-G(fP~aG<jd!-!X2d=ducjl2Vawt&pV_;=l2|+m
zlcSIZg@4|vDvE$3hn+m8c+{OjDEup_b$_@@dUoMsXcRuE3+dYO5cja#nfu~rBPSKx
zUuO&}@rs0cgEiUQ#_^QKQ-4BMQ>Yl~Dj~<dCTvEpN}I^yb?<%uhNcHR48SVaC9>o8
zha2~W=-Qm)7C9R~wqICQM&^Rv@Qww(Jyc^;npfwzTfJ~j#zZLOeHu?7fnigb@AtW?
z6c7fGvIEr1>_>?Jg7ELmtdai7D>h}rs90x1D^4Fs=@{VzY29LlLo*VA-N`2Y79;Dp
z1uB^>w*`PXLU!51T@<G!c#8>S0>1CZYmnh2&BNKBj~ZWHQTIwp#K6YSUs)$M0#i%@
zM!cp%{9B75ubi3huh%<(2TpMeeda3jRkDuVRr{Pnjp)o(+U{|&h4W+-{)C$o7>T5h
zH+cWxO`Uu8tIzMmx87j?bheA10wc0GHElN*^<Z6*N2j0QNK?Q(9e@_Dz0nwZz;zQP
zz9HFq$A`6)wT^vWsD|57GYew0N!m-Npj=2g5n2fmYv}5PR;{#M3=I`*j2HVA21}TD
z^e+#H52#|@4hEUJ1~|zY{E_G+G(CM}>HUvhWmZ<1-xhNODx=hvoSJz{YP}$`BEm%(
z+1-_Yc}*cDP`sJ~$O2fZ49*`+3cLa~GU9Gbk1u-t>h>P9-ZDGbjWB$0@b>zTa&uW?
zbj@#ndG!bp%_*`r4_AD?9Ztw~p)gaZj`3_~-Txg}m07L)M9|>Om~F6Z8Hba@b7F_H
zseUjrck-Vv&)63Z^+Be@cVWy^u^g?-m=Lgc8?V!ty6;!05Cf0^YcA4G7F62AP;uF7
z50lE)Bp2Sk89@0Tez&OnKP~`<?BL>lwX7hHr9%LE20xRHSy;<(wQ6C#ySqpp51Aw$
z=oYV4<P9lrO(6=o0Nc@qlHU;iCgNh1E6=W$<s6J1dt@FlANN%(&x{t7rnA*uffu0W
z-x~PYs=)g;oXRMqTr=VZIwv3HoP&3VTUJ}pm9J&;Kh*Rr8KKakJmZ$8oPAuSV774w
zeIOFR7WnKPSOCKR`g0riJCp6RJ)RPO1BNGIKf(gosF?UT(+jZg`BA=-d@h^>W#I?}
zMVr7gGmqr_Gd~U0D_;#v;>w<cGX(h~NB*`XidBZCU2Eiyc=)IB9MgnRXbq<yrnG-U
zorEj^o;UnjP3Gx63lmFa^Zf;&J>l|8h7_g%bqqjbO|06yS;<TZdY!68&kMj6TS<!d
z$95^-hgq(cB-cu2fAL>K>R|fz3l|WDwEx{mB#LCq(Uw@bO}-ts=<Z9lf2j&1S=C(o
zO-Y}W$7<fwYmK{02r=A5#Bo)OblHj1qV6wMXw@VxnAEw`VG}adfLI+_j<o&ja^@WU
z<jO+_x>kWGt{X-gKa;O_`%%?BS1#h>`r=OS70ubilK+Ko<jCVU$_dd*o!i9oFlHaW
zr<-<sJ7WsXb?<J<)sYI$TN!l8Gc>B4s&oJehC%ZcFzze$tA+-MH<s5iQU9A?x;^gE
z`uFX37s5x`heYAXkbf%k2&ge#PL)Y?HgqSx)@f8k_Mqxx(Tql}XVQ&Pa#3a-m_@ul
z75|9jP~EzW3(xb2nZBd!pL4~_ooY%jppgb}t2_@e0?D;aN6-vA#Phaobd$7F>Kr|3
zZDN4(S4V(Hn><6Cl@=pu45wiI>NamE41EpUtH28TORa@p<e=FfD(OR9uLdGsY|XJ}
zTO*ig9)yw44X>a|-Du{r!_#0dXI@O|v}6lG?>)|zAf3(cY(CHV3aS*um6>xb&aA7u
zFa(~dACMj^T<Ch0-4^o%!`exvvrQiY3x$=(<31bip|#m);pr3*>77;|#GtL3NG*LR
zhb7hYvfF6QKg7as$yqI4xJmv);OPju>_N}MJLIk^Y?z8{jlYefVk;DSFDN!AXu(K`
zKz*Pp<7;KHghXZ>kXB2~MKG=h+{l035SuZ}M#iu@8he?e%BSZJV=dKqf9jt!uOLhY
zM)bGMESYhiVz5~1PN~7tHyQa2tRm!|x1ha!R8bt~e-}`ZBZ_mN&@#S4ac^u}k447q
zEOi$pys=Fl!SSN(ycohaniC7N6iNooQa^Tuq0(=AB*T{{kX_i+9=}=G`!rA;C0c}d
zZrsigHv3gIbn->_v3AU<nd@~LStkvY8bJ;AJY4CfVoDwG=s|gMRgjfW{dGEfcPA}P
zKV{Z6nJd{!0RvZH=K8i-Z^X?=o`edY+wVgva76#{Q|STgpNegaSl)l=8qTQmxVW^x
zE;!be{7q?B1RrhjV;?%(e59*`QT6l_Mkx^=V58Gl#%i_nLTI|$Fve{sG~CFZe;f!N
zA}xX@R2IRlraT7}h9~k9c$?*<HjsztoHwE1jp1!+!@-^!@A+GJwC4hNMO>8}F-7Y2
z<RwTD)-Zc>qsy9~c)oK>tCBd05`zbOH5ZnD8bW?^!#Kk?gt?Ib7f$8l5MAeS7L@1n
zz5Rv4?l2Tu&qUEZJNOmf<M3;?VA5qm^|toM5Kh9C9UAcH0qUda5gkCZ{D%N}@d{X#
zh)8_8(CSK7q2DzU{`-zKs;n)gRzK~TnbH!b83J%S#t-Wzm#yEyI<kXuYaud!dfFPH
zLLt{PK_}D%*@?a8px?GOE<B8n7C5OuSvSw(aj|}_M7Z!s)-NrwC{^q(U~1+%JCZ3h
zomQ`2{{5D9gYmuKMon|&Z+mfX*T889e03nt9sQNWA&hG#$D)e`cZ8$5>0WBFS6_JB
za{(HA1tUfwprSuvcashXN&+Toj`tQj8<fbeCFURCzd@|ZQ~yR!Ku*hX<61bwztpG?
zuVQ#F|3N{#NCnVWWlRTu2n{nTUF#PBp})6AOMgiHc1=<pggGQUJzs#&r`O(w(LBYH
z%o8sfoJq0Ot|d}f#h|%E8`Put)w?<`*V_wm-X=x|;i&B8lcs!4?pq`S_+xX$Hc+xu
z2Dk-07%n@dl;84~)+oE`@6Y-(J}!bp(`9k1(tAQhXVTiWkFFJv&&>rL*|V}|c|6(L
z=}GpeE%^^p(y{Io3T@YU({=aO2&AiljLtDD_J#HPzUKG!{yl2U0z*C<RI7NR=XqA0
zy&dD_f%RFWTT48iOat)s^mu#{eMW^Xe1kzmhZvjOAwJv^U~H*21aY1~SmNH~n=s)E
zi0->ls;W3shR)Y}>jOVJ8_7J3-=0P5MN%FI1y@q$o@Z$6wu>_}i|v+_2cT(MdR#}K
zvU)lgW21WKG0@>3RQ=x-1xXzMuC#WU?jHdP+<RuLcW+6Zo6wIq&!TgcOY#7nq>(i4
z=u<9DO`fBjry_ybC{;_BdOeFW_zw}k;9POSm+xR5aZZM^$DWSiPpz|Va>3Aqtg^xc
z4i(+muuVh!IZhI$fZ_CYnmqV-@UDWYxF=f!SFzezegzy*8$hUgUW3un1TY(3D=Stg
z8h$+mj$La}`!El_{ChL?0oBG}D?l8=gCodrWb?n24G}Gv<!6+oztm|l86iAGs*jkL
zA*g`z1fOfysNkqAs?nFPPuucnzK0PqJxobvYJpFgR>Rv8WrTJTHJRYQ7LV$fBM~XK
z{U&i!2=0^%hSx-TvD;)O1<)n;o}Q`*YRvVqp+Q5$Cb*ezeyT=p=^>7RO@+rMF5K|4
zj!tEqISwBx%=J)`U`_B(e=FVeN&Wlk+sD9sdWCUww=R-L=D`vF%{K|12~I(s>aqLo
zm7;He8lN}G#=Ye=PS*U8FvaGwcU7lS0;hs=D1DbY_ke~#jnb^e83(eU+Wk9NMgf*X
zjlfM^_9Ho)%rnmcz0(qfIHw_!Kkd9Sz8TlX#=HBv|F^b)$WjR=Yw=!&>6$NYA*PI1
zB;l!YQx&%U`CaN0T0FhOx`8A7!?-2Y)#7)pPd+C)U!@uIg-xjdGz?ZFmG}1&8!Oe9
zgAQCvUHA!l!9aGdH7c}ra-R@o_P*1_k>Z5OPygeoKhjw@lOL~|lhnXs$y^K~>Bz6c
zGCr!i{vCs?hUMy(UVqc8EDl-qnbr$n0ys&UJgG?lI6-+PIN2G9p&4qgRd#UVBH3<X
zZ^K%Y@#chm^IP2;9=G|LU??TjC8B?fP3AFQ4d>`zTy9lIUmAqzFM;M;0Q3C+SZ!^;
zckz*TKK#+fsY}%1o43gcond%S!jp{4ro{~(E-3BtJvm%)j$yB<=dYnVd23i3Vq&E5
z^2+Zk7A5)HvH`THhX5S%>IWwevLMJ3+;BnPOnYYZZU|GC2j~ox{E*6_GJB-b>P198
z)v+ilF}FxX!H=^g4DUxxg1G9-T-Cw6HstW&kH67lU*owKfV0NA$BbdLSt3{2ZJC~`
zwOp{29<De)^76M}IhBA}g~8WeFWLwJsf&30l+o5}f8iq0HfQsAi!67AmO68}c`v*>
zE<Llv`1+ulaKX_7Lf(owBlEz*yS#_Z0tEAVFAPw+P)vE&1yoK9h~`zNOkw!C#yWA&
zuP-_YLV!hk++v9{H4^i95A+m`VFlN94DWBk*~H<nGe%-atbpz0tE{tgwJfqpqbwl+
zCN@W!?T@RI1hsQ$uD$5~&S@@@;-tvnUU6ePzXYer=js%p<|9j=&<s-fc%KfA^$bwY
zxIE5)KgGT}9q5;6q=u(*l>a_D$XJJ`S23)<khmZl{N?e-zL0171AA=;3Z@ik&!P$*
z>0+Jrr^WChTdd=V$bU4UF*PAoSsRkrEU1JX+QDt~3H~tN8^TZ10U|P^_%{NNVsHx-
z?lELOf{R-CC;g_OJLfD!!Jan%G#XLld4PRz#!PF1x%+?LI!9@3Xbwsa8Zr~Y%N&@Y
z_0*-8{}v)&F9$=8;>zN9&$G>0eo_lNgd*rEps8th4z%X1#*UgWeVwnjCSEm6bjgkv
z{wnX*Z8M`9RE0I$;0O9yo=AQX%{D*uy}|Q-y_Th2;N(7kE<ExAG4W8j+MZ`HJ@e<?
z_Y@t0lVK_;GK4s@uLHjggE!6CEWGwk!_@<0HPbns)kAn1nP@|EaW|jrdU}YNrdGdk
zQ*YbbDq)T3V_h>vIwxduKJ+0+q475-t*0HRt)6a|SfLIeorHJh$Y%3qqUsizJNc#h
zRVHE{Mw`ME6Y;WSEl0WOC)Qo4=qvvl#RUq-7~l3H7T`LoUpSvSNJlcDAhTjU^OU<{
zA`a$BrhrHKR&c{sVe#nNR@}1qK|9amEW#mOX)9#lZ=GJ$c`4}Wp<bQJrhX99PQB6l
z%15o7CWnpp|DiOs6hQY-34@_JIRw(N+kH&FklZx5E4S7$5jD@@NKNd^JjBy^+(se1
z`ir2H2L?|ve;a0_G#j92JRX@1QUN~t1<=mL;3=w!*k!qT{SnF>z5qjY(;u9nFfR^X
z**iD>3?d5Sri|3dDiy;ovVIT7(iaihTt>-AT$rOWF&$ENmAiEy%1GP75B;y%(fx0v
zy;O6p@ZfGoUze>*(;;K&<h9QC_IOtt6|C2&^u%7#I5FJOIgYZ85#V7ZdE&rH`#*9n
zl390FOkoM`Mf)8ys3J@vPMiLQ&R)}m2V<l5<`idr$Z19LxtwA0<`3K@f+<$1m&c9@
zQhc~^6dKA%l$vZ4qpp<93keeu(SWtOu^%B{+y#}(L$9vnh)q3Qpc2>PPOV)!t;~8Q
z+~nC;vG@hIBAJZ+Y^s)DFm}$Bn(7T%-6C{l2>sHVcDPU3!wzOrlRMOB$arUX-GeV)
z)bHYJyj3~q3(H%>naM>y&o^=c`2tE%Eh^Wwgri2;z}4cV#f!qv9b-UPE>F=@fkCo{
zO&yklO8dyPID*Yqk(;`ihe$`<z^$2eB)WXv(h_IZ;&xqrO+~(xHE#1Y`|_vC0;kNx
zGQq?#qaa(ZHdxQrC*hx6Hsa#r$v&=;+t<KH?<U?beRolwKya73HIlz9U7K8(+wYW?
zUJhT&+I4=*NW9JUO<|L}nc+CVEmj~+I3OMQ{tP5Mmz^WWtEa<%r1utv7P`J!#Bh1@
z)%68WbQ^EWqb0phKl^D=!VNQ0;(tZsH{+Up?3{vVvO%@b1>|f}_0q=A&z=2jXm-Lx
z_f;u^3f}&vrtv5_-K<zgne$;lGwtil3+<fuEQ(XEQyEPD4bkNZ)Ow}MlspndDox>#
z3b=}SSE6>e)g}2R@)i?cu4Mhy(ZUZmkIZBhgXqf5FBEo@PNRWZ(}xQuk}U?t?Un7O
zS|+pa+ws#HKRgU0wO!_9kWA$E9Su+EhMa3aqmVGW#T^7rsv#Om?-jDTJr-MHcsrbG
z7FQ3tqi>Bt`u+zei@qU$`MgIpf8&bH$tV1-^oCb0iiwrH3p6;hBDgOsFj#_k<5_l*
zl~=F2k6=P>e|ON1scPp4&Jkv{>?n=RV*Pl^*N}P*A*khJXU@&&OMXpJF3)>wbxYdp
zcEFiqPG_+TQhk)%X10o6KP#h|cN;|u+1T<(!Xg>ZOg!F%Ys61>$FQm22sf{Rf_%l_
zm*42Q+v~_cbsXz7A*Xa@f!m0_r|*xr(6x{%s*w_a|8w(+J*Uharj)b?l_~L@nNwKV
zNzH8@zBVk@bm^$1f2OJX#afKhG;3nc@k#g_^7G7Kl0UMjI1uO;_x06XPyjmlgFzuN
zXbK}l>0{-u1H<3XE1ek{pzyTVasQaI^jNTQ_`Rs5x7O_oQgaVf=_zjrOG66Sqp(||
z^*>f!P4loATp-?Lu9XM*Bhav*zMd1XP(Sr&kI5V&cW9Adym9QJEABz{kHU!-!0WT~
zk!0T9RHfhPGpKmzozMJ@#}dkmIo#gLaM^=a$z;+~>yfY<_UU}Ae0CzD(5pc2h;VP~
zx6@F5#F)gzagV^^28#sN3}|mTi-SKGh`kTZ-Evv3fuV(dp8v<9`2KO=+Y2UtSWquF
zI=ORu`evTo&P;G+IQi+3l-Ay_!zlm}U*`<ZQETA({<s*%M^!(Yu}@;eOQ5!Hg;d0k
zaT$ITGUi#%jB}%-ztp-r{Y}u6`w9eCLu+!Me6EuD==yLFihS?Sm$%j@munlIgpKA_
z?08S}{Jbc3_QB95gfc6RK09yLO!>U1<`SstH<_CtPmpXG5m|`e{$(%z$=_^gjhTM(
zYD8}MdEN7ceg7xM=L<pf(eUGcaYEid*~6*h?@d1!S&c?5Z2r!56ek2sE-dr6eD&0J
zf&ak4^t_)MiNjH#KkN^`_J3lxRFcs-bW4jiJt{nyHBHy~qnqv7aivlHvMxRQe#~D{
zuIY<_2ug#@)Tqm0z9G=#yZCQO`^s0ye-QFL>7Zzx@XkX6)WXa@^jv!EIgue$3-v^e
zoVvNlH+WT*S2cDCWSQ08pL9VWF4OKirV#qDM+V<9yT!xoI#qn(hXH?R&P6Uuy#~d^
zF*{vZP)hOy{&3-lYFDtN`dkT>o)f7bHTp9bKFfI4Fe==ga)TgkxPQ1)GwJK#;&6y1
zD!k)U8~9ef`95Eh>+i(gOWe-zHWPaeJXs6U#$v!gFSu=75$L2nInrqC#Cj4U44Wd4
zZ~$JAL$T$&o+J9w=Ko{wy`!4yw!YDXDjh`x=|~q45R?ugNH2m2(osNqQF=`frKl7`
zuK@%BrGpe{0V&d}^b)$#5<o~Gklc;WIp=%td;h)f_{P2C%gB%nv-e&zbIrEqnrr@+
zt`ORSkmJaWqhB{qhww68!G+}P1A@4_vHV?_2k=d;t2eJ>R2lc!FZYv_IT#!e=_HKf
z%ZhETSWiWNp-|_I%OL~cLMt?S^7l$Bc$|mAe`s7Iu&HsERC@lY;ruqdEBN%O>mP3_
zQs;bg>k7bMN>k+aRnfErev5GT`Vxtv33*$42oa{dggol<<LIlUHG=}&`@E_E8!)3(
zV7^121C^rGfTc`C`$n5;Y>Eo5`gtZfKtS2@H&1gP44&mTC4{$0Y<+ma+xO&$kv2E2
z&-?oaJn}8qBr~t{JpU8;$BAcsl*skNxLu?n1<L{+t_y5Oyt#dM$PkS#wZ)ow<E4em
z83Vv)Y*#@6@{}GJ71ph%qr=yX7yP_@cGMh>T}Q(<5uFbdu}8mJzOjd@ufpXGy@DNA
zFz#3HR>^<2CN8ghfr1GpdWu(9kn@Kf7oZT<qY~T_b$6=6GZ}iW={0H?i7CO7q5I2*
zq+N291kY{g*ShjM?_th9I2+PtVkK3%;%bt|NPoi3*>OnpvrWPygx&gl-snz4wv(p<
z=NY?TFhSE;C|J<E{xvPGKzHQBkH`uQzff}Q;17<r2eTzg^>j@+^6!)}jjvcw@r?e*
ztOFa%2(bH~*&@>(mkHGmE(l{#h3hU?3;u;usJQs4najG$4>GZ~86w-NGKrh-NI}$g
zbc<l7yg>(;@+%8=S-xAUHnYGcnfxx*K(XRHi%}efGDwzH-ROzaPSpDlVRN?u63mz4
z(NX-5US+w{bZRcQ`{9vsj#Ky7DUdOm98R}gawamL3Fw6E#KoDcU5#1il>6xEQ-Sl}
zSAd$(Df{-3-J$-dFhn;5kC#ji(yrxhd;jOuN8sU0$)_I;y>b*=oX+TLofXNRJjk0G
zto-FTZ^~cyA~$xouCq#+k~&JmEkbVg4vMUe(Y>`kk(R47a}yoCJL6%B4P~r`%Ok!9
z%C6bXu@R_oI=<ca#=;wnnBQhGX!!+`{z?2v_9Y(z>}04H1~Sd+@Sdm|WFR!s2T&hn
zB%E(~*~B^uuwDe%*M6fhH|D~4y>8CZ(G{KdLEA@PJQU|ZeNYMZuYaesr!?e?dEs%E
zwTr03yDceZ6pF3}7GhMIM<HAm{uoExUbx%BS82~(wvO&>)#(OTo5@|a_CA%UN+%0Q
zooPN~{UOqQv})l?QHKaG$`#&{RRw}Ry$WQ!=z1gXe<8%6l(6iiWS<^dzHMa1@`#VO
zR=z()xD#Jf*LWSDy~#5YEZ3L+qH2+liI5!iCeO3Ir)+CzP@9}xO;Mzw5syB#j8jR^
zwiA-^)y^Xg%@GCJGCM36fqKXy=#pQw9}3b|FS%!Is<YHz4j?(RdHB7lq<BJ!)(`bS
z#Y14yH?Dsy%Fi3Na5W1f&(IV1%aM&&(<DvS-lpR{ZtXH=Jw|xUydCB9#^@wgaMkmB
zYpI1&vYDvoW^e0w+8yMj4H}jn*We=uM?T;7;79Y74R<IhMKlgb`k{BqS94<k-jiWf
z5I@_I_mM*$HY&-zOnEJ80oilz%ARXchdWj2cyKLTiISAKn@e`FJSei$HD-vJntD+m
zV!bX7yiY6pQ9b7M%EG~PcJA=79jD!%xP_LPpD81+F}TVwocN?0dwX9QdG)hxrxXwQ
zw5~Z?<RE*tF+-9nq*ee<REiO22v{t1V2h=$;sAZ^wNEnSuw#KXu4)k5-1uN9o;HZD
zIGGth9iGIsLQOtCZxa@N5e4*;LYvROe=x(aIudBm&?PipB*|h<JW{N+HGs!)<~;}=
zphT(~qAMLuVfqJM?P!syh}f3V0W4mV#Yv%>qAJ9SlTt5a_2azmiFWlnwPY14_qnL*
zux;j)9gbll<@CUoIAmcMIr{nb!@E1Pne~3Cu1^cWRV81gn}mP=IMA#{iq^h>2+K_h
z#O?a7SxVwOjWgtG0Dosg+$<n5LTO3t;a6{6c5nYpLE`y6*Hjxh(g2!|9aw9@c6YX|
zZ0Yd(kk%(b0+MZI-<!a0+tLYT=j2s|DBHXB)Xxv&h|Q|1=$d99*ZdmXo6K{YzWe@;
ztJubI>ZjY=dX%@-EJhs#0(U-fabCqxPBZbj`!FJ$R3C%Ih%03I9vH(S>j)-bTV`x~
zixbX?s_7H8Z4UzPh62V2HCP>$ZtY5y?~}_SJtOQa1g6<Lg#Ow`{r;VkY-UEz%$#6#
zP30OX<AmT_u5MuuC*6tSQ8k8-rm?20KZ7eWm|?anu@*O6n%}anyizAsQK2Wf?niH*
z=$rQRx$R_*5+;60Xlvs4*4F2(a;2{!%ER6{drB59o4cP+vk%Lect<I9$Ef?e(Ke)|
zeX%9Y#6PP(FkO*mRkgR`ty$`c&W!1$B=Fry;?QwlRz@(*ikvOQ?5k2J>i(%r%t&zK
z-y1CN<)^h)QBXr^G`CHz`IQeaiR&8EPq^kOC*eYQEuN3h+vkjz9L!RI>m)mQc76cw
zJ8Zp;_lV}AW>SWp3-h?y(OSS{-Ly8UoZYU?X)wUt?Fv3>-Q_jxm{w;t&lIood+#$S
z=*?k1^gUiN)`ptB>t{_oa#j}ivgg(1x?SDF<ZX(d)AAogo&R_V(qc@1>lvEeY`Ga(
zLc3{fR%d8YQklZcG-G|=$VLal6GX9LHKDr<VPWhjqNkyGI|(8<FSqq2zBkNHzzEEE
zwcs9Ri`jxlgI`d*Kp8cP70=fj?-^{22ka?<j9@-Ihti(*_>G@CYWFh+p5aC=%}t8f
ze|P6`%e*u2O9U!9FjE&3I9Dg`=p*T^0aofDR|oq4db8K^vTk|XAn&bklv&3|<+<U)
zXC%Ia&&62F(GhJ26-%g0RT14Q)N~c_oRh`L#yc$=r#F;Vmx3pCHv>L8hs^BYndM8d
zlC5WNU1bigTcjO!GL5v(?;~+o<+c(@+F2H42T3`+yjC+KV~0QY_CD&IPe6kO<V@6D
zRe7L~>=A_v?U4`V`n{|&%ukI=-f>Z_&KBsNUvI0X4Cm_%-K=&q>%=YxsTb?Nl{Zv@
zwBFBMq*qoo7N)zYoH&1ac5K$T>#p;3@~-;y-7w~5?$3$%yFS9?1x()eU)*1dO7SLC
z+*&rP-|6x6&zF;KU%8to7qzkRJAJ^qi_v+Y$V<^Hbn@Okqa{rSWJQT+(=6z$|FuRZ
zW-CsuqH?uZ`D`v5QGYh{&|eGaf%IHQKZAZh<kru(2UQ&?o=(M~IJN6@lsBFaIflGx
zceCZ?OIPeeHycJq-Ib_FAMT(vV=-^l7~b3VB=gW;<#6zP#=NJR*3dJqKl9S^7~@14
zPtQmkPrpPO|439dgnHVAPbJg@<fLx@-M4f=G+9d{Q|ZB^i_;y$OKsu5JZe7#XwGQ*
z5D&z6QSN_BzGa-8F}?upvPjQ9lSIj&X^&+EZLn>UbQL(e*2*m|kD+@0Q$;T!uNu27
zZ5Y|SWK4>DAYsvEo8We%l}R!ThT`Ov(E2Pz1KaIXUEvwrC3LDDE9AFpwnmw)Z(jQP
z2AfPiHTcFsCmU{2T;~DT`TFkc%4f!_I?=&Xjg5u3<5dKBjfItO=J+W0!&{Z>t+WQx
zYseZx#rDF^3PW6vsTysE#v4LLZ%m@@Z+}PY)hpJluSuC3T9eN^eX(}lC^SUNum)5v
zkcF5nBk$Ypzr(gWr@lC=s#4N=1pDJuE{7^o@R#mhQhrbGK!cOm&k~yphZD8cCOmB6
z(TaIlp-1C4d;cnQ$;v93_OgR#FnuysrzGXwx&>`WYT8_p5z@Mm6KEtFNg^#lWB13=
z!KIBjELA;z={wNQy_bcOngNbmaconU3&Vf12!BW3A~H+~^3`RrxjKA<lHzUR>DQZi
zrw{HpxSGkG87^Yhwq@Mvqh{7)Qgo>TLdIh(eQ?6qSyavAHqtQg{5s7mUM)bPTPqJt
zkei~}e3Z9s^xp?&_$x{zy)%YoMD5&3+?iK#c6gFq*ytdzk|%9rC?3C+pnhUaGwyKh
z=W^(AYi0{p01tz#hzZY44Lh!S;y<)~ThR-7{`!|yRdg|;xS;9D$cPYyL^GllK{d{N
z6K!fDs4dC!#*tXyImE-@<GuY``S%XO^8Z}LKL2+ufVhWzrD6<O7?0W;mBM?TOe?`2
z+zIpj*4|;m9FPWvFK|KVd^V2rL_$CeY(1lXh4Y5_x$?*&qjn4bb)LwdzyS<B-_Q(t
z8v1MCZLXO#_u}zPrEG(FZ?|`cyyND#Qq=4g`yVyDaw@e*A9g0)rE7Wed;MWmRxw-S
zOG>TXKod}L;*jlGMCk85+oeFra1&iq?79E9C`S20UJ8WxM+*!RBi)=M7V<P2IAai;
z9c~FLU>O!LpIQ&C1ByVUICh44Jq`{hK}%mdtX}O%K2F!gl%L`XC=Ao@!e?`c4r5GC
z415lUMUxk+-c>I#T~?%j?$~2;Y!Gw5`Y4gtXH<Pdd3`V>XfE?)0JHt6TF!KsXluCT
z;dK@vWy>+aucFvGhpE&Gdb4gU>ciU^04RFYd2J~fp{I3KNju~D%;wqj{O7(95n=vm
zND|riT&v+c`R4Fp_?Zk<j^G*RGm%hE!uCPlr2Mr4wdECqPu@?SmiY4t{p79}n5(~i
zpswM7Az&Q)0Zg#G(CFJ;3&pgoi{%d<>D$NG5Zz<{umIXw3}BkPeCNZL8aQjQUwygA
zyJ!!Mcj=Kc!xd|3UGOzA+H+Cdu8&yo&{E${kF=zJ|Fw<MxxO8nr2!TA30UlO*Vsk_
zR)9~WC9V)IwItOVFLNGqTJYyz?YV68j$vO*RNLtDcN#qvxy;$g)8suu@?!m9Eh5v4
zIk&@7ar|a13Z@#P)hqCsdRpFsLwmWucP*I(W}fu*7bIanmW2QIQ?rQ&Sq(c@Sx4mC
zW>U&+c;@ovbJDY^?|M0qB<4**JwjA#-cfVw@{DBr;0)0{78<sj->-`1%{JuTwp<O_
z!pK<CS^iw+^;Z-IO<bO#LHtBe%zT!}=STUgakUmDN1fIB-A^p+B$sb9=6BPf?0=3N
z)i682Mx7uJP%pO1F;%#f*Z~_jth8UhEspsmvM1pQfIeyzTJ*@G5l#em-$1x2EU7{t
zT~`+R9f~l`iZ-~ZQ=Kf>5bSkU_aaen6Jo;h%ME-hGre|FdqoLvgdB6*+5ZmZiOEra
z&_r)XO>pbJB1n_s2@)6$pvFhG4qfjjFcDTeX++qrI|TImJ$L7H2=JlU<~X~=i4`%d
zw7GszI(<)^kMh)$d`~xjGoe7~0}Ydp^qoVtiM|}!gG6>Za6I)XmLT512&xH>fD*@4
z<!&%o>7d9xz#fQePUFi0ojdC_;kZv@pT(#a85Gi_U_&=Ji1c|WMUQW|ToV3HrI1#*
z-;`jv;Bm^UrPz*Ls{2EN)O(X`s`Aq6`^S8NB}2rc5*sdmVRS4no(}d+|D#sY_|yEZ
zg-EwHBfim(*p92W_*=K*pN6>Fx2=*V?nB7z8BdlX1cEkql0}81apBtS=cV&cU7=Bv
zRRKraGLGx<GACHkfk5#x0HdKhYI=PEchJNiDf~g|NA)YbC^lPZDEBdF06;)$em9%@
zNQx|rrWt;ndHb`C!$K%T`F9-q2inV$G0K@v;gtfbQLGJ#f@g%AybP0E+OVLUeTeoL
z1C;KH-3ycVqq#>Amc)~6wX1cyQXxN*5u|lLJ_J3!O{rk+ck^xa3%n=D<3}XBJj-;%
zFZVy&x5qzScRUl*nv~!C=2ikQT7kNN;`jah>b(;+x}2GbAyOYww2G$4Kw^Oqa{(n(
z7IE{L`tN?|FhjeO=;*hQG-mS+5E6*8iNyF{B}ju6x%HwmaXir!(I)ngVWEpv`tKK_
zhS)Y*Eu*&Ecmu744#HI#CYZ%G$K~%4&EWyja9gG@O3SU+Z7?~Z@F*x}QZ{XJ2X?8j
z`aYp0I_I?KilXn^77}3$C1+G;UI?%ATe1=&WJNR*Lg}Rvab8Z;dY=SuLyY3%2>F>j
z7T8JnYV@m3;KkuXREwheAI%2V1~P1%Gy`k`^Pb?Y19ap>U^@t&V7P}Dl5&d)@5x7L
z6P8@7PypxE5(PTN-_nbAJap(B(GiZOX!Pfcq??O4{Z1Trh0)*9Q8AWR>yeZuBLLCb
zs=A7V`?I4ZkOpWn+sJ~N=-v<k^K0~FQhxj%9DV})T&@$@44)b3mmUmog~@uceqE*1
zhMgwvw>kV~L)s(DCiJH>iI&Ll4lw2*a>Hqx5|V&4<~_&9M43g(Eim3#yEAyT6gU5P
zmc~5YS`WHH<S=IcoQdJh2~~{au|uaC!n^u>gMRqX`?_+VXP4mqd>D9ie53=SmBAlj
z3<fq%h)3QoIrKOi{T9EBWO^uCYNI<(EizD+2^aN_$%OU|&vfFd0wfdQ?ZE|IYOz-q
z!cbh4=M~Vk<!i&-T3hbjginP!(D9f_Sl6~e_#xCl-y}Q-5qT9)8gQkOrC};Bj8{u?
z0Bd8KK|j2dy!po}nh5u1<{LZUBkcRd_jQ%t1@OZkVP(zinxDII1#Q_7SzhWT9{1AM
zG8*D~46?k`d?kQWVq?kvO8!p-U$(BXzqUqnCbUp35C1G5)OLWova(?dei4yS(K}pE
zCitBheQ9)-V2-TArn}t~QiK(}yiO2E;#bqP(bAp90r6J++y_B5*`}FAZ;DcZ$hkz`
z5^7>*!uZpbz?L!k3Zt|()jiA`weR1|5X&>IlJ4&X|JWEvO*auWFkguxPb#3j4hT0;
z_@t)``oo{^6;L`9Mj>H>G(%7zYM$Q=f0WAKiI$OPkd{q$NJ_){MK`qOxuvTrV8ys8
zahMzBp%VLx%)RGX-#e|2`lQt7W#%<>cW^pLKigQ<GffzCG*<b86u0@N=nqk<7z#Q%
zl4I`3vCN)Mi7<IDCDE&R0HutSj0;~x1znAKaU`sYmU5MT{s|4aI<2N#+6;a|KCd%C
z`iI^tn-qY34g_Y{i^hu|&9|a%OWgo&NW7%!9rI$HPQc}R4|IxHS&X}ER}PGgxEsF!
z4!F5~?VzZWrlXEbd5S@r218GZueW{L#_Rlyn4U2HKrOs$Pc0l0cD-*`hMG;wm_+xv
zFq#$l1#zd_Ov@<#Iutuxf@pW=Xlm&_?=E&XPJ6yR<~F0gwz-P1n;AYrWVp?M+<K1#
znFB~*W$vHMU=yEIK7&A5N{P%f*Kku%4^h;5ou|MzcqlRUhD4ICnCwf+o?8wsZwNw{
zr|XS&11lvgIK<AZ&F-#GdJe>~_h-lNpkg}*JJmI}zRm)Gp&!EH2}PI7m;`Es6PfVJ
zv}Z(E7G8hdJ5OP}$;C^ngdukhpfg=eK)1EFLX>C0X^3^8bw%kHg(XwU6X~^-!Mj%y
z3ihO(7(t+TBKW|y;<`9i9wHmDCit(ur8tck1?C%+OVUg!?rN1bnWC;Q+ba0X*G(x2
z^_>QuB)nQyAK|NnF@DUPrQ?F&&t6OzaTa;+)^F`JUM;@uO)NnenKZv7ba|YUFK1Zz
z%~8Ceg;!gbqpuc=NW>HH!-;JbWfZRSNl<^x3!HGj_)m}669@Shc9kg&TG*QsoCvbd
zqJM<@$uu+46|`Ul)*}Y{1SZ7q2~@AN_E)~8i^BZdD`Ct0$mhtgrd0quS&g%+Y0WfY
zQT_cooF+L_O1?Ip*BSZ*EG~Hs4Cm_p#hGDm6HVbim*AFP-z@@kp@$C|pAl#Z>giiR
z^+xP4&o99F4&K3r>u(7ltwN1fUM-jJ_j3XQa;hN(a$5TG?^YLe#s-4rzAkmOiAc+4
za05`{EN<O|c6$IyynS+Ne~AMW8>>4W=$HNJP!m=#1fYn+qyhji@`Bb0UH@8@Bj!r)
zI^CZd;ZdKw&bpm7BY3qkvi5CY1?_fcsk}#`!601`QWJym#&Tjm^5c(<+Nta%?EQp&
zXYD9}mH{C07(+?`&pb_G1W7u|d+8+Nt~F5)kpW51-t^0d;dU|CsK+Hzo>k0((=_HE
ze&Q=C)Rn)Cia2r*C{W`8AeQsXRS3Fn09cwogg?j9k7$C4ObCynD+Ns^WhcP+)CT6%
zY6aQtyIVe=bTY5fjV@Gq@{4dEEO%Ek+y<|<efrMT1%U$V;h8xA*r9lX=v|B>VfSK`
zchQ)ApU0n%KC_1#+FH#}2#t!^3VAW2oqC<qb!~bGDrKNF8a5KPOCkauKkA7v3@WE&
zwmQxwtezBJPiJBPN$$B$0PtfK0G5v3W38oxHX@>+{+GW#WVlWYAnO{)uA*X-(HXFU
z?lX@)E8r@lrvH36-D(O<xCz<-eTm-toeG8vbZ=Ntx$zu469SIFx*)~EnTVq#4yD0g
z!sVf%j#ytlErzD%9g<?vi@}VU3BifL97r*SG7=K28N?$NsY0&GkT&GToYV1i7vY)F
z?X;g<Z;%5;dOKzvt_D%bGe%Xi7G#6>T`V%)Y|-6BY&W_=rObBXo8UCn0aC+d#~WuE
zSWBOps|_$GmAgOeGLCPq5(WaUePv+x!!g6NZ@<Ild|`I*+5`)Bw}?{j`Trt%ru6!4
zPYlrymD;pf#}W8=3#Pl9u=xB8F53i(TEj1D%PI=>-Hne9WsHzxTDjqZqL><Y6fMcL
zy{?(2N8M<$5A?Iua&~^R-`2B0mNm5^!AZ9F`<^*jJ9!M9p2LY(W3SNQO_#%E^vlFY
zZ_i5xBj)<_^qm(q+?7IV=hu=x-QM?f2y|)e9R-Qwhu;rQgw<`PWM+-A0T);rdc4#=
z@^Cp^LG|nDm$YiQTW6UA|MfL|w)x3)fPV?#4$}`QU6iM6gr#K*Po%z{pf`YFGUkd0
zqt1a_giIc0l1H4j9LNC#WW7w3oQ~vQD)ii;Jge)n3rl4e0DxpJX5b$!5RwF92qvP6
zf;yZ#4+WYG35zS#gz@@j{NJx2Fic0JfqrT7w;sPvyY#CGi0fYf;JMQOks{db?+yaC
zWLCsk6d@E18ELOq$&mWbjG`JlWvrqY7#d>16$1R9064go-v8)T65!`R&<c<;$XZ9&
z*jvQwi3iT_Sv7zOX+y%1{+mP_R^aRiXsR=s84%WI=$jxA2dPrJC;*ba2em^?R$t-M
zg30y$=Xqp{TqdBa*Z=%Wuwi8J7lP(R=6uCg3mEz0O%}-r`qxL^BW_@oBE(7wsSx91
zT$kh=h@d>=PNOq|^g-&R!5<y?3hcvr9;Y)s-gxj>Nvv+gOT<VSEq^PGE-#I)tBt&U
z^<D;U4s)M*A|&B-RtdqfqT?xwF1VNq=_`Q=L`*4JDYws|E=tX1Yqd!ZVzzfakjE2|
zU(9oJ+*a`N<~<{im%Mt3@rLJa1dCp0p+`h#updzp$N}IHpB14|G0jiN*??=bkM7&D
zeqfRY^Q`k18FS{HTJ4{}#&0+W4UmUvgp}8V1d=l{4%mp@cn_8_i0TUhl#5Ok3FokJ
zqfefOgHdg_WJMj!Naj`7+WC>#mx(+jkB&*KDN6<Ax%uR;$*`wc`Q6u+qRt{yXxVw4
z<mSI3S!D{?lo<(Zdcvt5kph#<b>ZU;kQg|}-0%K3#LF9dWg@cd3rO)BVtyCT(1AJ<
ziy;<&87Z6ThV-yo0#B<`{y-<63>Z0f0|f$IYIeevL7)$Dj^V{u@3FenF2tJuo|1ni
z=6~0}6KxwH6iGShU9dlxKrQ0AGHnK+sO*V0FJAJJ^qM4@iwCummXdprFOm<qFQ4tY
zNONWkMw}pUF|OuxS>$YEwuIXUdE=-zBmop&m(k2#bY922J%8>i@gm+qsj7X-!9%2R
z9%z;_OH;8Be)GWy{mt}(7+<-?b-incdD5+`Z%>7eRWV(J+ifWMP`Z(=uSeR}0i0Y%
zWbkv<qQktU)wW@LcC5{L8eMiPAk8r|2%3A^A>j~a?3GIPX-YP!-u>a>xP|&RQwqZJ
z8#{7y);aD?bXXGf;0|goJW#Y*{mUm9-QZe#T<0D4;UQc_XDZOL;IZSs$;JlN4G$uE
z!o8uQ2tjm`++W&fj*}WnPUKezO-dG+IJt_^KHsAb$NZlPzkG7EGM*HzS{Z&!qifk`
zH#KPM4ExoJwb{Ug7NNlAEp^ttR4C$tMPv=+myZ(l{WHu=oA>?~&cnQMj1DXVeRn0|
z&sN9jdCk$3zSq10G7uqW-b4`)R0nCvgg)qG8u;1C^swu&`doQ*&9(Mpc8C)nU6=~V
zw~#VIPG{<G_D-f}56n(Req%bki{bd}Xfb^7ufdSwf=wMYo#?kGYO{}QQS+xJ<zGyd
z7|J$O1P*nL?6Twh17n{l{TcGNv~KGP`&Ev$ImUzv{6lWve<Ame_XW8%`(JDh^Aaz}
z^|`aZZZ(XFrh8Hqxhr5y9O&S@!EXz$WWjPDc@izc$||Nve?7YqJ><ONJ3H#4R0|`=
zJ)c{$a-0MF>-rNDC7;t4bd_6h%Po6H$2q^gvhCy$ta+V>R`xj@=8oH*|MuS~eeWNb
z*cI|C8`bbO?EU$O)qdL!3a7tFTpQ%xlEkz~uG!M=NEqo2`L&5f)j&2^P1`n91yHF{
zF?rPtx#cgU(GY>@I)0p4NZXk|a%SeB>zeiUQq1h4a*i!dxEek+F>@@N{|VhI+H5Us
zJ{V<kko(f==om-h#UjCP{eqA_6s6zB#4#9ltF*?werGdf{+o4g)W`6z#uZ|y@}14T
z|F;O{)f>H$Plo=ibsWqv_?1XTaSp|mf9GC{2`Ceq$uEO1nUSc-TJ!y6A$cNh#$l;4
ze<t;1?%?USMXKbzFw@IcT;RxXQ-_|=>Pkq(+V`POnfN0%gMvfRMqfbD>es;&!%-%O
zxru@NC5q#hQ8zPUPDZlfyT9KCws7F`van0BC!lffzuIQ>SKIEw&%gVueb4QLz|VU!
z<FG#~PDXlhe0t9sfWSp~S2rYq76kMV;0#!n%JSC$RwPUG3<#0pnqhH?j~FZ8I+d@J
zFXEjwhwVSfbLMEzUm6f_86+@2Av??ZMHZmtvE<PkrQghC!eIBK>_+z)>qql<XKmqh
zYsb5VAprre?3h3{T-_D8pt|zp?X46AXzyWOvVxN@!K077WXWfxsWm-~ZsBNVPfh91
zvz@4ys}SjRVqKKty)RRfu7S&;(23xXV>%^V)bAm3_<6AOv4Ou*Z8fBRO~nS-%{eM9
zsC$mLv^~5!oDN$0BzK5YyU7r)pwxamy0bYi^r`5&aC&P9Gk3W=x0l^9PP!Uiaid!t
zcM09@MgZte)+Gc#WyT6tQlW{TQMa>)E0ZqLW@&tV-Z*u*OtRALZ-9J&;F-D5Ig10(
zY=2bDLGI7<CLzv(uo~F!^K#F-8>;@Hb*8slPDZ{|arVU$7Lw@1TQH#HDwoWby09M?
z3b<Ey1%3{Q_S<A#6b|@1S#hC*$>$a=fX#dG9Wfp~`xWPWwwzJAJ!KIy+pOrlz2!^H
z);k#rd3Csx=nfg?b%yH?Zm51#NK^HFl?BQ5o3DNPIq}BhZt?B9khCF8huO^WCevg*
zy)?;3V*!e-pCQxi4N%QFE5wI4yZ;D{R4QHKy2=-vznzli`E$7yb}7JY$t%x|Gad#h
z9;?dkx3M0AKW%d4{&d}5Z$>NK+Y-It`Hv`-w*)1e0px6JD#Xcw;L%U)Y@4MW(qKJi
z3s<)}3#OL_d5V<6i^XlJMjw>s^bH^@oU_*A(nL5}*4s<n=1jIQ&4EUB=a9`wUppn;
zqn3w_*S`TLHZj9>1U(Ncv2|6V=GcI?3UX_muVih-fQWP}Ku=Fbv^i?OLFzXmCOkh~
zL7B&%jI_XaOS+rGV$Q`6_uDK0TEqT|xcPrX-1y*W!pen1iLp7)pex&1iVYkYe=C{c
zDlQ6MBxRFpKW43l<7xx(=P{VsqG+xsW|M$9OF6d+%<hk#MF*jViEH5XwdpD9a_Mw)
zG5e^!f>|peRQS%Ob)&DtnrhJ!l}<I>RJZcHb!W4^(YGaujzD=vg$~TmTuqH8geqRR
z!BBql-Ut(nm@ptA<GrX;6Fa3plm3>%L45zs5&osyR~mjEF^tbkJONzFW3f(r`j04+
zW9bbY_R6$kMcf3*96Z@w`D42*%TEA??ANJ@rBWulx5Y6gxaJ%vErn=idxyTmzD%3q
z65Y*kQPQ0}-*vmO>%HYB58by)>-%wG@NSlYZk5!~AamVtIPdKsK2GbjvS7~zo^IFm
zZ$aylqb?jiw}*N4N)z|Xe*$jr5O90S|K08V^&@+m|F}Kv3qK;;LXUI2!6_Hm##^Fg
zO8<4dOKdL?ac8EEuPfpue=w-rG2a8)A5C{i7qh3r7-f$mxKdS&elA)2ZXZ|)b2&Zy
zn&usyn`R{cF7O|ZCp&XIPPM%WN_{?e@U^c6MmNP^w{mB$<})A>$-`=20!Mlk5+|Mr
ztXe6?MuN7IQ4K+-08jn92-QGv5%^!s-9>OAaQKuj;xZxh*i*lv6R*E{;YWDyIEy_y
zE8A6L%EHjM2eR)k%eli=i<-odUuWuc!j-{FYVW~*j((O!x_q1&y%F_~+)vQl$W=8A
z;;Kx|5dI$#pn_&rhu$Ju1OO;TMWqJg#I}w@(%@wlR9Q>5LUE~6eZIc9i(n!xMj=tE
zO24NmUtNOp8$sUV#`zUo9&QDazJ$qcC3iZS4dNDSrb`Hl0_K1GMEuhJk|SZLt0JsF
zD!su5Hc<fN9;NE9+zH0ee{~srkC}|=fHFWi8SF@aHRbE{lZ5(Ay1|Wh$>^CO>C2KO
za8(~jrsnm(!NVnBrSo#-8wUwK;&h?u<~LbR6Z$EInS=7bv5qfrcm*J@<X_}9_dZkd
z!T8Us{r6DKH~6n+UHfmLS`wY4Z`=Q_fcF3U<Bo9!NmnvHdFXeaS8MG3MFQp`sbdt+
za#J1*3z`)dH}bo2@y+f3e3S9mkBIl^wN-cz)FOP_ds~`l3Tz6p<15esR<d#bUdf&g
zc+F5GOpKQgUDJX|$pCpGUK)<owuz5yM8JyoU(1m#I}Kt^@D=2BgdSoBArKQ2W@H<D
zjN_h>fave?-qWz(Oe4*rv?O&Sx<wk-cPtB%f&VfJU_x8-bHwd*x-I!4&Jji7Ek!|j
zFy7x3K*r;rRcHbD5C4rH%F?(-S}gtLoh4WqP61718?}?>Z@(=p4XPN#<C!I_k@Dg3
z4AK;HF}7`_KnQP&aG$|*7ANM^RJ%(PbCN1On3pO*2#vpJhw=VhMuYzH0D4mz<O;Q;
zn)4g1Yx!Immd^@(S6ucJa{1>q65QtGFe!(C3+C{qG&f4XoY#1OyqrWm++olv1xRuM
zc%c7Ea!Fl0#j~uY>?bRfhYBY;GzovSIlshfPP%4AJFnx8u8<LwpW5EGdJWtb^KYJ#
z{rJmM-!K`P&4vmz=E?ElH#wX~;oXsVP|xL6GN3sdUZ(~grN<WzqzfYda~l*`hQFOm
zJv%+Dh6{jFrKUZLu4}&I)r#o&H$lX<#XBULMk<B!Naf3-#Qf&g7TR09T1EmFxqyqr
zl{48BFjvW(8`y)JpZ2q(#<S)QM#+>LTCfS{f3rOUYy#4BdC;O2D7WkxejcM0d7uvU
z1i|yq%HIN$Pq_+cX+#fJ?1*9$lrZtWY-ZU1iZoamH0g&v8X~pU_8|^ozx-#DxPm-|
zSd4hm7kvaI94>x?3Xp<4eXj^W0ojS7t@BOrl4{b;`r6^9L`4kYc+c>9kTRH-sC8y}
zDwS93FEt;4+x{(M|5G2lE<FhScP+sG^jQDr`Tu|BK#Alvdq5jm*e+xj_@SqkM5W|P
zK=uc^j))3(h3X+f5siq$7$1FmZnPiL2q}SlinKroAx21g*vQ=6pUpT>P9=gM<XJRq
z;%^RXT|ZCBP{|NuB|cQp5y1zC<}{lKJ?bf76mdBb2e}FHCZJxCa|%VgG`CmV_Ll$~
z0E|j(Yr{nvCoIeE`22U8QzJ>*qs@lZa8a7LrVK$YK{-3v)EAK^{pAcimw%3OT1VjQ
zon%)_H3fu$&wv`K`YxW6RW@9(IPF)KzxdZzELtpT-4Nj6f7Or-VG$iwY{J?1X?%G?
zX~dnh95i`t64L55|I)U$a`-1=_l<9=Wb68jhSOQD+)pCtrlEznp*r|p)07N}rP;J2
z<0z4M+uq=t)!B39yVX3aoOn+G^T2CUGgltMG91sO@vMNAg7G?ofF7X5Yf>SJTM_f5
z##B6!(BX%;+*Hml*ww}(Wf8YIrYDJWVNT0;AZcZ7Nn6Z6j#8dCm5LE9CKUeG?^d=?
zH3W_9-BF5n`KA#bQFBCrR&n7(AChWZ{JN95d5P^9f@_x!<z|Rh;${->bwVuRpBcNk
z&|eZa{w*>fAxNw;E<n7AF@WwTd0Cg*)Q)`23nMX7>-J8%DynGmBIQ23zLUqmbXh0%
z+qeV6&!~p)RC6W|aS8Phk2$fI+Zc!D`pr7H;dI?C3j*CE-QX^j_mE#=O`zNbR|UAs
zw|7?Oj)piwkiLh-#hiK!u^%$|*4e=3!t?=>3Op{DVesZJ+d~I$<!%YoKtJqY0y`sS
z-o2c|kSnRbZ#rc`-p}v%>Dr$;2+8*~AC;IHe)n=)eQU?m;Kj?9LN%uSx5^(I%NJD<
zU5#z1n{G=c*kE7>6<B{f%?o>1eKy^O*$9Jc)J_vOw8lu?1)vHS=9hR7RG)s&?>b>v
z7?5l8OUlA-OSET6TR1eyN~m-)!H$P~BaGmXy~{*2JP_tt_fQ4)EgO)$rCS$H0mgfj
zU^?3H*i2ZbCF&!DzuTxxGup)WC}l-0SWG0#)$A!QbGGHzEB}q5K(nWF+>P$i58X*+
z-U<FGE;yNVv+7c6Tpa=}s?<u&sl5-ynS4^>KLGHt?LJw6>VjkCA7<|HZ0&fPc;Kco
zbqziO#8eE6_HVf<%lj{)zIs_}Z9NBM71DGOq5%F(^(`rA^O5_{Bl~XUB*^=KVjU<T
zIyfE|zRpUbK_5CX{XM@stHE=+j$i}_;$67d#*ja8Wx-Hu&qI0c`Q!U?VMdz`n|G(%
zs^>Lu450fbq@Z2v)q@Tt;jJuOh4-A8XF{Y2K7N-OY_d6YKItr1!J9mY^*r-T{YP7Z
zG^;?l(?nZY^L4)OQ7Hsl58NEp-0{$)vK>rH1Uz}oy2x6mG*At!<{<lNNNF&Lb(PbE
zFh_z@N}MHvt{7E-kh^a}O^BDs_k>E@vS(8w$^`$)51>h{7TMl}2OhXVnz4~SLe*nc
zgz3voD}zG@Ds7{BNkEO}9Dp4FU_X)rCI(_It5IA!KY1lL=h^QL(Rez?T77iL(u^0$
zZwkPJ_lh?Xm+BtH?AuQ<S>tv(!it#c;G|h;><}TK&x4M4&pMSp{=@96#AkW^CGOVW
zjj2Ezmqo?4MdhVlgfvYx7(AUbrUYf2E7!Rwp+C-_)ZGpnQ_<RT_-i&G3V-FIpfwTl
zaL(K@xD|vvlbd5@1X-%>7Puds*|>1dFV4Gl6lnpKt_hKS)8D;|*@^OG))4fgiY-7o
zlXA06<j68feouK$7Y7U-cav>tt8uoadxahpJRy6$(70=QH$5c=A73UHJ!|sbwIo~c
zA?!39kV((8ioZ-4V|Z(o%NHvR^o7A1MTA;YHS=LGso@6x;(A8~h|TCpYeq=MxdK~S
zwuMc|Lr$K^8^tn-Ti?C#Dz!Oh_Ad>bxmy+r7F<J0E0>JA`{Sw3n)t_Wj}ofPC?q~0
z^$<BT-{4nN&F05e8hM=`{>PRaHLrBq!@7fAjt>PG+?12j8p>AGkw@&>{Wn}JQ0J1B
zWtGFj(8QW0c+$CuQkqA(_uiexC5`GrYn{IQ5A&$ZjOjKeJ)hY4%Ynv1mNV=PtzK@e
zT??@@!+wPL!B0n-2<Jv$6bU{ihDwH4YxO#T^<vupT2FOfmrKw#o^Se7+Z?TF0bb>7
z7@10);9k$ZXD$|B*bN=woPcRcEt4<xng{#F;U0Pf7dp(fzFGCV#g+*A@^Wt2<s1fm
zDFDvex)avcUiUN~kz0K(J97$*oTvpwt;MmD=FF{C!6v@ECjg~aEdSE|G<`;p!ljVv
zA$$!~xHvRHKCicSh;%X4!u+=!<k8-diD7nthm4NDl43=#d$347p6s*LcvSS{vpMfM
z<#hR5%6{lZV}YNUI@GKQke0Kxm%@}&)5W$014PggOP<cP&}8?ynB^N+@*g9MyyxoZ
z?IcWrqgAiFPfj~JUZ4iX3_YvJ(+a|#0;aS6`#;W5@e>0nDA*nXH+Cd28R9Cx-jVlg
zDU{(wRCiQ$X`QghmUy|XFIPA0ak%uHhdN}3)*0cD@GE7^sq1mVO7A~S)VX{S>3Z=7
zZZ>Tl;x!~(0|L#qsYll?bz(zAIdGeAR`oyE7)SyZ{KQ-&!KHQrN3CfjcKg37;8+Y1
zbm}=othk1t<Ai^@%}Lp#UkCyr%MzOxvyCm1*3Y|md&)lf4VXtyx)J1hU)XSSygm#n
zx$qRFPwwyHiW;&WoOAQeo+#@lRn!Q`rUI>1PLuza(LG&!2rSUP&8H(OqD@{btUUr?
zxqHGbpc2orR!~G@Jp8f2xkE4DyykcCqFyu-`}1VTw&;$yg_^_lLd0D33l{)`Lq52r
zDSVhdk+(I&No)Gcr=VD1&yBc(I81+S+dO{+$(%17X1>_~2*8m@W(bp(QQ~{RW&Sho
ze?y%M<Ntr&|6?35&MyEm>UlGWv1$fxE#h0G97AE+lI?(yuV0jJ6tvos-ou<q0uVrF
z)eJYZRy>f#$ZypQ`K`c4aKYZw3tyl(ObOkSVtw+C4yZrUYXxT+`rVkA0(|8}=!L^9
zDWXZv2L1ecl@rfbGPh#}lwAPjcK_?k)NIh5!rS-)UgwPgvXrFkOEglz^bh{E?_0tq
zpg)mO;a2_nuX$<GiilH_{~zBEyeA2`<wZ0|(^Q)~v_E#WK(!O+!Y$F+sRC4dl>Tl4
zg5ztpl4#&Oxw)MD04XhxxvsmQh()4P*G;X6ZfwOh;0F4K^7j4uh`df5(AR`VEo+gw
zG{SzNw%Nt_cG-KI!1vJ$@nF^~%-L~M34kyfa|nZ;cW;Le>n#T~<$iwf+tSw4mUAU9
z%rqYfVX}3!<*h%wYkOyO%ty5H`j`!|snnx}gzgg?N|6lPR2m`61VpU6)mz>pAhgLy
zxJ8x&W}_YGO<_d5v^DYQIYOwa4J=as&PIrZTW~u!<}1cFoLeHJ$#C<;Aw1#EC2kia
zNFA1f(yj{_><17GT<*(p%3PCy3QB24Un>fBfDET25g*)eFY^nRLJ<l`=Y+*emSnx5
ztJb2FeEK9gTB|0XslrnW(Qu0S#qKv+Misa5U&4Ka4AF$EH$L6sqQv_bhFMxAiQ+CN
zgWttyV}B-C(^}5y(zQyh;KyUIinn|BgrTNI=o6B<yDtnc)brbWx)jOPD8zLKf&1q-
zvuW5guHIOqI?;&EBfeJeRsA%-!<O4C)bA`WVG#_}NCiQqf*8)EZHM`X<w=@vydh{J
z@z=(9v|0wR6>?*55Z=)#nB=^?WA326JFbR)XL>VBMcuOFR9uZ79j1}0(9Lmk6wE3-
zwE*oK^(PS+KPcX=f9O4lOpIWz`*Uif6|{Z>&+)O2s)3}S2B<d1OwzuIJ&cw(+1#9T
z=jlFK(teK$7OS+V8y3k&L&%@8L82$$d?}72G|^&>w5zGS<c@&W2L{bcOh(zBTqp^o
zH-*CNnXh)CkUmjBV4c28%DxThZ(2lVG5cNMU<-yrZJ?}kt4wVt^nLWy_nF(7F7pQA
z7A0kt%1p3~r<gIBjjQ?fDvETMzn|~i>qK?EhbOM&lic#yxLL??QBa{3+6AbPk6?ml
zivt)+IkpgOQYb$;(-19q29Wxj;8f>E;=#XY;eW4M@r&9Mt=dZpX#wb^WT%-nPyk!Q
zU3n(p#8b0(;7MuG!FCU2MI#vNN8~?Ia3zPVIkyt!f?P_}jGq`30gdQN;xS*Pz}8^4
zB#Xv0!K5YXc^ZX#SGOwoTOlSYw=)JAyu>e8U~a4&QRa^ca<8BkW@I(O#qPZnQ;_II
z^+(G>-`*u+0=Ma}oU8i{xFKmes{(EGR;=%DIbJMuCQM5~zx1zxiUDy#t9REAM7Z?h
zqGGNQ82yQCBJ3YU91f9NtJbQKJW;3C5NCbzHR)^Ck1mZtyYUg$llj=;<qnPNa}V&4
zL4LLEnVB8+&Wh@|B7!M#R!V!Jn+P;0wF<G(p(D4>u#kunoM?kPg~P8f?rnml^%SZ+
zgGfdhJO$-@IMf^vEykO(fgzW-mD}RFa3C@Jvr4!ff)T+-G0`IZBNB&u<Pyx_!54I(
zD0OnAt=NCz(@(V}7e@JfV|_Th_*bb2!igA<zOW6dNmH<mhb5LS!Hr~y`z94=64VBs
z9D0U(6u&a15nJhQpR{ZSU)GBXdzZRLb_R8#c~Xy7KD6l^xfduL9TT1%r3RA%jXmh3
zS720C^K5brCohRBBa|_>I_7skpljwxSK}zEZxO}sZB*_ijkwj_l8b?#FPNv`xmWFW
zU6A7xXHFLv@&;Yb(J6NXjS3<kJG&P`0iiFr^_&zsAvtD-=nv1PWHkZX@I``|MctZ6
zuzu8>Bk%30&~P;sqI+_k)9z7Tt2^XRT3^c7T!BhN!s5~4GZeq4?HP!k&kcVxuX^}`
z%P?9zcwPlqXvIo_^@%q&O!hQt&xU(O;3{KlXt`cdL~0BrC^|1<kH`<+zi|%lH-O)c
z62+Z!;x|Uh28J-$7BZ~vm1Nq)Hto(mD#|aZ&!>0*uT?@0WCf323sBM>3%NDqIK8pi
zSyUzqA%{`1*54=<m{G6AT9A54neKYiOULhaZXNkozbv{MkYL4mMq|*(DnmMU&aY@^
ztQ(TNuS%vs5uj&oA$UW}NLlvEv@!;h(>aYhY4q~QTc_|#ln)nsxcdgvbfQq_A<ZG<
z!Qh=rJm0mbgU|8jl<6M1`>Z?%#XHliq}N7iNn!c;NvM~(d}{b#MmhkDWX;jiy^`iH
zjb_h(mMe1|eUe}~j_>E23-@e;a+WV;HXRd9zlpjlVmG+zUp<U=V6*wnEl!dJ{(UPk
zM}lAG?6$<8dyCeH?XgpDmV}OnluZR*#*{^c=)S8t+1-}pK<2y3heCo$ze@o)M1b<%
z=HQRSJ6CmZa5Zvb>1<c6=Ue`qoY_Y`DCIm?4Bl&q>pBG8TE(*KudN8fm_3TEn68dd
zdug+H(_bXVx1PndiTbwud5DyKlk6IO?MQ>j;Fenx3v5aE<gu3D!jH2zsO42w4*RJ$
zAmGH*t^*rnQ}DY^wbVU@vo@BoVU>d?AHDf2m11ALog!*lzbi_P=Z(#y*jP*H>MZE&
zB8rTFIu?u)ucT^GhjM8dF=;U%yC%AU(@c~2<PQ8G!nB4PJDXIBmtJh%a?J8M@OwTu
zn<~%oEtxoFskblN|JjkA>5D&iEYi=VG)+~;Cf8ah+zggOkb}mo^rA#LI)BuX0LD)g
z?tr;`zxv$ra(TMW|11k%I8k@=wcsVOxSw;UU(}y=T|b;jYmSPyQI6M}>t<$YdbL`<
z^y3T{emS4fuImaiIU!EhZ_lj*DF(t^y}L`IKo@Y$i*0oEDJQ>&6_DT}3ecg1Bz5$C
z5_<@gqsoJ(r7@V08)O>-i|0zb@w)2|jAT{@Lr$G5R<kn0eO|lU;ST1aCTpC&J1B(O
z3)e)&s4I)P^6MM$W=bQgJGSh`iYG3d??hoKDB;`OcriOCM1J%D!sA6|c*(ko6dQCm
z5;A9`31#{~BU~$h@d%WJw=S#GHlBd}<`Q;~xkeAf2&chBMUJ8ztbLKhr&qji#$y>Z
zV(1<rHc7qBL|*65aktJC=^Df^7*#ccx;h8O<6-guuyP2tD+?lzP}~VUB;N76&RY}y
zJ%$o$brYxG-envNx<4<Ol`^?V_g6A*06BIm;cOja(3%x>uT^IQL5Fg#Rr06F>vW#z
zj#mO~!wyV~o$WRvI~e*=7iAkHM9|AfQVgbdA4`B{W=@Clc(DSNQ#>wIoCo?^JTyQl
zwV)NzPO60jytsH6XBNDQpOT52XOpW~?)Ue2zDH<E;o0N9xy?CHr<@)=)pIY%t4S;c
zaTgi<=4TU;fhQ*GB0`YsmjxB<`Lca2BS?cy?*-9&X+B-*=WcxQSuCw%NNt<Y92Nca
zje5*JmabxNh-c>8StnHB{%NqLNA}Lq6=uC<%V)PD*3}9TDf89og3V7djeB(X-v~$N
zG3L<jSOsfaXLM|kgw0Jn0CuO&<MQiXg<Sr6_)~#oUMiVM^P@)l{+pkvVX5P{_=6}o
zsOK|Uj*Xt-vofO>gD@)9<=Inu$T-6AVyH&%<Lr|~f`<czmLL^95jTZe0)6kaT0|At
z&Mn`YRx)R;arOG}*C~0~((6mXOZ)TJhnww!vxxPXStDZj9k~-OU7-}@p;*r<TtsoU
zc^5w_l6(9MIAFcxSY}=}YvsG8xm9lEs5~5&Qyke*Bl$h`!4qaV+L6q3;AJM}+{F$8
zGpq7j+?2SCy@7TI5aTF(mBeb^%e)?Q`&I<2m|E+e0%J_i-+taESpT`>dkNck8-Jf<
zFSjA2{pnSbv@H2~8hQ=?O<&?#(tultZ15fxS`%7CJV1rkhL+8A5eB@kk>0tAml#Po
z@l!^gSN+KvDE%6Uua}ECHw1CrdB0){n&8I~?rSrU=DZpqn<s0I)F!{Bb_n#AvT(fQ
z_2^|z@;9d4Asrf!hnEnSB=D-~9(MslCBqq=+Fwee8GTgz?z(86Iv^7f%EVuw2ACn|
z4Bw>l;kk$WHlJyd@v=q{Q_Yk=$%nqlBR+L3cwM@~HS`?2OYUdb>~!~P0B0x1lEYP7
z3AE1J2omCOIt@|3v~+;JpbRke@I3tWZTutFVkI0zrqF>1t{JqGFfgU8et2~2&mrCH
z9e<IGnk5TodM>AZ!s0yI9=W9*=lJ~MrC5|q1ROT-HXJ;t0M`@4$?;?X%mGC1$s92*
z(aOXjlGi1_e59=*^nf>0<DzIDx8?W!Udv=`B^eM7^_h-f!R`%FdrUo*xhv@>QrN^8
zq&CvHn9RyS6e9I`Cp`V@tzm+R=jL;K)}Bl+*%oJs$}xfODr$O;W<V%RbDH>!UUae|
zkG@%%l=+=D(PM^E$cM3)!JcosAdm|VN^U!pT?Ju{4uDRMIK&Y`W0VM?l62mD$w5o>
zUaT$<I2B`J%==x%aFAOrXPKhhl{BK(kNG-mk+bt?6JuCW1nJvK8dC7FBiU5$0n4hS
z-R3=ordjWuIM8rDf4kXt065I%;QN4i)F$N1M;WrTl<P8=pTHY!-t7=R3en6eU;A^I
z#ZX67p9>4C$hEz9ujRFR?gAwva2R0tQhnYqkJ(fMHT<uHt~wQh5U~gY!NphS<7?Z}
zf-O$7bRnr@ZwzZVu0}olR!9&+iXhYGv7Rtd{LgQ%#;WhvligX~Rhx$PQLznC-y-{z
zJJ2!F#P!^t>`#G);jzagvoWpH!|b8<0hV_Fc|{?+1n#S-wBQD^8G>hF!z7*-*%-&k
zVNDzs9yYo8k?S6$cGW?A_A(kz8376&MVTHSYC<O7IjS}G{N-iMtSdmG(_6{PdrZRa
zSH&ppTjKhoo`hz{Z*3MGe|&J%?;RgZaX)M>bSB>g)Fvc-W-sXRXD+6a5-a<Lzd6dV
zrhQ!SY>yh`5h+vo=j3a$J<no9-xYiIsB1j)StF%K0sknAP0U2q)#0Lwj|=?CZLf%S
z9M>Z`X}j97ey<+@;8o;cGUA$%`GDJ$L`jOK<hw4|#~uMP8UDe|ObHi^O!qAZ1acUR
z>W$UAPas46Ldv3Y7g&BhSLCKGIG-h$KD6BQS1ydj^g-@eC2jUox|4DUp){XLd@jj;
zF0}_v$$#+1uy^{2vB}3CJ{9O(xI?3-9&!1*E{Kh!Ygi7U%v!(j;mcdOjh>G-a0Zit
zIAz>PC>8oXCWmDscbIX83*lu<E11iL=;5Lj#uU8jb5zV>RpP)V1n;L^katvWLJ2yP
z|0a&REq+;)Y>~{3Y>~8C6od88E8QlPugaRcW~w+J6Ig2RC~qrWYARXmx;nK@-oSI!
za~1UoP2}bH+?Ke{C+k68^T-=|fn>-I<&gT|HHh|a{Yc61`l(g6fzqb{Cxdz?*E!$0
ztQHMIFpola2~QE%8<(T*KV`D^pO5$BE>M9g_BQGpv?^c6N_A>q+vO~%6+#;$>3A$x
ztOEXOcFRYAk@b0{NU{7ky{4lNCen-3e-jt`{8;)(l_qcaI~jG<-D|Fsk=&$AmrT-U
zu1ZsF#G?zk)k~wyY5QPZgJwAIfT+YlFwKkqGcg+!!^?Li$a0K&U&F~%b@IXGTc3Mm
z)67)SXP&<FJwyTyWC!aI>(44nIq6r$@;z13F%Fm9fB8RSf5V0Fs1@d?8H*=XBHnMP
zjw`*_DJGchPoBH*zK!Xpz>k58*H$q=>}B9g7A2MvJtZjYtcVsuD#K9cjD2TKqz*5C
z=J$R!V_Fff{lB<+>!>K-Cv1F!r5glkP=S>eq`L*B1f}bPG!oLiODm;xNv#433R2Rn
zAR-OYy);U9@BSY6{@(MR^ZvbO&za|*duFbgx#pfDIcJt`lfHnuseQ0s>&Z@mAC<Lu
zH*wihKm0nudQUu~Fo8?^LSZ$Uq5kj&!fuyKj`OUe1fwaJ$=<AFz>K0NGvz6iJS}~;
zKdEYb0&ME6KsThEbb)=>#)|t~GB44G_lbHeF8`@o9wGzh^5^WGH5+m67nil^C8Ul@
zofuuZ*u!gcs$w0{1cL@u{{*A+z&M41vw3TbXUS#vhvkjvp~4bl*&OY?oh3WX$G9?R
z+Bl}h4%N_P?8`~N1w1~bVqBfihaT5M{W#q5y#JG^#{Y|JPJI7DG<^8M#y_Km_O=^Y
zHyn%Yu|cFA@x|i0Q!x#EUH#2VZ=y9B=h8Vgl^U1H3+Kb>hju#tI*hZz-|i{K6sLN>
zm_UYsT2+Ga_xqu8G}dOeYpyqUVa1^aq+kdb`Pfx+@SU+&{k^l68jkPhJDuKMvBCL6
zt73RIsMCoPAv#>XX$7#TJqNuH^6o4@8=7%={7o}Tj%quiRQ+GZ{WFZ)_b=U^#Vp;o
z=*^9RoPF{ab-+qX`-ZAk4q8N3fKM5Sxl|5~t@wE~B<}g^%44`<&VC~M6G<PGNbZG#
z2EHG>eT}xa`8(fpqjo131-Cp!T)pW3;Q}bcJ`%{Si`j;m|MIz_bY}ISbysqA$Lo;i
z%^p+=uF)>=dsj$SS6oRAF%JNZp39OOy@4<eBe&qg$NaQmc(MZM?vf+|w!+iqomr|W
zrI+0z8JV{JSU0jsz@sCf@KRO-^71!NfTe9GD>SfGL(kr@U7nxwpm8o}&(PU`(!rwj
zL;DcIMRAVhjSJrpuPxB2_sE$AMO;@<AiKM=!A?Y@(qC=a8;4p7CICr^K|~FqFOgq=
zu|aSa9S6#(<!<pW?Y6OY)w;Pyeq@yd1rIN9EBUZ9R>}m0CbCmLVk;M=Mv%5L2{>Ij
z2-a4c>wo3bPnf2D3N%dL1hO}Aw=IAKM>os>HvmNUS1Ua@m+l!$(t>7TiIoh^0zL|d
zf+ji*@M|G)4Yfyshh?fnkSbleP$nVLQ=y-3&-PD<Dx})R_42KsZ%^BUY3>9`(?Ym!
zF2VD=0^pH)<VcamB<C}^c@j_`f8HoP+#+X=yvb}UR%P^RVCUib0_kN%rjkD9<9Zv`
z!Akgiw&D3R$Ukm$DOyMGxt(3cMxiG44e)S%V{^Zk<w`GK+7Mgpbhg?`62NYX1-}W$
zYz(a>>WkReASo2e;7GWG{g0RE^84oN@(KS!EpPN*#zGGAauIKcVX_RjnKfBC2rQI|
z6^)Rf5<Px=R9ubslcu|<d@8d3q9=^;@VL4jfGzAMK?E;P*uYOvUIZ$MBo%n4lHE>E
z0JN1bYJZt!JimH2-~Fx+%SN$TIOt|a|1VtsTZzeK@oJQ_0hXIr1>-Xk4zE!vMyK>n
z5TVIZ@pqEvKm*YEe7|BmbjLTbR^OX5N?V{-V2P!*cYn5I?*OiD6h$P$T2iFxowc`Y
zarRq9_><v!qP{vkWaFQ(Au*J01XvJ^IRUK3xpu(kF8b4uW3zPN?4CU=wvbNR^*3qa
z_}T78KyGZNC@lbl5;kW7_$oL&L**jGK8qwT4s1*#MxU`(K^KI60#61MR#3JQ20v$b
z-y~L2JYa_@cw<Q<rgkdElPoI3u$_3mA5<J2Kyw|0GHdP;fd1LD&RYI{udxR~h<>~c
z?+@9D76_{nJ<A(``Hu?ZOVsU3IUO2uTZ{j+gt942o~#uFv~?DkwD088)6GJg-acgO
zO*53>ZX-T{3RN-gUCvi8Iy_7Slfu*>B*GLWszun%nJfAL9}k($@YA~(Uj%B<nO`9m
z8wgKfZJ=FzMOexc!Z|^|;EHL^T%bXCBG!0hzuri9tMneoUEayTSH7aPwpd?0e9-3~
z1Kia4c0;}~1)mzd6hXk&X4$hG#J`1NHL$es$%$S#38o}(C+G^py^&{5Pp6oZm(#;j
zy&H0eSS<&-7wJSt{n?%VU!stJh|kN28gT7#+gXMv&j8M@qjD10nJ+10&Ta{_e*9Oz
zLGbfEpozXqGRj;P*ya<{3X*%1L_m(3PjqxpS9tQvE#lOY^0U9ZQ4BRyyDP5wZrftM
z`P={5$4AWoP)?$-_X%hoI%l@v5ztPF=|Nj~fsaHm^N3J<EUXU3U$+gz-r^_CI7$gR
z%M*LUG~6Gp8yi`2aNpIgK6#O}BLMJAY8K%k1NeGIGk(4w<6))m9zJbKU5ttd*OLeB
zBnk5ZG4nZ}DJ~S!@e$?p?kbOBuADehSVoqQkAStGbW<ICaX<4L3hB(d1!$QF0ojWd
zZdcZol>498L23%jbmp7X%HzqA%ESw(VZj*|jryhaKhUvgdZax<2h|amJk-(ah0$8y
z@e<$VF2ReT2bU52#L@~flYXei;$ok*zDWoSaXtO3%j{5ssvtRHbx@v+$Ooa>Q;?w^
z<vFQzqT79BYy@UQB*ivkB{I0|W{w1v#!iKgp;1pAoC0s#)*6O*1#|u8CO%X+nAet7
z^PpLJxjtq6#Lk85qwSk+YJ>CLA2YkXCG&d!6Z$D97;)DDyt7mywl0nAM7FBLZ>ND<
zV{7@vlLk;Eubl@WzrO21{}}2?CfTCRQeb}k4eMEP&U+>U8E@!Fp5DFx^{SPKU(?W8
ztRvQlVEwA0L8p*(kp(R^a_FYH-~5<;wTES-#|)D?Ghuq?$!2`=180R?Tpw}eN+?(1
ztoPrq&77hWKYZ_W*g`1uY;_BQ3drinCQ=#9kPx<T5u0&^UVIqO9w|>p>e1H{Jn;Xd
zR;n>Qp#YtLc9cjm*c9j$EY+v;*}!x0wfo1lI?(T^&6aIHqEwR87m4rR)l}+FG4zZr
zhf*m$6FpDf{1J4Fn;G0V`qB<QoC7(j1ie+_-^QW|*jZ<PBnNJn9q!<q-#{OLcOXK}
ziEE+G-hTyJ5Z!>Y4=Vr15`-?Bxl(Nw35}L5d17HIJQx*bKRNIqTvr}}y>+cTf{wgM
zHSpT=fG0PSCYy|IZzX2^Gk;{==*o0Hh;^=p?-Gaxe9_cWq32sMLRN)-S^Y2nw9Vf^
znj+v6zMvPxB;`>}0coS!F0)fb;B8SHNy&X7c-3h*<woui{fM7r>mxr+1FoXXR3P5-
z!^zb>`AV4k6^Oxzs}$V9^B}LIJV{w`v)BRHE#gwRz)%(a1msC~91Dfn)6r4~Lo2Lj
zcjR`b&9z}YfA0hVL5H57mqQzol4!O9H4DQ+mpPQ4)=n`JhbK_J`l9rZy<3-gViZaP
zwk4CBvax^h#R<|eect(WX2k7Q#Fl|z-c>8y;lWpM#N84OA-JjEEg@E0YA7?*%u>td
z2M-@W2l+L}`Z(}mMh)N<#{Oxegt(t5aK1X`KA~5TnAxssaFgoj0s}iTJLZ{kf4^^+
zdT4L&duB%JVUSvW0Pi3}s_RaX{J05D-MShej4<Y+B<f@dZ9kYc%d5A#eSx?b5?i^u
z9>&36;9=ukY?|7toT;e;u^-->m+k9sf%3awnw$6cn`pFL4k!?tZWM_1g{teC<=r+o
zXn4nmH^voefvn^%wrvTO+8qnXGk8k8{=I&o3mq4b7i%0|PYU?5d~?Bk7NI4$08T&l
z$ra#UV2Z8!&F~Ml`<1pa79R)O2~s{;KAUXHw^DBb0R1<g#<ad`iA}UQO`Ud>>1Doy
z=RvHfw?`xI+8w^RS-{E-c@Gd2P<c?UI>}cOL~6-qCuQ|-U#|0u8gQOcsC<=4H$F)1
z>E(rI#MHwx>FCASmA-ZRW@H56UI>P!3w@_rLY!*iZJas22TcUbTUNw~#ri=<EY?uP
zi26;qsNI*J2dxYqgpo<!OM?TwtIJsOYSfo~RB64d9A+kVCA6}QJmv(lqk&Ie&?1=(
z$2+uF48m`7e%3Uo??7*JE@{W7=~S{|e`9S`c6z|EzlWEWm&Rht2y^s)IFC;8m#F+u
zN1mW!V_nilsSE-SpQo(UeVpTV=o4U#{I^`h_0-JOjwmAdS#y{`#(~yBE?GzX5u=I`
zkp~g+8)mSDT5GcD{S;6g>!LQH^S_{16!+Fr)BJB!^cMNj3-<*834zE_!z214?)<nT
z;Gu@}<R@oNIcj@r2U`lwYpwmg%F>5L+$WamKQ6w&^YPtp{d6-tccZdWA94L%CET8_
zVl<JFXM1yL7)?2+9mn%siu>aV-?EH4)Iu#)YhDu+JSi>c8V5T}<60MELpK<>^^J7N
z;Rg>J-?#kRmxp`0zgWd`U?LfrzZcKMRz@h2#6E@#+!vw*upH5d9wI&k+i1?(XsJF^
zHt7dEDw~31h+zb|^#_&HdHMkt?Qwj~I?ElN<kYLp%CJRqlH~jR%0^#1?_&emlsz&P
zZhXT=ktPHr2u=6e%kbq)+#n$@RP7AE&H|_b-}o#47xL2;`VV8%uk2SRmZU4Tn`6ck
zN*+~xEG6F@YTscfF(^?70i&<c_rEsEB>f4sta`q1Hg96ajDVcwZ{id<KowMNQ9gNN
zv#54=D<{`F#cte-NEWzX-c-VBw?^hS-+MTV)aEd(IsHmhV*u{KwM<n$UMJyafg+*X
ziCJgy>zHrgQ=5Qh<4j}J>#O_y7TJ&XYOyd%a{b7XfS+d_;cCv<>0E9^rh_8PN|yNT
zAxhUX6mR!I%zRiyF)Q~itH20JG%R!B@g14@3B_AgS$jZy6@qQCg8LQ-kng*$O!!6y
zK2$uWv{J5Q`a-qx7|gGH6)lFE1*7!Pm!EhnmL9zErOvr+{mP*%>Cp`oAS(y<W#+AH
z<FsFG^jfBbO4(^`sTepxs^5x*EwX@W;4JPI+t{r9^CmI@W~3j%?%!U_^XeEO2@x^;
zQ^KD+O25<C)-@6fRO1h2q6(?+w0;@h$Ufo%Y#iozmRI1QKGD#7;zR}dosRp|u@=Tk
z{hzQK4xGhy0$~~?f9*v-Xv)z7vUVdN_<MQF$~<C)mRrm$EFj>f0%0uf{T*-6BXgPq
zllaQ!Y+3@7=M-FN7i?B5w{28yD_0Y~Xot3q?bejx6`!H6#L-xj35knK#Qb^E3Lx_5
z`X>2hReuV9(bO+|SCMF7Eq5F10RlXj5*i6oXfWmH$5k=59sgNN@$d>3$Vg#w-F;<Z
zO#r++npzSxU$=~<#C&<y^4q5VqF80CMd(JGuK!#?R>7;sESPsQM%Y#S06W$H^6RiT
zi8rA-7TSy(&uq9FCbYL;bkF^J11|VG8t*^?@=S6AXJ|1~%kwp?x&80QL7081=2~Mn
zAlh&x>&Ay5_DV*cHLP0#jxe+SqBJZC$~bF=saYO~Mm%$%SC|sI)S~3BKKbctLR8il
zZ+<KGG3O2TH0I`Ugj^ak%H8cMTZmIWa#w8iMIm2i?yND*rgGL(_KOz>%PBmQh5ZxU
zB-rjJ(^iTl-gvcS4|e<7WBjTAK)HTlvR|p1ZV9;L#<iCN?zEdu&^x=Dq6FHjyCR<L
zWD}R)WjP?ey6fXrbXVCqDKk3+X8Mk3xpkw&)u`g9!24P^XCmNr3v@ec2ESTl?0t5o
zhS)_YZ8w(=lkIlVTUo<TOS+CZ1|O<(v@Q3DD<s6lprW(0X-uueix2<V$W`3;ln1$$
z4?Aa%X0L@-{r5NQ;B?v91Z!_g#yFf8qY3($03!J;W5*InhEDE!FNS$W+oZ!3rBhJd
zswkPJ+d#AXnGjkkQFMJC#y#sipbBUz)tQU5&1`*U<_joj>Apbnw-Wr7(EUl=V`Yy-
zq})9fjJ78IMKR=J&=n!hO4>xfXcS;HSg6*rx6}3Q?TW+LlmB5rl<s(l3~{BK*7kVe
z$eih`lXSd3-{7vhw5wZv>+|;bkxyj}m6-p`(KIG)Lh7R11faUY9g(T;Ki|mW{P*+g
z)p@Q&e=d<@Rbp@1zOu(2aCfCw-1&U1&oT)NOtJRxk2T>JPRF5lHp5QMF*LNA+(;j(
zO_e_6aDEr%2<>UM22nw7HQv?y`wWIa^#~k!<iP`H{7|4kS~U{4yg`{V+Cr8(QL1{D
zH<Y<loa?yrEuNn_?34*}w+megmt63(BIC*SVleqFh+D6C(8l~GoHJd;uV9Bu=Ek;0
z)eXUjP$&CM%fLmI-mHl(T49V@VZi=hr6aJFYHDYm*51xOx!1yMXCw<{Yj_yQMtJ`A
zYpm>r{<?<{c^%PWB1iR*GUO-W2{f;Zl#L(<>5MY<y!J<fdS`i~ls<<3n?R=Ea3R&9
z0SSKOcWd!+$;^P9mFN`bci0xG6x(9chZnk>tl9~|2CqUvyX2U)(VUu_!>cO&K;-TO
z>Y(QvMh<|-hrF$Rl%>At!NsBKE(fRy)k}2$NbEMv?0OdXWe2g*`9S6iPUlGvPqHI4
zK_9g7zlIz2U6_p9e+VHvOQJATY27j20P+z2d*tfIo4SmklFyeq5kJt9^$W!prwNsT
z_wPxTscBFYqnr~$(2buR2%Fga*u{gQ9`9XC15{~~NRB|4EW7mUVGgn;m|n3hAalwh
zPnBsDJtB}JGgCPU6F@bD#2k@*lQ&_09FM9qQ+T!#s(U}nBlF@V`T$6Ke>CZluiMbl
zSt@YC`jtN3eqIPmej%>X^d$?eDK3K$RvMmP8$y3Xhs&O3*){hagr3RZPJ$g-E;t+r
zrUa@3DL(Vz!dEMPWL^ZWA2fEgxyJZH#q$U5J=Pn;(vJK-!NWYkL=szCv1$fM2#6f_
z6IwdwE)(2X=hSB)L#5d@T!)_KzSv|##-mM#BL9>R{x|#G$vvSDLxN5(^$StWd^$#{
ztq_Z!&%bdsv<NN<(yI9wu-_MINu>KyG7O*ANAZ$%*BUeNqJqo_QOVtzADPiy!mJY-
z^Ib_|W^U-S9Mc5#sQtoa?`{Vb9n5H)rrdE1B!f#II0&6jA6wH2-vTPdSufH}RFhbO
z-1^A|$Rhvv1W(2~7|uPZ4tFCFVi3ijOJIZuM(X;s?|z`2pM_VXWr55<GJ0FDZ#rY5
zC{0b<OQP;$X$sM$%|tpFmeE#Zo_t1ekKo|BuWRbS2}EH&_2s5GtNCS-=|#G-gzkCD
z_G_mSUlZ55L~-2KlZkk`TY0^ttO^XB(4igUDWOvX%oSx|C3BZDZJ1>K$?%3b@$amR
zHQc^Gc4v5z-OA}zanVZXAjH46wT}uj$&)zAu7np79zcyuV!3nPD-7aPF*DyAs(}i%
zhGSlSd&_Qh6H*yc@j(5M5Jl_kt(0YnL|fJa_^9QfSn|%Zv&$AUlG?})1;;B~U*^SV
z8cs5MaejfngjSn|B${J}Ac&1$A&il%U)?*(Ue8EZlHpXh5}|(~GPE)Gjwu(R3C<L~
zJAhj`xSap4rAjgujzsB^oqL@;t^p&yFK@Tbm`R_u$zjsh?fuxCTx2rZ)(S;9`qwdC
zvwPFab)T?kEvkSQ{o>abW21+s%V0KAI8p!a@ruj+WRzn9e{an`lNXgg$1F($GsUI;
zY70H+yvT!=4iyb**KH0M#Qj}7&K+UVlUF}et*sAk@~~oPdlh2|{i@mrI{{WYyBNH}
z*j7*f$FqAsMn0+ng+e_GqN}*aL*n6MH}MfsX6%m@%v-~?@SB(Qr{j!G`*NtQ??k?b
zeJbXawOAtN<&XoL#W6w>)sg#O&>QGRj<4EbuoV^tA63}Gd^!UwftjkK``k4ME&KS4
z8dS%23fh&|`?a;E*Vh-`b#j3^Z`Cq;^WhF}lBsX+2R|_Sya-55eX2XytVAPZKkEFg
z{YxtzD-%bDHhnZfvbP40-RB`*LHuDyL#lZfZ)1Gfv9xVjtP<}62bKC`mP1U4*O|1Q
z_Du4voV5z);x0ZEng;=(hQDjvsS0;cgENGQj=HK*fQNPcb5-H~3onQ{55KBO!@B9L
ztPOOMfY$DsnxCk_eE#H(%#H?6A?rQxKzq*aO4&rE@J9W?|A5!yLPs!{z2}viW5>Wz
zljooB%df3$w=%!^-u69)l0ybIlS&FKz(L5CFt}&cNHU+|{hZ{CE;%NZ5`Z0!nY5-E
z3m#m$eR)BhC5^8)F<B{lX3+sgl$@W~uSR7HUC7@`$p2#<t@|1^3OAA<Ioio2+e=Bv
zoB!B&9~%VQ%{3zIzA>+wCICB=AbE9bw@5LTgLJ;SYQx#ca?tm&MZCWAA_>xroEe4~
zI*LmAl3Vsgs{H5`!8B8uC9z)c_LFU9C15uhoKQ%NVTWP{w6{gc>%X&cI%-YyQCdjg
zjr&u6%g(ZNpvL<=i62KZm_v;`qOA2kE&KXjXODbl%RrC1!i_L{NJ+MY9?fdNxJP>E
zcHtVM%p+9^!F|_T*MbPq)^RkvuDy2047$buyIFDM(hal=BNlZEjTCmpl8%3=8S@d2
zBpL8n|Iat{70#P*dOG;|iCIcxE4h(r$6l672FWF3^9i#eLwBiOBH~>pfd#2qdWvt!
zN1({}5Y*fnBE%CrQf3VFqn^LpUMw{}%ln|XbymM%ue7)5olHCc9Vl$wxN|7?fH?Ib
zGg9;Hna_)0`~V?wnbbc|AK#cg|ED733Wq&s-}pqwF?w5(!%qFiO`M*?bkNw+m*LCr
zybJ!siXh$DtqypPyU{AL3h6s4Oh`$KIrDb)quhP>Vm!)N1W={cy#%HB0D^pf@j@1m
zuKCL|Pd%RiV8F=iR_BKEwj$?Gp8}99{80$AyHBd^dG3>h_eWys5aDa1E)@U?G24py
z`iM9yev|KZfTrRY$Xxnm^0QWv^tzXUjSSZtJP+V0P{>4#542Z-WdG9v;`_J+VFhiS
zptua}sl&(s(pUrtmwf_ci;1d@XFmD+ud<U~iP^^TDElh9`f0--4!MFC2zNycj^X?;
zyeU!thdeMZGgbAm{;+BIW-VUd*-pu+*G;;fr2-VgkR_Sp0#^F;;#JSR?g&Yo9Y|C=
zyiJRmBrGVLkN@C-0Rl01&eeSGF7xgqJ1ctPl$<a+>!<YK^s&R#)W}l@S3f*rhy=~h
zxiC=f;Q6yJ?<mAO+wr~RjZJ@H9rG-Ql{RoNs5T2sr$YhuR<%1DB6EFy4N4hp==EYk
z!|{H;2ud#f`s!rXl!49VKL7=2@xs_NH*mCbfSO(Kd|Yqd!Aj`oF04EoKl-B&MDUqM
zlii7s$rB0b`c^ih1Yt6IeZm|w@;&y5VyDw73Sc9*xb3fB`L7h@HR<m6o-+bsK^Cn>
z)(Y7I=A(F^APteJSkLx>XeIohSRy8NEVZ9}R-%q_o|_%8sOW+{L~!zf?$`Q~kv~6s
zCq$`Q<`|plZW3!vz^9$^39-GjIVNzdy7NF6&hj=5XZ+S59hOw=wN^5v_tuU1)L<84
z#;p|L$IWB->XmREy&r#lYl;#jh9`y~i4`*1D*Ths7px55o)n@w4JT|RBA+vwzlQif
z9#-qSc<p<;y_zN}E8WcyfeXZs;*SWg&k#f!vdiI_C9MAKTCeJNc_?m4c3HHEQZgMm
zkV|C%dSCkk27zNYcRW~+lM}%Ong2nLv)f|ODHywu@S<u2B=`jArQ}ej#luRAt~A?J
z0&>Z0l+7?h)<mA8xfh6O<goK~GdSr!Z^hlJxOeh}1BgVf2_3j%yZHWt)|#+{ZXDlp
zt9luh7r0ht1A^93**C(+>H4Q-4{2dwfe1fWP69F~q09eVPxng{!mo8{-1F%kX!o|s
z^24g9XZ!Ng2rrVrkCQkkg2c1k6T)1w?DvJ68P645YlrU(>@sGc4S?Wb>|&CgkFSN0
z+;x(k&wcF!eNsBQk$EHpsBrl|`D$f8-NthqH>7jzS8mC!krtOq$`>4j=UtuP*)Jgg
zW63_lTiq!>_54GLc|)9NdwnEh#n)P+-tD9n7k1`4KnNZlYZ&EuJMDXerigT7&v8|7
zSJvsmcxWlrF=!rjinC#=V6|l|S{cNLuti8DFcO{P&Z*o5KXPXOSw$QET3r2U7ytX9
zRkU$wXz}3DLpElJPd=+KFRzRpKzY$Xkfr8Y9T5F>txrd!-@m672kKYy+`rEvQ@@1z
zY<$*UzfgtJB)dMK%$k(zY6vJ#fo;T1jsEeHRNaxPJ*e{-kk67<9g>_R{dWMjy1c<f
z$kZcOSV<|^P4?Y=3GEcYQWB)$vtTPi49S6!YW|7>vmqtI7>S(7_h@&?st?$AlrS{k
z{OO;6kp#S@UAiEy>Q<_!=~E5mDZ)<R0iX6n!;pYJo@(&-!b;i6_jKy0vomE%9oL;q
z^9<<nW5E`YGnfAzo$z%6dxEy)3dQ=<C%*hZF@cNSz6!p8?)1lB8E@%yW(|JIl-E(P
zwIk;`;qAlvE}^rC8#zMgPYENf`T|D$c3r9~>gjL8f^IMRvX7LFJVnG)VVUYkZ6odG
z?{Y(1=@)7TL%P_j=75%2mbv;7j+Y;oJb<-K9x47#ecg&Gd`}OBUs}c~tsw6FG-c=8
zI;L896w!!Cw2w-BoN%J_vVZy}bnjim4KgFW4Vq8kO!xcR{{8Mk=h<xsg3%32)i)Wd
z|7xJ^I1Z37m-<-oQf^n#N>jLw8V2rH<Rq*`_4ZLHUlK($iw1h9iu<v*Jxx)4-*?W2
zA9U~YgT?kAeb2vq1J06*uCp)_htj<}72^Y9#!pKPet0|9C-sHoMXUZoKHb@Tot#EE
z$VON4Ck2-x!j+&4S{8Fons(Ld7Z_ov#G%oCBkLzX){OL-^Q>WHC|EAq$s-h?TJ=ps
zE<vZs=kK$Qf<10K9TJ!*5YhrQ_#!Wo0N*FS;-P_$BtU!W=!?NHHZ-*XzJplB?bLsg
z;L|u(rrPSj63qe7bM;fkEkSJy;<z~Xz`;8y#&@s71mA2&x*g69AQJTO&Jg)fbx2!_
z<@^7Vx^=94$uaeg1_5+oB#0FemWqC#;=-|XSGb~PAGLg3UbK^E|BX+H|B}QC3LfPc
zY1*a61YxKQHhc}Y@f+5A5;MJ5X^C5PXmsPT!ChA{jpW8xE6zAU?<KSeaWo<iG{6m|
zN_Cowq$AeW{M89CSu}2dk%8;FIvScHa0tv98g7g~bXEk&Fw*;A#KRa%A0V#h^=!`U
zE|IVM7*3D`v8Dd)!q@r*6+vcmSQH#f%}Ox3psIrwBlTqQZ@hArg(w)Ihwen+;u)R8
zDpQ@uB&woi5-I~MRg7CgwVh!~W%eIP2moB@@+)-m2!|4QWxpKt{>svGfzX}6Jyh5o
zuUXthU0O<8eTPCBnW!YwxN4t&E3Qm>Eam@j0Xh*o;6~5+7AYt07~{J@60OBWt}frM
z48LS{QzJQ!<(0rh(pL5<evIfAfK5!bvC`uCF{JiB*1{_)K+IjRlmA-<vnh8md_tmc
ziAvlwqN<_{|3>bluD7iRu0zggx=uwMNUj1u%T?vpFx3(H3GODMnMY`Q%OD<<k<<cP
zsklj_AButiS%ReO^ZH>K35Sv&<dlKu#QiouWlMsOXHKLh^a%Pj+>h+%4^erNBaTyn
zxBQ|*Y<`vz3pF9KRm1?iv-2X(Y8>9gnXfndX;o6rzSL_r>+d6zU&s|mC2SlF?1Nx#
z5#ypdQJ7OlbTPY6W*6Hb?##;dw3jg1$+zWyx8YL~2u)NHwH;ehFczP=b4BgPSodD3
zuV2lFw;@IXRE94%*;s57Jo1qE>R|%-SpWC3g^??uRWEcV03^($@{1!Q9l#+JCNGR4
z7$p;df=Z$$tDoGiWQ@Jf_e+KE0}3Rit=(jaKs&wPpmuHz@C?uTx^W;kwaJxF<|Qfd
z&+cskt9A{MXS9qb+RBaK1aJt!;<s6e2^r1D_bpgr14VBND5CE*=~7xE{yn!0)6911
z4|1cK(r#oujr~0ERZa{{5AX|HwDwf9jeT>{f>^>dZC-?OTa;n^+={Bb6gXokpV#g(
z?rITp<~&FBwkrbInTvrfflgi??D3!0fP!(t*()7=0GaK3_wrx>5p|jIo~Z5|ICA-`
zf)Rw_L(I8HF9!RCON}P*FDbA3<#af@77Yx1JS28KVz|8mXkjNiU1U0HV6AIdSXJ~4
zXgCBIBv+yMhoT1~4;y)8KUq*B%L_Mu`GPEWW{1DE+&wOSyA|+cU({U20&4B~*HcO|
zD~#~$SGi7umx65MOZ0c$T8?#`Cu@LPwyK8~mIgVX)bu${>gVhpsN#;ekM-f5$RmIQ
zITrT^NRo(Juh#7^y$MZwWXAj*Sar6bxuqQ%CKI4bprF;OteAz!f>5aAw%H;jz6_%r
zT>A)ozG`ZU-^4A^k3igN*He3Cf1F>*5ZuQIOk*X@$x&^H6zSGY$m74%n$Micv;48w
zugVbdNw&^QL6GnQX{sCSu1gdx>7b6*fK$G;BRPQzfI`?7N3xn4AadWIo*HY;-o8-t
zzvZkAcmsxyK(q9o<uE$fSSt_}Mo7OLl}8krB>KFj0+Pxk9hL<xSCr-&P|526qI8&_
zNp$qvi)5P*YM}9Y+8gqoXz2}<M3o$NSpM<Q+RsK5182(hknmI=bEkpci(#m$*hVx(
zfH_WyK5qlyE?M)K%J@BYJ<ZH7GbwEZShX+@5_eIP6iZ|lR7X8Cw;VK`%pHd>hws7a
z?nbRraye?PV_>tW^At`r%H|gukw=goS$TeNX(=~=BjZ0-RvbMBFcaT_)tc^1#TEPR
zlV`>DTZDqj%wRJZpK{)JZHOJ}J(vJ_qI7vH!ly?Fe}ej%)6SpXT~dq|Ed+OCYg#D4
z)>=Un5KBC)q4YfNZc0ax@~yQ|LN$<SlDwhm)DjMiEqb#iefJYb2hjfj(5MDIxy2_d
z@Tu7w`cQK4!1Iuh=S&L>F{VRptCR+R7B~$A{==`JGSaYCJan;=`q_w3T@gbH1laoV
zx4lPuXD(pT7fI}B-c;7<GDt42XYH)@+j|h{M_ue1B~ZJ?P5zmrsY(K2JL(te6%Qgu
z&HQ+XfOp}`lqN5ShqMuAaJc*3pGZn6o*##V|E}DF>^8dzOO+f^RyWsN?n@H20!2YU
zJuZ%}8T)_%HxFMrD@DIVZ;48@&WXL9dI*zg?p_N7=RaUcq<eD?T^fR|*DeRzrs7_o
zOE%sU-|PsFqDq5UDzD%_%|_O*<8XX$HslgPa})r=nD}gA%n<Lp9CZQS@zw#Aiu|t-
zGXbyxYxAvKjipa@k3(w@N3EM^ii&3w!}dVs1?C>=1-YQVM}3VPxx;hG;c``15Q(=8
zte_d@S<0E~_>%R66AVJgtC_lq>rKdAtT});@1py>%vkfwfqIp9=Qzd)r|R)#=+)6t
zXpdd$SOy?D@_WAU1L^xOF;l4rZ~+!~!k0SVqo``3^DtJ_uCza%D2YW>2<!e1{V-A<
z;ioF96&9ZLXij@CP*EOJM7UP(`fc<bQ04N&x*0vlH#Y;E|6d;vdlz<qU^kEKAB&AF
zmkP$vfgC?|%`^t`pjbbHex7Wv%qKDbL%g(S`_fQg&b6T$50`wv)BV@7EJk=6e^bY`
z@6OXY%Ca$Kv<fB&umDM_(}9_KBjwjfrDJcPfk&cob-K->(W-p*2kaR}S5Q_EaTg7x
zOh=ilJ9ZVlFEKtKa^P^3B*sUEDW`*~IOYiWN<Q%EBK-AsVB>m56YB(KBsokAExI9?
zPmlJi{O%ciw{q+Q>We1;>CKErcPFj-QisWe(t@I()Nb5(D|N>c)GE+M1yH-?h@Yq-
zyTlI9HUU=od22KIe-XP%T~wHlgu{c{EhGL^Y4yFi*PkQ`P_seBAZMg(*Hw5IzcXz^
z$!DhrpSxVZ+uN%)_Q%VwjkG?HG%F7dTu>2<q8<wfHSTg%zm)Z+6l8r^R{>cZM{FRI
zlYfXA(5=9&h(B{QaUXnm2bVs76{7<ZRFW_oD{Z^HN9ZowVT|j9uq9UD)yX?hebA@E
z=N^zI&qxVkiGQx)Q?UB$?@QT=pGagHU>`u?iAf(gKe0pB@vYU?_aj(eP^)X3V$63?
zo(eAAbP%_|;9z0#X0>M{nOz|;9=W5IxAE)6FV{EjrtnsV>Kj$HUnfh#yR4X8ZPo(;
z3=^%OlUS1CdYc=*35Y3*?!g$!r2s}djwjlaof_0<hhqbeDXjafI#dz{>H$6r6j&;+
zx?yasQ@OrNhXE*vY}FDC%(kiI#c$tbFviX)&*SeoTDeM+p`cBa-c1|uU<!Gdgrw6u
zADjItCO!S02kUD`w^QE9L>)gSfYFxSeod$HH2cmB%b{=8*fJ2_G)7lK-H$`V2`|{O
zl&7@BO0H_SMEiWnmm2#=_*)phIU-7oLWC(&M?a&7Am2jz1gt}_^Xt09S|wKoG%!!>
zurqm!f9U<eqlB(MpA2yTJKZhwcfqZ)FVU^gP$d0-z+9$I!R5fu<&9s<&c=PAdd4x<
znHK~IZeDru(sD{<G5c4g$*alDAAPipDr_GH*`-98VdMglbs*6T^WpdV3L0Hu;|N}O
z-)-^|#iP9yT(FcV>#31}M#P2E-}%mK#;voG2jA3DN*eAp+_PzVB{WIx+u0`0{{7Xw
zX`6P8JOC<CgUn5oN6__XNDzQlbqRIoWtqlQ>*(u>{4z=rLsZ*D9Y2Ep=gLt3Pg_>j
zNoe;D&-mWGU49H_b&NORm1&$BLp2)&i)Sm<-xdpc?jj{bYnw@?Dlcz}Y)=`Na3^%P
z28d6f3n#DK=i=5ZKg&^(;ggwe%!(foQ3a~Q(3G?%40t*3XKEF463-uPZXVt`n)7y!
zC`R($?;k5S$R?k4)L$4L6bR==bcTmuzxZE6yp-+5$33`Bg`K{Ylt7Q;&~_qI_7I9w
z31Wy#AtfFm`VQ@8J>#Mfrod%T7M)Xq>-5#1saguctP{2pe)O#S-)-D*1!3xX^9;dx
z1g;y)dj$7Kl-+3a)ah~KgZ}(;^=SU}#a&5Y@Wt}vbkf)|Nr0I-J7_jS0B(7|IE7@_
zj#lB(7D4W3>|vR21QadEAprZT#(@3gMdiWXEyw!kLQB81ungOiY!22@@M|76j2yoc
zM35@#&(uA#lkjlKo4-P!zBxCJEgoWq($)VX;yJ(qYPyxd@{HW>ThHBx`~Zr1@t?wu
z<K(_PQ&RaX@uF|iRcmo`r>SgY{7o<HE*uZ6tsHw!G8E)}Ux;+koO15>CWVk-n>HpG
z?w64T4iawD34^HJMA=|9UnpopBos@&`^dFS@bsa(^sL*@6Y5>t5rbY5P+Cm*{Sh*4
zdXwSEaMyI)>jro;J1~&o5UMwEp0X?a(-7ICH)bW4E!USu;b5pBUu~6QDYFwz$6?Re
z8Nh5bs^_2oV($tdy!?8%ZttnIx<OcNwI_A%AIvf_p~1C8M;u!>pz4RupR3(1>q@%|
zwsQQ#mT&q|q)Z3MGy3ePdB)SyaE{8cKW7DS(5Bam?B!KBCe4W)A_#LlHk1OD<X@#n
ziD7qVC5Fu4FYGc6MA=fckOkfD`}D~MIiPlGj7ar9EB#+jQGrtnA)g(@lAxL4RW#kF
zyJ2<51O)FSCKjFRQ;5Yuhg7Wq`CFm<K|ZCDP9q1dMW-kBY$h*iiYSkvzREe#S%upD
zvKaSQB<szGCCeKY2J!(Eky8JLlc%Sa{YW^fP1EiX;cY&0Eg#!-gF4?0qT$bJdcPPc
zsOxj2p@)3KZAMq(czsLX0!4{PVa-$3W=qQ~rrf1XoEl9nkTdJ#TOEm|lz4ENoLUmP
zsIMT^fXNB*x$h8{N2|>g;LaE1GNecC^%Q=;8s~44X~FgJBV|oeyb$oF5`*abS;nq#
zxW2cQI2)Vp7zQQlqc%q0rzFqhXdG94wv)5{&`rxMs{8W`Svmu@xhzdT7F()0?F(DI
zI{|{hEUAV>@ZVk!+{uN=QeljU%ID><7SP^Z%7L(n(K{}Ec%yw86hXIAKX;2OGb2K3
zaeQr}L$SquKr!gZ1E}T#K$I@Bpdz-s@dzhaECKQX9i>%7(bS^eKmOaFV2+k~c+y|A
zYrH9eIbqMuONseY^^AC3+QHZSxxB%@o%4ySlUp~g#X`S}93c;L#aqo+3y78u^=m(s
z#`t@^J8ndiqzG6c1uk{)TOejCj%n#NWq!0KZ&}m}1UB(gwr0NU%j;#%q%TLM7pQ$5
zOHsmG<?IVCM$J(SPy1sy3=%{6O%CP3Tg(}8biBv;W&gBjePGPUI(%DRL4q+wiKhYU
zjM$ge6W(qk1DB#H_*GH)f7I91ugr4^EPN29_kjvXiz`nzMknx;SuM8Rxltd>%|mlJ
zGqc;?k9B)tP4LsL;N+{LL-FhZ9OO>TvzIKeYYFW|_(?~+&zBTz(;2$>u~HvCfS>T%
zJy4g8<>qUnXd6Nvy<RD4`dMEc3ZDHfhj$6G-L5X!n&Z_wOXYAV*m^SZp^EI3wEObY
z_skM}kI(#)+#ApKg1kKZ=(?F_pfV)Ak?!{C17oVII|W-ypUlzYO@4z;ixSn3<YE`q
z2L07h3h0TMQ&!^4vB|vm;SStC-Yg#nFi7in(}_ms5ShKE8<l!>8`eOVAN3=c^R_#<
zBePjYf2t9c>WAI&&&35AG!b3NTOtC^fcXhfYt(K~@xaK%Y$x)5<^n$&M^DT-U1r@8
z;y{J_vW|Pj?oRwK)sr%{@3F7<Hk1{zs_2LMU_L_L%yi$Cob+5am>3pr1}O(hb3mJ~
zs^}@8>(^-9#~t%a=nP~A7DVeO9|@})p)!cRdXBlMwybP78Rx71ohis9R9wo<<&ASI
z^xKn%wC*kfTg9`HjqG#r5M>%YJI8ZT^WAwQ9W=TyV3cNXz$zd+Op~@u^#?)`9)7mG
ze0ITJ^aIi1sAkiLr8{Q`;b{Ah*fut*=3o%)u<!!k{hm{SEeA1v-JbV6z^Xd0kRX(}
z)!a(1i7#G=SnipXK!2hY#6o8_I&dv_s=*Z8jjMkKY)A$#rn-Qp5FRZhgfpn=*5WX_
zit?Dih<Tw^Pve_MX84J6W0ZERyZT^`^R=%X6XGEP$wXD!o`2pdi8?Y&z6giEz+>I~
zl3b}quZ|)e(#ZS~kJcn)nhw_d)&u8H-PhH{|6Eh{_T^OL2O-vIR4?xS`kLJxV7p%=
z9hN))WWgGa5TZ`SGfAKMLwZr*ptt0OVT^f<SVI^lcf6gt<gbpwO_#)V>EAyTbkGTX
z6Atb{8X{B`KQfeg0Dh=t(;(&O{qgQ(VhEQ-*x+AW=*Zd%uV;G?;$w}R#)hLc!_RW4
zIBLL~X^=UH$7F;pGHj$Ab#_dQ`7g#3;g7741hSvU9+Xs8q`P}1=qnMli8ztiHR@5*
z)ga2RLZ}Z$uV&dLPcfN6@)v$SMwAX7&tupB^I+W-A$%`J9@kAAsP5{=u!|ZL7jgWu
zy{PB#<I>BZUFs!94)1n69iJKs))zxpBZlHzJr>^d&1d{Qpgb{p*S_pPlBqc_{_ijS
zDa3fvv<dC+iMgIwZUL}v??LDQ(0;aAR3B{J$sSspABBU?ya_j?gNuA5nq@^s@GlJ=
znxe+XUVf854?2EM+-j4oXNd?|mvx0z@SpsQFR@22UXe1$WkARCUj=U&Q}n_th{=VI
zMF9>o(AfQCPzlArFf{9sqk%z!!URnTm#>@Ey^I>npeCYJI}Jf<@%-3!RTjC&B&3E<
z1AbJH)b+*U#)7RW-1sIs3*h=5T*UaB>anD{a6M5*{{M1FH`R7JrUwTsm`uv*n1Sjm
z?5NV{Sx*R9vHZlUk{5Fu6SpkZ5MF$!+{mi^J8mnUn2=9hv8u1XX!99RDM)<NKJ@j2
z&Ha0=E(-C(jV70DijK+vbM$n6Jn}9Y)y3?`;x1Exe`Us;=o8eh20pq4cPEd06_ieH
zjq4wBF%2MLlRDSj&Ymkjw3+>eZbb5;y2<!qX(t785;s@&9&lwZWp~rN@<Qi2Y<MTF
zt;QVeP}ICYK~c<W0q3G_YfuP*jThL38>=Hs%}?fe=#L9#xewSA6n}iG8gd1u<6iOy
z;8+sqX1McTDZPZVzfEJA(y+D;x%<-|NzXCj;&I2g!jhsaaTi34|H1b4eqa1SQ~_6f
z@x2pYH5eqVAW1RW**kZPANbnxtk+?2oxl&MlkPRCz^Bx|TceZ)K)T@*(axc7#-!W0
zV&>JPmFLn(1=OWgxC*S$lkk$}tVI^^-R}?1XMA6Y4;%R`dj@$Qde4No6Z0M*hklei
zoRd^A9r3VG{fHBjtay3-)v{lbNDix#O?`uaY5wfTpoecK*iM<dtF`G5ES0r%T9aPb
z(aS57z;={HC#%Ro@;<YnrOgtAqj$7@KUdq|`y4EryLgfeg;Q4Wo;AR1d0O!JbSm#T
zhyKb6<pLG!IkXr70g(pn2__#sJ$(9qukGE%okGbVJK?wZ)VPmT2z!I7b>~;@zOanp
zyTy*pd<lvxtoASPFHmwaG=tmqUgNVH%iZUPDE^$^pC6>}x#Ms@TDRW!ku)~$6=owN
z9p+}&trXgUGC1%_S{)1l@m(Y!vb(u3vqUG$L?s;k0LYX5vT}K-K_ZB)qc2QxGD@Db
z!Rq0Lq6eQzlf<o2k^a^J3WZpPPCy6#FKGh^o1%nY(1V;g4o8%v^;#{HlcqhoUg{R>
zB+ahh33jCzF*=4I0ID1=E0!CSr=<Ajn*;L55ko}i>vPJN>UKujfx~Vf+!`V*t74ON
zGn<0$bSjJt+c8g>e<w>Nor?1DYguscA@%3>-N6xHKfjbJKSl-C0+d;xS=q(0!Ser`
zpaf_zqX{%f`JX(7LhtOIaTi9a+V-NAql`bGn$;I7^w2l|3z|a7@vdYQD_x>eFsP_-
z>Kyfo;t~C9=g?kI1&OF^5E;K$nR+V$livL}Xr$?u4~a_Z#_hbEi-#DM>eD~(WZ}Fg
z_4)Y<`~XYiA3%8O>qsmxS93g=e_4{J8r>I)(yR?iTf|=_K$H{o{T-uTB!W}AKb4-}
zjD~2^=m`M>jFS_PO)a*Eh<3ByCFhnd``RZqlY0&=TbfL}#95ve2VW5uK%lv<uxJ{}
zg8oGID$_dguhEN3SCwyy2s<+uqeyB#bzQ~$s}<-Iz;Yj9=kd)(`X7uz<v%ZkeK2Nd
zW3>u-1zjYp=4EeYQJlnv>(y?2>&mNAyZ0S6xP!(Up+~U*t37T<*ngJu<VUzXa@A4o
zGfW=pkMe|*;WJ0lISIs@NMu6G_S!T$WN@Z5G5UkdA*7M)Ydlhp*z6SU#<WFXk(g!R
zUh8PYDEcHx0qwu;X&tbYavVMtv5MMTUvps$z)h37NcJLFC_E>Z|29d1{dPqYLcOV-
ziwi%zbA-QrJD#W%np>kMtwZ0i6MvmtJ^Ne=LX#gB948`hz46}BzPF9oZb9AecW-L;
zd~i{q_oistgQs5BGw?rI<yR#Obsp0lEt9gQ|18?1dtzfUEy3R`!db8N$aPZkPHSMk
zY&)`)Ljb8e%p%y;)=6xo8ljFP(HjU4%FM^YV*|J<IGhODDQPEe$I(51p5M{L%Ii3S
z2Yf9XJ(QUESow#RE<nJMa>E9SMVg0xs#p9tSjHJSigt=Q$+w_h)63}!CESWLjRqRA
zA1NK=BYc1-yuV83j*mfn75v0_N~ECthdLjxX62F*Oi1U#)ops@Gji8!!iuqE12CQZ
zazDdGY>j@;$ICgSxbVqWuW4Tu<8VPGKiRHxc?y03(k|KWWPX%>ZS)Rf6Kzz)v2+7m
zBz8-KVL$&mP`N;y*qc5z5^N!9`dZPf7;Fg_nm+#z>(&uTM#jv(9ne}?lJ!Xno}XU?
z5KC>c?%xUxS|715D$Pn3t~*3}JM=rfWNx7CtxWyu7S2srZqzBM$&5T%>SPZ-w6*{=
zj8|PEj~ARPgU=aJMo^>>^boB~xe7f<wn8&F$T7|L$5JRZ&i8|TRA3M~5Ywgc@O8@3
zut?hcYnc(<%!7~dMIH<ty9`3mJ!r$b*39m=wk-3vRE%+@8y-b=vb^N#%}{cP^AP0M
zw-UvjN5sZQj!{=3nEQl@V@G^P=k^6GV4KFX>S(CUOoxpbbmH33w0<HOak+R+Fj!S>
zq3W<b`;7TMDgSy1<?S>Exrl#Qx9vUsS~<WIA`!_$17_#a(-eK_`f%-rHx7pArx+iM
zBw=)t8y>BM-{V(JUtJWXq0FQuK*kJsz%he{DX`Vpn<7*sf=0xq)P8?|F+Jl*K&pk^
zl=xEkMOJI((MxM_w-gdc61B{hycidFMfap5Iv?FWtnq&N6P!{gJRy9h6yEP5UbQM5
z-zJkAGG_i-xPq-qUcza7zAq(2pk){c_B9o^`tA=XFvo>p?ee8?<zS{E`yQgdG=pD}
z9&o>A=0!ndCu9LsL25*IbagQJwFc!dZ-hx&uH|$(`7x7^ZoW1Y1`J4{nN)R!R-o|8
z&eFgAn!y<-L0Ly_D@$>^4Xd~`>9uZ_n~l|y^L%^k9L0vm3d!|ZO|AKQE_j)qgABp_
ziA;V6KiDTUIehGPk&HkZ%lcu<8^xV)6pAaM3dVn=^^*4s*@H)1TYaB03I&OwNo0d6
z$GqCWbL%xU7X<^!Rl0v0Uc=#U`B^g4u##kcWQ%Tyd8LUYyu9t#?()R0d9TBsu^j!!
z5G&Vi!JOKRExXXY5nKYKD;NsB9N@{_d8*|t9GNOkf!UR8WC$6fJf>hmFMvginzm&x
z1E!5?m+5hLg#bD$dfP(6Tx3IG@W1)~IXMbvAnn-Lc~QM9pQ<!y3v>GK>Q{f>hmr1E
z=#DMsC8`DiD!Z}B!6={HvIswUV!S4Xt*6;fJamP~=zz|)Mog?;`$cq^&uu<fnp*0#
z^eq@!(NM_i3TyN<GWL#4Bm{ingdP818LJ(PIL~;xa?lLYwb7f@M<q==U`{RFhfPD;
zs@4SUr~Hz>jG-oL17)&(n)C8n(xiVnVYc(_u4lQoGJ*fe;?jX-8I!oX@D$U@xUfSC
zyMJXAcwd!e=_{vnC-c@eJX^-3+CnhKF`mdJJ<R=Ck1WfdY>-}U*SyF&S|d_E@?GKz
zP#cK`E}OxNne;Ys6Pcg$$aY;My6~&d56&Oo^k~nom{Vt>I4!4j-O7%-UfK-0N?*(F
ztV<3W=`=(*|E&wLHw<X5Rztn+PRWYN`l6*6uR)L>o3{CdOd_YAMUh^py{gmnS9QU6
zkEQyJjE7F#cM~DEdy&e?##JjO+h&q+%`XFtpF=$nkpd^coS6CUrKM$h!2onv*oGH5
zt_QVB+P7j;h|M+zn%Xd1lkB6nlt@u@BF9`p*43Os=f`0D0vJ{#>4Iv4%;Tl5e6}t6
zGXtN+{`lq9;U_+MX-a~cAwd<$%R>?ZhLIii@B~RT<}tZY2CnDZ#&=;meWt79??XFi
zYy8c`h;aIY)BMtUdpNU9!q@pK?cbMYk4APTl&nH@>+jxtHhVu`X<h+5xma2hQ4Ep)
zJ>!PD;&0)<+ohHjLw7Me{Wc+QDENZm&(6S^yuBbM&SqT5pIJeEYO#yYTs~8j33%;#
zZ%L8a43Ez{+_Ea|75FnMCqRlpgg1H(*AOFupeX$H$<Yo6EbQ71ChS3JLW4nDl)tB-
zmvD4L!22)8ikr~^pe6#*D;8z{afb`!7W()H^Mqt_I(H8G-x7g)-KI)8l3D7NO!-#M
zuS%BUD)&y~-g(KK*_9W-%rsqr-A~?g_sTe6-U-kl4>{;zQ6~Cv<^_KZ=&ehiJfrR3
zkFi+przjY%seHG1cxS2mZ;zzoe@s`vp+S`=8~Oh1d_+y|;p95~vjZ2s`^pyl0fh49
z;~K8J;{kf7kR@b|(~<&sMW#N5P4#NpLg(-Hcsx20>QcHyl<mz<uGAM&uFwqml^O`L
znpPlxDfvD!oV&_yM)&yHj9C-g%a_Cc-WzAVlU@uIZmeDXufo8lXSmStXV$<PMJf#y
zMT6<a6ZXFT^NNFc--oDoYD(19i#!Tu_g=)2PDDuj5_>G>^Tq93Pb#07MP<u90k=n-
zwe;K*1}@C$JzPDN#zr|)o0SP#smT)<D{1^B-3t8Z3<XMh&sGgy1$Ls@kMGp7{7(|s
zsvC>{=4R&p-S%y`IQS&xT{N&2+%Ix<ypyvgh0P7FY;cTzh983ip}+n|$WG@ZM=X{*
zvp#jXuJ!|WHhnNOdlqF8w9xuDGFomtd7%C~NlQ@Z8cU{lWBmgIwA<K|1$j_T&Fb27
zK^K3SL*@9}xS@Z@;-fC#kpJRpu>8(Bm!JM=RRvfzJ|*E6GyPl_5ZX83D8m-2Z1ZY^
zmx;_qdxYKYx%s&jfxwmSfXs10%R=qrhJ#h^llrT~t;gxPPPXWwe#rnG`nh0VipG;+
zbMNK8E3~A#)XAdq?8#rZ?SZejQHzjy+u);j@f_a;OxA?cpAY1vHLltCebITAmmN<m
z1p?K+W(C<7o6kbr8C!80@#$IR8;2y96ym>*n_JF<yie*bE%`*{oN=F5kZc>`6u6#{
z!}8|?|6g@q9@W&<_IrR@`hqoW5wTE7eXXKWz=6pyq>5q{3l1caAOTF3nV=AckO8%d
zLnE!ef|W@`ro<4Cc}lQIP!SLzK$wCe1QH-XNC*kZxhML4cdg$0_qXo4`D3rM&aj_n
z|Mu^h_Q^SinKQC|S8+F8rQvfKg2O(01DfX^uK9i_3eyI5IOU>{rvv=mI1}X{@)YHr
zA7x%8At~1IC!jIsh-WMQY%BA44RgrJ%f$tZ6t8#o-4yxStZZqu;UeV;qMD6&3LLem
zK{8?ACHdC4R<iMiozqj#!a7rA(Nj0fhI{e1hljUF>sek>mp+b3)d2EL0H%pBI+Mps
zn9pRid%xRu8x<J2@~xIvQFh>`RYM@&)D0MLlr5ZH%<{BuH1@fu#IAGN_O6MF<&a4u
zH$-U+TyHTsF}&TTBptg}n|u`nV*1ox^vi8t7ES+RMvwNF+i@NyxUVj&pHV1LdeB6s
zotHn0I%XAlM?j&6@Td0?Io$e3C%&F1(k=BP5PclZ+kDkBrJuul-0x&R*`$15$*CbF
z;x7xW?01@5aylnE39en>JwKTb2AnX}EJ~;)7}q?ggQY%p<8Xg*Su$B3ML+Tzw>Pnx
zn|EUKL(Dp$dD~lqv6)IH7AW1Wq~VCqc4S?A()|GJ3|l!ces#<`#F08hM!VJQemUr{
zNq97DbSizdPYesMzRTOOruy23l*pO@Q_!c)9`@#xd5gYnRX7eSWzUg|H8q@KVcq*#
zO;w0c(5gP)ksJHau*-$5hBrR4BP&WMlkeE#f?U$kT(?H?VgibgMdy81X50Z6C|xc_
zNz6vsBCVL?c_mX_m<w~#u7f5u)a6ekd(^jLg}ijMmW0h2RBCe=NyJk`ynKS^y2+`_
zv!FHpYfvvG;1CR`er?~|<&>~^8B{aUFmPsc$(&%=c17mJ2yxuGIj8&n@kHTGgLOW+
z;5%{JdaXgT;xFhM?BPkL?-GrTrII%s%yNYf(^7tUbf5a5Px9+O*X1O8A5IJ47ZDKL
z79Y|Pw)%znRS^DU)Y0PVgEFs<=+e;>vIKNmgneRNJ?X=aEHW;l%WghTTtz=x7h&NM
zQtCh!?PVZKqW_|Ys9|V$R@!ohRrQ%~q4&35KF`M&oqI`k-)73G+a0_dW!;xx{>m5_
zd+rOA-Q%zq?v$l(u}KKlu13ba6dsfnyaaNGrf!mx^0)SKP=?gH?nQ8f5BgX9;%x|&
z4feXkoXZ0*-OZL=Ep=9MmtwnWOWIs#Le=MUrAIR0(_?iNld`siah51S`Aj-zlbcMX
zT^zdEGgqRGaXjlb^L0g1zTvDHqsE6GT>2p0ifAOPF||YnNDIqf3kBO}Xwmp@#<s#7
zu6>Pg--=g<6cw!8w^!acg(#i*;AULd9_JT<vHUJGt*fc1`-*aIS_msWZ5h7SI_}ER
zpMW5{lX}hR_tOmJMeK6P&+<_*(uZ}$QFh8MvNDboEq~&f2X1A#_3}OFf2S;?Xs%WF
z^?~Xa<h2vA#PO9ehU5Hqs!{D?E7%B5OQ4CWOLYO8MGcFKh6w3!EP^z1V<e`%btFkS
z=)QQ5%O?nH-&evgglu{awJfz`F4R@VVNa?VhwVvU{8&{<i<7{d6e4D+Jct<3fpiWS
z7w}sc%igi;vz)i$x0j|JYky^Q$;xr(XO`u8{ucCPOGlfueJ{&losz)Uo@sa>#m!?4
z;D6S!UQFej_&(4dJ&)wpcK;0y6?WBEo@g%1!Sh-4jK7e@?vXFxzyMaZ`-oH+Ih*H!
zZfMwTFWdy?NnNF=rX|1J>5>doZk7XWn6ioVN^6OT4_+Pyd)?)H=-H*_LnB0)&vMr&
z<jHE#3HBs2>IKQ%7x4^1dEo^)Q^KoW;qc~>sf81(VRfwC8$fKopP+qg>6EO>RNu^g
z+h%GU{^a;!N76zxZ<uIkPXA3izU_=L5Y%>DuNh4`m5+MUEgsLr%p++B<w9TK+tB^W
zvQy5}rMwtqZs_kvRJjp_hrS0Q)6C-D{|z>n$Ldj@)L-aZy)0g#AJtOm3mkU1?~-t|
z_%)~$kuJijz;yhqn$^EtBOQKakunky|70PH^Rg+{XUkT+b8ecCm)y+bn%mRepS@|Z
zpcz`a+n7bqZ&^~U0Tkbv0+D|SHpa@hSpw8#1#g$T4`J=B?6oKJAs@fOp5S<z3DX?w
z-Dpqh{ch2Kf%phiy|DXw2b|5>6@kldl_vPoO1iPjKjm+o$NQK5R7=Go-qyiOv;iHK
z1NDy$ZrNM7-RPf*+V#sGa&UVK5HK)xDCS`~A<T{G-oMd&EfC}*V44CV$7{Eeu=po0
zJm`k)nG7FCqu-l;j?P?p=D8Vc?YVd)|6UiKQRJH@?e$+C9toWWFL`=_KK(O&9p1YE
zmpe+XGa<WAgmJPr(h?r<8z<8umu8YZ+wYQbz1x>)zqa4gF57SWrm1OnMwdSmRy>f}
z6!G9!<?S#%An40vmi}0y*p%<(PkY2PAAim*BqfFH#|%^m$>-WLW#(&sw9cT=(N1Pn
z!%@qRJcGeQJ^R$=HgV~^@7%h<uJ8oMYl<YFd{h_RA$jVU6nc}T$7*1iV0$H)z0k87
zOn>RlEuT1o#5jIl`}oAZiVrdQtstf_my6XzcW}M|0Klp;V<)iA(I}N2zJ5KIvs=(R
zOV0~eI2jeOOL4xMIC(!aVn}uskWuJni(VEsGTW%SufRt<FV&M1DYO`DXTXBgxNySv
z+U!15V5v4jKF2%ANnQoNNFiPajrStC6Cu|MLatkzx2&G2JzG~?B{Fn=J7qIp5I%bU
z)YKI|ZPy1F05q2fyi5UPPvQzVm)?-ap3u-Yg!eld-48@--lrEu=S>=7qv^B@ZLrqZ
zBjW{{)pX+}jM=<h)lzQM3H>z3ZHBAI@S(X#7Uy9*5-9p<RcJ5sYs1V0*1i`qYgmy#
zl=U`9-&9J6O~^0GJttmw&~H%FAm|clkSmxQnQ71mfPoIh$u-SBzO*+t0<{>`&vFSr
ztqN}hOo~ulCANgdU3NAw2O|;2Y!EJ1q7%IEQntfyU1&t##@H2YSklAYoze`8@jbN1
zs?!~Fg7<MinR@1ul2$mWw&*g5FrBpfa-pX8BpB3wS5|svoucZ?HQrUh&5h&68hE&?
z1-XzEC>K1~<kFJ)8Klcyoi6$2YXFOdm1*eiY_f&PRQBy#%#iHKIvH=55$u>GpP62e
zu3E;_IA9k?+*#7d$~jtbHy>0?-%qZZHw-eFxvHcYO{&dJH5pF4VQ%F_Qc}*YD53EF
zmSABe4y#}f)a<ws_FU`YPu<}BcB5FN;*}wt%qI<@XM<kP78s=@_#%BZ<orKWzn-U~
zuSf@99gKR~mh-I6Ec;jRPu>q?>x_av)nRix*)l4@2`84~7Njl=cIY!$vpys{u){9W
z7n{{TB}f>pJTrVT55#s+%urp?5!V)6`ug2npgd&;T&yIY!_vyn7cGJY5Dz||f#e28
z{|*Op#8yn01MV0m(}8D1CCVGcLp9cyGRF+~fi85}Z3nCi?rmr}_u}>(H8a`5PU5vj
z&Ft^llRUMy2L@zD!JdOF73L;pxt)Jd%xcR22J`7RLBoDO_UiH``eq7eaB5_kKXER8
z$^Na8POfge5(d0v^v(7!oY9oZZvJXa{=MOGCLuKL0urrBl+UesEVu8_R!-)YT8>=I
z&=e4DYWiqFtrs!)+HAZ7_wnwv%^rCF<zww%w?KdA^q9S*6xI+`GWjq%KyY2-<*~xD
zeG`ba4E-QzqNlG~XC(Vu_7TFre+X4AK=UJ=0t{iv=*gy9Qm}kjJZP#;LrYnsr#12U
z_XgMPRUgfF(M)G0AI`6`1ZK81xM_vxfOw;}5&x053#>&##1c?@7BTq@)qaxDT=FjE
zb3LeAmzOvuP<*0UGZ<EasOb)$JDpFo&=GS|2^#Wc7WaT<Hg+yzy;69xBH1T~o>_zV
z6S^#t+jLs5Y|xuFZ3s9<o3(-hH$L1BVG13}qWD6w;T8kqu+7x<-fj{sI_e^j67{_n
zn!U2)JAuCCvh+5<eR(Zg-p&{9yCMq-QLaGt4JB(3tsQ)Utbt3v)OY4PP9uQkK9i4o
zO}z_TR5+p!aiub=BvA5tTYm$=<SisCOR4h>rFn+Us!2X8Gme$!o5so+EJ4+uK{qA0
zb}dJrL;xu!pL&7%j?B&HLvQjQ;Cp<Y<=+$(d$%I$Jl_c|a>AnKP$J*7;<+btfic((
z)2qD4_iG$ECdvF(I+q*hCb-lve2yQ&l$d4RmXkG)Oqxr-*WLo(bCukKtMKPEXoQs*
zYoL==T9SiyD@6EiDv1tnd`zn(1qGL=u^GP7({cX>f>J+ea!P?+)11V&W@I>Ge`SeI
z%O8j*Y|gKa;Ek2X0Pokr$!xDx!?C!Pp|!wFXmW}ESdjdkm?3uKODyoRQJmYri(9D}
z491Q@tyf9-RZ+Cf%`Co%VMUOM^vgI%|2TbToA*c{<bbX?sT4diDWl`i8mxST@9*<Y
z8Nmdx8j}2$|NT-qIY4gAH#JtybX*LUXPF}5tU26qX#7F=nW3#P-&C{zq#r``ZlvO0
zQs{Y3iU+q4)IDz+29s^t6%LERFaPv>HX{hfHL<%dlkKPp35W%n`UZGKiM$Rq(4A(W
ztkpfBDd~!6^{$v_^K(0HVTN}__V(h)*}E0H?536E#E)Cto`se9N3GjJf=i`Ja_z-G
zZ}|PuMkHI3FO<Iw&M*OMazFKmJd^ve*B_fXci~SXTp6m@pLeaS##sVodvAac`DXld
zXl>%dP=?+MN<X;?zoGiaLqIe76Ydo{r)lR@pUo-icAzvR=Q?cDU(FXM^&|n!oj-hJ
zXvl0}oPKpTk)P&ED0Y8yKTeWQUvP$DM5r`2Cl3)a2#)g`%>F<FL9tNf{$COw@_zW0
zP=Y(tR7Q#>&fe4<W-g-%PpAzlalF-6z(5={7XXm_R%dGU`J~xO-|V4{>&8wKe?4&n
z%4h%ck;FL_X4U{&_7(;tr+s2|O~y98&!G!H<(0Y(VCKM1VBly1TeVVp_3_z-4=vxr
zuklx6x(=KyKfR2WnIffp;hF)~Y7<;{=<|g@hyW=8A!yB!@%=QMW<Upf2FA%@aV$qH
zxb;it-{fQ>Z411VZF@R2e1j+itMzV7dKTvQSQxPnu6=j8M?cbbf%X+>_^E*dK>S~^
zOes63Vd}RZ2PXJ1w2xohfekYcJl)Sc)$Nioq^wTeuBj|>GK1hBcOb=PvljLq{2pq!
zx^exF{btq~GILq(zMD#;i?z$*-9<5T`}oO>w&*e0I^W!49OW`T$4^5gV&np`h$(ZN
z|F#J4viR<}e39oif@-QGsSs_p|AvJ}$b!+DpX8n595GMqVCBq-_jG^Fkd}-h3S`yd
z1g0o!Ol8r;&YsU62z8G<%+ZsW{5L+I;GP&3=kpsE-Ya|<PQHhkTlfxE`#_hX*xbfT
zU1WJ&XGEU>6#A&SzD*2L&yp;JDRY|NXHmXsCB|<&=eBo0HDJ6T97h>v;LZLttg2es
z#iSAj!5iXxd;$vQ3qy5b$Fs8kfK`ME8&5o3@UZ}E5`G!dJ~QtZPElrmlRo#91-mSM
zdHTO?<~pdO-weVo%=IU3C||)WoP}dWjWVXRcrFsZ2B4ioP@D8qZkgL{goie!{J3&^
zIt!EiUKH6I|Dh$<5f_K4RyI6Z*dwpw8K1DR)dBR5T^+B2cPCCalux)8XWxZ+O(%Ke
zulaDtgI?Liv??9obgKt<2yf6QThJnln6sazAL76*$YjbS<cAAytJ7)yiDkHvr<1cb
z+~;;SMJIun{lo})b)OxeEC~~015&`p3;LA#_3jb*)d+Zw>n!wIPes$y@xb^E_jIt#
zOL{m}^=aJmgUi=VhvJXAuVzk*y!xCKemxv8skNFu+V8#M9cd4*MhJBw;J-Mmo1+0!
zg$+V43#H|XbJ-Qr>aZqN!XzARcRyV?+8m-m($jHB(q%l(nZ@?U3&D}l9ZPZx<?B;b
z{FSn}z=%!D46)iA|Gv~UzCre|+7Ui?zYhkTm|r9}u&$o{4>{4BnB2Zx^Y+Ug)hp8>
zM^d26giCohc|c4y@ID5G3vj<KW-LJFr1<@oIg2(*yE9IGdMZLssp+}UZ;#b4?bn}=
zkuMNS7k=qULdM<c8a$SH4<?OTPtCenVFL?~T?{T^f@%v7xbG~RnuripF>ecZti&YY
z$)YN<*OqK#)!89WsU)&JRZcZ?d;s&$CNxif{SORIDC{5%s%#0aKLrM8SmHu)gX6pU
z3=~+ey2|@NFnba^tQ9w=`-4xad&N^1GXE$v5!@sz3q3bC=Xu_$gR^W|qaz#j#{zmT
zgm_u@PKsJs^aVU}c%yXrbi2(9YL%+Z`u50?m6&#fDB!rJomX4Mb&49AqGa2w#O&l6
zTf9uAThgjO7|fvm!4H>oBcLz#TwBd6^@v!zw{h}`s)lX`VsbtV6x?W@3kU%{dQ^>j
zkM+DfBbc0Qxx{O7d!iaxk~(=y;L*Pas;v2T@rn=oqp0pPY$}WNt2~`I@GJvGD^jWX
znf}qJ>K@s<B>DliHDuEqL=<E+I)%;^EjN_m+#{Ei6<gs!HT8QIxX;0@m~Gk&W#C8q
z!K#^r=t@Rii3!p-_lZO;jH>)lceb@!ehqNm6$OqHOXZKlKqO*6$+74BM9Z^ut2T9@
z3w5|xW8-sEDbJ=P9nZ{wv2zp@5o6ZS{Hh+T%60#kcSmX)nh^DP_L^A0R1lUpoz-M#
zzVPbZi5AZ^<j|sCy!HR~zpXSVL8{`^EF9MuSUFD~v?HfQG_HcPs39Nw&*u&3w(ld7
z`r<~xn9EC%SNC?L#uiczV;c-RPZeso2o2eBa(}j;`;VrV8}Zq)(gzCg@0(FBJK*Q}
zUglhB0?IDBjz1U<Ikt-3@IH0Gz|SAC*bQOv%|B?fTamr`mEyV=16u4o9ReOMlHc`~
zsW1*yAvJ$tv7vI<eU|oeNp={MCCl;js;#2><T|pDNe`y|y3nGQZl1EHzh)SfRAA>L
zu{tI>1Wi<w`<Xo~VMC|Z?iUH$zdXunqD8_0^GOu^^(#*HT4DcF$eSF2yvYX$lfYi7
z!HnlW2va6GGaTw5vyjYyiZD><`pm(*cX4(H-B!xX03#5ho}Ml{{|lYMOobbTF40iO
zC&pGy9*g24iSE6jvx$}6-$Yk2T{!B+e=q8cIUNgdBoUn(OaD@E$*tAdcVI#X&FWL)
z?ZBw~<*HL06N`>iR;N=C<)2DeH`R3SjFKA<t?-FG94~Zk<=-!H<o=6~HUqX%t}nQ(
zZnwmjJw>s&noC`R>#WXHYkzO>)rMn+7Iv*N-XH&}s&y@9b)*U(C@3IA5GH6p>o0f|
zSp@lDCtiyie-xz)I*T8J4w0j0we25KwV~6*98a)}isNby)6HZq7BIdr;f~A+`qm`R
z`Js&G^)3y?4*eZW(!|Z72X2#<YMUA|`H=h-?+qXmA4YF#xc*VRF(b`KE%ein<y?nr
zzuor{mBv?#TJokvLrkRWv!|pd!;o7QXg_L!`M*Qo^jqF%ak=4@;=1zg2a~=$(#TEy
z6ov=|mhbkQH)8~bN#OG}Ti~23QUtNKp4Gi`?g8<3z(LF2KFE;8Rt#jsa6g6tt%*gC
zcy&&a6UpOLpX$aqx<qFglb=GH?EegH|6>`&|ErF(I;f?HlL2K@zX8CIr!IxtbXNGt
z1gZtpOoLnAp8;hosNw(yHba?~e#)kIC-`psA_g+BT*DD>;U_ck#y^_@WyFu?ywiF4
z=G%%(&}x!moi60Wf8hb>N<XM@vQ=b>F$_OBi{HlV_hm;NTXJIG_6Dh{xK5!&c?8Ja
z7cdUxam-~*DiidjjK{s^^Wl!y*4O5d{3Ni$iq+`>7iJ~{tRTJrYjis>Z>AFo3>Z+K
zF5Cfy-Vzrk#vZ4jg%2uJVlLA*t8>M$M^T`kV&;PcN%+RGToz8S{Rij4L;>v~2_@gl
zFDe=b7tE)V<dsqrJ{!gD-fn*%BW(KCX)W+E+b%Qv(G(a)yCv{BhksF16Zs0c5ilxP
zV5vAo6B~ZRum-3&A|anBJnD7zD?38v&u1<cL(3+|iPI$`8-N)7kBRt<o)?t$B+w}q
zHbA0+F>aev4fdJ$Q=yg5v7<CJC2u@Xn{j4ufTXRz{$D_l$44AnKjP>oipq?k+p$zD
z1C`nr=53bKN;3_#;1K=0#VRY}J8(XXGW*vLUjfdyb*XN{A4Wh>Rl|nW6_Jlbgpwkj
zT}T_aRtYrYHbb>Wvu!1z`(ME6b79UYw7$@9YeyH~Uwym|7%<Xl8L4LqGz%s#iSN>H
z81!HC!#BQfBlzD0Qdrw`*KRnj2Sh%{-d&8k8ptROKeng#?IkgtPpDMpK{DB2L1lAL
z>K;QNX6Nl=lH92E`i%wk6l0_`feDXXYcP*=#HnB?B+ji+Wx>z(hYIi%ox(0TmQW>L
z+eGrqM~!kr<64AAW#UwZl+ZW@o|K>B4>2PcPEiXY1K=Z&A4&B8chlUd01du*>{4C*
TaQI8zP>vk(K3IAn=<@#pky7f6

literal 100881
zcmeEu_cz>K)bET=bSZl8HKKPy^pcPuB1VW7y>~_=O7z}I7(~<wiEhSYi0B5<nW)h_
zgE4a_&wcNE?_KX7@cwYG#X7TQjq}~-?7csGpU>WB;vXC9P*E^b0000gJzXtR0DuJl
zk_b%pcTnh#$;S^Qb_P0H_!0keso5ThAIQCQpZfv;Y_Wg;fGO)o4gdfrKu=5KX;9(r
zVqhWUS<xTdXdax_DVK{Xm`u|Vz!wMp_?A;6)%Kxbi|fo(XLaawXW)g0PyKW!tR%$Z
zV!y7gj(q%)x_Tz3Mm)FfNN|6!E#p6wQoj2f%axxii@)q|)ooKO-UP#JZu09+Ce22-
z_N%UnH{Trq_0oC&|L=dp4n*76t{9~A8nO39>ZS8@tipKpG7Jgs65^_t?*L^<u<R|s
zAmSiOd8#1NC!?Qg2*d8S&|4BHQx%f4QX4X@6uyjd;6cwdM3)EvaV%)XhL@!E6LrRs
zeN}LBczxXRP6ixyuU@Gz0GHg8M~IO_DWdpLOej`VkgZigy^vl;-qAAJ6O#k^`2#=5
zVGWSHO){xb;aAov3<)8WXZ}q0%X))7NwTSZm(H{ZKW)EzW?g^&?Fp&cM0feLLQC|t
zi-myNsv$R+#((57Ro~y|%%(Vhraa5Pn0T&fl+Mqy3d27_LN6obKc1xbCj$}t$5}1&
z2a<qUZ7c^k4M-4PU#==_mTc%aF5HQ_7IuJ@2d`ZwP;z`$bO*3LwC{m%P=Ny?`-I@%
zvNjhChX8^(;iWz{?U6rz*XZ~j!cubApNRSLgL#t*|I2dx?s#dG?0Au!(Hx(AmeZ~P
zESDZdcNJD2au}>{S|qzuDO)IZ8La1T-dndY9gUjVc7+)HawHJA=O2=~!8xBoqTa5y
z$~C#u$vOiSG=YOA7*T^%KmM-svH+=&5D#-5mdu(~c6^85%?DT(ToNu_nu!TN;1$Yz
z|KL+`<(|Sy^r-~+UL4Ivdz8du+{dCvqQfDn{TTs#hc+?PcCK_Bg&e}{0@~S99!zN<
z&zg>hH=1D@>oXhD5&feQZ0)5_yJUt{UnoXAT%Wo1=+Z*5CUlrun)D_@4lsAaH2TTr
zVA~W{C5`P|GFSYyzw9j^m_2f7n6R3Yl^hxQSFY$0ymAlT9f8vO1n}yBG4w?m35x^A
zwI1BbwF!ff9Ce3jZK1<ZgIY`9Xb5s>4JzI&S%-B-GZxj&RiiJErn9GaM_tR?T1VXM
z;w^_O8D~xFh2Wn0Dfi5iVdlXYyRGZ%IE>d=h*oRi^+6i)!j+ci9MGKnF`4Ix`|t$0
zx&?n_e>%74eBopW|5W}1R>ukgS{vTren%j1?Z#NV&2*jKOo=6f2v{G7sUfxB^HM4P
z74+rbO@t-l`fveEI6fEFP_>>E;gpT)sn}IDy4J}z^panewyWF5N0dbBWh#9zH1ju~
zL{u$8hO32t1PHUWCtZnxtF+20%{SW<zUJ5JZx#VC<dR4XMcVcLgkLd(w{Q|H#by+M
z7tFe#FXZ6YiULTC4Yx(MBej6vM-GwjVHa=APWiM-;O%-5z33tQK6>W--A9T46z|6u
zDzNY8l<j>(-^bVoN1uwpffmW(aoX{g=JM(5ucn4y)>Q_78@7}M$PzCRV?LgSGy`|O
z2kqNPdyqAjnx-UJ6<(abDh6GLB%aFw-eUh%s$8A2)5+cXn@i-;GPPOkIEHOrk}as>
zbgqQGY&r`PA%(ZJc?Z1T9u5AdNYFQg9~B9%ZjjJ>&NkT=OW$r9v+Q5oUB9j`rQSid
zBwXJT+Xh0{b_dJrJFS$RAlSGKTmJ>xQ5En8+i#000NQO`VH5xZu-9cS>#Pn%jsFQG
z{J2=+ZL^MEz>AWHJW`6S>T$*<m&p*ze>-hUV7lykm@fV2Kh=ZzYybaB<z2G?uv}UR
z3cj2O7E4|>Q^X2n*$doTcct{bsY3Wbk4e!i+BP@e+4R;jURbxDPm7EywfSJG;ZePp
z7Dz^nW6p?(Tw#~CeNWZgAH}c3C7&ip;LM1j1apvU3JQaVf?DApiY|b}(*t=cSAu()
zHl9qw1=M=TAV)livET~u)?(z>7x`C7c}F1RI}7Ybn_nwItw1m1gw-AyfOxt0M7B9L
ztB+>iB(Z75o)^ebKlv#;)Qcp^fV0|wx{`Kq;>{1(i{QmM3z3}ck?SV~{s&FK!<d1~
zvhZ5`^S9~YMlTAlK?XnS*<QVjxl&60MJ+vIXQ{QSt&uKG$M_cjaN2nD&4~IJl008y
zrc=KdeB_mR!6r|J{(XN`fgJsM>qrPL*KS#9{@~EvuBmy0=0oH?Ky}tYbyNr+^fq)u
zYrDIFHO@-NlfUs=a9c?i@9sHIi{%O&*E_&8i4EeKH*7k>bS&hHKsWQumxiJaovxYn
ze0mvK|2GV}{!`^)$Xzwdk`O7uyn+@-^ohwdGqg7Z_82nj9N@@<YL5#+gt2#*J<TvL
zs0{1B2kmX@FPlhoSq(!z=i6Liv*?r~m6^AltRG^M-y~HVfnSJhX6r0?N?_eD4&;$-
z#X#0)!q=>ak&JCoJbG^&F&xEs2_*Cw(vBj7{;GlZ>{)X*0PAVD^|X5ZRbw3dxzYij
zZO{1^HSR`d*4m}oUf+mkiV@~dzPqk>?BPA4xqS6o+Fq^5ru$<B0iaw?a-{#M8vWR^
z5m@ntDN=F>>Y^3>DFEaCC>S3Fp6S77%TKZiJvn~2rc2u};i8Ws{Sxs&-1Wa-j`&vt
z4|xxh?v9pFSZl75UPkcL>}<m$MZ%LDw-(sScNPu9SMyvRbq=0g+HTFHn_jw4=2RGQ
zX(w1@CP$X}*(d~}TqfnFo)=llT6Wn4`{oUcsI7WYl|{$`6fc>$WDly?UTBEkGeXpW
zJ+qU=_<SUPHJt>$`yE+-P*Zr(TFPGAWewzaDZT&KNnZVR-QmBk8~YbcT0NuzvSREU
zlPi}xW9V1yH9?E!peep6U*S!2xpk%yjVxc+?sBX1`$@sLrRf!gL1#)#UT*LGN<SMJ
zMh9k{Z?2%Jj<=*wlWCey!qiv*W7tu$Cl^Z2*iQ4&o7I6}(r-Z{x_Fqqv8kp){_e8s
zB;|Ec{!efHKQRO^h_Ntx{nX0kkNd|o^Yu?h_8>sO;np+j1C4E-0f&<3y}_en9R^_p
z*aP7ar7W^{4y#qbvHelYM)4$E<QEXA{EQwMfKArCv9Wm3>pBZam45O!9PT~9Cl_c3
zyn8keqU-gX^kE{LR4oiq%XDQc(QBYMwnbdp%v3mETIJy;{0LDYY0@Nrz=EiWfdM@@
zWs8I9T2h^N-5#EZA2FeMxJ?#eK$<(#p=9X0P(37RTprQtvr4tq6L>N_Hfk__d561n
zyx|`7wnVdh{UXd>#Mk`hwjPq~KjLzJ7#_3!TIGL3!uS7No-M6z3KRMGs+MGEKah{t
zC4CS+(F-$`V8roN;SzD%GJnBg`CTH88bDcR0>r5`IKo7@^;Wo0x~S|33oV9Xyn~S7
zy`<9)?=#(!zgJ*{fM^#PTRIYaZopefC~(?NpFj53+k^1a$>OJ@Dki^!t2}Kdc;h1b
z_e0VJXYB;}hyV2>__vwUtgu3QrUx}7HS-7h&ZbgW1JeU0+>|FCy0<&Y*>G$CMvh_<
z+3kO2pS{=eWjSjM;)oYzNeWxNjGuEdltB@z1Q=r>4^nF2;4x@#GEX=uryF1)EC84#
z;Pa#yEuLc%-RV=Pk3Ij7&FSxm^x05)<%vyUJ=3fs5QDA_$IKneXf&=c&ayC$5o39D
za4Xek+Ld8?k!Xg%l3E#bt$%3MTrZ*5b02G9bZiBuM&K^qBiwer#hqHjJL@PDREbC`
z_hiE>;Dd-_mQg|pe#_aU9lhv2Ab1JQc(_MBQqo}XSo+s73pg-~!jUe@!}tgto*D_n
zFhCqS>t0s+NPlc52>=EVIo^(?{wZmC&Ek2*M@72|<B#j$I8?Ch<SK5UZ@CpjA4GNB
z*n2UpX(eAF6O{)iu$8lo<|=L^3L<L}?Dn+}4%cG_uEL1aiys=!DJAB<?BtTvJ90qD
zqRI!zML5;;DQT4%aTV)q(*ywoOA{+&A6TP-as(C?x-dQ?U24E--zT=>gJnIQR43qp
zDEkn<BKCz0dkR1{7DJO-y!~uXG0mRSUafQ$HbwH>415z+^q{Z+g|+!*k{)A=nU>Fy
zBS{7(+oN$QW^vEM4dR*neG}>zIhSu)Q*Ir7ccqgeY$iM>`2K3D(SDlW<J^KYVr<`^
zugVCY4*LYlpmbL{8}3NX!`zHeRPaT_yQuss#6XuAm=MgDT2EBB<K+PVu9;~<{EL*D
z-&}HOwp#`gC7ZtPyuEOFQwQdM-xKF>+0M|o5>ue?6@X@5>*S=_rJtajkYXjmNH)mT
zD&Yn<02~8ee#y)4dCxMAK&ww}XpYI@PVqo+nx8hG70smQtuSsIiF&o|e7GOV8Fka^
znpCl9YE;do*NI;sZ}v1%mqruU_sZ52n4>WCOQcfFet~rXqZXQ>|N3F^l4JQkQ@5}A
z)HG+3YqID?Utknz9Mb>~liw^Y+HVUwnA$`Uti&C(uhqGVbFt&M5$}>MEq*}Cjavi(
z)ZT~Pg;<1IEdRbEbrZn*_NL}l@fCeqfNhcfB|98Lx$Nr+&<n?>G4h##M9Bhkl((GS
zl+-aLb-l}mHHwAC3-a}Py9{0`!$Ow<Bmr;3#JqrzJJmIM5Y`YHP0yIt^p>i~_s+N)
zgC-XG27(s9XVa+=@%vcU^Z_X_j0dmj=lUH2+4Otw9JEA196`AovN}fYoVJnu7JzAW
zX77D0QQ&JdHPp&mcHAb{C>)8hhqLj7tGFmL&*Z@}7R2FV<v&E?$DdTaS7yUUi1ktS
zN5Ab>HdwgXMjW&>j}H<yvpNBVq?l`6kG^&>okKPHTsef94d2@nF)bVT7J~mp=r|_t
zT=3A{_p-|9nm*KE67yX!qpcTRuR1sn;NY~|Az-#(P^Bp(4ug4TEO)3u{kj|#$Ekx!
zX2RlbE#4}WU5L<9;3r_!+!+B2#D#S>L-esI)`PrnQG7xqK*;z5Z1r>^E+23_rv*d2
zl3vQG=Tw}DB(yg4>)<dxdAHUJPl>+?|IWR1PjeUzA}b`{dY}A;)je@4PPXUs6{XK)
zWMt{ExBDAI2gs4j!yl0r0NvGPAc88S)?UqQwEV(XH)XOUAD5YXhST0o8kk!GrT{lj
zA$^+tbjoTEF)z;Oy~w@H-X3eK8$^LNoTWJ)4SlSd&76@&chx<$Oox#3cv5$yw5ZLb
zg>2&J?H+VL)2a+VBe~SKtB7!?|Hl=P*tQu*PUcA{TUXd9=0(3|#?^^@oZwu-YPiEv
z=L9Y(XGD2c{W(4Otf`*Po1|S#7?%5|Krj%ib}Nj$r8Tn$2kcLw6Txg`7As{qOBGUV
zIriV9bC2qDG3IIFJJjpnRDTn?5%HnxUAcQPHo4RT(C#EJ(p|gv>(RFC1t7KWTZbxC
zVc&4O`{c=r<4s53NF?}_^{`jIFQJ^U5qzN;p_wDnA?)k2j{%WeQ-$$#ABYs;>J6Z;
z4+{`A|9ovu3>M?l%YS^V4Ub0{wu{Bj2GTN6)bTjZmyCzBh7Z775nnpKF2=QB#ZZ+9
zinm{fqf#D5s3d{X2o9aC0<p^!VXiGBxbcC8Co$nCV?r^!ddP<3KKM^93ixp+#twDB
zdD9h4xO_@9uKL-dO^GAO?5Hq@^Cc061Led0^6@M)Jd}%Lyn6{yH1qNFMjC~tBz-Ma
z>#VK-;7od<0MwzCTHm<y-XK0KgXUpK!&&gwBUDKgh*rVZRhD0NYrcu)8`~$Hs2qco
z1467!H?fc@j#A{?$VIz@hjA!AmjTZl=461U;#K_LNS?6U`J*P_DFvQOgGdULv|R;g
zb{;IoxkwULpu1@Ho$C;4x50@JeoLISSp$e5FHFn##Lk;TSe5hog)mi=*TECwdPwKB
zv$#Cm&P47%7tc{Tk{OYLThB`O4j-?>yHn+aR1O?;?-7D3gTHcU38T~m`K3;g-^2!%
zs9B4QewPE+0AT5ey@zXkBBNMXO3M@M*<odcMbrVq;Q8RCcQc_Xad0j4vu#WsQ@|h{
zky_qwgz`{Dmk<drHjA+kO@qq3%|^pxK15klfl;t+^3UOQSs9dyAHjW#zkbUWZ+uAl
zxn5AYJinsy+vx)@;iQaP^j@NCQ(R)eE}gYKu>z{F-Fo!Oxt9JvUwh$!44ArG*OzX_
z_WDtp;GV_{9lU|67gMn67;?@ehhW0qTB-!?e*y8-5{*E-zb-#Tq3tw(AN%G2)Cj~O
z!6_wF@fzUpg!uHAIj8`Xli2oGrNeIPWm`*TGFh*3Zy^&NpAn64NI2c$?Vb^=gh4&+
z;Qm1HHhlxl;cxDbDP4{Y-Z5nl!bs4phaPs?%hq;!%ddKVQP>zl<3}{v7Is1A!5!|&
zHvnEe`o{~Lu-K3~Ty2z0l+NcsWQkN8)j6^ezLoXQb9kC}=c%w4MVi$g{|QUg6OE0V
z8~R^1olh~$htapswqmXmTr>BGJLKdxnX`HIl9njk9xiN-iVv{!Bd-eY3598MtC*bF
zyz_M~zO4>nGqeCMbDc8_7iiaPD^=y9#{9`W2qJV>*GS{y^Et$1nX*aV5(1FuCK;$9
zO!hjG@2p@8l!9?UnjsReV-4O<MD@5ZdBx_~H(k@j8^uhsthiLuJW+Kt^|OBJ#Mnlm
zo(^eaiyHoLFfG;<`S$AgYKLoaaU1cw{5q~PbRdAshtLc{e`SI?YXXFN@5vhELC4jO
z1WpV#F%NnxU@z0`%1~x5OBl~kR8MQL3AbK4k^0hynFKGz2BRlvX(&A|qSQX@HgXT~
zCF<)Um&F9ZuVNe{6bGu_aia=jJ{B*l)(kb>jhcL;Zl{U*Xfvt7Oil{GW9RkCa&Rxq
zgtlyiCfV&g@_Ac^Zgb=3r!BIFT>U36N+v{qI7aG7g<%)&xr~f2i%mTPD7E}7;nf2K
zQ>0biwd-4;MkMXLd^CFkCuXxo*`a>w&(ftPFuiplhPq)lYFi_3hJ#+4feU8b_eX;+
ztWkIER)5#&r)~dHLfER$R=|G-|M^LYCw)zWsqC^Mx%p`f1xO|ixXpjvC_8Sa%n@s8
zM}Otv@+3>O_k@*6?XrcQL&{Slv}=@W_V2YQrfL6)D;=JA*8A?g=R`+Ix2ggO0i??p
zpw1lV!%YGhXvMiotwk(uBw+Jf=p;x-sN&O3wyJBMxXlDo4j17>;-`pn=t8N(RAfkd
zJz7ipLDn0>0KIf>8BncnD2rYERj1Z=CnZMIqh>xUo6rd`Xz*SN960TXFvBVwwC%cE
ze|>&A#NmT<^jt@}wxyoH{Ef;QKTVPJHpXWGeDg=sd0ie1kjsDglaQ-Vnd^k2=H9@O
zAWJ`uau|ms{TY&IW-}*L4Pqi?B@27!3}ZEZ&HA}0P~=bo@iAh`1}>?>Oca!-OZv3<
z>3!lLC-#VR$Rq<bKGKcyBPefzdspuu3_IVKmE3?b9vm{1mUV^obSgoiqp3ckTlK?$
z&(FyVslpVx1L>{PQBjkHuFb9=lB2hU!9`Lv%higy_*25M-?jmAnoyQs5e5<;sEc`O
zR5l%RdK>x#G<)}U^br#>4TI?)J;@5KfQCKFEDw{pUOIE74w=^F=S<AHEpi)XpmOtM
z*wBvFc5$PTS1;f191tMfwfxTV5&Hf($I$~XBPWTHVuBHgoZWwFm5K*Qv0-$LgL`iW
zYOb0X(Lm^jk8MH+1gLj*9KChIS4t!Gw>^n>7xr8Ys#oxciQQG-6$BNeowJVr#_o*R
z$cpit69zR{lZ6o-XV;GO)Ac8px1QimiR5@Ke?2c!43J9_BBO7JEZOGPLvqd(%;-1^
zH?Yw)bBHFZP+32SnQpt=QfV5%xfZOo4G&iG+BEp#d!+)nxR1jM<X&@o+ee}qf7xf5
zLL|_V!<X-z=%N8CvyUIAL%K7k%eqWZw{3mx0G+ZY*nGRYk(CiN7;0NNkGU|N{al>r
zPVLzR=B!SlapQJrH129+!ZgND=1|PmdmWVwzYdXZMH$;ky!#rsW;x7xHvF0upsV#`
z%PQ+Q2`oX|I);0j`-zvTbkC)*7fUI*jkVpBD7p&-m3eD+Ohw^<!EY|}6cnkF%kZmB
zBx(+{$r+)7QrJ!_E2v#VevY6e(kv=fX}gHivN!3t_v?v9M3EPXX_@hNq9n2?Dr@?_
z(k%2~%H9zF@cPA2aQtUj4m#gczPSesZiZG5ekuR+RK8$C18k-E(Ldxlho1dEXo;kx
zX1Yr(*OHo9{596l5YryuCZ6%Y3kxmGYhK--Bk55dsA$AS!RC%0{kn8{V?1x6iqo@u
z29mim{v95l>Xc}t65A4jQ5`As=KIGGCAQd6*Cd=#%iso-RJ;^zAVLd|bIwHnQNMX`
z_yiRRycLykZD!;m7%VF(QYQ;YjmM2PH?@u%-`a9SRc8I;9+lE*lR$>i8AmgvM@ca(
z$jY8EN(DY%<^h%#s~CuIx^Ob$${RSKAG}B;eLWS<I)m%po}y<!55%|>r$Ei%f@Yj>
z$$TJPV<ph=)%r1y!bK#Kv7Sp7Hy^aSBi;1W#6HM%eJlhX>*Fg>3hBnP$$eTjDIwb>
zT9mnqw>nHSZ`hH8wIi%&S)2~YD76(t)r+r(msXeh6*U+-0ae+I-uZi>bd)?|+$6v+
zdmn<~r0qr2pJ#5PxAl*@DtOG*j4FWOq^O-Zxdj*<)wLQiHb%F|hYu?1%-?V;t$la?
zQJ%EmcY*BWsq9ZM^@WaQnZD^un}q2wnzff74qUeg-@GZpO%<$niFmA;@R;&-QWTaj
zTgdC?&m`jL0l{o2qu%J+_-k-%TmXU|b-e#lY0)Un%K5HDO&8xSxybCZUzi7x(31z|
z2Jk6T-=ho!(Ys@ccU$mluKLB!wr`)^o7f_>4Y>v90q8gpAjG>HZvjO&pX)K8lH5g`
z0_ldg8zcM1yVR^LiuH9E5$x|@R6!v>DJtq(V!}Nl;>7OK5eD%Dl~nX1IN=#tgar)M
zOl^A)n9b;nda0tObTV83D2We|%_OSCtb~PH(tvDWEPZR64X@*U4y}&oxSHuZSus!*
zNy6H@(#1gb=+l43Dw}x3NV{~ywK!JzF{36cq%6on9o!e9=0tb4KI6BQ^kHo1AGhTQ
z()nptVYinaB98MA$QuWESufOp(?8S;<Ad5*Z1vs|swpHm1iTo=;?VU4D*YWIt3q3P
zsC!u_j`YEy?Onm3=V{b`{PG2~+yG&sI#!%|(S!K4$17*dG*>#_+O8%AdWb8|E5Yb%
z48lu*jz|9-!`32g0(sg&M`Q&mppfm3v`)K__}m4w^E$jMwR!vV*XlN(uo3IWLQ64!
zCelJY9#8>TZ()dnzC1lI>4hzq1~(!t;%PqcwA)CzXdN@dv!%sZLbIJ)0yIM2<r142
zi#7G$xLpT|<bBnyaubq_c0`r;l{`YO*<y0~8^LmSX}*hL!&CpPMG&+=ul;o|xxTl!
z(0DlacZO1Mgt?2b3;K&fSisBX#!Hm)Z>9{eKg=b`4$ZiSHD=xLINf9OO9%NG?=4WM
zgk&fiH`a)-Se)Gn^#$6ZD=CF_W|kWacV}R~KEH9hp|bnNt>wo4h(kAo7|oTY^84%|
zYVi?=f!v~Tc*a_ObGyK&t5*+hDhrIl<4uQM5%kNI`6bi{s<g&XX~7K!sf#x_0T8$2
zpFkG?Aa4C&dYawPp*M5W^1^+26LQS|+=7oWBP=p~lDIEY$Z_9@%Mm-uJ4f^FqL(<;
zJfNL`3Av#h#`EjZo&VJfa2UC1&^)s8-U682aGntOwTA)zDs@n5MI6#P5iW-sUaWI^
z_90VC?Ne*#H|oQ2%7XJY<8*BVTEbtnUN4*gh}gx&R+-eaGax67D!(~-epD2}{X>+Z
z9zud9@gg}C)4fR9FQ2H^TGkGJh%!XQc^Hcex|3OV?UTUk7TYSMGk;104nt8dv%VU|
zFL!fk7W$HTY08Q4CH->_XWc%ScZ%ycx4+}{LRTcJ9+&{wI9#Zi&Q)Cv4RdQ&ntdgP
z=V}xUOwvi>k3bpL#t7|Pv$&^>k72<u22+1WtL9y5w+^5Q>^!*3XYh^aiwLP(Z;CMF
z@}Ig`)oOW>2g)yklcKpir^~Lo>KKX8LSDY>W6X#?_({l10ln(im>k7sxihBA9e^tx
zp@WQlIyt9K{%W*H@5MxFmz>ypKZ7*!2C6NMd<&7?hdD~CW|Z<vPlZ$kr!Ai8sz~OM
z0J^Am0bzN$H%(Mt{NwQ<M8gGOR$#{+^acB`C(I9vwMA#^D{!s1OmimYAUfa@L;K+3
zpDTF)lly_^7HCYJ#MhB5kIy15a3-A0>@eRPO#z`)Tr#dQp6~SDc`1QH-Szq0DqG7*
zd7$@*LdsVcg^~klyPoCY)mI^R(2{A{WfiiI8$Rx2)d?bjH<T^keVj?WOk)VGkrGt5
zS7DQB9n1|TRkl)g1RCyBsOU!^4JV8EnPAGay^oGUUIF(^(Kh<401|{S*#8Gw@fnP3
zj!vdAemJqbIL|O?QYxx3N`lrC9U4^>&-6r%NStJZ6yHujB%d>Lx^sQ=x#lU9uR&4r
z0d-oVkL02e&g}f;Wvhd?aQ2Vd*8`cZv#$G&%3qCB>Lm2=)V=H17wQX7r;!&bI*{%b
z5~Q!nmmXF4T=|c|<sKI`&LvNWYM{WwC3*G=r?|X}b!zYy9GXgV(9;KJLi0gCdCxwz
z1Fx1Tn7it4tKnCcCr2-`$Vsy0M3|D5oHC^fzNXK(JuqUBq;RdnRq3p=7CgKaj`ZP9
z8ap=qO=+hFt~*Pq=>O~$UXF$@$p7K;pL)#<=!dsQX-3r(L>h@kN3i0F&e2{dP9qUh
zLb0p2z^Qlhov=ZiE{MyquEQ`99{7^K7E6phI(pu5Xj<#W^`g#lXh@UGQ*xm-ebNa}
z0NMppKs&<4&&oGU=-uN6LrkMvS{Ex#!=Am}<ho&R`RL>3kK*m`C28O4x;q-BYj8Mj
z7rk1nH(uZ+d02fnR*qCP>(!{8l&0Q|Jtk)^yONIWM23a_plD9?eA8vA+6hn9PS@7;
zwI*z2DwG}ZqZr^iD<bu?LGmg_zLd-9LAbyP2We{M8p7SFbyxlR0z*OoZ4&ra)F>ec
z(seoNgOWo0*{JA+379TBSZ|jV#$j-^NSt5zMAttWKfL=gJlE#lPXxPPx*A6A1^)`C
zo1vCxKTc$y(oCOwg_(XWlhV>H>Nwf>Hdo8*QJP)ly#N6pl4;R$E>h37e9LsucYaB-
z$3qn+FyfI<M}15y;n(E-^N9}cb+~(ne&BdX%myk5bqF|yhNYPK&a#g{Z$=61G8Xcw
zqI}v)SQL#OSyN`>lPGbF1-K+I_sXdmdGUfkK(o$c|3)dF^#0PEvgQ2%`VKN{GodAz
zZib-$;%gRwWtK%epHXKUSGu;HI9-@4AwVRK@3o2^&P>!{*`ImvgyXPCNpADE(_P8-
z?8ZPbTLDy2=jQ}oe>|8lOx0<zlm=*COG~zY%gFQRF<(0i_bo@aO=4`?6E~g`>;O2T
zJM@$drwNz29NDs`NNy#^blg?RL{XzaakC$u%>x8B(Q`j<fRHFWj#!C|8iEQKS*vT4
zf5MaOy^>yjt`w+Ikn>liy>MPX3W5LF*&`Pukr?4J6`8I9vQGZ~X0fi%eDT;i!xGBr
z|I*~d(b&!myb8Wye7+jl`AqFmMt=ap$jkk85FMTL{=xWJV3}oKm%y~xQpS!)TWVm<
zBR2vf$#yT#{y_CN{)D1`Yqp^9%u0snQ<fZx5?*uKuF0!H$!3N$sTN1#!}*QMa>`3?
zI~wqt${roA5_b2*Q~5PddKZ=N)gArp@Z?kp;<~*l;h#hX4=|7x>?_3R<L|*o-voaQ
z!FEhf^hxv>67YPprA<lS_a19zGn!@6s$mSK_YSu?yQxr@Qn!0ihsSRVMCcovDmSrG
z6>nxpXsMuBOqc>R)?~OiwLbdv_RCb%I;F*g{Mg|J<#Yv}s6yuY0BoIiXkE-dwsNUi
zMH1m#Moxyg6kFXje6LDEx%^es9H63}8m+sQBbeVg=mSzZxl>oV5}p@?IvNp-=}`uM
z>-E!GP&UcNn*Q?92X^Nes?L_6@tB!<(=anAQP=xSE?m0qr^e7NJRKgMIvRbdq+73!
zm26N;lWW1K;%l*4Xp(*y_;hRrBi$onddIrtHxR|hy{<mxW(T$%rF1IqdrFahHHwya
zUu3Xs!nH)R{(NU@*lixFmLwEoJ{AHik%d@rl@PER)x4|8qwe51ZteZ@Fm41b_qe3y
zKC!&?$bo3@`29NmbIC8YBZa_@_cNK++srwo)Sim=jUpF7wqj30TjYRQT;BOQ5Pa{?
zlPeCpKC9r3rDM6{>m6{y2gS*^E7rCQs6+NCgY>qa==ykD3tPhNR{s)wo7ctJpQ{%W
z^3nGvQm<iInNf`{#-y3x9xe{%P@ZziEgw{<mQ~r8q_|TJPl<EW?fBc((lAm?>L)oJ
z>k{2teqI3&vZb99ai)-ss#$QHyNtxR;@)E1`XA8e3Q~;w7aDj{R1>Q_LrbZTEybaD
zzo($e!9N4O-#-Os{hpZ;B{BPQHYd<yj*s(1`0|?*`O&N9ZpMyZTWrP4wP`w3T~uKI
zD$n8zW;(xvt?{#aPuZ8Bt?k{gjYl;!__@8Xz1OdF_SqwSk_wNXfM6<H)#0s3kKJc)
zSo585%XxN$E6;S9ko90+mk65OTln*9a%$y6D>4j=q^p<EFf|}KdZ97-{5B8u1)u!;
z84gs4oumD_%PvhA_YV@tze-D!gB+>c&*_^9XMnshWT_1tlLvIYo4Mtjh~FG;RE{*P
zGWo6b&Q42)Ww%9ubIv^VG(CfoEB)a<>ou+J%ilE_pB%LPG9GZ_;rKhxFtA1P@LpD%
zv(H>&UtwxO)ca1|MzbOP`Bu{~UrdI1pJ|w^@OCpb^qr@;7w_78l_3dHJ^Oz(CcrH_
zD$EWuge+}6yDBk?y902TdVW)#S{IEW@FGaslK(E+m95^z-olQYO7-MBCbtlf*d?eT
zyc9@N{Tygo(<U#_!>83&D)+SbD-g<znuf3cyy%!T>5HuLKF2X5aJ@d;zmw<adPOhT
zM0)R)|B6Q2^WIP&+m)(3d2uHXY|kkX>!-AK5@liv3YUu2ooM`!mEiu1%x2K*?%^k-
zj1z&*Yx$mYa9cON=GPjP7HPTv(4J5|!$OAGL}CIrROwB0<-nZXSV_vcY`j(%$}`@w
zm~Z(Z@-?(O8SjR#oiG+^qMRh_uEj*`?@p~#>Oe&|u!hdUi!Jv4WnxuzO6Sa4qqQLw
zz;P$sA2ED-hA_ffm2Y~G7gAp8q~0C37jl#<&iEaGdMUl3Qv8sT+K`KcfY*q!FEU9=
zVO$>QB`%rz^5j0m+~q?1e4B1_>A6tE?Ut&=Y1URhr&E(do9$?yUdP2?1xPy=j(w)l
z4;hu}8Pfz2k$Fq%LwP@Q-mo1Icyfg>aM^8=0vw9iX%>L!s?__+roe3Az4-BSnfRn}
zc7fx(MQsZm&T_i>bY0-nM#dJH&h%34G5XhJ$Ge+{b{|n1r6-Xxr6VG73q2&_E-Uzg
z>gq$Ad$~HR0Jzzy$2pGGj`-nondv=2$PcbamLRz!Um2As=Y$~va|LS}%x(GfK9$U0
zq^i`HPjyqigG$c%TI6)BK59m(2*=V49KLzGeE#VT(kl^%T)UqWavr5sa}KNX>zEjh
zydaXT$t;>~pWWXQ?{@<HFe~v_yAx#6MAee{2Mp?zmnD?X&9xpA%b^dY&YXACc^xF!
zee$xSj{j2t3E-S$1ZZwL>K^QV6f{;#jS)L^n#{pAeNA^>!|+M&Jl~G)xI3O_a}CRn
z5^(OEU7WE|VD3E;=@nBvnc(iXpet~wH5qcAPjoiqpuW`5O=!1zuJ^Z`-WOXGZ;ZB@
zE9^xWt+A=JVGCRq;KzC~OSFOSkhkC$oSyi~p0&EY)yI5)FkyTa@K*dqXfWTp6+D&`
zM+y;8sx}M#(3v?RZcfOm(^&>YHQ3Rj`XVNcnBN|O@>vJAA~$HOr?c;+<lxhD_n+e+
zFWYi-#ko0Y0QaK#+IeJPl^6421nL^0-_fK0BOo?R6S_GzGfk51Bu|JYI>XK~AtC_F
zCabHy=Brvt1%Oem?9YV+g}Otg&xyj)W{CVA4|6KshXJe!3k{_mHy9e@zZ0KxM^v1E
zaRdtw{od){G%zJK;&K2NXM(jOBhs?zC4AYN-+YMACbdN3N3-1HNiLwNi*DBWjChry
z8^neSa7z>}&zmI*dBjHdvsOfYcm#k#Sdp!8i#VgVC5+qgi0jaocF*jB)$K^UxYzQ;
zMt1UUUv@`o*?TyE%Y1z!URRair+PP?iFLAPp1>_2Un$wiF`J!;5sGi4&J<HUxaVjc
zCgS~~FUHZnnh-g54*;}r4nZ^udiJ52m;b4*eMM1n>m9A)FEk9Es6cu)aO5ND*Z6u*
zYG0B8X?F>IH`c2Zc*21gdLJ6k^zediu)vrvn^J=FxV|94<mRhF5+OihtV1ch`KjDk
zd+?uvi>5>}O9?X*lO}fW#``I`?*&{i_$CKJ2@?UuZvzKCPFwhhg))+^yB`rn?MO*2
zCYEIM_LBebK+Ec_IU7TK?%4eV#=_5e<Y9p0L}>kdKVJ|dnmkG9E`wKCawY)DiT5Ff
z<%-E`_KG{pYA+O*0+dSoZ$NuPU9yknx>WV1ftwZ-)4b*c)}}(s;QhcpT=K?Ldgx{!
zIi;zNVo_duX3zwyhBRbZM^p;Se@J=Af)ZrT`F=k56F!UhgVkLi2xCG!Pru=%@-$e9
z`Q|8@<_fF+u+Ujx@xdZ%5ND3!vLgq%M|Zo&gZ=R&r~Pvsy>KT?4#`9|oPW}j5yZ9P
zJMFu)83a3cCe-z%7pq|}xZb~6GV-ZQ7<F{lB&@jKK{@5>z){>Q3}zu*@#md#QD6LB
zv#5*pzC!r_T%DQ1&1eWhxWoz*kvuX-H_s4GDI*g1Od&iw;+D0!P$40!x-afWRO;p!
zU+?R|Mi)a3^OszL{;zA8JIvyZ{Fe+>Jkk2xI-3ewWvB^zzuEwI#GjKC)yIVbXO?xH
zl>-LD>}B4!yIy>0#Bu140~oajv{n@vs=(QFJbzx$%FX?F7@$xqeT3&Py1&XK$;~E!
zSwb^ay-c{*wTB1&x?Z6KC(Vf=p7aF(*wf&DZt&+pmAG>3;ngl`VVWsG=-f%Z-Bg)P
zo<C*rCSN-HuX(01G?-P{YBEot7JhjsmF9MLfy&X@w@c`vQKqO0{n-ixU{j?0tBYT2
z=i{0<JZG(VD2du#a!;-K`7wiJ-ESocW5g|Yc>q=;i6HNY{(~tX-Zfbx$H3q(l&aif
z#km{>4M9^3x8ieU9<$M9hP)DoR(TAYqDDNH;rD=Tih##dahkd*3<iWbPnMm%*r8m@
zhN#?#Wm^Mgd~dRgm%GX8cK4%G+mmHGUs$+EFU%;spflTpM^?D=+tb|m<wtnR@r_jG
z#cROv?p+p~`2^h%A5vH^8jhAQGJHXMu|vzYDCJTZ6y!R=#M7(*ZGJWTjICpd(IAhp
zsiWS<mi_Qex!Vio3q8i`IXDA(R|e{ZkhzyW-uHmu=(s0UIdKb-9HbG(L^t2S7qpk~
z9D0M$I-(_;!caG-slEPXI-Xz)x+LtUh;t%09gN}iolU(?^I;q;HzeZ5V=i%HJ-9h#
z1(Ogr8uLgCu8-LLD1^9vQv$83fT+rot-_??$8!k7g*n_2r82<fPBpp@u`lhpALvd~
zr<8a~51EL<@hXaNsTO!}Md*Y+u0P>Q4crT7V=4b5-C1sm`9;}y{e+{+ix$6C+#-Q*
z>P;Y+=g*ZRVm6Gtb;=a4amACfJN_D0z(Nw`CkppjENe{z7}qp(z4?>onx3hZ|9p=~
zhe<_w0&^ymADDkD{G+uyMe@3K0Z>SWNwrMwNT12v)v$9J-PWqyAz0abK^~TfzJ%CR
zZ{QB5NZ|9&*q5`IbGm9DhA<r~AT|_eXrmdWqH5!woN02Z5p6-CRdo<{e79e;y9hJs
z5r`Lg>_0_YaQ#xQ?Oao?T!~Wx-#uc=E5@TZdm-56!FNt3PjYOK8Y21IPW~QkBjvOp
zy9sG>{~?bV@UzKZW1YfdDIxO(a*Ow7^a?S6_$o3VT)S^zz=exrM2TMX81f)Vf1N8>
zv&g?=yr0>Zg%6lZdL1qncTNW%ytEOPdRR$saNC4n>DDe)rqlJLERRA_?)q3l*p8`4
z`<T4J`jeH7rui@3H+|zUt+rdoR0yN9Jp}S<IR0Yd1b)qlQ-a)6s{R6D$D?MnI}5YZ
z2Z_>Py->Mb>(+e=)Hb_y;BDk!stNIu(c}eR>1w3Lk-w7U;8ABP>{gZHt%g9E_k0S9
z)+yq5JPn$ts;dbhHh&RwnqR9@b`l>T%oQVE;EmLxi1m=Mp&@#D=j$o=8lEJx{q`Iy
z!BD^r(>AOD1o5oJa(x@O;65`^5(2YVc-te*fAKkTQI&2+dD6C$Tkyz=yeLBlGDEet
zTp2M_a=~}UU)eO$w^Zq{pkEpdEr9ezb2=-Brxi+q>N_bGo@MT-cRo3a=$~5mOD8Ef
z0x?_GkD^2>e?QmCv7osdO1cLU>XMxDOefW~GWI)uX?9!Lq$V}@q2JRHK2;i3vespx
zKo4F=p&Yhso*!+Ku6RRSnfrcvbA`Ar&&F<Ut+-D5>=L%TC`S);gMMZg^P7^^iQZAw
zui-3wBCJGzT+vyRH(7;Olm{`jWv_4Y$Ht_GAudegdaX6fchV>d#^p2iABAM#N{H%t
z(W-~*MEAaPe4H>viLPKOqjch)oLUjuGoUQ{i`=xttD5mfg0f1PLDW0q>#vd0QUUXl
z95bI!<)s=1uRrA)daFX6=(N~Zu0J0<ZxWJZIF3%e{xxgU1dA|+JgIt~(BCS4mXFU@
zS~haJmQzE%u>9e?p@Gg-D0{V#wUiA{Z$nQ3ULIApeB<Ub@mHhLNtQA#caidv$lK}y
znO!CNZX1mww4+A&HZB$#y|U&zNeA^d>$u3pnZ%(=QZA5@=uLY{u8UNTr5%payDa=Z
zbm%407VjUcJ-?)l*@L3aEsm13{m*fGRnQY#Kslu~z6ho8A$PHZ8G7ZTdUjn#qmBha
zy8P(*gJgRg;Pd*u?M^L1XDu&3>;1&zT~M%*9g4*cmNz?TaiCeS73qVR>=f>`sO)=w
zMkxN0COCZCB{7>;C^6t8HZ+gC5O5jWtNjmBon;FEcggw@`8bSfQ$h#Crl%!7cRYmF
z;p`l%AMpD_Z}$rqqZP5J5|;$F;YCt*G95RjMl8dG^_gSv_QLVKTmctpVL{`FPoBRJ
zf<@YbhoM;>U<}3hcrXXFDN@spTiZQm`6xA#pbAuU*MhzkCnuW4mi8OeXQA_<j|-22
zNA5<HC=CcMIKD-PH0X<N>_qa*s?D+38NX2FjX_wP<*VoO{b)VAER^{q>B=Tu^~uPN
z|2kxx88111`N2JcNQy#*p_80f)GAd}!T7sR<zC7pgAc8!FP-p_-vhWrwoA64RU<7=
zaD0*)0)6rcGA{XV=EJ8LKNz~BthZLPS3xL}a$s~AlvA(Jhb-#dfFIz?t)GX%3vaJD
z;YwV%uR5WUGQX(J*K~<bC7G36&H;tWpFHr|V^|hO;u3A3a%06=EBNabhOJrV4^^XQ
zZntFbA4b@DMk`E4g09sgD3M0tpCQtAJZu0#xeeGYOpNz5Jg4uyQJk&m3B@YE#nNNs
zYN`YYHJ9V<1%h4U1#1ajbEOVB5BNlMcy$HD{I7M-zte?UG*Y&d7YvU3A`Z7>N^<;n
zLKd_36V|T_b=U4MN9X=-7CJM&*xZ^}A|YoVkz{(EVtS;(<iS+JCd@4G7e=!g&TcL%
z7+8WOL{&c30s-6s9@bs7t08bjF%f<W3~*ewA93IaOcLQwev9FS2wa(&xJ=E>f=?I{
zOQ<3sM~<${C$8co?7Jq<`X-+tjzi*3)dFb;*B$TtPww~o<x@REd=-_d!rj>z=j)kM
z|Gel1r~Wp!Eo@j^p9Z)rFeljj81`xT9pfK)qGR#!FRbu#{2iDM>ca}vsYm$xVW9k=
z4f=vXLO1o@-%kvb@IpAvg)MIH;zCU+OH=H(h*6c37kXhyGa|1-w;6~l=&pf7f5LAd
zY0swsaOU3Wrn?S2m+8`q?PfIG02+-DC@70bNr+I5NBbw2EU>FYm-y***uNYPb!UAE
zgaU7Z5kr-#AE_k)W8a!Wm}GfK*ajjdYIn|zLnzVg<JzrxA+gul2$JrDm<|Wfo4X7T
zBVGtCN%Z93d`{t)#{auQ<^;}7-48jf*fMW~Ya~faZ`f@MuyH+vyhqrE4b!89N|PQp
zvU~YK+a5B@S2F{GfI)eTXZ6`hYX-=;b)wy;LLIFfGF!<_?-+?b2Qvmy3I?L_p=4t7
zc|bZ|Y4T}K0HukUibyac(Vy(*r|c4j4EEo+Q;P{A0Ln%~L3Qp;?NpM7N2>ih7gatM
zv{hWLj$@eSR7HF-mp}G+;M(7Lp{H!|RIPnNJ_xP62uhnVu%OLgbS>t5z?5n6YM*F#
zaZq26=i3(*0QP&(h)nrciO0*g4UjW=kL;<J?<DR}fnVrdJnW(f@_lkhn<n7JV0Rx-
z%COopL#wbj7OSuS7sq<D5s8iYIZ^ZN-(I;V2X($I4<Z=CH$j<bsuH8gpoqvHEARls
z7cd^j{&tvJi3G{?tizS`dqW*s`27RnIvIn@_|qtVO9u(`YwZT#Hp|@7Etg4_5E|&^
z!GT#gXE=eb2v-StSo#bNzz%Hhv=V4JpR@V6KjmQvgeEI9rs)e1Eb&e7+DcBCJHGZa
zDSe!B(*@1Z!qKiBHx9?&yQ&jK6%S0cTI$vkEqTGxQu>08DLn>H`}W1UgB8kuyfngT
zo+O^c#@``jFbttQPKi&7vJjYaj%TE_)Nnfw4B?m)fDxX5Dn51*K9ofzvQKL=;BSHD
zqrX>3?j1THzBI2(R~E2T)jAiZHW1{3GcJ+z_(MJicSNiRzHjUjzNf`pF7)>-`)+MO
zkSZ}i7Fd_3y&Bv19qt4!dwvFg3vp~8M;C_>Aq_NDf-(9c;EAe~RL!+f1L;lba3wr*
zH=_7&6Ff7k(vxKbjU5TdLUGNv#<XUC;-ui|D3fwGgfz|w$4<m9K~YI7b#q=3Tw#CV
zdb>y}=jn&Q;Rf&eozj1!1?9Xs`!c~6f$qJ)G2G*%_&`X1h8nP()Y<w5O>xL8`4j4R
z*Q@3l=$zWb@R{K-5Dubmz0V<#jaYgXGLod#gZ^eLBClacY)o=)|2kTHC!kr8`SBD6
z7kFBJS0$}DI*i&Ytm-hBJ9bHWo8<tw3#_?yDXqJ{(3~DJh*k^fi}ZhDlhJ2{d60&!
zd;$H6Pgb9_wa?iCAnlwb#H>1Nx;6ztpaX&qf=fEx6q*Z)mtIm05=kyRG+gK6scqf=
zPC@m#(3u_ZN6;iP8Esj_d0#^o$A)y!ip6!T97$<a;zA}b%x3P&e+;Ud4BVpxu##eE
z>=aDg{SG<em^w%a<_VqX4D7n3Nc;=Qu#{*Y(oLl|^1s)G5T+ep$+K}V;YEsIYN%hQ
zBhwj6>{j`4EjWV<M}}12Y#nX@ddWY*-7?$t<uoGd!;Ryw02F-x=)Ky?2QOM|d7S(-
zd@7!TA?R;QSyg|IG9H!Xvn-)4fFOyr$oEZ6PT!&me{-Xv%U5cp+=}=ew{_1-jv`MJ
zam-Iy=-AF>2LE6p&ZwlR{Zyz#cMoMqt*w}<04(Q(EhSeGfN$Y%mnbfO+f9KdE%qRQ
zD+z=%t{84tN*Bg@IE^s+F$*7!r}KO-7Z8M}ZR0)dnLSr>BtKaII`(}~SNgMlc9WP(
z{mU!(4=&GN2_O7XWOi^nDHOy`T|J*H5m$z10Oj2E?B{<Ix5@F~J`Sw@hzf1<tJUzM
z@*+->DxT(ZROXn9<-=Z#7jAFu&5ly~$Nq=tQ?oc4b*rW0Jf74t6X+H0rcLjGhAnxT
zaKra;Z~i&$ld6kv{3!YU9qAH@Gjk<EKsy7A(G)ggRK#ZU80jlscJfg!EezIkBG4VX
z326p98%`@fa$A$@E~hLs@D=#S8YPKJ)4#-%PFxl&;?b|Qx30I~<gMN^(SNUc>5QW5
zJ?7$di!4q=)3Tim*K@_>^_+0i5K0VvOSW^#q#Hp;bneYe`yAo`TdIFlAzf*k+V}Q9
zWIqu}QPRf|x={~_szgw=RP{`=>L4GYR!qB1OKI#2OlRj1oazFh4f<@Z0>onplQ}Jb
zUHlz6EGid|j|V^Z?`ehP$Vq+G$QcGpWU<c!a?t^`%3bX(vxP)R{eCrK)rSkZ)4N30
zk7|f!%AcyT1*xr-JYxP7<xAydxOEf*e}xzlbWxVVrr>Y6yiB*?L3(5^Qn)<2Cm!&4
z{b*Hs<Xwrc<)2sos~3P5NzsoZAL3F}dQ%d65=EolRh>&UQ!od|S5_1DErlaq#I&1s
zmYFl(`qwsH0qaxo)Yo5)NdaFN7CrRNPUSLspCJyu$&$h1(1r_fGpvkusMA_$>btrs
zQzHO)J@^6`X$%p%G(!!ngvLD|;;>usyppt{8S8i&DA+zW0~kY0ukc4a&>Q3|p$IGN
zGiq3(wALTGulVt)ji}0qt9sk_z%$mgf;c+d@y8JsDdklD#pP!ViAOLMyyy2+NLS#3
z-;4VVY+9x@+n1xKGhFemf<HzIAr2xyE4~tRA`16tivXQ|<m0~|8{GQ}9LMHrHA+ko
z_TkPMaWc-|E4<aW^CV|L2dXGp?u(mM>3ZV=BGZ_#yN9s|USOOG?VD@hBtI!D;F3!>
zL+E`eiLR2t<1WJ?XMFR>&|Jn9iQGJ0Qj{1tl0EuLFGIap<3u-1Mgq&5nhmI=En;-=
z6dM?+E(WjyFU7+9JCz`%W^2}{)`3rDKAjqzNjiVIiJ>xYI3{?dQ{fl&7v%`bmiAlI
zOghL6t<ZJxZ9*<He3!87%Y!%cEMiO6LJ_VWh9W}#1NbzZ(!K~Ads}WrjkqflRX}dD
ze=75%65q?qe3tb%&soA<$G)x4o~u4#k`xbF3)CG&{qEmX4)55-vmkun;oy3L%9KaM
z<$53Udjcg7AD3(+owk@zv4~BBS*5u3T_<Em&5b+!6QL|wL#8uXmYL)bAbE7Jwj-S+
z3ZJdBDvh;l4*5!NR2*2wB)N<Tsl_|ta>|vWN&s@S*NiVoYY17FV_9La3TP)(e8X;f
z=5JiLF_xwQB+h~(zSNWDBajS}YAjq40vB{5VQ<`{W#XEmREq9%YJ}no)q_TxbZpNf
zwQsahqNPXI!5s>z+HWofR<C&^u&=Y&c^rQ|W1-WbEqSiSFMCfol;Lm$$U+pv+Vat~
zgWSuD;|ovH9eh_nV6ZT>rejxD*S;w78jnCa@gpAZ2jer<QcUp-dC4Eay}aA4F_9pl
z2m_BoYwo|>n&j#0vM^}Rlo5b@C0EV@t~MZ+0E~|18hdM^h~$XP>yyqRTkchDHm6HJ
zP{I=zvpH6jnWy*1$MFQEd_JQLj%&E!*4{9fiU2T8&vKg8DI_=gIkQg*FO8CK)R{p_
z&N!>9V!A!NtxYYh5yt|dEHpY9NVws&5D4^}caSztwk&YIr?-f|5Y`uq&s4A5?|ZBN
zwveu?`%Km3hQAXViT@^#8D5Lvk2lwb#pBMIe3?4r<e%R0zI(6Enf^15^MNc$;!W+%
ztAbvV=;^+tGez;G-adJ-5t@vI*wQ|o_yd`qpwp*9U*;r72>HP!Ht5ULl#J)LCENb8
z|J2mP)YfTU#7>psZ-r~ZdREJL-nJcprJi-z(09H3`G~A*@dXm~w)eR<pyFS}U^!qY
z0H-hd#VmmGaNyMzksz~vm(iM|c&KKK`y0XZ$y@kpluhVFJ`N|j0UKKj>umAC8Hc2b
z1Ysaa2=MzKA~YS;KK|jA*eq^lTZzxR5gN%F45syXUUK}|9>7Z*v<wkK-9~jGM_i-!
zz%TW$lcO*l8rg;!wHeglUPQ{Jx83imSy{ctfzgltuV#kf4l;Kf$;w&ct*@psb?Z4z
z1;XRSlz_!Qy!;99hchj)$k=swfzN@{E4>?hvKG^$0so4;iH-js%)Mt=Q)|~Q8VC@P
z4ho3$Do7Cl=^zm4WeF-sCoCz_dnZAZ-jOOrrHM#KIwUSa5kw?(DG5zFgx&%<ll6Y@
zKKr`%`M=M<kW7+i&S#AAjB<~0OG_)K+6BOdTkmTThwdmFQk&gLF1EbZ&aua>cn(+J
z*z9{jEyt1C1Hcs6oP)?E5mi5E5uLh53}7-gx0@zt-t~&uBt(>{dX@iA?XcnRu(w=-
znmFrgp#@?lzS_CSbzZjsb#v7%DWPYrw192$a<p5Eu#e3vvLTrxcYPZCWmRVgD^a>S
zA@VK%RybLEdqao^?h)$N;alC8KrZF8l_BawM*ajd-*CwCn~d`S`uIs_O`z<9C7m<=
z-Ewo75-*{ANzCaat5MMsde|~dT%|TGn;W7Z{u>BSi`?Ka!LY85l=0E0+22z+-w<SQ
z9nk3wijPb<Png%CRxB|qog2=H36ttbAEnCo1rgXJ)lbbD*)#5zD@Q-so@P?5^p#mU
zL`@?~N+7g6;x$OxcCgvw;LRW?bHNUch7hRsFPsXR(yVLumiys9mV7|aHs+xPqd`O&
zpfWmw_t}duKQ6GTzBiZhBNp`(1IVp!=YzrWEb@ifJS`f|*bS-B2amcUM=5~_=(j%6
z<ejh3k_XsB++RuU%W2>pugjx5HDrN8OYo?rT#@y6h8Uww^`*H7Cwj9L)_>bP)f3iK
zZ_SZAdJ?&yg4&3&Jl_yW<^Z>IdT*8c$MvL-IEle|jY-^OE-H#Y0Aay;6HDbOapKps
z;nK%_g7Yz{>@ohLclH+gP1w{twUgT14Js1&yK&lgTZk;0g?*VuAa3gGr*h&zQ22wW
z%~}~x!kv4&&@E#hq$kK{C;}l%-C*%NL2aQ#*a}GTu$3FL^nC!aPU@94`Z#Skbz7j6
zQ}s!Zts^cGy<7g95}y|+DpBkp9854OIs7No1&5ybjpL;O#+p1qMmkwd9x}ip=Bk#n
z?T+*MxA!?Nut}q<oA8=Fl!Usqatb5h`hU)>v-d9Me5(L7eQQo^Hx$TzDbre^d8wtq
zbz<k-YlY^H??OE7?wpPFr<x`oWdXJ{l=_5zkZY~E<;IEj?3-WEzz4OTo^{zn3#<l3
zl+}GnpBPzYGQ@+4d15RSDmMbDmJBfkhfm_>E`9E{(SgE0L@j7<0a?aF#V_8(VC@dw
zEpM_)=c58tPArGE5kom;p~#jcsND|{PfCngWfuw2a97@W`{xP-TJ`d*@Ey1OyE5iA
zCENif>2~ac_ufLNZzw$q%X<bfh6*5YA?DiVG_%#0bi7<^zSDPtAEZ{BtT*wA9D8u4
zSswCnSAsjQSz3i5K27w-4AZ<A;$>dE9Tkt2nyRZ`DUo-$uiXV0?lF<-kA8n}H+BQ2
zE);uZzU_;&!80v4>owNYU_CZd*C~8CjhD`e{x6P?CO^IZd1Yf>ToON|{Ui^!ocRt7
z=`n$}qBP@rS=6f<Z|YcW{4Kp%2!j7bIDhA(6p&q7MIhfl@1%+iU3ftRu$8HDfGPWJ
zt}9uG(d-jq8d<S6+}L%H{Rp;=8&CRq5pn$RQdWEHIVm{Z)+fpYVI$-q{*b+J?(KXb
zlWSFlD@GK6Sackf$`FZ*C<Eo-9dHcALewDY5%7{SqbfR-!c*OOr3$sdT{ym@#zpBq
zh?l~(l*}Si=5v=4OuqI454N~nrZV3mht;pjyre!}T2+XC#ln<Si&2WCig)~4Ajgsx
z&#bwHq>R798b~f<PW-`xJ07rsZX*Y85PlvLW&|3w58|ouc(L-r@9p0WfNKmn<%B_v
zgCu8{(NB&uP({){tUKfck)>+&R(MK@m>%9KYKMN=J&*zqgROT;yI$QwfldOiN8sx@
znWbkVzpQf1St?B(e+shM(28s+Qa96`d<UrzeHWU?SLJk3@1zL!zr9O~rI`RV&;49P
z-1+Q7pVyCh{zwQFA0ZSQMdU+h#6*0cqmL(rGhk-F-ZZf<0SvufyxS-{H9-PLBo7C*
zC$q{wJ*3CKz%@Zkr48OIJ5`;~N<N*w1>(?adUz>>t3)rz)7*|_?>T3ZB4?u_^PmDC
zIKnC6D!)Ev@;8fU6`^=aW+nPOHELrDrS-4w;q%Bh57E8B00FI)eU*&am4NfkSFE2>
zjKJX;;|F&nA_)(XLu!ml8S~hP<G21ERC(F+8{+scL4Ze7^!-3pLoA;%2wa;q&-PJy
zbqZa(p#ZBvg>b`3FkyZf#*0PiG9b+{P1x>77SdjjBLkl|XSFMR(VL3%Vrzw0<49x(
z_o>B(E|4EGZPb5W6T$kmfopgdQ}cH9scPeDSoHv$o&?E5;_;)BN>X<{Z`TGuZj|}`
zu`Ladf>g8+Jq$cOlIJXjb@pXoudySF5waTF=f72jF;FYBsW5I&SEQ=NXJBFplk4vU
zFkrs&eL06YL6Yl4hXwZe=V(;VGD<>?b4=!$G_-XMG|7A-fRl*Tl#S{6TG=M{q%dEv
z7}P@A%z0uZSiuMFny=}<*l-c<p455A9R@iOsVt{@z%a7@jGk&!)VUpW@(6d>t%E_`
zG{IL;9lB{HkflVW0%vCX%S)mVqDm`y1&FGYg}e^nLnYBMGixR#`H2zN?v}z0j4o2L
zDW4zJvG+oqI{|yUIvGDs^1>w|<fTZwLq<47XguqTm*Sv42tuDPXTgvUnjSMzNsKoH
zWA7p$5fY3f2OZs)rc5TQ)>yyqXD(uR2OK{RPG1D`ulNeh<(41LrD9%q{X{uEoR*;M
zY2fTA$fn0hmQ?Cj+qFWk=10#mZ*fO+1?Z>@dy>QujJXG#n}>2S<9_a(C9>rUNrKEM
z*TC1DzM@VFYl67*jQnX5kF8_w*EhgCojMz$ugOss-Wxl*xd4ukO>F4%46+s*KU8sh
z?LM!ehoA$=%yXq0mQ+S33tSeDht$PJNsAwb1b9bd=Q^+I*gM_ie9Z*5vj??*P1Pvo
zuztng@M1W;d&sQC1aq`4!(13NgFLy4{h;*98RJxQZ83Sl5GLmcwl(;XATvaqh^5A?
zq2D~6+Stb@e+F<Oz|!UGyonUrMuz7<g8GN?#F_qWD&vhieuv{ji1@qb&%o8|0Sf=z
z$k!};>R-)^p&ie;{<sT(v?ac6M>LO%wN)jhed1VS!)-v%>TWy#oSe+bu597MTlck%
z8Y?BX12$VlQ?&S09p>xS@@<_2SAwQHWU|U~zsCu(1`%lS5~zjX_##TYYp$>px8`7T
zU5HIF_1bk@J*u(<!bu^{^J+B}>aHA<8pwrcCvxjirttnM)O6_@c!qjcKk`<F9Z*ZL
zz3b*IdjqahF*C7|h>Vi++pL3cRJ64FQ31}%LJ`@qUiw&7WlSNVxGn^~&@4U(Ak-5x
zJDd`Cj2LM8*@D+7c)QK{#_a8Pv2L+n$cI34bS=NX@eP)zYnCuYNHD1Ot-11(uDK|P
zJ)yaeZ$d}xEu&P&E!Y!VR-SBQuYDi(4+G+iBnwA+_zpeG{Ve*C&7_Yx7D2*5?riAx
z!2(LyK`8I(R0mt&F82V>AW$eCv*-F4**Z^&`_0+z;delMA<RZOYU(V7r{SG7p1KQ>
z2F>@n=BTW{W=BdnD*%yS1G{kx-0Dk4*>K7>g^HH%Gi%cs0O_O@$BI6A)+cexH{^|`
zg%Ae_XhN^A4|<httnhN18{?UM{v<4o$-=N#ib#A|EUfKFVgus5>+en`F7#^M4n><N
z1g?$~qYFN`kcZA8>lS@BR*lMO0au(e8Z@AL%vC%<Aj+4fS66LzetLKOL=mHab7}Bz
z=EoI>m+|UL;^+T{aYIk*v3V9+3Az9h&-w2qa_N1dhlVI`TWROsC#f}tkM@TTV1J(S
z*#mtt*A|qv194j^S%&A#8oXWo)NKxCPXe9`S_u(J9GQZMmxw>wzdy@@Rks@cvUj4y
z+u`K?Osv6v?(BZ0sc~7mykXeHRpWbm2@aZjSghKIetRMmbD%##a?E0+LSaUJdPDm*
z#8@G_zt;U*^>DCgpnVivfAA-jGcciJ3rLr-d_E;sZ>i}kw0aZAM-#%>VtiGX(jx*e
zKfWQQw^xSLa-AFejti+>?&p!Vw;fE~jWt9_w9&p6`Om+LSHz|++>PjpT|ABF#)8T#
zzJb{q8ff>jVir9FnWa<_a;5D01XXY-{T?yVKWRSAKEFGFseF8Vsn6f@xg@I^cw7g9
zBJ!3dh2J=dcjhD-z|8w^x80ezH<{ZvNLzXOTXEM5m)siIF%sO<8-a?$T*yGHt0+V&
zqX{0+%b_f0#T~M7CO$!CjHT}iH((&(m4z&eVkMDZ<uzA2vi@L+*KuOEI&D!;4F0oX
zy`a0WgSipKm^Xhb!h4<mR%x$mt2lmAhM`G;gWBsFd=znRUc*@ATN+CDe#s5|31&MC
z!NMTSR6?-`KTjZj9;lC>R*sEYI0v4@?3(w{`}LFuYL|k;4!fkek<3~2i5nIS&OC7P
zJpmM4)1vp6T-+apyA_BDlHL~p@Ux!;5;>;xGF~bsGzhLKYI~>}cLArgv{LdH!EbTC
zt#*%J8~pfiR{YmICF<nUA0qNr+Nr-w>iuFL#Lh4F!2^0pK6@E2%xDdycJk*D_^y>N
z^~I;U5TjfUX3YCr#AzUH*mP6^I7a5MQgc~a0R@ua(}MLGa@Uqm`_?EINXqoFJ}$Gr
zb5Uw@Vr6K|N%FD@NZ7G0y+M0;F#OUMn#i6Fz&1z>UjvY$YiO?{o`t;m-Gy%)u@6dL
zpc96%h|oUzN}7AKo0zTG=QPLy3rNeSA%#J=3mzOxZ2MH=V?LxFVReqC{Y_38Bq{DI
zTCJWXNKRERfX`p6LwxLsRm#e`x$cKFUj_#K+Z{tj^zq;wb(e>~IZr@K$`oeJKmz2+
zk+Q1nzJhL~1Hl6*Ld`0qjo)fJCOP&$bC74t_#98P-uzLgP(mG=C3bru8Du>(VU*-k
zXcDt%?JVojU&{aZ?jk${*eIYO!h1o%pQLtfmaqF+q9mI<U_;~4xuRJZOBP!JNlbX#
z0@j0MjwYy(XZ6|%rR20{>#amZ@J@fFaNU^Jtmr@ZE|`USacN-PWX~f(=I0x4q+%lR
zl0T|PUU=Z7GU}X`mMfh&o)*MLIgE*00bYzJqnBmq_2z_Z>tj(^B0_k<?06srNxN3s
znRVSeY9!_m)F`a%3yV58@yV@|*6tym0vc$c^<>-TR{ZY(SEB{lse9d-MME;4&QePr
zATE&q9c4Uk-v&=Uw#g`|A+ijFg!+u`K7;=LCZV8gBk-RupiMkb9lGPr#UHMhbu)jE
zEohClk4V-Y88dWDGQ_`dWQQ6=FZ?;bIen2LK~{xmu)~!kVzJ5I#kooatdqtng5vyj
zVlz^M17r3;sH4Qx9M391Rr6u8zfmB#aPx(0@U@dUr9ul1fvVjaiOQn#andjk=y+(!
z1`qtrwNkwp>-D=aez$4CRU$`Xg3POm<QSPSiJM;r@CLXr>BwbcoZBsxtVX3v-3u6D
zCt=ryqu+w5=(J#t#By@^-7_tsO(9H~cT616=%*i<Y=hP!OqugRS;t%H^O?L{JCr4}
zAz%qUwosX=1Y*AAr3+$D9ftd@o9jk<yzdJ^^k@MEA|yzNQMgcCkRou?K>EzXfZBFO
zKcNmThQHsJMVj?6*cK;{<TJRll+A#LeGC&|s9ic!%W^aZ+z8K`_+Z?5z0!&dN_E&W
zBY0Ajhxte;CELC3)ye#<2H~ViP!puralmdKqPX*F>Cy3DI)I-{iWgrv$%t+5JEsSV
z?u>l-%%nnUmjT*gh25#RpX|Lerdj?w)a@!=!rvdOvrz*MyBUO@s#HeFkfr6!hhX(=
z-?;5$=Pw<)n~@CRvEz@lv7l37?F-9_d3M-Vg!d(J#jzRaz4ON0B9_cINu;J#q#C~9
z!{6~Vk!E#iFQ(~UKuLhC)xU6^fQIebq`VT^*qlVKwNuMIXbcrfRbTzAR;`u8%Q5EM
z{XTLQVPR75^E~VUmKKq&OSIM&oxEpa-3MK|#@=K3aR6aN%Ei;!t6z#R0vCeD50PfM
z^wH0@fhzl3%xsFP){>N)ei!mhv$kAX_RX41x$zFpz5CiLH1BOy`AT~^p8XEE<|*{6
zX%v(T43;Zb87lF^p*T`F*A_04iVu&8^hx6!>T8C5M2hzGKg0=AoY=5@%Gn+~(^jN6
zR~xD6(mQfHt|1c4CHcu5xY=RvYYcbUoG&?1Z@cWpU*OpLj_HGlJPskQ8V$GEb$mUa
znY`P2>{|Ugl*Uuu(<Mt-a5^p5#l#8mXPN|`*neRo);JaV@nYd%%=qK{HVCJL+ebUi
zX~oP9KSDQjJ2DE*APo|H<pB86=-2kQVy1G)hNLtn6%1dBDqVX%D-A^QFo;Q#3HJv`
zh1}(qfP~Ks&BS#FgRne?>;0$iY*Ob7ZggB4R|88uW4CU>?%ud$PbAec^|x7~uCvQ<
zk9!iC2bGc_^5QsBkk{6#a%`r%%l=`KPdZ1c*Vp<|ruyRx)A@>pt~}jaK@5^zFSK&q
zLXVk{_9GPu9lrLp;;8A^VPf*lX=B*ii~Aq9J_i46%9fi@VmAYyUWZj8&LK~|X~shp
zLH>6y&L7|cVi<U(rcih>!&5~l9xhYAve0(zM1<i)f<lFYV29N!^8(T}@y%=r(bndW
zQ$5seXYo(ITtLdp73L2h$9K2ipg!{o{PMj!b6g6o;psw>aPgA@z{sCEw=N|g1z#)A
z`q!C!Ag%bGN3+a7ok+}SD2R4N=b<#jt(aDBwim>lhA<){kJT<6wU3XMP58EO$1!=|
zXcS3>98RM0HJQ2q$1>?6re{v`tdIr@#xK|QwfEQrH}}Et6uE9JMhsHQEcCVK*11$r
z(@I_pWykyM$DYEw=e3LQgda`d_wBs<NK!}1V4%~2>*T?{P0Wfe4ydf%D*%16h#o#W
z_E)tq?8f{8LaIW2QF9;H+I=3F^e_)JIBFv<EE+uB({M4`^jHG`VR*9Dnt65U;#h!2
z@57c{vn!tt!qj^dC4-`|YRB%299QcOw%4$kucg5<=X;(1Ji}6&k)7_GHK$;^p}`0~
z-SYx+r9v~5(c+c2pVSrGz1VbjTNXpi_Okt0O!H{Edkb~f>F1J5@zHZ~0>mpkh^!K<
z(eE$Tn5Evy-b|n7W@`UF{Sg9Q*=(ibL$<CTwWd`^<wg1{$->FtE05o2g+yX&cxn7&
zi4tnU`&CLfjb!V>V_wraJG9e>w%CS;Az-dM-Uml+H=exOkjkTBxB1=CYEvk56A1Zm
zN=@FxKZc4~KKc-BS>5@oB;67P!dq53%!#U{rZidtK|r2+bv`cUAkZzyHM^{R$a)cY
z$FG+4t7aM?jOIbNc`uBD{yx4y<(GQrK!w394?E+v`OMsxY}6VYEhz|m3|KloihR+6
zKZBX1z35C2il49k{yopoXuDnJY9|Q{L3z0lA>U}XXU{7%?yI`F#t$1sO@-Y}6@*$t
z^;N}ueqZ$In1Tugf=79hBC_b$^8EQ;EsCR#phBp16_!<f^AT;Uk(HtJtD9ke)`A_p
z)49G%oHo`6PiSkKK+lkhdWtMe`5>;S_SpuBjUy92gHc(heBm_UTuoZ<b$sb`Wi1aa
zVZ{MOU>qgyMRE~Q$h<Y(X$k3-3rPlrcLdHkqih&%QMkkYNW;%bkxGaMw1<y{q(Cav
z?Xi6tXF<GNM|V!_Kh|z5@xqS)48cEd1wPgunV`HHl4>#(YHts&1%ix7C&r&J`w1vt
zKczy~&XX;CJYnSJ_|=F$II)VArrC)gaA=3gpX%PrnBPD}pliRp8>PpD0tvH-zG;Rv
zQ?^pXzG85hHsxPi`kn{G)l;5sSUYOFyYme+_@89;ODXN-!!r#grL>~?V=HMEK10;T
zeOH>i<y;UG&=5~k(w@DVGMmI&Ni+Psouv75m9eu~v0rKqf`wF~)DLtb^08xq7w4;;
z1<um2CTFK)H*gWrjhon^n)$u*s=`RYfW-Hf*FW-BAdY8Qj6&5mf`(+Jh>BRQo#5Pf
z7Y@z5zgXg3y$oqyj0MVPkFV7w(^u5E@RtbV@?dN;?_#<U5<+YXg=)TJ>;`$E9h7@j
zD7v#+kcrpumUJjh93V#td``+9fp4|-hGY4hIWEpvdlDJmmd24sOfc}hz3m*tOfRK$
z2?PV1CJOIwONB<8-H^-R43?}B=?cI&b(4Js6pN;;D<<i(=Y2#XDhxFyKKD80zUPCb
z+Yz#i8`m;8Xq0a6oubH7#zBtiD$%Snfz&moyEDSL6|ivQ*+(GXWrZLPI*}8js!6fr
za~cWn<)iwrvo$%aH~sG81z>hY<rZ4cwz2f_11}p&+PH~e{3}>nQu94*R`bI-rpZ*Z
z1`n1#EaCuaRgw`5grzh+RDfOBSZe3dTEt6c%&OMc`ac2EeY8Owcnj(Mw?Z(oRU5TS
zKJ9=5Li+pyyR*QY`o`g6Dr)>zXATTq-UG~)V^ff4;6UAtz0p8V<!8jeVQs9_dV5cm
zG5@%C`1c27UY|Rj@R1J}s=X2WoZnO}oI>)~uKPp=s~g~t<#lCtXcjF!#(Bikz!bJf
z7ap#~Rs@Sm2m~7r{2;16z-Z#Afst(0BK}oE%rJ9pvXK?~k^4>2gs;~+G1@?qEi5am
zZW$F(=zrn>!L*)CYr5WG1+OXfZA2swx0<bWqOcDy`HKt6!nHlEq9jOn_<wq^(SpMH
zy4FxnOUc;ufC%>F3xz+$k@&VF?*MA6q3RajO^hIn``aNZ7AhE7SAvR1*z*uUnC|Y4
z?e4EdJ?cX7AUKdQPnCMf?#k_5cbhOqcRmvQPdn}YZsLandZ3D$+Ic%xxYy_bC)p9t
zUhE~s65mVVkE6;!nGc?5xEOVaYT@WK_1neXHO`B6S}M_tFsggOH<X0dM=+i3m3u-`
z!I#7R8q**T3YC<1<)qRofYHDQan`(&+XgFLQzFkZ`N5j*kg^TBB$11Bc=u?E5$Bf@
z!0A*w8UQ>~OWJmp=Co{zR*2WH9QSm=Ss&-Y#Sm5T-20`##1t|7EnF|E8y!qXGfgk_
z5=n7t(0#|q`7NNrmm$xWFTbX6<{ahJ_enm*ywUscUo1d~-;3in`OLaqcdUTcy5`b7
zlsAr!8(FW-Fg$IPs98G1P%ik{A2OFU@79u&re1U<=o#%Ck^>$Rx#gDI{F=^KGt+Wu
zU5qA>k${5<F|3}(U&m|*io-Yf$$$N?Tzb=rP(EfgP(i|Gsw9Qo?fy810@v29t|-J9
zq<JjLX$dWU{hMO)_TKl4zOlQ9fSa*i(H>)#P`UYo7W*am(n~My4T;P%Oh<|;DM1DQ
z4+cyXLB~jSesQb#tMbJYvv>10v3cKmZxosyiQ-V~kvu<je~k=H01fpe^WN0MQ`6eu
zV>vu)(xvINv7`&X%|mKy<xDfv1#@61$G@5Ko6RK?<f%Qs+|k!^fbvz4OvWQ{RyghH
zPJ}Eyl>(=>&-j=-`q6$;3S2wyOX9*mFLOnKV;Gr4+9);)kIOEY&14rQp{?BPd>`-a
z{Heq*^|;W6Pt!5WF^~%q%DG2G_`WbwmBEw|xVwg8a_&M*3+e!XiX3mR0`p4AnhA~5
z39!@;slf3<dR#m;@6RHsxbeGHx8!lw1CVM&C}+)$wB2CpE63ltRws9s0X1H7PG|pD
z!98cOrD{@HkOvuUk$i^xO)tCln>AcZM5?uK!TrL6K(p?8b=I?O8=AFU+PM{0^PycX
z^6)vYa%obG18#>0OHL51X+FS-qsqA9<R`lN=R5O4y{B8UaWh1ub}@;Fh5naFU*DvN
zds$_ym7j-!(o<CLJlb!ms{4Wlnmg~q@mJ|~NkaaieMLGgNN3;25|FcwyjS`5t%vTY
zRx-N7^{_D);80fY<Ome<<R8r8(ThoMuC%wgr1bQ!miQ^4(<Lqe8)#5=Y`hR3E+-Ix
zoF&uboqG(JCLXMMv5T<8Todcb3mhKizmV8r$drmZVs`<vleWJe@BYH$P})u%=$zC*
zYDzj$y0%ZNm)4lV!sJFBg2~(RqpRJvG;$2mVN><-1Bptw%QCL%^`HOgVp1^p)P7y=
zQ#)9$$#I8s<ib!yh0jq==GH;9Wh={`V{8rx`H+AW<GGYL<`aLL1&&s`<n-*2XMYM5
z>!i)J7&f}3UEfC@sG_wu1WJJ3T)oyiS(=8eq<iNGBl?!m!f=H4?Ojt;GJtJat11Zh
z&fBrhvA0_&La@{>^S+{-ESt18@ug6r0Lq*<@z7eC1ta@uZR+6NnNProdJH~TrFQbj
zz57DHn0Trq$)um=AZ-dk*`s4pv1BSHQ?gv*q{TXiuIBqERV3mY^XFXZ^^rUbMg3bK
z)?jDlrD_JDm(|>tbOCobX9~0_effs3`GS{-uYKWSkH(6xnhy@^`Y$Se8xYfr_*z<H
z@Kb-R3ZZ})B{R8@wkMBM%(%sVyQZafNrErzFjUL7mGZQU8i-Y+18lzJSWOBmnWW8z
zcb?EJXy)z5NWBgKA{u2pmua>cMwd%fjeS&XC-o?+SblJf1pj?Oe)&z8itavw&pEhv
zxb@m}CHfyFtGmX;yuYU0wog)1?|U*YwMED>$@?6GGZ}lQBN6#==a7e;5q&u!6;h<q
z4-0|(E{(RkXr{Tp1H|Ki(VVEsMJ~Qx^A}r?Fc|wf^d?@?b_Wwlg&0GH0Nt&HRD>It
zS!FRb(Z9o<C1L01(F>z(KQBK&ujC##QAo%p2@x8;<>~ec(kfOB0(SF@wc3wQzw4Wk
z5l;3e<!kl=gT0m<Afiuh7<?r8mG&Z^gkY&z)<r{fV$%>X!dq6q``o3W`1aNDO97*a
z9z(|@L4RBi6!*OCm8Sw~X4wlue@rpleS!qE`I~(1`18oy5nNyQDdgFn(p7G%@X442
zI>Aa@c;GQkn{dj?m6W?N`~4jVKs{cu`r`Mk#RYSCONN3_LGl<=c6EYvK=$DqpH8t5
z=4m}|>Zq1XZB^gu@>dD~OeD&14kE`oS?J(6Gl+h*?th?{%fdpE8SG(W&SD*MT@K0O
zk!OFl6)(~zp(A-{#3uy!eR@!P;y^@5Y|o`z?Baenv=`NZesm^<FntJk`P6Aev;58n
zPe+KBlxwjMu!|)b=f`$0A1aF}Q=CenTN!tF*|1IG%68GPX)Za6X4<`D9;=D{G^oZ!
z&aI!kKeL)3&lveEOqr*>7nDA=i#eW2tJ5elu1Q*7Q_7u>7FGcBJHK8gM^(hwFB^9M
z?2X;%;!+F#T$QH7CI(cfdc0u@FJ#1i=s!F7D!tyc=roufMCTcjJa^vGU|r)z`g40Y
zq=NZhOTr+R$|}&rC&}h<Q?lRw7#70X+%^ekUs8<COE%TG-&-#Y%;nD2u-2At&di@0
zS46`;pPTv@7Qmw3KheI`%(hoZA4zMiLw)~_1g0%b`%$w5(4~ze7vElXMU?3=@g3~k
zZIoY@i`v$qek%u>e34lOR1;;o%8b8s3D5u?9vyp@F#1W7k}z5NQ=&w92{cVJjRO?9
ztA*O}+YM$hVY?+yGE}iicrGEHcVCoGQzpY|_03?jHl4;L=`j6)G3@n>$O0@tYeB9i
ztkz9X0qoBsK4(LD%JxwT*Qi!?GsT{bsPZVMfS385>&#px>~)-UIdDS*sjfMfE4}++
zUP;)#fxB%U0?fJ@rKtsEmmGU9(4R0GF+Pb~x<=;e)X+dxJgiR@QU!7Ue0Rck_=BHC
zZGOQb`&tgFK4w2<jqSPVLQvhh9l?}tsWRp(@u;{Z`@94F&kS)m`R8+;#Q|ExH?!I0
z4P_Emjy8H0eVTV`v)3-{ohia&G{=2H4(8adJ71UM+RH#aM#m(}po!<4=p7dZ$-A~?
zumKbw+UpN1&iyKbrLpd^7?N~@)T3Zo*yL7c8m=@vQ-cjw(H7dV3nyH_mM=3>#v6Cf
zA|qx+&JAUWEq^k%5riRV7p4|%ej|kO%J1elrDK#LD==|&;fB*-ln>vwl(T|-7(5>*
z?H6l0qAt-bn5Sq_%W~R(ztx65@}I_%mY8!v-ecmU4(kRh%t<}i_MQrPd*--s;})LR
z-Ira%h3`TG9aCPlbh5>6F%@S0%5S!_De7@Bsod>7qnP-q6_SlurcRDn=l6-J!OJ-$
z5PH7tWHQb>n#We8Gr!8-L{=mkKBbCDqE-a`yETvIGR0W7l`95iuS#&aWC8xxf(Dsy
zW2G~iG%k6(I(L<9Z`+Dll%|Rp(|hl6>&a)`Sh)&yR+>AlsFj?xvS0p9FwNSVhbE{Z
z{!o`q$3)od7YN_cp}A6wH)!qIt{fX$R9J*W)^-T#J%yx-86+ttinD!(%E{_vlQOuW
zrY7q7Z~X+@w(EQDC_890Dh#Uo$??W?9VWg}yfF;Hhmh^r-#0>VY2=z#%?7o;T0JnA
z8}QlQnqSc=>N%;NVmFemd*Gb|cvp_^b{=0^8Q2ve6f!wUI_XZSVbT{7(5?|oKhw0;
z`{4J+2`RHr0_3H4fidaoa~y#jl6aSB*op!$x=)p-vby5$G|enP9qJ2CWhZkhy!n^6
zVtc0Qd|JaNSl=AZte=-^lN86MZVO`0382Z0lS<`yImwx-$tJ!^mRovGrPme6fSkG_
z!X@Hv?3C~^*hTTVEcZ9%TbC4*$=RoM4U?G+MLXNP;)r4$VCaYR9uS9b<wNI)%P8c-
zmjMpgHK^8(F)muzxKY~*0I958Gi8o{vSZ*&v8m~6od4WrsyG!!Foo=m0~xC~A8u0-
z4H+gpni%Ty`ODq>LE0Vry0DkNU2MD;y1jAT7uH4_*mZm8PZNXUw-Fzu5L{|=4018e
za@go>hMK!0M)KC$OZ-*_qJ-lcsio`TeRNF7%CFKad$alp5Bt%vj)Iy8s><~O-TTT6
z8O2BKF{?Fh?504$Pz@}sU{0)RdX&zFap1kc4f<t4(|T;Iq0*Ptqy+{(6yt}|ESG-%
zF~v!fB;1EkYr~}=(Y8;J_?k;n$=24*FH55ifMwPer3e#W%0M?<M8>pZLWV8n0u}UP
z8N!w=FK@@vB7|a}XpAHZ&Vo;#N$3LHNkTi?#lCkH>&>oN17kz;@?siP+nD%>O~Rbb
zk<EgN)xG7G0tT##pyIheO3C%sdEr(|Zn0kN?X0K%*z1<#gXb^vO1`4Mr0a1@aI+DQ
z`Iu8w;!X?(UwGBJ85GmD=4J2MG@4fHis~?}cB840Q4mDMhHxCe4jXANql@NSh@-N;
z4|ApqL3=%p17tk0$BP?mu>N7Av$VFc$x%ddG}s_sb;j>$9`{5P!;ZP(X;hnUksvY!
z1|XI}zgeHxy_#a#wfdgP0ebL6$5?Am94LysaFjc<Nh{WQ%j8MwspRE0-gl0qZH`(c
zJ1JeEOly4sL5z0TaC?82b)~qh93X^B2x;16U4+EVz>Oisf~6^QBzez#Eh`0$DYi6y
zB<BU7u4d35f1fhf7Ic$-Oy9U%Q0CX!3h`RL4D*$=y&6aTxt&aoL5|f-bLqke+KR1+
zF+^0WncG+}2qe+~Pp)c-tUb|*-L09Vxp^@l!J23foWOcXt<Q`~3{Z*P&-$nUQ1dh+
z2T6d2xvAC~zcXRUD#G!pf5H5#a9~ez+ph;gQZ)l_)rL<q!WxM(P+-D(%64^SOj^&?
zGYg>Q3EsSYU-ArDBpfWyW@GeaGo0+i+Oa$RcWP<O@TJSTcr5>+d8+kJ-~ud!ujWI8
zo9L`Nzb4j=tbjWy)}>AFTNaaKTlJ<p?(8Sl1h}7qcE#ZnB;#ZRD|=0h99>JC+>RVm
zniQG9VZzn#&8fLu9w;Jw6oQJV^<t;UEHeC!NXkmJyMZyf3di1!d3LfkCbz~hoBWmT
zF0qD@gUgkp&Uj;6Y=4Nt)yBl8HbJbRUwRuslaR@ZKu@vZ6C%GShvzd6fw$6~o1heh
z%ww_wi>aWOspcge0ddao9YJPiF~z$%9#(0^){`INC&WR|Zcc|TQ_Jpr%a@%Ym-wXQ
zuP7!w2~5on?4KU~jYq!j<E?SqeGQ2{L_A<hB$g>Wd}v%fMZ)d3Z3R_`h=QzgKQ)S4
zMuo(@F&l6%>Y5@0M$1yIFvS;2<Shnur2Skx)!OP;7MkhaZpu2U?v4-9N$YuH2=p4s
z_jUp<6RE8WFU|Z<>FEV`h&vA((FhX0ei~66o;^y75J0H`$ufXjE!AU-lb2;<EZh50
zeZL~9MAasNwzx4|Rz5srhbKJxMF6SIhB&r{zx<yQ)o+Y5Df874v$0rpKMPCO__%2S
zboZIaLK7zQww!IZ<0+gr2c{Mzh!w=Sq;PKiY12;o4cIIeZ~M{9`@C)*i;Z4Ce@!xj
z+dv^Bd6Lbnld-IuO?3yQM)Wj9I?`mYkVb1I77N8J<u=eFUZ8{`lJA|bRtp1jyoF-M
zCQ8!U*-uOsjz|9@+j=l3;C(@4^h*27AA%?59NVM?_nb(87{8Q7jw*JO{$l~2{<V4I
zj>p{??6|t^eB@^Hro_qWw}AR-dsDBz{E{Yxn;0vra2g#9ft5HG)t;iw((F^a^<>6f
zD*;zH!22;3zJX~3=-B*FmCz%EzAtGWsk7Vha#nGUfiVPSjJuBex7dNoK=Gk8KZ-Yk
z3n<%}PTX*Zs1JdR@5`3opmmlaG9cmC)?$M?H1d?#ueZN;(!SG1bD|r^Ae>gb4RCCH
zy@AY#j(4LCx`}_b@K5KaR!Y@c&^*7+2vBvS1<K|$oyzfUvgRfPLq~?;+J^SYXKyb%
z{?44J<f}T|5&B6t&R9-46uI;Tt`jD!!vR8SkP%I#LJE%J5!H+lL>NF2z;)E&%SIwk
zLcAGhy&Pc?0ngtbO96NzvTBz<!DHr#)l&dTk@5?|1ZX)jVxo3A204K6f!J{B0Lgk{
zY!=h-d*s{I%_+*s+UD6<G037CdZz?xAAZ_6ysBynQR{NBrNKMZR`NLH&S(>Ux+5A|
z(me!p^0SL+5yxPx!)1JfH6=kFe>afyJ>DAkG$V`-#Fg*XZ!Nh1H^Py@c;2tg07J7_
z?e6HTb1Wh$xFrOBG9+dH`39^s;IXO>+nq%?b{T59A<j+FCbJE8U9|!SSnC!~yA)46
zf9`6*Ky#2tI8W@kyjjwFJLqhJ$vMQa?wVYCAaTu=m+i3d&P9mS@F$zO2lR;-E-|T>
zCeU}^zM@J}I9>J!^o))4ft?gbQ$AF5w>wMyohx_x6%FesZ`k+39rMgKb=TYbqQcRi
z(!tq5yQo1FIV#yCPh0z{j|(6~2myPLHRb~=MR>^d1M?k!o~Tx~XFKU*ODCTVyYfp&
zyqq?ha8O&r#>Y?`_cUfd_gP#?M5ULwT}V$qS=(o88qiywj$*AuYU~D4l?8<d-L<JU
zomHc5fA_NrFr5^<%0o;gZX8~-Dp`Zj&!{%N>5dfkC82Wv^idMLJ#GuNBZYiC9?L4@
z>90yvVwaUg;sKsB*!Vse)$K}Hq!Dca2U2*c*M-$?UEBOQLq%>yl9$R$SHfSe3V1(}
zz)?UZ7Z-E-^j!oOiK^ccl79^L&m!E~&_{CGDsGy?nIn7L?7*LLkb#i+m{31kwzp&1
zGE1$bEgdD{@aOU2ey!NZP}K%nyulx&PjXBeKx)b)dGTF-&R)vvVs4UC#4R~z$OcoB
z3kZTL0H9(9opF20d!edz(J@;_B)ufPbe|Dyf<cs?jQGq)uD`v6I48q}jl>L=8d3%c
zg7AQCk6pMmReNKkERs@A`J1w`SRMZr2<cGUT04PWiBvAZ>8+FRr35b<!?s}QSt04#
zT<&!do!pp{LZ%wo8WzbkWc#2rxWkTQ7^gh3ukiJ5?WIAJut($0-k}Lv|2z@xcaj6m
z3kL<V8cR~C3Vl?A6Bk;NS*Cw491Sn*_PQ-)NyiD6gMo1%R;a94rU>DWS;x;{>r?l6
zjn3_$MkhD$D!4qh{?EUcwh*Qu4*^Y$8a_Jrrkh5t<SBip{GCDHfje+_-cKy#H<S}M
z^z@O?6V=HDX56{;#cC_p+)PysWPz1pf%xNZfx#%iAw^aoMj}S?wk34FzI%NFC(REq
z`%)~MNNlN-5#`ss3;^n__ZT7zc`^I0p9pPv(Qv8ZK&<v`eWo^cpWl<qlcCse1g_r?
zrB>}2*we-Q1w48X&^3b<@NPlU$Wh2O90c-loqtDV#W-7je(J6-W@)y0=T%7A$eIst
zDbEA(?PKmzJlC~N4gGjmo!NRsIXQ@Gs{K?Z^thk+6W-=sz~!$BpipO1|76Mj1(5#q
z5L1tQH?4K0C#CmjP4{m~WZPMowOHv(2Y=!I3CUz3&P6%VKiBn<{MNIm_vg-17k~us
z=eiBB7|JDt5LA7yidb>1!{HKMTU3(jovqnhHwHB?$}wAi`X7w=ol@XzoctT58<#WO
zNCqGI-EO)alN8z0NFzC77w1H%c26{nNUUn%(;0c!Knsc~w}4x?(zEzdIaN(o$uC2F
z<_mkePwE!1%kld;jXjSJea~+<jlBu&m!}Z^Da9tlv&>2C`K}{xW4EZB$-r{5Om!}=
z{Et5!N7tJsc$$%Qm|T{sbx<Iam+_wGi#An1)!%Aqql!Jm=gXB5Ta(_q?Q7nb?xfqF
z_1}3&_sBMuKv&MTGcFQW44Rs4gHpfsY*E-6BzG=(<DR3nFRUjY7Vo!3vswkpr)rs{
zNh|fpUt%@U+iUs^K_@#xnn<jTYl+g~>E9oj>~0OZtgX8pV#ksqqV^O>-XSb>i?-T2
z#I31j#+F00G0=BtTEyqabY}uViZ}tZ?;*N@W@QAKB38B~qd6#m%RwLX2}n!yuLKWy
z)!dQ`P##bXVDo5CXT3Sex>iG(tt*urn8HufBpFAX?XM=C)A^AQDK^COIUXbqdhoC_
zruoFiEKCdbY<5mr1gDHTR1Pc*TkG5f#(Xa=g+VWI?_zVpWdUxEf#XbowMwp=xI<5~
zIL%n=Y~XC67*KU~!0o9CkH4eCcGY{(BZ8}ree{O;$F60}W1sv$4ll21y=l<U_HGDm
zz3{@T%a^$L6!*xo5Jdo84G@zIzF&Yhsr3EeyXIK;R=Y@cB&OXT>+wMG0N?SasMZWe
z@$nm6Y&hFnHue6z8qBKDf+x{LBaOdYmf{9+L=fkg-|;L>T4xiKJvA=v1Nxb9>?aPV
zCcI>+F5GcCuWUO>$kCVm?D4B#is_G3k)5~+`ptzP%<H{kZ0Z|Ru&Mfezhr)cs>cb$
z*>&mhp_&v<`ZauT#>*8uUI*1j(tRvJ;ZxB1+D>5dVlHP*UMo%=R*T+pday0qcnFPq
zr4>s{^y@xi#cR*xnwq8cO!5N^TA0^0U|awFeg;*#wu9BMKi-0#yd8IC(T|QhP(K@z
ztnWw=OggS%&8C{l>#BEmXrv`pWq(<=m(?O$oB#5=ZXXnZ^QN6AmzFq^&A7ba_cCsZ
zYW`%SdfcF`gCS%4dV;N#AQfVe-y7JICc2{+Au&<AQS0(R8E+`IK|S#V`Qe~lGgLz2
zj+qsir&Q|Yiiq|`8$iGlDmyN!U5ZcVe`;V($l^D+rU-IqWS%V7+BtMY>BVXu82)(h
zPZhrv&FKp2$TepIUP?8)#(b>TAdZ)M<ny6e6MXTOWhkv!Qu@^{(qipKAAl2I9H>R^
z?!2@QD~jx>b>pn`dG%PNg5UX}AOP@96MVw~k{W;}O;|TGPSQ*ko-TuCNaV_ViLs~f
zCodCP(}7s!_Aqce2izU!bAWT7OPNRnf+Jf93g;@g_hU%QuOG*Vdp*ZoR((W%^dU#X
z%ea!;TW4<Qhtxg6Z;Y@fYkcU<RGo~N#}5-{FzK!k95e1uE7kpI9dnS?g_O=NT)>ms
zD{tD&=lOcWjmb<oy5CQLmwzk2et~)b?ONck=BrK6uJ$JBf~z<YOeOS<3DM0z<@4mm
z4fek7Jb;=nL~n>|5BhV%i2T*n&4}!$(rXh$y_m*Ys#}-AtQ#k)<s2YAL?i+nVHL}@
zNWnYgYA-oi_VVs*W+TwC5x@h)l3sv7qzM8DyEN%x!Ah|0ExZnn4_dwXA1KiK0!|T@
z(v0Hy#o8cWfWA{Fe{WK$72sDZHcT9=4N6l0Ws;hLAfOeJLLu*cWp#ZM8)&g`kRaX=
z*Ay`Uj=q1Aj)26DvYykP_z37oDI&K?Kk)&-Xe<jfHv@B0bk=*!x`|H$NGMkAQH3Pc
z6rHlkD_qwtsR92iI!tszzkocHeX=NOXuHP?8a@=&oBe|wVKHn@XLK*;G$re%MOS+F
z17~I?WNcKT{+|Ww*uMCw8io@cgBQ+P6ycf)+!i(-k4H%siME@N&F8Jif_dGtdv^w6
zHOnYAc+BYRj5qAaQ(u+Lub^gSGD<jr23>$9GmkH(;Sr5X&X2D)jUT=yx-m|xGlMF0
zkIN7;z(t1E#ORkHzbd;QY$g!-wkdEHZk>OO5Q=AiWqSE@?e=0OKU1+8#O?etr>m}P
zlKXnfXTC9?ydh@jW^Lf29NILuFzO%#ycKbU?zSUvB_N9rcg~O3feTa~0T(m_7yKD%
z81IU{dh4=aJorT7kh$yu5ej@*Gq>QR&Ps%=_?i=YY@T^|z1*j-Jgm{^=q~S`YZ(9L
zc6Fr?#++`8Grp(05LknsZ`J3T_uHh@zr<0Az)G0~n@B9uioJjt?)=Vs0N<aE$d$*L
z9MLqAoB84aWuU63KUNpuEjk+nwg<BWfcVp%T)EDw6#hcYpkP`l_%T$J>|h6~a;{_W
z9H6(5yV51;AV4ViZ!(WrAxK(p1hN3VC5$l1w2D~rJL611Yd$4Ag{UYeoXZnH6=d-1
zbO5EV|Lx~NDAU+(SuD8yKR&^U1CA8;H%<@DD-Oh#4OeoPt8^u&+3kS)pcR1>dw>9r
z8I-WChSeK@x-kr>n*7hVF?ld5f<9B90l`1#6_sQ<(B-3(E0^AIt>(YHl1yDdZS8*u
zAA3?bK<i(7^@wxSUn(lNWkTA$;Ra$Qp9&A@zeLv*$canqzf{&<>|3;~;e2)Vo-Od#
zAaeJws%QgLy5G%WO_#gJ7h+@^JXhD*uPr_lZ>%1(%v9*OKB+z0_4NwFZ}KiIB04rW
zkae>9YVRdIX&%$vj+hv=$TJUn^G9-lNha%&@eaFw-l!vv#P;~=A2ed(^QmVbo94g<
z`^XJnONjvHg<pwK5|DRTpLe(Iy}D|JG4j=YPnMdnj_%Y^dm%=exm(SoW@cAH5iMQq
z(`gCV5?{8#;FJ8^0pqL^6b+Qa6=eMw_`F^ff9)|H9m)Bq1-nj%zY_jso}Kp{yrAui
zS@QBJPN?$c;{dgnjcK2vHLeT{GktIL;)Zxkae;*a`-<S#;Wj(2kDLYnUL*0CF8%ij
zwub95o#L2PN-oF4(XL{DpxCqRiAzMqwNe?^v)EbmspGz)y!=>-9W0o<cBZaggfJf7
zE02u)@2>pcW#Gh|Jii{w+RSc7H$y?-vYn8(Zlaf?o2*P6GN-nLgucvNVW-Wz{)Ge}
zV&XCz?{W0~JQ9pJ9qYc1V?qa*gF~t00y9^VU6cnXzSjR23&4+q;;3!KJR$hC=hjjp
zhbf2&5SFk1>>i#5xB6VAD1YZc)#0z5;B@KwaIT$!1Y$Y|aXRthp2FcTcXpTKkVJcB
z*7K6HAX@sfS?2rKS3q6<7j#9B>8yzmpw`Bl2{Hxe$axYe@h@zpYDnaqGABAe2F`Z6
zh0c}5O@szV^<vGH!e#LjDB<plcZg-TP{X#?a!7WYxIcI2$e)EURSkB1DbFS8A+CvF
zZIo>MiY0_Vj(>y`wq^2!E}s?|)0GD*F$qb-bqHM5!bcb5(uKo618z?QuLgAJ<UU6x
zA3yCOeg`;Yd%{`sgPmlPzMgmV)N&?V7+)wsu3flO-FoALuxq=F9{vQW2FQPH*m#g?
z1d@*WIwAr_Tpnq6kjH!DB!5#8dI9=h;gf1!TG1?eNI_H<OZDkSP`IIx0<c`TRh}MM
zv$qG46RmdOuwXlR@8=<&v6M)w$OySqbQ<rzGp13bZzHIwCjEuGzc8Xh?|k~yebJKc
z;)|?K{l@Wb=Z_QBsa<6LBwf){x-E3AKXRgZJhIm2bKwf;1J;eF$dmS-{sa+t?6^a?
zOAb&ozSjI&&dF#+5zY0i#c|6wm!!RRPzdiqu5>N5Fej2&&5~>&d$A(Q8z`RN>*)7!
z7LbIA?#^y|q~f4G{97iZ)vFSMP?g{;bU&Lci|_Rp-A*=ECRo&ZOgk6Ap}UiCu`QRC
zwnQK0gZerC!76O;ImHzv_W$<^3xO4S|8Fb2cbS|*?#J@*Ya7SOCUvs1<xD1A2nosi
z%hl%PqR-`<)%ma;(V-v<uaGtA<glbwfl&Y3g;PoX8zg^hSQ2++W_>Q%0y$Ets?P{s
z6%d40{B&hm(B`e~EtBr8J7ZnLC_ZVV&6HdB);gxszjIn{LPK@K0JC`(*6#5yM{?}N
zJ@dm~hW`qv%l{J4DdzhkfPne}0&4O<0!s0WK^e|6xR*zFA6Y<@y)mR1!%BY7WQQ1x
z`+aL}@za8zN8w_9u$#&usD7fG9349pn&_hZ(C0Ju5&A$2@d%|DH*w*%)>8C@PLV;L
zvur=~;}@j$?+HKzSAtg^J9O6dtCNmx0qWvtf0$~*Gvx-R-rpD2i{^Aa;m(18)&vhT
z-(Oz=`B@UXR!1Wzh`^W{b>Msg(PZSusJ?e3MUu^c+dmH+&ybub;JskOS(OffbG$Ht
zooR=__TlN~kgVmLWb28p$7o8_^^Z-24%87!9({nU*Bjzk2}Wd-27^4wQ+GV#9O(86
zNr{yw$DhW!z*6o0cX3qE?N$%w^WXmp;+6_LdtIGPwufj-pzElLJ_YopBRJdRe;#4O
z`HbCXh_fZIVkO9@<1;SJ3eNURx*j3}41(x?HuC!}kp}l&d`^O-Y}j%u>_OtAhXvX5
zy9V7`b>jU}HWh1D%A$gu)8i-^fILmy9jmq^e!hd^V!l#qCo?j_tr{u36)zk|i>fB&
zBJF&1Ad;-1%&wJpvQ5%X5%_5m`|ksFIzH`s!h&gjp18}EK%}Wy+uE_~tmWz{;w+7~
zPh|hP|L>FCQVGajS7S4+*kuPPK@h>!M7x8HZyUgkWoh>D;#gv=zlLO=Ccms&vnIY-
z?Vc`b@F-n_$H!j~CA&hlOq4E%<@@LK_A4RTE#jFNOV$<maIS<K!ytJoTgyPtS57vS
zr%IP`{g>V7?<hf7#QVop(X)shhUGSzK*QzT|A=GT5LZuYN{<1s`2brqd@9r)Zjb#m
z-sS4`Wp+mfu&CQU(Y2h#Ud!?iS07pQeJTWxdR<^AW#g{&8*o+lHou~|#$u(ERHna$
z2k_$I@9W@=1h^4{NmuEJ*-gT1JxhzkpU+oIWM)2>y?2rD?jo+2*wEm?zCn^Jq^?>p
z{}pm08Q$Kk`Vz3|Snlv+7sT<e&6cW{|0;FEIl-=y%6Z1ur(O0EvIuvQJ?dH=>k3^o
zYs0!&ZhV+sp>+YAh<+c$F9VCd<~%U<+5fiZkF5qR(qCBM_)t7jO5uK9ZAOrtFwpM#
zSqLv)rco=d7hcp1_gj-$Vd%h(T!_<&x>xVK7xh_lR0blFxmjJGT;r*Z?k($*uthVv
z#N+6`H*)!H@zahsn~I*%6_Q-io+wFn?+&1eMBo2PC$b87oQb=ttVBT7Kg<Gl(v)sL
z(S9qu+N(0!ztCkfD|o;Mfs;q!t%zMtWB+QBL?Xg1#=0I^IKR6X-)odRzdIEia?DrS
z_OuBuZtbJtMYJ!OAUNz!ej91WmZKx=P7|=DKC|QCAJ!~|g8iMoxxEG7GflFt_M8v&
zv*3>;A~LR5_3;c>NpRrP+8z6mPMF?SXp>h}boltY;+(_%K!BC~{$&%_xn0BhhkUv5
z?PIoy;Cn}UhkNYY2lZlPnv3kVlT}1J-}V2>BRRC%_Yl#+-rNxAI=vo$-G{tm?Z<-)
zkC27rlXLXw`>05O)v7dc$k|l(Ula5)cUj`tT3q0jU8&#U-jxlPU`|GUsDlgR(-&$r
zllK7g_!O$re){wCe|GwRnQxIr_(*X#@&7PG<>3Y3*WrzCzBow#B#BxBwz=$I)rLF2
z-h8w)_gh~dd3uK-n~U%AleO%COUNt<C>1Qq^bKc<Fk);$x{Fy)OU#afe^{|_eUx9H
zPe<9%kB~~?tDFl|s}5^!db;iANg=S1%cyMT-1v#CM0MpacaD@v3E1b2FRR+G)MdPj
zd))x@1F+uJ_b-jy2_AqjI*n=w{i`ePhqjwpQIY@eG3W{$gM|O8iBkxSB)5=BIXHm6
z!`Wv!ctZ;ULV2O!jHe_!WS5GOr;FPSoVMY9yQHv^{X_KibsPTDj={ezH|!*V>EWRZ
zZL<&pP}1FcHD4XA9J`}5ag^%(uhZjxc<<By;q(v=e>wka5KXSml=*o;c>^r^?f}{`
zeW*$upd5Gsn5F-_KH~qcex0Zq62K|+{3&NBA;7^d>Q>{eOcIK}vBw<GNlD|bU5`W4
z81VLdij`zbT#|9unTvU@{mLWZV{9zj;;Xo8B6a;Axw~k)I{SGQ!PH$}RxTD>9${Ob
z&W>1FSp-~k78->Gug|sqKkU6%R8!s8H=06`F4CkURiugZ4$`{>1*8j)^eWOzf=DmY
zLAnx}bg9x3r72Y+y+#mGY5?hheB0-J|7V=>jdL!(b9FA>I|gHC@44n$Ypz*-YyR4d
zj00!<c9k|1U;Q6?>6dGc=zI17^{hZyvjZrB>0+=z97#8sX)B|Oc_~emyS$~Pla8zd
zfXV+7#AAr0T4tzHC2x&DZ9%(4UV@%*+I9e2WHA*Lpqq{y`x}J*C6dSRO4R;+?v0JP
zB2ciz{u_Dz{XfxDa<2cxbiA%N7MM<1xxlfYHn20uj*I;AAN!~KdxeRPS1-@GRYU1F
za>?09g+YJ^4GCqcpW~2?wSb+1eD3sC@F>&-?QmmFu)}D^wM#8NT@2utr}I+%8%sie
zb!()myg%)@c1lURy<+ZhK?6iJmi^+KcUouUL3_eQ#acfo@Yis`Z!;RwuN~`0!ikhA
z7aw(9Wf;!;g*5I3VExcc@kZ(-wlgW-7U{fy16vl=zkw~ugIXv~vH|{drORPu;m$(r
z7RdVkW#v^P<tW!|U2`~pL+;Hbv1<SO%4#Dzm9L4dp*5hNDZ2~6n&kP95&$tvEJ?9Y
z!{g)@XoLB>{}$r{q4UcJl<^4p?N0KbxO1#slcR54pv(n7puF6_u|EqCR=Xk$w)~;I
zM}@^L-`QS0EULfLv%L7DWw?^&&$mcuTvueQWbkFd>Xz4<YYDIJJAg=u*A)lAput!W
z0{Lzu1b^kuXh%aL%KF`(_4ykSKn$RVC$DZ45P4G}1iQTuypJZcp;r{*687zb3ZWQz
z-SkPae+La@oi-t)Al{?=GBmr!_Y$XRfddAw<rYR@&W!+U<-c<dW0D1NxeDhotp9##
z3|*kri~&fGfa}`G6MY6?0QCQPQyy;(n9Ef%o54;6=*1ENQT+Dt<U$n3@4$@Rqg8`5
z;=BO5mr?Xv>K;J2e-~cd-vzZs4$>{%r}Viaeh1FqNLdf?mzrUNxWB?#f;fF{HSpjr
zsZsy8AN>Ey2>g#Q|D%EbO#_AEok4(Q{rGQw4cMi?4!KJfM15*q&kFH?;H7+&fF@8x
zr=mEPw}rirSx-%|dRSsC8I}~ig%IpzYhrEipnsJ-uz$&m-HVh!><s!AnI)fu7LXSZ
zS*zkv$|PZBvU@s~mvSlph~je45PNYPuMaiO$=)(?BNM6S1cYOMt7*WOlL^4iWKp$(
zGRu|>?gVL1q<rVd0ZZhu1zvA>{&KC3NB-I)f!d#!Zm)vZPK#3R6mC@d7F0Z1JL8zS
z1v&K@q?~!;akL?u9u=Q5<g?M0v+qC2>w2B}zuybkEg)vLODFky5l1u87;(Gp+wLU1
zL=0L-QdH$Zcs1mk4-&dl|AD%_;{DDY7VRQJOd515NC8w|-s6Z#q*kq7S)<414Y_-9
zJApFWE=|_k&EP3KkBbMea2oO9(vH#Bpma%~GwOeR<<<VmsW(0c3I3yx#dEZMzn~0H
zC1k!C`%I5D*i`v7{DEA>ZvLuisHZlvLjNMLwrgWNVaNPlCbV(Ru~~n58f+vu|6C<m
zt(A!#VTKwsmzdmTKpXWR6NM$5gY0%?ThIO2zEtQBDr)1c-Hj>>4quuSnWq2q-`f1o
zs5d20!w-sDE8=M|rjeR`q=$v*LFV%kn;eTgf0Z<W9X+fFC1;-#9cRJbogx9P90@Px
zG`o76?7M$%^@KiM`8f&y>}@_DsOfq!EqVS8bCq%Df{|f?;9B5WKbJO8va<EBTI48O
zIy)x4?;NSr19u(&PW6vtj5w#qd%}Cn6@g=2j<0Iv?=-9dYYZ9x1<(s8KzvG)psy<V
z+JLy6iC~*?YOmbd3|GKPa_`N50WQ&`SLFfjy?flm?{}98^*NcXhTaDqPVf5te_q$`
zb6Xz&l^tRTjxY(iQhiD|7&_e0C-lUArXd~G&K&+g;J9-cYUt3%fz38)_LVFmjj#?;
zW)JerA6rTQR7u2U6B6P2l%R*@j)|b5;g8rGI0SS_T?Qw-`*M{Ssh$Xb)oOZ6_i_vn
z)Eg*ur}K;c&9*!OR%af1dEj&DkYljqg%94GJIB(HTWcy`>0NX{N>zco*Hc2M)n&Pm
zGE9C$Oz0v&bf+`x&0~;hypeJ3XdvT)Tk9S6z#F(*4#;4#w2NJ6l8V6U!%c!phZ3}8
zYR^|>bD>*1W2V4%p3rU5>WnO2(^<9&8S#8iNWuh<v-bw`RTckLV=ZtNVfvl|Q{4u1
z;L@eDXo31rtCx0ke!ZD#%%eO>hZ`8`2{aL-9VXXqQmnCDT$g7tUubTE-sKE};*Kz9
zUc}Sx2lJPeV%M79V^S#c*#duDtFKq$cyk0%{d)25l=5fIcr5=|gxk);r>=S493F3E
z_!cRkzj39SKPIHx{dd>?*T-{!b-r>3IKB<-r>Ir^iT<GUFBm7<h3l<bE{%mQT5q><
zzR#1c(GF)g(KMU~H~!de#yFz*>zQj+%Ec`TWk=j;${M-xoAMyIW$<VQIvDMJXJPM6
z#80%vpAP8P&6D=0PYDBtW0xN>R}0!My%rHEj#9*@X1}|}#m_(8PoLaNA`B4B0LRF6
z^&VhJtboiulB{MCY%%c<JoxOJasD={x|yJ!6<$HVFvIr!vf{>Y{}3lO^i|K9h(gL|
zffMWX?(xd5-P56~Z;`zTYxPLw(@{Zo^Ih3Xi6iz3w~DQHMpW#60BD-m$yPG7Mk-;=
z>XM`F-8QQ+seQA`U}GL}ft~!yJ(fgjU5v(mnKM>E9K(QrO#uq%#DqbbUzONS{{#O3
z^)br0<nafIzUgt%mmnBjlAdrTw|@6bN1QyIe=w{h8+Fc?I>i@j8n^D+a%opx(01(d
zGachg_~x)MxB#B*e1^9epV~aJuz|>;zBF<Q?Vy}8J(+hmWX%*{Dv95jHz|m9G5`F{
zaj_D%04b;-z|vhgwF8=UJb9F(s_Q06C(Cv^aIHRA0M~;#^oH2&+UMte0jenz%Xuoo
zJ-gI3CiAvblhtu$aZ-Esw{K=cO(fPG?uDkj>Aplx?d?YoZT<XOS-^zrF=>6;UcBbQ
z_crybC)%mMH8HNNCNOE=I$g37m=eGY{_6w-AFtP~u*}tfm4+wOGf_9e&ENXykKTVv
z9x70ye);PX1sN8q|3EL>ae5-%KjUT}QuXCSvFK%J$i9no##^_L-z)lB`{(g4?R)#g
ztAxtw5A*cxWzRf<`@)~>w@d*7ES!g9kIFnb!B^4W6=c(qS^%Wu?>0%7WCq#_|D6y3
zBjxW#hFzi4ugDXjvC6qt(6Yt|Wo?<pn>@}rpV(B&>-zZubJdjS+4(4Cq34TlvEWX9
zvWK^goLBFVI9YpDGvQAoTwm2Y5d@rUirz27zr|6jdlG2-kAWkgIs?I(m0|^ElN)m6
zbA5XcNZ}A>fpgT=LE``V^TpOfQlR}Yg`uqqLwnvi%|Vp*-1KX-T{ef~WBEX<-<Mtv
zmAH}Sw@d5C1=jN?_gyb)DMbwP?<X<V3-1u_|89S_i{88U@S!t*qelZiVtw^qzIo4J
zn!Hx`HYjN$^yTek(}lKl3(>r|$6!0r9V@*b4mGap8`pRUC=OG(T^>z^JSA%|zw7(I
z8Ck1;&G6`7>lXdjy0sfU2mNFkr|~{GcdQ~b*q9Zx&6RfW`gpC+x|tNEvDwPCGFrb|
zh)@q-Mzu)?kKWsa!ug%VM?~1oskC07o9xV&=9%NZ4qc_)89#Xs1=Ye|UH3#K4-f##
z4LD407YFS2(Q#iSVa6@?Ia0d`1qJ|4|Bt(tjvW!mIPNSHd`Xsc)q7lgXE3D#3c^KB
zb*)@dCg08p_mLV@(sj%_CBiSbrCc!5UK<7|NdVi~=%lm>hrfIE&9&!cb5`-1524<@
z=BIT4h}X~zSc>p!ASRRa{cq#xjy3r04@D}UP-AEMDuYzcxJ@Q54>ca0Jkluo#0mB0
zPZ+?re(|;Aa*_&TX)n61c&Lc;d)w3`nEYr}0yn5p)FB3@Vca|6^5Hk({k_v?@FL!=
zYh|qoZG_$f=g<-A$FCR0T*j3JefZ*Ad&QRjy7N3oCtd6Bxd*Jpzkx05-{tv#7$N`9
z^Zy?-0OQmJ!bz=WlDic%$*D+iPNYCWJxm7JomD3QfKVDRQb)_kBmy<)PyGp$aBXDM
zU!7N1@fsWH;^J0}3kC=&VT8LtvmIdlOseqP-~&M=Dz#w6sE>B(=U*P9fw#Zpm-_n_
zPJkf$Z~gny8P-qu8^h-1{rk|U6Yw`ba6|cjd&KpxTxHpCz4S>XchJ^R2Q7;W;9zVj
zuc1QzzGq3MY7GSi(E&W+UA8j!(@92zBJGa<U!Gb@L-0gKG26_S%^9ZTnDGeMQ`qxh
zrLn~Jq78x#{J7h=>&Tw!e~O+!?cK3gM6XUdl+q{p2M9%<S)qdv{m3oE1((S0$L^~e
z7K!750zn<-o@u=P(fja{b*cDv6i>s=prqW#irqnq&_yy=*YP}y+vm3>B3mN;6YAY*
z!z(Kg0+AcA8oe~hNIB3~_C1mw`II$*og4Hm`pZPJEVEhdAMet&M-NbM^rqH>b`j0*
zIz2r^wa11}u6vF#kLfp*434HF1MXf0J}ROrfUs#>yQo_?h`zvse1v=ie5jE9Hm<ZI
zAY0=;XT&4sI$}{nUsKg|$uj`gpl1u6C7$gH@}Ss8Kd^}%M06q;%A(9vd)R?=5G!VA
zw?*b3S$f&HTw&Stbvb@9MM~@SmrAY<&p;&#DI6R7R|S$uUDQ5A4WuR||5(L1C3&c>
z3HP3%tq>Jlgd6R&81fKziPfJf8Vj$6_i0GMyoS#Q2z8&oA?PC8pj&1E{Ht0+AsRK-
zM2ny^p;V4dmud&%<^1@mASqMCC1Od!TFqKVZ0{chUR@h!Jm<||Mw~K0k4tYaw0hp!
zmEg!Rq4FX5SF9>NDNuw%?0ye(drcx8z4R@!s-%JEtD(81+l7wz`^DI<rSfM_2y1^k
zy{ws``ua~sg{m^;DOZC4ExOIIvb2#tDB>kujq!*Mp6+w^m{_n%vW!Z~aDz9^Jh$%7
zKlr`&E>4^j_EK^g@9kuRy^d8+J=xYt<epg;il4gX7aKfBx%%lMiaB6Q0}xcn5zon{
za5q!lnbUz-wTa$HB7W<*Hg6jsb^vwJ955ETyA|#)QY;dms>WnI`-d}C5-j_Ek=+iu
z%OfhQeW=z+i;Kkg9oAND%G&R(uCU%<f0YMtsmBaQ<{-4=R^E8#6*wzlPq^Uh7;lA{
z#qWGU%DpFXS!AaAohieqH>|%U2I#_AZxR&l?2+R|d?mWw#2X)mo34K9z!Oxm+OJjU
zyg4}x5a}NGjU@2fU+{OD!#Fp%yjTiypL7%Dw3Sk&Ba0-iXi^3v%^$!WRp>5LiG+FT
zhRyFy(#+BZ#SM^!x<gSkXAal|>Y3e8qEiNOm$ZadMRP@4u1#2tonnkWZ8JrKehg_O
zYh*N8h?$=*=CiQqx8eZ4XCJ3(-tNCJfeP&rv)L)W)*oRr&=6ST0K(x4bT?W;ojtaT
zHf_y2kVWCGKB8Vdq`5bl(=3I7(z|Rw5U6p3Cwc0LA|8iCJZd!Dq+cUcu2%s-)>)HQ
zNc-L6GUvDK`$zx50^Azs6Mb;PM=~ZO)*T?|18ul|%7J7WTzVJhs31M(Z#Yrcjp%iW
z5D!m_0>SJ66Sa^;pfg)$3S#mt4U<b--H>yDEq~}kqty(lt#fi~WUOytJ0m%pOTF)<
zzrA+1m*Gt0c$_<dg+Se~f33obY*B5JJB-cZpFDNLyBYFVoe<T$2L}V_#{P_>xUp4o
zhnO#=8ef$tFrLYS<%5w;x9lKY2|3Ox5MQ3Rtv9e~DN=AEvfUw^CkrV!Wwe1;Q~Q28
zFGA*P&>=w{OHk_wbTJLdYp;rs`@s`$30Ao7>qKJ0WY2)}`s{q)a@sBG*$Aj+84o9u
z73k-Jq94gQePM^N>4bMF+*3C)VFm8}9JMHt22uGop{zsqLV%VRV&Fiv9@Zcp@m^H9
z83h+zcyvpZ*o`icv>U$^QaN_N%G}*fCIEFo^d~kc$^Z~`#hEJlq5Z}xST)75n2$u$
zmfK8Yk|Qxv!L&v;b@B-{NQX|VNEi9#HA#);2!6b8yT895HSQI0h!Pfgf#-3Uh6hXf
z@dI5bSAP8?;MkX)tauHl%JQKulGl~%ha6f#h}9daOzt{$jnvj)@ee%hrfTtH#Q*^k
za0(;>150#>v2Mv-UPHIRtWSGtEI8n5DUY8CK;1I~IJ4HIs3AJ(lWor+MkUHIVm$n7
zOOFZj&jGFx0nF*H!BYWoW^cwuJzqgzb^Mvb%9m6W`R^*~FgiyVCCQ@gA3I=l<q=Px
zN;D-7;-4Ig4deW<^HWq>IxL)tAbq+|!|X|NeJeXgeN+bu@%aSMhFXW~^9}vi*P2l^
z+m%!hk9$|rZTy;-FTClhncb$5m1JM@%>Xpl=g>;Ax=yy!Pi|gUrhPrdHyZaWW4uPj
z`AMc|UUofA^qh~e;IIp@J^Bf^ap!j4m5FC1d8++Xwl7ye7q%jwI?xV}SQ{ne{ozLL
zLF`ELrj|ofbS|2t8}ocrxc$E#_i>?XjwM6!u*HIQrH^osGbjg!f;b3IbCGBnjioW_
z!eW6ut|29<c&3@X->OqT58AQ6wOHNqZmVfIrS)Gk&5Yi=?`8I|Q7AzTk$5LRkju$c
zFP+3}`Srl-V<vUW+3l%hSfgc1eq{d5J7#UvzPI3tKcr&h+Jh5UwJvG1MuDTwzNHqR
z_949a!Ue)zQC_T{&jI+bs)_>=$z#o3ei&_?RtMARJ)C+B1~fJ4MyK>pvDnDquIvrj
zvP$@*`3=QF2N0H`4}qRSU-l8jO&N?-Hc{asDMRJJ_ctj`LAv2$-X9;3mYN|pfE+*A
zM+v!UaAl&$e4we}b5)2*kpnuXr|8xL#3IjciwP<6_l65P0hcpK^oppvURl(d9SY;J
z8Vdz#UjWn?0S@I$w*4vQAY)qblL|-K?<wf=EKjCa9a1>e6g5QP19%gh=k4yeiRKc5
z7)>Xdmcv_Ren8J%3_r7-*GJ3s^G2Bf_pfOW?qd<FCe<{l@mqLT%Jn(BilFhiR00^I
zcWy8tb7r^?;u?Ip<{vj74V>)-^Nx%SGDnw1%;<e|EA!ZU6i-VI6fil%@$IMewbOG&
zI>~-QzTYCm(}tYcV4~#EdR<|unmOo!QmL`fq%E1*dT!ZSM2vE^6Qq7T&2)}Bl9wvG
zv}(LCKXOks<k@zGo)8zDX0{>7Lu*8Kfi~!t=+HMeXB-n#8$d`yVOoP>4R?<r0UPl<
z;j>0T$6_3bel;vXsqLR5B&RAy0K{#f-`(W1Z8&*VG?|;LoDcC2p?KEy{7jDFc=wf~
z1SVs!EyYMAK3J!=eDQZ4{fSV~)7<BAa{PF98jnYFEfhWqjh3dc{9K66rnl7%3D7Mk
zP>nWGZ%%@zN>nrQi?8FKXNHJDA%zr?1d(Hngx9%K4^<byp~oe1z%lcM^*ZtDaEAFV
zA4!5$W;RF#vg>%8;j1ukV~yd#kye;QCoS}vm<)Y8vTRyosHFZnadNDcv`W4Od!VEV
z#dooLU3au?fI8HqcrXtZH#D#ipZQ`f^&V0))hqTn8PXd}(5o4!x833d70eq;fh`|R
z>YmFtZaxcs+IGoOHOdA`Jzu!$pRv3`3B={EUe||uyUwsa#dI(|n_TPk`_?5`^KN}9
zA6ziI78YerC&4~!>z>91+(*(jRax^A#M!NJ5*Y_VQ6#&!Jr|Pz>awxzdeg<{DqZi-
z2Iw0*TBgx!WIa$kg%30ClN<GGFdNVtE+)9Pr=C~WvR1tcC4R~6v-``Gkfa$B*)De#
z*ng3^1MO=0%y{<=195acNsq=P*xJb8rOz>H(g`rJ;X?Jp555DPfz=R}Y9;IF;GIp6
zAD_RRkZM1hsH=E3-@Mcw+=SU3jmxJ{G7)B1%s+n&mOx~%6Qv^6Qr$HJhM78Mz#`Kq
zSDc>%1G=}&QvnjIx?FR1OLda^c73`P8U8Ihl^)JIK^{V&EFfM~cYK3=pB#DOBrCaq
z0(Qgbf#qbyEuvFh-_jm+pk<F7&SX=7$$$?-w72l#WVepUxJ)PdG?1*vF+{!UvP(7f
ze5R*wH048IO%D@7v%g3GmeRL@6Zu>nA+*)$Snq|;&g#&W`YrFdffLpYB4a3%nY8z{
zg7hs@F`ab&;<G;c=c9U9_<}i+m<S&Xywv+3q5E~;Q0I(v{*xRY1`~fDd<W4^(kQRR
zYiu^r3djgL#+yB~qUUIdsL}wh1v*=x3U^X@UC}$5m)nGjIEzJM*IT$(svpMC=jbn*
z34OY~@^IJ&<tj*ZCuVbd=r|ZX7)c;9BDS4a576ZcW@&5|qm|*ZuPniO$G?!<2^O%(
zhwZNj)~t@7<}@KT^u_eiTR^7%*G%7g^n3p}R`&dp5~Xy0#x+9Uk)Ff4`RE$@({Ybx
z2WNZw`9sxwNGDzYuXo2>tInh3+)s@Mc`dTAX+-7%ryM>cqdA|wNxBBrK)Mis2dhmn
zdSLFDLO$ah7z4kVaJGQTwgwWCH;uop%W}`<FL#uFS24GfX|d7sYsZiX^^ux$DArHo
zR;y{qH-mVaxce_zuNO`+{B65(`f5RDi#BeKux`}r*XesG7XMcL>6R(W?O&}9(&)gr
z0>o2y$lj0%n?-t__?58-(A&q3)K!yodBee><m?Nkl|AkeNNn&6cmJ4LmLn7LJn}F_
zzk&DTN#q_Tmovx^JNiMnM;}unepSuhTB>}s{R|tkD}8IKF4|kn(>m(ct1ci~Nt2|d
z!T@xYXvk{JW=>q$GAD^L6+oGEJ|CZJ(j{aP_oKRBx@y6#AKoV1bCrE_2StYIHGJiw
zma=lP=@BzbSA%#iwl8A}5co<h99=mw?5u`Lwz|vXTG<MNb?-=A8M_3Fnvia+Q$6=6
z54SX;j9BUYdgBjsUqTmDUu-e$eD0_kVdr0NSe@e_)yGtY5(Mpc5;k|+bkffOlJ*@~
zAyFU&(v77@RLA(be<X*P=OR5oMo`<i!l6s|g^LPQU>UZ=0zkDPIHR}F21?nycAt=<
z&+7rQX^uOkOJ0>EHburi6zWE=Vs9JdMh-9FonjxTDEQF&g#SL{Lt`+9Qt_zdBEbYm
zv9{xzenIp;bIO;aGiE+2X<)Im(wJ2C{Ctef9_zLVx!I05e}T>i38)qoI{tL=*ZgGY
zTB&7hjLe-oXJ<L2<xK5<&F6X-xHw8eMMn#&`>m-@o<o;GJv8%60VHtz>}?SeU~zzg
zydgb?moC2p6Y8CB6e8Z|0A%nfTS3FfkQQbVD@MCwe)7vD$nP)Z+RY`omnh?>e&QD>
zevxwDuTN06Ossk)Y5O;@0+Nj6@*r6`ZE`-Kf@|y%YgQyC>|iQjqc_50V>{o<>f7X^
z?~vNFJw)B+{17wj3SlLu<t%<;M3V<V&Vi;73J=YPRVaU_mDg`4G(%3oX^d?F?23sx
zx3=G@==QQ8!3a5YX_n`3{9NBFnI5*dW+B_VRfz{44rH6>l%ql0liLLKs$xOA!b>|p
z2JyA`PUPX34>~54zlArwF(41(E}7neCGRYxvLn8@;n~iM)|#cphWO?eaUv%<(qo|I
zFsci7)8Kt}vZ7yM`C0<~R5eYNVmD+nDoL@Mvbs@yr}e%=-&_sE7l!8u(Ez_Kt;yiq
zp9~ru%e5ij_;oM6Qb|B{#LExnq>IzMoqF<q%4n1s%_Zu5X?nUJM%D<izlHoR|IRVH
zG;U%!Yc!|qoRuWH`GrpGgI0Ca4`zeLe@;|f4NwgnK6)v9Y3U?6`|#EItaDazvFj@1
zqB=CRG8ntDeT9siRy;SSY|p``p4>~jY=>2Jbk^u$Bv0s9Hr-8t5o*AliJRDK^ahLR
z&e2e`hF%|2;)ZG5aZlCF_Ee;NlqAT9#H4vg(w?J%s`2@zF4h7YH?;MN?_CwV3?QLU
z9gepED5WhrXHTyc3I<sKY^1_Iz1QmPbS^2Oh|*gp{mwcGb4tz-nuh^~0!~VGD?yn^
z@DqG-JN28M`Vh{2zpd=V+YE(vubgK-<Uw1{$OgSEJVNe%jI!|<x3DH&d^{KMz-z=&
zST_SG_N+Y;xe5imw4p-|UH!ObcwnaV7?G_mo<kp2*wY?7!>0m@APOt6m=BZ@zu%#N
zqgv(Ob4J@inV=gtILir3i34t2qXJH~-j5aSG#6)k*fEYfAsK9YCV<$9Kbr{dF#ap$
zp`YN;)@qxzET9<b&n@y`?m0JMpDTP>m&E~JoO<w_#c}G9rs8IL^NX9-5l-^M6{Al3
z+v^6!$0uRiXr|Xv(aHsvcBYe1+5FoskB?*6ekbv|zL&UiXFCXEalaCFIUZWpZ3C+a
zm|abIfC`9$GnSimVJH?Ps{?Zoj|qC`Io1?C-A6SY2qo0dp6ELC{juuv*j8r#J_VYM
z*YR&K+Q;j@<=PXQ$19`MFITQ4(1z72!b*!j{IykWOFczcD~N<l=}DAP@3$B<eC_`t
zH&nzu)hQ)sgoz`jE}f7opp~NDZ(B%<s=?k7JHlF61O<!p>`4~AD|pPIQin@}vs7&J
zR0kHzOpxN3(!Ca_9?~Ho!aO$~DHIn08*N2%s<s}#LgoNA7o;J8-9Q&Z;k^&`iy6?Q
z@p!Mp#QHt1XQO&HsdNnz4VtcN8R){)cm36t@SL0Wm1FaEquT^|CR(_dl>0=n%yd0)
z!_h+l^H3Fo25f*LDV<Qe8%LS=sFgWBRi+3<*Cr<DeZjfh=*bN*Djp|kroH`@lQnwb
z)Sy#>RI|sMQ`q)Mue`oVKcCxk`)a|!k_<0)M0a$<=Vm!1{o&uktoraqn%`xat&9fL
znWI52t%#KSa@_QJ#4+JplG`!d&{6|n16H_skT+_(B&4|+#c4`tQ>2pp_)s-H@9xz}
zRsrHth_zl&^jYUF(b)1uQ`o`gZ(%!d!wXlTq9<d|uxpxD7fWRt#c#K%n9rSvi+PXY
z_U%STcwm~I8U{fN*}KzRETw;hO*&WiF$2s1*SP{#j4Y1TVY5pR9pC5CI<Xk`Jo^AU
z$9#L^cW9o4(<$z5J!cbJfAmSE$CYNn#~6SJD3?sUDM&fvbg9QDOvIy}X#DA^VB?=Q
zCqJwzi?g~OOyDNLs_F55<zuk*a$*eq*};iQ(r~eI-l4Zr&f2j>Uq?&VdFKuBpQPO`
z3od*XMLaaeoPDNTW<(sieY(1^P-cQa>rPxgL24Xxax#V~1p1qY#l2z3WCaz+hTTDi
zhR#UqUY3lj^o6*2#yIIG2gXctLY<S=PUG)(8mu+3Ha0i>)SX9-Gyqhig4myp#rz_X
zVZ=TUUDv>=^)Y|&?Uu6<GPF@b3Yo0{IrZ4&Op`3{QR+VG<e1ngr-(^#Kg5AE0lY^T
z7VzpyfNrexN+;@H7)OT5Aq7Idbxr8v>|?6q)o^*xd%^}vdbPk6A)qY}KoD%Keri=x
zzZUjg0Bzd`f{$vURn)F1bf)M(q?5!|-PMvVJg2!bW0xOW+W{H41w&W){cClM`9%lz
zTox4iYSIHnuIfMJowL3E77vc;fgvICOt&|@>dGJodwLk3BtrMm(9mK2Eq^5?^^~Cf
zXI4)8tMA$llO^s|S#v#;#xX5<StW}8MJOmJw`~O4r1L}92<Tync!r3l6kIJKON*MW
z3;^>-au&N^oEoCAV+FAKpTLOk-5?JO#euCFUykC1uOvtHHV8xg@nUeDmtWmlZ{?nj
zGT9>L{5R$q%NE@>G}vovV>^jx{9Ep-y~L(Du?{SGA60R{qEWZ>UMO_ph66Ti2Xpw8
z^GBjHSf!@ROXYL@j%|}=hCOo{d8LP_j|NR1kM1P^+e6gGPs&M&VS?1VUzKe{5}?AB
z+ZLp!+MlO%BhyJ<3L;{1T<sE4_aK4d=WK|4P=(q2+b>(Ho8EU7!PX0P85_UQ+h%iN
zzh#{WGuEVR8Y8z6r%MqGxJ%!nHaahU63Zc(OPWz1?Snmin9>^Ddz5rqL<ydjjhRX|
zHo1F+9s22R+Pop8fzP*_KN(D|kNP7UF8uLQ4o10aJ|AXJOS89U(!^WEd6PiwJ8ID*
z4JU8!)b?~Wv7|hsLs!IK>5X~-T^R{tI71zw5R9U((sz*RiM5@gmZSNm0w}NffX#mV
zfW+aL@K(A<7vOrQLx0*YmacsnMuscL4A_(^tA=>sH@u69<a#wHX+R^St!P$`ba{X>
zSGx0F<_#KdC#9vRfNg`3l={bH_&Zmxv(>IhvCm_?wri-uNDfFBx?a@NiSm-O)tM_A
zqgf^Jp)cwxEP)%OL#Ha=e-LdOU$z_;hFffS(PLnriy>F!vR%!NB)AJSZ7P8o;{w;_
zGxVvY1X-)nH-A9Hho&`<H1POJ_|VxiHVU7caPOVhqA^7+Mp6G5zrAZb%8gA3C{!H0
zG3`d(`|mn^ojZ50KBy?}t@?`Fx3dNQk85&QJ;PC;SOB;uX$eq&s!KTm_r0nmZY8pa
z$3B!^$;EznQ}LVR^9IEGcDIu=aMe0TVQFArr$MrJ4!tM-Jj2lj+tL^C*THpv;6c>&
z_D>j}$%WSoAm%4X?RQ?~4UL&!_XmrT^nCqE7MFhhKnM!DfNoruxsD<WOS8~Z1Sh<`
zOLs~M#A(hsk7~K)<%+qoL-Y%u(I24_Prfbl#gM{OqP~mI-vfNV^|#+}t(bwGTYHpM
z$$Q_V&0#8_2G1Bn*z+sC<Pm^gcDC*SKh4qngw=ixb4ZDQ|7B~83%2d~X#V&oH^Wyc
zc8`^IhAKqe+aBye5wn-EPiXx-UzHeeMBQ$!h&Wnl>H|#>Wt*v8vEi=yjw+Wn`;q&q
zIH-KCP=7alnjsnA3?m8>6<<5aM=Qi7#yTTVZyH^!R9=ID-XH}x`<0@9jd+lH=r)I|
ztcIQo;+*@oe?C@nsV2E^8+-mo1+<}^>eV|}+S?iW^zbSB=U})1E`O+XjceTN@FL0T
z1!=tYW2wO=<WHb%)zC$KB(rQztJ3ME(0qVzos<-NZmG6tVF{a;Wk2=%e8^(A5-19J
zCq=QG`lQxYQVX_n577U%e&2qcwrkf|Ap#cj$F6S$O=W1?AeuIS6XP1+b{gjOHwM%%
z6#w`EN}HZmE=K3^fw{(;{7}RRWUH+zdFS8>_be*q7aA`h?)Tjb3x2yq5jV*<NGoF`
zGiV*QS<oAgdXv{$PPj$FG{l~}fmoOiy1GAv{?O;FJ}T^Rol8okyy6{Uu=B_HYmW3h
zfWqQi>zk{A;lt-=pA-hiBNB&x(mMEsfMheyUpKHq;nRCe6J%x}6=cs&6JloLd@3rP
z1fXSva*cDD9c78Q%>3x0U12VNF0fl38;439pRlkKq=gGP-(M~}^h}%d1olCu2yp}n
z8$}JYE35s>=~IXDNioC>okUvb4n0=8@7I;fB4q&yUTx44sdfT*eC`X6>??n%Zkx|_
zl_Y;42w@!iRzl~DB!52JB?bDptcin3k19nl8c*@&0uRk=m2MITc5Y^6ZJZ!2q*c;6
zCvv%6?698U^4G84a}zI@WsNF;Lj6%v>HI*eOaAKGNGiQ?J0h>`Klwj%n@_~C>K}Tp
zRAT{(rZB8c-^w#71+)+Xj=+n4c7NWZv|~c~c02QlzJR#SXfjZBbh=~maDyx6Sz8?h
zXnjnqOd3{VxHt7KI7zAPJe-n2CTsq7J<UVy>2c3BRu98V;esr*zo#Nx!W=|n-r7xl
zFn^SPAew4N8azzhR#C-G4D*-XC@NP>wFDX)DePDWFHc$jVV>q8sUNPhzDxtol={g*
z^zS~Z_ip0fBc2oE9z0zT|DepC!J7r_hq-O?%zCTh^6r*1mLC=-eryf=6Qor}*SAal
zK#<=M#&oZ*4LQ*YMJUThl7N?CiV?*F4^z@_u7J?b9R!zhBAF0&P&fqtvhuES;%B~q
zYiR<w6@+<mMdR9YadGu0=bN?`I(;>f!}8PIRPTE$2^XPk?LEmGE2x98Ckd!;)Oyv3
zrua;VlslxJGY|Z_jnwGQ{K<uqwQZ1)$z+uii^<)`m`xzS1P7Cg2iZ*T;0YI|D(>6j
zt4?2UBL(6fs;*~+v4Zvx)|Wl!dX4fpQuG1l=Wl>btj|{$0L<{$7PxFf7bohb&Uj|y
z+o^ZnJZDyK0S{G!%WuGEig!=c5Nz-@g-mJo?6oExm@cquVxe}VjC2w)%TRposZTBy
zUepKNsOZBd?zj1Qbyg<1{=?eAV|(RBjZm=|!ZMx7<>jCT=|+={Yaz)fIBSC<!TNm7
zY{Y)C=l$8tj_k>1nLylFz}zt<gF@XHzaseAjPuR%H`e7(FV?Rt>V%J~dx30)-%qfg
z-gwSLKLdz)8gyLe(Q`*iG@0|udf$rbnbvs#x|8eVDB30Ab;`yg>2a{p$c~k`u&aYS
z8!*B)sw!htQ~GVTCD*Cv9|HZqE9L7LLpX_h)bNEl?3DO$l^H3$iM^0_&8~MQyma|$
zlccW-PNUAj;@^Hh!<J34H9;(NLY-w_O&tN&j;3WRZqBF;ast(lV}<9_ojF%9UaS~~
z!C|cHRgxqZ0gQyDobx#fRGF_UT);+Cm`pVAEx;<vwlC*%b6<u;+>&4h6oNt;6Fpy^
z8psyBd66N3FuZx>VeE<p`daCBK3EY8e;@RGxM-3Fg3|dB#Szs^en7ac`SOI{VSf^!
z<B7&|rkW)^kFnMBQI8poi_jQt)mZF7%G9P$k^;Ruw*x>wvx`*pV+C&gsC?6?`Sehv
zFl;=nj3E_8vIq_<aIqULLZj|q$s(QeIhI4sf#XbRMWB~K5R<0+wgrrEj8mzbBEEcY
z;Mj9$d&=0Tv0tbW(QsK+|Hqc~w^Cy6I_}{S5sY<PcO<(in5?~=U0`fq#-{M&>6ZAI
zJzE7J>>T=n+45N@$ply>|CBW>{5RqV0r0<;>hMe!nteK8W5<^!8dkfEz;C+(&V6CC
z)%scYjwN%Oyk=wj=1`YAEL^YCM9-KZ3gj1An~=P9y+4|>y!$9Z{=0LZTlc8m0`)LX
zCm|n_P6A-$GnRTLD-t#bh&l?ZiknY-8d>UNVI*}g!|vb$`}{*jse#&y-Te@xz$oin
za^K77`Gn#5JBSHJa1MXBS@xf+m^X8sy^mkLNi2#03J}EHGT&MHYF4BY>%Ya(#KQ-$
z^<!0xWgn<82j%r_oi;b~Ca5P&IM*l#0(6X_8|KnRx3a<GM2ruQJvD`Sc3Z3C^4Vpe
zE1TgJ`kU^669JR*U#hhU+uK1L>5rRZZdMGehByp(oCUg4V6S}+l~zHj&_{8sl!+4q
ztihLrw7=No8Q$!QiFSFo8WVr{BLU?<KF9D!me;~kD@`q0iE)9cGCL9ZT4u=8+PdXA
zff=TNY!XXgSGF~rzndS#m~-0mSZXcu+DrqO4*Ts#!OLTK{M0IHrsCYX&nA=ctL!xp
z&dtH>YgE<_X~4yRPC|mQIp1l-i_XroWegFJ28#W`tickKBXeEiLr0k<(N*?CyOF>o
z7|GZzvlEdjPu_&fLQrti_mpkY{%2=<H-y&QoUUs|UUXb9L8jcja=*8bSqXm7lu3P}
z_H1(YwcXs}Xgp16Y6KK7o;<T{1%c01guiC3w41PF#b)&VG6fYrJU@Bpz_{nb<4SgF
z>Of=!E3_85n;EIU@a*R=jfV<1Jat7TJN^iL9*24(BYu93KtnZ1-L+gVd{c`Ne|(r&
z1ziDJ9sgdl$%=tc+V(b9?}E}(+;rBc=WTorqq;K1EaK*AvEGR6Qd%V`jzK3pKS1;&
zRL)GitF;-ZM4VDO`^&VHAUOcpA9(0r?j1Kh@44yi-~SPM`MO$J34ErGcGVy+$+JPI
zMB+7drob<XbQ`#}J^LN?^G`(Fz+GRf19UGPnxUYTT17(ds(89_Qnwri8F=&P!xcy5
zAD8^GZsYZ{rOPx7{}Lov9A~!aeeA7<gw9_NdfD!ht2GHdy@%wZsz{&o{5uYg$-#kG
zu4KNMKS2-u2r8-E{4Hu7tOxyl*a4&df#gIuis}G<e{+Mm(wvTtV#t*|;FcRR{=I-m
z`lQGJoGKx_axZF;)RnCoq+KQt`Z`$cjFCuH(VQfjEuWQ*u2cxpqo+1HIeF)AXl-0$
z@s2^fh=fhUz%7+z-h~u2B-g_MP<`O-U(Sg2vW37Ts02gieKBmBiSMFYLliGi!b?mt
z^0KebnHj}st3k&CPNB+LKdvi*5865(fnb;hz-5am6dK>#9?1XME{*~myhol9`4l%T
zzdx`7w^OCL6Er@r{JvlGJ)8G4bA-VNzSw*o&R#M2A1uJpCgm&s_Iho@jLBeNGzcKM
z?*5Ba{yatetd`Bx9895bR<xGG5fvtn6XzzruAztSMgW$LdFG<qyav*qe%scz$u367
z)G?gLwInG3RHCSU=8hJ-v%p(xR;1-fx;7+T7)HA3JcCl*$PrC30hqDOBikOpP4;EQ
z@1sl`FQr#6g89boW&+5hdTJmY7hyMDyLxo-V1+4yvAwk0J1v{33Ng(G15F)JdK*vv
z3QkZi%F4EX9J2@7<1^b~$d?oB%`pwHv(e3n-M5Rf=Quhe<b3-M2C+C3R%FDz3@WT{
z!J~4&nhTxUbjHA~8gG=?Df86vB}B~da~KJjfvvwbG3thwWO1Y68g^nP>F3k7lN++O
z5Mh^ZA5JP2QF2pEds}p_w75raWgR}0`q8AF_}Hw*jD(;~Qtc*u$aT8`H%FjGd*p8Y
zV_hZ^(6hz9<<9qKn`3=re{SMLR`)8h&A4cVI8&mnJ+<uwF=P?N=Us#*#L8%Cu>tR2
zWU>4E`S|m#L+8i~g(D};6-P)J!PSS%ht?MegasS#VE_5-7cx=#j=oyBnoi_od}1S$
zWO|bXG%d{WEzg5>R2`G5erKF%U<AGYrGZiX?9|PRZ7)~>DC|CG_aULpHNPRlz<8rc
zw44sgVs-rp<U|w5ugmTTu<~sigze5Ksw1Pennfy6*R9pGoiEES;$Q!fp{?5EI{)4Y
z@s71N)SNVj)!gZ7ejlui>)376f{_AlnhuNGFaLIui2?L^*#?00*`w32)HJx(o_4N=
z*ty9<lsgp_BjqP@6b-L{*8euKLiRLtfPs7TiG7uD{{<J0^ccwzNsn*Afu&d8XW|`x
zhyN5{U^+}7g4Uu%5LtV7YsSq<S<vYcw@N8;7jQ_d0;&~JZTfQi)j}Iy#*O#tqcA(@
zleoDEp~Pg<#MT1H8IGF|hU!QfIHwHV7+|6P7g<lDVI*0;;i(_qL;Da|4uz{MK@eSi
zgf|@&L^CaYqgt?)k6)^0qI95JxWDgHqELR|ScAXVJkL?ol*q!R!G7LB^3W?fCZTjK
zk#VTBkT;hoa^Xz~-P*uIlmTkItl}1&LY6*=VIr-v>}u{>kVG2KS;2C%F^f*|$J6@V
z*~WveHHq)f=<BL&{{*_OTX%pD@-&n$Em_Oy`r~GZf8JLn#U$)DJO4>|i&o*<1doX9
z$n*Cqxw<@!LdI`N+$2cdl;(2!F;m0qlpNT{t3xnN=ZM9JV{0|oKLu0Px>=KTF$0D`
z@{J?!3|+P2&zx5FjPLrcF5EGacg<u#xIn|!OFsb?Zo_9l<IBsu*jtwYT>_)D9x`<+
zlP#<9o|y^YH2iJfUO0OUOk#v&nc|F02etW#q6`eI4yGSflnm+BAF3$4yLiN}Jvf9M
z>CY($U%s>8E>37Q5WMQ<!x`DDf3_2InmfuO5@XPeB*eX$igkx-gH(1N_wu#?3>8rh
zGBF7Y>XV(pB~g2A)Z$<5n%LS)qxNQ=0E~bt+~*K?kY8Nnr=__3t#Nhbw)U|-Lkh>$
zPsRupVbA63vz)k|7bIgr7&n8`tC*jyXwt~FMy}k4{;KO|{+{yQK$b+wwGxdb%G`zY
zP~ssLX7i1VIPuR@uivJ30I>9cxy*L1y2k?#?z9`tHyz;1$LI;{6a9V3e^}%4K~jDI
z(C(G`?(ox1K@6p|P@N2FZi~;5WdLr-KZPd*1_T`~!8Xl@mNGZ%0|iQ+L&Z<w6zXTk
z?nij)H~SnqGkih01CvIv8@XLvRGDTFe~2<g*b5!kgWC;CN2S1lUh1|c>LagfKx_o2
zc6uHqY9>(p5J<ouu9=Ay3`R_!+Pq_=+Fn~lm1Y&i0MwzezYHlw<ccWp3_o6}g)}>l
zX8bLR{O#+s5PhDni;8Y;&+yK5Jht(lQX@?yPzyZ8GNr=RN6uAJ-$3%*L9t*3)gF3T
z8jSjq_gb7NPxkq^V37d!h<R-LqkRH$?B3O@!hICs(YO8Qk_5Ob@sd!ab0lEjBbGAi
z?!Z`$_}n<gc+bT{`)BYE#lO_&i1ND8Pnv~|fB=}$EMOy*z7B!V@o0f~n$Po^^DB^7
zhX?)S&d@!`RVkHc0i`l|fpp5+#RskeGjB0_v)^Zl&CbL?DPZI@&ug?``$aS5!Rx-{
zXddQs)>+C^`X8pZL^dBEB0v6v=TulDe0~?TwxrYBq;N#2e}z0{-%SHRO5Y;ez=QWP
zIk><;$q#&?-gC2^>tOfT_Qg8$7$1q4n>4@E#fKW(JnJNaqceWbJ(#+YywQol$FO2W
z>SK4G<NFVp?H`nrZxZEdQ9`!K!srhKLg(PR4YIv_Xz{Yq(0?pau`=CmIY2;<1f5Gb
zG{fz%at_En4XkP3=pfMclV7Wg;ob1?zz{XM?-)p?MyJk;E0pJ5ZA?p6IV~aAHO3}3
z?Y|hIt5Th%Ie4m~{*{Ast~G0eeKgvN9$P&yrfeaXQ$*TB9u}GV*h-pemPD5Ml=p|s
zecK%6R0njqW#F&4Z38pc_u~J~4r)ge`K~IM+?(i?`Ve|NjtoQ*-|_M~YES`000m!i
zNY(vhVG#)H5`85m(job#`Czz502dyUed6`FVB-MeF6xL^^S*bs=gYCu^hwYy(JimY
zeK7PB+wk<eGu#E1U#qjMUqk%;X3bY<B#nB}#HF+@6v!Y33=p2I3!;B!B^Qc<6gg*0
zfLvO#OhYIyaE9E%!Q;v{#~<1@@S9ld?Vz=}Q02FI7u`3YVp&DT0-}}O!F+}s1B6D2
ztCO~Hp>gk5vNh9X29zXMsRQmOA_xgC@985ahJ-&aRjNFeUk=3I!XT&g$V)b=j1^!)
z+rELW3mvKg6qN>&8??VIsU3igl20q6<h~6uun1PdGOSafA9>2|k4=tSS`aCIjX6Ai
z^N0S&ID`NisrKZBK}90*d^%!Oq3y_ke{~2C$~Un?cnTI8gCBN(?iA9r>`_kNjl#>>
zt6-D+-nBh@^MfiL)$NbyzGVltUMkOucFe{`Q!;_-#XOgbELwa54@-Fs@Ba6IZ0?LJ
z<PqRL$VX_tR8D@k7!Q+qIZoMHox@N~F%s(=i^Zk(z8|i7Y31$-8sJ`xU^ZLsfqcNg
zwOz3|m2?_9VncWMUDJ3xoML_gt^1>ARh*?0wgUdz*TR`ROtAofMS$a^T(U|b;kVjn
zxc(2;+A`%_k}ehtk~8<z>_0WZ82h_dZ3Cw(*8cvBJgM0Mo1@SAlH5L`H~>?Y+<c>*
z{J;=Ri=OC9M$Yw9INgwV#>s0&vJF1#OfCAQ=^p<uJQRj{>*x-^dH5;T;O<F;aKU<Y
z%JwdRsjWjh`85(KmjKKFbd3&`P7+w4s|-Ln{QEbD<nDd$jQcybzvlZ?m&%k}5+@%4
zTST7fl(4-N*k$vyaBqJ+!xUx|5}B%XTbJv)<!wj;SKPJAIPw?79>3U~yEZG$2boC=
z6~u-gX-`ZrGn83ePrY}$gj*+#`W6`(nJyP^<e#E~4AbI^LGRk_mYKPV?T+L5i;IZv
z>8*d-nu?pAja?~i38GfOXg81JZH;@xUQUjMk<4fs{jbVt%?g467ts#Ro!txrh0L9u
zB;N>qANnl}5ia{Hj96jp&>h`qF4!<Vp)4~cX-PEDq-$>0KDd&>!sO?PaJoD3NvOxc
zg{bgMfUNFnLze%bBXo}wsv?$VHjn*clh&#>x~oTDrJq26=^G93*WR~6%-nF(r}DFk
zG@j_}`6ju2ehH6Mp$*eCC6S+1+e@OXxY%CG=9a2TB}0^1Y$gZPlne$oh<MpAfQ`XD
z6pNM#BnikI?~)U@URX_%oM=F3VN;5}Qqc8Ns<{kMMninI5W4M0SXNx-i&a!6=1Jqi
zQxTBC-t33j$5aS<AJ)iX2f6lQG>k90N?OY?xJ>}O2Z7%@y&rJOICEd9=0w8d=~8kX
zo>0sUllv?5YRjE$lC!O{AB&>c;zsdQmv^74EcDhWBK7HhwJB0=Cog(^3b0O?Y`#zU
zx?6EV{trT-K}0f?+%r{&>a|ut<3}Tp{IHKN-iGY|QSr}g`WlQw^^R@Hp=Ys?>J8(C
zG%rv_zY!WkTKjf-CGUw5N=YE<S4)NO3;pU?PbE$tS$HCj(84v+Ms(9ML9|w{6{K{?
z=*4)gml;1fZC_8DFFh4&aq!v>V7Jj1_~=PO5CPh7lc8T?BdgK`Y#>){YIYTPkrx3p
zH1>4Y-#98Mey@S^!7|9_z+a1(Zl5N;Gy>)Ew8@y@k3YTshot)>KSMVCnUd*%KLiCS
zBKq`^D%Ss}6J!s%QgG424dYKBq`zYR=U`6#i~_`|7El~WQPqw169cOZI1ngi^)5KJ
zHxwZG*Hx~yljGHNTVNWSF{>zK9H-$P@wHcc6Y1xx(oYnai_zyI-&F|@g_CDJJeVA-
z1_dq-h366(*?hbjurMS)!y&2k4Xfydm0PfJqogHU%NLJ#)*RdHd>ZOM@4de7rGT@t
z7r)L(Lr$SMMzox-K~9t=|K?PzKY<Bk%Ve;dt*^+wa;Ua^p82d6Z8)2-cr>f&(SEpN
zkG<{jVx1gk>K8#ln}!#>(w{FhIKDYs&NW(;S-3z1_&j8an*7A$^ujH(*4Vdth7egR
z#r=xerC9q6+P^DOrH)^i^tabNtMChXjm=E&H-R0I6$$4lcaxSi>57GZhL-1B5#wov
zv|4%En}}I|$x#LcK8RJ*&{GFymyct3sP%ls%378;;N);U1U2N#v8av~v7ii#8ChB8
zt@)`&k<H?5dC!wRjRhD^3wIHL5UpWp(zfV?mG=<%dA~#|2+QTcwAP4@W)P<AN?#8D
z>K7i6=8~3Y@|k>Y9Kz)ITc(cB^reBuJ!&3>N8?@zd-fy!^4K?p=wZ!>@#awa+$2LM
zj>L5?_QYCl99jXrQuoP-n-BO{*$pv}^%w>2!#~~|%fGfiHEFf;VFMvKPh2CZ`(eV#
z1N?nVP)-m*$j3e1y&?{oyNa4j2)ZK3+ZyE1w=v_7&<wM@=>6IIgJ04F{9mBj7H4qO
zU2Qg?Vy<uI1&CxJ`bV86NEU{BV4UxE)WJiKg`~JY0c>|ApM^^FZxCKp5Q0UQ1S$1C
z2IB%DjtwR3bwLFN{8BJhALDyFbX9`DPK5M;SxZvxY$V`qf5V4PZOMGM&;<H?BG48q
z8)XuZLJ=We<}`qGr9a@&<n0E`70Ic|pD_X<=!OU_AfZ1XpWm$P;Dc=zWwy8#N=2LF
zY_yhlbcu1?Vt7n!w5VydDw3Z*9r8<6I56i0O2=q9ja?N~B*srbwH4&;QxAW-kFxXh
zwXqQWK{YE!cE~l}Cr?{}cx7A^r*j2-j-L9XSxEFTSS)KTxod|!(O6noa4?i_d<PUk
zUDF#tT1U+x3)0t}O)`Y^jmO4)Rt0{;SfFuyI^*QCmEo4uE@B~HVmK=mrj0{KTD<#w
zh&F!rz>$MDv;i*M&19E=w7o?KqH+A7sma^+#e$>r0h4sKb<&zco1O3L)iwuWT6pxj
zn(eh_Qks#EW~#QJ2)+BsbX5$1=Ex5n+GZFf)RYmWpN$BImY#z~Gt_vN*#tw-OIJw6
zlf-Lzh~bZ%xjKg$ytRL&>SOvPqK{;_@x9d{1GLrXe%hcC1e%Twl{vltKKsWExqdC)
zV0%|0><P1}#yc}WEIT6Q=yk?gO$x*~-_kYRmd6%)d0(j$ES7#rob8k*6h5d_^J2xZ
z+S6nowzIOH%CY1VwM*U}6XuRK!RNr@fPMQIz-!(K=+rVNXn!!nY6&y;+%G5i-fVs6
zb<>Q_J`7j{Y4k9Bj^%Y@vdXRB@35^;GC=hZF)a?+YqG?$75AxvUv8iCmrjU+s>GuW
zK*B_IeI&EX$+dU#W(U)>w9~pnta$AXKO1>BFjKfh%hT&SUf)C?oau>qiTJQe;~Ev)
zo4~Rpxi4zDZ~yQ!)&R61Nu9hTI!vzUO}D>48SH--3{sOXv2*DBKAk#8=0nt48_p|D
zcK6Q7mZrx0+gJKSM?-|Y#<$ZjhQ$Dn8}`Hv;s<C0tOKrTXc;oAxt9l>C+3nWOSSu1
z4<@kTFS5Z!TO>hU={;Yt5z8c4&&G?Z#y_KJ^FLxSq5a=ixs1c#tp0{)cNnB7{_Ync
zCTL?tW=Ae1g?BzSTI^!{5Gs5#%4w5;M!467L0y#w-i)vMw+ePXSJQbxPxBEcjtbzs
zh9Fs{PBgpotFtY1%hB2YqV7xlp^U!&XJPCnS+h5`tl9THLb605O9rK6jci$FhGYvN
zBwIw1Z6sT^vD1QzjBF#Mvdq{AW9Ikt`ToAY<@Nguet5m`dhR^Wea?N~bKduybMNst
z0bWbu<d+w1tkj>Y#n9%`=mrP{RTPYX=F{giqN{Nz?%85xjh`Ir;T6)YdNQpiCix4F
z<p}0{P;1(!OJ%BUe6n%se2wU_iviV8(fCltQ*#ar%pJDN2aoGteF+-?w+-RoLkh+h
zJ!ORprVW?~?$CqiEO{=AvjZsU?;iNu_g-@g6e=W=9cbE=PvPg|yV+bC+E2H?U`t(-
zXQsxSAq(_`N={jGR0RW;CTz#6B@eTc{LbWPv6(I#<b^H!tf2V|Dnd^?38r>y;g=Sr
z3GasI0!gGA=w7z?9?(xGLv04pRmcLE3MPw@Jdok(tNsJQm9g)Gt+M%*3)U&L8LV;i
zfKviD5A5Sg`>SE1<9T7wBTuOCPCN?UCkfWZ1wvY+d-jvd(|&ey$=(?AmTxV#>rQS-
zCQDC;Px}7VZ4p0r6a~u-sN#4A5`4U8;|(G<%mHqu?o*l1Uy#rH@GfwZT@kfUjDmc7
zXsj-DXI`ZVLR^#NYbaj%0wfK;?X$D=il;R>B5FPnW$nWDp0c!a%HG(80q5cx$xSV6
z;=#?PKqPJ$)X}$1gJ;aPopEwp*60!e_Q`R!PEQ*?VjYeo+>q;DIxTks$1(aD6XFSp
zdt06yRYEDWNiXa5{S@U+8;UW?>d8#gjT3Hx^#do}*<>rwV;12OHrG?TUM41Y6MlA&
zn^92}_p9!1ni9;MP!bc26ewA`&E38%{88s_9enFSNoV{u@tl-=5FbnQDKZZ!mC)sL
zlUKI(Vg|Chh{wa!*uW;n>Y@4J<v}XUcn+84*)pl~J1qAAg^7<bZNhG9!YOc)<}}nw
z45QL8!N5;CHIf0y^bLX@{YMXMoA?J+<_ah2AmQp~y7=00dRd?7H=+q?$*MLcUyXs?
z_`m6ttJ}cMP+!iwsjSbV_{o8y&X)JYgF<cX-xN|WhJ7C3dHW5LTy9s2A-W=2g7kO{
z>sdDqL}rW+_RCxxZbYHO!9+0iyn3JG_k5Eomd4>xTDOqpv{>>F1JafvR(~63E_Z&r
z&Enl9@d{wV1LYLgErQQo03!qc%uC*0xOQK@n4b{l71^N|rqsz;!1~Pc)t%LRg3Y;b
z%Z|^ZV|zi5#}q!CLrhu%os}u3q1v`-m4kY?xrSqJD9UkEBh=xgOqb>IefZ0dhh2yj
zhhaJ#s~qt=!Y^TRAaf{S$l=F)@`;B-yGujrmoT~dqr@!#^zvE3w^E>HPtybOdjJEJ
z-=F5Tf_{NSGgIy*t6GY?0sZ6)8J^eg%=>A-yz({vr2Em0`|2dI);ug1A5wOP$KxL9
z$#Av;o%?3@;ttvpJN862djG>^jw*d%N_m?N1<2A+cY&cn&wI)6Vo(?I7bt;F?rIZs
zh4#M=xTLF%B|%+DpOZSlvag#Zw(;HsKfj!fywm`<TYj(`@MsuW>pf#;9GQo}_!fn_
z@lrQl!0yx(6NQ-!?QGeN;il{sp6ASb0*td*7lfD-%vBVcyK~6nhrBHv<XE*JwjzvZ
z)M?<h%ur?16m1P;bl}qLa(Vf*tVz@x6u6V`#)z<s0kl*Fg%5dmm;J$Nc;1^|+8C5X
zlwuc{d|&HW#^~?Lu~)drQWVjf+ae9<tpX%K^cb%AhLfe7m+QxCx`!&MDrgO_wn3Ba
zdzy-8gQGOC?*xHfvQ9C%rus`g4mc5nMhj5NO`c0+u>;oeab&{`puI1$J7|;J;tQIN
z3<6Nj9oXR&Yd^z4_I%xWE8NPh$H`kc(L<ohyg!{L<q?1+9fpuq;0rSi^pGr>&1-~2
zyu#Qx9U;Uhb2a!n$S-LwyqHJc+!^3Iil_1FxRCv<QHql<<`~VJ9gtK$>i3|McKypx
zh{TmTmNk+AUlhTR5cQny)SqlXaKkz^*47ho;?4+A!1-LB^KvyU^iByC7L4Ci(?Ihg
zHN|pG$HO;_Sc95dik|+LxM~5K9IIr`Ci^Gke5d2`7_{!3;}zs>PIlDeIlZ*`RA&>r
z(gyUD?!%Rn<J)f?$1irWMAVF5oP6O;XvF`0)pPm%(-0o}OQRY{)7>&!;EXMmCiCk(
zj;5o?uBpC_yGqa@7{CL(l>IrYoKt9r_7n~KhOrr2jprcqXD$p>1wv7eP`yXBcm{;W
zleg0gYk_u0`^r!2^y$O9!TQXsW3!(JgzG|IkY()v1n*Rm-G$Oq=57ERG|K4;8E6*>
zipou?{8-3}Yy(r=*KXO4^+kHZ@bKJfgwwCL-<VUccK(B9{O67n5ki5MFfB@MY)vOb
zH&^l{_S)<dw|8}?Xb)H9WuG?lU2FM+yQxgG5yJn4R0zGkTgOSsLlZOI-ub~up(GmX
zX4B0dl6E=&=|rG4&<ALpd}glsLl?$a<)FJkLm!=V3+d(Hn*|Rmpzp!1BMSG0{Pu!W
zNMPl7ug|z!!m)pktkn>RoJw+u!2qrQ@ZQZ;8WCK4>9vyG=>P_@h*zjZ?>fuNh5=?@
zoV@%u0UFbqNP-UAMX&5A_|rcpCLhY4a2djpe^k2IYhcvCAX*DHy3PO;0Ns=`I?noT
z;*$dTEWf??7j(5H;;6*`+=<PlRtLXOYYJ`6mT;FF#xqj(@5!i;%j^3rK-GodLzP^r
z)H{faC|4Tl8`|<x=fM2*!#=%+_8|!IVEROc!y@-53gD;N;gEbcZ;%e!@$}u)r=#{b
zH>b?`aaDp0iSy&uqtUx&{o$HLn~mWuDgO~R2&e=}k7q69m_3L@@&V^Q&h2*UDa!97
ztBWsX!YM=eumj_Y6W8zBD>NIm7KUR=N~Pa(wrCE@;}0zs<#(m2t7cmZpOWWd6*3*n
z+3Ofub3VOs%&;1NJQUWSz|Nq+{8#Jg=G=y{eivd#l)x!$ZCk3B`KFXyA~HZ*F5++R
z{1gTh4A#n+RC*A%UWVI2!{g6~9ay^y59*5<r{OBP$aBUBZhLd)7vm*yFVDzwHD|0}
zZ+*A3t6=N!ngI)WDW%eCkr$7lJ#I36lO>W^%L8{zi9qUM6&fThz_aWmcNkSjV43mr
zfCDGR99V(x+fzhz2T-Zm45ma-vGr@EtZZK>iZeGGbuP^4L^nt<td?X)>W?&(B}c`x
zknGxcpP~0EjX&dNFw?@Dc9%pm-@G6j0IDH*J8f5C;s{uFWCjP-5>Rm!9_+mNszB*!
zhxqP_{u3_2cTtuJ`}@Ol*_c$_*`K=_veLGIY49UlQx@7{f56+#m+GGw*+zObSy;O}
z|7$ky<;Uq$(pBWYv?AO#(zK{NWrmUJi*p>l4AHFFp<-Sz{(RN{Tq;X<fYv(!)VfJt
zzcwp?^S6AX^pzJ__yl4;VollRa0s~sV`zR+S|z9*V|HnObX;fx+WKt$xh}{I^i6Me
z^8s;!GxLKj<^1K&A)2@lv&pxfcTgM22_9*d0~vA&xTLRc3ha!HTdtNKqa8qMtnJ>3
zl3e{9C*p5K7-VM2+=LJ!{k~Us%gpeC40@=b>MJT?MoJ7WIqHzL{+gkbV)>r9hr@>6
zUvy3~6m0E3)`9_N1q)u>kT-WHP?OYyJdN~4Ef|$Q0v6_H)6)Qbmgz7o+M0(0w)YFp
zFpV?G(WMVbf^)rOXGg)^yt8pxohs`)BmuwPMcua{K`kKS9*L9Tc@l3GFpL%j-$uCn
z=x2($l}ek<4|Uln?5%O1ZjWJbW<A}WER;(_n1di4b!rH5xwKK&FZj(piE9B^P7pDf
z(hhbhRe6)Ksx4=RRMH}TxWyNwrf!LMKieow;3s@E^gS$)__JKlV108p(1;UMwR(e&
zMh_HjI;C6!7Q@KmXtQ>(<O%wJj5o7khK187fi)|N9#|nWU`L2Ccx&`hgVEtFd_Qic
zrzOti#@q1bk%ZeFwX;!D4io1m7%v0&MC6nKL^8~7eP5-*nn`RD2ds)gDh>P+iwk5w
z0t=y=PNB=Csx|y!I<lwt8s5b7cad<jmX)dnK=$ZxpBrXGSkkG{TC`tJCvDdu-aMpp
zLGoBWm${kbqSSl%#DNuDWkU|TQI~z@jo#V>kGt{e1c{*bY~QS>k7i)hZtS(BN)ute
zG%E-NGF$ggUtR(1B%MLwy7Ai=@?>ekjYiEOhK3PhC!U$$g*D0nikcAg2!YtS5Q*q;
z{9mBxh5$CY>jE~Jv4V%AN)Q+hX#*aY{w5kbA&^P!;)l^tJVftRR_EG%eUcW*{rm+z
zw2;x?UU*j3))!XSUyHXD+MTfGlYGF-JRR<K2kS#;zmM(cU~|SNmsOld1|`&xEj>00
zKMQ3H<lY|~m)b96MBFuyJ&LF+Pv3F){lwdE#p{kBsT4Wx74^?(@d2Z8(aA*pozO=y
z`D<hT%nGmTL4$xYtiWBto8YQ(GKAn6CBpZ--8p}?&G`AS$Cnu#bgK145jlW!#79kJ
zribT8x?SUhW(z&#08^jY8AWnMdST<cC$d4+db4b<d~H#3w%yMq|2r3ebQZ3mS<NPG
zcMWErDJ-XPBZ2~v4Ql8rdHEyG0B45QITL-hcmSt?hUe6pW_flo$&$;M(9eJA2Jtp;
zE6EhnJ)@_}6UIOW4Zs4^&SuL!<zDpqKC^$14QC6ROR|Xd55KeD#`06x-d^s#yDWG$
zmj}nrkg7&LZ+Gt>JE<H9I4>W{&sTMEQ-V-!7@bGGg&QEa5vRBg)72x%SJyY0aEYBU
z+*L9OTwrPhDO5byzTUuvnM_>?7X1z$9{PBb_uv%M=fSVEho*nE(QHnbvsKb{43rK0
z!#}gbP%jM8Ll$uW350aByVR0(Tbolfu9bHH%*`F3nCGp_x$Wp$RPaovUM)qktsM)-
zanBZHtgHYVWyOL7PW!$|dYTG21|J6YhgF&C6`NR!GX>C-t1*FDXDWUA!&J0M;TYM_
z24YmhJ<RpY<3VILNJCpd>~s}#{2chy<il$}=Gpv&39ls&phSrsBgIpRwfqsU{p7P$
z3lagRhB{c*AM12CV+|G{5k<OtPMv&DFu&~1H#7@2IzA6way@+#hF}6*54vPnjf3@s
zip4vVl$G~AC)_h^SRqS>)1Qw~Aax^>6{%5nL7zif1we3W-|&sdId<JlTNiPT&Fh3?
zL&xROFHSKT8Sy2~U&9-Ju4;fSJ>g9t$D{cstjTHnVRrOCmz?Qroy0M``gsg@1dlA}
zK6mpajVWYppEt3KrI5YwN)_yqGkFmwhzwpHO#`fb+Dtd+3r<b2>`@2Mk^6GFF3Pji
zS@Z_aGnN_#8Xs2tnfWOT*#XqZQ0C2gE`R!5DL_FpweV0-;6NrKZhI$@rq4${OI|}S
z<ahw*;fn*V$FD!{zvXSKD}5hUVSbAOUMq3E><tNmediTop@Ip470w-hUd+sh`{<}!
zpW=r_<NB;d|4Jr{zZvH;90ONo>7X5|>2b$sfrfNLxVmKvzre7-qZo6+c((aHkO)lJ
z`bDkXTzY@=BXx$B87yX@BDu+2rhgCr-AEf_k2b{<dvE$VnmchR9Zg4^`hFOpLYBfv
zshvYeSCG#iYAS_A<wzR&Vtg5>3=})UXQ#TAKbcahUWbivc*>@<TABF~U=W-Ox!O-2
z--S<#t@gNQY}o3aaY=(WCpAJf1vI%RLb-pjbVNqI`MV6wUL`rDk_2UVcc`<X43jKi
zl?>&Gq0i$`zo(okHsPQ=PZ8J-A|u?(r9Xf^;DvVT8H?>-9s@*U0Svv2!qxMI45;(!
zOr!^L^$W0d#7{#4qnuvvJf6kOqbO|(gF3bL<gVi7uO?gW+Z<ywkp+JFs$ULq?g<qi
z$U21A!|_&5ooEX}Aic0dib>{GdrBS`$*bW=^_XawtrhyIDXv?+RY^dg`<@f(7I?pX
z+k%7w7t5F$3}6mOQhsq;R+)+Xn1ctc13hm&7tG3S0aunT>t&hH3o>^YNbv-aOO-0!
z+V~xiBaC3?5zp_12oER=-wUtu8D%I+a3Bb-tK1Y(C&3-WYXw}axTYY5@InZ647Yu$
z{xVZAM=|fLD@1px<^*v9fZu4_Lp$L$J%IfWBauiZvJJQdq(JwG0j`dW_mjs5pcgPR
znU{4;%gq6e0+Y%tdJv>T@-+6(d_(w-=|0m+hkqnL;oEy8H3qYJARFXoD;g2=g?NY4
z1|E6rr*(eY78;bCI3uX*zJ8p<RrUmd`{yO({`#6;2#*6CtgM4yZq`#5*mU96ZS-sD
zI|N|~doHEBaGvFMEH|i)8Ek`N&W>TEJPmGE{ovv<&wiOHT?1OOM{)?<V>LMfEM&Y|
z{Ce`XwOq`@3Gpd53S%qGGJ!Yroca~21<B0e{Vjyu>3=={rwd_I`R118DXqL{1JCxT
z;@-NMM8LJV=-DZayG9s;8oU!HGG{=$Gv@T@BJx^}FbHl(fRY&#E&)6L0+9TQ4H*~8
zzZatY<JtUgo~owlsYX8A*bBcBKdVjX#*PiPUi4GDNxrV5JQU&$5`m(?WDhOj^Hs=e
z#?c3kDh}TW_QDE#)7h>Fz!7$)CfloJ>h-=wZHVx!o)gFCgq!dhp@)+XBMiumJ-opu
zXAvX7=*X=0x!t~!A@$`npu_LUO*%5{NA4q<U6!4c+b;@|v<X1eOnfSQ<k)S|hhV*a
zq!X;cdpO_p*5pGSu)=VMQnknj@=#rksW*<9Pq#TSOT(YIq6oK;k=^xgjTU4ZtnalX
zx>qwt*WTN}Z4ewirBD5oW$1OK2{|=1cms83WGOIw^JE)=Mkgl|;{2i=bAWUaeJSX$
zd}I!n#V`of$b0Ghb4T35-VkZ;R0UW*WLAaH0woS9)JW06ANNT<8U)+dsH1%?UuPn+
zL<Ue*VP;v$8ptWv=b$IAfnl)E0NsE#gA%=<Te8@MB9IqAo(z~~9sP>T7+5;e=sU7M
zt-gdR?B+^toz2rT617tYX2a|wwg)K_1rb{dLO8J#?#Ek3bsEaDp8Qx<<L98)76D3R
zHQX7T(ylO)h=WbIVBayyJFz8OIsJISQxg)TOK$J+m$8&xG=<Z^pN1T$t3UFCG=o|b
z^RUTi$6#Hvzkpd{jT52&ls0#02r&5afBciNfKHP+Rf;_|IRpbmj}Yq5I~H<gW&zIE
zyWwZ2vu(~}QY-J|y#Mu}Mwr9V(@R?j56NM2ntQW_H!*Z1#bs~Aq8M7s>f(jum?gpU
zreQp}jp!@Fn_++C-YQ7j12}_R9k%+L9*z^haAfLeJdqaumbZiDI)7#D5?--NJ3uPP
z(Ux$)=XVR$5%lARpTJ&3O`j4}_FTsqz$caLZx9WO0L{*4wS1i?dcZS0psJOS>=4~O
zTY_yMCvOJ$<AhO_OA1YyFAx#69ir1_Uv6O);)0{mRfje*q^f-GiRZ+GV9x9dwrG}K
zOc^Es&TQ@-G3<o=73T<1tyXIiKsN{LdV20c*Fo$(JVUjto6wo2h+9|z0@d^Db#14g
z{=tiSH#7Q*8Q(>ai#}e~PkL$!1O*U#zeU%7s!hb#QWxf?SUM(IG^N81EMG(2q-OOA
z79<-FC<UnOn)o?W%D?Zf`6tHo_;|1L>OB*5pMWFYND*4h*Xf1Q5`=j?;!gt056Zcn
z{+9j0vBPa3W7@JWBoU|g2Jst}afk-Ha93Ht;C*Xo4P(tmlo>@Y`4R$Q0pE$qNl9D;
zRylvm*W6_Le~N4M2ad)sP^rZ$z_~F9qiT3fttEP-s%yY9isCB%o`_+~$FFv>=SBfl
z$yuv8`Ls_*bSD<vkC9(d9j7xSTTWYetKW?f8T)M7rdQ!mXddQK?T=eHdhoX8rxS1g
z<~(@z&E&Ve!AQB7QCUI(b`oX=ZEP4L2qV?L_b|4IG$VyFF2o&6fu!QPMVw~ce^kY}
z__Q%N(o3VfJ>w%(LUcprzF*h=HpBkR%<WY!W`Z4{t?0lW{e`kk%Q6W&dB9z&NWSHi
z(yoG$&Xjr#t%F$8(?WyRWEU=SB->K*-bMPZ?*p#rOsZCf{Dezd#fp|<1X9YO^}u%B
z`NS5fN4iTQw_`Ztu6DXGKe+9*w8-o9y#GsPqBq2mUI%>R^Py)Zf=v~S<#Z|CIl!R?
zHw9R^!CB5q8S4wHyuZby9@y?4*eMUh^=N<5p!abs4`{QNn<CheE?~57Zi}^eRFkAU
z<7s^jf*1z+p&G*BM;vXk*rhocorlTm+_<PIcE`q5rFm{&Jh0uF`b%yyaRLUsVt@Wk
zV(<?p-R#Tu<?mZR6HqYS47XijBT0B25I1vWq4MJVf3S&W0W98v!ysq<{RQDx+0sGt
zzpYjT6mwx(MIXJ(R(IT|0Q5;ap<x6Kle<P-!g&$)mrv@jBYjOysB_O{B6LX{a8wV^
zbd@Id>4YIC$dO?z5)r&WI~#?xFCRQ?z`uy)-9Nx`lVj>?-RFh7>G-&#ep}y`b~Az0
z2?>zg>0<f5_h|CZaA3J*t?w)5v(zc|D6t4cpuVJ>Zos-iC5#pq1Uk9>qJ`8AIWq*<
z@a*}V9g}Jicq1cOMEM(6AfB*Htik&O!^1cQ)n$&=0&4n3Bb}C+#L(T)Q5a1?ZqKdf
zFwC@E)Dej4|G5XA&bBYWa?c{-m*1le;@ls*5rxy_ZSk0jPCdQ?5TAhH9G*BG5!K}m
z(q#^Cxj|&s(W3~q>XNCYXK@9XKHb^P7h_MqAr8N?#)N9~3z)O)y|8wSqe$RC#GS?2
z`UUMgC!1Ea&|X?LY<ONlg^*S+y}NoB69`Tp+CFWK(-yKHsd@VNN1{K6YZCWwH~RH+
zL!%&2U{&%jsFYW>XN+#zD!N<SMJd<25|VK-z`)u&G|616T$6h2fh%$<WQ}2(^<q*p
zfAbvfc!J_d4Y{|05BL~`<qo@~CY4i8KA)5OlU%~J0HU0pkrq0VZxxZ65Upa0<q8vu
zMKmj72y1e3{vKC#a_hpy9Hxj$NH={oNQV`&$ddRY6FDsw5SFXpB=hpsQg;d{b@$_#
zcw;S=!SY+MdwLJ;tdn&snLv~AW62v9ZR^B|5kv-JORNWD8Q?I<Y`uFrtd?UTd_hXh
zc8k|Y^~mn(qRf|S5xQZm3b2vu97Bm6^aEa)sc-FK%UGtzkwy)K#G>t<l!<ti5n;+g
zS_~{{nV^e3MKMEXcxXkuoRYYJS;FEEBuO?iJJ&1RBoGKlA=*tK;l1sVi1hOrwplpU
zLym2i$th+i0bfV%8{Q31gs8-MDL7<!ssU_R0`~(W2`zN1r1i|Se$g4OYg2}__V*nW
zS*a&+nR0@1lmut2BemNcw+7Sr$kKMPv5cLm%$<*F14eZ}Qf(HJuLWlOc$h-H)Z+@r
zequM^5Mb1zuRf!gKU*TP_`SamyK4-cEc^}Y7Ixwk^o?gqqPlnst>2EDrR@NA&!VAH
z4Tu6)VW`ZyrMi#o68}n5hrZEtYcbAeo~=P*S*d3U-<*0nf4`Z!hc_Xs`l-!3kjA`}
zHYcMUrgKtTM#w4+KL>@^4)%~fkoQ=ogW8@3HhgWwIlCR;PEsrWH_tzIh6T9UrDZxV
zX=840kg~A@fkU1R(aO2aQqKpm^qS(D)&p-Z*0AyH$%7pZS}YfOlmor2c)jQ`7sRkf
z&H!O3^MSljrgD>wIqDtXyt?^G%AQ02zRj-Zw;9^!8LZLx>slO}^PybDVl(If(=D=g
zweZmjtnkw2Mo%BMCySSXJjm9w^NN0flVo&TqVN>b*BvU?v;5sS5UOMb3G7Yb8=46g
z{_b;K(J;#7u2oS#%7Cl=T*`Zc<Xm&-HVFbvcm_tZmQzTeuMpxPzRIe~3{l1kQS9Wd
z*=A*f0|$>D$4@TSwh|uH=ubRA+GCk*P3~SO>Zg^j+k8768V)dt2L$!(HXZ|3OAk%n
z0&<BpaiVpxiLAPQpjyZp!D_j;XIE#c%pMlm3U+Y~hpFE<c6AR0#WlQcY}hsNzxkEJ
z-{gdO1n>>zC^ZgGRYPs>#cFmsswhBO!e}b%w3TSNIg5!f!UKj#$A`RsO4fRs8R{S?
zmG59JrkUpviuv3SkPv;>NI!uQRqxVMKOhPH4ymV?MJU(4&3SCux#6VA?!u&{y&qHg
zf(cY*ppqIL<NT9$=4O4|kN6cM?vZ$3hD~OLC+GVe#;v5HZmmBg;8Q1Y`44NcbM&!i
zu%~9Lx&50rIzF1I<TIt&X0(9grm#$yDKy*r2wscYeA6y{`7HzKz_U8nJ}3|{4}Ty+
z`}RPx>!fgS2zCUWjeVwHJfFJ5-yj6dZt&grclt1IhJ9yzJsaEuX;19n_7qw4^Q$@^
z(l8P?{%N?ovjsS;4_VbXmt?o93bsA7FraYLnK`<%naOcJ|E7CT80a8DBhK?Rk&Zp1
zGeAn6T<;lb=UrDM`T{x?UxH@q5{aul9YMEdsOk~V_HwwKkw3NKS`z&0sxoA4_7hXU
z0eS&#1=9rw%usozkM-94Z}p7;J}sjjg_?LVyTHE^)FG))C7&z1lFZcskI)`|fEei%
zFb8dzZC`A%WO0yM4f0bvfyuopqNBnI#dXMd5w;)rtNjOjaVOCX$OggnpL&#Ob)@W(
zI^AA`N%{LpONtFT98YIB;YE12b3<l06gkzA+Oxneh+!KpdC0bJQ+{yM5ZWB%aXQz?
zQx+>}&iM{c^>31lFQh>$PR@KE^Gg4u@GvBP1TBG&>)g0x>QtiI=&IG%u(1_C0**Kr
z<h2|IZ~GQ)ew_?O+dMfz?ePY|p%LxFQ5u~Lz}aF`TF0Wi7GMKr7dK#g%-`x}Lw9Wq
zaqpT2&c&l#8kgJEtl-dQ-+(C^U+RfL9KPNn6ALgp!^NuI{Il+yKIs#D3+TzNNTDze
z9btf#8WOv!V(_Aeb_AE4%S_AuWqOZG7r&qfdbVzcDO?t9cM3Zj=G9JDHl*P8Deoze
zsArruL@Ot2;W9=F9pTZZXTL805I?>B8NaaOs;qNHYMh3^DERRb9OM#+m*j+q)wsGY
zo=Lixi~|@kFGC3Tv8UmIVya_jjBpxGQmwCE9=|$kTO)rV{w32|_Qt^7)0=GRifUKX
z4jk`lQ+*i>SKfJa>&BN(E+ucJ;Y@FaPc%a_02*dZKUx}Y767EFxX0775CIzbZgg})
zxjr`~>0f=J8JgOYW^ZYAx7RvHwB7v74V;&rL95KxBX{%8V8rbpbplRwu}VDIHxefc
zeR>sMOTpX)usJ~tUc99xP!Ydxx4BzmPP#Knl2*or;;aI~>CB||H3V<;hb=#M3fiAl
zgPy1wxd)4@Uk|cfUTekN5KxlQ^fue{-3v47mI_{PYAII=V`WKS%<S_zne*@A;E|jE
zTl1_TX}SLU5(?4*rDE27x73{E`7&h)+F0METLdDmt}3JV?}kv^p38Hu^C_>Gw3$;E
zXBt}P*<;~6dEuFVGTv$3@huDaGEG@YIN164D?Dg*Jaavt?((L`aO>IK>md<xxtj+O
zpumO3qIGIU-w}gYT>X|h@7=`275rAp5noOP6Daef1FC>?kTW|OSsL^(o%cR?kRPfY
zaP5bPg~1fKgG&$%w?BPZN<~Rge>-jt_dZ4aipiVL^bIp((kB^Di+9%RO_zp{ytPB%
z0xagm7LITIXQ)uK%!cNE<+rW&5n=qiux<;Bz~ye%0ux5!+S;2omT1|qt<LkBKjP{a
z2SO;ndliD(U4JbU#Dz7-uWhM|qE3BhVC%lPv|)Jhi5`bzQUjh~ay74oJL=0EtRSxB
zf=T(;d5EA;I}x=#y~TaGUpQ$XeeYTq|8=|1d!XtKQn**c$vHWzO%ErT7p^+ci?E?x
z{E6GmpY>2zDXYyJqag>uH^_xChBJ!$p;lgQeR2+}*0(pX9Ph%|yxgZ4opsC_tG!rt
zbOy1FA-Xf#>^HqWUUp~!u9c|3Bvg4cB{cUVU0yeUBCexSS1JJsKITl~-qlNyDBuZR
zZ*a11aizj$>+>zEtm+akZ!T7@>n|94-^Seyk_nTyYoX{OW7kKIv+pENaJF!gWyuT;
zap6N=E*mYoJ7qzWlpZbNg3H9%8@%~F$LYgggrnJ{n4*`f`y@Gj%5J_iyATwyZv!K<
z=S|CWbGmG`c<IjE&?%;8t6$ps#p$F-G0$xl9DH--^Zn)QH}y_HG|OpL=rVK1qY9}A
zZp%H#WDJy}%B73GRVHlJ5?2&YWzX$B5W*8=@lBKx%dfhbv07u)XMbpq-E2hStze&U
zJW8xrRm9gVcW;X;28;ALjV#tI2R;A2%ZdNXj`UhGX;x>e=!`nI;g)uP-z2#A_@3W_
zMy7l%!`p^E(eEwRUT!lz9beV+*)~qrCC?2i*6{S|BHI&h14mgu*Vg1#1&cz^C9pz~
z`R5<1y{uOnk`0z#rm@`@d9p<Bj}sbgR!E-UXyHAXPfcn|S$|-+hG76Apc6u?(<Oc1
z7t$(V&lA|zm$Pr9q?(_IJr9gh<ek3Nl(t1lXw;$gP9KM#Q5|m#l$&w6jeX_I^YP@1
zoRD(jO5dx-<{k0kAJgWJ5)NZ($ZEbykQD47GEkZ&%!7BcJ<cm?8DBXWgRpy;k=?2C
zMu1F3h6_C?I2S9BJ1Tx30x%k2FYjaoyjgQm(NC$@&rcc2nm?gqIvKE0KH`s57I-$h
zLYnLI5zvnN{zZP>UwEYWQRcdjLDHU1i%I}FY>>^pr-tdvl*ZRdy<8!G&gr*%MYGi2
zEjhfQWD80*KW<?0i>YLct&=QLxB1S?`mRY$f*(OHQ2HsAuzqaG<O4CJGdSJ)l+YW<
z{t;d{rqW1FK)EPBDr(c_TEjjy(2<$G#KDpIchP3hVco=ycU|ywm0}Bg#<)xLoS?2y
zdy>B3$CjCeH;3v3A7z&<fnlyKf5uJUUWH!Swxi<Vs0_$&Z$_7VoP#75z0`IlXnfAR
zQp9>@{(4Qs5lDUSGIH%n!#dwrliMwQ;roi^oWIvS%x{ai)u!Fw;4_*@_ju=D(*LWL
zcN-|bf>-5fH-xA1h@fl51zF9!^Gw};bXql=*o%dFtkNPGs<-uZ5Oz|OMH8MidSOsY
z5YF+(L(-v>(XZA$i93bn9Io*ExQ+viNP5K*%NqSVq)*78r|hr3Sff_kv+E^~k*88c
zbVpm1J>6x+_WeB&>_q<G!e32Ro3E_j_+oPVf|wf{XoGdb&nx1#&V}}Ur5YB(0y$5C
zFlF?Y&_QRK3a9^tjCtp(Bm4-N9omOk+lhhyDUk$~j~>DZdL{}^uNuR)kI9qUVop3I
z_gvw+ta9JfZ*y=|xzB?|(&pOuYs<3V5Z+Ihgyb2}$8-Yg&@fG$-t+lC558wfr_4+l
zqV5}uI;NDIy(tOrtyc&}?QT69UZUR_AGhEvQca6|bZC#=)h)%k^@Vgl=oa?#uxxdv
zNf7c=TfG)&OFI|fw^vM*o0tZ<@Dru@a$|*|NOD!<J7WZbV;)R1-^Wh|5E5g$di5Qp
zJhO3?NF~t#wIG*~=hAkk>)4$51v%zDI6-~G<s$3JC+*3f#Hr$bO8u;U!J_&e<E*NY
zt|7c>eeHJyzYdam%J+rNygN%|0^P2^4wKvY&a2Hglc18f?)Hq%ApYqyKqTpaNH+Sn
zdqzDDmsUf<ca442)Je3gW`)$GBofST#d>0_>B#*}_S#?yEkI~?=fM3Ye8{0PVSs$m
zwCj-$eHp_|W@w^WRC@geUd<_lq`$~>_?wy1fHyqG4l6yyI38>W%#PtLoy@hWNhBfH
zej!&nb$m0MlUlr5)cIe{zgmLF#;3N(zrU=1xLs-J-dk2^Y6GI9&l!Tc6czb$ShTQU
zxdN|DVLtvy6Jb3X8?@^Yvfea8VJD1GVBv2~o#lx;`1H2s*y0HQOfK1NAK^W)Z<U_e
z<fY;Db4fg+T`Z|Robh7s#e$rcBhTB$37+)C{Q#UFdb9D97c{{x=Li_HQ~)1VoFKbO
zN)X|+Tue`oQ{v(+<d#4rNwVgJD+{)FIz^F#gQKD63w&rZnxo(xZx#4D@AOQTZzBBL
z-^ZzfoDG=m-zGgleh^QUT+d3uecmfK`dGV$Di4KA*hOa4;=Zp^<HvOGaU1ybhB-bS
zlvL~u8)D%9xet;}^)GMP**RMSUs&|uAeF~k4a5d=Sj34~5`_|FF*A+Dx2Cr9zC_(K
zo|Po1GYDMzZj$-t*8?oI$D7dh>15u6<<v!}PEQjIzHFYlVup-Kw?5XdE9bmD`j=66
zbCId0n~)RGt_!H-(BTXQg`sACAtUdC7HD19Ll<o<Tm*QnkuU{^e%Zg36<vI%I95lv
zzQsr!gmm<U(LWZAsx%}MgTjqfH^y!K(9v}Ha}0ey7qs9nUC!K_9XceAX6$4eEc;dL
zHJHJO+^<9#zj5q>>RWrm&zpP}%eHSlBWo93*0OUTo;%psPDd0&+**}_E=WX4px~q;
zK^=BrQA_9^sv{CZ9HLChN3}Azr-e*P-KJ95zUE!SirtQCTc|7s6yx8=C74=vP3Dh@
zJa?hR43(Fb5D0|U;YC6Uz9<!^mt*2<(jTe(=TQowVB|)vLHeifv<B8X^1TutbX@Au
z+*4719O%gT4=O0gZ<}O!J6i@}*F;&wX<*&-uApEC#_7E4{a;H=zLkf1R7E>NJ{G}v
zOv^|AJ@{cD#2Nw+6RWT*nMJrA!_jM&z@s5J9W15vES#cR+MHxQ(#l>FK^mOT*pn#}
zJkf@^%q>`lF0Z+z?5&RB?05!*&?|gYU-}twMC8^ErhdGgK)y?O?phx-zOuNA-;0%U
z7>;ggM5QJ%?&*NXA6LZYLj@oDqc(8cZt@JdFUqc{^=cpwDGNz#&VokXzKZySQ9FWl
z(AZvVXHwF{sR3Ht;WtdRXShCFz`fJ89GRE?)F-_S?029N*^BdDBP)CCg^LID*KEqd
zKPblxPDu^W-mD5>8qqW>qL8U2X6&HFWnnt>g!pkvpCj{rbqf>P12>6V#R@))`P1W_
z*L|5j*v&$VeAY=~s}X%2r&@!vn7CWhcpwn8Ci6oV(3|1|v@-IyhRKL8wR?JZ_PWiB
zLfo!;ly-@Z(M}_iTdvWhAojq@^LCbB!PUZ_sL7or?8y5NthR8UmrVqO%&ciO4)}#j
zizPviAgS0Mf*HmaGvK7qYQ66IxG!bMl|7n2LWk$!1d};E5j%R)fRo1E_nV}R<2l<K
za)igg8W?+FTaL^h6=e6Ph4D1u)-$uIBkwFsZUz-j(zojT2{T7<(XKFY0B0hKf-Fpo
zu2$!Yq)jgA2y331sp7PmS;2IJ7U(Lk>LU=QZz`G4W5^y_G@&@Q_&}B9D@QVbMsu{2
za^5VPA!+qP$at!KLk_jgAp(Fp#1~7OX`jeH#ZeWRLZjcQ1k;R(J+v7#&AQwI080Ez
za)~Api#geH`djhG$!HVWplV=0;#U8XrJ%K467M7aKOdO^3jc~CWjH^+R!lnu<eDV~
z_BT`ZLf;{)RV6*7V_ta{B~zP2;@Bwl3^#hV8T8p}AVBFhm82x|oGX;E0ukm8da{mO
zOfQuIJ%bUEd6WWNu&11V9&X4PCa=#bU~o4{NDNCK;f3v?T*0dyN7SIn<Alfy4Qq!U
zPFp0`_t>@lC~VOqehM##bmoKSAk9;2pyj0%La4InLJ~chr-pMpE);Uns|dZAIn>ME
zuUI2?N|<eTla;kcTj+DoRP=q1nOH|~2G}NUCTsQJ6o4cke@oSXCxW5bS(&;MMPMf|
z{JVQ5mn~&%L}&enPKUKfLhWFN$ReWSEWY&%xM!mj-xNs>eK8Ob-ykqkMLQK_(8?x3
zE=C|uU$Nl?0jUM_T@P0%0!e)iY`RxRFJ6;cc=-88Mjvc%2njv#LbL$Cjutsam(3iZ
z;-w7u6u@=N?gW;Dw45*92SoLji*~#$#4~jp0(^TwA)UoZY-O%5dr%x47RlO=X<eD0
znvU-Hi5h6*MA*rzjo1K2#gh4f$kZ?KY)8aio^?I2E@OZf)4~N46U1L12%#F+=mdUI
zjZSbPB!@WF;fRn*=<Ta-Fata6mKiEcYs(MKcpb=LjQcxmOzEZg1e%i{mWCt*)%1-g
zg|hND<ezvS&A-Vj7p_?%@ncyniH)c6bw<n_9Hk0BQSKdmKf~|$#&S83g4lx?6~XPV
ziL!&h$)oWwnyHDn+h0-YF%Lm?U0eThkNsVFE)DVm{X36@oN%ZPCKPwkkA!M~QB6%g
z)C2R12SkJh^lL-(aHst$A_SV_VzeDDH=m85^x^uL-Zj})LRY!SKQi&3Vzc5Y#h)IU
z*3}m|ReCUBd7Rig82!Gzc$k#5Tgv{kTO^Im>GD|>9}(dT2pc~j|4ZQ)!&l5kc{`KS
z3r$%BLfW2TtAqt3_K_mI_##HE`5ew`^+R@BY#?+W5*1h9j`r_q5V!*`Zajl$u6y~(
zpNAv!8GtqrS<S(U3dL36MsOX&OLO@BMi-0#_NuZ^a5P7~kl8(8R<o>c;L$M*`RRtf
zLF;Kk{tc`5v##fGu&eIUWo$ImD@?s%1Nbp@;sFIE(EQ+@D1$PIK)XTUAhZx}$TWKZ
zvyth|9@(G@H<iG$7Dn%Wu&n8fi6Lmm^Bb8Vm&_J-!WoA2<xc}sId)GS3~GKrVdw$p
zPu|nRzX*L^JJh_*>!}mQ{s<XQJoYrZzjxA1I9#0Cz`sEU<;1qfTr)E?F?n-Qeu#E2
zKKe%815LM^DMUSnS#W@56~80`k%e_mr`bSO%>ts7r`B~wSSN^QqsF`Rx_v`ipq5Sl
zM0YKN6Lil9esl$e3;rYYaD@oZxTrkC|I$q3yAnP<^mITkj3X?w(-<X!o$&@BU~<=3
zet>3%4u29Q<<6U56fTQa!$s6|lOPTTw>z7cQj^S!UYAC!7SjQx;%MD0c&0j?#f!;F
z=G<TD3~xP#b)~KR7E#89^rY<Y;F>b)3KIYsHPzT{(X>b(8{E)Bc<T;!9o18!2!x#n
zux$yXFq?TEXhP37KuujOJ7LI2Xe9`Kh&qq3i>%B#xlVC<`#F>O3NXj30#j&v7F+Hn
z<z18aDO2ods#?zPV?n1ffC3;8E0tv6Hg+g3;2p{(*}Qd}JAXhgGpBpT2mv0sH5bcq
z7(h$T06W@%ISb@P06OD&g%Bua2bX&~rhWj&MF@gv{87fNyHcYpC)k0ue@gzZwoK>`
ze_*-r?Y1rX_|LfO75>(yk3rUh4vJ3^d7jQ`bdS7*E|mniWT|o->)~_mUpPHM_jrgo
zbn<wiU!QR1&2iw**X07DYNf?LX>v$yto)xx;-5T}>e*bbIFE51C|^Ws4QgDOYp6dx
z=(+0Q^OfyS{6Bx0-T(>+bfLmDcxZno6_KoA8w*AM9>RaDr!ZOW>}&;kk1n#%5I17O
zX%c*pX1)Gr%&mrK4~jr{VCkS=AO;ViZFZy{PkA0_Be|)rr2%(hp<!%dBEIrZP$qwY
ztr~#2Z+ZX(nqaPM(d-2Ye=XrbY@}ORBi&QkxF|s7({`r|Z=}!w;hu|8ugv@R0j@W%
z^b6K}AjorA#+hO&kaw6{5U4Z$FYgdYa1ADW1JY$9uXb?)wt-3WT`v&;Voa3oADFq(
ze};!Ih4u%C3?UE$udB$!tN=s$7!I;_kbcmZXt4qOqs<8$LH3rPqkB3XmFZb7E-J70
z6M=DqCH;X)%KD^adsVk9`T|xET`K&yksW7S!((EO;{0`Homlofk|6Cze_d}?)k=1N
z_CYg#-8gZwc&X8qA3N4&%e&{oAS!(DqiSJnoTo<E_|AEC(DjVLCoybCijy-M4ip;=
z;6qxVX98_=M#?(US?YqRF{KY`6EvT%>n|$Y^^@2Wy~EnC>Lop8EP%en)FdI2LX&QR
zw2F;QMYB1%JZheZuScR(7q^;j7Gys&y;+jzmE~`Z@Xm~3cMzW`?2k?cGFkWjUBi+B
zBR<(QYuPby(PX_urcpC}Kub^Sv&Xu+nSLRs_afhp4bUu`wUUqYrdNOV0-t$3ce5n*
z%87wnIWmQo6Cyp0n(DR!8XN|`9qSv)20rcZBu3a<Vy4jKvedEQ#M<fyZlDKiz?B#Q
zWs?2fK@9GLuKP|fNgDvSh139GpfZnu5F}s-UjzK)7;})-dD3jaPG%(8PCEk}0<rSy
zf6pls$et>gbPXxv0QwzqM4c9LVEnB>*d>Gl?Xs-@!2_{p224&9@$qN(p)z;CK-lvl
z;d%|@&}pQa!-NN4Zsi(Z4cY}x@~BbFNlHp82R1?c_f+Iyn)pTs!qsEiF>qwUU2E7c
zQ(@v@O4A(kEXc3+J`_M|`SSk(SjIgKFbE4J8wG(>y@2u`==5@aARLFd1>~CjFD0-!
zxkP7J7}xc`9#UoS9~6iGtzyIi9r&ML=5yn8=>Pr)Xv_AM4*m4wCVs;$`5hus8?Mc8
zwus^7j;LKh8EVGg(9w>~;+g)E@_mJWKV_?B&<Zb(D*B_xD%wEtTH&89Ydtq`W&vyA
zJriq@r>cAsK{Hr*=LXJWL|sZrWMCA-uxZmOZ}Gn1QzZFjufN3yRwPRpdl)F>-}u*w
z?zEE>3BU%Sq^H9g!xVp-BhVxZD%>}E0Z!`rhk`1c_qyvk-7iUCoIAOH_m2*+7}~%6
zY%$TDd05k`z6dNGd>(^t3{vZ6Fe;q>R>9T~=Y}1UjH3SZFY#Z*k?2v{*~Mn@KkCY>
ziz1u>mUb7Rz6~Mcu)<uw5R-D(S72V~x9kA(UTfGqSXFB737r}w5Nd7p{wz~L+=<mE
zbKnmhrW32A34Vq`whnr$bvAkh$S<IwOwcKw18>RM^t+019bU12_Ye%?2@?L1reJ3|
zh%C8hCOP-y;pg7bzW#E7EbKSz8f_5x;{}XhOxOc7n8#8>ck%-_kTfuk|G#}T`JN+P
zN1h#rbwHv;H1C|M_Xi2nzgpgkwZbVA+GBKM<$?sTEKc;%OnkXR2j{Pru6kfsk#wY6
z>$iW<&w@W~tR7kvCI-_KGzka-7zxMnw0fhb$`^Gt2@pc@i=r8H0fzH9G9x;+PL*P{
z9L4@2X4IrSa3wwb3UnaaeCW2TtNcLh>n?K1@@h=mG3b!$=cC2@mm#S<zV5`TZR|qr
z5X$X1Pa}5nXXLqXl!XawoK+62yAP(S5iY(E_zg<uBkZlh>h)AJ4k!Vn=f1(ZVHPC3
zFRlf%K;PmB?%RpTh~^FvhJ~<8A`R@rNij6eW7|C(lU8v_f8@CXVjMJpsko7qEE%LX
z&VC;}v=U*(KN`;>eBf01%LPv())BgBMN4|yhizX;8kt0`a|c?SnO!+L^8<UlJ7<MT
z?zK%^T2K6&O0$Ta&ufQ2A&%wdhLpyl&9#63Sa^D@J8^+B)r30y=_p^E7<`ata`Tja
z^2u2@;dYcm5Up4LhK68-3O1Xf4X%WZjZHG~rN!WMQ)560^MbmxnfT}oW3FJ~Z7qfj
zBt}7d&4BG#F^1{bkX0g7_`^v%Niw(vh1l=u&b?oph}F~Zls#)KhHZ`_5Qa|<DXu<;
z1;Yn|Pj?EF&1oNmh-F8qhfj*fT3<V-p4hr=8T=VL4&DcEpG<$fuP^~WWx?QfTo7E6
z2=KD^`{x5GZzgwaN2ys!tX~`PiQx*}LsZ;P6~BW_rsYMYXV!)+So+qy*&?(J-NEV)
z$gU`lNh15t80Ji3<nXGXAipm4P%tAVdtM~nqUawyqQxR~r+D7_#tlhc+_d}{N%jv~
zzzdEUC`7D_h#jo#gjB2qA{IhyM?)kLC$hIYQu)vQy6`}M$+(H{xb*!`)Yp=Om38Ry
zabv|F*ViJVS?urkHCA-tO@g%04R<Q}TJ>sa;fcf<+!KJQT=QZhE$Iu-@I<H4x}w3F
zvYG+IOP#UlpZ9M}4=v@v+<BKH)^Fo4FJ6i4aM$0~gC7>glU6_Mbp%2K`+|QCl(_%d
zvE5uq5bBw_0t6}qqx~~x^Ky5B5ZIzmW4wEpNTJ@#T>zlp{THAk0YIZx9{(4h$8Nj6
z=J7JP>PWFU+UY8cK}VySRf|~mpeA39rl_w9S4(r3y6a9pige%I(|y|5EsikdPx}ng
z*Te19c?MO3<~ekUjDBi{b=N@tIF^`>*jUo@19p;rh5EA{#c{0&$Qp&!p{Xd*?nL3?
z%t`TJvmrzkl1E>EY_xvpqXo~#>SnYRK5E+G%sd7l`({JXaC7GgfZA06wY^Uf^9eA{
zol<RDGmt^|RrvU9KG{8O6@2L%!|=gv%WrMA28QBu_csDB-4g~v$ZJ2kTkQJlo%P@=
z>*q&6{MiBMs}&Y_k*|J)c-#X*tpvG?6J3Bt^)ln$Ek{_3Iea<J5zH|1{(kQ+(+`y=
zb|s02PG(3tb>CPCMd1j(85r4ixot97_`_B^mj2_}7RHZ<Qy08>4i_+}WL*ubRgVyn
zdVg?T-}cnp>Dj!ov1{_HAejEY(TOf`w^}tS|NeBTF;-DY<jq=OyWr~Cv<8u^G}`2S
z;TP}zBj<PjC1<lK>P0}*6RUv0|D>H%0C&*sb{;jd`a*|ZR?c3}KP<q+n8w;L%v!(N
zj&koc7L7WN=#V9A*9`Bo4qm<#F8mx@;?IRuB?E~fQy;N{-ZWUWL_};;VPgJMYIx~C
z!Uqb{$h`p6z)=DUU<grA`<uG1Z{O-pP&*$ST>d6De)+R_BO7Dfx%kseFIbsnQXqrq
zi}4q%AGfIvW#36($a_}!viQ#K>wCSCCSVROQxhM<{MGybe#|59J9mD!O&u;>zg6pe
zEsFR&(d~fv{_l_Wd9?`_U${+ec*K10*!Bze#J*&BmkvY<WOH*Rt)o7+4K_X-%D2I<
z`W|KA#^6&nov=@>uh;l>0d?X}G>)q8dodXM+n++LTmx;Ku-bmO=0c1rV4Uh51KXiy
z=!qviQM4?#J49mq0X40=KKX$V|8)14G0QAx)iV*&U)uK>y=eE+H1^g6%QOOqT(8&q
zpzYtjO}(f!RyVy5Pq#k^dvl=uIF#+xl^dd~oL7HY6<m(IQf|lI7~9UiL2)DBY(Bvo
zA9DSK_FVjsO2_s$6ffNR?DdQ?0VoS~ao(}p4f>j>{WzSBf391kK3?_9JF$7A6D5(O
zoxpZCrND77gI!7L!BQxCKN@ytx07wVi4=lXb?P{PZ?v-aanU*t0#TIbuct~gPmVYE
z%`YsO{LENeR@y(aCPHKHXF#{`elYW8buzomQC;y)nS92}u}y5So!!eVKGSdJ!QHWr
zNTcQx+E~usQs|w%@X}q6-*2y^&xy@Y7ea2u-xoSfnCmAVtKjmVMv+qaBi&CT*k0_1
z-dX-9?d}_Zw0$Jz)%bTxQM6{4v`9{g{R;<6|1R7)Q89c;^dGNFD1nyu{OUyAS<z;-
zryR_8t8Dp}-AKH+PaN`998Zyb?yLEkULPbUeCybMddjUpmnPg#{IQxO>hwT2#~&5G
z-?6K0nKVyw?k<VC<xaf&R>AAfawy(px!dJZ<a@jMm(x>4!M8eF*^8)d&kABXy`OHt
zZ@V9i<gk4WCCiaYaCyl-=?kK<`%S9bU9dlaWtqO?zp0(?*}s~68=>u`4My|T0OQN|
z6c0b)r9~amy1tcw`C|g(!{jM~qI(*|695;b*;4>Y-WnyH%(b$otp|R}(?+s?iP~?w
z(%h{(6Z@BMm9KjBfWCjC<>ylOTd&QLV$``An%?!pF1686;jJ-`LB3EQhw3Shjz57V
zl7IDU2b=Y(ihCzg*T(-WBz!%Gmc8o-#p}#a7hR`cxF$N@z9sS|(xmC){L~AImFfdF
z?a0sKX?59Kbzh#r<97_|a+6OTm(qTs&eImckNzX*C;t&Nb+r@qdpDFXmrU$L#lX()
zhc9X0-`>e8*LJ(GYt2i$HwFqAZvHGU{8VM0ElB>Q6|~TXqU3IT87?!eF+*!)>$v#R
zxs83vE#B?il)*x(RvKf9bWZiNgTLDMkHQyc{f>QER=%NS8~qP&tQfVix1846`df5;
za`>44X?OC)dLya&{PKoS<@rfv2j&+>ha!$2O_EQ>-M6}@XhfaEGcUdzsGI+rd0&yF
zd4v1g+UvbR{V3wen##F`?r&42*(G`6lRi=i+ILA$=23=ZLPrXDN!zDzX=aWu$h&#t
z65%{1<Fpno*Lx`RlI&Dzu`fyggP4duGetm2Pm$ibz5#W<?m{Jsw;8lKynm=hg5xTa
zd`f-7qOWwjC+1(Dw#U(X>82cR|3flW<I8Wg--Bw|s_(6b+RGX|^>=Bs)-RX;A0!0*
zLjoH+@#I2feV^jL{_5~{Oa2e`-Ycrf?%Nwp2t`182kE_wAiXI8LYE@FN$&&!>5zyt
z=}nNT0wN+HAWbPzkS67&2}p||(joK~lJoHX_dfgL+?_GLF}{mEh8GSqeAZlZ&AR4q
z&G{>My?2dYC%C{!3>qjm*4R?C{;*nb$wrVB0z9GRe@v)K7@4w*&6JB^-|QgPrmilo
z<2E-@%ge{f7TDV+7~&=L!6U@A0*ljR=<3Om#fRgg(~oHOOZp-rU-fze!g_J;|7Ohl
zpBirg&!FP8kg%2Sk?VW-fU=jcdd>?;-01;6s!|OibFp}FWqqcg&MLNenPBLUki-Gl
zggqZ@0c`rU4Zu&G{~sg2|M$ND9AMRoq5445ydJE{r(-e=j62t1Rqc)OR5#cRCgP*`
zG=R#nw*Ut81_J}HvdVuuWAlh7VRfQho?K#t=8zCb%iqCL5<hS)SoZXlJunV65Ey^X
zM*)mDm-(u{`|lUF8WVt3IDm=eS@A%_an=WKLI1;efse7sbTH7%22O(UL`(PFR}=my
zd>Os8oqzhE?Z!XgSp_YitEV^eE-DvwZ&o#efQj0aEDy}-bQI)L*MDM1q)ZSAo@tws
z6Vb`S^#H@r$1zeky)2n$xaT(xwA~#nz?}R~uf8-iNJ8oe++6#x2Ilq`Ah_EJb_Lpr
zI)&i1B4IO7!pmS3o*2j2DmaJd^qySHo^4h_#t^A$M|!}JD|7qt{5o0J+vj${=J9#n
z$W;X?leU+C^nKL4GaPp?rV7Fq4BB7O38iz*@LH&{ho-GBjEY9xIKM5$^V&rKL*mGL
zyQ{MW_(-qg^#{)W<?{UsD$v^G%+GZa1cdx@cwlg}CSaG7ri|dAmYCzsowK~{;^M6j
z*8TwV@BY6#({OnVFu!Wgc04{%**3IO;WAt=HEP<FWxW!VKpO^0e^-xh-`<L@F?*23
z_we>LD<V_y<J2D^>Lz2-D$!Rgv(w#3(cT&VYKckBtnx>X{)?ar?%P$<3>oE5$AeIK
zjg*_rI{|J+UV=JVieLYmt9Fhgi1O%0%GP8{`FAgfpq<||?W!~RbT(yxk@Y&=eP@!v
zNOn*btFgCniQ(Cek0kZ|4;Jas38=O8D>F^t^yB_pgamF(gEGw0k7uV(={O0ZiIu0m
zPG72G6rEVn4%GKz1mKJx*)hb94@n#xviMl8MN$h8$)^Fti1Kh{qR7PcTs)5yMP0TD
zS_a(&eFePKkU;057E$Sn*o!Q_fop<q-vRdizgFRYIq-ik2ZDPUXEgw0Np-D0fUMF1
z4<jL1?>bQe9}=t)n%I#7a~l(dwtQ1kHcK1I3FrUF8hT5fN}eNMW>D%9ytNhTGx6<G
z2N6S}O0hlrEv#T_GvXH5kVt~Rc8d(%)o_?rF<ZJ~`4@r**^(=hd?1tnErRs@4vMm5
zhEnjoG5=YX0=a;om;cmvEU!}C=aDXo@VuL*bkkC<H(;8Mf>5(E?J&He?;GRdMCsbd
z;hK0+rBzLP-)s*B)}t@wr{m=IXn<G(9R)*K@nN*J_q)vUM=n@792=msS%9T%Pi{@}
zI(qw|_@we6-cJR%yEl7{{Smj)43T9#Y2#?Ed+_A$6z`~N?~L608d=aS4cBg{BrB*O
zDm@#aMt{=wk{O&Y8R7oB)h#B?*-HI*7$4OmH0tz{7tCK3eZ2#~q5!;$N716!RBF4^
zg_ip5^xLGIXfc{1{znIG&++0G(K?Uj{7~6#JIUmAVLIl~{HwS6n>daQtx741il;n<
zr{rckJU|{NDjXVZi)_fKh6BPgK@oq<A%*rZ?pEJH=DgG>wc2!6Hu;)z+en~p*8lMC
zc%Dso%`++sXp>I;@p-G%7q&C5Rkj+1p2_VBXvw4~Qcmi^V4JZgb;71`jr4dGfFu5y
z_RYJS<G*|&>0<+3HCW_m^X>YPyQkH3y4qZ7J<}{Aiiai#OU@Q<7UJyXKr)vclCNy{
zmd)@qA~E)LL%!Z8Yw|CY0{hxqdQU_j&yCg`7G+KD0VLtguf0l^%xeWAGrEBrxA`xl
zi36OU-1ocWK=mc5wHB5AHx{718knC>WBOK)@&`vBZP~<g%7IRtqJ?ZlZ*&tq({v$_
zL`j}|Zx(K<ke(Uro;Q3up@>@>{oC1<-cLo|q@A-Gh*!-4zAoiNh5o6yX_*eRoemtW
z1l4&>v=xqRN9ZF~9*;^&`YaJ*pC9b~{@t@@C#u$0zny!Dd(S#22}v93tZOm8mvvFR
zzjU@F_FrO{At!zEu0e~V$uA8T+h1XwMP0V(rShD7r{goc_B?JUW@@`oA~a^VNRMpp
zWZ3sh>VXi7=FM1{+1i)JhAg|lzG5;gh19qYMdsznE$h)F<@>A0>nAr*Zc8g-B3n`F
zsZpvg=Wwx#Thc!Pk#%3?IP~_x2ko}2`J&siCP%M1=yvUD0I@~Tvw<XDDZk%}Gh51u
zDQd0KLzt4yE$UW0gNZ&EJ;E$6`%Rmw%)Kq904ag$z?~&ZOTMUi)=Q52t54yDWXET&
zsViw5&iL__M+FWb>hLAdKid)WwI4^{u&&|0%%4>Ez0=@T>vPiwoBth&GAtYs)#-2p
zv_|d+;JE;KhLq2M9^3*4`}xg@ysuoX>-UXucr-U+bxyM~`A7m|_B+@swWr{ojCu}7
z!sJY){odOPx8vio?FhbH3eb1$8e+sL!-*BCEuolX&XV?O|5vS}&)`?Hr>%OAE}QR;
zTrJL@DE)^7EdzqiVx>Ll$*zyD_>PqTed@Bh4%C2vjq98kk$Sgzty0aw*7;JUZ70Tb
z=E)OtRp*pxPVTqKYyN1;_gxnTN}2tFm-@e()a+*Vn&Ng!3GFl4WB7CBx(*-rt+E-a
zYPxpo)Q_$92rOw-Q-GusGxM4_8aL%T0cyM-*@#krj&Zlx?$Vuw?M7}@*#|wyB7O--
zg+I2=4G2sCasK_sfAZ(_Oo>0J^;PdlJu4{MVD*h}6V^e*e}2)*|N9&tc+viyhVUsy
zM#Pkh-jq_G=WS(Pb4(L_alGt>!&=&<d3LaW6BzyK+kR?E_3WvoKKP?Xsxw}-<%%l!
zr|YLu=dzQxkLg!T%_QiBbm)-(S^YZD(-|+NuT<B2FUu}ZcGk;$Swz|{dWtPXF)3PB
zTAQHtdOYf~LR=h9N#6b)OqSa$Ofn<nuQcss8CgpFndP9rt4T50QVJ-r1=C5Pc2lH1
znq>3C4-d7zaQ>eAPeVXTZmwNCCRX!P=r%r2OBQ@8V3GmOQ?vsWwXF9Y@|PDsjGH9|
z82v=s*yXD=f?pb}u6N);vVd8<B`<M_BbuFFPTly$x)y!s^eSiD*qhIFENA}Idvhj+
zI`xFZf^PcVWNU*8-23+oHu1|`V;x}4wraN05!sJj_egM5WpZ#>)_WFw-?ivn$4Vv^
z!6HPcE;`dnZv(1Qjd}UJv<S+Xw{4r#u9UfDQ<U%OM=RQP)CLAmL~pgYr>>|L%-OTf
zHrxDets(E@Lk+j>l=@)B!5wylAE!=3$C&2?=qQLQcIt)(3$GN?1i>St>@76%pGTbe
zJs+^(WB0qH{!7$u7khy$!I5u!alELzW{dso_U?9-5PU|2%$;Cd+Wgjne%SA}0jrL~
zV+Gz8-{4xT?4ap{AWhkK+12J4fe>5O5X`Bt<a@bd7+Xe62X^^!GTC0hhxkl;S<t5g
z4sOOhQQtXec`_`~80paOYoq<gNy78*+6@3DZ^4un5o4ZiP34ZzDG>VC2K;M9fkEm)
zZ@`WT+vmo2Pt=fug*C^sVcsgwg~5NB(!?bO<zAGL^}X<uNOt{EHFOzaof`GiG;F1m
zwv2ggElqSW<*iwom#E6~u()(FAIU}ewc`dbs30btI=Cx#bIV%!L?vz|?R~wG3Sbs%
zz3cCJebBDy0_(6s0hbojc%4B3lCQ2fcAcn1CbM3a>R-v(lnULN(}jRg3M=i4CWE|#
z2a8H00isnyR^(CzJB=&1PJNfP<oz@L^`88JRoE+GRaKoGdfH*^E$uJ0rhN)|6rd~t
zk^j+ggV5}7J8;!#F(tNynPJ*Um1O%SmJQDI@nyehBsYfGX{=S-`_?Ioli7+kf-Ny_
zt%NB0p3D=U(qc(TqD6y0RaeRw$CGd>0tY@uok()2Eck<_3yxA^pmIZSCLUl{%EL-Q
z$^CY@@@^X7wJd*N_d_!IrUn4z{*Sy}@A7~D{2%6kO3u+|0@evLNBcTaZXFE?B975c
z-31^GEnk!dw7#75D}Yrwo(9rix0BC2S$rIXhJ=0_!J0C^0cThhbFB^Is7A;uH3OoO
zZ-p^Bn}TYDhC9LhGOw?H$N&4gdg0_2*}J!4p@)(>S8jdw7V2~*Kwx$OWcz?U`!vu|
zi`X<7R1?<OEU7}x1J?-}!+@h(aKE;(bkEsAvA+WxAN@<-UsqJt|M&NWOprDt!PMbb
zg}B!($tJF1DAwLrr#yd#E#K%*fj#jD@K3NeQk#4Bp99g}y^BuN4Y)T!p|wgS9|iMD
z7i)$gkBR#5AcqY&IcX_20x6g%-@#}}H1NK;-7DzM4((C{YiRS)U1aJ`D8;ozb~ux$
zhZGS-R>iFpKYdSJF@Y>b+9efU#6i39vZ$Y^ocR$pL~ajkq7u=RVNYT^;k5E?AN6(i
zl%x`K%xACnZ=xFvTLYS!T|y=r`gk0b-q3aYVL?mrxYfl43BwuSa(tfMb!3f6l<y(E
zGwuqQZo*&Vvs}R9RhQeK^|Tdh<;BwM%?x0iIKj4#3@cwx;zV)7+Ju$@T(>z&$*Fv)
z8+8+1k+DcAhWzteK;)DA?xiM*&E(9Ups-LrkB@=}n53RD!pn+tfXV<Z!g_whO_p3q
zDMf4>1`!W&E1p~255EgHRgpd$PrL*Ivhv63>f*_WmU7H0;m_&)x7RXx%;GQhWrY?!
z37iMrLq!Jp91m!}+<_6d3CWNr+9f-QGFBXE)40N<5w3=zgOzxqZBIdxBPvVep}}No
zdOX7jK7*j%n5hgu()KLoFb7?VCQ+%tI{U`PmA(FwQsy;%$@o7nj2#qA4%zjb8}%3o
z1S>I+l@?%anb2z{)DcioQ+zu916m}O7d)LW6rajS1CbDD)3iN3bOT+|v^~?4r4GYQ
z)160g5;PB=uyErzhX~%%Zak#~`3blYZ4(z!Y{+q0S-BI%fVW|*1TPY(Jqe>N))e?W
zGg$BsGe!s&@5!L^e2m)mV|Z6B0c_;U>ZDK=g$RBA<zUa6VxUJ|Vkj;gCU@^~#P{l6
zsXn?!3a_vf*r{|fGvV!o!MnmIMD~=}w__n<*C*FvYg?-(PJB?-B5V2*Vy+>%@Phf_
z=U2VGv(OcMI8P;hVs<4AZMYn=6%CVhExz%ETJGeMS`J5^6n`m>SVI&F8C^O+BFVE7
zHs8BWuKK)p=Pt^Fu05R`wI>!EQ0Y<>%~t$=ruMpm6eHlcXZ!os8{=qk3>QXdY`-z<
z3?C@X$JQ7OO#ECvHiWj@LxWiPQCGA3)WtU(M89wPWG=~3dxIEMsIaW95~Xh>w{FW9
z4jM)x_5}8=?ici5`y9{v^?;twf?i3LA0k>z74Foz{Mg=?&n4fLAnGI;D3BUaqLYth
zP68mJ%$f4J9v>vxzktucd?C{h=RAt!$@5D@7YQRplZeTy9DuL*N#l$>HgKq|$3zo}
zT36bMG-twNzj1Wvv@xa=11gZc-Vi9Ccw0S)SebA>Z|7*<r{w%V<nK0Fd%>nvS0U0H
zA)56#vMw?=pW=*C#GS5|q%#K^RgpA!9V83ngugA*+hp6gi4J<thSy6ztsUt6%ZB$S
z?o@cPqNaYJ43`+Q0%7>zyi&<%b)FjPo?ir&_re`j?vAjMcIwGo<fVY>bgnpoNSAEG
zhlj{$fX}IU!2m@zxKT2ae56u#7YE^!122^oNvubsCEn0E8X>Aalt$6#0RuvZl;Spf
z@wq4_+Ou<yR6Izh2fvqH$fACeEHu!1R`_2<PM<v!phyWK4CVd!@KN_L6mlsGbyi<w
zK&A6WsjFlKqgoQ<2@{Evz9bpOT?wT_iwWBq$78n?y@N!stq+v*JYX4oo&<VyXg(45
zZ##x5*ipG72TGmdjo_>=2!C335Or1T3-Fwg`woZpVKtM1^TICv8vvm#sXAxrk^?#D
zznOT1wMaUP3V0oAe-eGI(%0!qITBOZ8*4|-=0_B*b$o;yMxAHa%AN>)v(BREN$iSD
z!&N}NkR9YooKzdT!NpCs)8qOGwygdz+ZK8)lNo_js~f<Fo<fi+k4HoJ9`C7@ihfCs
zXPPm1LEymqa|;~mC{S(T`=czO_R+-?(2tq0UHtOi>93)33RF6syqa*`mg!P8iWA{Y
z{{6Cis+9C!Q^O6=^%c%n4|80MMV&+Y)B~i@<Xmm9&-3(TX?RZ#i$PR$yu=|xxdsG=
zaY2QjN65N>JC8_-P0KDB6YAn@71fMy0iHuV8GKbn6@kc=$3)VJFC$}ya#ak=ah@N$
zIk>_I&qOd-LuL6r%V-5}*-;-Xuabo2_aYXMp1jXb(X-*ZI^}F$KflK*LT;7zvJzeg
zCKvhaAd@9MCP<way{b<6l>&9>qepYc8S^t2TLQvx?#wr@T=+93L!)bbSTT$+_o-lW
z%UqPRj_a@3@NK=eGH2F$=DACgPz`JQR?Dn%)l31D_d6fHrA+1>hv&eZ_MV&YSwYSR
ztn%LH^V=`PO1{NsF0|t7g?(mPbL^UNyh_TadN^$SQb3R1*3O5zb26!jJpz$wNIu1_
zg~*QUOf@(bpwL0D5+cY*HccINv1Y9Bo|=aQE@k0Qo<#xG)=)M0N(Ut?VQFh<>-mN%
z)*n8ErZ(uHoX~Y9@(Y@z{5nxCMvg7>hcBU@OR(R*Fx@i|?VJ*igDeg(*<UafMF?gX
zKI;Cv(0W?$o`1DXYcK1nVk$`INwD;fZqdk33QH3qRlb@c7NKkVH#!66I&R4`Y4bn}
z3{o6an%L(sIdP*nT~becb3QyjG*7v(xtD_d(S=I`XgU8*JZF(?nShQQY)eZ%0gc>d
zM*^pJ%QjRs7FKC-z+A6tIMI+FB8IAN{j)y}{~87QF?&9uJ^O@|&lGu;M2Ya#y=+Fy
z^@#Ok5y}hKV3q%)b=O4H1SyVzGfvrHFKEK>$>)2c{=xLW*Rx0v4)kQ4ya&u{Y*;3U
zcJN!u3={=WIg=qwpPq4zK!5u5_@0YS)@QL|^6e7Qe4U`lHM>RIMLbyoflW?H(O0>=
zTGxvZzh%TK)cxwY&n0<MXw;4JOPNb2QXf#XwU9lmd^H=8o5_T2P{zgdYqK<*1H=}x
zu?SVgwpVkvb#x!M#SdyCO?kwPr0_UVu7yU^T$rOX!89E?X-x8_CTR$fedz_>AZt>I
z#CZ0eDw}0-L?=OOSuPXIS?lYgxMfkvKjrwr;!D5f!)|5Ay${<>86%!2!E~*qn?!Nx
zK&g=AT5cx=<Li-P+<p(5g>obbRc}NRol$=@&D-K%EuX03FbH{ga(s6Gyc<;^-sN;t
zqewQq`@EEqXrxkd1{OphCbAt#R(@z*Ghvz4e>aE<E-_UMx&#Tn%}tfeiU)>j+hwIu
zvxE4htJ1W)JiFKN-S0Sr;0%0)$U{Uz;v}+#&J)W>TMW$lnsz_-`8ir;%jt0k{=5nS
z5LS+p_F)HNBEYY|s74ym%DIpb{6MBdVo0%~A8*Qkk(<iVV!Ij@O)`GQUzjwhR%jho
zFP4B|L8*cKjjGxD?+g$-O-8BW_pH%YB5cDzx9(@3R;og2Ela1W@@dz((Cda>o#jYJ
ztcBbT(a@mE5)xKuF`hPtZ+vG+7wSqY^-|0}CLwWpl(q{GQi9CUNL^04DS@**u=udo
zp){EHZX-a!1E^ueS3w+=(AGTe=W9vdp?Jo#8*l|+K(kk<Y-S&p6|HCWa#FJ_4Yx!V
zsuLX6@P<M%{3hn<_T^4eZ-X)4D0u1Bo!qYzP~BHMk6q`CRcv1KHpD0TgR032(_qyT
zyf9V?U{L8?a^O;);^ZOwabC-#BYY+hRvu2F_C!N;N+6v&J^iaWN_(e-jMTzN>x~aU
ztGkv5Phi&PxNEgPp1B<iiWIv1%!Zi;o)5f)LIA9eJlfDHXVSOl{@YOy=lT=PyGLHB
zxULtLXL6<Y(ZX-`-}8H>5Gc1kz|PNT^Bd7n?0N^r{YW}uwOJdae0qd3-?B{S!`XM;
zbJZn4IGfJYu?1&L<^khK-vp_yQU)i|oTmRNspE=XDu^M<*9e}+`O5Ova)r^J$<~MN
zu|rez6Punj^SO!2Op#_l`|x2sYnkmU6PuHWYl#eZ$yWr~lSJkX3ePP4=#0+XEPUI9
znnutm*-LL2jc=G4ZFG@^(tPxNjeX<uJm#>DQ6a*?R4DpsKa+Br3sI)MqjN1r!jZ|z
z<<S6X#$R{hAbkSSq_MQ`U2EAB9@y;b8rSCA$X)!#EU%jL?tyFyE`}(#?Zg5+o!S|3
zK#k#D4!P0-eE75c{X41oG?=#nO5hSfT<1ZlwiZIJV(ja`#-s20+&OZ2KXUn*%bLy#
zWt@6EDVh|Q<UUx-`@qjSh5ztzNj!tTazSnd#v`t^7^mY}hGWihMNuVo@g}bL37WX)
z;GPIT{1HLA$NM{ZJv`r+D()<Ny8*Yhe)$GA4!@17)*GjBvP;SmPs}$v5UNSXIRHUJ
z|CdN?Yw<j(t-h-uO2#RE9;ls-k^Te!0KlfZjmW+?CK@8r>WBbF@_G7nPlDm>Q@q6?
zT(w`N%5DYTglh<j^5<5VzAL`Odx#ujYb(Nkcl0_?G=tNG=p9N?<Gmv}^O}DqB~m>i
zDO(df%<>C!#kt14f;0(EO&6G@801x&jEz`j0h#=<m=lQ82)doixU_J?b{^<AaXz|d
z1d^Ky@i>TB4%F<Vb7ah<6yjC{FOqExX^G}SFFq_WJP#pUbehk-K4)*>oK+3x^P_HA
z!v46cq{CPq6JQ3v;RLD6YVp;{!ISX0<jpbskEE*eur7qlU2)io7M@SU^u`Bb3E$n6
z0Rxm#AdzTdX4)kUxi`Tr;tk5-;;tmtui|DQIYR^^Fi|!SJ_8?sQave2d@w2oq(-P(
zI$YpwO(u<GVtgJ`oa6p`YQ@Bf@qQH@TH#H0^r_D!584M-==rHDWm6Lw0$D8nmg%|b
zP6_Z*$L@|kJ7|rjEJqC{qra#(58W{myLBOorbWll%H3AuH>oR{OCIzRVqkjTfrOFf
zv=p7B=&C&1B)W)x(et$1K)056oIGb<yL@)0v>#Fgsv9uj3(}<JWv>&Vm*i3eyG_bN
z$9t}4Z<doI?8xJ$?=>)H*>GFU#xq&zyYgPp_4Hp7^A#p_B_#-l5PT4D`<|n?!36S8
z2)c|Pxy4SSMo=Q=Q=mWO7HPq{MR-l|=9@>GAQ;fnbOm8VfQP2&h7-z@pN;y=snPP{
zI}K9)czI?%tWw?3ciN0v?CyCt)&9fAfHR{v5nD$7Jp>%`>G5KQEd<LG(LmZ-*t+ka
zUoK@03;y-m@RCT1jmuvNJtV{Eo+DGwf}f%wp)Zk4aM3JuD9gVD(9lKEb>M+;v=U0X
zXYV`}W!fW#AeuF_0r|&%I-7ObmhMx38dz0q2=_+&3K!*@zm?4a1Y+_YAg-W*>?3mY
z9wOb?9_!DGv3G)EC}*c9l4y)g<hJ!zmsWJ&o8p!A|1|qgW%hiF;om|e7kES8@mt+b
zc@xJ>bmnmqUU7Q(7}nxB>OEy#80w(*`r?<b(GX$2c~nk0Y42H`;am#_d-0@1*`~$)
za*)T*50^5~2n5(xAhyOp6|MGk@y6n}1pe_srST&-s{w7|q}#el&wjIAC9bBR`U^m4
zxz<e^5>)6<+SCu;5q2~b#aHB`c_zhcr!BtBS8CatG|#$po#-9e_$@$s4T7gM6gO=k
zbF>@e_T-+^I6R*e+wU*xQuu@G$j0=`LnfW5fwMZsavRgx{EnCXjZ22#S%|4zD0akQ
z=Ab1(MLH8F0W@b)&a(=Z$JvN!jnB>+?(s57-;?m!zlyoZks)uV_Wr$*Ei?cy6!Fj}
zhdsY1r#BiLv34>6<$=gzvN#))^DhVTJD&RT&jvkDs}nWJlzv-4YnOJA5sARmAY^4y
zX0v3)yv{!Xq#ulUoBCMHYGXdqVP4R8=H?%0m_)GP6Ep&j@K+h1KE5@qkckd?4k7aM
z8J*3J;O@8~FD2Sqn$Ob8S^xJRt>x(7;u+DTSx(XT`cV#PkOZu(QxI|(SXc?FEU$m^
zZ8VEn>|gUaybWSrc+s=$_0PpRIs{Vc_xr5;`?N)j4S-X~@(>!mv}4o48@&!U{$1TT
zw$Gm7^Ylm=T?{4ZT>)e%Dls0R7JEbqE7=^{ntn&~LGO1V>SfP509r)Im17o8D(+{=
zyh_1)nbkwkC$gn6a)X{>T&#n0^4(l*vTP|2tjpHYd&J_e5{<EiNi5GQH7;CXUKPhI
zmkj8y^quNKGTRn2SRThh`pzt9LhSXC%ibaSXXfocC7Oe2KKr3;Ac5xPcb(5**`^0F
zaPLBU_vHD*;7<zL0ZlIatD49?Zolm{YF-S`WJBu(VpnH5Dkj?!B;zlM{>Sg9rpeMB
zT(#i_0Z;L8DH-C+FD%dZ$*7iK$4}Tggw5koPHxqh?P5_IP!X}GDI~GsEP?|Lo}###
zo48m?zNUrsns;7KpALTd%1=|-C@<0s(3yYEu$-xLdA3PUZPDv-!D!FVnn@{2?@+jq
z67s9*`ceVC_`%1*NoIxoT?u$BK<CdnPO)A+U{T6*l_&XMuuZBE5QCBp1nww?v9_I^
zO3=3mUm4kF5v<X9Q|BsEKXlBqoi#)W4?kIS_lDR~MP}x+<cC0im2@dxTN{P9AXYow
zawpG(Jvxe4$A3YKlxGiKOBCb-k;SNu!5_5mREgOWF(2Yu&+f3TjPi^Usv-@UR)6v#
zeiQH&IFzCMv{~BWv0c;CyPje<u@ZyH>}k|hh|Yg1Mrd@d`p)XKwoOA9caPMPtFY}j
zdW>VC18=s^v!e)<RSw(o9-nIU&<}eyCxg;lR9#^-f2deo9kpkJo9lO8l#X!ZR8S;9
zqsiqA_#YLy2QeDEP9kg|8SV<w+_EJ>8o94VYlAO(D8hbFN-B5pBi(<&7Zv%dxjR7e
z6s>IX`Tc_JDw*DA%sl++7W|kX*1F4H+x<xNGx}&B$g^_U+c+c34P_Dc)QV4c<<vlc
zEnJ$MoyqxtJhDM4C4NE*!`a5x@wZkdH6#;`qG2!YUg?g}#7M!tYC{FlT<FhkqoM!@
zt4yTXL3y;~zW^?dF+)w0HYOuz%68mn<z&*S!)*M!$GSG#ixVTy_yds-+t$22x+&nS
z$qTz?LiXdEVQhKxq#LPFi4|8U>#O1<D=^8ua8pqVQiEwSKO<g)v9E2*SIEl^SOJlz
zBMZJl_xl*0>i&Vx7sHa?p$^Ts<-%XVxG{bsjSF71zLVRs*7Dv85tKx`Ah(Nh98azz
zil_1uBqrW|DXJt8_a!TzNcfs!&EX(W`!i;#FE1@&hk;Pm^kLr*7M1!Ws$`LfK~3{y
zS-R9)oWv!2kxeq?&o^RZA?LPux%=a3@$Z~=@t3rgWATTwD;7OM-&T9xBuOXrrdFVI
zGNL;Tq&ZlR1K(uxb#Zi)3M=P)l4MAfSx>>6R5;=AO1~m;%gbFqvL3URQ0na_96d&l
zHau;DCR|DM`~p$OG8?9hJT0%XA|)*s#-=dSKd=ssP{fE>J7@s#2_4#DaRTL1nBai{
zItcBx?DoeQwYyVoJ!b&%BUXj|QB^DS7@CP~eL2f6|18vq34*6+jq`Yw^Gl5Vb;EB*
zvZ-^tzu)=8<-I2P`}HJcc)cw5#_1*f6oF^!pnNhPqFZCx)?bO=yndun8%l<`nEXCo
zLA%m*B)H+@6mMrgrwAH9(|>SF=${=c&n}|fA1X1=6O#@DU6ZQ}O(!$>OgQb$h3t!Z
zm)>1#Ipf`lk4;G)s7rDlx%Fz~5*QzAxxTB4u6t(-|HO$E58w?@9z*OiIQ?1ffGTEP
zCdzOv5gYRPeP(<lVISP<hf0}0oHhH4iz#CEoPK{*wQXs6ue^Dqf*;E9`yr~k$79~E
z!B_jfwSLGy45Z3hGPqS=?O;fX%m;wcR`r2b+%WU0`9_ng_}u@-0w^e%z7&8%b{rf-
zZw<JimAOiPu7lsKI9u0lhr$kv35HwaVc&B@?z2J&fer9Ay_(9b*+S?c=n{#CCKBT`
zyVT2v&nV!uQn@YDBg7+>pOOADEg()3T!Zu7@!NBaT2|2)*x_NrN=YHyj`+sax8?!G
z<roOmy(fIItyLQ_!Bs3h__g>6WQsxz7&Yhk*~|qy{ZaHS-ZNwc8y&Y6BA{Yb$`h8B
z&uyW5-1Kr?`A1CJOWhA7x6Q+A>K*t;BDix)jsFrz`kfR3qL6j}i-}DNb@C>=R@moV
zw`Gk>B4`;(!8g3dO`Olql67Ali08FD@2{pFrm!k4jK5|2LLAC-iTf*C((-T`Y(z^=
zzGyIHGIMvXnF!+s`=Yujxpk{8XRwL+?}Ik=kL7n(y5b)wCR@r0)D>^U2$(K>{ipk^
zSgh=h=ICLMdoufDS0w`N<|7d)QB1-8{YOu}T|4m_&kBemnN%*09OZ>RMMWgWwWslG
z9gc864tJo4d~tB!sRPa;bw6M*RLPCjX%fxAUY!{~C2MB5<)FU$Ci(|gSb=OmBRN(K
zQ%72qcQou&JRsn7&oOcA+YdxdQf3cmjzG}8n17+h%C-@>Rg*LlonVJ`51rLLYN5*A
zIOaf>7bhR?8m^)dYgI$vm&h#`H!-(q8V?&I(-1*^$#sE5xm)`_NmXap68b`B0t~0e
z1oI@E?jc&}OrNClYwdPsGT-}>#;h5^4-a~>y=7hPr-ysVT=j`iK4WYt!ZYY34hrd0
z3#oX*D)9YC&Gem5zJJ*MQahpSsemu4)sF<1AB6|i?|vnJJa<ZDKenUmTFYfZmpjC#
zC9l`L4U}DJdp_i~8&=}m#H)&4{)Fi04s8y^l(gm@u&Pq?9;d%0(g}KEYayhn?7Q^A
z6%a$ubs6c36Q-UV{2@Y;po{2dTHuh@^56A%0sA^TKae3wC)n0Vrgf26jOZsdZ$2YU
zS&CM}jyuhskym9T3Cc$VA1ym}tceTf*WvD<54ACe3y#URWLSP+Mhn3>6aF;U{=8YD
zS5}XCXS@Hy<V=H>x1<8VHRZu7O%1Iq#z&6SCJ-HD-=nfRC9Q}yMv>DlL~Mx4B+6!!
z1YC56J!C*by#1etC_N#$3rQ|FyP)I7>+260V}Q9BFDJ95P`RrxVazuH`|5>GgK{Bx
zx=^}+z)ViprlfwA3k|276}s<quo|5fu7f`>#0WWtRJz!*?k?1rO7iP*cNDJE&l1k-
zVhvPv6o^m*8LE(h0||>=5j3@vi*yf9f}|iW0oL6Q45PJXIv-mX1@evF!#$1FwZlrJ
zUtd@Ky!y~Tb1D%yp|LPL-`Hzn_^on>0XGU6G2}zfq4-GS+OvLoKO+$)(iZ^DW(jU@
zCnN;5`cqvcrpn4LzE7as6X$zOlQPYv8|t!?Jn^MKRWP`hBj=D<u%~|l8+6h97SBSh
zCa6WKAtSbE;HrxhMr~yHO$_V0dS*XYKsgkBYxn!9vB}#TQ>Az{TI{PDBXu<@el@jT
z49r9OfOXF!c~{i*F>$@$<NMyb$fY@xmjVEmRmY^bU<1_3Apf{1%CP^0jc_YTtM7Pa
zN2uwZ>eXl@dUpP>38k`7W2gozdDZYbiXqcYuqy-b2t%@B5;O0%61BRsXu`kbS54~m
z#!$2c-5JX)oo=Q?(7S=ZFg(BH1Hz()=j-i~VFme`jT*Jv`P<=lPsu)JZF(y$;yD4C
z)44l&`4u}kf6CkA6m65%;7`z=ULt8Oa9kMlY;eq*ysGGD3>zOCi=hQUGICr9uboJl
zrIW)1TciyYEq93j)8=O;f=m_0vdZ&Yv^$XHxsWsH(iQJJh|I<1tC%ztJl?&0VyPeY
zf(!MWki&cBe6A{=Y+V)2_2zdpyOhl?HBYg9W3=9?{?lSp%bX(%UEpx|s`jSlz?iFV
zhaoQM^^vXnn=raxBhdGxiSeHRO#<Ie*@(F_AOMpj92sFy)sf-R|B*XPGJ-hr&iFGy
z6*Ml6dIU|NfUXSP+y?yvTu$h@0_q!bw{R_*9{5WXwv9`tu{Da4rm~d8oLZVsw-I3v
z-Z03JVUVN?m5;luDq;xWtA7iiyMdP%Q-uT`m-EgLOv!z#{dxwB<WB7e8T7$<0Mmn)
z)@}^B$zZD2E(&np*Km<3-m7NF;=HaGGTw_pntFs+E1Uq!Ev4NWpRn&hz-6%Rh{MaM
zGHpK)+C2=i9uPt2wLPtXUUb#~6;Pzbr!$@|apfgx@lnR>4AU$job4Zufhhb#v|QXB
z9cNuSkwd({=ISS$&7AUc_=Bsgbs^yjW}U>e!Q~q$I|Lm!piI-^NCx3Zg8ULxfnit4
zq&DClXOR744h<!i*k(IMrV>d#C%Ago7e;x|Oy)YDt=14h=yMp!lN;sJ_r`rh$||4*
z6_IQGhDZKh|2YPk#v2%=jY}FGlcm|`{R!mzA#&-h3KE`kLf^i=_&_8fLBFdmBcg08
z6hi~|kn<wSfJ&yN;#u7QQEXgZ$4C5X@-t#BUbtj>ehJ@Svd*G!%{T~Gi6Jy`oZdd?
zL!bF@L08<<1MOZEH~ty>y9t2Z`6w#)6+vz;=>vBMH^X3L=VSQ5q<MTr+#L%DRd9#a
z)fOeX_WrCr%34Ip0F)McL<IAaTj?^Za^8qR2t&zUWf`D+eU)Pp_(c)?v)RL_4(M8x
zr|go1S)Km_{H!zb_a3N=q%-&I<+bd4^@p;9fF!eH(W>l2NkVzX9%!au@YX>{0Zvd-
z2U%*eFw?~*MG&JA?T$8sF>zHD1EBNLavS3jZ>q8WC?hrpN4k1U)PMm*eq~!OXXQk9
z`D$CU#}nbqUA(wGK-a2-Eal(n*aemvf57Zdfk=0|w9OX1<hngxMW#b%q!`<8K_0-{
zj_A*{ak?lK8EDLirU+$Qoz$UDPEpkmqqN<Wd1)v{hOH2b%P9XDMGx$>!Q%J%_><m3
zLfHz8FPa|}h<doA7;?(T#sws}yrQj4^3N>U92M%LzZs+Wfp<#4Ny_x}Zh_A7SY1mu
zp_=8SK2(FeXg4}?k{7I$KgSj|2?aA;7|I>^6~TYkWQPQ5*KRx*>}qGggu}WZ18#gx
zKxHyF_|6p-NGGKLZ}nw#&TgP1WAT3Zr4>XO_JIgPC?&nlpRt%xaTC~Q)+GP17=i7b
zmygQ?P`juI<n*2BzNa$!2OBZXZ|+=;E@b%JFN2kUJ9Ss@#=^QjLh4KI4{)HoxqVmh
z1?BrJ_`A??s~(|~_GQJpB9{uQ_b91oV>wRvcNGM{f-+yYJ_1XW|CI?WD*@qeM&`G2
zhd=W`4=vqXfayCgJ_tU7vjXpOC@VV+^V(MS-h9nSqfT|^x1mX%f02)Kw~f&yIWl*;
zdX}|{N=_{5)=SD!Cjn)1tBj^GRPChTQ>AfKosk?7hy>^O0xvytjc_SA*=iH*2`?i-
zXDKYHRSj-h=XG5cq9}9&u70Voyl_ZnB|RDs;egoRn-{uz=JDb3G25m*O(f_nl)O*L
z#0$Oy{qv4w=BdZcaM_5?%>K;k0RGYY;SzAgk9Ow3<XH;&L>NsFNv0Dz4t1kk?U{>n
zO{0$2^Zl-XUi?>!A-sRz)frItUlkW{!>EA8co>PJ3oNP6IAFH7Z2^97K6Nv3#mYH-
zTKd}KRKyF!g=#c%nrjdyTHZx^`}yU4S<ube$nLK{dxrfz_kUe-TTNeyP`r1!JNB-X
z&!RPrIY2q?Oi5Inxm^w;05j0$X-fZhY2*t-q9>yC1pw1@$ZQF7+;r#lB2Q)$m?J+C
zVcsuJ8USmvs$OSRH_m9JE*PD3_xR8)ZBezajtdu){>(F`8}%(|UI`k!14hH%Ih{9t
zMxiHr|9~951%>eS+>e!`4U0c5Yj?zOpllk5Z|T%X<K#xp)MlQPJ#OBf=t*Hp6}5t3
z^Uv7rA!3lX#yHzgYUST&8LlNv@zu+*`cZ<$$D>rkoqtVV4snkc3fV&Ep&VLK#S%&c
z_^b#&2+Hx-^T*CdL5>2MbsWp?n?m2-@}Fv4grT8M0biS-f+#R_5NVS%%lHNCe(!=k
zP*zX;zbH5nQ|=OFTd4ia2Sg>b@=q<PL%ki@cvRKMiU>Dit=y5<5=iOk=5;=eTuGe&
zyM<h$^OR5PZ%(I9juhvAOi&$=j07%6<mjd!Likl~$qPg@@Q$&-15YAZ_#AvMhQ|7z
z5J(m5Ht=cvq<Ad@w%5fP5#;Lz5e#wtBnjma1)^uFtW~?ZH^j3)-svj@2b;?RiB7RQ
zc8|@I0Lz|RY`=u~)5-DBL%o6*HJ5Z8;YK6<OcpXt=bbc9IFKG;p;jeu4L&Y-P@n@o
z3UE3aF>%V-`I^JjI!D)<<#Fq5RCdyc)#bcu$Hv$%C7y8{Up^KP!nzM!YG5utwRNqM
za_Mllk)2tYe=Nx-#oVfA#+X~3i=llpIQ}^n(WL{tS+@&`2pfKyrt8uR){q2aZ$_@C
z#AK{b{z|v(S_8q7rNsdr@iiGsr^dHHwCUvTM}SMY-nMF#L*(ht_%Pj1%pS1nZ|FN3
zpSX}oIjJpy;o9VaOR>y47I$J_8_{KAInF#!Tp(kl7``#m+e)IGU_<25C^!2!C0!Li
z;9Rnu%8oj$oSRu+0O<db5;0!IjY>KxMQ#(Y!JhHz_4$z2y}SzoN$@h_MK4G{@=JZw
zW_62rU|8S85YYHV7Kjr0j9TEkykw($y78S55LX-H_%KQbCE)bzpcZ;jF&C)B0rnx?
zFdAAFyL+8;7~8%&*)5rL`Ua~_2vP!KQpP{`P_~XCXbzr^ne&e7CA|0E$O^yc76cz-
zxW{&++o;hfBcs=N<4vMdPAl^AsSR9jl27p^O#i+N-5B;xZ%Tk$&`?UYlvdKaL^26P
zJc6KM!T_?c{m=G4q6<gi8xf^ToJ*!=3!9%$&8KQQRQIO0+TU;a98Ye%zpObFThhMP
zeY;~>P|BY+;!>kc^9mJ`(>M1z@TTR{^6z|(gO5w<PF;R*@uLBZ#!xqk4z+j_9}%LA
zQ7(6wxxsb!*}j7c-!S)bFS%%<u2<?PLy!L(O{mDL*d#MO8|Of4irMEU^w&zhX}97D
z%GbQP#6J8vpvYm<#FJ;NjDqKC)b<U}XU8eO1hc(Q4Qo^`m?Qa->DY$V3|Vj`HHD-F
z9fQK2YpCj4HnKRQ_7jcz`=ot`j#}lb(XtF*Z#|uGUD+EOJZNWNNg|ZK)8rp>9)^Fc
z`%8SBxq(_>d5NhJ$M!&vi~T6ryU(&bI_UQ8yTE8n9>J*BA=|9|n-_lN`H_vv{KyWx
z=pYABq|CTda{0m)L2xGU<<vx#LcZg8HTyI0y6x)Qk(xoaFQikuT(g~xnMZ4zU2kai
zkOS7Jr3yUk(=n;pWC6n^Fi+5KZ|?hAXz1|;ca1-fJvdTJ-M+#Tj>Q^29FV<#K?zJe
z*BXeoxW|gvj+bT7-(>8KIY!FlF08e+wonjlJerlxh$d(+Y;D%2p-AbYqabkCxM<Fn
zxXzSQjg26vKI9zIG>LN~%Rb}I^C}40MvVE3?T?RJG~O$C>*ut50#3{jXH)GX4%Aer
z?kL9%Y;o<RC=l|cqR#4~;mo(||D1=%%l)b5Zur7&N74QqCi;X+D*PaHxn+0@>F}xU
z-%G>2cY;L!ETUR4e;wYtNWK~8<he2;ko`M4{j(64@D46|)maPKtva-gl87KXleqn$
z`L{vF;Djnn1i0npG=t$+l%)(FQ@}+r)ZdqMCz7-fdG{2n?1-+b|KU0+6rgF^+@h=}
z79}`$-UrOz?EUxfEU4r`bNtUm{7R+@ab)RlVf{Pk;tWb3O;Xw+L(FqRy$Q_XTa!xQ
z`+XIpJ<mQ>=;J43ta!rD>$pt#{6^)5KxvRgpYC>G1)1n$8Bh@kmQ>2vLeiQt^7BI6
zgK6t5HA5aP0vR4WUKpzQ`spyLIJiVlvY&%M!Yx!|&;{L*!6<W>YZ|Gl3MvA(L(U#N
zogI!er~(6((_oU@k!krmC}&_URcSvhoOJ)oUy0|%0w*9tE~f(;-dq3T+sJ@bKY&!B
z*atC!0gay=pH;B2L$>F3@4sWKP(fZD3dW7%KVt*LUA(<idG*2eCe0V|Lj!Vbd|yb;
z!e$3qP<!`&l_g0k)8#uHVAatpq?G>4-nuKUSc^K*JzpGxZmjvfEL5g{=_}?az7gQ*
z={*t7m)wATtkON!C%_V+>keb^@RX=V#gQd3hc};3m1sguXFscXM+QvBzR1BwaF3aG
zWw>`PF5(4g&*9`xK!#>>4`zMCR)$6O%tyeBR2w!KJ3il%Rke>q#0}(RAfYkOWkkD1
zh8k-}fGS;`s6XiTmC+ACJ+2VUNpeZc`0uxP*dII%;&wT0seX2?1TF6?!RS6zcd7)8
zyuElcVp@WDL_Exo2=P9^3}*?$A3hjpE;z(jGQSS=n%yjv_Rxw8>bks}>GvwAHZCer
z3$drsT$3sJ;_HK_dtiU((9VsU^6c`rzkX-&@%^)<yW)bipwc^JQ4M#!5&Fs?Zo_m|
z-Rz(X9%JG`m`@_cKMuCL@e(h2O$o)OC&;oya^bHBniGj9#L}MK7$a|)4+om^7};3J
z1b4|m<mF=;5byfM%kfquA8J<0wdqur$ew`Zrfu%03g~4-8_E#mnD=uMV!`Em_nSKp
z>VrFZeDvVln8)a;n|o4GH>7~~nEI#XlulR+NgI=s$h$5|zl<|*FLn?|{ym9BH--M-
z9n|Y-(6s$U<kxA!dlQvFP*C++(g3-`J^l9=Y&_F@fv)7JMuwj{WeQK1Pd&NE0wW@D
zk*=x(c2g$vM<wRX;hMxYy$^L$dDwWy3zUtUA90M`fWx{SXiOM|V>*X-@hz8Ne~%+?
z-KhSvI<Fi#(7b7YWRrR`01A}a>yP4RJ0VCTS|r^_x$zpHAglLI2{A$eMS_R{b>`M7
zA=dBB4b<|6A1~hf^W-Y<qM92zZ2V4Ms2nf6v+rPp!)q|BvbYk|*425u2c8RH*)Xu#
z_zjdohM~`3^*1jx-Yj$>+hhpB9SmIhhU*t-M#k&C#CO#+)IfPrcIuvs_zq-uB5_9}
zw}Gd|?`wxzYqFQR=biLPU`fF5m0C`nS~l~2{NI4Q4{gn$J<uW%HU?^)d-cnjC%9fF
z&MA6BcO&NKWWe~rBL386f5BwJbuLMUt0m7jAJk4X{xajOEcW5b-uNkk!nJTjx0@}t
z{0U$qnApE!WPFV7HCNfi8Q(j(wcp~#50{gYgg~8#kv;QzoJg53D*u9!K5GExh=fKV
zXb+UeEif&5MhfSJ3x6*vF5Y%p7H|^82w{GoFlyUv-hO`QxeeKVl|SP8kbNWOWMI^4
zvt>o(wO7s#4=(LWI}VuIj3*rrNHtW;`0l<&0v*hHs)L#p*5Zck08(3b=gO#WP?jbY
z1fY8kA?NYkiDLk6h%?SbqwU&eJq>3CjX;`dH2f`$w>NHPp8UCq7DhL&NZ5IQ;1~Ff
zrN>C3{#+BX{tg?!n#o#rq!ggdz7}CMd#EiP+FtV1Q@Md~wh5)VtxB^e_U7PFBjdzx
zz51IpYIM+ilxVRe{`5hv_3&98N6uSN!i$<u;8$fXY)MCv-bhz^B`Y!R<rOh>YzC#5
z(fRc$jhF+j#wMw#$VPhxryRf=W!@3z6-yYIy1sD6ie}ppPy4K00IP4=QayL(#^aqc
zYUZx;jYk+MZuGsVX+Djyajiz|(RRccxr8c~08iL8!tHnusolrmL4mWI-=VG*lK{r&
z_|3R+BsJZ=>Fm!&2DvJb$kMaR#v%{IP~wt4fazdfQy1a9cU^Tk?f|&sRv(eJ`Pc@j
zvLzq#MTOXwbbfmG={s*R9zR8HRuC~_`26tT*1=S$aCcYU7Upn<lo4V?%b<~rte}q|
zHv$<F;cl3ONsjNI@qbCj%HLJ+toj9%*zC=?uZ(=FE+<81yP==!?>Rk2_aTx{0X_cp
z1E@<b_Pfi%*ZfeG@!Yn15=fDGv0Rr_5}4Z#X??wT|M-<Ks;#G|cFJsVXKa4?9g>WJ
zBrvgXv@&t;yM)vD=lA^i;C3K}@^8%Sa`!4`>G=~&1u>$_RxQ=n*FE#;9JNJ5GG0Ks
zp852J+TV`-O23hyN;x!Sp_X1sa-BXs>BXnL5cqgm#PAls;W2BMaRQ)(lwsIKMMC)!
zW|rHacJguljH&E+V?ebx6#xeW<dF=RJXD^D51tQBi)jwXsixZoe;CE9EKYKif>q_x
zR)#MZGEKDV!LKrKl^F+XyNV(RhhEI+uRv7sah?%H5c__N3dM{ewo43@#%H4W;jJZi
zI1gOD{0FnU(fCc-N?6-tf&J493)vf0sVVp-X<SHiEVJs@iziMd9>Jfs<J)P_t|;D~
zL_~K|Ca8rjmB2qihNWTdmxbZAtCF6a0KxBON}eX8IB6IhKzQ{(S1PP$QgBbm;Iv<A
z0X1elfWx0i-CYs!PN?9<w&py-Z>VJ|2PddWwU?Ju)5$#-bzs$`*$mX^xWnZ*^XqK2
zu`!$kMd~TULNXWI;C8l9(D-UbO{L*^Icn+sX=oMlzo-O4vD169lP|K~nd1$6epNs}
zx7$I>u=PF4iW;awF8=-(O*J{!u7l*AT;5SBwTAyRLFL*Ou4tY8G6JQvfwO;alc>AU
z!3wX%8<<f}SI^bC)`|SdUr6>Kv^1KB`w5tQpInxSw~+S*`?2_gr(m;<s4jwO6RaU6
zfYtZBk>+&T#gqVvxw(;|Wk@U5l)2v}3!HH7oj2cFb2ef=19fw~MCO~BCH@RjBm-mg
zY@>^W8_#~`8C{?5t>YAs(AmRIeeMJ2y;n1rO-y{xH*9;iyd3?}^zXfc6qwJ#t*j=E
zO^9!0s|g$78{fR(cqT0HGcsUybE~2Mv^Fh^f&dTR;O1cp=XN-Cb0o?%bdtqTk1-RD
z$sduSLt$M6=1-OhvxP<Ar#ypKVjqWtG+2@QUE#g>k*19NFBOhVWBJv!nH2Xw-;hsW
zE^c{|_^6LMP-$LIhOnKVlk}t0r^br2!Y7LRAv<p)39s*qCXWFpEKd+hFjJt$*`H?u
z=*D~Nh<m3E$EekZyAeLp<Zt4I%#I9sRrg})1GUn_W39gaU|GDxE1-&+*u^8{jYyBo
z9RoTaC&H5M&{;ywZuy8ScmQq7kE*$@7+WC?&E=uG#@r2RnY~_J*rmfCPuO5%)EF_T
zwqtRtqZD)e9o%%^+0x5FU{o5R)%ltgyCfSTh#T|hjj_9m6fK&WM`FpSr@iHGDyYoh
zCYIQVgA=hrixo;fioTWui)tX4kw6cl7|ON-88I<Q&p#T<#I^1#(az+pZ%C_~JzgOH
zK=`D&CTNpb@33!t5g&_k&Xnv_6b{cQwQatzj=+b+`gq%KU`)^<NhLVy+%hVfY{Jnu
zGraYVs{3+Bz~c`>zZ&On>z`M1@m(ahWP58(69;qRpx^iO=zxa^3TzI#bitB<XHD9O
z3D;_7AMXNla<g(z(=4qzC+3F$ZQq)izbY)l5Jdd=62hWlr%w~9X^B9XZ6a6Q3`W}l
z|Jsk4WLmw$qYW1{r#66$kf<8eH`Hcv1D&efk-o!{FFxxPz<>h7;`vGew*KuvOwuC0
zVySx=kv$chDh8HQ4}Z44pjPLyqWGK+O7)<K>$4-{7#yxxo{M_?23$3pdy_=^$!ZM0
zY0~_jAb+c}^h2#K0LDjBao)3D$z)EQ+p}ps>jq<@P6bIPF}FX@r*b41Ly?T#xXs?c
zCR+4IfG77KazXy(#d$&tmu|Mh>zu9D*cXD*bRAVIkSsg*+4=7Iz6!+-{hYRi^d|_R
z&3!ydH|-=bHgOz<Dy3Cn&vW8O#4Fv(H$1viXc!qSOMM<|VIGl1FZ{?*#3sssan0Cb
zpyb&R1zI0{mPHW@<cTcMiVedjMlbKg!4%*;A34T%dX#Ak3(aN>psp6?&7B#i$}}<L
zf~K<s=puSgn&_9B&$*5zfKwP7Z&*uLcYEGgn^%)?<cp34f*kATsUQ}qL0)iOf()ef
z^QSSD*e@nMopMr<BWb3H*)UzM3Fx)q+cKL2J@}=!NdxptOV^%_K>&@04EKeX|Kjzv
zM!!dh!np$iL)qXQ7&!yU;?Fv-mmQaYWFbH!EF`tc*8VGI4=wP9sr9o-Gu;_!`6Y6i
zP6bP@D@wB`5RNvgf@hthSbCJfqR6HDCqWu|q-=CpkUY}ECp@cv3upG)iPwluCf4JR
zVtzh-fc0!u8UdqVD{<>$e02$sU!?1>z-?y-#}2+#SG#y~MFI9~Cf11;Welyri7d<6
zsLuL-zS+POe&YU7`MbwZV^VMTQ(h#M;8`?d*Yve_tTkSQd}3V!b!z<~*cCXX>u6lE
z?v`@k5@1k*fB1je`|^0G-?!gcEZGTV-<K@evSlqM`;sO5mNjK7$u=>l%v1<jq9|D_
zONb~;QL<+#A&foQ2{X-@Irr4}InVPuzdz1-oqwM5$MN!-49(|WuKT*~YkgmW{Hsny
zk>2mvpO(p=?J2H6Fk&7c*7q3@B?nSC*r#V^cQV8xK6mnvDZlG_yW(RJd30}Y|4?%x
zNAAVoTyP%<Dz;i#iS(!yw9vEQuG0@^cR*>q*%RCEPHb*plL8a0B*|@zA-$u!a&z%J
zyp((eK?vz|*X`4|wpUjaj=VC+GggK5_B-}Uwt5Z#_3LH<={ZW+Q2hmmHs#S?oOiu{
zMKJaZ?~4$trZ1w@GM|dFkrND?ojaY*t5f3f=~^NGkqc1o_9gR41AtK}Q4`ux`XI1z
z7{ON4h1j4lU|EZ{<KfC%E4jA6j%nx5WNGnNTqi>!UIfvHoh)AHC@wF_Li78)H>gUu
zDhP}Q#BANUilU88WVD$6wCpFBylja(n3VXuqJ*C;Sg7BdUg;rNC9JA<miTxI7+#9U
ztSRAzS(KJz6Zg_+@a)QYm7CftnmAuE4_)(O`aJk@=e?HANV%kd!z6vN<KOfor3;`)
zkwO8ylr%kvQrc>i2hd=Bal{S;s!ub%&dq>?)4%bFpq+j9N6DL{-154tRp_^uLqv?u
zprWMgwy>luxtiy^LDP(J=hDyM)q*(O><ePdmuIpv;<67@AqB9IW0o3cYp3rb+cGhj
zBx4l>a&aBD9vJcA9GO=S8pUG6VcXS?@Oh`}UcmQB2-ObAGJv194-^YiP<~c{qrW#f
zSqiII9t!f7z`jL(VD_#mkU=PtjkkB-pv?|k&5S1&9;t)ybVb-UH+G=zC%LNp4Yicg
z*17_$q^YBi($|5AnY3-%iT2{%@>%!kNZVt~3iCx&;i@5J948GVMfxOSE-60FJ-Wv`
z`>}d_iYQ$hyFj@jM~QXLMw5no=5zZSpc<!}A6$@+I+-<`X7N~4){A|}2XT{Vh|Rpm
zqW<DXmKkr0U`|=C#TK)-*vfYIl;O6Y>+QBza;Sh+nXz5q7`T?K@^(9(K3&XAS8jr4
zIlp7NsvXUx*8}WWLo}EE{=9Y}g8;*_?I(4(Z_SB9jisw%uRJT}60RS1%lm-TqTVsa
z&&rag3=!R#ndM0MWY-kE+bR3UEF32j9N`m~t90~-v+6i^*5V<^=PV2L^j^khW$YTs
z((LVBjj%>Yqru*ea$4%u!@_=1?Gt>s!ZqE}$r5HwbU9k5=shbZx^1R}n+!*UeoWq3
zSoR;_zvmS%);AdEv)WvM&wq2+X;?A^x(5;H$4?_aQcBG>yrLK%Plynm-<UmhFylm<
zksJ@d3|HZB-52v-f_H?fpxc-n;XR_UGyBfB?XQOh;ODXI)5H8IPhxCUz<&~G3QKuR
z3?V}mXcXYUuV88M**QjB^U!x1H)dpVc})gm_F-N6rTKxvF5yt26)cxw-}Hygjqf{H
zwiai5KW8!79bljdYu>^$ub=tlq?3(z3#Kc8d|ezky2OONW^(f*Dwau?kKe^CN^BtW
z?fu-5*T^_jA_1|`vy5%#%98A9E*3rY4qvU1U0M^%m!mDWR^f4<1=`kc+}Dz1LR+7K
zf76hg%j6{*<{5{5fRz)SG%2PXPk-0KK<15NQ|7A&ca&4SePW>>M>zFDN~Jp9UfnGy
zY#!ESLLT=Nk?m#Fh0%|nI~LKbn)5@nIlik~4SVs^S9=}b*L6H`b%)Y5t_AyamrZ3?
z(pO<5kJEtqD9;B5s&uQS#WXQ0p4S%Xt)ykuQtA3{Toq2wrzCXmu#Y%5<B3f$r}VU|
z(Hw*MF~1Kb%R{E_=M4XHoNUKehcw*yvFl!g@g_&^rmUVenTAZ6lB!~F;pOE;y@v(7
zRk6c}va}PZoTX(#c7X57Hb~sBCEgg?L9vxyuz35PE(F{{5x4)CnQ?HEM$uAW<4&pP
z7tL&Z>*^!@X~aBhRp~y(05ox^n5ZjP%DQ;JGT|zxP2fqETyuM;URqYEAv)debzSxo
zV%OiLW2H=bP=5DxM?Vjp@#|h-sG>;~lQm&2xx$xuGX)u}1|s*9FOcL%+WC*fEFZ}N
z?(oET5UVBP1}3dk4+&7W!nk8F1<_sT*pczoS2+NiqMdtf)c#rlL$5+?LJoV1`E!>|
zf9ma`{ACzY0|R0&==T8nAzIt$TcrE-4;fmfd5B{5v1D(_#S;kHZ@S`w=e*Numo8!K
zxolgiBx0n5&X5kS%X)g6*r)+@P0Kc>VywgWR`rdm8p=5z#8X}{_t?2kLcga28I+FT
zdurf=b+Bj~3<i8{8+b$+8dRYy<?k5ULx-`6h=hlV4vp#>ky^D;JX_H^%2$4zFPfuA
zwC@_=eg&?6)5%HM_xwckZ_D-+)H@$_t)90sS{ShR&^Mp}<6)blgtmCikFC2FD)JF|
za~LkUo4@o85t8KN=Spp&Z}3>fY1+u_CG{4R@#O?bhjA1~Bj!dh9T!lQgl(GioOpD3
zn11#JtgJ>t`F(vlC;0LaTq=o_A{T5?9;F=?JH?}PG)g@wI0Y&AXWxH1-HMkgtWvpN
z7~|1!WT=h3WzX|bH4@EZ+NeLs6D9^U-$Bw#uYr_a-DMWH{_{GTc!pLzVlc^fRLo&|
z;24XAjNTn0dmF7ruB-`IMIY4}kaqcs+-{0$I?tYNnz!DLG{&R4=h;a^x;KZB8v1bo
zM=~wVSndTw;uw}Mihb$7gYMp)5Q){9UN$vplay7ll9DA#J8F&@h$&K5L@LE%qV@0Z
zX9T%no+^9AiVsv|E0dqQo-?6{bNd=%S?;aVdKlkznmBV67xmDD^7<ZLfZJ$N*U~f~
zmHuSVY!lM2Cjeiac}*!8;shDX6n~2w>o}x`ijc9Tt)%(-jR&M#7yFl&LA|0|O%A+d
z(rrtWwieix5hR6A4L~&{-Rhs}znBq(%Kg6Iz|kOGba?Kvi6i?{7b}1KjjCk!@I%ZT
zEDS5#2chD!QU$1>;`;_a*+5anldJ6M0|z&`zd;_~u4TdQi>z4}cWFQTQNQKeY9Xa{
zLtbQYV$7tNc|3QcleKhBY{*iOZdn2MjPw>#jJI?yCrRs%E)#0?-nnCIU*uk)ZY^IU
zU`T4|BYO+w$v5$pKriWY7G$}IQlL3-7Uw6NK`=g<Lw7Ty>rd@$FQAb`z9MBDZ_(0R
zCZTz)V_(#hPut~G1|Kf6yFj1WS<%9~vb3uJLOvd<6T>}tidDI>?*u^wnrmIWee!nh
z(Tnh!0`KHBpM<BPlZ+&(4b5!<4sE8VX6Aj;Qj&3J65n8=9fTt7smWL$FTJ2cy-yon
z!aS)dS$TFok3ORADJEYTeI7TAIX~{WsF7@YwC8Yo|H63_%4vIVfkPI@CoZQUADG%3
z$V(Wp+uv(|j2c|C3-5k$1UTgAi_diE94n~`G(0_$Rco9Oy2JvN=4@kP+PAt%c>MLS
zIDAiR_Q&~->9z=j4^|4t1?*8rzc6z#qSWR%avH<!@%n09fwZkE&38XT>imAHIl!?}
zv&L)E3Dr53&u?Wyo|%{2<w<p~;l!)^=G0f5vgPz!5xvf#d8b_7HCM4j+1rX+I*<0P
z(9rdJ!2;o-5g=Ky*{vHrHBlmTD=31W@o81IE~8Ff*=Oz43(QVuYuk6YhH3gFXMZFR
zwsi*CYmR7gYU+s}FxN8KsXh}jwf`k-w(t>kkPylK^jF*<eqRb4x;P~;D*{2Uq~#<f
zqg(%MiAU{?a>#KL?5kkeJ0q~Mq<gJK_7??Ae_eGmL$d8j$SOY5vI*}lJE9E`A)8&#
zdt}-=Ha->k$vviyGcwWO$RZj;@9<9X)THO81&G>I=%?`T*02pen3wV5D88ifny+>u
zVPfYa$_JIBgp4al<m7ZK`0xo%wKt@E)F7l$#uakx2}ZGq)X5l%p~devt%4=#F4E>)
zc6@I1fky^c!XQzJ<=AgIxo5+z#oRyJ7bW`0*1%uCo1*Aa?@x!9`xJPzbx7dDTAKN%
zFuWv?OomwRqM1N3W$S(|F%Jc?^`_EAie7H6&r;Ss1=^6FPYv+K(hEA8Tv35>t%+qX
zs2~=$Ee~m&#wTYGk^wTYl;>=IuIOk*#_f}ES<~nNO0U3&jPW(4qU388o{digelm^G
zdDo9v-YY#^*d~rKBX?pxqS|Fr<@?Y!a3&NdDp}-!=C!)`1cS~Yom_9CG9u}HN@c7*
z;HIBy0MX{-Rk26!13NlAd>#yr3|-&~R8dxTt#~8qI-ELWGTDARz5+HE-*ayDAYRxv
zt?{YqPmM9Fq8TK2F+<0(<`S%fS7?!qmrED)=7_#AG|ligZJ4C*N*&MX&3uP!YM`&B
ziw^Y_FR*!ZPIKzvc;yb=wSkrxOKnCm)1Saq^3~PC;-SkkknByfGnbI!tYgLkG>_*#
zrW7mSCZUN^RyHW9x2gH;v}ue@7M(OFk>6?wpQqM`eCY$vbyDr0tAt#@)N@7IjCD_l
z#y$fClZa?JGp93!_smpHn2e=N+X;1YTF_0=2;@{ZGx8I^*pIQ|j+ivU=pl;IMh}HL
z)-#`5!&3)x(hQr0Uu$@-(-qtZa;!7Uo89OaxV>}G+nd8nm}B2$R(`T&hs|TA%ReUs
z#!o3^fmCjB<5D`(7Gu82RW1)9i&P$uXyIzrxr{PK`0r9GcePjbk_W2I3njk6qZDnL
zoHxCLAZN{e8KRi+9(vGC^MppIIr>(`6+1N7>CX89&_ia36Zis=P-{p&YH54Tc}zvd
zQ>yLm%2He6rK8q<OE*NNR(ZVN_k(`~d}sS}!YA*KhJU!q^DXTq#b3~Vd*dCAHD^Uy
z$bi@+K+??6LW4P8K9TOQ81<|NiaZu}x+`1`k|%2kn=<|G%e@LyxE1tR*AIA%A%Mok
z;t9h$(kBw2UbMc}Z8gTlbTMp$ht6IMHotw01%z^|5`SItW<_bhLAef3^e<~8WL(ZX
z6P6V>hIBUk^?AC_qQT1fQ(mU<>}Q#*hH8HfecWUc{U&~~Jt}N?@WuF5SYv3xhwPUz
z`_>Yzu0(jXRTx_s|7K0aE9z8}z*Or5U%%|WuAxyUGTwLTTT3TS(OixN#DOxm+LniO
zijMBm0Qf>!1pnsb(LMBTsNeo{>m`)AtDRLvM$jQAUh_PI;!>?7o(Hv$Z*y9}_quYD
zP_7l94=LrOoRHc<e(0toE<MoMI?wumsFd8>k}>TEGc-)z!QQfa9lf@(xe^gu+m!ba
z^&Wkbg`uj~+m|Xh$3M|QcAM;T?^TY-&Ms1Q!g^KOC=DBi7zrCfAsV@C41I2EUtDl*
z21i5*`_<H)xtaCx9t(Th>Cl_@<rey>x!+?rYBaW-g}a!&)N5WM+xd>@HD*?5m)3^Y
z&p~;7XK_dTb@v5ggIWqdz;iagV&E1Jbd!Sj+wSu_QUnVp)eCo-Fs!PvsbP81*fECI
z^bc-Oz*ZMI`5_}^L%u~z6d6B$0Hu_-bOF>%3d;F)kaT0=^oyV(WKaXTVoUmZJ@{r)
zwRDWMqeTv|r_>LDCR9xznn$C_rwP-{HHlWsGVyD=bAdE&rFx?TIaviQ)rp6fO<mRk
zZMN<lKXGE{K8D$<&&2LluF@I@YXO0a$ok^^%pyzggnGvfv$mT{)i^w18aFqQuz=;3
zk&^HL2|K;jVZE>*$UQJoIWW=Wp((63>ZTw6EtIf9!_!@G<V@9Klx?;&({lBpVkHA}
z2oDN>!s%N}qQW$B+3(hl%a@IajH!E=9Hk-XIMB1RK<&J*@6DSSSJTF+&NkW6^-2yM
z5=5w~_6nX~TE_OE>`n|FVEdMK2-WeaJ8WWnuvDobXr-vpI8YlZQjGGv%+@00!veZ?
zL}!ibKT7%`(kP=fv2&yBo8iEb{njZ%>q`YfaVC*}C&IRlhw?l_W%do8Pi&#f6yY|_
zDBmMY;jP_;$k@VL@^a;OyP1V98CR67XQ+7FTZ)!Z-YT=cX2Ns$%ACn78pAfr^G!CV
zP|=EZxr`y@Y{e(BTLjzXML+S&?b}i2cYuv_@xByiV0@}LpK+>W+`zh6%$W_Cf`UC6
z?wfviVd`tnxu5>du!2-c_cInMZ@Dznd7x+4op2i1ae2N80UJ;ZG$IN_Nt)TsE{L_r
zEu}PH+Av;UE66cmA)D7>7~=QgQtW|7{)c7a)}akhM}xgn6St4;7Rz%)lL`~Bs@aUi
zGb%HLy;Jc1l|p5{yDX5OqJc+jmgBi>mi1l6ZwLZQ<Bf}r3JA3cEqmdDir&Z5*%0*a
zlwWnD4H6lWcGn-`l~6O-)2R^{TEmh-_(B(zzkQMsJZ$7Vm{|^(#rHNvF)7*DxVvz=
zgf|$W!~ylDFpzBTHl%|)hxo~XemHN;o<f>Hn^D#f8}o&Wx>=HH?%8ODM$0E4iZwC=
z9&^-;2{jD`fHhJvf~W1t(^d;JooSrQaqEdeu}qBMfBnq%hgM7Q5v1pPyl=}Gf8}8L
zP3<&6j2`d>ylQI801n@bEdVLH^}Eo(Vm_ALhU09A%HWC3CnQ3Q*l{f5r>|;lFTfFr
zlM$cAU&h|Syw-b=i@d6Tn=`Gd|0tRRq93X7Ctv&73r~JXDa0I&+#()s@})=5%<aSE
zprI13QkIgkgn*?U0WZ<5_;8*uk@TfTJ~p@`M+<E90pqC*`}g?i^dSQrkGJ<AfzYjs
zF{9k%;y-AFu)+92j%^OS=k6|;YldoV+$8{E^N<ahHz<u3gtamvvp*?5w<$42&>jeN
zs1?{YSU^;6XAyVM{KPTBTs^W7xYJsf?~*BYA)G*#@`j~wLo<q1K#(M34{`)bzm&#=
z8tR9Hqdz`KTmvUQV0&pL;f_cw9d&Pp%70fty=24a)AfPpFu$W_D+RHRH|>z9H3rQN
z$y`w<l$YHK9bw?2XCdem1x)GO&ces4kfd5he8t=c1iYlOhWAER0&(iV?a2D+H;H1B
zp$c?`iQ;9h_!C@lh61OG2@lTIu9_;hSXV$nA3;+Q4gxupKF+WE?~YpOt3i;rUlTw~
zTSYI!HGv7k4?2JDpxOdsmd9l+OTY2QtGC;2agp84(@2R=F=km?d^GBCHb@U@1=YE>
ztZ4cyajWgTt&wUuxTKvWDGHvB<ED?5nnR6RgS(F@svuPRUx-Tm7NjYVB*$kPHoO|#
zFr)5tj58RSq!%XhbefdY%!!Z1wO_TQJ(WlIkQqeKl#XV*1C$e)juheuY5D|D)2O@n
zs?9>2A)<dlkWicXX@{tqo<p$m;`VjTvoQf)wv{BjzDMBUt=Px5esQ87JfD=2?Qdu9
zWGmS9WXp7frafcCZkkctJ3^D_QLZR@QGULqd}_9NaqEaR3nZ448}cejQ@_gfZF=sh
z>Mw_DvVN4brSR11pDK3HRGDVAX=+RO(bcFN|NXvjqny6KPwRj~EZ#hW@3HV~Jm^K}
z?K|y(=s$?=IMx!4Xl`vmPv7Lk+`rCgo^2{Q!Dq`OwijB2clAP<?AGO(5O;C||G@VI
zk^ud2vN&C=CkOYiPZ!!FX;iiDEK7qY=RiwVdG7Qni&n?-PtyVuweKd`OmkpF_Dsd9
zSM+TX6)e9+$#9?#Ur<ir8PhoK>2LSqSM{$qq1HIWvyDp6JEGpO=^;P<!EcnStvZ?d
zdQfLI3RU3*g{i;n;MgS$X=@ZN40mAHC<U<=$4y7nFlyNK9G1?Bjuax;$-==+xLHzf
z>_K+2R(vbZA%g_hNRA$wvP|PWJIfFupp!o?OQDxZ-G2Vd>Gwte=Dyy|RQSpr^O98f
zpHYlf%IHuFSYiSb`Zh|`OHz3RsXb`EehR1QjV7{ha^S|*@4vDdi@3Kg9W#0K*6L9q
z(JL!#^JzCrJ||6oZ_X(87y(^49nX3$=p#Om#KgF#tz;d49L;{zp*CUmYhKa4!Pa7>
zkvEF9);%Jd%1ap|A_pT7NSG+#LDtO>9G6#M*WPKI6wVu`jZl20kg03PTn04zC!=!|
zI<GixVleGjmA8&$I&T^*l%O||b6gv961klU?;jASK5w#knhzp!>uLOPPe>_y%i=Bc
zR4@=y`@gP`+PI=52M^Ue4H`4D$5-DBS~<4lG`wo?=Ulr~3-d#|BZ!tD4fUBj+XqSz
z)8?IPIRNq<%rH6#av?=GL6I|)(IN9`<i+qiU?>;tUyNAlEp605|3Lt2_c6L^04^-p
zxizLq#Y;a2fY_;1`pFA&9z(7q_18rv^@C654gB+ahA+&<8;-81mTkYct)K8xqZ40I
zDkd4lpz0r7bP}ez?Hmwsb8Iu+qZKPJeOwp_zCK`cbKOQ>O6msbSVn(v`@mJn%nxx1
zV1M1%eZ9wy4kUbi`8Hc`LHG6$h*UjWb>vK-#Y%Y@=l2X*4P#7|K}Zi1r<RW=zKMGw
z82cTGJM9&b-=m!2Gwu%Ar2(LnRM1yNB{DrVBiEK`_Yxe-<;4&tQKP5?=YbT6g$y~u
zv?m;zC;OmP7b8W%@~-?v54H7NZlps6VhPwXjF7W9?;k0=H$N?SvHg-+hSAdW;<_KO
z(Z2Mw^#OuB-v!6n9>?mX9HzTzDW<Vugo~ba>7`ADOwp5MHvPS1DywOn=yhlW#%D*f
zBvMvCUORd!g4$BxV>@;Gnq9uG!|FdrV~-}BNa%U~EglpCZ|-BKRbRS^pJc={hn?d0
zcw^%amCn%-rYX2&=R6)Ln@()-lCEj7^b0KGMf^@l3`YD+iK~-tmxfPTP<G|>Pq`&0
zLW{f`>?CnVis+;D<09WY+G+#-_jh<<>PzS!6Eky?&CPaB*%Kmk$-w#d2TTXs()-%C
zEgu5&t(8bx$QJ5POuTV|Q=?O}^?es_c_yD(lefrU9oW3IiHN=PR49NHsyi1?-Qjqr
z6tLZ3H6wJKAE!zJR#|k;&MC7kbpn20|ATxBs9PC#5%QWbicP7h```p_KjBiIgBbV(
z0N<dV5!>mXfMW+Z+pVglZ<VI~nmXYvhAo7aE&S^<OxF}nF&(F$VsnGYi%U(qn9k3A
zXSIN{(xk$r7AN#!ZWFg^%`%EVQX3V3Bmdh0i1||%2G{<bDY3sDj)F2UwTFKZzejR@
zS3)5;XWS(MOzEZxy%G&`{d4g5ii*GJsl~0sX)KV1x63&ae?cm#=eoL33Q8?%qkdO}
zeuGf0(m#t~tA(EG`{g#a(-$mrh8QiFI4XMIGx<1p6B|rx+qDySAaByr-M0U3h5aq-
zKLp8$v7#g0!apfQu{Abo2G<?k<hsn#0FiD0Z^E@|l3ueuPk0b~+>(gAJsUZjp`KIn
zD9&U{i*zQxN37!I|0FJ;FqnjG+5^~ITFsJ4+Q7ONR3kk3)f<uEh8wS0!^hq4scW-D
zi7UMXe|w6-I%DO=-Puho&y(0bQoO*+KQQk9EamTl{<~g(ve+erCvao00o@@UMs&Em
zionHqwefj`s)Vs^X4&kW6v8Xl@J6xY<-A;2s%_F7+>&{*0hs-I-q$pmF*mMHTxNOP
zG%5<8Vsv8(f(pS~wYtfl1;H=+VG091P0-heG^6kDjQ(zIp3$~hiMa3~B2!i6pF^$m
z7j?e4K!w(cL1Omf8WWZlE20FE=bgikE_MpJw1X~X`isxSgY>nh$r77>IAg*NeZsQZ
zTdU(PmjxF&Hv*OP_eU}L&uBnS)AThr_=WX|PblHVi4ABg4U+#G!z#?GJj_DFEm?j#
z8_k#!FG3F00QocFH~)7t+F_Wb|3&Qns@OkT%`@|Q8G8w3IpP2fyVy!G;fBN>*5K(u
z^WRGxo$o|8b&pn@ldEcv=tv0(OlyOq^DC0k{9+fT)liEb$&PkSBqnrzYv(Z!_>Y2>
zv&RN#5p16lc@blrVPRdrU8Wryk<u4iSo>N-sB-#Wj>t%OHit%V7U?1Wy!%w$qS<YY
zt(zrzjHm}#EgYj!6$IaobS)9Zwuy#3E_2v&d3A9wLQ{2`yu{sb9W$e$z19pGkgaU)
z6B^9gJKOR?UC-85xmLBaZ7+q(uErX)t__%d{Q?NH@P{yafG}GBN|>1UH-Ib^T(H5#
zr_}@-lAXv82u!$hgzZOs9xI;c0uNzOcNMvm)*}~0j7L~M@wph)cTtU~#Wm$va?r~B
z7V`<?+8p}48mhXlrIW9}ZgYM@k1*Gt5ixm1(27po78ze6)Fpg0>&5fk>k&PXMPFAw
zudaZmLVwu=@XBA*;p;xuMf(GlKCAo6&sL!LY-F1x!Wx?%0|}vD7n%`}Ug&HIxI22W
z*vLIOAKjgH@#?r5>Z4oo`?C~R%m}D~LhUbhA}fO^Ax*5Y9E3-5(Y!YO`tr_>E5x7m
zD@sEFPyUWuq4%%jMq~k5h|>D%(ZTjZ9K{B)21%{h=q30xy`G^3Sv~y5ffRK3*dAHN
z@kc~`JiJ>q&)+RM6K#g#MWi*cL3dD>FkFz)d16F2n2u+11cLlc8;xu;<chs_Ah_vi
zwzpQgQgiLboIi|WHeY#gzP0M>{ui0iCwT(~$48%hJ<wjz0wV11g#u&_Rf-o^`N2^}
zwdk3aZ3Wo1yF0bBSGph&22oSQY5Rw~o1C*BDds+_uDOSTRt`->E~E{AbS`}<U-WYH
zyvo+p9>igR(yAGcR7`LC6`IiYB)kf$O4hP~wH}hEQ4qgEKL8*jNDN^NA3Xdf;{>h5
zWm{Mw{jje~E0?Z&KDr`!#AZ6+eDTWJ6*72J?D6W*oIVeQ>BxB!J@p4_P_jUc0W_w2
zz9?fvvC;kdd0$^DW?SHFZAdW}By(rCjc3L!d47U&<^Hy+?bE7upuW!?h_~;QS`}%^
z%qm}<0^8)CTs=SG;o1<VV#8Z=XO<>mGDdI@t{~mLcPPwzY!7ftnS1g@eI_6GWS9Mo
z`W2B~*?&E<#Ra=Yv&Orw#$y5X{H+I=e2V@tSA7>&+@lWmEi<c6X~Y(r?qTM+7FN!k
zV7R?dLby}w!N&H|qbL?H<mi*L3|GD!>O9`x7+>n^uHl}{jHlU(o@Rb%_;S>x35+i7
zk({o_^vaDoddI(y-u&^#%SfBTTsV?X^V*Fyxt&hKpE?xp*_}OWP@{H!EH{*b0IAen
z${<%}M)tZ@TCea%Wu1;VKm_EG*q*K`)wFLK879n~CirLr;_Penf^an7d15AE&PH;j
zd07i7i1);5V7TayvQsv)Mga3JYuofhlt)mmT^B0$@cm!qhd*bFKRNKmTG`UXv|9H`
zqZPt69_9HKBw>k^OWavn$`aV=LQV{F_wW16E_(ewwoqYQR}G@`ou`YW-I~OC^pH)0
zqmaB<K}KCj_1GG^Mq-*g_J@C`?&<4?-Z<#_R{`>GlKm>@pj;*CR<Kr$&a`?N>G5Ho
zNxXxK3nMDJgN-1Eo_R-Wywr(oxiwmGrmnhO?Dm4<+B?;(*D_X!Zkm52Xz0?!lx;(t
z?B)d0H88CfaB(CDA?&%J`#HeC*&B$;(I;c0mO}uV<iBcP88Q8*5$k=Yz42?Fhcy=S
z(ZXLs`R4FOj*wIHP|FU>ndF2W45{$>j!*s*?$H_RV|z#)Vf^8AJ`w1K;zZ-$Rr|vt
zmnN#|$c*?6gc<t2uxr(Db^Bcs!3bCORDPl93Y9U)kIHwo>VGzy@;%rBe2_$aN%^B8
zSmzmAoQf14e7>Yyq+zMn<dJU!g@o~KZXi-|oe6Cv1pQ6d%Yq)PZmdD{>g+AE*>`I~
z|B~H;E>8G$dX(n?WV&d(W<AlWm6iVRM&yfyVeWoUu<H3gtIh(e1~9<?vg&_yrT-^%
zi~la2=KpHkip+zPN|4^5$kRZ#>-j6s6qKJ#(%u{SA_dKya6bf8Y1==lbcMM^P$X0$
z;$}<2aqM-hyW~nEG!mu(H4Hum?9Yorf7+vrEJ(e21IcI}uSJ3|!NDI>L?VL04)UL*
z(Ovb$g(fgZoeb01Jcql~frw||>7`u18rLh$tH*%bBAk?hh0=u3Ho)6xJ??LW(`?#_
zm0-oO(RUxFu4WNh^HF7}PzZh6ehN^?kN%l5FtXsn;vw4upM*7_OPm_S_}pp`uZ;0>
zi0h5-A3Z7y>F+9}=%y>;-Eigz$#)``>)JT+Dw{GmGLk~EeWK&p)i?=9a^R(lhD4wN
z>a#3B%m3T%#q{9uS!fMZvm|I@iqNp1T`61v!(F_!|5N*pH{1IUk}G9USIpG)62>g{
zLdh73BfXFi=r!ohvshQ)kmUMngJk?Z^#q1XlM8zO3Wn>bJ@8*zX>*_Xt}hlto88P2
z^>9XmLTkPqe!X}oAGo#Es74$x0)lMcFo{&mxecPib)5H*$Kv1Cy>Epxw+$9DzrFE^
zpR~<L9Ox6yNi74&m4N_09`L|O5^#*$hlJ;Q#>bGO_`>>*5szOu?F_c(=wA}jTV4Pp
ztgsLj#8O0J=u+z+4rdR6E6o3|U1RA5M-^MualX#U)vU=NAX;bPuK&fDssQk@_w4VO
zWp1mrFQGnkNF<d=!2wPYf@sHWk};zA*Fwm-!U{3{!t+UZKvEW}09RMv+Esu}SrtR^
zBA18t;h^7|;-;0C9p`_vOas^Z(joAPZ-*Wl|LP{TbtsYi&9JekMs|(gg>W+a>%&ek
rQh8T3Fa!UZO#h$R|4(*cWRe_z+h6!$4~)eD0=`VoSRtwmyb}Kh)I8N<

diff --git a/articles/LC-MS-feature-grouping_files/figure-html/example-2-eic-1.png b/articles/LC-MS-feature-grouping_files/figure-html/example-2-eic-1.png
index 5d237c61292015de494aaa2fbbbbfd988b57ac91..214bcb7a30273287ce4e89755acd3531440d5554 100644
GIT binary patch
literal 106326
zcmeFY=T}qFw=TSsKoIF5BE5rjl-{dAsDe@y1VjPp9YK&JA_z+FQWb*It2Aj*DN0qO
zgAhSP=|u=N`)+>coOitUAGlxcWx&|^K-OA&mS@gq&beZ*o9fe?XFCr701e7O*8%{L
z&_^=NxwD7Thv;nRfpjv`*M*+Y&!J8$8hTLr8r%v10M2J;f3U=*ZD#=B0Z_VHH$!sQ
zeuOsPcMdH%4ci+qmN6SNj=cJr=knDUE#%85pGzF!y75Xq{2WYsGTV+rhm8MJ+zU~r
zg-;lmg>|o!X_3LmUpl9-X0y6QyF};DZ6?0$Yj?L?Z}*sZ=iZE-u(T#VvL35)CC=R$
zzP-8C#)C=`_<z6u=Os{zuDBU$kb?Vmmy{{s_Ad(*M6q3A-GXnKIa9zZVBzPRVIiy`
zYzLIS0@6ias%!at<+py!7QK9tLw%sS<?KjGE8JM0$bG(!?KV*zXI<W_JW-u~JiHya
z@K7g3w&#R<0Nfh@Q)Jr)zz-)5!zWTG9nSzBn2(53f?7;hQctL=W&mpO1&w{H9#eIg
z>&D?`uG3okN%jrD3yWabZ|+t35U+BL#NfFA)x4u0Ndoa#K#E<o#BQemjlwQk8l1Qa
z8jU0!8!hMr2FG5hK%G*YoZ>$sy_n+ny|os}u1t~RBk-#&%MbZ(JzVrK9Rmv+?#~Vl
z7mX1)%-%*Q2%s7Uz~5+Sxp=PK&ck`=eEfgh2@5Ir4hrHPCv|EL7?vpKPSYKI>OEv*
z_4Qg(N@TIW(FWKsE3zJN)L@Q&jT+W%^h|h842x}^viGd0%0IdvHPNuU`(RJ4_j~L-
z4a!38zZWj*J@FwafJPPtO4+s9Zo}~J7&G`V!U!9NF_V^m)X8Sv?*lUzURS_hi39Jp
zGH!o3B=z~}1?<gSL{J3%QlWl$WuKQcK{Mp=QUx6J>(G;9Z38&)T`a)N^@3M3CgVR2
zvezGV?94f2Ev!x-hQ2$>{}FR71(|-Rkp<e`BW2<~+$Yif2FD>KyJ>RvPClC{$CJyN
zN;bh&=-G>8mC-$l4(Niu!_f3a&)f-xM;sOhw*yMsH5Eh8&%2UVJ6b**e)nIl_V$gS
z-LhLc2`e${>88Ta#M0a^OMG#D*hlb8#Eti&X2S{;yC)itb{{OOsc^m!pt-t>R*a9;
zSRDK>TWB5tf65Ax;uOFM8PMaG4Z`q~JFV}Q;&LN}_e(;9FzhXYcyW5(#7<T<#p8HT
z(czx8qt!;pt)Jn@fdNjl0-pTBlkSfQnxSQ;gL;}F^hF!aW~3s^<f9Z~7`3AKL)E-2
z^KS{>;e){yna9UQcr=ANwI_U*HaxOfI=uaZhwW+TyWRXnHzH)B$h<S@<eW+8r{kG)
zexFGPv*Zd0INc8&>ud2U{c&H20Tm&i95;VQkr?6b+(qFcW7H6QIlMBbgxk}$Xe*aa
zo+x4JF4Q2yPAc<__&(cWBI=bo2wcv#4~Y=gTW#Hzd+Rhm8zv>zW${?mbT?=FrpdqE
zKjUGIjQ<f%=V7PiZ#0U{4l+b^`I)!ipR)kny8DB(Ez7ZN*iiK6Jn+HKS)ufOHR0I1
zK|L80mh*4TweMh?fl~G0ykD#G_t<=cH`*O|RxVmO<EzD%N%MBCEWq~eP$&2uKO8f~
zb+~`*7SY&qUn9}G&9%JNV1!2o%mHnk&#b~3vdX4D$n+87XxP^WN>iBu%Tn&Xq`BEc
z&|dMOfAQle$~m;MuUZ55oQeW>gsl~s<WT#jlarA4+=ui4D%P2El+lZDNAiVZuYWcz
zu@lQuw4J)^-FLS?*4Yf#Cc8v%<wXn!J0;_^T%fI>ZC^lU`4aLrACiy6MHJhEL%P9k
zNS#tZ#0cR;mj%c~SR3z<z|3j7bvilrketTX!J`LCE2~F?LK)IRj-AMA?tfI6JgVw~
zm5Dz8L_Nvj$L*WOI}O=~p@BzgGx7iX^7qdi{IIjZM)t{bqgMiBIfykxxW|*_T;Rlc
z<6s7l4!$B3)8!}XD_mn<yq?+R-`z?z{DV2m_V73plM|3gMK#?1enj2giB0d2Dq2tS
z^XcPWzkHeW0{yoTEZu;1yE%NKaRMn0$F~b^o!RsS1)=1dKAC+5nF#?ADujyJ?M-az
zdp2{=KRf({uXZz~c#ZFhuYaW;Hl~Jq>Bp=N4xFBj`3;#ZEq{w8=KcxFOSqPzmwf1O
zW)^C{(Jw*4(!?MSO2Tq%@r`)BF6(=qZg=4J$Z<D`lQ+)z!xo~Q;o4=5<n|IyKkwgH
z=BP0dcJRrBL8{I6V90A`vsJ)u<nmj{BZ&|y8H%UK{uuxlWx&KUK#m+d(NNz74+cOs
zR27gUM0D*+=~(#lH;};ClQ9lkqf;I0n#e`url@qBJGhg&&2f$+Lxp5Z5OuVQT$U70
z!CS@^e4Rfxo2+6_74qhbjnGv(MfHc_$idT@f7743$nWjVZ_dZCa!4^~y#0<@4;Tik
zKd#zB__UkPblQ661Ep7X(H{RDC=>_4$}4BUU}V7(cn^t>PG5Y?rV2W`>-(`-cX$`+
z`=JPIQ4<lqJ3gC<RT6E%h5ALC^(D4B0JAHn@w?sdmoflP#9TMJ)tddC+;rUAu>$?_
zO{H(09&gbK0&etK;2>Ce7BmP~|H}a;KKqelWNS4_PeLC@2FS2;a8DCL=~-Cvq^Tw=
zbsUWDbx;pbkVGfs-)`k@!Z)Sf{WrKTZbl0PB^;t<&thiiZ}6KVY2x!st6;#QAdK6W
z(|o3y+gM(?dR&0e3G;!IB5ewr+IE($W8+`VQHhJSRptHXTwKthN&^PKq7RTJ`XG%U
zp;+{PJ`MQ4Z%0|4x;n(yr!KI7ep>AmJv;bCD^l8{Gc7F1faMdAGvDlZJpj@I|ElQ_
zAcalsnx+6sm6`)SYZc7HcKJJoAFC#c_UTE$&%D_RM_CZFz%j8o$ZEk$?@u-XO>{Q7
z>a#BnDMw$F@aE1GudzGk1cyFotI><Qo$m*a(Wn&-opdno&UnFZr=3h(M7{YG>;Ylz
zRxpr)YkUcgdHrWG*0K|gqBUSN&HYpaWwAK`w~|shNAr$*gjA9IR+%A2pw#U*I{iP(
zwZ9&Mj!PXKg@5l%!HeQ|c}Q5Pi~Op2%v-}WxCyIK-zuHm_J>|<)Z>5uiUHddHxoxH
zM<MCFKWVmPv&gBstNf|5X%W_;viDTnDF6XCxErUV%tF4ma&6}ff0EjDcbgP}y_?f9
zpPw0FCotAn4b6SK(z~d~T`&N}FW_~a!)$T5<1i`W+^rQaV<@w~k={r?r}Js-+TZ1*
z)5Zw~RLW?Wn>B&pH}r$X4mP#e9%rzkPYW2ZF}TSUUL=MTg!xpcZv44raQ5AdIec>-
z2yg&yy!VMj>C31+7kaS(A({FatidOdcJh!3GP6l#uD|w#tDp1#nAll>Zg03M1!$UO
z5sGX!ZYjcC=iE%`;LuX;TG@0QU%H5Xw^W5IzvW~JFb^OisKj9x6}CfgTaPyNaj4>^
ztA5MqIV(H`@2LpR^8KZ>(IyTcN%lM~TweU$Dc1o_QuT4?`7Po0^M)3G$KfQ#1GX(^
zr}QL+A1OySr#Td(ONd~v9syZcf5oB4mz-c@P#?xpfO|d+_i<a?3%+j1WaxvW6_=G1
z-1#Y?S@e>`TEnz8k+lExKDtEjxgbV>rsW+~O~C7Iu2&YD8WM>8p}ZO95s4(V@p^i}
zpzrngJv=OGzX-EwuEM%djb*_!eLv}GPZ>qHJto+=tAMbfXB9i4>Gov%{du;}vdFd<
zxyUZN4eZD}%o|Vfxak(R$Zni~8#aGi(_psr6aRM459g7$G1;(>940j!gO+UP5<*aH
zPHvVfDyRA6LxW!XVV1>}q6k_)J&dfZ)tio!iEsi1o;!v;Mh-iMRf}QOnP4bK2$a&{
z#;`<&yo;e4oU11k1b4#z$ADnWI~Rqb)SJtj*#4Lub)%f{UsJpy@td0N!ZK2@p{gH5
zA)}-p6uFe)6gHED)iz4SkH-=5=h5E;Y3ff)+HcuYhtt(f^kmlw;)l;y+ZNbxU3OD^
z{Jgnkj-fg05?OfuAJv)k=H!DAw{NHs?qX)_V(f>3wimpuwT&W1_(S~nLp_4m9qQ{A
z?ZTMH*r_gLXR2D;W>R*FXwK1iGkqNen}!B-0ICy*g?upmBU2;){(wkTRcdqs0!WFz
zIelTW<Ld#{=ln9Xcd&-9?Q1_eSA>DiDm{VY)rV!66x<coyC=(sDoce7A@_(GHK7@7
zvN&<y2qQ)n@AKp)hh~z18wU4s)aU5<oJsonK!#4RTJ7e<Zd9rd8ST0z#=@aKz-WZt
z@Ow;lQ>YnSK|aKPOseDueF%-?3@wFFsfIDdMh@WwzM?>ziQ_VyfpZJ294!B!wY&P|
zg%8EWr3BZnbJ4gIfpjma!uzh4@Q4ht^_iz@N(6R%KopaT_B8GXP#p0az~Jt_6Wdc5
zU@>j`V73qUHG1t=-G|gpiS8%IZ!qPe+4sty2$c4ZlqCg62XfS!ht~cE;e@s?Dk^nL
zntoCEsw9_(m#?Ez9$mI1uJpCi)(O8QHPl_hqs?3Cc<`jTV~?!CtP>IbyAlKgM;3ym
z(~YkM@M@EPe5ljQm1J#2Uqh(w(<y8(%+t#$mdBzwo$9}~sz6@B=t&n2Yi4fFEI9_f
zoakRoq8cfDFZsGlA<|@@kM!Z+;Dyrl{dTnHyt>|Wb`NUz2FPp$J9hVNb1c{|uViSa
z3zVw1(u;+Z+ruS&PWzpQb(HB8MJAeg&0&Q7@y@EzPCYk(aEtJWfYClS#}iBL&}>iO
zsk*KOFzs~->?J>>S7@=UW?Fijour2NOY0=_s;r&ohfG6pRLarkP~E>T^f-?);bKb(
z!*Ej`!{}3eqE4!8_K57lBWF4xhvatZ=5x({&z)&6>b$3&6rN}fdnczB!-KN;G9v5P
ztG!sw;}P)DdtVx-f>rp%Hc$Ciau*l<?J^BeVZ3gl-Ze1C4VVGKj0m#+81qgU>?(!^
z^L8eDGp@vXeBfGyoE3z9<Gd~uzfRBBWxm<ER#g}bIPyCM&qaDg36+ZLTa$8W2LyMH
zT=4Sg`g4@JUh9jo%-EFCyKeHnWHmudX7=Y!F4gtghS#sg94Ed^U$l(58oTJpBaX6|
ze`)rsx~#jaY|gYf?Q`!~#QQPq58sB|=&#V}WOqaym9AtT+^jlvUsyOm5f<^nGvNh9
zCb^#5a!rH8OucSWT>2&~B)a)9@5vh7mBoU?mY#M$K$9)7KL&RCHZ~`EmtED>lF)6B
z6m(0&wc!2Y99D#3i44Vk8|3*9e8`>l&*Rv!O~2WQPXYsuWrPJ$E3tW0y-nSsO{Smo
zv4Lu}x0~I43_4$Mh|(mQ&yyWME(d{IxxIiPjAH^zRB9faNIcSvAAR~xu*UiIHcOI(
z$IlmcEEVUdM}Aj!LlkcEP^#2uL{zu(r_(ZMgs;YK0tJz4t4Vhl>PadV5nG~8jrYf%
zcj+xdk*wvNXpQCT?H=0BFU-1Agfb;vC11CaxJSJ^frJjqAt@etxZ6Z=mpvloJ)a4`
z3%D%tYH2X6UY|$JAdRy0kua{hg#{z~)rG+s-H03)R{nUAsTkNqZoJw@YNjYJnUzS~
z1-7Sl?|ioZGdJTRrSm?!Ck$p>tXn96YHws}5IZRl@OPzfuGW0dI4L#JEQVv}v7MYE
zru50IAOIGi)y$XVh4<nNDX50geLvv>JP_QUIcI)eHt`g8cO4ZZEXb>@N)_*eKaqI$
zSXymVq%*BAoB6oU5!tXUZ8+<JETPT;#hEf*9D(?xg4yM<9f<_0c>8j!h&nl_0q>Jr
zbaS>TxQjTq(M+(V)`)4RihORs=5T$5?9*o8&vu+e+%$mS$FsytcM{<&h<PC-?0L@J
zbr=UScN$AH!4Hcix_Ih)^%mj$irq$~zO^#f$-5%&Cu$w3ef#UsAcbqaEiNWd%26NN
z*2!GkJIvZeAO+wV7j8e0tljJl!zMBuqQf}Il15p()t+C3YdrA`^_f!-;~ZheYqj*;
zDctZIFS$@AfO4LJ+ekcYdpczN)r-gW{9hLQx0aqt9xw8#_8*~UrA=qfuIh9z@yu(l
z%JU-&_d&|8okH+W#DkRS8l_r4*D(u1Cg;&~r$ArSP3w&ctpV-k4jl&v3u2yjAc>pI
zPWHpDH!vimO!6U9rwo6y`VGbA?`BlU#J|bXzIp-{vB#p!4yWRKE&JG&Cq+;z3O=G7
z^WT<e);7tw;PPJ@*6IGjgaP+UaU()_eI_v3m0mD-j*P=}mH@umnq|TmVK_0?VV8<<
zrDFD%X6IUNzn_x9n?&_A>12<+td=UJFhvfzE)0QfGnnsKrq+weIELgm=C(I35wx5H
z{S%%0OA2_duqd*P=oWtyK&{lu<{b^9Ul5l#Gc3g({xYJjG)eoKikyR-_@)FRU#Jh3
zOBbGSlyhQ<MZ_$%q#Rjgn+$v%ZY}=g<4;~YTjTERt}v21IzLXnk+a~zRncaeGQBOd
zqG1&At@x;WusPC(GO0<#*Kt!PS-{PXoyl3X+#bpGmo(#I26T$ab})@Z|3Lecw(9u-
zuUMrrI5D_4o<^Sct@l)GUe)^~x1T#{{MxL73(b4Xjk@E{>MPz*z|8a|#u8y2EWwB3
z=W)j8NA=k9hPA~#(9Xb6#WGTawrWqmWvqvEo#UX`(ex~Gxs&YTa)K*;`H~ge^5KZ)
zr7;-hmTqW45hjt*8DmPGz`qhU%mo&w92wM}3`^X3FMS|)rTKa~41K<IiuM|6=}^3e
zw-&Z#8i!s_Wl&((Am1;6Uh{SXLygNuvpHox<9#SKdHdqL@IH*TG)*E!pwzf~M}zop
zqq06T<`~;HhAA3IX1ibXj#3)WL6C*OP`5!c6XAeVr)d_+7|k59b65C-9m)htKifb1
z$K-@gDr>+zq(6=_SaDE5Y$7_YjndvRP5WS@xwb1qk)>};on^0tX7shBtTklgQn{G0
z8NGHGJ-pVk=4>!8Xr;15l$$2EAs<mnlJGvcjY@ehQi+=Nu$Dk<Y<f{oc_qmM!NYRi
zhHI+<em?h-mG7k?0T-qqu!+wFmapackH}#@ABL=D7!N13<9jimFzbHwnk)L<Q()pv
z`StabwFm<RMulF>R-N}ubzGzTX`|8gv2`rMKrhhBaM<q7>MP+(>wEEQ_b?;slhhzt
zqnPYPkMo)0#|WGclh`Zk&v{vN98AlGGv%5ovhSbQgWS(4L)9LN)Tn>kc4$z+J>zLD
ze;Oj|?8aT#*MAQmDtOAZ7jJ&M!~guYmOp*%JJ6W$$mt&bSqwurdyMH5otT%H-IyKC
zlq18w2R5X)tCFufE_}V4By`(cpj1N~-l0AEEB~4?wax3DW$J}#g3%eSKc2bX8G1L-
zRZ5;eEUQ^qw90t$$q_R(gyh<_Bs5(=_{+H;!wr|ZV9G)3w#Nt`NyE{^?nIZveSlyI
zxm<N=Lj)j<SVpkO9iIGuj%GE;Z?$$)LeTJ0;H`b-tyJyB>8_GG+=seLA@S2D9Rj6H
z&r>gWR|X3Fd+o`)6-4;*G4_<F(neb|1(%%ea>L;_aqcP~<y3!Hhf;q)d;o?&k8F}C
zFV2y8@@nub<Qom_7d)pBt4<wtcwP_Rh+Z3;YYPh2*y}$amfBSquo=r><sgDMB5hF4
zDD!9csN%%Te4lSitx(*3?742`Jt+HfrgA5)hQDcbj>t(?GfYGD&BRR7T|f0$`ncU+
ztou#>4F9eYY!-w%Sy^k{34K-91WGM)m=nV8L~dq|&IwfOrqLzKX6<*6Xrw6LqgYTc
zT!}4dn%{&nIp(OIJRUDQJz3~&MunHqD|R}M#ly5kX%EVw7J*_!Y1-wZ&(LdxXV}2~
z<!YD~(VMK>uXk@0>TK4l#hO~Aj&L|eYA;5%$Mc3IABcaMX}`K^mLd?3OEY&+=^S<i
zCz_EqHV;&~u2Q7nHjG&J*a6}fKr;igTQW~axNcByc+Ma-fL!xKTpi-i>l`c*HV^ma
za}W6e69?=CS3(B)UDl1Ob2TWN%bwZ@#``e2yN*jdt%&<JTaHzbk*`xtwLykNP}V|D
zXUoKNKKo^$d4;8jTa71Bk*;D-J8N$tF7Y`->xgx0xuqv4Mk}7(h!T||D>_~jJifn*
z|0LKwq6y_B+zd)})7-Vn-za?jZkfB;b;B^p4zM$O0bCI*@c(!TY-V{#3Q<Kq1RMJ3
z4_Ii3W`KXU*TAy1AcIWAvtMivpZ=TqV6y^&IW0xRY|m=_mjb@hM9&+F9dGNf1mMd>
z%_GhRG1fWw-#2~K29Sf_-Tk<YxFY?-tnS%DM8S`Mc%P_1_WOyM%;{`(WY<5G&;S7B
z2o|Bh0H4c}1jWTn<}}<JuiJ`}ZU7K7rt1S6ui|jUW0Qt<gl?^PyhVilPAGToj<^#t
zhkDoa@1WhQ?IJ{BktVgl+nYwQg^x3t$_vZIGRJv^996I1)=k0Pop1zRN!_7s!)Be*
z`L#z}!d36hW4cQWSu=pL?|ocACqOG`_?tSi8}9-69}VUmU#;v`A`M<gkQ{NpA8SMv
zpTi)ifx9bXd<Y^5t}VS@>HAbv>%&cl>JyEJk0=<6vB_45@j(TEn0~=Hr*0Xm0udw$
ze#X71yM18FkwrG_f<oTrcNoU=im%pdKed6>)BExbOmmrz1wKY%KrkXgdmsyx-`d3Q
zr7vnJFxilYG(r^_xp!_q55AOiTbG5!yhuHXi$d`HpqLs-ae{=JwF4?eEh4a=FzB_d
z<M7zY>j7rjIe}30<a@XVU}P1~u<LK6M_RBQm1mAhKAoG%XYpj^N6{jn(Tdq_45_(n
z`R{5(%UZ|f;~O_D!nqBawD+A*T5iUXI~rCG9ykMzG=;yqpT(F2&6}+~J#{AJP_G%Y
z0Q=;U9|egPohzwFVCbzU{{je7tgD0^qXFP@22(O}N_rXezGb>55@?k=JfAQPRZJp_
z6an9dBIbcZ{P@MVbc4u0=Q>JqeQ)yZcH+D}Zi1cjDxPRsPs)nxO8Jy)h?92}n1A{i
zpxdi-M$LXqIaY&T&vNcsp#LeSuavLGmX_lnZTDrAC1ktg&Sd6NVb3p%q%<kuoEDOy
zs=Gr9n_+zk#EV?PB=sc0mHw=CvM*Pqq3)DMP(8L&Soxfk+rsvk|HrhWv#3rdbYvnF
z_HvSr$H&!RHH{(O{lf8Gy$FQk122GQGl;L0le*$OhtUA?z>MHVIA6hHBgZ-Uyo5%5
z>fU#M?X;s?x5;h9CTemmR>6rw<}LS)kc(SG{@Tg!y(f1-i(FQ|(x99ht#wCA3-!KV
ze`u|n?20?~rZ}0SFEyUbOB5hUFhyXpI<|KsYQi>IlV6}xbKPC$&&&}R<W06yynS^I
z@rQ!O5Geft!<7BS%j0KaOub`<TJFR=!b$<?vDZX4-5FRSqYQS#eZk!B>MwH4O%3UN
z#rxHn3<me6#JYGi=Ms|Uu4_wMR8t#~I@69`@}-PcI2{&*AzuY*G1V21R{Mri32d6k
ziXV2`HhS;wk+{EKo=TgZKYYu>o<uw2$?C_QtiV0;Mv$!NN(SKmDi4NviRbEi0FjO$
zygr5sdx~Yoj>W`pMo^uwZzTjjS%309cIm}fZ>RM%`cB9Nsg<`YAa5|aPdW==;V=$X
z2Ow&Hy3YMf8h8J9-%XQ7heTnMS90$pl)#?e7>}_ud|@yG7_#2ud}WESC%x^V(WT5P
zYZ*m%veA4>0mFXIV+6h0BF-V{i{HqVfEw2~;9QdL);AqPOT}TMnqoP?CL&i?oS1`(
z`v;yoVDn|(y8ZL8lUIJ?j<Khwbw#q}8}^tV-Q`ad%?<x&7GTaW@T$Fm*T|s8TV}I{
z!$(p8?uZ$9tOz7YL<Xw+@?DWRUx@tkEvbWcYGKDIb-I!6Qz}Rl_CNh+HxrB3!gs>;
z8D%aBx?Nqpq2k+|MG`fie6uEK(1h^mliTR=rSX5@bS$lU*lArwHV&IJ4oH<$5JXy)
zwSg0tS+`&qajo}c2QTavkzJytD*dr;ziz=xuvSPwlGi(HSWv#)E$!cn40tDnxA)J}
z_kVLgZe#Ix@fyJgkJWYJG}pBPSb+MH^|gO+O@7Hr>WX9!8>Fg*g>>p?uBgB(-;!+_
zXnil^0-hBKVDv_7eu44W7qM9h!Wn-HY?X+MhBJrh!79xHCU2T84oHT7bMN|dHXs07
z?b}Rsu03Jc=sxHx_O`l@-IBIFCk8cfq;#vIV8nLB%yKnqswgRfuI<W1GYhV@4U~!!
z1K1$gAl&IHA>Bu5NWo?rx9mSHkh|+6gzGtdZ8M#ck>A=rZZjn{)K@glR0$i#t|fWQ
zaD{L*ze$V1?b8^MK!L6EH7w?ISJB^Vld>8P=xed_1$(N>sK_^3r{LXGHRRtbj-Tun
ztGZlj*dic%5Zo(RSOfKsJqkKS4w<2ov(wMdpBC%NOwS@-VXgLokf+mvyYUsKkK4Ub
z?v{|po4JD4&XmoVqaBCiR*@RyB=mxxu=STfZC^@XAz$BNz7*M4BOHZf*XP=ubw>~r
z%n1*l$d;!WjC%aneiBr@2rH5>gga-bF?^Me)l1|Pqb5zmRlrx}iibBJg>}BdgN%jG
zNu!BKPXG|zBX!dKa+SW1XGoG46&5O$fYz}_=I8xVl0ua4^ISwldfnq|vQ}f3+kG~|
zIXWji_$IixD0)Kr7kH~{bJZNLA3KJ53-5)gFm(>6JQ6iQG+vgaJ>og4rs~ZJ-JB1o
zFRcu&pKQORQ!j`~=sSI^T>ElZbWv4-KV_v(enGUqcvOlx`DU4Hq>Xx}0)PPM8~nc0
zVTo7e3Cc{0Rdpg;0ZZLNa?Y1_!SpWT_QSZ&-l4D>yG+~{C<t@m-**r8WoW-=m{gf)
zwtaVNKI9yvr~*VTGCVid`ieM3hjAJhR(yOH#V2$)_=U4mJ8gi|5i$*~SB>w!+2bD?
zoqurQ7Ri-kkQO+3<$HqdyT@0M|46aeCU5fKp^BpyLANlNxQl-w*dr)XVYW_w%=a1j
z2XqPM#!<x#zbs3~5VelVioUG`e98hl0Xt+k{J<`9mm3xgoBPrKN`MAua=w?U?_A+C
zZ=OEZo5FGx2+!{88)4Er(Y)X=(pxY>zZyBjvp*G<aG9up$TC`6+_op&>)JcoS3Ji9
zWz;9bk9DrTI`M!y0gch`OpA4s-Vf7YdcUm!ni2c!I(QL$3zI7P%9e4t*4|eVK58^r
zGXES0*@>91z*Z{bLW-=*;>T^w0BN}2`-yDk^wM_^*d`z;2`O3xI<XW*2mFL->u!(o
znz~t<bO`^pTE2sxZwm5Sw#nmDkk=jtB((#7CEXR2X1PaBllB(Zd>Xtj?SLY>%b*ti
z8k6cKlHggm2<jE8esz7%m4~Z5-x~=5<BAsH%oS#AYkg|1D?p6|7a|bCM?66KUVB}6
z@Tck5{O2Cs-kC~#V^>{I88@O+(3`YORF_35A#A|T*ZncK^h153Zmy{&9et7hjgGM(
zrvRu}#ay=)x~~3OjHLuVNh6@X&p^?>{mNh7DzM<8U_(Ko2h2xO<YLe;WSgn6$A_`X
z;I22qR9J}2aHR~YGTC>)nmcih%Hl?&o8((PhHm$Q3Is7d-};2`+=ZPA2X5oSvBi(W
zV4W2CLwUC6){N{1{>@nExjn_jy=DbIGj82*H1A+gd{+lc8QqK)1g-CA2i_hOG6X6#
zJ$3gv*G@8JTTG~WsSIOYlTD@K2cvo(^Hf67Z6#i75L|^5O}6Fetn7tLTc0nXxgtHY
z41Io>c=hgbtXQj-sEEU46TT?tS(T`|$UgA+{cV_*V#w*_+bco*OI;-ZwUlMF&RErr
z=Qey^jHv5|D)RB{Kx;aEoNV&lA@JIZSZvEgAFb|)=F~giwpULt_4%`IJ=zE%cP#MN
zHh)HQgVaf_crk10*RG7y7(V^Z3j+%XZX9j;z?=8`OSu()ypA)_OM--PWeULWc8MR-
z$^GXLR{wt_h-Y>MHLOr?5aO*Y%|%}igp;hq_}SbzK4pW;Fpmq<jWAPw<f&P4$D1Yr
zX56jOaf4^CtHFZIb}~E52eZs0)6Tj@32bjpz*~h#;uzjEZl0J7RqfE{*UUrna8;=9
zv6fAtI`RMjRl@Q%-COc&$i6BHm1_mvUQ(L)J*}>_{-z~z#jtP#oWDY{0ExiFKXE?P
zH607m+o9puTXgkez4e%fuz4{FzS40wYHJdjX-miH@A_*zdHVkFb#xI(?$kpdhF#3V
zY2By*LF)Tb8O+A02%|J}+jf6d8b#_6bPMm6rb_(JFW`Tw2jsJ4;=)2d&sdP5{{X+s
zk}1T3VvjyZpz`DrYmXgcXs3lCh2(JY3vpTtbDL1iX@+x^hW$sENPtwUHIw<C!GN+u
zQ=ICAY^w#itN~wxx3y+!p~|k{5Wv)nB+25&u1Wc%#<B|fq&|7$Lp1w>b{Olnnw^8&
zeCnT*&I<=VUGcrDgw-y`*+INoJFV4-<|NKnRl6Q6uPaWHge45g4I$<z?d(jAS1?dM
z|36fT-dUk(IL|?kIyhG+jbfvBFs}ISjfyC=gL7vaZz>f3Eo}LGC^&#W@<u>>QQn;r
zS9;3&Bkmw0C*<lU-T2kw^{DY6sL}!>Q65O&3sTWA3+!uWP-Ic;j3WpXE8&%+di48a
za7NxdE`mO39$_pvip4j#S8O9QKrlJRO^mOn407s!bbR}#umiO%S|@434SZ}n)r}G0
zzfJV$L!Rzbbhzu#MiN)_+E9v<1(EtTsrt6=ED>GoulmBKwk~&)!ob3ZCiE0!Ycf{>
zYTVLDgZWazxpy<KLNc!UvJX5EK#;JL2X5_#GvNyw>zUZyn5m#a@5UQ%65$YJVB5vK
z&^qE2s$7aG+x-aCN)=7)a8B7<p0I`rM`bCau0acuj0+9HYpILh#Qpj%sDhO#SwDH!
zd`Tmxb_gt+z*fW1(|ex<XkN*tHtT(1|G*4fZ^<~kV$QF=X~sJ!!OIiW`|3BiEsnJR
zG=wfCbCYxY5svUhqdxr8+^04_f7z!Miopy4B@(xqE?w1F6}!y}gwSmf@e*<IxDy|O
zNMZiiPkaLYScSon&Q5!@I+tdE+=@qD;E*<$WaJFdTh=#IAoCPcmm#LDou_i{&y>X%
zq#USWa%xPES-ymBTqqxv;!YQcNe?^B>0j0h6LQ!;Z?5hfTIEOU8|Yl48|ia>7d<Cl
z_^-#jO8{>M|4*#GVVh`YLA=lps$R8AJz;y_;M=AnVjhR=2aR8)!!U`khfI0DrRZ7k
zo4oQPtuHRW+`U2gMcX|o_9Au(Tb<4sBLK)ta<I5=(!S`&So`<N#HOPxI$1|gKpH0|
zd?nSoE5o*H)IN2Hl)32jl;_WxE?i7w&_#dY;uW%7s-)wvyOmJ=LB8aW+evyQaN+jx
zMX0OUN$-s5)T|(DtMw-t;h!%(6^V~q0PxaKaqoTPwAM}#Z>6P`w9e3u2Gbe;(<kW1
zox*)?A1Yff<p&7XV!bu`Vy?1xC%+DgrNIshxR6~?;TGI8&erqt3cr+1gLzSB9cyB+
zI|sPUTrEr%k}f#}Uv!Cg&tPs~8pPJ$E|V%YVr7{QTo-=hzUjrJ$ns4H-w2kdO*EWK
z(0yBroPpFl)ip3<M&DT=%#;K8TnneOg|{-Ml1cI*mAyY{0Z_|>+HqHV@f<a9U&>tL
z7C!8B<1#JK4Ci?Z$&{t^RR6Kj((tjrSNXSs9AF(#HSjzlGJX?Df~hdDY`$e;C!I2x
z_t$f~UNI<g<a8?a9uvdYL$whqg&PhFf6f_nQ$gj?$ZXI-Xa}T;qd}@t*cnuu9MALx
zZ15O@;TH}RzrEx|DTx$C0IzIj9_l*BW>Jy?e|EL<DAl^x(72N{qW-LDoSTfy`twkY
zfvBVVW1Bt35_m>Gaeg=E3noPwD(n!?hr$kYP__T2Iiv1T0PmsP@l4tT7JO}2%p*O4
z@p<XS*rC8el0$7((A2oF(`~8)x&`>&1p*r0$=C5r0Vrq8@wcN2ZyymVZF=6n%3V5Z
zLt84&ke7T|jQx(i$Ln?7)Yu{qiiiMJEKTDwlOPor+5Y_s<g&OU!fAJ2BDM}lx%p1H
zp+1s@sXz+O6UKR<-jGvn%D<1WDbCFrg+UT96WTbGa^E>@x+2ee-C}uAiep}7mUtky
zf!nu7a~gOg67M$*Fa!TNd7xA92AmM=bjvk}^)!6J$?+b+L51Ey*iwH-fkNHb(<%^t
z1*pj?X?e_Y<<n1EM{XzGoyB(An-)bd%-u{i(^=fFL3l!w-m9Z7$O78Ju;4Y%1BJX2
znl|_qb~j1C$G&K&0lxB(#szfkXIBrn?@Wc|W?{1ya=Ggv4GG;LbK3I!U`Wl5GAd8u
z`5q`%XB{(~*`|=e@FgjL4nz6wDzFvi=s*Z^(rK3hG|F?sjFz^q5pqq*Cbjf|dHxKN
z_2W)xc&Rk66kf}^g=Xl73GR)bqXmHbO#lnD5t-?ech@Lks8C2S!aK^zo5z)&SRCp8
zVZodG4=rrDxt!2N3u|UH(U8nD8N^szO!vE<$Jxj^Qfihij%0V)=UVn-C=AcP5zm0F
zbS${*SOdQXpoTKd9X-$78-ZNgbaRQ|=K!qmc`u_FXGOCy4Ni=(ShdH5D1JF+I&zTu
z{lH&*m?1*6QhOb2b;d6WlHYi;#%3S}4_jih7O;Zbk)QNUMT>spfhOu!(iQ+sKaeZF
zCD$gj{chDdj{%)_BMz4}`d(`QEFkxWS0UBU&1!p5CX~fxy*#<L5b=1^p8Rrr62zpQ
z@rapz<P3huz!X>h_(}Rz&6mYwDsLyzttC)_^kz8%SPZTG<WQaJO-74QPxhEeIv8%^
z-jbrRujj5+a&1*NN@IRtIgY)A(=u;siEtwGxgf=_eERT9gL4<n7gd90&ENj=d#Mpa
z(3lR)4OwIV`Wp=6=iJYGM>DX^YSY<DxXL-=)@_@WchJB5S9gIL$`4!!l2Cpi^kWdR
z(CoAB!d|(xK!eZIe(<Jqbr+A~VfyI|V^~U=mm8a3UNX^+HThq8FcPqG*?GBQ*0jB+
zvfkfS*YzDM#jssq(iI}TC9M0_cI*RQ1w$7|XrJ)e-J=B_q{TfrENBzA^v*62R)4lr
z0Ly_1>m1SEA5LS;*Zc8m36kv7y%^JvW!Io2y<%BCMGf{>xlq@5Y=-B@N?b2iHs%oY
zBxpNrcHYDMfCd4VGhq{qr(~`$ds`#UOc(RcCx7x8;eRO}5&+Gn3d8lF>Heg4dn@VF
z5(v6p>PnpQXNP!e1lbiBdDa|K_E^E%jN^1860&0bulRD-2!Hz<tgtmm+eZA`<^Dv@
zHvF_?Elv`P0GMyVtDgRo_7!MgpRUlNhKIA%F`N?3DWM3K=_OyoIhYOvR@0TDZ-Je1
znGEclR;xRv!iKOs%a(C-kp16D=eYyJC7AS!&tLjW=P4jp_tNv~U8_&;y)2pmyUjlG
zKI!yzS5|a4Thu*QkOgujU(w`~YQ&JEul$e^zBM-C01-ol1qD{X0^Nakrv^(LU7lDv
zhGVjKs~4TdzmCgJz-k1qI3_3FVLQ_}RQ>EyX|5+vB1K*zo0NU%bd4~~b6Qe2l|SkQ
zc~}2G7pEs5$|O2ExK%80=FX>J#{FuY_gT#erPFbjw%V$5vt+M*$^h5@Wj^xg|G53Q
zNCn7khyR7SgC@a9Nl}qkcP%)8OJiDZ{2)HQ#Y=dI-;kZteOn+%Na^GE5!sr+&(n9x
z4#TU14%#y<I#-DydjCFMcS5XWP7njUzHxoAfAUX)aNV(YHkK7{p6;(n%w)>Z`^>Wp
z)`jPetAHP1Kxu)ejbbqHD)b>eCr8fm^nO$z_$L@nlXmwrSda0Fp$uJRVsg8PFrzOg
zZ35!Vyg~YYQL80%q&eet0H4EiNBPMySMY5=gd+uT9!gehSPoo~3WsLD<Oth!kx6=4
zOX!~Tv&vKLBe#OtSU3EYqz=lLOXo&SKIMX~OF!iue0sA+Sw{Ls=WHY?^2J56U}=H|
zojl4Z<>SJJ2ssrk!zI>VCmIxr@t2Yc)IaQIFohu_*h0=YdJ$ix*80GjbC=$1(VF){
zfXmsi2_N^I#ckbB9~d=L&jOLC<GLvYA4(CBjunS>KKTdtwb|$rCp=CX^1bI6lzfO@
zG38h0*vGXnV#Mfku$D>y&)zc#JB0H}6)dNilAyT&nGMdQji;XUDjE7;&Vo9=vMV2L
z&9wBwn_oF9gZ?pZ0QL}n))=3dnEC!v3zLUYdmsr`&tr+HN9GVC*GY;GVXg8M!Og>b
z`sFs_5Wu=WIbHw~)`5U8`4$M13Y6zS6^S{I(gE*O!h#K)6|}O+IkG(L7!<iiUS^nB
zWM+YHH<$@%<DZ8!40%i`M=$n3l|_H2jC&1jQ@8OsG5+UP9M0k#uQEvIv~V=7Ko8C6
z(9IfzcNVzqwkIMx=6hBpU?q3WIe;2Zujbfk7{-LBl6>mBGr8=fMRXWoT=w$YpF&%V
zhOqbbL;=Z|3ASZk)L=GMMVgAevFAu-blp6>Q&M(F5DEL8MxPX2aOf~n;t#yyuX`JN
zdnfc*)7{x3`;jY@5r4rS%67K=_QL<g$8v^j@(ZK;o^k+R=?!EsCRFi7_FyMp7+Xlm
z*X97$Wa`-Tn6h)0FKTmb^Pn0bR6k~!ymY{8UU=`PA|1-z%5NVmlB6scl`1WeYm)Tk
zz`YdgD*EX{%9MR#f%LroAaFQu{PcMV>`nt)jjht_5)>qZ3Fno(c!)PoRFzMMw+jMt
z)W6MVM2`7PxB6RaecOr<3lbjJNy57U?_Q-lR0YI#T}I`-^ZPk@1g7>e;tCk~=XL{x
z<zWd=|2?KKAJu|lP|*X4!i<oy^5%**nyKBSDvG&mk6D!!Ue}@VRWoQuDMSLf`nA!M
z0%N@jw&ps-=4myaNw$;XYj&49nPsq@+a)M*o^^fe8Lxt!h%v$N8Xj$ErcsQHLM>Fv
zN@`#>E{IJqD4!i@60oKH`wUMxN?Yhj!vT#N**_@dhlW7)%>59A-hpzjR>sNXTGK=&
z{ES=2%@>b6AaT3KCcG=h(lHjtv6;kuiD`#-7<f*%KBSmafi2(U8k{_9TDV=Z0nyr1
zj0R1B&lkB3y1%4L;u9AxPyqR2g3r|_3@Ld~D>bqdHY)R&y`Z-|w(#)GBf)ePCYeM5
zl!XTyHBh29B{rIAFeN9inWp^Ph48p*Z_A`h6l04mplAP=U*OTn9>U_wxSqM0tb03Z
z+W$6X$Sa)V(NqI|WOkc_P-A$gC0DIQ$-~?r1Jo#QMG$uJ3Zh=@u=028X}IKihK8Iw
zYUEQM&g3##b=FuP*ja-=>fK{Xm|F-mP9M(*C6<e~-Ch5fAoj&XF3P#m?}Aghp?Qh$
z319|z2G>VCV_1@}A~z);zq@vvGMX=OzBlpK@5<!0TZ9qq#f$hHqjN4vP$8PKk`{nw
z<Y~g04%i99V(S#JVfrsOQ?i{83lcutlXd8fG7hM4j0g#m;eJBbOk}l3s8<xo9=F9A
zM4L6H9hoLp@_Je6$!3p22zm`~9b;KrAETybRdS*fdHHx(aN!u5ilJEnJvb&TI*0p6
zRnK!6jR3jVP)Iy0R4)b0!1aF(onAVSsLl0`I5iS1<y8PSvY-yNn&Yi1KV@HP+*p4s
zO6TK=l-p>9{;#i4bZ);1ahN~EQHNb<#5KL|oZcc^^ohoctabE-txRZUAQ9+g5Pag2
z2wI?Sq`tX(3W|JY+-5y`ROKh+%ZFRgMUA2AV`<XLGhk9_7mOV|AV^Hbi{b6NuFZUJ
z382D0j^*6c_z@hKqNkayce?Y40+>#b;F09pu_8-|lC39hKQMrX=HuD=!9dH5n1rRa
zvDnpgR!BfSNF{I>W-7eUNqzv^lEQ1i9C!@Le%L61g6=BG0u3(wJ@TQ?4nSv8?)AoV
z2cfV11PQ~u>tByUkcN<7^L=LXI#Az%Kv0oGkt!rKch?9bgN?MMF&r_tm6?@(WACcu
z)Cy+Jim;=-_^4|@S%KL??93Mikx}=c!`SU#_Pz7kkITH{>uNd)B4=p%TKAt!y0P)8
z*dLfmj7cnOUn?U!xr1)S`DPjl7%te$Qn6#0cIwSZ!ljad99j;T8+X7$a|Jk$vvRSj
z82AOgf-2aJTMdHo`3rT8&%|Utj&_K}nt}?yrW^T{rK=<b`JmfTojCKw^O(8jOEy%K
zsuPWzTO7FlHc9axP++(blkhn*hL*HF3$fD4d>fiA6hY=rqA8m_)ZGg+ighB~ryL2z
z@8aj8QjdXp-aAgF6*5h}YXDyH_AQvNpgGZ}@Y`FcO!e+UGPDa4e<xLtph_q`n_cj_
zp{>d034sy6z#y-3F*u141KtbR>ZQ5dvQeyM+2S&&UvVhnVjDd{Uo~e8C_I}=it*PH
zhDLYX{uEorQa(U?;*Y=^NrImx5kuq|7fm;4IV?n8S6uN1W^q;#ejR|Pj(D+0Aj1}e
zBTHd)zJN&&3{%x-FDy?UAu)fB<L?{o?6V%HxrI1bPhZzvtb3KQZX)K^Wc3A8Pl|A0
z<n~^*c_KB$Us#%bo>zyS@s~;wZ_@GQ``yDqCZes(2SN?RE&76njL+rHl&^%&It)r%
z%ub8B<an^^9h$4T$bZ9bMd4K%?#VmR+9}(yt4>T;f^P1M`_d*=w<{t-suf1F;b%vH
zI)3^6`G^jxHhaZd`7JF2TcWzqmF7bbnx}AdNgW>k@1Z5PgIlDx_v-X9U||H9&WULv
zs^*gZv-kVJUqyIW*VRFHud+_w;3IXY-ea@Ek75!T(nNFAEm$Z7=>~J3b+^e02Ys4x
zqts{n{sc5dpw+4^96Ee2`ikS)a}RBvpV1IO61M}*TGzp2ZdPgT1Cbj2rJ$5I;wBVO
zYeO0Mnz4q0W_({nF&%4W&mV^znPj8ZNivi+H!Y+eSErj`$|XxxU0`C2MmL`K-w<%)
zV`oW_0KZ#KnWPVq9{b0aBQ~G?qxt;2o(iW2IY6{sA13~MJapN?c%K?qjd(^c!wH7K
zCs}MLM$C9@Eeu17JihLLgZN{N4@`kU?CvIy)s;3wYWo5Hh~78WLj_QERl8zDdNS)E
zyyOvONFB~5+F)N6Cxh(-torg4;Um^!<)`{%(W4n}!iFe|7a_=SRQH3~tzW)v7y7Hx
zRQyv&&ED5&Lu6ZImigC#vKK+`*%L+;o+ALdbjI!nEl(~9&MQqcUi5|Ie6;u7mF~I$
zX9)7#5|bX#YgAg0TvgR~QRnE&)@5Iw2IBP@gqY4zuruOr&;*34^;jw_69(cOG4lmw
z&iDr1(t&!6>m$xrvbt<(;FlvW=dYN){K$VggwwdCPfA^V&DDupT-UmvGty-;c4N*m
zzF>khVVOyF+0p6mZ$l<k*R<Sox!-zL@BA*~O>;r4GPB)*eGR(xBoKBbz$0fib}z<X
z)74U*Hskpjx?F>xi78k%lNV|TC*>m(_zYRBypS9hCS-%9gQi|Vu-~Ol?oi4x%%}V{
z<Az2OZ>eDEZ9}o)9SzQO3CusLLJr$8Wqj*BUmn}JgCL#3=ZrVLLJn@;(rHGKz~j;*
zo5COmZ#UoyaLlQG2xxwK!0(Q3V$t3=y!9b2!&3<tfLSc5{NBv#djVIk{|~Nxb|DuE
z|9UT%(1z)1z+Vr1Sghp>OK6T-rb<Gv;-~z;;k+rjba(vq{%)w!mL{r}x_+LKxHhvi
zJqN5&*x4EC7Q5M0LBy0&Rt4CK{J2daniw1YXyqq$<%`nxm~2tkh6k6tEEI!VtjwH^
zcpoQDHcNo@lZ=xEv96(RfoGcI%S<EYtxXxy{Gy{Tb}6HcVTMng3X;oL=!t;y)Zr8d
z8qO4aG&NpSy>y!6!(mUUOT@dm6f0;da#U(naIU<q9XNbn&BX>Wgaf~P*@R*?CosuF
zBtC-*X#v6^B&Ujk%!fj?Q(yCCn<((p_-6sV5P`kKoaqr7@FQR<C2M`S!2DPV9Pb<a
zM<Yo7h6G)ImUS92cGoz@vpPTK3nL$S7#mMt&TeALZzYva`SMscaOm<9=!M;bt_!i*
zi4+CvDeL}~4!YLZ>`w=>j8+XOpBvdt`f<a;vE3@SVmn3<COi%)iK(R%9XYAH6YG~>
z{)qCxdFfl+dI=|x6m6VsDt@Ow070<0Uv2IEMnlt+)-H%1=gPC%G&nnIdzK=)=(r27
zsZvb6a3(J*Tr0I>{p2nJ-6o<;8p3!&6Z$mi5S=P0Y9S^3#lMbf1iu~n;$AnX*XV78
zu=-k~YU5(o4*PiUv2ippt`I{Sidsw9aav_v1<lN%ApmsA&T?N@_qcAp_s>-$wun`P
zOhB&`Fr0be#b^5pYIttf<ghav6a&DGIckn6-lPAS1t^mSjo8SmuCIknsU=AYx^r1q
z!rMCQ7zxEGaCR#wbxq_kw3WXnMi`x~MF1ANP@`$1;lxbpzM1SL`T|-GO1CHmo56WU
za*WW3u~K&7OMD`RCHBaWNll<M8L2QKIU#DjPqt^&YK^27<-N2ge1?*ar{D))x$6}{
zZOB?i$0+2m^(sW)I~gEmzRAK=9%kjWWuu+8HrRx6H~h4tAvmaRCc`T?)%(&a`kf%I
zNo$&UWaUSErEOgprhu6ptHZN?=kSpF^*hEIm#Vu;TQ;Z<)Nxi;K(1lj8wZM9nE3hw
z-4s1+{pyiLurN*x)3nUi6dA~Jf7~w(;wVW*tk6BH7-%XXX%dj0KYS;-WnGJM3ieT|
zr8z4q8bOn(>}=rE%l87Mzm}-)^EqL3Mn*l!Dl$~;;j=$lo6N)zDpp(`mecn9@=qUc
zgRYx{MN4JXD)YnDW!^tRzpC;H(DMkuT%=P!IO65EIcBz1XZ&*8{;h^IsQC5%h|P<r
z{x0S>61u%YTu{u9f9puig?Ef&^`y5k&^{+1OC`(mqr6;YoYVcFU1MLV?8e*Qq0<Eo
zLuFy*rg=w}W2N=zPPK3a=@rl2m`@=LBLVfpXr3Ahj2E8DaY61<c$Bp5harRyzucGO
zi?42o{=O|a<=?OgrjD>{eYsZ3sHuWkO}hM|@UkI;O(YUvd#R75LU~2+>(%JrM6|4%
zA6H*CEU!<y8Jo(=W;CLU`e0L|fmCLoWXX&moH{-P=t|P=PRX{Of;4se2SN*3CjkWR
z2Sdcd%kj|~$7mlfj@n?#1wuzmpmVCED?6WBNdZlyL3AaBX{>6M(v8<2#)10pl;LeJ
zZfb0nV`yuL4L0<erzpYehJYjMmRZS%n-_{@FCx_P!DY;iu~3njXNvU`D%Gqy*^Jk)
zi=WA^vFwXvHP>g6_;q#Y1zI--hjdHQg=h`^N=~gW6pwxRdc7)nEnhB1=N?cyc<1vY
zZ)(b$aSG!h9@yB(Q{F8Y!EQdXT2qwuA^)wsXEU3-rRORg!MB))z$V!SFh|oyev9t%
za|=d!jDc5ZY28U84bO+q6W6mf>ghA!+!SHY_)}*Dvf3Rx$=kRo<67-#p62y6yL2Rk
znV*7;mzOAZFIikIf)+Uc^5O)utnN4uRfiZu<*?`bhdI~@tXlkigc{=a{iU9716em5
zx%nDe=Axh*VLq!eLdCG_{}KHS;X`hAO@RwYjC5fO7xGbFC0(+F&5CM9{H4i7yt=E(
z6c?=z6Zx^1?#C@loJ(Xh#wd6F^5mvRDxuI#@E9%3M;@rKb4>blM5}L}ogHM*o!+nN
z6jt{1&Xdox6@86}sR>eq6CNeN;HOkNw#gGrEOJBxzcnc%E17;@tC<JX4cteT#4u!&
z1@d4UJ{ptWp&|!d2a3r6^%s{0J1K-jRXDDhPh&(gJf?UFm7nk9u3zL$SYQ3Xg=j>y
z;s>m}?M%Kf;gJ@3yiH$R@6%4W8W`sNms#B?O4+IRkFn-Scy{fkQ0kRm91m%H#iU{T
zO3)=A-Ce1pfV)ok>ReB7Ubz0#Be|WI8n^qC8-Pj|@EMjJqj4=86n$>NHgQCn=Z8(X
zI}YoMKK#har5bfJbY1rysKT&J<Qd_JU|>|$h`iaKht#`~O22+yAg1j@z^pW%@I$&R
zSJC5$vRI{`WH5DI_v%P7%(*Q`H9bo$t0ZODY(U1o`hzS}v9W9cO#O+m0LWBBewgg6
zT`YIu+O{K2&G^IR<nx?Ff5ILvqm{W|)HN)xi2eBHu_4(@vdLDA8Q)a|4=Xbo*zb3|
zeHT`8;n_6$n)&TZDwxl!&0VF<sxdSG3x6!d)YVI3UjV*zUry=s`)-cxSufezg7c(%
zo%27+;Vtp-|Haf>M>YMv@x$9_>24S>x<O!c!w6AO36T&+DBax%2nb_HNvRC^pfpOy
z2<ecJj!8|rl!iU~{(jGM&U4PT|MuSA_qp!te#LcN?^UV?({r^iTy?L#VkCL80L&k+
zCX&q{fn??F7Jx{L(%BXYgOgLYTm9&3#D=O@bXecK2>}Ut7^R<9-ED}tLx~LteW*Oy
zov!c@yj%dfmc#U*<X?6G_wMHk?%c02^v*bp8oiT5n1KLM1UDAg!>I9kuaBDw$)CQF
zq`>4fq%oEuiQhEgTHY_SIn(&))b(izxOA8<n8961_6gW{+RC=CT?hJbFDixfu4;y@
zRC>Z-?!ny<Njky{^M5ib?}&BxqqhAY&AFI~i$B&iJy{>WQ$K)bS&+68PH@|qZ^b+>
z(<Aih7C=z+^Fs*E@+^~}@}7Z6j)rJOn=MN=26}u`=a~<@Du@Uv9SjO?Nh5GGe>Q*D
z8LyPiP|9#KkJB%#jJ1q|uG{7R?pnrouZV6tiyZ;x$S{HUy-h@nURhO9x^kY*=)Yzu
z_tmH-Q6_NKFC;c6me*)mb@58OrCf_s;!}-V$Twqp9|TMkQKiAhbL?pZj$6XZ$D;>U
zcKr_@Rirn@6n1(I(fc|eU=9H`BV?j(Wj-zLmeOot4x-mdmm6Wvt!v0@GX`A2-X|)m
zo<hG{C^_p4o&k>`zZ!){)+?GZl;)Lxy6cP}I@0sqJ=c#m?JegHXdsn(dta(h15Wk1
zP7pH310VAR%2W^j4dq0O#(d=n`OyEYkpx+rv~0Hcl<O6a5g&VD(Zk_EYNqpWMYm|{
zX!cubz&XJ~b4kT<ccAj?fx(bWqG}*9r@;q-a+EBeeP`SgpODSWm7Q<Kt_@)HGgQlj
z6Bik=6Pg%tFUc>4bgmi}>v(O3Pq?l1!pWZ)Qq%#D^t3q_<NTY>bAIZ_kVIEr3KBU}
z&GF}RYQ0~7uKUHjXxI=kCq;2)E4BZPCY129pgkr#0G)_lnNXnzml{vsF^@9~EniZz
z0Uo`3=38}$y9Xy)lv=wuS2lcp271Wzi#X%NZafue?NrYR%e!%gfHBp<_s1ObyYT&F
zgj5Z7-%HE@uW&|u(a$Yz>iF}HNWhGx8O_O0?AD-Gg*E(svfE=gs5f&iR^;Xxt5mCw
zNa8s{M7j-V2|UxE^loI8>hFr=J}zwX0}`}7m}W4RecE@4ecb8?fp`iiHtqteP?y)-
z?_OuW7fm+fb5Hua)y7W@aF;wXwXZ#lF~y`!8ZL~o97)afN56%|M+GEs<^Z?@>$bRD
z(-UM}u7Tl%mOCN#-<MoER}UuqbqBSGHnkaa%efk;U-pHaQ4LiAYL=7YZ3X&Q{Lnr2
z3BzG;pXUk<LP&ysUmKdqVxLR+H%!?_k*Jh(Md%8ryoG6*mA9+i2Wq^0g#Rxa+2UE1
zh(?9XXAx9VB;lP62Ekj1k`f&7!|Ev-d}LZf&G*#bWwkYH)f3Gy;nip8eCGhMbS?Dg
zg$0JamLlPI?NT<NH0+qaG%0>{LkM8ktE#m1m2){e|Ian-axEMPu+Wz=^rg)=vYt#g
z;b7GDv3l@I_zLo<yL&bv<Infi%P5e~T~0ZH>l!Sx-leOEU`Zpigf;pB`gNMh7=6|+
zGEW_2!aW59*-yi9g5hX&;I$}x?Wf)eO#JNvZs^DYhB+m^Hw=iM-bLLNF_g+*loAe)
z3^YcdNepJM(?RDT<yWmg<!Ld}_?9XM$XA)ilCio;2ge@vK?IAg!(1?r|8p6|A-BJk
zKtQQ!bNuwHOGOcL9ysOEE6DZgu-K7@l+ziN7?B^{UMD^yindV`!j|e{RI6{8EDk!5
zWnl<gW%3vk(b;Yvoer<=({m2wYi|EDo5Tx)o2Q{ZqFEv7@An?erbI}N$K3`&NHG!Z
zs{~^t*Qdk}J==1p111T*p{G^&^l+sUL;2A<Cme6uZ4lf&#zb8oh!V;@PyW<&!8}Wn
zZ1-pZd_;m$`k?7=(0K^EdO~wbccfb8v-%dLP_PZ~B$6PxCYMrV`}qwd06NQu>6KWB
zd?fH;aVNf(m6Duw3O3|7y|!aD0W>n!;+!FXav?*HW0yDkqPLF%NiS&XYR+XSZj^H$
zmvXm0H|8iO4W=5rh?2`Ie8<x}=tt&#Z;uas%0}u+K7D7A;LGm5>-FHJhP&XW;X^0o
z+<VafDm7<*DO_;n5_CM~#&)K#l3YK<%mCi(j;~|Z<l36TlWISLk1^`UE{&bV%C=2~
zcEOqKt%V*TTaH`&N%u86M!L1pfsZqK!S(EW)X_niY=liTQ0Bcj^c3a$X>1N|KF_g0
z6y_1GLOy-W@rvSv2%vVjcZ$Mxbb?itN`;b<=`#OtmB1C9E8bhcQTvWFCjSLSqUIsb
zvkuX4Wig%iYg}4GKhK^x^G+w_DwbvZR5qwiwWPeEohw9Fb6WA2DqgDmFX6o;mvMfx
zFL`W9PTQ=~IWk5IubKq=5je&$bL`{%XcQCQ!mFM<ONQ9Ff<LHU;J+f;6{QW)zE=sL
z$P`ii(kWd^+x+H<Z~?TY^rq)o93<0bJ<OhJ@J9XFo!h%aM*Y*;un;^LJZu4Ljn5~v
z`1vPv^!3YoU{C0AF_Qa-j$iCAi$Q;gRw&%-jt)QbJ~elioIB*v1bBHj(jNyr-<`X*
zc}q`P6Nef$z#?&jh}%`6!_r9C_zvk~7l6P|gPh6fIa`eS^}$V;)RZpAx&TJ5L-`?K
z^}q0~s4*YHXi)^kYj^$$>D~};3g~P2ZQ;vGqIx;>CVDrzib5GbRx6=X{2<IzsEUB=
z{HF6q^Fz}wukl!-T=DEVsv2QLc;RqN{-a_H7lEecd~iqHcgAy?KO94(7^p&VLoosq
z*Lx~BeJA=|Z@+UAGgpc(gHgJw5^q^JaVbPYn<URwG0u0LPgahHdnW#_CJoSu1EGU;
zN4T{qb}HYZ&dGG5g1k$<j<)N24G$PD&0G+dB(U_D4vOnk(5fu|ppZ+qgr?fHoK+x;
zW+<0@^iWJY?Qv%l2f=9;$8}Wd^!dLjP(Eklx7%3(bn>U^*wb)avC!mY6ma`lv}4j%
z6lEa}10`&Tjo>1KJ=YDDk|T8Ymd_E!DvpEn^VM1G#cC+`)F6-T<Gd1S%o>6j=LBQ=
zzkmGjNhq0nF$S7?o$<q#8~HZNyMe`<qc-28xcLeH4c+T&(LXn<L^pF8D=)oA7><PI
z`VnHSqO@g+QEBx|7mj5llU~-HBuG5#X?3KmTRBz2^~dxNsdx&Sz1V8(k%i&wuTIkJ
zl+xCt2<lesb|Tza1+h4isOD3umCh^Am4?t<e#bJa`)XtM@YIiaQS0=6SYP;Jd)x1M
z%Drt(jw^yOCM^*`eEwX`iPqLrdwBj`Iod)oUqL&H=ZK6vh&A}*P0-7Pu1L;fmo}zy
zos&8iQ^h&r8;~(kJ{@x$CBf{~!AQofS1YXYzgVQGvNbs+0M8=GhY+}0ZS<mnO|%{&
zHCl%KE<5fGK%M0J{v;H;r1J2Ow&7kiaUVz7#pBW63<Iz8{QU`M`JQ3#KF7X5MRkjI
zsin%_GiNCEsL_PiD!_b?XCL~~LX+MFGIyu)$u=?vH(Yw?Yny()>$K5u#^Pkrs-rHa
zsk&^NoP@0`+w|UK<)<~^HHLYLvod?zid!h7uIW2AAF^$?Wf{)yK+<i|#_4l3!%=1E
z*>J6l#dXAFCd>E%9k}*1F8msI&48mz{dpSLYjxm2ReLa$GT+V>445-BJ}J%+*rT?7
zS#?8YCOfYjbN+0OY{~1xh+&vx%&H<XT7ggLI&!{FH2S-J)bq0BR5yb{N&O7M6!!|g
zf2tG;!G7!SC#rzaQgoxONpQcU5FL{ZB8(zx17`ZJt~LQl0FWod)5~(sl=~BIh)KUw
znWKO*F<)63&7r4C)KY|PTupR)5%$S@i9cOg$oB}s059K#lW(UXBy<*!Wy(yw7ZrY+
z?!+!cl^R;ciQUbB-|u{l(S2g#S<gP$w?eu^cs2_Ie;E8axbv11{9UuSv<dQB4fWeA
z1&X`PrA3iUmLyT0yS$VDsu2f@9@($|g+5}?H5G~D)TzSRo&m##pj*Be0@UzQD&H8+
zH{UfUHA}AB+Q0+68<JayF4=M~IkZJbvJQUeTxi6B&3=2c>`A8B;MiCikHk?|-I5vp
z!rn2j?Lgx!<w8OPvQ$g!Pk<}1P_R7@9x9y2&OcjmV~pF^6ya%8`3-S&?oqdR>1$`S
z>Dma;V4j@&BfBv(Z3Mtkgr}jXESlpH6eU{k0hfiP$&({u{NJq*);s2;og@{1m(Rhu
z+F2}?PSR!3i5z7k_v{%@R&t=YAm`vo2TYuO5Z~cf&&<Lhdpa~#%{b2_Pyt!A+yc6Y
zpL%rJ)i<kS{wzKezlKh=z~1>o!;VezJM&~BnA6ff`$I3|{JHxV%y0Tn#6(o*OnPH}
z<5sMDFAsmDMz<3f_%{{$FGe!fbd~>ItYQ>N{v=e1o`<Vc%c>ABS%iXOZDI+>ZmG^3
z(dRo)e|!C~J6D=4ycp|IAy+~A--Jn<Vr=!<&MQ$Q@-v{QKNT6=BF>{;VCUb0c*+~|
zGv9I&&+(fElg)4C9|!Vgvg(3>o!VRPSB4S*M*?L^M=@-pexBf8=`z}1^g7d12g^io
z>px9^(I${*K|tT`aK=m8VogJs)&Sw17K&ke9Kk)s|L%~miQ>(OxbX4iLWR{!XZfo;
zb+cUCigrRD-mrP?G4e=0d$CWmdlO>cJOSRa$fuajGh4Cq#eH<isNE;$PiQ86Nj%b_
z!N@HUHxUc-yMK5^RjONFc5CsBie1aFd)W}F#UEam0T57KWS|pL7|?P92s@2U`$sAs
zz32~A`4C}l*v$<PdUV&H5Pgn#rV8+_d?}4b06rYxK=n1qu?ycB&`S&Sm>M|#@-==z
z4Ild{Bd51EU-TP12jP(i7>};88Ig(s1{C#mRKd*1|F*C+-d-=>292uK*nvE?;BGgA
zo%zuzU23uSA4#}Y*+T%eKe|m(FRSf*gJ{U8mfU|U;O|23KZ=oKjx5Sq5P8?#e=WBx
z`x*(ekzx<O|K`>4=;9~lDE~X~J}dAZAe_zWrW4pC*&k^gbI41MD~lNiMLpee!x5D#
z<1`5XhEoodC)Ou*t<hL<4W<lR6M`TGtixn4;=G;RU6MD$kEpzJA3%51M6qY0leY2b
z_Wk}7kSz5L@t*7e3SRG(=!E8uxmC>Nl;bX%l$f5n6BUoy*BtwpLX%kAY4A)!AOm?u
zh7Fv&45NGLtezJIbR|eCxnzWW{vM?n(wsVe$6*YVH>Y(+AnST7G{42nUUJRp4=dr?
zb6FwEN36ZHwF8YkT9`t#hl%$q{Zm1CAv}9#pDFYU#@9Z2yy@Q+tcR~KVeUdJk1HW-
z?TqD2XGlOK<;~2%Dv`rhG)aL@lksE-5S2I<Vt**or^W_-_1lh?<Yp8lPfBbM+QQg0
z)E$SG5Db3A>mtfd#oJEWRvpB1B$x(yM8rntvf*qNF+-0&JW=UauU7t)Okl7uSldQD
zaTZ}sJATQIDS?FqDc<m{MGG++y_nQYp0uC)g%(r^>W%3f=luR%6XS?-X77!zj#jjK
ztvN&1dd~-TRWH)ko00S{1*Y32E#)MCRW5iI4ynmB5Ljo$U@4i7n$o6|GsVXu4E;<W
zUfYS~XY3H>5H|Cty!-Nk61g2E_u2^BhcK?qDcBF*lr7oQf+X!x|JM@EZOq!zwjZxb
z?W5OZOkg*17+=!}e(TfDq3fT*KA7+ez^?V>##Z;NSEjMLyB=M%<4kT&@f~MaiC2v7
z)0}}Y%P8Af{g^26%GngTfws`MhcRU_#lLzky?!u<k~{R=6aXXPy$w0nhFF5ejOG@5
zVgqlng&0r%;oGM$-z0S!j4wq~wbr2{(UI*W0Tlcuty8RB2C}~Ep>-t~d?}n;cB*o?
zj;dNcAUV~cm0X%U>ns!jWCPC`E(^3#IkWG&?f6O;6&}rj>N2MXp3*z3h~7gK-F7xc
z8}y8#lz$w0iDT7a^x$hRm@q~h)>lvrUW-<e49VftPHk>2bmdBnQv{sZA;8hNr6vkX
zqO(L3bt2tzhJW91hEp%JVXtT8P%iQ=4oq_?u<8BkyMJ@3prF_Yx{NO;7P+1uD*D1b
zUfXp$M*+uh_@Be+M<2?ki&sGeh;P{Trjj<gHc{096i8q*S5j;QaE^+@ee8<~{**TF
zvrUGK23K#IViKM1hjQ~SUv`ok+<i$rRZn_S)6Uz|(;5?&I&vryt%!mD^wPRw4=v?u
zjX95wpb2iv9q)!QZjV<!5pJU^zjrjOTLn^kp<8sM$;bM)v?HB9?QpFfMItMaJ<6qz
ztP0Fwrl=W)JzP?HJHqL!>czM{XdPz6t-m{gPn(G}m18)q7onEK7tkYDBHJ*!i*??!
za~FV+A-w|u%0*~Pv3g$#@@UJ6ns;;}7JnZ5z6aPP5UR`Q{%WUCTok}Yk)JWpQ9zTQ
zkQaC^>3+}4^wp>0$pm}#bvs}(?W2Gy%T0%xFJ*RK;<7$>vB80-&rTFh3$@&*JTcx8
zY(?*LI6P+93rB7K<F2~3d{4XhHhk5#Xz;Wk3nd9Fi#}vnfi^vrrbj>QQcEL*$!CNV
z&c}>UDiAoP>HL(J%9LaOf=8Z6qA<!jwbN5PiOcc9kfDag0wND}@8jGg@?3QLj%`pz
zr@a{+g%7g@^E2PnyWRB+sB~X15a)$S_@X#c-alsBwm3_AT}E>qCb#`dI|syUjn}o>
z{$djH0M~N8E<T*wKwQ??*|Qp#=r|Y=rlZ>_xPqR+;Fz{I!|+Wwc37?hN6j$h-OTA)
z20$}Md%zU?1v3mHKhqC@OIL77@1`&6!{5EPr$k1CNYIIM>G;pLWTa?v2!BuKW3C}J
z(~^5D*7q5K9v1Tc{O_M7J2C*J*MuEljrraVD+{T<KaX$pnh7Gzht`ZvOsP^g6WeG+
z3Nj2}zmLpUc^{-itHu+WP${fnUBpfz(|JoHRw!m@%y_)ZWyxOE4pEm<;idNz`y4)`
z``zAs$)2Ph9FfjK8L;ZBOMP{*?iD@7V{iJ!2CSLctfmVB?a$<{UI+fCCP;My7_nsR
zn?>bLq0<XB?)YK48UphK2zbIQpgM5ux8U@aD4syDxzVVev%|)V8qg8;Cxip1FV@o5
zFMJ^^NB*iv!HSD_D{rCY?K;titQlSvyw3ObIFn9q<jpw$SODTqf{!~y`8sN@D_u7_
zv;RU8so)-dSy_;>eZ3pmA01{47p0DktHyzf>qM}q9?=*p7o3Ng0fKTd@h{uawZ=hK
zZ_e|%NF*6j!rE88?Dl`FABed{H`3~BfP^0^rrub#AjnrrVZCB8e-P^CiTJT)Gg->>
zsVqEEWRJ^xyNZZSz@`In?_WNlm-u&!Yg?}vh?axhZ(9u$^tK3d`jrP=Yn#W03#1@+
zCq>|&>I%Byy`eM&O=K*@*9?>K)m>~DyiL#<3x9u6$}$6{j4cLnxgD&K{2x6X-j#s(
z@kstnAM{8C`8430birMo6kig>TQeHG?O-OkSQphPpG5femN8U)v=i)MF+y?D8Zsoj
z+1Wc!EW5{{181=Xo&k_;HH~Una(Gm*A8+&K{`d1?ve=QBND5~QC)?oyJ+`SCB62rc
z>y_QGKNLrtd~~vMXiJPwjU-<d>B?3Qman01BRVC6aE{y3ps%ga-(ZRhduLq<aZw}R
zQ$Rkq9uaM)IQyEE+$D<m9@fBBZ<Ay^A=*`>vr!xvqsT|S_~G=zjYiR5()x7wj$pE8
z!TQphZO-ne^-I15VF2LB24jnfgKZqXG7I=`9i53TL^u97ulu2}mblsS`8{`(uSt}o
zrhR$6%~soi@J4t2W7)<pG5cWEodTmf>@&#e%&-@FW1!D1AXS$`iLc;SYdsHyv}UM&
zbLu}^P#bL%a8G+caaF@aOqP0y;EYtp#H-m3pr}yg4+j;&Oc)qSh44&TZEUDnZ1l(;
z<fEk~GXKmlr8ar`P!+g0P>ZWXStP0e1!n!;)#qlS4Qe{I)cQpSBxwCR>K`aS%@_ws
zD$KO$nlh9!)&Mcuc;(3E6Z%}Yv!mJP{jDu!ThU*5Iz5_GZ<_B_s^TB$&sF_bePLAG
zm2$9EG}P=n&8RasB}d(nQ}5Oa#ox=oZE&6xOYNFh@B$f_;fjlH;Iu)R^@2Ua_KDlT
zO7GHBi6+LqZxmTHyrUafT};Q#dJPW*fY~%CR7OpC;u|rYS?zq~5%l8G#&?S~Ar`7N
z#6E_edK@_I0G3`e)bWmZ+PV)r-Ro)hk|A%Na^_=$x*fX3EEriqPEK^9sCtESe=kiT
zu4M4qUBQgAhE2NRt3i4}9DMm!&0BoVZU|M9A*Ecfx^~EeP0yi%uLUyBwI`K07vI@t
z@#2qj44id234Z4?bhS!;LpS36AhC2BDH@JD9JXbBB}rU$z8nIWATv+yw!|d4pD_=x
zVc2@}ewQ&{E>)9^LEDjM;4G-fZaJP8tJ%P9We*6r+=&e2MHxG)`D_+Q;T`8ClW=d5
z*Q|kTyB%Z(2`U;hdr`{KPlUoET%^m8m=KJj)Vi>K5Z$u^0rV^nU0EK-zUr7`b={}j
zTB13N=aPi7Zy=9Vw;^#=J}XoFC<~h=cq=!?33cqxvwJ&=GJ2RJZ2jx~H%9<EUE<0u
zfl~{vCeJgfb}AjGbDIX}H`uuu=cE0Bs1M*_77yCh=PigE+r20$j2sJSu<r8IyuZEU
zO8N}cSuk8RLN32Z)|NR<dMtLt7jVFooTn*3Zjz8d2_`<zf-xqnc6|q3(9GQqY*ZB_
zm@xJg06WOf@UAk3nu-`v5laaxj>96z#%Tjch4??5b?(=sAZ%sx{u&Bq2n0z^2O1lC
z)D{kCO&DcQ`nDqee(t0n;L!CjvC3ubHDJ~?NBn=a07uHGs2~5}fj^_+dM#bitoi~2
z8Ak4cc`!VH(5iLIc=hE)T~*J6-s6bpiO}RAV5bLdKh762-&0L?cp`NK+swbsj?E*E
z1RK;T%Faj%MtE>&A#vse)dbgcN9;O!hkLcD6(L-;X3L@CbkXCGK@#4<2rlmhL$G~f
zMl!--<?*}vJF;sdNb{lLM!|>x{f10pxcWoJE$TYWNqV`4uj-Pr`)mMGJ}OiSjf*Aq
zNeMLOeo$Ciiyp|D5^pBWw#VOGZOSHTdoeTm?aX_kMh4n*D|t^}y8zzMtZw{C^D<pN
zHmmt+u<uUIx*1iWDKmMSXQysz_1Ra747IRUlPT=YiBvt5?6IaUIE4RFrQ~?Uk(dwy
z+|xw8J5-ts$kD7?Ov?sO*`I^+I1)LmcytHj7jO+56<gQ423f3r^T4G;D1X+Bv+Oow
zGV6T@50X!#yQ68w;Ca^_|0))NnESIU`%mI+(<+!{49ucouRR0h$<K5*7TUR4W_<DP
z-~u~-Mp&5DZnJDEG*Z_7faID`$~x~~W;<TkSPXL$=7ImP_0ro4yA$Y?@I%|^?dip3
z%tk(7@J9GWpM&&u10|oD%%krgiIP(*Uboza$jmS4CaE00Wr`%)6BD<0ScB|`OY+ix
zV>Lkkqtl&vLCqxG8A4uOmxkA%9?2jiuO?sl1B*%2v;!ZAAjt0235vDr#@t4o%{V`k
z`KC8&)@mndM>rO7Iu7Vwuwpd+$##0A%YGgnDDv1yb7Q9?71AQFT(0HT0*OmV0_}_x
zUY+UbSj-NaOR0a|Ia;7=1zc9#TBY0eWt^bf(Y1YK#B*R7>E@!5XUmamuwOrI?H}fK
zp?0JDRdKCWGKQxg&pJJ7RngW6bb7Hc<WE~;gQM9_93?PH>(yFUi9OfJz6)XmH;k_@
zBFYv2#qkkTq-K_naPQKhpW%gpdTHSYA$6&8c&)T_29HUG(1GBm|CpVEy-9Lg-J}w}
zX590f?S}%lxW8T{pFG1N_QW%V2YWV&Fn4SB57^1rHIT0)^)@_12fVncfz)w|$bbJ0
z+5#+)iwc3Di`TU5)<Aa_G7~G8{Gg4r6ve(vSd@a~c<iMqD}md~>s1p*Kw*;KE@*@N
zR>T254xac-W9AUbGnuqLjv!j+pYA97**ga|d*&}YFWF6|W4EVhuaG;xkV;uFQdwCy
zubXuDXA_k!7Y{eF+lTXcX>%m&+U72hwPruTs-v}83vgJziD#06(br$)>dx>f;lN);
z#n$MRdD1Z0OeBQQ%%QWmzBFw9TL!+=fk3&yIr=M2;)rvd0tGd);;ebOB&A7Llwg7B
zJI*7;CJ}b`*Cp6nf%mc`D{{Z*J4y*BKG`-l{tHhriuGlOUd6xRo!u);gHMG*q8p3$
z8n}B|>oVYLrVM-1c;rPlsjgAIji<c4vl;W3+=7n$k^Hs9gI7^mhcRFH*{!V(Z4w|!
zo0-{aKo$F5)|K8|!lVD{e}^9sz%=Y-c?mXL3f<3Wj##|siZ`$W*L#3K%ks32F-2Pe
z*5Bz$w}f0moj!GWWnzlvi;4dk1m^(`#BXyeA1dPwhqq#Ps~QI;lmCC+;NQe|ia<uZ
zQ<&|PNQ)uET=oflS15}K!^@wrH_zk5z&A})KSF8}GcZ3izb(zh$)S}0QC}L37YreS
zkV?Wy0CjJs2*Y1e{pd(~E1gxnv(=XhgGdc@k&evHt|h0vm<v02@jBz&TUus*RDcoi
z3E0drM|w7V73N_yeb>m@K!BfenEAUaWJ6`{<7`}DHx)qP@?dF)aG?FVQWy29EfsM7
z=hBRNgaaw|w;=Pj>0Wb@oz9tJsu2nDs75{{Cj!{iDm??nW3c8Lf`NZymGPk*BfrIz
zU0&^l0t|YU%?poX*5f1i>I6=aAMT}D3R>U#Lgo+;dR8_8GR3RQEYEOReU_-$s1)}=
z*0q#D22x)T&{#?JZxKP-_1BG5LZgGaBO?fnH|2H~3{O9lWV9Aj|8t&Z!7J*x_I5S#
z$r6M8FscOukSn+Er|bNX%{8y+bYv^TXSVw<$Uac=u;QwDPeTGgZ_E!~?h(DC%)Bp>
z=ih&;gQ)n#oKd$<|CX=N%xy0eL(rkzd2fq<Tul{r9_oc!0j-vo>U{v2ptqiHgs;L5
zkOq>paIMWJqzt7K7DVnec|V^g86%lD<}Qnq-x0prrbEkX?MGR^u;5FH)x~@8R7>S`
zk^4yzg5@&7vgjN$+3tPtkpb%4iWAkvz1$2o3Nzt*%JER8G5sys!sc`J;cB!>whQ<F
zs~-RYrF|uy8;2rfiEnJSlARZ++xkPsXfd=>lIA`Io?fMku!p_xN$xmDI+EMpLqnHW
zyz-E8HUbFuvwc00h((%Kwv0QQ9G8BfwG!oIh*GNJViP+{O<h&T+r80uE{NTkK+NU>
z;j=vl3^fcFp+d3s`(B%d(IPJJ3&4_i>G6giRfc-2PFGPJ{8n=C2tz@qxwJBu7R4+}
ztwI(kiX;kU%o{F$9*X#p$o{f0z_5afcWgr5N5+z%tuF|xR(#q?AFRIZ6iQuw+R4RY
zRA+V64jJAh7p^@wCn*4>DK7e;L_Q~EaR{Ul@|`JTe``TlYJis7GGN72xPj1qBL1jC
zCuEG2xxpUl)S$g$>?}IR5tG4l-%;o2%dGLs)i{Xs->b7$qe9%<WL9l>UXlM7f^5F|
zJ1o@N?hK2UUhmTd()0GTsPvogz&W#d0QLf1ckK3G@>5+Y>&)BUjHn@BVyLA0ha*c-
z9k{F{%5f$dBS099;9npgu2FhR=;%R>8-mcLQ>EUBq|V%Sy>&IVkht${MPZ`)hywSu
z-`&9p-ajEnMYb&$WQ{n_4251ZoU|$0Gbsw105dm<NGmkIP~*kzAH%d`hJ3=!u4VV4
zTWMVmlpqUP)Ny~{qr9fo&n6+PfClbM_Zy{h3FVmUd^@{Y32lmTy<|J%0?oQoYM}m3
zvAB)+o(JqPn<Vc2{X6-1SGQ}*pQo?{4SK@F1tNA00e>tXeOnfTi>R14oAlp<KsxB-
z>m8QGl(s}FdkeC5y=O`UAKpH?UOky3&mhwL?0Y1ew4wj{efB0tq9)Uy!m%xL^JaO8
zm_NsTF-b!!zd(zxpKyid4-{+EUC|dssHOf{3=L`0nui;X))KR573hC0T;zdF9an)M
zp#N*tMJ+nag1Ck4`z!4u&6jK8J;WJb;@JndP{H_SPjS)HmE?Op&Zt^1i3l9aWXW-(
zUS^xG{VaI*10_5=StJQS=0~*1bf>1y)$+cquL^rCtE!W1wK&cj8B>zfbEnHU*d0;>
z?4{1xUsA^Rx{T=1Iu7do-$=ax`Njlu2p>iyj7(_84Zv>rcnMB-o>A6>)NsNQmcI^$
z)8+vppAFVkUpbiYMt}~qnGl}p(Q+G-3bxVDy{nN?`!6?QgD9FWC>1*4i1uWE{DidP
zQyeG$ETQI2kqWp|U=WCbJ;P5ET^LMODMt00NVG>F-2OPYEdFU$-?_h``g=a%YyMle
z&lzs*lPmw-dRguHLSXf3!&DCmsG&-0S>eloHRI<y?kO1}1G)?yP47vrX>!H0IIY`j
zdB8te9?W_u9BG>2hs>Y1_*IXeUUnSTV;`fhfC8bRc?V>&tF;-}#LfX13Cbh_-ZUuN
zl6W%QzNPVQw0(6n(ytNbwnL-9+sE70DGSrWTZD8h>^iZ9q8S0Na_V2x=8coNuxo}k
zmpgvHf`5g6Pntf8w3p1yaZ`YJkL57J>ahIqoRvHhF5-rFJF_)E(<|5&GKw+|T$u7!
zRZS!L7EZjsFeqiLKB6|$o!h<ff(cACcUJn?cQa%op75mBS2c1DD7nz{=36{*aEpyz
z9(r14hxt_gKqY8QvCaQC@tN`S!70A#t%)geXl6CD2J>KY+%B*JLdMl+ko|y|(4B^C
zsak(5<+WAXrPxvZ(D<o!4nF??1~u;FCnuC*TtVD=Cf_7BAgv5<8U&*s0_>`LpzQYv
z(8M$ZM{ISw>t5PSi$9zfvSo^B%mNwo>h_k5m4krtBqLA8yW9x|KKW9b`L?AkO$z7b
z>T`VTS6K``h5pzl8aNyZ6VKC=+3hRn(>Ua5*ExdHl~?ZkSdbp#95_@42W5DBdiW}U
zIz~r-rs7pO!#}<aO6aV6&#bn9>o5WI@BksYoXIu!%g5|7fU7M29AyaPkzsCz3{yD{
z#M(!Z^~(?NdPZ+(RWbX#r%-H65o+@?l%PYeHkk0^djbC1{n4n}y)M{yEfhhkB?(%K
zX>_~WkN?KbedDwYKssTgGRBv}<NY2pW$!}$x5#h))!U4ihmMfJ)9$~8ihovJq*wzd
zAt+czre_R31mkNjr{9arMX~D@47<S$9WR~6PNZWY9{U7Tv#YjNadQ|Gv)LQJEGzZJ
z%>fsAqmvdDt7W;boB*}75lOK%isjdTO;R^PonQA)ybW&<r6uv<F<DbZN0bsKR}`c$
z^I`9rH_jiV@y@Ps3Y1WoT|yn7#S->2floWQkS2Fh<!W&z*apkb2zzwy={bYcc5|W5
z7U0*3BP+b9l`T(@GNeK6-VuFb648ur@C2Lf^A5yBR~ARHVh$hu?!0<!cF&X%OOai-
z_kA6)-7px9J@W47uoyMBgUz9*1l9fvZF}|I*3W<HPvae>wT^{2YlS$ESw3E<{I-lQ
z8lYG{*VztkxffZV-GJL?!ar`VT%ELyed5cB1ZFKMdjqezAbgRbKkcca>zx!7GOV!N
zy&>o1Qi;hb4Vn_D1KnS(0b!o487b94qNsTbjyDY^yv%c^F3RgmdLls+HN?%)Md|R?
zVxyO<)`UI?Zle|XPyUE@(!srRXJx$@uV_W$K%%vqb0e8lfSst5*1FH1utJ%GOU&JB
zI!l!rG7aYo17m-(nvH*)LbLK^j>-QB<jL%PI<{p39^#L6C8b&q9~*Fq`rJo*mt6$-
z#`6l7&(#VlADh$JQmh~<(m+|Ch*5daSpAXEWx6)PJDxm4LKzZBE6kyFnPklEW0{=Y
zaUNh}T#}<>6WbfAF^-Q|945*uJU)Mgu;P#7m)88eDlXn-<{HT>g_~bhK!LT+^{hBt
zcbUA&WU)-n^B=XxxUs0k`Ti#*;ew{fu2?eUgR26)h}7@rDyUa@Ci3dRo3qt~4EqaT
z7aX|GGq5l>3zG1B*iI+5O)$Fr?{4$H7TxPBgQ^LNLBfhDblTc%tURENdC*}jo}Lu^
zdwyvg68~I*o6t{c__$i*ip^R1jL>Wl=e;?*)j*z?nfg?ac8G1G?MNw<q`-qF56AF+
zz%^zW?Ug)n)#!Wk3pTl-v{5jMP51wI+Xy5XelT>(Pa>hQa;^pEYAV<>sQL#stV4Rl
z!qK7$1>-MVQk1XtGUWb}VV-Q05y~gs1TM!j2Nks;Iqn)Y`&yY~7L?IC4@o8-;a|Cz
zKWQ9ehzhn3cpS8P>Bv9=dO}hUp;FpG`l*kkcR5-!u=*>*=TtVm<e391uJNe7BPl5}
z=M$a(ojnwQ3!jo;vM&q<Vll)%fVpVdY>P~rC=cQbxlaZ;zhm9QU=m-Ld6*Z}CWlLQ
zesB!hM~B653bKZX^J&;CBZi{C!=&u_OO;FdGu~vMubI2K{1;~fSTd<QKAtCuKIC&s
zw%{9Oxn)K2p&MnL$SkGdwt|uBgKyphea&+G{AdeSBw3F#1oXnp>6++Gcb9T_$pOaS
z^*$Hx>qMRo!-gEi_`>F0Yl91U?yS$lauK-Uz%Sqb5^9)UuX4<R9lWY^KzsAhSO{;y
zfOo1V`~}vts;85!*rNbN->q7`6k>u783p(jhO&&rh@^frnM()z6@`z0sZe~dPd$U6
z`}3QO8};m39qJZ=FQqi?yB<Gp-;!8~TzN6`S+RX=b70O6FJ<(M2zK<4jD<wcRvOX3
z@}eZxH>PnvqL!rxf;(i+^c=I_X@nTo4L6xfXUG4K_`YKU1dDoy1SCA6>8CNU5cKJm
z0I2ES>9lLxcUUbM%`>988CAQV@Y>%=3?HkfLO-9@l8+QRrIp;XjJvR6;^|<YJ(t;7
zVnV9qrC6b4UXbTz6!2vP6f0w&#XypsGJOVnrBTm<T;%h(MI%vug|n8iZCD=UsaWIH
zYULiuCB;G)$$O0(-F^ds>@=XCz@-(#8%kckwiHApNSDZnnDy{kw>Qo~kC@rpxh7{@
zjSuT?3xd$)9FGhJ`C^Umo6HXLGYi1MBm7dHY)@`YazOuov9be1yw$V%t3cDhJ)zQ$
z+*99UIbt}n^qp{4HgG1W_HQtR6If)LAp3;r$!ciLj2hY(b$n$zh)Uid#iIHLe?k-C
z4S!$VlkDN}moJ`L9>cf}UiK**%!`2I;K~TH3Ed=$C!gK=BL+f_`{AT4;_Dl=#Yfrt
zRa@l&0}wa@Rnl3!wwcLB)Hv^0C7?_6!pULt_O2~2(ggw5j2qU@Q$%(CnAnJz0k776
zRcwi!tQS*)Wjk$k#|FfKdeHQFoN<Qi`0C&nieHsmS!)%W-8T&|n)aP@N{hc$g5(Us
ztlu`n2<`eZ{FIkfWFi@{g6Mq2rB;=$je<<^(a6g^mJ`+%d9P2I&CTb2`m?*uQNNX_
z#91BO=cRd#oVRy4oL&T{eoUjU__H#8)x}q!4Al<Ov-*4rg<ch3BrXS?HZXQ>$zd<*
z?_B<*{gn>bP(Layrjj|3KrU-UGWhVu7mQ{VP=tBn7jrFyd)JrCPt@x4-%sHziH(AD
z1!-cKW!}Rk1BttPvBH!z`1*m&hmy<bqUvEF1BrY+(2ObaB#4|wfA!UgV0EPy<nnpq
z1wX@<t6JHBPh3UDr+0!~vpjyOsexBQuZaLRzxL(2TYhn>Lg+WI43ut|AAibJ*I26<
zHKD0FzyaxMxp?)*u;^2}-_+E0gHuekheovde-;{T$D!HKb&>4#q$yko-<b%Jevu2K
z??Hv%-56Fwen9a#*?gb-2;UpZD)>9mU97-ea|~<af<yEU2@eQXH!Kql)Y^q{WG<`@
zRLT_Sh{#c+t~rc3py2#=>O4Pu5n5Dr>HR*|0!|n2K=s{3U`sA#C8if6V<KTW@aGYA
zhOL*zhs1#!qCNHqB4xd5r{O%^0*U*klZcAFBf9z{FTE<f9H6Yznf#HU+4Aj72uHvG
z;9EN^b37#j+3w!A!b{1(t<A(?n~0-(6=q_!v9{wpuUj}9#Qfr*`>Xi~(!aY^smk_&
z1x`MlCHiTR2T|6zcRJEtNp_n(F8||!UhJ~WAQe*Y)c~t?L-Y$8VUh;`Y|ExyT<ZCm
z?EF;!lf%lJoxf2VI7OrpK~-`u;!wcEmk#4G<OmaI3<v|oBhJl-{e-xskw9*RspWj4
zL(lrR+(P(ktUIrVIzhkC78HT*ng{yci5CJtpnmq9b(+=`4tzRL(Fv3)p1OmaT&pBJ
zS2pKMLomLRyp?}wmh;PIq&aXpi_h(yEx%k}0%GVV{{VLFyFnGD4D5R!(U^xOqdhrv
zsy%5cLOb;zrOb*tx>qB~a_|hSCsdO(m4T1WhD6E~=RXIzC4Pa)9ouBYpH1pr*C4|J
zynj_FRGxSU3p^|>X{4LUlJWOi)`-D7bo;AOpv!|c1CkX8^7z$DaRulroOQa?+J7yi
zwfG*i!rzkMwEobcms!t%-sHXy#xB4z)rx8NGb3oBS)GC@``QIA1Osy6oDOz29^SZ?
zm(<?Xr{|%VO319$L1=G)e9ME$#<)wLyYn3RhEs&$DG=FZaW2nIi;0vPzer={4Y_Tn
z2&uNU(^44JcnP}T$yM!ej5iS2qf{FAA?P#3gh&aJho=w6?8ZTu&ZU-T$Kg@T^#q!0
z%P+RHId!xFOpfAxeWX8Jwv)%`(?%y5AJEh=Inool3;KQZbnadri4RNiKa*Kr)<(5v
zHje?`*dv0aG<X3tyxErMJbc8ZRk+}Kode%`InTm#FJDy&+YuL$6v?LDe9jsV2M!u`
zmOe3Ymgn|C8gP1oimW%)jm#W|_C+oI_1d#m1Jn&z1uR;cRvUjk{)f>d=oAhQT1*+~
zvEZAvIFDeI!KHvQId#g$)-k<bL;TM?mQ>G`xN^y9{h|_IjBlN7itv}O`4CWV%$Jq!
zcM=24nmGp@s?M=yP9)I*vaUIqBRX~lTj$F$EHx$@pa_af9?7A)C}AA~+ega@%C;d4
z=eU<<g|qdS^?UAu29&Z2ZgX!*B|da8&p^vZSC%yN6w1fWd61iK(yNt0U*GgJW;P#i
zIWx!S+Rf?qiaGm7B98I<@Ib=3HLiEO=<ACuM=xyEE&gq&_lzDACwey;5MS(36jxxk
z^0`xDrqjvyy|tqL4R)Lt8;I!hANyf&+uqdqQE|L7=ya9&s~w>0YLm0A^ee$TW{wMo
zjQ54&s=!BPuJqTdqy^%pcE)A`?)>xL?UywwTJcPtgVk@$Z+LL${f7Igj|mTXhpC8}
z;_R2X#Td8Ky?5xmMSq&YL#LL;xC>rwcj;VRP90^XJXxC&;;^Lnf=a%THZqz<O}aOL
zH?itb6Ucqi(J9vVAC;j7QD6;+&nffA;Nx1>7OeB1qWn)@@`m~<Fcli`3#GjLFcS69
zKVFxTk{QMn1A&w^9oD$p6`W0fbW}>1XZtBNGtP~@cdHo)RKQ0{9Z!Jx7LOk!PT3Qb
zk{`St@lbkotFS-)zP>}Z6`o^J*`Tp8<B01<B;XGJ2wdNy+4p7y03al|BX#3YIvM&{
zKiO-+%}ej<&Z&jz4B<|<kXT{mLPK&;;i7JFH=AXdsua~aJ130^vu|KlXm~QVd6X>I
zc0p@|DX>zar@f*(CES|zfsglJEYwSDb!qk~XZwn%pml;7`}K%(=U&1`4z44zP9y!U
z2(R)EY{tolvwcJ+e(XIbnSg4<7md9P?eX&`1}sr*;d8-x{oRI`O_df}><AZ|H{c&G
z`X)EWcuM(5y5laa-k-{!hO%|+(oD*d9!%>^K*Zy43>0u?@^%j7DXC9^sx~G~`3*xA
zTaz8@MHSVU2cFYx3H&5diOQ$QUp2Yf&*z!fKSZ4dMQ>c<$2k}d_Brz<%*iz>jd4y5
zqrJE36tjA^pZ<I>iTi0Z3FP8sh53DDShk<y?FHI-;?i>{!oQC=VhCqjVCMP>uzya9
zgKa;?n^j{zl@@uB`BFLM4==ye=@VfV_rN^`{a0F(fSPG*6q_IA^)>1VHSkEvEU%bf
z7W0jX9Gvf*lhL+_a%V~B5;ia>%zNeZ{LT6d3o<2pJven*h@1e=G(sD__q*vUSl)l~
z(u^<q7}9+!t<^5A72y+GNCq-bv+!|=w&p6Lbpc6Jr&=0I{&bTjNB<R&vaxR!1FmQP
zVYSkKCUwv~AN!ryjS@m3|Dr}-qnnKm|3O?s2!qncTY_P{t~D{>H;>Ts6!Hw{!03Hr
znm;wU8cD7W{QZ47G91!lmW9VHo3UFW4MmaMtE#n#2`MwZot<N%h^R0@2Geo*!`8QJ
z&@uxP29e|kD;xKint(ZkzP2H6tFH5Xz1j~a>;h<x_SSFPLpVuh{C0zwE^ue*Q_fI5
z-r?iPG{;O;iX{W2a>)3%fgst?8=wg8!p0YBnHt1F2G^3mx3-};Hgu#W@kecFr-+s*
zD7~iXl!ClPkmXO&r%o4S#{p|gnDe&brAOg-jJ7&vxteqkO3l_RAZ(c8z|1#mqwsjV
za?iIK=n@1u)f&Yuj{2*{ZUl_Bk2TXr1l>&-y!w1fL-%gnzy0sH?$RZUrulP$j*Ohk
z08cX$_(itwPDRy!Kjr{433=?S_71^hJ%Y2o7t{F{^W}{`=M>wsnhFGqxE0yk;y~5W
zt&MMk<Ab+L;Fq`?a<}oR$s>v0IFxc#`{9rL+CR&ZUh9_m9W~SdXWEjl*}ktX=@JrS
zd*d|Mx@c-YTsLMu2tM-)dR|oLDXSe6{C=ehza{GZU+Y^g0a?bM#{zulAuqIhJrKFl
zmP_KAXuWM%FX_AxkAuHuN^OEpTh749J(JXP*=>O*s6q0CB>pOg-8bha?R=t8AMr~9
zwzql(8%wSQ8!D)&`hQUC+`k$+pA|J4E_Ied-kS8*J`yQm>UaT(#unK%@=>;*5*@c;
z>w_fL3)6oL-f`Ki!HU1E%2`Qi2^X5!3!1(C3tjse%4C(UN1I;{@agEFzK}(c?eR|w
zl0E_n#Qc4%Xm(fYp0qa>1ruP}?6JAVt_9X850N$4bv2A9B-v#D{`ge_e%K&7mpwS&
z0lfHp^D-=EQ3!mwPRzCZv;WEaiU-uRp)%o+uh2dRy#%hUEo?z+i-Z&Qz01b?J%Rhp
zr^SqDw-}A2TvoVy4_LT^q`!}U({$@kL0+k7WsnKU#D#+%FJ2&NN?^Xuk>c>f^ViCZ
zBG`*w`Hs|)$~vD8*#@)lgzV{i0Sbze#({$^X3*C{v`xOLWwh6S7C=D<n`50@>|I^O
z)ctzRosEt^)K-Vich79SGu@X^D_Y3HGsf%pdD7B#f4zmS2V~MCSc_5VS&Ba#Af(&o
zkNrXV_q^{KRoDRkQ4*k;V=PQ9twPI@e2oAF=cZM#710mlCBCG-Sp)cQ8=;;7-60Qo
zUl|+r&KBqd`W&I(<Hb{Q!;j;`);ez|r`0q$zI)ACX2rneD5TQ&=z@a0o1VfyO?HkI
zT&tw#`qN=3i(^1%hWs1Rr}$AZo0t$KkZHGop}u`lO<swtls;5S-vxd3eksc^x49S8
z_lcbK?W=$84gX&)0F_J;DZ*QJ+|087<2*j5|0q!fWL$Cf!bWG#oeNvZ=k+WRn?r#3
z8}g^sQ*D+@MBzEu8ISDc2e(!8((o6@W`(xg(KOOCN4##q(u)#g1(-FJ7k<g5zaNY3
z?>Q+20<X1zbd8GFR{gZ5F+Y%>Hc;kgt8uE#7o>x`bHotY8sslZGI-=sEVd(3N3X^#
z&)O64Q~bx+*T*Zv{ulCGZN6kZZ{*sbwQ1U!>>-g3eB%#;q!(Yq4YEHeKbJC(WjN<q
zy?nM2W|sna6c3{$oP8#Lytiu36x1Jay}>!&-u7aOI{5slv=$?KSgeMB!T|=qMv6)V
zvi`&l@efiTf%TFUmDWv_)=ZTge6)PXkHNJ%Nx^N2dBF-Lr?<vq0kX1QzQ%k?c_}vI
z>-{T(h59*C&*A3o+P`-+Ty_dK%yaZqt^=Ls?`4hV5;d4Dg`&r`Fl?BdF<gED&TaJd
zKzMz{h-2x?i~DOgQ3Sp=V}0hB)}(dZOH4i5F6(k3pa$o(ZmWK8-_(NYbklr+QmR%4
z=`<v5WYI7lUYSLq>1Ady_U~zQ{NTpTct3a@-+cT!>494JgGQvHY#6`H!;(xz&o$a>
z$SkV8sB6K~nQnOma+P~ekTlg_MxiDC_q<=QurC)0-*8BXgvKY53<I|Di$J92`YBD>
z4eKQRcP6k1EmQNE%e^WA*aO0R{_j_9le<??nRAu20V_@!nLDkbS##)936jEnF_WX^
z2X#Jik6s|@blL|B7L_}_L{bY<Jd5dnBoYH&^)N@Pi>+V2mZ+0=YK}_kF2H3Lh%<``
znJMVKayh2AclJ{+6=*|tvt4YIrp))gN*k-(a+P;EptDCi2ncLbn@G{HxJW^D?Y!Tw
z;6VZUetVbj6^J%;={k1mFYzb;cZ^a}8o43t&MRMk7sp!eA@a)b0bZ6T$dN=VW2$7F
zYLBKvJHVitpmHiiSho)Si!FMa5v4_A-o-c2f#5xSW0TEr_BLaI_dn!q<Nkj<)?%Ke
z+6}9ANz)Xa-U*{}f3hI)#&mXpeX~dLd5wyZt;9YcSAa&j{P#MXt$c8i0_;rSe3M2)
zCPmM{`LePs9NYO-_%IvJT<4mCc|vlt8t!@7dch+KHGLRX54w1aq?+^W1wQpkJv<C}
zR|9QH5R79!?w*Kv{!?+N`zJK+qjW-S=3kPsKjbEp+y#Qqr*8x_Vr@UKFOQ7$laBV2
z>6Q48jMEpT(25LiiM<HE(mG4C9Xp%&I;XUvjap_~X+cKMhnf?@1UP~T2o5T*&mD%m
z3b35oa_ll&d~QS;lfK~`O^HimQIg{z9Q_4;P61%&1#Y<4vSyXYQ==fpvwKFRoKf!X
zo|;0>0z%~RZe`%<7rIn}PLZH*)xT#ugOjXR);$2+;1`nXc{zCMYnffPZ(<I}v#L?o
zqZ0}<sOnimec}IG0G}UpDtMmlAqJ#tC?0Ia!!c$VrNwdhU!kwXM3LmixABbyTn>Bm
zRz+&b;e9Wn&;9zhXHSNYeTd_1M)X>93%7Y>X0_pC_BM&iuz+C+WW>Etm_WKNVtl!h
z#ZTc}xzH!gPq3!Pj?u|!<>b!&uyLrDRy|X>-@pF);rkQNgh%?RYAZH&GT~bo029Sa
zJmysDx9b&rEN$<_TAlgJ@<}cqDyTI9HF1`mIL^Zy=_Na1%nY*N9?kmAqASr|hMulB
zHIsO?{lWj=9g97$8d@h@PM<vVR#9mba@%FqVjN%ezldyu@8^efI~RF_4HJE2udrnK
ztStn1;Oe|Ba~ePHeBDo)r;}hDMRwfhPEs4}HS^uhhwx25Uh#4!68voL<G!Xsm1Z}v
zG6kgb>{)*mb$93M6*V6!@IVXa@_0b)ZIbf#@VQfqFZT**XApi0niEpuB-4PcJWuiY
zAy6ow^w(Zz4m197ALd~nH`)al)-410mzPeCNM)!8;rM<JqyA-MIRy<V-g5({-%;^K
z8TL-?<$AjIrNRNn3NpQO-4-XU6*4rYN+b8%fr8szI)4m|r`%lrm>5qD-*10Isv4NA
zrmSlh{(hyGAtEr%L{{I$yU1H;sz899$Kg$+fD32;@U&st!-~zL3%#OZ%kCqJ?Nj2z
z%Eojv31ye?#`!aY3}fD>RfD2&f0n6N-NLZ4Ro}8Lc)w2mQt@<hme)$kbdCMHL2wv$
zGA6pz7hyAYn@G!PlT!$jok_vZPIsf4Gw+#wyfZ8V0X(8e{4aICD)F`&u*Gr|7n=Tw
zMr}l)^NuR=Z!d66y*d%*l6zUD*{7|Y`U7VB*rvFjVy)-MNj@KoApU7(0ARoH?Z?Bk
z*G%>Q4}0$&)MWQHiatr`y-4p6KzeV|tAr{@QKU+j7Nr*{8UY2QO9yEpD1sCb0f~Sh
zNL3J!7Mh@dgbost^Wg7&znOc_ojLRU_syL%!)OL}c3FL`z4lsOg&#(VVA@IKFMq7h
zk{k|}jKqbJT5JbLv4-tzpX?E#xBC@tXAQp3zm5@sdv`;}DhnS;td6TqGx9B=TTGvz
zUuHj@JTUZVY3tpaNj*aki)VP9rVr$PE?jH#xl%W`Z%eFoc+hOi|7}ZijfDT1+6=3-
z_oITVV95&JAq=)gXdA8|wqd7tv>jOCNY-iKf43>h9;%io|Lo!eZ;CIf;630{s0?Cj
zKYV5B9dme{j2lFKEqt(pEV3Z}lAYhPpJn%LsAdb&sTQ1n>8dD6;Ti7g{`gp;Eh_aW
z-S*<d!+b>5!6xDMy&f_dDfc}ggHQc<+GU~(gPcXm+Qb`K-)59Qpgz(G=aMm`5*zk_
zXCG_j-8qro3v?J9rj}8Botxh8_{<XqpS9GA5x5y3^3-R2E>Onc-K$Pp)%nOl6R>{Z
zopmu=XaP4FwuE$=dH46iad83SbRPV-AF}Oa66vIR6(;m)(--^ePGsuVwkbx_W;C>r
ztO2>Oq55P&WRX@fEVYO_04X~YJ-z?1W28TXiV#hXiA^VV)UbO;ex=~+&BKM%4q(im
znvdoUW%|de$;I`IjeOGTPM4$B+wjr)qLqVI+P4hX(Ghi9jm1L_n4z@7^00#Rwxpnz
zJHGk+`S(>Ou^WX)8;<v79Rzl#!w5eD!g5~F&);5A?Iac{>1$O_*nhbe6__p_Ah!_(
zX7%iAu&nH<o8X;gD>h!;H@vZ*IZxW#h7Ki!xHB^pcXk94ZMQpf=cdv>dfXs=J8R;Q
z^@Z}VksKoCX1)eTsj7^*GBYpzPY(|dh^NQ=fO_V}G>;!?LR?i2>c-BR-^(@XKb{o$
z$VkSMNNIP~5?6GfL4K6Y^4Z%ntveQ<)cQ`NaogR3G@Nn|h&w^|oFo)x%g1E?0Dh4T
zBh$k@_r!A{YP=sWmhy;crj@?`+fiN4TxZAR&Fu(jtuHivY0V}=4@>Opwwrmrt<c~i
zN|rqyiy5hof{$3{U_iA0<+93&WBYrC;Zc0oU6RuJ6F!UND<(ZvukZJj><pCUnPy(|
zCcId*&dUG#g%J=W2J>b<zWzBK;!Q)i>DzX|`C@c!dT^~N^nC_>QQ$?t$KMIxFn21J
zT*a^m44W(|1$D*+Y$}=~&NlN_lCl0)C-|}~aO`|>!Pr}f!ma}E<CMFsVW#d0SA(gL
z#AW)nhJjC>J$oqNeb<n=i|pN0fYm)ifoHXb*DEYl8xE{3vTk}Qoi4;#S-quR{Qksf
zgxW~O@$niJ@7ko{!)}Wjb09rw>apQ%(PrJP7@vBsF&Tm)1DPT39jS!$P0tn;T;-=I
z)i^_fb=L{)mM7sf-7$*sxlN=%`3B*KGvteBBd3UEWtiP}ybthHg%!$ObHoe2%mnK=
zh@8JV2a~je-c!y}M$v^SAr<&XmKjKk_5RpLCH0cOsRoM+fp0h;fiW+UidGK^uW)Lx
zvPqVZUfob_(s`Me3xXUGe*XTwu~I4=aE@!j7d!YOD#rp%z)5$!s(k|aB7(Mh?#mlA
z7SUszJ4j2x_hJ?GmGiJGHSzy^9Smb8h3l2QvM3}xJ#+%UNAE3iy3q4wmC2PXjIy5X
z8Z_#{wUi+1x~h<~k7%b$uHK3f^k<qm%6_AyPsx7@G*fs{O!)Z9cNy|f=j6nx2eTWi
zg70Y|C70!0v%hI=#jIY3*sq+#a3_v?iyMWI7@9)!pMI1rw2P&2)7X-QZq|Ig_MHu+
z010o8)6U}Y+ao|=Fndx)QFO(^gU|4#7Z&VEd?O3YVHJ)rZ&#33dFymJi@!BOq8P3>
zy{|M|z7?XF{wGhvc46`2^{~S6?Vo?Z3Pk<r%Z6u<-&>4)xM)NPrgvV7N**8C_c`Et
zobjNeBHEngTd1-T$-MTj(EalBf<+eug$YUHKHsV~R5=AH&C~y=J2-$7vaRzTs6IL$
zD>gRFpyd;L8dsFq{b#GVFFH9ZV)nk|qjN-7W}lZ(0<-A;{xnf<aRSHUmQSJuuxgfe
z8?sch6or6>?P)1k^aM;I7X%Y;{vCVKPgj)`m?w#%%pKkhyCC8a<2qzeaXybx^Lgo_
zf5-iua~)=|w<(t*Z*=wVq6+`pqG1=t&J<J*!%4Ave(_*6C~y|)|83AQsWS}A;uZqM
zi68~Pj5&gIXpsft31p&wR{Nre`lTLwY1ap1j9!4oP;rL*CaCHig$f{9kml+nRevY8
z{(ntu4*{?-&^priI9t`DGtZ}##{ibl7`m{SkriCl*QpR5FE0mAmyd-R4zqpK|E5J}
zdj-9p*!F6};ri-r+o<+Q{f@S!SBsMK!S|CKyKi=(E|$eUl78eCUl?d6hl+FY_=VrG
z39Nb1HuBTT!NGcAWN$z4*?i3_{807RgBLYQT(F*xC?Q0~otyX^I3fc+p;HtI$m3+g
zhoC<oBf?UM3@Zn)I$j1?Mi?YWX&3k1FPP$CaVaS{h3Us-dAyo|C4P|5y{z(rv!PF<
z?d8&mLxz=5apS-T@Q>es5qfs89)KbHlsMR@{#utbPz82^?E~L%HzrZm%0}z6YVeWF
z+akqpZHMXHdl;wrC0f>m$p?Hz(j{XD*>v(%w)~k#*lXryTzk0|d-jO>G8ayPanegu
zN}0J?iS%s>GT8=NCjr*p&L1B}C482_N@E*tc3Y5cJ1^m5$ma>EFQnvd7afie_wi?c
z<HX22Gch_n52>3#v*bMwzPjxJy$m^J=1LT57Dmc1ww3NQ(H+Y<&y>N^jRtX=mo~s-
zQ5}*#?wLm~`#N-mdiQ~UsgU&5SN8O?58}-qcSN1WaC9WsksV*fTzOfM{zsPM?E~cf
z7;+?ky%Ltf+1Qz*s~<~*DRgU_+2M-?+bn~Xug}3;O=8A3I}WzJ+6om4YCW^^r$=N|
zw1btB&mmnCcO#DH=(yONzPBM8j#zd32Q}ZC%3!ReDdd^n2TNw0-DVjTvT_7t@CzAM
zFin!*0w!iNDP(V1l{H%uL3GW7btnH~6#W||6DwtmpQ}F7CG!2;qZdn>3crNN$RgeA
zJRg8ixDBn(N4j>hzhzeDIF`gRVvL-vZ=FRxBEOmer%nfZSc1bU@^fjU(ndRC1Q$Iz
zckGDmp;4^YI-diAh%cl6?7Nwfi3`n|tItZ~OuZNfp0-BgpB4UY<b0|B)=#z^hcX-3
z#}+b6jiapZhi~VRf_@R#H5wf2luO1|C~<VIZzLF9b{?i{sk`r(dR3}#y)!PG%W)(q
zEVb{sbxWTYdr{27&&dNnjq;p@wm!ozZoLwO$l+TfI8BghGX~(>xa|%CGMDhVv|w4j
z2DdPBm~rbS^CFxPW*ln{Ss=1uG}D@C2_N}`hV?5h5aF|tWl3@B-+c+++~#2DwOAhb
z=N4FCLj7H?{fNQ{b%T^7RMJelq;F#fxV}{1#j>;QcRO|&n`>o*p8LCzN34ka!MCe1
zBiFa`2z_7$cFVk#B0|Rz56xbTN>F5>K|0}d+R-uvN83fdh%z}@s=(zdZnv^t(+~99
z^*}Msa-lMBbiYGoE@Y!JUAtZS@tQf)p>;Ep`gMKPIj-^&(|w&mY)O63JEGqk7qvge
zs?N(>fBU>W=L+{0UCtt?dj@;pd163@cF*v{AI-;v4H8zRU;4yq@5N}&7}w;9#kq5R
zE7DcAsPLg^F&n_Y89FR6{pOb^8X05x#<pN9#@NuZk#y?4WQ2-Y2Y2}c#B^j^@<-Qk
z$GU||vu}J&y~jx=;ZmKUsp%Ef$yf15g%f!vF*uMuvJu563z39H8A49C^eagGe-1i?
zV^yy0eiuMz#_Ll#pz$QjZ`R|KI!=IWZiGF$$XByt=y1ED+1T#<gi?Lgldes!4QC%5
zdY5Ew)lF_&3NHY!%38BY_7DF;c0l<YvPJ&|*}QYea)XeiT4r}<J?Sjj5+o;^OLzjw
z@f?s695)N`Ax63;1V=KM>GV<8-<@k#JwKBGS?6RZ;6z#saGT=i{HY96Ednmx&G|lk
zAA8ENfAm`>W4w-kw{z{~{J#rP2@oQ;OEmwYkXriXITh8Ud+U=veci~1)dsae2l_sm
zLw079TwHKaIz|-9!Z(eh3bXc@VM9&#ZvV!Ho5M9ZmJ1Kw-E}YxE{sqy0A>U=JIQ?^
zI1Z!+Hbe)iqS0L2_gr-@DyM-TV}emdo8|>~;O5m0G9s+q&Y`TnM>qy$foP`7kC^SV
zJy*O`f@2*dKKq=U6y(a+0r?PSNLAI7PExfJyJEz;SrZLXcXzunqoj|qm+4d<`_S#w
zV6xm(HEO<QWvx)7rS64(kOl}20kY<y5k{eeR%PpG1{agI00;|^ijn#CjbDf$*&GO8
zyt!GgxDj#E5t$4>xJCc*90PMVP$x075?~GVYBgq=c96wlqM2RrQup=uL;itJ1`%qY
z^o*<-MJszdBiD3Y@dB)*s+Fam3d7R>jw`bvuYY&sQq-_b|DV*u)_*Zm&6)2@_~16}
zI$mJ5Q28=()N;9yW+U=HYrjauZL7Qp1q;J?;&ZHox<3)-PuE6pPBe4&Fe*91QDX;O
z!8@5cG^jyM{kkLKlp8(+-yrZ=A3~9S31{2AJyU^a{&hc=$46xH2P66qoy@btlOuA@
zB;C2PUM>V=-2D&xviCgS196Yf{fB+2^3REk8pOQ`c(%>IxTjwOMx!tSXfv*FP9?Zm
z(UVs8u;eROTWf|6X$3V~46`Np#uZymBYl$9L-&B0No|m{OI+pi6#7SkY-Lj|UO09d
zUyr%WTD6m!8D;PvZt73IagOKWf4OP?FE>ep+{8NxbPp5GeCya>L$zeEBWTh~FBjBK
z!kub1XHa%4(|&IPOfKGWsM#b%bvB1I=69x~Ig?Dkr-@KE>)<SJ&O9rbc3G%1iFV8q
z-D7S3wbnLX&+LNhtdqU{VZ!YHu()pLBoMwrh<=k}C6~i+>ru;a)(G({qi;jz*;4$4
z>-g*+MitX2IQ>_(#$z3w+I=aq<wAwcwC}VqUniCh&MJ@~l7|rkj@i>k-%)lOtLS#C
zR#13>=cf+f#m&<c-z5Bp%%91coh$^}c>ast@N+Ux=mXJvW)EBWN0|6(D}SgXd5_4w
zuoC3Vwg|uNR&Z39^|Fmgndw-(bmlr$`?)+Z`Q`v^7Y-kxMET$Hp>{dFw0@v`a3}d2
zDYxObqN2^y&uBT+Yoa4x#&Z>-xgZ!D0N<G0T9W$@jhlMRp7Z|UAYpj~q;HemVUW&f
z6HYog|D0&7{^d8zA#u0~MIdEq{H~+UmUGpP5ub_O9m)E8(rqV`-U}1Y7qu|9DAXMu
zthbOR1Z4WZx<kbvker3uR_OG70dlN<5xx_aY<Ccm%-==aU9SFrqyk;3K9TLe>g&I$
zWB*$U`Ty60t7q}d*T`R52k~aW!FzAec$!E(aT2Wt1vzi^xCbDYM5B`dPXS93XH@)=
zAs7oPbgTpL?5<w499jY$T)n#%TZyhj=R=o3Fr9z{GORRz#8lt(xWE-#Vc8}d4%$$E
zz-rF_`E>;+)bPK+SySH;YYwAF(jf(qm#~6edYUrzmrjmo`XllAKM=7YriU_I%Fwr*
z*C6)L&2-;J38Z4F@!}j>B#8U*=0yJ+u$fmJ0eHi?uPKVia6Gr<tE2xOVDyv>fgfF2
zlf_u+G{nSMhLB(*oVQlgV=}XpB!T%n@e%`8DXbf&5tW3RK)Dg1SQSh+y0-t2@WNEh
z$?)48Q54Cs>|z;cq$?*A>fs34^Zo0dI|R3$--cj`>digodi&ok%dHkk0@0>{wfIGo
zCf;j>q9rgj<L7hcKK7`n!cqQG7j7h5J;k-Dxfsx1H3c;T!tx9(Opd52?IlK&et)@f
z0CheZ4A~G|bqZ+Q=ApAAt~i6dP5rU5P=B|fj&bHp7_?~3|3ykoIva=-?@ioIvePi$
z#&)5T)~2C1V%}!^a&dRb!OO^Tk5l6Qks?y^D;aQw|3U@HOCT+)xg9f7Qh-yVd+U|4
zrO2u+4&wR2zc^Kpt_5)_Qa;9oCD7TC1VwZ-rmcf3bN{<CJ46hCrxdWn7LYTNHaBXx
zw08|%W=@AA1`7V2LjbI`UE$bNkR?nTLn|(1BAcbis7!e*`!Uoy3tlbx?@TphYe3p~
zN_PK9COHd4leBnnXYV~Jj;)jf(?*Vn;RZbjawJu~c;5)<Fq{MY*C(8mb+S4D4_-!K
zF6h+aVi?{nu8NR6hRy34?#Ix!^C<I7aUW|Bmw-;&-&Ovx-2bmc_`eYN9|{2x9v{K;
z5ChL&0`EH)3qyM7I~X~AQm8nQ28lflSSF05keU-VxPFIc4{R?JP~|vhKdT4!s;Pp_
z99}r^I+{)@ipzFrSNw&=M?da+S~A@)jaP&t7zYV)C!N_KSJr&S@#GS2fKEtt*Bg?X
z&I|XJmDAE#muTu#*CoB9N{1p@LU!IyK6?}6m-;&OPDzLIli^M<=(Vi4y>%2uwVPK&
zwGp73A^+ignu7CbzW<#&51d={_Zcd4eG}?>t>hac4lEGnHz4$4YcBqf{ALYhY<Db+
z<qWOB^|jX{j(79%89Tgv<qIR5Zr5TsGAcZN#O!r~Q~c_~1>ltk#+K9)kYO?9_)gCy
zz4wKqqNR&o`R4nVL#Hki+2{N2Y5h>2ri=}aJA#%F|1MK-zf~mvZ!Un!LFUW7skNuk
z=~0J*CdqqKznR?Ul|JpPAIYh*^)AH2tFuoQz(sP-^;Q94hyeBopC9PS^)?5-f*l;(
z(V`e;Ou5PUeBEW%&)bq+%HBCZK>1#Oh(y=JJ%|1~dZEiLwM6sESp*j(Bct#R?VIZ{
z7PrnORX5Aish+^<Tz4sBx4cb9?+b0H9jZKc@Ar&(-*-fPNA1tqpxVv!iUwH%=??;E
zG7DB0{)HT<7ZIY>%X!MJwZL%c-8-$!QxBwNjM@%7gQf_4{eOlsSD$yE&U8HfwCMvG
zVj<h!S|sGnZ#(*ZE6KO(tQh*Eg3j#!(=wW$+{gkHFbJi2w+|2bE|@nAyO!L~E%>u}
zw0}Gwo!ZA+qOy4nUOhNS_)u*U{aA_60_rcz@I1o(t9dxi0DOQu-gZlwB&+5Y4Kget
zSQR#DvOCWh_civ0+-iW3z5Hx}Xj?_M#vzGE<r7=aZ<o=7<GcHarG6tP@!2ttyO-iL
z$2`u2<4(}qaXio&)1B}m*yYM6Rg894RQ#!TX_nlpV>DqnYOF2#2|5@(NyZW{)f67k
z+2@)D62WLEH9}N%FUA?a{rH?p|B_c5*t4pF!MFuTBv<&nA+!-v5uTBko3`S~n6XnJ
zw=u8Zn27A!eD>m-WZqo|gpd17*P!C|#_vw$$~^fN-(xM0v$FAb5}6Ivc7U7dh_7by
z_djZf(@DBlOFAww8ZRFJbLPEYrKfj3ZOENI)Xx|g1j%AM2V8#O^S_^r#3LbTrfyE=
zwKoh{1*EqY$Nyx?ZoDxWpv~uxKQRfa^NBKRLtV;2LcWA;seg=alc83YJDC1VzRaws
zhyK|AF2*oB5<>Sl^v@m-EZV?pD_yd)==TjDlVG*o<H2ax^z*1y0pZjgJn63Nzn>V4
zoq(ndvyg~eXRUtU7V_-~N_oNCrUPqcyC+}cK6`0|Z;#zQhUJkK=KPt>Mm-_(P7vPU
zcE8CFe-2s@_FaUQIiD@b`I7g_IQ}7)glWppn_-W)f&$gz&+_i~@Sg|Xl+M>m`sdI}
zgd>(Dc(&c2iq?jqKl{Y!X}@pgArPz&rsZlBzH=<<3;we|jA!&vev(}ivGF##k=%jR
z6;rRcEp<wa=FU4hdRKSx;MUq?_1W3ch0|X@!l+K=NK^iwv*-O~yKO1vi7Rg?vnp%?
zV|8N#%o>Kk8k(RhPK>+!SVr5BQy8WMEf92Pp?fDbUpO5S;H<s7Ed{l$bkrR2ZG($!
zXXVyhy!&BvSfx5<`<qLF<Kw=1jmAtzSFl5uEWs8GB4uQ`<Lkgs+bv;LnTHiNkwTgy
z+x5}H7>?>yj#$RPl#3?oyD-crtLc6;q4cf}LU@m>YN|z{r`F>czN4)B4>+T=MqtAc
zwJI~BMqsZKG#<`PS5VS<C$CNrV@t>!s=iMo*7fO4zGDqzo_gMiZDWRLBJTbW?MmLw
zRLT9pzUJ>(WCtX>vY@qni%vGF<u(K`-^c$*HC<oW*kRq5Q~fE-{FmB64{rE;e*eG!
zgRvTG<3k|r*`dEq&U@djpFx-$lb~7{7*0)>JX9e)M&3HH$NZ994p&E7-8$KhU4S%x
z_ic+$BZ|JfhI$V}-n|tL!|X`9<_-Qpt6F~c9G%<t82`;@FRd6ILoo9CutT+|Ttu;1
z_#IDhpI;c+VO^0EKJ7SUOoCUd@5RIn5^B#k!~It}oofgY((oo=O8&Y=#n55GSNo(k
z+3A2b+c(CTuEEF!rGuDl!5oiy)#(>=1Ua!fL~KJXjZi`O`%laONbiTCEoG+H+xo-)
zsB@o1|8VP^bRqZ|TC(MW68G#M!Snaii;ztSujy~|jFOz@>q?D7yXk&!?jvu@P&LNh
zjc<}NRpy#vKGv%7HS;J4fNc+C{YR3T;bzh*!1&qZ{d0_a;OoHf?JZBcF^Hk-?=3VB
zx$9}l{pwq56D6$2Jm^>GPR4iZwTNiEunYfl#@LCxf8_E1*DU`-jf4MQu=u~6ssEuL
ztQaiDAYPV@vBdhMd<X>=SgMd_r3L5@TJ8)I*cHG)Ia6K6$pS^}&Vel$SPs_*Tb?d)
zW5}Gn$Ksyvf>bK60ght=$EBnWtj34|y<f#K2orpUyxB1HvBV`Z>wJ(ddO?>p91H}0
zIj*+Cm)HaWw|pQ%F$i|00w%6lQ7~eGSnk8sonJI~<E%spBNs0L!Id)jEVy@&2V@WQ
zaGSvkii$Bg!V51@N3(Yrt$`~IELu1(m)%dR$e-r9W*J0RbH3Y@;u9zR^Oe@EFA;7)
zc&l$N9mLrJw9q3RtLnTNL4>GHyGDI*1NPc(L~oQBBBMs)acmBx0FcYHC7Pue)73W+
zGW|#aAeiXQNok;tLp?a57M&<=y;LR^(BpAu8396oLx8z?8snXq1LWv5d1R%7B(W-0
zOvy>y-Y_&UsfD|m_=^4Z70XG`v9p5gkf00VOy$IwVK;Pfq!^t<n^ZM<M6e>YGDJD;
z;iLZk!6EU>x~Yc0uB-8cse&OEj#^kd%@kybM|c0&P$v^~;vCx9kW@u|2_+=$<V(cp
zx-F*a+isvc$B<M7Bc-!D3%yigDy$Y&KwT(ThK7@40%E18uz2Eb3cX+EYK+Q{`+4wH
zel$3yp3d0yUEFyffc-(3la4@wKl@%g#Wi5ZoYyWpz}_<8e8W0B^3KMXl%|-Auk*1O
zO+WKQskrt2_VUb%9z5**!iGH4#m8h0b57qpR@Kp0(IhA}HMwSg5+qU2h<1?fZx^F4
z$J4+u#oQ4v=r8B9PLKFy!3jyAXXo@#*L%zSC{Eq`y~hGuF3HaIL!wdTonA{zL{9+_
z$tW~Coi@By$3ZS})&g~;eOZ)s8I^Ebb41Au5z!wS>q;8dsYEUM1&=D2YMP1N<BbR^
zyXZ)&%rK?cxJ6I502`AFClNuou75?3!P;ZyT!{Ird3NM=1QO@Izr?|ics)u&n25`#
zk??B25Tas0@{@u|HK`3?$R2oG?|HChR{xDrHBub(TK;&GQw|mV{*Q6RHeDP?V)17&
zJA}~*-y+G=8Rl3iQjK)GH&89&hb=-)5yp{3#~ON4@+_Clb#Z>qwmSM+<)riN^_($%
z%*c#e9<`8#g&~Fe74|)iu`R%iyG=WQBP%P&Pbxp<#Y=TXAvUUPIH`6R&8S}9b^#G8
z!#pC;RNVhYbnsoq70Qyk^)XzfOHaohc?i|BAjhQ3vnmj?Y3=V`GY&BUmXM{NXnJ{b
zoijCT(YFKwwFQ+u_0D&+&sDKEun@?renBe)yLkzG-+uo6$k7dNQ6~Ok^W4*ms<_2Q
zbPI1p{n=@mFfU#R%OY<JaFg#`rJKvst+zCPdHsp}v!!No@}{rd1kbn|v31%(A*VKH
zW4AD0OcvkM-8d0JSVp!%D6f8&LE5}+E&9TMd2ce!!5gAXi3q-<iZ#;l!BDq_aTSJl
zG67#ceY^qLE<ac#Tt(ZeYT^PCi=pqSD})I}B<)N~54IKrQXbQOfTz6ZjwQ}(U_!~@
zwqw_)C65A$v7*i)4eiV>x%b54=_@KMX8`4R>F`MkOeor%ytb&4{gEe7H+;d7e9UjI
zL(>knlNx&@CnJO9!~~;{5-wE0v-Opjs_3|-jYLh_Rb5!SeFc6dkV_Cse*0^V-R%#)
z5<yenir%3sfutX?ik?R>QcA;IyV4-uM-T$d*cWbYd?tK)vm1Yw*-8wzUSLFr`wgTD
z+K&@Buxn`2xNSlAEg@XSFXw=K;S#X3{F@dBy;RtmCq*bKEHjQD<XiT4VZvRg&gr#s
zCBQR?fbHaaSNyF%Mu})X!$eY?sY_D>vAljIzmGc_B|SdUpm7u`QZLlp4yn!jV_5U7
z*}Et?;D!+QVQ2BEevF1o!#ru6Cc`kJWlB$@H&^_UY~qun*NvJC?~T3deu{}B*xyr=
zsb{~37Mgiokggg{gfsCvQsLN!Mz{e<qW8380~(=GGDyE?zA}p{307q10i(PyO1r+Z
zPF}W=PV+I@E)?3EbiY#*A$TOCQ;G3M%|pzA&t%Iv+tlr3Q5E&;6*o6Y2aP4_;8aNy
z2^WQC)vEj`cRbFv%rhc9XtfZTi7iC)277^H3g{!=cNbfq{JeUBF`lVIvvEl)@n|!q
z(~<zn7kPsvvM|4ASne*j&in7+jqSue1+*@m83Su%7;&BLvX@nH0RZ^Yh`;*Ej+&iV
zZ;safETGy!{Z{F!$7dx(`Au*ot{x7E@Ti*9Eh><kIO6x%#*)yX<VVcjm$llvks7+c
zikX0n-u^de=DdbI(ie9F61j!ck_}an*fS&1(w|Mmv|oNh6}e0=ez~ShmRn@!2*E|y
zJG|K<u4n4GLH))F@vA@`sQjKZDWrxgDZ0M1C%u4?*8d#0P4~kpqB>2QCw5YoUE~~s
znVVDHEm}b%bVnuNx4C-W>{h=Fvs=H(#^WDA>s|=k)a?HQi{)DY43|Y6X{Us*9;+rg
zvPLb86`{HEx-=0HIK-Fd(_sEibC?hkEEr(^otVzy$%v=H5p1C<+^p*-NT8i+B_mm2
zA6H&)lO7J}^y@e&*4>0z{k;kleWoUnRqGZ7C&Fv&z&~+PM46k((-Z80+(u4&G$LSP
z`2F{33@`G+B44Z(Y2C9{O57diT%`}wjz9tI2F-0k#<)ZzVARZX89mmd5%{)H+B@KJ
zSwUQ*+=Iqg#*BDY%ikU*9S`H(e7bZK{6!tqAkQC%jEFDfQO(HdSy1f=qGhVP=6Lab
zU)0Sf8XSHtfQ`7lxb9m_9Jx9>VOA73hO5G(FacmsG;SzfoiJKa_L1XDeoRbe@(oq^
z*j>5>3UM<gZo{xqAw;<wtycAW8sd<_tT~!O)k406Dh3DY9Xqs3b(foR+3}0+pW!0G
z9>i!bPVHPsyORys^QEPqsBhoBZCCeQlmK%mWyb1Hv<zJgSJ{Nh-Di(phj>gU)kH-2
z>;_(1j3nG~7IQvb$v!m0<e*i_T%Q_S9|@!8k<_U$+cm5T%U63yJ?m0Rg4383!jxSd
zs3Z0l+y`*Xr3}2PBZzWR=Qa{phLOa4lJtVbcLDU#<x?u&ckpT=WnSyIGK=9oEqr!g
z7gN5VUjnQtq14Ct>$(qh*qxHR{U4mFAzd|im}=v~$dO?+NJ>yRft6)UnBX@cSQ9=A
zFZ-7y+{TWRP)+Z~p?6cpdd>8moMG<Iu=fi6HMSwY%Ufs8An4RP$KnVmh$<+rqsd-}
zCb|)wN2l(jrPT>#{xVDFErX4S;x6S>Aw_%HI|cz?a%UtOqvntaJYKhnYcm;s!)}ML
z5@p+PjzzhWF6x-)C({zFg{Pk&Y15omBdP2&L}sUN2MDgN*P>1gJOB=c2o4_VLTHp!
zqhC2GMsf-y&$KW`%(eVT)%S*B#{?@G8GRY7Zdb7FcMZCLt6Ew$<dHyf?^>KNnUdX{
z)w@Aq6Z>iDdpn;h%6fcYGaeLs?08}N+PGtJ%w5#>jqXJ{P2#bllj!#b&dkzs&FvHk
zfaP)9RI#*Xj}ThTS<vMUwjgd}XIrnCPO75$fbc*iH7=6TC9(7%bJZIijp9>*i%1;;
zS+_($o9pskBD)NdS?`F2@v3f<EY(T36Wy>Tb>02r1uhE1@Or4TybX+y{jbu@-qH#D
zI3>{ZEq;wRf@WzO*mhsi-l4*9(Urg~hO_Agt`{uE7|66{_KtI6tS|zoR^ZZgSRqXm
zXP6|RzM^Z>JkaND0vY4t>cvNV?@`jdA>GAvPDJSv{(?;X2?Wc7_VjNj@*@w{fAHV)
zW>Tl#2_?@4s5u~!&>SYl%$h;avLt>az^l4BJW(Ebg+rZaXBvIO^1;_QCGB>|Ny>wl
zI2C*+y9c8kPJ-kK;0YL$XU26JzGb%4T7MV>#qxE*28v!YXk}lh)A#x9AX_#YDdDoT
z*sBu3?81uPbSun|H#vul;CvoZq9jB_FGid{f4J<F?|uFczH}$pPO#$R(RxspR2y;#
z)x|5I5?}%sBp!r0VCn}S!N_alV=jFUlZV>?&lkH<34{Qm;Uc%T&vA0(N|d&Ay<~X3
z-(!gK7Nwyh;4?XGrva}W@lGYWw7N^34}ZU-3BHQvj<wwC+4C;iKjQ0JX<J^0I+F^;
zi{_KIvmfib&{t;^5?!-QfAd*<fyM$td#l`-EHlnsn0uaasrE8YYCM`eb^`0z71o)l
zjx)bOQApA*+>%())dSw2$UcaP^*SKC|Mn5KzSafioJ1e}z9Pa4<kE{sk;MpS9-Z=7
zeY6-EQ(1tpy+BkVl3Kbh_L?BIzFLa(^iiSJq9`Vzl)g*(Q89_+@P*%*$%eetDWIBo
zlvNA7txmfLB>DsI4K`SxCg}PGRPVkfjb&FP@!!}{4;9+ZtnpCF=kf2*THle?8Ix=J
zajJfq<9-FZ;Sr!lcP#2+zcmxvM!r`pA3@;K&du=+)B%;$>b{W<vD5cD-iV6WYY{t<
znCkIed5ksj$FfHaP%TUCrRKOH*#|8b==)LY5KNm^1na#6a?rwM5d+4Nd`M2l%)91-
z{7nq)7eNp#x%J^YIm`HaJBDp8Z2~*9(4{mh2N2W0JhumTx6rvMVCBet{#xA&;eHjR
z=kr^<84_P835o&VZ1dF83em)exzLY24>-qcCi)qQJT|myui0ErS?Vn7%Fm<zbUon|
zlXe>HZV<%fS~CBvcA#+!8~PRTNB-*kO~}OwqjrtjTEd9f7^kk%XoO#rvG(id_s-pg
ztWB(@5(|TVB+ICp2ac-iik5};&hyaobws<4-$w$r3_t#W8C(UU%WJ~by9IGdfH1UE
zjQ^U*<t@?q92N$|TecCKH{0ID+46el2NF~*#Hb6h{t*^@M1BPKjtujwZ~;>YKyx<_
zt(ey(Qv)R00Bi0PEv3!+alJX2fY_Ev7t&FfzISStTbTi@T5_Q@c&HNHk2XM^XjRno
z+W;5THo0xTs<C0?Q9CVy-mK>bGEloOJ!0`aaZQ>1(F<?x$DrvKZo;0+c*97v>zjN@
zOfj^#HY8eP(Zl71)wBlc)dzN;+bDHl{FeyExS!IQdC>*U?(A-%jVPxdfv?Orm95Q%
zvoC@7Z-)w-G_zN)Xl2w?j=``6D4S-V*p0ZhSSB>5PUW&Bdf%-BG%oIZ=g!TsbJS+F
zta2+`rCZS1^8$tF5z~b<(H?G)D$}^|YfpAK3Zx#zEx1v%B!qQ9?^y)=pz~j0x6xUF
zIl9*p7560&Y7%O_xJg{uut@FK74A3II%K^+3M0CmL2b%?I3VtnY4Y-h!L&<k$i1)-
zzVzpo=LXo#+dW~agpv^2AzDLzJ1I%|Xy3Cs$04G^^&8gTJdR6VKM(7?|AX)QkI_qd
zPvO*M;3;@@wmYZMvny*<UG<tyMBGxMT8R5zswV+#2?X!AOktOr)hMT`pI)=#JF^0C
zkFNM=cz$|qSd(Dt&0TUvtYbSXbKxV}J+DjMw@CT$CGhET_8y@$Y`JdW5y$e4DkgP&
z48I+0Hdkc!b)q3W*s2*903N{ZyD9$Ei!Yr@zbRfCb^SZHzFyg_pH)z29*nP!PZ3xi
z_^_1+aj^eraMzEnRuZ1^+urys!s{C4RO!=e<6L6EF4)g`z}wY%ZhSW>c`{0I()~$`
z`490ldG4A*!WQIW_QGM_s%#5B;FAsDOR;0{K(B0IUHG;XJSg*&*~PL*6ErJIo*jHB
zhP%7B9WK}zUT{dVf!caf9EtEV1pCzv?l$_hSk?^^&hpH>?snW|sjI!TsmrbxGSxA|
zXJ>5)-<s16)ZY?Ut1?`!7&ut~Wn=Y6O7B-aH%xs-x)>f6!>KZJ-x$-UKfm;+8n{L+
z4w*L%`?O1fxrvEk;9i-dX-}1&ehjbd2_(}6D%}}plm}>FO>8}2mMcSWH#cx1O*r!R
z#6Y^#gq1>eE~)O}z1-4k26}^zOX9t#k$0bszV@%&Id5wH>dmnDLZ&?C#XTR{&uv=v
z+4Hud=%+DiQa~p_BOD~5K?7zll44o0H)<3dpH_SjFFR@9yQue>@=+z}*fQELPe&NW
z?PLS2)NP&+<h#0)dN)2JPCn+@9)9qy(;%Vx;xI`8P+u&*?DO$d`LO*n2v5qLmpc6E
z@_7ww1dHorQM7M)!s?SI>ypT?igs+SLt8+Rq2UjwnwhGXtj**ycHsD?_v9<N!z3-h
zEbQ(>mU+9-^tk+tj!|hU<5%?{`5)Q*HkEC%omRSzsh`xM$xJ}`)o-{Mm3r~!b{?1G
zq7vBc8nq5nO>Q>q+0J-@Vdj7aE{3eEit?e`og)Tre#IZ6xrpbnpNDCAz16Z52&n^A
zapxfgYM?^o<5#aO$dzYnlvfjNbfRGOJ~c?&0hxREdPjMg06Fqy=T5QkAwKvS58wom
zgxb*i*MQe2Vx=TLoVQv$=_Rkdfhj_MK*Y$Kx0M*}_&Tm!3$hH(ZiQkLyY8>_cCSvO
z_C9^-;<pqA;rC+jq;v7*;O<6Td29^l?poA6WXf&2xr29wrh%Wi#Q^!|L$aTbm>E81
zm9Z{ufJMa~sNN)wl{c$NBbW=Uti<=G6|;E0QBD64Z+s*4m#L6D_{U~b^2uT95eO~Q
z1!KoorQpKZbcB~W-JISikgSwZX>_Rq_Cyt#`0dy=xfv&&TzgMJtmlPDh%_mffkak1
zp(E@Q`!z0=(TqQXgjwoo@V(+TN{VN+W497>zTl8IZ9%K{vKh26Rrg{5UTWY@*V|8X
z-^oWr*kAe5#-36>EK{8vD|BT^XuV&y*{c7=i%TbXm+(9Sh!&j=?aYMF09yHW1-%a}
zCf<Vu0nK3>@Xsl|gOkimX#+1xC}N!$xQhvN$L8OK&t6)isnOPVqvq=eg`yD5j6yP(
z?jtkL;H61RRL+pnwjFYmWYT4NAuLA|gR?ruDHitg74cKo;+jVdb9=ufT!s!N<TVSY
zNS1;>%cN>M^nCTI1?oxBND$q|3QU)?UO+)!2|l^!7EV$RL%O@rBktW=C0HB)?b%p+
z7lC(egaz)m(#`hG3AvWs18VQIa>^9woQ%VsOx25l`ivHoZS8SaUMv{<6y}bmv>5V$
z?qQ4o@;7|Ofu;}<hK(UcQj}5>cr~coBV&(dh_6^*i6i1tr`oxXD(&WtQVK0?o~eH%
zm-B<8b`6vdR?<u$2`{8rPhHiUXiP(z`GyrsBi3>sEv*P0O;mR|)I+sI?Xve`if>{T
zC0iUcN3<$_JzN{TH0Wa8*V`HYgCe%p5A{4c=qg81uv8{(2~4_VG~2(ho#mK3QxmkG
zj93M*>bl39+Klg*0MXVk_S!WRgAD-iL8`X$n;OQ$kT?Pk1WxF3lpP_YqT%yAtLBw^
zuXd)H{gQaW(!#tMj+U!nHd*1DvJ;hQfr-1k-0C{Hb^Dc2=em3MEFK<Qg&?_**b6h9
z!^fuIn`K-WMjU7J+5@f-(9RC8`p!H|^9O|FTXgKT!-dzc8?l}?x#$WB(OumEHGui`
zy%^y=@=R7dBeuE1S5ByRDr-Pe6*qGCf!<i=`fa<qyiiGg282|#9p1c=^r@8@6RX2h
z;pxTk!?>`J8Hq}ocU;i*sqxM#Mxni*LH9W!!t_GNoWd*&LXv-2T?Pr&dMHO23-w}S
z0PqHQQ_*IW<44X&uW$|WwBfVRx@uQgi2DU~${h*R>q{IVuI+u+uNZ{^?pLBKgb;jI
zn&(#u_61#5AuIvf2Y9m3<;a^%a|7=ES0_j9!(UVOu;e^Fx@RS5Jw&(;+AIUr{!D-^
ztj?v(VwKa`ybJft()L{~aO~)DB9$@erZFF>-1-OZLNz9cJ|$CP+2YnRK(#{OYq5H6
zSnK}5DWJHC=W<mKge<GA+(Fs>!uu64#kS;>jkH`AZ2$?_nvdY=2PwSlh1jEUR_*GE
z2Q-ZaqBF7<EcIf)V3ydL4sS{r6QhGs4^=o)JQU&k@TjNti4}jUZ{h=T&2x`Kg}Y8x
z)ln2*-O*o(dj+e-cA}iMYn&C3q1qpk4VxI2jN^1IADrsqF0Zv+Z2@;~NE>il$wg<5
zCuEA*Sr8l6slB=)irGRdkuyR5S~Fzb>4OkOVG~6~_3JVG*f!|RY*Bcz_OX9u@C`P_
zSN?)c;sL?F<BI?00yHU>=)Qf3EbCeVUu)P~e<ED30v}<-%Y{Ap5gx%!Oe2VM(xhpR
zTp2@j@5tsyzEvZ(Kny301=1}TJxBRW5sLv87Bx5Vq@WLMU9azP;+Xi3ZMA$8Z|GH_
zCU@O20eN>tjx4xxh9Y(!#pnc!E)Wq*k|QT46rpEg?|O0D1}Pv#kjz->f!p)~DQ%|S
z8k6F%V$q9F_#WEmMhOlrm;R>Sv8J)OqAUi)zYsmq-15?Sye+Rbo`klIJ4zMj_dXuD
zgfs7M21BKrL_1QSRimC-U9uh|fXJ^YH6_;K(i_o^BCdC;!D=Pe9sV(#TlZwZbXF8H
z;^ftcqsYN16Y-!BRjXV9Hf>i6lXQ%AVt=Bax7sGQn}(yvUt60Z0u?M6jb5qRJn3}t
z!#osu6V7}sl<2i#xX2<A+SDK7Xed!<%35|6yx0&VR7C>wfZ7wg$#uvv?<(7XVa*Fr
zrygBk(3goe-@Ec-<Extd>+vxwCLl0#GFj1mu6+PRW7O`|FId04>+fGA^2SR=Jd(a}
z<@-EnR0-JtROc3%KB}Lp9PuWcauIq%^hL%Ikmv}fn#g-&OP_A2aCH^8;75z@2Ghza
zxIYHQrpP{U_gRsS^<pc{ar@3kcl7n`4F$Ogo_<*!i4&(cA{##xUU|^>4qReZemKgG
zal?c;)9M5O2v}0pS3BJpTS(l>>h9pK8C~k>koknW7;tHB#rqD(P(X=rQ`X7FEink5
zeJXR5<BDi$m_)vC-wj=rgx{ZjOxbF+Kiztg5nuTqi|WMBVDI2mfUfqUXzo4xjO+7c
zQ?-d&R+3kTAjn${LzUGHsbAE*&GH8Ma3vaDD})3E^iS09nB{0bfs{tmyxa;BXBDiQ
z_Un~J#jg6Bdzq_js1={ji#lM$KD_!KC@)6qu2$RN%}*6vE}}<(n2&muE1L9nXIomV
zjK{+IM8eeCA;%>NpH);G_kkW-X_<CMz2}9hoo{n122`_u{U8|7q6vsS>0)n?R(fxk
z_7TjJxE~y|J5O3ImS{zB={@!>d;F{hIPOz)clV=LDj2o)kf|q9hFvoXV?E(t41Enu
zF)Krk>5vy(GMvRA6?pLhaa(sBL9dBq>2uuVGm4}5W7>&7F;5KMxUV-zyEOi}Qi)p_
zMxa$tgtZqL4V}clc}<{ULPb0@hRc$wJf{8e1n)a+*k|Lr+;jwzGSDLxuyz2^zZOSH
z&IHz-Y<v$|{6pnbMwaHk)bKIiAD|zbsC{+dPzH`oEg;@FQ)`IyM5L|U?fW<i3^hs$
zGZK5dO`0IRxJclL3lyIIeWxphFpNTIm_szuBZDjztb=d_AzKJ29^J=voK-Q;y7+7A
ze2LLkbL|r)4XXsR=)87z^~=TPj+WCM;*B^hJ`1Y8VDtuAM;Kp}ZRpcr+zrHuSR}p$
zR0Fqmg6!6mYg}jnvOq{rr?APbc_eR-t=4{!g@t$fft-aNQ4~d3+(V9JKk`hMVj5iT
zsGX<BOeo7*2g*%jx|`5ciS|ZE-bpg7X$rek>#+Y@lsBzVf8@(0i$!fZbHxwOcVA04
z+E*?<>(#!#N3gPCh~M2^?5w$&Ie3Q+#GM|P0^vzfM0Jh5R&<q+NtXR6^~)>_NmELS
zkpTl`?6w?Vql&|yL@aDOnCu~2<jAQRE8l=-SQsU?Gu9{f*!6S)EdX|CFDG|(?(|`W
z(2Vt>cHTwa_T1G#bVRBQk^k|onc-Jer62KVI*h)8Z%_w*MQP_IH2fem4SCmI{9Y{m
zW5+fw)BA3iXttpm?KLUu`grWq7lRF%kO0W*$+a)*R(yo2vy3b0$_2<RWtJ(D<AA7R
zsgy>=Zo+u@)Z99;T>@@;&Ica+#he(O0A`FL0##h5l3M8E`&%E54&bX$%ty;BIvC}5
z0eN)vcv_ZK=5#B+B-zhQ90W~?2-dfPSHqBP64_!2z<}-bJxvs%!JiLo&ZSj@V??K*
z7Z@XOcZ;R%2WAYop31|gDPCS4&~%L6XDzib`yQ>Y$3J0}w?}XTUGVHAMxf;#G)%mi
zb4*RMbU!0n6bOMuYk64^E%GhMV?wL_B>7FA%9&}9`x@l}VWF1@3j#Vhy(LgmFeEeH
zb}1QjVsCqdb0ju0%7{AAK(gvUC~d`wk?%yOOPfH@+9*!&-emr9F!(g>jQf~o(N0UW
zVDp$D$c?Y+ch6gO;B%}ymn5|>Pc(=AI%o7|z^p+N$+7h_q)g8Z7W$Ol8=6djIip9e
z|CG{69=D^XdT#*3(-*NVut+*?qISbvtxaD{8|n#V$}Z^ovlkieJxHl|H4d_8(>gci
zx5;9hzMiEyc%vHqY?LB)QPc-dxU6GR$2*JBEX*QsVU@84I@_^TaU5i>o*Lv)zANcQ
z_g{=26HNWfT5Tug2VZ}A4jO3p!GJ-{wbapTRW3{ut-WmH5=|Nj%7ZH8*%KfR>&SF}
z4hl6rAxOL6HnRKw{6y^#8kU0}-<OL!AkQS=imz<Mz_B)%i{M*F%P!cxl-N2Z#%ER$
zQlw!1SP7q+*Nb<)Qn6gT^H0+Ze|f6s2Yzqtf1Ai&3<7f!>N)z|Xq~Pr_(M|OE=z13
z5HO%F>wPS02cfGiZMGQZrmLIt_d)1dfBWmQUuvi&Y$tdH^Q9*jU3}Kef)+xX5fhcm
zd;_=XjY?P!g-p`-Le(U$w@57Bbf&}*X!zc%feP|r+A-)zx&08*^rW+6yqg}O_A+a?
z!P?_w_}62?!iplMbo`E9*r(?>6$lbPkWb(#Z(Y6Fof+g``xbgl42+L~7p&ye4vwPE
zWqrm>_LSnr5+lH&)J5iMGjKHM>RVra14Zu;3iOU3lb#f`55f%S$Gp_e4OkfRV(Ej~
z5Y+EGn#SL@3fkWTg$l{QWiSf8RFZww?_#x&pD6H)_mQ_>>08F0$0GTfU|_-a8;+pR
zMHf5cKG3P1+ao=rt_U>)qKjg@zFcwdBgAl)Q8+9ZydbA#|9Ti7=14zgs9^aIgkA!<
z--%5IC!j82vmF$V^j)7Xe1A$w?(nlAn(C!4>u=Y>A^srp*~g;_9^9o+jCR*CVUkf2
zfFFn#BYf_?LJ$LV{u$h5C)$Ih=$1$1Rf-LaZR%qoV$?OsCj?3~DVXhlnS9~nW%62<
zg=`#7xilMmnH7k>i5W>Evtie=76#1PD8iT+J0d*jxi#!ugjZiA{|!wD_B48R^uC5Z
z)!!%T$&ryUr$m3CbGlwj!uQKoQymc6lV8rmaexu1{Sw$K2$&O2iDlC<j+cuQ%Q$d?
zTTvPQ8uF00mDC`%NU?LhEaPG_)?}+yxW!bu*g&fFx>tpuX`H*fc1On@Er9AwEn$o&
z#<jT>Y>+GAg&{l;Nm2f5-T)KVxrj-ZHVge8O*bOJ&yH{{`is8qXYw&lJPGTqW%t+u
zS;o}f_1nqYl0EMwVWYj}*9AN1S5rje6Z{liHb+0A8FWI?m8b+jiXobXl5Cl^7>G)g
zmcb6ADB`pwpfpcGUtlO68*XWF3$*DnXULZ;B#pAUD0bKs@+eEmrYV3(&<^uG;n7JV
zz{n<nK1hJjAo2L8{zkIQoR|lrFw7zJW)_L{eF}ir>IauzZZ~C@0v>{TwVknh_03;C
zU~ed=bW1q+VqE6QhliE6xOK<TInj(gDl7XEwhl(Hga{=ado+Q|?Y|FBUBp`IB~(Q+
z%6j-~j^@Qb>o>Tyetcu^-X}QZJ)9CyxN>PrNNMizGW*Q!boc4Uk0#s?j1U(lE?U9`
z6<<%T-icH_N|kG5r5b=<%Yq>1*YZ4B?tj4O7VO=v!qI^5`2v{x;>&-nuFzWCk7h!e
z$1mHIX9L3J6Mz#Gf9KxFT?OPC5jRie(U^2^Hxy8Ygwq`dIj=#jB7E<L_gW1l3S5VN
z5(OyCE58$X3t~HW6tDD^>aq}it^ic!(6k6=53jXitIIYT{=f>tYGG#0-U0ir_)f}z
zbWN^NGa;H7v*jg?aT*)w%z0=F;33#cc7j(DtO0g|3o>Vf*)R&_gfR3QHsdVmPf`H(
z^7EY-=Ap8>EaJT?i{+C36P^_&c`*!$5I1n#L&U>V`7q1HO)Gztp@Ua}6#$R{_uAtx
zo!yn3xNAdv#aXEIv3&_&*^S4J{Z=(+BdCK@nVJXFaGN~tSGy*4o^g8*eS($BtRY9)
zetPj19aarzEfiYa&%JtFAhDdc&m7BonBt0_Ty$As0**Pc{Fi=mbFEEXl;NNTR=SwC
zZ!@}6W8=ux<_jMfRMw~Ik@i54mp?1yl40rcEmlZ(egMYxAN5NS>1P4jPx!A?A1+AW
zibDpriw@zBZr|rkg>92<s&TA%BOO-Rvt6u2CY%Yi&{1j8%sjYJ>H(uf_3P)mR7mcJ
z4;gaO?&>sIJXm1_a;b5u*`7G?asyA`7^2~#O|YT9P)39;xuWujvi}t+`U1uwZsx})
zSyAFl0s_6YR|A-lZMKt90vFhQ0DE@P2|R~^#1*kW1Z|u0ypGI{nFI7kw1YcGx{H<Y
zgx7yIaR0(I;mNv!hBu?2Batm-pq>En2@$|8eBHH);aE(bksyL;L>pHW_b4~%VVx`z
z&ICms_h-{%X6Bzz3#YRJ(#b|2pg_MG;wFH(mKKf)$@bB*P9z45bU~WA^MwpYJ}nG1
zF@-7{FkQV50hFkAFCtBy*YX^hVioQwC3m@ikV?Meva5O0$X+{}zclw4GI=kG#a@9p
zC`G(q7{Cf*EaYEwWElB60UA&oD>YFvv<`YBPNV7NI%|`~m7-8W)hKzY3+pm}rr{Ib
zDZBVMJ*pO%n^?Va{T@GN&7I9GnBfT^+h>0Jxz(wY8J_bv2}ENAz;>NC7(TPWP6hy?
z#&_s<w2)MOG`@=8q$v`n3A1?=IAL((@)tst<1Joa19$^wr9U~K58{dfExeyr4af<V
z7lJ4XBR__JhaaWJSwE+-6)k%<bZ`~pk8wqljtx%<lm6z>Zf=Ms$B`*t`vqY)Aa2an
z^PJ#DcDS<}kB8ms0np^UT$oMvtE?}fPvt>^lwuL<)r1Q$1JVsH%=zt0Ux{BCu$Q*c
zrzMP$RBDK6Gsn%?ys}_3$aXo@48_OmRo!@%d!|HHLA*}(8wh#gEzR_3hHO_9dDD4q
zm3QOAbbf2^ClG>@UIrX!sa_SuNAWKhk7)y75+kXlY=rx(>{%{W5FtWlvC?RKa&-JN
z-Qc9{i!+Hel$1bLyip5b6|;d3UMZROWUn$%9Hg9krSP!THa0NsB<^m&k<sjyk{c^y
z-)78j)Y`Mwo7?3sa8JUvg<qt=<=>FU0`n%VkT8}{=Dx*NH$X_eeJ~KZsZe%3{PP$?
zF7x<FC&yH*5$w7vZ}%r^N}%E+`KL_uDOA;*IGoJhyw$*(%Mi9hlJ=D5>#!{Kv+va;
zr%t!Hbqj=xO+y%fq__YQ|Cx%}Px0)upwMci&HlZgmtb@9-o99d&A|FEbeR2&-b;`4
z6@c)`B#6+x;#i5sd`Ay2Um@4eRIB6LB?sCue%Xd{SSCyo{E1Nl8}WORC^l(w#%@*?
zZ-Lh;Qx}~pj~=#D*vH+vch50qL`T6o1((PEZmu#S!GQj`8%Rntup`g#O~gmeM2qI}
zFaSXNkzOZ#NwpdE!gfi+>Mx?4K}0#~-v6`LF9~sG!)~DcC;`@ZVt}Hb((9(4wD+C_
zZzS4J^lNT1L#8?K)Y4n0G4@Frwbc8BCsaHn@uNpK*>6SFk>Y`rw4!8U2$icEbAtai
z)YDp?{H;ZrXKV{;NO5g8<VSrm(5Tx|3#oTuRX~lGs{{AOjbIpQ+W$q=R|iG;Ms079
zlF|)IN_TgMq<|tF5-JVSuuG#fBArWvfPm7lNcVyu62c17wRE%J`g`AZzCYL*h8gyG
z&V9~xu9J7+Y-ecdq(jqG-m7NXXhUMGUn%Qf(gqd!1YhG<vP$(}OgCT+>+d^cV+H6z
z#mui8#V+u~PlTAF_|%rqQ*3$STUHd<eFan)QXIie78Wawt^@sRD3ZnPNOa`x2G*$4
zV|NosVkbz3Rx(70#Xn%>MK^wBh^PQqv3!}+q>m#Ch+oR0Z?mjUtTQrl!Pt6L76y~Q
zzWZ6*lq-WJOTG~BXsE<1en7B_v=JR``|<re+c`q|0mG9YFRHNS@NZN59>5%tDl7@>
zv}uOg$3%b*B1yn4=_NTbZTY?$62bObM=eSp+zcZUZ1_WoWQNec7k}Ja!SocZz+<5>
z%s#UTeF3Yx725tVZG^c%L$Hdr9f#vs5<%VaJO1?&U9^zU*gbu;q4xJ<1(_OK*GC>X
zS|91bpV>kbhE&P=5+1M=koZ6HVCF9OevDd%hctt#4K!{V@RMrm$J86|yln}FkIABK
z*;t-IwxNuy@N<l^lKDsXRoqw{yOWx(#j>`73V-HV_m>#$ciS3^`~fH?JXdO!N5&y0
zyp^6>^$H#j?sePJlsYPHeH%toB$h+J7KMyLs(FRJwEc&E#^h)_8^&yWWFxbyS1Y}1
zAw4@2NM6;B){ukYr5NhPl)3}WtditSSd7@xXrv6*Tf%+WItJECP+8G@p8TD&H8o5U
z;}0d`Uk`*A4$FSf#=e{ZXMA$ujW#6u2OhAQ7hq)ZAFNBtVJ%jtBCl=r^i+KSQW~kD
zC>=8O$k-aa6O3!ZcK;wVNgN=bcUt$mXsS*IA%M0{ckqceY|XsYKagc?>jXKxG*dU4
zFBuB{I*oX2^v?lSQMj<Q0_(8>qWFJO5EDO?96#C${@>R|YRW8l_c=vo*Lm=Yx{>aO
z(H&HyOq0__1zBKN7`$)>5A2fuRymcP?*;__%ZNni8C$8dT1&uQeYn3Tg3Cbli(VE-
z+m@&@PyvQwT;cf`^RqB0`(i9y7GnAGKi~INIH#}m)>a=hhzLOU!1rlTpM(?o1-Tjj
zQ?=Ck+etT%-dD6*CTocx3^h1ass(1{uL9MhFt&&kGQJQ<6Q8-!obX5hDPT)O1Pa&8
zffnFfoQ0Mx3gd#CBt(1_^GB!5XhWf=tey=DSbMxbUU1`avRSk0<GSz+I&;(f=;G%~
ze~Tu9(S2cYe!0ZOjiz;A*L-0F7mPm=x^bsbe<E~MkoxBwsGw5K8%zhic<~Im!&*Rc
zYHcux`9mNXsvVvM@7A@k=r6$_H}di6hu>IFkMDoN8Whm#c0ZGCX)5uZ(A_RmS2boj
zcEgr(9&T9!Mo`R=3L9+yI|2;0Iyf>89}X{m?#Z+L8$Tb`E2ym{38;th!YXK&Op*DB
z344QrN<+1O-i*e#5G_|vd1>yBRshunUi=Q%-Z2J6I1i87wBp*)P!3HaT+`6QbQ0m`
zY?uYrn%-5g1hz65)SMZ^H-f9iF?^%c9teq-1Rjv=TL5Z!du<3|ssFucTykJ)S_j6_
zUCNSRbhMq2O?FO{J2O~Ex(m?3_>071b^nIEkoJ-b70NP_Kyz1vMs@iW(Gt$t(nUuL
z5gU>4;_nA{VieoP3J?fx&Os1z+Xov-cGUnq@GE>UvNQ=71&OH0=;H9<H&goaHJyB3
zW*88R-}@gqXGWL~+%N#=Olo^(i1iTn6+MoJLdZ@QqJOyJ%prW$xw1$Hr4qdnUXp$M
znpfopa7p~F+;ATaw2>23@Ysv>TlrneP-tsRN-(w3J!wcrzeu;XildYd<R{sawZ6#u
zI4v0|qB3@5BF#O85yMxZ0@Wvp5y(Bi%xqkQ7^L7*lt_~P2aiO&#eWa<1#|EXf7+Bn
zv4bV+y>aMHL+WMig&%0iQvyaIKjHgI#?KX;Z_86_h^*W;3bPlTipetyigd%6)kdk!
zp%;az9PUbuQ3oKZbo=^fLtx~3n7A3vJ1be8Vk|Ryk3ftPXl0(NZC4s&Br_W;HZ)4-
z2YA7_5b#7ZNJ`zFhBEd)Md%MlxaO>Uq-X}FgCBT38mhq{F&8g}|NR6}CwY&}E_)iT
z>O-ZQpi_%e6+eT0fM)?dW6ecm93<e)LfUa_-a#8880ZC6TW2Lko%}7Bht<!$cCnN3
zG3vnap9zZD0U=BN6rV58TyU0O5kZMqdHw@j?zf@OLJu@TisXgfP8d@<@BmMtaI`Nx
z2<l@dLOhGdkOgmf-xGwWK@&kkG$732?gYm_INR8vu_J)HWvqfAVU#;O%7NO)g#k_4
z%f%_oxP`Yz0~F725uAnpfSW<;>gK{fEkxx|(`n<z5kYPcPeoxU;|L58a;q>56a3v@
zr_owR1}ZXPFH`+$Wss!w|AIMycujkZ)x?vrKGWRmmx|XS%m7LNyd+RP@QElmEgL64
z9i{3wAfzD1Eo@*>_-K78=9#f0*uCV}p#+XjX>(^pUl-u&cn)69f%5;*L~HoZ9a<8p
zy<%Cw3YM4dk;K~bsH_9BfA@vT;#*Mt3g90)K@xv*F)-way6+y=r;K}DkB&BU9a9zg
z6><#)V~8UAmj{y2KIE{BqPTg4Qm^Ls|J(&f?ovHLE;iL@?;Gdi(_?rSkCM1^9RNl)
zEDj)@dAW>p`^ZBldJPv^A@y0jXyHDLb|Fk;xUEp06J-7XwIsrM>5m~IhOANdVNDOm
z=@h76go`)g3-i{KhcLd$H3g(+b52#X6`UF1TEv<1)5b*xHDZANr~>DoI;p=u_9qBW
zOrpnbB}%_P<vGZP`bFIKEvxiu!vDva*=~I^k-)?UR+R8BeU#eS&^JnM4N}iurMd!a
z{E%qoLh0=G9}M&+p?J3<F9*E5ZP4=W1dVu?cnd84#kX=EQp0A|^L7gStje0K^^#5A
z_Ov*ul%Pg~-_vQD%ulN|!cIOwMpf4Vrmd-t#%+(1t-*%`dltY3tlVLqghz>g@5vWO
zgi`5>=rE6COkb9iBXPho6akk3$t0<4;wPm$J&k^-^`9QT<^_;GS^CGYnX2SRAq0rO
zYOwD^7$N@kUS1I=8%=ulu+-~s9meS>fbYpKsZg6w63yef23C}POVO=>$xf<$$F4}s
z6|2191H7eg0LVl3ME0Wv&rCeWTf8wKzuM@XcM;R<nu3l`(OO3LzHdL8uU4&^f(q;0
z68o6X<JCv5_Y{p<Gtmsig6S^@F8{|>V`6dgHCh|&xM4zf;5OEUS3tA_P0GMO;i{=;
zPx)GrVyw2n$>7z;!m!l`$r3$gkh1R>s*iw3Pp#zie=$t0rB$R431K3_%FW$qsYY<9
z3l!FUF-ROYM4s^`edU0wf7E(}epX8+SgFEz>je4%9WU54U+kVZtc9ekXEDW^-ZLok
zc$bw)xz#pX5;IWvBD#$_MgW{F7s(Z0q`~_{C`Hfw##KBgwEH)t<}fC-!XR*DQ(@{m
zd00hcUW%BWq8qOKhjXms0Er<(4=Q8vc-8*|{+24)|LzxVZJkAgQ1BAhkbr?gm4efI
zATuhJB4|WHob@?|n%_QreOrLtrC~<zv@r%inYil=UO>?xX&qq<x8q`qb{xlk5^>)x
z3fcrZf%os1$pg-IFmt{41i_yV`DB~c<d)iG2AomQ8^|CuL9Y%sP@D)*quO{_UJ@`(
z+*FJ-ntB_FYhhU8w4|3T=BxlHAW-_b8zQDJ2_QAQs^l(7{YTaa{_LSUkvOBmsqY6E
zajs}lo=uy?%PS<|!ORxi0S3JlCjaw7PwzP!T3A_{L36n8zU{iU(po&v!g2ho*+r%(
z0NPY3cKWO>X?y^rQI9l>{e-?I1exo|U}fRo;y^bR5(t0h4OYEsoR^`uQ4leA--sBU
z`{K1zfYPVW-d>4&o#Cd*zP2Wcy!hxGx2P>z??JfTn3)K4BTdH%g!kMlcPS%7k^ZYj
z62#+AZ}nWA+I4sY6`Dkji7KjNUpHf*1RrCzJsyi4W$>P5nQ^#=VwV%a%U+t^f2!0!
z00N{8xMBr>TDa=H<xx_fNj=O*Mfh2%F-I_lbI&MpA8q&{m!j=1TfB~s((5tG4J2&#
zOmaH-p?>a?viqCMt@0L|=_}da-Dy`62TrCJ{!V^^pbk*SW4f&u(WbHt8nP;aLtZwO
zm{B2^-q{ubWEZy;?5cVoLePTcUx8`COW|b~;%Z!E5M}SjJkiGj4?tlofkUMV&)<cP
z&;8vIm%i|BjI+_`_)M^*c{E_6xx|B^+RZrc^pqk13I;)MX~dHLd9wGhGbt59Bq{iS
zSnx|jF$jk-*=JxL0pr;>eh;vVzM-6K<b(<}_W4&3^F=mx9rtJVp`r9(5Fm+ss6&hp
zW|~YhOyq0LH|0GCeqKx_XFq1Ef7379b<B`2=L6bpmg{Y86mO`pkBy#sE*a3v@##P%
z0=od`>QPHgX(xjZ_zgs?IM9y=@(jT-q-5vJc8djFwCv&FaU&o40C$1rQAfN5U#Q#W
z!&U{a3XU-wY6VB$&!kvjn?ybUWPLdQDMf3!w=u#s<Y>%a5+4%Yu%RW2@HDkqd5ejB
z6e{;eWO@Pr>6*nx>Q>I4GWi4Tc*&+ZWUzZqgw@SUVY!i@w*m_arv}KnW)mG@C<d7E
zKj#vjN5uo512!SDI5m~Is>utF3jpv8gC_}L%3+z&wlg?`#Sg6tLJP4HowK6ovgYF=
z=BGWu3YquVUFWwHiG*N_am$;Z-pZ5a0rhUw-Z|hHeeg2zQtpNnodKtYKDS0xgnX0G
zB|-6e-1F4jsCnoJNDf!_@9A9@*G17RZfF4*w__y@ByGV^>?r+zy#SDt=cGn4BagAa
z02E#RF`@NW#29Z3x})kZQu?U_4Iq~_BkZ(dCSd>t;Jje`-Csza+&;w}V^aX@hx@Xi
zp8=H!d$~9q(L|WfCx64VU-|VH0b`#TevM!=f%3W>eXCCbBu6-ZA0W07X?gI8(eI8$
zL2dIxM%2Mu+*<V;>g6h<59%NPrA5e4TE-k}nAA;)pAX}Z`-^B>)nsy&W@{cd(_B$l
zg2#&T)=OXEy0I(auuXZmfOu5)ZLJ*itb3t<NxDg2QP5n$W~{$x{k|*ZVseG(m<?so
zHn(BZ?%aqXhxn1sZC@O0Ff`F2t!yl-^1<_K*Rmn2Nn{yjnOMOzf4g7T9IZVDP3t>H
z4b&bnbNoOcr#{g=8Rz}?X0VP(9>g;57|5&)cFt?@$^@?re26uH)k<!ms(sztV(5E%
z;z>q*$XT?0cai?;Mf5|W|Fq-jNFiDv0lw|rzgF{D!7QdZApNd^`bAyNqWRxVevlqS
zL-Ab^(Q1wgE+(=E9s^KK?1ri_7%K=Px^2Bc|H}|rf^~ZZxgS)rn%z6OKFl91N?aB}
zICf~wRS@{UUI;@4bvo5^<U&s+vjYSIw^IrL?kYOh8W0W8U_PNmrX9f5xfCA2LIM;h
zuf!v@$#cf;OJ^Iwfv9&Y06HEmu=)FK%;n)<;0#5u?T}9@H<E=Js^Ez2WsWGDDq~Ph
zvC)tuFrAzVYls;Wp&43ziga_&`?bx6W3lJD{S&Jb(<Li`h*(gFv3edchve|-k-+|1
zYU>Cr00l5z`ihSJ`JMcH;r@w@Ix_J@DKR{#>M2+x-s?X`_xMp_Dh{B8Qumh(u5Qg}
z^dr%h`TaYE$cdbVPFLZ|8|*TB%nTP!1Gzub3-f^C5I5GLu)@%EZ_Qkt0R!tXA(mzh
z@8#z~{}Nf)zl^q^G|>-TJ|#P|1U=dERirZrx$_<-g?8r`9214!R0v$H2|ZX^pYg!)
zV1>(BsbnjSAJDThloE?z`;&Nph6F^6!Gr+6_vgwjIO9HqUJM?&aA~A1?#*DpvX5+>
zwgOg;U>%k`o|n$LE|r&>9ozFl2UfH<t@IUOye)GylF?Bt0fTy6^wWyRpa76?@HBBp
zL7%#ty<gy_RRWH=_s4^jt1V>|TPk&*6G-Q4+qmn5CxRV*7Ko44t!2lu=4wTXOcy2^
z6C8>A4?Wdz<LGX{@UHLjQ8Tb%hw{ROv#i{+2)LectrKTG_10LfKZnpvm>^?_+>Z|Q
z-Qt(iB*t9{)Vt(a_*(ipub&+Jjzu4w0dR^TmIQ90u7%D&2K%dnFoA>>O1xL_bxi$<
zZ-{MEaDUMRglC8pnj|EBIT`%r<l<_lh6bkObSsGR&X}UjorZKpngTCh-QT~#*T*ph
z%thStrltMvp@O;>n|-7%lbqDHiGq{EVf^>OW~-{Kn`4S#V9!yAkqFiAsig@sLtge}
zT%eDgc;cE4M+$GolqJG&mdO!)VoxVcJ0W^?VP>&ZL_;~IOt!h7r<mVq^VeT=C{KIy
zos7p^pZ4@`J)~)}HEi}xc@%LbVerx}5enkCtbsqbjSX~T`GSA!%K}4M@%?}k!YuyF
zoxF%7f4EG@ey|ShL-^h+AKnfoJ5Tmv64YP;5WRPX)H~Q#?w+958)Fw$##;=xgsYX*
zetfNbhgTjU3zB92y{uK~p#g1I6i?wL5G6P>yCu#2JUFT;4+LO@_P~Yh%M&K&fg9Dy
z#Yo5Tf{%X+CmgS>1LK}K@!8wI2zMsEtQ_TmEfN7%PICKLooTHJcHjy*s<`)LNP89u
z$M*v=a)T<x*A>sx_J%$b86Ld?HB0?UTg!|QCG9EK@`;`V%Y%<2xZMbY9k<?zv!{+~
z`m`M2I<+;aSGH1IiXoCW*L1aeP*VIpHwm|F$iEiGhYGj9kA+)|bXL}5c5&Rm0S~`k
zFcyI)uoF+X_r03p`Um!p-X3VgP2%rBE_QGGKR;RdMc7*bC<CAqrB55Q0M7rSwF>R;
z_~zuv=!Y?mxz=$U$aoC}6a(t*QR?;*%S)xN*!~P9*e?>V6+b1*GQR=Vf+o&4G$?<v
z!d<bh#21o0-H?e{eG!1!+!h*_0Rq4!ZL5lNnX8Aj`02Xr<6gnyfl4-YkD$z!0~GHJ
z*O;c~CKrmX{@Rv*C6qSFLc<FUl}D<Vmj;l-vFi-w;GI_Lt7jD{$45<n*Bw1#g!3Ps
z6Wo&e{_Mc^Mc`(EgVm8h8xuF(a7)f1zs@6U9v}zIgra-w-XcBxsX7lJT()3mi1g3l
z*!B1La!0`%YnsrQ+dQ-l5WuSU0<A!D#n4J>Nf8qOZu#nOX6D8-Bw=XwjX>+xYEm)c
z7YU%ccy?~U)v)sMGW8O^MKwOw!SFw`d>=QD{*IDxUi%x2@sTrtRj#319L0^I&RZib
z`e>=x5r*0)Az_kpqc2R-pQ;2g=PP-cY&=E&U}+AouJ^5xAv_L2M!=7tGLD5x5N4>1
z(WjzW$u(?1Vj;oY3iiE7lK}mg2d%mqE3u?PllOSv$gu<Q@T;>z+Pg_Wbr0jiPWcc_
z#u;}T82Z;ZsSdBDH(PgdP&5AJB#!*9%YoBO!A0(MU65|LK44kP?q?{Yv{Du)il;7+
zW%0;2U@eF_F2S?<Yw@lRnd>l<&GU^IP1d+HqI!TLrd%OFSf`6Rdgv8Rx303FNqGH0
zYS(kL6u$l^58hj`{5GZgjT?<Wz=4%pfnsdz#C6p(KVhHm;DuLD=c*QdM@}E_GLKRj
z=)%-AuFE$514c}wwf-ZNLCgs+ZTRD^(RcCf6FO13#1=XmE4n20RQ_fKtPk*U4+K#x
zLu(b2>&j%+P`au7FcNT*n*G?kaG4c~p6!}f8loSd1c_LuHQyJ`@7>~A7MjF5zkAZ^
z+agW~^D@)3vtN9OoWIrQ+;=_?{B_k$ec9@c<-ps^KYSI2<$F&(|Dwf;Pk$|tEQeow
zS3ZEy-={;Zn;P#0=T1cc=Yl`i&@h}ATF{!?l=~avWFVlzPSFO%Z43)8Wf2l(eLR%S
z%b}p7^>ZXKvqSao)dNQ~LC;Fj_*g?a!1*Nctg<k!%TV?Eyl1<&cWSxMJqb`EA}GX8
zWt6g+@vnIW;na4KOp*q8Um{}Ba>4V}_fD+oBo1~g*U@?S+hpq(QtgIn!Pt%HDC8j`
zAh3l@(j3DUzAC<0_Ru+qi*6uHrzMm@Ye0P~yn4iuiav~6ZlwMc4nzww+cOAY2J9qZ
zu_3u9fz*xW?<O0kub$E@WvE2}nMQho+9%ksRz+gVB$xQM@JRTQq8;sT9i-d%i3Jg6
zQ-lBGTabreJ$_+DO1r4f(xaKHm8_MM%?MG@T#;^Y?;f@rZiert@U4Jac0(il86|%K
zl$~3uVxit~oVfOT?golWO`Abdlt5OXkbhE;LR=?%K8~L?6IhbjpW7pda+p#m+E5eo
zCQp)gHQ$!pNVf1sbZN~Qq5`)nMD?Xo$fwkTesK^3yQlYVxE>9Hm^sX<+v#=fG9hkX
z9{x-EN3leoV~uiP$}%3$k>BRSU^vP8$-lrzJ&$T{|Jz{W*E|+QP53R$OZWK1^-qW9
zosVrU7rxM6nyn9SYBE+?k)Btc0=uy;)Q{~))J+!AZ{t%i#h17-PBWFmZQ03ngd56s
z6D@*tBay<i@6TU_kzOjf<e`yjzruh^F?xLwW9q&iRyMFD1{87}9|hz7qHe3kPp59N
zA}bAVkb09YjfVc+L;IRZqIb1;MX=J)LWD58hZ<hd7N4$LglW?k@r41mCASk{fD3LE
zUn>2*7~l{bQ0}Ljgy%+p3Q663>{?N_kG;J4iY6M^_U;#9M}EH}P!%?itWL(Wue9#n
z7;@_)+ViM|n0lrt*nV9a4P5euCqX>EZeqsVJ)Uk4iauPzTc}3$BN)>lNl*(gnQ5%x
zdk8jwuvb$<avBrob_sK5q%3{(CEt0$tYW-@TtIksI`^&!r_ad%FGLr9Ni4BU0jUd#
z0?Rn%L}%xlQvY-#Y!ov~;fHv+kHoBt*c&xJK}O214#~(GMcSbf1Y+F1nB<#CL?0YQ
zYG(NZ6X=^`fA!Uv8bjkrWm|B*wthBDkWX;SF%&<Y_=Mbzgl>pL=7}AA(cts4hgyU!
zO|&gd9Pi^VBkRfKaRu1ZvD#<fA`e>Vmlnd~Yxi0{Zw0gRru>M0y=84;c0wfaU`emI
zBX=?j?j8{tbV-ZNp=>6G2?B7l)V<<NyXMiOS`%9<AJjJo8|JX70L+ht1f>X<{z{FH
z#xLR=6B##4)~qYw&hm8;Ew$(>FcQ8urOG9;(6ZZ57NNtEQbQ(wKb1+Q66LvPsb_7w
zj*(y0<PTcXf0uacQRotKYF%fjOb4ZO_84}T6_5Qi;|8!&6Dc8L<&JDbHwO^;4z3xc
zR$nrinp24_;dw|5xgjmtq_H*t1>jZ^e|fq&?>Pc5R_^Eio?8=?TmJ4JHiBGf#6Doy
zTrAa}D_wqJ!RN*JWJ>uc;mvBiIsD_0rfhFD6fu$kshW&yn6(k#i4TL4CkIVoO^#|l
z^~VhSQaH$Y6lAd2QPBJ=QKr22OWq3}hZz?76izEB1tJLhzG!`e^EO2a)Kmvme=t7(
zwu|is48Zm^dWyd^PZ*5%7a!SkluMES6Wv&Rxa-T7p~#ODs)Kxl++GR-UuuF;Pgc`P
zzOwjXt)TH5QdQ+NAu$a;w~;!Jb?*lOcOjwJmyff=d^MnTKx>ZHy`O?3J=2JNUaFx6
z(iXP<wS2Cz<o8iQKnLapUbolCM4Xw8e2^DsOMSc*tp^Mq!6Ji*3J@f4r{l&yrJ(Z}
z=CHA+QTG(Y1ejB7)&aex&SaL!&mp`)tR{4rUKK|q9~&oQQ*m5`JJ2kFL;Ita*dz1Y
zLZYM<#SoT1iWB(oBbFqfzKOaG%nReTwqL|*S2?h@8{1}Esrx<(efKn((`&LVO&u9g
z(bYN0wZP?G<Cy^WLE2Z~SR~)%I<BdbHV;?h_)83@lTia1S(*dou!IZniSchG==!_E
z`(8r=jg4%w;I0AZikW)FwsQEHB*z~t=x8#CT@<aolgHsL>6A>r`>fqLZD=#q$x*^P
znjsG^nc55_<3}oIV8rdOoTZMK3)8mKgg7CFrX}frZMGzP{3})HcX1l4+MwpJ6gdSZ
zH_oAEgBY|{kqkAc;DTW{NrN?cUmOxlPxKP&hCvFcZ1ZFY3@;n;$Hbgf|74Y-$^jMu
z%cw*7%w%h^EyHl7u$Byei~q#W_NixVdC=*G?V=*GBLFlbkj3ZcWuy>VB6h#`k=o%T
znlPs~=#DM8wB5Y|VH#?tQF_P&eA%vnukeKFM+sXOVnaDwf2A@``VeDivm%)T5-Yfx
z&P^R>>(RbFzx!~J(I1?dxQW(^UTK;o8DJ*J>zTNre|cKINiqA-bFl4Qd%v?7oRQdT
zR!nx!wI}YoXp?&n%-KOsdp?dj<%Jn0+)C>$Id~2?pNi}4{Ul}s2#e*vuLB(ndn~<1
zmT!GW2Vi!dTT_MjM5p9avMqcHF%$jTp4_J<R&k=R)XVX`B{UhWR857UGrl&2U&+g%
zvREB7AIW)14(gv5qfL<=h`dHbzy$x~hFgsrXOyk#15AYaQ1|;hT4wg#Ntq!ViJ)4-
zUrN=Tw)91cLvJ~y=h`zJ!rTbcx#h9{`ijPRzi}DNByhuPSs{375v8q+n}xpwI0m5r
zE%`=C^$YbJ@X|lg`0~pjfO=)iM~BVbyjDKb{vy3S$|4yn4#cV23~B}Nl$h~?7v*t2
z`CKm-Vvc<a`HRzY0-5wTBAn0DWjvDrH3LG5+H)Wm1Qsdg5jGNdw^(=*9<?eu_zEOs
zKjZkxh3WGUSOFXaG0y>hcuSK#ziGxF;KB%l>0G7LU5+sQR!HW|mVNQ~I4`@cb4e3V
z;fTNaG6IlMwdWF#B@MM$x`1z=dMeUqSKhQb{}PlSJg{miI!d9y9Rte~-ck*1vwVtt
zMbrPUq^+73e2UiRY@J)Q&(^ZiX|eT40$FR^acqY(55l}63*rZ4F$hlmvgnM*=(%+s
z@h#JklvQzM3g`JodNei1rT-{}ur_2&ZL}`DW&{?z=hc)Jl-!A>eS)Q@B8JXU9_%Ds
zO6y(@Y4vait*BW$B@b8pky{~%j}POJL%l-gl;jgy5(XQj-mp7J?<yg?5n9Tbw9XE9
zzgcW)%g7#ouJC5tqS(e^dNS$}|JqKx^+IrdRc=CXL{<B$fDj1qWRpPb14AvhQqrmB
z@RzWb2z^Q=ZNy*q*7DU6op_oX+}zn^gB^>GvBLmfMVvuS|BE&$Ds+bPpIO$h01h0(
z+}Q1q!3W|s?|Xg2+vzW<ja~mLW`n6=Cnfq2hK$ep#ix?MW*nloHCmpEA}`rdQpvBT
zLKvq8mYb#xNP?-x-;Q%^2}ku2EOBM!P|*%xrCh-^v}XPsg1Yj}weoUx8a(Ae!6X(L
z8LgrCbZmGhgjF}z8%FnWEV}R;Ew=@{D%&vqdUQ{V(~pbYx5jo{xqL0um!&^&##`f$
zu$qD!*!%N-Z(-SmxUfGcVY<W`zSDQZzA)MzNa?$(aIBf+PK|I-KW!zFB=9XGe&yML
z8B)16)Jih)BFlzu*`IeirZgJFbWK`$tc-cwZnN#KE`#_r@VmKK6Dn>eu}eM)?E??6
zf2ylEYHSoz(*B9>XVHMfKr&!Lj=jBmVZJu!-5bd?*HVcKVq#T8mGB^tOdB*zS74Kq
z2=1i$k%|ekD~LbuM?22L7#vgn$KLu=a_}^Go3qaP$YUQShEHElrDCs>thhb&E2_Ss
zsQ|GY;y^*|SYG8bm)`kHi3sz8-T*xnnO59w>@Jc6h%;)}5>ESsDVVvw3bW(!*(0GQ
z7u^>lI5srn-@_OzUaWH7(VCDPs<6Fgy54Ljd6Bmm#F+-+(YiB(oIh?^sb=63UPT;^
zR~E^bg8kA1j;&_NA#|%gqG=9dbn#(vGf?#yoWpy{bfI=EU#yR*_zPj4&SX78Z0C2R
zGeuJ|RvAGE?B~{3XFn2VhB!5A0RE&F*@v(uP`8d9DVhk$IlB!=c@i%cqciMi+=VjO
z9{7l#m0)^^X+?RAVDcFg<9J1CHg^9USIhxjP3r>QD3)8}DAIbGhWPrX@nUX;x+#LS
zLUttdl&&53SKRrJO6A2DPlh<BYJ3VfdtOzTe)KDXcCa)<b)LI8RbU;Jvgz<)ONC#v
z|NOg}!Nq43b8g-df4YOK;#@jc4KD*>fvXno%>gXgzp&t3K+KCTe0A=fVbO;XBY8PB
z8ptjIadRIZA!Cs5{Jq4OS)@clX<B9zs^3_|4Lz9o@;i5R0@(Bzb30L>xCvFUUMEQr
zIZ}`wvS|5wyyK~pWvnT<<b6$2aRP#g&+0_e<vd@Z#=DcYMF?buZz2b*5URxujh9|g
zpL#Q4I`e2bUV2Pl$#tI#mE-K5WRzBZmm%I%K4_GhQ{@zl*>J-LTRw1Y=jAeRD<QhH
zCv)8iD<K&f3DP7iCI+q`HkxMRd6iv<;wk)y!TN;>2dvW!24%&%;gc+9%BG07-(reI
zHnRKo!p;oV&%5bTp1{3C@#G8shI*2hQvZq$CHYoF`HqT|3tyP%@(wXkS3dbSBOUG@
z`K~M8G;YyJoWFt8H(y}b13}7de)-ld<U%{H2{!y;n@k~C9l}^4U@Nj<Aw4IiYqA(d
z08k~*0bojaWkt07gXJwJo<2uepHIS`>|_#$YG^=m;GO*0Uqj^<00NrKD3NmVI&QVH
zB4`ro4aq8fm4xBu;y0g^Hv-9CL>^x6+uI@5l5+a|_B9$h*a949zK%3vh4dZWtXq$(
zh#8Z*@v~ovM23FT5DtrR0CH!DiuuIM)v=Ie;r$aZCWK}N%j4y$*de{Kd^_&QE+1aY
z@#0?L0OaHOKM^QNw2{6b&m-J<sWPGDs~7Sv;^tkpeUcjcOt>uIAZ|6D@3p4fOHV(c
z?N?CD<l39H4lWkHE&s_!XNNa<3r#)|Qj#Ol^XQ%8_xoXAS)LheF5^2`Nl-?F`pE6L
zxAQ$LelfJ}?ASI*?14_-_(kiyt|w>{ikE)>q~xy-5etFARUi4#9(Xz26#H%LhC3$l
z5l6w9`upq)$*;(3A@NsP7KXFuD<HsB$p9xn{U}-ev^YR$*bJR)XveaCMFnE!^qk^@
z3&1~<kB{7N4eKrIAqSSuc6lDg73n4ku(lI^vR!@kmcEyqp-uq)q{5qvcTx1cc<`8t
zo|a?LEDM2Gkj2Yu$nWKMSmVM>iXs7tV5(H!C`61jdi3k%gJ-kkfC%+fTBzahw{Zyo
zs=Z_)JfADaaC6(#TlEuJEAR!yuatozfit)}kP(`9gXs5c<*Cq)_bC*P_%kppZ&^V&
zl)Y;5b^ZNQmC|@UZEv`E6P>ry_JD(b=3Sjz)zJNOVWT8%SYEVMJHdJm9eaD%hiw|g
zw1!(=yEH;>yj8>=y3iU)>@is``3?W`0KREuQeg<TAEjrqNup2Tw>@K{oA6*_*n3g|
z^~T4B3x;MDKMc|%-)>=r=0+71Rqc0U1B$hX^Z@azkI&#)6=kp9ya%g8f_mgKdZnCz
z1&_^EH%vVQ;<ODrGF1804PsFGdD4+@@LUud6ZzSE!OO;kFvDHn$Z%ls_3|O>*WOS@
z?#~}}dmE87jC6+G;i`8iqk3E-2DSbx5)*l#;A5>WAVdVQ!xD3U;1`fnaKxZeH_NyD
z^S!B4X)2dazc|v?!@N;;d59j9Zj*01c~e;}lq$TfNIfWJ8g4?ciaWHT7bqhWzA&^o
z^}ZcX()X{VsDH}DftxBjiMYD=Bdc^wE>>HY=rg($-*-7r-z%)qIKFT<5?m3iFOV%d
zlO(;18+##n97rFhveGSXlh{1260!RStLkV=y{^|vtZCFU<Q10|<p&jbH)NNDaTCFP
zRIj@C8=23yf%gX2MQI&x@{xDA9dowq5pN6LwOe0N_inngZ||um6AG&ZcR;G_|75EB
zPp117Yv`xbyUOk=utS%o3!{9$2Pl_Rq5;cr^AG%8`Z`DOIm4`MI{w8pn(VtSQ2O4`
z&nc#ltMh<b^V2pD?3dHjj?^B{LZMWZD%~Q~$rhc4g!#ZwM#g|>p{hWT0Zfz(y+5v&
z7a8Gf;{|>h@N8oD2i9J96O8yyW!@=FM=sm)D`GK`^W?j~3}&f=x+p{WY%Hcc5RckN
zb``9)U^|n_XRf!<E#;{(TB)(5A>Nz?l(^I3a-SW6P)1kO-=<?tALQ#_pUtx@ECo>8
zQ@*P$XX~w&!}uO%lecIdY!^lHq6t=SP<Y{SMO{%UTM)tXSq>plntB={99}jlGd^;a
zt&3L)iQ^l}^Q?#LhP`*MJeo@+4?sxQqCA13gs}ZPelDXJE9yzLFoJE=O?l1xcP7P8
z)ovIWKGeE=@7|Z``OLJ*X5xzG1CZ*zwgK5tP`jwI+vz<D!FOQhran=28!+Z5(u;{4
z_;N<5AgtOee;DxWm3IyBz~?$d93b#o4%=06LuRD`9}lnU49#7y=W(L-+oOem2(*_E
zzjOqTcDT}AUtKNQ*myfXv5oMv5T6&jF?9Jrbd^xIWqgKJIN*%a1qx0XJNKcuq(i;(
za;vf=9eIl`eK_7C4esPGwU^)~I(;cBrp!Z`g<-C7Cnq_nbt%(?!HKFtQwieJ1h1xN
zg{R-2w@7^$D?JF8&awjj%(#&IXyy!SFEx`dB)F41P_pjL`O1klroaf6#1YL3Wk@<N
z7;T!Ek2+_0`CPhE+!M~b3WQ=R2A}lRINQ1`iE0^EvK}bEgetI1QK9jcHV~eJ#dp%-
z_aEe(ePJe66qH#Xa6oz-CtsbYQ4ICs3k7j~h_7Q_#vH!Xs25bbpZ=F6{j<sUNgd>!
zka9b`S)U;83$@y&V)Xz$kHZI6i}jz<=yqtdW^Ju7T#eIfuCz|9ugFY`GtgXo{sQ#2
zJ#tWdd2(^W*^Z?KEFbQ;C%?Vw#iFLp@G@M2M66}s*`NY@AGe-%oaKOb-2b?&iW4BF
zO8)v;MsdOBMaN>y)UK^u8;fPMIBz%IU1I*z?L5H42DBF;w?JBzS;wRjQcLs$&udCq
z%2<mCQ|qfea90*h33bfKZd6WBIUWVg`=`ohZj&5etd+Ud5i967{<H7Ym@McV@aga}
zbJoB&aS|Pir2G8=B|<mWjF@b*%=Ckfw9`&K97E1)I*%J|@tUSsm9(W>-c|cy3oV>$
z%Rrz%-O`^`Onb6c_KFY4nrPg;tZ<k}3BNlODO351G+~uxzd{sY=1`lstuf2k!N+}<
z&hhkroOxQT9-yyig|;%`n8!tp6T%700!%`J+B)n<sICO*_6db!_4Jq~4?%osQ_<b_
z#2bOp_J`|Keyq&2RZfy7r9JY<cew0)l;nAc8|gW|E%yrC@#MYgctzFlvftVb6@<=Z
zVMv|Z?sNLf(6^5kdwcTk;J@KNNBr+n3jYbVF9Xu8QJr=$yn2~t7=j7|+&9FoX{%xF
zhnYelFlTYZ3Fa0>f$2oYL2QLXb9g6%^nqy8D$h_O@_>H_RR!{_c(zH{^f#nq<x}Aa
zwe?MFUjj>i5k561Y%cn#bY~@v>J4K@y!i3A*Lj#W%Kzla(?IAiQ7N($WQdR0EY6b?
z@OAcud%2I!co;*@w$4_^7f=}v5o1b>H%cpOok&ycL>GXY;#`-#%+%lL1a=zh4W8xy
z>jfxDi4-j!Cf^haJ;s7RYI;1C@nng7f92iF|D~Q30U#D)zL>1UPqwqLsm^!qZ`4?S
z>n^WIHxcsOhTv;M31Jhntj<zioPiG|<E3$;r)s#5Ce}Oim8~dB@c9<a;~k$~+1IIv
zF_Hw@OUH^ol+MhgWlyM>$~w&3^iN&)?s`d^5@JReOnPT=0ZYuv2-kJ+mbJ!38G`h-
zqTUv~Bk~)&bn3cP7rFbRw$8;(a^x@6Frn7{Vc&Up<EeX2v}0CtiquQF@Gj1qj?d(G
ziSLT+&$>6yw%iNnZ+6`imOUe!2Q)bA+gs+(?AI&1-fegIXbjN!Zld;5NBv9Agqh}j
zrAV0HItWSv<-(=0eu8D5raoc)$Il*#7kQc4mb-A=(^<wkr1m_)dH4=6B#X3b8a9=V
zhJJMxFW?LHR<2e3ni3*Cwb2cI(=?{Uei=R=8w5RkiLG9%Nx+@v&pq!n&#V*yq&JKS
z#Bbak<i^^>zMpt}v#UZK=o;{Z&g@$XW_;scAjYf7TJ<N?v`Pl*1dZX*Z_s1dVU+~y
zPE<{Lp!l1*=RX<6^R~sR(4+Z@_hj*cS^1xH5))BG-Qda-9!#w!O%26M&Ir6qypW(~
zK7M%GT+g^K(^?KneTX?pr#5HWS)Hosqu|(we#sKWSw9=SWTN)4LP50xAMEe?^<}Pq
zyDGv^U0{EzECEddqudiN>DJee+qF!iu*R26lk18P4gK!*GQT+LQvb0Oc)9H&o3@!G
zQ&3KSpW7WG7IhLv;L-ey^Ct2r_bzR!zqT8dr*nQCZ@YttZ<AVM%mnuqk5_#wdgUQQ
zu=~O<Tsq(!loUwhW?C$)bP}PFNS0=jdoiFS?btc+{*}b>!{%Hf+)2ALC_%un2mSA7
z2CVPo5v)5y{zSj_L<?<cW(VFP&l#8-`-5_Yq8G83_FO(%$4;<*S2_;#0s7G)@Q)t^
zhRj<6$?t_R33yrEWKNX{w2L!~{MIcqDAZ*F9qiG$RqRGN^mNti{D+9F^Mh-tJR(Bq
zTP4Q$oc97Qn^=~rhwl&5q<!vXsC{dB*$t&14LO|p1w9&K5;eGZ{fg}?jruYM*`&M(
z={&U3g*xb4dUN!RS@a;E+lJ*Z-5>9Ks&;Au;DeY^4mbVbXc{@w<gHuW7J-@S4dv84
zG`#`7Ic!hT_H8Ze55p(rKeQB9@PzvpZ~qnnlU=ubf4ZYwhn3k&lykGhG#-ioN*-P|
ze$TD8k;OYW8jFcl`AXa9`rgzT%b#M2^;O_K8SB|r)HR>K8CppFH-1e)dwhb9p56W*
zg*9|7rtpyjx&YQ~unLiToa$?ECWsie7R{HqaWlLMoKl2yAbrb;GCo9Nx?SJuvCjB6
zXWd^nz6e>ecsGe1D<mylkY_nMj|dR1^T{8o*5LfHt`EnVvVlvBYOmNEDb3~q4OEt*
z@6{98+gUPHcrve{%~YNil@*1AidpaPnI>4gBTNt%UIbO2JFq{)tIu{n#7NG33A3<q
zB0(J<5qi4-<#iiW)-gO<-yiEb&;8LCvY8c*lo$A$&!Ywq@7th=pNDs6^nE0S;`NI`
zmYq{y`Ro@ac@R<t{MR?QQ->FhHs%2%%k0q{?YR&f(S|IUeN~Z?ITJ<ed?P!|HN}`>
zuy6@e3!(2unGumcMS9Wa1_wZ;ib6YqCZpsvWs6uT<1^#rK%}4M$1fp_Ekg3RAD=-l
zAld}Y-c6Me5uJX=we6o8A4N|nWKRW@31~19+?`1j6l(VyHWeS&H~X5M1wBfp$&U2z
zXdY)(Sn*=cR&1TUX_o-a#s06S?<WNGK~^{l*YbDqxo$5u+&m@|PnsGLo*CI-_y^~R
z-dbVVu(%ZBnJf{i+0Rar-;(!#WPI_omC@iV_&yrc-KWrDI$7ThxzLef_G&{+-V$a%
zlExLI_yJxte(*e<t%AI3@P?lCV9c2x;r&<mlZ0V8oXK6r$R9UWw3y&t1YZkl+-dV;
z>D!lxo}NS+SWmp=HqvG#_#IQ;`111n-hRPgZI{#@ZDxPhy*f?^ew~&3E(d>s52Es<
zJiBrSUy;aDnp>-p?vfWpR>_GbCK5*TEa0KYW*!=7b0dH(<6A+&{PSY`xaq7%+0Ya9
zsMnEHc&((a=X8HVJ(j;}&`sUT{NWxb@A;iEE7Awv6N$y0FE5<9AFhB&nlKmG2`XD8
zPGvaA0vNd5>SdW7gadV7B%Tuwy-I8#Au4m%*aYOUA25@uH1g}KfkA6xU$8zqoHpG4
zvRqS=ZbZ*Q=h3MwgT)D<&z=AoRBz@#HZfalW0F<b6Dm2{5w&?;DP0%&O~ud+d1JcL
z_diX@z{^RdH8=73lRs91-V>-pwpUh9%5OjJ*0~ln`;vvE(b2qq>)>%IaG4`EI&V-D
z@JCP@S1PjBGpTrYF?2PK;BNW;ev5-wmB>XEY}2Lshv76Zd3VW+^s`Sexvs%4YzDrg
z;FAB1MZN?9o~q<U?r*7i;Q21$2!w}n%q9DcIag5pPHMY02e-wZjs}+%eHe3bKF_S{
zKScDe#QGg1IFo0366QAntZjlf_d`Y7J$QC(PzBX_m&MIsODuhMsh7btmVkX#YFBJ7
zom8!5@9To1!st@+C1l@(oMB&4!4&@1g_~Qvp+x$G79+#koZ<V*#$OVR(4uInK-`t_
z{T$3sklj)12@^8$scxv;V6TNJ?BK-=7scw0Y^3(;o{GW+c+2TP;#!whj*95*n0mrA
z;VNy?eqX}J!WDXjG@%xvz%_T(?xd69PN4$y9kXhfHQBGk;+0;N<=w^XpfF&ThDCu_
zjci>?PaDRI%NOK{lzTe|%JpXD(oQQMB1DNV1nkV`jKqEOMszf&>U&9!<H;ts_S9E`
zx+e#0k@Y|3o)7;Q1+*|-3SHAPtBF`O7k{XAG~T_SIBk}DaLa*;5RPw5WdsO`5^PbN
z1{Qp7oXHTMAIa<$r8k4WR-`J-zVE@F7DvWNZ>CnlW3(o?EeLM)+pglau7b@nv9~VT
z$+))zX?Cajl#FG*{a75aNarBhVsL!2vot(|CrtUfkGe_505>Q<B+E;UZffOkr85g_
z&Y3j{e)q7Z!+JUyM*AY>(Z!E`7<@+ibVBo0d{I0ha>=a1M4vf=MDBvsIbbVdVMJ2O
z9nc-7fuAj<dtxG<SUc@5IiX^HAxIX+@^8~WWaGLvv`)pW2Uu^=AJG>a@YfUBx9+!p
zDpf?|Lk8{R6Yu4%L#~4J6Sc(_1mkv(ZB~`bO~~S7W?=W9qMQ%o)f_nKDaz>YT=&$%
z=x-n@Y9G66Uf;L_A<jlna)=QN!4$loZ58@Ke&!0(7A;9s!emK`%kYK#-MK^&`LJ4r
zy+6m!3q1{qz=uH{lDgq+(N@lPdlyuNrTKr(v-Yf~wI@BtPyL#v<SL_<wbHiUxRr$&
zN$T9nn(p6<t}IGNvEE0$BlV~8=SV*~$-O5)93DULvgV4ReJuavSZ0(tKK##_)vobs
z0xYS#-UKJYS$sLr<{2g*37zUUfCYhe>S=g-;)Qmo>g{)e-%VgZh&F*xyi`Uf`3b|e
z`S*<p@_?`24;KTS8p1qqaB*TP{ctm<{F4)-`8gy`5ijTZ|NiNHx=UZ^;*oh>q(m3P
zIL*?sK?55fL~X#umcBR%TZW=~Gm^Hjf67>vq^DKo17=vBL6eGzrY9d6esjd~$GK!j
zvHFBt6Ti1Xzmng%e8bMa0y-*O!#m<GV49f^)I@K{w0`1Poy)TyUdoiCjEP%3k67N<
zLu#H$J}M=$yv42NxQ4hxWk3yhceT;fI_8MutH??9Q`te7k@BF?_3LIk5zW=dOC%H@
zi@a=}aywTxGpZbPKU1=vlpmmK{PVWYE_kk#NS9a=KabgARPZ|gMvL0g8M483<Pev_
zT}6j+VU{x$cLG$xMCllsGGo67a>Kn)C*C|Q-NGcFO^NzSH*Y^jUABGeU!2=xWIlb9
zKsvc`Ub=O=Yb@|uBv8e?fazHkZ?F`jYjDmH%c*n_Oa0ZkJ4o)HL$9zw@K<h^{l{OY
zVSkb`<Ob*QXSaSxF5M*E)`>ln6b_a-%nu=FwvyN8?O8!--3R?PS+Y!K3TL;;^jCxz
zWbEOdZ+<haV;|Vi8w^~uJ1(ex5SS;Ws%V$`WyLdGzbIpA!}xzw2;9yf1LuLZ_(?W?
zsc9{T)A0TxYHkI4*$T6jp38*TA3}g&-(Xs-I3P9b(2x;n7eBDD<(B(&u7oNuXP4Fu
zM3?6nEG8y|K)?+`$iWHYrxKW7W`=B@4GkBll7oB8Prm2BX|cy33ijYv;+DB>H3@m)
zyx?(-JQB{7P72jz`PEkexOkF~V&vW=j=;$=&8k_rl=#5J5Z=!1JY>X)VT6y>sEkaa
z<z(ux2Any1FibEn-xv|`Bb&yVsa(3yCbw@PBy8BaPRILVGFO~bSJ?-OwB{tH+`cjV
z|H()_p8C6t;fnj$rob*#gyGGub<Bm(H^4Z%w;99B2OWKjRKlT?5TAH5KiPL=b=NyM
z?<&?h$dcC##}_?BZgLQBFqcS`vSsKUDp!MXI$nQ^+B&<HJPnk5c$?~nzT11$Rk*|1
zGb~f~**qz<nJ!MoE?fB~xLh}<dHuwafu!3D2{^JzVIJDy30;SX7Ha_piwXwcK4=^`
zF2xJit71A03KKnYmp3aGrX|A3jh!&(cz@_~LjLADE-dL*yNC7YvoR%UfbDsATP5(u
z+PSajO^o-{m$LSO&PdaXa=bBZxtAgP7Xc@8llv8UT<H{+9Ko+5ZITOfTB19e_41q@
zyT))r-+(x2SmY%ZlN>b;vFAj-c$TW4woBrEr6lO=#@9-GBWZt}5bQ9V!;ppqXMRt)
z?_NcB2!05OSV1;t$o8mdoo`6HE&9kQgrxbi+;Y8^xxE@>xz52e9glIpxqLu*=1jI~
zyt+ei<xsde@=)|yNO`wZ2$4F7Ol4E&1ETeZ5ZmMjf_|<Yg;u@~ts4KAB+v)EkF$V$
zB8#5Kxvda1yb`)9_UNTi$|DQ?*+Y>kwo{&xm-t#fm}sfU$M1n?M>=lpuZiz(IDgDe
ztNZED|Ni_$_}qJ7I|#(`;n%yr81xIky*O|!@}mf;i$;{m<ru=m<NQsO=&$d`*l*$%
zA2v~h3LFj3{Mp(@>oExMoV>tv<VVo8DN89{8^j>2R*ApB*hVN50I*6%mT6TBnEOkc
zH|`D|#|~}TIos};s@C@8V4}~bqRmsk7u5aUe4_-fO(`ED!%l|Fr<d;Pf0WtA!G2zx
zKI<EG#ElzOx|%4kI&IsQsR#~ea5SDkX7T1KV?tO=+P4Udnbg<Ub5L}_(+<6Au)Y_o
zkno0E?M!LidoF*{fz3X}^Kg2Uy-G#mq8@QcUDk3!JHv0KRq>jCpu)W8O?&8gdfk})
z+|#D-{Lf>$1#&i>Y8W{V6y<js-Xu%eb|60&ND8smW0#T<XekV9Vc$$iX{3-5^kq;U
z<*}U%D{Q6ng~Ha|$QZdo3z4zQ8T??knw6d-=fYYVXtp4@uRG#TVOlr#rbs@oixycp
zrX{SDqj)#WMI$IaPtUQvk-P;L)86n|4%_nNvZ1<hYB!i7_m;e0KN%G<QR(mzcP;G4
zm4L%TI)kjqzUuXmTS&{w<+l$l<iAP&-Y6|Vjpv>TWVuRv>Rw<w)Z4db&Yd#K2i;1W
za)c<n+@`%GxIgb_9CN>CUiRDL-}p*`m|4-$u_pQ4`Mc?i&vBR4&tqKfqoV0xPJ~b4
zo8fn>!}tl%&)AmsRMT$EY)qu6YV-D*K}FId6G|%oMGJSHzndtQmZR%-<F9uDTG38e
zM_&aqVAfSsr;l6c69_xsryn@PNzP#I+Gu&ViuaOgee{$8kU#WyiMXD3{9bL#>5~yz
zUwh`HvO=?4p~O67h){IREtF~v;4{T&KQk-B@W!U17FyJ%j<bsaCFip4p2=Ewhx#JS
zGFoZ%9fnngRlE2$j1yF^pm#1q!%2xW+Ff=PzJG-3m5CERQwaE&S#8huE!H*Xx#x1O
zx}^`c4!t?pd@-oScNk`oturjX{9<Q^wzkfxr6@7w8uKsh<A2Ty8Ljj|%;AYeX#u4d
zPxBA62L3$hYplMDT6=VQ(IA_#NsPoLwE};Otj4XPEhn$1?id%yIt{`TEu6DepfS;q
zMTwH%$vI5B=BqBneR?>Bua5GQ!YmrRHfO#LI0Pp$mB4P)<%N$ayQZ5Wu)X7Cgmx5f
z@Shgf6-6x%V3^UjsozW5lO?8sFFB^>miiRES?WK2T7Rqg{Iwj!G^Zr=h22V&x_j`v
zqhRgn1X}siApNX9xQ%pZK+lA%9kJr`h$lr6bK}Jn(lOJPbOoQ*NBtV@SCnOF-Xn+9
zO1;Faei~K?S|}nvl?4H&6T3dq45tF`Kb5a})-)U$8(pY7&v<*wdi*2Pc`AG&k_h>P
z-hIsYRVZUK<0aurIt{E9OXoWFLUjSSm!Q1%P5^Nz5I%@vAbu8y3od&&xs;=QPrU7D
zoNr_0jIQs~IVT|%_@mGlNsn#~T^;uUV(JTXCvh@Hj(@}MKR$gd@24|}r%4TklN_a9
z$@_F+CR%8&^Cy*8nz1Q3dFqDg-T8ZXfk&IdJ>!0DSKZIizP_OU$KIQVv(<fn!w1FP
zmg+p2Is~_()L3(SYY~)|mZFAMQL~Dn1W8A26_uC?Z51`oG{$s5V~CcT$5x0jAxK0b
z=RJCVzt8V^|9bv@-s|nfbwzs4-g~X}S)aYuUi+La8vWsr_jSEKPjL0_V%W6C1raCh
zFKcfjJofOpKWaFzb>)cM?X$Z!bk}lYLf)K;bkfu(%-LKPzsx>avO?12$JuvX(tqqh
z){~@f2}CQ4U+?z!R(9y>eW0t3GYC>Se^no6APi*@9Km1>h!+`IE%WyW79OkCLA;$2
zmg1rCFoI9FiV?eCJ51i*I^q$SiBD{~V&vfzlE8R)?M_pKxPY%y2eCBajNDNA<HhBg
zss{_}hQFauR+4y(><FLK=Oz%EZeK%=8r-SQ(06AX@bY|rtb8?RpNPs~p~Rz`k!*MZ
zqG1F}L=|7{{b>h452){2zN~EY^UpptL&YA$Fl*(=OWr3pUAD{*Q%0U70Xgrq`^<kl
z{V4-~wI}waXYR!DdvC&|cqpK4kzVi~q+7qrE!yjw;-_~`yztIQ&Yzuh?%uC1X-j%`
zTBeLUTkbtq=swtdkKKhCiMn@hysfB4THv_W`3DUKCDIa?R`_Hib2)!)$CqAe?p<?l
ztM|Q$jjwatv=`gI+9v8!MM6*qT8<WbWsg*U*B&<MHWnQ_UA*?Hrbp%Hr%&0@L&KU`
zI9$-H4JmGuuiZ&yo~R$_VY|t7OV=ZY;>okmEm|bNtP6@y4~_SqE6>aQG}78PMV&n1
zH0c%ouyp-RbG8%m%HBAh!y-NBQLkHTPMdic$fRZOs9???J(ki|GRPaO3b~K({s>|0
zY@y`4+fZoCb+p^ez5k#)3O^^jR=@A`*^}1<7yZLZt}o&U>`AM8l;~?iw|GVk`mP9#
zo%j}Vt$z5@Xw6?M1Q)&)G(vPmc`m&2$NURzAC+H%yEvT_kyTpmrq1%%7w+!W2L^T}
z9z^LD-?IJZpEoLh!!c)eTjT;+E}WK8CF861KPPH<y7PjSUh+jg{wna|+?&RSZ(|Lg
zxmERyX~a1=EQvYvn^>62>H0+8n1F3Q)9bt<I`OG-=-%(xcdJh|8qSHIa^$}u(wX?c
zHoR<?^4Q_sTUnmFC6YaLPQ%n2MFeaf*^U`F_B?aChS@kf&$54Rb9eLNqZilr{5;TY
zF069GMOS9O7}?+}_7-PA!(@~PxA*)@!CRw$tw<Plhn5z^e_TaMkB`3oc+gPgX|qIM
zJhpe&b0%d@s)XoZ)NrRi*>XkyaT(0ufv7O&N;Qj_bNYNTF)8B2#93TT?X}z2^S2~L
z?$55bhWL_TCDWPn%tJpX{Vt;N1iv$v7I}r+MaKD;48D&u;%`bF?DQ1TGG2J$l>JC|
zN7@gy*ZPzdkckvII90bhEMiT3bx7$|Q~#H*rS)B3+lHUTs@#h*<Q-8`vg|H1%PqXE
z>=1*U^ISVnZ8wYC4toa&&KaMsPwnuXp)2Qgg3X+v%IKjqle*R{zE{5Px4Wg27>w^>
zZ=#7`j66z3bAN%AA+&24Tu|e3^Nqe_JPb?Zzk|N;@#$YX_;*~tx^}|xqd4u<^CrT6
z+L4=KPm;3*gK($A4#qAQ7&*-HsGd>W*mdeez;)go@qpkzXn&E1JB*c6Q#;MN*8*=e
zqI7PoQPT=Co301l2#V12TDsAwSk3Hhte<BtQtCqV<m8Y6i$9O7-f*(AD2%+gr|)dL
zkKnvVpQpzFgR((%YNs`A5m{`i7%y>tqG`fQqxlD}!}DxmXojKd+@7Vr7%APDiLlkD
zUNs@fiJ8rmV%cs=pJ7R~Z5Q@>)T5>oqO@sc8_l+`B)L!w)r&@XlisX_dyhw+5%Mf9
zA|+n`k!eco5bGR%Ryp?eZ|2=4cQ?J9i=D_|*w6v*q%*)_Z1~?f>g+UFsd2t--RbGa
zJzFw$I>cTK=ME<CePF27PNhLZFJ?ERr9bFOd-9^IZN%OJDZF0(*!f9ak*?HZiDB(2
zxml)JBAa{xUs{+X_x|Ezl6TL_bs-Rxd_K*(@DoNB4zCUJs7dJ-x6;zhM>Vpv?nhE@
zmL~qtwX$Qm>*XPs)xwjTn)*^H5d{q_gbb*v&!xsqzuf9lo#`f)mwLPpEy(OQ(Jh{~
z8}2~Z+S+#J&-{~Qa~M%DWL(+f6Lw{H{xzMy#xnbOGepfERqEm<Bdjp#!^9EGUD_qa
z9{FLli=D7IHa6TIV-eh1&k1zTzc*mjey<5LbT9ZoWD*u!>hB43!3@AybqVb)T{WL?
znt1r|sk{l+a`Jf5aBPWH6x)X!D5?n??)=gfrfzrbb;3rF5v6X-EFy^B{M0Zl&(mWf
zCvh__XyUquYZ5F!Z>S(=i`B8ZqJe^mG$>4;oEx`gi&wm+H&(o^=1;=mIuNUxW5&bl
zc9vf69qLciMsVm2)6aA@hX&Jg@Qt=<_MQ1G7ySG?CV5|6ah~RcxsRXry6)1?!wVi=
z3U~2AS>c=CABs7Y<V3`^uY0Xm4abq{n>ar?>qok~aJywgqo`ke)^FF2{8KcN#`e`T
zEo_GOl^nFY>vPhORzvM+pZC{eThgP)q6nTM?RF%2e+kuK|DdhHipYz>G$d&Q-)ML8
zs=~e5TebaHOxGj20>e(J_TG~ueZnhF|Hdm)N_N*$KDQX)oqyHbs-j#yJBaxE$f9Nv
zf+0Z+MaH^=nbz~w!JbROGJ8T$zPUd<^ZI>r#-Sh2obeKg0S6kl0fTo_iLnlu^B$Sd
zHDZ1=x2r#Ndd|C4Djz!1qSn%{%F+*JdrxW=9`P0@UxRIVKd$gfbVJwD!Y)U_!Vcbu
z-$dX;ihtVU0-5z8P9Da+#)zEw8LlAZ3p3J*jvpq!V(u~MDaq5Zx0X|>Z+s~#%Re%<
zYmK_MWXxpU^Py+$)#}>i@VBeu=8pQcd9TvS`iPjgyFsIy)F(AjSg{FD3tX_p+pYdA
zedxAU-?#!_ojDRKggY+pFlxNY>9?)8IQ9d;^zGv>7-9Pqa&MW|_ZwG^8&^776!wP}
zht#Oq=PPECIQTAAM)6(W36mLydhejUDY7KqExSIjw=(0Yok|R&Jfk<$w9>&sB}R2*
zy{gApWjd`m!K)hco-wa8^$0nG=!mJBm+J2IRdB!UA&X2-stB38NwU>T<6CVDAVtK@
z5SrYM$LtQ6{V9cK#jXYfPNpzz$S$>8*SWJZrU4z+Mu$@l%Gug4H)UA<Xx1of)T}cQ
z7A{zmP>s8ILt6@^agY00h4jNwft_z1c3sY(F0|0|cv|9WP|&IpF*W9&Fch^feWD;&
zWe#Lpi@t+xb*G!88mxy`7mY_xCHhpm9WORa%et<vW`79iz9P#({bQm4o*0nEpUIg7
zBjA}?<bj&dB|dnhhHK|QjE*Yp>&Nsj0S+b|HEj2-<Es~4!kHhct?E8Ugdgy@smNz}
zrB^@a?gMI#hACpb%5J@#X)m|lNeT2i+0lhBQe{qmor?|_>2X+{&M~F8<kU~^mQn1w
zVtQnT(YvlT5uIL<2=A}sKd=M~>2#g0fxu9kclTSG?4j!;OT(jEpJ_F_eD1FM4Ybc2
zI9A&SRqviEhdC7q&*pq$1fAildn!(wy&J4(M%*h_B+srG?s3&CJjPx+-|_YG<JN-`
zF{Fv#r{Q`%ENwAk*`j=si&9n}@B3a`?fbLthj_~ja?f)I%=2`}6BWV8`jM9#pNne3
zwMr{q(L?UJR<~GT$dN3vgG9mtt@ibFbj7EWTq>?xN7MD)=bYwlY$-B;+mzb(k@<S<
zSXgmV@DdOHR}(wn_p8I<N66n@6uaBHXUhY9sUNbEzX%TJIIF@dQml%j0~Y9G0jv3}
zq9PBmYZirDeXQ9Rd1T=zMWVu(;a-Qnm7aQ2HHO7^kdzd-yRfDZ*EfC-&Z@s|A08~X
zQ#$OyOT#q066DIGnSy)7giG(`C7sqRy$_?P?V&hN-P@z8wSX8~G71<UeX%$)J5f7k
zUL2u}j1?z_o<4RsC-7-TDXyI}YBD40np&7e3eQe8bdf@=Lx^-cUTy8UZhd^++OwMK
zo}U=DK*~T;it!!dTcio&%1F2`ejzhsv@*r2qm*c64^JA-iEqE5c0gG<O#2Yntl;{4
zBa)iD|48jf!PY%%nY6KcT-IHai(qnpQ~xu@(1H$MI1&?;N5_p@6+AdatoI@gwQiZ`
ziM_LbgM3f1V`Di?m`z6eovCP60SlrmmI40FeO193NGumw<bq6PUQ@=?VWI%{v(kU}
z0xS~aGeS5uw*o)jAo6)%I^EU!Nvx?O+}y9=t&MhdMrUSNh-PBT=3M81jk1kBJiEX1
zR&lDKUH(9tM__UWwfC3H$vkVlWF)om9J2sN;?M&Qo2&NTWVKV<C)hpV)bHZ>N%B8Q
z>6DpS;>G60h*hW%mogAO^Oi-xrL!8PZ`OYHzG_hA8(AG+-3YOvS$3xAy`#gjSz~cA
z+9yYT{JrET*=NqJQ=e|`2=21vCsbk^^cIeAZ5)!xE%40LB^!VfhLN0TWy^dtz2hTA
zydjHyVO`H-W50Ja{->el47)kRv3@Eis~6A6t8|l}9QJ{8RpEM9?UXq;NqQ{jTJ@^b
z>al@({>jGDyfbYB6~Psw2rK2=$#>UJ){ZC_J84w;c9b{6WfyfEkC8s4@57&Rr<o6u
zJbzKd6Q|{h1=xq4)!5Lc+o^MgHON3J(Ff=1u+*2fHczP-j%c>O(6e+V<V4O1+Gqmr
zs)jJAaWr<==GoTG^lMQ&%Xd$qic^D{k;BopJ`RHp1q)+u&7aXHS0}?a=&X$KpLlEK
zfvtLKgys0g^EOIW$!4AoS5I~VDPs2GQ(r^ps1$y>kU@Dcp=nD^Q_M2YLaHTOvoEr~
zmD>jpf8VkmJ~G|kD;N-%!z9$tbacdhL&oWzp=m}cMWx%~x<A+rmi7dA)#*x#FuiGe
zB5sIt-`1P;!4j*h5YQEZh>xMALi6(J5D12s@cvZ)J?eWXlcVt&1Lxx-)~OW<jqB6$
zKP_C#BhN%vw`6Eht{=9@?96=2d~&wS+oE;`5oQGrEdA^;Y3z64Dma1h&;;668GQaZ
z9!cp<c1RgXUhv4OI}rsr7JgMWbR8+p+Ju2xqZ5CT%B)sUSLNm-WWPN3y<7`4Xlt~G
zW%VW&?fG%dfjatdXT4&_P}HgMWb*wba(1<_j!N;Ahb-%{{LFQ>BmO?i5x-J2;Z*#y
zR6kWW3K}!V2Y-+2zlQ|P<WIk7e)KmYIh{}A+YQYnpUX2;P&(`*=5E~_P0e8c%BLXw
z1wXR#=Wu#XjB2N6>D`i0t<^3o+k3dqJF9_*?K^DmRa@#eD$1CP5kp1mP%%}Il53j2
z)MTb>TSNrug^kpbotmrSHTi?{sb!?b?gE5u#NkrCWSgjx`8_e(XR8wogEibkI)6Rz
zxDIr_inclfs86ZP>8sCLmo#?t)rs8HMorUmHmM}c`-2ZkZ0?PDuvXia`d(bBAA4t~
z+a_G?im2D|*U_4;>w7U+Ftux-QcH93h*u)DPuD|x{H%d(M38k`u<vx|qsQ6R3PMaP
zdtFJZlA7WAZ2O1&mDCovgq1BVR)3R}IhC2%wQjti#<$WuU_<9ySR;82SW5WUB{KRl
zO`~lU9?%%h1;N_pu8j-`kmIYGjkQ(&>4h4-9~9l+Hev9s0f*-yePocrT{L&cz^Ydf
zZ9Q=_H?j&~x$FJm>?58;nKN^BPJYm?3Jd+&hpcbhMAdn@R9e-W!U0lO`OF=2;a7=*
z0ac+Vj0O(Ak?8!D*g+pU0L9X=j`4Bg`14L$mKKFtnv=oJdG%vG^<xc%UuM@t1!NM}
z$(>YkSJBd`rg0@LqNl}YUxP>EzhDKWV<vOBV6De;arFj{fs*a@9*a2Un@v0M%zQSe
zeeKW#WI#m(GyK=u*y>Z^ZepO;>VRa|!!Nk;9f**bQM`+$yRR&gl2lP!p3oT6n_3!@
z?Ay`P_s^zbacWp&&u|qaa~N}V*uf6xQl~FsVxh34+T*R!<1GW%EvBYSdWm($w;*S7
zpTactLoz-(>DlVGJ|RHIm9>D~xuDqXK31f_T&&uJPiL^s^ib7=$`&z!ca2)6Tbp}u
zO>%rLli*U3LyHNstB%n8F`Rs0J9+4~v+gm`&8yhxYpVpQG8LuU44yCF4-GO>YqdJD
zv(MbUsRsHEV(8?`s|)h$*dZ&Z_||XkT0U~%ZO|!p8btd;W>ae>GuGD1A$6&(CiQub
z-1PEPnjgxtwZ0Imhx4n`JXGr!;O3H@zp_J!#Zk1Rbdd;y@vWyu6o*>1p2e=mYcrR_
zqMjU6{pv^?)!usc*CMw&#AWqOtKX{bNI}~gN$#q{%$bhhAXHT4jf!sfrRC%?NABYu
z?ZpNo4VuQ1oh;%Bsmc~QaDxJ~xI%yA?XBEnXHxpGFv@>v!X(vsw!}WXS@SP-k8Fj0
zFYQ~{A^f-@@~t;HJPOm`b(6xEvTwl6=hMVM-Yb4Ew<tz(WXbjvmwCg>z1o#w-L}sh
z!*M&-PSLh}c6wf1V@*!vV)0KX)Ev&tDB2Sm;*O~uXrhOdRAU)VGI&30%4DjkC3rz%
zlO#!W%J}$qt?rw6x5lL4!ldAOhh2KZOF!+3f7Vas)(+>ZiF%YkG1bpo)=5x?(RP-o
z5OuceiV<V_D^B*QY~Q(?$lW9LB6WQ$uJs2RZ^-;8Hcz;TyBgwIWrO^%j{HuA&}UyZ
zt`Jx{P8)~69MHC*4GuK&Z*Gy)?(Lr*Hz}Ylgp<rrw3TDah#<3(i=o5g%(b(RrF|s=
zyXNo116IS$olv)Z{cfd<uyg7)6vmbwOQ)9PhQ6_$cRJR3M{~Dr`?z7q-_g_nWn_{*
zjvYP=c8N69$}GjeVxSp-n@3TU4Ds--TX0!r7(nxY=Je`Xe##F2{-v4?a3<kxaowRk
zV-s&vmBH>qvJ)5kG8fydc5&*3C+iYhWGaKd4fq;VXS<GkvtRS;cegqY=*d5hx@u<<
z(4}q<%D6tCmBWNz5vct}jdTjxUFZit7VrFBOx68*?m(~SQdf@xvT|)XA^b`ZXwp9;
zQeYaC!OUr|z`TbDwK!YzyB=oMGYXn$5M>cCN;Rz$k5{40JYdv9c~m=;AlS8!Q@gvg
z5!Ch{VmCE29_MM7%9#$;-@^OW{`$u~^+o#|-=>^e%iAUHbHz$2lKy6D42F8{Y)p3_
z$27Z}UB6Nr0^Z6(Twj%ggAK$0lea8kO++ScJPX-7o~|(;!qr0!2gyAxpou0(&S~gh
zRUx`-xw@<PytB)FHE3WZ9nqjpL|953Wk+1>vuv$I!7(z@tQqfnogXVb-&g%TVqE)s
zH?o8^-Fa8;A9VTVvm#tGc_M&O(rT7RGYfq#ak<HuGjVy?Yq#-0#DmnDM-M)j=W0%U
z(SHe68gM3BNa|+@f0rbaK#UGxmg67}y5|cL1Ti|g#6qhxyms&si+oK-lQ~O-M?2Xs
zhQxJuJr7My`9XEJAGa|C*iT^-TJHlF6XYyg>uB1VM-KerkP4Bic#_32gkScAxX{4k
zRvZxg)b=IFgnPj8#nxL`@8qhCePG^~r*OU0B9$`m^%4f8-QEixduICrOwQF|PiRq}
ztPRi6s$@6zEHTdP1!e3Nh?iwoFGSw%^hi~!*Wz|t@?Ws5^KFzRs$e)*F-Fq(ii+mc
z9)-1PxMLzQ6s?=0{A3@Xr2Q2Ir0D+E2MpA_4gv6h^TZBqQ%lc*1z82IjB9||SPLK>
zN1GY*l8TVVi1C3g0Jbe;l`6&MQ~Fl2={x@IL16t@fVIqx^n<lIW=Hu5<c-?<GU)hF
zdkmr#_xw2sU?Jr;LcItK-B>~GdM2B0>x3hN<^gf{W$t1<;SwX2OUvi?f3DlZD5^KY
z<}oF<7S8X#h*7qlL1G*Jb|oDRj(QBw#E?J1sTgvq>^Y_}M!PMg#3KfbBMxWg7%7iH
zdyLPlpy+*Au@6Ox?n_Ss7sXakA^X87$~8E{x<Ae>xjJR+a=}V7jt-K@9x%<?ib#24
zwTnrI`~mPb-;9->{Mhu`1FV50!WJ24-I^}8`N4fj7Zd;N*Cr!@)-+F%hi76f2N?<H
za=g;UlB!YN(c2g8g)F37!O)9{Y#=^#xXh6(iF03?pwm~Rqj(NiGW2lA{Q_>beW_$P
zHu^OXd5)U#6UbJ866-oZm>pG_MyzvTB4%`5TLs{wDW-7hWU$<Iz)WuirS$<o|JVyg
zL5G3_Ugc5Jy$%3Y2LIc=oZIbBP;BjmiG*r{zD^muCQ&ZryhE=!bjblQ4?1??sPGDE
z18Pg~5YTJD51e+{?+iF6g@<5lat`vYA+(5DQlY<NbN&L3s;x+(Q*(F$AVrD-(+o^<
ziFIFsWdMZ010YX|?HK?-+Xe(-Xjd%lDJe8lIu!@s1C;SW)^tnS1wd_b95Hejj>+Z`
zUXFMBM!Mn{@e=e}7YH4I&<i+PCJPQgzv$alj!E>Px^u)4ZGF%jv!Jcte0HIBFWF3*
zCo}vv7^t*U=~e88{w?UwJ<zX&G><^;vt4Ln9^g#RhjhVazy{ow+?TB2nGzzP{aapV
z`U|l4B0umdX+vM<43PVH7kch3FFHO|Gz5~Rb+Q=edXm@#lhNJ{lp$Y$K&Kc2Q1UK@
z$(&g*YD?TJ5D+Krq)l9DuS4$0uFN3&>u~%AaEVzq82K2vlR_}AMbwlm3w@Tk(-0zn
zChhJD(%pM*K22t~Qg1b8on*dzdyR)sriN{(sNlL!nf%39jp2m4j_Ic9EEP2>Quhh9
z{ey{uu^z(ov&bxXW->Vnt0SKit!lsevb`IHtxila6M)P-8c+6<Tc95B&f3sF-bc&R
zkiom&+IibS?wX;Bf?hn?A#oj4%MAp)jTKzITiROAzr+?G<C+wxKi(Heg@%uwKKe`-
z?wCObemc5)NQZi~zXrU=U9dVQkEfU^L+z_#bYzJxt+>4@0l&f72m_2v+j0-KB#Es3
z?v>kdd1+Q0T;d${3G`O>s+*njuppWVy(mFweG4-xM~uLRJ7DspLS3w7_^w0m|L-;B
z2tsH$w#-S0+N2FnuGr+h+RYI?Z3=OSCqy3r(22HyvFfihQ8X~4s*kDwK-+*Lh~DE;
z#S7FAwZQf8L}qyIsvh911sMbYl=!%3i4<(7-|~hXCPX@P<eTLV;8og&2=sp{lW%2R
zfacoO)i>jD)v#pupi>0D;rO3CK=ug;S^(kI13Qdc`{M`|A{2F;Bu>CWBQQ)E+x$h;
z%md6$Kq&s-W_~+kvP|d3n=|NT6Aspx200=Ed+U734qm`O1|qOk&na`Lv7_?ug#^Ew
z$nFJzVEq+IsQ>@-zu?9eShp$t6U-FuYer|o&mQyVG{F?j==m@sEXM&OSc2e|!s)d*
z7yU#sKb#@7vSF*#j$BKICke{82{g3nYV{gYujxJp`}?Lh^(ahD-e>T|tUbR$^2n?%
zMi6MdvX~_SxyEZjAh-3O_GB|{q8dE%G_eq?YSn};nbqWehG~RATWE=r@56soV|9wU
zS*Wr~+6=5Yog9Q)w~!jrelq3$=>zSW26maG*P+kU9N+Ggc1>XmB#lj{ku*!Fn$6w&
zl+5mcfzSjp;YN+u0hG+|fNK!3J%&Esh?Jh*{mVmlgf=n6{gk;xGpoQ<R}-1WtVis?
z_9ONK4-*udnR(%w!MaTd>co;J(=9IRP<DgSI!QAU_1NQn+~&xl?s2lVCATT0)O!_@
z5t3?KMp0{y1(^r^V?R4DYC~Tf`j}7W!i)qKkuL2=F4yKR>*3a&sE9?s5XM+x#gdHn
zNK;Aoil4<!%jl8wk=C#15)R*o7>mNWEzrY~D$IMY5mt70I1jpmvran?e&x=c;94}@
z{t8}_0FLUGqr#p;WFY~hoIybg5qylsGt;~8qTILmaqH1+F=R5#!elP8e_atbB^nWn
zG9)cxYKErVTmGi~c&yci>J{kyEwNskid6kphzUR8Wq%K*XGB<jvDJH#lNJ8uIXLDr
z&4a%7Kax27x?tinQ=RBTgXgD?e#n7IikH!T$vR!I7;9ar^gYU)h0CG_B;oXSSQKV-
zn1v7V{`6IMxG0kOAuB3}v-E65BP0%83^n+8L!YI796S&TI5$8-o#a{wh62*&5u-bu
ze~(HLyQM-yG3l6547vUcTQZYemvC&xDS=!m`)NiJ6OQP{&se%0Z@y2c-V@|l?~>ue
z*sxXWL?*yQW*eHch`LzLHB7BkXqd_J#;^8r>W|ipC5d&YaB_y|B*VJdtp};{=YLe$
z)%r)3BL7t-`h~g%b%)aZv!?Wl_DwHL#v(^2t0!@G6ec;x;S7@&MDkwAGScCZ#fb3Q
zksm8O19o!_)S=%imGc!sGzqDkT-M$8uS&EcZ;+L6Zi#z}2%6KUKN^V!*$PZ#M`mVV
z_94TwA|O3~a5!9!I)3Z$3aZB;sjKDBFycB+F)X4xPH`<BcbpOd%--T*PFa?w4@}#W
zzuqxPejALL=oprOA>iXZL(Dg%8Ww3&U+80Nqc$7m8r{}icY8FuGV~T@BWnQ)IMRD*
zkkmO<JrAbPhz&6w?aTl1LgPOOT{-YS2>mH8Nv*f?JU9YR>TPCzi_m6+UalkgFv;Wo
z)Y7i`L4>_-=F&v=dd7ezH!;zN`eWt>?OJJDAR}T+iv~G(FlJC+{3}STb^Oyb@g&1S
z1Y+rLd`j0kc?h3i%?adBd<>_F(8dsq8)n4PQ`TlvjNL+7(3Kl0f#>^Bna%y;Na`~m
zG5y_!5pn&!^9|PZ@pvQm;ZEK8^(@QQdmLsEiH6uAg#AxPxQ(oDe1Nvmf2qW|-~;dm
zM{bs3&GE1mml`6QRyS>AtG`7@#4edj(|@(s4CHXK*3*A*qAYnkgJZM!A!nupl2CnU
z1S*KKKHsj+oHa{tG;&~DwQjL1i`;59SN=EV^86zhPU^pc0ag@U+3FVN+)$GSQBaG$
zC%DETiGG<WXlwtlDrG-x*jBr6gI~6~2TDk$iWVLwuHF@G>wbA{JX<5UCIUimH$1;u
zFYT_AYd!KO#81}-LUK4=O>^=T#{Y21h{?ab%KT5S)0whP^e$MbdFny3D?IjnI9vx0
z*_-AKUcwxaVl5^3LQ#$d4Z(E}WHibVRkUfEi9J_*z`g%Ud_)?LucQg%5a4Z$i>%8|
zy#|Z0(QFkThJ+Nc5R$@sk^@0g=|2=;{{noW1C*`)FAsH|zx7)ATxXk=r>vhzzPOmG
z<B$ucL6d0NzBp5A2KpL0#F^7AWO=4n;gKo}D3&+mn9$r%r6JY3%q8Ld%!sO^uDG?2
ze+*lYbN8_C*jH9WEc*6;DFGP=5$64U|3&8idA9n$><aXK_B5;m&WO`K2@qoULFWrF
zo`Xs`^U!Rq)Hj3ajMLr+eZT*|-yh%p{?J+T0C{2?HvD77)+NjWra=boB{T=UDaHs2
zLWWLrznoQpC=0M!+QRVzDdJGX(kZbbnVWDD{jYMIkFe`t&8hE`$migCX%K20GnIc>
zXu~U4SNXJQtf3;@j#)1)$~{=9CB_RYFBNKoi3*Cv>kvgP>kf{mkXgrc#D-k7+a%xt
zrWk$mfYK%IJAhyV<v&Wu?G)=`H+&~OZlOT}PJfr+3}SIUrN(p|oHW-!IifXEe+=y&
z*zc5~>^QlK;RKsVzfIl0Xd%TKgLPk_zaXtN56kE(&MmTUVkS)JQL?Gef-y2eQ%3}V
zKTU57^Tt<T63X%zi}fEXX;H`;k#>M%+TTB~f38sBFRIf}lMHbsVR);+SDO4(R$+e~
zLn_n{lL$NFy*R)N^h1ZR5f--<EfAyv=#RNixTydlOPbUM62pd}$y~6s<G;RTkLh;z
zS*F#!LzK9$v!wv^mH(Q?U<S%yoqspD{zdtScW>+qZ7k_7@30k}rC@c=xQ-Iq)-<^s
z(n8`0`~*-2oyH1O8gZhIz2OCBg91ATCPqRd0x^XHn$&mV#=C*eT~G|>7Pldy@<tD`
z*xGEcAV*SGflol8`>5$WfWEf(U&EL2L3t5&<~l~yQhLuk<#WQyF2%9(gHVhp{jW5)
z1wR0QdiKP7o$O^N%&5LA=4}BBR(gdVbqr;58Av(w9}ooZL;!$j`Dkc_qlxt76*%=M
zv?(4-Mq1VJaBrG?pGYR#!q39#TkzOd%bNgtd-%qjC<j9k!1*-Qn9fPCW9h@^o@tms
zoN|ROX~nyRg440^@g(v~SgGjBIuEcd(7z$+|7D)|{}1<n(}2=(Hk42PSup{C!D&x;
ztTe0mz6^}W4|jY`9)`7HN0Z3H@FsoYs6cnQBJl>CTTtQs-ak8u9D-?(Cr+oEtdlYk
zDkWwW4G*VF_{N(P+2vzidz4>+!hrP)@aYChdOQ7aK7zNv1Y?Y1G{xi0LzTQmZtCG;
zbXA+)cOw-C3GAlg08Xus@{)EascYk*5Ep5k9L;VBDP+264~X5p35o(m`@!8SD1|?5
zC9{YLQ6E~)<^KAd!9p|hVYQ5*8b_~Eb}q>G8#>BSB8z-|dgHcEQQYs0^!4<i;V&D(
z)Hdr<&2ME~4dA9J=Lpz>+D`oZuV7GOh~r#SYe{l!Zk5U6CbcNyJDbWw`5+Z)K}ktR
zj`!LF{wIi)SWYmed!{q0xA;%1BkM_|>WRd@`<ZNo>P)5&CDMC6?Xl1liVv;w0{q>)
zovJCh>8JFavl5ZYmkEu;oQ7Lt!gYwHcdOUOLuFl9k?O7)KSQKgZ-pWUhqLIDYuUJU
zJ9YKQg#d^Jr#0mDlKP@PW-U~L^X*)lJ?D*>x0ml>Xx}#Z0UL6s;M$Vjws0ds?-5HA
zBV=|BLcb=g>)LyMXS;2kH>+)DuY`n+UqyPO5U;yz>feg|n`mGuZMaYi9^xZ`3b?K0
z;C?T@@d>|B%WgcWdzkl;yWekSQ6_U?a}jmT6;z90(BJ_^?jUozGiqg=7hRucc4EL`
zd=$a4mgqOSPg)wZ$1_wkSmVXxaZ9%lgC0C7Zog40?}|WiMQ#=xO0As-R0Lr4eVH%|
zu`*gaO!5|`vD<|H8<sUY#MPC3Dxh(rrqPR_DG`77ZN>QFkdvvhldWG7Hq){1^CJu9
zkW&O&|ED`xDx;Y~9)ok$iF~AhMnzbtiECqg&$|F^Lud>e1?}B!7|uVKER%pzn4~$K
zBFk@(xB0)IQx8F>*PNsH6^wz19RPBTtNER^b6kfWl9~ivmtcscgeXo+LevUvZity}
zTbM*ARK=GJi_B>bun}&qBmBB;I_@sbziqqReL7mRs6!}Q*NH~05mG;InN2oj<91cQ
zXp5%C)cr&+4P;M`gD<I7W@^EX7WryE%^tp+pn3po0gIt&eY<Nvo(r_ci*j=8*!QpY
z-CECS=VsA{LP|-DwhSp@!OG#rCL}IW<R{W<*NM80)1-j$m=7}ef`FxL*{Q=JeVB&I
zN?NmZjU4bdvp#HV1j8{f@#>yuSh!9Vt91wK8{Y3K&gFMbgYXAJo(DO()q6YOd#N_7
z;|zOZ)E$&L!$!SzpQzLxVU{4a3^7c%r7t62)H69gL&;=Gxd8`rj~6%Bzml5lE4B9B
zsnFg9FmgFfTS*A@Wv=J1@1~VC{vT?c^Fwqfk?H~Z&my{(AiQmN*SyhO4Xup;W{L9#
zy#ZNUeY|LmwV`H59zruLlx<rA8ribHGY*Je*-&?guP$z-eoLL&Jew<Y`@8wb&TQz$
z69D|u#q0D9Jz5rn7ag_UfU2Yv+(?S#<^(h~>vZ7%)7^ih`%iaDd|~w$_}quFWP40Q
z7|ta#o-8A`fQl!_8KvaD*Ez4JNHoGiY0(lYPsqmKpu0TII3SWjNtR|^a1FKUE*}W2
z5eZWoiz(n2*lH-JuMX@v@20|Zxd{<Yr3cTX*>9-c1ue{!Ev%cH31rOu!VT044}z1h
z(t;usj96)y_D=cj$O1+OU$GPl=NfJ5U$vH;kqk^ic%2N+Wwm#k@-h5oB88^X78?Dw
zo#XOv{Xbs0<5=Hbpc&f}ROb}^akXN&x(P>qfT{OYV2N6r(l^`HY%n@hc^WIX$mf1d
zyzAB71h^Qlvo^ICu~fgdSSrQ3>I#}}E>0|N4N}zvhfM5%T-=6!%Jz>4Krs8x2au?3
zmUU~)noLhvsF!tpzYMWfhD8@3;bn*jxdlISI!ZkylV(OwXx~Jc&|_h{BVv7WbaphI
zek?gFnL^%d-@FX>op846M|fZ)^NX5KiLyr)tGaUz9t{E`QLlR*t=?>p2)l-nlWPM@
zGnszVk=~TFg)N6#%wb^j09b3t(Z~K_z#ZB5cpflGo3&7^^ul)Jr57GC9%)K@4IEi(
zU~0O9>rKCPb!iTMb4(=TcpnViSiDD4`3WMnU6I_`JZ72Yvn4$>GQTxAuQtH0-@``n
zIUj%iUvw{qg#m<3t^sAeF@n*>v_%DEO<^oQDhcg|OK84Fl+U}u&sxxX6Mc|jv8+Ov
z{&1y_Hq)hWK<q0hKv;ebR$lzWP=#^YJW9VZGszQHtqj}_cQ0I9e-)NvhJjYP^sVY<
zYqd&I80CHBUW_VoycWkU@AKO88yd)iui((1b-Vv<GX4MlN5YeW7;T;+uaiX9l=%+A
zD=!&*_3AEdNR>yoR~ZKnK@0RHStj&)I0M=~1CSB$%mi}XeVHc?797D8F$n)k>(C`o
zi8|!#7kvUO)CVJZH00D^-&YixcpCCHD<9v{oo`B4r+!H~_l&wgt3qegFr$j_2);JQ
zqbtyNQ1U^zZpEV{UN>dITZTA{5fckJmEvh54HakVeW%2nBrNTKLFiHd9bbPw=X-N~
zxCd0HNP$nhuAXS2N%#e%{A1&~1b%l*?kC9{bDn{il2B3+Pn`vna)y(omr6~CDKXko
zov=e2fXk02RS$g~4`_@g&=P#t-U}_4w_~9%+sTp<rr9DOAmat(o5Lc;^e?!?Y&bg(
zGu(6D>T4&fI^o=wBsNB^?xtA)tvw8R@CEPSSpQd08_SS{sdWbdgwC{_U1$UNcsf}-
ztOmS*X(kN|u4b7As1jYUoGfv;IPis0*4`~>`OJpB_yF_yRh<ga!u$7zpbXyk62vEm
zoa|YwQCKpJ)tS=63n&Fws_yu%31=|9>vf1;mUXizcg$`p>F}^Xj6H^fG@(ILP`5)^
zzeSAHA%I{$FU}~{OoGx4v*%D@e3rs5dtfob2ZJ)3lk$f_>#`E@<Uowmar=2-@Ezlp
z$y49Hr77c`&;f#BZCm<U|GBXgcwmVckY)vBML^5lH>}&@b~uM+WL2M|8e7^+*xUio
zxbME?PREIm*wa!_W^<=s4mb`io0GWR(s(ur#*aQZFW$6NQ~sV87=%-#^KbOWNvRnC
zJb;Q1bSadBiuMnIG3Sm+5>zdrIkd8ESsws&-HLf9+hNVg<TlG)Oz<4BHWwvD$j0hO
zxgI?M<eK{NqgSAOim6@eJ_cP+=uC%7D8b(+q(ZM_DC<Manj&#I`VBZVVMN2Zlv%6*
z;q}~h<3>=_hL)(j&a?a&H(=OKejQ>OK!e!?CKb3tM2eAwS5#T;+~It5xEe~r(6-Q+
zfj4dPH-JDjt6h(LP3do8r8Gq6!`ZXDl!_TLwnasXZ6R7fC+{JM7<Aprd7V!mJji2K
zs#gqpEtS820<V1Cvq0m4@~{fUoJX@sFbx=9kC<Pcvj@GjE$nv}YdPQ`04zE_1K$p0
z!W2#Drm_Y+dwNXh70|MLz@0TBuk$xEULokp{xg5BjBGpv)W%62GJDyQa1x|uk+*l4
z2RLA~F&XB>-MB0=9cE!lkKQglwnM9EX7RxaVbI_*ZbN(0p`JPgbb$|Hz9x&Pe=r)Q
zh|y|ziOCB#R3qcSn$_t)d6eYY*q?F>GT34I$NhYOlB`=k?*$%!;C+?nLf@aAac$fJ
zXr1CbQ3Y~lvjfFY+_U*1YwNXxi#9)mw5N=4%xZxsM(nuNofM%XGAy$6#A+&u4Wvjw
zWjSCoC1)1^T~|NB5BE%5{{mlH`?&whr9YcCl$$?IdCA5Ilt|dV1dj3_0Qi8lbEby?
zsA3na{{{TMFn}G^b>(J7x%rRhB`J%1pO6U+q(lH_BD(*!tO-Qw64_R}fJ`j;G2AX9
zwlJsj&@v{07wtEmzk|2T=fA4Mh`>U_FgAH)zU?wNeF!RtFkpC9q5(!#i=CB77I!h0
z1MKs*jUNYj&Q$;?W6$wi(6^BawZgc=SZ`r*W&y6Q(7Ag^3{=D~eE0h==i~X%AZA@4
zE$`o!54Q^t(99rD<D}q+EuZmuO4&--yaSG=jcWpsLjrdh|GDmZeG>U|X3QDj=%r4<
zLABxPj3GV(i~s<sUAxeZ+oDGyEYAehn@efx<yI-B9>&~?C;tm5DM^1*Xcgg^fR0+t
zPoA<H(YEP=H{r)-4gt<H`M^yty_|*FxA$dsps#2*yKh~!^oo;`GE0L>ZwcgN%Nf=&
z=+)BDdQl4b>nxU^0Bw&2og|<*?HbSbh5PXFjQ8qLiLK1z$Xe0Fxuok#9^20|5Vx?m
zu;fGV5>pm<9)~%irB9rOXC{%4Z2M*GiA1R0_Gt$+cD@h5pk_5J#5C1!(mPD1e8=Qs
zm>jS+a|)iBZn*(INigAFEV&YJ!{igl#^~r1OJj6`0HMb_(L8{_Rh}{xiPvNW_}P))
zY^JcHEKqNqbrOA|+nl~x<h0=w<bXLM6Y93jYP+`~B2*UNKkGeF>b3ywUZ_AB&jEv7
zYrvbVXl!@@mc}yM28iG`-3Tcy=mL%z<a3GZ^+ApkLx|BzOnA8HfM!Dsleg^L40I35
zoc<H0XnbrH^!Ai(gn;M@ne5EAZ}WAA@e``yvDf>X^@xqw23dH0MPpc}HOI64`@uM3
zYnVePls$Z5yy4?9#xTZVNI6mK-@mec4`3VgU3n)dFqSjMI%~>P7Qx>DBW7d6SIwpN
zqJ69&4p1T#*#HeS?pxZB7KtcPdJwcavvx%coJ=_PoM$j)d<CQ#d!Dg?V?^G6owUij
zzM~3rD~ntSPr@w113aLLtw$h3Uyi8V`q~<(Y2C)o_GlptK3)NByKl%gvLPCb!ssXy
z6BD<(U^KYL(d7bIEiI6jXRv_r4<;f}xER*fmT(k6zo<-vqd{3B*2)fA+lMJ}!PzTF
zjy{BA*)J;!Fk7=@>YnFvE`H6%HmL3bd2+A#DH2Y@izlx7LDf$WxkEsl8U0y;^`TNT
z`Uizf_x;6}y!N3zezKQK1qsX0mZ0Vg@?FXy^juJxyTfUSU|$vs@dRFMHvlc(zshH4
z%eu%n$>&^{#2T26k&u73V*CZDLvwO=t>|y?AE;)pbAl)(YhWlZgO?%-%OdzP{>)x4
z^;RJRf3HqVLUzHRr%lRX&_gLUP<fWs4cCh&L-lto`SZz~`#<;O9=yXb$LQ0%>y1bm
z#!#)^^{+QS(q}fBWr)$(3}dJ=3I&C1c%&SWmeaX!xrxM0c3ZmH1DL6AqhMjscVNp3
zK8_*3fbW5YdYX6<pfyu_Os#uG8}!s~LSU0JQBj7~tTmGLT3KM(sue2_sF&zwRVUUI
zC#;Mg!+E>}qv3i9WE)HA&4}Q3e?z84fGzYQ>(EOkrOz~ESS_%D%TiEXd$t&WDyYDR
zqy2-8F$rhaCsODaa0$FAQF*V9gsT!QMCR<i1a!l|6wAMt-@qp9hnS!gs}@#j5x^qJ
z2&1peYOtIT^0KMAy?eo|bWDxRd$TpRI~0g0-~ACIt9zv6${hdldzi>JekgIL%K`5+
zO-R1}a@;O7D?xA$l1iQk)mw&OekLBa&5FcOxCGWRg)0be3etWdUO~wVTnBSul5mgm
z!5}Y8Th6PUoX<Fm<3PGI2gIrQtbjwZhc^AjCszh?gpRCr#^?NnKIt)V=dTk0I>^U$
z|D+!rs@YSytFj7Pj>%*TY{O|HF>`7n(K4keqo<wgRFl$wn@xq4ngoOsxV^H)J5T{_
zmOBX6M?sW(xM{8m*D1r2r|ekPT!RRjSfei`9fUKfP|dA^SmAx~b)Z2lrvsdjW?|P*
zv$Sdq)FPk#$MFvt*vi`dx|zx~=K!p!2pV^;cm>R|I-}tD+7gdauTF5}vx!r_1_)iI
zE$c^N?U3@aJV44-`+Q)dB*8ku9XVU-nfIzvA6{ba5x<dT)icn|lnP}+J5tcavrC($
zyw0k*YFsc3s_EmUi3m)%QV#Q<QoHtBJc{sRzh~1!&O8LXQ-7fVfOH4)6^)!K;8ypi
zgw8?t8h<X1j+8CtMM)9;m_uNtEcR$*x4HB#^p##dXLmC%4ir7+$%F=(T)rSjT!-tW
zE1P}IOn{>#F?4pd<uk3Ywxd0}2s2R70-TSA6cB!%XoYETg;M4eWgP;y80HzpF~uay
zu!>>&iI#g`!<w-eu3)YZqHO5?W1hitONQ-FDsM<|ja<^EKxl*eLH*N25vkC-Py)zK
zhmT{))v~L6V|wow&Pua#5+wN7g~d>$*Co4^<RD_-1BD|1aKJnJ5Oi=O3m>l>NW$~Q
z7E7^M2~l%!<|-Tt@|s<%3qi;?0-+d7R!Qj@j6l{>^xADrGxHQZM#04lU0DJBO<KSX
z?S<X4pmuS>i<d^fp28!8L}{hMawmcM>ohUoI7F9ggZKCeUSa*HhRPz!({u~c&4%hq
zdDu|0a?OfctA>Dcc$mGJDA)aCFoGj_<1MQ1&`5amE&#oLjQ1~KQeNGT$G^6{TSGe3
zw|)8iV|n)nW0a8gfZ$vu0R0QEOEj^b8WMIf>+r1_ff&c+Er$<t3uCzg5m!<Ih+DvB
z9RIz}&4Nwf?nXw_26pdre$(^?I1V-arVTnX$~T|Gp}@z-D}9FLa|78<^FxdcSh5r=
zynV+fqYxce!26%22R)HNm?$Z{0<k$x^J0CM?+Xx$BK~&O;jT+7`$~n{nkYUyVDORJ
z2S000f9~zQ2i<|;Tyu5c2MEp-S09UfmuEaXfb+f9(mdQ3Z}aX@Ze=Q$fM2v1Xy&Qk
zm#<^ur18k<K{xk_^k8KOZ`Mn#3H@`g`m0*O7MP?-00<Fe=w-};<+B(6PT^8<%T=)E
zl>K07r~erqs_`*T|I-pWPZoGOcTItOfHz_248RC>w8#@GzdBa4&(vN2Zd>>bj7VPK
z7nE1~`5zYM9$gQDFkA4ZJVa-XEOrMMTC~#PHw8K|V+SGj*`DxWevXunURK9<-by9^
z`}C4wrWR1sz5xP&L0Lx67RP<ld*quk+%c`E0OU#Zq2vgy_L|dUblYePi%3@Nd*5O~
zXp>sG<}+2Gv(E3^Fzt6poQ9^hr$!6-|JuKk5wjsuc6-UsbXc)K9@@cppgVvgI0?Gb
znx@3-=aDp2&HLVtHNBcPi4f!~(X$_~3LPDJVm=Fd_Jo{kfDuICH54a_2aGd!_)Ze9
z@a;D{dz4SW%3G-BQ=>Cdr=~69rEX)Eqf{{^Zel#UI07aPH(H-%M>aARUe(mRSL@NV
zLzYh1HD`U(Q61TZ?%rtAIXKLO@1k`3T=x_7JyCywd}8^^q)1AtMwrlsK$5^@oY#*F
z`}V~PtQ?!M0GoL~;=i_euS&|CIu#>V$9sVvefiR*O93egz_NxWBO?E}aQ?$Xgd{O`
z);dZWF<OgVtOHJ+I#ukfw+@Ez()jhB=BTE{2!<at75N)#!c+}_Zol+$Z~7OP1A~_X
zCNFvCa_nVGKl)cU0u5)|xHKh|EsIGZ@567NKYyO$wgBo!^RLnbNhd*kT+|_6?m7P#
z;Q4*~JZ=^6oPheVl#)mi)!;9pGjV3YnO|dWmX8_IDK|&Pt;j2Fx52h<o8rqXS;4aB
z##^$2{xxpU8SBtSDD7FREFkj972c1>9thd*JUWW<iAc-mI$e5yzT`9p1W}LDJIFOk
z>mbxa4}JV8?k`<|H8$IK2ZJZ?jtU@N6W$ab=sXuG=D&AiL8lEmP~x=pqW_MYme_vR
zUH}bs<^i0ruVjD^pAMQlMW)(*KfiNSFg#;(jm@*&=;ixzr+Lu&`IFc}@Cc;qq!2~*
z*Aks&&s~muXg;DfV>D>VHtgl$$Sxmb?dIk3r}D%GReSUc9S8RA%hwkM?t0B|^^=3&
z^nycw{{*9-A8f%L-U)<%cw|um332JtOkzO5dS*=5YYTSm!0pF8%qOvU(GAqh&x-k!
zufyCp{#cgQ@(t%V(hB<_Rqp8;1qFx8`*!6yd%UVE?*ou%3c70$|LFO1)ta36k*N8e
zUHhO>N?$bxoR7{5k2QS<S$FPD@Oo=TsA%1y2|$)oed9(2wI6mn7jk~{$C~mv3FryG
zy3uS4St$1&FzM`IcXC^oJKALv)^<-e-qaEo=}1OOsz2Vd_<8dS;dJ7?ZF{RefefTw
z_O3Z=hs0;kgSUW;C&hQp?|UG%%rj{{*^QPd@8f}5NqM(!SH*h)67>Be^R|GK)K5w6
z+LK+}2pK++P4>7$a(|adJCFP+OH?~gMzJj$Tqp~6xItbKEql{SDd>O4+16QSd+20I
z^iSaC)lWyRf{M`SR!ubJfJ}iBLI63Fm&3Pg1vc&8T8btx&ulxyK99e;hC$T9kH^x-
zoL<R1R0IaBi!Yx#rF=y9re`hTpwjic8c2lastp0205JMOAw6#M<afbW2_q-AZ@$?z
z2$m9fft<k^ee+9~j$6d?5;6--O}D)PdY2*(I{cX&Cw%p?=yD!G>R|2syVANY=)U|9
zhE$OD)$>UBu9UOCABfni`G=^9N}W388a)Hr@HlVVg?DB@Fe`<eb3LndOeKp~Pdlm&
zndk`(<(rtAJqX%Po`cS9b|FZgz)hVOkrU!vmUl*l79WS)QHogpM~#>0ve0?n0;51f
z_H!3Ce;-oD)YQ7(%jd?8YF=RTz71gciu~T!tuK%I3AUCyx1J-|UKni#q2oq_?iD_N
z-g)*oaPR$_N~lNUFHl3vI~RY0w<o<WRJ>D2%i{7skeNQ@AaK>(6cVQvf=M&~C~x>*
z(BY&!WhVnsyB_={=Nzt^cIC{eo^J;sfbZKUdr^}IJpr8na#otTKP+|5-1(6I$rgT|
zSpFulx?;$`p8vG3WP*GI8;<ZB@da!z;d<k7NGCb|JDqRe&3jY>8S@l>!U5nW4@WU=
z^>=e7gHQh~eGs231mQiuiq#K#eRsO+cwgc{nyLz@Ag$dsdg4LL!3U5p?(?vn-T_c~
zl5W)fLF{)3E5>~OyP=oN1r3@lJE9PiHGl}O_CcYP+61oAHU=+2H<a*EdAUx`A1bzu
z7B7$kxuNIR(3`(j&U(K_sU?{s6x8{2vU<0Cj(kKCkTt^!iD_2A9|UjTt<KXdth1A&
zsovpss=Hk4<bi$rQYE)LK7`L9*bV~RQz!JIWbFAG1|9*U`F+4@JLa3uy2iGcm48DD
z*PKLcfLyA4TUh1jHj=yELZDB)V3Zmck&hAH7k_IspD;Zr51P~q8Kc?wN5JYcYhW9!
z2_K*>%2x<ACK}r~JZ=Y3mi~^j@~h=B2fyu2b-%^?;I#k67b0dn+UR0}x%1JZpDcGc
z?8p$)<lcEBJZf=h$Gu>ZiyJU`Ro8BNFg<4NI|1)CjcqlGL*C-FtAS@@&v!x2k)$T$
zOY^M?0^>2l!xWX6oMWPV3?3wUjM1CkexGAKGHN%;cXJC(U|l%=(Ut0xtTiu%ygYF5
zFaQ}0LO9gy_Hb3BVg7RJgBvQ7UWC&5p%d}_GuC(3FKK0mI2XRk1Ty&*onqo`A83Dy
z;M-984z1Sm`m+hhss>7;!0XdsvvtsRBb{z%dC*S@_Gp!du7qlWo%7zVuMp}|OVg^s
z=Dcy?z553E8sbtPgGOlDquP4q(y1kWrX<1dc4aG{&EpY~@!0l;*q<qMh(jO+p0u8s
z8h?)Shk7*iJ}g4$yD8CH+lhDG*2cXwJSh3)5pa6l_qnlGz<qBi(xY$U-2iBZ{{*5#
z08^Hl`0z2oalVq~ic%@Hcbx6{=9DXBQGa=V(Hb45(4_cVb|iYbuQ4^1|Km2+Lc5~4
zV8CR2n8rWvFFWikie6-^`Ma>@BHgBzT6R5*TEBA<T{GX1d;Q1DZ%ceZ9#5e$8e<`+
z^cPo;rYcW*R3tk}040Q%fe~S(VhCsPhHoLPwv$N$XP+;Q?%@<|YK}H`b(-<i2y6)Z
zI|ez~pJ@y2lt12BE1>|HVxNaPgx7I_TOzM!@TlV(TK-q?F_4YopJi`1GkN%q$LG%i
zWf@@pakby(ldNw|m1tAXwN2%a-QP;Y&%Nl2-FEPJTgW+Q`S)}ObpHKjGz}7ZHEI*J
zPd0uPUCX=l7S~G<o}At-ZT*mfPEWB_W$f~g$*3_?wl?&o6>{gdhY`<u3n6#x=J>=v
zrfw5Eb*@@RZ>witK6oVF5+@=iImAThYoKDFlUHjWu0RKb;U<#?nshDS-Gfe8Lv4fo
z<#lslQW9<T&;o<pI~j*wMJsl_b=WrB&27)yIeKbRI38!8?{(6bws+a#g!u3fPlMzw
zi-=*^Wy^nE*9=lc<j!#Z4IS%-h0?7pl%O|*R=@Kjo_A?mv;R+bZypZi`#p}&Vvsdu
z$u30pEZLV*Lbjribu3vzmNdvRGbB49Bx?!DHkK?UM)s^p_GOec!;F0yX1<Tt`}J9V
zzw7$__xtO6Ul`ZSJkNbU=RW(n&pDm>07znA$?pyd6AO-ty*BG0^*RV?s#$vJyHV8o
zzW*{iGZiohr3J2_VghB5NaS-8eLL52wkSWjSO8aL{!(<_U*JYPO%Bzf@y=)R>REJf
zsko_39azfW^zy~;&lmwC0Y-eKnbYYry&14^X*`~un?yf`f13l|a3-4qq7Bx7a48~B
zK-A5nIv3_ueCi=_==0hEo>)5A-2Ds}eN73$dS+Bc*}u~04j=}NslTbn6{En_GGq?9
z8e9f|MhY^3>S7w-i)tQPXC?kTDkjHlnPL;hQOI$R)*MsdiNbDEga81_(ExDC{br1(
zZ|ppkbvKu2fGg$b>uDR5tGB^Krm=#nHQOR6Z?IFc+AR3-170rSMlsN-IwwcdQ3QGx
zurlY1<8>aypV`Lnw3evUo=VBTCMPY?uYTXIu9SI!4*olmE*KQmIzt=sx^LkFQ`u!=
z3+XzfL<aVcdbyN84#FlRPqnOhM}~afSvn;G=v|6dKpe`#boyiO(_db(dRH5Fu_ehl
z^$|)Bl15*3PBCq`Vt>8OF+c@`lN1A2k7^CD9q5%u@u9nJF{d}E5my9ZRCfp&s3fQo
zn{pVP^1a6ez{1{N`~cV#7SI;Z<oA#!i`Kg6d&vebH3FG1=L6p{C5VkxqVHo>-iC^T
zE1{1o4dSRgiCTQ0HE8V=9;O2Kjb^>$34)uf%mFPi*(LxrTW~w+h|+j?<2(N`4Gi=?
z$R@u3HM25qv{T~PI(iv<!fMG>c4o#`tRWPq@u2v{r_cA3r~U(Y(|3+gkVDHd1rJko
zRoR1c4xS2w1l<$lYHrS-KUr?XdsZ3Nv^IPbhnEAt{QdE!aWe(i3=Lr*RnRxFO<yD;
z#!4K@4G$@&D*h568m@8sK+S1=6yG$80DGhLfg&?w4?T}9WVL|ib2JG1TZ2LF)ZT)@
zg}oPibaN(6!Lw%@xKSQG?{oFxI_Q>1`v#2v?yD|<5}Ih+V=aadMm_UdKyMNRHEg4M
zQ9@WFF8>m>T}ojVgk1Fe&mh$+mz!^*)ncY2kzH#)-hAIggp*{6bwdMTxO(h~Xkgo}
zyG5z5@Auo_1yxkNQ)@(2>z9__MgYOFxC<@<*u&Gin_1!c1b76{v}swsGlM0liM^q*
z!sr};h39^&qLg$I8z9Bex3c$f8cf5&Owu6Gf6f9thVA8WSZa&VUNOx>e0p{DTJXvq
z?Ug(2f9!NcuI%3OkqNJF8BX|TCiV_V-6ShC8)T%TB|7?EZ>ID9-6vx2#nVo?={LJ1
zm>ZO-p|FvTGVqt9uWYf3l^(J)@>ZjK>3R2hc?~73@dD;zwb}An01m^$fOMONN|yTH
z0M@K+oHRXab!66{+%OhF5@>LXr4!2mc;jZ{Q>=uL8b8R}gqO=q$ByejnrO3-_z<gm
z4Cyx2P=SDIRM}UUW^%7Rg`Og1M;g#Qp6LK+Dp&;IZe-&_?AB1Cg1jr7{g095=CAYU
zIjVYi`k%s6BL~uez?CQGM4{{fpj^cjVg#`C$!h5!{@tHc`W=l-a4+ZOIzhfbtA#z_
zq%WX13JzC)F9BoD9C(%K&E!Fu9pxqbR4nakmI`6mAXX0l2NpBHtO~XAg7x)iCG$zM
zhO5}ax4g3X92c_9dabg6=K=RtHedFi0=f$;*%t4OpfafTc>%lNj>-^?4gJOx?(4n0
zx_;}f<j@&^+NK9bKFhN5Afn;JEI`c@91Z~>7Q%GEPi|>Z>#efm9Ms~srJKHvU@0|S
z31v+-KW9X#X{?R<a>0jjVlF4@4jrly;Df_k(`T?c$vMms`+gc$X4ZD+rWvtQ7}cw2
zW84iy-o9h#q@qk(D%thwMCnKoT8hmh`IPQc#R2JgzC)_zRlHWwaVv7@T5NP}eC&IQ
z3Jo_E?H3&}(30Uvi=M%g8SIphS`s|f#_BonG@)?ThAaNMkU;(hLb!$Tch$_*gYI=l
zppn!WDw#e)ni%jPf*pWl#{8IjANV8tZJ{3<Ag=ZPWH}x2k1;lBR&To#6`(#FfHbO=
zTn2l$n$GgfoH;(*RB}{XVy8|6uq#?0os~pobMGY)V(v$isU`X``sW@J%5aio(%mEe
zQ!%LyKuT|w><y<gfZa1>MA=olvYnD-xntrgGEifP?Xw2sv|=%POmue;zzsw9Yah=B
z4}$kmSyUBFGc3?q_f|K;{86o!UGxBXj&b%tGU)-6B-4yBc+jHu{^=taJ*op;yAmzO
zC@e&Zm7R*QDt+DVC(5yASl0cnTYc%MYGSsxiVzD%PC;jVG^|B-<;Hfj9$!i)i{dLL
z;;3&k0TQl#IUQng3$a19@$rTeSN#5!26xY=H(yr&gjVbgKG`5fz)auxZBc-8VcuZx
zYVGw?4Z!_T4Gct$Rs=*hG8QlK!fz~2FWLaajj{lBC6m;?p{JNUA-O*~tPLj4Hri?*
z@h>|-@6FGZL|~z8?hn@P0&)=*2PFuTypePb`cLmeW^eiEmuYGT77_}v5FaJ$b&y!%
z_GJ|iPTi$K;NCyrWSR2>|F;W8va2v!FWH7CD>GDONUOJ%uWpAFh*QF@Ll7VmX2&uP
z6?si*D5n25A1s$zF<j|R0tJ0<>c?C%#&npH-PiXDT<?zh=$UsNkZV{E^!?<}dQ?CI
zUxK^MY-*Tlen2X(<OE0(vagYo$Lb}95%5=w$)RDdev>Idn^)@d1t9X9Z$^RjwDeRF
z41T!*%7U)+i~c?2RYf=ZkIFt}Eo{T<EuE#UUa$=sLWi84$=_w?>No@AyZ#_RuqoX_
zJO=Nm-2vP*f>ML!Q0W-_bbR?xDreJj>j@1JoG<F%Wrf>sJprjR-X7eT?q1J`wY=_Y
zx3Cx2Q7OO*Y(OUE9+l<IkJuo~Fv=LjfR3kuJFAhu3qmE4sqzzZPjq}Kx6K9Rk?#0J
zS<8<Ub8cAcYYKi&p61vlC`0Gw_eIsJSBmrgb~mslJ2O_%_1x1IXy%j|nWqo}bRT5S
zaZL>E=XVB&?;a3!2{NQH*OJ2j`2I=%l(!w8Vldlpb<Z<<tK_=NgJ^yHH@pLpX)!Rj
zaQq`$o?cy7));KOBU48hGsN6lVh67Ysl_z~2H1}OO~Q0xr<C>)G10Yx08xpkF~&i|
zK)G}wgWuJD(|~7R<itI<W#AzcQ>j{#>aI5Z?3Py6N;F~>bA}E9Qk$yj&OC4}z}!BW
zGlFWN#g3wjrI?+IW>`K+`vsS4k-9puzg004RveHxdY;$csGkyG-j^v3v)6G<Q_~RK
z1xeT)tt$*XoAoSk2x&@#rI#NFay$E$(auHU_&Ok9n4R@s;m`g^j=<p95WGh>pX6(*
zum>@>fc1W;d96}#1SZz~Dv<5(;id_ID&i>5s^g%QRAJ21gBa8&^Lx%WEufDf-l~-O
z0jDLE_FKeZw#TQ+o1U4+C(t!f^KqigWt;9Rj3Jx{3i%J(1XUXOaQDX}EnLyipV&Xx
znL-ZiHZ}S>uH*qz$cSP+R|6eA*Yi(c4loBQm~Wi6DWAUm*2|g$f{>0TLhO`=?TELQ
zW~Et@%@^3g7Jy)dZIs<`dXv43BDJoAW_9*rV-;~qs9jtb)$C0Tt8vR|Q(Qy$z=S!p
zLQw7LdnjM2bLlB2FGZNsf&Aer!X}Unnr3SCn112h&e=E}%ZuOlp-Ac`n6>#w0aW*0
z-EES6LV0eFXpGcUvqZqhyhlW!U!baD@FwFm?*-I_x})WLaH{Dx?2StcZ!*5CrQsc8
z&a>JcXDFp;l=^`J^R?H{vAD}hQ#V1?T$zN?1PRv@n?N<}iM%2Xd23<dC;Atz*}Lmg
z?TR5e1FLrDx0BtDXc7RxXgxg%7)Udbk!;Th7^@AsC~pTS&~ju9jJ&@)8lwOkqqm7z
zbnCwaQ%_<%*0Rq-KOF>TVKpLbFErDw^!mbOh^jb2pM7u(wf9HFI;+-ee+TE~^Bti~
z8&r~T?mc=&)0pU2>%aPJt_N7NL}273yuWRcw;>3ra{~`r&Ni@5-xz<?!(LW%3Vq2M
ziY*P|E4%puo%Si0A0568MkPU%%mRs8czql__S0pMKzN?r@=0IzedfB`OAff?PJKX2
z0yCpXSIzGoIKtf%p<k+gK{z75Km%<K13g?c%;zqaMaKnjKr7<$A(<2OFu?y~yrJO2
zeh<mV_`&?;GOFu_82Vd@hvRtA_X%8BWR6MUqq)w{6W3NdkTQa5FXiH6DS*uqy)Xi-
zhRqU`&pLJ76;rER&lku%!C?iLCnI6%mJNE=8bJV~#X4cRB<GGZor{%0^qx&@frV^K
z+jbRRYSszcC<#F|dyqGhg;_HxbFRN$%tuJjl&Sul!O3|=$g~ZKaDObWeB3s?e0Htv
zXzik8{}jkP74Evt^fEXK8_Fob6moZAK)NjbM<i|olABH0jkQs$6O{Kfn`aBH;I^LU
z-0MC@R-WSB_n*L8)zJHpyl;Nw!Ok{w#%bIBTQLfn0qjJ@>L=w;{5NWvSWb)=C>M&D
zTna<m`rG&QUBf=Y3E>o>^$`K@>()Ivi<31-EV1Hpt4CGWQ!%30sr#8O0KuLcqi<rC
zxXrES6hzVW7<my^h8Qfd%%AP`9eUHfF6JZSl={?;dQZWEPFd#6F~hVuuKyZx0o8$>
zLib`%&ILZ>qiK}6q;}tlF7QKi*Ym--PBe|gwDjqY2Nk42KzT5}HaZZJVf?u*?q<+<
zUx5rH1C%lL5iOX|W!u+xPXdnAtydx#N_ayK((I`Y?pt0FI0_A<_7QKOsmg8=XSF#c
z1jSQDV~=$cSK8gYk_UNQd9U~8O{*x6F7th<=xJa2z^RYNlYHrr4Tz)wPmc4IEClW*
z5~>$I#mEC<I-ClF=U34Td|hr0g=M|;5tFyDl7`JPdcyY41BTw`$X|=>vQI=P&`44H
zhA{lT2%aye_>_Mxi~)d8z|x?+9nmNS4Ci=aKi6tIU0x2X{e+|inKp~6iyg__#w?C!
zEH-Sr;`&n`n9pADQESk}sDfN?GDSO~u}-ZguG>Z2Zpz%Khi|8t4i@puPl{!0K)bv+
z#`|S}3r4E-lf%={cREOZ>dM$4iY?EYhm6aTIHpk#4vuuQ&D8fHp=K<@4l=mB3sFjp
z)9c0JA&frK@=!rh5E8WU7}p%+uen8CU8$@4`ebr#tJ~=X+TNn!)Jp((ivfNBYW~NU
znYHN;Ry@l_o1PuU+`yNmAw2A8!_Yw8TLXYC3^Z%@$=mEx5EL75!Unc*bVq$lY)${e
z|KVxx692Mn&Cr4f(Iv>+R@CRI^Rlu34;ewzcLfoYHNgN+SeSX?82;NMqD_0BztY_=
zXerBP%0<sY3oQ&SJzu4e+CxJ-+2Nc}hP<Yi=o${Wp_*u+b<s_4U1{9>@@!Xq-kO6y
zt$};PR1MspH%JI;f}M%lO|2AA<28k3W&3h95F!FFObKNI0qQ=_oR`-lUAMk;SfYgS
z6`kcb{?W8ASV;i~eq=Y<kWlMK+-)w$CcjGW3MRCE0u$P|h%b!5N~rozdtM(e5ia>q
z63eMRhxZ9|&??eid@_~*GlPiXD!wrTohwDEX+nmvNm)vaY0-Kb4OEH;$dH9Z4Qzk}
zDd(@cWB}z5o5|X|p$t-osLPW+N-yWMb>1%>f3l%$E1EDA><N_ELljM(6aM3Cx~W_e
zsN`hJYl!I7bL}tUhtJDuEho6nbX_uA-tM0DI9T0y>N&TL)5qyE|5e9q@Pfl=H|#vM
zZ|&4r!H)SF^1w+~5wQg!HZtb+;+`Do@2<`{sh8PlA8f0~mv6D35*Nfu7Ie1oI+6us
z!wi_eUJHq`I7+Q?^8J4}dIAi>4(5&<lv+z{9p-LFok4Q9Q(Gh|EJr@mNpK+OygZt3
z7NnEPW6tKC2#^URiO(nl^tz$$!gQGctDy02l9QSLj3IB^z;)h2(aTl8%clxX3Th?@
zp_Y8dn)w`-w6k3BOPOd~Yff2CU~Xz~Kvj{4jKr7y>grIf-YL=5Zfk4!tK7XWfxA@!
zn}3}~YZklNrhSJ!4t{y?^ze@7Z<+Cp$W}H5oSohs4-UG-evP>n+=4ErZYb#HQf2&h
zsoOUA_vnLx<bf{Ji;*XOe1$hoovbBqP8Av_`D~L`>bIY(u3`CbDrJf&I(>b5-#9=s
z(~#%(93}1cxAZmf%qkGw+>=h<cH&rSj3?>P)q@5HS{%bVh&&D6a444$Rz5N*pvxf=
z*~SD3uqb#6(986<?sYmwvKlkJWOiJR6UXOHJx3^>IXoln@y`cCCx(){vyVK7kmeo}
zDvjwm50*FF9q}$tyIclVrrIUgwc6YbZmq6%Z~A!CuW@)goc#9QZjPx5IzH%aB-Jvi
z2el`M(W!BaKgeUft}A+}3DqB#RMfXgfANbeCd}8{$wWa}HBQSa`1f@pf<18B5toph
zGldFzoo)(9f%jWY6U><ws^~!w63w>XC;E?mq-gzVbf7)`V$_t`hp&Q9aFZx4=cRmX
znCB7&zCJZlHZ3+T9_z>pxSey0=4B?=TBSgm)H}}+#$bDxH<FRdN7e_qvKK+tOnwAd
z8~2ye!7T})m5*s2Q+YEje|`*3FD3JM5}RLwJ+GX|5^p7GM!@Ke+rnvn{5q3%g_`F!
zi`GHtY$n+xEO?#IJcm>C*U!N_1V<*B#G*40%l70=v%n4Fs=5Y|PEn&JR)g8G6mrY8
z`KFXIvo&J?n_{L|1{mM;TLrG`#HNnm^ASsTl=YM#+8_QjX}qwf#w|?*A`S%c+=7q)
zgdl?o0fE9huCuE@er9IucuE)Lj{oh%;v<*Ny~~Av?hpOM3;}$nkGgYE{Z`CV)FCuY
zF{%n5Vj2-Z?3mL~tdqr4@t7XlYc;&_88g>=KUu9AIxAU9Eq)6MGWGaNDWtZ6l=_;v
zSgCR_jOQg3I*a*h%wxjVS#TXhfy$aE+uK!yU+-Lwba}vg4$v(ptg_-2^#oWP4O%F-
z@OPYIDX+m;K$z(Hu5u^R@F%=YRkTo0#DzFzx#l;3juo$wC4!)^uxk(lLi`%y5rfIr
zkS3V&<8){$ws%}*1_2K^8IZD3ORsjjv5?q?ZQ-y(|CV-g8N#^`KW38b0TT?cpCrmz
zPkkopK39D_Sh~EiE@GkJgHrK22@Xb(9%G}XC+u2ri_>|c5aYFs-cJtbPR;z@xpb!z
z2ERhS?uT^r*P3b7wSup=ABqrg6kFU$^$W%yAk@g4BV<o&d54Qul9NOi)2Qe*g1Mvo
z$>6u6I_>b^Wm)e+U*|0?t9ue#<hz{OimI-Yufs1G@CE>xaT6^P*m7#SGue%(*5>m;
zzKdLlay+~mKE-4Aecp?|crAuQd9@YzbtgVzJ?q1V;S1zNgx!sIhl{3f?FjhL2HSU`
zOa7K?`}d*)puK_thK>;w*s$SeMy#3;pkX!NU%a0R*x!Jo*v`EWgr&E9OsP*Jz8(aR
z$;q2M{NxU7cwc$@*7ptS+|mu5v~A3{(ACk~L4VZCUd}%&BX|2c(_I*F;b&(Uk1c0<
zU2*%o#J7$unLjdrH)qCil<z>W1FST15b!S!T>8cvSEv=U$I&x{INJbi;=LbuqO3SY
z5rnTx(uh{2c_y)l{aq_QR5`<aU~e+-@iJ%J`FRE!ntQ@Kg7WA(T9$WuZ=(8w*T4k6
z_V*Hx)$a%Vxs>F<$IoZy#U|^8^*uu}o%m-5K68}jg?9yoWes+iJH4`=>vxd2)Oq5?
zfQ0F|sWI3kUmebl%2~EqsFyh+8#zcE)v*R9?v8)N69aI{=|{zrXm-FR2K=3swWc6A
z?SYac1zA-ABc~k)DVFL$L2}H1{F3nEWxu)BMfW%$^@Nn&NzG_V9`tvk-!mdv`}nG&
zr4IS`>aW8;G(zYq=nYg5T;}bS<8gbe>>pI{-caY+mX(hQtb2AwM_OIt`+O?QS&fZ=
z63i24i<jxr%;IhhZfsvR1hvqYMNe;@Q=F<C?U=X<68Ka^wilc(MUS!j;sgTiKjPL9
z@S5K;*du~A0-6Lhd12P5Oq$W)BYo9O`&M&OjKQ`#<^B@*C_fgn>)s_bKUC7SIXqE1
zUh^)iDtChK4Ex#FMP3a_^Bf#<#D%J+rcJKLK}QUVCikoQ*PTT;d6ISq(-dQ2(Bp0C
z7d_d*fn*Qq)&ba7;f?8}93j>>hb>(6A(54>rj}kG%OBP^2mUWiVy-Ie>~ey})NG=H
zG1qIxEUIi90rv6cm31`Zqr7-q<@#U54$-Pw5YWDNo18B8cm=QRCyVg?O#9`|`hGnV
zo{V#RLPx52-9F4fHFY7=rtNz3ygubX_{#c(;pcM=Op>%dDU_OQo9+%hM~?LfOCr|f
zQmX_0U6|^50UuvajLXG6I~c(tP~z3F8+YK`<#Ip*OX~WJBLF&!jR0!=0H|$rg6wES
zR5wq(NwAJ&yOFn^9eS>5&;qyn8S`L@!!!^Z#t;Y@r=4vsNUDSHJ(FzEn9A3()H*s#
z#8{BtSYPs4w>2zldLFD)<}e21+lasWEd9>6Pxq9ZQsb1j{P^1j0g$Iw!haT2E?U5-
zTbD=dnAhq}umfGA4kc&Syv`GU9*2J3#weJ1J1$X`5hC^%`nv8#0?tvuxnf}@1_FwY
z)~vALwI7{t%wP>37ntU#(#B>up;1=DAe6$H46I97-imjx$4t1f$hi}SrwG~E*`Ax$
zG^t)*hN9tK7on?H@o#Lp_C|u#RmrLw>a3+<ifGnMELK?Y#>O?bskI4s2-85D3pVdL
z%yiWB3*oQu=qzH1uPKPH$vbWDYycB+bSSYLTfg>tOYc8yT%xnQjV`JJwg;JGpn2{=
zJUS!!%NC}}Q_tG_rG>S(p!qre)ipEhAU4>Osj`Zedq*qdn}mHmeEH`fY}G3pR&Mgm
zCAYG;f$XG&FeblpUH2AeYWmU`GJ_bkt<FeWUU&F)WLvgc@SLoAYjRhvx<03*U+g!x
z(v!|kQiHSi!!TEu+VM8Xkw?{B!6ou&1kC;8*U2CmgvPEv8J@k?Dgw@a^JicR8=U{B
zJWKz7*gB|J=+G!}iQiRT@ryr?+Mx#Z7i=tPpTb1sf9h6D4SIjQ@dCmZDT!cdOy8o8
z{mGg~A%B1|Af)H-5Cju<_T)#Vx_k(pI!EuM|Kw&+URRDd9tK_%!#f6wD$-UnE5BB|
zjQd>08PY}4;`mjTwm8Lfe4*(^p++CzX>+0sE~D&68fVeKJfkC&!Axr)F3ug#M7k6h
z@C&zl9vh6wZ)h`5WOM-clO)WV#)8HTNpNl%nX#rghtS@`@qCZwwQzUIj=>u^-W}J&
zN|72rDJzUd-d&tKyudYdu4z7>5M->mt|{H<#@&@y6?b**#i~T@ZRxHjfsexngqXb7
zsB)(EYC_hYWN!_#uIpmv-qICqW^Mj${n#QcdEbdSlMNMjl1DGLPP)vf&a965CG-6U
zV%cFiWf?w)Y=p4vqxa*^+0~x|oYBC!k|xu~Z$G%~O5xaXH^w_AVAvN!Ct{wQTl=Ur
z^f{W!n@e#vkv6v6#Jeu+u2Rml+zhjcjDbmx92j!`#*c998T>oBDpc!%T0_bH{ppqv
zF`A5e*{@9rs}HfUXH<iASB&C^dIji<s2^OBGo0JGnC@@%!)|1M!YsRyfje4k%G`M6
zNSdIDcnMplzmRR!p4{^b&l#-IqCDPnQ&xWeH<k-Y|K3)l68T{zH-IW|Afgz5g=1mb
z9`7AG9(=)dDKZFR@bhPk?NElog}cv%%_HDvNg?)iQmN3-3U+i2jiw3blg<B(LRxe-
zu>R#Plsq(JYy{&68uz79$GMLBzLQJpx<!h{F*c%#fymgoQEjp1&hx_jN3P^4`Ia^d
z4ydbB>ci1H$5-1g5G6;Wyebjb?K93e27Z#@W8h<G+XmDpHOr~p^RrD3^lO7EeZS?Y
zEd#Z)@3roP@eQmwhR9+XvZ}T@p1eLEB&%|3f90}<Q^8KWVO7vvSh|-6wmQd>u4M)b
zZlPrNQUoG%3KEvIIy*3Pi;?6(*f_z2n5*V#`s-qgSCn7zkuUw`ngqNghJ87x)g7mZ
z`;kJHI5l78Zj=JYH>pdTe|y8_jl$W$M)gz=oHuN^6)#}8vO4O|S<*dbXb`*Sxqj1y
z|2SY=fVfRYV1_)ex{k$%I#ah<IIMbqe!`Qx|4qRt{%aXN5S{wW?sr2jFVD}KjI^&b
zmZLc$2>(KK+Y^t8LpdTO6Y}}SIA~9KkN0c{Gu19gwTgZkzDB094<XTh<vo+KN!2p|
zSa$LL$LtkW+YiYLz#D!pD;*g0NnzuLfBLNz+Q6ujy+nu_rj<<#^DS^5%*?T{V}IdU
z`NMJ7AtOW{4&#6AhBM?`D;$Zr@2mJOT7hHP;iQ|+Q@cZ9@9WXlG>vE+f~k>)!e5LN
zc(>L@T{q?!`oj;ewl6~5OCNusy_|ALttx@-=<b_kqA={+OLYJ20=4S}8>o!eyqVbq
zEXza_Vh6f|rx_Gz6qCUD&D>XmvFYB%wWgXuIh!b#D&&`+bE+7paD6IY?a~W2OVeR=
zYlGjehKO`m%%yJC3-5j5R)5TTp+B!qrrv2>g^99tg@~DZbC5QTI5uU<T_+Ccr6^xA
z@=S9@Dg|BA6FK6v8XCNFXK4TZ&-r1oNztekem>_`B8Z+Q(dQsiNOCffZ5%M@Ct4}}
zhXkXBf#hu^%2%->nf>AMAndhG!t*!>q=`ornBpW>p}Z-LCzzyGojSe5^M9FwuF@BR
zUWz~Q^@R0FVCZW?Rs;9Y`-5aBb5P4*A0fd3r-Ph&<Gokk+xOfC&3LYS6S>znuLVK0
zhP8X^o)vx#5{X*LBeQtBe@QD<RrE!4DUPml7X<&D-da;-K2QjPe3*$pS-Y%3(?i9q
ziq@HWP`~%&_}A68o$|GZ@4seMIg6hEsSz&{%$fQ3q4TqgP_sA{^-r%NQhO(~KD|V*
z+TAj}G$QFkrf6OGpcUTVPVe$3VOi!5UR^E%nyT>hM_=q+I#9_cl^u%obrBuiD`($`
z$bSfWU)$oIfenw)7>92uW7H_{<}Vi-<hppHP{?wli{AU<xP-U7YE{$oYT8}xRbk#&
zGq3lAK)0=1zcjZi24dH$Cbs6uQjF7}+3i9N`_?h5@Dt>iW8Ph{!{x6XJR5sI|L#h`
zF$Ai*XWlrDFO|PXPhjRXLC&V_RE>Y8sx$+Y_Hy|}<guDDyK1>DFo~bxR`RW3$E3;Q
z@}$noR@|=z<Q?E=4b~M_33*tu=C{^JXEJ&8_#$O4CG3TC_Pdj5u}KqPi$D5){$9}{
zZ!|TcZ2Jcw{e&MHve%bKsB(U$`WqVkOwTQnx~dcWC?(N3<)-S|+r7ic`im3_3zc@t
zIKc-KN%3wCFow`8z)GR%PUm%RA8IAQL)W2K8F-f70paP>^P%Q8;5&Lgt-cMjjAD5n
zZPLyWBjQy<EnmCSJB?QgrW7($gERg#68A-22M5<3$FF&FAX>-Jw!&0{%i5LPea&>#
zlHV%1nY~)!pSF{xmIh;`mrho{4QomXk+kLAvZDH#T55u*wQ7b}s5pM>6RGLdS9tuG
zM{Y*CGzF2cp)|_8Ush)3y`inai_BXA$qxjZpZ6JVFOI`oZ4AM|$6l2!yhQxIK-5KC
z%dwPFMiAlK^_L2W&0Hr?`xD$otIN#++86OZLdSdGyU?0T;={c>HfjC@U+}VTZ6*U7
zwdVSvJ|npU_K35=SR}>Qbd^e<LntRP84-2|+o5+}D98AvVZCflVya(j72t2!LWE(0
zS#(vnsarrGK9|$F0O5naM@)8OduA066b$XfbwbpEAALvl4#oogb55!L@rgrfRe6g}
z!O9AnhW3x%)Zx4@9vIWkW(ZLH)iNKvIT;UFF>yKMWi5$9NRAzfjmH)|mYb6pbohi8
zSsVRa-DrlcH&gB(bMT>@nZ3g@7@(Q1t92AIXr=YC2B|JTm@3t3`r9-2<mqx>o>#wI
znG90or+GaWx1`pKkH#N~JVg2_8y~v~MxF|7mfuQFD#zQmU&imsPXFlgan-spL(jRg
z{_uTq)Q{N<tCnhJJ@Vrr#9z~&)&==^XkpnJ_-$rtQg9DdgyZ;J4=8iwY3m_QJht-8
zkYPBAyd{`q^2HD&-1H7ES99{?^4`q07ykGn`3grFGUkslXA$9m<=Ge$27`PF<89k{
z+L!(plduXDBAx<d%GPZb6QiGsS9M4L?w&vIZ=5(RFMDf$;j%`m(2jB958F~MburIh
zE4OeRe_5}ey4&2n=s?b2APs3nww&|EUDTlOxf=K0DKspj2#dLef1aV|3L6TWpUh)J
zHG*H*+YvhH;@PXuaeoL68IUUF3IO|X!d$x=rLgr|efd^&icm~WNgz#XjAGHB)^q_*
z`@VT75=)u&EAMyQiMV3N!0k`Kx|r~pxcCTM5ctqeu#YV`bV?ei<L-1RfN{*vSYCP8
z_RthW#qA8S-RutaO3n+S3AYhR(D9kypy^WImn`NBOJO{rc`zCDx+BNSu(;y)D3m*J
zOR2|L=nA8uZ7;IfQD~VbRX7=dL~(R6bCt83RxizPyG{1JvbfDXCXp%COg93&`fF-T
zqOpi7uU(g?JH#(2RzAbq)pl<AptK)kt@4&qdORSJD(qKrK=kpKyDoZOa#5goTfyCN
zm30%KdO02S#W~|IJWv$hKwR=3DA9GQkOJ91BHRdk;e0Un(;0%iPL!d=KZJ?6X6cK`
z#Tnz|vG#WYiAsQL%X-nQG<!462Uat7V98)ZyAdX7$w9T-)qQ(xX%eNBOyb$RC)u5~
z_H{xZVVipG&6{8Mwo;r2CwCh!5T&>JcD!B1XFCF|8~E`~3qw(doi&sU7{j+Yr__$(
z+RdUKhL_6=WpfO}cVCEVwjZCg<_9Zf3YO=f_W2$+9fNwS2ddPw#|G|MhdaLm(1uAw
z4J30*l|4jPW)au!+oTYtn3QDGeM2b6kjmZ4+$q#)Xps8Em@X?*166p){ay_E@G9GT
z=fC26;F<VMt}wQRAIw`Mnv!7EsMWQ(p2@;s%TGQW9+LspbtexK*SX$lC5y1<q%q@Q
z%S>-Ir&QFJ=bHdUL-f4uA%q$Lr*O=w*ko~$Tb?nCNnvzxig04C1$B#=>~K~0lFp!Y
z_!AmtrLV6dXc89}HC!=4%UKOxV}5%H4$HN!FI70A$vM2>y|s{kj`I4`f%;@BFHvx7
z=R9wjhxbX&ULz8Sa`=^&?USan%JkFHNc)5Lp&CA1i1{WC<vUxHZfb=@p<wsmKz6??
z7#~$d?oh>I2UyqV?&B8Gs<^U(8bP(p74G^y89I&8<dezP7N_yZt$2Ew>bb5J@%Rw;
zfOmv#ka;bF4##|0OkU#*Ds9_#eE*fIXIvI<S&eG7qi9Pm%{M6)4U*-_W3m<#k<TzA
zn&4WE{>-@M?ykYi_XSY6Y?NrCLsz$IT@vqL-DYZ_Ls{P2#c`B=!`C18Fa3}@qV(<}
zlPYfeI?Jf>Z&NBIj>~hy?~BU2Ea!c%$^KpghYGbCQXY~O+vbRKQ<U5m%g+%`x9Oq}
zp&KOt^U1!aZY(};Z_`zvrIEdySkX!3!2yokR_gQ9EbP07FzOm=X!z887j;tJdndtL
zs-vC$r1(y@p|%gYMQ4|?8x8LWWQPg#_oW%NvX@EIg)3kG_`=t-Xx@zLrW+?O_WU^E
z&k1%6qo4*DH3?LEwn#Z*7?3JZ4=?R_bBXrK?OZ}`^q(pP_8;VBP9LSC>nI#SGDDy;
zP`cjfuqy&mb%xTID=w52R@fNHAgPk(I<_x2oGQSXlo~6lCvr}B>YVV``*=>fuDw<O
zGAo9;Tz0(!mhn5kRPrRZ`O2_IW4I7FkHj-AZG_zseAkgEe~_!nktNKcS$xLaj@mD@
z1spiG8M)!T^)}6{whL4l=R4Nx_Bl>=tj+CXR+YbuP>JRXvlE6w_9x0ud1Q>9$}(lP
zuC5UJg-1CE`xO2HGNqJnE`v!e>Wn!T)vnvO)qtWNJDtDkK$r?FU^6KCA?~bQ8dsSi
zqh=QTp&RivJp4wv2n|du-DO1kuT$W$Lhghw*C2jpYJ_Ho*Ln3@hVtW0KJ%x5Yn`NN
zUYjs{KCh@ht4NCatmxuo=Iwgfb#)#l=BO^JoGP&v$!o$Gg2?Ixe&^P=h$0tyY4(w}
zm61mX=1SLI9Ca?wG&Z#@EVe`uuU3K+Pmhkm^!#%S+!O)>Ju2Wo2*#b;Vm^A27W_^_
zdjpJ;_Pk{zTWOmJb50M7f~j0<b(zt4VS6wAw}Rv+y^Fn<HIt6z{A7&m!)@;gJH6lc
za-ESKkN-YV)*)^4(Y9b-fXIq1J_zXor}OsTm>e8>_qfxAK|EMFK-EM>v<u$9>bb;!
zXhsk)_-Ua03jdHwP6tvd_F6|d7G@I;rUYr7{SvETM#1KJ^So8k#C~fB(NDctVGgTf
z{4uB59li%<M=${YX?v4|d1HiQ084}c|65)E9WLcqx#VWJv!Jm!i)dOyG>vXmOG`yf
zpr9|r`pbURyOq|Es0_{S;Tf^tIvm&VPJkq6^j*X2rVC%lFL};5PM9`hd#35`$=^jy
zsVpL5e+XggF>^_}=rvT!n?D|79|6TJJwWp^mYYF1>gT#w+i=U`qb_$#t(8hDpScbZ
zww;$^3SqQ?<9QGBB9s!odc$JZN2A*Q%pF*gfWo7UxLsNGsu#9GT^OeW21VxFj`M9_
z<6otCH$j#2rkDA+G21Z{=4Z+}&3V$-c>qT6J7%VBrz4nCKjSV6uuoE>PV#QBD%Buq
zB@}`$?~MD(wRm5D)Zp{1LqIP@&@3cCyq3S(Vjuw;acFkTw)C+0b+(znpb6W!(NYrv
zjr<aUzaL}!_T-nkjQ@DBCItWB`}K6cE7{d0+*-tW;-qQ&)Oq0ulwP*sI*f<U$-j%*
z@&Z5H;g~k=T!dsa^AG;K_d<x5g(Wv6ISo^y3uA&>sbAJro|CyQ$19Zc*hzJ!-_q|Q
zG}S;wxhfu-TD;jue1tGbztiic$dP}V2okN=!wUM4Fz8s~^Zh=)tbiX}*+vceRx`rk
zz~S~IlQREce7@;;e!?Bh5!4iA`)a+5y7<XNk}XPdkG~!|uDyr5si0$^yo}^B7&~tx
z@WOQ_YKpCoTB4H)|E@aKupFJw{*4j4!5v7z(e0fK3g$|@C_ne1qYP8nsx?~uWUK$!
zxtG_*7q4vV1w<sf^u8#IjkUh_a222F)%)wncR?jFV~gDu@1QW#=5{q!j+$dX)j9Sz
zZ0aVPop;4Mc{|=r`i0oFhr+*l^&Uy7#of-kqVq+mJG@=Nt4eJ0tNWy*yQ8wYj8VMs
z85w;+MSWe*H;Ue|dzL9TCH1ocBxMqEddzewij%L6LsDASr@K`QbXV4Ir|1dm*P8qs
zO&9*5s1s}*8<^D=r`ALjc*bWIP{VHQ5a{jX8xFjkj@;JKmw4XT^vuRw#dB%pdi_~9
zy!z~p>m@o+ZS3iBlH>dzsaOt_M`kc6|JF%Mz=a0iy+6~cewAf!1Uz1j1cQ6uYRQbp
z6Or7cD`Ru=#uk$II0|iKw@lP&sgbVSo7o=Rpy*V`i_l2w9$o|=+xshY;Y6VY`+k-S
ztpim|a44CC4z=4Zi!Y(oS~R<YYyR@Os{!xW)hgbHL!4D!X$z-2aW>~-&5?6kGT=Lj
zPb$jZ@)7o<e7uB0FpvFcbJJxm$nh9wV|~g5Q92{J_|xEFT?yCTuP;EB7?1_?;TIHS
z50Yk_v0JA!Dfvz7VqUQ<e!R<do<K`H$t2k_9Goqbsw#h#L$2dEn8fQu6aYss#OavB
zg{yX$IvVIDX1R9NF`VlAkAur?7X}{kq`gx_>OSbWr`5NsrrC{L2Sq1qvRKF5sc>Ui
zD(GG5Vfy^M9(RDNn&&xW;~b&=<Ro6Jv)xUDejz3&PZLphBR6F0Ey5UDVp9SaT!n`F
zgVq;J=2@kzfc!$~;{!ys$64WSrtrulZx7elhddwdU2Jv^U+9)Nd|CwtdRGi4Rn@Zp
zP>hQFX>o1Oux(&yl<kPqIGKg%+6#W}inTjpKSzoLN)%rF=jhD`!#P`mg3H~xw!d9e
z=tDSsdtRLNgO>txfShkt-ZNqG*OxS}n}5wHnA2&^cYCQS63H05Z1uD;X}VOfV9yEK
z+}qvvGAh%ZsaVK*C@E6Z`8h0fhc#0mq#-YxPQ&7ZfRxkQKe{NPSv(lh(*|@;E2!b+
zUc&E}$Mqd*(@ZR%wRPrjW(q&kdOHqhBdkn0GZ6DF)YFGaZt8Ee#(041P*PL`=lpij
zN|1R&g~^JP%F#Y+Q<$>AIZa)<;#h{SS9w~^#{dG4){A*lJMPud6o=>e#o-Wy)H2$O
zSs%-sice*<d_2AQP&4e6O0kPCA0m0eXipNUt<?r?KBNxo@#TNJ9J3U#kBRI1x^S7x
z%6Tfdijz1WH{+xFk{6edK@6eqmPRSl;pyuKWxfX}0udI!MS^`~X!FclCvB4_JFYt?
zbS)e|TBG3t0)Z92k|;Ue&oW03BwJllB4s{@)1R#10+eN*#p3HCX?)GzPPIOYS-~RZ
zs%6HUz71x07%V@F->Mm;)qYOT8Aa3g$C>&4@^>{}9*z^<M5fQWNL8E;v!eNlmNlxE
zoRzdRG_XIz%Ggc?750S}TTFj^8`mR!`6$1|<V`PgmA_P%*)+C{W;%kAx+N^Jat3tL
zg8e~AzGP|DT88$nY?zC~1%>}^ni&`G&~a`~x`n}B8)|t$>Ts*tCero!oCV?$@<-h^
z%P#ZS5H2?348exJ#%s0SFg4~Z!u3||f8Dynf5E5^N)BR>kDJZy=<{OIG{keB#EY1`
z&MckFf^u{2Ab6NEP3iS*TB0oceB;j=s>sq7S{ta1F~Z(P`ZQ|Ruk&9%_gu(_bW`jT
z9X{@_<KA7dn^#k7s<Ml20I5(5zY_ax)D6$n@E0zz!=Ea?sxI0a_z(bnFQ^FN2Q|mQ
zAj9KxKM`_Q9TLF<f~j9lgzmWzfpj>7KX)l}JR_8`q<=#stROcgdbHZw3@vbIL*&#Q
zUB)*MjH1WvsH0iat&f7SyV$o@N>`ZwiRYa<Jw*XiQ=-w*B5mQe5s+`PTS!=1QUrx`
z%n6uz${GD(Rr^JF4IIJ#5pbH!sPM^g>qlM6lzY5mFb%qV_wn=SXy+izEaS`AH{a+^
z6u;wuj0g@E_nZI3z229U&|;)2k@P+y7={LROL_)K+f&amcPw&$3W&I(=x@~jyv*z)
z^mnHot9<1_ah+ZmeK$F`nZvq7_|qcF9-7xmHoy51h*{iAJ)$h-lUrm@?+%25Xn_1O
z!!JYBL|M;UIP1S76)p(K1fkePPO+%v66wj2b7+=O5JSp|iqr>hAnjqx^dBfGohID?
z!&J6B8?s^vxPH&z{>>+o+HK-gZruAPX9AeubYy0kGl6Vq;=vhU6Q`g6C=E&irGYK>
zm{a)A=cj*@8{HQ{*eSrr|3iJ>1Zq!HQSu<})Svk85PzQr1%1WNKS<<+9eE6$+V=mB
z<k&~Rl|Z%@vJxvxpyK;rbef*=UqieYxQ+tKrVB}#zK+s@3_zRj^#TW8|9jliemBeb
zp}-zu57uNpR)T+Np%5@llwzn7SZS=l_w*_M9Z!CImjHptR1N2pK{gu%l0p9Y0|BBm
z_E)j{e|-gbUG!LF1EKRs)@nnHh6h>yJx{dW&qup#AO|xni0e47G`V~I?dMu;7kQq>
z^1%|>NUdUbc(olP0IbOWY>>)-Hi#v&0D+Ul%~*2fuc1sp1n2-v_3v11(7dkYU*V{)
zEyIDvQ0l6|6%eMbxj$K5(_qPaRqO9GJ4^D!2a0#dUrJWQKM2i|EcX;*4@z-1{`ka8
z3};|@TR_Yfa=?hY4-gHjKM#=!e*Y`CAnYr316}#}x#Z4batp8oZ>GxABjmu)(^o|;
zSAoeSR1&C?%D7$n`6zT5N}vh*&0i0Lf&UBMz$@3})A4-reI33Qa49{p@#81G|MlS#
z9)^p$q`CA2n+V9S|M}qkM=P=71v*^EIsTe{WcL4nEn49NKST{(Cz!O43Pw+rjtc(Q
z#sjaJHBT^Hpq4#Kj`wZ^O}1gtS>nU1?m$DM{~oW_Pp;8%kDGu^Kw6My=rn^mweXF7
zd+R)x)6=B?4!sg~2r|BKKl2+kDMIQro1Vwgw37ONeWNeUf<Uc+LP+x7;i9Mguj~GQ
zeAP2s1!;mkA)V+96q3rEEznSj=nkj9LUnvr`73s_67+a0X$BHPw^D}NEYE+F_Y){b
zwB;!jlxv-L_i;yvvFujM+`QM5#O1NMl<0f0|4$rBele10>#Pi9YCpj$p;-}NzekB(
zl3)a>E9CrvC|<h@yu74LRyrXeph)JBxt6c6m5*3pn#nw~E7yFy6W0fL2FnL8*E?#7
zCQTEbVrB?hT{Fr5A5-l{jl)_=LL`%|eeds^uY=LT6Y@QyjvUI?RdHF@v=E!yKroaw
zFoJT7<PX{~P4Z^@b)4-;esb)T?DfO-mAy)ynY4u?Ve`$9=805{-lY-%3I7L;fn@h2
za754s=?voS*<3OI@D_s16CO%EIkYQs@1(hfaLm&+7ipW1=$k(<zuQ0PUWFtT=+x=t
z-3==YGj7~*sF6h~Rof#C1Won-x$7x6Jat<Gcst6BmRV>dJi|lYiPjR#$jxjjiT2b@
zoYv3y4hoSW>Rp@sj!S+!|1NoenDn)*yKY7Pd*#TW*qko~8NAOSICj4ae4OQcgFHDp
zVjAK$E&j`r?GuZX!f2E!X7XJ${05^V|9e*Js)J&I$!lk&$k$EdFJErIoi^O`${cF`
zT)V@~z4ac6&@nU=o%ZV0=<Aoxp~|j(cpul9iJXavtiup^WALELlcfRj-qxRKmHlZ^
z`Ct*R-!_mJ24ruV!xr8Jpx#dY(*c9bo}ECGL-eZ&{q3hfANJ6{bwv!?k8N&}^p)b3
zZS%GQPW)580>!a|O31Udb(jl`753TBY?HV4xjW}<vr&;RDO1-^?+BEgtg=?wfWjX%
zUheVI-_YOE21-9ythv`*?P(QWOT`;*?Y}}Vwdq%ExT~c(d!^h6oh7&dA$Jwt+LQNO
zi0oPO`hAg!>;+wYpz6NacaUGOg|+OPmc#avsaWCtONaMUT%Y~THo#u)p{xr~>LZSU
zny=MzS5IfmG_L;QK_LrMRwwOq{M0trQjW{X<ZCmJO&(bITcmi8?B!S|*{^Mf75_!7
zL#9kVt}-YYTmN0FGz#*8cdd(gOae_y2r0+%jtbGMeZ0kXO7Csr8|&T<G*ZP6`!7%w
zUnd^VQU6Q^iu$1*J{Ipu6RUZixPTXMF#7J*Ca>R|F%j_6@Y00++U(t4^RbgCQqy4J
zCvqvhr`GBVw&F$`v#DA0{SG&uliwp=I^QMxqmFjwZ6N94$8$_uNVRUD+6%UN>;d=Q
zpQp(CGg_HTUvy_p*IFj-BLsWwk9H`z*|x!mx9gn7*+(iZbu&}<MQ+jm<Su?Z^=KyT
zNz-g2l_cBghEfnOfdyWW7zDlc<PbhIO$pd5p?J*LS=AS(9!oqd8ZLRUNBJ<-Z~G9b
zm+3yb(=z%XcyB_dHjm1Hr8sP0#B0BN!EwZMNck!0_IIv+<eLKI^0_OL|Ev&zI8YF{
z><|L4?iW5ZPO&D9*qhczefb`iDHq{sbb~Yi?a2KrK9#YBtSfwbQfx~7I?i3|%q`C1
zgijRG(ay=<=g`_x!}vE+!p8d57ayL?<t^f286pFh;FkX^96%zVvRC>is8~(=?6g_4
zbWiH1jqr-C)mP{NW_c3*`>SFR?U~uUaFL#(SDfx7vr(1xhNjXy$#*C3FFefV`$)3b
z-}x&0aCA~ae>5O1?k+N-Q|z=N-<MD#WoI40d;R~xd-%j;wj5S1XJPhLRQNhj)a&6(
zpBrr+k1L-JicVxMY`n4_a7~#W84A-a50>n4UGp%0KmD%Rfp%TA_3|HohZGKv!-!7=
zpk$!Y1KmK68*>2EF8Uq?m8*I4ov3A?9FqT=CP0>>iiPTDHyU4Y-j;da$MndmwARTw
zWlmTncfsxZjo)t`U)^XB@9{RHeifC9|2_?Ud40+x{BEba?RQv&=%cwPQ|Svf3c9B=
zyi_jNYlIVrLI3Y5fZ|sFcMZ!&_Qrm4JuiIuX+aq0Q%2<$llDgwl>e(JrHpU?ItXNN
z8^9J+x+OsUAglS`!j&RH(miIjixgwOD?Mmo*G?ZBpex@gVGa42{c~mR2cR74!YP5|
zUy=k<{Rd#nnEgu=Y0f$+Ui!Jl)oMK&V$4Tb`hK6uVL_q&%krd)C_+VO*=@hjl&(}B
zlP${>iAP!WB0*Ok>Z5PyZx~V*yEX4k@fr9D_avn@xmWSCf)4lrMhF6lV3$p)564E<
ztAy3es7duB)p;U^J}C*iSM5J6=Fo%G{*vp>M(LY{dS=8p&KMUn>;cu99;e8N1U1)D
zp0#a;w2W#1)xGui-rFd=*568Y2!VsHoe~I2f2a}UqtHVN{wrPabSn5e$rf}F>N5D3
z<TfxM&{S3$;*~OH>azcI&F1-m)%W=J#9e)I_QJgsqivEU2+MN{Owel(u!MTW5yod=
z+*ghDfy%{T?xWcZISI@Ze=5%az7^S(J~`HJ#18>ohz8g!Xc<%D%MZff@%-BSGJE*E
zlBozrFB6x&YsjZS^~Eyort-HZYioP9aVpn9aQ0IO3|wypfu4PpTiUm`s0ZqU`X?g1
zEyJC_DE*57Bf+Y;EjBQN@^xKRQ2g)v81VZwPLjGbp!o39*@STPAb6un^1{R`8XCpM
zD-XZ^Hb}=u1HD30f$f7+g#bk>->tq4De$pK={4n(8~1veLX(2;Imi$P;B$KaYX(8o
zpey{MYkP^W;y^9NTU)eq1hR0R-UDi!yjTQzSO2l`R;O^Dp`Yh-0yF3zJbAVO0xkVn
z6b>lr`%cpT?sZV0kh;(*%x}QCGMBcHE}SGhk>WfTJ;x6^MWX*J#^icS0ZlKW!*3So
Q8VCaE>lojx(1wTqAMt09WB>pF

literal 106836
zcmeFY`9D-|_&<JT29tfuzGvUV6tb@wYh=xueaV(AAv4lK64`g9Q4!hqZ7NCGBC<0P
zvSpB+Ip5>;et$lH!sn;22aj`paqjEB?rV8o&+EDqu9;q?rQ)Ol0Du;4plbmD6woUf
z_T1S+`NN}J=t1FXbX6C6LO+K(tr+M*8DwA=0suUZ&;DS^i(4K5zz?8xwJam@*Jr|~
zg*<ZpoI0C$o|7kop#@XZ`CnkqU3hXXRTmZlw|3lrd-dLomGl03tlIL7-rTFdt0JL#
zdLAr{y6Hksuvj6+$N7)M&KdB!%-<t5wG^zb#uUCOG94W~J&sX7%_%LT8|zc}VPiG3
zf6tL0ohtPI|Ng(%fKNoZU6es8ZtEr~OUUOW8x%%!DsT{mh{+yEcsVkfqKVTMt_nXP
zQ=-ZqU5e^Mn!-qu``*jHMG)%r(H!GpH_Ykx)ZVy>y4Z0gFPjmMOk;RwVwQ9xw*qxi
z>v9h@vcbF@P}t{A4j6aXnRDozsy}Y63Ez`ae%7AQl_C-`jgCSuBuYEAI&oD-IsQGY
z%=282iymB2%1r>(caoECv^D%Fx~f>D{M_kuDn+QmlSID+>Mvo?oc>Fo3n|c@l!as5
z&+5whGtJ~?TeKv5pn`pDzn*uiZeOp?l;L^-zzs<Ct97z{ZQ<Kgb_5d#QiByWl1(*2
zF0>yt3ZQ-6Ns3?|Bq6l-e(>zOI$5yg4DL^$`!3sIT37jl7D4ZmVhz{(`#o4m-@!wj
zUza{JV?(hE*g;tMP=5Xu^v~0X!GHq$<8>oD$K~o#LCLoFc|}*>oQZq_iF`=``hGu{
zii+ujzQ5BcM4QuC)#5q%3Qt4$foH~35v&Q<%X!-|yym@|GR3BoC3yRsgWfQc`NSQy
z`i-Z1tBukKy1!qGE>8EY%5EjsDa8c;J!lk-NRE_`U{Xr?Y6djCQPv92D?i>UoQW*n
zn{duvT5CDHd*>+UG+r;Y?Ar-nA6Os<8d+TW?-UfEU6#FVtUc=8up_*wGnMH?hKX>#
zs=Go;qMXHaPHy%4O-+zh2JR}=5uK>LRkf4?I57h!#~p2-w-!Ra$od6G^Ac@04xXb#
z7v?GPuXPbQM+^BMcU1mI6Xnm3*WF$pPMTtPXQDkT-Jatl6u`#|IVUQ!Cv$x#vglC{
z68#s2!|=BM48rXT=QU2N9juOsQUi0)j8LT-s}f^VMHOh$h5ppud@swC$lHntxjiLG
z{YAkgEFfx0XzTo<8SCF$!Ky`gKL6koG8ad}ZNeRwR&n*s7y@y6&)8L8Pw@y+#<<e{
z@tatXF8bU!QxorjNU&>D-Df|Q)4O+ea+p{4<{k-I6`YyqlpAD4LoR2gwFkLC>Bnsa
zfJv^V&2(d&4s|&Pt(#h8B($K&@~r)_KF8L>i2;j>0_x3O`&{JIPf68f#KtM(8&qC*
z4lTw?spDe!t=`FwQcvJ~zm#ex+pAs3JXXdv_9rB%t4H`SFHSvGL(@)Gk9Knkx-Out
zJR$oF|Mov>`QJAdYgL$-NKN4KGc~P|bA88de-Exwe`zS+nDctV{?lW3yX9!P&hV`t
zgDC&*m%rEB0#!VLH^0@7(ZvBb9Vt#)!q&mwjeRoynAyWY4A}vP+>~G^NH<wjAvZes
zimzzGM=I6Q12RR-n?5kr(((Vk-8^K*TK9hK$XAImO<ZVPd<O1+tX}Xv9?Sl{;{~@F
zI#_JCKmHZ&p&sP)HzbQufpMFzlqV{CG%3XeWeWV{C$&}Z`qgh8oqYY3<*lkLRcOn)
zrra<*F!*Y$Xu{-w7wtUkthE&?{q~<TV7#ggEJ_wTQCam$g&ME}bmBr<0pM`X0rhzq
zR7rB4<h!Lvl<_xPc)R6>8r#h?lDcV(he<m*?=oX&s49#8MtL+}m1J9*-Hf{4W}~8!
z?1b@bgg%*(bLPz8fa`z8cjkTtpNr)z9pJ*H6RIv61eG+^<r%#5QbQ+obbYd%me4aZ
zKoOQVx_5ft_I-rWujW<+wQncL6y*OmIP6z**65yDdbYkf>A}m4qNSM5Ub4X)Xje9n
zi?4KFK>K=<`0hbYW%2wF^DhUy6!s(uEp#c$ZexsEM5#kC=yt%dK4H8=|2(Z@3>kOk
z)IKGl$h+X$_Sfjg;<HnX^PaUKotUuu3qKC;{+m}2jdDD5nqVlhLE*AH`Ox_+D!HEC
zfFIm-U9o(}xV2mz>Ej5`1O_*yP8vM$In9F(T6fZm`rdhS2L~+64bfn4I>5)1DyWLu
zZ~ls)I8hHjNtAwOw!b{I;bn?TgS5Ze2Qn*wX8(DkC#fj=5aYUpNfNdJ46aAXR^uv_
zBCjwj6^$CL0Z#O)Bi&I02=&4#b>x9#-WE3*ci|L?i@@Y!nz;P_LMmR^IRCpdd^%-n
z7C5+`S8ePWLU3%}2>d=f*Z*QhC;#G4kVMMJ{;-AG3c12NF1~*f$l^5jb?{Z*VE+1M
zJtPgfA_r7)BL#XwPPlpz4DbG*FmdmDG6#_Nvti$F>qN(!`?I5U)SPwld@6pIs`3;+
zf@#xeEhbKW<*gv)>z(?-E%QS!+ASZ-djXm!ZHbuh!G`g%G7hsh|Hb;6&M_x&Dg`HT
z5sJks9%l>5|IgdlcrZB@)jakyroHTQ^@MZ-z2BB<d|IMYHe!9~)&8)rGerYANun{(
z>6bD{-u9-d@_%AGBj|5nv6&5iiGvO)-HXIW4=J#K!vo1r+Z+zJVX8lU^uhAg&Z;q-
z#6`3qXv7OgzVq#jN%AJ!rE~H46*_aGdgZh);Q&gLkzY<Ucp*>3sqTOg7eS-{9|`?G
z&;ZNx3?8h=C8X`OFosmr5)(0K!^Vf<EFP-wbW-#~@pzHz>&Yg7Cmt~nG1x;7<OTFR
zFX5Vt`9$nn3s2xGENU!f?wKRj#&qE@6-SZ-{jRk^cJ)i{_Td~1`wNAPph<&recBJ~
zt6n<ojBRs55%f$QVcF?C>R*FAT3s*$EO5M>&+4+Bg}Yg6HvxNldrw>iKwFIUfvbv@
z5GFZzYDp@x3IlwN+DYj`K0!;sS|-l}iH*m1H2i+QbRvU7Pu`)k$Ig*+Dn%e}rxO8;
z5-(PnPiPqoXHQT!m9O(#`e=}am7MYdc|g%u|H_ZhzT!MdHWO!}e&AYxHi*T+jPe+b
zN<EnryH*uH|9!YVF#HkP(1GK|&@&z%ZXgA9dsx*)#l^(B2<d@G;QdodnBvleN)bLf
z7aE0KII8OQBG_xL_C7VX2m&;Rfs*!Egottsox2T<6gN5*7mK!M^!V^In?Px!d1Xus
z+FvMc2CEZK8B_6W4&i{9J~y%b;j^&#APhhwQk}ly6hW90!*E>6j9Pp=R1_S^df*H3
zv0b;ogmsQDKa}a<27)(P)4-z2yqQf++^ms8tRf&=G@oEU!W)Ra!xQxI9US+;?+tH-
zP^kC@d0R)4=m7BSQC(OO0<tSkEQ(r*TPc1ofb1`f+ky*~Vri0x=!9i5FJ5~cIc*<t
z+7_!P5h_$c+m>7!NLua}q-fEAuSlm{-ni<XdLx7!i&HnM<o9e_lori|y@y`bb5t5w
z<<1fPd#ve1^wDm<2!a^F17C|5#@pdTJSc>HEOCeUsf4^(T4ekz{~$wuN4Q`le59Q_
z8@3{412@RH15X-L{ITdRMq6FS8${C&nr0!!m`bqFaS!K%{1l&@nUe3Y1;O0+ccmTB
zSOk!IB(Y)I7tw1^mk*zqpWexYA1@WwlAFu?nN~-N`TSF(a?sn+{5^!<_i>;-WF(*o
zqjhslHmNrZ*5I2LC^(^i(73<;b2@|#wpkL);G#$nCw%O9BwosA9eNmP1Naz}*x~KH
zI}Sm1BBF;^N)>~+P_zN>3*34{9qaZAq_t80BfXm}j1}5NPv*a+%x=8@Iu8U+sG43?
zZL*|<tzQpH?f!BDHEW3&XH{Z35R@}$R+lsK2_(P>hPi>0Xp6kzZhSBbldMX0;<eC^
z!LADs?D0PyE^l7X^ARt-K|&IZirWUCT;{twL&tA$dYiv0_~slBek?bTgm6#R3{xlS
z5oqS8TfHfQE(Vba9p5y#HxHSzx`7{q$EEE#Ip&!YWQ`R*87oSKq>HPwF1nwN>QL8l
zaM*<BU!6{mf;Z8O7uc5m`#M~lW_fevVVz7;75l*zclyQ{$GwYk=c<GnLqs-quce|K
zc!53S64RpTo5=MWrrE=Ha6IulzuWS#n_7G8zbpH#HN$v_eE9H$eA|B(u9xsFGhZZ1
zdmvj@%akXaBiwz6(0H&ukl>u9hByIw#!ic<+unbH84kb1``~q?pQ(}O8;@*c;x3mo
zr_hnjn&fx>b~KApG~gAW5Z-Bc{IA$aPaHk`uVgp=(cax;nAcB{oqBPjlo`5-#sojG
z<e}&-c-x#7P`VE}>WE(ye2QhykzzvndMD}_5u|l<Sth07kBN6GSy%bwl&EqHTh)q-
z_H#>r@pt`CPA*ol^^e(f3iy)PjG~R*8L!=X6v2SYnfWY^wh$ebDh{`f35pVy`>@$O
zFsIl`d2oej@C%=d=kjBw7T%4A%rH{S-w?Ig-^zFJ88!QxD%Wd~m*(rHB)Wb3`n>*(
zt`mCtK^l&AI5(x?@g0B|Gv94c{hJ;?7`yX&Y~-R5Uio1G7hHBz${L;8vhC~Y3E|cH
zuThTQpuNKaB)B64<f^Y(B=<?EU<ImS=9MS^>r|R-uX2W)&^z(q)k6;7LD(CuhMHZ!
z8(*!pjyTLU8Ux(sKkk3~yO5506bA1API#M=#HzO(T^R`5A4B#<GL<l{DKJA$YajLM
z3*OB0(eBJHAI{50avyNGH-_%zOq`~?FR7L}@Z)!+pGZw`5=L7DmRyRpIUhEwYgXA?
zTascbZ1~4NLxp(7pmB_XQYh3Xp90_|e}ADRbD4)Wx~O}Ufe%(Fq<;mYnDPg=8inX*
z>+;gx8~-;a(aJS^Vg9w_jGw>^;bA4t=SWU2?(Hb;DRsO@cc>^c8y`KB$v2^-3KUla
z?bhi~z+pj#uEJKLfc>lo&<QwwTur6kiGN+Jv9f(n80|YI8LvMYYQ{Wv@WXcE=QAm-
zE6fKB&o}x1+H_D3E507{(zK9VKmk>(No<F}6#NaGo}o6W@`CAfIAD4~|B|z2h)K=L
zH>?d~E!C8`Ta9&AktnV0Bx8in&E<^xmY9MoMSp}ZzG+ZP_C6Qw$+ju(bJA9_w`S~`
zvQ6GqGr!@N+!oF}r?z>)n#_>X$cLJcgg^QbF~oy!PEniAW3PE5;7Hb{-Wbw2=^n2V
zxx<kvWFl0d6~a+y)cWpNBWIn`zzY#LurZz9loe)<QNPZ~2MPQAn-|s!FN6BY-3d8p
zHdGN1emKSqL}3F%w~k6PIgEr2*MP^YVd}cmx@+2d=W)(M5x-jfq!I@IuKaBmPI{Tq
zg(V5<iodXqOIQQUXywJwpM?5FJSR5M#)O<6iTNwPoCt*nm!fLuVUDnM{dCC+74G5A
z{Y{FeLTJrCQc>EJ6Wn2DdC-BZt!eLV)Fs?h<c|<dPUnSL%bjZkR4&&u?P)-XBO;1k
zeK|$OS>rF_66XzZ#BA59M<j#dL=4%)6W(Y%i)OIM3}Zd@9rWs@K$4N0WIue?XqqTC
zdxBwVdgRSd3MKFf9CNxaYli6*1(2?<y}TomB3n;X<rKYYkUHY#!`ojA%5G}iZ54M<
z=&3)f=~TcfM$v1o4m8c#7E(^Df8A=07rr{vg8g%eKPlISp#-%pgI!N<zk7Vkqe6gq
zd|PEq&~v$9DsM0arN6}BfoTxrygPu*WGn#!VF5s9CTjsszMXZv{4?oYY|+)UkMi#O
z+AUATgl{x*;qtBo%1fb#aeFU<Q|D@Rzplf2GBuAVlhT3}`6)Uvm$TV%LkfQpfCl+q
z?(YN11Yv`jqbt>_O*6az7kqQjf4}L&lM!kk!D122tSdpW{WO1F#=wTrqj}W{wz8xo
zg1(-M<Tgv+%`4AYmLT-vy<)wN_%d<7QdozhO^8ZF=AOUY(W)jgLKHGPC0M$+e}S3s
zRH0q#Tpi<oy~q77$d9%5{HK{>aW;)n73aTK#3kx*@+@2is+fu7#e19n_`6UzW+wGg
z56*BYaUbZ?t1B|rW=s`Sw+SX8fZ>Z*;=8$g24P+!nH#&4gR%bD(2(O<+jp_8<@80O
zK6m<rw@H_ElZyfw;*OJHy$K;N$?*+%6>Ku&Ax4y&ETvmB_iDOAU)Jee_<@AG&M=Lt
zJ#2-BAZyFBe5I@TyHCA^79{P*zg-d&EhV}>vsx-_wg#5K+jnz_h?WNtI76;Cynu#?
z=XUgc@17{};2pkVU9;J*_{xMOCs6Li0geitH|%h<T(15T@2G#mH((ENe=izLLusp<
zPM>IGvrlUz*nW&w2~rF~n$ItsFIN0MfM3GD={;vXjOWoE867;{B2J%F_Wl)pZ?^43
z!|k}{%#XIn8m7th0-enVd~0uU1YN1tZVWIX5O2t8qkmj9ML=fJAmn-Ef3EGk()}j9
zMm=#l9I?!-2z#{t+nJLpimid|Dsx-Rx4*@Amp=SWX;|G{HY{seh!@FSt~)%@!zbY7
z?j;%wrx?gmgo&@Wao2qI(Eaf!5$jr1vJu-j!EK&8g0P4AEjMHh?J#xzcs_IWcSoDQ
z4=*kS^)Q$dcxz9os-2EdVm{#RXFFU2kZ^kX%cxSS;iImnwQWYof4@2bUg>oers0-~
zLMb#}AA*A`whmDRxaS!^;dGMg(eX-S>=qMLgiddDwn@&2s}%vYLg<3Y=Bj&w?jBiP
zZDdW2Rhyayt@V*LxS!p@zzZHTr~*KB?G;y>{*BY6OoAPRh`)#~ioF!usFSOb%`Rls
zCmY)c+ZcG(A8&<V+n2t8{^Y{P>Bn6u6v<H`CSG{iBbj;8m`@x^R6$ZWS|o3(kPjKI
z8D;jb_5D)=PLFr-7Mh<$x*lsuyq`})#Rl^L**Z^e@zO@8iy{6|5;k<Rf&+yNrAz9)
zog^?mMVz21f9vBVt`II*;QLXjbX*%V)+B)f-zvY=Y^E<IuAV3s682FO!YlVtd=Yjx
zmn-?BUOaHQlA?yDfuHY%`O`!*4FCOkHB-psN1{*u#T~LB8|9bhJ(6J>LKJIYTbegj
z+jV6*^r(CQ?W-l&7eo=6r|EW^6+SC=5bhxxp8Du(Au1+NmF{2j9>%EiLgd{xp~gTJ
z7Yt!pxJ%F{!jZa-`GAgSf8ooVro(qtqkOJgrp7ecb}BJa)zh}a-ji|NdE92OtgQ1!
zkD=gyFYk|I1NtvsiooE)W#FaVfJ-L|p{;g?KkjWJRfYm3CN@$`$!EW<QmS#tF$P(I
zQA~!Tmmxz8NoWD57;1q|E!0sIjjid4qZGd<*p4`04^paPhS2{=CfQ%7@w;~Z=vk}>
z<#2KWSbLZ_+b*C+GqFj*mxdDvqYAnl{A#_TtH1N)*2@gtw5=c5v{((=E1KxI)FlRH
z&TSb!(y&H_```!#jE+1^ihG(hLIYQ+a*iLJ>k!FXRaLdKFH9!?aGllQ4eaWW>Nz5}
zJ3_*w+}Gfk(aPyBGKR*k`ss)*$D7+6e4+s)K4Ym34a%k`-ZT-4=ATY{=S4oe>@vpp
zCv0n`x|W`_P46jaCMZtUdguzaxeh-2#P9E*t(p30Y5{K(-!c~xp8)e4i7C*_=ZdC;
zlEte4#d$DhCIo*;`13GgA_M}l3f}LN+hl@p!c4;G>*G1y>X?0<y&$^}#xkwX|4^nl
z7hGhH;KKEPOF(}T^`}}#mg}T9N|<``s4U!(kKhgx2{O$abSBF`$Ni%*9{0^a;S9|<
z*|kzS7q6oU$C@!eXN_Gc@Jbaysi=Fs@mv^TOfs4?AH_WWu4?Do>fpIvCigKI2%EN7
zYGMh7Nue#M#RVyV?MFd-7^g5b;=;#)3A~Dczpb_=r+OY+Ly?W)c^!8M6<xmId2IG&
zesT>L+?c|yTwzb=U}q|(H*wlUaGKqT9ZP3;iu28-Ew_Pc@S^}xPFcW)VX;$RcaMFD
z`{3`#*n0y0DhCllXqGC<A?L~fmY{p+YGa)+o_^-&F<{nF(g(NIV0yFFR@rs^=??%8
zx$Dq4(NCVE+Ltijp(kYZC6Rh+?CQ+K&F9ka_fU)w#!2_a>su_YCzoJwj9K!<BQ~>v
zS1Z+H8dCpwZQuEY&*QJKhmHpK%lZ~6%0BeATNru&{`yY{k1o|m#CBf|X84m}QOte;
zLMlP<D)Av8!bXHt>!tgnyf!}Z0W$?t0jh48#^w0qK;E}_cdF&@2PK7!Y^@<iPp+m4
z(I)%*6#k01i?04>w|AbXGe2nCNI_&Z5;9D+Hd62jxZs>`tGF&BbaZ#gAK*6UIvI<Q
zcn?I>q>DZ3Q7QnO{@4HJ9~@8+WC(V}%_~c~8?`z03t!+_=?w2s#+MO4zxm6O|CJKY
zM7s}a9mx@Arw2k~<Gf`Uh0t*|JaR8z6rkDi_LNzD@>#sVxoeU+3dJ?80;g~#{>B}0
zykzjhS|M>MvY&na8p!DzreX!(m~gcTv444PBO8!T50wKQn&B^8po0ZWVFMJiv~4Wo
zac{|QG^V~qL~;hnXChKwG@lGsq*Qnx_6>U^3Y<Rbe)PFjyM??aulI@J@c0$D(yPQV
z6+iT9LY;b*9u&fQ<K6Kxf+9+bHn;!MuH>$!ho=bb4jEBRDbjA%liX{pQU(B4hPau~
z%GiI%p4U_KmKm9zfFD~}3;e}8U<iSQj7fwGot$JKOI3OK3*4`C`CKIMsQScgRu;=P
z)YJzC8SIZ`b{P#P;MJQqjvi`nJt9=!QKVkwacpCSYNkk!tdDy$DW}!x1{4xTDFb9s
z7SxCLk?)l>2Nw(Zl;b#1&TDhm^o81=S77f*5MeGc_e6t)aFt1WWg0ByCJUh>slIf#
zcCAI#fEYS%{q#N#^4@q4)7g`(Tz}*Ap;CJo&EcQB4f)x@d6iKrMSqZlepG45t8}oE
zA|^9Rs%%4}wW@Qfq}b-p;nB$An($}ZXRdLd<}w)3$~9g-GMFkA-0rt{{S1`hq4Th>
zea(@+a@W1;&roa-oLJD+OFmKl$BDNW=$8={$-p0h(syd$F7yP^uKxN>I7s23x6t2E
zmTU6#yxNU0{^1lc8Ok=E8Y-Gh4-EE>Kourn*;@B>F4%TkcK4ImBKU)93JH6R0KYE`
z9x@RxR!^Q83^i79xq$~rX=zO9rv9*qPg;Cnk^2s-W~Wx_MiKn%->2aV78y@1NLh6^
zwr+r)J=U2_Xfu!SpOJ(%nZ^(|LOuTOViyi3`4bEWumcppPeD_D4FlBZJ^HPzLl43f
z&FMM;y<sNxI5e~Be3>;#bkkz~HAo$D>_qJ&=f0nHKF!kORX%v^!7_CH&+p4coo6U6
zwZ%pX3=woUGD^UqCZ0_8rwE3qgk$gIEbID0lq{#RbT9aHm~c>ne~lt>IKx&vd;ZFZ
zi_n?|ljFw~fky*|K!p&eVSCmeRuYA){>B`<5f;vHsqyEg`O{Rx=a<PY_iWgbPg_`A
z<ms_yEWs9j-N=V=lluff88ic8sx9SwWcn4A-b)Im3IwSUHE+(rGOt>4^?^SW*$(*d
z&G<uq)edXZY)-d4gx_LA7Cb!jyb+%}<$+w1VvGXeoIwZ_tr5Seu&h@&fF658p?aNo
z#>_@WOtmvPFHGiRWNfGcgiJRkBc!B6bxXCj?jM@=6iN!GKJzeRGRoXx1s+F!kPP)0
z*#@ax&=BLd+RPKrR1|7<?eM4@@>}y>wLpQYU4onW!=xLe&8$H<hMxNz_xEnM%8zU9
zrskb2z2LNbq=`jsNf*E7ES#__O!n%-ZE{WO89HF3pUj4Tg7|-C0U+N;Pyc?DEqx+C
zzciJGci#O|TjqV0pbm8|moD>qu{do~V!vHm+}2Q@=+`Y6ZY#i!LAf;4-*OuyM-ALh
zMR7jq$5g(XImdT>qM+;6MjOJngdexUo^diKI|VULmF6Q?#b4b8T++XRzugmKC8<=L
z>Ars09P(r)u(l{q<e8vGP9%t^AYtN7!Yx-_fA#8hS3ZIwz7t-#A~LS7^xU??7IR|n
z4^~nq`?sFtgIV5JhQ-YQl#36n3LS%#Q?jyJM$ohDZIG{T@?%l{uTnsP+?yT9165-G
zWJf=?-YC()g1%%ID^0EPr5=hNgFd$|lA>K}Qc)S^U%@zHrS7LK)1)oU5(<|P&cX1k
zAyp++Lm7Z%IUcTryY%gxKEhlI_L~a0;s<Zz3xW_);AOdCP>`^O7uKxS?J3a>neotd
zYo%6sTF>|X^%mbp$@oVvy?$ihX+8a7(YB@JSO|E3Z;-t;hp>Ct>*n4oP+%Ax-ap7Q
z{#fjE>z#aYs)VP8z9`sP*t4q;a{LTNM?oT2da|^AGqtQl=@@kCd7=n(U)$ry>EQ>O
z3o2vpAlxOPn5*M|_FwehK6eo&dwl|j7qwkoN(R<2yVryFz>45@)v}sJD8xQ<+x<Nm
ziXDS%NJ>lXqI1#z-W~4H5lU0<b4V1>0QqJP7mG|$?|d@;C0a8WJ$b)D*g0CM`?#lt
z|Ap81cMw8)CMI2_xKRM~y{x9kiUOQ<{!~*e(qLMa8hpjQIMyzv4x?cpDbR|`Z^C_p
z&b^a7h?nt4YrMzNo3UpF=2PLG{$Pa%nOLd)nH><qJ#^)9GD4oGW0gj%3?=sOrKB&=
z)}9^2o|4>76VLmPL)fR+4~yE?)eSl+oHC{kZyStt;hXe(uAr>V?@m(Ey-b5)6&YR~
z{qUj4jTdoPSVa?LUvHi8#OmF9qL-_ktFx1OIQ1<aw_k%4xc6A>_<5kIwJiCauMuMu
znIe}Y4#6LIg)VYDlgsIJ_2UG)#BgTXlUWtwAE>Zn)y=7FVZqZ1f0lnx^M&7$cK!KO
zvn*~-tD(I7Evimm7&N+X;!~d7k^j8;XVI`;*Zy#1{RT}00vC9f-UVk1V&Db;fTpZ&
z;hl<dqx9kDmTI!NE%3k%{|bug$O%5|CR1<fjuSPvyBXl4{VEkq224c)NR2hzN|Z|O
zE{y2adN#I(z(=>PD9V2M@EP?}%sEl)u>?z!At{SYg{2{oYhzpEE>sIme;kkwuDU()
z-;y+dgV{n&jx}AWTBWb`#VuNxkHam;@ecog`dzih@mkpVuLFhnJm1g-p-Y_539(+M
z_mz2hcP}1iXux6qQ8JL`8<(-o9%|_ArNBUg9eXu<H~;h7Ab0WR=OO&ngi&k}FiW>i
z(I+j7^0@}pn!P4Tc8dw?*10ySZ5`I+TCYB^0L<QvWZ#FH=LUsx56jg)geY7nwXW_S
zChtmu`52cxU4mh!dgF0mO_c8<c2>(x`9(qOrD7jqQ{ST$5!91I&~<)rtI>GcZP|dZ
z9jhl)npJ2vQdepJp{KC-&W1%X+NWJ{*ujObc&W8}>g{jvE`AFyniI|P31S&^4?E9T
z2FoWE-nX6|aW7nl@{Gg$W-+OtithPwjNJ|0)slHC015a`@&HQpRi~s$H~FH17^M>1
zEbo(6!770unktTJ<1W)7PQ|q!(L-%bI9m$-R^+f^!t0{L#9V(@=Oa`9(if=r`f?y9
zR_OzHp4iO#MZ*~NDf$P0T<VCz=Ls;e!cXJ*dh?CzU5MK%`w#B$xMQRwC1@X-_kba>
zbIi-TS+F^NPZbUQa(JPh!~QKJ<8w#s!Okua@9^FGte9Pha9Z=k-2cbrvN=7_?sqRp
z|Lwo6aR?_?kzF%sMF6rrr=$SF0)8b;6~jS%V;aZ#VhrK+Ly3#PPH-VOV_#v5(iyTW
zGw0XA)~guS=)J23qOeWRP#ymGNGuZae_<&frKhf>PKE<2Vlw_Qw$=Ao!z}$`1m%{O
zYVA{kP^ka$$G!d%U2`6&$qBA83wXk>Y8_4fDMI)%K~YuXjG|yy@=9HHq|Mi!KInWT
zM*lDTUa~3YTh!yF6D7|@45tH4Tly%ArYF3TVn9s#!tN7EK;(M>Jr;R8Ign|lmM{Ca
z+^w&j8PeFDcsrO1Loesr0KczmA-eBeO84i>4>#oO<crkwQK!*WK`!A2t?3LYC@Dzi
zI_VNh3=!<(jSZWX2^O%;3nIW%+f8O4Zd|B$TW}86-jM3`g_=c_qTX~$w-t<20}?k`
z@cpuA>tRvjIi=^>3}Q*S8}Ux;p=OM6QEn3UR3~j!Ay*BUGp<|@ZGyYN3s5i;|EC$k
z8^<D%%)NG~-rF<yJN^kz>iXJOCGC-ZbjCEcFDIFs;MaW!Q(ZNLHXQ+f3vJEeDvpd3
z=WuoN<475ExC<>o55G@b>a?`_ysfc6w%sa{F#|Z*j5i~NM71f)@<2t`qqH2>F(=Wa
z+ybhpM1FfKL!nP@j485*3e4l}INK3spUW3D8r@h3cU}4|5rE$aZ|#kBzh$|;XMib1
zrg2cTvYg;S3gjUfYY$N*XH5RD{Qe(TCu;HZ@gNvli*3}-ojbZTY{0pA;#~Lipd;%2
zx~5QQ|F%YfO5PHhA@D(+^*d2N_aqAvC``QU5|%eO_`e)T>3JXZ{*st4%_n9i4OC6V
zRXdCIc|B#XA$xkm{pW^%84N2LZ@0~NtGcq_%SHRd?x7Cr%<m`k2Ao!()9z34bA7U_
znux>OJuFy*U)5f9A5FK^dzA|+zNsEVcF}%XNYDb!fcFQ_CB*v?tgK--NN@ZnEC^S-
zZZ)@DO3=cV=Vx2qH~$8T0(nfd*l>^eC_WDJy?v=-@jsM^1XIp@9`BQV@M3>S;Fr;Q
zlqo#dETq=m>hUfpDrdz^SnwPCn*8M6lkV5eR-tm{`(R*BL@D?jzlcaQkgy2`T#Q%2
z#?({ZVNX17bAjvyPK2YasqwHT&!G1rm<87mI?Cs!IH_v@H@!Q)gK#-FA`+2kY9V{e
zWFdT{$XjZwRNP}<tL0pTNx<@DyjIEp&tQUp+j0byf}Q`jdUFk0`=9D<b8xAvzvKT_
zZzE<E-@t==beat`!<V`U&NGtyr$0Mi8P<?Mq1T+i!DH6rj*(<dLn65=G_-{stXCr4
zgu0A*6ddW6&+62ZYF%xZE?xXGaTnjM`-#|cPhf_2qCohw5bZYz)^8+GgBe{#zse>_
z_u}(FHULceb6l%TEW+<6a3LG%uGpe^{LD#T68GsGw2dQsiokVry3w(oIHOOBzBq`;
z*G_MYR3uuv-Zmq7{xNi|T0j-660<WjM*kSC4MS3Ag5oVq<B6m1gf0th6WGn-%cX1p
z<Kn#AthMorw?pj_e6ZP`5d`%of;K@CufbFckiz}J<%ArVGt3k@ASI~egMU&BZay0#
z24_|XdD1NBb-P0>etmBOnv=?^7X*QF{`nI!;+IKMr+~~qdXa#C8syUkebM)54*Kwd
z86z)IY1yTPUWY}`-2da<Uf->}|F4V*K1_UjgH#!n-_$UN%trJHAF=ca&(G$u|Azce
z-y8x#?I|qB?N6K6Dd#!)H(5M%i{CfHU6>%y<Bp$07$r<pGhEsUKy=WYWM@0tJnZJ+
zBc~JceX(IbwLjK%8y^n4Di?g0-zWXNY%b3R1%K3&rgH-r`tpQR099C)3pyjC5lwHZ
zFdD9&Q2U8gf*(uZ;!G{ODG;ns89*82to-un{dG#$%l~R0B93*oU-ul8`jU!z<cpW~
zv(3VW{}^$YQz&fid+1ugI>m1V()xU|xL~gM1OdDUdRBf@{CtiFf%k%@?*%*uHOSi7
zmEhy#5oWS*ZtPBM7Bd?yb^HV%3-5TW2wq|xzLlG|G-IvJuOnekb5d3nA}~a+$th~p
z55Wt+`f-rGTK+|bSHcH%^k|ji&Eubw#$lmSQSomJI_iQ5{06d;0{j_o$__OG>CD$a
zZ&xP5n*^)aKxEjfV8EP{g$u#;yJ79##Sx)adWdQL0oz=yCi^Y~6*mX$)wsRIF6SlV
zc^qLAo0pKP>;?O>E7q-Lr!Pex9Z)7}DCjAW)EicWe|=Xk>-#kGO=ko9492IHsbx!G
zAd?A@sCdpyWPXt9Sb@MZPP8vebrt`VXO!Uj+a1F%qQ_!m$V1`N$ON}gh7>Mfy;pQj
zRfonUpXx-xoZ^7G%MkD9`=yhtPgc~pR5ncJ)*QuzLW!;g?Pn`MMJLzBP-rz>7%MXL
z6~N2rqMsg>=RU08Zx-6mPfVa_`6MQoTfpkI=dX(*Q%}f-I;80rlFH`wcKk2*OZwW;
zD<YoVuk(RQLWIre38?3qIe&q$pz9_Z&jsIgUW<EC8;iN{t+?Lzftb%99C=ssy&1z?
z<i=?AIxnF!zH_YYlfcFeh%{@1HrlM)oMy(D6q78N3k2$%a5PYaiHcC>>(Jbh`ltmY
zJwU}R9f6r(&a???fW62hgi@Hc0+q%7ufqQ8s+C`HEe!j{+rhBYYhx_BpSUcyRdXc2
zVeG(glF;;gUvFMy_I=mwBIa1Pcn(;&LVhhf&Z(A!v9EdK^A~xMFM<!ZVMK}MEQjO}
z)+Kf3j>f^DUNf^KNP4}O&lNwc!AGu+p6*LQoj6&Jcd*I(MeSRLi4qaqWb8JO0r82x
z51|v`JT27yC(%^U^s_TgNl}ShDP75CJU^R8%xZ?(=LFMwzNsE(D{;YIul(G%srH{n
zTAwQQPBEelOQG@rF@)gX6(6ZD^wY?0M%PQk9Nb)WA{*DqW03O)PLS1>TMKwlXf?{&
zaO9!`a{PI=FtMoq-f+C(+OP6vpB`2O8%r3(Ncmn@O^PjB!eAfA+Ax_=gU=2Mryh?Q
zy;=rj`E$VllXUVFOWMyctcoFIJsyori`Tc~01mlx*Q69)4q1$zV7P<(F=u3H*D7VK
z#VAW>R^Wqj+X0jQnBLV<Q-*zdwoK+HufHCF*Z;T!1YN=*-Vwj6D~&&rKk`pVi<QP=
zR<xqqE&0CAQ5`T0mH(OP0`T$#&EiMDu6Ie{PvbEci@5K+7zWOYZ-ia-L>{Lj*-1;4
ztYGeZ`^$!D3r;;qn|B*wV50YdVw$Z;3x!-tBUVX+xb)JJ!-qnyS8fpebT4g2drb?g
zs2!=LEawA2Mi1cy$vid)FId?^g^jLtFIXoW^+b$3y)la8x%Xr_sPWJ=z;fo?x|x)~
z<g;wy{o2WDHNRw8_?z7ec-}|_;yu?D%nbe6eA=i2`)VPgLr)5su5VR+GUyK>J#oAH
z>+s@<iL#F$)MDwiYaYkzn_KOG?wL%7G2`D@4SWZwz03EHXoxQ8>Zb|otc1s?LPS5=
zEZisQwyQ;v)g=&2wQauG+Q018RR%Kx%2pc$ts8%;q0`$fqTWl~3ZyTM2(kO*HS@7S
z5}@)lIZe-uL9cmS)S&qJy4rlH$`(+)Hr|-8bobud!xc>jO{qU0yees<pS@KZA9#^i
z(0vK;VTGbt1vuBR&S_u<1isuN^AjZ48gwFL-tN++=<7j4SGXZ7DX_N5`jgGQj2RlJ
zU)OWSjVkI~`JCG-SPIn^WX1VkJn%^Dns;FvB6tDA<;vg|=rX58F1Cw?6mcPBp>EnF
zoqndOK`qW!?c`kz+`_)ld7N}z+?=M`6bXoITPDv{X1#~eA=NXc8pXGY4go@uecKX1
znB<M#qtvyFHJ!Mlds>y!%^_#TkYMWH8}RrJ6Ok@AD}O_~kF^PC>Q20w)O2$IOx4fY
z1&z|A@7Hm><NByQt$UAU;03Sfr+l!Ie3ml$@n~p!Aip7pnvrqKRDP4)mD;cqTS6+6
zYeu9{g<eQN-N|5`JLRyIE@G=CORI&fCWE=%uw<&LN^MAYvwE~2z~3-dyMFpyX0~&z
z!&u5lmGOXy5fs;b`WUDb_Pjj818<{<z|gBlrO7x$17b%nZ9y<B&?5Xo6vOZscI=1h
zlq;{f{w;e0jo?dIecjRRPTceKC3qf7*cITS?J_fS*|x^xG0hhG8Wp1_=+o<;PcTqN
zX6Sgc0?&&#ZC;R!m5e_o6J2;}ZZ$`TvMx9fJgb7TsaokGc;49@Rhpz*w^Y}~qY_YD
zwv|;szE+S!>9|kCccrsAX_@Xv(%O$$g#;F0*9tI-&ygrX#K7{A4wUOU(-5t25Js<l
ztm}H|!x2x%Y+lVOat+)!-a`dl!TH61_%y#=>mS=$Mhl--B@%J7Z+%I!Qw}ayCnaIF
z`PITcu4i#-_niCXXs)-ngmHXuKC-2;a0sU!g^Q{`eOk5*eu=Fd;bl<#=gFWzO1uY@
zlueaZZLEXO+>HcI3%w(q8rg7p`6%V|%P-_WRf?Rs%`?%X?;DV8;BzmoB<(N}MdtNS
zV!IOXHVGaAUzirn)E!hP<m?G?x<%LkN8~Nzr|b<O2p2zB`~|+&5gi{`62jsg_Omd#
zf;Pm|-^<8${Kv@|h3G@hQmi!MCsM2yTd$YBgZA;GH-VC(;4&RoH<$Qul^9*N@<QAB
z3Fed-vbMz0IXge55@u6rh};=|yNC&-^3g3JM^!#;?RaXS-(`iO1E5}>I<ey`QQQ<?
zcVb3;CN;wWPZe85-XEG;T`4QI&MFK!k&#fD;31d{AKX2twPCn~occwJ*|mpIl$zYP
zioYI`vAqH_#tu^cP7xmi@JA^f^b@rRdb3rBj2(=fI(k_V`~U-d<n37+q`~XC>vRTe
zu;azB4zgh*{#pm^-6U;N5#pAaanIC)?Ce^a2(0Zc20^G#DGT`WaZm02(=~mlT5bGc
zd*!Y(H^8p{83sUs?mLvY_3;|!Gru)*ma!V?JP%O7=d(<?(!4HbFh{m0VvYE11Ky(K
zre29kQFAhk82*=Z8)8JGTB6&T;keA<o%qGs(0B3Og%X3Gb-#XM`@P2lUy<o46c-iR
zHOSZjRS6NUnlVd4gZ6s=XilUEoi<ihP_1AtZvHeg(Yq~tTid{1^XG<gM`SMx;D44F
zm~vYsjmn;)s3UIPT;71VtRrX^O}IV>adpTx%twbH>l+8qr!y6z)Rw>%mBOZ9RrDJf
zfJ2=3to1U09k~I8oA&qaQuYm0!#Q}Q?Y&o_CIni1vLUrm`?b*%jjXqr4%#3N^A2xD
zBD?ugIBOM_y(tn&lp_QcKZ=y4fQ1Q*!0XdM#gz}sKIJ9L%yFTi@2YP4(?c^JH;flf
z)9%G?W}Z*EA-N=m1a6!O{G&^39HdvBm<kf2#b@d&RJMq2QP&LCAo)*4iD52@)3Nnf
zH~s>uoVbgeVUWyG6bG0yHZ8azbA)aUq%T=dPLUC@R{yy{wi4MBATB8pT2XcgE_C&S
zaj_vHcV|a+!OjA#U%bZVT<GCMAS~wl=QU~A+ZBlck~7Ujs>KBkHMSuopz0FNM4+{^
z9GjlC1YcVU2a?mz_g(YzLbd$4Lb%sOzG|+U_MBEE!hlm9I^)KvuQ3cYUa@IDg$Z)V
z2qbO>YX737p>cT~eh*ZodmRZR9$X^2+?;!v@>5KZeODLCTEX}n<Cq<dK*pd8LB>$5
zc>7<BdoAFBq6fxL8og>-u@Fn^?5NG}TLBTPW+_(fcK4j`KoDggztSc3lMlYnM9X?Z
zO^B9n4a^d;*NP5wa#Fg8P~sl;^Aimkz9<hGaZ>D36o&2*IVC0x=U@AEIGs#(N!_x#
z4lS>5+}XF{%Gvw&&T1;O@^2}VFCzHJ=8X)z3B)Is$;jLar2abW`02wD=)a01`@S;1
zd!y`a9d_yUU70+-w^FcWtw^{MltR>x;cMDQ39o)~YC{|T^g`(v6I`+U?2-N+fOBSE
zz7(pW)xsj*n{Q<Xt$H@dPz25SnF$M8)1D9f7dr2ZMS6E4zaQ!H-M@kaP&^Pf)u^@S
zApZR6dzIMZc$RCSNUbZ)mx&eI=b?U@|HVjeNCiRu8v><~XN6j|kCR=^9MnF4hTCe5
z@N$th&>(uf5`gZ}plBK-ZmB%KspM#kvYAn8++;A=*WA-Z%atlX*hUoS6y^jlSTT<g
z->Qp(hPW6ApR@)hE*smxP2Z_aR-g5DEb^p41$MjpcNmES(*x28ZDE<h8Agj#*vp|7
za081_mNn3x^u}_+1m761|MIRqY%><sxgwl;HaVBrL>LGGT*N#i<8u}cS(0|Jidx&8
zmT^JS!L<R3uKX^_#3+KOZrm)yq6fle0dAxH#VDNCB-~cCv7P%2q3X|Upei3t(vH7-
zU|V;i8VSwYGXGin?dM2P!!O4M{@nA|ABjju-5Hk3&FVL6lzAN~(@K9}MlgGb$b)~;
zj+`a9rHVjT3@kJIeH<?BlM%hq)!v2^I`~0q6}I{Q)Uq;HlEy=^${q7!<k%kmCf*+V
zDIufL1}Nh9d>&J~J7u8r4l+E?p3DrUuRXo1l2N>p+)BXko-movFehs1MMi{^(n|<%
zKvy&1(Od_9gnRX;NM^-qy3H%{==KKTiz<z$LeHMjC9InhgmUxP0|0JhsZyDA#^)2+
zQe%j6N^}g<`1hXduh@E2ML)CTn{S6;FhNpRjiiisa7mi1tc|bHa4x<p{|xne*x2Rq
zN7G*LpH;&HcH6#v+a9c-fWNt^v0H^LX_+xf-nbkh+RvD}#Sj%BaDYbzaX?$0A29X)
zRoaLVv{qx(kg9QwaF?D{u-&gK_+x&Sv^<nIWU(HS@7*Dw^1YXd=th{B9>~GBa+AG%
zdPaoCPmr!6FqF|W2N#K9#SgwrZ)>Mr*}K5+EWf9mi*y(bnfQb+MeUu>F>$EvY`A>!
z)rTq)h&ppopJJ!ocUZ2393Ye)am>uKsWBCLCpEHk92h?Wbwur{qLU8-`t6w`D&qhN
z=r*E7!mw(ws#75X$VgGs8*v<bS2U&McKg6e85~5I3e9&uYx3D2#B0Ev`8~ItuB(uQ
zIRjrCehOIrxJO%0l3Z<FHzllp9N^Gp-rCohj3CM-mCT<$e{7Ucd;zL_mEcMow(=jg
z7InoFb*|7FT2z*h>ni88SVat};%ZoQ{Xh6oA-Tz}O+eQUzSX7JdOGhjxSM62Pkb%i
zn((&1P?0`D5yzInkTMBXml|2N^+fUcx!GVmZ&%r$hh4wl&Lrr6SY~(!XZ^*Gl@RDB
zZ;)I+xySMRY2`XJ|G<j8EJ6Wuq7?TFc4&7)5PYCG?;`y=pseaax9RWHZx9*|G4pR|
zeq_-=bs!#O1x@yL4S!6c;7x@t9$CJ3g(@?7%fV*6$HRNIvEiDNpSgr}-u)NTFMa=@
zTL7lThpbJ-wkBAJ+c3Wa=!8A52bz{gq&%A0L+$ND^VM%~w%xXSM)NAaOFzcu=x(Mr
zj#ytTTef&4jD1}K*-T<W3ns1L1rMdethsLL(T8n~j5J|ayI&a(5j(MN{Qk-Z*{G#6
z#f%igs>+0ops`_RR=PeR&zw%`;Yb9%(RGL^(RlX9<iFt--E`_JuLBv;?bjiY-$RP=
z_!>Y?N;;qukCdzC#8XDMH;N=lHe@1_GtL_cpIHjIYu(oMI(16O#N>+~72<|V>`;pR
zt?f?@P>yyNWNwJukO${Ke8I_`^?Fn5==w`&HYG$^Lq&~J9F0RZ+~Xf*V#rrU>vwEu
zA464hQa0(*;Mt&yK<3ZlyVs_-HA=n}K3u=5Tl9RH3OR-(!H4PMlKpV?Z^2Fead*f#
znjnIGNNHJyfEnAv@HV{+TfgI4=V)=*fD67$NNYlsY1imV<$xx?@KUbmUw`A0(QeQ!
z!BVNu2{ilm->Oz?Bx4n@kg{sAV*rNIXD9Y?i8WLTVkp~^y4PNOb?+*RwZ7C|a<fJ-
zDK3~^dujLdT2BC3hapOTFprjdkzdoyj^vdK2?8951@RBvygN79(JA6t(m}0674%z=
zlkP0Rs>t8CWCtI3b@yvD1vlIWv$o``d3Gi)r{ME+N92i0MXxe6t3th^X|OSiw*oEY
zz#`z5;lG0MjD*d5?$Fqzoo*%eY%XS`+4Ydm3#89jmm_fD0~3*0LC4Q?3S4Xrd@{mM
zs8&3E77y#ca2Nj_4nfI=KDR8h)OgtsQ53=+luhiz*YP(|PF-zO<G?USinOG#Y7#S4
z(ZLKk=zynH>x>4X%!+kRW+VSjceGq&T~wfPv#@Xu#?`^H%>S*Jniw0O*j5Lz5vAlJ
zs+G3_13fnms`0|YT@v-;{y-alC!(nX=Yo29i=4=8un&J+8~M4L##I<k@7s<Td7jSb
zf!HBILm*be5ZY|OpW;&|_i82u107vpmzT?W<4ek;e$197&RejSUSC5TdiqLk6*te<
z_z6hEp1lr6?<B^Or;w;?M874N2o6A}NW%r3jY1;!GtMU-T`<CjrUVE>uuWR^rW)a1
zN({xsXxuZ?i~;Zs!>P!BV>XC!@$hKjA-^>g!JS|hM|UmvjB#(}kE(g9Ir41%hJmE_
zEu1=Ygx^NX%<R*GyXr~Hk9x6DMo&uQ&up!b{lk^;)MajXbW%45_rf~6E9JE}QX4N#
z&xi<*{%P}+8LE{j{FKj0;S75OX2X4+bldgEIR(&2LMM1{+je8&|I7k_Gbt=<-F`yu
zgK#hHvgN$xMSA#S{SZR=I2-!ZYWD<G8hS&Jtlv=X!nIkU=-mx;q&!e`F0qRk34VJr
zACVy4fMsI>H2N2}{yG)@uBL)|D$(zq3@RZE2|OD(jS-QHtluLW?2qS+EzGVAyOvcM
z4rmtCoDBAS`y2P~k}CthGZ-LEXs_48-pqSgvI<>mmaq?&kCWuM>J)~=VGvHpbtk2W
zYyZf4q<#!LLvthTE0txFxwbyn8{x#IU}5*pceHTm5~eb<Qj7Kc-xf%SGS+)Cmogv$
zOr>vicJ5{OdA;Eq=(&>Io%VKxv*D3p*#i{)cum!hD<OiyQSf1Om)^u)Ga>{Vu6<>6
z^8RY-e|hAV%gvi_(m1-c8_pka;HDDR-kyZ6k*b-n&ZFOurTg9DPzxZM{y+n|NM>?#
z5Xv7Zv>|+SmwMipLgM64DI+47am^`?b`~xyX-jbX^H44Q{ShS7wu$~<3U#`C^{+cw
zH&5Hz(&9aJj=ww6V`*dQ7aP>7twf!47djiS@<-K`yNt+A)OcU$B6Wou&qx(F5R&;@
ztMkW6vQe8n!girE#~4$q<(0o*l@SUE9q!xg_sPVN=~9&BMm5sCF*k19ju_njaeRml
ztY}<klbTA6TlXTDbeVv|<eB(2nK(SzDY7}(AAMC0hfH%b`Ci~yg?(e$yVk(biYg7i
z0;?I8!HS@LWtHeg3*_5VoAbhk)@iNIE99nEQEb9YeQPe8+7i6&?a2WYlX!I+5$68s
z)m;4=kPV*rs_;Sa`7Z;)&U?B%-IpFukX=RxxVd1%u^RyI7=G_TxyaaHD21_^Wm>7Q
zAuYhYb8^Z<2%ZZy43;8hD;WN|a2o+Uk!LyaFwdoQYGkKYr3z#BDO{e<WSM>A@$lPu
z^W#b}!}NnO`M^KjO^k9;A4otGv-j?F5becYbl5e!te{flG1w{AKldwZ5jc27eDg}|
z$4A-6VYX2nqY|Ey<8lN<&*7IvRu_)(s1G#c>Xxx5n%SOUR*z*a0KN~IpsX_qa_;|b
zo^DX}2yaqrF_-Oy7fI2-6`&sro23rBh6-iz_;#j~lKZwgF|vXtp5C^7u&Ozvy6Jn`
z8_4JzWn_JtsE~J2zN4hhqUfvU0zU!FpF?K^V90XQ9@`sj<GY73adQgY+u##vY}rsH
zgqXTSVT7)~^DWmx0M}oiSLW2{$E)FiAm95h#SBqW<em{EJ_MaUimjwLn+<#1lR=jV
z>x6}Ap+YQH$?Cx;#g?guc;i8c_OZC`W7~IAh^u`r{F5so94}LGlFxbn2Grb*4OGHt
zJY7$2LFh9*U*C0lBR1M4{F`F#5mq8SqQ3uG_o+Lt8$K*f4Y_XR5(sboxw2!ioIQiz
zkG(lZ>xd7$-#!r>A!)Y+uT=%tCWDxa7q9L<HbC8DCQARNg_O}k^y=H7NPFfkVn~*q
z_2D&^KDC0(oBS`?@t_OEtGqmmPygvG^+1wiKNhVo&s_54qNuy{r>kE_{WtaR15>fi
zv87gON4_}g4Ghx(eE9ppoA{?gDwOYqR`uZAPb@ra5e^%zx2dxw4tQ=Lq{w7|fp8MV
zEl#v1x6XZ*Q_Ug;nt6Qams>A{3169A!eJ`0j=PdOo^<z5q3ieimm4nO^RSC97<f0I
zkhkTD->1`-%TfiUhCn-+1Je7GuwJRDCD{2(P6a<8)F=v#It#jhdbB0-<Rc`>Cl+YS
zZ`Z`JE68J>uAB2_k<I69>G$*qKK?Z9RP_O1kv6v<2ydf1VI#<v_Pd#LCTCz9pQs>i
zC3wrwHSj3&IcNu5ZZP8h6V$yf-eEv}TfOrfJH;`!XK6(6HQ#zT<6}_&b{?V#c_K58
z87rj6eglmYOUVFMI40@;qv@-oqVA%uhwkofknV1!Q;?7l5C#zGkgfrhjv=H&Qc8x9
zjzMW*5Kv&qAq456oB8H>-|t;({{F35XWe`DK4<T9E~9~Fjj4xhCxgkS#1eYX<-8vP
z4h*KvxmqwLH+GN|r6Rf_DBsCEMGA@Gb<Fs$NFX8{j_0mk<#E47(P-9?x1}r_$bA_D
z;<}UT4(I9c6f5R~C`$3hlvwAshckz`p^x~%f-}Sm`ovauwgi{h8R}hBR{nc;N+Ej#
zk7S%Ava0y^<kKsWWIj^ky8L<s*Da`^nB62G?}7_ZMOMOBHC=31ZOfw_KsN6x9X$p}
z1FfW)*@pZrbj2*exGUwlS}I*|;bz~~!~$eG2%r^Z3PU4JlUk~%;bri60JvS8n+W0P
zAU{wUZYSa^Lghe)ahHx`C@oD^=bN1U0kZ|<uPj@Sw~BQcxq>g4sdm_3_9T15)-No(
z2y>R{53m{iH@H?44$7COF<Zlii3cxfGh~pN*kOdb^8q>{-##A><`N@Ixu|@ZItj!?
zh>CnAeSVFVneh*fnHXQ9yfg^^scc<cDoz_|n<T#8;h0gd<9zezPBQcU+K~Df{JKaS
zZdYCoWvEC9W~@2bBkowEf$H&r53jlH%eBP9ppRTgj=|~;O1#y-;e3(Pr3vA6qZhBQ
zIM<jeHL559*OZP@D6$)pTx!h()+qjfJjv~PU%NAzsy6vI>N2-G<NgbhIw40!h;)T7
z)k(Us58_Chtw@y|47ch+(>13`>i;=?{IzP2Bg5(6#zHkBU(<FnguI?17V;hb2HP*e
zxnzfX-VG#Y?|$0{w4m`4t;$a7TsV;nlH<*6X6IWKJ@>vuQFFGOgjC!#mYE5*gZB+o
z6U=^Dk}nY~tM{EJ>Jv7-NVQPq0%C1?>O5QW;toE-B?|>d0(xQe<11;_v1SeS__iP9
z`d@8xJlL^f6gN&q9CUF1I!taP81AI~ha2XNrFV;4PqKe<Cd&4L0MJ7mPPl-3Pl05n
zY$AFp8A4N8wqSu^UdwhS1}yxIe|?&>YaG`Jy@370>&99jW&T5TZzf*u5QQJkXcgXR
z3&O@={e5M@kMTc=4d5PMp97YLaw1GMg5;iHHJMKM4x6jt9$+}&A!}u@h`+ydP+XW#
z=Z;IFr~zztJC(~Dd@p4^?8oT?nPJvHtHq$MOoRxIO+7~SG>6HAu_9GsV%N%{v+-1`
zqGOo>!{jf<f636+d{o4~bk^u{|GGo!i|~YYIvm#}FRE9HyCFEV`PX~5EU3QXDQMz0
zn#^If;WX{rnLWB@81Mj)s?n(J`-e?}qkVfH{zOx|IZi8}NCY$iX+JE(!9egB-<qTD
zpqf$XgF9vy&w(4p+stpHe&W&PRWb1Eof_GKWpvI!VbL~*6nm9rKI+tso9f5VbapG5
z^c()qrV4LS)kcz)y?igMa`^Rno8B*f8=&97e<F#VnGHOp;;+)joW5n#O638`X(@nf
zV@!cDj?lNRJy-{<gACm*D&NwpV|BSmB-*17)#iqGxXjbF6M8~i3{9sl*L)=e^rT;t
zWB_;miar0#+tEhQK?9Gy)8&xRFg72)@ko`=kn_fqx%RTfx-ZTo>94-`Prz*XVWM+Q
z-x+pIqhlZ4%{VCAi~)Wbz<_aSbT8(f4otLjjrjG;z`6W}N~6xkBx`Sza}BlhV~P@T
zmj2|vtdt2X^tl=|OVxAf4<iT>X@=Xy{p%VcT8@$_TMLz3h=uJRq#p!UN=;bt;UoF|
zS{NH~X=8e|^+;)W$iJa7U2!&qPGo%qv#tFrp6Jn|GtpI6(5#CM$R=H@nQE8Tm7E|8
zV-Q^8r$q@*4%KG!`9#2UpDIgecf_Foq=z+;<9M3Uf%qU0q1kFG0gxo_*&=q{L!(mG
zcrz;)dUBdHUbyMWH^Utk0!_h-4ij9<9N~_>8#uyUHT1&TFhWo{*aIT1N*GSI<rboF
z(QsdJtb}MmLrT@t><kkJr!Bm48Dk+3a1+W09-QR}Au}|#s(Q(VJ|>Vq3}}$1X)6l%
z>^tO8=~w4$^yX_HhyEo&n22S5krSikr%@sCz+C#s{NqMPjs4#{wZgZ8(e0oWOeLbK
z#`Aek;Dj4w?O6<frI^VAvwvTReNQU(X&&!e6yKC(Ue_6$6rfa!cRw7Jql;Q&<o4ZY
zp#F5r;A9&ZH>(zP{l{;Z(>$|L9;^Cy1kFq=Ldma^tf!UJFVYggH{g}o$>^C_{XSt4
zRYUP?ZoX1VgT${*fMhJVvNxy5L4f^*a}I!wqkp;)Cz3>o!E>ksj&X^P#5j$Tw8W%~
zdEkLmJ_;-uvMqRX4dj^@Gc86d4m{>3>T8%MU+4@~B-wnBpvd9eaY)xl5{%aVQDp7;
zxpF(vRJG99t8IY+5iDoyxd{zR&-_wihlmv80OIDEaKC9<f!}{sm)UX0SFxAolF+b}
zzl}EDuWF~B(B;ENi1HL-gfkzsjAD9oXYBIpJ)|dwVkVl}EZjgHk7=@a01h^wAj}-<
zjvtW@k`Tds-Y{N{#=}b$h=2OUuc0j$RESlyj?6OQr5t5zusP;`1S<CeDjHq+BD{we
zASGhSnSCt6Ah6NeO2=h=_NW}h=40%PNnYs?IuMCxg4>a>yr!q<xYXzqx{M=Z(6%8t
zMsMY0Kq!dOYXybyFB4T7<WfLX)l~TUSs#U=ln@k#OKJ0&0N)&YFGp_=u?yrq-rZVx
zlh3wUw*fW4NBUC+P`2q1=ZNXOLK=0YUzYCUNE>E92>AI1PO9{rm4P)=t@$WWE*oPX
zT;-)s61NHc3&n*ra^}Au`Cjru`E_dS11~btPt|6A)Q<F$p`z)<52l~E8i!GG_Y7t5
zUhs?rfTR7i2G*QI?fmq<;9CTb*C{XPReZgBO!MAdbX>Z}Pl53=NmmdbR3$;~R#ZJh
zg3w$<$Zxn{pcpq$VN>TFOOEM>FArpKqLS}+RNnwf)te;AN7{0P|B`Gi(o-t+aUh13
zp(}rKs9hwWjnE^_!gc@fpqJEKXL&>+e@A6+mXG#Bs<3v|<)~`B-e8$p)E<9K*FRo+
zHp;ks_<8|bNMq}UmdpL3^+S{_GBhOZPnxdokcu`~g<|z#$QnW9Aj^>ezF?1Qr3_i?
zRZ=|p;X&7#O(|yQ4DUb(@ZYn(<r2I{SkNo}FYBgRi4LymYpo*KH*8A@zywkBL<n{y
zX^&Z6N(F=Fxu<<VGeSg@`SI1GA|~B9q#INnjnxGge$U|7jPHfLs&ORWO%g2WBfL-z
zqO0>hsm`MKhFbSYzLz+1w0$piX_kdM$apGwYEN);BdT3N{18@3J=6cM=8eRo{?JPV
zHZq>eAA$>ZwSBhEzNy5G-%0Tg=Z+Kdtxp0wo807%jdZp_XRcTO$7>R#iXWDjQ~bG6
zr(Fxx?qRzCV%Qa=U2CH^!VI{(isytHic2)?hrAn;AQdefc<!2PaNvCt9*RfDdQY#a
z^m_nu^)f%*;L!?XtSs)!?e_IM2Eb<yb1Fs|HypoaBgjoFM&44v{Gs5j5;+2me!YXy
z3!#38)4mPyg>nES&5p`c*w;HMLfJ}FY97XWz78q}Bg13<L)(h!36C`rS5SAEA}GH_
zXAq@3XaWJ>e^HHhISPULLb{_@bhd4N1Vv2^z2vS#lADqs$Z7-F;W+Kl5d<~(Z-O`B
zfN-4u0q;+Lx7>R^W;Hk$fPyH5oP#N^PtNBJvxh4frViSRpg5zUJ~SVSAT8;LM`76F
zKb(rOW(6jioW{~iRpLigpy8{xI1WUehV~b-@Ze>&NY(yo?$>q93)%5!YyDUHSSXAt
zt^Dc&Y9#f7<?0V()C<Wm=~bH-QbTO(mwyU6LQhzh+pi9lf+c9}K%Rrj_>++kRJ@t=
zZGl!hXR{j9z>(U)!}?<LRiKLLAH?50N;c-w)X#6;_bHsX0$hG+=9p67xiym*0a&rl
zwxo)^IgkZgG6^_xF><J%t>k%xG9NvHy*2dC62(9MyDi2JO?S{DoB7@NG5K{TL>MYP
zirLyQ$`Zu2IO=T{WHH*>nYyew><@dFXexpxbtK6VbY&3-%`6iu97@Lq6*Zsx^C>F2
zkOO<L+Y+HRs$HUY7@X=d;Ioq_NI{48uOEidg|8XoHxx<AD~#~O<+BTQg#CA8JEnQF
z{_OBOF~GfI?z8ni7iWmih|AZmo3QZH9x9=8JQ=`C*yp~qv)C1;1%)Dk4)qG{DJQM1
z7|!7?zfoB*+piNp%$mSu3#QA}RgjKaH#2UZ^vUv}5F)8h5M_O<ZJ_gNDasSG{)u$j
zA@BSR*=(ZEVhEIPEa>rAixxYfMm0!YjG@8}Ny_c+$aQlZr!NSBG1z?^I8qNFy^`{>
z>d^%l$E+mNckQPqbE|s2(-d70j?YpXWt>bbdm=K&o^2;dzO+ReE2mrUFQ1HEJ2f{v
ztTpCXKu(}4$}x&sFxPP@j55qr3#I??MBGd`SDkzI7^gaJhLM{cl8WeEqtrM$5nHe?
zr@=sfk7h8CJ@4Ai^&WpFSk_|)C4?Z2;fcDGZ?yUu)|K<s@$3WB!eW%@u}uIO+e^Qg
z59P`eM{K3;z}(a(jIQCnVK&~b-L)IHzLtx|n~f*UeFx{SV(Nugrqk-h(=Z=ixUt)+
z7%(mznC;uATtABmZ5PrIcx=SD^SO`;B>n!C6CEC;LBc*2D)iJR0fy?*76jB%O#+_t
zf_h1V>vj^k93I!GOWaA3Zl_!$q8ydc(wrqumh2y=>?JG&MnDbmM-eQ}I7YNf7`7lG
z?nGy*3JUGX%v6mRz0c<v0D9(Sm#Sn-31$_OC4;eA>4uB1xsz<2f7Y^wsbo|hOAz1x
zJ)dK_WCTWRP5CI1+<!UgqibUU8}(#t^jANN-#%Z(CB;Uw9f6cZ)W<<agc0}i{l6k&
z-A^Nn^uerh4H!Y_`)RIh7HH#ks>meg4$;-X-FZuB`QuQd;uEQ*%g1bJ8>S`4U2uP4
zOAd=pNt}Rx>JlyDYP9Ii4RNf0m%6yd)4q46E;dEi)t=VXhDx!<`1MqV%0)bbCQB{<
zqBr%Pe3-MKsLUWW{e1F3XmD&J=-lh6QNDC5ffOF7n|Wlqvsk?R-;ZXYr?fh<u@R&T
z%ef>zlVYY6NMg>G0AYv}BGtEoV7=0z9m6;}&+AN5lOOLb$~~<vxA2UOdw;CKtZS$S
z8OjQltGSu<EZXpX?2g!jMKNKm#w-Y_DuTiWq!7o;$$+WYmG|d19jx+fw^pR#m~#Kj
zun+1Qq)W@5=m|TJG_Rd2qDn~NH)}D#heomvU+Sg=3aMscGA)UVyAf1uT;bkjdwL@#
zz!7f>uba&++!l|r-Gz8J2`ROK)k!HI7=?NL0cs6FFZuYwX@lZD#iB!*l$}M1bi$&D
zdKecc-WF;R0a`gyQXX?RC@zSR?+`zAa82o@IQXti^8rOnGwuTK(u$saeKyBQT0f>@
zEW2@EvfKTdNP6e~$`Kco<NPvK?vjoJ*Ev;kn?lGRYlAslaz<$fhnxJ@i%U047G4q$
ztZo)CVFw|B(v7DM@t8;__JN>|XyRj{k-14s!px)KELZZ>014)mocfBe(X3ph0K{|4
zd%XI|q*DR-^pXhE>z;<P^y{>=;Y<3YH~MVY-SlIr#D0nOYTO)Tu@S7n-vR~HY0y<X
zgM0K6cp^YFF53R%wZe+dhk>$arYpde<-XnIc;BgCMH%V|vI1Ot#_s?Fpa?ISD<s6e
zcq0T{3*TY>wQ*Oqg1*!k4r$+%?JByzFp)$WIj}60XrUq?KgWy_3)tIg31}@gb!DyN
z6m}zJ3Mi$48`P`oDfE5Ff?oi1w|a9^3LTC#^0j*KI&yctRYY-EU!O8~hzR_tp#I+f
zMo#V~v+Yn#qW9ipE{8l~SpoipDG7MP`m=X+INBM=qM<*An*7F|=Qj_Ez^X}nqn2Xp
zL}8V}RmL=0IU3}Xib{of(nUvL1{)Q5>b=)+LuK!_Fx-S&VDN8AGH3xKB}Q+C*+E4i
zqw-2$dRTTsR!Nr;g#~T3&sqdt8hbe$2|2p+CL97(5`mB~2CPU^V}{iA*Qp#vU>+TI
zb{z%ZJjvC)^9ML(Z>*XLb!Am24%{Y`0sDy5`;{Xd$Nr+%8j0?p?6*$qt7Z!inBtg=
zU!&$+MCEt~HS_@I&X9W_rN!MeImL|d)WO+pq}cIKvBZw%n!WROu;d(RUkHjji(bRK
zV@K*qt&uEY_vF~x8;By`B&5l{uSzv3L5CATmOhD6jK)h^KE^?a05{Q&D@vs6Z1|z-
zYxK4G1&F4M_k32iVb^AKDsP?s0ff5iB)N9`CPMgq2uMHme$z75<Ck>2)6%hq#3&|E
zID-Y*b=L7rdR5GgFqZ-$&-ZNODZ*v&X1fxngJc14$sy)#7MziYIlUO=yohefJm0-)
z$CamRP|J_IDH>oeSP$X=McY%swX}VaKT!tjXJ}#o{MgyYgd5EkBd_uoE{(nYgtMrF
z+SlNQhhL46YgOn6tJYSQ9iGwtu{Jt_W(QhZwtLR>*Q~jH>2^*eNRHFG<uoq!kF^IN
zC(Z9CGJ_-QSD>40!m9WtsJlZD<_<e3+8iUu;~DecoTL}a2Shs;<-=(9Wjq>oPwV%<
z;t9P=uNbk%YP4MF>(2|1`ffbHtrvce>GxpGfcPfHYa#ky6x*Zk`1@|Inn|vSDmeB6
zSOBvt0sqRZ2{!!vQf}EZ(SRw3ROkMF^jBJg2tvxa@?wA(Gfh4}#FEp{J|+}z!q>|`
z|NBc>rxR+Qna+b6)YUYQJfxIGM|XWP{^UY}?mCxBEaZ#0b5<7qe|MK=<mJ|V7~#SO
zP{uw5JMjM0xCNYI#D@&vajofC=-l$|X~$V^TJWFXcsY#Z-nb_l`#gi6%LOGH5P61)
zu|v~QZbRe4<bzw!B@T@g8zVF@Lx*k{Vi^D9kNtouVL02wSh}4nfhoyOPH`klhbtxQ
zccLQ?G8}K<?&vkQ37<x?CNTX{&M$bL6TbQQZ37b_Thx0LdjURo6>8R6jqTZ~LsNh3
zaP*qtOI?kd(up;uFs>VOaZ$ZLT4UD*7e~IKi~^)7)PR@qasb{RDlr!Wyg#F~%kr4`
z8}QTwJ0k$-TJ{<u&aJ=lvueiKq8?+_FtX+<68QT0j?uoUSh>zw;SP8R`e!R=Jj6GQ
zFjyC)Y$(HyAR+z@KVAezvpCpeeO2SUKI~I$z=Gl2QO_Xsim>%~;?c1KJEdY?fg-5k
zFKrDUZ-bN>0r81U5h%7|TdEL6GT+Cb8?Obj`_ZPq;}zm@AU3(<V{;v8Lk_owPFM%7
z)ztxBAcms3fQ|_@PNBn9-01v<F+kO4YvKEb>1l+~6c(A~ChOXG{(;QtyR}zs82w_V
z)0Vsw;s`rP8oSJ~^{MCT2k2V}S#(WiRdW^HK|lLQnBdL+CJ!_YB()F=wR2@AvwS-1
z6=1ifuZm34RT_D^)qU)o_)xKzqm}WglWc7JRm#6&!}l1)8l27T*}it_wi;fSFFtM<
z5@mn<VfP+*++7jfn9aK0+RlN&FFFD`-gD_BpJQK~vE(9(-o59|jAo*jOX#`c+v6&S
z4t0@aHL@?vowJ8VHd^?KZ7FC-uKtxjl_bjA(?0Cq7zpGt+Ci_7Z8we$;qH=`76>Fo
zhWN421jX_<<xeYEg@nLk6w~#z4wb1G+i2h~?Gx)i=>5g#q&eT)!c+v8U)tA5y+!R)
zx~CHk27yIKGBs*{U&h_%=Wtl9uqRZIf%2*XBSRpZ^QRaw1=J-NkuHPlwy!<tHq!g*
zp)7V<wHSM&-i|w%-kkQ&RO6{<EKeYkmj(~L5Lx!ggkJ9S+T3TE!NrPWhF;1z_dlp(
z?B02zm})Q)b<m{IxErf2ZKiW)F`b*~`Dmu-0s(1tO@0#hQB{IoC4wE)#g-x}EFDLO
zg`o)|{>hiimF=nCUkmoznI!&(u8~T3NOEHtl$5jJ?|k}1C*z~igcve<Rt(9qweK^f
zCGpxwo(5y7R$NfCh6v5&Sq)T?lG@vz!ZYlF85xp#I@f|u##lTqf8d%>J`m{|i6}>C
z(_7=GL&+Egzy+_0GH({jm%`f>wtLAo1FY)JCvC9|0%z~hfXX=tdx}|=!x6@+s;mPI
zBP2QI+j1T}>8FR=XKzYZeV|#2lHLG%U#=8l6a$JSHr}KHB%{sJyf9XcB3lsp;J4Ao
z3X|^y6SJ>8Xf~dxda%Fu2=~YBSOVX7GSs9BYfv?}2VGq;K<?6`QARrt1pq}XE4<qW
zYl1;hX@(sX;c2!o>+ClNl@v1!z|6t<-6+8bxXwX!#-xYj2Gc%92q+2l##&2oD9*<y
zr5+H`mTKs|XUToRqcm4Qfsnus?;e(DsU==Y79w6AmOOG!|M<G5f<k%a5bFM|XzUG1
z^YgBv@5j1i8>K52nxM2U7~qW<I(~ZQuU#?h*<}GfWVw_v;(;7U{!{1eHUOEKuelYA
zqYaoH8Uo8T{~jw5e{|{&{VLE9Y4*Rx&^u63!stP<{%iZCI-Sxk>q|wNp>dTP<*h3Y
z?0+~%;$j3l-knY28c$CRw@|!mfU_ToFs(Vm5*{0A@sDP@2e_+ZoJCZ_g2GP)zD)7R
z#hWKP#TTuSFpp;wHLr}`h2Y!_lyGw<ymO#DlY+XA8p<fWzizK;X03O>ThpNd0D>@7
z@l-%>tU3+XZD-<D7SVE9dNY@1jm$k?sjObWW8<^r20GdAwQCn{uQQ++uE7w=Q4)M1
zpO0&96dHvdA6Lj1CZXGl6Zw*~!2}0Jz>@&!o1_i&0v1ra!c`apKi}`sUP@fFkW<;!
z0k+*iF>s5?Vg!nuDqb#4?@;7ny?eBzmK+WqE|Jy|jeZBk!Bz}XQ@N6cK^(}qCi$<}
z{o>zVp+z8b6`jaGE&7DP_zN@fZR2Cs{TNLs^#SgeY^wj~1!y>gVrg;9)!}EV-(sRW
zp4wh%5Xy&|HXMTVG>(y++~FVxip~t(hWN=|05mZlS_gH<tuiTcNW~%>T?>?<#FxWs
zI$WVrS}<2wz~>%F<p_(4<+j7&=j+ILnd$S0m2q>7npB>N?{6}F(+U9Sn4K`e0Mm~C
zl4#!mS63OZsm0gmh_zUF-B*yJlje9#hUC6(mfeJ($OXR!vQW|sW8@}UCBYR67{(+l
z$#ey@1DFQ9u!^z#%M3C_xLl}%$q$G|V-Sv~bO1vg+TWT`_66B=mzXdX?btE{8gw_p
zlx87sYkyBgylHbDrX__^jF44|^}4JDI8iWE&q8NYKb|TK7B>o<vo}j%wqtM^QLRVl
zm=5ui;6F(;eTHqWUYV}8i|&zF1gn6ZhFt7ZfSGsBcWTTVWEi4MV3At1E&)N2fk!$q
zYdA$dM`iT_2#7;sS?9oy%F)>US8|lul|F<aQ;USlM|#IAUX@dbK!*ZQA7zOuU>2(f
zL0l;G=nQ*FG$DA3a5<HHcp%15Ee=;DX5#9%UlI|+<GWHyd4^upbN{c(zJ|w(Th{*;
zyIEHt<`r>x*791nrrsf6a6P5UFlH<aDJ9)3&Q0!v(QH*rg+%Z=#1eP2@#`*`hZQ*a
z6L!!YG=kNj&e*Oo0TPit6$4#K6%o*rMMba5@#ko5eHsE@_Kzf^x_k{;iQn#5-9HHV
zePdsli!scC8^}v}-cMZ<FQlPZY@p>ZQ9~~GFZqYjpGtJ)l-DF2PkI{;>^IkQIi-qs
z2`3&KTqQgJNSd~}S#YEpO!@)ScJw@Oo2cB*ZDL%)r5#M6P7YBx3zYM$vYAo$xc@i{
z&T$uB=#&)FC7Itd=~!QD?}W@hxAj|SpzZjh2U(|ahy&}rcXPM}3&dK2?4yhD3ey&j
zc(O#KRin9mJuoD*BjE40zVZtp7pRM?U%&YcP$N3Uc@>vc6~LYv?W9AnM_Tnz{yYoP
zR0sMQVX0!dj7*jtJJXJ51H5o7SFV)ECH5m#*cK9?<zV_tqT^OlVh;O<(~Tt`#PU_R
zx?>mP5$j-cH0aKh(4BY~-2B785VayS@DLAMOR$Yl*CfzDYm*v@hrCh3lIoZJhZ=Jd
zk#|T=fx8!6xhfh@*WD^Mrj^O=m;`U#G=X@qcaO%g%83h)vCt-gLypoXdXtVjT=2b|
zjFo3=fXjkZAw7x-7YK~Co4L7Rj@{^jsZcP|LYN@JB0Q;%f)A;Xs91$Z;Rc(pRow3<
z&s_Ar$`pjZh`-T!u~jhOg(pK}`c^DM&bNd{`;N1@IrPI;HL=y&fKC0mL8wYkB*wNB
zGK*Ld3*J}~x8N1^hd2IUQxf{~#j*bzQLS5|(2z<Qb4mg2avZ>7bR!_j;G{}{0loer
z2ef`_tZ%~mU51O~Lr-DVo8iJ&@gBJk3Y}>nIOX<178*;J9iX@f#+ahbmUK<D(H+8-
z*TUwTSW-hW62FTVd5Dc^!A`VvfsLnPdv6t^3(+QzgGQp)1-cLjh@kTvk6*odC`CwA
zY_E=%_2-cQ&Sure(^JnkKh_urUsi-(U?V%I$gSh#h%Cf?(<<bw#@O+e((3b4LyNmD
z`<W8U3`my}-}IE&BS`frbh8PbSoX^-rCwvrg?$Z|ZT3pR7Tn*L9t~^5)otY;eqzdS
z*-AlBq=j&eADlED9R6lvOIb|-C$ir@@U&N})j5^eDncY|N+JjZ%`<km*1*}=t{d$f
zk+JR#^_9iq<YiLX3{Q)L3R;f#eYRg0f%77<Ji<hL;*Q6~5tg%ItGCr)OhCgZt`iBR
z=Pm8&(1ALmkrW5)+c#(tln!h-KJ<l3Euq+EH0S+|+j@M>tLXni8D`8iZB^+a<^=Ue
zcfAX?+@I7}E`fk`wtM%g#Fylh#B#*UMUEk_K8SrfaT&jK9sTZ1;%7&ip>aB6Ak(tW
z%dL;Z=cH?`ktyI4<oJAPZ&Dk(4^9>nHKzT#${6%`$Kw=;dG*I)sDLtg7f<FVea=%Q
zJYC$EMi9H!5-->KGh12QITGBC&5BD(Q|e0m`4<Z}s(^#92;1<5Z0H>*Z+dbdF9F+e
zp8Zewj;T`9a3vo2>&jb3qnO!qjH?|x*Mf9CptsXX>2Sx_V$>3}FXq5CmD(C5$M8@P
zZD@4-frx^jsuCk8Z4;ZZUgt*XwH~35A+8e`p&h?~=D}kRtCgTAyji9r_djgt8sQpi
zOXGqRF~E5hhS5O;zxKxH!@jqIyH-EJY0^)}Nyst$P*6Y}@TiKVRLfxUEpZIcI0i{d
zSAz~wtEL69GdTvmDD;&X#B2FsR<m8EE51sd5C?^6((2anE9?7wko<~$C;Nxx{`q+P
zVw7eAb2l@;=anD%v^Hs16vqyV3A!S1W04kINglE|#$2dB_o=Z0Vo+LQF}10*Py~HQ
zsTyf`lJen@^X{%BPOgNFp94Uj)vNM;p=PHXE(!Olz(Ztjx%%^g@C0P2<k`H{3WwF3
zMIyD((u;pc4p_hj$LF%sD`T~>XfKrnYL30sX|dB*N6xU^tE09D_(ip~CeGP%u7vD4
zgTK;S*P7S|U06^G6=RT^S4(BSA`O3kzqS|QaH%2oIluh(Ht_|NDZZ_-->iZjg>^0U
z;altYkmi<NH}HvYbM0>y8Pagtv3S2-tVfk#3MWf03X}eAlt1*x-hHF22UCV&C#kQ`
zt?HGjoPOI=1bGIB<zJ%K6QKd~(VGw)Dp3uTGa+N%JVz3H&Q9Te*5cFuNp_xiD%xJ#
z8~8{=K1o;<N<`zk*ys()`^9{c`CGQ76!c~1e88fWxbfc-9Z(V<h(MFQY^_67R@0#l
z?YDv_)LUvR*xNf)N>f5Wwq5|7&J#CK%gNtE7DCYxgE$@#N3d##^@g{Qv!Ifm6lxy1
zlKXNVV_!0iO<3p+52yD&{MoWJpyk9Ud@8U4V>bt2!NG>nEvDj8{28{^^<9zVyD66n
zr}A1tb?#`!z@bj5N){U;Ge}qWzx0i0Ot3dEJz`72j;3;ES?4hb3BrV-^vCAsaQLxa
z>_okC)B+t!)EAueWz?%2X(%MK4@Z$&!Ar3hGDn|&?^Pl>w={NO=hH`8@#*`}`c2do
zuD|+CtZ-${NQEXGlnir#V;=i9L?8O2now=5qefU1j<&?;G%c>{VG96k`tp5YJDtS0
z_xPG7Z_)OgkW6=qk#miN1|FTATssJA>z-wx56%b_ZO5q55~c=(G7<nk4Rs#LJ<M}h
zt?!Tv8E*b1wyH+0?@cf_(iq^@<S*c$nHXSE?d0RQFt9QP6QT<gRRVH9`87L$Y(YOg
z0lURl?ed|ya~(QOl2j;~@56-ZJzp0xHS{N6a4E&y7hg(dMimgW?Hk7(+&6|XqKhe&
ziY;g9N=*nclH0T^z=mOm=DiC*N@<SWBr2-&UPkN+d-=+x!Z^Cb5E{F8@2xT1vXvw6
zBu};fkB%#a{(%*U;WKv@<b5IUd|-fXP+}es21*&~ymAFdk^h(ODLsg-MwSuX>sDf9
zc*{{NXjwUSe(&Cr7bG3u;{Dk>fHOPwD(hGD^TU2qx8IM(glGfD3FWa;@<KY6bKx&3
zNDFx!3ODiiPbio*I_z1*JF6?$GYQy~b__Yy4Z%qI@xBG(i^9Xtgp^FrgD_I6p9rJG
zcRj0*Mv*dSjKyi54(`XD2Mnb6_}3wDXzyznSCX{O^EHz9G=%?heomRsU-p9q0DXi#
z@!`28ICi`*<0jZm8SS6R2+5kijY;HuiPb|h_68|IthCkbtytzB;R~fwPSnu+!!FA$
zZ*^MmS3xnh9tb0>2#+kQC(V&-mMf(5DZp8ZH6tw4s>OwkuJM-f)>e_pPYEV2MpcqR
z4Apu9^K%y&bbfsIkC0_npX1x?83<|UA5rZ&Wba;`f$7-b{qXcmZfi?_WB2pxEGkJ=
zeik+w^)#}*5wpLrwq+nuYB0P4xXQxCd(Qh#FMvKtZakFY22;~yFAYc2OCL)c9KELQ
z5JjNgPw42R<2PsRO07tn?hns|>c4h=P_%md>M<j)!*yZ0AbN)Q^VWNNA6v$n{NY~4
zwKeQfrWhmRTJm2qhXX``JK2B{EU~g49`uQQAm3caC(IfHG;bw)s<5wR;pYGoxNNi@
z82Ea&pzMRucwq}m>aByzQj(MBJ}(|kOj{2GdP^tQqf!-elsot36f^5%2Hqy3c#SYs
z@ml(SJDD}3TWb!2=sCE%crNzX`}9jLg$mD6x+BWL>1+&@*g}u&TIeu&UYZoAwrO0d
zCzH0`U(u75^rsS^l$gxAX@g~d(fF+0K|Ms*l=z0_ce}ObN6j~#>KExqE)p0xZORf}
zShQz(e1J+M=FrokmxGx;b@%sf=F)kNAcl`)xSGv6+SWLs^ObaJF9uu#<B}S#brkpH
z*S)ayfdoPpDNnsfZ;0ZFI{$rqcC`<7IQ{-Z$mYcJsK-x6naE`}^Gs;|ba`DAdGW)r
zVmtI}9|eI&zrwDSF3+!S!sSa`Wcl90fgoD4qCzre2<<H9Q9)TN#P>MFDBM5AZg&5B
zDv+_AnJr+W;TGBPiy38M=gq&-p*)|bU-Ooo4`fLW&#5spGiDHk9d?fHEdTBFcot6l
zwed5-iE;<&Lx=xEfPGYMeeomtX@6A5d|FmN<#%BEBXwL61LM}iOGI)W40ROtM>io`
z1Mg+XB|vaW2AnATceI{|xkxc>R&d)#rZ1i9nZBGtN8z()f^u?&9Bo9!-gsI`_3EO~
zvBD{Gr|7T#FtPR<Vt<><mfDLc`S}B<cL(f()}g_D{F*Nm43#p=_zzR1P7E@(17be?
ze&YE|*P4~#^C8@C%w8HZg&Si_g6JB12J!c81;s~w#@(z4S)}sdYZ@GD%-lH7r_6G7
zMS_$-$&_B}zrBCjDMV<xso{vm(-kqod-pbv={78Ht~&S;y4OpQ8~b@ehq#7GI7qv|
zyOnEYOHnDRJ^gO!G!s{*peGTR$GTpfmgo$uIT-Fb;qn#)N}yK<z7H+>sp6MUD^A}8
zYTT!UqRkThrm+ih$ZG09TH?&fFA96f1A1RS)Q4*;J{GFJx}|To^N;^)*(NzNJ^V(W
zsk{Mw2Vs8=4wCLKj_MYc@zP6&?wQ!^ZpFM`WtZ~oEI0xWSeCDbwXdhAF-1fA!O&IU
z&W-%wiD`V8!O^iObzxPQeqj_I2*WVuOX&ZH#fqk{{vBi!YB=_?9}kkcl*(r%UDDr~
z<ygBNW@u4<Yr_nyo{5_Dy`0iKD5@jrj~|ODN##=iR0JJ^uq?^!S9oiM<E_M>bun2r
zcAM|8yzcYL)6HD@amfV&Rk%in%S^)JIT8wv=$~b58DaJEc36jGyo|+ux_rEC0JK-7
z(_OYv^DQAUtZ-a8Z+|ELYi`V83I8zRp<W@A!v!p#s&HnkkmJT4NK;_x<*@HeVC~4&
zD(f0=c-an6!~2my;-P%P5g8s#psfvl=dIe=JXj5S#t~t>zxXI?iW99biX?H3mJ@Iq
z&9S80{PR<y85Wg3<w*c1$=0W=FYrA0BY6BwX3#vH1Tpeeqk{L|`IMb$h%H*4w2frU
zt~=Gi^*~o+4N|%}4zCC+x{)lLUbhVl{kqQ$Opzz^MOYJwGFUm7kysF~Qw~P+kUJHs
z(W<^<^^~E#X*lO#kYe><YIg9EeCUnJv6k5zd0@Sb8d;q;L;%gNKVEflG{fnh{kc)J
ztOa~S=?Oc6{cS#<dEjMC?+A~uVulenHS(oPPv_N);>jlfRz~HcWSYfG66z_saR24s
zNP1a6%E^kZPSwKw<$f~5wv|TwNH5T@$OJ7prt+9d<?(g6!P0Ji$-Cwjw10Br0N)Cs
zS~q!%|DCLDLvm^qca2Sz0OX<O`I)`fHrHEc^@aXD=r;8XON7pkK1Rd0Ce)hzvM%pW
z4E{kNxNyAgHdybuOKA|L7pEs~&A%{`;7YR_Gr`oA&iIHteHoBDX(pKxHwbI=@%SCq
z2M@^IqOcClz|&^;SorAOrpsnG!wq}bzhr*~u6;lw-^S>jnBe;zJu9p35UN+k9Qy;<
z=^~yH4noas44Q>oDjv-6xJERMqa2Wr$TgL6hDo?y3;D<N<%knvyR|iQz=xnnTi66i
zS1-M|#HCVTNoU?j+S<gXtF=zY$ljxhEBl(NXq-?uUm;ZYVJ02}pxU+DNOB8iJJa}B
z#vspdZH!MEn6*qP(kSU0Uw^bzYHOKFtgt_BSg$5^MHt|*BgAP1#tc*3xStYqP69X8
zE%g62@mgxjk2nLX#S|nUf{=Gi`<e-rtse_%e(2RK#t|(m0oU=gD)1P|CH2~>qsf+@
z24X0XSw+L%4F?<=BuVl%<u>a!_wU*{NajY&7RH%Ip;hs>fJ}@#rwZ9$rh5Jg6YQ!_
z01Ftv7sqBVQ|wbGHv?{Eudh;5l%)L`;#?=of{gHf_@{{}xM#I(M|Kt9{(H+<T6cp@
zv_#recJ?GmF)Vk*&clc)WbfC`_4X=Oy$32RGHRFh>q{v$OPKYr<G~Y8E*4_8JLa|U
z04`$|2PI!8rE;MUqj*NzbtrY9=ePIu0cu3etZ*Ycgd(*C<uZDYWx<flFhhY2&AjoY
zKNG-+{>J{lbCKxz6n57Dyc#<4X@R}czHE&Eic4$;PLE~5Iz_ZxI%3zce#-Y;CEfo4
zt!<Z3#h3gr{C=jl+hu<B7!qjYpFctluTa*x3=sHdWMQ||Kx`jVuSqK=j~8hcsqHOM
zvVY~vvf<`e^l$aO&U_kWwxb>u9QRTk=;1Zo%Vf*I)s4scGzL%6K)obMmusN<88ovm
z8v1t7)x7m-DLH?>ME}<0lFQ7zOv#H^{&Guy45KXhVv8oZ{fk=?%v{7ra4nkpk1_4`
zm!NZxorQVzC8LLI5LbaSPijUJ%rEql@tfxd8bLym)|mA1ukdWs3h{~FVUcAL`(a+l
z#4>0a)!Y35pAe@{323#|Puft*QHkeLJzN@b6-@ty*2-wvbnklaSM~R14qTFcTv}vs
zvl$rJqPxBY{S4+W)2h>K0x}(RPDfhicw+~zLW!mgMcIhhFwwrZoI@?{r&$>g+#6W8
zQL6lw1-aQ?&nN81IX7Y@hP^KA=VR)J{gf1Q?Cu55qqpA;wUF+UsqBFP{BZ7=N&^YV
z+VGoMq0psn`Zvc_{pFbhsYbZ2t&%gZfg+#DvARJGBQ^}W$4(a$PC`e1EDz5n=vk|r
zD$??v^-eG<{48uCux7=u%S~KXmv$O@eHxa<H!DtZzYP6Ex#QScRto(6IL{^wy@4ul
zo77{$%bB0$YaXk?224I*MlCLUgVOj)yz)#A2#}B~&u(Iv<W_V9TIcHy7wF0`)HfMr
zlngDPQ90(58Tyx4bxz+<LoDttKKE|5U1IL{@Ed<Yr^$w7>2~o&G}m+&bovkHhIL1r
zfE(fxP4syuXA$UBFM>PlGAr3t_$+p(*%gUt$MkYEK|{^ykKb(EdH*8TfiKoxDs!6P
z56LCTZFY=<8jH|4bwaCxtBi(E{+aT<um|qb0^d85D$rm+^FR@y&V1agrp4(3i)47t
z?2-I)-c`U+e*Qzd_*NJKmy0(2&dy*vH&f1iiGi4;a0&*lgG&Vx^&73ixSU#ceIr~O
zfX*5DVE>?kop(fFj7OM({qVQ1hy50SEkD5wmI@Gf4@f*_%cRxUeKgiaC2rs~mdAu)
zNO!%o4-pOIJTE@~`IM?An8~R~IC(NT^C*AdIhw?Ac_x2#b2|SNsI_RV(|TK8G08nE
zdv^WV_+$mEIOB_-!>HE8h?3LqtM-0^RQKfGl&FW71<H6SYoZ&IyRQ@R00N)fT>suh
zkZj0ZiTd>9Jg)cl4xU2&UJErl?DL2+s;;(<+Q{tvnK{skTzg%ZR#3E4_hwU|vI^aV
z)fjr0EPGv)PMms}0@kHJsFCuO4gTwyiU<E@%&0T98E~iRVgF~$AuPEyC39?Wbb&sP
zA0AKL^fF^RIPufV#HuO|1xK!cTwc@ihGddKT{7mal{|6lOs7^>rq2&hG^o=vI6;Pc
zZAbRIQH81zM+(I(SaXWT$Gk+;-)}{TR*w2%iaq;k$6aEEU#SM}td9x_CnvkN4Y*R1
zB%x%{f}=7!f}Q+p*n^;gJazy=PJET9K8RV_OKt2`W2e>oqe@gSlGQl>{z5yAKQ=t0
z?7apPk_M5+|JeN#V2{_Ad@r+ytDWdU)2X~H1o^46mb;7gM2F5ca55bPOl>|s$QCaN
zXN2^43f$2MViaAdjR@LG9QU6<J;i4x4J-0{jjneoxv0+YNYEr4ebDC&qsj^%`|{3t
zT&4?hHN72Wl*e!inQbPTnn*t;$(OENjSA*$vaR0Xz*SIp+bV19R{X&BDQiZ!O#`k-
z>uGy;WIX4>mYz>4IJ?i^83|)207sbUcF8sF!(5*-TX9E~J2rAHrGNXT8-njU$Tsu7
zo@6=kRxIqef#d2<lE47>hS`Wq-IQ%JTiVMzmUk<6rcWHxrROvh->Sm{<MB0Ri|<x_
zZ!huk;CFhYAO3&dFf72kk519CoWAN>+@l@W#UeZ2Hi+Yd+|*}096b7gGDzP<*#Rb8
z>qUk<HwX*uEN-i-m?y5aa24D!x&0o7Sn#b+c`cC|;mbTd7xPW)5k_C%ZPIoc#2CwC
zG#t(IaVR}?s7X9FWoUG;=qGP%iXuvu!J|bohc_t!D?nVV>wP)aE8=8^dfyhL5<Ga=
zVUqZxsqE=iP0(qJ<jQ$9SV*mX!=Y?Fa*lt^C*}F){a5AY6do|m)vKQ<mI-QxRl}Z=
z(&HXW8!r*K0O{3(BGyD3GV)aR=5hW^fwunG8}so}k#h#x4W5#vLNY(DM>P&Nw!lfc
z!$N3-mvv;+OzPk8G-}gIsgGDafl5w$ugaVMRT2U9jCkR;=}rSlo69cn^=*tHkPGUw
zrQ2Ni%*ezibp@NSyg&C6Iiwz&8{&Q~;`d1lu~PK`nKMu1nC?U1QG7jx{}fPbR4Gr5
zUJyO^@b_++MA-K1yt%SHc-3+(dr4Ka<Ajm3dUw5+l)Mu)f3mOKSj7L0FL|;k4n0dI
zmL%U-j~?$mlb5bbFjF#GWD(~p7o+7n{L1IJih3c%bJcVURk0|+jt4U&Wg1x-Xy;!i
zqxVHYSU$bHV}^Z|8;kk$Vq(h5esDCGo`H$auuw`|0_K`6F1~X(O5TFTL${4lXDa@U
zm&xDgmkcDPHBqCcsl)+dh1`NI<T_L}=R+Qf7c#60&7cM%ik$%EOpc(UKiXbc8nQnc
z{zxv4@qW9GvR~~_`SA0}<B}3yaLZYT%1+3SiGK!G!B4)_&VEmcfA3tA-L1aE5_H%^
z7Yu=9fy%a~343DXpv4${AAs;H_O)`}!Gu=Y4ytQTBrC#kj~gFR|2L`tp?t1IgaT^t
z0u^jYdLk0jdI_z%Rrh;m2yDM<39Q)ixh+e60(d_?a-AJir1SOopfq>w*+qx0^27~B
zN`=`kmY#A-XZB6Q&$;K2!y7n-9tA((zYT6@{N)Pr4UHFdVM$zTO$?)*7x+{^^@l0O
z;5Glw@0x(Ion8*}*^y1}tF`UpC3OJ1Qt;KJ-qA3`N@n*Z?A4Uj;K&Qh-V1|3QAJO_
zR;t20<F>{h;aiHEdUQs-4lV8B>eQQG-34OKihS>ASnGf9&rzXYpdns?qHw%QTd5wh
z#^dVg#}w1!UuFex5edus$rl0ChQF`#b?zR_-+Xr-I-=)eDft2aU|4~<rb<-tMh0Ei
z=;j+{xMY@%al0k*xJmF^ql?&=-#ZPcL~VHAD01zJ;F_Xuh|5fY7ti|J0bGe3$*~lr
zKmFip%7*05Kcsf1z?RdM<cHE^WMWZi^_YfL;NL;Y1#Z0b=5;+^<-fb|bj+_1B;Jt3
zWxbGu%8M%bEDb7t*lW=pNc+M{7br^Fw2x%t_gfAdjx1cG-v#yWO(o67AMF$6OzOUU
zUQGWm0-B7^L~u5P{-W8(&}1qnLgMYcQA`$1wdv=2NXE{C{%X&y&HY>ofpyh6TzGq-
z)*tQ<N$2{(A(=Ki#y#02ZtX~5y>ek>F!%K2<wrT)e9w$+gN!PJk1rK;l`=o`lK?j=
zB*i82cun&HwPqTUc@^o60>m=nGe!98Jb3Uz8+Y<H+fNtx`4T>x#EYvx$A%Yn-wvv$
zxpTC8s}sS;7ZiDeA3EcoQ2~{&G1o8E+%Ua<&;cN)OMTLV^PSA_b5+(2e}RihHukWY
zAa?V}55ef>{!Nj`f6aZklzlhnWk0q^HDs!q;HJn(7V!K_$*UTn5juccOq{pCX|8kd
z`Lw8Gw|v^lWSkZ4waW15g2k|eXR^y8y{8**p218{f9AlT!EFYSym)i2Q~6JMDTU0w
zEUu{*dN>(P#z()*LAAnGTDL7CtC9Ak0#GlpJYSM~S#U;9`|Y?PjESetNma{jEO(YC
zR_>dz!Y01Y;MQWDW5>vstaxxV48wKnH4gxRi`g@fE#r4*L!AC#B#?*;5E!%st>q1G
z>aXbZW&bNeu^Y`m#y@R9$Lrt$p%f!=Mi59T^Az#NE+To9`s$!dMq3_Gw*W>g&Mo)L
zak$XX?_&o#@y!exCU*FqJr=FI%vxpk&dFp#gt^gP3Pr9wmkiC6cge}G#z0L(!JgNY
zb_`Y%P2yZ6$**FMJsl+V4&^ShXXF#JW<DBHnMy2ao@6{|npw{NbF6n#0yqd+h17nz
zSf;5oy+mo%qP@z^yupgi1U-xX5t4o?lQot+|K|m$rPYq&xIc5(lDftk|I9an(=&H!
z{&=_ks@`+IT2V#KvcBZ#n!V_R!)5T}ZhYD=R;DVh;xf1K;}@zc)vnWu{YBv5{9pmi
z7ZQd_dA!AXGle?SSrg<$p0P(#0sHntkB{FD_WIvj-qQISFnJrNmX<=z<YIN0MYIlX
zF?Ud61aD!ssSz$=48<>M>?4UfN;Y6oqD#^J0F}ve^mdnK6qU(DX9;`Z$Gk>Bj=|=M
zdRstQMV+Kh=7meT@?+FCidk_q@e}uC)d!8*?4~AK$^_DnBi1^+Bn<O;USR*Zh|$HG
zgnFqArhs=;iOanimb{aOU(d!jtD;}UZ`ss7|5Kh_`Y(m5mxDqqQ7-9jbR4&n@+ps{
zvOxAg^3S~|<6Cde0Z@AYlQK%wSES^0<ydl*AflFfZ<R&n&jvPW7?-yVgZ?8;6vUaz
za_Yd@qNSiyM$27EGgJ2$rMFXniJ(&xw(u~+W)z<r>(p(KmXPq@eViReu4jS?27&|%
zKV8oyScv&{d;)<g)mq#BZ{MH~&s=j|b-^P?GvnP?Bil;f7P5Ov>D48Umm7e!HM>Mz
zQ~vLf2x=?cf>MeCL>qgY!*IwEh`<g7C}Xo~W2NV47TqIwBCXvEN7*%>s{Tpt2g(z6
z)6%|{Z<(5Kcyof=X7ofBTmes>anX8{Ca(?xs;+2<=B`LS4GYg|yME&7dvz>sy-8rb
z>C`G0)m|7lXLy1we8N3>Lfpg;U*R>~WgzBp6_b*QU2l>vWXK=c{mjG6VmmPK+5CXb
z!+D~P@7Uh>n&+;^D*gF=w>X>-%)YDNs8nVzF)b=@t|Kt6sNHCrt!Vpd?d{8@gFr3j
z4m!D1v$2qigq54yH6oOF2v3rA`q@+?rrjAJ!>SkLb6#njxs+>c!t2OE3CJbB!ygH>
z9Dg((q}^g}r97eO2>Z$5P|a_$)<cgQXmicPEk@dD8olbTJK>LJAfwQP=;{^yKu_91
ze56vw{Top0K$V3b^xZFG>R*xL(DI=pgwFL!jG1s~OT&u)=Nxs!Iwkjbv3f1Q0-U`q
z%qz&0rOd|d{Lk+8+x)9{fUB7$ESH76pnigO0RxjPfl5Z&3X@ay3ZY+rtu?NWe{PBF
z^&3K73uGrzV6`|Trm>Y_^A(n#T#?4mF@FdMd6>T=@Y}j&`Sy4`&8)HHC19oS7{Uz}
zsx&?Cb{gH0FYw5F^~=Tc-g|7z`#XacwGYqFwO$ntlB0Sq28|fn%(4XTNniyg)Fs|o
zVh~L<#3goff%HJf1v9EKU|~R6f>~{XSy>SnoOez#RTlIm2_(MLr*4?|5OsG<SFu!b
z`l{se;nkZn&)BbmRd8DokCReR*e=UZC_A>NTx=iHqhVjrXT~j$EAPbKt1Cs?d(ya4
zCjvi>>yr7q(8xH=Uw0dJ^IP3fWXN35A;z&GY3d`wdaM1QDE2OVbzboN)rq1`PtMSh
zD!6>8*Q=j^FN{BafApP<(vEHQZ<PxdR5z#~{T+b%(+okE(0Th+*I4pP=q4VBkK7e;
z2bPWri?n<1@jg}{HAd=YML!Cl5U;jm{20RX#&9#r?%3&;VrLLYDcZvn(%%;%yD^BA
zO}|)*od0}8HW5}L`%iqeUekN61bw1+sf4z;i6#hW{St3nvf3uZSdxDgGgQo48wePM
zwibb#h?}OJnOS>_v_DQc1)&uh9X)oLy8@)b!YjVe<)9+V&Bewbmey6|Ra@?x3i7x$
z!N1CTwjP%YdK~|jsWPiyLt@sXzo9M$B6P?S|76W6eg_zHlAbPA-7L~Hp)L}+e49zF
zxUP#9BNgeWmoMC5VTW4;+(@y%X9<%>P0Ri_6)@vD`No<Dv#i?kuZL0Pc!2vJ5`GGy
zkvmE}Sa~buL+9wmkH6f1GBux5%IxR9d%Rec;*U`wLiukgS2MoQG<jP3l@6ds`W7V^
zKk34Nx16O|^>(V^xTSFX?7D>O&a6^7=c^`lrHjd(J(~x>P9tXOKUGUka+zHh!yaoj
z0sD6edcqy}-;NFRIOXqe58eR;rFt1Iw&kHu?}z(7Y#91f{j>#w0zRcPwwkE}YnVdT
z2&MfSG`HDHjJ<zT5fH@cSR}K<KVPUZ%U7u+?(ff2Hb8|m+*_=H!&N_tx7yE@hNeR3
zEL;@yie5}!M)7PE%Dot#A1$A7e-DSbpBr2y)H(=nO3=xpE;W_G>0h+(*s!*_PD822
z<^SMcbX2*i*D{rqUyIkEM>E<m*|7T3urVd#SMMXY?q@$rmG)xa^tqDn_KOvwp08Z4
zO(|&jx&A-)-ZLnwu4@!L-A$AXijs5Af@B2A86+qoQ6xzQ$)OuTM4}|gpac;li9{u|
zh!P}7kR(EjK$CNnrq3LFo_Fq7b7!XN-l~~jUzJs!qW4*QowZlqd#^R!_)gbF3kNGJ
zr#cIXGD{mvOU!;eIT*4s^=htV)tPtq<l@>dXsZvZ(9!Tw7?XL^NEJ%0-As=wf=*9B
z53dHTp@LOL0GlS`u<d$_QX+Ts3`JP-1>&yw7k@{PUZMaNQdRd4i+Iz8!!&MQ<T2Zy
zvFbRlccF3f!b|MZ^sCFqF_J2EYmhWsiVPSXD7=E`8$R~ErRAd1d(HWq6~$A7t|cvb
zMbMv>6fYkTq}QDv-GCB&(bM1`PUvEmdHfz5v`qlrRORw_1m9@SN?mn=ZNWrU>lKtK
zgw^&%i8=Eo2+>IYB@;tY)qN%F7}Wi2+jlb$)0k#f%MQ5$-zg&RKCmU0rF<BSH=^6v
z5+S#;zLQDNP?I^4v=4YQz`k(i=vh2Wy?b_efzHjy(oS;w(t?ulYLsRZIKhP#$qady
zVSK9OcdsYs3r|#-VS~QXH0V^)l%SWo`kFt7vZ{v>?U*p#g!CBno+4w;!$>>;GjvS^
zPPef1b6|s~f<lk=9J#6W$AVsyWN9bfmUiv4LG#N73%r`&e+O;~#2v|)o!UNWB|Q07
z<qU+;pL}fRD+)GGEqoS=+e!|Qf4NQ6(p}Fxy`rBh?VDW2%J|k=XH%k$dgGz$uh9k8
z(ZHg7L;kSD8HYcgT*~M!bg&;~&C+f>Opi4_rFzn<gx6JI?u$G~v^08nM|E?DIT3>B
zeGP&rFufApdD^4&Dzx|<Kmi2c$2kC`;iklyYyIfW^a>8OO|Rl{bIwi9qoB=%0Qu3S
zwguLqvtdKNi<;1jhx36oA?<pbJ7I!-z0v3&JxT>n&XL(q3DuW`fJ%-1=?#~!1Xwle
zl*PCwcEZI>R(+If8+T1jA^x~ns{aq8oIh*aWQ@_I1j0xK23lx>E)D<mlOvqDz>H|O
zIw==`b_?T7c>gj1|M_zfKZP0>F|Y}=YuSsu;okPZ<X_57`TN77=6Rire+I>kh)ekb
zFVdbeDO0BW2s8N0+tdHk01T=6QIkbfGZiodN$1~J&7jSYe_=oe{62l_fuTAG28r>}
zc6?CvL?Kj1#X`wntObP5YmJ-(hW@3;ycAd^Rqv@mc>zdD9Bb`d3m(b!zGE%25IEA3
zko!BR?-6v~=HC}CmfE{O-Ih>dfES@5Oh#_v%=rbyF?44Fttol!pV86(bu`EM=uBJL
zV!~xWGymoJXX>8)E31D|C=X?9gQNA}%cVQgAfg>S<;vD*K2GihkyR<sL83d!g(<A2
z{zcs-Qt$_0>9jq6#nb56qbwq>T#X>|l)Fz&sIYHB6p7a)z7rh;>BrF$Mf|w4cInx{
z;oX|1-;py8j2bwW2K}~h&4U2NXQBE!cES};`7_E0n+cAoIT71D7^=u*cqXE1*q~~D
zCP`LVc$?_JXT4A>sT-~wA0%qOV^G(<f0<^8hn##{%2$yAlXlKZPVgts92jBM@#maL
zb5B^ro7RH)oV(k4I7i_>P(ctc3MNd0!btt`>KT?DtIqkFUe?h$b;ZouV~Ht;KVttt
zgzC6}JLen>N#r!A+iH`#LcVbp6|-8BD|NH*Ik(GmPr5~pqIX@O(n#i?CVHl`YqS(I
z#hXce1n-s%jV5S2ONLvbtb2C2X-L0XV#RtQuy~32mgtWMJ@`2|apa~;nWRS!ilt|#
z1Qn<SK|dCp+L{UEEQmMZn0}U)xP}vu8bRA~qymZP$LQ1YGrEk-wP$qZZ9P^DU6XUG
zIBTkUcpz|FQz?ZyI|}!F{QTigBJ*g`Yd%`eO!!vup$viU3UuGxiEHGz*#>S;q~qS4
z|L1Ln&w}K8iPfjk+K}!M?Yd<Hm4Q{g5A)T>s-HuMWFAd@%~!;49u4y>Zg<C?^mv8e
zIyXuaU#u=dBRkn65}Qnv=UN(ioTd06FkR#eL84DXI5~XDwgb#a%Q#$knB>A4kIB6+
zd9&8i%3hv|ogr1k%y>~H{w6R!=k1X5L!SQ%Rv$PtoBzbXWc<l2IKD3F{;B;_ZSs0u
zU(r3|DVobx#@jERO7K?KJpB`rbR#_N#qBOX@f<n|!6J21K@1!YyU(UVq7aJ);C<5l
zwd9ih`fbRW?}q;V2YdUstj&MU4f!8jGt~<`D`+n)Lk3Q?={R6Uqs0!i9tPn(qwQO5
z`Gq-kQq83HI=FRGxuGAgLoNCW_Xs=lW>dM#3!ut=(BDhmUNiJ#^B~7nf5j?hDH^3h
zzQc*_-(MuWS+_&&LY`o?(BILek2RptaZr84b6~FIt*PR~+ljs4kHm;r5}_a&dU)7@
zaZwt#j)pj>cmBE*uOV=YwtcdOz!M<~L=far&5&1WvnaRZiSy`;0NEhVIZ}*tevwpk
zF-Ky2VN+g;w}1(Njd)y4NCkS|8!8(LjTo7H+DG{U?nM0xVWH8?!LF04S@;iv=ZO5T
z|0^*WAh-^jxgbo)3dd8`t`H1mBf{vL*(~6B#Orx}HU!LsR#|yR<ldCELj*N##M@~L
zxYNiX;GI0)Rr7Y!;+QMLK}h@JSjFho9Jufn#>aW^+WX)+k>apV{ADpKP*pBhx;zKD
za$$|7OM9T*#XDMS2Vq~!Gfp~K`$3cIYl!c+8s4K8M=O?bBO9Tf3=OL#2}CL2M|qUs
z1EOD<GDOvc?LZ!3B{%Ng&`|K;AGAX+)Ir0R4+r~<wkh8$S_mSC3c<?B1?tpo70cC}
z9WG2dpp_BIhK*aqPDeP?aohibWA4>B)EmAatK8PXjBK}ud+m2MPhxHwx|Z=^^JbTt
z7fb3cX5a1yqi)`3shZdmDZ8}<hCu6gQb}cY_xmdRxzU@VJ-w!_7RU3?WmLx#uK^XW
zJ9u$L`Ol_0)at<TJggN(<%<@LAhJ%jK}57NBT9z5#)n?~n*Sy<W@%w^y!_0K4i9+o
z;v%J`Y_Nq2xNOav{~+kwzL;K<*-;um$<^AOTXyJwj@`y3YeQFJ?wTplle>n3rYXUn
zq5e`D<HI3xbH07f_!{E9j`X2FDh=l4qIM?>LRUK;o)$pgL}ad+wur2yd~x9egT>6!
z<Pd2GYK8u_M_M>C*AJiRVmPm?+lXDt`{3-m4(e>2+SQ49$zp0qbzGftEwrp<7RL-2
z4*nwa143w=b3g7Rz8~*78~F8gThZ5YdkTyIhI5(^`BleJ@M=vKNL<__Zz+boYk1NR
z9@ybCcki&hZWR1ZC|nq(v6SgK=l7L=j=j&N4o5W(GxNd1${OVkl{&?i7|X*8bq+o9
z{O~d8;b@Ng)G;-Rqe~4F1~yo?^6*T)(^pG*NY0HXm5EWFedW1VR~N&+3fMLw^)|w&
znjeoBF#K%YdVVuhBh&O#GKrFt^vm$wraZG(oO^L6Nb$JJ;=kW8gtMhuLJAZfL_T<9
zVvXMIy?sAo<xr_9VV9yPf%hwD$D(qrLHgz6^;i9P!suHyhVI&azB&rew&Tw}j<uJf
ze|ikgDae8OEH<H53mK*|j$>a>o|^_vpNT9^#c$&m2Nov>PCXg?y^I$LwqK3yTyrVJ
ziSU^Zv+u^8bXY{a{)@#J%N|lG3oH#AC$6YUIA{3$8w-Ca^I9SQ`SAS{wlB4m;-%}r
zSSD23kk0ZK8;q4WsVDE9jJJaQY`ZDyadh+!8RlBHUm-!kV9OvQI*@eF=FGXyi_!k`
z+LP(7@nt#jVXM2p@74LPzDjtnH0Qgj&xh2zL*p&It=1X0<NB9WSR!jf>h=X3^)d1X
zvsZ1N8Bf%p&=UBO=^;zi4~XFhv%s~FII+cd*Nr_K$D#EbB>$JE@>d$qi6#EZALZ&u
z{Gz?TxaILGGp2Ox@L{KTMg!_YDp;6{I~DB=Gq)wUF7xpkB`40_9mf25^}-E@<F8~M
z1x=hMf#VCxUwU%2PJLG!|G}|ekr#ir&yT#8&VDDe@-K(^?Okz~(?A=M>Ofr}T9Xz_
z1hcy_ySfTXAim$B)?aE5DY0aauZaqV$f>wjd)?S^kG}?D|A$VjQD4P=wqu{)bY2>^
zW-v%U*CG!*8Mg&h*G7L=;umvut|ZYSY+nX<bxYk6d-zsw(9M&`DvKFO66CmayuUCh
zHr-Wk;EOZ;dB)ErQj_;yq}X$)DewL0rk8*a=YHdHaPIHFWIQ4=H)P}7zdy~{Q1f$J
z`Kyk(RBUUiBc84D#`ocV@bgu00&2tB=N$fGy6-ia#r2;I>i@9Z`%xVRgXiaXX?6`q
z+qVBs-6-!p^xf48b<bo5pNzGW9J_V0b<uVf(Fey{1HuSTH$V5Gg@(D&;A00aYr~(<
zZ;ClSkn<NiZA$m|_QjdgwQ9<2eQ+Ghjp%dr>%`6!Iv*NIfrX1D6{~zTQdwa9cQ2L!
zoHFzOP-H)9*C0i9^rwpWudeu+`#OPFsPJtyqM1mgt)K8Z>H752aei;d4{orUpn$U;
zRn;R0QuTJ+Ej<ipIlhb`x8uT8F8J?S>A6@Bo+?cXoE*(c53xA@OgC+E9Mj9TG^~1m
zcDL`3g6iYr7}DzDbJ<uFs>=&tZ=IWCH!EUF*67{ofYh_TQn#<kow$A>3TS07T5Cyy
zR5AavRR~e}N8c7+&8ZBlM7)+B=cpGAFCN&L;O<T~+EB=U#_zP0*TFUL$vjEBQTeC6
zA~G$n!*B3yA@@Z7&tGndsh(f|)g0yKVn{uT#PMG>hQ5jCtj2GfvRKbzzwCavQ?(F$
z#PefiNDWeH0`ibbJG~8wMCSj!4OoNy<+3+6Utg!~zNLI&c}K-%<EQjHq_nr5ue;|1
zP_a{NZNk;_Z4RkbHnUr$ML4m&XqBN>Or_SbV@IsGS*<;G>{_?E%%W8C;MdpZvxNP`
zi(Hcu{8Ce}aPZ;Wd>+P$mQ_BD0q^`TsQnOuVw*i>3u0pp4~;7$b*bXaTc**rBcCBg
zbMP(Lqq)`Ol&@;B@~8a%+=n#3CGH>bC^Ful``9UYiQw65=-%EnF)nhwyTyfU^-J$X
z$_)En7oHQK4JQH%6O(HXAf8xW3P0DI#d#yVx_CReaz%|r$f{#5nrp3L|KF{8X+p@r
zxvIU?$ZBt&IY(>R&dO0Q-;i5I-g)oJWw;*c8>d3jej}E-)NL}Qso!aGHR06Wr1sU7
za|qimW%zP?9<NYw1SE}?{&zKLIB1lnKK~a#qc>IA!cfqN=G;D7Xp9a2f%M-4$Nyp<
zs~`nYFDU<yeXM#fP#m!f5$f6hhQsvVubflh``jG-Yvi>)N_D8Udm`k3mh3?74<0<w
zoq}msG`OPo>!KjyszIFqxLt9OqRz>&3;<Fe*gIf?F->Te@-s1b6M_t7&eH|?y>Za1
z{|uv%tXc=SzfTYJszD76sLoyW-`zN|@9Cz%|8Q-&+}SYkDg?~rGZcB0VTrZhjL0kX
z@g07<c^xl<X<jI$PGntRrkEiRCai`X>JCoPG^4~zwmC4P^@U064ZSV#k=>C<MC0ZW
zaR#>z^4`C5^XM?Dv9WWXsHGt<Bg&{#&xq5)oq;C^H{PKWm%5IdFj<MI>lxy<7(SF3
zjUcHIX>SC>_7J)|@-%f1yltph#?!qC!;>Bpy&R!NXi?4+Kj+ptciF!@JExAs*lR!$
zv7RZ{=;OO?oqe_;2xN}VR3I<D38h|tHmyt+6#4Ca;rQ$5u;<^nUfNT7bJfBvYTmIy
zv4a5`C#ExU4Apv0m5O1!grZtxUTcJJ_w@&_ai0$L2XC;^g%FllQ{f7h7Usm2gR71f
z&#0k@M(FSC$xu6RuF*b96_Z6PDbz%e@Vhp_lcU2tzs%9R<ua&Ez!HU`EEeaVh74Te
zgW`;_uK&#DfF=TT?&#^Z<JTAFqEJG7vR+XfU7R{2&F3MFR75{GawW?ITGjO7E0o6_
zq)LcsB|~8zc__vr_|L*oL{b2$6;2G^V)nmYD6Y{h3C;Mr&_H2U(RskbsJw)9C2<Fy
zjU~2f1jj)=Mku2-Vn$veuu#7s+{-wK1<3vbDUoCX8uxUux1*5V3?4x$EVvzl1-v-O
z5n*B+D?_kZV)qNJ6c$!cD{mdBCUt6w4^+utGl2`>_P<B7?W+0)Cqw#aJMJQ43o2Qr
zy^o{{qi(*KMTKj|tU)yXJ4+Ar<^MY&)PHM)|E9qIKNK)axR&|^$`Y{sN3<cuLI=}B
zAq5qv?UXQ*Xah7E+7wNQ{(*93Y5&5EamOf|BpV9K&}=(m9@=$R7qM=-1vtpq1rsMH
zPAhg~?k*qYY>>|fs-@QFoG6QfO;G=yk>#r(@e+6onuF2)2V9~S+3Ulsu1)Tf+l_em
zT*~lk@MZ{$;?_a=p6#6D`lqk~PiCXjBgwN#S!_$&b1mfwcW2`crqq(}{b1xtM>Zsj
znz`q&WuSukBeAFKztkQ&J(Z|5S|i4tq0{6}c^_>2Ou`XJ+#!T+zt)YNL}`y}v^q7l
zYkT!J7w#?}KnSiv2#)pRyznv*2uKs=f&(C=466wv%zep`L#tG~Jkd6!6y3wP8%V3%
ziC7e2uUxWYL0T11Nzux7Ok4YGd|^zlPIUg-ssByi4L1v=A%|J6KE;}U;hz&=tndE<
z-%P^G`5@1A2rxw%A>t!1p}iuF-o8JGVCl4qAS2x?u|~8;9kMmLQD2u8_GQ-m&shK?
z-9|3?+zt;p?*)~JP7LcUWGqq^dz?}J+qp3~QS+(=ocT-FCDchh?{Y+%KTXY4&F9{R
zj*-Q^KLW`%HFwm7Q_XHKfoR+u6Qq=&<P$`>qc4zVx^^IX3)w6}7-1OU+d?s!Un{OV
zyCXT>is$Iv5lP#4J+QaOs^x6sW}PEMvb8mj)$iE6F4t|O!_py*Jr=gR>5HqX=l$(Y
zXy?}nzna+nID0Qt=bX^WPG!~MzL`TC4T^tSmR>ldl*mhh#AV_;&aEdM8PMvje#Non
zb#@i*{N;&67A$Jybq*o(_oYcsbQPWAQ8>CB9iDyW#cRgMs<ZLo)qvYNS--YNO+Rxo
zYNdRmoUSV^qgq(;nV(`ly0BLcA8aS;&9Y5rSEf^a#P99?AXM<BmWFa%&_INH>K||q
ztrgh;ZzO_}^lGoKh<*+BtXUi$y7pt`2wJ1nK^$8ar~tVOCBt@js(vsZvWg~&rJ+Dq
z^z+4rG=~r0T6^^MwRa^+dyfaQ9r%WYIm7G%u9~?W^}GGlcXRj=lI?f>+f=e_`HIw$
z`?2xTaqHOIs<iO51E41-M-E9{bpVk>hSI<~)RFi3m){cdwh3kSMZ5f~Ic<>X^LK(c
z@CfQeqI2N*pQmfjtAP7-Q`zMX@uYqPug)a=aa`FBcll1K3b<ohDp_qNnhkAK^9^33
zq*zZ4ci-n{!SnLpAJTTVKBl1Y_V-<9GdZmSqIV5#K>^I==vmg4v!l^quf*QI;#jvs
z#(3`U*(Y<jIiYiI$oxN_EZ)PP$>go?kL{58CCz2Z<A?j-9`T(z8+o6VqZ6f&L0jL0
zkHfPOI6?oPi_rxnu39|Z1@{kBwZ$P1Vk){&X6vmJ@sQDN$_`z>Me)0D`v<XEWof%t
z_OG2j)kt0aN7jz$|DD|nza>JiQ_-;hWNA0=_sRt{p<Fj#1z<&q3rSrAg$CHBkyEai
zaH!w(n;lL|>;V*LcV0nw^#n)!6aqpW3+6W5G|IIKzM9q?Md{eQ`DdQQ|A%=JU#?U0
z>r^?xo&e4U-h)rk#HK&yi1AhEEOzcBBOnlQm-`eSFK)Fn+us7U0pkThL2td)ov-~Z
z6TGKCcwXI1uNGf830?(58|BJ!cdv{=%p*;&V<NN%E<-9PwXPp59>UeUO|#BYH+j>R
zh7|Mg3%iX8E#LTnI=aMo)8gA=H(CP9OaXYA6H=c1{=n>o1`iI;Yf1-TsYg<vDQi^o
zlKK$SL*Z1Zq<-*RV?rwFE5sC+ZjDi+#pq%T7aL-0KCQm=m)M5a-t_tvC=4Govi2ji
z*Eo6FnzKZYHDuig0nB_z?35b^tU^Ea?~pW$_~YvHe*b)~)iL2-mTAFVzS$3MWP?8-
zj7{49{@>(`pRn*HKuNV6(Z=jM-Zd(NUw#+j!h&Xs<dY8?xP$IC{c#x%LJ{y2ME9#k
zkh1wgR*K)Ipek*1+#zYWGiGq?kO)1>^5trZNaRsT<IszDM^2Nhyk=LJw&o^Cm1GO}
z(nkf*p5=QTxxo=)=TgTPVjBvvL;BFUHbVYo!6zpjGXEF{K@}P$K-#F0s#gx=&$;W9
z$ybwm)I&*XCAK$1rH}u8*88a6k1u55Eboqn2t0kmUwpW4&+!mI8P84~dAbg^w|))^
z5rK4`-0-RVGE};)0<qU|(s3KII~8L8J|(hf0ort@?-u5m{E`fYrmpj16-hGUZ)CyN
zWnN5-JSambz}a*+ax>(x%BZxfr+;<mR1#!<q-2-C9VivaU78JcL&+Ei=kw6y(vc^I
zsB<R5T>Q>jYX}_zNwE&Y*>+O5nfG?gaSIdp!)veZwMO(6Pq;Py-iqO-NIF3(W<jp1
z>%Y+c+d=&Qo&paz1r-1T7ou9vsT<lOS@Ds4X~<H0DCjVO(#wp}kRr_Rf;@H#;QPUC
zDAn(G9)xy%LJy-i9EpokoFU_|Dh|3?KpQ${EL@qO2k`n81DaDGpK8W~g?DGNL~G_i
zX04(dS92bPl!s^vf<^)gz@p{lv<`0b0eUFC`41Ec@XdL%wu*w3KJYw*VsY(PhN6}N
z2J}o_)HjQBe`Y%yG(X3PVon>2WCB)y>Dc{-Lf;YzKd4{e)D4$PXa?8geo!zMnli?#
z<Q(l(BB&z4I=MdKX3X;ER0I^N@y=63dXUKN-|5WzHl(_5Qf?1X+1#@AdM%H5zITlu
zV=e75i@H%dL$(>AG@#A+Ea*Zl@wPMCvUl(kJD$QDD$J-$W{_Ia7bk<TNl}z`@f&=H
z_D35~S20_<Kp{;AeGROYHOMiSZWh>SKrd2PU9qt;X_y4qU4eOz$WJaT74lKy2@jNG
zy$D4kT~=y~(@IkHVTJ&uE?zl%Dtp5PLsPCg?~2Ziw3e3*7WF3B7K{bSn(2y~dzK;B
zw?e1@zOJ^6l68AT$E9i=rVxvqcL5TGm;mkrj#)<8XmRV0>aXCTuv}H#)}c+h6z}y5
zdE?!C6D(%z6*)nS+mv=sL<^fmx1}*Eu|rRmlMz$~5Lrg%6G`*=rB~*4asf8fF)fBj
z!Ng?-!n#~<=b;@L<~zF4!_X2Yj2Q1(pG*Iy+wr1yo4Z_|a(VA$d{om!ks?UpL<YPK
zqXObd+we@u<=K1E2t0Au+@s~}`Fk1M2X+X|-G|rV;-P~Hky_k|(XyN~MzRUGkli%2
zTJ$N6P~=ZkgK2jeji#9=$&Q#i!8h5R_|g)cE+*R4^)KiUDZX!tO|VSZ5F(UTuzkg*
z@`*}Vt|?515ur?wwDtQWLV!UI`TnHCVF_}|Gtl8xClWSvK_ay>12RvY5PT|OF>>aq
zf{XEJpKwKF-PYZaD{)YBc(?+V*^0>O#`08Qjt6@?qwIyrYkv6aX!6*hC>L{@VO?_v
z1gb=Oo&oKKLabC``w(d4jeVB~a|YCNh_LAIN_U+2YEIw4h=g>^_vt!uoB4;p4EeFu
zgC9hw6k4t5mC#*JJo0i8cCYZV5)}?UzeXBA)5huVjlcxOA~E|Rsb&mQH^2Q{Eg57_
zObHUgh(zB^xf2L<JX&rUR~t6LSjBGfND-7wT$s%NM&ex^k8op#)<l3S36)Qx_IrE<
zLrMj|t+H>7|Fs+(TLY&w*qEW3&t(sJ-(kU1zPqW*3N=VuA_5Yhhew`dho{ZpjN%Jo
zj+s02nm&|Kx|1<e?ufDUiRxr*2k$EHDJ(j-5OC+yQV||<2`}X~G9NRjHyPpOhc~D}
zm@KJ<kiXsaZP8dH!SHTmw6J+PBwi--B(kZ-_8y_M+rrz#b})4^g`i(+q@UIqDK40j
zwp~Gv`MriRSelBN&yGe4;DhVyOGvegYGE+@2W=vz#D@kpL)QYh%fpW7G~KGSoZC*Y
zh})%&9>BBMCLaBhw23vSVdTuF{6$S8Rh)+^mr6+7{01seElgVdYe@3#3TcSzOE+}!
z10YXy@EIGdU?U<Ya@jFfjBqw*`s6}vlc%z;O6r9sn7!`vw#d?Gu{mXgji4ldVva04
z=~eK}Zb$T;lTY9I=&Df1PtdmCaHYrMv6of%yJsCTxG9Xw;_H>~x?D7=>M|vX$8F9n
zMH8ZV+6;csi99-1h}J|=m=HG-%_4S=JP5wM_Vtw0c^YKUS+DH+2IUr6qqaZy@^Kqh
zE@lg09avWoZjs>B(Fl&gUDHY;#*^TLwy{uUMZeVWP)ZT`#HgQ<8<MAxKJeKRV`u~N
z3|zEbvf#|QVLni5t0FsycUyGV3opi*kk?w<A$MBLgAODqCK+iYFw~`IdajvdOU-?a
z8P>1kQ3DL3mghg#Khlz@lKCZF9L(yV!&raK)jGjT>(<=5GP6bDx3oX(otcqLE}NIQ
z5BP1;wiTW?QXd<DH7rW7yuzs#Ir{=D9g|QdfQ_J?Fwb!LBE~?u+-zqW-HWE2LPQHZ
z1}_y66Z=&y6JkQ^1B_?#zLAKBich3ee1ei#Dy@>WdQR9&)Q)ub3X7gZDo85W77A~2
z>3rH{=p-pqBBe+hf>Z=o!6Bgy(e*o)4;gv1fEBS6Ari823YQ2hzgNnNTG};YZr40H
zxN${d5MAe3OzwS3VWM`5VT~x;5*f=2c^X+uvL2F{`>_+Q;Ynt;keYT7D?l)3Or2c$
zrD;JUN;on~QbovY@vVVdU}zZKf~J$}F5ucH37IhF8X<|pj!rj61P79h`!A^GOjKo2
zGH+ZAP5qL`1(aIGVrfF~H5qc-SGKpP9lgF$8Sm3>hhu*AM<$|_jTm`LQtTj&@X58B
zxrSiyiHo^JW;{Jj%yubDI{8UjiG9`uoadw8WTa}SQr6G!NGS|llMxdXc*|iw^ylpS
zABYdobE=tJ5|A#)GYbo%5|{MDuwp*wu$0*iUsdHoclJj%kZ{2og{~8o2c3VH$ZX!Y
z+;L6xax$Uqu`cE@v<Ha!_Nb=(Eu9zHjs*3G!RWdON@nqoJBY|1PcFu5j@n1hmQlRn
zulLEn>vA?95_GlCDD|Q&M9PF9XG}i2zCUs;(vL`aaY(&r5xOn+L6X=Bz7;dM^3g05
zn(y}prvH!)^(MZ<y`H>T@tq%4tc(zTnvrUU2(5#jMc-wgEnh&HRsV!4Z*B3+%z&4+
z<#(B2l<*{;8G>ga)LSZRiX*pg>_3^WX84RAB5@;^>W5JZZ04kzDV%>a$81_HYxh`x
zq+Z$yK7LhP0+LpYvTJ^o-34#vu2U*yURUor5f8z^#^_wie5v=%ld&=Ev(!NvPFNo5
zs(ytJfDGc4W@*ArmUeziwVo3cFzK$7!8HM!B3#rNn~HWw47ApwrNI~Kw_tp<jB23Q
z&P&(*u6HPhs|tne31LDO`%+yUuY%QoUZ<WT8BY6bYnv)KM1wL}nZHf*ievB(p}GvE
zLXOk0AciycGM%R1D;C&}zG_)g(D7THv$rri*(BcmzSRs0xM1gv?2Yb7;+Vh_mrGq8
zoHOojlW<w+?nule^~>S(vl%N(-HR2En)pCN!f`@rL?dj8FS(wM)UUjQ80kEgD+8=t
z^SIYJru?bTbpjSj6EVT@P*E;Hys3&iG4~4zlvg?>;Q~60TnKuqdq=LovL@_ROy=fe
z#o<TSrV!|0>Ywtstm&uKQ9-2J^f_7rI0kXLc|8vT<DZj&I5XeD3625rRP|3<5Q`-i
z-w|@GBmACieh#kiDqvIHh2YqT2{!g*y=X*F_o=FR9c_?c@C;F2q2J(iv>e2{U3(;>
zNu}0g!^NNCIV@p2pV0~?3Qbg_v>HbgVw5_9`H-yi#g+kD@YDO0l>~#2`4x^L2-O2s
zTN8}Sath?Vsl^fVw$mgp<b5bmznaU}fszD-3;q<DGfwCgvl!M%(NceczRNK<NXqXo
zr~tR~5xDu<v-PBg^|Pr9%OGCL`}4NLFILOT1lRlX1YP@EStfJ7t@x<fQDI~;qiDHo
zKPE09u$u-og8FPqs<iS>2Fye_-`<0)%f13e*zV2}x;>gE)5Jp-ZYSFgS5&j`Zb;M2
zY>^sbb9+-n#9yH1M)ldP?~V&GdkAb^G7^MU7RuL3ieH-FgUX?%e=~EQA_BAklFr!o
zC($XIxwP~;)`nT|4I7N1rO>en-ytB~yZ(Xwj<IJ?k!<$8Erd$?`V4CObLhwI-zg;~
zkm?dNV5WIBK7Gy4PGB1m8=;KLdVfiYd0b#Yy)*Dh%qNZrk|K9!&GJ-t;7U>kx`Qv-
za9wu|?mO~g-Dr=x5lfQsi&kT(lo2+r$1+KJOx<-^W&ekXswu_iQP=9cs!lQg-=6f|
z7q)eBusi*hgwW;;eeJKCEOE}?TDL{fEYwP7no>9RWxJZb%_wY8$9gs0v`)3OFiMZ}
zA$ggmK23`ejxAu0kZp@&HDZL}^S+xtG~(;G+}<*@u9UBZsW1xdklS_i53Y6QeRPa}
zUIO7^z+yos%h8hJL!*8;2=zAG1C?*jl3X@N>#K5MDbWpR2J{Bh3CmH{+QqullA8};
z<JvcyrhwQiJBqZW!k*ARVx`WSXkGP~3CN;o6`T^&`P~IA;J42;1U9)-EhMsZMb3I8
zuZ@Q3>}QcldK)w8wY5H%R>Mc4p>2yVUqz#t01Caxx%7*ANiIGkb$CP7s|KMwwS-;H
zD*Gef)J*f6f!dMlhRr8PZtq_N*q;P>2F*JlQz!GsLLo)@?^dmp-gZt{|L@25n3Sk?
zzDp$=PO{CfvHK>u>HJ1R8bDxv!zKUwyoTx!cwsChBtYm_=x5RLaN-*{g;@IMbg5$u
zmtoJpd@ab{m@})<3a`%+w;sO~-4`?+gU(HDAt(*70xVp&`1+3&h6lj<ACab!X=L%*
z#S6wc3S){?R@NrM!Svt4B;9!J1yyty{S#?Wg`8pZ?S<mBA;fm9wMzCJLxFUwBy3z`
zu%(68=Osy>nds<$8P0iG!EnUtL+Ny*FVrQr{c;qtG$1`(kPkAKGoDjMvz4{h-w7+?
zsD8D@g-881H6y&v577?yvT?bW1Q<_$L2tbe8;t7RD<SR5MSO4$u(?$gbKoLpzxt;B
z_6@~12`=2Ou_Ign6}+8dM*B6LBEihi>M&A^?)p#@cLI<;mZmK7M?tcU?wC|P`5X%k
zy~sk^i0YVz$lIB8$zu#mK*>&ubr_Z=wm_68HAy2GHpksRlzpff+r-bo{ymBdNDpAJ
zw|u*jmY)?%(ne4tEX<i0hq(F!Nr3;5pu+3JW#88kbEm$Vm|Lg#3&axJ9#rlHQ-Kod
zff+RVK1IZ?8+LE0x{}yU9H#*G#K*El4X_t+G6KHpVhAj-5)76B-MMz(V<-UKQ^jBL
zJq^(f>eJuDIXhL+%3S*waX8~5Q-ja%NXT?wS(^+Hb{9LM{j>9lRTL$++;tdfAiESJ
zn1oWhs<1^^%@!f7=0j+|+8x7CA3&KR2>CRwiv5VCSB<tro5FnQHtO-Hy{ifnu-Wwx
zpzqg372OYy#XGia&;!v!9Y0WHC^ni9V(G&<GG<sSRdYpxojKEoZ;D4CXXis&bAQAD
z#mYNUKU=+#*S-WVq^y1nPoIKu_h)f57u)pL_)%w8WGd#42q=A77Z1)-4C}Qx-@5gX
zs!HifAP^wRam73rqNTrd?e%S%PHayY=1M2rv72`3%o+%TPm&xP%Gp<F))lXMRvliR
zS>~|}#jud<BwQ^@DdF&jrjzm^XqA4qXA@gzdXkiri@I$3L((Tl!+fm1*C@rzYe(PG
zD|O=jpF;Xf_+4uH<p)75U4gg#rU@x*71Ne9%5HtL5*R%Wp+C0VMTzSNUZ}lbVROs4
zqBuy&?oI#T-LJ7FdSoPOqfCV<ek5EDn*J{BB|GC+U_tcuC3AwEB+HMZ{ck$aw41I|
z>Kh~}z>QZ28l{k0#K-;&k>t1QndjF~YNiUdggXbl7ZXezW#KA};SlNe{Uu(`-&Gsd
z?KD$a$vtDD7Tlz*Kg8RbY}m924*Pbd^UvaLH`TR)Y=*BY-ZVp<>1fzJY|BQ{q&k^l
z{Fg+VjzBMtkjexXU2nw8n{Pk5(wpl*j!v<d!r4&O!KMhCl7XAiS2gwmOI7iTf;?!s
zO-3UR`Y2WXskQ2*fiYLJpA_i6vBN*nHle9f6^xLa@Fw7Mb((L;q=|EN-ay}uEc<%F
zxT3|wg@r+ms10N)a;y)i^@iwbC`JKfQd<r<Q^?ZmATkvUZsdEK)$h6Gc-i&y=sU1G
zUIqg2;75i<9Fjj`=eBbr=R#i0CA2lq=aH!8V9VeB)s{MR3=Oy~5*624o3yERsv<4D
zrzMJHsQq+hKUAAuoJQYwJp?)e+Jz!RXQS~-?MWr`8#z-ESBji<pWc5_<QzG7KcJmx
zekIu1t+E;L_JPiUuDyOAno1b3wh+^Qm_$7^pk$h=sQ{j@hhD3wxY#?tgb+|hUC*`3
zmIWhBSdnT&dlV9F?>omQT;ZowOLELvdxlX>lL9{6Z}bx4iMwY-*qS$OLx!C;T_XX?
z07m9Lq}MCB0__q@!TMVR9f=AqD1m_OACwoy=N^P&cE$ptPfSfhp@zvUF#ZF+(2igB
zu@Sbq+>0LymQLy1M*rZoOIJ!^c0Ar9+ctI-7!u3_3=)0>J3_Fr_X`M?OF^#amlYQD
zPaZ;3)UurA-6^V=ASZ_5*H@Wg#?a(P0SH#aCQlgN@UWSEFYK~&)$@~RxniaQKHTht
z{zZh0Q^4jTzr2nIlxu8AYx@|kC2-SBbst_eD#Dgk$;OjWJ@+#7D=HpUOj31si+jHj
zCxa2_{8o4~N6fe)&0Xv>aaxI#ZuQ5O!?AuE;M2_`98&PtlE&|y#QWg)?Czv+8>}Q+
zGhA^_jQo3Q{;e(P*jpL2I-n=<^H+_|4r14I(%yvP*`6Pu{7_n?T-CgxRyKxNCd%^%
zkp`m*ufM<7e%ze2eO=1^5m%Tw%_z~nJQPP4Q4$1bA69YB#C|}DlP@_`;vb2W2Th6X
za%CkHU*>3M4f1$^g;9&4TpT|*yPAE5blR21_e)@|3;t)&rI)^Vy&R#9Gdx~#7`yko
z*J#+g%hWah_u(NeoZ%T1E_-p{Hl){ycdB|xU^+S4)_Y)W7>#UuGWG{d#s68`TrG7G
z_A5z=e33JId+yR;7BQSE{o=EIzB4I6pHG<Np~-HRhV~z2qIN<n(n|KQay%bqE&A7n
zxy#Rj8?S5E)m`4^A%ywr+vmt;bo^@Wl{;yM|0(1pBhhJ)@c(ET?J86EX-n$Q(zm>c
z=f9kTX;$+<rojPsHcDnP+;3FPo?PeF;>U;5tG>JgGTCi3cUf|W70-sq&}_PKj>evF
zO1p!^R|BX4Qa1u^lX3n9(3DiJ#U9FIzmdeYQ5}1*P4p&dZ#<^}Dx%yG==@Ukn@STS
zGeE6U7n9n4cO5+SMH3N9jf8Y3KU!Bse#MrsHC+y5?3|bX8rfv*Noe=i%~Gv-9B>O(
zhJdc5311g>#nkl%T`U82)WuZ8Bx{imu#t!NnDyjF!D9t+l_$n?INRq?Mj}O_L8xKV
z{Q5}afV;9WIFz}}5R2?yo&7-VIpfIG!=2cqb|oLDTuFnn)wcOCp^@rVU@P@!p2Bt0
zyE$%n!+WxH`G{JX+&yDDn0<RR6`qQkWduZdXB0uwZG+vki17=Z`&>=?-fcIsNoO>V
zQ~gA|iGJxFa&}UGc16FBK8Y}5C%>ck{99V{oF>X21tOT$os|IR{a_{68Oz@_h(pJN
zxe3kaj)RDh&DRlm_%aLs0dRzUK?+ESW^(ZaCK=}a*?0F$O+N82g7%3(YqGbkJ(v6`
zUYV^~xWNN~s_;yZXFgo8%wYloqK-S*eZ%E5US$%YxJau0r16~9XYF35Kb8GOHkg8b
zOqUl(=Pmu~3+RP1qKnx+`W=rBJq>DC1HL~FQp>rVZ$&s`Y9g*it3<YAEhY9Xh+P^;
zV7d&5-m>09^4+3-^)SV^OA;j_si0K~gPKRDp<{XvFGBu#yk5Mv$@c_byqF33L;1z-
zKN>`w6t|d%acxYR9iQx*HVF%y4(0-iBm8w-X<XO+M;P;=G7m<D2=UC-s)uJ0m@wY<
zlfnQ6`y_N4vG+y<-FpgP^~+%P#$$&OMx=pLC~Mc<bedS*-z2De->-pz^Cor!by3ES
z9tM8-Co97%9BiM_2if_dYJc4c2iNqyj*?9e0{3Ur0gx%Zsh?HK?$IIMXCkNtnS($8
zI(3*=oVr(W-$7NZ%u`W&uhpsT4>4x=dchX8ioe6mD~PF*N2j^=&{sKcy4GD1)B?!Q
zNuc0I8h<5%;n=Xwi>y<1-emf<Avo$SG{9x&1P{%)7+<mqS&G6X5|#4MsZ)w0wfg#W
z?TU?Cn3MC20R_)zPJczhZ#(6n5s&cF=q8yj)C0w0O3O;4Iic2j*qOu+Z^X-)z}V~y
zOti;~gSQ)r?<gqH(sMq5f&rGke7tiW#4s;UGzWy(ZP<z+i>IvyH92-s4_SWx;NmW_
zK!$eL{k_)j-e^}iNEpQAeS_4eJg1HVpn6WLVPABKdQRJ7`JFv7_OeQSMac{93^%hq
z<Z~#9td>19!WLz6v?tUKYJaMJIe1CQ?3Nj$ZN4;+mo|luUyw){3ScvQd7WxT-y-2R
zB6_^JRqPs#eHChyB>sny&gg&60vxtGsZ7{vU~R8BQ!VE(D(^}Dpi&*%;vDzi1)$7i
z+I<DieGB=T`dBG_Lzq-h@iXvUBRoRGB5PX-Z5aKS0v3_8uAn@%CC}*~W|Ouw{6d0j
z0szp7=#Q0nPw`CgQahH=ecjbL{v|M`C7Q|i1MRr7T(&seTVmVlsgI6-Ek~`*gHRr>
zj#)cfJk#XmNvVb{lGqMcJ~E2CWfVYeDITP@zwxuMUhcA;oc+gm!4|?G8E^WR9Cnvp
zmiltkUd+<4z<30ziajw$pQ7IA#+t91H^S`lLO{Bu>G&ELl+bJ;?z$~^!EOu+kvFoJ
zE~=a{J;oItxMDO^e`Cm`YDV|Cz4?DRp9yczzSJtK=pA%DJ9eS;OSVBB-?G~p8$__T
zOJ;Z98VO?fjH7~_-aid$PN%Tg_jlOC`qkgCw?FgtdTUSDX%;V&L6`Q$96~y|3=gtk
zCbGl*eWW>l6_T_fCaESx{KI%^0olvPtOu{Vw4Vlv*zNTXhC<%o{XkmHP4_H!Ewo$q
zv0nVDmqAXMMW_V%2eU#=9<Yo>t7m6p2^Y?AEWQ2lrACP$)4>TiT3y@%v-4BMW<B{z
zs%7$2Q{t~DGIyB0ajN^Y=vL7aj`sYe&}8KEhjmvz8uC<B*wDl9daFvz$9FJp(e{Iq
zBxLymhWLEVtV8UZpgr4zcMeXcH)FqeUvaCC+vef%Xu2OxkXXHr0}t_Rt?#j5%(X~o
zUO6nGC9+H^U_SQk10c4b=-}1FeKMpjV|Z~YILZZV!3%U*+!BDWyg&Mm+!S{d(`u_7
zr?>iIOJYP{28)cyN!7G*U-GOYLQVBLSPr%uvijFD2ML_qu6UP!^~dkY(~e(;n5Sa_
z4i{0HnR_7BI6oE?B8_Lzq|4j8rRwF9*yQhucfa#-`6y-xKf49&(3h%nmXkv+kpc@W
zXzUwj>Z;@2oGp&)6vOxP9}uY2M~|<3K8gzZdbMacD5nx;C-E`^C+2tPzx%p^A-V;h
zOnhB+nPQt1V;$Yn?DYI{&RBd9QB~W;?q(Q!T%^!dg}}(5V%kT;-cV-9qBsYHwX0f)
zEt)RKz;P7zjOjI$lKkc-(lyt~hvQ5*844cPOZ&D~N6T_&I6%q_5>ovAGVtvYGO(Jf
zUweca=4H6KsQozCodRW=>PKI47#_Ez;kSC9{+II~ZaR?s4QiD16IfxBk^(dqpKP7n
z$q7X#7k`LUrly~Yd<OMU9~2N`zw{ug|L|RlbGd1mxaEoRhHs}C3M9Z?`W3^6&V+W5
z0(kpL!y*_ymYaNW`BNCfA9L)Ra~VBLIbYSu+?c94Baq;U_eub8g(@FD4?vFOVjP#{
zUw67%T}%xp*_l&9u|`AB2L{VYxaR4P1KY`Rr>oOkp1j^nztq+&G2~E(3mW_8IjK(L
zEtW_+R5GlamVULCM#Vf+^2}(g`kn%B(Uo~2R3MZtBj4erK6_d6!Ypr{k;>>f?dRny
zrMN;YS9IK=GA%s?@$ubGirFK;c@)2d`l8AVgW^LrVUB0C>Qt(&&z1^7e{^<fG?#0Z
zrW}YQY)<(0Vy86kHx&MRu{s%>7`SWiEoWUqtMPm`aTq?Mo&(l1r(3AHj#TeQ#a;OR
zl@m=d>cT4Dp|ov}G0>l_HI7N9^Tf01SS%m?Kqp4q>rp|Tu?$W7KAS-nr;A?azKY3X
zNI_TwK?!A`tFM^7L{0stG$sW`1|Nd^Wr2)GG~yZ+cky1>dv<K9#F74_ewV$7aex<I
zC_ykMZQ4fD?K*>FrgLLcWKgaNXNuV+{~>UN21OPLicNuzoqz0YKS8+?g{fKS3IXfJ
z>c>*_Xm)tFsv_UkeOKcUS0EN8M%_+w?bY~XiZ~AY5Tn?cc(4*h;*FX$(q6h85QN`=
z&M@^tSNbTv&`Zsg1d-nTO?Oj&%^)EdEfm8Rf1#d!z^zFb@jU(FSN=C{_`-fD#Jee$
z!4fOzxYmEjMO`JxlWL+)H7vh4%pv_k53;wrpcd*A36};WxXv`1Ks$v3OglIaMA|_K
zZX;KG<sGui?6*Hcs09gaccAW)#xJP2C?Owtn{LS38DUCp5Xh=9jh;OjJI{~^el-OB
z*wlrfFe_)1D`8tFl&j1CIJIc-ekiECa?#@PYd5^nk5usqr|GcHyzV)o>o9lug)B3J
zyN-?BZ|#+SpxUot_yJd3f-2s~kKn~vMi6h+0cY8p6}|mXOhx~oHT!Hsf{PR@gwF95
zd&|Ud5+&g>T0z1>#3sjn9mw#qJ3PZmGmETsG&h3gmveJnZEVTMo@~(ejfR9bjlIuS
zOba@4v8xy^3xeKb*FXexsDeHyYpLUGh=!gV4DF2Vr~#5;TUP>56?y9<C<%8Umi6TU
z8J;PvoR}vuSsn{Mg?0m0`=zDC2=L@kTIo_h$;q{<Q9V||og7Dk;^7HB!U!WwcV&CF
zv{-Qd!?o`j(S#2x7WUB4?ib2d=d@6XliU@@09oK6DjxZAz8#|I;5epRzr0(Shttp$
zP+~=`)sSJ>1#_O+gkq2rC^Zt)p_}yzz}TrKb)<iL_FlS{3_6ykW;t<dcrq3XU%}Nr
z`X(-wRNdq5rFww>!gJ|O^@;KT#AaJimhs0alD#M+x*Rp~PNttI$)``_(o4w2#G)R?
zkKubeR&tZpsN03jdqvP6BHKu0;5I~mZ+)sSw|tvKib31w*DaqcYMZBSoOMuhhQ3|+
z@aX%e=Nt({A!>3jUC>rQknCy>2&Pbv9!85DBlQnK7JFDV7ilrnzh@I}Lo5uxF1fF5
zerd{nGg;NFVQ!7%DJxH9d`19&ms>hbVu=Z~2vCNr^!jI-LD2TaGHItfkE;UB^$#VH
ztB|8g+i4~*N!qrkOlhTdMU~jz<(m%L01^lEAkjFVFy4>l<#+H3W&`~ey@`f*QanmT
zZKcz_8hg8bukk3IRzpNxY&gd%@Ntk0=|>m4c@oGO)PusawMa=4Tl^>`#~7K-;JPhQ
z7+H|%uaxo%$c3N!T!~kBQfsmF67u<6kD+fn53{}0EVVR5q5M^fKg+hZ(;Us<trs`e
zApFca2X-E#S)yA2BZH(3?bqb)gt)K3yin&m^aROMB+*MwNrjU@M*Hc)8rl#--AnW}
z6z(Zgt_XCO=S%;BHV13!=J9dfum~Apu70C8A0dfF#w``ZQ^dQ>-SrW$qr&cFcmO@p
ztclrBs9pvS;pxiFq)sUL1q}>@#G}J~3w%uNs4(<e4BBx~%u(PO;GQY-7G$T;Klhet
zLe7M_>a&o9#~Qxw;DQgM_a-}>`O#wZW8`zAfFd)AsTCya;~u6$y0}}g%T58+TCca?
z6!?aKT0R0P3GEQ~ZBf4h`Xihol51n_8}G5e_<+pIY&jKtHc?9r{`V|H9TSTQqrcS0
zkjRdRy&AEA#QPt#5U4%yGG|DRxS)WCLg&S$5f%Nlo#0HGm=VW(;z~lrIX4xxt5M?I
z-h7?+l>JB`Ux{uB>_{_)J*yCgGD|<9C{tnh6ST#&o;w{3bOrMIvAnT?xLM)M=<RA2
zXoR?~p4E*rMIW7jQwm$MEq!~l$hoPhidmkqTQ446(1@M=C^ecGMK<|1G4ZZ1UAIql
zxboqFf8GKecDCy+i-ON5Lh>(m;sp9Ks!|-0&!XQ)mt+{%yr~NhoIUOKOIqsB&i=XL
zywtyU7B)MyndiDyzt_5D8j&1k)E=cxA4YsEpUPuFh7%fDUzfb0e~Xm40}pzNzvu1R
zj|{(XGSIBAF8%83qwKUIex4HA^Kzwvx;B&Ysc$&V!HytdjWET50*BjCKbOrMpdS`Y
zVBgrf`Hg;WjiB83e?(@W)zC+gNX@GZtSZtH8?ofZrsAV!>5;XG++v;AfM;@I0CcCH
z!kcW;Dym4=`&bC5hk1*qzdQ3%3&FKMCO{l=<bq|?chKO<j1xuO4k&~XLVx{>=6BF2
ze(KSEcxcN7cR}LZBL43;<bEKl0JC(Qi)KhzGcyq-p=8Dbv<1e8_Dvcib<0(a-rTS}
z>!=hl2PE?q%pVgIs@ACcP{ZGJM765nJ*A#d3HY!?E8>`T<jS!WH~TZO0rcbh^bdE|
z-V4PQW^+em9x6CG=xSz()<RWQ_L!N;tR9K}5u)$_tU?3_@KIr6qUz%S8|*r2AMQI;
z?2p#bUI9DcxrnS{6ISTUDy%@|r_L#JDL)j-XquCD)9}U@OSjE(&L%FXX!KHb;D$a3
zHdwm%+t(sh$8U+g`#1>uZPaO;UfII~7?UZI%sl_-@m1&Yz87^=gi<df#%5Mk5N4=W
z0c$4;lN`df!-B3RpJtApzD_3=1k+@etZRi(l1jv4un&~|1uvXR??+v8Ayg44H#^;O
zq}!$%1av6HdzO^*nNcR!CPUO2N8*V9T5&D3C_(iOa~|;_i#StMPDT;GMybh0$6eU5
zEG_x5sL9C@<<14P7)L7ayJz~&=j4_ZfCIXu78m-tBqWYQYVUr0NFc?>L$E*#9ACYM
z%jCSXpHXLzm%4{L^}~nWd3>KREDnbN3VZhOi}lY;;N9aX?g<bpFb#2Tl?vHX-$+;!
zyY}HF=ao2FZ|nEZOdgiXnpK=_QGE}%-Kx*)Gtj`Mf7duoz+qNJ`W0Vr(od2@t!c3v
zH&u0b{#pIJTCT{H0z|lqp15DRv_APd5>dz(QO%gZZL7#SS)U%qmkFS;s7q)<mCD*t
z0m0ur`0l#zEw=5g6f*_gXWwbcgR%|hy_zn4OE}0AXR#|(mnIN3MajjoHk-_rxB!)r
zxtu2DP}!TS8jHxQon6cfZ6BU`=|6#QsfD=l$I*&*+JOyl<C}G#3cZ`Hqj05N>xk0}
z-q~18Sgo6SugcB3a|C2Dp=Jd;NE)BrGcnrS2PL1_%@L3JVK}6AO=NseB{5nYqdDa0
z?;>DhcwlW)V>lEhzV^{m^=gz=HS7b8J}>qHkE|KKtR5oG?R5QNA~b_TK$LYtrHAz(
z?8Alr&>I7~H`9x#HBW1j^fgx4GC<OT+t0exw+fD;?2v+O@QXYOzNA2_YTall?QtmN
ztovi$p><8f?vc-v;6&@nTk3(KpTxK>hHUZVE&x>|JIPoAtG%27_7*xvn#K|@$8g*+
ztbke0mb?ax<FresLH2S4KB=D$AfgJRE0vw(7atBZMgV7fv?FOf86@Ip5j$iN?4btf
zrSm?46}30*a<*O&0lG;m{w1hQPSKlFBw<(9mEMn=)bT*{nLuf1-#Q)cG%R~9OuR|3
z3H5}PP@g}qrNT(PF)^A#q$08K<WLvzGf#i0uh#xX)C1)-y_hHm1PFMyV=+7<+>>S_
z!PV2>E}%xB)l$}OhvBvqH*1toy4~EV3kUrvFL+HEvRJh=w}zC6wGg*cf`>#jZuYS?
zfK!sp$28e|z!-g?d|c`!Q*E3ftJ={WnEfo`l#}z_@>SNAw*4%@8~jtx^Z+}aZJiRG
z%{k6QwEXDIp^E3IAR`s~bO{^rb@Dv2yrq_QDhoJ{6)F^Gu}Uh~2OKXVrDN%9OaOWg
z7@8OBD{{{Tq@GMjkEo4V08qDpi<!EO;Ei9(&d!iD>T12)^3?<UUaR!e*LoqGgDDG$
z1EQ?T_;pE)A@2(WW>O%A#S7g?RmE$u48V2;M=HY_v6BpF!xFg<MfcT3VJI&y1`mLW
z%KL|}lvNcPDT^pkRDN_H7f5_fK!f3mn|1(iInL-K#OIgcGC`CDutNiJg-1CVsn6|u
zPJWcq^%~xc(<=P|vLADJ3$-{Ot~kAUD0M2@L?He|ckxb%s!ud_vdRwBbAJQ%wNIJ`
z*BK<X4gP4%Y4K>AfgP3Nx9ZLqqRcGcTM-AaWj(iBDm7zlAplqj8|gzhd0tWalnlao
z*%>;1qCM>p{|8fV84zU`t$|K4ba%HPNOw032q;L15<{aPEgdtUlz^l(C?TRCrF0B}
zgp`2NF@i84Ln|RM_w_sH+<SlV?`6;0E1$LYdiW^IQ_xxd&UA@?zuH+``P47og%|3S
zHeGy2MpxKSyBZ|nqc4+_j9Ku}RiUg8ibJUK3_-F`!g_1<%2o!lEFtC3BppUqgkg&4
zm>N@=pi&bDnSnls$paTl_7phVnE{qg*GP~VSWX0XjL&FcZbDi8CQ99cob&i9+}`6z
z1_h%oDRXyvMop?10Q0yJtS>kTiXcBFD?KVnW!Z;fw>85LDl3Fpr51uiGiBq<$l?>x
zk#V|eYgGv6{gj>&L&I2kkhog57a{hi<i2fObCLB(qVNO53c?E8Hy085MozpZ=5uju
zrINYkyZ^i^k=9Q1akW9rU;3dNVC1#^4qO8ql=X7oov`=yVV8dohpz}&9j|nkID!Hn
zIVquOHko-7^wPYFcv*`hv=(~8qZ}8H=K9&mf@WD!K}moW_|+h?BrjpcMrS2|v_=Fj
za>W3GHXK>bJ51NROQtVm(M|8=Uz;-sTIWaR+=t}qumA+dRQ=(pOQ0sAkptxGa=?eN
zFb!Jh2YN^WTWV~u)F3Y^7bYXN5#>4wkaZ|kn-(g(2T$SQWJHu^D67obn|@=CKz=c)
zBu~ncu~(kdx57i!xs85?asc`8g}E)H(J$OblwsM^-Cky*Yl)<6^bVw$ShiXhfyaX$
z;@>gk0&dI6dTcwwu4p-gYaFv!_zli%Hv*$8&;Oo~L^3kL0#SlP^3%a?s{xN1B7h4t
zdQ(q}by?B+PLNO;d54&MniGayLkhtqc`LLB<v_lazMxaN{a}nx(b^1@IDO>Ie{Bl0
zhM8hrj$+*B)E&1li|H@9?A8m!X00NR`51NV--8M~`EOO!CF_MXw*-R2A^A5ynk52`
z{%_ATA1SvzK>30|NS>oK&Zx%V8!$;$K&#?`jZdT+^t5wcIaEk%%>c8%pKn0y4>`R+
z>fmYaJ~IZMW)Ow}YtyGB&)hN^u*8j0pfNZf>2SBc+n*l!+y7mFh8`&!ka4wXx|!MC
z5RiK2C+zDDe24D99Uv$ReVuXpU#R&D*hJ(j3xY=UGU!5ZQvt=v(C5OSH^lyrF_g1#
zwY@{{<gV)|Fmtg|uZ%UV%MW=akbV<{^-Q>SC5X8vK#y_N*jGm?Olme4hR~jpq&<Mw
zYQ2SDXb9_FmkNR<hP5%x(tLG!OEN`v;}2&C%*&ak(TvXH*v^|2N%bA@?$LT%&)rtm
zUw}4fq>U*;lKXhV==ZOX8)d>q#lT*$tY4T~LhYf3Yf3$0vUJ7W2D(TUq_u@B&yoSG
z-J58aXnxXmC3bUrC2jkkg76EP)IyI83G<X0o%^s!QS9DLX%#)g=MGoyD!GB(8WnXN
zkial;1)c(KLCI>~Imc!(RH6lc_P@}f@qbz&b#5qbQbIiKF!4s%=nUX-Hq(CoNcY0k
z9!MeQoNweZqj4A3AO&=%B+Br*tC4K{IR%`G(zihVOKs+*Kh_jjMf8XVUach`#+l&k
zBRzT1+~0a7-UogothGuFs-|{fvP^tHd@kafjl+f_bI?jG->TmhN&$rSdsKkl>i?69
zVQh__0IK5yOftRhP`UI{qbC3luUqKr5=k=W{U@0H0O?iPpgH0VkK37M4WB!xnmD%x
zR!x9bEZ$G@>MG8NG2<o?W;kKDWD6(H4P(N|maQIwB*{)iU7{)>3Se&SDB()zO_T(`
z$+c+K5n0C1pQ-c%6*}KsML)2CgfMH!VLpOK^$e#8EAQvC1j{mj#ZTqhe+bW#wsZn}
zI&qslm=sgFoKoRz^na=vbs3#as8Spu!v<^C?j_yF$mc|<8@2vZmKB<TGi$?6lGB?=
z@cv%t$h+JR5CYgXS$6OZF<Vi$neH)e1w6WI`2}qwa5Z-PCZP0NE?36QUv4pfJOxB5
zUd;u!3Zj%OM6<mKjGu4)Pvjy|a-$xa{}6x*1sm&|Wi9@rZhh6ymhp#Rc~!|L#f~~R
z#P?c2fylp!+$hf?2egHWkLq`{zPu4(lY`KmP(Bt7dd|nHk*%pvda%U#TbDtQsPatQ
z^Eg_v8r*cAYDe63?R!eL)2qJ(FugqR_0g_jG4RMMwh*G(b(J9K-E4BRtS_2?RGt(}
z2pEO@c^O_!*6&5cb;Rif+l8qJqe!p2YKwzGG^9nI=u1q+Sbeun;BO811uOEX*zUl$
z@#=EGTUVOJTThN68&ltlc%foL-)GF=jIK0PtxbWCk-Di{4trqQJ+}1!5N?`OEJW-1
z)jbC8Cs8sI!;QFM=%Z;gX-S4K6M_Dt+q1uz1NbISU0t%0K2<~Pe{*txBN)8IJCWHm
zY2(C%LYZG_amhtv!tQQ-ksv0;9A^G`f*r!`WAk=sNx;&bio%zrA-M)lc{`?(#4-8F
zPC(X2Rk*GcqBLuXoQLFFd`veYbW5y@WjmSw^IaFlf@TCloe`dqzL6VWg22BDAOn`u
zxJd>8A>igeKdNYUN$@=+Tm4<K2YB!nsu1Zk*6ZkhT4|giBYEFY_No{C0MG`OWMV7+
zF#--noc_A|S=!-eEetV=6_~IO!#)>KDP*XyDBNwYE?S7i=?Ne=CqcHHUv1x}|ButS
zJ^EjsYH`}Q6xet_VwvZxBa(8K6g&d#DjWH?#`D$Z>5Ps6^Gm{Q<c-zMmDToDuR3?Y
zmkT0+FkB56XT{1ZC^m}<btSok6(zZFM=}hZmQX}+<sCT>!hLyt#eFGGjc770-*T!@
zvEY<wbk?K4fC4y3*}|0Q@r68YckF-*Vg1jSn!f_0GMsNTOdrE1J=6Vv^8!)ZEQ{iw
zX8v#`OZ>uP0J7vHp?pmpLZbMyeO1414Vs%q$CGCSTG~D9hPO~6nr|y4=p$Q)q$5B>
ze^;<dL9iBTU>M?C@Ocn+CV}EBVeqIm9zvk9)pgU#X;Uq^RTTkTdI2Y{kxunynsl2~
zv9kht3B2K<8Soaa4~|vK2}gK~R527w&PMAUI2!Qb7;@A<l)ll;wW_<tf7xUk2jK>v
zi%P2XjNCM1;Qr?Xl@aMQjvq=4-bfZ9xgqL0i^Lp6>agQV|4wsC7dBAHVDOo@`|p97
z#u8IAR)b&Wum6m#SeTdwrOg5eC@|cWJ#hI^>z<_}r7w_$e<5T6KNnw?0Dk<Zj2(Ty
zXStI6btzy`Jv4AYDp42rniqNkD>^F7VIyQh+xUp_o<zhQtiiWb5hI&_yG7r|vUUxN
z5a^9Yxaw|Y<f5eLPC>K8k9RCNfwL$M9TDWQJ}86p)<^|k$r(e&M3aOjf<oj#G;v4v
zhI#<HhA3oL3c6D6APc%}C(A$QE50OP^<EeaR5X$4x|f7ZTKE1{<OV|+PzrgII?-7o
zSn%DYr)Fm**ib4KlT~*q>e*Wj--=6xD-8Y_I+qGv{bzBhj6bW{Uy)X*qU7ItkMoFZ
ziH}(+Ax}z-p`dm(0)fsZch59n!GC=Q(+%O)FT4^zFZ|&y;wv=Cqv&wO+5#E;IFxrq
z^Nt*-%atXU&?DuVcS)V}QU;rT`V-j7e|WcFFInkz5^ye^3&4xi-N)c5-HaSRQ(aSX
z4!%PpQ>6`*fsCJHQ!*Ccm_tE=1AUk)u2b2>L}_bO3b^){zJ(b?@84m$qqY0wvtvos
zZt+JM`lOwxmRFsB(sV1&N(K&Uua3LmOji8)x2!Nf{(|QE{Ro;<+I?p%`*m*x;?NWa
zkaF5ft*&bLHD1<FdA~uqq^dt!tJz<^1rnAOEs1wAxrSYeri3cQlhD)vyiMk!kMqj`
zR`Ljnsw9u*jHlmXEO2;gaUP@-l7K1{T?}xfEUh>RkmT$c5zQ*pUu_MjmC5>7Jvo>G
zm3RHjq?EZ(jIiE7V9cc$U)9Xd*40Q&{#l$H$x9%0`xjk=qRzNY>d4ey&rDHc7994Z
z>xz&Tqda{86!1o`aI@JJs6sN{Z$ZXD>cBqS3$|_X8Ge)G7yPyLw=0sRq{k(vu3^3p
z8^2r*I8$`JALmXf$#s1$xUsR>)bafFXrVM!7<1ZPa}wX|r_Cn6irKdm<68vG=_MCf
zK?Btn@8zKdDX<-*1f^AvtxKTX7RJURxzl*i=Af~CJNqirSDJ4u0Cl$iFlT4uNQJ)S
z-1nn95oG4mLJ56)A!3C67p}iLdK$h@04TXFk@qTsc}sM|SlevLo5)8N0NcAR4}_<#
zWEMgNM$H8`NS-f_Js(31NT5ofHG+BRr2-9@;Lyisk0Kt?`-;=)vaff#i7adsj;YA(
zZn(<@*fbLORa+wIR60@^+fiQZ&0Kr9HeO7G`&}WHI+odcZC^l*Z20SA9Ja5a*1T*3
z0L<oxc)8UF7Wp7@4Q9~*?CE9xFWZ%O=MN)JG^9B~!z-ogihURa75_Q%(_H;hjXK5}
zY!%;#eccH(+SCbhXo4JoNLaN|e9SO@Pp2u0B2TM6uJaS#Bo+Ri#F;Xv8)`$_XKe0F
znUNStnOO0cqO@k$YdK&2MWi!$o~6%xtf(bMW*{=nV7c?@z=>`GCxmOLM~VQKB-u*&
z>M&QY^@YV96kY=`=(^iHN)wINl}!dP89S07iQH_#kSocxy~KK=k$O$H(kDw-qOS(1
zi26gRZ#P*6eUAeex~vm^uJDXAHm>JJ-hQ0OC@A^g&p_}J${5qXKb_;<7Twh?Pu7YM
z!9g4KwlE%+hy(x&lcYN3dU(ND@wITR{eml34Lu@2szTP1{!?1Z)b;XoE(L}-{l8g&
z>H@9`g{_E5y~xp%v+Z<3-^`4+`xl90nWk>9t1Ol(wvaEMWDwUv8r{n?huMf1$!r~e
z2qG)4|56(MtxN|Grq2<Dw6Nl}q~04H&yoy*L6mP`K+az=d7wbQLTlX3wD4r}y}VKx
z?N)%38kDX=HY_cv#d&OHZ2!@>QSZ8L&4!7%(Dxp2_3L>L*+5oVp#Kt}Pi=_-_97z_
zp6qDaL1mF&tZ<KD{Q`=q>|Ha4Mk0JKCLV!rRiJ30$KS(RO14Atip_8TLu+DO+JL-L
z(o4<v^&Zk{)c;v*4UsrRyg#_h<E2a(_*tdH;P%9kO@owniCurp1beRB+h>3CIgRIA
zlct)QrQWc!kVf<jVZ$YI@)W5q?8ZWtjIzSWR=?&Y%90<;P_z)W41RisX<Idf<~f_v
zVDI#;ykqP0T^3EH6ukMGSNfLt66P;zqZBX{krLF;=xQymZ(c~MsvXOwz?daawZ2GM
zL-vg5oe{0N6CO2i2d>}%Q>iC1$naJJNI>)ct&0Bi*#p^v^OvZTd)fLsHihSiW1E$q
zb|BhWBlx$5DYI{%LaK6aE60n+7UH>oQW@4p2gA2rQz_c%L+I~9UsryUu^`hCHwhbq
zF@ZDFxs%O9Hqzhtaq`<|ls%yfr}B-(Vf?Twu`&P-8-RXiIPwXK6j_Li$d0ZekWe?J
zVtr%!OjS(6eby#+#{GBl5L1#820anMC|oKZ8zyot>-@fwkR3|#?qtzI5HBORAc>iz
zt}=5`pa%MtM{|010Z|RQ3*M-wGsQx%kH^;)3Z1_v%@bD3j=Mc(N%l1ivog+EbBpvM
z#^B{`KI#pQVoaQvd&nw}wm9%Y!f#kEzP*&P#X)8wi?8fhgZ3lgGZ3%Lwee3As%wX@
zNuChP73kt=1X_KrWl+3$zuKMFhq}bryW_RKvAi{gu1te-_>aYlG%7gnl5{Y38CmfB
zJl4n|LuDhHH!U$Md_D+ep(1%&vn+6ZoXc5BbFAOhg0eu_RRr=aVV*=Pfd1CE$Q+52
z!V#uu)+p+IC;FY6#p`V>0HHZCDzSKhB4cJ-jwwjem(zT8Me|eWkhj5@)Hd<MI4`Uk
zR!!7w`v~1=Xs=lEM3Euro|J6?IX8*qp5i5k_#X~0*cAu9cGzILYO($gLUQ2di$jej
z(lf5T+j9vqK!#A|l5ZGz)eO}m`!cWJM^$7%==3Dh<FC_<vxDA%G}gk#@9P@JTa5du
ztipU*pA3R8sYW(5U%wXq{$GL3?0Svh9M5Cu5RiTuR+Y$O?!K&U7HK+tZsp#=d?f;&
zqEsagVl=0e<o%qIpVs$XrjQH#YvBzKU^5Zyr*TJwUg)BH4jELhO8#|*Ffk2uS!8Ky
zK_6@3UZ|aa^a+`Ia*@_8%Z*cKH+%tFN|8PNbK~zQxH6eQht<aUmE5Ae9>tyUCnM?l
zn8#xXnGbdO4|lBO{)g<ropr=?2EkUA_JM56)F(RRBXj|ggs%PG(_-foj1cDPbqj0d
zD@XZ6MgarVi;+EVVrYq8nDp<M%lZhRY*oM)SV#v%<S0-v+#Tb(ACK55x<z31WKw+M
z(%8(;deJ$!*H5!YjIH3>Vwdb=y&_Y|X+=~WvmqT_ht~n<^4{JxI#;@Fn$S)Sluh>E
zuB1{1ntvq*n-_E_F$xR=d5Nt8fZu_ZOZe_4(Y^Gy<y^Ri_7Wd$I{0Gxvz4iA_F{-F
zJP>6^U9D5jOunTeEHmO=r=de9Id;uPn;uEGZ&T+3lCX66sOy?BO#Q>`i#fNknaGQP
z$efwj>;9oNHBdL+)RV2>4o+<CPpq3%$HeFjd}nJvy3-n*KZBd=_L)%Jg&kY+<oAb`
zBu1_HV((#9ADGM3L6;JGPyK{wd<BQ*e)6DE3n)*q8_;EGqU!(=D*kAUD4GT<UMB&r
zWn4aEV5-*Q@UUMVY8csvtWxHcB^>RoLyX_&17(ASu)MZE2x$W5Ug0U2-A3>h;Jy-j
zY00N<_k6$|M9{h%sZ)pAMX5YootaZePeQ2{OCXH7X9s8>^gUCv)<sbp)y>?Xa`R3y
zm;0|)_mxf@wWmfY+2&U^@u&TY%|p1lUAr0#LYBDW4d|{?U7Hj>z)49c0{VSjo-2L}
z@IuVD4?NHn;KV`gkk-&=y#+`&C&XUoAEW_q_G|adTr8bd1gA`){d2~`KyywY4Z6Ea
zp=Wm3Po-A~w_m=odi0&&=D<BuJ$xS~_q+3)VTQXS4mZZ@_X^a&8r|pefmbrJoPq{a
zI;4@_@*@RoS58GiVZgCDp7v_DFkSD{UUgQTo&MoDWd^J7b>CNaX!qZyCy%Hi*kNz9
z-g3Dq&pnGj3@^SdP5<C@(+SO(x3V;ywrJ6=OTKU$@IMOrJkmZ_0Ujdx&s4JCR<O=p
zv<C8xokEZXkeN(n6|%k;9q7#?+yiS{<79flt>f!I3qx5GaVBYdi=l|T<KZz55lQFK
zyI7*Sfpnf{Ii5YKM8%V@1Cz{U|Ap}o;q;5=oKIX5Z)r{o(IZL((tQn(LUj|Y;YU;X
z3QSckUOIolufx9R>!v#v%8Xb2UTLo?Zuz|mM@rR+mFJ%-re5MW^x{r3@wWf~8LCR=
zrZZGo>0%6&>HzLG&?=z(^f)tn#PIdqSQm_gfvTu7XAS0WW61fE>y5i8Me2(Wpl$^^
z*dTo)F+P}9C_>@un&S9?(s-nDBJM4lZm>7mtdO*rf(+lC-!JJ&S~#ju`}9CC9c#!C
zyK;+~)D(&VO^4b_WTGV$87bKVqEqV}DHAX0LDOz|X1*eak^Wl%D8Nwh%@(eD3l5%<
zeUqi_n7iHH(Iu(;{@56+Z(duukVf{Q8J_Wen0ku1I^fIXEj|2#;z|lznp-pu70EPv
zJFmdv2tF&1wW1Sc#zE<O)ESnIRrSn7k{vN8X@8=xk%B5b0^eGwHO0B$9-*P2AV|iA
ze9lPMxp;ee6#u&(2FySj<!k8d1_iMSg&~g;tKkPk<vqz^);*=JM6$h}Z`>Os`#iHY
zh{I<JFYi!CFyQ%g>E1wQ@_x9#1fk9agXKPU3`oDy%ri}*)5l9@xSB5CxEPwGGT_*_
z^61Wq#XQu_N7jB$CQ>*4HFzS5(bn<jM=eUdC`ZQ3kzXjxYVexMc3Vt77>py3O0wn%
zs-RIT_gJ2~tdAK`yvWhl_4hzOby#QKlw=Cw!c*xE%oaSoN4^L@cKC)U#xl^c8|WNQ
zs=1$ucbDb8vH^mq<ONZy3PB4BBEsk+UZ95WTn$X~fd2%T+p_9NUokVz9PAC;T}7i>
z5fyVJD=GRwv$Y8!Xea1xJ}OeTGsyONTax{mONkokMiFx3weTy=ooL5X;&MO#_kejP
z^$ZDI6)zb2hT)Rr#+)5~UW{g0^%3em&7J4aZjjKW-QB7_f6cc&G>5o#O(PRMu0*vz
z^Jz(P)(h0mXc(ik6uj<=F-)D*div#7Tn-unGDu6=c7SLcat_1fY?CsIMGY?~xr9XX
zGL7rJ8y76dPn<C#?sz}WuT~t}3Yixwd$23y%RX1T*OVb-czG-nox%lwdVkw+I=q*a
zot0nfx%-Wd1X=F3CqI8BOblvo%&FR=&Xcz>UGXq*$9H10JX7=D0TBV$W5QoN(~}?=
zo$mf&P}I>|qi<KtaS@`oes`p4j51hU0ZF#{ldO;DbMOnPy8~_Yn1_I;756uNw7<0J
zf;~d!@TJ7uWO*YDbCa1Z;V<d+7-xKpxE;x~AGjq$!YIh_=2C*9?69M|>$#ECzP0ZA
z;T5w-o-`{xtgMTQan;%KN^6>~0zItF&!?FF!4o;+Fw?`Ox;Xky;+?h9KpHFKX2r9)
zrxtl(@6_CQdeY*hx0c>yr2+z5@%L`*3&6koy&`GxFG-<Cq`v67JG;9TB;rdvG`AzA
z`vU7qII>A!42@1AUZ7NUfqXYp@Wkxu$zM1}T8|OCDfz32Ha>XosT;IuBHEL4lzuIC
z`J@^ZHnh<qJLsssnpFiYo^rzodB9(vrmh!UcFQb2t1Uec10)m^tsR!{A_A7o)oraK
zn`OvWBg!Jwucg0i(>Qv^e9E8EV|A@0Z;QyDxIHFhk9#jzGG?}8&Qx%zy`scfJvR~=
zZS)J@*owKHx@g84c`eDDHiG;k;VHHxEueNqEBAWXvN_qmo=b!%-*#GOXq!5Mn0DaI
z8%bSPD;-W7DkUW_;Y8Ws)0yj1@st!{xjKvy{ZRu4ECY(?!UoX^hQ#xBvX@i{i!d9B
zd$_c!=Eb|Xw)cJ2vredkpDHG<`<<h>na60oZfSkFxSUg7Knq>YC0dOU<v^^<L+V^H
zQ)dr;Fw@;9wl#UUMU<vaMxLG=S52A1*!!5YA#ILaH7{BRn>FqNDg#K$j_EP7<7gA)
z?)d7J94040Aq#%$BiU*NvPHp$ZI^vhuN&3hK6kk71v@26WCWeW4r%R8r^3`tRyW`R
zl2`4A)UxyS30`JxlNZ?-nWvr>wj$M9knTO9$4*mLE=!pv+(2*Syp|z`Bx5nd5P~rl
zNW%ln7uO|&u4Qm+PQowl?}nQ#z!nV~0!0xo2-}$ceU+#hEfI(}k6-6j193?8#Z2Pa
zq|Z})m~PE0x@o6ZO@w*aT{ENczzhZKIP=#t4RWQkzsxDMJeF`${aL>&2c5&j8fw2N
z35k&ovZq_{E0Q2azDk+?;8>}_?G>V9B1s!VH5MBRwL585#{^@h(cJ@NW!mONJid|O
z4*cEI54#W0mK4d}p*iDLfsrnr7ZLU#&n@Dn7V*$<Za%GGv7t-w80hu0rRwtPP-nEO
zX-Uz2#8^`iJc1T4ig}k^5(;!7h-mo(3f6KS-pSI=G`bS923X=;>}2PIKl5VEBvt&w
z7mfPvA}ME?1Xvx5_uy-Z*3b^7<*vXD;aLa1#QZDuoJxpHkGM{lOO>v(rpr|JeY32N
zdfK1mo37-my25>H<JxR%XGMf)$WbN37=g{D7ax3e0_qmiz;!b>a-K?+{wR5<6HT{|
zLlP*w-s-Cv%^}MW(@%10kjuk|Xr@!~T9NAN7eJ?0AN;}l5tQ$IEjU*K%=Lx>*Zr&>
zISnNwx+t6#UBsrz^|%*uX)`_n+&stD+}eum439a7L!gMjKmL@8q&l;V^=N<NjE2~B
z_)HXz;EiSezJ77yG>^I9I^~>Pn79eEiNBDWeTMlfn6DGF1wEqJr{P2Y5{wa6ykMLa
z?Qxs3wZBet5ByTNV%>}G&$d{`r31=XMFlz(aYB4t3~v<Jn$CFOuQEh=M+JRO>9QR`
z37R7Apg892dsD2{@L^6>;N8J3OoNl*?%nkJzvT3We>KymQN)NLQyFkdkTo`B;ofX1
zp6e48zfH8i&F&p@Z6nSGaQ@TH0D0bQzV>uD&J$r@C&HMR0&0vTo^d;4`G{!kAY6Qa
zqz_l6-3ZlqO&l^dQQP_SXO}WPy1nQMEP;L?b^r_L>qC7uh5lqWv|mYvsZ#}U?rT5I
zDiuvKX92RF?<0Y9;ftIG<><z~X$ZHV(sd`%eIAxeO2=vRQk(^}aHHK{GJ_&a3aGgb
zX?C~4jni98A?!nRW{(qo|2{*mz?eMsG`GG20~c8oDqMbnVWGq$yAajmAp!TR``AQ_
zzc{P>hDF1}*WHvY;|W%FM2`2+1DLM3-%`12?OM<b$Edi#cM|y5E1y~ur1n16F@%gh
z5Vw8fo}23O67pNQ_UOBB7R~&f^po*n5lJx{_1iLuJta$Hw;9p`FwHb-r&qF%oq=5t
zwa}f<B`?f%-O_{!Q(9Js8q=WWjK$mtey97Q-_G_v<6-CY<>)0rfj#mum}vk6Tx)Tg
z2P-+d!b1w0$yodlDla)a7a|TR>wc_9Dggl2f~|c1kB$xf<h38h@7~irHV`~EMb4uQ
z3U!|Wl8l$-8+w)#yOav!nTb)d`eMVx96*B~Pv2WUM^LeS2Q!0lN|Tzkaoc}#gxEmB
zoOfvQvW;DGt}%y-8x7S{B7(Y36lPKeXMUc~jv@q>3E`%=ixQNJEY*SCW0untG}%~!
z%sD!qV=IO|edd=d-F|4@_??&Dzpa)}Jkz>AVu0N*Ojoe#8@Kf^H^vEOIHfVrIYr$o
ze@?R?;f%;%>yrB}RMZ<Q)nZkPCwkW<|6o)SbWR^*xymPY9d$(nl5Q9+ZFo39y!9J`
zzLDN;J%v7-`$8zZbr=__vhRB=t-YSS1!Y3pCC*AmgF4~q)+$q^unzdf?L?tGXg}l~
z##>vy*gca)AQgn;Z3<*H=e-NAy=j(p;~!X*@lj(q*w=)LI#6yXJG&s=9>IH$nUSt>
z*>B(B$>7Sjik1lmcVN0g)ZDdAoFg#`_+HVMVWI@;-j!RTN!0^Ps7P`f_lG<$5gwqI
zvK`zZBG1py1x<kN6Pt^a(q!;w`m>yfqO{AZKXwN$Y`ADnx*Pi{gkYY0!lSd%N?S>l
zo+}DJkey*^eGSojg9ZXuPMlomL_92@12@Hs*w_rUV!OUKa-`J6^9(N0i+(=%ILMJI
z><k>h_pI+7ff*WA;(WH*jv*g}1k=|^?_C#h{Na-}O8A4Bx?J29!-8Gbx+zZC_D4&{
zO|L_O7={ideBAs9RRSm5bHo?&9PAAixmE(GDCq`o&b4+lJBi|O`fO;YG(L3;cC`tU
zSbi=533;F?zrl=NiYX+Pe<Pz~ru`3$raj}{vp;3A879mIAa+?>a=r@cF+iuNRQD|Q
zOK`~+Cf$_24Ki(*_e9I^Ad4vTf-UF^WzYHm6uBWMi6Xb+g5gL=4JeB3bG-74o4q;-
z(_!kH?w0kl0qIXGbDA+fGuH8ui)R=Ump5nBNWFnm@~+-(l!1x=R7V;zp6^iOYJ@l~
zia7KHUdN{qpfVIk(N}dTvfp61RINDWo=0qJ;7|H$xIMutfJoN90Uj!bJ9{6(HR1_2
z7UpE;VRaVwN$WVZ{8D0qcs*5Pvxr;3R0>DxEvpKsObF@|GKh?f@E--Ptb$Qzp>Zdi
zkcia2fJZ|6Q@@62nrX3tx@F%>I<T%PleMllUA;_ewB9Pe*N)lT@-WEfe-8_??DxDw
z25>a_{%hXUHD}(PXn`3=gQW<U?);?j_R-&@u(%)JY9-{GxN@IgY+M?WR?R+xUAI;;
zq~yg+L)>{uS#+ir-%#7ccivn56a41iRA^S*`cG!i*A#=#PL=$B8|{xl<3A5_;d5CI
z>8oFGlS%|mDqj7*M;<LKA<Kh&GUd5tNN!y=Sa(Su9O3PQlIc;;^VEPt#jFB#B((Tu
zmIFLtXB}=fAtOJlZ~9opMlw&3tLFT~j0ds?(eFDshTUX<dZwLRc&#wrN60oMnG5%W
zpSy>a)Ced<S&BgJKl7^YaLK*&@l5-ZukrP-#~(#P4{1Uzs68o!2UEL5X5Mh+hitbf
zoiJY>JEX7J+@@6tyI$3}pI^8=PDG`g=iw9YzE&e_84G~DU;cG{cWjD2UlyRLGk|0P
z&UG6X{XSQ`uj&Fm`vc+Y#x*qgS$r+6^GFV9%}=MbFdW_~NAsHa?rL@Im~Mn{Z-l~I
ztr*x*kv*9Y|Mm#Yc}>!Dx9pra;e8u{PuYglD941dNj({m2$g3V{Tw6EU2j8`{M<#<
z(U#|9_EbM2HVBhoZY=?8Q!tf<ed9_*Juq|ga>K%!Z{p8P1sa(;&SRSK4l$hyKNIT)
z-;h4E8~kI_&&&FI>Tl}=8<IB{+NyEuRXwdP)wb~p<0H)T;oEx>-&jGGea?eQ0c?4e
z%B)yMf$j6Rd}E@F`QTiI;~PI!*CweV6dE@IkCg=ywq0p+pAr?=>p6MS6a?`~c9pO<
z@nn05eJ;7fTlaWy`yK{k6S9QR9?$qi2DcP@NazJj5RluRQ2pv(p8ejD!hs-FbILW7
z;%8v$n7O+cLlMwyJ>t_UYkBQyuu`(iK<&tYkD2<c@D>zopUMxewq&T)mihLN>~K9%
z93;ZFKoS9ZAiR#_mGlgVSPT=8@oU>lvM)ayYg^=Mp09)3ker1>3tPZ-;JeyfbIDe+
z20qtXS0D+!=R`=oD3h2+GITEa`B7}u5GG;=TBK*%%4gMoqUb~(zlbu0()}XRWR&6_
zVhbfF?XMMh{5?}Es(D3E`jNOE2e?9aC))mQY()R^jknf49cew0IB@ia%Bgt9`s+D;
z)X|lx4&A*AHKq<PJqUQgNb21u<N!Y!6)JH2+OUn6ujrI6H&QS+myY6p<&soyg5aj`
zsl^QR+!)7p@uO~K;uFU9uZ(tQNgDd^p-wa5Mm$l}O2O6OJ3lTt9x;%jY>j_?xg+st
zFf8C1#nio*hEL|NJV7s2Cx|rloG8tmf+>gQ8<-{(#?4mysEBt#{eRzYYUb0q%?%p(
zg+XnJN49YK@ie`YeEjVck1OX_mHW~voD}x6XnTwJr+FADt$F+@wmh^bX8R{q8Yktm
zRpS&1Cpkgttn`3fvi*pCxf>Evc^Rd}29-X50iFlW`j1k;;%f<LZsk0oqb@UTe76sb
zJ%{>`;aRZWbg`dm5_Y4*(L$_rcwOBLZ+#5~|B5AfYtyQ>6u#Om3Y<^dN(nnM+E5gt
zoj6a}|A!Ep;~6oo#ZUt@H-#<jgLs}i!WEaM1S<4nnv4bf6O|q$j~r2(y+;R?=Vv`+
z7rEc`RX>ev^d{msx?)YZPR1{Hm!nwq|5F$;&op>rPG*~I7w4&3u9%Ygl`ghSzySLk
z)P3C<_uwhgov*>@QVtgQK(FvPys%z0^Cjx(ogy5<y(I@M!7`WpGeHi~ggHml@{$}z
zE^e?6(B7x<(`OiB@Qc@bYrN>k;W1nn7|23Lb5k%r5O(@FPsLQb6ZA$qjz4UjTmM66
zC#zW`;`{g>Iy{|jiw79jWImSfOJ8fC9f3j@E2;dh!l7A_7&T-NV_4jeUQ=zz*Go$n
zhwfj=>NnMtRgbjgmv9+v`P4rZ+92uFH~5Ft1aE!2iU&$gqio0RG<>v=-a;diiQ2bY
z3nOG`Iv>{C8vPV@)Z3r<79~T|{YkUaHauO#N`WOU5q6K&iPBfPp{c~=;h(s`K9pTw
zB-eeJ2yX311Ub;{;!Vg)KE+pw;ovRK_bRCP0G|wJcis%x$1c$zK46V&K)9_SEjy8^
zPC5GTl!EDJb%FUV*4Y<hb$6xiGAr?CPk2L%iLExSlmE0TwIxj+yY%~8K|f2)q4`!|
zUd>s!v_h5i%&uJIzG|`}f!y+~^$kEwx8Lv=)8^icLr4LFkqMG)nETV^a1uNBZ9I}<
za&b(;<QSIg`n1@wrT$KuZbIgHs~L42*AOOBp|Oi%9aW>%#n1eez~;zbm)qbISM7oV
zLK!gsMXGcR!QLB3Q(?nRq4R+!4<PDV{`B}@l*Su7QD!gbYOHeX)jNV+8~S4c?B31w
zsMIE=blv>;YPs$6?nt_tghPzxP&`m;LiJJ^^b9{yX-R(?yA~PCM1F&)7&?`+c2r2T
zsO$gITg>T6c(vkv34PNmG0c&^v`tpEWGBYIPLxLN@+j?Xs&RhM^T+)}ni8`+3HTm2
z<a1A#6qc2zR{13MCjM3zJmp#kTTEOE8N4OKp7oe}b=$OOF-Ke7HooF5f>W6REqe<H
z(yA{Hg|+U87erM^+8UNN{*=tP{Kkz-$6q+03oC*Dw}1}sF$Mfh72Z^g&P#)G6VeWz
zm|7`x&I|fCcw|(!$CDRqoz^B0`C8@}y_66^SRRh#1ZvcD;kBcb(x1<2fU&yPKoBk(
zYA(1IeC4AXND%$@fJ5I{sFbU2Bt0=`#&WCN|5qBk-5p}k1Edcr;iB6Ca%u?35F1V<
zijIy%ku`S^)3x&hV}s!Cf2#)tC1ubyk<R@d)sz3Y#nH>8B-NP}e!WF5VR<h5`uhmR
z)6CDZJnZ<R1VZWnG@O>WF@wI}Uwdz5nxAUdO+Z*O9Swz;t;TJNyyV0;ojNy+Bc{)L
zt=KY_NRkiL`iC&gVyY(g-$B>nWM!7IvI?ryTfu3MK6Kp1Ly94d{4a6V3~74%RS1+1
z<m(efNDzN_j+~|jyFi6%y$~;3mHqN&w6ik+%5xGs=@9PPu$TJxdO@QogQWRBO2vP(
zss4j7JlR6*N646S*40N8%(7O3>y}W}#I-y&{H&{qU}k=F>Q3J~fcNMc^^k_1nLjX9
zW-eA}T8j6g&dAM$-=~g9cF~IsM#sx0*p1~{VHxXc`L<I-2mb=xDzKm7KSb#80Wsir
zJsjeny~yLYbi#zYh;R5qAZs+yqq=?ZYbW)ZW27&B0QhIpqy*51gXX!$)<T(L=5hv(
z^wQj_aHqAP;t#oU2w$wW0QthL+ha>KPHkyVpJ!@4EK6|Y+)GYG{A|^@{_&b_s(VvF
zD<|Fx*@zxxI&v6v551m<TMH{^!&;EvVw+eR$9=>DTxaJnCyq)!bV_F0-bX#76z2p$
zSKV3Y&LI>+HW_a<sQ6q1hi|S#+%+@(`rjDzWzM$&6gSAhK%ge8VSugAJ8b+jaT<?j
z+qtn;d349g79`5LRk><C&QIR$*hXOOm)GP;Pa@xTwT=J8uCCYKwelI$yg4d*s1@2S
z7N#h!ISs3NWAOGMVl?9s`=Q$7Sja;9+aBrgnk4fB0J42caz7p~+SyU0Ugy1yQlZ_l
zlDMBPXZ95F$5N%A&+tZgPgepgV1qXsycNchq8CQoMU591AZcqli~(a|upp(irr^pi
z|2GR@qU1)*ul~FWL!ztQ{%c+BBe7;^DI3Z)wllNb<UYhM!4_jK87iOB5KG%ha9d&Q
z46r%>`EVmuv!UgV(v_BJ$O+qoyR}iwBV7?~Q$PJ{DNi~%K+{W~C4w%p#_ezGI0Pgu
z{N@=QvR81e=dFjc#QkPwi5*1Bh3ZQ<A%h3sY^pQ9ng&(#{h}>UhMuYJ`H6GuM>&|e
zI4aV&KuQ04YcY*hT#Z&;j<ba;`xfkBeK6(egnDOmJK{ryRC%qb#w8@{q7FVZm>;h=
zI%O#QcT2!tubFYyI*O4^$bR7|f`)(Q(vmr2w26`wg}>fCa}VfhQ4<6F#v;AT$*?86
zLX6ehE>EFC!ktKKFN0|_k!;I@S(4Rvicmwuh(O}FiIJ}ip4xKmp1}VbN58<&nG-Un
z#?=L4T1unUI+IH(S^chryb=mL_Rbkg79&)s6*9ZF&!mvFbse_92{ncHf1&mJh9(m@
zw)d0YU*s>nV4vnGNVr48TJoTrdRl;%H)u2S;_cqe2%kgS=|F(H9D>)U_}Z{|4i&cf
zFAF=6!<L}8X&=;}6)Bavqlr)5XmQzO9|<;&+bDvTQ_vJ#vfCu!{1wI>F{T^Mk^BXh
zO~I}e#$5C(*((OenGn0k>H0#>ryxxZLxEJDToJNP<+r3MNK+>jxn-^l#86-i(}>X#
z2jV{Dv?bf9VrZ_>D<q5p(GGM{)k(@fovo!pd0_1_xlJRW)RQgs8^P+|(lPBo#Q*=3
z3UEU*8N3Z8^eF}-R}i%*2BlZFXq>>Jru%P>WsWmv_2>{P&#58WKt46_3zi|L>!em2
zN)!_>*nhCb(i(IZkg9(@tDqjjC_NeM<ABc>PNaGr9;lka67S8N#f9H6vh8}!?3cSo
zOvRefx%GKKpQg3QZ8Z`GryxoGd3)-+s>T_g{dQ}@3>7e4G*J5cFI12?ouWca^I2lI
zwUHy~AIQt3*(ZhVKBKGZz{_Mk@09mPz&iu0*%RJ|5b;RNa}0-(-@*?>tV;K_&1Xn^
z@o9=|bRE1tyOFl?%+N+%Q{kZcYzH%TnC{mQ7oq`be&{WG$T@NxycFNjpCWA@Igkx0
zJ=Wm(XQMC`hEqrbbFdQVLSXLN{5@<LqHcaz9BW@WR898~&r@!Vu@&IuXYTgpUL8<E
z{WhOV8I*tVfV(o8^yB3W*FRsk!CAM*H->Bal&vv$4+)+`32No+1TN<VGD=WfvaQU?
z#$8s)*K-agdKgNrM^)cYHsE8QkrME>CHLEX<HzL_2<CdH(I$r1N}i-bhCh(rG)Zoe
zvc<z-rhMYevWX1oCpyVq(YXr>?;UpyCH&VrcsXQ7KYSQ-v)hh%(LCnjeE-R;@Xr86
z;yNo?!o9-!Dq9gPfbg1yShaXPvPt>m&KD*XaJub5gJQ~2G{+=%tTy)VwS6Ngw@R0!
z`h=8a=#9XiG?n~4s7%8^xAM}|7CWNNfDc~H0l?G~OrQk!{pTMie#G}!c81&p0n*>r
zVz*;ExP_oIXtgJUut)Hww-&Lb<p5=`IniBc1WGr2j#xG0)7eUbBBAn>%S9pMtagm>
z7W08O<yB4$XN-WO3p>Rd-d+BDP7SZ;d_CnwW3v@2?+&#@L0{-cK_p^942NgMH<MD9
z1D83ET)R|Gy`!cx__o_A6ZN}X%Nazw^9J|2NryJLlUxp{g^DLMGWMhP?j@=aH-U+b
zBn({GX=<>o32S^?ovD-TdBgdMQ?tqyFN*J>!rH=aInC5pd8vev6+9nB-dSL13J&X%
zI7f$nVAj<lc?{QNJYbnb1p+1IP1e5ai=?)eeW4ZFJJI^H=?SUb*7AES`@KIK9xo(&
z1#L%CH&BJNu&fWKMn2N?!t>y2t?6phLm2OO9yP;T9r{?MexA?ooNaRwqaJUGfW&gY
z<jkp<aZ0A#5ZN;KO`wl_FiVOT>+X*OdW7hzJ}BE(IrrsnUclDVzhIVl#fBvW|F_^G
zt7Fc*h>WBf+3s4P)XCuO2Ys8clSyc5Gvn~;EKw;@&Z&6+q*mo&`kw^edcKu%7JH|`
zzrA(?9K@-7+aH`Uv`dvQM4Pw_%?t6#h61%E0Y7Ny7<@DNPhdHPQanJy2!N8e0>LzF
zrdiP?NwS^W5@qFU0})X;-04u@%0ZCsFPs>E!PQ8Ko&1-zLZz0}t5?N@XJ?(i^xrZn
z7j?gAS?hfql|{~pkVGo*)dqkRb9c^*RGMf>;Um0yk}p$t^XIBn9;k-vQ~#K~&A;-F
zUUw@c1?zjRc--_H>x;X$p-OO8Conb4`nzO*(W84+O7hPE<WhG~b`V7{-8Fn@BN^+$
z8O{6v7IM8}oXYxKgC|IG|AD~moprOS`I{{}s=9xB96mO&g;akvMH=w!UDI{=$Y9yT
zjw0?%jS|&Lr{-mJ-$8sXP`Ov+$1;bq&b6zfby1V~F5^Ej`<%DOzggM0_QK~2!gPF^
zr`9J<rlINuUQ-0JbS<jA$HNsF^gnS(=PBpYLh}3Q*mdSXF}Z0T=}$R--_N$CiSBk6
z=EU&~MfMi2<!?l<`S|=QbJOMLq<kDWGxOD2Chl^edsM+Xbro4P1GO7$`$;o7;H#5Z
z;xI(wnfA-3mhFiO{3Kw&SW&;j$e&v%j2vb^!Se9D7K}R7Wd<=0CuWu_03@YK*h=CD
zihuGy>!!n>#->p&dSWJxIVMuq991*crHTA5+^6nJu+P&PobmRh`?C1SH@7?Uu(Am$
zN=>qsrJ?$GR_8J(`-<46S(ot!(fTgjhbmQi91^i8(lO`tq=j1;&fV8NMb^ZJZO;{>
zsX=?|)~_75(@h4w%m=yu=^J{fx4+T7Pnaszqck#2gJ)lvS$t(kw|m5lnvOzIcX4)?
z=(zHp=KV8cCZ3YmZB^x8sgpzo5W#QcD3LWPtomx+cTXB(8{$LG5{kmqLLH~U(9<07
z$eo?&?%6PG=3wj6u#>>TsFK3MweqdC4hHH5SjcPNjHu$d9##ZvR8<|OITDOIf#O_4
z2y1<@kX@y#+<*%f-;snJ(L2B_-*<*pFgjvd&A@Omc~?ii%ypk@)@*Pd$#fXjs3h^5
zFE2P;)!8Am<9&nPPISoab8a$x6~cER`&J5XI2%rNwC7Z~66x*eCqfbLg`9pfevo;e
zTKb_WwCFKq;Pm$?OO0gyjq{5?hlY4XUYYhZQT?l!_KiTZlC)gOypwy~JsR)1dVcuV
z5oySnzsg~>nvUg9m%Az2;9a%<Yrv>S9+(q47{p0}4>Z=pmmdmlDeU~sH)`<y-yo7f
zE3fTz)^*|oZTJ!R@lZzW7O)h0X4$u(f(7&GB%N{xjcz!qX6e{MyyTskJiSuXR|Rr&
zsD9x^=#Q+c9e;f@UnWBx+E`iAIoVqR0lgO6$zBOIHW)8Gd6qrb9M=bJY4~s7i}gQE
z6-u=hw8cF#R{M}r6|c#rU()yfhx435N9ykUlCfg$r?b7T7eB`K<s*MV+jCm1R(kyK
zJB#BFO4xdSyk?=Lk|}M<{xMNTQKL47&8K4X<a%ATN7CD-?1A5U)l~UBf7!mYQ;LuF
zrMh-C)f@Kl;@yI?y-t3mZ^MbTic5nh-Rx4EKCEZnK!-L26hig@a_*=bw<A0LCo;kL
z^aRs6=yd10D4qrzE1*JH;hi|sARfXRXR}^G_U}m*5~wOntLF-x$_hq3*#TF!<qhms
z#KYS<En&Nx6yA-GZZsIGqFl~aWwi^FXTud(`Kne-*%n2HSc{#RstYPYX3xSoLLKId
zx>MCeU*l8#6ncCV(RFSvI|`gO<i`eCr{_PMIE(Ck_~Oso7ANSr_0C6yjq1(Lxv5Nr
z4s5x2A4w=J7&g=#&ex9lt5>#~My*CV_49P0e_*~Y8GBsZ>QeN%aUPfIx{`RVx0SLq
zI&RfIM(9mWxXK;B<uG<}aD`LM{&U{lA3QBLVK46QS2iRKLh3DV>#zK<XD3uI^~Le`
z?KHzl$l|sH#A`Q^o-`%1y6%NPZO7aq@qa}x-m;(`&y*V^zB3L~?v==%+hVm_zbKFp
zya&Er>;c!Ds9QtbOdHKOzf|}Ws7E|urI64Rr?330^!GIQBNM&c?3<qHnMwMZG8MK8
zg?l$|eJFn4rrfsuaI0!taK14;PmDeHqr%P|a;d)S9ZYvKP2yr2Sy&PkHLmfqe?!Yh
zLMQZA`<ItsO_2EHLJOUz*(v*m2#gEap*@qb?`j3LRI(VN8y%^6-}@Q$b`-H~tTNa}
za-9#I_Tt*Srw{fyGjG7if~m3d_K)Zv??lgAhs?0?P)o+URU0hN6z!<8d1T}OP?zj+
zpA2NUq7*Pvx+eILM|YviR*MNHcJN%5s3lhyS96nlAC}Abxj^XdS+zRK<-x$d#b@5*
zZz_vC9?S<18(I8zY@17Km_nC!9h!YKMJJ&=uU1FWL2XaJ?deEue&QF^wOu=58T-q3
zw~F`vv)N(8gq=doh6sySzM}Lhk-d}*AI`3Ppvs3US@X0AB!NM|2Klmos1*x`h8ac_
z?Vij<+6_dOnJX*-(K%(Fkko4rv$cCkye-g$dmMj%F6u)T?gqP0x$-jWT4A>#?!aWk
zBdqZLwXRW{weZAi5eoSpdsjE3TK;-1E0gW>S*Nx=e?IQzd^;oRy!FP?nPAt6cb0v#
z@}g%+TSD!vw!~V)#go49xxeoNPEu&cSHc8xe^6P+%@!m?uxvR5tClH<UAY9OFOG!p
zgcadbG&?4JxEpFMsya%M+i81$V04$v6{_8zY4WNxTzI4yfUv>{GO|RvzEbd_+P>n`
z_PRi(D#%YxeANwiR<olt3p@HN288G0{<n7&0=`_?m)4U=(et{2`1->5`tI@dO;eUW
zx{T`Q3NAd88`NKV^kE8nzVe7|_6f@uKbP79ZXJq6dl$!mZ!)P2rFp&;KNM4z;F$M;
z8SGVDDK2-D_4&s0+$fLZ->s>952#-1ziDM>kXR6k%U1l=<40nM!Q5S~@a3xUT+mfj
zaUnUl(f_Ki;I`Yvz4-^f<_2aX0Y@c&o>MSl^_o&(NRr`ftWs{!%6`SJkGW5Z-Jvdg
zJ#q1~Gv$H*?v5jc+x5_Y_>-Vm5L?D*i4h@!IHO((+CoCQacx5PJSsM_Jf$Z*N#2Cm
z{pj?Zhkaj}ee=8s*}QP%&T^3a_D2FVLinj6?5)AH(1UOe8$S^aK`!C_)ag}tfC)8S
z#L7mV8NZ+Vvu$5iovQ*f1!ShlZFTrGyRd(}1`_*4y2AHOL2p9VPw7g3U)2MVgrPzA
zB8gTD-53kjFwtulXi0~-;^to%N<|A7kk)mdZ&%h_K0LU@JBcSv!`blF48f?dp0>#*
zZ=Q3-0}AIEtzF85wrukUjaxN^8U{S!AJhAGHdx@ZWu3iSvD`oTr<Ps+zOk5j?ijsb
zeU@YCJU4SGpparWxK7_J<LW{Eamnr*_s+}uDbB{(y3J=7xKF~JHJS!Ao{R(3(j;dT
zWtd_P;AO|4I*3Nl-co8Ipkw??vA@QFT3Z26sO>xWVNW3cWy10^?nb#mpPn7dt*C3u
z0!!L-TIzdtSzX;XZOm#IL5~v-a`kNaJJ=t<n!poqib<*b05~dJ?+USMqqLL#?dNC`
zZ!WsS-bc4hA&Op8U8gy(+Tq0jp<AUTmRNj8Fu(Cmb6i(GNqs)s-_m6Qf8RrjX}se_
zcuJsTYA2`6Y+#7`F!i|m*#um<lN@qF&{oX3{J@{LAvS0j6dk&K@<L*4RlumDQZwt+
z-7P`F-qzrp#DzA)q^!#;k+;O1JF%0rqhMWp#bf*R`$w$gZPC30r%ELcHeY++IH1V<
zK)#uX=a7%-jV<t~^(ilVg&zAG*7nPDtLhoe@aefqW(si^ki{98AKO8(c35y`n-n|9
zsv%lp{I^nMAXq~xV=j~O)fnhD-5b3a_qiwovJ^K^F~*cgU8|ucna8u~PXfgkB#M{g
z9rw^HtzN}R*{rhIIN$0Y*NSIP9_-HDF*u>cD!^~mL}ahIa8iAKK(l$l8Pidsqe@^7
zuSL=dJ!7ZYH=^X!`ts){wz^R+;W9{}Y**MyqQZ7dZ&|{}7xTC}{84uS;_I{OM7x10
zmY;G|W&X$L-T47KLQ7fBZ$X0HV0a^`v3LXU4*4DNM}li&jX_jES%Nj|$MCB?ony!M
zo&MopQp)=b-<yT;>nK`j<gTUv_@E!>FDqR*y(yWh8}C;h#A5Kvt)U(#&$XnM%x|+Z
ztRPjkMc6+f?_edU0UJ8+%zQW0nem|V>dwdIh^;p8`uLB1;HP+})SvdAiY?qpx5IQh
z60?SEn4XicF4T=bL#!{5xHFW!)rgz5^TE!a7yX9sJm<~6TIZ{=aGVuBR5^a`IJl>}
z@_1NV@)^9v<jqd>M@x__Xa&-xO@Pb6FF>ctL|>I97@5Yjq}MM7?xWo)tJ$$ve=u@`
zhTleiCuW@5+x#p)d5wNbl;3s^u$@*Kj#;oJDeiZ!@5vRn>J^24R$U`m1>F)}Z%bmn
zSNx;xfuay~tlXbKIsL7g#}ZK81U(9QI^GTCGk%i~1^UWPl!5*E9?#{zvU-;DPa_Og
zc;@Hh{alU&g1id)Qpch=iVpXmmQfG6tiZg$#}A*hNh<D@ZCi}92$9SrU!>+=ul}a_
z1piL+O<4i?-4UW28wMAjy|G;(+{7vgx`w>XqybZ=*BUYtYffE6A8u~5_BdXvYj<NY
zc*wSyJUw}5N;mwQlat!FpvQM+mb9F4lBpaQJkPz7XtuA}v)#L8+g}QuP5*z`d(Wt*
zw(ehe0|FKl6%_>m#Q@TqfYjI_R0TzPktR~5_e6a}J%UIQdeZ};Hz`4C@StF*N{2u|
z2p}B-0RklZ-8s*B?(-kx{cu0uac^V{VVAY%n)5g7npykH4M~wUTWy6Tskj|!232R$
zU8?P(`O(f>-J!8pjookB_nkc!6zhoGyD;inmk<#faAn{}@S_r?_zYd>%U*(W<@hf3
zE2rYQ!>(W?who{gTk2QxFMjGA*VKF3iPY-tnYbU9vJgIRzr7isuDyJ$O!+f^fz;xq
zl>n1%om+Y`BO;lPDD_Fc;WQqxPo~|PamG_A$sVW~X7#)2gOBetNt3s~3BrOizS>?3
zepE}CzeEOx_r7}%&g#vr`c4J)G8RXxOplAL{Qk~a!)qeAz8WC(W+E_{A+pXx;JGDr
z=-}#W+>W5Z(@1d-^WbGnNBG#$E%=wJSzmjc2GseNIi~lhEc%&m&$4bthaFL?rjBfM
z_zG<FP?-=7G`gyq7-<C61WM!ILD7-G*w_5~d<!2FIbvpk?2%5%CFiM0pS3s!-Kk*_
zWjFalRBkamj<^STAjnG>XXK`OFD&pxx*iwO4Oaf>t+FBa=J?zGK9|-{Lose~Z;mT-
zd5C^p;*_!5;N4%9y{WFjf9OhRmi&PIDj0t%yH!mzlknd2NW&YC6{~L^Y2n;w3v~Fm
zSGm5n&5-elAv4**pH;Q^-!#0Qd#dzhG(@4fgsd68)W}^JwpiqC|88Dv#)>EXu<pZo
z2L+<GdZS+oEpz(l_w7o~r?U@L22KI_x-eze3nwzd-(TKyq_oY-<-4&51FYsA9uwRF
zwr&Ieh^xMDUmoSSIwHq;hG*?O*T|BMt0M;ScsE|n2@%B1y6&Bpe&YKf2h8@Q?f#=Z
zBdt&2+!z7TuTa((CaYb|MdH}J>Zh%(vPm24?R%&jETivZFEk#hvdKVht!NqID@*wq
zHDRMpK|v?=lGz2{D{!4VaEt5SeP^u?lfWH$-&?0{|N4FL{HYhhoDJ-AsFv&UQogrm
z*xA|HZv6zVUx;CgS3eGWuO9B!Gj6*bS0mL}SXWpN{VS?l*a*!GEA%LN;L%t#QMlXa
z(P$Q~P>8=e3hkw;YZg`*p^4Mn8^Y(hoNp{lcs}PQ5ZB+ay?U)3+ITFds!4FhXI0ir
z{Mud;wIb*$kzX)Qm6lS>qaA?36?fJRzeNr^+#p5h5GV2s*@G%KCuqTqbam8;{2KBp
z!u)gU$51dnZibCl&(+vyVI-qvC<-YzN|4Z<MJF8Ii?%|iW8}$e`3;Mj+wFPBdYnU%
z?VR@h1X`FRhl>31!nj)e5Z}Z?1)1?@-R^H_XM=RpS$De+vN>Z1Col;|N`%~(V@Iw!
zw+Xg*#EPXK>sU#nRE5?4E?b#vGn?|v+YXK>EyK=?`rNqRYq5weoz1qC{M3LQ8KsTD
z!y~tPA{6Ir=xOvUdyV8s%|zwOMd#Vztj_&RT<3~qcW35I=aOaT*h<)w@I>W4zv<s+
z=vxJ99Rqu4XLV&-*b~;z!o=GY>pCqr`|zvbgEC-sHPiEn`Ckv)UfPv?xk<~tPjiLd
z*X;VS#VN=|g7!eo!r(<p%8XK>{ht4#8#GD>Enz=Dw~^u6=ZBt2k&*9lolj-dGBY2F
zix?H`n7X=(4rtzY@TA3l4Wl>O7kwm|qNm4u2hNJS`Jhv*=d6RZ5;N*a(F2VIZ1%L(
z*}7Tkh)h=%*jVBDv^F#4C9ZSXvVG|tG@S;;aDO9XYa8e<XV=n5Lf6m{bwy-YY2&xp
z5o<Vlya3mvm4@qdR<E=y?8(Wrf+if&Z(25BYO--&CuNVkCBM;6Qt8#``)y=vQ1Lmr
zmanaHdgs%UR+NlpoKQ#Uw@V``$jGhL2A#dt5YZ$49Vsd5vO_|I&{B;Mt5QuHgT}t&
zchuD1ca(<fGk1$7GE7#DsEv)Y(+3`xIX`aQ@-ThW`wrV<x$&v&Q?b>(ab|E$g}y|i
zIW~CAEfV)vjwQK}9KP#^j&s4?8xM1TY|`_|svzR`W^|>g)a%p)=dK>N8pgI&qGqC!
z+k>*2V|Do|f^>5EpgB!xuFtlj_wffjc4THmCNfmu6~SB0=rquSsHKDXSSzNt9D*8+
z6$C|0{G!nO-qhhe^uA;J+3LnBrk#qVL0qZv_Ag!6%BYJD!C7sB0S0e;^7_#$UJW{8
zX&&0P+F7%_tu(cs2SXED4^`SeGUkh`q6KnXf9RP<eMsXs8SqN2CbUT-)v}75aKeal
zn3c(Zp~F7#>RygHvsm5A+cV5!9@)D>evS{VJj#k>*SHFtChbf1EE26t)@#9l>G&`@
zc%^<<)jBQ{?Z;P`nILE(f!WtBapL6W88`n1#up`hQbArVN$?+^dXSe$mF@1B81puH
zw+c_!*4&zz{Hg3eQ?fPN`hagjp7NDCzEOLLw+AlN;MJs<Wx7`OWYAI$R{Id?rt&kj
zq>;NM6tc(4)m8Y`CnA|oFEWxOwsDGaCM`G6tE3RLnvbf1`lTUCj15L}Ryx#KZvTl@
z-|)})CA&J+ydL|szPUJ;IoZX64FAh{ox<&s%C9Qjwpy1mY8AuXcxSL9f|bEr&a-al
zoXk<`K<!9PqEh{e?@3z!f|~2<M+2(1{bvm|%?9$8Q1DLyvpvP|p$KMk$wn7PwX>t5
zzp_C}hL={2R-&?UWx7>KW{MSr_;=6{0g+o1zP84ip9k466i7$AwkT_vDKay`eqUB#
z@ZqcRJ~=$<!%pf4RXq)+kkhV1)25};qeBl=!q>pQJMSuu1-cEXy;v@jMMmbtUHMpr
zw2gGzOiGtztb#~<t5N+!ih50$S!ReaeEzD6TKXsDwyl1Bj@+j6J?~nEW~z#eU8CqB
zvvX%E6~ZCpODL;_D{WNLI{Rn+{d(S=C)K8p$hhDKcVgR3T|H)g+G<ztW{!pITncS2
zFi6p^O|HefP0<Wh+K_;Bu7R9tl+7PRYYhJWkN`1pX$*R?_V#S6%BrOd<?B==^^Ff)
zV|9NpW8>v=@9V%Hsj-Mbe^Hm;WgcejH<ugtQjZF#DuG=J4Bhs~%2<n1iKKpX@B{uq
zgAs=NIdb6l7YoF!!NJ;4f+bDq&o6zwTWOls8WB1lGWWeo_QBl2D-QrIZ6c{+s|zeN
zz%;;!W|j2AD1I?I-Kc>ubR;`CXCF1LA&qHU(Rcs28K*sVm3bGvs$csCpU5Y9TI_`<
zBg@LKII}?mHvlrRkejYBJ(AMyU+G%aNV2HLZ+A)Bev8XB^s8KOb)Am>W=l#c`=w`I
z_f*I;yvhA9yti!3LP>Rj8Tp%shKkOsuD|Z|EJ*X|s);O*(~3$~Q%m-OaHk%n%AP-i
zj*!L8G~>;!Q)*`ZxR=K{SpHxgnImvI&Ez}q>iS6hbVnZL$GVx{?+~<%>q@TDV$*q>
z1`i$N=9o<wrIfz^iJo4Z_m#Pj{JYCdZTC)5W<vC_4DYe-oa(k~8_(<M4etk!)bdFl
z6)ikt)*dD}KMI#wOu%7a_5RMG!SBu!jG98V#4>{$uiKl*wpIAg)}^b4J8gAGNVwUq
zFON-)z8I<Th1I)3>M-_cDP23&7QzJ_f`{_klr-)5mm}lq6^l*uo#hFPZbZ;siQ%{7
z>Pr6G5i6?I{Yj?Oe0kitUxp<=OY!*hlu`KXudJ)7+HdEzPJJ7*q<426qg3=Q5<i;x
zuYLD>GBELzlw0*LMp9R%V6jut6qw!nYZl3^iP6eY6m7m%_RFk$V@umR{7HM`A7yzF
zZoE17e~fD<D19}d#7^vH+W3vT`Nu_aSNbsT8mABx!{>P2diH-VrD~$b{dFtCnIS=)
z8Q=dJ>poAamevj59bb0VkqB1JOq%nj-pXUUFk2DQa;|nZQAq0SLPv_FMm&86V`-t0
zLf?JUDVx>#7nRs4FYg)^sv{96OZoc7qq0z_r7noF)BEoHs(_#|@#8eN-v}+ZXRdTF
zzG1ttvG!?6-9Zg&!|lPD%6YB(b~VDjEu{*>om$+=(0+UQ2l9`Vg}X+R4i>1dZ{3v!
zOr>rO6#`nCQb2C2rQ~UVv12tI`Qqx<j{*Hf`nGw}0H)oxXZ%0C09IvcDIUZD8f8ZE
zS?%adzV_?>SAo-Y=H5BTwOoz__T9(|zk2~r)3q8SmK(0&f$8Mc{xHG1`BgJ3uB^;K
zZ##)Tje?GN?72xbm5<(DxYMUMaqA=Q<^gPJ8)&TJuQcz05T*ku$fRR7EY1_LFT%&M
zT(J4IVID5uGZNkuPs*&5=Z?$D=#VGq>g>@p=Z<K4r%1o+cCOWUw-C?MBw=6ec)jxH
z^}xsJ)pgN%EsI-<J(|neq$te5OsTOG6>tAuRbD@>r>lMX`nD7)W~O*EXvtM)=C1%%
z-1%X<cUH#pLO~s^K?9j$18H9eK6<2XT;=JwV>M@+rV^$6q##dmyWbDJt4^PqnYStE
z?Rw}@00vqK9+tVES(KG9d)H#%4)*})zSnR%Bl=)!M#oes6pVcLb9FT+Vy0NRtCKuu
z;FYezSPr=%jI*i%ZKhLZ$W${`B}{aQ_{+&{j><qNC4Ix~%}y_G*j&~%Qd>u^?J|t3
z!RVfxuhhgl2OzT+N4{upTQlXWuB?t8cgy_#SFLWP$Ly^$Pb8G>HRE5VG3`2Syd?eA
zQ))*aO+QuIP}4r<Izf9l+bJ!$nxdUTPvNsn_PYOPw~VWIkYr`Y!#UT9(fDc2#e#t(
zrTXEJfChY4&!btN!Yn&`r_VJrt-fv@149rpGrM}Hn$Bxg$#e5&WP0$9t4@4g`9hyO
z9o#j?s}e0$nrJdAlF@$IeXqW+&Ugb>tYYX;p~3ZV?%or^GqTP^Z^SDv%`TO4De`^1
zP*s$4dC|t;eJ9VnTAW74kd?9XbP1T2SXtKzeZ4SkgK_K=si*~kf5g5}mXmIs$A##y
zbTjUESAgOUFrAWz)`B+%+Lz!q1wKCp+M=GeAGq1r-yK}>YsH#D&iC<ztMa)(07jMZ
zJ_RMv5q=hX+zv9f=t1+#6KTC+%eL;ZX=z>sN)<bVgeN$Ib&^>$FsgT?!F6WGh(8H3
zQR|%aMqV84Q6*zGJns`BEde+1bi9bKSrshx_OI~T`-WKg_;5ACETZHMqmVy$Sq+65
z;NKfq)>zJCoLydZpRnJX7<daVP(jb|RjatIcGQ!))VT-RL4%>CQFf&!wU$5Z&S8)B
z7*%5rEqLp>V|$~T1_I>scsv4aXBwABUk&aWqqkMXHSNm&a&b5&PT%NFp4+S3t@%i9
zI1wzP=_a5?_S4dIg*@jh6c4l;)sKu4J9}6THLJZ*XKQvpWP`oQjPvW*jC%M=0lkA;
z<hx4^@~%$6?=8C3MsR@S%xtr6bR;>Z8u1of)u#>@+yC;F!40f_um&Y?+@P0f^t$CD
zm#a4J&kgGYmkGmKlS+9aZw?PwPu`i5?QpBGnMe~PK>HPF*2hOLY-M=I$UoTfSHQVh
z23m%H+xJ7`aGi1@?UL0kBW7@{S#QKxawDx}1jG6hJzkwXQXK%RPkQI3%kbTj5$bt2
z!#Z-MKJl1aPw%fK%R0sIq+>JKZ-RacQ?VoURmrtDOp4DtMpH7{O;@@{mo(>+{Q%b%
zvA1u$e0nrPfb+hz@My=)jP~mpQz3a3cWVBide0PES@jJ?WMEq>1{`zi3j%lQ6RAS<
z8Y>c`$;z)v@nl;t?!HC9lh+mGP`tc{ceoamoE?pPRhVdFX2QHWqm{Z{mquW-FQTcl
zx4=fv*miJJ$OxVIlbc!by{7qnJA>8OT6Vik%1=qY=_m)FEz5|0VXJpxAqMYcmooq4
zHnio6aI%tnL?1yVQ86^?ip%R}N9@~Nac0d?|D*sAi61+@eC3u%d1&mab60S&&lI?s
z{Kw8jPd?Cc-$z<x$BVtS^Hv+^zQDFdgqa|yyVyTaGuZWC@&*O>##2n4B6_e6jp@BK
zpW|vE#&mJ>rErZG@757Xl_~3)6-mq?-td^xlDh_e8&{&&&ONo_a+JfNLL$YT&TM2z
z+IE$8JM+X=sk!h>6)AJQTw)u}^82YISv{j>GQT_Iv6xrSt0-$22fevESfUoGs*deW
zu9i~2B<)mZ{Cr$eHovUX`D;=Nc7k$qL|of8%=byV%Gcqqokmd-{H9jZc4<rrQ0QS}
zBF9%Gm!sztGHK%Zs(uQ;Q%6)K{zJO2)pTv-@3LjjlRX9M`_$nWEkz(NFJevtH<n^(
zXQ?4Mw>c9a;^O@0GzdjhFruJo9Wf(`B4H*M)KgMyuRnigHy2iB;MZj?QT~=8X*&&|
z<}Wfinm12?Y-jV`ivT@++36yIxSak{wYI+Vn+N;^ank3Tch*T-wP(!4^95A06jAHb
zHKBfQKp;T*9-*r-Mptd#ILEM9Ne*JBQpER%x)}!nt9?>FU-OR!d6$D__xM1z*0>#b
z?N=mFmy*8&cWtRQeL4gKFA=@)YsvlyVVh9fo-N^354jS~x^JGVp+HbhjQm7^mJ9Qb
z<tM@Vyy>|{vD(^P#)xx!HS>4z+YMZKxhi)&=M)9p&`U)PI;l!y)@2I_okYrW&(+oQ
zC0i1qHt6u2m?MmSY;yR`_U+*Ncn#)+9C2B?EKWt3^gwoIyU1Og_~iu7<?a?NrK~|G
zeOX(4FQQ#tC8IN>s(CzSD>?Kge%t0~y=i-fk#1$<T;}Lw^=tcChEquDk1AEJ@fCt4
zf;d#T9I@PcWqz(|d{s@Xs3W7>U8_2Y<DCS3O@Q<kG;HWI)%o4k7L@gP*3n%%3wh4v
z5!d7h^xMOCs_JQ5U=A&Ux%e1U=`LclNFPI15l5<~p|e6BK01Rd>4X0Bfj3mehvIVX
zAAShuM^Iz1e|&F?y=<GkmgdfY@Av!^ckA;`8QiFSR7E`ZIyje>F04ID@f?lA(IUu7
zqZFR6)$ZDPeY$DsMtg+5Fdc#-bT-uwFq%R>y;=gk(Dv|YlC7Fs3=&xhQ!SaR%n(de
zS=L`vH~K4ffC1f!$j?z%E}09Md5lp$?v_VOmlN^wyVG*~8qj>D)(gE2qZ^{@^TuL4
zclWTRnLY-7-GdTGfwL#qMNkDeb^u5gpUdM>-+B)NGnH1n`yL>KGkc%xb6RIz0UY!o
zG~^~81^|atEgj-wZ)E2>2Ei01A(Kx74lj3joBK`K0KgE_C^;QiHsfn0@drDASNreC
z{J5AsYip-85fkdK62vh8dK>Hi8n8izXp7d8&+uV~?CG9(=r?om!@yZ7*3Bwv3kLwe
z^W(Ne=mRvr1pON<NhJ}y-u{^oY(Pzf`3T;g0#F~Jx6UV9g#iH2H#>q5j;9pDc)euy
zVquz?eM7W_ctM=FJJ9_8H3%4rX#)VteUsekbdk6MYZKrF99XJhD5MSKYa}{<qV7)S
zpH#Z=YSd0MYkexB)tVOZPTW%9!3{2lUTX9gjd;qg;p#UqCldY+W?9-mUYWxf+6fzt
z(qScYRmsa>xRQJ575BLo2Sfk;vM7mn66)%O|KhnzOFx!ZEmJNuVAh@Oeln_Nbb1<=
zZnn4WcCx#GF^W9dJ@0E39&GfN`j^~BxtX{3OsJ$w#vQI*FuHSltqd+;eyqD&pSq0Q
zP{&g0S@;Nts&1)aXdm$rI153clAK*(=(l~LexH^pI7_<hPTz?<m*aAs?ezVr*L{Ee
zmVwo}p(SuU9KKQLIdS?tY#UMG6o3#+;Xsrk=-xI-6yr-qT@L%U=1r?Tf{+dUbx)E%
zi3ZCuQTqD<?_l=&>REh<sK+em^1IF`Z`ZM&u#)Ka)W*cm1#KOts?M>#a+%vud7Ut@
z>Nc~tHU~}N!@i?=_k7%)^u+6-51`*%SjQSPq~|uLpxf^<rKINqzUY}LUG@3iuFi;K
z9V4eF`0x5$GCmUXG6fcANG(5+=X5tnEc6CCJWk^xho382#K7;>JqhP)D_B1rw;1Rc
zRU09kMg06aUo14F{b~WJQR$MTTBmRl<yp4z!AqZ%i3X-t!UIDgGGd{(+us9yltD?t
zH#gR3N@+?Tr*GcLc+aMxCKk8#4mwmJKNGQ{v+ysTAKQ7@j}o|7v}ahYx|!E~ikI8j
z`Vjl8#fcPd%>HBX?DQ<`#?D6=98J{76G5n#AvE%Ib0V1HGwI^X!d9l;^Y<)Mq+L#S
ziiaAaxuK^401p{rq+x9{co&@~R%;Js#D4X#V8eoja6feL75tul?wmNJMWX8>5SVkD
z>5}}kxTU9i{$|l)^o)<VgTPtUe@-Q#8>)M@_l?mVuil>l96TU!|4@*0y3Y;-_?e^S
zx9OQ>Dr9cJ(_{(`4Q2SOMx5^2e&Bs$z8KvBy?s>nHqcxS0UrPqI9D>1xYpVPkim-#
z3j;r6b8+^F?}<6bfU{ixe%4nEMh0gyEH4I`j~WSjJ<P{|`Udtkfo39Pq{N0|I|)LX
z@1VbALY0;r-!sXvSyB^ui_pPR)DhihAF#8bo<SZZIjs7g1a#SfGp5lT7RI#@GYob9
z9v`CNfAeV_d<W|>qRzu~F%jBC(Acj-g<PpgJc5o8BYuJT82Q!6kp|IP8pM~eUS&qq
zR@g*kE0dr}v=$&C#fTf2rKNdK@t*RHyou!*qar)1n>g`}uyEF5IT!^_P~C2qO=NUY
zOa>y*pdOW~f{%yCynXtX(Ii-OR5NZ@R{DGP0G6#a%@3$rP%HgBD-!EC+jj?_WK!gH
zJK`SKu+|KaZ=pv&K7_GJrv#%Lbcm+y-1~AuWdA#89~FFPBpPl^R~o>K<#~~;oYKP3
z9hm*=PJ1t6T?nI&ApEDrb_oobOzDapvPV>S+;pprJgSM@9k(M_EvO)$Aap*rPEWP!
z$UPw;37`+X^3SZ*?mHWK;~j#e#)^(GddugtHN&g*pGM0!@+T^uYgrkcb7+v{zE%`9
z{f7th=UG=BFXwQ2)u{Pbeiu}i*)f!Ak3PLwyVi@IA<zLI|D`QrSdlzvHATvac+-?h
zopIm&)f#1&<&3Y02u7P+85}RIh_G+xr}9bK4{tH{9eL-jAfhwi3{QC<bCMo2fchE>
zE|*Am>@tuYS>`X=({lWoRpAa!gmLz!eaG||v!(6v-=4#v{XKYUzOoByJ@-6N0_A%D
zW~3Q`)REU6Eo6^n<GQI1j7EqOkJ=74UYdDfQSzqlrqhT|z%UbMDKrL_e)2@VT3#Kn
z*q)H;$Q-d4EGU{J<Q^yzvhCG;^B;m1)sPzMVz|K0o*LQJZBQ0zbyU;`y+SvumLLuw
z$Qdha_DluiR>`eB?*P?Ktpzps%ssbuVA@C>u%c~b_;$fHS~?}xkqZB%s6mBjP*)so
z+s++e;xe)!*wMXE&gjBuiQ}v-`yQIMf{T=@)+W0LmRro$ijiZatfUp&KJftvZCT4R
zT~8RZCN;}CDFY?S|1ju#+AKbUnzyiOYSlR=I?nMwn3Npa?-SLY-Tfc5&1wEhMFvEt
zm`s5VJep0HIe4P|fX&DD`d%YqULA0>>E1w5TAs@5`;qqTul7?CTp;~u>e$-$@`PIW
z&m&EbN+EYg?8uv?*37DUlXkGA&!U^bB+QujgmtAkFYmkW$|rbj<%BmU!9BUNUqgsz
zoVz7VOUt6>!~bDQ;4Vv0@E6}9rhMi)4T&k*K;41Q^-c!BH9L!l62k>Csj!EzP-l|r
zwvnHEdq@;NtX4+sb%QeTrsmQ{A#5nXqDg{ikBfOm162~caGg;Mn%ukxUdqB_tOTJN
zvp*jg9$MLCN6jltq#Q*sRb6s-3jfwiFUG7<MG*5!#!m|nJ~%86>1EBm)~JbCGLJ^w
z?ufZC?an_?r#11F(%V`-Y(SNihU;w*(0|=McJ_VkJqgq^_4k9Rg=*sJ=AqhIy|y!0
z7)ajPGm4$dGCo(c5R7T?&6w7@JY(RO28-*~cl<bVde%2-TUIqQxU>>P1YUU2l@*0g
zQd`t~Ukxdp6J|`^r8g;v#l@b{V>yHd#9((;&Z&b;9kCyTfQ(3@NbAV67`ZLAw%W~c
z3(_+lxwzdf6M}uC`f(T3)1V9OY<B}Mj1ySowm~PezJkB5zW>t7qWJ3Z8Q}M0IpTJ1
z{|(#t1m0iou-=k1n7o0<YFurxVaU^MxxK~*=p%w8$Otx)27A;78!J2Zd?qZw2C|8Z
z=>L%R{Q*-en}+NToxd?h-bAYrcj*`y0o$88ubSX8Ubjya!N0io72+k6N0dcjXurrm
zs;LnvVY3mJb)ACQf!n<JL%tno!JP+ewZvy`?(DQKj<hcF&5vzNzl)_2hVTuE6iZ3_
zcP}LQYywhuL=jLVZI~3e{bUJr3o~GTC3!M8>l+*9O~RS?wLgDa2_Fji+lckmS5w6;
z;YjZcYWS8tDX~mtHf!Ipes}>gWIqXAl%IE~X2LX+f26w*{Fe&2Gis>w2DEmnTo;3E
z?%{SeLFZ$rWe+6U-o!6XZV^5d(iAxzDoGrWRB?POga?DB>3FXW@G4P6EYt|CjA>~4
zRAww^ToAajSzG5nO+R8w)^ul?4R3iLD_R=-0jQ00*#D`rHL(g>GmD1}$*{$_o(xlF
z4-#$q+FPg8qZnyEO`DTi{U7c@IZhRwMSZODAFD=ZDNBaqEX2CAE$`&tC^cWW%zZ7*
z%<M@DGpA$smFgh8pkgUO@{5!1tK}Zzg12;>r;H|{?Fz%sX|F!eVmq<*Q!wpw*|N7Z
zm1RDRief1y6{gbre+2%AuE%6p<4li`SzR|C&TLpG%d6NuL~c`fryq?yBCNJPY3a_6
z+AtF?$>mWXV>v!NAWR$@+Rs*8iZ}_X`3$p|clH2;v&l=6D$T)<t(&BY5`v3c4--Z7
z=s%cGtvpJDV72+5N;UjxZf3263)Okw=Osn<gYyC|dDdCaeC~4|ruX%1QC+K}HY6-1
zFTkBMBDeRzM0klRS`D#E?;~2-G9hZ5(;;f^xV3a<N8ybq_tgGtW3L{E=uzs7SO^ZM
z_m7lMzg22>U@?i58{lQBKcBnZfaP60Dk<_;TTn<(Wu<9R;Pv*Hp@(XT+PI4GmDY+y
zbVFpU?%rqXITiBuN0^D%`RNe-;A(9_?VnyxlIfN4|7KgyqQK%7?H>>S?G9CZULHAG
za(a!A-m>!_hdGTTxNwG6r7AUtKz`Aj5ye7&YdO_D8*)kmcu~mWdS?0o0ls1F&hqw#
z1)TL-xUB+Ghy^0$)bw94T{G7oQC{f!Dlh+__E+JJZDqm3;0yG=Q~PJj2}5dq3R(VZ
zIn@msEX|xjF_9HJXgkox+++F4(~PKKRW$s;BIo`srh-nZEaZ@iRhM=qd@Qg3|6~^v
zG1|FBb+0aY4Jl(<-ueGcwDf=A=|(75lRz(>3%miKWH0@@zTudbdM=P1<>e(u+|TD@
z13V?6XfZc#hnLj9l+Oov9&r<^RV4CbBJdUadT0Yo1dK>M*vkp<K$if0y8EvVDYJ4%
z(x?P0$qR$xV};1C$l+WbXyePjKYUB!=zVXr6`EUu9)W%+9{QmD7Mun%h$XBJhV;ni
z1)lt{>b>NoN_0Zopec{hykfN`PHFz=25p?duOUo&)=Eo!B+`i_h_;ER3?RH~M))`Z
zZ_$6MGY*HfFNuAJH>^k8L?1DI&QKQYu)|PvF-u+za}NbO(l}8Y0%i7SCA&#~Ov4r2
z`*Q5br8o(C3#<o0#kFn`1}<vL`xNiR!Q_pp5|S^Y!_c}%r~btTu$b^3nxR>~CGc65
z$RG`7Kp|0|S4aJ6fSfgG#n5%baem2=>IfEBPdBZ%ACUQ1!v);!8+|J<5rc|k$|l+P
z2w?Hz-#p?*N%qSC`=+k~8E1f%ar}5e>`d8m`Knr8)#3cz8rX!P$C^)_fs?(-V!%n@
znmv@VC+keX4dWrt3Y^tk6>ipP8rTY}U5{{o<z<qi@}oy0^jiDpXS^%r4ENYMe>KcC
z-~&_~_*)*5AJ~D%!M<PHCI&*H?xNog4zIXcL%B=x87M_D6j>L^ZSI5w`y+QmXbp+0
z;`zOG)dyC<0YqQl=SIz6Vz!*Q%Lc@xO(UA#D?Pm{;g%EgFk~z7KRFQ#8P?Z0fgf?k
z7%ESTk3_9mJM89z4FnlWF;Utr{R2>t{SWyE{Q;oy>d%lUZ~cptwun-}sdaKPrOmXC
zjcKF*V<Lm%j_^*Qyg_)Mom&4l$K!W+`R?!nANYsF)ZsXDW{+vzucKz#L=!Q(yyP=x
z4Mt0uxQ8ASBc{EIhyboZ1iQnS9RCMu;{PN0zZ~Fc%WUNUl7;`BL4USFzd+b&E&L0n
zt_|gskV)`G7hxK-iF}w&+UK}&1o=J1M=H+F+q)mVe;u+X>Um}NeU^3ho_eGCw6(gI
zsLWVOn`_)i*99Fxp!hg=WF4u(N`M`nk6lD)6QS>#yyM2%oE@y!YDy)`3D}c=HRf%v
zHHj2@f;S^`Bc-=T;v{f;VCC-BqLy&ARFO1wSG5<ro<RAHefh7TIg85Xe}%fX-|l~R
zBKKEJaZ~f2KDtOl#=59_NT`p_{<KjLfTvSJT=uTp6!SV|F?Pp2mmWnZ(?9)m3R2GR
z;Q~s&BG>KrxuHTJJHJF|&+EQ5l4b3C&E#IGPxI^UjX%O8k|O)j5jmf_YDUR@Q5+8T
z=sbx~cXUU^?o_}f{>7HzIoHPer%SvpF6nc|aaN=~Rg8$BHbmwQL}u!aJo7E9;&5q;
z&+?oyZ4{=A=r-}YiT(zaM5qnLRb)*S7nz9;(^drIE;`F_`)!G~B)>776k==OGhLIZ
zN6mkh?zk>uS^Ih~;n<>lVKLGxDe{Uu7q58et-&TX-=yNEYZFV~b``d2E!D2PV^ru&
z)THbW|5khze+IRz&*D-nR5u3z{F|)q_HSK3VXJy`Z##OR(pdY~#~Fi!+zzJ5pan8$
zI<TZ;Pg7gC4`0^T8}?#yJ~G`ckTB8b$ly$%@dsNFl!<|{G<fC!|MapN;GLoL%x2@M
zBWX*p)<UFJN6qQs%&^JwMx$U?T-SO3AMR|!-Cx0}=fw;`d1je7WEK?GW7vo}qbZc1
zwu2dIls=D!ZFhtnnquJ9k@eoMZ<5+x!5cWz#z4#5x0A0fDHDS#-kG+Sw3iFjjx>!r
zM$*`sqsJEQvCc>|uSBQ;TC>c*0W_q3ZeJY1o=HekjEVGam6bwHBvT$@_8*~1VsuY5
z!LU{_py7W3Cyn6=cm_`)FEY39up~Zwr%)DHl$7n*QW+fAF2QsTn8{Itb8KSZAHBwV
za-j<R^Ma(((QQxCkVHjY#quWJEAuBwveZMTL{<=4;Kg{>UPn?%EE!@iANM3}y?9)r
z78AKO{VTyqz8Z(4@0f>W#gw_1?4KP}R6|43Fx=+{)%5@U<WV1>%Km#3zG5XBtN6UA
z$;=X5!UxuP$RO7=+=sIgR%l(*0d*h934R=Z(mzjV>@9yYM62C@7pdSfHkE6I=2abJ
z*tygS0%C-sHi8CG)66Q|JP9e*STkrN508G@8Jo-Lpveb*17rU#RpH|oj-rgmiT0M3
zNXJ(+`@&*y03T@LBQsIm3hlNWAydMyy7)2wKfM4x+w<D=v<gjPxKx{VeDBFm3qNB6
zAG)}Resxyz2A1^ICye77wHpXw{=ZMVH>&qXITb`|%^8Og<F1SNEB`HhmRe0%SV7pw
zCoJ6Vdvk2DNDUy99T`^8e_0g!A$4X}F73}{{xG8*@o}+|lGf6Wg{P6hIXsScO{gtG
zWWu-)FqXH>V5}6GhXzoY?M!A^MkEa_@<M0UwU50#;|(}P*dctXV%76T>#uPXEklJ<
zSed;O)Ni0wN!<mFs0$Z!B_;XaL4`Crtj1ktFAw|caMKU`G6S7HxU+W}p<@}^P~=5=
zE7(zQn{52&!qFGT<a_izJ6rs`1e`0d(|gzG2vGC}#F#*0bC{|&G2|)^nrPkFt1Bkx
zqK6*Xf#K-9qbj#uZq%qrgnGJ+_+$3V0-8P)+jUMvW<+{i#E(YItY+H|Zev7VT%EOA
z6%4Qfa^rVc@fx6TCtBq|vq(-r#!A2FN`G-cNtuLtI;AsKjNn!c^C>#-uSjee3hlYs
zJ33XIq+4?aHGTC!v+LN$<zY1+YkLD~O8ah7u*HN=l;>9RxF<{&AZ6@K{QbiVG~fTh
z0Ss>YOWN*|tWCPaLw(Ha1|^7bu-b}Okt>GO^!CLgn25r*&HcydHfSE(FcH0qAC9CM
z?AvDEoIOs)I69(~y6cK4omZ6gP}d~(+_xi$H!^(iqiUw!XH`HXGQIctMk_X=?gFtk
zVHqr3W(X2l%3lI5-sxq(<~9Ykz6K%QK?vYbh&h9*@otE+c(p8&F=%chA+Wc+`B>!S
z<R*Sw$BEHbPt+YH*AHg*h2jMVVW{PjW2@?@$YUL4t~7Av@hg`M@oCcID9_Dg4~9KA
zV+0sJ2AOSSHWb_d;+eugIbOt53t8M!M?Y33LUOFn8M8V2ua~fGksnPK>+P^P7#-JM
zk>N$!H$rE5S>yjUkoD^$z}fTbNDU}T1`Hcgpy;l3%nTiI#l>^tX}R5^mTGP$#S+n^
zNdyt|e-g$*n3dQ!7qZfw3zj3(h;gyPm4aDEQD07gd91+l|FVx$q3Xhc^4yoOyjYDB
zfLUoKrLa}mlXG3fVH1*xgJ?c*)}D||;l(ToL;1~jHMEvE5eKbH@IiWXRs?d}<;w-Y
zS(77((zkDssHfCin2A83X!7GJc=HbAuAmIayZ_LK?7pcf5r?rc*ZxBe)~QBRTZC)a
z@Ky{YuxXY7$E%zMNH_iW5dgmX2AYv=e1J4siy$XaI*pd5+u3lOC^Ad7JmZ#<YK*kF
z)fiQEz-<1RJ42Sa_88Nmx*1&+6}=lN;}eNy%guy@Xo$6|%jRfSy$_cyn1B~dJnENp
zd{WTNJ?sZjd@bQ3>W?(B&_~c}oq8&T1HrsV{B%$JFe<h%D;25;J@r;B0B@V_4Zi04
zwi)W!p3LMm_KB5)N(+%Qfstq5uqWHFMyYo%WaThzXstiJ=os|Amz8`cRIvX7o5-S=
zS)(-ou6eTMo2UMgY7fQ%haw?;>XArKWE?E%n4~PwH<{@k3kJVt(OT;+!2Z|=t?kv(
zji9z-<#yqzlo;%xEMY9=515G2yb#d5<=259($z&D1{|a_$XmN#^`P}ig*&+g>>n_5
zd4}b|Ww2GmyqNDz^jplBJdp?^=v4HxmET*QX4)yD?*oAKd=-9ZC>C}a0Gymhr2%0K
z#RyTlRK9W-JtRh-mo+*8q}_*dhyM2GGq8uTmlc8L!Fy1!@xv9m*cB5I8cRzOtECOX
zI8kiwJ`ALbsr7k;K9uj8W*JqKdHteg>m7pL`<t?L7!?c)ofKFk$cs;N4115Zf#4R4
zSWg@S@g!l0TLsdu<fg+YC~uS@_Ta|GkS1smy*Ygpw^*o6yTUcVQ=}DyGA5@T5dd{;
z_D%*$c5@KMt4AF>wn$ELHkqL<qSKRlpKO9WaCd;`y)a~bBcTZb0j#bg_|K3`Bgyk*
zDdI9_47Z-IAx{ivt)gaICtL$mi<;h3V(>4eps1Bv0NwK2Ayo=UO+?#_heU`INgAF|
z^UTK*!&3mqQ*)~}uu;4~NaVpIXp-5`^74ll?#00GmuVnC!u&#7bllJ{BWUJDm<LIy
z$19n9<wr-zy*YddwT$2dfNR`+R&0}8K(n(Fn?lBtq2G0=(0A@|dD;xD%#ivVrW>6l
z18`;Ay@gh!QSWj~*qe_6C}J)*upE@bLl4e_K>@50+Q7`4C&N(kNJb(h9cE%o^*}&t
zf@7CC08jpU)2P)nV{vt@KOvtPlK>&nmjiJ4bHELgK7zl?3uIzW?ujH&{!+o$v)VGI
zm~BwTPkDLML<o5C<0B1OqJzdlH>Uh^W5u{uNn`ynF_jW9Bn(~SvOH`b=#EZ;g+BR;
zL@ocd1_uD!&KM0g>&MD!0B{!B3g%%d(Gj&h+Yu$}Isl#<f)!wEXvT>W1OQCe4j4e~
zK}1OrBjQLyrUFN8u2h|3Q+N#V9jHsrJq$o&h^Gky0n|TXHO4969C_Yus{q%@vB`@N
z?hg?GNG<bTWo1IpHi|&&k1JhlYk$JB^lI7GVqjUu6=RO|R_!)y@+Q!MA0_%10Ko@k
zE`hwjX{Z530;45HTr_KE>p<uvQ7WJnH;DH^=<wXmi=Wb=yjVx!c@=xPB5SQZKoNqA
zH{>r$`!G<TDq`f<^1Ao0ozR=;U@>)hqKSAd674U&n1qHZGim4sIiiYn1_Yr(oit~A
z(qD_Z@h1V)(<{&zZ&*_<jNHlck*MC)GyUsxoeq)J({lf8SI+&Ntg{dP1cs{^l8v9!
zj)_-)yC=p4IE)!ub6%T*-~n5JiUJ-dH`v5q{6Yf2phE0vV0jFRXPMoe;U(-S<1?(R
zJi}}@W;nz2*k@59ziBj=*59gqg2m0j59}YHg+DX2jpo9?fFDVa_DMrVqIIAf#t);@
zu%S_Pmd4a}n7p2!^A3^S0bnUg@paB?MXw}k6-u1+*cApBj<`4Olu6Wv-FoA=zobV^
zfd%XPJw*?}YOl<2{Ul;5ez+qbyF`Hz5-Cq5{SOozfB^8%!yOj;(PL6BP0wg!p5NWs
zl3{8-;gMdZn(An*!n;h2-v)S3GGTul$~KQmsxjf2lzo`|`LG3GScU}{kGpK;Dl6%f
zal|ptHAX#m(kav%NbLuyn=qe_Q8lng&WMVJnZzCjP>K)BZbN$2oEqjH+da!LSB=_$
zBer-h3jr8CPQXK5nwYc~(K95l%HhzktbcByWZoIwa!LOP&}@;qB@D<TsFUrB)+(9s
zVf=U>I^ro`TWu8IDyM^&6kyQ-m!KEGdQ55+t1_+hp%;`#r(6cO{=B&eT1CPn*?_ZV
zeRgcV2(+M789@!QD-F=1I5BrYRdC%}@?{@)UX309hPzj2{_!I}G7bZS5U8{CQh=Rv
z2%QY)+LoYyy3ej7gzFw0I^fX419AZ6@7&9P5Hr%Eh!OQOi~ygUhFTaTS>6Gw_?pcE
z+QABKCUpcdOqzdwRmjvyIr+jq%k)HfI>jF$XW(~**bQ@&fI1>ZT6gor=*<Uv4_`z6
zEm<eXSOdf4QESj1)1P<;eX?~#&@^nD)sux8gG0d`%>bc;?p2r!M3jy!9AGcMsOACT
z3^HzAcf{G4ZLl~acSUX>J75Dp6b+usTowk)OsIv}Upr?^#bZ5<h1HXkmU^V8X5r02
zy%19d{&qvHs-|qdAk4?a?<&y)qvh~o@0{*i$vMvEEr)%uMA>~EyoGj(bp!AjVX{Sn
z>`-a~Jg$tth%z>*{SCgvDh47|A+K3xSbK48<{8d=u~D8#z9+$s$}9keH896P{=by~
zGp0mb!9oB>M_hFAMonKb5CRy(Nz{6BfEF?KZ&?qI>UA{YAmd4|g@vQFl!#EHgDV<Z
z`PWCg!PK3uxu3@~B-Ip|7l`(%iR0oW7=3nypoQvuU|64|&RjKw(58!y&>#w8_E%Yz
z%L=SIyAIg``~&+38_;U~>KF0}Ek5S};1J`5Uu3SsR*k8@v1iW06M6;K1`wsj)UtD{
z>(7xeu?%P(Ul748Yo+SqV%YXS|AZYu2`?cQV@ENH^a=A3<8`Dix(60&*M5s&Vsftg
zw=v=vj9v=cHkcpqbwt}6__;%^5HXr9P|&mKaMx+OJ2#*JY{DK{fLR**O<>ob^NQ1P
zY(gG)z0)Y#l4_hQnUprEMWzmsAQozk_Q&hE1DhWjX@o?iwU6dka4sij`JN;ngLxYr
zhvTu_q+v6UB;{<fhZv&6#i5LVo~<VXXnT|apTl5~V?67hRsbZM)V_}jY`1s|VYCB5
zPUx+O3Pzh@w64y$?utmzpLM~4n!8Qr7eH3S)gb#ctKUjVeC^2J&u(6WXMY->0{hIQ
zHszrf5CgO5BVh|aOb;|)_|1wcUc?%*@0Y_&4CYthS-rihT?nQYksE#9gT%Zksn%zb
zQ&oWkx%xC7$C*&;V6W<n;VHfPYbn-=qQLG)b_Wk_8Im$HE5VfQM;xnBq<LpS0V)Fk
z%McRR$XGeh2SIh4kN7#^t%yrtNT~TXIF6)IMg#Hc=}LG-U|N^6d2qCJ9n3P5lFiIA
zttI?d@l~_Q?!q4g?_kdYs89I#!zkU{ss-!SKM94IO5Mz>wU=?Xr+jjA$}p@}Fj<2W
zfOg?fMcpFdFzOd;#M@BkDP;=Qqeq>`l8-Duk@V+m&~ka5{mL6^CHZX~er_@~XK=no
z2yj^N%3+_RV$YyFMzSbXSh>yV1sBN@_N*idPG}JHYllUf47cB6`+<o>1VR4E!0+)b
z_mndv6(&-NXa@CZ;6xqNiF0V1M5wr*A&Fs<m1p)7&7w7S|F#D{EHm>XJN7;hp-P-a
zH-yJ}31=v2HD|`cp1f^<6N}IdSBdhuV+Zxna6JS#ZW^urJ9x|ih%SRXSrOW*hHS&r
z8uSkI(M!qMX@J2~C`?2*tEZD15A)Ga5&50UCKRj-MK$yH6#81W!np%%PzX23!=Qv|
zR%~lI?Gu%fH|NrOT}W1Fw>%rexOz8<P+jyykN!h&LXSECgJb+!nvaU1)}0_5p@>)q
zp|)A@p^BY+&KTC-I3+*itlfJ`jp{mc^utMZ1xLuRl?t;00{k;OX@ESuwL2#mSKaUd
zq{Ms;7mRy}eUtMJM_7+#8xG6EH=464oX4C8Uj_CN7Ua?*2#^QqX04q3$cAqAen8(_
zL}X%=w2ajxls>O)SkK(T&$<FY+QY_Q1T5ukw*L2jrB&?^-usy`yP;aT={i!Bc$!tX
zS8HBp6JCu&Lp_B1ToYydqRmpY3`u`B6<LYepj&SsUlHC2y=MffJIG`9JxCSHuDbd)
zV76CB)K%g$Xgwn3!p>x39uLYcLTko#r34)zRbZ<&wP{$0cmrx;Y~CvggI0X!<RwB~
z^yQ;3aa>)wjp)|*BW$X2H$Q_RtwJGF?NGfn2o8x?KNX5P1~^bul@6`R^Jty6o=e9*
zWk6qEWBkG%ItnetM4+u7;%%Q4hYy*HA4l2sFCgTE1Eq31gJMBVG!*S-nCd8(A^hCY
zP&oZ}jA?o&#1FDOZ@@WMg#=&-#{<3KBAjP31i*A3Xa;~ti!yeHhtqUt^f$aq<JCH*
zxb^HAgic!TQLs^R{FX>sZ|eG=EUWwRQ`d+j=0KA&PK7*Fqf-kgC}B>4S$_uu2ss<4
zdIM8|L+NR~E&mAyekZL%WlgT%xAeC++yX5IJDk!RL~#$m4CxMIqjR%)nb1aQ_=$0g
z?Lt|EZBcj*4a_%?ZXOJGutL6tYeD^k1&BXD2$+L96C0WcIguMygU}*eaS@Y}6Lo#&
zX%Cx-L+UgfP<S@D$mV<9DJ|}PqPDz8T$W~OCRFn_6Q_IBT_i4I_D#`ocrPT3=wTMk
zwv$kzHwpHTBVQqc27;N$tipx!;u2jkcLWgGS8uQ(E?-cxa8S75^gPDvSJu&}rRp!L
zaW6-*_$|sDg&dA9R&T$0`o7%o_>CK99WP=z+Aa06&&6Iz{Ygn#nIE%#Ft^>dO`f-|
zDXN#wV>V>Y>5K=OWfU%y`mfS6H?&1XMUPJTR~a)1hd+P%?u6%Vd-Gav`rhFSCr+G*
z7wzN+cHU|&Hkrmg?{AiJu+F6QaKfQt%Q7t34uCFr`5|EeB%yfOvc8xKbhWV&-r%@T
zorsT~BB8$hJl*_E<Mf?dw{GP~!rHH@z8r_WGX5IY4xYpr$=P_x13ulAOj4&I$fqH+
z$axteBzI&1>~Z?CzlW+!VT`sp^07@EusMHTLR2)?<y>vx58(5S=k?HaqNWF4&9rs_
z9?t{QzIS1E&Z>@hU!q?s_7suv1&<DU>^~aNzoIz<oH<bdp8>iDifQm>-EzNIfeV;s
z-g09=jXmzsCAOC3hl6>^h2ZU5v$JY#RM7t3ONo07d1&|}WO*n48*%K6FLkRx_%pL|
zVU)0-&V-&uc>w&E!cTaX3!x81_$ID-zBMeRKUDy@b!)-oC{UYP&E%DBy(2Z&(s?Mh
zueq@~FW;n5#1tA+NdnBZ04A3ogq$1zp(j*Kj=xaJd3EzQX87nr+oipIK=8@LV+|A>
z$aLL0#0DnMbG@5>Vpe>vAGjcDihKfmALyhZb(XKWtySL3XY!v&!GL#V0AJ(NnYU<!
z*RNk!VNLyLnJ;PYd${7_S_C*&-Yxm_uff};ECUT4TtJfg7F}Xe<k_=StR5r}^(+1i
zktc0;q|;+?iVLlGIF`6Fxii0%o`%jTmb}v@n<be-gTB5JZ^{nbyyin$=ls%B5+C@s
z{MTmJ@4JmhE{KZuqEc&_s|R#m#ZnErKiQwzky^8KsM`H@@<jY!bUB#}>&r)$ip(4#
z@!v|l^N4Bdx^rmlqD;oS7U~T7E$stG{U@f%<S+eGccBXeZd49}Lc=)(B)b->>z=QU
z?@84^e%+eXnmjz|I*DmE8~zQU&Hem38QlA#^5};p3smzN(K9#3Vvi<lH_LO8RN3H1
z1|vVn;XIDEv&ld_YiCO@vRnj<%KdE3Cz)xVKD6=|n*TXcz+zqkK_L+YBOio`{w<AO
zQ~h&nYcgwRWOP@E%@Y84!n6K@MH7$B!#C`Iynd~B^YP#=sP;_o6PfWc<nYc(U+8;8
zFi}yQ1z&jaK1-aUoR?jW0Vq<r<Hl1pAhOxFfb%4z9nk|ZG*C(gxEr(l<tfCmt5cGw
zpds8T)*VK<{Xny4v{Xcs`A@EWHebb;v30+Tyq`idlPZi~1WDz!M|3z0PnQ!~?i?L^
z&~*b+OT`GmaxJ{+B?xb~ZXLdB46U*LPIt@Va7ia5L~!jyoNKGsheUWQ$FR#BL<)zT
zg5IU>HF@Ylnj8F;5zoGhj04iW<S?XZ8&(K!-NKg1*2M>BS$ghjst<fPx^rfY$-ecV
zrCL3^bDE~}$7{Vt6%Ph@)|b=w=$V7xA7RcC4#YTqiJ>*8TvHlmDo(CwU?1GNMKxn0
zS%aKRI0@|BT$%;1PYPbE`%~<R5eV2AyJO@yb`B@rhNwE#WdqtD**~~*dF}dSgoD3x
zI9CXe*1Yp%(xK!CM3ukk{}|}%+Ew3+^jAMR3~L-4&E-CnUugc+{P0rTY2f&Y_%M&H
zT_mt$RSr}+1^S&yE`InC8vnvNfVu;)aC<5r3p7hjB7df<fwU^#&jrnq9Y*IMk(s)e
zn1e+>?Q<SAaO546HWkx(HgNr(02_p7QPU3)*n>HwrEHHw({X<cS%x!LCf>z9w&gz0
zvB_q8t%kAb)V$G*Y3?|DP>)l#W%lEtSfC|P>Mis{y*^1X!1uJT#M;B=`1UipkQ4;4
z3&{AyS)oLMA`XvX`)E!^B8T*gi8+0j5sq{|8Qo?$=NPZ?(Q4DuodeDFryU@~oRGAU
zWJ5`yCV)(K=E0qF3OXH4-1^dqELA)UK~?sfJeTp|MA|ipNvQy8)a0W{nY9VdZPS<C
z(nl~<kVOdgBQ$k>6ezII91q<p={Ncpbg=&+q+dm!gu8GdwP;SrDrT{Fb9d*eN;Za}
z>^YFqssh&<@wij#CxFBM<{xViTJx4ZKMQFBR?yvQ06#Xw^Ae}ht#*nU`|4NIUF^c~
za=u~CC5}7K6g{u};Brv*m8zwW@->s51y1@LP`lT<(eZp*NLd$n%<jMfh$!|X5F%p%
zjp+P~rBNp*y_#!2Aq*g=A%D{HLWIa5K&VA!<dwJXAMz}XV_4{yF`9LJv;EKSJ(q0~
zG5|NChqwArE|X<kYNlfz?>RkBs7l;^eDcVxH!KM}(?9e#@mJnn@l{@9n`~=VLuoEp
zUWgxEUqnfN_quk}<(ZI03&Y$2{@Ih>GcW#(^ktQu9M>g#+=_EwV+Z-*__f1W36Nll
zV2yVwI~{0#G^yQu&f?fl>1Tn>7ZmPC{e+0mVS}^+84MXgQ|l<9wR^7J&wf4AeMTxI
z9bCnuKteM&(d(=2o9BS(RU<%#1+3g#Z`vA#1dr{PHRnB>I5<e)yZTy%hefl8`|K|A
z)Dd=`q52D=18N@A(CfbhC%p@ert395d)n~B(FzBDm@$;}FmFwOe{9WdDF|sYYgrHw
z=ZAlge{6qhs8eH4YnRJ_M!fQec|Iu5DO?(S@m8V7U@KVvtn3BJl0^M1IceP~eBgJ9
zS7=UX2zPN@$$j7&Yc_hg13>9tVLuy%L(VK7Iy=hSz*(=b=Q)DogG6@c<F0+qFS-iF
zXP>|^W+IK5CY(CVu7^lu8;@_}Yi`m=fE4>yjyt6KA@aT|3Z(L1l9O4e-~_A7EFHQE
z>5xF`Du<3CvlY~QVwZl-^(cEW5OwJ?L9hSQr#8(+{;w?9yW2rTggv{tv82qtndIxu
zI%LSYH_?Yf=W8dlE0;j_Iy~UY!EzP6qq?uy&h2V>%$4#Gz(Lnm|02ZVUJ4}qv6nN9
zK>sCwbzdYxjAeiC<UrJ441HzP+^CBG<rX`zBcdU2VE81d{FLCK^3Suk(RWVwfMo!R
zC5l7P2>VBU53fm3d=BhrPufgMyI>8V*$KRb^td(P(Er!ndxkaHbPJ;ip-7h^y(m=_
z1?e3NRhl4bklqBPMS2qg2!ix3f`HPM5{eXQq4zEjYJ&9MAwVGCjqmg9Z|`$m=ik20
zk262WO}LqvHLK2=wbme{D;;mcG}ycKt|t>WK;#&w5QJDJto7wo7|`K;;H%vPwQKr+
z1#0f|#~_S5#tz9CkPKI(9>eqQXd*VfG3{gtI+tAr0bPd7OPbTcRM|nrb~vNeyOh3+
zSTFlWib+(q)|)*%lsvO%{?yf7zY>7=m9o2l6$WA7Vc3y=m|?S-yZ~*?ivUw%-+$2b
z4FJ>}>xX$)nrMmq3CQ&)nY7guF!mi+`$s|ygi^Uj)^WsN|D&s}#od}m83yLCuMoEf
zU83)0GBBH8`=o`tjkn80$bCtsedFCEnUcZ&^1?)E>=T2R++l~@EYz9Ej!M9rUe4zs
zpfM4oD`9f!gv6coC#Do^rYbd;e<%g4lh<sF_xAlaC16wQx}WxH1a}~qCR7L&CJQq@
z+MubY*gzQ?ec_qI!3*G-<~0$3`*dOx#Q}wFi=`(ff63tz_kpJ5b<&m2vM?uhvZoLF
zA(D{eR*9Qe5Icpu#|nqS)nC2;V_Z&L`%JI`iABdF&jK0@j@+P87)|EIOZ=ZcJr^2&
z;!6UPu8GXbLLLVZAMh>9VKWF`V<KZI&97qKz-rg@#0q#9@r=N*1(^>aAat1gMUh!b
zk4?Bx)FnS%9mOvi5~Qh`XWfnd;i#Kk<Mko3%z$NT2Le-Fdy;wUEk`xCXT4LX{goil
z_{}SWd-Q;U`lqiUwg@3e20?L#C0-q4RuFtSr;yCkMmZ;tA3;T;XWb7ogY1(pL}|>1
z$5*SD*K|LzRC8mQDG6K&8gs(}3{!MKyRJI_5y=T(2SLcK&TTx!G_xUK5pxx(9{PLv
zB@eUWAAmv9-6e#x%!W18)j3qb1vJY|=zw!hBbJG)QG{4NDJ_aA@8@O#>4QwLOl@mm
zb~LX^gNZ@k6gY4jyvKh8IgJYeiB~~9if_AJW#dl1;F`R5;%=`HWh?xe412e_;54D&
zc^Zko&yHE;b-4G3swB_Tj2AC0{m6l?Vow3hH%RPDo_YnQFkDMy^lvP{S`8n-m@N8u
z#_fTCw-{QoR7fUcwbULo(u81a-(T5&#veXq8r}SRcIf*3r0KALsQ^<VyaR0N5+%g{
zpacQvWdG0^oZ=N<h}w$^CBSqBos1%d%qeB?q>dV2$azEyE`(-zb{Lg6{E2t;6T7*z
zx-x<nFQUEKrsofDYMrxvVM0NiKHA%o6oc>?$e=ZW*$rO;6qBKW6$mMd*+jlZN?<-<
z$`Ws0%1^cOTjE56^T=SR4jTZQ2$pc?HgedkO(+0wC%z!Y1}I_i#c_g0Cc_#|tS&p*
z$o1(j53XYRViQR*B4*dHcu;Zyz$OGyMBZG{{O}jyavcUxc_J9q4%rU$L>Z8&fUpau
zP>nVW)*{SL1q`Qx_jD<aI1;t*yNx#h<FtSoP2w9m-7jN4i|+!~UAMXx(G@*^*eZ=I
z3ED1tmy-QZJ6zt6!>BZ?uRmP=Zca<;<+L>}%@0Y@{TRSrO}OP(ZnD){03`$Y-^JcW
zn$l+z#_R!nSwNaWpm61&O=6nR(UhI+H+=dXve8n$H7klp0eta(GFiQQo*nkaez9CR
z-I5Zf=}NTo(4g(FfCgV`2?9UPl@b_Wd|Nf`Y9MGpy424wt&E^;V(n&UaE4xh%UxH1
zaliM=VvLc$P$51|@&+FW#D%Mfu&S6RfFP2P-Z5<Aw==yfFy!qVWD*ah!ds-V`qWWQ
z-#5tlXBdhwUji(pecUfa0zcx2fwJv@;g6$n4CahE!Gy+900#onv%Thig~x^C$AM$k
z`EdfL@FFAqS!p5wH9uisn7jSVb4;GwO$|EOS(86I7E}}8A%mgCNbQUl&{PYw`rw_w
zlFzVn4eTg3!xOa>>%Rk&0`N6}aO0Tc?f&u-y}O?@5Uf>4w9(^dH@WU~<{{Ghbg2`Q
z@n{bV$_GOx8S_ZdoQ$Ugm(X@+kkxTl?p@!Jhkf?3eC|krlMK)Zkc>GyFQ>vvVYdUO
zUMBxh2iIA?y1;Xd;9<mrG+uHP_5ykO026Z+ef!ta%5FPneC&iH5zhw90vXDcC^-(2
z^t<P-v*2woLx;rl703L8{m1tY8`Va!hsebRlYK1^-jQ-zL^7FZ3xuaVQc!u?<qCcQ
zZ=|b!8s96KZQ03QAQ>F#E1)8GK46Ivu}*w^DWH}AV})hI7PS>}Sg+B%RxXK@_hXy3
zVqTVwPLaRDNqD?3(l0Qgj}O-mHl^}?KfS&u2xFJZ8M*b*hZKi~Y!gW+OM;vT#{wET
zrfu2D_{+n`XN2fI<S{QGc9yp}@04p^LUt!B0R*StZnUH`Fn(z60y?meHhTofGPUsC
zA8<?Hy}R?iKT=X;EG1zYtIPB>1rlg-qku!b=wGBpHSB5(Hzmm~&;jA_syNy2f0QtE
z0GH)>CuslK%f$;wz&LgIt0|}FhI;^`o@@B!5;^F-B=^(%GL^+wS<3AB5Yt`3GLTd&
z-E4w7$a8ba`kqp;+bzf^h>hpo;DVrii|DH4*^$vaL<03t9i9&9(Uh~#$-V#NTu9HY
zfqng?k*9Kw3dZ8_lwldT8>o>CCo)JkfuaWVpijV5B46&-^Y}BxnB4s7^9D*!E&Cik
z{562H!#)Xtgs)V98%c1+_Qp37YE825`VvTyN)g#%w^J>b1+gy>sYQ(N^X}-|ql-SD
zyvi@0F!TG~MsQVb@xeZHOD#Xl0~GmpUjn;N3VfAw_b~mF$$8IeHZgdBcW|dG#P(KO
z?Ph_PAUA<@WbZ6~3aLf^7Wc{-NUH~@3)TLms5@2-Amno3JC4x8GdgD<gO6%0JM}2x
zm2nICX|h<51Yyq<TzAI@E{~iw^YKR<fWM$3pI1{@1_Uj`zceZNa=FxmPKPYqc-&xc
z%o;G>_FRFgIHK$QC3`(vV23f+NUt^V$)A4>+(}%Dg5w!6eb93{$|oEU92M3F?j!c5
zaLkv0^@Q!vKDNV@*07+<|I)^=05F78PIogcP<fz~tv_pC^Lv0qI|t##w=_&?Pm%P{
z)x5q$h(HCYNG%6)bV^x_Nm}qhG>+z<J+3u4HbyB$>q3xyyw|k8UDD-%a<67kC98&E
zZb;8}7&C?+Q$?`NT@|lHmU<nlxuZx>^^=OoD8HXa@?0n6j|^cNSTItqUvJF!;U#4j
zsmFuD#qOY^D18(-dv+C*)y;emJf-Ql`lh$LzXcXonpCc*LHx*JlGQJ6dHTKfV~m?R
z^ZzvLG=^4p<c=a~Ftu6((ZsUs<ymHJej^UBJDhn&?kJ1(2hB#heGYv|hP#E$fK=Ji
zPQ2MC#SEjmJDMVsy>RzWtpAGtI{*`iMP|jHgw*d#3VSGQ3D}FQ=E2#^k8d4>m!1$v
z5G~u6JPDAT_+v`IL74txpgs7m(>C6xNh=Xn+y?~-`V}W@+)fQu`*}i%JhQyr5#SGz
z)c-(m=ZPad&I2K{z`me;m4bmuiWK(_`ZBZYWil)P)!RFJ=E3+Dw~E{oiMlj)41^ip
zfHK)r*$%DDD3kWAXj%Qv<#s4CO`yY|x|R!-fz}Y|V+PnD*#6-tKiRtSB5~k3Cz<!t
zJK*Oe1GQ}AJZz*%H|(8%hMB2bu71#N3cT{gD;-m9#!Q4KOiFb#1(!T}IS{-+{OR$=
z`8FeX`(x2GWZL9SlakW+o)VCP=9U}_GiK^L-FlN3rpv72Pikp`=1>h-54ycqj>Hps
zzH|}Q*Q=>NuJ3Wd(XsrzH1gxn56MBU7Idr^_gr+WJ`pvL`7$sS`IRdD@nw{V-A<uT
z`qp6?zoT%aPXAQvdk+Cg&|(S!wRby1Mr$Cbgk)@KqJ-5B)NmE&Cx|*?Lug2jk_T+(
za@FJOTlOwggS1{_yd#BaV%rza;I}_9^WXz=vSUx*&I32RC`J(_&5sM6BNvf&O=t>7
zogxj?w|-EPG6&<74_?f<B>G)!$K~!@@FuwcXb>zHCjMlW#22Cw>CCh~m)7j9;6$K6
z;7jXCEg|>p0jPim9U>mr01({jd=w}e5)Ef9eATtIEP`}t;z4$pmzY<Ey?;U2!;2Kg
z=D|NSu@@?<=iI&f4VY@pRVjtat>=fC*MND-BG$XxN-3)Xe7;c;#nny!z~E^-DR-W~
zimILl`Lf|2AvZDYZ;YGAVylk*X@Iqa%w9D-;i-PKn5=cF3dHYZgOI>>8}C6NIZ%pj
zLH>lc;JJ~{kQvYFUCWRsgfNC36=#0&NOJ!D8REJtZ1#1t`~5Di&E}27Qmm|9!iM*=
zl2>EFQwUcj!@!{{!?ZXHOdzUT&l_JtJWFeM!nJe13$S?#SOB{Np7T50oAB_t<Nba}
zG)X+jl&mglnx=(nJ-d_{7leF-39&Qn&-GwgT-;Kmgfe>-&i?hR?L*bTa5SMKAo%e4
zd2k`{i*F#fexhk0rI4v)50wFi$eyq#X_!y5h_Q+lMR~zspAIiCot+9Zppv=)r3z}~
zOclxuGb!m^VBZN}(1sd6-1g6C<{OBRXpaC(ikV#1RBEjPb2hJ4<O5!l!Q<!LE_6s8
zq#_xPU@YL~ck=ZBzqi6wL+2Q(-_<c%-Bo}oStR)FxM90hAwECv@Zu8dy$&~sK+ny=
za0`3tD7pe#zz}s5RDL!s1rth|&FdO@QX{rgai@V?iiZ43n%c#910ieCzRC`@S^dUm
zoIDiK->R^jKc#VHqb_Kg!ylwb=#jk;7Z$1??9{=Bi_E+2a}Dcg8424LD1<b>%;Ya~
zHP52@z}54>em0#*kFz!i*%$yg%0rA(m2%)N4ntSk3F?EUr}1q^v;M4uj9zI(Ti@?r
zZqAfmPCg_9O@|z|o>C<3`zuhfif4vDE+9Zh9h%AT544L(q-@U=rC5$dnseKyld*7U
zlk{`ruY@`l=An2ooGsoH*FJGSUmG|UF3+3QKxQeYFAHus@vM=;g0=PUaOrWYx?DJg
zI-sbH$Q{d$tUvBiNfED)`msGfX7EzPP(x}1t*H@JKZ6Jy$+3*3HTHM8{2#Z#2<K^b
zs4dwUU+%CJvt8&5n$aGr&Fw3m%<r@L3ss;0L#sb8X?3|&CGQN5CQe6sN483D<gWA}
z+j`ToA7^X&T}z_#^h%F3MEnN1r4Lp;H)1(6^(DaFgK}@DEaZBDH8@=w9F7$`0=aAL
zUg`_9Swn%XAxm{Z^>0Q$%A|)k*Mx)fTmVv~L5I6ZgppJn6TApwEqIsm(2p0Uf)Kp3
zZ>15q7}0qW4LXB^&4jVi@V;li37Wvk#9IV-QNb^B!M$lHW_yl2Nz2s$KVqC_00@Mp
zuHy$RLzqTDN7b_Q1qfzS!+WU$46ZQffiU*~=QfMr)fl4mChx2$^Lno_(HZO~B83pX
z#JG>URO`F`%%b>X5tP<o5V@of!sl8w4bB1ezO=;pm5}TQtqXS%tUu(~B9dHyXg{^{
zH5&mYs@dlhjCI5WEnGBa2RH?>3YEjTxUMZkH?iQ)ELq(l<G=H0rjNm?0pm~36Ozt*
zeK*ae+Y#-v5TrK7`mCP0#ABLg2Za3qe_DF~egjPeJ4z1BNEiS=3#uVvxV{ZPj+?r1
zo!qx1l;)ljmn2a<iQiSiSC@ok#R14Y5bVwlHzo_3HGG1K)V&Fj0~wJRMa~q4>ch8@
z&gUJhN==TTC6KDTO6xcJj#sL$;bu@zw#V<4>)r^fPbLG2@oXcc*}l13awEv(RS8nr
zWM^CUNKQ=wk^E<%)tkSg<cEiOBwm~Xda8}F+BiKT6oe^cfD3i~6oqL8Euq}*oX<&w
z2*%6;-YdlhudSI|St_b(uUl7_M}IBmWpige!AGUFm&?xfX&LI6P2I3h4_Ho<_Xu(^
zSN&!a82hHAO@`)YoBIagtnJjfPC!Sh=#FkuctB*STNM$zA{l1fay(F%I&ez&uNzQ_
zAtaaq1gh;i`Rj(36fv+b6OAR?Uq^)WVi(>5VqNjOzcxl?fe(Hbs16@M?ro{~2yYc?
zlxAPm=ya$au%J99(vDbRm+(N6mQqr~eadKw05g~Z|7tHU+?R&6NI%E~-9RTr#CpSn
zKy+lKEhm5xQ4uuNL`lDE<N&>Hb_c=oBl(3=MtS`1Z)UlAKV}j}58<Jqc730~jfl)9
zP<Tyrcz<1VIF8taLoT)B$h6zI#M9kxq6s!<{avz(wVD*CgSd}WMzN{r)&pDW0t){+
z7)!}{4bsr^b`|89!P8f9=k@||88JNLc_!@AbRutmNC9dsn0xlUl^;HC`3N~L*$@vc
z#`C{1Da@T0L9u~#EaW>S%<Abg?p_!v1mu{@5T{c35}XM(9}u3cr)JeyQ-D5v>KA=y
zGv^|~=73XFtTZPfS|$UI>WCU+agq3Ah@ybpiovkUv8DXE4mir@!sLVl6*Br|pilf(
zli~NI#rxD_F*A{eR%pl;vkkf<me@Xn;>eUt@Ra@HqRb@3hpJ4_>l5$hYrXx%zWUyf
zHlcF!UCO%S?3qG!B(Eeyk;KC)-I6twm1K)_81f~Wh~SrnEIY)P0!NN~gdAJD`0_hm
zF(V}l+yUQalpt%af^fF0d{vssjuO3Gfi{DDFx-73WYwyW@u}q4m1<T`4%Xq?`}?ks
zt+Bkwvr;eKKSu;;B9Baf%H8kU*P))btD8hG#(oLnY~vG|4(p@Y>u&CwGQ5>cgm)8U
zZw6HZ1q*>KGq4@Ks{06q*UXK}p{EgeDyM9~g1u4Q@nhfBPddrtg1(T%@5T{G2c#2_
zI`m_9N_>|TdevMRanS#^LAiIOep3YXv(E%BG?gz;9mCu7gKcD&i?z1O*#iAs+o&SQ
zl2L0o{g{JCgX@ef@4#%I*O=Ik5pnIunTLd`Wlbt*ix_{Mfw1~h%CQkYQIGi!HDrKZ
zk~VATWZb+7mrntEXa0a45CG=KuXT?DZZFGR9@|+xcxelUYhzrHoTA4&D}p=EFy;5z
z`5fYxmTKW>q+{GX2Z1B7*&g9i&yOLFcLdDC8p6&%bR?w#nrdTk`Nb8<kiG)F%&bWK
zu3Tdllxp+q$7@<j*&^v}d;!K_wst=8A~SE9ZG`u_RpFe!R{#k|vrM@hrQ>!)`Ad05
z_xI8Pf3`@XEH4i3P}68eq+ailan~-iT*l-~AF{zb1ET_1PjXt1t9{$3qGrTY#@DF$
zCP60V=K&DG=%`jhRzW9Fd($?^q)leCZW84dlnQbr!87ehU=HMD?~qp3Q{q5lyBJ=~
zHO!O!K1NJDD*uyYtArII0V+iA`AyQ&e_z)SZrF1>1F!))n=4`nK<vnPs8i}0+zF&(
zE@}T1<x@(CBJ?H1@<@E0dPPqd^M|dcw;fM|QGpq*J>f_>k~&f}i69)IyqXMpy;EW)
zxgrZUX8CAxH=}u&FpX|Xm<)Hj`W@h44(;@vmcrtLavBPrSQa_849|Rt7=7V#*d8qd
zC*nybI%@CvQ*>7rEF~*-?7L6cyEJj)4%>NOCF3k05x5@Oqh2Eo_T`5H>8pjM$6`xV
zzdbCr)w4N{xlwj#PJ<&bcjSoY7BPIAn44g>f1~8>J=k*uoJh{3jW3=mh$AR+|JNTJ
zF@~`e4%KJDGE^54r7z3bx4!yqB8VI#<i0#dP+}PXYztp9JDfj1Y@h8+p28?x9H&TN
zM1(UK)!;9(;&jiN;*j$L+u@wPk#4?B^kKlgDCGcDjLq+W#bo1v!{W4UpL@YMkEIkm
z(})_ASJOIWS!z`+vZQ62(114yvFoq0{R}u<#%dSp){v*BF2E2GFN$eKz3iz{M6h=^
zSPtniO1UMROXYN){YfdWiR!h|elge?DaNPN{I-d2Vi_eV-Tnic*ebl|ZPFjbF1WEe
zeZyYAsUE{<nK7|SFY%^?A1cv079D{8AkMbkJo`iDINw7EVoK7;AVsZ_C^jyx%FjSC
z$z&kO5b=0Ay1o(8Bob)sqy4)fb{ddYM+4pEZ^^J8Fd+sQ$A?HZPfPn2H5gTnB|V_>
z1!8}AfZwbT*-IVxgi9@vBNG<}=1sJ0%(Cbn8a&=E;B}iDH<Luj2HZKUnL7ZK|FXmD
z-yD{at^5cM&ko@)#QB2><hmb9&5+uKbLLL_%O%HDutW#w4lA5mxcq}%me_@9Xe1N)
zxgeBZ@6LDp?QYYq-Vk%dv3R7oYhG!BbQKY0#m`%YvSW1qqGKzHXr&sn#o2z>EPRAn
zc~xWbAyjxye=m<ag7r%O3GEod7d7EBnzG%((x}s_BIKIvQu%B{D;-(+m43BIsBE|3
zOQBCq12@)Ja?G}Y^sM*RVPkrGh(Uy^Uldq%Pw-LB=z0U)+1X@-TWF_;j&w-QxmN*v
zpBL@{UT&Aj4)8Cqe^V5A8OJ~qn^YE~0xBeM6x{oCc}lMATwD5e^4Zc<z$i56@E(-q
zhjw9=C8M-tyuua2n6Nt>wL{eYA(UL^^J;@x5n-6)dxZh+JXmQ$w%3uMpdP?I_W_>g
z@MXJ=0BkN)hs};Y5hP6-CyW+tx*32_U7#^Tjm1X_>T;}0OEMd%{NPmE5h#4nz^H*n
z@IjwL*fCY0@kB-HKt-bU6HccCnfEzlPbN8O1cG#88+SQ3eE27O8mO@#Y-53=@e=zO
z)LUZgKKeQG0Ue5`dH^Csxc@^lV~E8h*@ALm&Z;3e;<{8QQK}!+hD%d+ePkP@IUSQC
zO`XT>QoAqWO#`F!o<c6PLWRDRI%#Vga)nkI8QJd&!<{zQENy)hRZXBB{Wj=JW5vQQ
z())9*Zei?6!}Y_FmmA*m(2Ya4B+ozkhy!{=35nu+;~q*94bOLs=c#JuP1(vQw?XG6
zDySSWj>j`{|JqmfYBusvWi2qe=+XRTxQJSh0kEz5PvFkoo6GaS4r}ZCtFKBQ=g&%s
zBQ1lhd4r=DNWEe4nFy5+7FNg)1=pI14}zLRm^O_W#GW6ByrWw!tsuCTV%PFWaSW@_
z<qGj;8dPIM{}`gQY=Y+n5r^_>qBBcKlCHtxct#$47CF()bsBD&!+!|Mcm##N;;N$Q
z`H<m7JvKPXDOewyBZQSU4QlCve*fsTe}7>9ORqi7A6%_4i{$0flS?LUytXcH5?Z__
zesShkn0G8q&!D6bD2`Pmsk1NPb(T15P~aec%&B8uk84_Mu|c#?ykIFG@|WnhT=hM&
zSmYOZBcwT3MNoV9<%10w8~Kt#GFCj8ZPrbA=nL=L;p^erELKBPrsnd|Z;yU&J6oAj
zbC(?4t9U2yp)##^f1n{a<GHljpWILBpidZYOMwxWtHA--ha-=!;aW9K>GYcwfb#?Q
z-o_t5NsG4Bl$7*06x9sVx3AqwzSIkWm4IHzAmM29p2SIvW!*G=l-a^k1q7E+7`^;K
z@Nr>}bc0XA-qs{qTtPCzQDy2Ev%|JkT~Yk*D7$7wjv9XpuR7mElahtNxK{Bi9dF9o
z5j3#ZkUbd+FbT<$pP=@}{wg<L<J~{PpDuYF?LvjGPnP=yPU8kY1e}Wo(e{h7V?4#A
zEUAws-akmOyeXLYsNh$S^&;Du=g$05y!$P@vFWRAf24|{(hSk`SJDfm?7)WLoM$gv
z4mu0ldMN^R2owu8o}RfECgW3}1GtmGLlVR+b^u%w(iD2#wt<QK!U;(*s1bg%lb*ro
z9Rn&(mf5Mihxz9*?%*`5=@izA{B`4WOmH4k2R?f%Gu6II!8&6~0wZxcKrg%?QRr{&
zAuSWq9KM(?Pdl1EH>%F_&f1IWzw`F0=kS=<2R9jL@VVCe9xg#o&VC2|X!?@cQe-dp
zek1>^xqr_#@<MVRaKP>L*<Ct_xN;2CfN}-Cgw@1$X+0M8nI_HNDXtcgo;DWsZh%(_
zf^%W-E?`5F$C(D7Qv!$HUPa$<=_UrbJ{QS3U%D0q<f>lXgS?xGaCf^m_Jzw>uVwGn
z5tV<ij(i*!`=-{h(3b5$)vzb)+=y<!v_;6Lt+$?_S{g;dh~8Pc!76jQ$>K=C#5IE#
zPIO2Ja&z%G;Sh+3gV#rUz<vDsXVxD{!*j;;@h@hEPTuOFd{PP9|I}}2$a#>A-Nyem
zOL3~0p|O8I5sxyuH#>Fjb_x}aQ)4dvI3y&%0h)y!8+&sg)VGW^T60bJYcYBLr7=~U
z^|;(((>y6bE)U*uLMo#wx1k+BDXqO+93(h($i?arIY_77E!!MH<2Aa(ekA#uM(|71
zaEq%Benn61bLIqFb3;k`=f=`meuQ1t(?ILW>tiOE)RU#E{Cj}27jUcVF}ws4IP!W`
z1E>j><9%F`npmoeA;s-1XKN2AL3g_NUv1B9phF_11lk^8M$IO|XRsgW8ge$<_g4_d
zghy5M7Zm=~tV+#?<}GThgR|}W9r&eVg#?~*=g@{qbMGmYmDjAfRSq%AU*=sIzUZxJ
z7#$*2Oemh#M?n173wwEe(zP;v1stb)eB%95PoZ;kPi(Yf@b}v~F`W070vowh0yo>e
zt0_v++?vbBiht}g;M*;Z?q{Xx-se!E3nlw7GN(UNK{?z~xs&KAgTf)MUXnUSRM6p;
z>@<suJiR6*NYCy&KK0{vlge6Ai%N8{%>|9wYs`c)g+ghNCH1ijkESslF+Q0Q7UjD<
z9Uy+{1UVWLt&h&w^2D%;_r-ju<lIT8jCL7YB{6UZq=*F7P}SWBl|r9y<wFf<C*I0*
zw6CymX4(0s^$)2}Qd12buW>xz{+xY&wdv*~!!>ErdxS352bK?VIRvknQA<Ss9{Lh!
zeWmt@8CmS1vS&(%h5hdS-04i=qVp$>c#CAK81F;yADzii9-{{eCWe;~kwaTvN^xb<
z+?C2H@cJ;L6fA9<teUiC#)YzLVh>}B|F)C*kz#BRFXCR_2r&Y`c1fdl>@0VDwU+3m
z^)l-ax>F}gQFilCgjL3)tk{nFN9|GBcOTB%wN#}QK%rpfrU1VvHiMqx<kD135R!XQ
z;sR0^AWO&U6{`>ur!G>jF49Ff@xFzB>(B5zeB^-BR}sn;cBo}vVf=d^au6YyfKeME
z86y6D+eSuBltb|8zGMQHa71$D9g5QZ)3C)ocb6IEi&*aXrv3L#ABAD}G`AM~v&ai$
zO_knfUTB-6znYh<+ETCWoc(0A6E=PTuFtDwb}($7+j84r^O3EpJC((qBU#)_2o*F}
z$L9<-UYg17a3u{i5^ysi4O5zl=WGXmWvtSR&e5_f(-4;#Z}`r!$vtfn;V}HRf2e6b
zesiLemLoNaO_4Yv(OI6(S(r!>^`>WN_}9Lt?P!?2ytMjKHbqzDYduyu$Db{X^r0Ft
zTCSp#BYn09PV+nnUC6v^6y=Dli01an^>7uHiJG(PJti!9E}SE)5me^mSDSp6%O0F(
zG|LcYfAY&{X+OFC#k};eN<%fhoKPXyc)qxm0BRoOe-AJVD5uFBGb=;@l>r$041cb`
zUuAN@wK|-LO~S>m40w#}i^3G6*zsozoK*6}b%EkXje(0q>7w{wRjfP8O()H%WiX36
z2|?-|%C>1Kl4YEfC~I9FB6E@{Nl4LtP~o$wKY8kOQJuJ(5Tl1&A2T}%^d9b(U3z~1
zjw3#=H(jC&V!Bg}uz6SE@lFz96&RJ1u~af!8UB;s9e}sXvod~HQ#O9}Zu_u&#CyfO
zQ>M>r%bmYzsdVlv4lY;_1K(QxV(#Pl_{9cg)w$`#`DH;w^y&7$u>g`A4w6Qw<)wHO
z`s=)H)<>2@4FI}J_Ot62BAN|3oZilkh`ML~2+u`ad&ac3WnFDUS^IQ(AA>Z+uFj5l
zx@W^jcv7fDDnyN6@RrJ$DN5Y3Qyja`CXx0bex|mQR{Me@oqy7nZHq1nOP-MzJ)Mq@
zbfo<u*&-uKQK1tfJD?+mx!T~A@=%q(E_O<G;Lq;J<R=-+kCj>gZdF!Iq?JsS^{Bkl
zGQK%Dr}gi?mEl)eVV-Iy^`R=~*<gyzLFnFK-gs%Buu88qM@v}=s8ZWZUSFYa+v=_r
zwyY!USG`lH0`KHhoOMw!)cyMLfdBN_-%#Q^);FR+pENx8pIQhpy;e~0E@zv?yzjYL
zP2h++C&uh?rHh#g)jJU+n@O&nArfx;l+MoY&t`hdZs`e@tkEtx3QtCe3p)r@AgRhY
zF0P^8*GrO=rePoLqwhPBCgDYwP5gc}aZRexbel=8@Bb`BM`{tIZNG9QsTO8$B9s>J
zemHMqF4?cc*ZC`lZOx>4E?t@8=s<XQv_;K{f8JPK#HGHmAt+4qqpQRNiPY=sEkxI^
zP+<uTXsyZxh85KLQhDa-tb)+x4dFuv%F|VN+AV1|pA41l#Xr9(>pa@7hx;bAS|znA
zg)Ar~wR(muc!q38*r-Lxw=KOhM2T5^I9Z|e)2~*TPKprRTsR`sGK<%r{R$5hD}Y&y
z#h03VpMhqf90g5JMHkK;V;J0+A8c=&WUKLeGrisht~W`Z>sJW*<;}dc^{V1W{BE&C
z?j$JbE$qN5UC|Xxy-^J(A>Jc=--6pPF6Fg9_^>);5d?iF+j`(}p!YGAusSjL0gjiK
zE%k7O^3M<r@euEo>vaP5PyCsx>1wri2Aog3ZKMygA21Yg=J@p8ocW=;Vj!kudL~@I
z$$e0fqu$4gbnzVTw^EgjXFo@QhA8XmE}qYxVy=g`3&`>6i#(|GXP1rMd{to?QQSo;
zldH0l!j@%sOY3{~Ea^3|C@r;fHpI2G2D{7i&o@e7xRDdsQH7iPWtt+DC;yns>-2%i
z<<%m^ma?5>cfNQBP2QFab&)XGYnwQ?4d2xs+#JSWhsD{sv6gTw%Sm#_nN-?EsGZ<S
za5w2gB^GoGN6?V@(wvgB`L9Brq-{^`L5^(l&^$1F^;lvuNKmHn{((((Z<kIUBov)b
z5{c?}G9PKmlg{#ow`~Rqu1Ph2i)v%zIJ?#m|I%{bUulF)DY-ys2OZUzeq5px-|d+;
z5Yzig?G<&|MyKC%3gAh;ze+S^&Fovzmo9zTc!!~!ps*;xuqeWbvPM(R#WeAAHm?QE
zbT+E-O`EoF{unw&&B1%iCv;ZQ@%-;T9wgaFJ9yMshe1g<S{?t`T*GPI>F}m_)upwL
z;Z_uFii^UfJN&L28m^ED7y$eCPJ@U4x|5*X=s^02Q?e2C`v<!SlPh@t7W}F}u8)yU
zLVQvVvg}aUn6pMs5qxC$G1+VJz+ER+JFQ>TF8M9&E$h<)`c?gpuHPdIj?9LAmUmQ>
z>0dVnp40u95PeH4#3r4rsxnq~JMuiT2pOnpe$LRpNbAu1rI{s3wbTwihk2_!eDXUb
zue2nOPBYKrR@ZB}me+E0*|X+C&m|x7JR80uU+-fu+8CV}PE}TD!+AU22Rc)kD$f1^
zIvLA5S5NfYWvz|}D2-)rT>oYMn78v6tWl6OgLh7pr`&*6LF*Q)1FG?pkB&)S`shh_
z>Wvij_|KoeIlTE;=^A3ir@;k$(-qT8N+ngN5#vT$7Ir=HTR6^j*NaxJpVRDQu0z^R
zHAx9I#FeIa(}=29f-$K|EM(W26YRe3{Wq@Dmkgw)8<@VtFXF-_q}V|_YZZK^94vX2
z8x{Gqfi=hk<LS(AALUY-!GXFaIJE2*&WyYN`F9+l0m;>AO!ZDKU**;&zE#K`Pvj@p
zFDO}7wG-9lztpWoTGqka$v2tOTWp43xpfp(WIc|3=^k0KoVfdQ<FjU*G}s4kWFIDB
zCbw0@_(r5siGL?jXWHj3;>N(q9}PZF9R;S8Y@tb8^a`!!F}BB!4v-V_)i)+b{5O61
z8$Sxw2!;VyQ(Jo_DXCL30%p{8&lvR>EuJnOb_E)|{N9MPZ}Xh-;FbFo_slqN>ohGr
z-hp$PslSko@0m)WI1cK>KignS>x$3Q8cG7zj|$;=a8={>P`<QHkor!Qn{-ltHjkgo
z&8yT*?;{`2iBQQc*n2jpDO@`4bhZJ<omo~Ul_e!Yppayj55tY*4>&f<%6Tbr<ti>Z
zJ*a*OQ7WRzNVn7=bU4i{$Q0AYf~@^?5!G!`KV{lFJB59JFeG#L<#=#MBe5McW7FqG
z!tM@eEXc3&ba1BKFzFnUA!QeLB=JE<a2Ywg$hNX=(d!|y6(k0m&~7gWRHzSA*MUgd
zrnOZ0$NJU8jnv0+Z#5-(ge-IIcV2!#(D+$u=Jnm?QPZ_E$Q@Z_`+`z{s&r`;#nV+&
zCS}p#iT!PcsEFp!FI=O75!%%Fi*Fh4D&p@mg|1n7|5oBAWLm1@TQ#@x<ra!^T4;NR
zT3?|`R-g1SUHsHmYcS#iU)&^trpqAreqRsQqjSYd<l+Epykyh&qibbkf&E8USEY=e
zIJ30=1$u|D(T$0i5;o?zr;iTU)3e`w3kr)1;y4ksfcX-gsRC>z-T7@<u?LU62df^8
z4dm#D%!>Cc<;nZ%JODB|Xkz!onuA@qPl&#Y5@d2>?Fy6KZ%`JoLxVi>Z#aW|YIjfX
z6q3NcCYi4?EWN3tW^ZSC{f+U&80vVgXNn&vH8D^<XE1Fvi%}+8tll)^I`_bBou9L%
z{7{|mCY8}fx2#n<s+9Lg$G9)f<WYd5>Af^L$AoE#4|6~DcX9R+86w{|iD=tc#3&b-
zd&jpN0HRx+YABfn*o?eS|1De(d^;7g>cIJ}#35m;l2k@7z9Mfp`!QKAAvoVk`0ZN}
zb-tR=2N$LnXHnK{nluNkeK8CEyIrF~<zF*dv7|C8)a6#DSgf_ghG)#asUT3#_YShE
zsqa*g=3wE>ig`fyqd#nH3K5q8sTMaP0?v9)++4tCPVEgrjf6OZl!2L;!^W!9KkpJ1
zoO)NGmhTiSkwNt1`BiCu0zaOWnw_Q%XHWY*uef6&Jky3+rOGIwX}rpuyFP7OFRL<F
zA{uT?&fdD@N2*Hlo^pFAWLH?rP^`nhVqaNGhFjA*KFXfFO|_X4!cCMt^`rKB^`N?=
z$*Dqa5ZO@1P*++OHw&w>Ypa&jV=E9-U)*)A?i{tVlO;x$9NBwHXz46hA^C^?zy@wX
zxIPS6ym?o?_SLk>u3rxE75u~5w9Ii-<kzbaBabBZDhUTx<E_dBh9@{wa!;j2YCZ*Y
z@lt3OTQ8$X?O7v@xfL;FXX*0YE!!ez9z;mI>yf{{1`tvQL?-0e65e<4<*<_ydX`1_
z=DD8Hydsj6ehf=J1}Tr94#!z1jAKjFfQ+gvqw+;`wZ({e()$PStP@mF`m+>wzVdKp
zCVxrE%};s2y$Do%zyVmgU{;zggqKb9{Y<{A$h$0bM05K4-S#*HqnQ?=yaNA=#fw#X
zt2lSbhdO}$rxcQ{5$`;w6YVnB=k_^4BmS}G(2@wx484P@zfSz)3egn#3=$HCdrarR
z1wxuroI6$QIxg+QU<Qv@YPljjPhaa5h!s>)bp57x03H?y+2d7L(3}+89vmGLa_vEB
zoRq@Ks*(2{WzrE3R<5Y}1Tlzt?MscBUE>tc*9!I<=xp9kMMl!k-aG_g_UBzB0kQnN
z$oNrOq<6Z6VmO+;%LjU5U;q3GQdW7g#M}ew-`Kqo|JgfGOtop?p0m$B<L3ciwlhGe
zEv3o47sqlE5P$x<TTeqM-`sMHzhByMwhDUex|OG&?#2!^9t$U8_3r#q`;``-HP`ux
zIdYn`tU}aa5U(zoHQ}hj*`YFfAR_;r%UA0)*_Drp<`ka!@8^&1op}s{BBk~I=Jyo{
z9)OU^oqef(nmhaHC_m1uP|~OS&}Z$Xkqg3OtVMaDmLI!_*r)`e$@m*?9}?RMH~i7t
zlJO-ymW$z?3+yt`kUP``HtrpBeHyM(zOGPTF}(DGPO=z%S_!@R=K#In6I2WCbZ!9A
z#BKDg1>Z6oJft#jzu;?LBhuS@$Z?+XDqY>rGM-R%U$-hY^-S8dq`{WkL$2t4!~W03
z34XK@@sQ=y`-Zad4PUz65Shsdr*O%8F+?*CU(FckS6mF2d{*}4Emga`5xEiaV4-R5
zjPac9Pll>5N7zHFfR0c7^@6ljWaJ=vOWDMNTGpQz{sdAP(t62e=;Qg`w434=#w-$X
zAsSfYIz|nx<bfab=s|gqeaqha4``b5(Yzp}hH^S*;f)Kw!d<aPj_gwH2R<=T=jnV^
zCTXp+wXl@yS8{))j7#g0IC{6LbjkMwt;g9yd{7vU#`I3fRhPQA3ZJ`&@4f3|GI`Oq
z-oN6eR*|L2;>oM?mLG0divN1odHFar>U!8m(<$YW#pd-1#I%7~^z9-@H+AWCZC5H#
zw3~uweya-*8yu?E`j8qVr|AgUT}OO9R^&S7@>Jn=j%^6l5;n5mK6&zJ;kvu<+MFp|
zG-aY(GJQb7sESTQ9C5Sk?k(R8e#G>_qf!U+{)uavXv~70*cBgIm1Pi9E6d3IdJ6}B
z?&^u{-8tVz)qRSN9`;ENfk$-@EQ<G4o${N*UW>X2o&=JXe-TA*muodo?!*`WnkKa*
z-+XQG)X!Txcty*r!dw<j>Fb=1&gxj`vHaqs9XgTJ0yWwA?sC*<s+U!lao35-Gi<<t
z#$hE@Kj?GsyaeVE(&lC%f5G1zxqxNI&Qo-gAGLo_U*w+@$BJ|=Tg2TQes95L9}<}0
z(PWkz{_5BuhZ8jIN5e;dlGf$Uu$~u+OXR=kPUKCTUa{1hB%G36D139F997N^Qu0mD
zU%}6O`mskilS4LqHf^%)e7QTXC{t0tGjj90I0*=)d=f@KR_?|O^H|tbV5wsK8mjg~
zj|wveli0b*aDaCxJU+7zScr!US<>m5Ja#lTx^?MU>r?(GXERuVATvn>xlgpk+Bo4v
zGOIuFr}vHY7}Oa)wsJBSvbZm=X?H;Dj<OL8$)U3)(m;&6<<Alo#0aaxq)xLZep+Ve
z{y3+W4-hZzUp!c2RDJVbK_zW_W{vOPCA$->mkv2}nVogyYtAd?Uz9M8;v;nq{Vjpa
zh)WM27d^<9D!i1Q82Cl6&aU*pkCR9tbg>c5Hj&;mY-tR{Z{*~^Cj2Xufn7|tAQ@=k
zF;Lij%y6Q77)TIoTdl3Bp#+(dGazP*NH<&@5@w4|S?OqA%bhwh0BSz>xthXc@WS)V
zs@Y&<2)Yrv{SY9*sEpBCke+Tl>U~w@FqZnVcb3?-%Ot|rZ$1;ELdF|Oq##hR5osZW
zIN+;H(z-rIC#dXq_bYpX=D{n%&svsCDyTAV?|BmRe3>eYyKlPi_VdyJq%L%N`)h#r
zwd{x9UF-(p5%0A7zKcR1&NmGg>Mam^k(A<g_IH6Z{$<;(?W?b;zT<oHP8)X%E}ZPY
z<0JnLAIq@gzVT;Tl>v@!_a<z}(_Z|ZoZ9(LXx140Ct-7Bp&`BTyVKilcW_O~a=<A=
z9v5HdJ1zFZJK7VDyvK)iKO5QvrZQsByc?cez8?v^4<w*5#S(uzZ7c=G@%@fxonrkb
z#~Afc8|?Z|8W-}6d#WLt^01BurGNJAGW+bGmSA&(1c<v=2^fbASa^9nIt!=*)dmcV
zHP`|O`?oC`bb{ePyppAt<9-Uk1*zwYzRUE~f1BP7a%s>YIooHO44TYF;=lPHz*1kf
zg8X1MP&Ika+4d4@*%@;&Nq?ER_)p7rW^e{!oZwI4iZ}ePVvgZ10VTi7p8VS_@R=C%
z-2=R~@}E2>G}wd;$a6|M#*1NqL?>^0?SDH0=!T8lix7m|hHQh5Aj!n*3Y0*G3lLO|
zS$`__pP5|#j2*tSKm+>v-&2$RoyzMw2SB<(JQ74+yX$}R2&-oNSXbIxK2qrPb30tK
zo=opX(Gm0NVW=QQ(NW{q@e!;(UQ(i-<FeP+-kZo2Bnd+7{tK}G6<;PPl9hbZp|v#;
zGcpaI1G&NH0ID4BAgF!hgHUx;eZAmg%MyNaBVF_#{Z><pk`9Q!>2jHVvrmB7k#{;_
z+$FN3NT|$9mW5#t--SP|B-~*-@)kiZ9^Q#kYk6{xMgc=B09>Z^m&ZE^hDm+HR>==0
zk|=hr3Gxm;ENAvq9DK1IsNc!5ame;!*;)TFiF>@4V?DXv4JuD^3C0gUTnuh4iX+c#
zzidalq40yJ-i+A~e<22zH@q2@vYYb%5-e-R8ooXE^fDRC3cLcUG^4~F6L3QMA=-oz
zX7Yw6<UrJcIRY*5a+%BIzafIxI9Pnm^lw767r5ccPul-RY-$rGAjg|~KoEP0{@2UI
z#edtg#C+iJ+V&da=RXAk{w2>)wV+>M6luk^>PE;G*s<tk*#8|$Dcj!0(gh?(k5iah
zAtAB1@aDpdk7v97+bQs=R~8LoE;Lt!$dTeeSP2;S&dr!8`^GYbw#y^)e_E`?9pVe}
zq*)tUyAA4OF__SffSE!4Wft?lP4DiDZ2~d4yCC}$2J$QaHMjr02em#|kpC4qa0FNg
z;!j?fGXKdDQx0SeU4cPhFVoN2cJ5-9%o-mdJ>cX0Z@0!l<0NNJ5Cw=U18|eGj<{lm
z-to9i!jHj$n_x`_gdCl0G3wYbL#OzSzrtGwqzf|HW}<jNl_mHuLjM&=mrf%kAkfHS
z3~~%M1r>uuP{CUQUQq_bbu1e$lsg_M<)0+|VH?YVn;f1**d8-9chfJ$PM0lHAaI2Y
zC%#jU*_FurvG8J|?)%9}!mmb!ETcAG6{>2RsKo}1oQHChz|r~Ga{Q0U3>^his-v?v
z=?<s9*+6lDOL6$$s7r}wrxw+=tZK2YLpG)uyZp%qE><JoI%du|B4oih1aczpe@ht?
zu`xv{$6`Nq)c_rBoS^v8xZ*?eKF%$mq##j$e_z1b|DB3x;vRF~5<BXwR>IS0<E&`F
zCijh>!if=6nxrWNuE6=42N*XMSOh>S1Jd#`NS%gMj*OzeZ1V$3`E|vM?sg#Yaeo7^
z*i~PEci-`RoBnQqtnysfYKA}1?PM4-9X6Bg-m5qE{8`^XUc&hybD#c(jaaHep~r@k
z^4PUAzxDl;g3lQ`M$|A6_TC9A=-@<+TlfFTv7u@hVvI()eJx8ZbTh2*OcUYwEdBfi
zs4<jMfP{6}+cwEXh_rwIXXme}Ah#~LGyjB%57|{^04O;dB?Gp(7K&LDMwD_Yj?q09
zXs34yqR;mC!_Fke?A*owUx)0JJ1sR4rS2G0gdYY9w30bHab(?XzU!_?V<1V4^G!M*
zpgh7q2^g4JyS00gPzK<#`5eDF<B-qxf|+E;KaO=Tw=(z5P=lwPI_tM@Mc;y3Pn(3s
zD)p6`zC94+$|?GoGnr`%c_4;i9VON$d;TE1^4iCrL_}Yol-?rCePi0d^fKa!gIej>
z`|RBoX*>6-^5d;pA4yrC?QN1CtA1%&7x(lSn^(?Dy_rnnHByVH!+}+~wOxjcxjV|f
zzl=vmUp@Wo1PY}dxqsm#dD3{kfOn4vimWi09cIfS<Gifnbh0D5eMEq=Tgit%ff~%`
z0{hr4?Y*i5^3Jn3qsj53RUZ5s)Icr|S|^2kpZ}*lU)L!Q3GqRZXdJA#Ml!iJCu$&n
zv(478r8OV@ok#KU-aJpVdt$VAU7L-c26Kj^&F|BJz9AJ~ThWuonFah&Jbp^+f&>K3
zPEq%O^NOa<(%kmoJ8>F6olH=DVfuMN=y1Seba7O`1-hla*Cu^bD!-_7vcZ$Qyg(V)
z&s{n|iJdL|Wu%B7fL6XLuh?b&S<lqxB;nGL-<7C-SJR8es7g1w3fl`@+|HO0^-SIG
zv-4b|N7QC0wSxrfNn_tDZhX~@{Q_PhH4+Duc$I$zT#6vVwP>cD9#gF0rF5tiSCVtQ
z8l95nyF0qhT6yxB=Xm@SqJDSnhpFNE7VBW%{+si@tn&ox*9Ct>-}u~PnayN6G)xp(
z+q0o`%i?g_WuDb*&71#~Fw|?~ktnVOzOJAN`5%zO6<%bg#WYkEmlUk4j^GwLPJUi!
z7+e2{Gs<ud#MyU;7k@T^Cj#}yaIf#rq^?&g%K08A*SEFX+}Neh8an96M@PN%?}DFR
zf*J9ubXWTJBDymAFp>H*1nT>VVb8y|3%|*auSx~r6Xx0meKN`8uMy~CCH{@CxjXQq
zXI}jxMTBb!`smF6_hOB)brtW4d)!of-K{R_b@$IW8S4sz5p)6h`d%p*Z5-9ko)O&o
zG{QTQCCe{EtLEO(6!`EG%vJF%ti8JQ*qNSVwwuC-H|K|L_2rzqU+?@SuqmBOYU2cf
zzK`gsq;0T10g7AT@4uKR5@3v4XPGpOnN@`f{aCLJ*%+Ywd0rXv%V*6`lA@+)(QJCx
zK+Wgg+Rqp!H&>Ogca~9`m!y<8UOn&P*L}czp0Pc;t)M$Dizog_?Qr7~O0?f!kbzoa
zi*pG>rX}mQ(J>3rKerBw_a+kPP*cOQAFYk!NbF_|es07joQQvT_t?s%n{)>v@RY>W
zetmNJv1mV5FuR_q9+iJ@ZDsnpPY2r+$DwEVKzO96LJ-Agg8RO<bpI2DYtsp;M$vh4
z>HaR=>g#g}bi=n<8}_@>*k*wG+Or9&T{=7`PaMiii#Z=VZ?k1w)JS@aZObPtWVxNG
z>Tx_pSA61>$!Wiq(N@8jWkMdN8;8prYqU@QFIY#c1}Wi6zP8l=bJ)A7PfZSvJiBc_
z@jq4hM)8?ZR6wZXmtxnxdb+x^YFv@=;c|A`LjQ41f%()_JdVqHt2ZxhqpMrt9*3$g
z!<5NiG(Hnx%+aB)IlrWz0~Qb8(>^!IbsEmeNB=`W05#P9|JBU=GSq>BiMi??gdi#F
z1)#|0C)ywUGhnmteLM!TtJ)w?Uc#kBcxus?Eep~N7LCWrcFn7Etlq-w#GR{iXzw)>
zo)Uo~qW`|+_-dFPlsn0}21qjL|8$lM0?H=>4Q{<rq<+L>kZ7eE*T`2R-c_Y%a^S($
zwR^hka40L8eas7Y`aT7X2TJ-j>g@T(eTXwMnlg%6m`xBMqT-oIY#NEJFRzqOTQG~;
z=JDFJ@V^P#@!wMfdM*TU>eO`lzcA*|E|Xofz)!w(_lfi1_fkLa&wJywOpNURb&;5G
z66mB^Cb`Jhw3`rq&aK|?d0|97J%6)ta4@4}{1<Swnsiq~LMkzTT%C=8pD^e;P#zA1
zd^`vgQ3b9=JRJREuwJ}%o6{A7d9lhq>fxmwb$xpHS%X5Ci$K=dFE_`Bz6#b#79H(8
zjT+8^T}yL*3?N~WOI8Oxhxb#$*ICw7>cKEw)7p_YC+w6{4YJnyW>sh&uUqwucy`0}
z%Hr3jG$5qf<(fg;U-8X=Gu=3#feyHOr}N-n)9Aq?TW_I*aXv4f?z$#fmdVDtyXfjq
zR0JUTzyA>e)QBG3o*>=!e`M^avxi~ZG<i#QN&vbYe2JEx!8LUcP^ONJ4BRO<f`aW}
zLOOO944S`$5y=EpSk82a;_(t(YzxrpS4%0urac|w28ys=_RItUv|_}!)g0e>m`2<k
z0U_p%N{KFG^&{)Or`Le>xpCP=!X8j*jY)MKsLh-TQq*y_jX0|XE=<wO1!L8^mGSL7
z<2+t&amW19a_#33&?U|^<#j-y7O4cQ7v1@F+qrTfirT>a=E7Y769V_r)JSv*#oKE+
zAAdTg26f^usptRuW(3^O28;?-S9)++I`n_HjcoeIwVPP%RUb}WcC4<ftyKKbEb#vT
DZ5^*<

diff --git a/articles/LC-MS-feature-grouping_files/figure-html/example-2-eic-norm-1.png b/articles/LC-MS-feature-grouping_files/figure-html/example-2-eic-norm-1.png
index 67778e54d8f8ea55914415ce03ec69e36052a08b..1ec5c7311de005a52b5751f90381f7bc874d71c7 100644
GIT binary patch
literal 126225
zcmeFY^;eW_`!75pAW{<2ogyUy0z-$u&~;0Pq%=t93?jnZfOMB4h$u*x#EgO<AqW!E
z-8FQ~F!K)2cR%mD-hW{KvNsFXbuHGl&N$A`@j3ED80u?M++n^0001bUTI$9C05Re9
z7X0?jL-yP2Ji>$6K}S=a@Fe_Ps<pug4^m$(3x5EBE$-$In7nrA3;=Kfpz127AqBg0
z!8v?pp^JE*z`TNb7Y&|YiKRx&YjBVjTnm)-T-Uc-W-+j75?2-M<yqZfOS1dpFWFzl
zXO0gUV}4qIUcF#kla#n4kyqfqy?=Fm_;hQcs#BqGf3I`y`dZ#VY3`rq{i;7(o&RjK
zZzym=(|G^y@Be-YRD|mGDe^)`FP>qudEKz$IGr@fAzoC}A}l#T4L*g!lgyH^f`UMH
z#Ph@rr^izK;9_D$lZ(yaA9^4x5t?gp5ZD+NV&~5}MDpf*#W6~3m}Bt3@#Ds>o&xSM
zFZ5my-X43<cOeFK%sy0N#D_n^>5MWqv1&yNX|%AJ(MEIat*tO!F-R(!ue8=<4%^M>
z-@w)YGl~V(vUh#T&x>6T9aWMxD9_HtGyKm!*uq%SO7E<|9^%X-2m}3f!BX|$-FskA
z=9Q?_5Zc0rFTHZqhNUk~QE!+r`j4mEUQPo`s>$FiaL3q{9qLM!!#wb*0J$h5)!MoE
z-~SFs7+?i?XqVQ*L3k_>*1GbW4b`mCLIKlSONvlR%}fcW9rro$*<O>PS=`^v`Bk=&
zAqMyWJ|MzTp;;AT{@A|^<gxYt*qK|133?F7?h(mbwn6c~W?;Wra#14wW)9vYrT;Zm
z-e9s(MQ#(zb6^zJUy=Zj0T`*{d=f=$G;knWqCdTdlH0ug6?EsY=Q4dNNEN@`%tm5s
zmVNM09=fK7HFKB8b$wqc-d&SAvajI?krF@x%%2Pc6vpi5;wM=@MNYNi!{Ea#{RfUU
z8=mS`ylxGTa3m}EMMgZITfqPKnh_?DDc#J2GOJ*-W+#41OZI*&47<|NjtiVx(W&1n
z&EA;23W?lmbTNvIM$ZEyihoRYH1w=++?S8iOzIT?h1c~iXb;<4cLL0Hg*nq<5$^-A
z|09a~{P%{gCBPB;_Am}LS=WvkZs_~mL!}6KI<lALe?YM5jHTjOTs=7K!4ssz99#)3
zt99r0bref3q>>Lv+rSB<H&D}~IB8l23KlYq_raQ!C>U~Q#-^;7_H9SaTv*y!#yv6m
zqKd;1arnWP!wU|eM6_JA7{K+^uTu}jLN-&SuJOZAa#<?Q;n~to45}K4yd?}~CsI_g
zW@P_-IQtE!`1-}rJo%}QTAGI34W`bMZ$Lo&|GsL}IuI*sCkh)3v&&^~KXEbU{%OGZ
z@-SZ~`I>i}W|^X1$)xdVw(SlujI!yMDBJPzHHIEDrXT3ih0#eaej4Iw(NtYc4zHUS
zJv?Qs^6B=V>=wGSvHzfv5C2ye&#=;0xO+bG(Vy~zNSb6G6Fvza<BZKN9T~Xr!Ey-S
zpv7U04E}-><Fr&8sh)%?@rvz*HJ<#f^TbO<Haz3c+zPx5btf8wb`xpg{46-CXE;Pq
z<zyT424f<Fe5mBQUR*=X0vY@wl3^H0d%>K#XEci}!nt~!oMN7L%aY8{Lz#aOfQkQh
zqm+d6kj7DE5hS2dyG*29wdKU|KpvOGoLns;kMk(&ozLGZ4oEw02AvyUuSfdf^Obnh
zqkZlTg%|u(zyDP{+><0THpSl48lcPFxnRj*f;q%4|5#cujPx__tP!5{uPjqSCWjUK
z-}g!_9T_BSkuckr|8AVHhQ=pp8jZ{O>xazA3}S#z>y^WF8^G)x?AzpHywazI2G~0!
z1xLw0rpe3)?_6z=q}_BqbS>|Cjx{kvs2^n;R40vT-!(;=>1_`du0pDCSjFwgp?6J$
zK_&j9)JiQ)TBFZh2(z-_Nhax&)Yb$I)57v@Ta#dSpl>co7a7QAib0CgsK+X|-YncI
z0%SR1IyH68B0dG|I-@FgcaXpdI>k;~$k`iDm`}8u@B`b2V@KoPCmGIO3@S%)oI+fE
zh^OhH#s)q3+F=5RJdnU8NA|#y<kn)Lyr%)sv&*&)`6za@SnOVJbg|d-1P7ID{H=Mz
zzvua=1!K9lbTOB$k?l=9-3pBior<FaE8xq>v6TXWaN8U0SxG&#yP-K)BA#$q#<U(|
zbSNa}2mNto>S7gVcbhw`pmnM*m)BVUy9WjJ>E=c_r!Qq;-K|O-9nJvB6*|;Ab6d!H
z`{)WO+IQ)YY9b`!Ei(CCQN47GTLN!VP!E2d2_NoygT7xE3j}32AgE+^0g&@_uRLBb
za&6?#)5RIn4->%lJAdbsCkt+4HJ)KuP%+n@!ya&H`X)S!FUxPEmA~HHw`VmDfY+>x
zwcpRI?ah>CUOdu?KBsF^(#J<V?(zcreapdPaYa8DWwfwGihu$;8Qh1_W0GSGW52Ay
znLnGZ6)&`r2~W~{v5<ZPMXLX>){WIL21{~-?fse}ZIV`eO2g>RL(aZ7D)yUzT(gTX
z$C0pbv9E0>v+0pzSBeA7jkgVDcNU$5hLfF$0o?~}*m^l*N4u&YA;cIA<L=E&Rp$$@
zd6S3<bd0xWBGBT=qZ?ohTvYq@G66>WI^;1CK1sdP9-hc|J$`o?7*2l!t!>LEjeo?!
zC8Ez6&h&)Yj+qt(bTF@vZbl4WAq+R3d<ctpNU)&)CgN9O4((R-2ehP>-CrCK?{m3h
zq+&WXHl$<((E}(VlxDnO*xb}cGBue>2ifKW1)uxK<Zp!Iz<FsIUo+x+NeJp+@$*7S
zkwDr1`<;US^JXkH;%3?+-YkW{EqLZ`Lzy>>?KRIVm?>{`F0=<mA$>!~?cz*$Ho$XR
z<Sp#IhoPk`+)xKFI&3jD)jPgNp74Pb)e7=;0s*#py*UYZN`PcJr9Blf4C>BbOfs)H
zn)Exzg`i>Uj##tz1jtofCRSW5JQ}4i)T@7Zp<`TcL3sb2bEAa+|4$whK0&UYmh(cp
z%E39ZgmGGK##wU3PHNqFL1>mghYLXX$TN}}`%)Gz=*plDTu$RFHS=U5K1~u!&}6rp
zZA3XjaB_f$@-&P0P~#`nyQ3M*!`WjAbo>BU8oXVyfNu%<`Ap_)pq`M;pGn`54_&(6
zMC%}uczhMT{Xt*^e#ELZinQi-h-kD(*L9>8vwtFf^nR+UPz}iZb^s_n-s5F2oFx6{
zPcIrbRi}?t#6zGaP&nc|F)2k};$>306E^#`#n)Rhh4BlT(LAWfT4x2TvU{l3Zv19f
zO29-JpV5%V<A-q<SFyJq7p-0pnO;5m8ZBy1Tl22<i-P2|EMetQ^cUq^@~}aDmDJbV
zZ5GH|AxxQM8=p-sbV1>qg=ovpL*$1S=tpoUt6O-1zYDuV|A?q1Rl~@oFGE#a_91pw
zr?t8<KZ`;nV59n)6Fql}@=f<Sznd;WW#20^BY-|YeJ;p+7O324t0f!~n-7=MJadsR
z1h;1}ew1O8iPP3x3R<*Syj{BRYV}L1+k0Mu1RiROfxUOvYRlmq%5|+ZVqC<-c4UJ;
zgKPm?Naazf&8O;`I2bsKIw8!53v%*@PB9;qE>+~ukESfKd%wGbQi$<NHeCmx&>^RH
ziw_Q<51=dS=Nb+H9nTffoq?4X-`iR?dXI+ij=In1ex^N1mz{;=c9}%Cva#BIO>49g
z@NGX1%{tVx^<NE)P^fMW<*IoQYZV<%TbBxiZY!-I$GS&1<&mr{(HF4%+XZ@MYMjCR
zXjhbkpgdq#z=effEMoT-J9K#`tT_NQatPxafj-RM9)iX16o;x$Ebeg50sF45nx7aG
zeD|_vskXp&p!pNw50Q!SBP8*;ey%?>ztXqF#*b73hwgX$e&nFM_!;c2!b@g-2m6>h
zfou814MnDI9re#00|pG;sLV#PQQI-c;uYP5shM;N@i6L<d?#spBmuQa_uQP;I}&VP
zt{L&`cOZ&!JM|&%OGf<cTOasBv;M3Hk30>;`az~A{CZwyR?a}89c}R6X3T7V?S;G4
zh!x|cQ(NxlP(_rbE~J-=6XFf2irv@OUh0|<l;k2vKMK}R&_W!VH%%5`oBzE$jgg$u
zoy-c(r@rxn2sO4FEqk(y9p;<z{{T3EpYWvYvWHpmQluQ(xwo8XFWX*QmmG7HtPVMq
z6mWENqIx_RSmG%!Ha;qt!vst7xkBGAh{Q*|Yz8qHZDEHSY3on*i5}CEk_0;sTitF?
zPq*{e?eA<R!OXC4Ik5s5*^4<ptlFyK-2p=+^LBrm_Ong*!oq@Z!C;ZO2J{AYeo8Jk
zd_wn3U;cY3n5rHEb%VOpoiB>>H*h=~eYLf#H1@4_$YwCNv*;e#u>IadjPBmWLww-;
z!EyQ86`rBo_wyS3e->}cV7P^i`^oMJ4g$s@@Kb%Jkb5see=UljASF+Iy!p}ff)Nlz
zB@(E`K+Q%JI=4tPb(!E|RADyz#7U;frMNBn2VuLyi*<jCqsY@7!5#^vTm(x@k!h>#
zbR<?zQ{IM<^0SPOM!Tt7i)N2m0^|+I$usbi-md~!c=NA3Rl2u}xfQGLD?}v5|4B#4
zbkJQ~)8W$1k-n6_jlzbmr_SrzYYrZ<0t^df2<mNIlK8XgS{^6=Nesa2JlfJ~V>=zm
zfd(%w=ZF50llI;2uV5{>U6ah@Z@J~3`*Xy~*1j+6iD#}v{v4TW9VUdN&`VIH(8^qL
zPydUj!q0{opLM<7r!}}}j*W%k_cR%_%dau}OIM0~Nu)in3Q}i}+Z_!x9>jiXxLY+M
zGtHD2G~$NnboExa!^AariOCYR1Bz3c>7s%T$AicPMK-!9)|zixM6$|ch9t`XD+OFQ
zOD$LooT~j2wFSC`7uVEXf5J}YD4~G_s$=Fad=h{Y<Ah9bLGT=0?gtw#w9Y~O&exU*
z+$%LyFI<`?26O31KpO|A?6e0NH$O6t@wrU8_6rAwRL+{BkNm!}nI3U&-B+<~?(M50
zq}Yc^kMY_|vI!>U1?o&(Zd$B>6v=68tLZ4w(AuYVS6$iu1NlQshfILpoD@&jsNO00
zS`X>^4ZS5pE3&y6{J#r0)amTk7fOH3XqZ6R)F!)GAdiJUSNaY_zEA$o5i`0We?>}w
zR+=Pz<p)ST3}`f?bnp{)=i=P++X6?KFNrVP$kK0pYq<s&j&y|X_t4l$50WVPSeQ-;
z?c3J@4rgHuhHV7~BkCnFXd-)XrUi*zw3FUh*zn!l_&*;rxN?ZFGNlV~rBC~~p=lcV
zlXnkAIAjh$?0gP~X7P+IkHS5rM=CmBZdS&{pAJ*A9lAOXhw3bymeWdI!%TNp-FV5{
z<vhAnA9&Ds3#5^BYLRPLYx9y7ScY6Z3Zn$f63r5)p!Owg-?wOdE0Y1dg=`K7NJ8E3
zSR{4)Q#RLpuT!xqKZ~y|r@ky<f|f-laP-auhs@yW-d@3TyTK2#coZhtdJtSa^{u^(
z;b_s8+lq7ex0P;z<H$d=Mo}>9v@>g8^EBQhZmBxFRkZ%6`ilddJI+9wO?`($6p}#f
z=XV2f&CbOeG^ddXmT-KU#zti8y(*LefZ+v!yx(_AX)}n=99ldm#4o&D#r41Id$yR~
zo%nP4{1J&n0w2^-HBV%F;K76ITn~hWAD*($Qp-DTH65iH9fy#K<QN%=Qt}lWv7i!f
z>AcfY9#C@ZUZgN|>8Rd|qkF)1Y8Bm!$ceNAX{IwOUmU0#e@oqM*)r9p33T?ldTMO*
z@8<ld+ps}+)~yWUK_)<&M!JXgOn$WMQPRCi*tmSq;7;jYw}kLQWFjJ6|99v6gb7i&
zG&=M2bI8$MWs+Wj+5S^nyM8J;_9d)Bw;mR>{Bn7bS@i4Ttv!}$StHh3C8~>OJV+I2
z+L$4OVd1q>_V(b{>yxy`VpNtSgw9q|_`)nC-XvcCb*QQLYB{)?zyE};<CPZVet7hz
z@?Slt+x*?CKLoUU=Scesr7S}jM-A>WA$~BWi^ZG}#ISS8N&rkuL!tReuxO?yE>e91
zW5y{+(Jf#F7gN4U+P$zL{Kvq=hJKTE5y~C35pzVwRcPjTHXGXQ`LbzuXJR+7nA_nr
zTZi<PSgH6!7`?b!bzx4VFRkxG#I09#f!)hLE|pbSQXl$iiYYNx*b$;0H&d+@mkxOv
zm=z!DL%qSX<3K;p?;2S{+a+ri0x}qcx8+xcn!UxDi-_^%PKl}fsDIBJ>`-cCi0D_|
z-0{U^gV(e;SMeB1`@rdQ$w-1=Ub#Er)n;oL82Ta~B6gzl(sqeCF7%V)HfW?TepKQU
zMn6&&G&@il>3@7KfIdL=a<zif1Q1LJCC#z|gMFjLnT|%Wd{VTT`i9CEq7bpt@>85V
zG%atsQN28*=xsL56h2w?$nd~G=Z$h5Z5mygLO1tZxHd@@Hx1z5Au(^KYrrH;@i4#k
z?E?YgeUrs}DEVUcT$Vy-aX)AGD_aP@?~BU5&|c?Q)NP7PFaaS6pDUt<r&SWiz%_{*
zc8`nq$N)_eFiWzzEf@XJBSq23GmqJjUP(%<8%v>G37xLlG5Q;RsJPMk`*p%cVXQQf
zISzK&7(3TMSHAL#&xKSgt#o7q^nfYwF579Gc5mg@&J!9f+7`A8Rt29$DjGKMk2e_U
z(uaaYJuDxKVj~$Z)F-x-C`J>wA_dSvZ{MFx4+Et&1$v-XEmRkp57u4+vH%A!1KjNv
zig}^G_zIyXB!j#zlPZJrazEj&n*!koa_kG&>&nryON-ed8?Em1f7C-e=t5q`H=YX^
z*?!!ZvgfH)R2k8^dcMm1$??L|A*fjLIf4D$!`z(y*qBdWzgIoOzbf;LP$QcY5qR!1
z)$`Kgk4()mQ`W&Pnf`7JUK)Qrbc_8bJvoj$cDJ+n@bZ_>@;dZ$=tmhxpT4LM#b=@O
zK{WyM$Em!~_u9;V_N#^V6x9RI@b%E)N|Y^M>S-|TxJArd_sMlVLO(cuC&!G|IYwWI
zbtaW9oUQbLB#JhKCo`5JfgPww0sWQujkyynZ~oVtr=Nb-eEW9_Z^e|2>frJ~JwBN>
zX4yccuA^DtRCJ1D)_QtMHDGtc1z2ACkS8fH9GCM^D@EXzw}3ka35a-~Oq(@uR#>cB
z4L*We7gvhxDZZbmHHOIoINc;eNXMsMgd+se0Ac5qwd1Kya(jvLL2a!+3WH(t1M^B6
zZh284Dqu1H=ZYCsY2?s?Vx;W(ka<7Z4>hxEBBVln+f~(3MEL?VO>)~>gU{pcCy(p(
zt3jNY&T<c)ht7o$L;|NkX*;}A4tjV{_ro*HYR#=cKj5T#X!@z>Y5ER#8bc?b?-1Cq
zD}%mQ62sg4>b}3Zzn>Zb=LXZNykLEvM@w7|RsyxZ_c{4k;>mH=0)B<)<zIN~AY0V{
zoLyca@;O)>yyV}Jc`O)`x6MRSrM|1gL;^51yE$2Rapxi(oi-l>+X9+wej#)G#M7^q
zR1&R`RP`V9T^EWP3sNUBctzed;i$BqEf`*l^#xPPpleqA$*McMx7~L#G6D}^P5wq~
z+qxikMZXl-glq@cTSOvt2>N}0s<FGVY>uU(lL4X~sHu$PGwRu>M@?4fPQeJt4msK!
zn1m0+`JTa-lDdiCG-30z*?XK5L@m_^bxKlmYSss<5_J4`6Ye30jFr<$jh1ywjunP+
z2i)4{id20)T%BjZ6}IOaoHdMZlZLW9E=Jnq-10StG5Esn<*s7AurF-C)t*k`_eb{!
z>ZTAKf_6N@&}T2~HqydY0Ps?mV4`s;Un=9Dzr!gOtRJU$Y=Au1Go}D>xkn)aUUS{t
za_1*!-z_HQk$0u9t-sA-ycgw$2OTxfohS{82ABFsw<?O)3(aVEK&wL)f6n=#Y0L-i
z4tF_&6I(*X!i23BgAMBff%`yQ($8CiIzT?bzA38uupds8zl!_6eMU&d774WB?rR}D
z(ygMCfp|dQ`nM)QtGgG94neB(-)a*#wL&|}-~>fM{ZJ<_&iC@+AqKuqxh~cT0-n<c
z?Q#<XrmL6V)KAOSE=Gx3XggoORZ}m(g)eqlMB;IlYJs!Gx`Uv~xjIJkzm^>d_rNm6
zrayoHsGd-gZljQP+kvott<|rtgXbIS7)2Ld2&w^^E*r7wHZ;gEbxFx<Q=m?hKyoc)
z@i2-5f3ce@Z$sXIP09Y%QLOXj9nFQA1u=oV*0XyCJA?@65g0%VnVTga^P$Pn9O`ix
zcKZXz0=Y{@wsOU;eiMS9p0Qio?uh3cK^*#^@X@QqoR=yASgHp*$lb>!0n7DML2@-2
z=H0(v5krk>b;@bWeCQ2Z@$;$yiIaTZV5PTp?1s`c!A!$;op3KQ0Q>UGf2JJ65wTIo
zOjH&0lvy=W+hOkIk?J?Ivyp0hCeHC+A*=wc_^;1Qv@r|-BpFr$9d7!80m>|!ZCF?v
zobl5M%o!U>k`tt-{hgnF81#2|EU<TLbNR=U=&PrUw`3%9yXKGscngVGu;x*3@<YLE
zmMDKe?K<Wh%>aI3=4C>39KwVzJKTbNnxuwk34d&RTO(HX^(iie*YEo92f?y-#C-nX
z?X<P}SYMamMI!3|M!wQ=`3s65PdiYpqfa|Zgoib(-}NO)t<A|6G*=z`5+PN})}dg+
z%tHvCHv_2;l?tlU{Zq_cH;Nz>Y3aYRFwZ)rGGj0hna-n!>C-cto_4+cvAtU|s+B5#
z^~fu!CKODwS@TE=U}-A@V1Uq{!>Q!Y5Jn%E%xA=IwoNxnF9||AJt4kA4n~P3M<Wq#
ztp{wmF522FF4ufzbHNOn(;Uk$oA;ge8wv*t!xBl^Y#NIG{;U&N-iq+xWFVU2L@faE
zVbmUqkW5F8nc)+LolAxOR$8C+B>#BN)-Cn*g3^(17h*|jZUKzerqkDvC6rMk6Sv+N
zYC_2$MR+jyQ!WgRd&%W2%xk!Y%Lz;muhf!O1t?HrC3qtzD8IzWA<jG7=?sO8?mUN}
zN1w?cC5;J^_?X#rcMbNzs3H=Xdmk_HQ7T-2+1p`)-SxP8_378I()A$|rj$bjkM%^;
z`WPhciw7Orir**Q1+xq7MLApA9JOo`g=JAZ7`D&2a+aq62y;J*+W>I@J8xl$(Q(~L
zs6EtFcPu)zn+2uhjNzp~>cra@%c(qDDxez*fhUL=K>zl<%mrX_4kHA-df{<gQ{Mr6
zzLL?Wr1Ft+W>Fym#~gou?J58W2XSZ=<u*df-*A^nmB6d-33*QDxH`ep>JMS=*P<oF
z2U}%b#cqZY`#8yiW$_`mz8Ud!1=^>_&o{U}?CRA&igj}mE=c_@WE>KCM=9&OoJj~Y
zQk#;$H=c$h4l#5qUL{2k^-s#s-d^tLeL|W`qxl|+{J|971{Gv4!W+oe5LZfjq0dq%
zk}%caON^mpbLb!I(#;D^sR;L8DU+xJ`{-~_`QG>D>&%`FaHuLNs`^wZVCuxgNuh1E
z5SPe=%#O6xZ+8Q<$KA)&rnjib6z#zhb41%IF1!YBrM&*<DF#`P8U&+}{!*ISdbUl{
z>boEH!2qHXK10Yw%>CMRMWkLHP;}huLwJuOXMXJ>w@Vs56lXT}?B`dJYi0~sO4t-d
zLEiTQUI;C(l-LSVQn9LZ!y$A*@2$|DEC4M`k&^N#e{Qe-@1Mu9YP~bEBen5W0#RxC
z5CPxa2ln1~r~qQ`gWSH=$LHhQJmeo6p3_PQ6$@w=>vBRAH~_XJTgZv(f|p}-H|i6l
zawf|G_^UFzqcr{YSiz4dIJn8bB=T_^6`RY^OIO*BK=FJkg&(>8=a2l)b<Y$P#?pCf
z<YL=C?w=qbu~N`V3B#tW_{mlYxYg@v$|R)~H)50(99B+)1uEu83Lf_P{I?eX$C$ZT
zEO%%2R(t1?NV;hd-*_M-nU-vIO*AT+FQK{S`IWbe0Ten~I5?+gO($=O+&byIMipXm
z{LqZFZvscALnVCj&Othm_`%ltZGUzs1tCc8$(1*H+klZ>?31h`-Jm-Q2D31aWtap@
zU<x7!{|09uG`Y4qGcad%Jsl>|xu{%Jvj0(F?6wry+XStY&5|@d8B&>S#{06tn;R*4
z{}OQ6IQq2YeK^0kJw>hekl$77&;p8Y%kJ1JE;@i111Wz|<E114X|A=m6z~#(eaFul
z$L->0sStHet&bNoJ6PM^Y#)nie-`4?x_(qd3~-_LVSFMGpMcv_pdxW_KD0>k%qVqt
z8*6O))PVWIOO_(o02aGeocy&@A~iL9(npdThQ3c5e3#)C)p#qE1N3Iku?I)nP%P8)
zY67_$tk2~=F}$<Vnx;`iSSn1EuRm(lc+qMqDO?tVk1ci8wC`vq%K4fgi+LlyWX>WG
z)EP7Et+%X*6Lxr8A||mVoBphZoyP^V%lSF5paZAvK#v-O6l%H?Y8Y-Okj%}C;Tv!8
z^tkxdsydH;&i7@ZmhkN-l~a&dwLcqm$K8IAH$Kiwm2-po_&0XBrq%w=JIq=27sEOp
z!Wu;=E{DAs?g45KRL>s(>KetOgQ2HYieU*09imV#G=z3-SUz}<y_;|}XY{v|HK*xy
z!ur0;S%&`f-ya1X-KO)A>aXqC_^hJy&n2#Zi}SOBTVg)D%>P=Gh)ys@yI8n;6Bfw|
z<@tR5b#$L3U0ZXpCIcVcK+&Q&ol!cIi#+C@sI(c92nJvfL0L0LGH5CLNH;v+shdd}
z$WEXiP{jvP<(j7*2F~AdV{>8A(z{_`TUVO;g>~ndw))|u2YB$vFe?1=;x*H5{-c-<
zlCN%C`-IG43YkFB+iP8|@{?0;Fq%_=M9^o`dBQGqqANDgGp+}5C<#DwIb`e(Bfe$}
z!Nu+tzLo}wiLt?W{6=ueU4J#)tOq`<**%?(w4yP20e!~p`|B}%fTz<|;@n_f7~G8n
z;<m$GH^uI5cs-qOoMwZ`?L;O_3}x_Mav~cq7-_SXH`~T`395SWf?}k#NKQY?IeX5O
zY7&GpyGw=v`o4l<7J8Kk4(C;P%1=pKU29AwSTPT^LDQ0LUeml;dBw-dYlGT|_x!$b
zhpK?M{Y}y2v(e{-DzQej1bnJ{7Zx9Q{^qwoyj^tkzEXrJDx%3ClBIYd9kox&DmVYc
zW76A&L%;<;Sp`yVLLEr!v}M!8a5z($*bqPWv&?0CXIlmx^w#Y)SqQG~`9rlthRTL7
z@QJoeQsWmtBVys5n%LuOaldr*KMomt={zImZUtdX;|b{pB>RG%$XhulJYRDCGf4MH
zawlr@r!a7EH>XO+0NyE$2c>l2ydF&eKOp4wTDsz|HvNHV-ilCyFX6`>OBv=eu#nXq
z2vT?be)8Q>0Io;jKJ;08hRS9TEKOS${54tt@gq_iT&(x09w1GFplGGv&fCjo1u3cp
zl<o@2``$s%q!fATf!rEyQuoG?xce*F!M?$?5&-rA;#!txQkhZlp<+uzh1o9_rY&}5
z15zqT{%AcaAZ>qNB%tS@VH^5v$*i=jbK28y@hI^}!uC{Ft$-JB9xA+x6H0l5Ir3g{
z18@?}>k8jKQ(2^fuXX=88wPO<v15w6f8xs#SMMk)#P};LpJHHvZs%2`0WQe~Tsxky
zM1R*km!Ywa!P&_t9@43O_0(xL6rRh{`oD1(zz=NkXS{u|qlwmJ{~@8#ZK4D|DF%Xl
zh_Y55q#tA>-hbdpKpv6{IrOo*VM4k4ysnYB)0#koEw4ZJs}D^)2zh&gP1vbtL?^gT
zX#~9d!bML!^os;1DfEmt$%UX_0g#yZVE?+kNc94Dr_18{x?z#2P-#3jhXpcEQ<}qR
zd+EC$gIZW<;qbM>r+UoBpL)AnVOauJg=an@?+)E%Qxgwq9D<&Az<k~hg)jr)7v)!v
z(N?HpLkny<?ZrKZuju97<F*RhOe<qYo10*k{gcF1H4yUEZHejW&}`o<B0Q%NneE8J
zxdp<7Qb{xuXrHJiiHKJ5NBEG!;A-~cv)Q!QKM9)aXR&c{uWasIV7xH2%M@Si@lkZj
zKclkJfM%n;Isi*(MnK;Wi{E~8Ii;Rk1krvBfEf(y$F8A<C(o@%<8RJ<I9I(Xa$dS7
zf${gk@A&kxTAY37#lB~WNu(x8?;oW_a5J0Za?hwq6mT^66aoXy#M3A}nJgcV<hd(B
zl1$%J{`+`Z<(@0;q#kX2Bd<Hfr$nI@FTh(dO@v#ECI(_?eC;H?x2D^Qsrwodx`)>R
z6giD&cn#(^6<)^GSA-OH15l1#zsJI<#82Zia{^?hGkKGJaVH19UEyqz*&0%WbG@qK
z)Fxj2iU_*`I7yERX%D-N0Z>9ZGgQjP5wJ_j@v>mRd;xtwE#Ju0VaQevM1<HnW0zC=
zZKY5%$tc-2OtGSZP;>c^rVq$6>chn`0z#P6_NRxL>efuJG&X1t&+nr^O$vYYjz~XA
zoMZ!)DBTne@IGZ>zhT5>iKoKv-pP}Ff27<Y*Vv%2kMSs&mX)`JH-pUnAXTzuBRRN}
z1cl;9&a4?~m2JJHj3%>5T0$m7*P9NUU2KPgInYi|%aPyGzIcy{5z1(pCNcqjs(GHy
z&w8t@Vc*=@J8u<rlpjo{NWLK~ami!|2a`uMzIstkZDHRUquy&*SoUWt*ZHhtSp(E-
zd||LyUePYCOFW<VUEuiE?Gp0em$BtVnW`Sfb5KK)z9=m`3EJ<0GsR5w!mATAE<viN
zS`b-+1Z9Nc-%Lj+;~&;u9HdjoZ}ji_+v3Vw_=i`f&RF37b&+1`y_i!}@XiSQ*4c;D
z24ClA=b3`D|FrPKb83U#o?#N{KJX9Oek!v*h8@H=lptgPzuVlpwRoJ^lGO-h__)_-
zEMx+ckvHmGw%1D|b4zT9GDQ5R!ndjV$1nO1g)to57-TBLEoZ8Ir!y*-T1)|@zOFVj
z?NqoJr)PVN%fJWw7#Ff}x}6Z(=HBZvOYg3^p@TD<%rcvb-3<pj^R|zaNZ@h<Bz!eo
zNr^}^x*##It;Q}%h-f&j!LyCcTveFa+_mqimXdzPlfH%UEZ{-RNesncdq!R?!9@sl
zr4g638+K(#z3l9y78tl<KH7<E#~R=iec|S=4`WXfP@o4ffjV#x&Pe*Su*SW(Ze8S=
zXE>zx@i*F7$H%l~gGWSZR`I9F-qj{7J!wp3dm3LDWFMG=Hdm)Bu*VcX8r32?834@n
zqalup^$nQ$?~^)xAZr1yKFEzew>zxQuro`?B43vt-%m&8jvfVcy0VbT5u#Edr2)DA
zLb5as0m3nT&8Wvq^PyEcQ|A_o|ABk}_JQ401LG7?F)6HH0wvKrVL|TglCj5&KYqXm
zCS}$F$+x$3POKv?!W@7KZt?_m=HdS(x00}i4FV1;kAD2ox?v)SlupihmOUI6-kx&b
zmr9&v(Y#3^*1UX?B84&#bvLB;#4c0>KA_!OYHFqpXnDYP^>)qN)VBQD`G3zjh7jts
zL)s@c;w~a2{*}tIQ9XUK|M-sd9$g)t!{xN$v{GgS8m<WNWes6ZTPiv_L+P*M;k%%>
z`B_f1<%5=8PmtE~_mcvm{f$F(?o5`Amyz&F-E8+?C+QA`tf1_AyurJ)CscmdzvEY<
zO+IGOz%MvD8?zmcSuVwl6#|vGCcr-@h3+@|hq6siStWdhQ<+4j%g25UlrKB;FBR`1
z@L*-E`d}Xk01{d|>(iAnfZr`Meo<aSX;4QM!aI|Vhlz4Qa;vKy!}mqJ#0)-@r~}x!
zU7oCdFtmYaJqvRS-?-VUIp|}s;3)=t$y$?zn*><j!rE4djOL_^6omV^ohLX4jNW$H
zSf{F7DzO-@!QRglj1G@>ef_>_JZ-3|)XwVLv>BW8@`HE_fdAjfIfG+_g!i)R#0T*w
z=tlKlOSAoSikU6HbH)a5!i~DL?ZM`l6KnZqc{Lv!8aA@iAmF&kP%!mrSa>Gto2LT2
z`0bJ6$enhc*!<1V!g}7=H@w8+%$%Y0qr*TY%{5y;S@HU%fh5`T|Aa3;E%j$NHk?n+
z?iL;zoL?or8yLV3ay!^60;2BzHHVyuI<UbR;vW$C*PYptZt<j$huXZm)%3w{zASbw
z8?+$I>VH3}iO!YTCR0PH|6yf<jJvLgmk&qkQ%PzP!0Tb=s}|sWndEIUfLo=d$|=jS
zxPy~IFjwsO^rWv5hTY}fLi6iZtU)H`{NYX$)2>Za0V&&D18-Z7#JTZ4#$2FET4nN4
zm@QfiIKRRPm;6Ftx*kYd!?jk@lNW@{LsCC~d)~H1l7VVnd9?r^+VhSy*hg@8A21Kb
z@{Dka%grpPj;-5SWz$cdQIvEk9WdsC+F3`#xu!-je|YZ^MvJ!4KYjT(Jf0aF$Lv?<
z8HB5ta#1`MsHVk-xE<P*h<I{c39M5W;u8?YQ9iFuIO=l))B?rE!2pI2zB+D;_#g*+
zd*ddB)E453fs*d`c97`)VdCZBsg2XW=F&Md$_rdiQAK|zqxCAm<`nI07yoeo#qp!{
zxml(jv5lhd8#uxw`}((?PQ%3dZfap!9Dcw2K9P{0{Z5HRU-4{3Wc@)tOz6lzQkXG+
z_)Lf_dtsPQN#y`sdC7@-9z`{NL5SeSOh@Az*8*P^-Di`C+fHMjlPo=!FH|{I?G^-^
zXNV!bA{QR)6_K~6F4>U4E62VCBM3cdWhjq&=3mTqgzgk(VJ{vCLG!)O00vU%(gB3x
z$g7s8GP-=w14L1rfS|8*s#D6awz)d_ZQ;@==MeUC2u`2Zisj=XBbsqM@~c_t_!?zz
zy@e^>^Npbe^4To&AUS*hu~j^WDu=^3H9q#gcl`C78x34+7IeBd`I{o*_C@3;{2}kS
z*(GsPh+5}M?IxCU|GqH+i+?HugXu!{dC7L9W~uq-nzN^Z^_PBsSJTZM)6=QTlHiAB
zPKvdqkWdBqVyTy}S0DP(7F&VZ<zfxSehR1LWgglY*UJ`eziG2HIUAV54@*7&7yDrO
zwL~*VNC2l;3aZfJvdiO6n?ZS465#nJ1R(Y`{U^V5q1P}My|H38Fv^bk^hO|Ff~mEI
zVzXfx@#9IA_+qN*VP1Ote3lo-hjZR8&1jao>Sk(Z7y~FDzQY;Gm<CpJGX)h2$}68X
zqQ(W~Gc)nIP;wVhK6=u5^$b3D65BfwUiUXRc=|nBSqF)Idm6tD0&?`M$)oOsNE?N_
zQ=tV+#lprt`S0dRA$1a`JF;7HWMe*#bfjHugi4F=olYW*dDA`Gvr4)8y?q%>4?f*#
z_(y0SRyesQ*YoPd8UFSdUQAoR()ow$^kzh5ewIGGW6;yC>&YMFfRF*tyjdp^)-Eha
zOPJ67w)T61Q8TU=%*Ka4+Om)fZ|TQdtIS4hJa*;)pDL6TD345)P1@!>#wob-jEvNN
zuJI8k<j5YLr!aT?M=-Lf*Tut+X282}C$j?srgq6I)ngz$VnQYAOL@4#s176=Q<oIj
z?a-c3zZbOl6(V)0_^H};9dMxM$dAI*`3KR~hVj)BkK9^Zt}J$L7ZNLIhpAhRI1&3F
zP7uukv2=%Vr<RMFC_HCf@FZBtAaL)C3!0@{>0&3IEC<H_h+%@d@XZbaR%ZaN7O1ny
zau89hFwz(D^68T}8vXh9bKWTa$$Z2sLh-W$<xk;hhupQFFI?F8I_$GJCU++8c<_v$
z!*clFA*y3LtEPH;a4Hwt*g?GJh!b_$HE-{RCTucKBk0(6e^x9TpQ|zWRmfdgHo76k
zy!qp%trl?sS0ROcMAG(?e`hu?=chLxDTh|X1e!ur>4C>vm;RH4Uij>>=K~c()AKsN
z9?1vRztnCmXzj+#c6zo|AJ1pA!Cr1A=jfKb`1T`Zh{Js)?E=RkxK~8cp0rJg6Y2Iu
zSajzB>hEV5vG?|a0T_37sUdQy@D;{d=F=opz+x{VH8U9`7tQfo{uH!87J*=#J4n`q
zE0IKfcqWq8(|nZVQQNs!Mn^wF$eFTv$a)ZuGs(6w)n4JBKwYIpisye6$Q7+LxKc`7
zqDiL>DjnBOl$NQ61m<0<iK7rL6~zW+*y;X;FhA!11a-`A<;IaVzORZ9Kn9bJ`v+O2
z;?mIrhzncKf$NyaTHY^yop#d-vD@nL#8d!)kzEK$X5H=3!j;c9WcIU!9+gk^0W{Qq
z62!f!jL48iqC<)XJdZY-X=CODHtYM9U%=y&0r`kDmC!mBP<x%e7uIsYi;J%xuRN$M
z!_p$m7G`(wT)M5HBlrd&+ozf%+D<R6pExt?NsAwyI4_qhUI#)Z?L{~>FeJ+spL8;`
zJ{9MI0dBXxBqUHpq!ovKX}fy4sKr_=*{3Sn7qysP*r3pe_+gd&a7#-&Eh68n5JMKR
zdn6*f%?T0I=W|iC){DS<<vb}Tbl-}(l@dGw`Al^Uc7R*El8q%~{T*@Rn(ABg555dw
z*n(hNx^}Oex~o?0EWzl3(GC?c%j7*jrh4moitY3UFT^Tr36-*@V_csa>ggv(`?*r;
z>f!#!$GV3iQyYO$a?r`ze*eHB`N&d%0AB&=mYbTFD-HMyUgKa=G??bV!m7d=uM5)u
zruf>T%!>rTI1Nyi8$p~JIXwFm4#(e#U<<i-*sBy{-m)yZco4l4H~N?BYwS<M=Lb6|
zT~#LvY|RZEDkLAh{`b$TwrGl}$ql2zd$3iii<K@@B;w$GSdTKTRucd=37x^|3C}kr
z5zWZC?O+wfY)%KT*HY4le{|YbM1tiGadnA5>12Fvv4~K-KC~N+*IvOl$O8<E>R^vE
zx2^%b(IhOt^Im<#+nwC!RkPHRWEs4;lgW(2#Gg-MuiS!|4;qt#IDFwh_g}YY<g0Z9
zlf(#JkVz?%_+jTMJcU)mDd<7uyW@CZT)Reu(f}o2{zvjR2fPjewX2Jtx5Z6F+vr<z
zb8IqqK%FVwmN9tMhE^H#NBKpQd@H{OM}yQR(ipvnB>=u`hnhk0AFt2-yMJtbj<2Xt
zB94e-tl9>7lh~#zoQtS=?5AL<Q>$K(_)>Sk4oHo{u|nNMLmU)3JEezZ1=mQy=#?po
z>$S4piCyP)?oSKd`M!jtYbfi*?cb4bx42^0sSk65Q+VFb>#$hqN~}lJ&Y-tt6GM?N
z#8<_#Q0l(a=N*Cv?M_*Kx9?W#%1HsgUFT)weFUWSv?=~0OVLhA8j}n!RCi~1fPD27
zk+5NDlze<L@^4JH18U6z;HvEY;pD=X?M5{I#N&O@6hip)Q|Z-lVt!YS&@-5H?H4F)
z?w^Y=1@Z9v9KMxS^iecCq|G!tIT0)ualG_0!;=(YpueUe{hsy~kWN>LOj=%6z+2<E
zFwRar-AfDzEx!2psxxPw7Z*KAz8rB{M;s!Ju10~omCha2tA6vS7k)46URJ7_T2a!8
zfTeAJZoVQoJS<z*UyHZSfk@rcO1Q#_wP7=q*Dkt&D8x?tQJ{eT`M#5=m3e?KPeR*z
zedM;mVPA}TIgHGgGBws0qt#@&O|pV4b@9`83-|Ktk7Aad*U{;n{qwe|K<`xe#~dm>
zFyeZqVvv4T&00-QpzdBa?SV&95S3q5U2;oBk)Cz74N-(2<4|=fA<;I*xkc^(;N*C#
zg9yYmom^$Sv(Rv^c_w-YD|Q`L%I^+WgDI%+iS7$5<GCMXQhbR$=?nYiq_rF<Jr?qu
zO<cJxTPv?X|K!wGpWO3}<{zCIIgcA=bpb3kcE@fiKYDJwIIB|dd0tUZ*9c3x$kwgY
zeHq((oLD$aYeD&j1{#=^^!5BPz;I~7N-ic+#5CD97_`H+W#HABdgQ!(eJZ7{*LBu0
zSP&*==6P*5xTTm-{>HY?CUgSt#O>b?1W3sdL_wWxqnSYh=x5tsn|igAZa&yV{e=@k
z<<Ei$YY$8%wCT7=3zt6v)MRY_v`$skp<!dB$nbm_Ks#cQ<~lAyJQR0Auf<$jQ<oP<
zu;A~71^UseXJ9Jn8mxg|)wfplUf=oZhp2CuQkKzhI7=PdFg5sJ7Q{KqB>uqX`mNPW
z@lzB|@{8YNYA}E_HKzs=dqC-H=vbNabUFwC|CR}u);fLSEh4cs^=(r43<T30<KN|c
ziY>~B^yc|+eN90Ot1xmgR^a<g8^V>8dhtR58y+9)pU`_ye)cn%7HaI!5J_R=(;^XW
z{{!;$C=P41L9D-bUw-K&ohZxxQ*H2#pGr<|ROq)rDqrYRYLvE$9rfQfPAe}D;8vLA
zZutykV+c({0hcj_I_)q$5dFlZbw_JlOZ3p>V{@$x2;)KE>s~NHj2f~DOG0>wLT!+W
zT&+!k8+we>_AIJ|oJ*Y^2egk`I`kku%!{dg6F+jPyeGUI-tL<-w4grvblI}zqY;$z
zeZM_1fSu^e-bxNj{fb7+O1DzqVJBsVDZ^TROwv@tbvE_;@z%ob=a<Hx%6<3VyA9y!
z_<Ey<X-t3|0(AhI#a*kA%L)8IwVq6O#&S5?pfqy98p=axcCBC{gy55WM&a_oOyh?E
zRDO=Lx66&^{QPc*lma(GU+4UjLbG6?TGl;6uo!^u#27WxFW;u)newt-z+B92H|`au
z7njQ#o#WZD>0~{}#pmy(bWh7_B=e=Tm`P)m^vdq^>n=yy8#HzBRS4gb2lN?z(kb+M
z8z#Q^hs#EFrHbbZ^<&qW`ugGQvhhgam2bkI@E_i&K4N!OIr+UnqwFIVF{drOZc=2N
zssp#<`|t;9Y>|AV<mObDS%2DDEaG0DxP6{~o%k;h0C!?2#RJ<DZuh$astOwl{aGdz
zF$%V|%SjQk#gAbSaq#jxYWGT0LYuwupP$|i1I|y_h2#}8KkC>87=U$F3)C`>2KL@<
zICQIJDd+VF!<$_;ax|S5l{Co#<Z8c;N@+(i2u5E>DzqbK)3YVRob{>We^4z_7S>Tj
zi<*nceUjZ%S>%tf2rV~(2F<7TFmsySEyT4>nNmrT^93#Fkvs9K+dXj&`qDAAG6^bX
z`o44C`+B$CDxRjBw!*qtvYp&0J=bhnQHHzfx#~3=;KJGs6FJq}hsB}tdE8jdpH4@;
zDVoC1ME7zQ>fWkZE>ULTlKs8-c*3ZTa<m%c!}}8L@Ya!nK@Y=U4JGtqh@{Jm>dl(Y
zrF`$2LdA`d-#ChWs`|z_l9r;)oFTS%wFxP<*3(gU(j29dS-<v1o6$(0-ksFYf%6Ck
z@OGvDir_FBS|ln_dv=E)J#DRne^o$9)HjSDBsyNd83d-HI(ag>If$42M6&9C3^$wn
z3ws&w+{#4`;B9lPU0~1wt=9vNfbe?GLaM&AZ^iGE=}`#DtAC!-ta;PlRm$$_UhEEG
z2QYRIzNxK1f&t}pol)!;6pLg6vN)NE-|41=JW+6MfdPZxvF(@ZVpqJ_yRSGr*8?)*
zWbQt2%DTx2J;1B5L9Zn}n-#-eopTn}nKulQ$^N<OVOjxldwa8CXe)O3ije$bK5U%j
z!e2YtlVwvXL!xu0%mQGJLqwzjgcj9W&K+v*89(unks5OInUDSqmbFSdtD8TErnIe_
zclqfRBi<f*{eDPD*10!$REX-}$YX@?{!)!{`0!)-)QB8`xxhy!Qp5)(1S*WniG7&c
z|5eruWXHZNdsSD&2>7e+3|J&u9HevM&jccBy@TAs?S8%Qd58M~tSLhuk}i^zMXpiB
zAWJRjvzwY}fNeR~NQw|{Oqd`cW`AFbgXuNjB?BS<;x=TeP7~Tgzf<$C9sD@nviPdZ
z3?&?g-jdTAtghGg$?0jgD7RyWk$%|~|MA1v7wQy8fK>8YxfyT1G``m!Yydb1O>X+-
zcSJg3X!$cIn-77X9ufG7Si9VW?T^XgM06IuZ{oL*0Eu#UpXC~!o!=qwQ2sLSm0E4I
z_Vsl{oB7oRs}?37u3#QYvlLab`-{9}SQ;qmhHlO8D;RA819HC!gRo?YMMP1&QrSlU
z6flf<_kclru|WGYG1|x!j}(SRSKnn|Ed&sHOD=$-MQ8=v)JfAqP(ktc5^DF-fJcP=
z@CXjO3<Mep;_K6j3KtjjnXS?zq#nMj=}6yx0Se3xG@OMxkEq6be6D8>oi`S0qH)fA
zY}{%qmvnzPU>)74)<q~4RLqOMyHdmhe1|!0Hv$|qtoLuD&q)bUyp_XmNc8|r&R2rE
z{@-mwb^&OGEtxCB2(Wlz!=QAd@>|#pw5&dyR$L}cz{IxT3<(;1F8#ToqkP?H#n286
zNUuw7V#Zt@b3vc10pYEtt~*{tMTug6A}m8|vkoX7UZd`3vXXSFPpJHU;&D#7&W2}*
zdeLLVzpfM)Dnq8sL%8BsG4&tT^_5J*<eFu&hi7d!0-4`gdvE$_y>_G?3rFmfwI6q|
znHN3{E|r`XQIFHM9d0SyGz`tH2WTh|&TGln#tkE`MQl>?A;Jr-P7_JYTcZ?hMRUQ4
zb|sAvkXt_)^?zLl@Evch7zAk{7+c%X7u%4+@(;p@M;wEH^~u9l3ztkjDwsG3KcE8a
zehD3u8Jh&46dfQtVq<?Ni>aU>+kzrHR>1T3t{*U7vAcL8Q-cKvu{;=BRT1%Ct1heM
zV)K^F&oQ-Re|=1Ls3ZHwb?`>nL;q6C`-JP@W9ngo_~VeX(0k)#Hl#ti|Lq0nEK@(j
z!~1WO*F0f684U7qYJJDx?7=>(e_9fx<Izx_S|vlc1$bIH_@i4l@#w2Qv(1fx5%zlT
z3eB@3!9GY`>ipEu<|5cxYE1c@TEUgvlWnurL&tcrEtYKtF|>(Q#OrFYC>Z8Zxdc!z
znAJ*P|80O=zNpMIBKQlOe<tZtsGNFdZT_CObt~oZQgpL^D(@eD2g!#iWlrON*W>U}
zuRCQ7>MtJoDiES;6Dkg8hmGFQg94Vz>3KocJYC&;?TOl1FC}Lc;RGXcEK^vt1}p0K
zcB&R+!}uCk7%bc$_}E#&{)P~`2Gg>UT?QJHpkKUpyUXp*;464opQDKt9Rhj7Pe`>O
zakV0!imB`gvNAUVn2^nAB8(NljRo67*sgAJRA-UNiKlUR6<kHTC^|sFsLx(W996^G
zQXImekZ(R*ZuIX0-o?SUGf7L)TDOu#Qd{T2xm73tCFTAok^+FWNQ#90iGQudeEjbe
zo^+(oh856KP8Y!-;j)d-<|E<kAiYp#sPZC|pH_*obT4irW$?`xJ5}An@NvP2>P`k;
zV<h34nN<3K&Wr(pS)Se{`%Qlx-_sVqjrq7YCK6NQL&~bZgfX3@?R*eBL0lbf==nsT
zbb-H4aO1ATfa_hQ{DEk>*X3-V2sda;HNu~`VV-Hdp<E7h&%|%-%(oI^vFanGCkyZE
zL`NQ%sLw|%TRxMbAhe?`8ZvKKE0f3X^3%`<ImNZV68}>^`njTPZ%-<jIgYsgw(pmu
z?JXXmXgeT|Y{k`&6nbsO3dm7ArEnKe057?zSACt0Vs3Alf7n*I8?H!zC!y;I_heo^
zm^NVsLH5e+#QL}4P1^&si(d#~X1n^>?tV?dHQvd%wa=^c#Wh~Dz5?I1<$*f<ex|$J
z6D6Ynl-Dg=7w;w?YNk-j1yc~NMU`@CTfxb6<X{gM%@F4WFD{GIMV}V+gh`cW1x-E!
zF4j3oWLiXreqVV}f<Kki8erN!eabo(61_e8-L+i7V>Ow}jVv=Z+rP`tM&<!Izo7t%
z1Bf)x<(N_(D~R~yI~b$B-ivtm*f;Sx(L9;89-i2y^17<_Ido~~hiUM-Qi16>g`NPl
zN&j(vB3%f58V`l{%0q2qEhX}ZUBI8T&|R%_iH;qy%R&=emL04p=h_!o6XY}_*g?1w
z>Psl~Zi<$D{o^N8Bc9}eI8u%3B6LeOnG}Xnby<ax+|<I*x5ooB?7Gl{@Sj$xs#j0A
z`Greg__Q*Ih@fMaJQA4yXPAnne@wGa<i2qd1L7%95wUw}sN+a^SEli`!86_(*Sj10
z$yV_%lr-bjd$zh4yZpN?Wt*ip+wXz_p{gi$Zp96n<&G;W3VFO%o<^Y3IRanYasDR#
zGvolj;2K=^&w-691rRiLmyP95zTi7z2F`sE1hQ$RTVIz(fa^^mw!lGPpMs(08y!mw
zFxNZp)<;fKE<?3@r(Z3rg3htHi`mbu=R<@0gS%sM8`oiT*#QT$x#2<mkwc;V8^ANO
z=feMJaslBxa3l$R2?rC(>tp3IYsNc&C0{ah7k6T<96EIIvrR~p!Sk_a0gf+kUvJts
zEBy=6JjHDY6g_&=5;OyJ5^iJK7eW6&JbiUs)8F_17>v=Qo6!w|G}0ZC0wPFC3X&?0
z+y+w8C4%%&5R_IxY6z0jD4@h71f)ht4)&YAKi}W~+kM<~?>XoBJau1Phb>z#H)f_a
zeR~=U<lWUP{4`R|P>4<F`3+M}D%9kQ%Q5aoeB1;m`ay^Lhsy}?VucsJHfZe)V}Xfk
z_%6>-bS=qY0vO;K#%lLkECDb>az(?)n8(cd0H6-I5*&G)%-+8Zzj|@0nEa(6YZlJ$
zEm16GbXL5kJb)L7IeC?iCl6FXcO`J{2cQ6GOA*A4q)Af1x6%_nD)n9M4ps>0;-Q-A
z=B<l`#EittNT$m1xAqhnfOD#6$Hmcnc7MEYVIgP=j3Sy?WZ5slJ^nDFo*ZGhmCVcJ
z$DPK0`P5^}*YTtOE^L_y5&5W?6m1kmvOPP7AoiPXem_}#xfKyr(4V49xv{$1CoL8`
zhsRyN2w-(w(JgUy?_z^}^+i63Se_5|=qxsgpnjh*=|LOP_Yhq)aK(dhT;q<r3&A~Y
zk#(iTo?tHABc55_{NHh9<XO>Q4C<oI_Rk|ZUn;drOtmJi3#^gY{tDkw4W$5?Hs2?k
z=1*Ljz!+rzE;EaI?ycQ>d-*GUO1ZG@BNiDFKytb^5bzKYZm2gS1g<?aSuLp8{5Y{?
zHoy3J{ci72<zwwj{EhN}F2#e{B(~9F&1OG9L9aZK#^M+vW@q|6_iNi>vDe3V=c{{s
z6ww0m^wg+Z|J#}+Pxrmcqu!sk1w*Y_G*3se<+7yul6$WXj6HIdN7A16LV{A$z@|f=
z8_RD}a1uw5wkyl}ax~)4+>V!|C$T-_mn#EA6j^vSQr(L$a#NuV7{ldTci*A&fi;ET
z8a8<nf1(gD*q~HC>P;4|>|t5RG?mg28#F8zMwP?Ec4zM8{#&RqFTLhP!V?=B;a>sz
zsnq%!B08!Gazmqh%_YMm81anAmjKrHW~>0#xLp-jfnz|>rA4&f@}uSRh=45rrjHDe
zu=yO*BaTH)UmfoJi>vIreNC3|A+OUtHK@MQtkb3zyBhR=JErhbl~!ojLxP+RW6Bw>
z>};PeC5>f2ZmB$)*-6&`R%8zCawGx;%yXZxj>h0Mj;?e$WSyefWv^=5SFS%x_@6>-
z7MMn)bme?<kLnjZR$Sx3XEiPOIv8h^WJ_o(gh_~=p-~g6uhTIGHE&5rQ?{Ey__#g1
z3+CR(Vn68f*`>mNG7}*}(cy?LEsYmrWyxvSeyg-2{Lms*^^l#;E?pm6uP`9aa_cD(
z!m_)|RUpKU)J^RL43)3oqIdGm{~;8oNHZSr4R8$cMs>a~`Vi-gy<*5`t=rsd@uR`U
z8>dZf?7z5_q9WL;CF>N!MS#-mT8_=4MsPdZV0v~8U2o&?cf@Cpk0A<+x-z;3;ENuF
zl`8H!INmTLC?1Vx!1~?FdrosoiLH$ZNBtH3=^Xx(G{F{vMvb?nBmz^4nE?Yol?j$R
zBU@zi$U)_F2}O$T=X<F{_JTA@pyk$|?YV^A3D@WF02k#z?bQM+)whBBF8=Sbr~uZ+
zW%XjBy1<R<fy8EJkB0@?Rox@?-ZIj3r%QavW6}f{QLP%FWY0-8Q_5dBo9hWPHlM_&
zVV=LAz%rG*k6n}mKyBf|ibbK~%?n2!oeqZRZ_>{a=ioikLxa=3SJmBJvJCyXZ3iCE
zG2YIo@5XE5j5EA>z0z&9bb_}88o7p_{-BDGl!012{`3w5c8^N6Tq8InDpb;`NyiSw
z&VsCZeOG;txcGe+?d3o=;>lgfX%x@zWT^Q(oAbQ*5NYjA&PCw1A!plA{{`{%4ITOm
zW<Y|4A%Gw$k<Y;+PV8grWl}M)Di@*MW9Bn!i7TtTx%E{AbnvPv!*K?#8gz+|v>V+D
zk(<CH4`JIge$(6sS_)5l6tD8P#ySh{b^_J!k5i!v{%uNvg#Fl!A>I}o_I_{(VJphr
z$Z#Pp?pVr7qOqviosjNo7+j*=vW`??01Fcp(-dgx`?(TKGr2}4z9oBB%SEeGYI&{4
zWE66v8lV&hbK8-mIbIFlxpn~bA?1|GoIEi#I@S+k!O8?B)@z|X1fd4XZ27abk23Xb
zEH_BDgO*2Bw0FUz1jCT0>LEkzl;X@Q2|-Vi)80hy%kyoa=7vuH?vvAAi2vG$`;g$X
zxj#+v3gh1iB{>3+gtJLg;S{@V?ETRjt^dT~G~tb`k3x%IXnb$-=iX%Ve<dKwoBk%V
zp8l;iTY9OLyHa3N{LgD@KWZItWRXr2C4+5<neL^#o9M(0MNd3+&jy8;@Ug9-!nFW>
zKnLFic%yy7w}H>!gJ@<)%s1V^RfKpIF(>$($ezeSS8TtHNC{$e#mS%?@FD(H&08Bf
zLH;=s*iAis%U6uIh{owIq*nr1^$QZqSK&NNeq(js$B)aUYL);`TC8KLwZGwH@pTvq
zsKaDZ&(7nf8Ts+=<HfJ9Xy8;>59}Uj>CPsSU7v(tW3G%@<v8;fpPy&rL*GZ-IO$Xz
zx|eoKGJjosLQUyWG?f-jrj?iipMwf^ueWjU9AHc2W$kXLFZ$P_^o3ytVIMA%K%IA9
zlcMd2hn%7kPz>?I<#_JTp%}L1Ot}@qS_z3orpNE%8REk;irk3;^Q2U@+KsV~%TE9z
zUYUkpr*Z&jz^#<J2ERDhBbR9U4)TpC+Qool9=B}KYPi4~O`^&mQ$fBXrXH=kTsy<~
zL^y0$7I*T!bo>6M>CL?x-?z>urolPy7l08o5x4N;(nH9<=45P22>->>LHW;>oXpOO
z8H{Sx@7(R%^7=eDc%>l9<sQHLS@;b0+C49b1NkFkT4%r#{S}i|_2otKHJ86SsLg6N
zAGAcHYJ>4zdxcLE%PmWMc=4v-Bbwet$GwXs(^i~~<`;3A5RGC<Cs_qg4rx5+`k>Ir
z`#!SeODY<=T!Za>MtZ;oXg)@$<HwCTs-BIxWbLFii>q~Mdy;D-^l!$>@k_+)fQtI2
z)DSp-m;Js5ZtXXlgkKG60dg;(^5rax5;urwl%{MU&eG@06-S(9a65CUCH0Jg9E(!M
z>30u~HNxPRXnj}JkoyBCojs#F@@&_hX4x>T{c^>G-b%^eN+<dB2Xo;W`P|a<pZ<!Q
z@Dv4IF~|PcYob7yjhYYP+b1kPYu*o%M5`{(6ca_67}ZjY5v2*z>=`8H373dH4Z}Qu
z#N1SVPA{um-z8hAF9)O$-9P0Flzb*|nLh14Dow>dXt|-z@cdrvfyQt{fd<`XLY%uV
z<NShMeDc8$;kt-<{Qmlu|H`*T$l-4HUDlOo((aCQ*)qBp6ome#!cA^=ecOMEq-^Mk
z?^?su#oHyW)IJZTgWtOc4`r(@q#Q;*M2>i?%QKF2n;q8GMbo6|Oyax*-njHy2xtWg
zp0-s>q~9z5i|ENU($LLObL~L{ggJN#vKo;&dhnJAQ3AZiMpBEwnI~%ixp`iNNz-l~
z;ooMnJjR+};|#=UzLn^X@NP8PQ9A(VZHDTz(-RXThp)^DJ`uBm=EIl_y>7kFs-F_f
zDC6(i3zw?n%6a3(m$%SNH{U6-9&(qnDI5mMvFB$0wA)vE@cFgt_UsY0`j0I?{kR;!
z;|(=m;ywhPx`io+y?J8iFC)`e9uUCG0!Bp3vMsOrRVL-1vgGWYE7w!Y<91>et8d7G
zef20W-ZOFJ(Q7dC^y8mmW*G$>9~om`Sa$q(jhrdvVcM-lpFy3YB8((tb6YodwPc6~
z!EtiL)fDX^8G&u+E`S468Af#K{QTpdD=i6(0VX_z3(zyeucZ4ek`-WFt8V^rsuIjR
zu=oa0(&BBPLm|1{`Uqz@s{9Pb|6C1y@Wsx%Q*krt>Ol@r4e2%U)KgrO_}y;6+4r9s
z>n|xYruYkZl;m33K<!|ho)EcvHrZcSlAT@hWIVg~=ht$TCB;@4f5$CZB9DRCCO*lf
zD&7}&JK{JwRib-Jg~DTB?%qgqhur38W3%|SVu+!~q8ikuDk=WyIZqU85L}ZZ7F?c1
zvzXd|=L!s4#)@BiA6Fv)q2um{4GGMwyEMkMA@O?0Fj?ctw>pDBK<;_XpQLw;ZVVSW
zzyDagXFMoXj}h{Xb7i&Lx1ZIcj#8143GfUZQd-V`Xlkvh?W|mo|4r4b-2T&q`tQB7
z+|H~!PryUZDGRy(7_yg2VQDa)*vny6chto{)w(6iJlBvOBN8J%hg1CFP-b0_IqM9m
zY5n3h5IYkM?-7dW8Rz_|#Qg+q4QOT)l$J1{;jyTzU<u3-l(Lrp>-$Br%P*Z+ynnK2
z-@<nh_y$AbFep^O?KbsW&CD*|4#dr2Wk7{4vX1VMW8;#1-34Ws>-ZLwIY^>bpX(ma
zC^K=+hrPd?a-)rb5;<lYMjJOltwfQteL``3!e4O}C$VnM6XiUYw9v%DD{Nr3IveDh
zHxI-y=a|=MleV$QOm|E~pRZv2>OC15{@ql*`)l}}=%PnF?Ur%%x_KG1wYOV4Oe?<!
zE9O7>%HI!vZ8BhtCCQ(9%KqijBIJw&Gi^*gw88ifFpMXoo(OhJ{f$9ho_x&mdkb{n
zAdDw<w_*(3Z?rv?(7hFLVKG_C|JCp$qWEBC8nBoFt#r!W{8D5>2=zW!-v3;U{^R{g
zfp7SBQh{OEW`Zm0S(ee2SI)4LdW1x<Q~Y*$yrMt(V)YUOoSPJA8xu|>PiLQ&e8S6E
zUmSbhMg{aIan_0ZQ5pPCUwkreA*v^$HSCziEx$R7Dk;=E+Z<@{(y{T3;{dl#R<qjA
ziLGgyxdas-=&6-zj>g?6N2)3fGo%dN`|CD<Sn6`lAzqLT7Bsry9`~dVfXJ;v6XFUM
z%2t2!&^VapJ=Wiiw-gGh_KV0uX84wRL`pOc1PgAD?mx;vik!)nQ)dB;QRzZxJk>lq
z4!>(Q`nXWl$*fb~A0D6<=&?NO%mi18o7gcT1sqS-cWdz0cL5#JX4oIsjrL6S3Y+g?
z1E0n8rz6$LVQd}oIu@^&>6>WVXhC#TsOo<f?+&{p6&9eWCH*Ck_Bh?d0*C`Fk7}9!
zkq4w6uutNa;>5)CdBEWTa?Y^MU>#zNQTAIbX@298)_MOHQXIWRcSI#!1^#z+pDcSB
zaL1Og=q1)=@-J-#^gx*)pXK>?jEhf7tRH9}-*|LBL&Oq;S8Mqulj<hwL!rtpXv@j;
zKpfH7aW3Q@UR<4sO88N{U?cYRjwJ;-V4m`==&|My`h9sWx5o-A=Hb7HM;}T;cb?dL
zGxj9?{puJ*z-e;EY|I}}u<C?aD3_`JjnFu08B@j)i;x9M42|cfQk{}-sHt+{wlIW#
z<+rY@vHuW!@XoStR)>g6&E$cGY+{-(%kP^O@7@}iJx6+9*8q)he*jj5!|L0KvIYDh
zryvo-diD`#q0*Bh)}ts}k61lO-*%TPYixQc+`OAG6p(nCoP*GM^<yxxM~ajw*yiaW
zhFh=uSzzjQa?pUY=noNe_#dtasZ8h8<9okjqKOZO<cqFJ6L9YxeN1HtK~xE<5B4i0
zm@Tw3Fb%-1owXvMf+!{`7F{Cn-5=L!Bo%r@wzXV3rtYaTFeE4f7Sry3;Og%7hcq)d
zC!3J?@3-tJ;TF&%-Ss&lSQ(&Xl{2`86fc<+&EeSlccZV|@7=>omeUna<0PLDWAYL&
zZN3(i?v^ZwqxMX)U6sfMNgwF3od}ooiw{m!Hn=fla<E4Coe5$6v0L3cRhZLcrGI|r
z>Al~0bE9pqL#6_Qj}HjBlNNu`P!OoA=XbvrS{`F7ddErOcG*y$a^)?)@8xHYAY&?L
zUR8PSp!;dhGe`A#)c`&@J(ZyobB4peMHv^q=gw-*=?}WaxvrXZ9%xvN^J%z4=3atM
ztf2x<OqOW{^opVLJQy?gjqetNMFuA=b@VF>9$fESts#W%pcDoe0ZREWqmy406546z
z+qS;M7!g;}OV{by0M<92QF}*Bx%Y!i4e_k2FmD4H%}a~EpVY{ZCf7$Gg1%Z_$7wP~
zs0RaVBG>T7#tvqqbdyfQTPKCn?~3=9X?Dau?GP35JLM>{w}^TB+%2-)Xgs6FN@4;M
zFZ|dMRzvd7ucdFV+}d@rl-IHa0(20Bu{dxPC+ADVkZE<EM%l}L`fRp~nWMLw39i{O
ziRFrbT@iqE%;~*H?>wse$%p>z@j*wR@zIjSM=e>e8_rq0i5lar54+xowD~iiy1UOy
zQ*{5mJ3YZbLoW(|dpbl!xHI*r;lMt<VotjXcli#pR2RC;%6R9EoD`q(E1}hapz?gp
zG~#Pj<1NYCe~p{pHZ#Q2jTR^Ql)OB8wq*5F;RHi;XzyBYBt_f!URQo-K1>?L)bYgQ
z@kbRs&JQ<6xQa|wBaFu@-sH1o%{IGFQZf3RUL*a$Z}hB-RGpR+t|(U8RyE8oE_s36
zBD|RT194WK%R*p}0OS?kWSJTsl5UR>Dz=O-_bGdE1^09!eV#RB@*7O0mvNS48+7#E
z>j1Z)DJ!-_*(gxKpZJP$D9_a8(anF3mLiU!mPGX_b14%IKsvQW*OgM<f{51rFc$eN
zW}}m9q&LD~4xAI`3%a7+exT{bi7fb10@W9`zUQv5|9625$mgadD~5||S?}{PK}XVG
zycNbJKAvpYhfN{yB<qHy&jbJozm;gG+wOA@kZE1*sV{Wgcv=*u5K96O;-l^mL?>3Z
znhivvDJvrcHpi>=qf-3o<HuQ6V{*7sT^~`trgwIS0-q}xO}XC_us@CGd`-$clf#p)
zcIVk{m+QBl(*e<}qLsIKXsL+{*54A}O2v{^jEXhc^lSe4S!)48{T=n7#@nJK&-(G*
z))%G`9ZevIcLOoH@@Y%j!HCjp2aMds8rIL&%GIZsHe`#;#HJ3-(P^Vcx~RARZW9v0
z_3A4=O<wS1oV!IjXw;WxapRNxFu~ar>8cqzC8>^6p7NZchSd2MxLE{ggkR-UkuAl;
z8uV2NNeM#&I!cw@8fly;_X}UG_anS8=!v=M2z3mdjk|Fd&U{L@Kgri4eXtH&KG68F
zIeQ9?Wq!pQUe6I?AfY*WDd=6VO31UIrc`$U_6a)_3gaDqq(ee=VgMBFzxxww*u)#y
z?JVDDU31aD*F{=D{>g}5`axz3WCX0QXBO)9_x274>Zvl-?!<{1dX+r(;SXdu<;^P&
z$R_^_AFDki(_@0=i9q?ej%`gDm%?d$SHfx7sWdi_ccz-tUvblI7ZL%$y*)^aKTEts
zMf|yKnxgZJjMsFa!d*Z_01zmTwZdpV?}zXcCYVZNssPYaA^Qjil8EMpp6kiqd$)>$
zfkE+CAvdm_sUTZk&)>df=F4B$Y@wr(-nR!_W@3Dd7Q{RUn9~;XPy<F$=QmH2fB)w_
z`K$mucJ<$Kw!5{Jh3fJ4eT`MQ-d)judd1}`!i{#VS(}mBC|B<D0x-%Lklu+i#awv6
z8Z4pz>7~(+Goco3;?t~z3(4j-LzI7Q|GGfYVY!ugyS*D-ol?+SXbUs0)x)2vz3?d1
z_Wf-l6XWdw2t8FY2|+bU)4f|Vki@-Qi1(ku%Jo_sCW)^nvwP5s%|Z4g<B{@O%=U!A
z`t2@H#qDj9u=t5SRmn<s^?Z-(BfiUj<m{~jl2jEaPAx3*@3Wr$3qH$M9tR}eruo3%
zmLm}^^$>tQ5iZ*KW#s(AR%O`8NP1Uy&+molo`wDsc`VtRr?DxO!f26O4G@0B339LS
zXRH3^zd%D>fEeKTrs(Dr9)f3(d)>jee|lTs=It0I$b8y2tkLKzTn<i3=-7)S_~uvr
z;Lg*KzQ*J`dAc>KDUa~}f*t$%1CP;%u{2d2lN_g<J}o&(Q+_V@<K0*VS76iG3hrPo
z_J}2p7X?Z!0zHZP1+UP`<udqY6e#lHYQasl-Mch!4T>)gam*JbLPj%ylWeaReZ}to
zCD2EWJn;dfee|_7^K(WrTA~DqfJ9%G3q1A#g28I-W>p7!l1EYTdutL9bCr=9u>@aY
zI1AQwn!|_bBiu~8fu@$<DDfn^i^1XMw_F)x<bTP{BMTf)HZ5!w@M;RLk#+Q|77{WZ
zUPl@Tg>m9SrXAFNH>!bO0Ir+@*DOq645G46uL?<54PdXSW=7-7b*FzlOD-cNC@J<a
zCI~}`o{>?#VnGiMv|0~S3COeU`g|B9yeTjnyZ+Lt+42u8(!7EJ0(c<oh>{{i71Ve0
zD@_vNrawy-hOtgu$`NGu1)vni@x&F#a3~IvUbNdgeG^NEtwkisTgoKUNM96d{KdzI
z!5R2X?`xJ~nfYy7zSH|S)i(gN!m2el0?1CI%((mR%XnFfC-w<|TMwDbv;kG6@Zh|r
zlUj1J*`AM+@L~EdH@xvi;3eK4Fk?MH37Y$bT8Y2If%773aJFRr$WA%!)h|iZMG4cW
zI@(JsPaW0H0gRzs1~PBGTPH5=iNItI{A`6@Z8U&m#JgYgMrmCNg#lv?HC2&E2UpaW
zr~Z_wc$NGYkqoAMaI2^LHu+AB6`~fle?IsDU8aF67Kz6QmoDlY%Gw=_@>Jr@ne9Hl
z53L7c`%dKG3{Hz|+3DQw9g?dgn1Ze|h7wb)iJz&Y`*O9OMxb~b1wPaO&f-~k2heRp
z#*AJ#0MrImHYnRW`|^7O1klo7DHO>o<+4%THAqd%vJCEV$6N}5t#AMLcT~kKle1#?
zW4vdDycGIXFT5Iz40UX|&9wZDSVN267iD<afwGjw2w=vhxl7z7fcE?sI!3mEm2!h(
z>HhR2V(PIzns9bK<OiZXiquQ_;MD~Z2lK~zPbpkA&iMqpOk%NqA8*H5Z|%LBkTo2_
zZ<Bm8NPZ-gIrq34{oYe`DuFIr;p+Nj;C83_3w{RSX4=!o%eecM+?tX;AJ}sUnjS?!
zgk&1*P=l$b{|=9%x{F_HAEQG9jS)aE^nJUVh}gv0r`xL*5C&<^Hy5Eym~X^OIx7$K
zB3X|4jVkN%gXpVBMB-522B-;_OvsCFAWy?4{k$gr;5+Cc6~fT)X(Q8;mp?xMG=v;y
zqGUvsF%P?oEi<c!a}Vv8v4gK#Zo8g3^<Ef;-ZFydM{5fpw4u}=7As8}^IEPRy<5pq
zKbiW!UVw}Uqw6rB-liEpp==?5!PC24Pp%S7#}gu!8yWchQg@4NdnX6mXgzN_<{^>y
zSw+|7a<kt~np=E82X#jgQ+%2SUiD*WvL7|+KOqRd`&aPPJWX74J+Vit-kbwc@6vmj
za)?>L$f~*>$@%+dcpJ0)2ql-SY>X4&^T(6Nbj|$fQtiMID`qgBMwh1aY9st1|EaV6
zf#Dp%QdB(gDMbUi%LjHBNXgjt^s3pkw{|+(Z(f!kjgdB8+kM@#gXwy9UT305J+qaL
zU-MWk?_$Jh1t#5;M%W_qGA>utj;b>e!|nnxoWbc;dBfJ+1h1W{mv3B(s*HU!m&iLn
z=ly+BGz8E!wm;AIij%+w5g8&tE>eK%`RJ4Kb&Y6PoT)e=RU*aN6zcKE5vn{7S;&Hb
zFQ%+WzjKk`t;)aNP=G~n1_S4*u0jgkSlO;$II|M$n_nf1{^Bbi{vl_OaG(EV<DL&G
zc(JFE&`8qE`(ue?p}$8AA+`acUA9Lr#dE6Umhqq;Fp{vQ^z9<I&`Z=c9rtv02o=Y_
z#SndS{uo@ExZDeQI#9SyKJu}X@2tBEW>9RZT#CqF_%r~}%%ia@%?Nmb<UI027n5iq
za`ZY+?aOy*h4iOUB{)jGPBY#wyhVW4iA-RFRRYQQEzGvrXsfDHt^H<7XPsh6aGw;g
zn0PU%!Gohw<EwNsRl!rvv7Z56EYCXqHwe~k_=gfO8<XS9^hw*r*&PqCe;PZ@kT%;z
zCc(qR25AKM5h)^@FN!k=3QI^Uzce92lt*X^0s)Ov{g{t@>z33KdJo^F|3N2~e~Xcg
zh`6!<DM9?zu@B{D$kz~>qrcq7Ie-1u{4gj&{Wp@?Alb91lQy<pg=a>}-M%46<vDb<
zC$d?`@$*}mI(WxhvWxN0pO<FF6sPRi%V%LWAX_IR&Z4bql<nY`-Y7|1-{%Dm;sE~y
zTo(bmJ)GS?@uKWgvR2j6xyUh87eB`iS`i^-$>kv#XZaqsVBpC>7-`|Sv%XC2mia}<
zUv3fT)zz#TG|uk+$B|u`Q$d=t8t^OZh1AN9eYZv&^O>xL0Gu|srHNP((bXffnGE<V
zboNu1D{0;!7(~=gWnA@M<Q!=Z-y8rI3hJ}4j;`unM+`^n{2)B%S?}HL04@NjT4xZ2
zB;2LGx<Fr8yx-u=hRLK_-wt!ec9A}*)Q;Kq|Hmo0^-C%6>H|gF6=*<|u9MPGaP5_*
z5&6%GkMLu7G7Dngo2!d*^RO6EC2U)YAdp64z0Tcxn21|O#Hi??N7-1rd+q!kzk3-4
zZ}1QriyRX#(@5_cM2O;lW@qakc_@*ZM!^$~`kwF#+Ytx5K$C=Wi-~Hq=>%Vru<D1T
z8YI<80|Hqk#Z+WZ2XD>zjIvjSIO*rvx#q3XxTnvMUZ5;nS6{op$=|$vW+%cbU1aTv
z`Hb!n<f^w6|GRyZbuxb^^PEmqspG)iYfN9RF+|a_-!)XJ`mjaQ{<J{YD*EzIcN9Za
z=Zmd1lQ*A=Gzq%Q&CAimYEyiy#q`Q_1B&nFNpuZZ>zniYmnv2-NQge4ioRS&GkxcG
z2ks->D1-f-Un&LgV?B{Gm6ioy{9S=ps_bWu-%0>zt)|6d$}5k+mknKJMYge#(A)G7
zbb*N;F(6r2tAaKwd-tgWF<`<7djr$>=q=q`G^fSaUayROxFVPDeh0}sk_tausfy(`
zR=?3x>1d{Jd0zJb=ht*+w~`18Shj-~`7g9oG$|=@64Zo|^Tt!=&(!5r(q_ws0CXt!
z1ZC7yEt2KWGijv?C?JYbX~`Y>qsVHL^e?}mck9ZFOSJ!yq2Ma~HT@ISZGI+;+WXOB
zh&Du8iOcYhN(vW>-~@kCGcdGvnpLGw9@)i^0;kLGI?_t5h!s&eLfxo<rKUlj{ar1%
zSLKmL>US)Wo!Mxt+!?ofW2Oy}n&gEB`n|~Ph{1S-hgt)?pGN~tg9*lJY-mH`Oxm?o
z0n)=gjWxlhPrUfP#-#6{zfq*T5*n}(4|?c+aUE;Dd?&bCVFiM9#VD>WCal|Fs?p$^
z3Vc14LG64sSs71FSS{wC77TF755F{t2<j+?q84mroU|?R={q9T8t$*tK#>btt0dl9
zCv7^Fa$yQp8!*Mwa{j)J(El`pKu|d<X)f_5)N98cDy`oF^Tp?p>|$X@ohsQR($q=|
zw(p@Ie^B7qFS%+<0H(xM-rF(1a=C#T&G&S+q5q7-wvTW9p-AJ`RA7841ECRv7*I>Z
zB9!V^=Bq~P!`Td;J79JnUeerKNEDHcQZ)uh2nEpZ30gxO?7E!*gHfgd%*vRF9g%Xo
z$^H!Me!wbj%ez`gKMNi@o{tHQb9Mi<N(kZI;SjMr#|0y~DCQup(UIPi#0+6G8r(X6
zd`QWKEi2QWPQKe`KkcM4X|nOS{PI{Pq8X;P#B|H4e`1Qj-f0QF=1x^EIw8J0q)^Fd
zl=vxs25-JBXV5~ad|x3e(=H!S5En*decfLW3zmc1J<Wu&XULW)Qjj@-uEemdx7#`9
z{!w~^0z=%BMh2tzOrMmNI}kMBA5FcX=0{()@gn}R6ba!1ejlhuYRxqFR$@Op46BT=
z{Su@xFFQrXIT#(sNE>aM5$NqWWeeEj?t?A&Y9T0iQKwzEoS$=u9rn?kUnZGdXcEVA
zi{YwnKFmYd+&sh=K2r8x&sF`67$T65c4XPy(+Xi+$iibV?Au*0O(xBrW6DALPe@p@
z)(4?gO+r3ztNUJK@~Qr$N5aRXD>zlf?`Lh6g6O^h9UWxXk@Q<W;^D&eaoj8FJ8=TE
z=Vby}<C_{RHpEmTrxgE%;rx_~rQv=@>TOJdxlG!(et9tk`Y+GghZP*d73PgG-7f2<
z?hIVN{eo3R6EXMdlaEfW_=<n)-4Tj_PYS+r7Tsg@Nz<0$Q;*LN_+pUq$X-Rg!{=1^
zC>BO@%KFUE)86`NVL6@w2FBI|Q;$pgB-fn<w8W0Ue{9xqj-Bvz^mg67Rx_Lu<5^3;
zb2EnEt<ydEgKJDq#xeI!{MNfQ7`F?a30Y@N+`CU>0geJGB~^0#MJo|zLPJ4wI*DmE
zlL!0FZ1P@N$jc4YH>I9I*!*~IqGo_WM%5@M>HP1X3Kb@!VygqVS5j5q@5FLW_m`f1
z+4A0AL8KwOSR5eFlQ!u8E;yh+d#ad?ystAD$u&siC6I(Om)*=;gx3-7dRGuDiT-LB
zcHak!X0)E@Vx9k9=0M(Vi~ea<9$g)Y99EsT>KpT5KhTQZu6TLmD<(+bJL;*%l4IrD
zcFK<Tc(G9!FK@o_)G}LN%f1KPa{JV;jJ_iwL}S)2Z{sW}pB`wq9>ZsGPg1f<NBdyQ
zY`_s~&x|EUJH*gQJg?oW9uZ6fNlrg^3iD^WREaXRd>FL>y{0QIMhJ`lUMQEI68+Cp
z*S{#3)i7*##qbJzw)8KXTPe{DdzJqq2LyhphU2a}Px}3<m@J~VXQyVR(~PN?;4e$h
zMy`u&9RVoFzG&0ma!q@45(B4zO;(v5{!8Z0y&y_{w!K$^UF}TltsYYq^$HmXu2_`U
zr)nmNkgW=1Hx=WD^I<{QV$vr{TT_Rv^xFqmf&JgnY?=X(uAqjVS3+~^UhnexIk+qd
z{KtEmtnzs>e1L19{?GmL95t>M`co^6m^;o{*Nz4?zC@_|MH?PE8*bv`4>W2k)Ob@L
z`%3Ye*|DUN65<r8xQu)gk4z%IJ5+e+0*7t-$Nm`<;*&uPy*X<AUuEGMgZruAAN+;T
z@ls&UpiEKDtn&mz!Dq9)hj{JziKPUY?W->paCMufzB-pwpW?E>>Fqy51jHW;RoQa^
z3u1sA`hP<RDt_Wl8bOf0`1%LS7UqjLiI=xsEZn=ejKbO2g2a)<=~q|WY0-e#Q+VS(
za5FML6{NoQ6UM!Fa)%=GL;H_EzG9$QYk+FZN8=f>He>)PhatR)X2C?_o;zzRohSZ8
zUtjsb6=fk9keRfVgYgN{vK(nQGamBdsQUOKj=7i39idHtRI(~|R(<2DH#99kBl63t
zCLP5p)_*iF#<G`dDBf@eU#sa+Qmd0RBXV;{BXx+@{b(U|74d1zjn9WZLe=KoyDgd+
zJX<ifi+CjHdD-^vn*pR6`)tw`O_ux@$lsOR?@{biT`IWHwJe-<rrR>P5E|4KGhz3i
zafW|>d|1_9v$6=pFQIJ{PULQI-vM8(e>Qv&IR5LWIwqzvU#wgskUwy?T5(M5sFmvr
zRtB`pVOUFLXWc%t*U3mvbk6sz1oC^km%S8Zl>TC0F`3g)4$qY;bM|Q)KVsI6craGO
z-%&rd%m=COXYk%ri@sWj*$uc%cSB`7ieZcV06t+22I^`I-@vS3#zaRdQ`pge6ThE%
z-r&2ETI5_fxQ^+?*ic3`^3l80Y9p37s!T)jHKy~DlUJl-@{YdJ@H3(}o=sa5s<Q24
zkdnGT0p|UG7syj$m`h0CVcbn9ys7FNArQ0<r~Lic>(R3xR5<>sUCKAai}9J~&mJ<_
zP)D)MeHZ9RAN$E?%_J?@qZ}(kWb-GM6X+zqy!GTRMIJzsR2}g#%sFilM`x$dBmI%v
z_0V~IPxbHRdTqy^B&9P;04(SWWR9J*v=Gu(_vETb2TE}ZYScK$d4$eyL>EXy<?vy$
zf9hX8@vOV4H>$xTB4z4c%{-^MYLXvIRm9eHCHaX+HW&j)LxERqeI$p|Hmb#RfcH8i
zGOjJdy8E60QKRI=!xV{sYK!D1FVHpig)*u-JwFk@pAu@kI!=4bG*+jG>Opn}!>%Nz
zx;I?*SdTnX&-c-;yB9n;4Cr1h!Ads2{PPo^5pki`d~Qc9O_oJUIw5+!Vr!8<#V-^V
z_=ZiyO<a<#!GBx!EehSrFDV28%=kK67LoG%qJj|A!3i(A1v5K=z7K_NI5^L#(L_lP
zom1?sM<S$R@G61K_xJ4nLDo%}2Q7D&(!5kXq5LBwT;FAEx2BfrHQ^AT@`fb36&?L7
zMI^vM94m3Up3b@K;37fc-tN$i{3`a>JsSp-?mz47Kx+q@+KuidxEa~7DrYN0b1>l?
ze|}3Jtk_M1*9hZotYAPf2o7+t+%C<M+AxBMBc~?T6{Rob7RgNx0u6p<YIf2nNd*Po
zH{Y1CruzQ{HL9M<DEbk2eA$4n2)^;<=*`}&l*s4D<DVbf1x_wrFkyqN{pfs(M)i^|
zIE8NQs$<>KpF0m&U=*-;lA8Vv{(q3PA@eXnh3termudge;>}7D7`Px^>D1?}-!hbQ
zd%rFnrtk{{eXb%u`;NE3AY}M~?aV)~O|k@0eu%Qs*b<#L;{HQtqWMASH$QCL2s%DM
zx!#Qjjx^Tr=0s{RwlLxF;N7`0N7Ja%i}qikJp|R1SU&7TS;(W!DXD-a{9h|*k2+vw
zXx<a0fAuqhQief=4X!kR%9lMNc#7PPZAxqkq14)a_L;LFBQ=IgE2fAc0|%*upexiO
z#@AlM&mE_muCx1b)8Hsif1tlw+%A~_{mW%bcSmE(dw;sqqe)q2-b%8d5)}FTQDAw%
zD0&Jlk*x%%x@se9>AJ@+BlfSouB<@YVYHU*KE5k&k23U~s7}1UMmZC=-;J*E?z^DW
zDfD|J9y&D8m5}{t)WZ5JJ~HqL)1PMt<U4?0p^?Hhf=+bO%q_}fKsm3l6q=UI7VdC<
zLO#IjEqSIZt8#|-)ei*wp{~5E%JA$<{q{j;hSzv!=7LqmT|sCG8zf?K-MVV$I}*>M
zR_!K@O~ceu4DD0(5ZY%rtP;kE2kM;P=*^p0Gjlg2K@0Y`t@HNXm$94>261Ul!jDis
zSn7ccRn_1?Vk`_PBjFlt$5qw)!(5CG!}ELn-Xrpv=&%blgL!U3ZGrLhZnrsg6>n0w
z$H=Qr>bIt<ax@^K^Kvy&@lHC!W2qNj&*)pH@_Jqe)XBRmJs<XD)jx416)jwxFx--S
zQ2v4C&dlH;;mL<15B_fn?+WG(!q98?tUL3FYsgQ92_0Pw@kD66&R0OsM^$5#{eLL0
ztXvO&0izAm$hk?S#9!alpvkqe+nZE=Q^jdl68X06Uk63%3nK@aS|%9zox(jzkk%So
z_by{oP4Y1~V)oTEY4Ww7`H0$5gmL(CH>G96(t39>d6eeCoa4s^^&S<&6P6V4;=0u(
z`HdBaYy|5hmFnyo(J{}Bm*^cW?r;6FoWmZnR45Kw_rzi*WjW~MyO5Oro7Wc71&prY
zSnL{~AJ9dp<F#;$^W3nswA{duBQaC7AKI-q2Ji>1g5XAU)iKTy>;FfwC$Ppm2}6Ky
z8H>0lQR^d{H<5dKLqv235d^z&Bncaa2-2jHjt;*I8?UtdS3CeMB{Z#~=J0nB28WI_
z;E>f!mnR~VaZ3hbSe%&J-DkBVEs=z;3xqDYo5%j#I6UQ{AepC%3XwMWeV!wzf}@;V
zD-Z%Ip2Nz#6|R|YcTxF!UGJN*+IR~_=f*rNYk)ZOG(-BOlXrV247pic$cC_UtlP~D
zGn>N*ZYq;!P=J%jZ?Y7#pt^}3Ddb(l1-vlvOz3Iq6tK3@R+;3=*E>2ySv1ncqOn+t
zXa<27YGKmj?6h?1O#bKP`yVWyq^b87#!70wIv+IQT^zvjycVWJ#JqQ+d105)<3+5a
zhEwQQ*Vgyj_xLAn-eJPY!F>awLv4XW&$9AQPUZ9@w?X(_trWPVUEPKf1V@3pfk_%@
zW|#zUCb3^n(>ID<kHHhyHB}G3q_J<01}K&y%m?H!_K*{IbFn`whUVYuI(0)Q?En~Y
zw$w{oQPO*?`Zif#u9ozxT!#-h*^ftf7Ee7OE~7t(usqT%%5I!hKqZ=%?w&NqFa=H0
zc~|U`=2{Wls6D7mhdLe)!C9f3yYr^4cK$V!meC&3-Vh8@Y9+PXScTP+c+ql?G2b$x
zAys|G3A8L+WJ8ldx<cwYWO`(bpZ+})8H@puBCYdOEf^}sb3thJD<qoB)n9d<eHY)_
zjQCZj@6^JAhEtsa@=+-QHVZQlC8%Iv4X}pT42+j<p=NWEIH~?r%0q@FT2_C`fV@$Q
zHWyl&r0!*Upuq3^Gy_b4D?&HmPuRkX3evU|k(<}@_ls}#Ep8GD{umlYSD1d0((KQl
zqAs*=#vY3W;{pKC(fu4%0t(q8&573;6o>wF*Egay0JvWDhXj<@=!7)9beo9$pw{rd
z1hk{&qSEqC>;4~}rV=RagATs2`yd6PmoahM@_H7_a8OVyqEvGl`jqWy4j)^=ojXZ;
zkKY8=E%nqQI1<4`eE|4Ro6#<kJGZ-eQe9c&EM?jW=|Hhx+WE-%Zl=sjI8FXVvR3MP
z$SpOJqM*VPX1Qn4KPav=##J|hvvV}qNA$Fih`fP0CrAV<E}&a-d(x(Ku4gP`-r`Xd
zOjY`8&k8!)S{}g-X;;~oIW#f50VgmL=y_T2(3AALD{sq%s$8uzW%=>kNew3_po_ak
zGeI{oTtf<J_a?p%9SF$O#Tc<IxFM?59z2?K_(Iprg6jfyfz^$|G?3~d_u#yQuSY@%
zk@$wzi_1N_)3E9FUvkU0WN1L67bJKgg{~AczYRK?XlLTLBu=1Qo%gqG0!gKX8=TX`
z#zOQ`C7%qPR|BG?-?%7=9AaCCl3#@d)%Eq9AWqUVpHk(qKFEcEf{PvpR?fT0C5!cJ
zcTE(&pmxy2wY4*z`gKaJ3i2MA$thCIq;OaQLK%#$WryMp(Lk)YW!A!+h^U8zG)?3$
zw+!oXUvx2|miQz%x29?}x){jKw>aBO*FpuWh<^{>Y;#pr@t^yQ?Hf=Yd4V(6x&tUt
z#qnbm9bujF-(|$2ZQY|h>$b*6@F@eLl)>Y_I1Jj`tC?OnH)*kluDImJtE5Na*~(?k
zQ;Dxed*E#?$G%>*8_3)knNP!(a_~FL<)!HL^i&Kyrtu36O4#MN>+TBI<hu@9kkv)0
z-WL~ljf=^BFC|*Zu!ljvPC#a4A0DDLkU%1vjLTK4%wYa>zxz6S-JcV!40sX?MaG^-
zAm;Ln#HOgTSL@of-QM_cZXaQFFh1G3>E$fgYc4hM!s&0bK)945LXHac5(6gY)=^M%
z2M30CeqC^i1l8&;?QITl&qd2zLHIX|?TrFAJX_Sz0ceewu*)y{52xQ$8l$8r7Tbv&
z*hq*zPPYwqLm|y5ik^wYcrZ{F{s~;mK6u<1-w@9n*34P9KdyPa(3y{VcNReO_ux!+
zmk$Y5s_eoNhKpC_e3x$NS~TyS?ph2!@hXVrYvuh>*U<TlLGJiMIe4}pS$$o4;E)M&
zQVPNuSCHFXtNl)+3W$RG$VE`+OXyx3HmcM>7or`13%jr`JRfw7=$yk>eNNL`8<0)6
z3MbjWaJ(jGHLLwZBv3N*PIx)`;NlvC(xq3~<i2I^nybtN#D?~ReyU`km}D)Zm*BB&
zwC@n?HJ#7NVk{@IFhjsHiwr;+<_T+=Vh3@kW&c|rj-tR)ql+0J=O2h6r4)#HXCvP)
z8hXB?Mk1(ot~+tRKxy{oc|PbJT_ghJT@y+>5@fDfyio70{v?*()ghN&C1y2xuT1Rw
zVZv$3BIZR@7bddt$D7(}RH3@u3ENoTbAy%#vo1CQ>YI~U#>q5w0L{=-^K$t2F7&#s
zUP$i5L#%!8lA-Z>t@<hc`@sDq6{-DWnw(5MJ7A4`&!B!@5Y{t&WCkL?Nd5*~qb+#Z
z)-iMmeORa1Z|hIm3BH127zSP#?2G^~{OC8BE{ajA<_X8^7fU<Xm~@(5<?@CFH(`Kj
zC2|hwvQ(;PrS_5KSleuZ+IzoEs$mBLr7V^un_DvJj=hd9Mmm$JQq(*+-*R#bSKK^5
zo2xXS#8t||SfpzQbcTtY4L<6Y+K7Z`Tb#9$-C4ESQDl&u&pK%pMTe`-|6JDM?%5zG
zaQ=_Pie*Lq6Nu?8{@JJ*1$`QlK9mohPUMX%NAPk0_*lvN(Eg+_<rn3j?z60kCpdZQ
z&=VmcItPyyQ%o17!_x;inP~kEG>M$lJIdFsj?_TH)6p=;6`)JskL2AGn@%NP4*s+O
z;;&QRLS&nz*uT`=w$_qSgN^E3Tf93o5Dfwh^+Z@J$iAGH15^MtlD2=4phY;G@?!UN
zmgdq?7SHc(RH{*_X7vo4^3}Zi5tsjw)h0TI$tfw@y!1h5`eOIhZw%gwFiNjFbhbc)
zaqiY2i(n=^0-EbdL%|q9>5w`SPmOvleX%5oy+pL41<`NZM<1PtWJO8_37@DpwJhAM
ziXUCXEG?WjH+G72(kNf7mw+A}v^}B^Ox}IJ`+go@^9egN`Vo0lB*hhkIKB9Vybv$o
zf-1C8Dw{YRJc)}bL|Z|fH0{^D=$xI+K>pak+`${HYJaO!pUK3|go)b<SHmbma~csZ
z6X&PGr#^0pN9?D;KhJ1rwuk!mv0(G}j&u?sB}={6yYYltMYk44QF(xZ!=s;z6bkHO
zZ=s3QmmOF0Q_5$6i0ab<(D(f#@Dj`#&AD2|4_=j)7JiIL;q-Z1@h*xz!2gV=xz<#5
zN7p4Y{-U7MN*_!x`3O$ktQCiW>%)Yp8Xb#14JKvgRDN~Kos_ckn}*8p;<4)-IEQ7s
zq`Sz^j$E~t4xDcfu%YDy&Q#96KaDc_ZixAVIju>Sn%rk%M&fm`=eG2CGb&&w;HnS}
zCpnLan<$jNaL$6o4;4xoP!`a+=eo}#nYqG|^5}DyA0#zoBNmQrPD!R1XU#d7EWI!_
zz@G$vGf@0v)W{du{f0%|kGI7KkeysM*QddERq_pz%?mD~tJ^s-vAr|ecP(t<GK97C
zHddfc`b;tti{|}AGhr?LVZi^6c#d4`NIh9*XMd;$K6KwHbF2D`wwlB_;OO<M8P*R+
z7ktikmqwl8CjDb-(fo_^k3}-+gS{m<EFRBF5%<?xzHec|z@Kiblt%+1EtoLQF?c98
zzUJihfm^P3kut@|#0Q{Uu=fc_HN;CJpEN_gf)g;m{ZkneRDI)1DhQCw-ltds>Z^0%
zpjWLU*2`)JwpG2-a+3e~p0!xB>Op|&exJG}DPKio<<}@g0sHY;*fiC(p(LbgvZaTT
z$9){@Wz=G9@~%ksrZUOqWykPM-&%lhhYDE)C5{`}#n-`(DMTk?y2w>syYBW#i^Xs!
z7%z_L9M|)6{Sum%K2~-CVTD@gy_(tz(Y4)O3FY7Qh0I<+lV5F+u*L52E}V67Y<>zb
z^TdXFbU9@y!Bv=QNsl_ly*NEO=Zb3R<c==b2!w2Dgu#+e>skkj5PT(nV$;FkATLNw
z4vnp0vxDZ7L|63ItqT=JA3pxvW|eudKwR_l>NEHHMF5l1YYUxDkf3pU3h>htTP+zx
z27|xsDR(-Q1Tc3&;MFmBraF-qwZD+H#G{%I8+pCeIT}`IQ^q24)0Qr|)PGt=FNapv
z>Si}0P#yV3U4=`3!bzC{(mrH+<7E$h>P-1;%$}zdVp#YW&VVwxZ4jr?c{d~~<3o5P
zjrk8m0rv!GpP|qfPyF$^dBQd~DRkJcNnemG8qlR@$Ck@;R*W3Dy{#?7(AYkdU;Jf1
zJiUD=vxFVPV8|hVuEt3*Pf-BAy1-Bry0a5{xDvUax%<^<=jn0<0bul})O8dNpSns8
z|F0JSRBk$duKDuMwhJYTtM5(NM3T|i$L|U39lPd5+iQ<bvwDAe3V^LPNwLCoPtuzJ
zLVC6&oSay!mGvJ*iG%Sw5&#zhar6l(`<>;!SAs&Og()vA8N%~xiPcHkxdVykW&%il
z#LrCKhEqt@p0iv<89BBmRHyKNO?;p~hvCm?gtOY~ME{~@ums$bePdj=bORKZlHa{U
zEk-n|SG*#zJVe(|<P30vxX0WE{op_rSTN)6$l;+$I2u{QGqMjQPIM;-Ur0mnZPFac
z^^)Gt-VLMP7&oWGBd>P{(tj0M?6V4EM^S7;zYvN;8-7c0ZD-*6!ZYT~pWo;o#Hat{
zNacLyYx#VULrt`2Sll95;q3k&f1MjGvWY7AX5!k_Q+5{GfYcW9h4vINp@GE+!2$YK
z3R7m(G_e;@7zP{f9s<?`)1KB6E+8Pmi~L*Z-j)Ow!sE7tig`k)5YsO~HuIefsxqY+
zz~h&@beR1~uk-b2Ur-SMFJ}M9kAJ1Ca|`AXSX`vkPLZ=)A^E#K6X4I&DgA(QbjkkN
z^<RTbz5O33B!yD=FfPutBrW91F*HQAw?nK6!ww7F;1Ot4v;dW`5OkNieXN?EXCril
zV<La=2Jq~|V$%`(=WiL0n+PAP+v*$<F5WY}xxp}tTwP>-G7Xd8V0pxJu=Id$*9zAA
zCa>M{VM`CX-DUU8Rp2I@>4vYadi%=w6A!TQkRza-)0+NBI}zT`YgZ}Fi+WPWO<Ma6
zbvCz`eM-+38-X;32}(HKbnvf&o|Som)t)`KZdtef>OurSJ;PpF&W+`X{8lX3X1i*2
zuZ?d)iJU%af;p03LC_7O9D6n_PBLm0WdpC19<*69FY0H<D&Ck)6<)x4p_r66oPmql
zmjEHb(Y)Ndj9*Xm1@BlEkry!OQ`2Ve<2jWhNBKJ@rd;HZ4y3rzO~vPXHp~dTEJoHa
zN%?m&Te6JBBIxP_os}=mO6X7^-b+`zBn!Unm3e;O!0e~6<;n&1D-#~9@guUA^<5ao
zM`J|Ig#}|H+_e{Yo5Y?rq_c1#Udz0c@HvEpt_o<5d#&VGrHB1#Jc~zAOm|T>=Peq}
z?PIK;HT6e+dqMgL^HHtCOZ{7F{L%4g0_U6*A3Rre?||RRc6bxK&$bdwB|W5&0K8c!
zNVC<GxfqAAqcznH#8!svF<GSduA>1*?mo8KdA*idVf=T-0-O={t}(8|iDrJw0(ei(
za6I)}N~;hfyG_y$;MEdh9jT$CUa`?$OQ(FSvwk(FNqOpN{i=a!E+l$A#b9ZQRIWLK
zVeqt_WOwZ?3h@})N$d~@gjnjAJWZoi#nHiPDimLS03a-RBh>6^O*A#8l?}gD$Ql(3
z#9K_2d#)lLRZc~~S0!pGKU{SykSc$cBDsz1;)Z?v&NYQx!54aW=S!YuID84|yVgXz
zlu&kXw<k{2$u6Bma1@?3-`HU;un5?t#p)KV$cmY`@Z^TwcNo)Oi;7{WC?rch!HQQj
zj7HQ>l0*nrlAb<X{ys@#Dx?l{Ai4^5Z?`2Klcnkr0cbFL2o)#&A~nj-a9@x@`-z=-
z%f(TqOPIF6@>kxEJ6Q~VHm*zPqLW<J{+|96-QC+t?OiG_+;dZHnNhli^cvq)>mF%Q
z)%cx{@@7jpCg-D56MMC?Wx)<smg6+Fh?!u_*U>+3ssAP^yXRd~X2S4tn8{|gg;So~
zG&}Sgw-2(*^HFm0auZS`kYL1Bo5m-|I-VNeNqq(`(i)vITf2yMkK-_^t$M&)jZ3$D
z*%E>Y!kQE9=@0Vpr+1ZH3ksf<Yk{2mo9Hc(v#2tkw|wC1vZ6FLzKpCujYk|g)k5#s
z^1xlPJH2;jNF=^&nQwdE0iZ?rJLybM({#-6sPx8KbEwqhGiIN*=y&|WF2tG4Y&svO
z8sO4#cIY1%Zqi-S+}Ex@n4+mF^f7^TWJ7;KV=My&lJlW@K~c2<yq90Unq|zG%!F+G
z;>y`mCQ(g~XIQ3RMm~FA`7_NjhQx-OGUfsPp+$(-bR~pHeLH8u(nYx<{Dz2NI|#Hv
zst<i@U`I5=8{aL2rJPuXzcM0Wi_75Xq*MMLplPQ+_!ckJG;nFuDwf(FeZrEV&f=q)
z^u|qXzSzjEi^|Pc7ny097!M6Hh9(rL03=V-jNxq1AhGTmV1(Ll585?Q;B!kSr_Aru
zXEMOetIlEZGZ)di`TKqZIB_LWl3Z#o96W66Z`?PI4*+<-Dy>0&TO=Vz5bdBpDg5p_
zRK#8Qtj5p&x|R3cj_|Tj1s})Dqr2PS-9Y0Tav!;Bs9`*Jii=}$k^B~*O_AlOS0X=y
zAlq7<Rb1Wml4!@Xqq*R^`7QZtLW|`}?z%O#UmfcSwm=%8A<)6_yf}qVO)NLDN2JF~
zF!-n=wP0XmfOnSmL1gp{Wi!&7_XaD7QUQFyV5!N|is4QfdR_x5M@A9V@I-z`V^c_8
zlfbvivHy>~w~mVPiyB5}XhA}dl9EOm={Be#1(8+|q+2=%L`o@15oDAQDe00Lx<N`n
zVn%8}YLJ+rX70o9eSh!$){XV9b?+bF_s6#uYxyA0bIv~d?7h#fv$yB*8(W!znsxDr
z>R4l&PgL@k{AmK+LIj=!<4q&xW{{2iIJfj5eY;VxD%hS`EvV+_T4sVT!ES5NGm2fI
zc-cp8pl?*ZFpkm7bJ~_JQ6v<u`s%}@QJy^>uK};*dl^cWk?u&uJxr`UoFUB{a4oYz
zg<n2`y(#^ZU&nc^)icuaS;Gy8%l;dE{T!0eNgWl(Co!{3QEPO?M6m=FPZ|h@6(&);
zFWlN9Fz=R+0+_3Ad+%)`7hlw<OQv2c<#WHJi`U_9AfPsO5?0POHMD<j<WfV^N3x)P
zU)+ZNjB-cgNmuLyYSD;zv>0Hap$J$4typY9Olvm`rK?dD_S<<xwOy5zL~6nl1wP}r
z=Z_OFtpQ>>GhRK<0Uz(%Vr#wGrZ-K~j|PI%PdgfgpzjPRnkeB<1qP&ui3$1$`t#dg
zJ)AHd=I-bMukA1BYZu^YQkQdK$%ps>vYT$-XS+Fm<l=9x8OTK(e;oK4;9{J|+{0pD
zp9b3q1A;Q2cXAIWs8yyb#?OeP%&o~sG$}?JTlPi{!=hXb)S2<cpNW0$axqQ5xwUYi
zkM|drUO86yO}QsG$q4Utkaclr)yXPmzf@k0s<^SF)_8lVZU%VkjVWlwVWkM?8UfqL
zyWCezyjA^r2dqNuVVB9zBpo-4*`u9|Qcn4xreYn~&4!?0-W8f4YAE<*MsYT_e=q4v
zFT4!t?~Bg0lB2u)`QRB@60=6hCf`ZN2lVA5zRO-G(bQ1YMli<Nhxq939WP@G9>jw)
zs~4o&7EZcQ^6^jVhGx8SS?Q~)biYG4nysR0A``}6@F>d39{|Zrj1%YIXY==hCu>{$
zK>-`f{hwF^Nkf@=3tyKLsJKALGdNQ^N;U5{d41-{AauIk0RH;@1+BKqfK2Rug0xmm
zRjr!|VAVAP*80N(a&1^5Ymwg9E)T*#0jb55$7zu|kpZ7V?3J~@B-k%w3QA1rE~z->
z^Jd^iUO~LqiCjD52k_o^vx{<31%mKawq}ihZdoqa;*0df)SbscOuX3_pWm0V-hVLs
z$i`DNL)g~`(6K=p@#s}UnK@?xZWTI_!n?m22@ME^$m5Uv5)|WjoPKg8)`Sbt21*a9
zXzPEw^{BV21g?P=%Xj+C0{`iBkLyQYTl~@WW}_+=5genvcz$KTac>)giR_I4g*s5j
zX!ofqSUMr`FRRJk>@Hb*6m^i^qe{T>u^rt`y8rh>&$owTpUwt;(#-#QWQ+rSz%~$f
z)E+A`pVZ>leE3y+<xx8?<|JY_B?}Y3<BscQLUKI|+A9*z8vOt8)fpl-yd7t!8XCn&
zG5YJ($Bu{7e^{E%D^`QtL(soSBV?cU|9u_<+)ueUfJ+Y;<_Vv>YxpE{yc(|W(DT+P
ziXMQ`2H3Kbuxq|uj}%B@f66MKZtTs_JdQ{nactc|-%nb!+oGR;^_;ClZN#{=?MB9a
zBN2a#jWA<s-eUEK>&gVB5zU396&w;M==wnI1AF2K)AfDd^l8cH3)){_J|SKohXPTg
z{J>M{v4>+nQovdwzYCGM(gWaAlA#UQ>q67nmbo(o=D`~hR9fDqTpZRSG=%Q*83{wk
zDdg<)Y1cg?SDaVHPobveA4mjl!1M<{ljB`;&NUiz%9PTuI4FQ@@UWFwiLzbxgJ*D(
zG11KqF~>)zKthCOKo0>(<K4lrR>66jQ|d?GZyXxy&6e-(n~}GYh-Nuiz4Ou&m_z&W
zJn~lA3NG{;xD$SJ@;cSfo6EaAkUH?9uR;K>F(1PTBi3dq*~EWA{!-MLArl`^YKlKn
z$G0jS8&~q@m9()On}|a)J>Ep|KBMO|Ef~oVc<!Kew+klOJvZpSsIBP7!XV+eznZIh
zrmoL{mnk6KwSUyXcP1Tdb$9gz8Qg+zY_5w^<gjTIK6+t$(2pJag*-5!MvT|r?GsP1
z=$8FNUi3D_%FH+NkD>e-H~jL{q17+w#jW8GNY;baSx5|-O=B$VQv%#4ew>3myCpX1
z^1(d8#iH|l4wGwE@u>o-!XWl<?QW_FYOMG{^D>pqGy6Wfo7yRzjexIey-ERmImZ8P
z!D@j(CFZ$t!tHCw2Rx8j=EtupGG1R|$cHM4RBvL5N_};g2^S`Ui`h=a7+4D{w;vvQ
zJ1wciEEZ-T{??mTtOCiK6`+t70%mvM0_l0TK%07{%)4g3Q?B>*e?HPYd(^98T8sS*
z*L>uB9MKac;?#L5IAm-6Y|Q9<rzsH*jlhdelC<AcpG~_?*|7SkZ$Z?in^>t~4~m5n
zI#fwne3^cef9(2brjudK%Snx5yz|{GOT=}*rrm3Kwe745MfXV%Zam2;zBpCl*7W`|
zHuc~WJjX+}`vvDKNOx)k>Bc7;bG)r}L6ZN*-=XE^)|$p_m9o6wJI@d0-FxmjB>n38
z8X3a02vsFKYk16<gO!;o#xV39;jC5sHl!D8y?GX>xEEwu0_)~hrAmRCI7{rRtbx>m
zvw8|`n1Y~zr||rjq{VH5T%NbvO7`e(pj5pIPtP7dqc@&SQ>lDYLlwv_U)LSuTAZ@H
z{fS0~a~>AeiTJ@7sJgXZK^neq-1#WcDmom)x%;&_S30rcdG47@8(AxpRDNDe-Pc&6
z1(1<>`KLeWg3em*>CAv+RKG?G8oU2Q>*J83yRFq~ZL8hfU42&n_s{%H(JjZoB|47?
z|36KzGJ{tkUkZ+Ar6xb(6(=xf5=i^x?Wk8o{Wte3-%KQ=3Nd5|I!^$#3nr!rKw&8-
zDAX4A%K}v>A(obp?j3hH6*^1HQPnwA20WfWc|t#vO}TeMfFOk`4mGrpt1qM_jRe;<
zu~X@W<Q`I6_q3iBnEQE==rF>@-gcbdR(+Pz8-!qWYe`A#$S2Y}7v=ZyKcN%F|J_vO
zUB=tUzN}$Vj)0pVzcKY6RD74Ym11O7Mm@UADB*doKC-a<V1E{d8u={d^B2uK!rGh(
z=?;)AhdyHVT6^Xj*SD<5G_OboieL!U7zc9xlYyv)Hqw!H+{sl_|1aaasQ|}8&HVB;
z?}7WxY{ZvhoKLv}m`k-s;-c!=+oMX$A7n>7bn>XbCViHedE!ZE5;)r8@cG>POHWMF
zT&z@-wPvdGM8eh~BmgJNJqWAg>z#>&#qnh1cU)$e9U?r`e_$*7WwDZzKnU76_nsKu
zp~iHzf;bNc=3xFFC%-?}`{Oay)b_xwwxpBq`+o-u{x~5TV(xS^O19wd>Mwe+8?ef0
zAvTT^5RLKI&XvBYT11K6zr4;EwD{;|;Cr3lmyc__L4$M^YKO$bn=%Q4U-+fL9%|;+
z{kQf0a*&+2q|`QmMaGiUQ^m6igx_VSLuns9)Qbq^zUwfXvM20$|E_8P0Y%x=0gTcb
zkFO0t7^dc~f5vx62RS<Ur1}@lxX_zKW~AJ^{phw9*V`B3gIzLn-<eiHvx^k7{W3EB
zs#TPINRGJ;D_t6!>$EocZm)@GQ)zAwP70#xgxr&A>HHo3;0}z-LV{ZyTS6WL7(i7~
z)7e|4e@YwAKb92#NmN^g8Lagi<%=JBor-H8`8H(>IL*nmbhGG*vdA72<r9w#rng_i
zdegQ^mCFB>jKC+yS5S6P4ysOP8jTV=8<SOV#!UR-#HqfdyG&#dX6rdFNOpSqD^ui}
zH{1M2k&uWswcpe2ZXwW<_T?zw6I_CiDegURQ;68HG9<#Uf43NVUwhJDDp~bHXY9;E
z(T)wy4OwL37|YEg`WnrIN43@^3Rpo|B!sPhU+&&E90=$28a*19H0F|)@ZWb8ob!`J
zM{BdcuVoJzj_Zhj6H7F*-1z$9xIU|?ot<Yahs9GG^KqB3AzcyYpNZQr4P1y0F`M(F
z#q6}laV)jPEoJdeoPPor9Y$d2=o#e+;noi(zqZjEL;LW}o!H8llw;7vC2fujp{bwb
zqT++u=AINK&_|tl@1ARmP#4wn<Xc}tmO){IBr+bZb{Fv5hbPGsD!;=W(vv0(71$MH
zw>>W&s`$++k@UxRlyy^@hNd)`4(Y+EP95tdR_0x051_ST3juLhj``<63ax~X0<O`B
zIs!x6iKmBC^Uv>_!cF7hP)VSp$+)AEpWYqiY$rW$tN6r)NTG=RkrV9qDQ;$N7VDdn
z89gH>EOA{U_G;1hO7JkMzjnuMV5ZkcojYc-H<WKXxH-Xg?nBuRJ|SQtIyLrpCT>+*
zjp7B+)-wSJ15+Bjb#yd-eh%+qKR=hul8wD5xfBz(P3esf)@uM2hN-=sAKMjfDTu0$
zDM^?rzl$4pc2+%jbjg*UU;V>`>q{76s-~laVZU+JxCSyLqakycy{hp^PRdT0;`=WI
zXS08IeR(ZS8YPzfrNul#6s;}9)yC!3qLTYrTpujeaRib|EI#<WUE&h=5TZAPRTY`^
zUpbYKEnf1nd%De{_e&?8aV_RTPsXdCkY&CKcHtEs68|~K!$2#UbBGeO`5fJ(LX~?L
z7DmxnTW+T~tU)(<`F6e@18U>7JGJ$vY=6a}^jEvRqdsb4hPC|%-f#uCdB?_dWl_LL
z$Ek=NgrO6GKeOx&!I5L#^N;_yxu6+9TS~})_~$R*P;~akXV$|a*V^ax=SQ^ZY^|33
zdPdi>)uQ3Hf_Up+2D%A8iV}v)PaiuczOUa@jUoHnT+{?Aq{ZB@!<fi3c>E#^Ws)ni
zvG=fAf_~#<Gvdyk5&>@A9zgYh!D!nMoH2DmQu;l=_QCjTgXKxvxf;Ez3j$%DUO>le
zFH4_>4?2inotuzDrk%z9i<Nyp)dRf082XV+r{%{&ig*arT21$6&f!gTxE6b*PR+D%
z0^TR;0QyRbGZ@7Vks`O$RgyiS@!?^;Z4roh2hFqU`JC4CWxoBq7DhNf?)_IeIHczC
zr7|>tFeidPq9XB0WXeYHO))+dPsRzjQwr8<;c=>AKC8_giFE64*VjN*cz=BsoHA~!
zNmG8~h%-28sk`g9p#r(;DjIQstGv9%VW}yjW)MDrJZe6i$NV~8Y;n1xn5jBa&7VA#
z$y2zcL=uat_4K;YLK-43&iTb(Tp^qbvn!Va^K45gYX}gx7*y`nx238J$vI?K`{4bb
z9FYNs8lUFH${&5Mi7s-7_rg7IfNSPJ=zX={o>TR8dIk0m8O1-Wwe&D*8QJJGERt<y
z#j`}F#l=S&j6X9lVg6yDJAJ5&+*PBF)Eq#3@*v6+40xPgbU>Cr+k4P6x@Fw=cR^kE
z=i_d@dmlg@nz~YEPTc7FJB5(cFKGwRF^B;XE=52*`xETaeY+X+4zY1)dZ_{nPl+2Q
z?1{QSyVbqaK}OzG(OcvY9?n!(OMcD4V}P-6(Uwh|^xjd~$U0+BCVr-W(_%O3C_JBZ
z&0Vc?JZYYp>aH?LAe}^l-S4%*mPjGti!2L`xBd9Kw^-rk-*U+tKR83)epCx*x)n#T
z_f$=+W=ZY3-T>o~>1x6bDqO3BbWr(8J8b9F=gyBy&CnWr^WxEmVzcT%MWMVi`tfC7
zHk3R2i{+)19qrZTa5jm9!|yvFQ$DYN&Vw-`Q2}6%&+0zi>Fh)#*5k5_PY0g7)kosO
z%I2j4ZyH>uoLcgya}--1kjz{h9n>+cE%A{CN)#4^?da>b?LYhJIheF?8IJq5dKtRY
z-wWHMYYiBN9*$6wZtfaHp6)27(4TQ^t|6zY{Z%i|sU}Ycz-KbwY^S0)*0^(Is;Uc+
zlKPn()P#++ha{ueM{%jQ9t$0Vurh{^f`p6sokTKeRzFeS5E*fJ990j}U5JCMg})6&
z3pd=r-sHiIo0Ph|qyHqv=i&CUf+Lf*d&8H0M?tI3WlqPp8Sw!)bO?JRC8JM7%6FR6
z$Jqn;i}O0(Mzh3hD*LA=rA9U-9`JC61>afxIX34d`*7imIq<KRjJsl@VxB@4j54?1
zFU&R2jX75Twpqn4Tv?5xjo+p0(jX*KC!YLsSI207B=v*8JvXNWq+iJ|5q!Ejr4o*G
z^Z#D>YA<$+zStii#crNp(8<&%wH?!h9@=Z$?fLee4Le|e{%V8np<q0lp0S7UZHuOH
zeq1}oQRkcb#%&2_bm-p-77U>*pLDYKM^P^Fmt!_7Xz2LwADQ_y(E=^;=U<EWRpYcH
zm8VSaq?{0nfycP}jzn8Y7w$_1VprTSWQLCcu}mpKgP#($-OT5%xm<`x{aR1=sFKb0
zUpX@$lRK4fKA}6PH8!ZEvc=Aet~A%9nH70gCJEsPpeXQdcf#cdM?ENCj~1_p_G11S
zcRM$w$O)=GDwQ-#_H9R-(1eo)?_a+(wwx?`_CC)ZR47^x7~l+~NVGyC@eo{yN~(aC
zYES%P%a^o?2^EqU)9RoAeo|`7I2GVdL_py2cx_flghAFH=B%#y5HoCPteNb-KpxX~
z7~NTW@Al86A2aQ_?jcl@K??vKxwdzg?Z19ewI)kv&#FmBuhQ0oxF>t#T!!zRXDXGa
zem~yi4;Y;V7g2L~rd_8YNbg0<_4Pv8i)teHt9p@pPoc9Xw@0q_JoSZ;#BbMb-*ugW
z<F5D~Jx!01u0=U)#?3Dy+|Wbc0aj9HC0!Js*e=m8lbUG!!1yU<4IJEf5l4|_J^I!J
zF<K%dTOB+|_B4ZGxe`1#bi=Cm#zp#}W(d<Q^MPeT&06Q#1Es3%7ki6-cLVkn^GK<V
z#Er2_$>P-@8`+5^PyQ-pmaH)H$=V3jjUUy-7y^!Jw6#%`1WvZzmlJ&#$(4_?ykiGX
zv|efS2$4o8_@zdzZwzK@-ptEVvAc=e;Zya;>DR=7ibBZrJ6=7D4_ib1HP^s_p{0p|
zNhIx$7A(`2;L<6$^no<W9_3lGA|a5%0Oqw^v_RZKnuuj>9zilq8mEf-Oa{T$U0bt0
ziW=XWjKS{QeR$l<uvp$ysbLFFTt{x`@L9<%ZowEG%F{Gj#1a7)`}z*A#WafiOGun{
zR(FQvv0)f@UB-)OJc#f}ujyy<)G~>0;>2=&BE62roaE=K(3GH$u=Pnwu{HkFMqOb>
zfe>qP-SIg&(w3O5y;}us?rhY03jE_n{?|wByyzrQBmT%S$jbOC#k}N8lEx$gVaJ7U
zD4%YP{H}|i+Jf7)%ZbSjZ+SANwdp~Pw8ZmSH)IJ`ds2jwX~06WAO1*j(In8(#XdES
zt@~;I{R5NG0dYF@YdiQ%j~V_+5{oL(;Sw8icpdgd(Q@d|IvneY9m||We-r_oZ+YT5
z#jL$WP?g`aj)h7~Xymj^`|p2OIH~@Gxt3(%ZzltLs%U2ikCgb;NIBJJAJMMYVqWK*
z+win!w69?YC2C{so+0J@pe{}>`Rk=p;=@=r1+oi5o`5INP{;?vKKz$7QQ|Say@z>&
zJwKy*+4*7?nonm!U3m@RN;6#t?D^8`H`KTLC%=*f*4WIa8vm@3GX09sxh_%l$!SZ-
z>`e&o*z1*Xlc*QD?;mceXeq`$_}WM-BB1kxXw5gZl1476sga3`Ts|hEEoDQHFpo!&
z<m5)8SWNuBtMtU%L*tcpPtKwcEe|dW9G-=Tem$*FzJPuO|3MJL!O78)$93y1C7pLU
zW7C_sj<UFY{@I-Js;c(H&|#eZO3>k%TiD(itW4M-ZeNBGvq85`mgBm4S+A<&$Pd?N
zP;Q88+g@OTp{;(6&6ynpo`OiXlmdku7qe2nyeoep5W}@{u~Vz09li-ma@zuLxI4^b
z_U>5x7&AM+|BHIW4h@MA)%gCMdGF7fjiYQgDQpnx87!|Fx^mqk4AgW+*Xp_bTg8|l
zXYy-`wT^KD!a9Un86sGj5d#`BC$f5wKzYW>ry`sHmIY@KT60;1&2$S>gH!xzs0~mz
z(wf;dPdgIRo4n7A3-tfybWy|?F>tF@!G@a7=<~kvH-9+HT7owV?fouXPxhH!SrXpk
zD6*SuWiJSGi_80be}j_phiqWVSa-b!E4RnaOOaavbo2y@@i}i73H`LgxxT8bbcV{G
z#vZ*y45}b&Q$ibGC5uR>$uALC5i{dij6To=`LG&BgepvwUU<>{@+4@d3T>=<shF4<
zUra+xNl^=HHo>QvEY1q>s=h#~<lAg*bZJLQSro%`>U2HRL-t+MoCYKaC!gh;!smMs
zM|bnlj-?NoOt@t^502HuX3V#mKg)MxRU1csV`1&@|0**Wtps^WO6YFe8B2pbf9c)S
z<}j#hwKjI>PfW3T*M=g>pIrCyh+4DED6)I-;pdH@CeY@#0~V{kqY}&+!XWy>080IE
z(AJ-A(;@JtO<Z25yD*|U`-!4V?VmwIg>bS#cl)HJ`^F>{ynD>Swf6E7Z|?1duXwC9
ze^30X6(nxS9@}$}UwZS=zqtU%MH@23tIz_a*VB&~y${%nkC^<A#UB!boF7boue=>|
zB>(Pfk@Z6*9)lB13hmO5cCtk&#k8gwzQl1%OFHxipq&b?S!5`y%fvh<SS1!Ydbzs(
zZ6I@f&6DU&N7*7`-0Q?gc%j&f^ITq|-LF&NR5W9YM%i6;C4XJ^@8T0oZFp}%xp`}p
zJw@#1)~!#Km!LLo`kUDC%$rzVG(<~gMp0t&`uizakV_F3K{hF~^2|<Rq6XdMCjTK6
z?9GcQq6XnftRA~pHsoXTEXgf@iw#Z6&9K+V@=B~-cMe@vwK-#9aN{NS^{36CK9BbW
zPHoRlpi9g7&QPWvWD!3C=-V;wW}0<BRig*?7cr_!P(*<m?}%RyV!&=HG_;;%6~8DG
zy8^Y4J2^KyX^&D<?>pJrK27KAsXu4>>$4>?IOY>1BW2}*cpRr4W%yT_Oe20@iF4%#
z_mrhsUc<+}-9u4i|6Bj6^jLZS3-2U-j_13bJBxO38+Vj0-QPBhf4|aQXM#F&q}ZtZ
zmCig#b8lTr`v;7~`2o@TleLz0Ke~#|F(UmhXAEX`@Lk|;yXCEBq$+=8#huX9uWjQ?
zi-~m9{R^oVOwsGhu!R6sLx}iuvm7ba7%xTdD8H&eBjhhiFk&>bczo`@@jcf2i@IzE
zzG2q)m6SNXWvuDWQH3&K4J9x1w2ZF#2;dlXvTjI-bCQ$%q{9$4dxze^hPC!=r^gdt
zD^MS%F&qev2*u#2@>4KcK%W*A`TTiYV1`g&NuTJ7!-ZnXqUCxB%!I3*gX`_VrCR6Z
zq?@@$4-(Mz-qMe~o6_Hn#z(hd$WdH@7hagVi9m&y7zP8d?2G6pLUDcknm01;_0!Ue
zh4lnj4<8PC<oj)lMYxJ!v!vqXZ{pL;Z=EYc;z@kdlEPrSA&lzd<C0>`9c&~X$>gzH
zD*oE60iT>cMguKq-q3?;?Z9XWTcgj~>;$g6`8Zo!CYTXQwRXtPog`dW$<35zlSPM?
z8WS+`qx!LfpH=&d^nv8j9&ARol&MyAg41zzZYX0fuiEcku6z)m>84TH!On#&;VkWv
z3`-F|P@}wb{tcdZ(k=o4WMqYzM5Dg_6qulJlM87}`qR%P$}j8I2_;kohpSkwHabd#
zTQ}$msOfX(v76Tk-%%lvn0Nscef_&4-xG$peX1R&thRl*h`!%U5OcrgS^U&-^DWK0
z`t1*pjMbu}dbF)^O#$+6upT}bskd=`LTvNnQEEj${PQ2I$>guQ(Lc1*noKBtc#*|Z
zq9g<w$}tGP4lmK~Imzltk*Y8C{rcr`F&YQl&s=$q<A--Q?0EP>o^$qg7hxcD`DiGo
zp;mS3#e3Y_T6eY{MBsyPm5As0=yeD(-feyl@w$-VN5VnJ4+!!8w!_9o-PY}ufyeXP
zM#R4}ZgE>9!}-0%UXt=q&S*>Kq>IV^6azVjQb)J3`twv3A(ii-t(~PxwNfv7;({&i
zYPTEROZ`(czuLhAu{Q^cH*b9?cUzZYLwT_m9T8&L^Ff%nonz2x*HTEl$z#mJ8c!Ko
zpl?f|noigKmCoSI7CUXD$1V+6>rIKjc2D*=V>C*9H$S)}(230Y_|vpa@<~3p?pzz|
z)OJZGVYvMeZkIA*yZ?Rb9sMS%?<dSnHI_W3uy1zcz&f?CrW_>TsXB2Ss1Qoj6v1>@
z$B>~ygO3o{wAS!VX6M9LDQngJSti`6eOt!^;gz<#5KCF4+%k=;o>7eZyvmTazm>6l
zBjwy&ZqUZ4%D`i5;N*^#kqzC%``fZqdqASDrt0Y%6gZ#2*0EuS<V40|I2@AdbcPj!
z7zLSt(z9qQroMD**xD|ymXr(OjF8`7&C=^Tq_Oh!Ic4!c>Ycl=*22v&Cv!1Ks-Pt4
z3V^R70AC?}alicq4HxaJC;}&r_#FFqdwROtdYal@bT4~&-YvxEJd@7z&p+O)jQd&p
zL3w@S$-B}CmxB&Extz^-Ib&4a*H+>l#H=s9zw=rBjfWx2+Tm6&N2yjjzC0}J#h=2>
zd(aVM0aHz(I8Um#@yiXE8AVPJhbwKp0<_7+H)u)C3b|>NoWene*SBrT)u<+G`o0{W
z!c#H`v@U_TC<9GCjC5S<$s^d258*Iz?~w)DRd{iPYLUw7zPw5Yk2HJb=;$zjgwl-J
zjSZk*AJ2@7{?i=G5}=Rw`%imBND`z^&s=CroW~zFHQ5}x$VLL__10T+@Ert^8!0=%
z&(Xg@ypZ+F*5zxD^nf-sg(c!RS$=?;oCP<GFAV;Cj4(d}14+n25gaB}coN$OOUVRK
zm8dO2)C(jYq&f3@@$rmi0zOMJ<`5*6AI@h&u-AWZi$#xdLR}tp2brf(RFqK!7mH_j
z5p}Hh#>M6WrmRWZEwq;K2$q@tVUd7OU%NR={%4mn)h5qgu}=eG!1NuFiV-YgF#%tT
z3gr6|lG1m}2=ZKoE~IORd&p(ljDh|lk3hA;^nu0ugO0%^>d=-U%mqCfd%e*9n_>nw
zc2FeXr>wl-Dhc1l)ZumxYX;g<ccegUT2jI&4X_9E0>lqK99JIeB@Hzt$DjJzhNTtI
zh9|}c`KpJdaZB^A;&)Do=qk<kD?DEs_ldoKZ!lg0JU!s2JKVGZ-fb@2MYoJF3q_Ka
zhK_g7pYuSBe1Lwdsnfv#_IT_qN(hiH$WgUKAd3k|%dD5o<btG1O<n!1p8bkRvrS4t
zSRET_iJd{_bo|b>3Hq}ZxcvWZ90jV5;aqPav<&htlEJp7RxwgD4#E~3Q4l#57jM`X
zU+q>|_RxYCGI<0vzx*F22LXj9s3UYDWN)P9xe3gvkSr2eQH5TclUTTx&I6emyJDAZ
zpaEE_+V&F=XomzRVF~&kw->Pr*J`8Xg`_&9U$OE}JAjbtTTK9tcK88Ez+<NWBF16o
zB1i=ABG1*23$qDG>pvb-rAW=epR%heH;<RU7e0u!Q;}9V)H0Gh_%Orz=CA}?|7RQ+
zW3<QB#77)<3nAzA30WV+Q}X+gOsjT%QlVJAs4bdI{0^Gp_7%Fj2$0@f$t01e8Cm&M
zsgmfmN^{bP+N<3>VE>D8QvK><VQfK-2EEhdhDQT?$)o2&XxNlXU>U*v=jtVvH5lF#
zj>2mkGylanLrqb2T9eQH2r5bPEcahK8I2P>(@}CNl@I|xqs}$kf30OioVg-h#;io$
z#I*S};DPZ^f)(i2bTGeKs-Ty!_o#*WKVR|X;aj#p(qv?S#K59(?OXqeh5tmrKN$I+
z@LZ@dL)l8*M9D<NPONf`RemreI8v3Eod}3_k?3c?z)wYbI-IMJh0<(h_gz}vuRk@`
zV-F){x>LI&cF2CX`6y$<kY&ANziWMr7WH)_L^hd6@5{tW$87mz0zUDo%~$q_?*(Qa
z8q9>xdB}SzWqHt%dC6NJC%VFwMj#7<5?hfZ++h5arrqjzr%HYSbr(6+6AjpW!qxan
z#|0Te4#IB(&s51qa$^2tc6au5587}O+ve{*xQ&U=rGo{pClP)g&76v3_3HmU1VyJ(
zEdK#P3W8sh%u~by`k(>VK)Zj+EI!opYpo!TV)0>5uZSbq`=BMdNp8a`oegiN-v7-X
zZ(``K@+3{ueO^Mv{JF(Q!KBNnwIHgfC!qDa%~NSaOAybN*R4CakWs;LECLJg?1w?!
zJ}B(6IQgJMUUya*ZoCQ=IoENKXat1}xt+6ne-<deW0h4A0*$}K^DX%6c76}A-_Cn#
zi-*zPFAX6-Z+zqOx8#?SMQTNeW3;nGvUb@Cv?TXb%{2ibVqras@nDB2<Z8UZKdT**
z#L21vZJ8YIrX3(0@mqEmo>Wub)Z%RavDEjTgd8HZyE;9@g_Sl-q_uuRIy)V#f%FrX
z!l(~!_!HW5>#omFCrg()o*1VY%xp}Q%;Qi`-ZLD2t#iR=4V}_@3DH?P-}C$L*4VEM
z_wB)3oMiFb{!YCM4u4XI<IDH2F^YmO**7S^Z6a$60#^0i=+xMqz3mP58I%yqRYf<K
zdHKxcy4K-Wg>_t(wtN~E_vNfDM>EysNiud{x-7U4-{%Ol2-E$-jXU2_Jxsuleg+z)
zZoR?R?^H5On(PyO{Vhz217S6R>VXjCy85{i#hliNIGz)EgKI97L$1RKL1o);pgElD
z>3AY3J7!*g-dNhb{#ZJQs7mELh`EXcxM3@ZbXNlf>2$Vl7}^U0`<e-|CT=HLd|X%Y
zVSh;})wm+K4_axS@n3lUc!lTeD?HCE6Y82?RN6!9-zNq0-=lOCL6^LmwI+=re-=Qy
z6u&_jVBcY#cNt0{@og?PQg!@f+)w$IfKU6xj-x>$07-rNc(KbNS$;!M+<VpyN8)XK
z3e(iyFg!W3;=N{u#qb(lr^rfoG2yAPt>r&dR$}2=Vts-6@T9q(3D2@*aYbqU7GK1!
zu;cU7_QI3IwPcUo=KIPOwb9aKJanY1e{GQ3q2)`r6k~*3)!k8_1Y2dhD@t3=X1C$O
zvkzHw-0=wMJ;WdF^~d!=zH*E&&hht8ekqmeH{*+JuT2=OhS)cl9{x9*k_v1}0<bA!
z5W2O#m=O1q@5&*{54PBO`t29+-))w}y+X2ARnso!eCqcq<4=us*!3KqH{cm5er9XV
z2;r;+{XH5kcOXfI|4{aaH}gpVWxD_<4h7Vh(HEsx7@vO*U_5P!P&eQojMJb%Ke&(v
z5svBe;f^PVb2})es*hhwSVgMwkLSmwiMKs&yX5FxwypQ&wFCg9_X=RBOr;MafhV<9
zLfJvSUjg!db17z<+h?u&GRdVq;81Jw#VwI59v=NK9`3*5;n*u4UI(yx#lvOMX5ddZ
zv5H85tT{ejhUVKWb=;^QSmHey*AsZD+4)^4uA|<l_CrTJnfT-R4tnDpgPAo$sQ*P}
zru@m^{J87Yl4DEPp?8XtnehA8gFu-QqyJd7Zb2_VXGMS)1O{~1lg!@2?$_dhQ(89G
zVjt|lg`2JEC}i-VGko0o&FNAJ`O|G7|JAW;JjGIRwpskN9#MRHyyROkPM4xeo<8|u
z`68s8?sGtL>8Vr{eOEer*06VxZ-=3){}~dw{OWka(0K+Yw%(eHT^>^$IVd|_9_5C_
zPH7z=Tup-hWjlwvSA-RI#det?5!rE9J6zIR&MAtvgWLBM-i8sJ&@jfo)oDdh*Scc}
zIS6nuoBo*uo<h}l-uZ^@NW|E(q^|Gd?SBQh9=ixJ(CxN8jqO|#9&Rh#>6oFF3~Yr3
zE(_G2?r#04Q@->cK+>G+uJ~VS6}l=3^c>}rWW$G@z}Wln8;^$ZQ@XuZf@az_lP0Ze
zQFHo{ivBcJCQIc_+sRZ8f~cOI4eLT*(}G-|2#y=z^=rs(-EeRzAj2R1Bf}5la$+oN
z>{PWErG#NT%2(&eXu<2I=W`~#|NZ28Dx<zv)c3!}`TajjZ}|T+@ATp_bpj~tSU-*j
z@<AHl?o^B918Wx}_hbEt!Y@C;zCRUjr1ChM{A<LNK<Z}|!Koj*v~5(qIJ`7F&j{ff
zNIsiZsx3fclI;IiE%njWY@@qfQnHjPlFC!!_pjPn0+A;IN)r65X~rtEV8v5FD1rb&
z7y<6x`%Ql0z8ZwV;PN3mpFV+auz!L-@(H{$;yyAMIpPMRi`=JqO3p#7a_{u_z_XOy
z7E<8y4kPk?N%$j3>XFoyR$lQB1Qg%X$^J6FP{n{axL#eh)cC;&pOG9n?w=L_j$jgG
z>dvud`~i?dbV2EzCaytr{v2rs4~aaly~LNR1>;l}{Hf75kdi%m56N3?6;!HtApaN!
zul`Z~KP^(i^^x-G3nmT=%SLtWz^~G}<ug#;QwAdbo8|jplDJ=&51MmQRCz6U){Y1|
zh%_R3Gs1)c>*3n}0O8OQsI{f1!Nn6aCh*Jic?0e(Z&DcXgy5$i<B|ODnJ#?vhmwwy
zj5r>o@;vGN7po$L1<Th`roF!9#@M-u1}o=XE}GdQZt#(u=_8wanj%q|b5QRSpKfDX
z0Z1y}zXpX7MHRt4qv|$46@2RsN*-RHw$tdwFv$fmD*HcQaViTS;3?7irifIf5g{BB
z$(rW%12&%T-<@-2FazH%nYd1fp`|0us7Yt4L~okwyaTAl`CqE}KnxfWr5YuWAcrn%
z5I;@Yb^1R<qNFMhFIxn)Q>O+0wVk9i<wzq6wG&>bA~Y9?E<khAQK)kTbZ|Zk(v$Xn
z2V)?q7Y~}mX8d}dbt=z4gr^R4icG4$`e=th<=*rVG;=dXHq8h*3U`gWhi8PhM3hCO
zMF=@yFH(7cF5(a)p{ou1FFO0*5C0nj|I1-uG@;!MV4sTaf0;XV+LqQv4E-*cpU8|L
zlm<&L4ss)HCu{|8enm62!A&EfLS}-^%@CkSl?rjL*^^R7%=?)+P=aw)xxgXL;;&sP
zO4v~WC&L2)6b73M)hh?UhUh9(9LF&P%;7VqaGN;K66KrwkSL~Z<S$JIabkzq!7?v>
z>wed(OR@61yR|_hF0JqWA{>9$)bxb7jwYAn*5+=VEu`?QGvk5O8!kY8jk;eA4g)$`
zLwpun;@&z1uuWB^KNLe#6D(+#=ae^K<g`iSx|cv!MAsR2RM|R%`I5{ls2^L3=V^ye
z2amn{vhDdWRqEXV0_%B^b>*Ot#j)UdONf_17mgpN(2Q$hYk-IIntY}GN;IzCKIt7u
z*8Ut2b_NHOa00*+g8(IaIeC@xnvM5J6#p68S)5JdKvhqt-fZ&}JT=^wAI?tdpoC0*
zPQdp;0eG*RHNaeA5Xz0&!Q|`)IwwLE&<<E5@;m~s*M1VtDUpfb?vZnzCG^a{=xs=D
z@@iKO`S@ra>kjzJzO2*5gbY~s;6L@8xKcDgt4AbWb^?23#Mc3qi+@OXWE^#@%j#2l
z@&bhm%-GNQXNB2XZ+WDJuJAtF2h2M<h~)rualln$lom{J`gY+qT3MSkYvCXNkG4DB
zDtom<d?DQTMB&aM&S<#oO;Y{)cv-Uj@)aKjcFa6)1N5uSIVEoc2Gmls+tK&Avk6NA
zc%343Jx-rb;OdFCv*k97M}*`so~EbE7Ke6%sNB6ez+WSYi=~!0m0rd|1pqWii>*Sj
zG-=kWiTK#NxUzLQlJt#Ic4UTqgFDD0XRU1?q(Z)EU5E4Q(d%9%-u@=$lTaJ+?bWQE
z`U2gjTBP4O&(<B2z*_X9_}yr5sAK#t^qMfxf~fskx$4NN47gG>a7n-J>-#rXz_1#k
z&%=?)5eUL0@E%CTV3CO`E6}szR3gJCQ*s8_92fs98mdgVB2bbmM44V8O0T?%_@4H(
zu-HJu$DtmGgAvz209_~6Pkmu|OzoO?XL(`SPiOW1@6iv17BX+{`Z(Kj98@V}&(-{1
zQeT*5a8u;73}|8i_5ON$0!#XJmM!NXEh->~F*CGyIqV<^OW=2`xp-gV^D~tJ;^F|%
z2DAb=kfvJ!I>B22r%XvaJpy!-v__1<ZQ)_U4KpX5&J$JPU;ercxW(?m=QmFG{F)Eb
z$-f@Pv7{`%{P{e|GF}E*MvPk4Nw9pM22vevxb*7`_5^WtP>gr5VO1u66?k1n882I`
zdYwwpp~|I2Qm04$YqB6_d?LOcIHio}gj&G@Gaq*z6E8r`Y|k%=7w})578Glb5BcPw
zifA3aXRYJqN~ez}DJ~p4gEZqZ$)cZSyDw;jbMd@qkp&&bja#~GRU$8=Wn*DBE2Ga3
z673HHlZuQ}Ii|wKN($9eKV<*IYn3AZ2>`HicjtG7UrYigSnpJ~p-GU6ybT8agkF^B
zsqS1n?0w(_jw6FF^o_o)ZN-5o729lO&Suwp0>+F$Xe|PiYkq`tNzS+_Tx9jwtF@AK
z!z^R1-xuan3XJLC@DDe;>Hr)d`48V&TUt!Nded?bqF_voet@ijoHP>!ns#oB5#sMg
zIlm%f-6z~5NgDcj`F`_Tpyx!pBlvACLKVDdQFZaj3BgixyiPD>WgOW3Xb3|d4t2;5
zD;-SUQgD=?`J$fcD*%KZ0e=J!y(DfttPfhByAluKVf$k~KVCNE1|>lky_yXG@4uh8
z+51}56vW6_9>Rg9PNx!x?!1`pW?i_o^3adrLdDD>MEX>M4VAkSHuVvp4M5qrA_u@#
z3V7iQNn{>B1>rL{6`qu8gmRvK^SdJsD*q~SO0N@v7mz~<NEBSW7_+lY^x@TMW&kD5
z5qk;#r2iNaG;u=I)Mj&z$eMfR9<F<wyFPtIcL3r4O~3=vl((dr89ywE%W}Lh)u|zk
z@D1#M_-+Y(Q+6lCy$zjJh{yt23duQ8qnBUJsuJN^x{g|}2l{nx<B6?EeKY0)a>)p0
z4OKqD2o}xJ!OT?!U3Iz}PDokuXV9df_MG_>ChzX|=QtI;I^>+e+Ei==CGk1*Gm!GI
z)^9u}Kaa7{u>is*%y=NP_n&j(17K9k^x+=@fH{+ls8K}B%x^Ph&fzn3%X`@dlPq2C
z?9nzAZnkBy<5rb-zvN@|8`95mdmUV8KdqRMrj06Dk28R1PM=;ZQD&N%y4=mPI4t<-
zqkqz-k3q!nND~6I&|w1Ti7wCnlkr^SPe3Vrn3Wxrn2y`F@WI&%?fLyhE8SR8pvbzf
z@Ol`z+>OA1&hL5oVK~abkcsalYnl}(;eJ4Rd{F7b{{%YK^U&YLHIl4$Sm;{<riOG=
zBe>&s--#<xY}rRscOxVwaw-za_kA<bQB_q(m+08UrhGH(@nEvWxOozkB~=PAqboWA
zu>;gP^`yK~VgH{s1ZBFI=6`bm{@>pD-~a#r0Rtrv9>B)rn;{nd7Qn?*(pv;?*eRA`
z0c0|eq7EIE0S>JhL~a7!1SIthL8wA<^iz>Dm4}`18L?Y=zym4N6%Tnc;iWCFq&VOl
zzGx))B;f~c5y$EnaW}ID+O^SrmLCN+II$mFne70ua-U@J3KXD~PpJ%c=&V8wA7kKI
z9&n8RUtR$$*H3TO`e3O%qj{n>hQK(PSD()VtlD<PASa_+SWn`6K651H<^O;VL>oRY
z$3M~md?S-PlDM*E03@mN4PO8x8-E@+F<jnQX2-}n9-|{Dq)cK4UZnIT>?br>9=~Rp
zgo<<oJ%m4nbPDg?DJMuGVb^wh&h6{~ER3{p7p{ODV2??aY+m0~A<>^3ebu=jx7#j@
zuIMd(iJHs7V$PFRJC)%oHjfGHj`bROW)csqKiqdtw-+VcpzC)0n0^_~?PX0sT6~)D
zwCB-Noru*M1dL~hnHCwYmG@!UM|*0RJYxQU3W?YMed*Ic@kYvT%N#Bg4?owpkrkir
zzc-5?4>7^3z?NSUU^qZ~`oCG!h&y#69XF^_x(p&+>6j|JNmHZ(nDH|`K(JpjA}whp
z{>kuM_@n?^Sz}E;^=(eQB-HOQYjo~AOL@B^O{=<d-U}2^L4wpEw{Mra7=21aHI0$f
zGJ48{FLF-lOA#)MBfzM`jq-mZ?*7eib&ZQ52YC|g>HhXd${;hlG$N!6ZEhH52ME5)
zJcj>$#dZR+j4Zx=$;Z?9GhEc!0SE*uK2zw$<QN!1wp@_*Yp$1!Ai}Wgiw9ia=Bu)w
zv_bn~lT~83VOWj}oyp#TkO;uhPy>X4h<+BKM|lKkc$^iO@H2&zK9Lk)nW}@=-oM6v
z89+@vTovth+gLYU?OH3|zS|=_qEhNK9B%KkJo8*T)l|;0t_Tu?%@LiqKW$9D4LJs)
zZeIivA7D)Ho)XQ6!K;ZY72{*V-$vTSk%UX%94ZDfwt7;;+0^8_@2HtS_#8c7LuE#|
zH5(zsnf;(BN=j02E9px~OATP_Z{*dHV*R*~sySpszT)3)M3P%yut&9lR_p4Y_6ND*
zG6IyS1C>uR@x4Dm``V~E-Tl-*L)c2bbatIO|JCM;&@k<nKgE{}7;ym-lQaGIpe(EI
zG#TK_9T=k1sUI91vS)w!r!45e8-{AMhIPI3JEKW8HUP|Ra-%>JGzp*Bg+Uy{X;I9G
z0?8bxRWABOaJ-!=klKsg5Zm!#^M<(<*7~cmA_@kAg15T<jNKVka?Nz?M@7cQ-|Ep8
zd};#P7Y%gZ{Va9=7u5yTgu=m=gyaIVtL5kyZjH^Ixsi8+f_aOSKL$I$waA~ouPC3Z
zdsR^uCjsW*!!?q7G6hG-7NN(lT_5??|4zyjpt2W}&ydIN35_57{rIVW1PswAKV?fu
z8g)cOs!;ik*nJ9-Xu*s)TGzmccSGJIsx#;z{G^Kg&MX|o?xA#LS>ai44HTkV!nxN=
zll3aj+jykeRy&Lf%F*~)xeF^Min!C52AXjUo0{uI(EJH+PqjX&M+tJoy=MS~qR%~-
zvt7Oiq<<vXkTLO+Up}^t%U{F?h4?2Nr-WZ?Voz<0Rc(|yEC;=3yW^$TV7$Vt+X_xm
z3+C}+nD~Ish7U$t(oG(42mOsJ+F>&9wlmnX!?i~F)^U9D?5|d8RpEtvTj?V9TrX3M
zzl{?Cg3q?d*RLLql7@>!iz)I)6y5>!_V~hUXbT&hyj4edZo?|3s%oE!eW~I7Q$qIy
z{_W-f`~r69De!NKlV9UY#bKGST+G-x{kXh~kdR5z#Imw32EEGKc63)XXs}u-R9Y#}
zCrmYmXH*bCLX=d@e2_971^zKo)W)>IjhlQG8p{-=ijv6xfxn(VqoVfv;`Gj9rkefz
zW=XNz!K~$jCTqcrtl$I?HMkiP4yCdpJ|Zjzi-GfWuB)c|@Sq)#^C{1$FNkP|8H`P-
zz&s_LMb`>Zpbm;$2+`r{TmN4eg7ELF%!}2oY}6jxYe+1uKh^cg+AUC9RET+%Pi{kX
zP7ok_HX6ZzXR3-3E9hl6aqtt28s9+yF2&DEq;dy5d-AsIQLeC+xZdHp?*Z2oNj0|Q
zno^WoaIce>YVtxz<H-NEM2w5torgh3VM{m<6`x@j8P7j9b$0AghkZ_-hyh6xXhvhZ
zVor|%FleDv-!4>{Zurh)<Qt;)ckn0I7IW3^<JEPAy*LQk!50+h+ov(1OKS6-F_5uE
zN~;I$=#G!?p*FnU-*V!*@XwJz3-fD(0z568%?wx}<SMH|7E(fOzV{Tqz`0wAE=zHE
zO%F=FW%i4rHJS|FygftCM{-^*>vh^U9Ml~KTp;KekfxvOU3p6VvMZ@x>3e8;(iJXN
zZ+~>h59(+}u)&@02#>_@VsD_GCI+oj*exb*Lnc|%$4Esjy2h3tH*NRu%eh=#L;+IN
zF*M3ic6IxwAda>Ga0rFv_|t4P-m`W0gygm)Pvv8c_HYB$^G5%dRb{Zt%kykR0Hv71
zZmrc>j6#?=By}1&>Ew+k=Swn5B|zvRlk&r6UvB!GCDYDj!Bvo1YnIy&+Zpf0{~ebC
zKO+<#H#wqFCfguvydi?`kNXU52MsF6vU4-XYlB63Cl#!!yp^lsv<_QtD{uV3=+t(T
zgCLZGyZjcHFtfVnPfmP@=DbB`@tZ8I-Y@PbQsoVg`-?a(z`?<!9=}J<VV!LJqPDZQ
z;WWymF<b0d_Db#8O=s?JP@rsZ%vp18{3?#WClu6+%HnbU78Fie@N|5&?{PrQi%zE9
zUDu!&A=>Ag`k6Zyu+Y#$1@>FQb9jFwBM5R-dVW_t^jem3I7y>Yu6X#TkJU-xC*z2|
zzuez`t5*)Q*8tfOZ$6InJP!i5!{B|i=j~OS#|65qt9Vnj;mbv<TRL%fUmLZ7@(ivk
z5*a?i3+LzlHeItk+9$!$r}FdT$p%f31@L2do0|H)2S<GGHLo)t<)nOrUC2$2TzO;*
zzQ=8nn0+M-d_`6&TPG3Tj|%LcA|`iF@KI<~zK|SS3RGyU*j}?ikv-jgnvn2?XUjb|
zd>9<8*fHEIRNJT|6#XosydJ);o9kOf7wB`#he``b-Y9B=*N$(X_>aV=JEAk*n$*~v
z%61pzm>;k8hGsFMr#=#U29jeve)qIPbp20HsWdPb60la_3Jg2pq)gw!cji1ilk1q=
z#|8cS07u-r)qT`j?sQxL=R71Wk|-bhl|N%jzm%o~xlU-jNP%4R1sY#%DImLcdT;c+
z=<HS#!bQ!WiNH~}=MCqJA$g#0w<o<MB^>YbON?uh8j_rAXw9Z8zm|Gf4oaDh0Pgm^
zR*25KZ2@xQr5Hf@>~_*MSGL%|s(W|N`e)oM>JoN`LH^ZE0c_NJ;SsAG?b(jXN)Kky
zGp*RogQg2VT03?^14f$*(}|#kh}kdbi<=C^hR#*wkkpNs+T``$+(~PL+>JG%D)5&L
zyNI!x-Lm<*9a_<lavQ{bBqHr1>R>80@COdH8i+oTJvdfmsB#o&bdSp`YGcKIWXAIV
z2$*jLnLR=W=Q>@&GUqVS>XjVY$OGE*C0%2#wIpvMiFEY$!0Z#cG#uaj*)d$e5DUvu
zMT!EMgVs-AF>CtBel>$v3$?ySH`-Wf^SO`}=Gn1Qu2+kX6SaA;t7IAP9g{p<;bf--
zC=i<>!H5daH5SWfoz#3staA4dnG5o}z;(+$ASaei(d^f`ov2x+d$`B%wbu|`vI}Lt
zWiOr{FEuEI)pe^6U6xZ#=0L(3q)1u$`bVEzjW9WlkeZAPimY1qeeos2kE}a#$=eMT
zwm!ej8W7s~(w2~Gtf2O@?a<4&#V=>mPLG}MYMhsuRBvr&Uq}H<4|*{^<;YmkE)kEy
zuY}El`WY&4knD#D@D8}azOp6G8wbm23ul&~h12HSup>Qe?>>(xi}dWl;-pWR0YuOy
zUf(CqWGL()#Zqc9&IE>*xBvkH7xArIRBQqGejuC8fN>fI<o{atWkie6^vB&h|Hb$>
zjqNmJ*UoxPwlQhSpC7)I+ZqI1ITZ>r|3+iWj{e)V7Xe8)#U$>$4})i`RjGjG1Q&LD
z@cqjn{qx>WPTLVn^8D*2F81c~^Dld3l8~S|BFPu;DI^b>mc_*h_X+5NB*pjdj^rqU
zLz~4T2K#JQ1?=QW&1-7h?CW3du^PmN;mw{XJeeszl=quF3s<?6FAXdG?%#Me{VOCI
zL4cd<da6I4pD*wxzmUbW+-^9PfGgTD+XyWFj-uiQITu~?BiwU#v;)55g$>>d@grii
z7*n(zc+er}wB^^&ebZRXew>jOEeU)t@Fq3SDheSse_cs+pjnnEDVb^Fj3eUR&N*A$
zn}}g4n@+_bH#V-p$6KwpKCwI9tHDFBZ4irB3({!BRm1lq(#GJ<JT&6N*`!(pzt7<t
zl&#MUSCUCUH`O7W<(#G6sF&B}K;=H9FE?d5R%3~Y)mk-zSj5@Bw!Cu`yE(Z;QI)J7
zD=Iz#Dh6@gJYsX)7+f7GU+hw7oc5AlS;N%Ko?1jRIcA-Xh23MIaE(;#KUx(8fYzR@
zB0iM!?U&Lw*ydD4DbWcPdo{GheY?>9Q1%UVx=PoyN%AY1Yu<}{W7qAX^RB;jDZ#*J
zRg6-a2c8OM&O8o#V0csAF#V|e&5l8p#Qx)?LzTciCDB2Q)k^boj{}k)ua3?Lzq%CH
z73OrQs-<3Ja;r?3<m*L22{%S3E?A@Nf+MU0_n<egkqN7n-nds7{|ZN~4k5b*dUTUF
z!z@m%x9x)6yUD<hl%Y0YC?x1dS?_GD6iNH7Pp^}3>pKdw_w>5iGxJ`Kl@L=A?7Zl$
z)_<eO{@Y`}ZP}Tvzdc$Ihy)thBp+TrJ$t7eaV@W!*JKC_z<ZNd<m>toz0Q0sE%N$f
zp~J0FOL*j~XMcm=Z2oCZhP8b7z6-C^Go}kr(_MM6t$cCP*z0=|F#z&2G;YX{wr==N
z`p>G}P%kgGfzoN$lc|Nq&M-tV$xPl~so%UKX?Kh>0l$a&AZlzC7@2p-)!rhwW)c|!
zSx3G}T=GPX=M!V)C#^v}9Nyq}Y%kZL--uX@?>$OyDtPOuZfF$0P0)Yd$x8U`FV*r;
z%nEZZ+J)AWa461U!-so3Z>jH3RGyDZ%5Crs^0Q~Z|15dd6#fBGEUFMR!}~=ph|%&k
z4U)@hb<98H3pL>$s7tHy`nUb)L&0mk8xc<_PN=F%6)ObZyu9XWQhQ7o?aS!*IAvCU
zYpt1eY(M|&gze7g1<s6E<o}@QEBvDBzOQE(7`nS_kQ5M<mhMsz2}x<ByP2Uoh7gbx
zK}2a#T83^APywZDKv244-pliSfB(S!+;h&3b=F?{>`-;dYtm%fZ^>^MXPceauSlB*
z0*^Oct#3Z+R9Z6E6e<7j+`G5CngoTfU(_Dda86&j+1Bg<PhdS?HC1Ot43ckBxme%K
z+l+p!DxLd_8-O+R4GtaMoV;b??8#SNkG?{->?wU@dQJJ9AwYcgaCU}YGh`4LR<FPm
zUKW19i^@jqK`ia9XzELN@cVQ*KGy6VCQZ`~xQ_O?r2L#8smYT;H~+$XW$R-2@^w;v
zjAY`QY2243Cj|7BlwYmal6yevS>eTC1N*s)3{D88=k*zz-WBv(C+>Cfa;!nO9oJOj
zh1ar8>xbc>@3;I3J=quiAFZXbH>R?GD>#{hktKX@1XX#q1j|2Pb;txz48^s{8FGFF
zkbzo=Lq?Zh1UBdw@BLfJvrs-kxwsOZ;r$OcMw#DjxTX>($|xsyNxi7`L-#gRAOkk}
zjEPN%TYJ6<3~f=4!Ne|p0Tu1@DYNj0*Hvnl7~sKzxm1P*(tuvcEu2N=5RwsOSOga;
zm`TGH@Ps)Uka76K#b&%lFeXU=&ybTe%^g$(x+Ul3o*IiUiBc10+{eplKn2MV2hB88
z?7p&aI?itz2*il&^|c&#U|my8E?fyMbCuIijQo?t&4u4KwYQ<EJ}W&sc#L-)8(yBa
z3(r1M?+edM7M@91G1QM$DTJfRU7Jg7SzlqO`??f$SvrdN$~xMkQgV|GvJJmYE$FPM
zApG}xkrYy5SaPEEk}RZZLJr<XSR_o)Z3t+e!N|fJ$xt<_PsKHPw<lCRLw`C&gxmY%
zdlkP%tI)VL?kiZI;fA<THAycig{_HS==X^kW{C^GWlLRb4+%PS@NOj6@C&dGA%?3Z
z%#q0aWol&Wip7HAr6nFq7^j~H^2}+9;eg(=BRk?z{)pJ{2?#t{M&3e|DUMqG-rl|C
zVC?*agc2B5UZO2HU=I3)?1}fH=9@%W6fm<qjDB?{oPQr1pRA*kb1Gn<9q0X`&G#j2
zK^!u%y05o{A)o3a>z9ve=>wR~<KxBt1+pNqu_l%O$>)zk0}DSAJ$<D|Bk?aK@OAn&
zusAAG9fO5A%?lmsMs1n#Q3@2*wCA|B3%#pOQ~io5w2<)iN-EAkm{>^qp8H>JaqHT9
zz!C|yE3Zk4Znw^XVXb<Z3VK&HJQYKV*Eh$P;nOt#pkL!8Zfn#|=N2Ca(Z1W&utnW2
zU1PXvJbeG^*`q=0(;1dZN;_39Jhy^Hfmqs09!uO|IoaVx`H?np?r(`auNjwBk#)~A
z&ogx3#+H5-Xe@ZrR$M5y2}l|78__YSYDek;p~$+jo~3-ADPRk}t)CYxHFJNYPl1c#
z`|>1!J2kirPnX?G#(7^<UQRe+TP?|r<>~Jy<EubGn|~!J_-3@Fn)D!p6@&p@Cx*{-
zEDa1cta?s}OQ&v~{*W5rgJizIhJai(a--8-?mR0p`jcP7Zs%FXvw2k$H^?I#d7}r+
z$;k%dZVo?Y_?;w$cf}c)Puv~W`keBmh&dqG`!eE-PJIXOft-p5d_eP+7gzn0ykvuI
z0nlRV?sn|QQTVKS`XW#5C-~46ZGOp@_lYHfiFH37zUkY2eD(b6*R$z)%xT>X-rB0u
z`rMkUybL%e6<fdv_EJ}qCQS#72u}O=+XB5Jls|KYR6%45DvPa;M@?w!Wj+sITZsFD
z3GxRlG5_q~(=A79J2_^iz`27k+Rti6_)I!U;2{miWswRw!TWp%{e>qNk^Xcpdsbj<
zPSr~t^taH$@9cKL-n;UldR)B5Jt$1#sqLt;h7{mRIJLr*=hHfsHCPf|Ns5VeILhP9
zwkuu9?KeZNm0_IVCfPSia%YEm;+KPey&)jA9MO}3kactX=AsfFf;&IT>=|CX3bMLC
z^`r(`vB;Xhvh2i(AHms?0S@Nef19YCv!E1aDJOYzv)(9VV{~P?k73U1s}r<}C+<jC
zG6rUr<<>zY*4iI1e}KAqP*G5ERogxT6*H7yl*h7^hB(%4+B@X$0C<D$3z27U@LY_<
zsZi)GO&xqSIB>rNov+(V-YMwucSn_Bhp5k5`M2B$*Y<CLLdJSz>l(Jx-@<LP(3~k*
z;10aZQauEA%qClF!lMt^PrsC$S`GyX7=T&j#u`o3a?QVMc<MqPUFaA*`e%L7AZ;Aj
zUX<@ZQ5qghwY(OBF_s)lc4Gx4K1lUAi>;b{3OGo>yHiU&{nV?pk4#2io8rTt|8k8B
zndw3Khr81~VeMS0i$>uh^3n)y=1DZAytbOAB?EWYuJ`vv{DK+DmdL{>isC_XN*_~F
zD6$^CNtKC=Ia1>p{xI|&gkp?)pOd{kxVQXTp}GafQ<*OzUBYk~PE(*ngS~fL=p#-l
z8r`s{T~LKGDAgV>W4)*(9iX>`EE_U3uS*<V;LWZY({f}uvwvLhUkoe=mv~?R<$1)T
zRGMqg$(2PoNo@Y#;$Kx_E6d>t1Vn!t0qg3S&T0nkkZ=Bb^{$e<E-{wE6V91;waIiD
zC_VAFmM%}rQV_-RdOR4pQ?1(@O06<<mz*>SUaFvPt5N$J=+A_6c^K?-2S_Tzto?H1
z!QM=4n-+NA6IS64ev17~LE}n0#r&L1ZmuEazB}8Mk~p3o$#oqyIyf%Ju#~SN(XqCk
zYTNQqto9At+uke}0?;-8A=x=p-|rzQZ(9p!9*q6D^B?fY_pMQ!w?U>r4}&s#X(SA0
zU4$n2?)7n<@*0&#Z-uNxSlNaY(Idy(B%|nlC!<KohuHjIA|oHdZjW9v?6)Tvur~uw
zKB`lrT(ZEL_{00s*PN(Q*qDlj;44g)3)-#tr}ZoU`jXPv5byll8;4x4aaHmB8W_*j
z(uM>IeQ|1far~uCME>zQ`X(neWJ*<h1wg9@ihnyxO7Up!LEs4vsi_9j7`$`;<vb|M
zLl3KmI$<^d)Riwv3i$`-V9s=DO(t%W8=TPPsIM(Y{!z~sS?-~6Yr2)ZN9Rp3n_>WN
zzKxR{lok8nQ_Qo*k%t8$W#>}1RIJY@!O}vyEU`i+*y>r}u@X3}IH&C63MzUOJus-0
z9WA@wxmHLt=4>5{#3vwzgpDw!EZn&%<@)F5ec{x|3-S`upobpaW2jhY$BT7a3&jXf
z)%d=TvY-9pAo?vVP6W2|qc6(^m~H<Vqxo~yr8yfs4cNrIy6>%?{co{hsTUs{RX)zX
zNQ36YuS>{tjdgsQYn`Y9fo8`)J|g};s=j{%*o5qVHeqE4zh>{P`@2yN6{DAsGsbl2
zEmmMR6AWGYbH$j3i(d}ST@75LZ3;6YgW=iy;lYZ!+5u1vYoAZ-he(#lPX$pLD)EGK
zRG?Z^FpwGG&|88@7HQ%TigR&vM-4j?Q^Y~uRA{Shs8=7Q629hz1rC43V2Po;eJ9|t
zNYndv<$j{`%h5xeO}pja=bvziJV<9-l+4bh>OVMoq<E7L<))3`X3=G?x|1-oGknmi
zCj^!()<9F|AnI3N0#j62$Me24<#qp@qYC@I`$+g5PYo;Ozydtv&E9Tr#=?xc@5R~q
zaVY-R&Jv!v*k5oJ4dTUc*dTRa|0$<JYk=2h@6<kZ8aQb!Wau3JA?eHCLhl6&HlYW?
z=e4~T1o+c`M%S7h1gDW~s7wVx{Y<?K9WW@3bQ-!$h+3k|`nRPrj7&h1zIZ_KmMJ_F
z0B~z4Mo><Yt!uDkK>9W<-PV@ep!`LV>l#1p!#0Hbsyme_aXYQcsE9mgVghX(6eO=n
zlKm-PVTyME<4>tjv^|&f3(mFU@n)x5IcS6G?@Olx@;^brgW@^{pRs7k^Tg8^CD%3t
zQl7aI|2DPvJa!?u$v$Nl@)|QYl!f0Jw=!VdFmnF!^XG%;(G9Bo=6Gk{FZUroIh|RI
zexd0OP2dcplFPyrL>@!_n}?86OJugcF&xD)33!`bYZy)>j7E^ulSDHn!xfBJ*-~Ty
znGY4lmDtLDA#%zmtlzSTli)OCK;R~B6P@DCCSVh`C*o;^F!*&~%OuMYT9JE<k<GYp
zwKjtSs|8iCz?z3SadLRmSnvI7GqNhx{&uFvo5i?TI5g@PdS}jEIOM#ZQ8h6Y3Ji@K
zTfllByyY$$mJbbcbCPbVds%*V0};r%m6@^3+X@6O$JkE+h=bp~o#;=Sp0jak@2r-7
zVUc+Bw_1VevO1M5qEecK{rQ!|P(&MKrb@Uz8-ssq-(!c;@hn^;%VFsy_53L+fp7dy
zL|(HUudYSKFat85D3hs5;gg#-yUnAix%}slzO=U*^58pL|5-N8v&bdnKdH+al8~An
zd#mqZM0A9wDy6N(Kk4(a&2ar_b7H_o@<r4OUq8hU)9R=mL)OfHen`rX4`1sq9aS9i
z1H5a<_CAGG&3xbCsy&~SG!L(H>7`8i#xHyq0#Dfiq?e!U*iGyndis+zuOMWLQ~b;!
zWZ&jB8qUhfYz1Vj(z~4ZD*x4a8<I`l+$X0A0iXk~-W@k;)W+bOfw5b;1c~dS^vAbc
zejHg?f`kH8j1ul#^!q^*!8-!G!X6x9-puLZNJJlp@(T%EBuydl_t=NHT#=I?)N&@}
zn5rP%pI-8PI2a?9vh!%Q+?J#Idcth|$glTk@tM^)$RAoQ87Fby5W8P(^4i~cQE?c;
z%~5ni0D~-()RJP;kO{VtF)O)aBBTYbRhj8E-re(%VGDT26Ipy}sS#W`TRSJATlQGs
zzGUBVp<qHQKq#=m*SNbHqa>=dLO}#^v(H_Yt_X~39_iZuW~?;-p}k?|$YLYM3as4o
zEs(2M2kZvJIwKqrpIGKbI?2$%=E>(`3)o);OJ@cis9g?>@Y-5#{iF6q+V+YgXvv3|
z&mKeO?N-{>G8kBR`y&>J%oGf+;0zpS<D&B`{@fHz5V4MI81Z(U*MdaN3=Sgw<!neP
z`BQd1`eMTI9aEG&KOyiiKZe)}99FQbYh`)8Y5yV(gqQ(87|L}wALZ{I!X%>h%}Vv^
z(wRO&257Ad%XsSaLM%1h{!a^VRZ}YMR+Rt7;&7DZ^_0A?f^=E+;y)Fb2lKtJM6DA?
zRqqz4ET%B7B_DUgZucYECp7e^`VxVd6i;d-S>rqIOvv%sn`q8?0N**yN{K}H2fa@N
z2KQLP>-4WQq-UhyEGS$JvZupt#(ewt+zv5};K$w3_`vf{0RoDgj*+;zN9@0f7vPaU
z;|#4!cN*nlC=*xD{L1VIW$~=vHCmqd>ZFoS-BpZoqx`XyN*hepvoqY^JQ2<hZ9@$z
zmMqHkURB9DWFQ_XkI0#ibKBh0d;wUXhI8fY%yV@^(;r|ODly@=4F&Zjewxxq`}S%b
zV&d|-r82ACSALDJr%vCW<!872GvuF(00Pm=JGixq$a;LMU7YSHb|V{@=&)wSZuxiK
zfrFxbNA2{WAg)wbzz-*7r!s^z37b0t9P!D;Q9*J#;#e&%0J@U~Q&*v<Wl&*Y(z+H=
zh2t|>tur8|5kk0|B|N#k9KCU^m$z;$e-o{wn9(Z2-uBWIWX5!KqzqdD`!pWXtKVu=
z^T}0E*0otyZ*W`w;V-Ag36gDcDntf|OX|nG$Af37v;Km}(@O$K!p8ypG-9?Uzo3#w
zrYEz|jo?b=P)_GubhLXFyO8BR@a6)4>d|->jxqrV=uG)i*!t=1?GS8KM=x_%6$rPz
zxJEXe<AJX7={G49aNa!0@lX7i=_G8hpG2}RuGMl$7PM>nanA@My8td1YkSicd|46o
zsRsk~z@5C%QN5Jvpq{$PB2s~XR_ews81P5-g+NqXk+o2KkT%Dg6*ze7-k(v+NXlsc
zEgUh<%aBF1At%1mu+0k;i&|0`qkfP${-lsa?8BSnKKtdz$|I!Ch`jr|qJI!186wv|
zILTqZpX;TE18K!&UOtc*W%%=0?3u~DhG)6xzd=mzu2Y6%3g-`e&^77TfR&l~;4g?l
zZ3N5V5zBJVUQ@`86K9tx;F3NnK<xH4cDigMnj2|LujDS632?wfVuRYp;CcBo;6pjE
z-zq7nvT;9w+(9KWD<QhC@T`WT9fSfar_2^u3VIOF;&%B&0|5v^uLW9mDt$XP9p|DZ
zrH=^7p=F;urRY`gVYn#XQ^-dql=K#D4|by-Id;UoT_m|@UFIW{bJN);pU4c1{S*z?
z@jC-Gq^T+p#zR3r5uq&F0&{O|;!3-f*Dqeg355Nt-?wqaV`8uy+*<@T^+4ec?@8#!
zx0>FQUBRL}($)VxqZ#NaT}F$nY=(BvVEiTjJ?1Iv_>y&)ywMm2i@jw+WH<ZaQ|yzz
z-r`S4X96O5f{6Z44iaZ^(N0PY842svfAUTJEl;W(YRR2$rSq?M+?K^ksHHNHu&>kO
z2cOO%Aj13F!g?DK*w&YQOQ;RyPru4JJ4jR+3=A^CNn9WgoPE6d@Bq*`DQxFaRQ2rI
zd%SZ-U^<|Zlkg={W!#9YLth05fFUr@vi-R@Tgt)x@dK&idG9dN{q%PuMzrS^@o5-O
zPdmgo$74b>dN#6wUPLhC*5_JUCz{4V`fAT=OJWzD<&Z3%7vMCyJDm4NlsXXl8XO&)
zlRcyC3m1=;WX|+qp8#G=S5M#cUu@}gev&v-x6yYa5b^})J~im%yUyas>dwm=fIQ}U
zbeDODYmEWPNQ&)k8~sIgJ%-X_L?h0!D~o1V>8E6m9mb}3p<Chlu=^7LYEsza5uYcu
znkfQYuBTbZO~bw8j>+PLRv8|GijyMi9+p=s-0ri0WTPBS#j)V25~uoEioGw-xU4Am
zd2JswfIRue)mRK!6@A(<L_;!s8r^2pTp78so%7XT39QX_>hiyz!;dV*IeHpogu_JY
z$OEP$Z23;X4*3dA4qe4#2xN4xQ%Sm<+uFO|jdd>_Gew0#-(Hut*irx&Xs#2_28=V{
z$ih>~&pl~X%nf1i7YL<DgE}sza*A>8g{E-R-=@6Mn=EI@yrXBbJ{`%!+q?<sX8&0M
zg`m>uOSjX`wp3Zi-nT67W*4mDT3_+?Xz3~pZ&OeQUfgjj)kYDa*M=tWq84mxhVU6k
zWOn3zLiDZXJ(1TSdDeY`{RgmLrzFc<%b%1#f0QNGj+<s$>qC><V03N4N1-t96}G9Q
zb+YvE>u1Zl%CBtiee5ge@$98x*W2YS>{4BCz^msGI%%f`Tlm3<p@SIMtJf^^D&cu8
z!o;~FM75vmgjR}=T1f*f3^@qn6_;OQk3BmO<@J~sD9ap{_>U!;;JRQ%Km@kj-?o=V
zQW#-6qt}Np&F3FU6SmF1YYa6Mvf#82{@xV!sWsQj-2d<F(`FrGuCYrr+#SebulunV
zzN9!7fzv>!2f|2li%!zcN)!Cs;!PIYh4u7Qb*AV(t@Di8_aef)`g*FP##kVW)^Khl
z$q#B&=pgQToBFc}ba-3b>x+9CPr#&{lQ+m(!gJ_;eAJk_UVW;RvgO46E?2ce>Z?jy
zLhR2NOHut8>An`=<~;dV>tS0O+KgMj;;DRNuZMXhVN(7KLf{?==Y&ZUwpT9mR=(lt
zL@%mN%qm6h?y&nj>Zk147{4a~IXR4G(^p@ARyCL3X);1$H(f(m)csWC%i|@*zA~Lt
z-uFLU?8ZKe@zH+Z(R>A97`;s6QI(%;-ZD1a;T@`_KC4M%FRehu<oLVEc${7F#ZO+A
zp>JfN_A52M5zH_5X>OOy**%oh&kRM<4xk`2FC>)Sh;mHkk6MXMDqCWNT>NxL|1eca
zZVTrI-mSKsieCSK43uHp)#?(FcOdTeSMkjr3bl(-QjNgH?^lik*n#IDw@r+G?M0;7
zmY|F-hhFfRgBa@Mps-8?VFi7mq;HR8{*Ntw$ztvB5$v60f`ooN;-pdz8ro4}@*hCK
zK658f^2n_*{~8YZt-HMflqSHJLqiu2SWS}Le5kpe-&q5J2k#XI3Y#3u`p^=$J45;Q
zhV)SKK6~rLKZVl;P7;s=wjF{7$2M$H9t%hChk7k1wl+5Y#~>5@FZX`VC5SMs;e*yW
z%Y2qpWF2cDMK~<DZYr`#S#gDD&Ztg^c#1>)nk{GgTj87@!8nC_sD=7$5*61~PO2Q}
z8;Ge<->pr=<6i?*s!FCjqRdS}sX%>xvzIsyruV3xheugVwr<}|EQR7P1XTO|@)LdL
zIi(pDSFMq<AplJD^eJcXjd1p9vQ+>I{aAgcF`bThv2OVj)?hihKFrI)v9xuFIXL;G
zA$Z8ND$6?|UF42x<)ZN&3=8|>P{}fDv+NQ9u-STKgWf55WE%+4NDJ~j%JZ{?7L>>t
zDvkn`7w-Hs>5ye@<@vgUcP@4%$-%8A^ucQoiWrre=D4VZV#VrIN($rM4_kJEjeH;t
zpl=>(IW0WU@i)7^KNWCHy$%tbop4LjUG7w(k$^J}?^1EC)%{R6DgK@RnN{M<>7QJ(
zqj!*`R-=$Sv-P|=N1TPFdQOA7sFjuvc+C)u9Z4gomWF0U1+F{zisCmCupJoSqzJ+6
zXmSUL|0DByhrW9II`e~1>zivb(z{7C%|d_El>BfRy2k@~=zt?F{^jZ9lu{U}JcN?3
zaP`X&>UU33VF?e5n>>&dc_O_b^_W9@OcQ8=YroKYpF8vsI2JWK8A4e;=Pumhp(DTP
z;Jkd(_j9R982Me=Qm_RzX4R9<t}hbTzVd_~1zk`5$G#l4ta6<!YCV~#JoY7dCG5bO
zai4v8U=X*N)#;C3s`5!f+J_B-b@UFdZ2WadLP=6l%XtGZ7FcO%LzEx6*=kWJd`KS^
zekkWo&wD_aaewe=CFw|t+VgM%VhZm6&4dRb=Qnv>4YhuJ*wAK@qjkSM4V4rp(7#FM
zDzo7wNH01h|6hXBPjpc|DwtV}1ZKn(m$CtwAB|kD{N2FiPq>pn7{}yR{N9DYf%`1O
zCafQqcCpI_Eubx4A?mL(Bz70tDJc1ef*O!kc_VttS8@s(#$q!q=$sP_WVg4-MVPn(
zYj%T)RV9L<delWQ9q)XYZcm&*TW)qq`pmoFHjxb8yWP8FIRxvc00rPu{9qm53a&#<
zJ(%ZTyr$ySy9UPIE)!f&afG7oEi1rA6zn*uxWYzzSj*f$yt}2EBx28A=#Td@^|Upk
z34n929NqJ1JK8fW<#<W4in0v1Rl<=}e+cpb&shbuu$<e?JA9;CbC;P(>~ztmm%fX0
zzMdO!KJW`(cgdl~A?D2E7eg@($_4*41`wb1aXbp7Uq6|!wo>BXIQr!HN3gWbLww|Y
zQOTFu*U!XF&yg7w<vg0rcg|raT&-5>xqt|Rq<fIL6f!=!$C+;9Hf4m3#7N4#cDFAc
zO6W#$Ozc<DoYKchIKQq!-7h?wF%EcC>N~VExc7rt-hdOFSQO>q&J{_Qi2cnIw)%+b
z`FR_>oI(eMQEtR1OpDF>gf589ya{D|belk-=0ZXN`fyAnCX0%7Saz?Uj<lt35$Il`
zYbG9QC#9c;xcHOISKgS6?+hzw?nmehyx7mCu&}=|A=C`v>A{FoDny74pZ}NDMr6=G
zeqi<hMzl7s+s&b!Swi;L)Hycg_wNO8Vxdm#Msr(5PI>9FKDiB2d-jQ}xi0r&9!;MA
zmJKreNQ!2w`6qkEqaSKsgfgAsm|!|9yzlh{{`{6_PTruK7PhkITDbN%{EVC81Q$l<
zO#-JV1-nh!qlO&;JZmLh=u)82+#@N(9LF|CgtUF4tS6G=grX|Ri!G$jcWg^AqQxE$
z?urBZ<BMSyHjq$)yfNh~kg>A$%k-n?jBD54=gbph=VggR>CvUV*&-#O_v8WWdPnD!
zvKi(ZyKRSZ0hI195={bWUO!PkQ>3bYMKAp|^@~6uPuEAa4Uy@BO@Rkcw}K_DLWO*v
zgp#OK;AW|3T}N`j+v@c-d9d*81lys1X4Qzs(Qfxp4n&6!`pQ7}by&*eOqGK76uM|B
zT^O8_&W)689+eM<@xHSyrI6n{x|XJO+I%KE*aAIgy4}hFShO*jkhy<fwn@*odzmgL
zg2xZ&&H3#@yFdye>u@G*$gFwX`h*JXj~pY?z*t2-Z<d!@X1yKl1)kjTHgR1sdB>O*
zpdYTT*5bp-^>%pn4qjAD2-rdj6R{PfNklq7C^3NANKOr=Ft{Ni@<xF%Mx0*EMq*E=
z9~NZvm4B!TLpD8M8tliRdGCl}Qo+ahj7caVN}wM4v7s)KpBLhbg<s7TG+~k0nI1L|
z2F22nRfV3JWE0NEsK<v7f&$CNUK<BYMK#RKLwD7FYOsBs|6MAF&}%WZQ%-saBKux(
z>pnj7G{o+B8Yu89<wg#uPb`<V#AHWOt3RQeboMt%=+xYnNxz??ZJ|sf<&XxF&pm2o
z+MTwFNV$##fIszM$>5I<LW(r>d6bgd0ybXD%_eWSd2RkXNm<~M&t-Kv<RsN!IaHOI
zE9<p@wj9O*F`X`xf<~86Re5f9J-NCG&TTw@uo76-7hA;Z>%h$n=Q{fKA9?1Y?@lzv
znk(IB!;sC7UFzgFW!}8~QCcr<O_GNw$K!{BM9P)w;DsG}3&)pr$X!g2dRFAl%%IXo
zkfI}#8;2FE0xVrCzsn>}dy+H0vojq($pGua&!ZoZ5HVqz>o<7UO6}aAcr>M~j}R>q
z$YZlhuh(_X-9ciHf^>g5LtkwQ5isp(haJgzvrGznRrh%*%ILW+l<3(7$RLX;pkh|-
z-Iu`-o;4f&Wn)o}{{QIKP_u=<nuurL{mg6?ow&KGZ{e&S1lp$h_Tmq@@m+qouWX;7
z_(5X^UqCK4C!^24(_oEMbrz+S@WkKU+jt&$o>uRWX_z&U?>PnXuf)4P+aJOBfMfFu
zLl3OPA(8ldwNI_AF(F2?E5W6`i8|+u%(Pfdk<67+T6!VCfqMKpI)N--pp5sm7`J#5
zRz;X6$3;QUAJUALeGMgWFdqmI+Rr@0;<t<P4d0{uKCeY1#MX~#^fpToP!uE;gNSj+
zHQe_I<-L{fhv_#ty^s=N_n1%1G_fVQ3%k*E>0tD$Jwx!WZM19o_fME5#Xs8^2vB{L
zp`|N?v&z8<9o@^?-^E@Rs(u_{S?>@{>ln#;Iq^)dp6oxn?RT*ym7bLJJL`|tUM<4;
zd`<~|fmGe_>@vxgN$K3P;_`?*+N(OJydDncI%vgw%OSCKGcW9tD)D}n$xg3Dxjd03
zOB@E2+hlzM*5U8H#QfQ>+UGDebi8FpBDkH(CY%sabVih9q9aY42?<+bu~V-b^jaz`
z7XE08ZFb~R;)zIJ1wPYG%)_udJn`K|jKXR{LG10{*c_u-1q}sUu8Miktk&ajm$tpb
z-0g7*px}0vo_z&tns9zELTc=MQwF#njOK1J0WI@^z7P;CS$P26MEr}A=eTUn`8Egb
z;nU7rpHcf+H18Qazq+R$SQt~^2(4A7GHIsqAp+SKEH!Lm2ticRmVpoYRS(N}5=0oe
zK~Lpja0>JofnO^LoUomI?M-y}+Racl=HLLBPOp?!iO?@ee5)m4LUx5qa%Jp@D^;CU
zr^hUzi$v-Dx?>NgAWadKOeQnpg|<N^qSP%N>n+5Rztwea;B0=>@;LXsjp{-f$(-Jn
zbfnII71ge~e027c<9WnRy6|(AA?Sa1Js!iLW%C&I<<==t?u{yN@S?IX;hcXPjBjW~
zva|`IMdDGOHKP;mT@DeY5RRqGv0rv7o)$I6ZD*{Mjm%=F7X*Dv@@c%)PO3wO02jy~
zvu0K-y6@f)ywK!#+%_9wtbR}>9@g{c6#)qz2p013U3We-1yn8*IUQ~nNw;VP?Xta=
zYf@l1=UFS~^9YgUZBU2b0iB?y==Ro$ukR^GD;>*9CcJg|_2ya)$_nM7OPz-;9LrIC
zv-W(Zzj*i;QB}Uond_jP-C4d|Y885h|17SLulKff4#TWI@!1CYi!#_bFZG<JeaI)a
z%ZJ3PE60<8UXF#X5M0Ru>k`rq?4~6Pv=vvVr#dKV=j@f_aZ+Ct-oqq&r-*l-aCJL-
zz2(si5-r$TMZ982{38)m89_7ippLWu(X)7aN(9}K6=$J=mFU!bGvN>-2a8FOMx_Qi
zUZtjqUmnm;bSU5pr{%xjP8Z>34YeWM$NV*Z2og5Dg#!PXK4m<&-{MzBQH#!*_e{~)
zV7bABEvqx-t#M&PerkSMP+O?m3GiGs7~g%|^L-8oABK|#fk`HmEF0q$d+6*N6o+#9
z6TQsy1M#~A&jr;jS>!{1TUcwu_k4N0lNZtXtBn%=Y`)prswm61->z`Qx$+o18ztv-
zAVQecwcd|CwktDBo3`lRkl!VVdR2Rc+SRyb%Ji{q2BGA7>8|`Z?g7F#RsGnEEMrAA
zJ)p#PB>8Q<eMV_FIav4#Q;SsXs_2(3ty|@Vk)<UHXbR$9G302atEAYhJz6#|+~E(W
zj!Sw<cqsaHx*NiK&p%VwfOCpW&|F<LLFB(;?Vo4pb_QkACRd?ELZEc8TRJ<|?acl8
z1k=H!uo+a5bbA!i_E#(9$bhv)^f#}od-3~SUbj47QqqF)cV`3u(ZX|ZM^5rY>Lp6t
z9(hubroqJ1Iy~vlo_x546BR07Q@W8gLMs#{qF`@yy!yQnk5CQxg{dQ|RxP4MBDSP5
zl+Bbbk8`k%Y<RW@12i3u_Olm9ac`+SJ&0+I0bQMRPBJBkr~Gqc(wAzxT|egfU}B$v
zW_yKYiN1BXnKmoMja;&=afE@)cQ%uF1-})c#Kt&_RVFL0TAFj%J{LbBpWU&+T9aJC
zVE5B_p}S!E$m0Z~l{R59{c`j5XT8^qZ&73C?ma?A+4T}#dt-}r6Nm24ag<X)H^6D1
zBJ+ZyBq@O_jRuN|aUlSp=>qPY!>?f~umB-LD9E7^>Y&poZo!e;g#w`tlo!7ZZNWu_
zLs}Z_@lUag=`BR9rjs6Ghb4Xxvc9*f)_gx-l4rZ$nCLV3mgbz`5G<3d+}lpx`4>e!
z*dvd0Tq#F5)0uqZ`tY)f$0=@1MGRU7b=Gtesw)0G^bXVMjiTmo5PmEi!q!E}OzIAr
zVV}hC4qo$zI~N>EyHm^805Z`tj&ucyna`B2wzh~LsD&5!!uMXKsr<4<pG(mFc$W<5
zq|^Y!#)WJKgTx)@10v|3z4`FxEPq<{&kiOykg}P%Jl7GQ8OQ%{D>t?o?Lw<E-OBT)
zbuj8st;PG)RmN8~b#a9L!z7TtL-elX?rexYI;(TrZu~7j#=Ad|Sm_oQHZ;z!=11a4
zwDPrvbM91-uqwpfpVOIUrshR7O#WxAFgGGQ?PFw6NdmSv0&kYtiX{3WDVGi%z}H$V
zJEO!(<|-}KWJ6{+S%;T67z#1cB;KRgU-3e0m}X+BR17kkS6}1ltD1VUFWmTk7ni!k
zxT?>YrQw`o7N6v1jxUSdU)?It|3Kn>e>IB5>YA9Tc3%h;w9E%1drB&Bk}pPYVkN-Q
z-e>UsX|ju|blTX@y>#h<B&ENYEdb!du=)KQBD{^+l^(L%opu`zcYqOwQy*v(!dduJ
zdOywAb#A&Jv91={puW@ay-3`8<CCBJJk@hx-Tz>pk8pxWUk~7QLZ1F_XL9?oN>_z(
z=e$qI#x6-;PP`%{<8I1+ycpGn?S+N!r|p|UTsw~1D_{XHE~EGq2|+n-wG-?V@)-p)
z72_d$psS*aMTjPTcAYxthJMw`BzDUX&5;<<sVI9+%Bh?Eq|YfO7xl~2qJ|K}#az)d
zWY7Dqsxd0XE%g@TMtT4F5yf}RlMO&JH#TTL=(;G4KdI6#_;U!_N+gqtja8`1ai$h8
zB}7Eg(z!5vrl+d=Xv(04#R<<>q@rO+=KF0b3s3_u$@zQX&-Lw>3i5GR<*;Bv`}jVm
zm4037_*0hymP#v!6B&{M>~Y}wzo=!gCwb3EX1KtYoPZ!+lL8P#0hI86A9=s>>led1
zy$8wMgg9jtwmR(D!N|H_8VZWC$p3gf1L<+Dgc-P#CJEKu7W%|*62m5K<V2oCVr#xa
zG(}60auEZjVMV>~Tr%D!g}!Rvl0@V5yB~|hv!)ycWr6jkkXn}*6(4GxV~juU6F;J@
zYm`UAAIz0Rgal>*sw}CF9mX>It4k0N1qd;o)2FjHW^6fxQk1Ot`*zzG<bKCa>EXd2
zyZ3)9{@`svNY;}7{zgE+sjRs+!&)5lO=UH2t&f>g>_<}tM(ef$7ysXBlitst%FpK{
z?`1z8ZF`Sp6jk2xiN_@V7iJQuNe$6e0t|M1klV@I-`D=`>h+(e4{$-<j-XvT@%m~n
zUhr$SuLzonfN)V3iubG<d@k(2j-0U4>HLXjha$}@Y*}rPpjRrIT4q0`Yj7XiWXZSM
zF><93r18xEND|l*ft=$`)CqzzA98-gSSlMS2!^&$H~IjcVBYJdE<p-!Z1+t&(kHU%
za#JF5rKC8{Nw6)lgj~PEUa}R^|0_&<aUCfmx@iA07P<F{D1Pi^ao9&FXf^SF|E@vh
zhJ*M`RI^QXH2fy#oBO+s{q6UdJv+ieeaw@?n>TW6mgBzqA+7Nr0p&ISgh<(;hFeWr
z;!?r=l20UaQ2@sQ0Eqw2{a#(>htse}>7jd49X*<}rXZ@IlstUTWE-|sbt;Q(`mfz6
zT|()7BTS98m7P4tTq+O2Iy=Q^^x#oBc>wM#YcwpYv{b?dA0$IOk9*BU29{0rDN2#p
zLvvihO|2~oI+$LrGiKPG(}EJx-M({}>{3meWU9F@G5uj=y1`s3tix$=Q+kH)#o<ba
z=cNfm@hRuEP#I(SapVvxnY*A~oIe&OtYG1hr!mUN-@Dqu9)JAoLvLr-D_V19u9Ocq
ztc*a_uGzZptF{*jzg^txccazL#lz0xzcrRXDB0d^W0EvA_gmTvR&GpT4f76{`wLty
zW<Yi-{&f19ZLjJ^n6y#}0{j{eupPvJFTGzL#$>WR-ktB<vFJTuf_7D2-dJ5Kwl<B}
zUxN|!hJ4>5!3ZhBANLw<AKG-uCc|Qxl(ED%;w-_9O~tFiH2B5tY;g4xb^1}M5@I(R
zj>K>iRA=AIFrXoT+uNhQE^%9c?_HHQ3vOP)8B`#uN+yv8CuT`=mufBENnxFUAuUNO
zjSmEw7cSXjo7w9T3KpiFBxlm975n{xgvt37e*GW85e_i*Sas4JfckAwwFR(!%|>Op
zD&eV#4dT67ql?RR#a1CK?z{?@zuW`)FE#r_=O2DbEGg5qu?1>LcVxhjer2WTJb|x~
z+YfWo?|D!8;4b(asXMtp<uz86*VL(g;>OT2HX|M`o=|wfIlQrEBaa~-XI<XpawD!*
zKY&q=yrBJi>BwKn61!icGn>ORA=S9I^Z~*wlm>tnzJ&@D_ZGwY$z;L>_6}8+v0snB
z+DC%^QE)~D1Jx4kHj;;>E^+E)nuT_WDAPD@78~&hGJw=&{+Ie8Aj!Dtwu&Mq649Fe
zcNkI+CT-j04X2(RWzi&FHvL2H>j9NzV_dGkKtlzwHjjJB$E^qda-fyEzqF~j8t(RI
zE>T>gh5>o&dG;|@@lB6UNqQ_>Jn7KKOTv4+t$a|0p&~o>QLaJJxT$Uhn7V!xkEp>~
zjoH)s%sOlkNQ`3)euxmF$l5-?Ur&C4wavKzKwGoAi#SzFrw0YmlE-L+Z-4$^n%Kz?
z*%U5{ZgfvAr4DH7A4sKM=T|3@4!(CXPO3;OHvc~@fcm4|{=N3XNHts&`b;axa!lzc
z$I;zy4wa7w-dFm0IOdS^A|NIH0=SrbX-M4{+r;mR>)*HEn_W8k93OZr1}QiF;P{uD
z8_kr#C+4v;YH|2;u5Qnj>VG&sy#Rm-O-pb5g;;+|*@)fB!m3aQY5$mQEtdn{bB;AL
zc8(q>OMAfISZPD*-ClA3((<I?m_8M&eZiZ%(Ho4M*EG0Q1nSqePykqCaomB|ryAZu
z9<UEP?p#{8^{6v-&I?vrfR0mFXdHg|@kokwH=jK0UGjOsOdsm=T9IY{J8IAnP!ukS
zLQv-|ziMs0#OMT@nE8ckEQ*BeZ3-0xQm3}d^ZqR>TE?YCtW{+;on%bI!{e#ZIR>Xh
z;$tclWK=5nfPgI_{jIUny`6tA`N`Dk89|OJ6ieM!lB=Q~k|ggrOM|b3zeNE|1(=4R
zA_0$&PJqbOTF*lMBlk`?3yjyZ^(e@jV)I=o8Bm7EWM_UKdkN^puW}&TAfC7Hnln$w
zx!Sj-8f~)v{8$y*-eC`IpjnvegxY!vBR_;;bgXnnKgrOwoc5visZ*xa=<}^W^86r*
z|9A_QhIqFKesVRq`(xj-F@vB9XU<SYkY)Ga#L5oN{*!AvUy_@#(Dg#M^2Uy$`FwBf
zih%Ur4kyT9mGt)jRfD+mH$fUUU3i+yDG8cz`KfT7&;vlrYY`qy0bgRr7yIt}gXk2u
zKZ6V;{1<q$>`B=oeSY*g%rJ$6S8qrj;E@M$!*m|5ihs`rIJW_r;$K!e><iI$jlSmN
zn~CGWb;p><ISg(RU8T=D-Q?&|&!FhfvG4g^@)hs8W)Tv;W}4xQYY)0cdl{gwjbGu&
z{})4sxMxAl_{D{-2>^al*c2-**S@ze!eWBrdozql+cO8Hh(DVn@BmKf<ok2imkSJ^
zbk$Yj#qV5H8vhT*@|cO0$cf~K#2+k@4&_h(GE|;I9SgIFuB<>-U!JLui`B4E<Cl{a
zDY8r!M)*snk7?3~yw?2&Cz+3YxOur0HFaVNboroYu(S~K)>4GTKeacsM@0OZz{$Ep
zX2&i=)<jZ$k>n>2yNzq55sYb24casUT%(wkpxWS#cJ{A?4~=JIEHBZ1gioR}oREH0
zYnEmw#T#4kcf|}Ct2pRsf{R!l__VIMB$2}%=={qWujS6hBWg6Rs5RY52~8CGe*4<L
zsmSJV+l_#mYRWtyrXq{<sM=&_&gb!JB@o19??@1V9?i|pyspC3<4Xy;P-%N>OO71y
zSBe2csOI0OF_YdJy$vQb_=u?8j!Rvni+(~S=ptIc-K7=vHR+gr9=B2<ER{R0v}K2E
z?j_gA9d;HS<hbU$))54HhU`~t7qg};%+&oSy!;AAkZLLd7e!9zq9>PhbdpcRzhJp5
z)nigYJ83FfqEnPsf}9x=L0HW(AAq-MV1%5>5EwqE>M7=%0*#&a-LJx|-{RHfk!3x0
z{!<J?vmJRu%44%~X=#M8_{q4j?HV_(nc>e5O>X}xQ$80h_we(AJfaRZmb%b+ZRQks
zg!680Fo6Qt6n@Q9J++5jAEA?3?w%=6@;*>KEip-0at|bjbFO}SU{$p{RC7!uq{?)2
zg>)zzQB0p+@}Qi&C{VrxjjH~Porzc=-ZZ1wEQ28dK&%4}Lofh5B~%1=fZkNuNRQsN
zfPd~m<}Kz8>VUmGT$jP-MciwgFPwy#+l7>>zrOi1hM`sHJrYmE<@K>$K}`QKYb~R&
zSCEh7jMLr-oHB>pd9-8=C9-scE}x*AyWIw8UNfILQq<tzaNwX)z8rc0wA)>fosD8u
zKN$z|0q1}DKurKB&Uf*RhJ|VYtx)=~N&cM7rf&&wUwKx<`tFiWl%r)Ar@Z)T29MHa
zAnokIan$pvw7Toq&N$0REkK!N$NThE?R$i}0?z${nf>~mq!S!arR~9eQnSwcNg6RI
z`-#p}&gUdJ;|EN&pL;?f+@Gu);BkHDCq>kNB5UQYgq%J?g#N*1haLmr=34q8U+vj9
z7Z5pum?B+KZfC`D)wdJ-1-BPWNe}gY5Pg%^xv&?kdE8(5S?pJ&nqKC?9gkHXM<>$p
z=)B>d5&_8ANO54K+T-72fq}Fr!ffl(SF3cZF8;hKu<(p78r!PbL;%_nXA!3xc&VfV
zrx6ciwy#7XBu+A~<6R(Mb9Oa&G%sH!deRV7-(omU3qF;$&^B8^U)2kB)%;%6y&S<b
zGqL{y-d(wPvl$34eGZa7FajHZvr~ipa@C$<ePG`-84ma_hxHMI3^-cr$3Vd*nKLS$
z;plcT5n<5sF)EPiN!ppYOvom*P^e(EWi`y}@_v85mFiezyZ^lxH~=Yer@pqOosJ7z
z97%f=>m1Cf9$`wS4b0U0F@}P+Yjmlr7<dReC#y5yLSTd%1n+4BMkOUuu|0?5tGV5=
z80Xda>F*Q;(uh<}WYZZ3T!s?&LR0PaDll6hK}6(_2g73KH$4MoBU@rqbF{AjZ$$C+
zgx`RbJ*a<mVo@$e#%TVqa|ESJ=P@Tw!ucjt@CVAykf@15g&)*?M>xaqxfv%GYFj4S
zX%I#&G{TFAhYeOD_2jLbl;><IR2pFNVA|C06YBv+KrjZbHlLsBwIZ7)S_Y%C?w03y
z_YGUx#a+xRSj%u0NSb1`FqGx#s-#<gr?{?I*Q+QlLyA-AASn7_5nav=iUqw4(kFfz
zdz>|l`w#kwqQ|7$_zeidy#E}%)XA_pZ*Zw}d5`ld&^x>h%f>Pe;{z0sb5hqbc}m-A
zlfI0P%<bA+Utxg^v2YZ4^R$r|&pGgX6x~l~@2svo#~u`@94rVgt<kNH{*$MX<BF@Z
zAZ|}fGb!h00O8@@rCynP<W+AzL4y$EVP3Oi>8q#~T*U&DK{^Si6>-#p{B40u(5aJ~
zWN1vq^T1<6TC#1q{>u2y)lkvfxIHsMnap)we$Cc9sb75}u+Q`po;jBoeyo9$lrNzl
zJ?JFV9?qfI>ZOD@pC|r1@^}+XK<Qe-B=&@@i;9f5xKI75H8%xOGgH80UeRZYb^>5T
zY}~@A=#Lg$&!;I-IH;oF6kl*LPfr8XME;q&|BCInxmkCf8`QcMi!J8{{{7q+M-()T
z??*dXbV5gQ?7_%Wh22K3eO0^!ZoQu`|L}4iZS<)2XSbGWUL9DB6|T{$@<xU408k;9
zz(~cf<VY*9oIw&j-@B)^$IkL+)M9W!AfnxRYa{oex9~Z&i^(qL%KWhK_`W`$*Nu8V
zuS^#^&U<7$zE7Pdd9J4eNR09>F>2MW?)x+UY8jMuPDN)%vg<l{E}vJC?L8KNRz}aH
zMPX-uMjfe_brVgTD=KHm;pd3eGNuI<D%`@I5hjB#{@B;zq8usAV?+U6f!LJjGk^U3
z`^!(l%MnkosveBD;a$_6H|)QE&$FF!zlYbAzvL5Jwk*(=!R_5Ew-Ce4`FmK=`uQ?r
z>6)k=%;g>fNFw#&gEt9|qG<3=8cz39Ch(?&Ws5qfm}2fjX%y5ty@G(LNg^m#^M81t
zhaY3(Q9t;l{=Kx~OC-A8PydsQ>`MvV8{GTto|1WT0n9?o7mv;iPBNYuD5`CH+Us6&
zosC_?mP|=q)K5dGxsv}2wUIO17SWaB^BvFM;aAPPx!<=Z-=rQL`mv6%IroEs9{cLf
ztlBUDvjby<)ItcZ(kd~MaW(J30kpHMQu_UxF(R*7K--$6Q|(()r2^<c<(-;G`#9;3
z8JM0?*?DX(Z((SsO$@Xh*7+<Rquc#Qo#xT?Af6K@e~5dG=veK8@_2h~V^|(saB_KP
z!zz}rmqnGv{NLVmRvS>C?al<`Kp?{iAYY1v1Lis59V#7`_XbIfi0xJ{zOH+6&=E=5
z@{Q?jv2^y=d9%ymuu;~^W>hbK)gT(6?alA<>?GxP%$ebKZ0?O5BYde%%+@krpDu9R
zg@nj*2(|RXj1XsKl#iw#i<EP!dm%0}LjkL-FOKS4*3g>*3K9UC0)0z3xA^DbqGFNX
z-E}W0^#9aT#)sVC%u(a%Wh&PaICz^Y8@-gQVm3N_!`~y4e?1`xsVLo`br)kJgrudn
zF~O{`p2j?6VN9SLKPk^&Q-t+ceSKDv`xiU!i(S?lzsB89lGPo~7p*h30=yWayt#@D
zvviRAwsb>;8X~>v6)<fsQ{R+2UkyZL!Z@lzhr@MDAG|Fo+?iBGPxZgsp;aY}YV?@s
zM;aj7^kNEp87o>^oxfy%os9g1M3))879-Wb*5-JRA3&y|Et3oz-vE!KEuH9CaTAAi
zr2k9$vA>WYgBKP@_bpX=ZBJQq>t9N!$?!MmJYhTOxzyBtiCBfGe&(@#!Br|pYXVIj
z2Iz0BCT|BCN9q-f$(qHOwG!0AouC#zBthghrdkhejx=i{#k57e#`JpU*2GA>nB{lY
zWEYM(&v#|1j|8|Z?$*4Yn_}xk&2IAIZ=CDa?~zq2%41=1bVa{S(dw1yD@o4>iVEWL
zvX9GBz0h`7#y%M*(iDK-GT-^_k%@Anrz`pD_kVsVSK*gr=z0N<cA@7RPDwJBemF5T
zPyn3M#E8CoD-86v7DR>)2GvtlU=QLhLJViL=!bzE)IdhuQJ}v0=W*d#SmeX5fSk<p
zHI~=7$jWH9UX6?RbzL>RDRR^J%p~NKCl)M^CB^Fwnvd+U<A1hQa<qN`Noe^=xLC6P
z!tMHDZ;_-tU(P=9<E_UC^TJZr&c2EB%@nCg%p;;o&|pkwx$PYmNRQ;lh;{|Kf*s4Q
z4M;B(aDOg-v(K`o+bSH4E`p~O{e?;avKXNASo=rs)y46Pn40EiR9s^Jq4)h^40Z!1
zHP=y?t!vf{Dz?8!Ki)Ag>Kwk7D{o}w^7iCPF}?2-%55EUgb)=2YkaR%v7OE&vIT;&
z5NE62414RsVr+Nd@;zA33IAH5xkNs0HI6K42y1*0kaxXagQ!7VK2+4xbD<E%1$4J3
z=Ny(qznky!FW2x}5PiZnPN5Hp6jHmw3&xf{u^$wymOs-F_qJLwtSAum{Gia4K`+^I
zriJiHj}zCwyMtXP69v_Vo0^-hPQ@D}+1-cgh9N8;KUcfSWjW*oT_`75uKx<0@R%+b
zDl94~<-J>DkLU4r2gM_W8;lYVg!v|;#M5rvHMu3|G=hliuZ2)TzGg87zW3;D!U0Xy
zsmXdS-s*J`PCKzn?L=Yvoho<VZCOqI!c!oKOetu<m^%wwU|$X?cWR2*yXP;JkZx!e
zO*tf3^2tN0oZ*$^hCOb;3C3Y%rgD$-g!AISR&FwS1RwpG2p_8w#F**1?**L;Fj@D1
zgEk}d2&54gC$j%qsQY8;8_^v^h)SZADGHZ43nfN$j-exo5x*z-c=#OW&L4!e17V1!
ziJ&<7@DJ-@9phQ8PjiAa0H^OHZz4q(gD)T@^^uDcAvqM0i3fY`j2fvYW8UFwPH{46
z*;EyoBIL{E>Xy+r$`GJ60yJuF_1Q=>=D};sH<414-lvUgw%FzL1K~@e!d`&rN4rvi
zX%?}Csd<?=BM}S%bwNXdlO;YkVA~)?w%@5MvH3rEnsDs~9n<#t2%G0aOb@D#_+nFI
z8NsZ&w-J9Zv3K1I_(iWu4}yv7+4q!O(9zi~_5`>3hoV+eJw?{9UM|nKosWqo&cfSM
z=PeM5t%+x~ji+)K?^3yf!xGYC?v8qNfFJWLzBAlaI?QBAxq-tYWVYhN6Q`tIKt@i`
z^}aPx@y$4%a24uFnnmQI!!k!{5W8t85xv&pemT)Hj`A76lnO3*3YGAe;Q^QdG9z4H
ztaY&utWHxHzdY%F>f*d&tM^i$4In*`OIhs5c>5LR;Tj{3NsiuzUQsT~S`m2U9;rUq
ze^5IRj>a9bmP^{IKIP2IdQ}2`4U{MCueoC8<dX<MT;Txo_!wXw4_1dU+CCPhX%kDb
z%H7>sIU5(eI$$JPc9^sVqjp<FUqJO_yiBMmGW}zyK&!75cbI27{h!@unaFpi@MB=f
zQ&7T|bJ1yT&M9F*+@6fTpP_JJEae(m^8gaewKv(tV&eD#72)%KRx1sTO{Uo*2&=+1
zWky<OY#*S%JB!djw(zap!6kPvW4I7_De2(WA^bBcxc@<oL2>?8Y@=2F(oQD?{y1gl
z*;~E1lHOImUiqh*!Jb@Wf1T&%cia8@*U~H!EG-*y9c_LK^eCflM<sKV8rvU`Iw^}$
z?<(n<6w>bP?NCh_^Wfs+WwK2y2>c3gf-J*||M&m>BSh;adzxFHK|Xrzq;r*l8cw0}
z!Vf&Ev&#$NNz#7+IPpIAx?QoAa6-iH5L+<p5_anR;{ID6PSyVvb4n>2;{Rz7uhz;g
z2g+&VhH!BxtWDwW5**yFb))F{JM`z*2Uyk}TDK%!A3*Y;+~m(lPw`>Abqohu23SV`
zcEmNX7Z<%#m^H*%EXrY>pk@jV;8>>;dylbYm|`!u@<LXpsd#cOWM24yi%a>k8NR)q
ze(@zO)WN$HEp$FI7;z}~dtnnb7r(9W@T`yoXlFKt&g1^1Eq+uIx+T{9S0#TTtMFO1
zSG{<BG=uW(pb8_18>qUx1F)lr#h<j~W@pM5h%XiSZd|{p=xig~h2|^D#9EFYpS@&}
z5RP0!gMk)Y3-M^KXrC(RKM$-SK%$iT|4Dpq!K3RosK!rk=C90AdCf+#ogD9}k>7N<
z3wtadto%Qkt}?0#_U&(@ySux)ySoHQK^QPfI;3NC*8u4dkWdf=QIHylgfu86-AH${
zcRv5~o&z8F#&+NLRlm4$j?wnka$l96N@OUee=YcopBZAEpMu48>QtvkCy%ih=a(E_
zC~`%i`Rn96JOry=mDBb!)0TkJ=*LedrGx!$b(QzWMnjrz^sxEkp9s>fVyDgmL_44^
z8-~`t<VoO^R-3Y?Ca&wrh#y9XF)mtpJJu;geN21$aF6;FtBSU<;#Nr_bb$~PVM0X?
z$MIVbKl477x69?%Jfy3MZ`~Kv7Nr_Tw0$l&c_xorgk-gS<#oD_e4+mu1zFKMSG$>t
zrS3=m)+#A4uS6Ed8puf+8=n{Lr1i_YAvoIjX`y3i#BhZJTDZcNiy$A<a>OE~WIR$u
zTlC$gO&&NoXTX$#My~_{eB4Y|ic$%Hn&da(f_aG*`Duz8KoFFvtxyMSazwPJtZcHR
zQUe&Glsgaa{-+44?Z+<K4V9{Y7poe)GOWT;e664FoNwtB9X97&^(8tH8iqQAZ5}5X
z;>g@d!jJS9zHc1()LH5*rbb|-w)PVMv~sr%2&7KcV&g<q+#%N9UPWvoA8P-1tDWDY
z7oKXM{(I=<P4j?$)KKNPZO7c1Quvp3+~*WxhB=->XvL&E2CM5+>hD*uu2}P$zemuh
z86uHJYUE^PPnmBLeYs6DD*tP_pxbTnyT}c{9EnBM3#UExtNqxqZl!EeqdJC67m^he
z66ou}6&0*0hIRf-p_!O=KS$&nk4^A>za*6=*mw#dJ6n~ETK6Few2Rr{*V$AhWz)BM
zC2y{7xrbD1*Od*8;?LoMx2Q;Xss2z;Jxs;UW}@55WPr7>%0mLNx!U$Y_u2Mt1dQX$
zMmr15nr~qH9eHtXjDNougc2tj0?Sw_TM+gt$O~?|%^2x#c2q}nJ3^-(+#hIoTRL}O
z1CMOxnfI1d4j<AteM3gbwOi#Cj{higHx$G$cW-~Ogrv(QEg(K23l%Ml4m?*5#+*10
z9}65WH#iyAcXCEo1_7Dg_6RE}q=Ll2Kh1zS<T7|CFfFzgeT>!CLj`;l8HROhnRkpk
zru;yPi0{0{PyZl${-iKYdil!YP2pmy40u}T2ULqO@)<EE(>&=^^Ss;L_R@g|EG(kE
z&_?`EQ(=>qW<0ARCs278-9E{?XRh^{`A?I>rTAeGqILX~G&PA$Z*t_Z?W3=j?{-nX
zMuF5$cu9NcF(14zpE_UICyoR#2kOANIR^}MklY9va+rKEgf8xGGf~)L2`5S=@s=$!
zQ|BWqDybdd4b~m@9-mb|y2Skou{)_KgA4n{)aYF5dNt-Wp1&LnZ51844~Qg=&`ZcC
zB6v)3E-7@%XJ^x6{T{H|G4>>iD;w>MvWoj3Qz_;-Od+}Efj|=g{@SCSV2aVY8^w!-
z&P*=8_%phtJx?ZpJ)}b>--fb`@tM|elVkrSNCFzB*}Opd&gS3j?uhsU!X~~>>7{se
zTT>|asp}v9^+KX_DqM4%29|5}4DK<mvtN6Ny&v%JR7hTtZ0sVBYv%ZO#>mwG=n_v3
z=%+cP1_=FQ7k&-;O1zTCLLTjmTJ@a6t%el1g4#6#9~EYSUqc&MB<PShQX>f=ik{Tr
zySRkT9E_!HS&S42j^fI8w`W1kawqW?S>*_Rxe|0wweZ9WC=S?W1G7v50GfnlbGKN6
z<Gq0*!E7V*ns}#p(DuYvWZV`1>W3lRjSRJ18bAW$m)96gjLFCmAfvUbX`+(~AIHbm
zqTlx=TLo?I#&=gj!<M=RVZNf4w}M8Q)tpKHCTu>mJifRAC2Jb*1Hb&BZF)DZRqG&s
z>e}Jh#IN#{5=hcYj-?n?H$K49X$(LTKW8+(7?TA%!3-MjDaJG)Bixy+KLTn=C*~ZG
zU%@>weZ=BC`ue_;d{MShf)&&^GB>iFNUqsYRp(Fq;>j532vt5EBF(GB0Zf7oI!I>i
z`Ya5;cRn=lN1xq8hFjxl2^GVw2$<ah{3FnmaqK`Z!OsZAaNtS15g2XZx&|O-1aM8)
zrAXnk7OsBsOOaJ!okYxPX(s(6!RFhN6ED7QcCpL=AXm2-t?!&vC-Af9pyGahF?Xv^
z->@)baY>XSrIxA?(=eyuY>jCn;zc*7E_g2@XRS6w{NGI=K~_>aejkDBb^2e=<qV|*
zCGy;GdGWDOI!$LTDJ0tqp!2FbZq_?(J4h#;p1Cwu`WZ(1{@eK&PNllxiPmf)!CwHy
zx|uT|Qr;eoLd(Fo01&mj0ogsQ^uS+)?n6U}zIW5#P@fQRF%*}l@bQB<%d6oA6Wvhv
zdpTw@N~VI?K#yK24PE08)_cR$m3HNzN%N%?$y<B2tYxM%2}cy<Wf(-reJMJz;wWSk
z=giR2H3+ObQ>aCTopMMV<V@kd`2G$SV;|s3==W^q7IQR2;hOeAqD#~Xt8kS1O(FKe
zdwn4f`gPV0x&s5as8eub5)AEM5xK}=`s9RN{`?r`c+0S_1aYz(LJRu-S``I=)Jo(K
z;t5;7vUMfD0yTmWs($h>5^vIXn6yaldp4Wa74?O)(=Hen0*|hIS7T$s5-q<_ML^R*
z>4F#PvW1bJ(_e=Q$~UFH0L*vU4r(_7`QT!*$Kn_Z?q$Q;SZhI89Ivhr&^*QyzmYV5
z2y=$HM_8EHXby7!vp=B?e}3JqPa60oUavVmtBes9*}r2d_#~`QiiE16G=%zAj2|Ax
zR}X+M*r&Z#rJoj!l+7SQmc3-3b69+IrZdrQ>BeL=4A1EP4eP#}4;_@O%hIMB(|()v
zj=7LnDOP;w;LyBdOHild01AJ}JQG?+=rec~rrHcUVO5dU>d&bguh*mJS1e=uGg^qT
z;ZVH-FupNhrm&+j9Bq91;#Q777Z$g3;5&v4Yd@vXErs&qq!pARfSo-wL7O$_i!@{7
zI2j=^I{K6+qK#;iO#UuPMfXJO{-!S<Uq5mAm#T3CbVs_*-5)lr-s0E7ys(IWdzl*j
zd(~P5B$@X`Aq_apU%(jV3W9uVHxkd#`x5A0;?dd+iSUWc&x#~gEK40=<xH5w@H*}n
z9g4sF#jQ>Iua&z_&?7I5tCLwmZRTTteEjW5vZC6N6F6{dzI~au@_n#~-!%0ddaIDm
z>`?>L_2iu6NOQ*DOjluHW`{Xk4!|nL6BJTiLJ1jWFF`p{>{lSgg^8Wm{Zy~yv?pZ9
z9Om%;{3T;q8R|kSZK3CjX+i1cTRxl1R&@0(aopStX2%=WoXy+F;0-9=t`4ocq6lr3
zcQng$CdU|L<~8KzpwJw7TES7L(W(A#hDz_ql&loMGzSu;w7?x_4Ub&>kuTGFmV*cP
zh?YAIKa#T04G5I~gZ)=(_hbAv&?EseII906znt^mC$0A4#(UYe%+0#SXd30Hw}8R6
zk=W9X0pCn2Iy5cg+dk!D#UvcIyW{4B0NFu{={GEkAss++kXcmciya;N__tSpgso(F
zJ^V#w=9sy6M<`DIO*d{Rh|493Rcefp1QCJp2F0w>6`BM)lpRkh?}qHu14G%4y3Wlb
zQ0H)Dkz@-OMN4Ae7sw4J1@Y2xLP)na#XDf@2x<xHM|5leb^gRDkGy0>_Sl5`Vg%`=
zH6M%h^OMj%^Bv+uAhei@N-g|fV><QS?JgIgb<p68smsV$LA@dAo2>y=BXO>~$5c=_
z?G9}^5}wIEv^RYYjk1FPyND)1$LRMlS7y&h|J`i+5ylq|l6YFIzQye_BEP<R+V*Y~
z84?H*e(^zAQ8oWzMw~C?h0R(>f}8mBUcKm%PtYVWW5g<r9kd!k3G9)tIDdEGBngBX
zFu&{1%~*sw^3X(HqjamYgNRo7`lM#?NP(B|c}<BUOfL!}{yNT}^w=o{qtjTzl4*-=
zdgjcpr@y3EuhEsslOlCU^uvCr(oOxG2*l1)@w0QxfWnV6bgMwn7OU3gf^8h_UF`0x
z0h-LAh0x7#r${t%?Nm{+F`ygQsR=%zO<GOmKQc7TiO#tB|G5Bmb*NH1fZsyx_9%d4
zV*UBdH192J^n!Ybut34MzsF8jR`*s|3y=9$P}0}eEa!P_gJB!Wkf$bYUl!Bsx_{v2
zCb@^O64eyO4y9r0H7@JXJofr*Ad!#3vX>mOmES!s8A=1Ql14x&L11od>ASDM^JjuG
zJ@V-(_im6nGR@!0LaP|}ta&E@fpSRR8Y<u~6ajIgipUL#Q~*W*J1Fw!1QLVfQ=F~j
z01rtAZjP0lE(0#a7f&3mPg9sB(`?ozpGEm0``7lHVQ)xQ;8iq%Y^f@vM7jiX8kuUz
z!>Fv58`i5R(1xrjxZ|6HJF$pQWhrSBIUQcu-t?oYqPm@fr~)JoMYUl5(|2z*dOlwU
zGz$Oy7qv<49GG5<*!jM+D9|s9bQYg0ikTL#cJh<@;hJE-Hl-8zt~vje4eIy-azWs0
zo@7^|>Q&;MJTkBf_%qCK$mw4&O#}bv=kucfCKqdUTN=$S^6h>)@%VH`KkYJ5WWt?|
zGstk{C>Fi}=M%dBn;D+uIxX4p_Ka3oJ^>wwoe@q<2FGnAfVzx~oMiV<>I1-Hrw-iq
zspPBO#dw=I`%3JhFO=&!vCM@A17c2c+#q|#0TIwBIK3mH=}p$*nD<s6jVLr1=`+!q
z_Y%CqXnnNivkLHHFv41$|FPK5MO)ihn9iR!)O9&6IR_ooFlNqwyJHb3b>D;;7yA<d
z`R}SQ)5<Leum~g7%I))5ZaE>Xd>HIMO8fgi<BMW=RYIRIhF<V=lyJN-5E`4+zwH31
z{qYxZB8HK*Q3jHj@1X*%M>ku^*rNZ-g0@Uk6cv<hA43wvA4ZKE@2%GFv*=as=(nhm
z7erIyZ9Tr$#a!eClaGaqS`yb%Y92-pvim1*^He7s=w`SLabz-~b&Br9FyFW!c-EA#
zCJKC_K9O2y@w3kd_)I_FXxOA9$L6bR^lFbKu(-*N9XH=oj(9^hpHBe#G47+MC-Qp>
zutXZbVEc28{S|$BP@8LlXVY&bfAx<8#0Jb%g#74%1Td^S!nEfRd=y7+)kP5-IyfPm
zuT*h;*rrgxQeGcoxlFl1{hTv8D<fujfJ%ygk4gHYD(Hy5l9qw>zCNesVSK)Ce5r84
z?YFIM<@@!oigVx~hHiK1($^*L+=|4DQ8Y5^R=HJ8309>Wcbr~zB*AR|D~?v#bM;ha
z-gsOgyoEXHze%p0&RG4*l3?Cz<+Hv1o$%o4HU$BCMW;4|juYz=7+G9W8gWs{94SMV
zFJ7<mfCsMSkovywSO7PWd`dQLHwKcs$IbSK^dGIt7!XEi9ti=Bpt3U?A)%K5&~dLf
zwZb4@GfSW@h69~Oo{wxw0E5?HQ2o$_yY5X?Bv|_C&EgL~6atdcP8-|~f+5mC^s;2w
zs`OaIKH(*s!m7x0g0@;<#z<5;fB~b+?Oof1d53}l$NO|lV(_cM3G*|uf8PwK4)s-o
zAH6Bd)h($}+zOD^yQH0Q?blLUtD-(jI+Gqci*;{k6hpv>o#!gZx&i6t|6P+13^Fpo
z@Isk$$fvF=hx`+wz)&Qcx~2)K$`Z#szOEhr9WhS%17Z!4ePcC^NZ;Cu8OX}qhL5uz
zl&W-CS&-ZE=9wU;aqlx#hf0}OrB_RPtkL}?!=2dBaCf4Si++_h`@I#`0kV_hRh-)7
z>#A8RS}SqywXmLn#l4TRpQpIW_xELu9Z~d{@3btJcxkgt?W$7=As5@Y^^RdB=wI+i
z;xQ74v}{qCc0Ex_kjnq~6HJ^YpaKH^fd7Cy6(^<d)zaP9z>r=LI&?NJ|KPk7W>XKl
zS>{3=K3L9Q$45^Q!==M!5^LsMib1XcxK^dnDIc7C$`wI?n36BaTB10A^{ARgr~Gt%
zsVe5_iCUO<QpVYq_OBPgTh#s%48dE8x;StJeQ|WweV&;llgnRp>CSJrDj4=*j)Kwk
z5@pPO)*$4yzu8G)F;eBMpa$yuG*pUi(Tx!EqL^F1NH@@eNXh|_yLG93IJN10fQ^1N
z7z)E;zyFv|EUBw~ZRWw0J$~qdcBO4rkIcD|zNxQMY7(^@W<g4qSW-v-Lf`xU3@P1t
z&xfb`F*s^J*{R7>dE`_iGuDE2`&5@_KHZ?OZpPv>J=m~UKL?@pPwDQE?InuvWL+kw
zQGIgaF?7MMHeNKHn4Q_yj=t7C<DLSC(=?XBX1pVY6}nFe9yZE#g4&+YJ}$1Pj^b8g
zmr-bWwPLwZrig=0&;BXug!?}kX?2p`?p={c|9!A3?rFIvUU!Ho+$}ULah1hfjAgxS
zD@r#8Tct@p%49;}!mF2l1YWy&`m0=x4UAfE0|;&KOp`^WKBC^^2y(y7SNhjnCy3KN
z&lXC~x(s{Sd&Itq#ky`C<@|M2h5Eka4p$+ePZ9yVLW!rPdh1GJ?V208R7)B1WXC(F
zL1-^6w4HdHK)+uKy<Uf49)PWfx9U6=fA%IK>tyKxV<xt^coOSSA!LV%125opCzqwT
z<^P5Xm7}88k^hHItEc1a(If~c6(KyPzw-AjM;UwmaEvT2JASm1^1UK{EgBdZbV*B%
zV0$_o^*nF!6>z%z#nBq>BoKQ}Sm>IGc&wk&nR6SOK*rhPe{uHHII(KMJ9;f;OvbA^
zw4voO(!{CYKfawob&>Qq-9UNuP*1Y29N$UHNVRV6VX;VBV(f^WuFMHwZWLafxhW^n
z_&n)CniF_WBS`&hjPU)BO3t7dvYlS{KAU^sY5=-Fjk8y{d|bBg<GzV3V&icnuk^&|
z1JVT>hp5vo7jvplo$9qN;SJrXUZhgdokT_kR&Wd#TIssR5|L~?VNkRgA>Urb#gBAn
zf@Q~z*H_K4=fw1$z=~94w4%zA+*%wa@eg1ovq0v=F9VQGw!a8ZL#of28>pQ&uj%R4
zh$-0x9CePJZhGD*2ZvEKeZj^Wp@ay@58Xa4fk|j6{_}h`UDLd>PN=(OmuXLD_DvB}
z4_=1YLC`d(Y<Yu-+(akhP=UnnLI?<LQiX_aLU3}cT-{i{9{{@0WVIiPzL2(}-cZh|
zYU!});gSP5{l#ryVJh>;Sjun89DMRg)*0sPPu++vZT4xMdA)fI1%7V&@pPC(-!k{0
zC3~{>Gvla#8k?+Pr0DCULP5D+y|)bKSOG8!r4%oo6CD@^h#VAPQA0awn2u{1X$<Hq
ze`GWIhjD7rB_}q#!UlCSc9GJqUj~xUjqa;Qm<J6zvI0<yOuSL^{J8|ahC)NW7;Mh<
zxzOTzzer0AQoYO)!z|mR{;xBs(O|XbVA8@QcUB}G7T)tJjjOYL`7KgHMD)#P`%2L1
zSIr;QI)ME6K4H4nu_W3xO1LPB0xCW8vZrB`ad+3a3%;AM(6>-WL98-UR_z;$)qAY_
zxTx^5Bp93IltLn#{|CL#li;Ri+!Oir=-Bp5v@`8>j=N6ydqc|i!i-gT2jh_eBrBA&
zII?NFN9-tmhQtq84A%x_-s|<UyVNVUW1qSewa-7YJ^2=4hW7T37o#WUJ93?V*`yS<
zHThp>2Gl3Fwj-ZaXO*;}Zf=O5F8JfmY}jeN;}FQ^5IrJ-i^|s#<2)Pnu#yWxlQ?m<
zr%c~_=jDNtH5+^4t`f{Js$NOz&!PmVkOOAZ{sXf%=_#0Jw@G)rNt*m4ek*d?R}e4V
z&xPo)1brD3@q+h8F#B}MX$pdcPhOTMGNBDR1tiBZP}Tk&_t7OhkQ=<-rv|?{kD)<5
z0g!<aMf)&=`93fb+>jpVAa6eQ9Q>>Q3n?*NAH-YCXuM}Fzn3SZAhdZy0X%BX!(J_q
zv9}REopfb$$g^(jMloe9Rlzvn%_E#!K`B*D;juIiy`c!bkwy0Ba4XivbIZO-Zwz8d
zFD|rY`?c!)V$v4`VbP{E##hXIp(OtjA+{rNME7epXMQb74-H+d@N%mOCc$bW(XLPj
zI^sMo4f(7k&Hl}j!obZyuHv`q?7zBy{=g{OpkPcs6;^dTHf>Y`H?<m;mg6RR{_Bf;
z(0TmbnWj=&Cg#IXYvmU+YQ3dce|C6`X*e}z)*<Hl2Qc^^C|jwt=0r4GR8fdgZ}i?=
z9qO!p_GAWG#f=opYJy^EP-{@{F><c;2%^ydPf|l&2+bdgNXaKxE~C1_n}N91#Ye`|
zeA#=XswlN@MPVV~)h@S7k1BKr{xPYPGvBlkx%L~-0mSb8VEze9asb>G8^6d1$K5>d
z#JHjI!ZD&-6((|-$r>ubZ%)i*!G>u#yrudl9jlD%J3t&NiZpR1v6((`ct_S<_Zarq
zg3R$czR&+gPtho1>*NWls?9r^I9xRQ1&0J=b^$&a?yx991ETnV*1r^6gya6vL)w1C
zCRF?$eL7a&rt_rCTpS|&LYbG<c-{ih*ya*o8On1Q{I-<N${v&1wpkE9KEkbwc@Q?(
z8l6!2j-d9H<m{~l)|IuP^fRMGaiD27?SOXHG7pJ=w=EdK=v0!dT4digVD~E0m-wnd
z*hj@|O23I0tC^scfg|K|@!9fD>s@O5?d~s|bBgCIla>h5Ren{_?sWnR%Nw0K$Op$H
z<cF*W6VtG9A+0H}>au+}`2B6vS?}!b&yfja%z~cPSAcv}*Z_|PSs@PtSuv61h?&pw
z{dmRi{QQ~oga2gU$_5Zw5?G;G?Sy(DozIeBLPZCN<jhhJqnr-{-Qq0KNy70rstx#H
zow$D3YssJ34UE1-U-OTQSqZ&TQ*+}+d~_)T0h!6m34;ca4}3cSGe9Z2jMU!V&l(|f
zG^nsNDa%&&Fx88RcdYL!4A0OtTyI|Xx$eIf`doK}EFT}c8QNS{@*5GTC;galwUwyN
zOGy7tx)VPm>cEl@;%X;FVR^L<HmeH$<dY;yMKla(;V80ky1a`+c}hQVBl4bbiDV)J
zZ@DJ6yMN9h=WxV2<s;)vDDk`<9JYwGJnYW-1pYdmj1hqaUt$y{93)3x7ndctLoC5s
zSZ98BX~=@3+WREXoYYWRvneLm`F8;U<xciz*NGQ~XUG?&E#3$^MQNQ!04qxu=viEE
zbbW|UXZx9|gX)me;!VWoROjxPV6cE!R%~hCpqir;JLrn2@0LNC8~C%Bg9Bv)DqV{T
z?ld*3LPvH3@ty!1x}MtwT`em9;oZWX$BWe$k5|ffj<+kWLLz&Goi`~&FD2Oa1^*Po
z+OJVsUS|8r4Vx53-B5@LuPrv5G<fP&y`>Q9Vo-`gt|($)4qmBMtc21VthCkxKXQZ{
z1%f2AyM&OwzR1xs5T2AhvL3V4U_SI<kd2t9JP#><LRbq@@y?G2i8D*Lcuc&*%GW@H
z`a_3Ab&t#3OoWy+pu<F3RcwRbUdb}g8)F+sKl0^OT|8kWl|z6?;5cR7n2ulN{jWvG
zrhB4zZ*k3Qe!ajttXM3{-Ifz#FTO^Ud<}gdK8sIhF5`GQQA$o#w;vN0A5Nm_3v#BF
z+2B>fJvwl3R(G1P)n{2Txe>~>TWR34F-jf_XHT8E&RnI`5B0z<&4=rw6N%dyUoGon
zK%+0~hOyBYJ*%P!hB<lJZk<Wp+_%hkayDBs6c1@K>@qM#HBAG+?guVO8TSe@@+k?g
z%OXb*r+o$$i>%FI#F1e7&8rSmRYU~W>8My?Mz{?B4o62Cf5!8KEUjzG)6vT-meK2^
zMLM+$D;FUR=n7~cQNTq&oi&mlRUYfA?pL~%VK%RQKjygEeM3ri&fvAP@cQ*WD2@3%
zWr>QG0z919t<L)%vBKa=6jYg&@K8pPnTBd^YhQ^yfZiV!1a$Fv0ELk}yh7SMdPdh9
zT<`kO{ksSe7}6SEUDBOr0GwEI$3M<ozcj6f8QBhEh1yZz3>Zh0o;c8PCZesp+>P9o
z)j}QUw-}NVS{l0D^8)Wdr_aFr8$HZ5%PWu>{lfPHdjg+*3!C+oHy-OnKCCL6j|oij
zl;_m#QxaDD96_tELhky@I4j~7B7Qh%^=^!gjrXB<@0a`ko|*V64NegaqiLt~PI=&R
z%vHPMQ-+*{`cNLX<av{Qex&@AwV!xJD5yCBmO3#5H_UV)RYzvX3Q{<ZZ0F7*p?mzd
z$qJ9bCvlva<K#?jsa$ct`drfL@pYWB?=J^8v=ZF%2wEr8hM;eya33&l#2Rw~xmW~X
zjOQwiU-uqdE}mhPT{My>LW;jnMDaIg;PkLfOE|OrFc+Hm$~^Bt#DnKGLpj7vgRc|W
zhIL;eG>I-eq;3l>Br0w*w_g6SND(>Ceh@LO_+4<Ix-#9-2?Yzps~QE;vDsQj=_I8G
zR3S)g6g;stUYMPYWV#03je$UTnll^tQ<m((RHt}vnc*Y_j}0k?$oAUYVw2s}$X$uM
z+3xaDrO)Z6yZGiWltdK;LmOM@<lA$tp#0_!6y7CB__(H6vFUVqQhV!W{It{NNh0Dk
z*%?)dnrLP@Db~{nU<~6vOpwJJn>$HZ9EpoSP<c$8kNxQN6^T1u5EcH`@i!*6E;<R!
zy&9p)z|&M#B7<c#0<g;66yF2+nPniwV}vBAC7^b+J<CA8)5y2Ns{)Zc@NkTjpTKlN
z<jqO<hzmFTf)}0u!;zd^7>@jqpza|z2s(G;!Ly+kpKLmN+p#TLuD-GbIJsl2=GS<S
zo@94Lhw>DH-!8!3CCr2>EAdN+E58pF{)FC&=dJ0cF0AAT6p?89{2>{(;dJh}L{w$`
z1S2q69ysi|byEl;YAEi=&UGT2^nII?1bu{yv*-~$ht^j#EMK~Ug_kO|4b`I-A{>~)
zLJDX#?mqN2vAtcp%fTA&uIrije8$@}=N}>=%F-17F*<^oxYGJYPF~_XB=krFy&a_$
zt(H7b&CkOOzXPQe^I!@79ITAYKf}liHkCwH+ar^)N|I0_oxsWqzY{s-sYKs@+wg?A
z(@lZYff?O?T>J0{M-%~mG5V%TM1Hl9v@I_^r~EDoeH2)ASQWX*d1eV(#JeJZj~Brs
zg>gS;D;o;O*q3>#&;lNF-1!ul->QU-dM8Y#k5b|aw_c0~X<kLtjlJMF;YsGIZXj9z
z9r*z(`=D18sG!!A$?Mg3+lj?F-_!9e*PI@6{T$%;{9t8pN=sq=xsH5Kx<Fg{9|~ly
zLRmx&UGB{hmwzUTJ!iBU)`w9g12w6*mq%t+7moY~1GX@p%NezZme&atQg0OAR^mA=
zUkJh!iM;(8)Z`UrgN~KR*CbFPJcR}#835^^Lc-f0=C7a)fEARGHkJ@1YFN(6(wijr
zEvH64R1BsUhKnU=0eQ@I*=A04d#Ri<ba45(nQBi06_!G%EWa^7=oIni*&WcF^NZk*
zUMn7~Y6|l_`*%jH(Udrje8hlOf>YZcM$jAVa6JEaKS6i|^?+yDpTWr)g{G*DU%ev;
z7Q;zkWRjF*(OutA3EyWF+G+JpM~JTP9v&a4!HYG2N75u<0vkLW*7#RD`!I0YDW5NG
zll8<v6(W}~7IzO8QzTic=pB46o9ixkA+9wg(e&5J6DpYq;Ii)w`&jKQ?aCiRj`zl9
zz+7`KrX?I&N(=GCOQy&V?_)Sh(6iQf$<&6#7O|QWaYKuQsU?Mlr-}VN(fOI=Z;;z^
z#=dp1tlgpu$`}c2zOzXgiKqWs-rb(!JRX;-2_C?U3&}_tJ0(AVjY1F|M%Id7i&>Kr
zVL=YEL7#~ZLk;w;`AwIW<s=SU+w&DqzBrzDE$_ibI)&2qy<vPbXP5aCxIX*7Q}bOh
z&u@QR$(&&_zchhWy1r|?f+W_1LqB@YQ5b61exiY?Y)TCoCkhz;B9QjBYnDZhWN}&5
zw$ccpj}uUs_r#KBk2LYMktEq;$l$#dwcBc*OV*St)m|$V0aW4K0Ow5LW_y-+f<!km
zs}*1dRC<%!nzt1|0J1*z=k~($X_gzs-~Ro?V^bSW5g{v^PWN)`$YT|_$=*rZmLDLb
z_+5cVBI0_ki8a1X`0+2AWps8+1m$G4-!TReco^em^e>uQd+N=73>v*bF=`4dctpJM
zKJzEcLDQ=Si_@h#^rx$p=o;Rbd+AWg)gCRkKt;u(N?&two7le$8=7F_(Os}pKv+HS
zV~qUMdZ2l_F8{dV44j&k(XjNSPr9<w&@h|H9&g>bPNOMGk$nt@@XxyU7r-3AF5nwc
zV#88X{*@?!^sGhE$&H{9Fi@%YDQ~+QDw$<(A-90?JT5IW=`%$=tKtY?8<+|w>(gQ7
z2ZV#yJn<X_PH-9xfThav({pM8Y{bt>F*?|BEPPwLg(UGy+Thxi4+>O#rlOG&)Dx~2
z$Ww!xG683AHIIw566h|8@uKeR!@sE#4meXx9Iz^N#xAlDY$xFs=5*gXuJ6#W4Zd(^
znJn`nhs;|lgl$=u&Q@l>wPl2?UBd^8UxsAHK=(AKc1>54$G$ya`qb-k$gfT@(6c@h
z#51?GW$>wY6FbEVo{pKvzOvUW|9hXYO#Vq%`3@y?aF3}GJrJ1kvlb6dc|cxF!OOnv
zLSySpQ4bTFqwYIh;rJc_N_EI>4$hBHzyOnn7<86~%<$WuBMO|GO=%Nz680?RkK}Fe
zLrb2zX$m7TvUCu0!m7szY{=SPR~%oZ)4)LM{p$5i(X0yVX$ymhH!<4Jxqf|>qT?1J
znWa}SYQ_T=KR@PyBzB*=VD5O1d#t;Lt-UaYts|6@O|7_MO-8W&!F5~dwSdoZgkdj9
zNMr~c0&XH2U`rr>JN#ixUQLVa`)9-#+{~;(oeq&E%GQjhoe=U*LWi2st)`Vyhp$*E
zGn(gp_Ii^x0_;XUDzNxVhzpgW-E>L4M2^FO-K=+Q7OO1atLnhkpH!=oo4KF}ot370
z@>PVyiS`K9aK_YC{zd0U70oYyv*5?X&GU7k@gF(34+SZiR;gB7BH0*o!jW0ClIA5s
zC)Q4EN@2^YZ%qLuvz1P-+k{Xnd?8L!au!8nV7+L-<>ub{&QZ%i%`M4ND{L--)-CQ6
z1<Fd=JP|-1P$i^-LY-yq9ljEa+Of@)t|Whai2Ds)0qF?YE#HjVJ=w1h=+a!-o=jAQ
zqJuGf6vAH8mrk}6PhZ*a07xB-KcRCG81TBIltH1_$}{|a#Rx!qo@E=%NgIhM63Kj}
zQhxJz136yvq6}mWpL9x2coXq5O7khjxAc<eYz(qeJAS6nh5SX+Drz_QaIP>bh|%*T
z;6(+?-&N}p867hv?Ugtde9jV|AM8V-BR=Z1)2FZPgp8xDGOYmufxG2>`%KDMFV~cV
zhk3V)>y4K$Y>NJL6%40$56yGn3wAc=DtE-_up@!sys@^IwE33$A--5IBQfd!sH>UD
zY{fss&^v>Ds)~`L&^N6_764WwI6T=Ag;ct5A8v&tWiJEnQ5(Z}`IDM~PLbk5;S@~7
z??I6qdL1(^<qzY)*H2CiW3;Z(IgpeH=4B}xv<C*9ZuLsakRCU{J=&_ACfozta2}iC
zEbLa$i)U^pGj@ONyKa3Hi-ecszBmK3g6Q|3p^YJpBuLHRR63tx(_JppekNBRhj1lG
zVN(6bjLE-^=s>?x87yMNH1Qju1EJ@kT8U5!WzHSeTHms0X1r&}xYgse{Cyj-O)2?m
zO+r&)2zh&Ju=B@DqfuM0h*{9G;aHUdsuHjz1v0HCoqoTa+s6vg2U{?8WP?;9DT5*?
zW+)gCnDOZwH?2|`@PGzx5B7iO0=Gy;S^_)#YDCjCeVl)!o$#g-;HxXE7o>b%d_$h>
zFD}d!1(7XE1j<G^qqh^Sl${51oV?HEP0H6K|KYYH$>5_>6h|V<S>0I~^`=ScVDFBx
z+v8huDI1SM#3U}GT4joZE~Yt}KhFvO=Kb1K^x%mu${4nhvv_k1AFud_w$Q2LpyOZC
zX&USklvo>=pAD@Sm4XZjyT^4TtyB%l@;<^<(_;Thkz%jyW33D9SAPDS4C}S~mYR-4
zA3~@BNxN9#(Vnlz9LH@~MoCJ=k(yqvq^aB|_CxnUX2#PX9YiG1jK`0fm@<^?z!Ofp
zw-fj-f_Qk5V9IR4NdQ*A1(wrFLlWY6`{A}c4!a17T1+qRViDgu<8U|OJpaaS#in78
z*J%dX=d}P2P`e$GCCFyY#<j|)^)>KMfxvo8)vep{1>q;nS_D^LkQjb#m5$Y)Q4DsA
zTHd+LeBp3p0#$XBg1Mde_wCvI$alC6U^GcS96R>#Z&IRta7e(_=OTVaIe5IqvS0v?
zJ%3RT4q*8W+>h*>&&FMoL@4O$3YV&o2tR$x3#(7)92ERQDXCz)la)`Y#SdPE#v;^U
z*tqL8rBGCqR~g6shF}6>!i9aedp#H&I^FyVf8DAfJ%aoBBY;?l28suEVmbeiyvu_}
zn#^$9t&6dYe|6l`QF7<)D8LQOE_TLR&FiJ^>v<A~xi*Od#?43MS3N<bqza=gE#r+o
zKc&ZxgA<4S8TBJe3Eb^$tC=O7Y|z}%aOT867TO(K^8Ur&;8SFm&FNdgl~Wa4#qNla
zL(cO}rd}8-Sq4hq8!QfaeqHW4rY2sl!wbZpt7O}}g^DGhtWlTWt?`?T`9AeEV`JWY
zaOOfLUEVL>95`>V;>eq~yyymG<i{hr#0up;StO;?3_(XorCx$+AZ*`r059T<ft31A
z?GuWAN=T2ln3wg7u)MH$gUq(Hj}pb6O4R9(`XQ*ecs$7s<5L`23j8{<poa-f_BSEh
zzu^k3wC6y<#4C#-;X26jBc2!A>8v|oE*ACIEa(p#%gSAb&lOEZO6T;sxPYwz-M~QZ
zgo%-U2##x@%l~r$W+nAsn@%lkvAIQt;pv7u;|<#Yrm4@LevIwcF)kEIzVTNj_y!!p
zMKX{9hYOOe1poZ$vpDu6ETbSxD=~vJq_}I9n06clO!y;m2^H<}yPb73lXu7DazU;c
z2pg&<^pfa0dvd1z%U%&mKft64qS=kTg3KhX;!5S>^%H~Y!3ZAKJ%b?gah~k~9z56H
z-w*KB;SM1+^J#sjjk>lB_zAI5RN6ThaBc02PW52=G1E8Wb&opB>`QVxoXnCJNZJM2
z2G!iaqm$qY{c6yl_k9BpdcbPH4j5`gPn3S+`;8^E|26!+Czxm`b~w>-ZGc#jmfiDP
zS{P5}9o5+Rz_`=W_kF1b%v+iG#v5t6Lt1V$P4Vxn(R|!*KhzXmO%Z?74vneV2pkUT
z2;UWXa2~Mo|M5o3pM@&E!4=(K-mWtt`}ezqjg@zpZ3=9zlBqof*bCajE&@E|9Qo%=
z_yTK#5b6St)lYP#YZxKPyILLSC1YM&TKWcn>^cbQ)yf-|+1=W=fAscMv<H?_5DtUP
zQWjEDXA82JvKcjoAN%azmA|qW;b#cDd~PG;J?}Y#@_?$v?KO|P-C!iD>yYgo1t?XA
zDQgp#$OCCrU_4(~X79wK@jy<%W0ASxpDqnGZ(WLVy%t+M8`n&^uf>Je2-hWDG~UZg
zFbQJ=op66^s;_1<;mbrp66i6kg@qAg0m%M&k9}YkF%u2h4@a%lKNet7NLXCKXV%}B
z*s9($_z98R$cvGb$YpxGW5*RNQC$=}dxVskO##Lbhwuu`mAB?t?>ZC0L&#ar=8uUQ
zvQKF9B1V2W`{412Bivb!$sn0xm+wbj-Bwdu#vA7sr7k|-Pq6+&w5rha;GID+<nq5W
z9N++;*fuB&S6h9{{4rW^FcIG@53%h6ct2g6`TW9M#Mp)`&Noq7rG-|x#+99y-aO4t
znHiN4&@;<3yuFoIex9_;#a&+Am{3w~_f5sY?;?A0QY4b(@OV3N>FdeFBxNVkXj<k#
zkEDU7W|M7(G47n#>4W+~KXNb+A(D!0Y+i|8{D6(U_vy9h-GKK`8)dV|(lP?L2B)1+
zClGsK*hS&O<EingVzH9|%k(!>aFJR<-|x;LZPIDY|4x_n`_&3L<Gi;SbG~X~s@}^2
zC<YXxr>0&LZc!GIw(DdT%d2R&gGvXF82IxdgzZFcy$}&DwlbVEK&?->sxlZ_U$QCP
zq6ER8i-zpbp+(V{i$7I|9+cPf7EP#W*>b#717e5TcC6X!x*b$Wo(Au;E%b%Zt>-P+
zveiP^oIIL)HOHg$oC2*s{v$z4NOe4YeNH%l@PRcHhhzfwEaB$~-@94#>_r9_HfZ}?
zBpI}#DC-E62z*K(NKS&K@6gOYaeU>V`Xdg}e-k)Pz+w`|F(m0ved4>`J#7POmRCXf
zsZ}?^Fs=*&4s9wKfYPaJHAlVcFtFmRkYukz06o3(l)L_)2s{9ULrNA%iF;23)8JEq
zU-zUqYMhI}XBep_ccvsNcrx|i8y@U4oay%L$R&y&mzaY?vs`N4Zv<XS>YI=ea=L-1
zVxIrXzQ>Bw-aDu-t8Xv5qF3l?On-LcW6IE#6+&P#MXgN!))TOw3aR4yN!S?98tp_c
z%!J3RpR%NR1x6!gr6kDCowk0dHr&eiRK@=43;O9X`t;O}$yWOz9j8nW*XvLYSZyd_
z;BuvUp<p!!K-(0icD3>iXY>P6ei=j*g@6U<$+za;I2C$nw`0q-{%lTB6h$BfvU!*;
zHH`B>Xwll{h9O_9_wD=7iAJ9#>p0%t9hen!)5`S9rrSN>m`eukH$51WCdRAqOTUe0
zR&3&qF9veffbkTX;lg)G`yx*AOB{2{+L`PWg3G2m<i9Yhb`|lt@yB53qxibm-pgPj
zsnP-HYYW%`(&ZJdtomZp3L-bX)8*ROpTxUvyUMkCYkb39)G~QHxX64evnGugNBL+#
zx$_ETJnbqlF8-1K^bU~<4|yG)MO}vkOI7^%3kHKBkFx8Eg=7xUzBVn430d_mG~b1%
z9algZz#?T-MH{Y7*!HGE6SB)ET=-b@js)PEah#&JG`dh?V6!Bh8xOA(Vi|hF6n90J
zu%<8qcbNPG%{5x#oYY5m{0KHVyp;U}VIHU=o2KJ>%VVV*(3<pSRjwmqMybjqKks#6
zDfmWqh9L~esjb@x1PoDHBE12hm$V><o3S~(#g=ydv6;tbI4CB+&4=LiAn7>su&ikk
zsI1x{g38{BaMbZ=JaR7WBDqZ<w-8OnB(!C4<xkOw)T^yq1IFYAqrTH`47X`*tdkop
zxj=-bs3kIW%H`?)EaPJ7o1E;#!aDdRZ5aV<iSgBgQ25f6ZXUi(<)w>&oqmb^xr|<Q
zqi^(lN-JL>#*?wFK~UH;?9^&G(4R%eHG{X6xl;Qr<`C<lr|f=;<n|d+3!*K40rA-u
zPuQ%R#Jh}RjIMVO@jw9Q*&8fjke|URI%GQgcv{r@F#|3sVB3(}I1AQ8HK@$(T?WL;
z0UuHivp=S;TE?=XmQWY;dBOs6j(2$bFboYspPS~^!&NlPZlhv$UPK<#+@?8Eh7g^&
zuZN{8p!<5pp=V2&-{LRnp3&QoYb}Qi-_F0<T%g^(T#TE)OxT<tt^6xS#Z=j}Bk5D>
z4oJge6*xp~Ps$9^G8`ed)S_bNj-ES`_21cu9qgzY2}&j2x$Mw9-;hIj48xalN1XEk
zp2cLFM&w80y+XaAKY1;@L8*6&3a8#Ir;F!(7B(1bsPD)03~Hr_HWm`HdmonA=f<?D
z?56fv&oU!r(Q&gEm#9<nWj-N{CR7F`Uvz|}z<mYpfl&7kAi|8vNKOzXf3D5`nwjJq
zyOXY-=IwKmygZr=K1C1MdFJ@4-QBC=ZtaeZMjY>lZtQJTuKqSIeU&@>k-g(qv)0PL
z@u9(n1Zq#z1=0<xKjH$;Za!fzyeK~PK*LX&tf_<Tm`@&rn!&1r3F{|P`@O$NQ4vl<
z*%JKBn0yTGOhPwCD>pAt8OM>j7`p@9Ren~&iIu$G!Sd^II8GZY0fa5^#<2X|wR?g+
z0#kU$MT0t{IDW*JJM&(jGBCBKQ*Y`|-yIa8eq_ks4mH4^4Jv~W=>J{_EkB>0A1rT6
z+Z+v6+Mbs2I#MB$b#v6Z#ox|3c2b{b`a4V@-1$CIOKqT*XSa-(wwD5X4F}^i9)^k@
zoT>Iy0e>Y}{o%mvmryWCSi_->gXej7cKMY;z_a^O(*ww=x~jI94-6#@<(ldDnSE7m
zMr?Ds@Lt;S*;XvF74hfEaUhZn3w2=%eQV17XQ4)ty1$+7(^l%>Hwb~3U8AZt&r{0v
zWOo=KU*kZ9d(;uRi0(-2Y?-P*II@(e;lvB|=2;z`kx@LE!iqK|T$<2Bp-i0B>sRv4
ziB&fk3-(J8b`N9Yj+i@?^?!UFn{dpR@3!uIYRTgJA#%{=zSIB+Rqm@9Q<I-)FCG}x
zB>fq`Ygrnr`e7weH!wHjNlWoP3gB4%>EmR1QcZsz4;P852X*HlUZsLc01{~S*#qn=
z(FSZKjA+m_d2wsV3n(<f{YBoh3!Vflygo4_Mqa|=(j)OX#I4w-Of%I$VoBPMw&p-9
zsB<3PyV^qcD(ML&i*X3`RS;JDZ=BS>7iRs^`d2-4xP+)GAQ;+3>8Fl<_;hrK2{JpM
zVItLYNyUkxpKZqAdgs!a#2Y&qxZOf@xQ$aCCwENr*3lP6Cxkg_D7D7H<(^y6PP!Y`
z$#H4WgN@*Ly4x>p@%52C?u^(YtIg%-iHjeSh}y-REyAIo=B#Gzub;{u)c6ZK65l=7
z7w%$CSfLoz=EL+eoSIfK^rUQ%LAWX%%xA&b1L*f9i;ozsD`d1<w5`jo6eN(xEbhj|
zOch8*{7g4{<WN~Yx-`;7^L$l3g-qjK8`?R-{lvvk*c`VdV(`q8sBguIpR}LGqv~5*
z?0b{GmGM6rK|TIpf;@SuS<_7liYpOPhBLZ)<{;Amz|D8Uq9T{xE{q!_t*slQS(>C`
z$cdIFg5~l0`X`39IP<I+;lDSJG*uH@wIP(s<%sioAcm~nxvw%9Nq&S>dV#<=&~Cco
zSy0~zufyuGkpOYzQ3G(aIhqFOnvpoaD8jJrLY`q2;l*jb`g)Zf>>9cS-y3<r?xSpX
z_-)NPyf)`mbfizB^tIep{4v)o(pc<3ogFnk@}umGyU7_~r@%5rw;$u1_}xajx@a2)
zGcpYa#VjDQ!5l1+P%_+)e9Os_L^kNwAnJjFqYuqqcY}oZ^%^idTqB3iCi=&-x81lm
zGL_uZ4g@E)eQj|t2Fdy0=b5Z|T7!0-Fr^uHrcg(g9g~Mh%c_PoSUyM12h5GjpRUw2
zWsDPg-T?x!k&I8z2qQS7u48h3wwXcQ1-y~eSXU24;ec!Ad7s-GIph0Mnwm!gDP&zt
zZi7r!`-$CVY8kqhAy5<hmiMFDpmmMR7I{E?%{8*+%%b1wmuk-w`IK8O@B5bgy|w(%
zP4Q!g8^L<b3i`q8dyE$rf2_!Tizu}Gd6nN|-QnoBFy|u%#t#~F$iCp-F51t!Akk$?
zS6{eQk5+j*!0=C=q_zN!dROj|RZsMY3&wt7z6Y<@S1R)<8lT0do#y~Bh0$H`IaSel
z`k<Xa{`L!QEe8M`B{XhL<B9ilrH~7El#7A!NjYiD06c?#_Jt(FJ3YoXgEw8|Er|uG
z9Ki8@Z!)3z`)`H9qz4A#ttcxHoC(6zvXsy)k3Wzvh3_q>t3e*7Lbb27ss>J%e#i;!
z%e-kr%{-v~sak>3R%JuqEB0S{Rz8f-{zL|^*YtmQ7ImPcD!pUE+Sn!2ei>0=p2?0}
zb~u?x6=B_XYnYrZcFe{taz#6pA~1QYKy>@@hxpSWs0+pM_imas)v)9j4_RWuJXK?5
z&b>#bnDLH5hmG*@;^tZJ<qL6c*xv_@6IBo?D)Of{(4gg+_r&_jOYQqe|DV?^Io~wS
z8i&3vAQlz>{Ws4u-zTCaC@`MN`x1XT?;eVMQXC8JL^qV(^$Rz}I5h~#NgjXs{oQOw
z?^!-Zp6T=kDhhL;>7m`x+i>ftbX4-Fg!6ET#(iFY@s6SSWbD1-_#2G(i9f3@27s$i
zlg@vorRSJe6%umAae<|Ptn?-yi})O=AIz*S4zz}e!#&PQp_U(e`@C67-PiYXna{Hm
z6IR@a4S#1uBBzG~OM?N+Im|v>iVq$iXS#VzJP5lj+kL|^gsYGI!;eInG2|(c4-<oX
zX-w=dRAUTDpuknIm(>v>YP@F5yz?oC$#tsIxcCv~9AiljsIIeI%`j3}ik-NyTGnPb
zUxj2gH&EV|pcTyuYi0``$Sx;BUf`-@_@L}o#F}Y4_4v&2^1l7>IqzsA_QK;!-s_0X
zm+b|$_|nHFM;+sQnxx<Qg6gMxa21pqa=pIho=k#3@%`Sfu5)jXz8RcW&~I-0<eVeD
z%2$%#{>>s|q<UzMbgh>}#rL^GXiKeMF#Oe{*hGBkPxxe(-SJ~cH*_-TZArW_>nMA$
zdwXis6_Z8Uy$L$cF*5KO>CzxEf^KE@K~IdaY6P^1!JBkM?dPHabP3x{wP+io`S<*8
z_OD|v;hHMK!)j5C%v~Nu@)NwluzpZ_e&so>Wt1pRnqE03OB1g^)j|5Re%7>l8Est=
zmz@xE4N=NxX~q?%cbH^0qwlLeSp5E_G=H~pzFbS-eQptwZ%NS?68yNsw9ZvtU6WA4
zTviU7aiPmD{-f(aB5dQHGfJ<z{V~4iaMOHr)8q`h1;N$x1~|(o9wFW&p7svsQ>+2V
zpgsT4o{l4-6uBuDJE|IdN%Q9SH44R%!%WV7zM)8F|D(j&U`4{TFS*h0<A5<9)!+iM
zpRSgcy4PpiFWiXamMsGZo1tG9s(ow1ur<C_`7#=fzt!a90F~?tvnLo+d4R4Q41~-0
z09`5_=M0<eWn&25p22`yG$&sYs}GsC?K~Cxcc+Vo*+da*&#b#i@@Gq*M(R<;?89OS
zk7*7Dr{917oM}J?AYj3n%Evv_ae)3#m=1El9>_@%dIJRO5DpanssH27X|?v?hVg{!
zh`?NNq9$(TWT$T3eL~!{TX=5Nw)xI2Atg#~evLq`=;Xv0OM+*}KR30O!_!Jvr;zGH
zo?Xr>qQ<Sy@nJDdeSud@G0&^yVz%EU!M>We5#%u9WGoZuCiqIDFs36<+~19e8P5;R
zOZ?Gn@JDeBWbZSJ_z~s$ftxD#PC;f}1>;!+7K{Jn{N3N(Drz1oJ+rJ11}%vb@ELpP
z7j&1JP`uMbJWs$1U|RZbw8<GdhfWK2?X!A~B7G77b)MCY$Wg&1v)PE`;O7Ow<;o#m
za{G3!Bae4Yomo>Jn+5yI<QK(b$9S58Wkf^oUA;?u?<1rVRqc)bDji06Ae{w_Kgn0F
zUmYfFL@=KsUD_*qJ!wt|^W^sEZN*n5rG_KF+*Gpc^Hc_?(+%Xto_1Fh_WB|<xF;kR
zNRTF#)x4x1Twe-q%~&pL&zV@-=kRKc_4ahPL+*Yu7H2-MAkhSie@}hw4rFf6ne+Mr
zykX5_l1fMFCN@7mj#mTVMykg_bTIFh#7&Db^r8N{fXd24Y^0n|zaNgu7}9tm&U(&I
zHBLE6qc7_uKeg*7ij=x-?W@T*yIk<a({kHZUIceH(@iDRlvm~yl>j!0ULi!1E}3z9
z$i~J`aS@(P31P2w)?RAZ1F(pT6hWBo2`W=d1?nH(!?pVnAg98GaoYjt3XMfJM^~>O
zlC$8Ocw9UI2*uHTBYO1Q2TP+nnB0i(8JaO2_^waE>(9Y&s}~W2nRS@YR-5x$FKsbx
zYNDIc0BAkPAI7I?{3=X(*9!b_2|)^)vVFDfA9RD4Uf4s@vT0V$?t^0csN_=&gZoJN
zT3h4-!muYe6M-yRse?gbkD95Ph=LZ)bsJODvv(vMl$MDP-RvyaD?^HVg+(u~OuBq6
zcyid>jk8fXDWB!D#0dP-a&f{jIUJLdR1JOhg`{nqUPV56?$G$}as3SAc}F@+ggx&-
zWH)F!Z{Z)aDr5Rep_COW4m!pNPt`?bPGDaRXVv+DIRulU5&A;JDx)Kl?WEAaNQK!9
zlbtR7c+whvRf?t8w_hs8uASWHj5emst}C)@Bh9(nu?%#QDaS{84Q3NnTq#uK39b{G
z??VX2G{_T&3Fo0AncX6)I`Jf*5@eSrS;v82P`qW2(5z9WL(q@+->9$P^3rZjq{2(@
z7CbG}By&7ewn3jnQLbywZM4Yrho0D-P~blB(9C1g2%^8@&6{<jylufa$yHrHda>M%
z$pQ?G=8#h%f!HVR7w_*v+><{M$z}>Jdtuse#t3YpV=s=fH-NlQP=(GwUr1_MCDBAf
zuq0>>YLyRaF`u3<O@ezexN(Yft<1`Q;fOk<>&C&5YbVk%Em5s7J7{YCbl&`$NgRB1
z1C{_p%%ewzlAqhkqU00&AA9c^)zsGg3vWQ+Sde3-gGvI1E=n)ag8~U9O7EaF=>h_V
z7HlYjAUmNql_njjQo~UcLzj+JA%GBRp@u;6ZqIq1|Gl5zG48k@-gkds03+FJueoRW
z&EK4Btq1W6O$~fxP2+jbGhYSsF^&!qmDtKz>(JQ*s9}BSd|rY`e$Cs2!Wl#K(T0nN
zW3yY@YuD>9F1lyG&5j1IAK<b!gwGtNvpr{up1<``@%-Nrj@Kon;$EuXbUgZsvLFNY
z#HB{a7H&X0x;+jp1nw2EGFTtT-s#=H)bsS=Fhr=?$LCmjq<)<_7V+yd`bk(~rKzY@
zZVx}r1}_xCvu?XM7PXC#`^uc~((HRFrEB%->E0!s;a~b~7UY;R!q3*fyEc)jp2bi@
zTugd5@KtMo;d*8$Zsb$$*BM{cJ9s^n$27MS8te*JOhXQbG(Bmcs8ls2eY`6B^qRP%
z3K{-u{p9xUA6UJg67?}ju}n#Ol?df?<^sRMuV46BO{#{*Ejf2LEJbgj=gunE3H42*
zpNnt4j_JJ-f5gdsLiM8BCaZb#ZUBR{y#9D&$3zr7Ecr^Rr4<(8*m6o2IQJ=+j@F#r
zJZ&+q_aoQm{oU@fXU$h!4Cm}z%rbIniXT^eltCY9K*${Xp_WZI;`JH1Q`}HT$Lrc9
zw@Qb{4DstLKRR4(?7MWRT0Dy5WW8^NgWgYf;!C=#w4>G!@!iGd3-#lNG8Cu1MVq)d
zAk^j!Z8Pz8#1D2AcgCn-$bL;$*JNeEa=J_ZYnX}$WmMT}@A!$g3A!0Af5&e#{R)|y
zH;~5<;{6CHb^(id>|_7O9(8h651rvUDJCbrSjTimzlbLWRGoSd^4#oBfQq$bxzIZ1
zgT~Q`W^K*uF4d<~`a4-=E?}qfU8Xl2s$Ad5C;m<%<5(iDA9?lJ;!sqw3!9}p%U7@T
zr-srMckk*AH9@_5jbcz*0B!8>>*LuAFM0AW%+2C6+7fgVuJTpv20Y#rZn+%Y)Zg^v
z;H;y>T}o<K=C#UBsH%EE+@)**&r+<ElWdnYPS-{6!FA{C1kY5R!TivD`AoM}Z-oEj
z30Xzfn#Dt>6ouY4BlEvq*i+QKKN_u^ZZ{BFh_0D7DKorL)z8Dyd}?4X=u+){<YhFJ
zgWV@Y<#G)qWjWbcis!OVoaQ-naqZc9@P!BZhvM2sjp~QgI|m|#U+j$&JydSlqCDU&
z4-9KGI~3vw*B=8#FO{V@e+wSpwmlX6eEU~ggvr$pV}cHS=T-%-!6-wb!{4k^z?BNB
zQs|4_C(TZR9N~)KxrnEd@vpz${qwq$Z>%T^XnS}{`%&`y)g>KP?pa60+Vv0>>P|!2
zShR1|^K*k~Q3;$;r_ZB~zqhy^NSXS3(>S5vqNSeU7&8<Emf}1w?V(eB1sU!HsI>dL
zL)P-?;th$N<J-8HEzzR~60iGo-p$PJlRw!K5)%8*et5JHw{vqw!_fBIC9$(vPND0E
z-J{bAWMZdPGO##4S%nY8vbm3F+1t^fBB3x}23wW!r2h-wiO=_WIM+GV5X(!iizT1P
zK9Xw1vjb?|m3{7WTsUG`iZ8n6hb~!=^|<AOlgdSjr%t)6Tw)m63CAg?*|atDh^N&*
zyMR&poN0(|5+cyPBX$(z->)fG|E5t?zO;g3UlOWr+V(-m=*r$;08c0X)P(YErI=y-
z2+m>c{r78eNYZZmvpQ^F#&}5UXs*G<1i^~m<&E3JpMPDs9`a3E`&S-iQ}!C`Irel*
z`FE6E+*u2@U+s@LruuouY1}7!)1Nq1aVI%`QZM6-pHzw6XT}2J&(6lky?+RC=I13Y
zT{-WPM*b!Haum&L=g=~nu4Vze=B&%F_lpgQ5Q(rv#_T}b`F_3&oYOD?rq~nviHs1b
zr;kOT_H5|bmPBYDCX-9ouD3B^x~G{QMup)*Y=pUIYblTQB~?2I-l7DTGdg5ykvGd<
zO}42aY$KKRKI)!os%RZ9=M71(XH@Ld1y_41jd^JXqc6Y0)ma*I3{EC^wmlbY6<`x`
zgabFBQncj>jlsby!{;oE2A}(vK6*m8Kg4Y?up1{|P#%@~=xpJU)B+jhYqq|k3UKLx
z(+b9xQiJTd?6!TcM?_9Zy@rCh_Y3?sRr&LWAy!uQr<-T@4zu&D4_(Va?H-3+HoawJ
zQ2&dGp6-LukN)YF7;kpszG&E~OjbLN{!1T+1q}G(J%gF&+P36A{2Gqn9l3o&%w6W&
z(uvx|I?1?G*XRFS89H=7CVc9IV>~hc%p!5xLE~#!EkBCzPWutyth;s9pwqvUd<Qqx
zC3Bts?nU)yok5B$Wp{66zjcC2zXQnkqc)<0!F2BF4N>JO`KqYLkYMG5$@dSx?t{*9
zY2!5$6d(NRYHnyEJ3QFQCCZm|?K<}d(nOABS;iwLRN}el8SHbmep}@>7j>=1X1V8D
z>%3G;T6))GsmvzbI>znRhZV@96=G-LUL$wK3PQYgos|;b%}NCxt+)0}EN1sd*3#nS
za><CvdE6t)wUuw5zXn*vD%%#Fa2-~L8Sd`NRooh6EQmYyGRBA(uya<3dw<JK?VK(@
z_27gBo6Eb{=_<$1%`)2aETtKN0vo~cz!z;%MT@9U!BOo;K6k$fy<Wx^q&bV0)qaIF
z;W<fGmTFsS)F$!yeP?*Ex3nBPYalEU@ITrFVb^1~VH>}AKgoPk(&4?y4-vL%KhPIX
zU+6xMymwl^V{Ov>mY+yl6mq8cHtB5>5;!!C)X|i?f$FxXWYST>z|ZUQWN{c)hU__^
zdt<`)`q?1v@>|^k0>^DdYeHasn%J1bc#}K2w+J@Sv6ct@*O&%l-2@y<>G?Vq;Ul*H
zd64#tP}e%D<%0$*>#-)7-7RRXmvzhyI`dGUS{s-0hgiDYn)<DSoj~%chgOcGs-UID
z4QS6EOfPViX(9M>^p?YAIlkmiy=a<8#NeekH61s>->Vtk9*cn%PC+*#A-}vnP&$7-
z#1{JF76-Jm#-eD<l}!F4QSis?EvDIO>4eK7i`S>p7av|cFk0Rp=(!kDz`@VE*QCW~
zd)7G^KpVcoY~#+#;ok)ogU~WM5rKhoT*BN=!RN5gp$6D1UwP>Eam5E1=pXH`W3*Up
z-I<jFk2hUMUZ8Dq)6#30G=AXX)tCYGR+Q}{LV(DPm`&^FnVvN_pWZh&haTy)^Dfl*
z3jIRZ|Gk|E)CfCFwgz7Ru`9OboFf^EyB@L1VlRGquOa?eWBzY_l?(O}Zus{rk<pWB
zZ8FdTFX-DpK;xbH&?CTURv%XXe|CWJYId1sz7LxEQ85wFv^ym!-jr2^&%9}*EBdF^
zxzi5;@TSzc=dUAPTzI`^v9_=(VYk{e)08O5xPIdn+Y399X*Ac30@P2$6X5(|sc#%l
z{@V)x%K;}lNGh%ejNQ`PWtJh^Dh-{P6$@LznOGOI6osj#;mHk4inZnxr*SwB1uvLE
z&d2XtfeYd)m$Hd}d{7|;8AtZ9$?-KSudtR&%^(5E^9+pmx=2KVh4DI(KZr&9XBUbg
zNAky~H$6YV=u@j<J_gi=B8$?e>K8~%Jh}H$=r-B}mHFK=h^~C(quKI=WniorW=Am0
zMeyl46e)q0KoEA?d0Bf=>hUg;KR=h~ND@LY)%4G<<EUOqPCY}^FNAhIZOnd__Y89>
zBowr@>Iiaqxs`0I^?jXaSZC-+E!L~h@u021?1hXiT~c<Db256=RlU$2oPpn#1i@K)
z?EG9-?#_dIjh}BvEcfGL6<^jWA1k-r>xn<=W}&%`yW(xg+!+d=l|Bw(*8U&8n_2lD
zU#ta_t)KzvhI_=aSYaPuix$I0BnE$Fiv7S4#jM-={vi63RmH@IqjG-<fHuvoJE3J*
z5kM}Z6$d11K!UpBQou58jIsm!w8AbG!paP(b*7Rhfb#GE{Xpky9cP)i{_Ki-vpNa#
z_m@9)F)+fY#i8Sf2(hfw03LJX9Ir6+a5+F~=nV7zE7m_Ewf619i~s{pL`2@R(^rdA
zmus6kQhl$1NpNfsxM3_YAcOLAZqGTcCaxkyLLpB1c%rC<uq+WROO}Wqe==Ddwx2Tr
zxlh}M)IHb&Y&4ms@uu5=`UOUhEs8Mi{v(2v&pA!L%nC$e_U&!80ewg|s=O?~Cpe=_
zSQN;6d;;kI^ao&ZLoKh0v8ufCsQ3gwAeMCm`W_2lmbHtt`TgfEQs@fMvUHO9Ooy%I
zelH-WaRn%McR{fC{-qe(001ptqo9WffyCg_{nx1<z(1F<KHI;)6kp>9cx$9lD_+u5
zDy5Ig&n?`g4x(1vq}6m<avF-Zf{j7?7t=q_!WfumAx5Qa!Rg`-MibR9oy9)QZ`vnC
zN#cv>NV3hPE5C2kh3a;fW*&Sv^6^m(jP`7(!v6|(X)}vw;g96|tDa2^<f2cjll<>@
zUISb8kygXB<hbLI85!g~-!o52zMl@b(mzuHTc}B7`s&XC76b_rlrUyvyU2?f*{)wy
zQcX<3{#ikm{@DUlrht&X7^HhbLvWSLB)3_aKUx&iQ^{CVjY!b7kwWO3*|9GCzX)V@
zl(5>$qkkv+f8uBP>kP0FLArrj(epC}kr=;TR{ClxIS90gCwIe`ndA>8g3Nf-<kT<d
z7>Oyvij6;ave=8TvCyi&FYLlWvTMQ%cbh+_XD?T*t=1oWaB5jU5MlWQ+s1H01}VKQ
zSjPc;G5D1w45{QjE<pdKka4rq5^MoZD??3YIQh5p^uEtH*fH?4tvr}OG34i>OAV<%
zS=w;EmCN-oTsk?di+TnX$oG*^=}_HsYxTm9be1>_*T|Ggs?Ym{ckR<t{;R%r{xLlP
zv8&8Kc*Wl1RG%~PK8S^{3V@v99p>hr%>2h5fCF^?P&>~@Owf?XU@`GsB_kb{);E<J
zdbgctJjl3tBzRk&MpZ-3b-Z8LoFBoBt85$XWSxPgS>qAAz7Q=H?SEy&sOQ^XnHf~N
zMox7%^Mst|9p?Q%nc`FBGq@Rkz{2)_8!J7F9Lru4raTw@cGzgQ*M+9U@r$n0RU>Le
z2`{kRNoJwr6ydtdRgVr_Kf7mUhouIT**RIJXaAk(7kjos2aSob|5?0)J9I2<1S!<Y
zBq_!0W_{p*OlAbB(aKz$<_D4`pf^);)HB4qo#(4}7u83pLxAH~0}i}w9GSHhT0e@L
zh&Zs%q-pgf)-#JOy8Q2lRx<HCmS-34mglBtKW&sGu`jN^iQh%CMq++t9iL&61+@29
zmnqOp4F1<l3`Q>JDr^*$1lXh9Qf!LpV5rAKul$@)mL?~5dQbiB^z6&V;bllXxfoU}
zR#$+X=#oxW*3nb=Dab|rNtl|D3t#9yQEu}eN!^+KEh%2v_qbrlhW=InPPsPt@<c>c
z;=o~XS~SdDyiSu8S5uP(l4l5f^HU(Xo#3z$55p`9de8|eU&^eeE4{rkLsYtdPl|tD
z1r0i;E(Af1B_w+|Ee<BDcUV}~YndV_-Cfgt%RGUsfnq3-YSqT}MOca;9sE5@YxjMO
z4I40fejCFLd8-c`4sX4XE39(9vHH}SAu17GN5nR1wcmWJ0fts$4<_enxd@fdjB}(_
zMfy&p&_rd6tCxFbklj3^a_Q@_RY7=eE{#Emy4~$og4049HsH|IER=VQf0c!f%Mbr6
zUpI_oS`9_YyO<}+zsJUmAh}{=c9Dy*oqhjk4R4&N=r$%=RFr{vEIJFl*1yKYv45c@
ze{3P_Ppah}<157PqxN)Od1jFvyJ}|B0@SW_h*(Zv7Vh(8mN>5<DUrUhHFJDjj543Z
zjO<*#`?2+f1Ur5ayJ=oVT*R#t^n55t($1b@o*D*PcB2mZwf3`)m$F0@40Lesd?cLx
ztV;TFOjv_j?n1};l!Tlk4f7~(lTtb~YR3;OX)51Sn^F&Y+)Mmp6HCKl>kU0>G;T|-
zLW&s!{;Qa^)&D4_*t7pCW<=DT;{X&dC@4b<2RgiKLksjY^z%nc&gHStCwtXJCar?(
z&>kDFiPd&1b$aIIOeu?H|0)IR)P-$1dB~GM%2_(4Px`hyS0XqCS(*NSbO-&fyoCNo
zcZ?y}82~a9$kj(Zk;T~NJS-!LU`Ed1_%)WMPZlT%Sk<Gc30%phVR8>8C;UPlHQNwe
zo)lzi92KOW=-ODYqNHcj5`$xd1NBJsot=~hsG0wN8_(}wU6uZy@#F%U34K=kz1@4Z
zpiQsmSc7GdQ#;RueIAj;EuV1iRzA*bJ<P}|7numo7B74l@~!xfb#p6GGW%4vR}dvb
zL1LrT)srgxo`{8a3p&Q(u9(R$?|-{XvH!SBafUebcD|jW=Rbo&1Ei<Mb87@yS{LLh
z<q==U!0~u0d>z`X(dX`veIi*J#klA^Vwpj9W{YJ~UB>EAcB7384+*#38@O&WEz46E
z?oPpm6~;721nJAD6+;sIL3;#MS!V-kHiDBHQoTZCW_BSxxqb3qJxTmmPk#PydQypL
zZ<U)ux#c#h<_6q*6Y87~S5YH{Au7|d68`=a6P!sP;}I-#sW9BD5*$q`;NMY`2+@%n
zX^!>@e&5QOX6>_rU9|*%lUy6A@v0}YbjanP{$q?wDt=r*O!R-X=l?`}`@efn-r6bM
zic2JKD82)7@(|4H6kZp}8wi}00ptd2FcqIZ9AyDw2)o=srOy5{M78%gJCOG=5*8#*
zGX#CAhNjZUGcbE}h4ZmF=tzyY|9noth3?A(B)cytaRUy4zcCsFkCKgy0G5Ad+rqz{
z484#Xw4W8|1IJo}MBS91`GysB*f5)o%wec`{I!l^n?(}26J~GVXAk=0)$2zWe!@ib
z`HUWzf=aGG?mJ<;wwpvrTPTJY>4fCLHI@jGlmEqokHas)zaq1-&7vf0qelm_3Y5td
zvI;8DXL6&cYj2Xhr6X;wHjjKX6xu`-U8g&ee)5-yF0~CcnM|&5VKU`ktK+_V(gQ)H
zK6M*X6jhS{vvoE$;J0wjt&0FIH&<DwhhT9O&#_+55I5I@zj%NwLtGJqFNJdLo{7gn
zH);jtHsM3DeX?(ej(d`6iw}j9@?dnSl{QwO=C6Iuh*41nTg$N}Tc9zpr6$)7P%o^U
zvg_Kb<PfBXA(~jfHz$)*H#wP^j%kif?!DYBfK=3fB+FNE@Zl(+pB~5cgLgO}vGz(2
z5BErr;~v|=ag=ZtsI*qaLND{l#Mq!%SRyq4^FCc7IDprb?n^j^OxFe;4E2_zFDpH&
z09uOwgQ4ZvY6OB<0M+16EX*KuPsCGUSrNTOAY$+w^qjv=FmE5a0s#2=hp?nPeZebV
zn-R1l+ya(S&<Di+w-4xDx@(j<xE6wnMW;Bu^LhqbHK1<C;ND#TI*<I88>Af`PgH=2
z&m$x1M8w>M>1>p@_2(XXEex3{e{9Xw_z|NNWtTvH4MV1oV@>x%fvb?LEMbgK{Eml!
za^6}v)*oH8uVxhaqa;t2qz5*hgZDD+KzlS*1QlyQZ8l~6&sY9~zW?tU*Z)8I|7!^p
z$4;bj0&;3lRs{eSvxTrAG1@0z9vDdz6|P1y$B*O5^N>HF;!!pB<V%4D)Jzz&FIXzi
zFB?ro;v1z&-}EH5?#9)f78e`+@Vt%XXd_N#!7(?h?$zymFF;ObU$&J=`+f5W1@arz
z5-d){UnP^^=V&3jw(EU1sDC$~ZL*=~!sJfp5TnQD&ZrKyGxx!*vy8AB5dn_Ipt=P|
z`0sli{P$krkf{MJ{2O`*rCVw*-yfs9Ft{eqpuG9Zy*-+mgcJW(dOlVNC~J_|N)Dd+
zRyOmu*%TtbQoX;J`)MZ9Q;zZtO>qDRt+;jjedt7;WELp@*iG(O3rT5+Z@8Rmey8VX
zKQ4RcBYW&EIL)ZCCm>_zz9g`%PhEoH2;}xoqKa*I*Pgqef3D_G=yqL7$A<dPTLF``
z;Kuv$9-FUn3`HOrv(Ez^UjoheQi$)oU*2S@Qv=M$&`$?6KsDpZwerIuK4k=(X7fBU
zN<)c+asH83GGXMMZud>f+|q_I^b(b+M#w7P4V?F$u|qo07p%s5tfqOJ(*9P%tE*X;
z0m_OSKE39_|IDpGEwzV%hd2Ac7aZ%d>iC^mlFept=NYoUBLpA^w@a-{OM}q5@?#$v
zVkS9<5cQ423%9`9hO??XD0iu|nm%|e>>3zbt$A?fP?_Dwo>9#eM-ES%o(CPvR<oK;
zu5qDboETjXYFVdW5e}1s<0{qYvI{O@3bA!38>%$v33{Qcem!kyP5%*SY?m3_dZOon
zn%#aD29mEs7oH8Qc3*(GN|i6S!g##qn#$1BAy`tt5Yqyk`}bf;z=F)Zm^QJ|_tlMc
zl}+dhEf4I3S$qFrsnqUD#PYJht~IA8^8gkiPWubo5m#{$$O`fT7itrSb2s1FL=7s>
zqXkET!-7D218N5<R+st`#(ZUTX|p9_M`kCqn0%S1qL0WG;U>=DnPJXA@eu%wb0*lz
z4HqqITq~D%H!<?&`QzkOg5PC^V}@6mE>U7G((RH~HD*bWvft_BIhJ49BP+9_GoB~Z
zH&wY_ow2M_QzP`Pc+8Bur8FSp;Mq58A4n9Zd+vs9&l_3>$5ASFNxQ){ftqHN%Spy(
zBizIo(aa7w1M-vl@!srLceJ;QhicS}{0`L~yskdr%J9;;Lcxq_^KEyJukB4P9IJmb
z3D0GHyYt`|M0)U!!Lb-h<*Gi(dfXVUl0UjVWlb>!2pbcO{WN?iEVtG*<lu~Ds85$y
zM%>_T=UgJYLuwRrhUo`UCBNrt%OZDInKlNgkZ}1y#n86>vS&y@Wq49aD5D-`k79O>
zW%JnYpBYskwOMC1T*?){Gsf3dlP0fp;!cI41Q3X#VL3Gd?OW+%30k;BM*AD@&qhjF
z8|R``L_RS(6A1qfN8BYgRzTzFCu|$^e_<ILI|GOamWO$qf#NlF)GL%>pUcc!R*06~
zJtCCGuyd?+4!4`iFf!sDKpBdxG2F_42lp2a^fb%{gO5=cht?Tbfi|m78t=AiX_XYv
zum9Vim~#p2Km=c}5Gz+_tv{F#$L^SIW3iw;MAR6AM9a_e@VPJRkclObdtD}Qc{!3F
zsw!6<@x?xNOiS>nRa{))wnmvHvG?Vo)j+8nFZol+u()h_k5PA9xu*YlJ?W#sMikQ{
z3$mM*)ZvWeW67(8c^dNUNcdl1P<0|Pg=|`6^;Z7P7#&nEt6nicF@iXzZr1qB$4z~C
zzx8LD&$A&4{p?6HVdGLv&-izGr0v<?Ar%53kX&93qFL6E#*v2HP?^JgK5}Rnf-Zh7
z)EC6NrXY8froS=h>N%@{F~r4HepWxb#G)L90L*6IWb{lDA+w-v4Q(TZ%EJo&^{jJ{
z*uO_HkDg-kK(GLiI~4;!n_Q0)&L2!n8dJc~3;tRZJlPWEZPR)w$dLMFh&qarv2ns?
z;>om<xhZ`r4W@Q{BuKB_V0xi)jqAV{CF3T&nnX^#k?8DkT#tIWWG=|XmzZ{o9*%x`
zYjtNQ38A69YN&$oLHXVBVB&uL<KrQ}Ts=kcYiDh7m|;?%Ga>0gDxg))uK{qoC1xo)
zFfaE_MSD2<k6(-PMn{CHt{!~$Vm?4_=5~kFZu4&X2bK)MA|~vfiF5r(E}ApH%nS;e
ziU@ctHfy~?vsiKUUfxmt_1k#Q{nvuZ=lNIw>z&Pcnj~AzvYZ1A%eJkDp8F{GE1B%8
zrFzzXGLiffR_0YQR#(|Wgkf*EPN2Kn{E+_VA-5c<v!BKFJ9v0Z{s!IpACllSKeYE0
zs=Q<pD}~7*n}h1|B>Vhh-=_N1iK5OIzGgu{d80WYJ#H?1{x2w)9gEvV3j7l%p?Us)
z=Ssdg_~B@a*>bWIgos(>W|v?}3O#a1W{?n%z=Uk&HH(Aafggx8te|{2%6hdq3(6*V
z;u_8sW1)|bA>bQPkk8^Ezno02`9R}L{tZ^&`tm$dFxclzR)<~)l>R`NnD=hj_&%F{
zs#cpVFr>aN;s9wB&|)$cKF;oNE0`wE!kZz4|E00=mt21<4oZxn+bZ0EXyOkDQlu{r
ztHT$FGi`)!n*v!xOeSQ+#Rz^t8v-dYFiiVQpUkr2^g)Y6Azla+jQ8sgG6yS*$Pd>O
zUnCTXWI;-kg(@s+v*G;>okdN=AsSpAo-tHy>Sy3e{{;R@B3qmKE%3hh`gLE9fs-gG
z6$*kDO@2YKwcKr#AWfA_A4I<XC{chv1*)GXT|yb^7pUuk3j-t^#Y|YvU}*t{^{-sR
zqYm10>I2f@cjlin%iB{q_MxiUb?~wL@IA9(5@!+lkqc2g|H@Z5>GBde8zobSE|eZ3
zH19Is=7ircn$YY%2_=YlRD{m#sAMQHlHY<oU=hBHM>H`+@OACXoGX9qqchZt3Tzzq
z(no`v%)wcR?siGs<t}<>c-)YhN_cVbxF6}BhxT*H5WFR>1IGq?jK!ktlF8yrPD!^C
zlH#$3l&wsZA;0-G01$h~5%Fzz1j>PqZRWtBEqD(7%=`Wh6ig*Yp~4MCTTE0xR7y>-
zWs5OBNxZ`^FOd<ba7}n_&tF4uS{+QbW13(k4x@cm9S!*T)G2*?S;|LH4*}iwNv3o5
zIyb60iCiUB`fV-(GFNGV)MHv^Ndx>COa;zk>+xdQPvk{X0_wQi_QCGE_BQljLoPXj
z@#6(7DXu`SQ)XSntnDbM?)T73$+h&BLjx?%J+lW>fo)+>@cX)8+_Actm>#=~Igyy$
z(*SVlFX$}|pfo4c&(1#CGzWlBK~~_is>D@8KRfh|mRJNLBG9nnJShN`c~d;eEcaRj
zDad3mBxEwxWGA*ru2X7VB(m5rwhKCjy$nt>Nf<X4#~2h0AXQ1iUEA6CP)QOGUND8+
zG8BS&hmSKS<(Wz<iyHZ4Iqh_}Q`&Bw1R_x9+5eRDPra{+%w%ZyuEJ?WFjw(9>EGZw
z3*@+nNFd1;<1@4^OcGarWXw^%7##s3^P>LlSuSoCClO6Z&kCukD9t3lto~B+)GP9L
z!naUdQqIK1Mm_YWOZ}@6n^Uusc$1yDA~`5`yb8sG=uD6?$>bKp`j|>Eh@g`mTDHX}
zpUgg)KkBo(iIHj}2kW}hmfjOsBW?!=x95~<Sc8p4Xqn|-LGBx&K^_xLsFgynesPic
zvODA~8qs=R<z1yGLj-V`ht%E$Z5vDt5@&0aBGn-zSJ8f`O2#(f#jKV!U9(WeO*Pr`
zSwLc$pw4-KH@3dBZR11t(ng_K_>Su68<_-V586%eL&fN(1NtfR(w0p&<3@&~khA=>
z!jFh<>art(9cVcO9mLR?nh(R?^{6xSGZSB3<*_a!CV_qqNkjEKL!=)?=GP;qr@Jc-
zN3i6X*sfS+CNaLy$H35<(eUdTjdCO(7&g!&D-Ay$Po^M5fkJ{x1PJo4Bn~xGKtUq^
zoNG=!7f>7fk9I(%VmS}H@0QJ1NJ(Rog`ntes*A54TH^JAbpu4U!UHN+-Oqpt?}9L>
z;ekpxDX7ha7cg7-vDmYa1rsr>@Xu&eCQ0Jc)1BJ7HPi6>0qsAVGj)bYcBYF!uMqn=
zuHMZtVrA%t1)2qry9Q<Vm$Ok5{`#(4mIm;v!|Wx)LAeQcUcdJMlTbWz@v*whLp?!x
zn*h6mYJm99CKx=S2YTk(mJmg8tWSFmv9@}sOpn?+zcsA3s$w$B7=Qq3F+}zxWBGvy
zOKSfjMr5sNGoS88cR|lx>tRF5to*9Pfab%ba4pI6pL$C!)@v4UC{O`8A>r?qlj0;k
z1Y)m#1x<Yl^C<&;Or1}UsAPm07T}wTKoLe^cMU%Bmcy64G2q_p|C+%Y!{cb`ZuioT
zbT1ug3h_pgszs~q`}bMsVzj5KD!~`ooq6le+%S)WBwTEfQH;Tr?Q7r#RUL*YV51r-
z%aqhs#Mo!><i>DNMm0$rls?33fniJ`UG1ox?e>b42NYagGC2STvqmZB5nPWnO0(wW
z&i#;m$W?2lWMRVdVX`Ioeu~q=nFR3mPs85OIprfPEjOzK0*VNNtIH>X1@P$^<kdp*
zi@Y{mmc%H*<>Pd>F9yZ8URzKd9)`|m$uqQtj}}!@BYg>n00;0c01P#%j!KjI)rycN
zTT^S?er_PAbw>zXMVS-=+Ki0WqNzDBqM10RbOCeytL}hlyNWv{mJB!qs~2zWz1nP>
zgJI1RLfvkD#<kH91<dc634Ql)Sl5GfJsCLOVP`Y{>30Gn{v4zLZf6A`og9DF5HRGU
zOSMKZG)ZZfiJS<zCOZOy6ufGX!h5f%u&-#XuY2iKTI6HyoS1Vz&6Izollf3~=1DJ(
zXnT|qHviN}lSdLSSE3-c&^Re1;n|gFH8B<WgHwBUH{rul^A-l}MDCo{?Mxwj<2h2T
zYJ-R@SFefcxr3eB9oXQPr-2+m@|nsZ;GPk>C(9Zf&md!rWmZHIK+(MtlwC?4VlVB#
zy#QoxV~H3$jm$f&pZaph*l^da{6eZwIqyavFi5H|ysyB#1LE4J9hXzc3!PkYew_Kl
zovDd$bZ&^@9K41KbhAUka!(R@{2}%)OHak(^5UwvYLN&ikG>b?F)^8fOpPH0Ik?ic
zA&&$&eC#x0Yq^GMPIV%}a2b7UKVPE4Wl64JdK&ozA|VipDn=mDV~C}n!ebPCu<dC`
z?a_O{JcC;CkzP$TD`ggTa<0`7(9h{=s}tv)y(8MBs(YozKJ`y|Y<$UC%8PcWdCruB
zm;yhd-~Dzb)eG?wKv7fRSLSQOq&F6*i@CcYCiC-D6>q*am&W5i|ItR>-)x}VUq0TM
zBB|sr^IJVIsBWGHrQ`luk7QBQDi~6iO8Iu`&pe|H!2~ikV{pNQv=yzPo#0v6ap+;3
zcL|h4;FmzmE`xfjPARu<?LA6(3imwSQrt8l)R_;-p||m>oBGyBt2{3XdN@OL6tO|6
zqiGYV1gBB`j$6P3Z24L3GZ+M4rCgJ}tn@zJd!ZToIts(nE#%+sh4PPa-ga?2)DkC7
z8}6Lg7bh{@GN<Ahp|4Q|1N1hSYiTgFQ(uuDf3m#+)2VSMXLHED@F+*|m;%WSm6<}`
zBt%A<!&GERI`YJi(<x#c!cpFgKpGR`B##Wv(f8R?iuN&Rww64YNOAk~X6FJ>&4|Gy
zCe}R&(`#?kRLhUy_%7ZD7^Hc#wRHXC#-F6*azt#UOi_0JCLBEwSXj-3)tx_}U44k?
z)gRya*opbEjv17v{%sD|*<~8IT^&i)z(e*v8G$pP#V(E_Nb7nBo{4WKWi})|N_llq
z$ropk`YEcGNn}$*!DjFZR4E376n(Kb3dt)Hw3}dUe(Q^_lm_h)Nzx?*7mI#em}&<<
z<VobMxd8i){iv(lD4I}ub4!R}&vYiS<w;=&bCVCDK^?AEw2dbpbZB@d*bvA)_)yi2
zkYlw%ugITKGME(Dhh@TCxMp;P=7GiyCG8{UAbwptKd*DKFYp!^Bn5vtb;kt0)e~Yg
z&;FsH+D!{w7(Cc$mm96;szvhtgLpcn6aa`Pdykq-^}EI_C$zVu8A}|^njB@uz>px>
zN5?|jfj#v;3oaQ-K?J{lzmEZN??6s7zkP{O3wV}++aJ4kL4PB$KbtBlDO$$-=VX2r
zBdaoz*MvS*;?VASnIYLbB9Lej>NjV?t@c9vnde9ih2UF{pQ-o%VkVON2;Z)0wB>hd
zwX_x$!9s$}+##WS^)z7JXAlk(!;7%6T%)yjzv(R`K2+WLulM~%-V9j-zoD;sN2Dxm
zgNK>ndm$0xaPze!d7=`d0$L_N`wY)L1i$fjPag5x4>BS0;c71K&Ua%lU8A_7cVTTa
zZtsJ}nBiFKcX7ri!rkDptdx@-Azby`Vs%ov&lrUTAe|zwvO8;}eORH$U%UM$o7kA)
z;_9JtxGEINU@0XW5vruP++!46Ao_FfR1TRLX4>xLQ07qscfPph(U<i}Ao;2R`+k}{
zk;-HcLibn8&{{2pkkBY(dvHH=?^#w>syzDj1mxT@Lj4pvE`N&sDK+)VWP+5O0=3xa
z$DK0P{#G5rw2voK6JlqqV<hUTWnM;>`9OA>@vG4)o0~Va3H&u;6@HGC;V5wwXu1Ad
znR857L*HVnywG*M%$Ed#SnJsfB6*}9r_x%xo@GQ+cux`9XVTOZzB52V=I4kk%^jV5
zMP4D8Txqg%Yk;Z9kcQ10pydJO-PCHxLh!o#&khf7GFEfROA5evqC8?}PE@MB4cz|=
z%t)Klz^lVIh{$9qdOosgSp!wQ^imrD`W+JI0gD(q-iAksDW{Oy$dp|wZdN9B6oNar
zK}lvvq5ME29UIraKfn?vY!KywvW2UBsH7J95}p)51pls3D)^}h&1X;xL77x9Q4ip5
z-+soF<w3R>Tjm1+)sEWVK^r)AcSswE=?FxHUm)QtK4>R}nb#_pD-OV?D3rdejje|;
zpOP<0^n)VAsn~77EKqNMV5nY{;*?dmspz7-tq-+TVU1$|hy7NZMdNJRLX*+Ff88ah
zTqwGzt&u`$GoA@7>tp^v#p<UFKO^10>S_tGBv-ZAT0U;;`sr;1x0Nh^_=5<Ny(AwH
zg>~dc{EevWwFgt~jc17FIXO~4yV%QR;0j$k#@dpbf@$Q&Z)d#>#$@45Mm<d_^yANR
z|76~7O`Y6oQ;7cULlt_G%;7JGFNT%S_U&@1!l=x%Tmasa*<Ar4jflw8SFLNcHY^In
z%hnDgfY;eFrx=>z2MNvQ`TAef6b;~?&DDx9C~8-CeYTGA+m!dXc_m<6F|R@nqLFz^
zw2Za=;JGXg{__&|bW)f|lu^Ph&IyZfcMWrGv7YdZxwJM2&_^83VR5(}ly7n|AOtVf
z7Gl23xV01f$n9PzC~DQQ1IOAYj9<#5R@o*SVZT9ujZ=*-XxC#P_qAOJWE>e&OXHME
z*yYY1rW_C*(QHg<gg-EVYF5jW&q!n3f5e&7N%`btHD%fCHf;wX4BaEIpAPcb^~B@!
z{4CFAvF*Nw`K%Dwe+Iel`7`a)$d_jkXre@2z9H21W#qAS3!#+Vg#Ns2<aY$;*!tPJ
z==iFXeU?eHn&t6z5$(GIi@4H~_UEO(yDsh3H+)(r?NEjgt#>KmSqViEhzXwj1Bi%%
zByeBmu5h&PUwe{3M55}b+J-+9_dd1w@KkdugCimvpPt3A`>X@Cc9jbbSvI>!<wiht
zfsuMGBZ*;r-6pbDu|pKY>^FwjD!ioWBz*3>c}tukiI^ADZ*pu`=R(@V%SS=*+0u@!
zMci1ditYr`pIN8Jk6v<S?RGQ@9$QXzas(-On^eE|%$5-}r)S6E6bd|<P=&M?F}6JG
z7l4+Xu`~R>Y}ChBC1ff-X1Xz43ET2sT&2~*`2n-XyWGW*kfw4W+ycGZZTOoczZkt$
z;epfQ39+0gw^pPgiRX*i1rEzB25M}z$0_H@At3pW4maR|4al8mQoe5t){4{6dD*NU
z3osuP)b}vt<xG@DBnT4m>s)FaX9NA_gO7^kCJm1SA3kmb59s$_D<~PH=XRR*rWFS!
z5u&(>$eEcE1$r({^J&OsB`oX{#2tpEZfeBorU>rz<!;CL<T6K<a_>L}R2Qa$K9cEM
z56ito90tlY=lHL)wK(w&0QYv1+gSlQ-^6~P90RtAV-0@y_}yr~9_^#&XC_vBveY@B
zApB7W4n=x3yiK}aHr1hh^-dTX9Sc#l9;nPL(B(QD(0^XP*{mk!PphK9rV8f!*!4l5
z93so~SHkMoY->tvEJsAbcclMI2^6h6vz7P^BG}n%G5UVN8+)6ov^!vr`4}(UBndUf
zjiQzl?qhXwu5fJjF(It>w{P65A}}rW%l@;TlJ;zh(+-+_1EzBE;AKP5=esgMgost=
zll!)5#@?@SiX?AT_!W{0iZMlqS}E0~Mk6$^Vk7}$Z$?&{Jm1t*{-#Z%I>{6<$8ZOC
zU@DjAT*&2LKOe>Cr7;$^327Tq#tBzAmIbRG_W?peX70pI$Z${5Qp};3y?@hI!X|6L
zYS21dD?{B+a&=6WBmnN)fLSQ&tIi4hc&d{zxMMyGspQ5S;mBMj*NF!4uJe#NgYB>-
z0C+if1*p<^MRr7W>+<c<K>~zZ@bWg>l_)419XK`%6^ul5{kAXc5S2;R_(s1$^dTt}
zUShfxdWQXT_^}P;6eMX?c+uyO^6Qlz`<_0b<i3uEg%$mwC%qo=yWD~q@^RY3rKs>?
znU`*1shLdK?(ruj6T3kkAh$VwupA0NtJRrMkO<UwzFq!Uwc|f%V*o7x1P|qsmqYKM
zWgBu#?qz_W>~1%#OsLOQdN@+530w>5qDPSK-9*cl+4-zWBb2b_u|-#-drMlCl{b9!
zw_r9YpVT^(V1x4yL*zYYke#LY+V=-_0R>V%A%^YOi6|Q5r#H?%A=HcasX$_3=Z$-}
zUAJbJ7?G}Dis)fs2me6PH)ef6xmLT}6cM%PMVo^`Nh`F<4ZV0!X%6-qQ6Dm#K@m&T
zZS$YT2#wQ07G#Dgy6o?s604|)Tvg43CZ>;p-;D%ML1S!9fdu~BWG_)-qJ42z$NX(|
z-Xy|50fkCAHIsVjjEfnB<!6_ib~+XIwnC-hVVIt7lwEJkq@JzzT1MaZ=AibVuY-ui
za7H#-^S%31yOa%ZeAr6!;s!4K%iB_;c~*CU9Wul>oN`v-0x){iQi5Q|0*pq3iJ<Ei
zr=*=NR>h6`*r3iOZ_xG1>cM)#nV=it$lQfBEJ`@xDQ;ULS83&$k%V$M4wlPSQ&@2I
zry?+(OfqEl_ZzSc7)1MhxN@~)sEdiSM`s`Xv{Ut*dDYiwkD27HPyInyji`AB$KJ7J
z>~#O~9MdG-KqN5D#Z_<tFH}z_<=_mnZ1L%;c@@1rwFHiWOoK%v#9m$Yp*uG&oI+>?
zPGymWP?}j}hF7DjG3X;!=fIbxB*g>6T7p6O;|`vvx@4k!?#H_e1QgQ{O{3Mou%_)j
zoH0{>L&Jn9byyoo1n0kDnFe7!p3<vnAm#yas2*=b&8xgZ&xdS%k`RCX7*r0~<R>>Q
z;$(`!dKuA2!tvy<LpA-Jf-c7BTxM)Zz?6`S%CiFWN=b*)51lHrEVCrqk(@NXvZns5
z*hL1c_GN(-BwaU1y2#YNA|Gp|1Frxv(a)bljEqi0wu^ydZQAdVGrLgD1acFiNpS1J
zQ<85e+A2Pviwf!H1ErTG*P~6~wS%Mp#7o|;LabYau%&obQTq-&y4HSA$7K8z7f^I#
zpA!Tbc^Sc8dSID2s{fC*hcIF~qy3;1YH6wVkCA+4-T8*D4=%WrextR(ddl#`5;To3
zF02nHbPorPwV9lSVpvr=M?}$ObC%jyO~!ptQH`(7-V1W-(ev?yH2kuW%(%E>oN{MH
zUhNNG1Q(#8luKr2B_7#0DnAUoBHRMJ=s7uvcw)F)48SY2E}}cX!yq<F)a~Q<%fHcQ
zQJJ+ZbYimnd+#xY!9ISTwjX`4DKzz`3pO7no^$15xUbFA3j%#OOh{0BHH6S#&ULb^
zy@WxHVVbT^;oTHe^`*JUPhCj;(#OUUXQ4MzDro*N@yFXu%()L?^SIUYEA=zQd;)z_
z8>^GNdqgWJPw@?KJEZ=;a_=Kc1LW6_zgzfD5Z=Vu!bS9`z1@scX^i}QiK#PCOHDZK
zF<!lCW%YH-U7<e9W|e}xTz}{+30<kcLoLx(Wy87p^R;cPbTicCE83zokVx4jaCB&-
z%~=~YJC^cWM<u>PD=N|+kDYl3F}2rLCf-3*sz_NVPR6Jy41FZBu$sz=+L}&2Qgt?p
zd3B5>1-ZT#RIjFxy`gE9ROeM=Qo^z3Vq#y*GX28zBE?-9=}espGg`^sM$rsIy)*VH
zr(dFN^o}n{P01x1LDjVK&x*jlwJVJ$wQXp>yDyrpx2)izvgRB|gOdAV9xTWu{K!Mp
z&KwNWX~Un;1?Bfm1BV;Lun*9wEHz$Jomsm+GYe0Wp(d=}z<aK=!on*~+$ITF%lhN`
zQ9&9z`7wVca}2|2;qCjEb@FYp1qtMbgs3?2O9!J?L04Tzb2<S)DB26SYyZnWf&3CR
zNlZ9U`f@0^<(?oYU)8MC81CP`(_z{_)L)8TDbx!l=qMJ|I5BqCM6+2O-fv@2v`Toc
z@6XKL39htV?y3L*pw#-yCE-a+#$=WL+cAB={c3`C3>RWpj59A3si^>!F^cVy`gr>l
zrC*`|K(um%BO(y<0;kK@T@=JFcSQV3PZLLk^=0QpoNDu{y5DI^@UM7ssS9mH)zY5i
zNaUr9W96K~pFTFywg|-*tvVF6;$D04ash4k6U8-|xVF!z%mlFJ=QqSgZ!yMAP`M(t
zlopjYVYi5Zl#bN(PXg3x)eUPYBW>!haq7QU9X+30(b#UocGZX4RNx7`LwkP@bZ*N)
zoEg;^gR6MuJWaK0ni>;AHD{17qpBSwj+p81KjSk?iq#TOElL^yZYI@*9$zlbER44{
zMOofQq4%4sEhjv9u}c<56dJY--69t2Wc!nZPXZ2uUPUYpN%QXyA_AhZKyj-tVhY}n
z8_KPb4BC+=^@F&}WbY@g8ZiId##8DzxX%s=ZOby5YmlQIez-^s`p^U&I*q7rVwRPl
zZ|qgktTx6eF}-Q{`SSsOyANJm!4LiRb&#Wq`^QmvKx9Esj?D<srKSF}I#hza)a73$
zzaL2;s7z(m>@$Uu@{yP|ZlihCpJC!<(YJuHn`L?h_uTpcIjEM^s}Qq^jQc5t^GI>#
ztMxF_hE2wU6=mqC*{EAUcGdX<uE8BxU5LIj3y^FvmdZ9Q-KA;oUP3P+v?C%+jg!*q
zKc#+dN?WjV{$!HJTyzsVv40n|H!RpJGnP1cyvJr`T4$KFkYD9Wz6g=3QuUGrD<bB%
zhhEoS-{;(e6e;n_sJpE;r_iY7!E(asyEduHj0Bi#PV4d(f|ykVsuvp`F9M6&ELttU
z9i%lP3i9e6LuOt$V#e0O<Sk$6G0Q#&2*tTIO-klIn}GW5O_Tr_Dq-9tseRBm%~T17
z^=!WzK{t}H%1S0X&LV!aLfn&j7vhZ=1X%8u+&|RvZBZmF)L3C+YJx|M7C*nWjL0~1
z9L*=R)~JfrNk~p|2@o2vWc>XPMEH?-`S87-ZNw+r5aCSn7Rv6Gmy#49_0BL*-_NM>
zgGakx+1uNWXlI($T1qFEDI@|e;qTuW$*QAa15Z;=PH+H=IBlp`lCQsYZe?d>OHaM?
zea2kDuY)P$1Z-6ExRl+fx`>q`RARyigcZ1SzYxp)O?h_I-6{k3!@((3-Pb0~T%DfW
zk}w&4>IcG6M>J;ATpVk3cKBz-PEF85h2>TPX%p<D;~eLM58K%$V1Ycl1ofNZS^8AR
zu0jE$K4AH`Z&>|lY`py@>u`h)wE9JLJ*#@AWm9Q?!YJXeIZv)(pst&gXWGy{)5{)}
zQVGFQ-f#!B776O|F%eHfvsYOyWymY(m>89E4Wl_v01<Dj4B1*RFq$k%MmOay0>!Xy
zf4X9f7sQjN@QoKpYR_(dsh%FjHwLh;men*JEk-PFA}Pia{2?l~`I-lqwqaul8n=EO
zwn;TtXKW{JIj#inUs$)Pa~ewMGk1yiXOJGB(fv$ysoWE<7h<Kt4*lI)0ER5}t<jQU
zN%=OFf{D~73pn<6#(qHVt@{eam<oRn+Cx}v>_7J5rGkhEm(H!7ax@alw@ETWtTiCC
zu*g?rEriD8MMKLUbW=d?@{-Ys$&SVPFfgp&9<|mK&*Of^H23!GqoG2IvA26jpgwgs
z{;d1VpeYn7q17Rat+=5V35tOF+sgfA_tM+A?_uZXTt^91+YdVPGrP^GOpQFhl7I;g
zNcXhw06_11C{W|^jk5j-<yg<y4cd@qR!6%}@Gw|uw#%sYdweL`*N?Ilj%Yd*;>O{P
zA>oNbHtA0O8^L;L?ivm2v8<atx!2$3pxky~O|qmWM~M!p6U-#Teb*A@oXW<Z91&s%
zJ*n-?DKmJj)5=1f^jy$Jd8fj(M&5SQ2mbAjeU1pVFaIEhn)K6GgK}|3?RGr_9oCd>
z^dh^~dORUYBP;o{LOoP0#Z@Ij0y(wRm%W=aRLHeh=Cr%VzjSF=%hb;&!zQur2uLw9
z`eS@l3Tm}_kLu>rS{M5kG07ZEr8gvBoniJxG4-43TVIFK&7~1xJ@27qN6q3MvKdo#
zDFSo%ZnWzbJjUCM6{}^-Y>3&YL{^lTa^?{)w@e^>A<7{ZN>M`Pq4v#vm-g0lAN{(Y
z6C<AXNg#8l=+nh%erl7HW&r8%Vi{C8^3?}z(}vY@O%x~lzK5@W)w`m~<5CKd8tp+#
zfhiQPF!zj5l<*H2l^GYkON2aK|H|ubYS-X6^WaYUxMlOZ)Mzh{zY=VWlT^YW^7=>8
zl`@}p8+Dv-#=e)S_Th5BXMu@6`h^NxLF|B>R0oP>a|p(nv*|$7Msp`k9tzUQH{lS>
zBr^s6qD72S-wjdNjKscCCOU1+OO{_ucJyU^ng6nx&+=W`=#NdoUdv{zRPOJ!a-Dh3
z3(PV-!H~PITQ?+SzB%n(Z@*rJT71dKAeda$fMUMe#FR&db3sw@Iq5S{8?Gf}TS|C!
z(K*%~s4*IpZ`D970>W8j48Adlm;I27maSM<$ODhT$jyA--MzbdSFB7IFO3NQC`Z7v
zyv9yZ=3b6lc_`jb#6dvhcLTSJaqZi!?+-(`w!dy-T*7RtwTYEQMvKhFp~5s{{_P$2
z618)r%Q}{&3huR?236?<k9WEIwyZ!WVa^N|CgDsamf|o4tqmb{?5r+k4WL8qeBF+%
zQU0c~9Yv<o#%?QTytk$wSWW@aw<C><2gaPkkGxi}Ou6y~f_EAn?!Y@e{nb2X-Mx$O
zR2zFkzOE6Ecg+%T(=OYT=()Hu{Hlp_DUj$&?97ePM{9Mw{5O<aLo4=5>Lk4x58>_-
zhrci46k&C5s`n?{6`QTSb#czyd9d1`ZIzM~DXiHtR9mGq;FmIe&qP9<-?zEkwHK;G
z%6+<L1Z=qtnB>3HD!&M9q}6e0_F0C7?S&MHla5lHlD-k#4ZbV+XWRgxJTS5UIxA58
z70nL>VhWy*erI#IbiCUJ{`YYRMK2$TRLJ-ZL`8)-wqiXo?zKVSDTro^8Zcn`e`;y6
zFo0fjtoJIDQq-s1#2fcZ3YWi>MR74G1XsoDURS&OqnD}z<F+)wbjzK5ojdE>80P9m
zb29fIfWtM2o*6k$q8kO%C#qg@w%k#FyS4Wd>Tt!DTS7hPW1_O(c5TubN0=JMVzyo(
zej58HZ2JFVFxR1?$F`KxLC<lN;m=|_$|CN&W_*r@>helw*)OCf>Ak4rx9@&;_jK4M
z<NcPQ2nuwhuuU7F!qwqzb83~lc2|L;DtTDVbGhcjO4ehV_6eE(m*1dQO7R`?J8CF&
zPmOd+(dLvA$=7nW;Plyh4iNp>@)OErLlV_k)<nTy<5p1Re@+YW?esg4zC`{7LylU7
z`G7*ks*0QHilka?%Nm85wQlMTzED^X!i0G(e*AzPzVGqEz~+t-yG2EP&e2iiT;A@x
z3W-|9w6ZQ{&tyg$$#js$#AU1)=2za|X0OmX%ZmEL!@_y$#I091#{axQ%$^Mu8u)h0
z_4=2o523bg-}r8@e|~zyDL(8J>!Ckh{i(|ea}qy{5_h7&GA6rM<9eD3tZR$vi;9Xo
z!jK|y!LR^I;MQV^;Y80udJku&bZ_sG#WUb3fYb4B#wXFX?#=Q>5ENxe03-yvm>15#
zE@Cc=1uFMLpy53|Iz8bdX|s~M&Frg4WJ0qwd*h*EsW(?wPkK&fo*n0I6Ea{GB;R&;
zx~bBB!k$f#SM5VRz;*(<5PYLBK${JDjfZ&w6nN%vEppL<6cO}3;d4pLM5(;JN8)Ws
z_D_2?2jy3Hq)x&QMY8f>KlH;N9dYj7X<HI|_C2D6onIf`w=ZjC8HsOvL5#ZoNVDYJ
zd*r3S2iIJ(`r=q*a%Q_9Yds!on9ky~W912vrj{myY{wgGBU|w0uaOqh$nOhZZ`%(?
zV?`Vm`L+Gnrp2lRZeG8BJ;xY&w9GGNlo!vTy-$k`*UG2EM_QK;o4=`U6{^Bg*zehp
z*I-gt<;49w#md1LHc7aw9V4H`M#+qT=5}N76i5j{$8aa#)07Crz3g~*tw@ABlyJRV
zJ>j+mqT=%3Wi6@TGbIth5Mc}RV<do6O<kQOokwMyhq8KVs5mR?BrtL55t*YU<X!Jh
z^abcjY0rxyn?J;TocxfO{j??Nh{hlB=;@Mq=F<_0PrYUi(t#}Yq?xXh`3_U~U5VPG
zp=}k)bLctT5xotfuJdBx0gijH<=4oMW8grz`q~NbF>}zOJXCIIbL*zmrEv9oGKo^j
zkKpX?kNvsXN?A17n=SZ%b#p4}tSC9m9yeGcHH_u>qk-qpb>CR=aKw$4!I7l_?mIqT
zb#m$w-tyfqmPA}ong4&Nd+WC*-|&BU!59r9or;9CK{pCg3Wy3*YA_L`1*BsG6_FAU
zL_!P@k=j%`wvmETN(@05Ass_vBlq0AKi~Qnp8deVad7YEy3Xs2*IC!OI=!m3(xUhE
zB<s?|?YBiS?<P)blAIsI)YeYbVUX<puWAS;b%`>duYdLND)hvgS>8e*t724GR#AS;
zw<9v$ntaN;cuW>C+Au{ruuEMvSmNI$gqMvNd?N26$mSIHJNMtTFi-U9@MwM=;aHy>
zn#1loUby&V)@VQp8G-3Irj!l$XIXF%l#)!QhXg(FHJtSCj7<H}JmE=xN5U!3MQcc8
z%6ll#z-5bmg8}54UZpv?JJ#AXUhhoQ0)w~IR1|{wQ9l#ehO;IsY+BL@V#<s@IA*ca
zjfXvL&Vd)rKyBaYjZ1HeGWt{)XprHv>O$X~VPy&9lTR`mb025!GjGFE7;Fm#CnL~8
zlvJM2lF#EZ#hf~W&!bnFGz9GHCZ1%1!`~XT%ByC$Qb$5o;HQt5eH~!9qi0RB%BnA)
zH~y)<cmtQ&bs7dT3Rq_F&LsDL!>tmmcn+*SGil_vnb&{`IF8rb)#m@ceXIM0zKHJp
zr`;C0F%(p$U&(E-TeRz37?9MF?)p}ggBiu%{=aJhgtYfTet~<lq^XzMyHD?s!^uxc
z&{8vwqxc<{qWTZeC~P62mmGR=_)VycD6~z6w0m>}p~6B1B9cpp5PAlNnY6qV${8oR
za+XXpM~9HszI?oimp@)UciJ@w(#2n1H6E<bpqJVN#y=Uvbd(bmNuH7?y3YHJ{p)M-
z+<D^mEf{14@#FVE7<UoH{2tP+5=8X6S}h1r1p409SzLy$iiU;Ai`*$m`1@eY*o}|G
zn`Z%qlX=GJNG4-<cY$=ct+Z;XsLwWb{e>J2uSJVtwY}uQxuYYDVJFh9TEoFc?5*9b
zN??xDyIS)C-tps5!AQ}j7Gf}GAEemTdBoD2!#Qykk`4YLKS(ogGeDZo0t2io3{d|^
zA(`J|9d4uqlby)2WxETXdFFjhN}n4C^j;RV)oy%dluxlO)1%nHA3MCVD|_9VBlbk~
zXdfCVqO&?>OTLr-UD|<1609~Gme+LcZ;YgU4&At3xw*Uk*FXIU+u6?V^upVR>zo8`
zACnF-63c<8g)1IhQIcV`%S8ib8<?(PZp6cWdg0nO?kSkEor51p`{%EiFFfq~t+5cF
z?8BmlFYoLI$%6aqrh>ddjha1m{}KP&2WEc{Cdzs8KH(Y>_n^?X?bg-T#OkqTmPY2t
zSFfYQjU)s4U8Nu*!6)Dp#_A|~@+J09*_vK_w8#}og4j}CxmY8w#k)tI=*_4HL&(0z
zp-^NdUG`YlyD4Prb37AW`*cKSONV~BhG?bGzHkzp^ixiSz5Wf3gpIEgntmEn^vPKu
zy2a9?kU&b_7tHmv8GeRq@vJSQwN7knY7F$0ns|X3ZD}XTQKt0VtLvvXqIdv`r>T5y
zqUgd&kOPhebhDO5G`Ko^K1x^Y_PQs75=;r&oUA$|@@vuLscCAc{ivFJtu_Pe&B{(E
zi{SVF^z*|wSU)`4b5Tuxu%p{36MPiPoy_>%SO@$@6!iElAOZ);5}I#EuF3=kJwK<R
z1lWYmIefFw{+PCqeTldy=!kdR{*U;lF~$SNP>4mg$tp2;Pb5}KWC?2%V22hTJnYN7
zxQS~!)GPBd6ipCQ-&lVysr`IbRJ;MV{`NmjD3QxRU?_XCR!Z_Lu#iao^SJzG6Wq)6
z_R3Lg`Pb3-p~(Xym-sU_uXWg3Z0Ye=keFItA}|bOXfBf{4@NIY6pc>hbaKxx{~UR7
zhjv@9?p`$WCP%|Ustx26{a8Id3H`NlOX>O*nal^0_$v|IrQ`2=VJT9YZyUsDs<lO<
zY=gCG{(+bRP^t@tZhfY@)t|{W;G@2+$q@@B>N7oWlr={d&ppn5QvPIYj61{nE6FO_
z`W)l_ZX&RJQ}@qiph(^b)S+0JFE#5Mik<KZkx@}e4Gr2I0OheAT%_2SJt=uQYl^)<
zarGVw_Yk0HCA`6R(}PyS((Tl3!7sz?S2-zXyUbp=Abzz5oO>~=NbiyC(@hr#xL&Sx
zmUPG~+h?AAvH?F%te(Am#I&-mm~(18WvDTxasTqw_EjjAEsH7iiP6t@U_@NGh(5#u
zlmt^^;r{WMB9e7Bj6cFr^l8t-TZnWcEec0Ak9O^+{{<X00cAPCc_20x@XDMu_;i#@
zPG@2oJo-s_v!*c3?U?4a0GYC_Q-_w>oiFDMz#LxYmEUZRpS{u*g)Y@e3gSz0<l4vH
zhzfZ1*oL=qq){v{IBB+EE+*~5e0kVf8)@omQCK6af-_O{wYaty(@y=lWTSVAD<>%Y
z{^heTycnqh6jtu8#9N40pYJFzTXTwV<h>Z{%b#NMQJ9*JrzxRPjoj%aY#r&6Z*7qe
zU7j3Esm&dL;B1A<cM=<9dsa)AvE{AbS#-J50?6fSYjwERa?<3Th4Pe#u~NM5#MpgQ
z{61Km19E8P%A9E(<K6C7lFQYAuRisj3y{jce%xJuXR7OZ>1(Me`S{F2QK;THOhZhb
z4zIotbDqLf>!dJ0R}F~)Ik9jwrj%D#5_6zbXfK2F$IotkjC9|Y2{?Ou?D!#e_K~DE
zdUk@R0n-B$`j0QnO1h>|bFL3#@P_})2R~XYA}Z_Chae*^rAZge>IXxjo|!&iUHav0
z8GuBJV-sg5H$qp#7azsFY>c8?CX0wc^5)2={lJfZok{K4pd?xNN;r5<YlDW1bS%Pi
z=jdy{4wQ?e0kHICh^`u>V0Up+3a=*;x6SrSNV6JZ;h}F)87Oveh(I4Qc#jaTlUViz
zNxgLzL^`8`Q(PCv#7=c=6t}IO&m{`on6NP+BwHL&UTy?WKM)#yy8WE%SgR=A%>dC(
zQy3_`<Oo?Yx{0@HKRo<gg{yUHxhtdry`E4Bo?Ibz@_*7RF|}=UTyQq*l>QosyCD2F
z`0KQqu$5~mE;U6YcSiB~h6vwtK>-oY@1i;dec-v2b$HS<u~XNT4F6ed#!&Acy|g*$
z_xbnjL;7vWIU%eh{au@*K92?`>(}^oc(qGI?OM09tol4rKX%oCEh1YaB{@ncE2);{
zAesC#Q)5}t3Sx)CtZYaIgV9n7bAQ8h+Pzu2Mk?$V`@Jp!DAdTN>jOjoBrd#el4x5V
zsjYDog*2m(oN;%CiYhqkN{E7$$-f<0K&D+3|8pgAbn4FsQe#3d$4RyqxXPF!nY`Wz
ze1a$uY&AQUlrMI^TwDHDOFHZ!n}?}imi=`w8I?mXx*IxnO-(1@W4?-Rc&z4Mfb(Gq
zf9~yD(0`m7Xlvpf+ol>=zm`eO>*UV1z`l7<`(+B`?Y7GWV6jn!l<w%8J9p|?EMBWH
z#LK|wum5Ui5lZjgKv5@IqAQMB{^w(FZV-t6XlUCPWXYToC0J^Bc!C`F;wWqA3i7W1
zatWQdlz*>u5dRW@SLIF(=_w6~Yk}2ThyA1t_2*>_Q76ToiN1aWvv-o-Rdp8-RnD_8
z)}g)qr~kNU7ayTQw~$S!JaXn&d1DpgyX99X8hSb)hR<Jz56M|h;2va+(CAgrOjW!>
ziY>35<kC6EJH~!atX9L+Y4%qJ5igIEG^=HO!$NywC21{}4IFK<7$IEiShJflhJWt<
zn?|-xRL${z_tooaXV?nlH2)Dt664(S+dthTD4QgN(X*VI!~@P>xVs&PhJW^4g3A6@
z?9p{MrK_Kv7y7EnPxC7yPOp61m^s@MV6ibNs0vcj>7l<mW`D_<Jvt&R=VOsCy+Xw~
zaT}VTHWI~OijQ%2vGZw&BKturU|A^kkoa`d*L^KVM+MzuPm#$rWI7RH9l@921d2mj
z=pGPnrx)2Ie?UQzP2$IV-(?LT^ZTSvudr9{1^Oon#WuaNtn;b8<O37?K^cG5q{)@$
zz#6+R-fjyv(On_8lj-rDV)o1F79>-3P0ECmou0+JSsT(R8VvCKGIW#t>sY<#FqP6_
z?cyGAgVXnq`cmd%WW&YwT)_(;Uu@Ik9>pAtdz9~e!4?K)Z+Uhutxa+$9pzjHRe%pB
zv6AiFWwQP;E2_S-{#vvU!r#EOmJAZmk;zATjb6oiFr^f5=)CUS;K@BNNzL_;#Km1c
zX8HQ(*O=e5BcSEI4gz_0Wu7?XtQ$fzt`af=7H?64?D6?}NEAtZgdRj|`CgCtdHeEj
zO5P~0lmE2toC1Q&k2_Sy_R1+Sg@TNihJSm(k|QpYkJGO_E(G8}q$qGf6x|nz$_bpQ
zqL0)b5)t|$wNGYYU_WRGdMqy^ZSGM7yZxK~XRB8|h(f)o%$BtbkxJcYH#H&Fy-K#V
zik7+h`;x)8nmg@+WBs+dM#^)Et8#<L<7Jy)`=AX{N<6>|)jpc_$-d6f@NvzoXKca-
z?!XzDkiu^M{x^mWY35j2JakET@wIgUQGBSGUelM_X$y<LgS!=+E^LJ|>>&~l<;YD4
z|1)<k%`<VvD8KFaF1ti8opOE+_>{7N6R|%w-W(+=88gfIU5lN;EO>D5B@&)~JB%VY
z6L?WS=`P+WI-oJ;bHef^{ES(Z3^6D%;XZw~NTG=2J1inl0N$JhDeX`yN`GXoZ|Gh5
z&U%66SYdK!-iMHPNQYi=_U_FShf`ahnhB*O;csCCKdk!T81|OFDgqrK<j+fik)sfz
zmN2l94P_@JFnEO$Ku$xDPc`q+d$?IH@$15SCV*ns6Tp<9P#EjoN8qtLWe=8HcST)>
zZ_@D;0Mxv|V;L+x#Jkv|-uFs`N$v3MP+xA$|2m&WQSgsZVtf@*!FwQz=#SDK-bBh}
zkr8MqZ6z4p-M`5xW|AoF7j?O}rMpy-F$KiY7L3{gS8P$_-KG5H9#wYiy*DZWV~?d*
z{ca9O;N&h{h+Kfd2M0u7w9R^#eQ<H-!x2QQvVEVMBD2y$cPrB&;Hi5HM&`em5k2eH
z+7xRQnz$tMy*9lH0Xs-my*l>P)vt@ww@J&0uioDuh&d@8Ie~zn5ga}H>KG~T{TPsG
z=fG#nnXRv??)H<WWPEz5C~UOK^<1=goP39)A-G0$0ZuV2>+CX6AV+Ipxgu6IprHy@
z(46@4J#o$zpcSA55$mcv%tLb}r-lhxofj~3jX#6<wfo{w_L!?AtFA_XypzU#Tg)T8
zS7vR!`RSknhPBh1S0Vuu2GbkxI>RmNGNF_{vrU+9NQt5UGOd6$3Om;Rb$A{W#wB0L
zvy`gpF(IgF3(L#xeCy`L)WuG}gAQR)$uj={h(x2PVZiN5zStw(>xRk&v=sa{ft&J>
z0CZCS;)uknEPZWgmkicX=$!ijMqI$MXb2vZH72z+`-P-=V2?O<erv@U?S3wq*yFwV
zEJ1SJ`o3gT%rQTmFgu+qO(-B%7jm?DMg|AVOTWjJH~qoM(KJphF0gU;ED~%#=w?ng
zhVB-$S7>-Wda2_sb#1PLqYpHZESkHdJ^V{i?S?CLQfcj0A+I42!4TZwv)0oLrn$K%
z?A2agL}aa=a2Gz(yV}_dqM{=-WuQH&@woc^RfInJp7+r1)s_9A2iWg#M>KQ4h$o2L
zIg#WQK1({kD`1hP!DeCbcz7PC9Cyqqx4%h)`{A`GOXdIcx0h$@fE^N&^-prxs|s*_
zNVwUE{5zV4YqjdIU}R`un8w`8q%S&dE+8h{ZJurRZsx0BR{Tm^PYzDzc?~eczW|9Y
zw}aoNz6-1?Ry=vU+6N!Ld`1>S<}-8(${3?Ff2Hkcjih9%+b3@uAd+Bz;$3vPmvA2l
zasO1_yu>|XAaURyB4qQ-h`7;-z1*9zw0+E)bdWK?269dxbV2Vbpi1C5RPxpxZT%*X
zNLH@}RZg#+Dy9gO_&ioUN594M-Rty9@Us`sFPb_fSHVgPL_^rR5;_>a`ZU;SdTjd(
z`?L2e|LX^NvqlNziKa+_#h$8qD?yq${kd#oq9L~H)T~$zGXsFyqfFep#qAu*L0wWa
zmt<Tq6hJu6(k|mh9xYT9<P-Xc4;M~bA2`U<?9E=#?>s557YB(XtG+(epyft1d`*L{
zPx-FwM}v0sN(C)IDvVV3w7n)f!Hpr(KIUUlN$-E2-`27B!59kjI(oGPsYfWbRiB-V
zg+W>rn@e%GzCR=l&$Z%iU0p?hRu>^jw+cPN1Wf*^1))1<+@fN6<2DqRlzE-8bq5+A
zl;Iy+C`nu0b;b@?%%e==AqsPkGnIF=&yCwQ+(j<I=$FD*);Imfst4?su$yS+RE3Ar
zBAx}BAWIM4^J%dDaJmvOv&&Z&bd4Ea<am;%53+;78IhHosw5qi@uPJ=h7nTf8Qku!
zE%`T6L>L!Ni$^A8p_($i<v#3ck94HvMS3z$DD)$aCDz;)u)zZatFz^a^!@_iTI*DZ
z9Vig1zWG+IjG5$&p1hIQ!)DAP?Ew)`hGNA+&$KVbr;WE&39^cri)kR{r#Q;C091nF
z&3d89=%r9AV!zm!IxQD+f!97?h>;GY&v(s%B3ry^n8pbUCX+wnm51U#cKOo|_8R8x
z9q2%4Pl3zvrAB@`A)jonb80mhLs#6X()Lz&to`o7<}Y@R=YUEmT6+1sA&n5sSpyY^
zt05bv_mEKmIk?%j-xDI2F`{MxCGQzm4v(FIr$Lw1z?sE3p%+D-+LSM3#^JWFEUX?X
z{GQelQmz+7m%EDXb)y|tj^ukGJ?SzWSz)#O_KS<yp;wg<6A;{03gZOrrSvb^3`Y0r
zi)Xnc^j@^bIkMUy01weTuf}NNKCZPgqrR+rfb?whu=?Wc)bj3J1da+BkZcJc8~VH1
z%=J~F@vQa_*Kzj<F5GXFZ6&<fXq9negQaIMs$ydB-j2D}6NvQ4lhqlOc_%@g9c#)s
zn`{4pPtj2o@R1zF%Ah$1xm|s)tof_bk+If^G6tQC!J&`8-1@r_*y%b}{eF%*Op?@f
zZ%~SkQ}tr7pQ>XjozxS^PA-?$>1^$S3EurYMOyDelcTk(p{ItFPIf5+9+=b*`_@U$
zj7E@$a`oP1XRqiZSl;`rROEL7@AHJyX*0{Xy8-dsmx_AA{QJQ8cJ4PD`Gz;p!hqMI
zv-M;id$(=x8}2?~syWgr0~P74m6bObVRmea6#(A_!QhbGG(bPaH!GYrCoS<8#aHdA
zvxHNb?@)d+88D2sWFZj82t9-X9O%lK>XUESKg!_Rj}#_x!-36@|HfPz!_*K~W&1-V
zk5y;FKil+mR1h1VKH1yy>FZ@|K5F2nd{1CD{!{YO4E95rOU5o)H7CM*C{i=P3sV5a
z7oxb4TogKnE<p3HD7_$zk!oFU=bm<n``e7TFn6io>?=D<Y3SuMRO%a>^kYijK6X>s
z3F@)*$$|E|zqETd3#UWnGfz|=;z^^PUuktvj3BO6c?%%(MdDJgdebTc?VPPTO;rK+
zOWX@x57PK^Ia7Yzy!U>P0{u&&0V?h8F@HR-(K-}AUH3{3@nxc2R&UlwfmT21Oe44O
zYwiRooDyE6c(phc6mZ7Qy8Wnvq@rU)Y3k{@;v>kefhlqfM}=K5uO?}s{N^O-+q`&U
zW>;9|zg+oLnK0b%@WFy(y(%ex9CS5*mM$(~Gi)glHiQ(W807?7!9D?V3EFMp0A91#
z)Zi*LnM~S-k&?>|#vS*1pozJ0(wsHOWNR6eVO%RNUHcaf)KVgrlb0#-?aU;<#tWle
zacuq>Ir<HeW2GbDx4x~<Ai@>lalU#_-)YZX0T&RHE7&x};+vZymDrlDuu6)=36@$p
zpB~qu6hNDkGEG@B@^2x2k;5ta)mK`Yrp>(5+C9pW3R=6*H;uH)xdYwLhx+?xu#UIC
zV>0iDtMQheK~edabl)ADv|vE7DzN!V*nnemhyXy*pI5+&>r)NL*$cR;tB7-Hs+}nQ
zk0%!zw-DMitTa{6J9j@AX;1OP?pg4ob*J8p+H1b~jxG2up}>5O+iRs(NMoc`%sut*
zccLV`WCvt%>!{en13;sv4R2?2a(T@r!s6Chy1R|~)>$3Q@4oge{&ZihQl;vKDr2s4
z%V&e{=zPMTrmE?t&^kWKBzS_MqCFUaj+(TDGse>OYV%&|1@%vt^Yja>9NO|`spvCk
z^tDc6#q{Gr^XX^b2I^pm_R4YU&-71_;jZh<JL0F-6S<EyZMTUUGw4K#f;(_^?Z8t<
z18HUmJ3p7~;f7p%jY!Y;z;BI)j=a?^*ch4l6&Riafq}(Gcx5`?4(wl=vLxq_`%4j*
z$!4U!P*#6jd{Zw(clR7aCLQpiG6NS7>3)j-4PWvjAIt->v#rzvs_?o$;RJ%9xw4KE
z3x$UYjSI=8Zvr~D2IAZjtQOk@*W#mnkYXNxk{5Fiv8X7Z$31q7SuI2y{-cfi;$@r2
z1p#mBq_&uEV0G;~M2|$6I@yxE4M{%c!ot~yLuH?w{Y%XrC}iY5eR0mOv~rpl*O_15
zo0^+ElnF<Q$#B%Y(2lRR$xxzA@|T{8WU_sBPvM-%vU>lkXvFH+CT5usBXW>N&g@^2
zcGp()ZMXfs-e!L3E6Oc&4;9$vrgs9l;L~6x8S@)^Dx;gu^8|n^+;dY5D||zsgVEc}
z&9vPNY{2<_Me@fHdVCDid>mGWpj1{iIYzjlDHhBeU<IKY2Ka=>!?1Md!R$@(OlT@n
zo2!Mf_ZDH0Jb9By{|H5wHZ{%5*=|h!aU<ujT;M)dW*7QzL6C4hfxWBGP_M>~3XNcY
zBb8D`CqRwq)8ks+#UK#Xe`CQfpf$TD`=@|*6$|SDXM1dS94=LqZ@l2K<5o|Hl6C1~
z9;@D(EBOn#7-P3N;OlRDXE9iTf@imPv4e^59gM+g*c!qmKVgijgp=SwgAtJRx&kOZ
z=6}{RQ<usPEN-5mn79{hPmAA3UL}uJKGtKGIj@}ep0mC8e&^g%z;z-P{WiA7jI$y#
zwzhF_Y3L+lOlYtEd#NAdE$JW^*<u%x9Q|Xk_lU_j5|D`t?I_rs=#JjGrJ_qo$DCN`
z5LD&5vq#S3m*{N1G-9ZELe_-{n=&yQtYX)F@xgaRlfk}fo%|W<H}kASIpD{^w;2vr
zv7J_>E7Y{`<uKlzJ)4sU_NnD+-fUvI=**L!S|PtwiTEB;cgJ~xD}^(5?Dy2DloY&F
zSLZ1#SqQt={|haP*ewtun{&oIXsU)y+M(F#!7z2}-GWQ?V#mfwYGE``u=>5yCqdRD
zKda5Rf#~Z4P<{$Q``T9iW2w>QxK=+G=4LqI;R5vL=(q$$9a)(%@OvBXYKC?Oan9Gq
z4@f9jJ0&;KOFGlH0eg%n#ZrB88xJG<TIP1Fb+Z_1dx|W%dAiT0$Hb3e=2=zvj*t5@
zaNK158zTD;{Md|)p2gMh-W7k35<S8f*BfjFZ-nuXy6M`r-oHJi%SVZ1X$6YD5UbwZ
zfPpcw@#(+JE$I@uxVwaNE0FUdC~*bwd1iUpAQf5z#{KK1^C)E(=B(Q!%Tzf9FcwOB
z1QSr&3_fv`XTh*Y-P}3fiizqqXL;DE!fhyxofMP;v6vj2KBjPr2u5mVMymD>4z%8(
zB5u!_$EUl)-B}iz^o66Ho5kfmzMM3UeRQA}uY7v#8f6mH{A6gFlf7k+L4@JRt}fhp
zngG8R)0t+v_K5Y&5lbVV(r<t4sC^ec5qmen1LU7vqi2xnc0Ppqgw?G1${)5e`TxOz
z-Lanc62C<gyXmQEB2QtVeA&6v2Lpr%uoyWuEoQd$7xky#y2QQmR>;;#jH*F>W~?x}
z2;m>bWV<s@=#BrjMy}=nqT#H1(h7$cRce#rb7(xf!})S=d2_>4Q+nRKQ~qVmf!ci8
zlVBAW@X<-kj85Z|@^jv;bLktzaz29nPoyk%aI_+}LgZuvU9R?KlQaNM(`zJ`ACShY
z*g{p${{5LTFM@$Rl=%=ED$I8fQ=Cz(0hjNA8))C!@106JB3U*plE#0>X6B4CkH4I`
zwc^&{1ETwpEYX^h-Q~kb_D~sk4<_)l#Q?EEY>n;pm<w}(Ejb7YOPf@NuZpd;&#lt(
zA~gIz)==H1EE}bICs=3GD&L}ZKg|=XinZ*ICFjC>27~CW)^6E6BOw0uW~SRK&1v4J
zVcTS(Hk@MYeJeXPf0BRsMF0HtUa<!k?Zm4y_I|r--nWpi>O>Ob+9lOHpbh7u1hWhF
zAf!}6Fz=5x$Zq~<9PGLpvUKQu^7<8Ia;!Th22?BV5Pxeqos2A~i3W|hZniQ4ax7#q
z`{+}GYT{LwAb;&Yt(VAp?uPL^2iDx|(wCJ(LJlO1O*ipmMa^)buF60-*gYb*^>tuZ
zC2sL`F^fgG_#>%ilZ@!sktb`v-~OleXY*P|aMUVQ^(gRA?TJ{DR7~pyw4Y1S{u*TD
zZ9FiO*N2RM+*7oEW;c9UfILEuVOo2Y7pWH7+WH=GK}Lm#mJ}b6Ck1H<Y>MB<IR?vg
zy#B=5l~8z6XM6?uP^sh9+YH>~vNdaUuHiA=!VP*ZY|b>0%-?v2U0@Y?8f4#32ob7I
z$E)p7<Vo!)9oi(E^~kCzKGB5!XgTH2kEg_C!3GpOPI#3kE4}ew13T-I9#pASaW48C
zp9ZrReZ9q2#kbWQtm?v#yMeu6WWq>KjS=-BrC{{IsrRLB&+CXJm!D;CdNQOU)*n|K
z1+6{vkehI<q2q_+8gTFec{@wcobj5zxX8(_7jtwLz852PAg_00nZ;<C8$W7>NahEv
zSaejQx|IVfB}Ad*d2mtm^ESG|FdHAb>MALkt8@=Q45u&5h2Z|3+a&0w6Gu+(hr#pU
zk*`y_n$*guy)khXxrJS{i=(Q4B5=FfzF>++Blv3#lD<n?`yMo>75c7t=6fDp*<1aR
z2e@|mA9a@sgjxY~>SnAB`iHXN)OSh{>bk>47Pow-VyVx~9u$u%WxfH+c481SJ&_~;
zau_0fGoyEnxa)ljyEyi-V#4^a!jwX1=O)sF71GY}JJd5ix?d}b*nQRRm`D5k=Rkn$
zO6EzoV7`TliKnyVu^dqe0aB2kchHC9_gu#41ufAgDFj=#;OdJg&K+&Om_Ioj4r_|c
z{W%(q_+7t23tfQ*)AO$eTT@34oMU5=E7u95L84gBe3qmQfBv1lG4Tj?y(cMdyFbHN
zYbS4~H_Fd*F$0so+l>!=oV9Pkk;!+7v4h{S+?T@GATL?xF>qof!%_ZhNA(bb5dB1z
zGq^N->y@_~9ZItNMoX!Gk9PY&2kPHXfy|5yZ;p!Dx+Y$n@W+l2g*fpnV*2?hW=Co(
zA#zp+M(GhU=uGwhEV$Uu$8OO-h)Q{%9d%22(La4AY>5S_nZFj-*86@DP7aV`xcUYD
zSPo@!1{{4Y*sMD<>Dmnzo6jlnaq6TcPe7LZT#2!^E0;D)9ER$9l0ryJxbL{iDCEi`
zT|%TcoES-l4dwJXSjW1`Rl8!fYBkFSf8~%Tj}Qd<T`g(WER<?VoFuvc?7$b+$V@v?
zJLj}7aiJJk$c-Gj{8j7{-R7rEj4iYjYcT&!B(giWY`*&oqGv~Gc@qxw;tA>QoT(V2
z8Aod{T5LT@z|_QU9C15HH#tZuPIA_AXk?_s{EEOhL#SI{@v$}SqQ$7Ly`Yh>eRL=K
zWy|DCq^~%p$^BF3#N~>II}MpKN8b@paz7*a4D3!BkYEM`vHKY}X!lp@4=4@}tub1w
zK^dy0C51MG(z){emM!N`@H=cVG{B`@+#X8%xo#~KY}DOP4cr-2OcG%0zv{Ee*x~J2
zB+W+aV?T1VB@yZ?VJYq0;CPn?G0-r|aG_sJ!T+CY8bki~CpDk1iFXF)7%4)1LuhK`
zlZb@k{z!k$+51l&uY(K%3^!vQQ`DTJ{bk}9E2rNap-BBI=})rAQ`-o^tRZ3@$9|ae
zB`UX31Qr;$ioLr(H|u!IF$Kp#gG-ds?H=^6EjA9+y!tIG$<5z@{ez@pB5=Yp8DATr
zy`UpyHGOH_r`z9m&(}WGJ-S-muz85E*3YjZOgaSAbwD;Al04}PlP#6R!?y-dpi<^P
zkz+<rtDG#Y6SV;uT;Nj?ZcJZ;@d#lvqbng}xOC`^gctBTZ7X~0o605%E*;r_*cRZ}
zm(daG=k9%}F`)Yd2PW6;dj=3kK%FkCt8i$pS<|J@CqhL+i;CqblBW^u`wz4jPfrM#
zwYDfG(X3Z!$^@U@zJ>Zi5PB9!F>>>iz1BOS5qJi^a~R^Q;R%mh2I*s}?DzTu=uij$
z&e9(f`O_%4OTyBF{K#kbkk|dk$SYKdF<f(j=}BtvCyf^)AquG9ce(%C{l?jNSAt@O
z^Rm5;KxE+{g%Ie@mu5I?G}YsTaew6^g92F4sxAIEHXv1*Ivcp1N9UoWeZxqTZTv-#
z?)@{`!~;Ir3*K!Im+Toa4kCnk13h4gJY&lr%0DcL<sD-@s*V#SdbYhH!4Nk8kV?gD
zht7K1%eU^OOqxvUff_zeR=v7zuj32j`|8-A>ukT{AWu=Cg&_|AoWWIo@mp!bAn0fm
zQJ=xrC%W_YYmt?I{ui?tNX~2T@;@!QQ`%lp16^8u86RmYXYpUBI0J=khIN^vM&0?k
zJrP$Jh%wdYgZ9tleQ|}y$$;rhN4EZ6@A{^Y?i8UgKHCeQSXd$IE`d~6GOs%|LTjS+
zDFYpRluRd8C*7Yh^S#*!vB;v<N+uUhPk`w#Z}|9M$>6iGnG<zv8mtG-n6WR{7&Ia(
z)ko}accu~86rMT?YnBVue!HCi<<E0i|686@gkLbWqvJT!3sEVMQTL;ooaL7T=V^3C
zik&<7agew+gVz;xsUuQTqu*9Rt>>QY6HbqJ6(17nn;Jg)jhh{68{S4w#zttzJDV#%
z5cBfbSEHF=W^k-ip%O?oe|DM_0Y$b^xv@2b2)x4CSd=)}tLd>t(I$Jw=7UoGdrebw
z!}T~VM|+>I3TtAg?!K-&j3K(&wB9i)d>$L1|B~CQKZ2_?7+N+<9%9uDxTbw;YWcRl
z!u97bKp-Y_n$GOhi4|3BIe&+J>guC0Kc(WY(FHMrkEPmMsT}LBfq(OJ#=<kaV7?4^
z5Y1g{4J61C$z98E@CPCLp=?hU7Pxc|`K9KwB|nn&%Ddj3o#=)2-Tb&irj_N?W_~Zm
z=mPwCfHY5E=pWMtU6UE{asbz{<xpRny?O4$GQTx{%nF~L)|y&{hQEC4NTr*@FX;};
z`(L_Uma!dKAw14q1y^gO$!3&I+-LUAUmM&0jORoTHVNY-md5WZ=Rv=T@qaV2_&(lt
z97vP~-zgD%z{zj*d1iv$ZLB{dw0b+d0xvRl9?NLIz4sv<EMemvo31bZ4h4CI3fssZ
z@ZOscbx=!7i3$yVxYMXByZY-DGJbL&C&>W%o;^$r@U8SIy2qHXi-(1ttQ8;sG$D}K
zR`-(V9CxVaXg2H4Bk*jMCsdzSjI~i=jljwy<Kyjdg?;Ukc&a9BVR>q5V$bpF+%nn=
z>m;zx38|TV#~FJ}49T<LQ#R9D%t!R{(wquIO|-?+CO(zxo%6ekfTFi<Ra$yqx=)1&
zA8T+6ebVX<C>0aKtRc+0XL1WF13pQZWUFA0<Q*12eG_(-)mNTCH<UVQ!`cl#T(eX<
z1Q^WZ`_)~}jrnt=o2fAAd#<It^AL;8lSy1Bx=h8jn&W+2J?Huk{q*oqJu7T|nDlV2
zJHJ{8UrPdEmeMPIWyGnc2daKPTqo@w0TH)TOl*3w4z^VrVLRYBI3&2^3?dJxUS}zK
zWwV19D1D&KFJ=r33ELx$hn7yp0cLMz(5J=LR=npu`gW%Xyab#iKE{!4d?KR24zM+a
z7i59GBPWy^8C}7>d~c+FB`W!09$T0;hyEipaqtCi-~!Q`2`bRZ%k)oxW`#x*_&Znc
zk<XPyT8BOuWjx>>pZ?MO*=D*L8p5Cr>22=@87IS^fZ4OQD1HME_H__wHy&(>x`}<a
zNVJ{)j<41ub-NUo){o#7u~11?b<vjG2%9`T7TPInyS>l!NUV*Zgam8`z|@ZZ?#RFy
z2=eupoWp^OH6W|Gh{(6dzw`*~0|+*ADym|}1XVCYBO-S3{c>a&`Rzf94fi-7nUR9R
zSh?soEL7d$1(FcverNq5G#L#4m)oIa<U{Pm^)r&|IvNbM9jyggN0#%k`V2Hs_2qka
zXE7nz&*{3J(F(`WN71?pt29V9b0c5fq-AIQnoV1yPq995f!QIRL!P%kzO9=GCc(xD
zJXRY@)jbi0_2}Z7E7ANJ2O;Bbl*zI6pJAj=rH4+_*@n)u@lkUD-57?4zcBzjQ8@iT
zzBV&Ia@4Iwo9s`%inR8}o}k^Hem7``a*InjwL4t}qhXYsv}fQTXO4AtcdyBa+R!yK
z9c^xluUlsOZ>#TYz>uogQNuvWz7E5_y%6^}BMRS;n4ekNJ<LpK{B}4WhmvSA-i#sc
zhL?3(oTPYKm+Ef6+6f=JlT_3lOQzS69AJd^dAifmdV0H-dZgj2lM4~4$%Gq(b;4@T
zrnQu!K8FQ^2&TttfEwXLuqXBTst#i{jr-oQ>9GP8prCDBe6Wi)*w%fsQC9B;ufI@r
z>svmJXGXgv`2;BGQidxn;{P$(g&>0jCZYA$7}y3CR@00)cY!|`+y`Q9bp%hO!#h#n
zrBDi8%w3A<XL1Oazh>IKz@EW+=Z(b#{h7`3@7Ze4jQWLFVpjyospR4L1uwZ1&74_G
z0?&rVt_?)?X)6p|gf4J4K0S=Qhm_yWi5=ruk%+QSb0G0O4<Qs|pmESZqAqwp;|2;h
zJ9t&A;Jpr3*rM=Z@Pv>zmGDm?HVCM6X}*7?EU=Cx&|#9X>3wnXb-WuBeByafk)M{l
zmFG*3{2Wx)cvA#4U)~TB642BN*lR%Cf0xU1pnPSYU&<kAs9W6RUj;e@lS#y<rn`l8
zkPoNgO@H*H#b>kX41KYQY}OxEiMX(H<e`@E??x_e;v}Ad8oZR+rwv3^fD$g-n-Eh;
z3q$Nb8*cfmeQHZ_NoV&~-lB_aKYGYrlt;T#wD!9cJS9CeJJzRe!Z{*J@#NBxxby2^
z_4UbrYBL=fjj%@MQ0B3eCMg-`Yi17=g9a|2W}5S3^F@n4QT=LB^Q?*YfTPTX;zge5
z=|NZ|Y?5Q*VJbqH5vsncywC{5tEv1-6er*FL~0F_rBz*ow6Ckb{okoz_b3G;BZ)|T
z@+AfEN!7QYxW${;R>~=FoFQwq1vcL(cC&vwN%sgKP5mRm%TDwCQw@UsHJLfXyFYFZ
z626^CD;`V8yik9eNO%3KgJn7hAUBc5HB*cvkOKP_JQvtBgbw1!!$Hem2$r1cE^boA
zGWu;L1jC8$+vVn&S2_8LlZfdwRII&j{}FfRApMbkFR9G~pL|-|?m`&Nl9Ao>eLXt@
z+?Xsi_DqU)3=wUDUx2u}oOtU-jen(-Z^2nbMh4E4ZzuYIt%}%biHQt0?;EA0d)sa8
zSE=XbQ)sFMIT|l-Wcl42bdj{xiB;W;5|f8x-)lQxeJpIYN2Z@zX=!3?AON<OZ%maE
zg7o>j!r^}nald5Coxn#{@|(;(iM0XeXTq1d4GA(j2X8|>g%VBg|C_<)KiFu~BBbkZ
zN-O$-=`*;I)`iGx&@=mugpJG2s5xw=h|XPjT!xt6TdPeS%gBcp-wWXiiDwla=<>_2
zzx)JG<l(J>ABYkHyj{LsDYK<x$&b>AKGOaH{l6A&c`N;iU%e-S$?OUwMRclp>p3u%
zJ*d6s#~x}`GjAH6stu+IU|a3qPc#u{9wK&|ac#l!syCK>=av$z{5#r$KaF!$<6hRA
zwT@~9@#NYsk8WRqo}o7O)ZX&Bsh{5Jmvw9u+deW_Kd@m&b5ALr4O8oqpO1_8APD8C
zrk~{;Oe!DtVqWuu1{>Afim)Gw<o0KCgzMzu$n?)DGumwSLmH>TBY^<(uJoi3olOQm
zVdSWrFiX%Y+W-bQRG+ez7u>8OxSRQw31~e^v0uKtN@_nyB+L+O{C%1(v~lxFe!hpo
zJQ+arpUG<tc8ReVL$4AKFXL>W55{nk2GR|8ZbA(HUE7SA2#wDa)_$2XIb&Ov*L`qF
z=3d;NC!H?Z@6so{6x==OfEKy$m7VR`3x1P^X?Q(PD;g6Js-R3VjK^nM-?&|$J{u2R
zZw;;yeY#4yx~(;kl6fkrH+eC7q5(&Zu|N-rsWaP`J@2w?;K&Lfq3KiNAN*W{bPa@u
zd=*bV@pJ!1I9dU&x6EXZ^0+VDUEKuu`L9KKhj$Tm-&aQ?o1NbLk2_~xBn|S!t(LnO
zROs`W{&_m|#vvAViPad4=AQPciq_{&j@HGt-+qpl>C@)Qjw?S%t-R{p9^2lhWpS=e
zzZ$sQG}QJ}@evQ+caOb*4a!_aoF7d&Ji;@Kmx6+-BgK(&)!R+Z{=$zPH7wU>p;Ub^
zmUn+w58eo|lpjVl9!_vD1Q+ewZ|>|x>xq63T>&3;N(E08z6TlX4jd?>*At~WIjuIx
z9cOOSYG5K&L%cg7U_Upfi5m3#4+{6@$QYICtFGIxyUIuxvqj74EmZf)dp6y#FJBv6
zjfz_=&ARyT;*NGRgq5N+eBi{<*qT1>j?^BPMy;!AovfhwwapQAN$a5&6oiL5Hi)mX
zX;WBF)19wU6k&4X<u+Yd5OK)ho|N60@ZK(fVja8uLw9TaOODwY>Of5}1G{$TznyA1
zN|j1YbcUZ|cK1Jj_%pbzixfsE&R%;*uYo0^pkB}FT8%0lfc&PnOTID~i2CbNq_psF
zeBno|G_9REjY%HR^TSw%RzuR#d>W~JpfQxJui>Y`cTOW?u(VnYOw~F-md%1m#*>h;
zIeIwxS~=4Sdo!a3rv`h2HD}FKh?_E&W9%=E8<Fgo7e7X|C&j#7xf7_nw6+n4BP@v_
z&CVZ)g^d{p%8F(vTWOE^3<$zN-SKlV%(O*Btd1<s$>c^)`}misdleIQCckK8O0f6D
zI8)<N*?{Mo$&0f^7nr<~v;<1uMz<x3XwSw+ZafYCwA070xOIQD>y2XRE0wy(dJn5q
z)KgspFURbx_`I(#$6ROso?VZ9tXGPovX!#x>zl4qIC4W$At`Nh^?Tm0nd}|7*KF8>
zKUJ$IyIS<9WDQ+IjS_&P+;MZZ=UttEJ)i+}{`HNZw|=4Q(DBLY@80K*_JYk`tY`MV
zUf1OlaJuHT@1cgjm{Kzq|K*{FZiA<Ot1^E<%ec=Uk$DK5dT(R<;mf%gAVe=KpZNl`
zkmz{Rs(qsWP`Vb+9+6=mH`JNO=<v9l%|4<0O>YR70T|SkUGs(>hDbl<=$xunHoSh(
zmoBHQ57PjJzQ2%rz?Z(B+BE#A=(Tiu{eX|0OW)HbVtIlQ^qaJTCYC?q&+n(J3nW3E
zYIrtgRqg?C!x52yE+>|s|3YZugxbd8(5<5+&)*o5*U~leQQp+ZU}@|u$r#};x%K=D
z>t63!khPc!ot3z}7J>U2y1(siRe%b6Fm6YiNZAa-egVa3+un%sRpJI4_S1P*9tNJf
zKesMi7d-twFN(`RUbWmSyVm#o#Yx`})$H<*gAo*5F2)&$FZr}QpkyR#+{`Br54x}u
zaUPOe>}A>rCnX*#X|GgCt+4mIIoEpr{=|rTU#%Yz#p~0f`nCGzR4&!JCIo(2nA(qr
zMKrCZX1->Nxb6*^kRX-Y8Ar7C?T$CnTj`c<6B$9pc&mmu;ra*B)j;C+KNc@!AlPBn
zk)_RIKgb8S3Ev4Jn3(nU2Sx8d2AHr0E-eB3qCaBecd(#EKKHT3FJTkE$(r%5{3fbt
z_B1NOcskZwqgr?0<>d##w>2NU;U=k7H;iw+D<^0Z9D9`rg=cr6W-5PjV-H`?cHcwZ
zr_h^8mg!LlE~I(_+5iDLwPG~@RFf~6I{curci0Mx2n*DS*9eUdX&z{WOEOi9Cd0;t
zp{Z9Q{0iTK48r;koChKmcfxnhD_n1$44Tj&OTHF8<ZZ6ev3XJ`zlu0HIG-nV`-^ZU
zOiy`vsnJd|l3cUfF#u#nKfKfoH|DyV9B6Q|cKxoln5_DQ%eoY22cO_oiNjtrW&HZx
zY_@OxN=WuD-4H%7HO}kFVNbcctF}PLJxZQ%o>V<h8UBCx6OjV5o}5fhtpn2-hJ!8u
zV=>RZIb|KHIc%hoze@ChS3lD#dtcAV`tFJ0q1KA4T%+62Rdx%FNSW)@LG7}NU9xc9
z&SllFDZg8u)?Z5!>nD6W^0{f(b8nzSnUay~=@Hc}H4>cpZq5W;?5Yj@kQP?*CMd_W
zEJ)n^-Bh65DrtxA8Ge+rsZtxWJQc{30|EU7!fVF=fDCEq43>e0cvJo?IOLqX$<?WH
z-1^_<`~vcap~KM(Il+a+y{ZuQ>hGQp8Fh`Pt)dmG#|q`gqI-&8*JTOAQR`=}xF$m1
zHFV@);t7fiy#j^lq47-x0w%7QE`M(M?N+1cS|+!%Sqtl_u*Hi};-5M9BgP&nUSMj1
z>aZ(ZtNGsF>ao(aN+7G9qBTxk$}1;W^iG6O)&3VW4XA2)&y&Qswm8>W#-e8_m-v2@
ztWPo?e6EuuHTVRExf7RIo623bM^~E^@K;f``t+f^{;^)qaU{h!YUXlY&Eoe1e9hoW
z<zQ_OK{R+X3R|!_Kw4@_AQ^|L?nK<-J<xY}T(AVut=~jx+t#}xa;%(;`$EK5h)OSF
z23qG&bn*7kdD~M7=INr`nxO`YJw9JTeSU@xC%+>6jv5J=$-Q`cTwtHPN@DqIhhw0d
zDm(y6wN;XH1;#0SFC5r){wi3HW{zM#-$ljl_aL}kWD;PE?1ax=yAN3pwDwfB(nGk`
zZY9So=4~QIzo@8Eqg1n)7ED}6oNG}JC9AcO8araAubS7B`F7Ymq{E12+Qaw-O{9<A
zUznrBgHy+@qNc5xHfXKGc2La7I_@(ohkeOUhP<J~8FOQK5bIeI*q?pZdG}Q>N14;^
za_~qh=;(a3jp#MiYlH1j5FT7pF=BEyY_|2*n^pkm^3_(J#@Q%0s=EA*+Pfka*Y*rg
zX~h1p(yxCh;LE9Ja?&<r?*US&_L7R#D*GKU0#IpoZ2BBD4O!1;^cVDH_s^k5`a9L4
z-xIYxJ(3yE*rSWFpY=r3=M2Tu1W9XUHxtkw^9fcL|Afu^6a~s-m2uy#3I%rM%C2v7
zakH1g(`#REks?yc5u*j=E+qxrK#B8lK-yet6-491f+8+b?|VPC`=mI@;l)EcQ)W0X
zyDRjD2M(WTEk~5RcuRU5!^?L{R8Zu*TG((3Y^%n_FqUtn1<%{y&*ImBODtA2Kq;K<
ze_H-|O+0EELBR`#S`z62v%yIGW%+tKt~sS`EfcFSWU7$RFP?kyQ(R0b7G#T}>{i3+
zoWh*yuBA6Ze$%Ud89mFIqf6>J4A>{ax~Y&-!s_zWrk1zrz6bZA(T=Np%;xr<zocv|
zedw#=o&rAm)@83xnY-W7aL3=716*6IVLkS%^b%4EI__7V9+SnVHOtM=gSwdpE~95J
zykxFJZUCD$GB=fy0`Wn_8GVTm@G9aOd58q_6s?}#_g4T7r<6lt>OrVgDLsU7cGu(D
z$`++}t*d`sO41gxHnD7Gp{&ZZRv$tdojN-mL${?;HbWyOdTY2tHX@a3pPO}!KHhHx
zf<f%D@h<$kQ!F3-RUg)>YOyWVAPpa%%f>6lb-Z`<zee^W3Vj%InOu%S*dSSlw;o2g
z1J^LhjXWBfiYlH#O{ATb;!y$j?)Y6T{CM-*!MYcA)``@5rY!%@lOB^XY+=WutX`-t
zEjRu9H=2-wrUuA@YYnCbXg_hzyxeJ;Ed~GMmpM5UP4^*Bsd)|$k-Y2?6Mo{hbMtO5
z|81YEIwgo_L+}mC7bG99C7s6IKqY)X_n)IBA(74r7<~}!Bn4!vf5?L5I4A!vRG0#x
z@RO-`ZCk6p(ex1hT4kJK6PBIL!$K67_kDca7|aeC^{%S@JrVEmrhtsY%G$Y$^ot*`
zD7+D4{MPy)jGOqcn#{+jAYyAz4N~~##4C7Np4LP%-U&o(Hn2e^j?R)%gm0(Fi#X7p
zxq|;1_DPymA#1dj@bVnl_LT1#|ClK~{0s4UGnLPLvg&Sn^!-Hw-1Nn+dC`|6aUcLd
z%p&im5SJD@3)a~wHASz)-9Ln_O_4tBO64ah+Ree~0C@U0at74|)nMDdUFokzbyxgo
z2lSXZ3!LjdDixYLmUjso2RTJBsM=@Soye|m%JDHQ-=fOWjUfI62AUq5T*gVi)1*H_
zV3D#V%frLzh@xnP;Hg7jb^T5e*CJmV4MEm%cWy4aTyUD0`iau@Q8p1YEXJuTG8=gF
zY<K9<{8cT^qc|8Hj*@Dc4t8btVCLh_Gg+cO{e^QnU9uMm_`BTGzq{-&wO6IzqdH;t
zeIHat+S&_X->Sd<F9B{;SS)uQ%w@O4e|Wt(<K2#8)5y(Um(rChY2oc-U_k0*X1>j$
zM5@y-W-t)QDxIyO%x}ShIxm5!a7924Jp{G+z)q}fVxL{=`X@gbig_X2pjQ4|rcWWg
zS@+3mlCZ_Qn@D-AfJ28{@Uff#C2ut^K;Kdg37psuv9}!aSRE^;EsE0y!RRYxE(;po
z?I>w>@Ap#Czt2#06E5hwwEn(D>aOur0H2s(1f7fxR39*lYqD%i$m9?(kPzg`I?Gb^
zJgj_fA1Bm)Xcj5{8~qT0x`7QRD7&Ax1qwnwM)uID5cf0>f^hG^KtMz(y0c-eYet8r
zjWvcru|r1<xLXIC8T$;9Q4XwXA)ZIarNID7m3YMvD}PHjL2<cCH?l$;>byYgjOX8Q
z%va;lfU<&&pgRMnqE${=Uw${>C!W5~Xb|x0kH>{mIuo%z#W9i(b&>1n87wGTJ7I5n
zfNVqxIrZCH!&^f{G{$~*E^cEOHm?YxhL22&Gvx?bc6f@$_u=_j{Z-)f$q#eI5|BKD
z0L?TYGG{Yk?tjF8B2DCfN6556S}#B4S*l%`UkIxwSA#?5<^$bY!dBQfHy?|c#d@^|
z^8(&@;SJaSvOd7Crk;0hWjP@%WioQ=9d|iw`PJ=9a#j<7Kdb@$9RKY#?ekkv&|(<U
zBIWh6Dw6Ho;~(itt!5SX@!BRcKvPdz@Bi~4gTc|)w6Mijp<U=ZWrEJKnh>w(+Asak
zw6C6+w}b|SK3vu^4rrK3zr<Mx={dUvDpc?|dg=Op7q8-Rq%88kaDtpF*qZe!oCD0k
zu!mwts=M+pT43*0wYxG=jNTqMquWgX2ValZZ~T9pv^Vt`oBy@2Sh_5&SM!qtpFK65
z-}>!PfBdXz_15p>SgcfeIyWc^V}~gr=#bSBJ`+Bqop~W|?JX6u9GR7{kG?&t&xC~`
zm__pD`+&ajMP26wu&M2Tudo7Y3W5Em&#~yDb@t>aqGSk}$ktl`Yh;V$r%|0m?6l$`
z&d?5Cav;@6pIrOxF!(0CR$vHL)9>`Gbmeh2Q4(*d5B<vLp^*6ls_ZKNjmj5%HXYVO
z-+lW(1YpmhC&v8;Gt;=gh}1UZiU&fm75p7Gy&N4A3h_0Wk3~}oH+0iuES+Un{3Vk`
zCFyyL{J|O$Qf9!FlK=&Q7g+qy3r=*s`gDo&RO%I|-@{FmElgYn*nN@N|33GdW#;A}
zpqYSqtBVo|kL1oC0~$4e%}oGJ7yl!Na<+iQYKA4X3xUMU_A39lB!lRTc|4JM0M%be
z|94Wy>I>(2OpQBCkpi_a1X{J&9#YnkqcsGc!L(wQ$QxehV%Bqjpq=Pq|G!_6lLdn)
zGGs9MJ&-~WLE6r4r~xr|Wx_h9d{^uZ@c!96fC!l$+*d&$*2Cjkfd3e=#56Mt1R`-9
zDAa-nUH^i9W9BZgn**y2&-{;Oz(XY(nc1MHzl?#Lrs-cG9s#Tn__$T3=Yakz76;x9
zd~UJ<`Ktf>>3+@)O7}{}0Rti~uq<|0kfILrp554*AB3#P4;lasUvK`;=^R%>8Wn#0
z!%8FXSi!^+3xm&Mq5DAN+zgAhh}n$2<WZKd90jPeuYb3TrvZ^*;6VP@cZNXxU>Uyf
zQ$I@*2vP*hW3i_}ODg2ZuP6(nX9aWgpU3|s4=8*5Z>#_R`;q4QlfRY&!LR}w2I<2%
zR%U@=rp|C%kT#|P)2gKzqedAakMk1nkcXN&c<!yazQt67z#MyEu=a4@*{ZA2mbVJ&
z(I$QtHahsDEF(!f75YoY9(oZ_AO7Ew@f6)zk!Xc(qJ`1+Xtfo{|JU7>e<hjj@t3T$
zlcsgMp_OSyDyikzqL!g%Ww_v^mN`1*8ZN0Rlewg5Hd*GWe1W<qYEv#Xxg;(qgkWWc
z3u+R{Vmc#PCP+hWh`@bm?m2Vr{0Z~?{(QgB^Lf6X&pGdT&-r|!G74^)tU0I0X<M=2
zochan?n8evKK;w^$}wG5X@~0HZp_W(GpjO`AJ?zcD?=w|T6E&3u*4s#T&+Am%Pypq
zz+JjT!(XS8(1%l?pBu5zhvmlqNy@Ro#m%Yi=TAr$xMK?s)|-9RAMlcUQIdyOaJ(eH
zTkg@LF9hfOxcO(!nlg=U!^i1q1?hn!9&@4tx3uR)!zV29^gh?>Ku%|SSLC*?87jh}
zt>SOv6JrOYR-{W%l0#c*{-4IihVKTDHH@)_>qU_oE2mVSg0JPzcYd#&8(Z=kEOD(4
z?@;-jb=sgRd^0uOk0fQB+dI+RFuoYFt|;`)w_|I8`ZnkOP17seOC}mdBjH9m7G3JS
zPW$$EDz@d$M}+iy;0-zTm}Ir*WUDEI{xPWF{gQK^9w*E1YUE(QHDg*$ep5=lEhaVH
zXKr81w)+OX7MDLMv>j`t!Dr6HE#95zf-RLqd8p&!<B+I5iKMui0>QNn8>c*wrxK1I
zj`1*vFsoP}(y&l%E^srlQ@i^YT3f&LBE5M=JgIZ?%Ap&B{vClzS?9pOH4>4?`_JvD
zEGKNQ<UDqSU&`l~C@zJyW%fJcCOGOj`MAh=UgSWSyj~{PK{gU5rqaM_B{#v><cg^E
z+8iO#IGMKdgKTjUkxAI>EB9A1ViUr@+Zf7pnx(mXkU17I=}xfKTQ$_YrO`EVD4H#j
znj+F=QBuM|AqWMJ{*3ShLxn?U<H0;M&x&6xmLYZe@(H1Dhf;U21&I8`XQP8OX$@|F
zj1s+9;>2Ji%6uu#>|3A<D6S&H#B+5m5qt2E3p5o<2l19aLG;gXo$uxx2Q@->b&{^a
z49JO&j?;E@D2=-KPOU<#hnkuDKcxN+*?)$pRJpW~GD_X2I>hKB=>h5%ChwyP^HwTY
z(NZ2bWRgg3;l4XVkdLu$3Iljesw3d&b_MojjL}f>0Su$gugw_e>rskZWn4M=T9aUJ
zs$d#aZPP$D_dTLO(8mAlDpNM!{rNqj@G2I(w5w5OUhG9-@)(VGYU`AS^XcW@ajGEr
zaP}Jm8w{?E*21WrzA-qq*!fskKn%WTI6L{GZMp3V1=CclNxiLsK{u&y+$XX6tp1gT
zcj&u#>sd1o-}ppw7-Y^&Y8cM+Imsd6ZnD!X?sNj3`oyPjLzN?Dq~$HmW+IJ&TXkxO
z9Dm@oMrsZn(Eel@%%0EjlUE{Sp93`SXpD+PKPyF0w@+n&fmCyxT0-yM;#VBYn_=>T
zGw1k6n#z}ZudUAJa1-Wh#Sz?d15Ou|Ufmn}@Z^X1kL&Dh|0?~oMzJi6(uaT|twWUD
zGojRcdUtlQL*6-G4<$#Zv+VN@;axetrScb*0aB;86Ds_$aOjml0~w2Ddyb$4CoGTl
zZCF>0;NPaGW&J60m%ab>{|$g1K`t>kB54wN=B~KEd3vc%K1s~w817cCuGovrds$32
z4YOod5lv#10X%1EYt55#+Xt!&<7o1hUzI_#GmP(|c>)_W?;`2OaeUq@B)ZO5LC^Z6
znURJ<S>@W9fdIz?Kp0;@c3>SS6~8CuUzTE0Y@=_V&w3G`Sbw2kJEzmc5Dpd;aR;(J
zYcp-x*r|9-h3G+hxSrIi!+QKDcMu-0$ZM7zM4$Ez_DqZS7==*Vn!K1^o2hbHytotg
z!s0JZJgrIaN;enbkh_Vej}@D8n@+}dwIG=26c_d&G5_V>kvLIuG&owNZ*6GBC$;IC
z=e2x;jYeF((vR0}SQ%>Hdj4j6=Oey(@()ER9b=WSzqf$ghW}aXDYu<0jsgwFQAR)*
z`!$qHIk3*AD+_RY;=4Wg;B+1U+#xPO`M7os5c`3q=SQH2kcWcRv~EqQI1JQ5a|(ZJ
z1#$+9AY$GBQ1wc?^I$L@$YF1V2;$~_YMI`jfN{cQeiR#a0YV76KoSCG`fr(WB#3`f
zsVIy0`mu8!z-WN-6iG}&_#vFj#0`kM@Af5mLzesCvV2fftnMkwYHc}qM|=!4vr7LO
z;c5+han%YT;FHr|sV}du6Goy_tMuoR9!I<S%8L-2rsCVb?>PVmQn{+6>KgbZPmmt4
za~&CDA}K#6A3>b)kzY#C%w8Q=WY=a|vZ<g1-4<g{5I=mezyfl#R|W|8aj>x%&~S$P
zW8hTFJaL{5d{Oq6y`66$m%69L1}OxH`pVwvDPEi2AI(IQ>!whs9+LeSQ8F>#0OUAp
z!36AK`{6)h!lO+y%UH?`L1|FW{Jt5+uIQ6tR9A#EnV+Y*Y1dp|Qi56z*c(AI3vj{%
z#!(P!_^ww_rzL#X0IKdLL0PXPR}I<Z`vg0tg?K0(xl%eDmjkeDwFr>&F2R7AjObn&
zvo5jTNB(KE^nL4cDA%2ENUaOx;0C&r%>cJsoD{(ISh2~l4BK_;<{H3wg|^#K793!u
z;4IjUNW(SfQrx>UI^nzcYhl1cOX&GndJD?{<HJK9ile{PK|$*-QBQ@R5rzmp;g4M>
zAbh^nza!TN!cZ8V)D#JaWtWOStZg4vp;5&YeHPY#rD}>GB8z>|D3W+sZ`2IX3h@0|
z5e(?<>OoKY6qDu&MBx|DSWC#}6;<Xiz_el8p|_wZ+u7}FnXZ6VW&aVYNr;CvwH~zy
TEDR6SUI={zbGYtM*yaBM@2O~|

literal 126272
zcmeFY^;cV4)Gi#nxE6PU7AXY^MT(RHL5m#RwG?-XC6rR&;K9AcDJ@#8SlUt?Qrsc9
zy9EfzP0u^-y??>?{d5^&>^+i<wda~^KF`#|z0y@DCt)T5008754OIgGfDn6&4|(wS
z^6bZlT<nF=UQ1mSd&U0Uyl4kwFT~y&W_|zwTg=~ooaD_@CjfvG08&*p3eG=R3d*;1
zd3$odGG}#Z{bh~MA~R1o$l|H3!}`A-+Tx54@M$=eNZzM!6TeO4_sZul?Iy-yKf}lU
z6wRC~Zu_{V74fdxeU(qkxG}^l^IPb$VvR(ugp0FD^-@93Qo+jgc@BIp|Cz-0velvG
z_TqJbjkE?72>O4o|Nk|>EZ^%y34%__CSahzCieAiAV`D{rx^!4bMOIR7>OZ5^48JL
z0#JZp(m!KE*NmiL<j9btvAp;^e55)e4MrR@x!2pZ{YmEG>_U8Bf~WOL)`~h9K35Ql
z?B)UqNuY_=!BW^&$!&Tuwy0WXR92~-yJ!mTwp1HiEjR9q`H^*k5C%zfS+85g)mlK-
zW>WV7)xvWBwLIO-#64UBlWn%%6n-Cmv7NufREc4MmOfbrKSrB8L2HelIHM?gF`B(#
z5c5_<9+HY}s^_V8$t_+4aa~t$SD{*Ha!DXWW|jc92^(r#C*n4pSy!N(vYBYHT`XEr
zxA`mNEmp|Sjwop~3x>Dqk`a@L!LIg3`5H&j!<PjOJtMk#%KhlmMxs%~pYHm+YYhld
zgM%@HwP6P!`#}dKlE{{6>PMB>kMHcqb2f?tqv2wkvdK;Cpw)o;oA8TEtjMu{MSjx@
zUhh7&ftn|ut{}azDp20~n(kKafjbXTBxuIHi*$)hHel~8Qk^`rP@4p`GX%3NfoGF=
zp3AHhc<^7;L@Zb%_Hb?}eodq+_9$Ae-xO^ZVhp5@^+^a9Adhx$a1p@Ic>dt10EOko
z?q=t0^!AG|D{uryxPptqJU8e?#;GDB=B^hh26~b9e{Uyi_0ir7<fxSSADhGoNqf6=
zb2zvw@48zi6kX{qx4U)O#9sTTV}?#hjA}XP!HNqy9A0&bB5&J1u0BP*w)MM0<yBGi
z{w9U+#1c4aCg+j%UGx6c@tuSA(c<y5o|&u5>#-s~3V#crtAGSLhGpkQVi{>O@>h<m
zQ^iFkc+qJu2CJMVRyQTrR%UX`i4ngF+~n(%`D<J-rM$^3hHtId?xp8S=Tb*s8?1jZ
z&}H=opD}g8^Z1j(%bWOUhBxv>l!&}H@)#4CBZ2`I6iG|z+w~7@-*4eiRqZ8s@~<$)
zM=NrWWI#9GFis(Umy#u8m*xL4soi4?If!{L0rLa7BKOyj*#GbC&od@~pm>A4c<2$G
z5JL_qj#Kg92fbcG^4Fe1ygfwDUgd8JjDu^{E7{%i%w@LlI`8Ny-{5EDkat@Iqq7(N
z{2IcEh5!O@#8?%nN-i&h;V*n#j}{A}o735=ckk`xc@^%T?*)u@eR~Q0@C0ig-L+T@
zO8PTljWFB^Yb0D>f^lAvIob>=v}9CLXSY1PMjbO0=A|8WBRD;>wcmzV?aDXd?lG?6
zmR`tQa~?0Sc)(_Sx0lr>MT9|+(1vX^U&{(91R1^!@5Vt=ryi!RbiZGAabrEW)=&1k
zY7&7;rvIOZeHy|V=JBJyNVU5mg_utaJkJS5Z!yCavqI54(c8;eha3~XZWbbADp4oT
ztFK3nv+7}be%ni0j@%noX$mccXDS+x<wXI#R4ZOz`;gWh&%r)OA0aV{c!td1AxTXa
zTi4=$wh#3F#}-)rZqh9tNbd9>G`hV!QcJl+FrV65de5<s=a>xcsvw2PLE0&N(HO1G
zm@A^tg_8)`xIz8(5u=$)mY3Rg+L1$U9TAl_fc$@g?>}{!Ktm^e)sODGLQ<~xMvfgm
z!sk|+HO5~#{|D~7X?hUzU;-B1>)@67VtS)|T;8hr*9d#WBL&uTQACU10)`;0JjvtT
z16D9TibUqgLJ*6%)oISp`fXgp1~8a4Sn?X_vpRd%aKA)QzA6Cz@oI?D5A(bv&w*m|
zTKv`a!RV*vmr$b-q#}w5!wY)Zi&=#uYcmZkvsDe68XId{0?$#nh@rN_->}gA`C`*Q
zn{u$Gws7pj9yQ@<KF==lO}P$y-#Q<=*c}vfj{1apPPmq@(CPVcfgAL-A1Q`CnZNY>
z*F<8c+(;0LG=>##5@Yp><(ota0&yj7r%~2g5Ej4fMZGTqD`})vtUgC0CGBL8A2I}W
zid*!f;ig#Z1an`bx+jCtufQDl3NAsn<5;y|t$#@Z9R&Xi%e>R>&|b_D6J`pUkHg8f
z=tX6C`)^6bAxtE_Vj9O1=ga=@&?G|i<CbuV6H~E;b92OH+o2-P5uw_&OmqAL+@OhY
zr6NlCjD<o9_X8)35MR2J{kB8+M0=4=i3q<aqn2W~9xW5AT<ppOIjpK}z1uW6+$MtK
z{+WxO;>=$p7-1#wO2AD02ZjGg>p#8-c$G{B;b}OGcVh*UXOy&tHeP?d6woEkHxE7}
zqCg+AohYp~DJZ-NFM4@tj$fG_@MmK1d}J7awSES0V7uEvbrYkW4fro3VjmQV#?t<a
zNGutznEVC(cdTJxx}Kn`H)ZgY_U3r!W@6sROBeTEHAy8!wYD-o;E9aJyRa6z1A610
zw6}$qkoJYK1qa#L{13+lp|H#0^&wZ7u~L`HuY=tlW0(3S)s8dHd9903E^c12^{yLx
zd#OW$V)vFkIqsqBtU1_siHrWYjN?ylJ@Y@_hz;YVF%A<)w-92_S`hv}%l*Ho=>K~S
zj8mmbw&U!LY1nEuvT%V!L<y!{Y|7LgqqPjqu}FCRHzeq?W5Tcj7&{_}kQhk)h-WM?
zZWFc-d5OLIpzxP9*?YlK|6+YSVX_^=-*4ut;pT}WwZ?7FJP^0e(kj1l|9;4^-o5wN
z|3$&;{>6BVctYOfzWafFKaVGG2|ynD*BHYRFrTIWuK4_y5IX3uU>OGYf*^zS1bjZ}
zcIr2-f5nks@6P?ZPNMXe^=>@Ci(etO&i~22^$-?5;`!eE@6(vE1mrA%-NicjKJpl#
z<N$<~!K!zx=E{7eti{NMT*D&rHBexyd|1+&`0?7%dU@T=_oW1;PAHEFI^U#raiEnL
zhQEe<ItZqH#QFA<{$W-rAyz9f0?~m8psnm5;sgGM`f^-KzLPN8yctSYmQw{*@_gI_
z0_(v-?FO(S5okXD3dF30_3+sjkBhxHdDj+(<{T7&c_F8_?_-lCB3-!V3|cbPGOtx_
z4Xzl;zi8W!%`A5W;?$7_^Ti#TFw?LCUMF1R7VY=##~jkpkt?|35BWtl3eJ4)kc=+E
zP0oCWh(?74@;`cxrfGD+NpK;mINy1{$trQ=q}JJy0UEk~;!Eul`6}c(kL&Hn)Rko5
zRO+eCUYVq9#};jYvti?s1EdXRH+OYs*h-BgI5GOV5%q(Y?S-X8fp!>W4E4%->;Xgp
zCRBV5Ry+ZeD*8|<F`p%1-f-=j!K8Ws5ns2n@F?-(x#moQMR9Zi*5s_IhaI@%UMgpc
zkme^)DhxmdZ<hEAm+TqNI9*p7EKQ*!LiS`o9rOHnn0-72fjtz<Irz+7sKJC2&UsUh
zj=Ol86WvJ&BAqVcaa4c=-x*@%z@aB$9LFtN%7?A7Jjt1|WKJSPUT#ZlJzHOJAPv@h
zQ1^(-3U44j5m^YV8co0$h?RQNc@vyIT*IKnRb7;x(>~>D*yhKp)v*()OFq+%yq^o)
zt!D_vF`b?j?)2vtPei$MTTt0^wXIj8BGg>&lDK5KW!G*loQZHsxkiw+^eN4FCOD*#
z>yb>Vtji$(FUP>6^W;%>6Xjc(6F^-?w)n{#TjSM4IY^80ZGcARMSI5M!q|as(AyL{
z^XYOARg@KxWya5Q5ZNfhiUY+-Ah-A@M|KLp>j&a>lD&G6{k)|IRkh=-%HbSDaJ6Sg
zEFJg9)+6v_uafHYrCw%d-LSYh!*inMiMi*w3bcU|TVx2PHNGzLR`zkYQbRbTn;i5;
za=m*pN_Zdhq*z;RAQ<64Xm<9tK&eVtNhP2{S%d*M-mq0La_RoHC*}Cn7ZJaPKKa3%
z6c5$D{sZ>qj|DsF*(zj=-;{58{%&~)S!*X+0O-hweOp1rwnug9PPds8M%8#Z?VKDR
zn{@HCc%rjC6itDc;k-Au8NzqrjThr06i6ytyeX0v4;|*2UJh0#vFYb0FV#QW)-i&6
zGyY*BG3a~-_`_%e$i!v%)6NW$fUJyogr3$qw`b26U_xV+WQyaTKK~t)9110<;27PD
zyJM_(d6f3FW-V4cFK7B`DqG%1w@GWYq}ieVq|<}hZ|M$NJh%L`(kL-Uo7$KQtrY*K
zYk6s<vf^kh1-ZDvQH;R{Fi~o1OV(M4a%g0ts@m$OTuRoQpFy;y)7M+<mHJ^aoAW=?
zVjBQFZXLGxV?K`s7nZfYbxXeZl!lZmD(H+5-lU25NLk+Cgh&FJFj5I+hLP7)^&%@b
zE1;gE_d7g{1#A+}Lx;<J=@<2TKB7s}AK`9f5aM)^3^crWV@IaaQLuw7$}N!d;-F4W
z`;zF<bBlM3o20RR9Hf1ft=Xgw38g60F=+g&7C~?ItDQswBWr+$PMR90rRi*EVfOcH
zeClBvj6;nXQl>(T5uOwW;F*;UQRGS5wZ&wXvHGp+_8u0%uJjw<x-M!@*SyVc><bvT
z{){sVa->WE6!Ll2i19bL09`Aw+2T^a=V-0yL80<uideB_RuHK-i+)xR_i2!?Q0EXD
zp7#Ky#w>jh#OHH3wlJLpbmkQ<?O^}$BY{wt=cpx<Dkr%tnz|eB3;rng^aInZPK-wJ
zgd>iAh!2r>=<$M}YszD^TehZL>@dcD8{>%({5-S8q($ZnFTNV?8c@|iTf{{+elhxk
z!vhIYxqYTFef2q-T5%<(z)v_8{}prath(^k-D97TF}0qnvAJ_%ajKP(5JDx(w8eE?
z81oxeK+5D-_c!P%Bj*}-oacF}C-}4ePrFfjc3-+}+2L7F7H3DYo2d_$6r%$2#5S87
zn<eb}o~s?)l6PH1t*rU!vGv}Hw=;%Hoj&`P@L;krmgQ_4a?WK1`a;bKf_})-ikr28
z(&R+lu%LdcPNUK8p^v=9yuFwUDA!~6W3LYOV;5&qRf)p7E_7AW@pChN&d~1cwZ-jf
z*wb&ArOOe8ILPZSfL~tSr;j!gBV~d6jfbI};Sf`Kab8(vO$p*H00+ch|0oy)&E3`H
zuGx~qNK>EQ@*>o@pL3{v375!X#J9;kl>0X}mFoK5E%G1L133;P*VuK6jk*KAzUWCc
zk((v6M+k#>-5+JLngFV(<Dtf<5rUDtu5V91;RQP#dGNV-i1aS&B6@*P&N8k(8~{kc
zOlc9abq}1mp@i9QiFLeRaPhLe-HUV2lYJNNpxzjY(e3S=1l^(aJLE%H<<!xlnC3m)
zBQt=oDco)WovPS&S%Ugi9ZLbRp)NU8r;@(aige?$A_R%2DM5kc`mX*Y`4+RC`Gamr
zbgfb|H#S^xKOkR;p24G3?^jG77RnaxShlvsm;?ci`R3<Y<RA=jZl%$%_2}&^+?98w
zbri}Cl;bRM6e*(ib)O_3JSIhwYG58fbFp3%%FLq>9lPWw5QMJfm&MHB*bE*mm~2@~
z6z&+x6vX<J&3TvoxJ9{lOvHUbdC^1xvbGE5*vDN{+z*-q#(Xf(x%azyLc8l#-+p?j
z8isbI!J`4D&IY@XC=#I|XTsfq_Bs%qwyUqax5}C97kxl-0}6n~XTiK&@JqYv^PYE|
zH$3X_?bV<Jr<qpW`D-ux!S8|D*?-y^)2{Gv?mJQ2h78{DtwmCPG%+T4^HJMz+>CeM
z>D?})mNtugT5IJFP!>-VuS*fJ@(~Dn(<<rtrM9er;>5Xl)N{EQ=-G8A{OJ6=QPT5j
zMRhYF>fB&91@=P9Jcr8crb!ml9_t%jEl8>2&f7pLhC9k5D}9Dd+gHN*BcQmH%HQdc
z4^pZ0ea#QF6q%w3pg<{e*@wXj;w(oy?%STlUbYIGqnLM&MBZTead({`j#^iR{CHl#
zTQ*H+ykny0>gh)VVvC{dC+qCYZ*j0+oruluPG$VK10>u!u6@JhUv}bT@BH(DJ>d(L
zo}-0^W4|n3u;k>WvobU2>n^vDSGrteUVP;717oWY$ss*q4=?AvRwvfBjk}K1nyl=5
z15KzkOyQB}`m}k;<p9B=-))reqGAC#ab1Ba{sa5(2x=Hp3cJ;M6BOvW^Q_3!M!<_Y
z^)f5yTB+c1oLfBHK*0g=SyJ<ou0RV2+&Ba_;A0*_y_9_^f$_0V6xDINi8|ix-BsU3
z9$$z?DVkA)B?#!J1QDNjYN-Fytp+U((UgPi$0gmispju<%TBl8+8it;+&gNuopW&B
zJUromefSx<9{b}15$PQq@~y%*EHn-nH4z&Co~9~57N`!a=6&Dq(JLmucIorvCZyQH
z38kHN-R}N?M+5-=19I3~xQSl6=B0jcuCANkckXp;3nqe5^d>L<+Svf<hI&$#T(A5h
zfG_@hxDvs#f&xc+;k5vg^&fn6Ios*6Xtke=lb49g#_g5}eZwx^dGbc%z5}a>2U>O{
z`i6>L)MKAgT4btfIu~r@;zr+Br5{JmL-q}`3*9zZ5lzuEp7_rR1`bo)qx@M*`ISSS
zrB($iWPoUS74af+osicC4NR?(XH8V9F8w#K1+GI^I~5R=aQ>yLGum-5>QXCqqtoKf
zq(*v9i3}I@`J|^_^<bfJK^j^)XTufkGiQso=Kk@qpwjZhUw*_F;kJ}PH{2FK_%cEL
zRrm-uxmjQ(=yQq)D)eDS5v4^{YFrRuvQ-e{elq^7p8A4NQug-@klul#rR&~Zp@%kV
zUU*tm2QdI7r0&wwt{)=u8{HUFuvwFqDY{F3Q?lRY9;BjupGTB0@o}XRi;?Vu+cokY
z@_+%)HNsBQH!iRSjGrji9x;E9)XS=kKd%vNf(ak|X-<rVY9qhLMn*-abUhqT24HA|
z<?5SZt5Ty)1V|y^HM{vX7<5K9HKW;w%TEQO290EappqZQ#{vE-i)k;&{?u}7t}0Ap
z{D5M3E5(+?A2(<aZ9lutHRB=$Sqpv=#bu*JiyI@P^yh28QT!l5RD%LsyY`G#1-s=)
z+u_>PE&`e_pm+f2zgva{an}>}e^}vHOpT`J<-m4gs~@R`eGP8w_FJ|6Y=I7$1IAny
z^4l01>3gF5K0FIdrs9H+U<=#S5A?VZED&LY_<Vq?iX~D7(ntrAJI_xc5X@ZZcl=Q$
z8LqqUslk}WSa&r@RVdSqdGUF-|JN}!M><^|8Xi%un$+?O?<8IOleLKyIyn)FspQzm
z7>~I?BA@WUr}{)@SQN+=F}4=zO!lv6z14x`-97EM(t_ej<A(am$;Dn0e)qjA2lQ1?
zPplL)Rm)XFFoUvJq5JwmS~R>yqH4nAW}>*lyAa9pDJ!$td)z6*Gwh+QC`2Bn0iKEn
zEPD1P&5yp1X@C953ST~Z>atoe(n>Hat+FwccWeyq*6J`}?wsW5f?4E$77)VV;Q?Sf
zXb)u9YF;KRsv|4{f?lXI7wlv`3{=!}-0S>Q*UdhMDLc<-KpIn#5)@L%lXm!%nppct
z4qc0E?~L@|Lilm_V+sBZbs#>W)gY|&*KfcZUx;NQadEM6aq;+0GoFa|xStbbAU!VL
z1MwpgCnYb=AB{i|YZ&`q4|ai?Rc;*+-xo)BVb6UBhH5vuTByQ_6nG`Y2)f?J7GQvX
za(;g85$=Yr8_GP@vQPgYj0O5;XRhhvic(snPyMfZ2h<I9a>J!2B#UXP3(8ZG=+ft>
z=QCJ-kP@J3Y?It@9sW3xAY@t7@!?5eSA7V1WABk8YBDQZK}s&{fSwPG=(AkPI9rGF
z)>YfAQ~^Ps+a*2!tY|px7Ae3eV}(&2P9Xo$nO}#hftcO7E59Boc%0R@L^QRDo?;@u
ztg%l|ukO4HZ9G79f2;F*tXtMYEWH0UWwC`RD4`PQspDyasPJgJ@t1LTx4FzrV$L<8
z{?Ai6WnyAiL*pXvI(wL<FY6x2JPS?N=BwdtF8XO0gAK0Ni?&K_3VSAQntZLf*Up%^
z&%ysFLJ5nw?C-|f!c9s~%xI~l(#>!+XaIA#!L<ShU5ZutxauSxJac!So_<#=g~bo&
zEbDHg8L}-tV`hlMh@ep-4Sq^AI9C!Ip0|oU7&Bj-YZ12JLqwXC9u8}>;7!Wqwlp6O
z&FUdK5VZ8i_CuI{bhsj@_}d8TC%SKcD%Sn9`xvjqsr`Fw9_k`1UjocRTOqc+<4|7j
zqkM%rZ_Dz+V<(Kz5H>~zrG(K#vFYq*-2#l#I4E6p7Dt#*mR*guEKZF&9yj?jyp0$1
z^%0xSv%BtZ(<7SPfy?9rS4Y-!iXhFN`LJ!YuI*Npux!Zlcs&h`G>(&79!+urPfdgb
z@nSX%kNC@ho_hGzy#14P_hR9CNynx&fD*2DndM5OBj1TWLNQXY%=eof^RzE1rg=Ni
z3-3sXy**~EHX@UD6yJ1cI8Kg9+L^(jXn8n&X^yK!rQzz;LQ0QHkPVRj5F|HSgl6Kf
z2xHbWa>NzgoXx?+CtCqJ9z_pL3_jH|2*M_>QDxqYC%He5Tq64#8<cLMFE!cD0@8Z+
z+Jb1~$&c``@lErdi{*1ysTjSv-?RF(v(N3g7XqnB>0UT4fSI9roHx9{YR)+mcp$xb
zCvIzUc4{4P;aMJ1K99t5zZhnaoPZhsWT8PO1Lr<!xu(v!l!MqzUQZ6_+GiqgXlL78
zB-<&zFkLXZ4A1v}3q;6)$o^2eliJs<SN%MiDSvGD5zZ^6SxpYeSDQSX*9h~zC-qVG
z=1YBxx`u^9y78>>muK>wHOYgXr0Qe_Y5A<K&?TV=YJ(??`4X3Z&r)O4_)RCZW6NG~
zJ;xYff}>+M9-19&)uSDZ)Ah8F&E`g@3&}6Xh&dTbkbTvC>|A4hqEM4Z-a!)VVFTx{
za$>}k5e^aV)Rr`C?09k;ij~RUfz3(=?|4&nFa);ihn5W<)D;c6ztnNRjr~?(XPx+g
z_JCz^-EaY8j9#=-9C123)|7TAoyGOQ197HE1+_?e!MiDCJaK1=#IFXw*?@*Gz&o#;
zZ?CP+$WAm|_qEw`TGNOfS{}b8gAFBUuE4?a8L+ef_a2X)Ys)9}!48@7g5=M>N&+6|
z^&@g9P;%l0SCo<6_(%9h*wCM!hGCo%P*l=zl`VNt`uIaX8g8L!*CUr$wdK{7sFe&b
z9Y{62mMC@;k%)MyUA=-ACYIMcBS!45hnT}J?1WmdTmSl!T(i+%W|7^&%L|fo^}1sO
zN33EQ7Jn4P6QvG)gH3lwPMOz^#g6;t5)>=ZEO-Ch=NV<f_PW<T=wX;`Y;JjKtS2&-
z=ye<yYD&zyzz@|AbV+$B40bf^lu5G6ymd%s%1+z_JERgRZbF&@6;C=}<HD)ga!O+T
z_+V~aBIqw=#xAI=+bzrD#!#U#le(gQ^l>T0+?4{7xXx9UrFX>k(9UqKF8rCg&PreG
zRfWvv-0^AoFRrf^QV^i?@{;~+IIFk}_kUVCshFE-gB?@Fr*_7-0%l1qiRE<`)8z>)
zwbPYuI1)W8Cv8m(o@IQEXC}qqnta_lF@yX5{Kl5Rfh7-zX>o^N)@q{g!tSahDeodZ
z(DTN~T?=dbtNDBXgO%PM9}dKd*g>%fj%XG9EiXA<FSzQ5flOw0SyoH7u%TGAV7eQL
zOFOe10N{-We!ECmUep;yY_*V6$7V|}<`K?O1P&qGcgF1ohPI{t?^%F?pV5NaI;&3*
z$&eNR0~H&vg1X@3JhG*1%8iZeqvJzwN}qx&x-n`$MsJdOne|Jv5ogd3q!?Snw?to}
z>(sfKe|k7!+!}$BH6jDni1vzeq&TswFC$A!#W!B;){rhT5p3eYK?VgZJ{)yl>W-h#
zQc2M@jRRl-Flk6YTlmUwS45^t6)kt(O+0LWZ}~|T$Up{1rljt!5Py0`_9&*q^J&T#
zr>#;C?ZxIapZz$}&9eFj)>h#htw9V9*Y1w<e30M5GJ9S(jcJjz6K4ef0ymkk7#eh@
z?}{4~JX;XRLPzeAEj$D{Hc7wA1$goAv0$~Dhs_U%HW$BdWGuGu%2l$p(jmi(C&2EM
zUbAYKAw+nGqP8OXnpNMdfBJKWzLh;8oWV<>_*)wF6isE7Nn*s7$ObV=!%)}CbYM8k
z$|BC6ALyC^_qiKxzbh4sDoK5-#?v%Le+>KS&8u;Wj%SHKf?WGwm*GILXXKoDz=Qrc
z^A86FEcYN_ye@6&qYXCEq(Hu1Z5@L(-R)|yBRb|n8JFtgs(7aTW}R&V0%tofy?mfF
zm(ONATz=M!=K^fD*<2K0Z_^-JU;Fs0upNa-PBYf7%>Kcl!>v%zTYalac7Unnar2Zn
z$;sZh8vmiyP|o>1wp5^a*cUpgv;RHibpr=dRzK60ZF9e_3Rcyl1arES>y|9VGSG`4
z^Q)%GA6<pWb~6gl>2Vdc=vzt~(+K?PW?XE}__-iqfqa-}FpQ00cU7V5g(7d=!`7<0
zP{Ljda&atGoIF0c{n^D6Tl8adsbk9*F&6Sv^G&gN8Mv#c=c23Q5rjC%_jK}5o3ZJG
z8STttw-?OPpY~%(H+d6XeH-wQEVe6Z<sO8yL$TDDH1PM=EVSFyOr+3sEJ&HR$v!h&
zcZ$U|MFTV(yg!->Rjh7;AwjQlI$I)GXcZ-r3Tz0vUR;yX9CI~rdht2)*9gml$CRYM
z88k%q8!Q9l_1<pv(AnEf3~?#6ba|wFz+dn7*BICD0fZB6(nz~C)Lu(>NO_ZaTN2$(
zB%j3DBsv~rgMd3S@dp(8Ww!L*)}0Gb)ttIMl6JV3oo**wggo7t8Gr=pa~vGUQ8una
zj%gU+Gq0>(pqTBMkHM04i3TtH4R5!XOY=d-vPUuqVaP(;#z}a>Ynxv2f>6fJtR2(}
z<)R!FDdx)hOy*>zow40T+943HqvxQCW0jWlT72z^3yy{m*7(R%eg-2x|2)cqERf#)
zdl_QlG#5agJ_?|x4~(FNW<e`IH9kN}&gBy1NWmx}U|qx(KD$P-1Y-K<9jnPo6M)mg
zZ#TXNehnHd)9%IF1{}wHTMkGq`!R~eb8cE(<msGhPM!jco?A_EssDrH^#u`HE!l3a
zu;uKlg|Y93n!OHgtb;DTXtaMAUT%OyU0dBTc_uFN@Hi0dRr+=Edg-h2Ems%=w$Lu%
zgM*P5zr2=2Jc4=j2(4rCTJ>+Z^exV{MqYn05)(`){ocGOTU_IQ+Uka5?1!w}&XEov
zP)}ThU5u-ASI3dQ3z<RJZu-F{%FF-&%7&9xz0+0FD$z%rRZUoQ_qN8i1`5m(h4IJT
z{&@SF33o&Qb8gK2tKO0J^qG$mt{g4$=?O8y5avP?42(N|1YUgdnc=~gb74Tg^UH|1
zY^(a%Gk9B~#szXEh1~`ds1cQUY9lLI;bNn;37&Sg;dobC{lnu~t_rAg<~*u4XzWVI
zo~qW!_%Ro{X2sx`koV+Cv>QP&{z}>9^Ojp~9f%DNeIVv>sff+;*L6e;8H->&=iH|1
zZkh<pB{m6uO;eJ~F<6L)WIj<QVYnH8T$eI9-?5Z|ZX?Sk^X6yJJ1*_?5I?^zPd)5b
z2wj|cACI$^zuR~9_;w}miRZXXVA2A%CXCs-HFT-pS`xImYjpp?4#L(v12TU~A`X5Y
zEkIjAa1QCDz1=f%QA}D=@pn|89#_?kbdizE-IOnb{^yU5)+FkDQ=5^5rVXUyV$^YL
zJaJM0O$ja+0ETS$Sh=CN>oA3o7fT~<yFl(F9j<A>dq%)4rTqAN!UF5r2+_!g)NA7y
zK0-m={=y;Ikcm64s(tNf{LHcw&Wy$fSS<M94(-47WNTQD^_^~v^ys<V64CL{Piq<{
z1H{HdEFg-6=-jh4#k)vK+W2Dj)bMcr&D`Aax)hO}JFH3Hxuv6@{cHIuG+lzyqNl1o
zNTNL5syV4L(c(+gjM9K6LQz`6$?B=V54Ca~6roo)!mXo3rp&~^4cq>@3=qt?ewYV|
zdBYm?B%|zh8n|iRm3U@uYv36wcCvCG26uwRM~9PV697^UxTb}XtW%9o_^|^hyBodV
zUNZY*gVN)CwVrLWxvQsZ5tu4Qch7uDnvD2H0uEt}gc%u1E`vCs1fuxvqCcc1S>V2f
zx23>V@Z*miI{Q~{bl!}c1T_}M*ocvJn7x?CU$2pcdurgOhrn}y9dX3m$ClmOoSDN@
zw7Q`Wmk`6et;Y$bf_6@cc~o*7+g(`6F}1L9@`9`#M=usyQ7vAG4`W23^*|@AP#k66
zz$xjymKlM2ys3RkDcQSjw%`|b+xICKZdBZHy1m$(Jy~zaiEd9A)F=f-zcWQj9zQ_T
za<VOCH*37WHY@)%5UikPJC1RZ!`gLjy}*_HU?UYE5Le8I9V(9-R(<ll4d$g1!WfXQ
zx~5c5_eT=37E)*ub1?feaH#g-oF*?Mw~9PiBK7T3o^qD!rALyvK8Zu530)wDTW|*c
zh%HFC&GjJ2XY7JVCe%aIV7{fG|3Pp{xz?4+ZK8*^%o|oYF_1=@RRUY*`)7TDzv-6s
zB-^8*e|wj|%$qfA>hnk6y!k?zt_I4rcV4Cqcw-(kTHb)K&u|Y%i<zluW8gEw&!f)l
zqsdd;KU|c+x?kn&2FjlFrJ#9$V$wAV218HKZUQ+v{wO#ny1Tr`J!ixS?&<d4zWmB?
z3HUH9I@YVgae+<M>|>Df?OToJ>FP*gVV%`1*t52!?TQmyn$6#nVH2baEN|GSR9FE!
z%lNhqSI4OeY_p`@L}3jawO%)=LgcX~G!sRdfjHOOl>MF?Hfbx@hBewV;*T)WbvVHd
zV}>}PC&Qw>OY@+G4hq?Az?an%LE71y+ljXaTYb;g=bWWqji=b7vO<o7KfB#xgok3q
zWUK4$RmAr9DN`i5WH;vh<ccbKZC}~|Q_8)?GVlQW72bXhGrXEtFK!uN7%olO%ns(+
z2fHroj35GRr}eWz=l$H=Ok$5YV=^>j-XmS|_K~i@fwj@yj-zy`4bigRjroR#J>uRf
zB|`L3nu(YqPfe7j6Tw29X{aF^ylVLo1>lRT<|9ICEx@HaKAvIU{?CZ6Wm3SU7Rh?#
zIZ6BRh;aj4E@?4wU#&c`HK#i8)y-;K;Rn6BV-U8;;)SRSouBJw_?ULpEH#|_EJuW6
zfc`G6^o1QNi-J4xE_R5dy-3g2HPJz?()v&Bm=l36=QO7PD3U|vNlL-FR%J&svz`-w
zSQOAAlBLqp#tz^6VLjoy(s^vxRoqps9)b$Cry?sx>5<4UW4TA$srby<)`_xDQHhKy
zeOl6;hY+tE&uf?vv(LS+)mB}1H!MkSV0``%Oi2O2Hq)bKDeRdil_n|g0u^mm{^*;K
zYcaffE8ZTZLR@4%hgn%2X}7iyWZ4wXE>NWsokX}bI}TCwVgx?zD+xx5XtOK63wIfH
zJY;d}<c>~~#J#u_w9GZ0W=fyL&X8bfLM5G402JrmYOY)COJ2Mr`8)?9WUs#n4S>Ok
z&hSCP8c1!sAfNE}grH0+rv3$AIeviGhl`%AZy@GL>;_-<0oDb?(CtOpICZ6ky#68H
zZ;n=YV$XgVwkyid5iR1#tG`V>o>RS9Pus!RF8+|$-Lws5@zyVZS4C+q2*8I0%&hkN
zjqQm)9+i<<CRTb;b7*va#gSn_PN-OvJ_oHkU+4ckD&iOaDs~)95<eB5uehj*KMK8s
zHAeaGA@Rg~4+<(yCKpAt3rXs7e^JVPDD$$^x|IRy;gyPK`#h~@SLx!hjUZXkSPT8%
zTeg^u;1(2PHZ>B)oU2ZoEpzTley)u^`rxR=rhD~BXrG}d|EJz}!FVIDAnKfyZnbA)
zs{)7+4Mm)RD?aSDzx>_yc^sxhqVe!jb?*BIF?8NPBH4K!fmAaTJdhGYq{{ex2!1m4
zW>OI!l$?(Wk0m~GM(H~<bbEhn3t!f-m9|y1<?e#+2=wLIiZ*SfFt{_>H?Adf6QHy%
zuX(JfialQ(&da_wBlK4<0`>upSC8(%Y!aFBBcMKm90G~{LCgkkK|83sJ7&x!X>F6#
z#(A>8Gl6wXcV=Mbq@)`K^z+IYB9)7KnbbEK9OVkTGelZHdEKoJdXa)9D&6Lf$79h;
zK|y{8bc50tyvdWY%}Jxhi)}b{(fy2HVzwhY(_Qp*fp2JPD<^J3Qc6=INM%wr!Zc#t
zex^b8IqwGnSxQhNXVeU5Rx>qkP`Q2$eO+O8nMwZ0^aE0XtZvzSQ)J7xBShGDHh_jK
zS0`I*q3O)F`<kp1fp`>RWbRh_(mZGfWM8gawF1NuEyqYiP6jnr370j8r|{K3HL}qJ
z+>^CtezFpE0H)^nDV*G?*4^`P%WVk@ssCHb^=N-qv8?Xyh}kTNU*GzI2(q#fUmcq_
zJ-%`)YR~-EN)1k=)^J9!wJZ}aRZTNaXNxz%kE0I$K-tw5-+VK9gc4&S?huLAGSDi;
z4!wavye<sQS(hohfogW4|4>R0>~VVFiIz3ct;J7P`SIl0-B{WMr40g<g`xdAMl~be
z_+#ZJQe`0KH`uz&iDZmGMw%F=ktw(sR8V$vrr223VLYn}r^ge+32gvZq^D8#0=%O9
zIj65Hp*;z4SP#cgnh{4tRQmX7^lp_~_~gXF&%jPX@Q3uSVSwoa6a=Y<wt;oyP&zF{
zExoo^{(@2d@@J_)wAgv8$dKJ9{X;R6(e&^>kbdA9=l)JjKbolUdrr0UtAcwBD=>6_
z>`!Dve0|Fohu)VGrgCpZ8q591GCN?4ImCk4T7FUlHt8vp3**oy_0(PUuqM|At+wmr
zHi(g{VHHt2mTGPJ=Q?YNb%8M7Y<6kq0OJ5z_gyG}u>~A0nrb`Y_M+%KUDu?3zkuEJ
zxF-mzR{W}0b$#!YB54=+?gQ0&x1{_CKVi4#JpoIs%v8de;)FmJPoXyIF!Rl=$)%l<
zmVt&B!SYN0D^XrTK)K^WBSZ^QMBzQ5MNf-If1`_$E<O1QbY_F9VfJ}d<vsC?U3hV`
zOgi&!-Hh$-Z1FTiK8Z2a7{<A$VsMv<7xhre_fn6&5V=ngYB+iT4qjstFo%!0pS+L5
z<79!1GhKu;f-;ser(WlEwd5;ZQFpNc<7&8E#!v1DY_=vA+a1ScfaVvQJID3wmU(xt
z)m!f;4STNUFW-0C*fo@=7nUx;PnClcgu_A9^#B#SmzQl=C2Gdmgq?f^#wHzx@tuek
zA9;%O>9;M9KDYlI#v=lhrPrLsqnx?PhwHaZj{}SImD7NCkrr|Ltp@GRREHtqPqkF`
zl1^=y11Zum16i78lwTqpb!eH>_2{#8(nUlsE#wO>-q?zo;dZ2W2=jLph2g?(Qs`+h
zgsM+r;845kRoHQ<w>fhZ3kPso==G2SW|?j|H$!U@mZLu#s6cr86!3?5kqAXtZ;t#2
zrU1imiw0%^I6j`Kgg!>T;GVV7v4o-cQ`=7D#mR$-Q#&uD5CJ`OKT7K8i#{}CtEMlB
zBA9a?GcY8XT`(VIYcJMii9&*CFVA!JNl#J^9T(Gr{8mj&d^h0JkKH!4hOOIke@QQo
z5FC1Kh4P^PWkHKX%F)bbDCPFJ{9uS=i*AvvJ^h+RD1ud3r6}h7H6`gSJ;QQ%gBrIA
zTUkl$_l}#;LKjHJrMw%B8-PxW?`b_56&0SCn87nsKi1R(eUt*|Z7<jr`Rrd)fhHrN
zG9z@wRzg>ZW{E1iZO)m9I_qo`OyRVJ_}T_mEa<eUl90h!ez4pB*@Qg7CDJK6yd;+3
z9XNh)W3b}}7Bg;It&UfMY_wiUHzs>>KXMh!ge3MRA2UM+7ONp$VyQFR8Rai=SBMrh
z5{|BWxe>>a*OOmg8n8J{?3@|?`CfShQ;Z)R+Y+_c@1U(j{aFMEM>^`U<(gw>#z1$8
zEIEN#x2uS)=(M*;*v%O89hwd18+szdR!oKihsu}3CD3m_>YfnqBHD9O3GuLb$-9P-
zX7z`GGacf34yy~ap?>ac1uEZ494WnB(n%`Ye@7qQjC{S$D^(S_PE-<X_j{7MbkD32
zYR&@^!j=$!{R&%q-trCoTa8<dy>wF7N_e=*flqwLR0z`;?Q0Ac16B*dB@cjvI7wx!
z;0l1rMRn6s!82!^u=u?I2m*21&DQ$l>!_v4infTmwD)1v54Vryq0O}Mv0YF`r<ffL
z`MgSoR&wMLj;x8z$i^QAF)K}J(jbi&Fi`X%!@1%1xGC9w@lSC|{!)wBgu30fdlp8T
zP4F31wD=o55%zhJ_7KJb(NhWGE4>N%Xa&RN*4hDbzY=rT4lUyvK#DbG+FxPDfwJU6
zj%-WV=2C6(XwJ1%2UNh%uz^0fLwP$Qb0<t7M<qaKo>0r0v!T%D<6`rAEK?DKbd^4H
zE|z;!GWdD`N?npmA-2>&^87L)SASGd(Sj6<sq!;5h4Gk6KhFxw<sed@0w~5B9{Ejl
zbEpBGS&O4sem9qzOvO88JQIBME#zA@!9y*f&dL9*f(gK=><<n9Adq*;dBU}(c>Ay?
zdIN~SE!IU;R9V0Nmg>p>gf680EJ($X9M$0|0^O^Ynd+F0?51KYT3A*`ffe)n3A0lU
zSM(+k`!!w1oSCIPk}x=q9I2itXE`Alf84K;31?e>^lA1>N72FziX>q^*E(z^L`1*>
z)e`J5Z&d|(tD1P&VZqY=z*9D34<}xo)3sm4+pMbOhcV#;dwSTADluIwR8#`u$0c?C
zO5s0K(C=}_81TFD*K54_HH4w!-O-fz!$)<!RJPGc7kuz%<%wW~m>71<{rTzf<z?<o
zW(H#f6;e1OXL}*CmiHtISVM`9s_aT>`PkM-5ECWI<t(%H^>ct@_3jV=_8XQJH~K9!
z*4$W{S{clGVm!!@Spe&VmCd=;?%Vj?ZPvy{OCRyjGwvRR>cyV%OH(t*fTvQXL3jd*
z@y9&a*m+bzF&Bv`xK^T3c4g0)!{kLWa0v!}q@)Is46woYV{<#+s?!!Q%>ZVw6eDU6
z_~^eVympRAAe1s@#wH?ITWKILqKG{%Rr6H}Tn}-zyM_XvxUuY#2}vBa&=hGU8VTb`
zKJYw9lTwX~$>?F$ql%I`gk{VnHl_!OWBEfy&<T_E{8$?I5vS|xO#e*NlflKBIaPNd
zCyZiNk9%<!^4Vr6Hg=%mwE5}H?G55sX*YrHm>?;Oh5sbND$=sqOhcaY5^9}G5bzl;
zC6~OANE+Oq(tR^*?D3K<Ja|N;hhQxzj{0kRoVlICoSr2>aX#jpJY7992-k+Ie9<3A
zk2>D}fDawT`qH1U*#+L6g+!NrDmO*DyyOb*`u9vju4l!2w;UNPw7=}-*q?!^;%y@s
z&w?m&?crpJ1}AalaR*Bd*;tN_SzvwCL;XiowiG9sbnM!LhQwvYj;Jhi3-LU3eQ86F
z&Ont>*7vriUY~$;K3H{C$F!La>QH)Iq6c1?9v~CFgQ%1`>`p!6B=pa8wCSTNP!=4c
zHn)_}{E8jTC1-I}2|L<ezX-`qy9d*cnUS7K%Y6Nl6!C>6{Qz>okgmn<#UneTZotkc
zLo^=K7<u|#W3VWCNde=_9EE$s^p5m(*g(%JP_dPujvRV4DBhIbMFy~Z=}*T#X#Y{L
z{((tj#v7kb(qI;S+|9|w^8y$itR+sENIg!{sK9de`Jc$55ek{^6I3|y!3Tfg)Q3g2
zS?zK2q;q7-1Kb*lBw;DpyV#x$VJ;~!FME(!zWs<v*HX5S%}>CEq=3956NySO&p5St
zqmLb}W5f2ry%X2xt3+?=1zf`pr%D88cw=^E+ZmusORP2w#0Q{tW}%|&dPq4X0GP*h
z&wFat8ofi8^Fo*+jJH&)!_4@vn|_|ec@(bhyig!%2a!+g>JNMQ?pLhuv%7{vwu^4c
zwdch>uUu2?c*Rfj|NTMi{jk(#Sd&P4D9wlhkPrPS!pCtnw;FvoqT~9Tb25uY=~3mE
zf`5h_2j?22M(7cSsvbX6dy07+8fwcZKP5Yy%ur>hLGlb|@2d~m-U|e!)ke3aWrq;f
zQ!UaB7Lm)vDlMjjv?Zo~;VWNLcsaTCD1eZabl75KRbmuTZ?%x!DvIGobh?r;YJ45U
zLZ-%#bOFxaR7*bk$#=C$^z;70)1QK>R(DYHNyfU@DfShEf+J#h_^NN;hpV!!r6Von
zX|gr_p_;sZ9j;8qVCcL~&#laFgSLRQ-!RhHC-S@#mBS&&o|&s$8rFnI!-qTEgirc5
zuSP!IL}kGB%{~X?ruKjA5*$@wQwuk{pUA?2@JEtT`VLFY`MmgHz4qdx9Y?T1L4oRK
z;%08C{NCFq;!SeLNu?+ZnB8aH1nPv54p8l&fM&qJ)NbtZnr|Z!JdtTx-iz<0&+Zwn
z|EXbmT98e^*(CJWbB^x_$Tij7ePZrQERmn;+Q?cWqnp&vrD6-Y;p}?1HR620S<|iA
z6vMBtkXK9$WmCoNrx=uwOyZ94WKfo^X*c=2iERqlae3g4miT^*5D8%9)U&fE;V<<z
z%H0prcAz?vwf9JcQhdhn6)}D+msbJeq`Py|3&Y!S$5SOiG*PF&r84LxOGJM4*hL?b
zOC`Ti?i9iC4?kKt?BO_g@0p|nS*chl34q>KI^PtzudY;+!oMPvV0yIn-@|+Wh?@|}
zh8yT+N(4bL(FvnI#t-)~?M`ddjQpk3a_2#K81~0l@kv!7EUv-WYM^Y?VCaXhbg~nT
z?p<zG9aYO(jyclvWcF$fAO`Os3C;OZADvUe;`4Qxwe4^D{ZB2&cCIG9#l+nl1W*0;
zIU6XlsN#8+Cp*fk2L5I&5?O<euzQ$8%9s5ur1g*{fW4@|K{Qp~q(LEZ&RdL%^gUd-
zBkT#5r&(eHhzXU(iFd!k7X0Lbs2V9X-GI74XW-)V$J+;K{Bh`5=#AjwBF2nLWB>=^
zc@YReSVSLQyCfH)Zazt?8-PnwoBqj<zz!UMnI1oH>85QJv<PRn7s3uZlWraI69O`)
zE1IH|8(7E10NuJ|xcQov5N$l&yHK?M?}v*G8v@3|Kuv-&!35BtQ&j0HwyNJ$_r;;x
z6uq4#wKrY}aw-UgghE>fJipx-1<Jb<vU{n>an=X|rm3Y%{TjfO(A%y7c8y~kQGR;0
z0`<ulYrOo2`}n{_>u%+!C@j1qM$1A@SMJ)MHP4I-R9spZ1$AU<V{1FYH7_2VJQ07K
zB1Uv!RFiDaDgT13D{FA?fNwtkT1bN<s?==LK9lND=<;_0?`^;b*C)<aI*;6EJW{tm
zmACLhDBucLI2=L;kDiy~f#v=rX86BSH#95pBdvR$c)v}T)X5*0_Xlc!Q#1lI(6|!+
z!s`}^0g6Hji#tyKY;5Pq9TwfJk6C@Yhx(GQNaX#|s4~NrYTWX)Hk*|B#S737NzZRJ
zLLHvGsaDHmwZ%^cTzOCI)8@~(>*6a#mO+*!BM#nE*WoDwajpNpc4HBb#rEr-=5c+V
zc0gr)$nXcIh8G><g|VB;1H8IYj^r?*4HQV445=GC+xhggQkY0|!9pA~jQ7weOWZwM
zcmiV*y_AWwmsb_-K?Hcx$p#1(yhh@Bo7mZ$uHwoX_zjD=i*dc1Yl^;SiEJmmGp|JH
z={Mvs;+ra?hM#!EU{Yoh+jFYVNQp(Y@0uh2m88S=JEo0?(O@CL+C0PDGca4m<<&Lr
zipS`;OMmVR>0v={N}$fbH@)KWQ+TX4W}NO~x98Hhk%4U(cn2w>Qmh^iSdl4q9Ao{g
zrS(|~3*^;Jx;2T;4JSn3ePZ%lxR3Z475CBC^1G&|h4|0a$yhgN8Kn(!N@bLw8R@jQ
zoHhFMu~NjE^h(%ht5Hp$ro`8G{_eAgP@d7O^%OP?R0-ehx#fJ}63J0DMMng%=A^$B
zOzK}rSB)#umi^GmnVm1kX8~ya2PhXy_P!!6hZZ2bo;^^s9nSCNWFUytEf3*xs(Of(
ziyqobJ{cFi^!qp1MkhUC;10CTfL1^_k_5z5{aX7Fvm<U#uhR8mJ=pdK0wkbO$UO-3
zC`Pr*y~|$mAt>ef9$IGp4r`{jHJ^OzB6#s=Ig%?Heg!C=1RywQkisW=s-XJc*!X1$
zHUbNRGO`L#e4zBdDeABggkQz9Zo_SUeKRaT{?Qn!6fThYH(RRGTC2~Xjr|{n0H#;W
zxfFq8&-g<9Cue8!VoC`?>rtM@cGt{^pFg|Qp$|mg_Td-`fcXvp)8oE#@&ld-7S%u{
z1||MO`Tn1vqP1Ur+Z6J(OkdNO`;lV1toGy(sMH4iuGV__%~bS^Jh{9%raWL@<?|3m
z*iF|^Wjghgr`Vx0Jxu(Asl;kFpHy`w2kj<DQN`koFL%AniCgZJfK>k9ym<T+_J~vt
zOacm;yNR*4vkTS7g#*QZLGZ75>ty1TA>9OM!AK&gsfCy17-gz7>~F>~VkCNzM-xSL
zGl$-mbTjYemi*td0Dt5x7%GwXa4$8B4|VnigUzN7%)L)=AYWBQ`YEUX+j;M^%t4mR
zEJ$N4%-kUNUu&Z;m?9?(|0Q!cM-NmACLQOi1i<#>3y;sX3^~>o4FZ4FyKY9mlU)3J
zzL~x0$VQTO4k4#ar+7RYoy~wg3ItKW@S<F-zvR^Z1*+KWuYj$5p%7mt&e)u+CVEwO
zDArKrdXDQQts;Qp4^A{Sw*9X)$shgoyKVmte%WDD-ED~&m<9+1A{bmq*i(9npWI}J
z;bv@tpHhQjFE0RXcVXOZtEPixkY3!1sWx&2L2SJ7n(5;U-(b5?t&tM*$scH*r4+Vv
zCC?T0DilwKqzv5<=1ZBZ;e8>nd}*_kn7Y;C8yYM!K%0X7br_rNFuU5Z75t(Z^SjUb
zRAJeF$TTp3dX7TT-BYgYdm9ec;|^l|q<)tIta$N?$;icQDO<7y?*w0nXgdh!TDnT4
zX;FDNTVQ0`9nU)JXN%C#lc$CaG~?~DNh!OkGy{^{`Wvzu7!ekMKPLg$84uUiw(oWK
zq#DiSuseD;Os#hD5!g$r0j0;c`|d&w(Gmy$u4}8HKUn!q4O{(qK3?kBzu9B1RpwTQ
z?P|Xy&2T=-%85RtuYNt&7)1_1c_fB&)T+M#URwA7Z22<4>kE|~*<v}mi;uI)4+Cio
zVObRZcrn==)P_gaq;(mfg-qXr;ci5&9>xWY)JI&<c~-N(O8r29-DdH4D|_%icGoY=
zzFId3-eo;3w8@vmY+a}b?X%SKO{*}|(aaX85x;K~Y&rKv;37tyo2oyw$MvL~_9#)p
zQhP2}GhmdtlcG`|u(oVGexpj{y9~a*v*Jm5=&}@$cW<+PS@QZd7NU-Lr9lQ=nPxAC
z`}341;}wP7EPl~))9z@lnix^k3|HLDn7(*fuNj30cy@0V{5fjSwfh`GiRT3q6f$^G
zrr!<L^qhJ9S>pUiTZo}kFh|Sj3PW$Z3z))Z{R!1YOgQ3kVTWYgICx~iJyWEf8{`sI
z;bS~axY!C^N9VtLXOY*N&0^)gfv^Y!#^reU*G6Sc`@OYDN%#tZc}j{B1V9`YE@cZ7
z?qz)Zba^ci4$5a-0HJ;-t#A0y@ZYeC#rbqNblAHaJHQU)XdZkbAU?$X6uwf%<a%`7
zSF36D&yvK%(!_+-l;869u8YysKT{?WQvpW`L&vD9^_VH<u17brJz{0At^-8((x^7o
zIKbu8j57(PkY&E{a~u%86={U8&IhMboJP!tRD~LlyBJ=}H{6*5&9IufWJ|(uacw^I
z;%w4J@aNy88yiOy#X^vQ4YXcU`i@2Z?DJ)dhsO0}TJ9EjhzFaDZ-`K|C%oz84ea6i
znW{-|$#s5`K4X#Nv5HJCA^1mJIV+M)j<tjW<cy)@tSM?C2p6-E4bpIwa|gH}0;4dp
zCn6tbL~oDThd(J}erAozrDYv){OHYLx7K{k#H`k7f7R_jjAqIC`9krGp;LVWFA6OU
zQor-7xni!Sjot5hTC9C#L@lX+lJd=e&&uEquN|o*Sq*&GjR(RZ=c`~-z(JBByYc=P
zO<&>A^!L5LF-CVI-616n(k+dE2-4CZDczuShe${`6a=KDV+c~xAs`5&8-~=7vHiyT
z`}zF?uibmkJ<mDMInO<F4mtgR+l2n~@@h>B*b|N1=_UPkTNT31xr<fzj@Q`?bW_5s
zSsvx7ytdFf&FD-JQp`jcLDx1u&oE2=3=(WsIW<KHUN<(Le{Ubh*&+xaD=Pv(L!%>=
z=mkP!!);W17YV00udHGH(Da8QgCQ_N=)$s3L{ei+N)81vfQXej75B+JmG{%z+R83L
z+5MHY-1q9<I#4t=;`{^AqI|+Js8y`;)W&ytDbTmy*vh3?Eez;DcbT+=o_TcZ<&a94
zP)8a-IgE=#h<cdtxG4ZIY4+;1pr4S>O$l+iRQ|g9T$HzN{U7R}-n?@Tb!tLSO2qOA
z`0e-IX%pBfG?KVTCiF3`6Iy5Js5cX50_UG*3+dDRq;AN_WfM^5#D?$6p|13A`M?SP
z2wKgj5ULorbsSI@w+o7DfEEs4$s1l8_1V?A^;Q0c;y(wJ(&6y5RGXV-g^Xqb{t5U7
zj4Rn>)w(=|Q@_q$jql-PshgrZjGp}++@l9bp~0uwGP3G@nL(+z_3H1>aa9e7Rv<#a
zMa^9P1;5-_&3UNpFS`!O)3Q_Ms9;^DYhne(e5A(DIHS7vcfw3fXbmU0g}9!<QG_fz
zi}c$trgcq=4B8sWk*c@md^u>K2{@`aZ(4U~q!}GN4Q%s=UPn<UN(m_`1F|{y82|T)
z2W@f<^0X_(1eSqr_MxtcxlzW6I=?-X6%s7q6zy-K4X-@q2#Sh`6!M>&3ucKb8Hd4|
z&P2xgxtR6!^4{3;(`77LzkpZjYs&zmj5Eaq#<-bTEO{)PZi&7W4g5+e`F4)9c+Dc!
z*staLX(>E*;bizW{o7OhygBq5iNIH*p<AlWg6K$4%aH@r6OiuLdOkhtwK*w#bz|Wd
zXO#ZGMZRlGK>W647)20g6ANiG_GZ1KB&teaNh#K38!xZ^;>lLbadJdQWO;D@P}(>u
zln@cPF~U{4lURbgR$T0>At`ceBkP1teXx0(xmI_QNnj<HPbAVhaBlF#W3svyX+X7W
z1fI2~N{~+2k>g?A|FHGE`p2lP09$xegY`@pi)L>60+yLM8SB6`_bx4HsU}sBafOdh
z6U#-)S0wJuQr)!ZGBD>e+VoT|f$^)evE|gZ3(SUNSrRB>!NHIAqe|+DzaHb~A#VRQ
zA#sQP=>514eFzLe?E*uX_eP?NQ^jB3N=AmSLM;DLw_P>4j!TNC5ZL|CyIH5M44f|H
z=`+jowd5B65XqTzf<segqnR_0kIQWeyg0IS+9({AWy?=PN#H@mL-Sak^oTkDTE$X-
zgP@cg*(w-ZUi;|(4#2{X&u6(Un?UO-Se`KJJld42NU<nrl9_bO{j}|O{4xk-4Rf+(
zBJjQ-ikQGw?EEb`(Z^QTJ~op1$2h}YRoomVPSqCFkkqg95R!na8W1biNp<h-5U3t>
z?PKBTvkO*NqSe!+3v|*iz#(VNM+ENXlK~z;-1GxGK<Q(Vn4tWtns)~0#FblJW)8d0
zv_)VaNU!-3j{`+vl=3oDl%@GXRy4jdx$>?negL#EHGj2uiZ}-6<S;mA&YHj80(?%l
zr10<z#g6;UCB-aZYmufrypL8KR2&;4WXHYj4bvrUVl4L-TBlKq0fc}50PT#9g`pQ>
z0RG8K=)WSjUUOMG6wAWV?gBv$t|<l{b%AHPmn~WoQqN?bB+t{e5+Mb0>=c=oTBj>*
zCWG8IIn34M0Nz+t8;e*(sYj6&@vnsLMo5{)_XYy^8cjaL_-ae`rg2wRPk*<?X~&Vu
zo_4SmfGbkdmDtgJg;5VMFnFY?tnJsUh2qvX=gtn1OEKQQ?t1tmDo`ULC`zGKYL=@i
zc;ypqs4%DpAH@lO7Uw+2;e;lJkl<?%8R#BGu#KINHePabuNpl0oPVkkv8|##_=}q`
zW7T^vgFZ_1;pEq)gx>?*I7p6q#SY3gAfj|U@_^<(cJU!y=87JV#!`0#Jw;XI9HE5<
zn0^E5zJ;pgZW@`C(Q-?_>85f1<s$cRc$^&Yuk2>=!$7PMNl=}csEf0cgMFyGo<^qq
zC_X>{OGra%#Rn=iEU!C(XN?9UYbikDL=e-#(y_?+tw<7fO$De&-t2*Z8rvP(k?GO!
zzwbBP9jgIs26v^=gB;3HsG|d1T@}KQlhFnbSo?=Du1E|~peH6u@nB&uR)05EZ?icw
zYw}9;tX-Fq9v>im`8<d)NUSw^9jaThHXx+lW)>C6eiF79Q%<?+E)P7QxZj^(E$ga7
zp!``?&y|QmLwCm;KOR<|04<I$dt6i-BtQdJ`qGSb%o~u%bw62e@q$^-JF!#M#F9Yr
zJOX!s0&LUw)rV9EzyKBifAmACd~^@uPn2**=h=z6&EDk84*xhKVs5ZvJbP-d$Zc<8
zOW_%#PFSDUKQtHor9#B=!p&{`zb1EKrNDxYYdbGthu3u)^8l%s(=m<3N=3|N)ws(X
zL)p;0JI5s)acHqFU_9F8YB7t_iOx})7Ensd2lZWPZ8f)-I&tt5TvNckTzcqdD1#K?
z%!F8TT$uCWpbWybe7|@@)b@)4l-$`E0#&=km4qk-G;nT5(zFM~g+6$mki~kXaixd-
zdBI6KSmA>w)PFA+>AlengcA3JDSG{)&t9~)$u=r~uf0as`Q($+C%ZYpAsoeYf(TE-
zh>5_nzem_T!lN*JF_4_-s_XQ*o;=Bw^?0}e@=dzDT%1H9(<-Thrns0b6`T)t2%7}k
z=ucAj4-?$4F1DE!eIBy4CYI<-?#W}i6f&meG5KYLcXZd^#E1MD@IsjoVI00PlBKd@
z3{6jb>=g$vv&#j1z`wVg(C5<6Gi>8f7XH`AXNkw(e#pYRV+M>VI%hf2lrT9s-5GyW
zAM_KflSZKeqzS<uuORI57y1kSs&&Mod611Kn*yO&9$C|?sF0^@>ft?X`<7=%$E0S+
ztiTvfi|e+K34k6+>MO<tjUYX38zj6Q<}X!<cwN^8qamj|J;D7;YM|6V9H)V*dflG$
zKUuu~wP`TfUag57#6uE(j*F9Dkp9i0(>++7a8tFePeSx;U32p`Yv~K>pVx6!?)&nt
zOF#HJEmSjIue~k=OCQ*d8sr!nFqPc>^7!8H|AKJ|PT~b{TR*ET2)#hFNYtlt4;-`7
zBJ~-5W-4xbJFJ<&5Eev(ec8}IymqXkoK)D-aYf?d&>ID`m+3CzWLS#kQ<IeX1f&R{
z6bQR^`QFi<dcQ$U(c4H`QG)8Cf`#U|eY6>%Vk}C6V{s8tf*f~W+ngmW>8rsI;WIYX
zs6ebp{4lyOXRoQ34wRT9H;SanLgP2xe8%eLIAtn-72N0ME9v9>U9wRoFBr__Q$x}c
zU$%SvWZ!FVmjt8;wr1sdxn=lnHy_F+@!-9Y`Tf2E=t>qNYCv`Oo@SuuG3oaum)MVa
z<WMb~|1R3yoq91qgr)5RGU*{{S1K03nY}j_RC`{<e<YBi9!{omg;nS#{%-DR8t0GQ
zjGZ|TldJp`75X?MChju);5KOi(c887ZPc{-ouqa$j6ueUd~f_iU(JVfpOIS9vvaA4
zpQ*c$UL!=&aat8H4nPU*C6WUQ(F=&AG~3@>E<p*k`)1%|f#Z~`5Ix~@rh96U&A5h4
zc_v!RxzO(_FxtqBU!$B~)LT2<mNamTjV9zg6%j-%MeQi1!mgu3d){$~|4v<s>>tLC
zAR`#`RPC1e1xz>D8m*#Xq+0T)Tl!)@@geHyMNV)oo!QX(uc*iemo4}*gpR-JYDN-Z
zuSgY@QllWgQhHGPxu$(=*tRYD8*@bNs*;Ynj8H>&8AZ78L8>nnpq2G&ZIhNBe!bmm
zp{|P{o>GB^`Yh*%{GZ&b=Unm5vg(~|fo<Y~2TcR^7o99vQ%aED?CVte(RY#EsfxV>
zxqa!}W=S#xdv8&3JkgN^?_c4J4g>Dtv`4xi?50E{7R4(*gy*UOJk5bI-Ha#NRQ}NV
zoeVmi?}Sn0=mk?0Jgfnxor|pIzH>AOtq{P|OJ(-OsQ`2|T|(eqmX+#njAhe8_tut_
z#s}~1;4XXdX}uuN$`bYJ4B|k<VB>>nc>?NQw{QIr_am*^APxO{q24!zHme58EN)XL
zd`PL{=jK@MWV(FKNPAS6j&<goIelZ4sH^2+rhQda%Zo{wu=y{=H@)b-aR~SjG@%)c
zWEQK{UCz0;Ba`qS&<VnF?!2YW)<RED|I*U_fm$g7P_&M^>O3-_X{=yYZy7ireB`r4
zT&CpxBEvIt(B=;^(NAO-@}w(sD9b2x*;hHQ$(F2P6Bw10;NJBL117eu?r(-nxNg(W
zd3!0e6P?vtT)M%t-6cdIf8O6orFeKhwGUMW*U8@pk>T?+o?e+l%3~&7fA7MmA<D;-
zI3+bsmHZ?HOkL|RDQIM7A-YDxFQIP>sYZJn_6ix|5N03hPdu0rf(ey!I&vJtA}I%c
z0{jgmV%I9WzK=8_DB9JzcrCiRd{qy_q@U}c)SSM1wV&yFPhzcM6{|#+-1XgtN9}KE
z-zfr71aSE(F}HqR`g^D4X=X!HnpFOHZ_pvF*wD>r`IJ^lOxkr`h(yv1TTxRpf*Z}u
zHy{DJX4~yo*%I(n@_UBcMcXBjIL>{K(U*wzf!jfGP%^)-qkO-PC*Ww?OV<f4957Qe
zNkE#By$_OEsH>WO{?}U@E1XKEOa5OWi%Lr?aD2a(cqia;43px}gsS*UZ({>68Wd``
z7^CztaF5}kGMxs4E<5wcAA%YmpG_0M8aOveof!I{Mp>IHC>ohK^@CX>_5~@DOy0&?
zBl%W!nAbp+5+An+hy)`b)smRqG2ib_F@+6f<;mIlyg;eTU<x?Xno*FsSZthaWb)1|
zrliwQY~F6`!EdXhpV^*YMLjvHuot6?$e3ly;{fteg#y!T+4sh~$R?ty!)k3+Szy+(
z6or%hkU28|evGy;aB4)Wt~`K!8kjhrHR)z?K`KU}gm5!ER0FB70Ww3AU&i;UqEDa5
zX&paV>yn_(p#;3g4Wn{wv`)WJjc<XEvh%S2(d*aZQU(>fyjkNIihQ&btV!P6%Mjg8
z|EeI5A8Y?p=&&OqwzxrR&uixj6j)l*CZR>+!a*>o%XJ;NPrBk^?yEt7A*ghhs%$qU
zhCikF&`rais`e{R33YXk!dhrJzKD3S<FgJ+&@s{@8S#p&FUfO<{Gd1&k@(g0+z`N^
zAxvJG6qn=i{YFhTnQnklZ>f<{4p-=cmL0J2)W44yz$?O9G2e`!S|p(-v3v$;pc>#~
zyBM+c6Vrw>>YucSMW>r3%u#L{LOC2rm%1ipi+n|&py(04@YBxU(vo`Fosku7XfHF^
z!9dTXS%h?)k>n8j(tCxzE1|7n;m-qg87TBjQd%t~rTBI`OOTDjH7jgV4mq~o65C5C
zK`Bd^d+Vh}%{0VjUSV4K1U?RnH0c{7?spNsha?g_?VIe7qcWrcXy!0AOY){fEJ669
z#TeHwUrP-3(@?a}BB!kZhby$`)UGy}=RrAHARRIs%3~Dg!s{=MKOUNfE2K1&aL?Yn
z==Pnc&LmKFzKy=Qw)YN4zB|TgE%EyOk-~LeX}ck>;v%?>45_O8DbgZ{94xL0AwonC
zW7H@`R8#3O%yE}$zWSZ36qba>`T&?`($uXs+d}5UXxHQ?&|UU>o;IWQ6y?lN&pws@
zD*4O6&ws61KSjm(qhfzpJ0gvOW1@NSyyN>{0hxill!AX-+00Avc|-nC>WSh2PjS&=
z-g<Nz$N$;7&(Pte3Q6YAIfddjzi`@7T?0yI&K6qif9ID_cj)7(HJIL(G)tW5O$hsL
zKZgW^)pXk>tVDIwDk<xkMrrUa_m=b$UTqc$KV<X2ZFA?}^%AW_Yg<ZBrIq9?{y8{@
zVyZo%mtz43<k`xWVhBZ6zHO1pM-SQvFf<lXD^F4o>o8zYRyE;Rxn@jT{!D~Gnl`Az
zDfWnto3mwM&Vh=)6hqinV7U_dfF58Nj$V->&XJJ!$}`w5)VW$pyq`A>3NYZbLH8!?
z^!DHq0MznHPi#L8Iy0`bQMn{aK0hK2Gf@UeMu+m@VRg~UC+9(WPkD2q+sSgD_iCDg
zXnsh#rC|bTC21Dn&Afi@n+!#R=VUHFKVzQ`vur+mn>NNyauquAbxbMwy4A2ZuHI~x
zo_VLO$i%Q97{_rtRn4zpHs<()76j1^i^La7PTQV{!+cIlkVyzF#%x*<e*~evLziO2
zB#C|`?7?jUVzfcAMT|C5IT`ijNRX>}_2Wg@c5)DHR$nks2CX`AT`!V>dFM2UNxv8D
z$;?ZOm;vUW@Fj$0NyA-BG}2oBB+_cpaJ5g0%s`*#v@eJ)da7L?GN>0lE^d?DRhTjR
zWkZYY=-<nsC^AcnXv<qaCU@xhlZK)x6=*-^6+5TO0R~}>;$ZmDJxA3ja`<u~3aAq}
zE^mqWy@Rhwxlk{!7U9=ZbI#sg`g7SuNJM1u1k3VdmaAz7>x68Sk+hJaCyj#XbLm;2
zYHph5f=7ZzH>MJ|N0h=P^letn(edmvXG>mNo2tx+`qpEI62YfmS%0)~Z#`dQUXsCn
zpH?1j9Gl$%4U9SkqE(QAYq`i(B6FUljNo7leFOfc1kmAy(IH4-6@5JrqjrOWU=n%G
zU5@<J-ZDS1N{WYABI0`>L%noX`@*kw5XbUzL;TBla#sSR+s0Ss;WG3B^Tgk_RA{;(
z)hY!1*IGQYkcKf*2+=VApwq57`aWUewZ_MO5=tRUpwEu_u(&l>!L3)M9+q-g*`(p6
z{LC65<*M;l_a%EgqfP`F&!;v$D5P&t{{sCW!2_pq<ID+u`TKqS>VJ2gBmWMK(`FwA
zdPU)I&I1AnUNK$#l`6_x?51!mTL3i~<ytQ$^W>n~slpl~pX&M4|LN1HQ0tGU^eWjh
z{a}WRSRJ^v5vSyAp}Wa(N314l8$`wt?P<o|ovLL2a{n#gnC3{D^#{X|AN*rauiZ}|
zf)Hcflo)5ycxYw*RgBWE>(ppj$au&nB^$+x2M35$LzlaaNTYh;ZM9H?4315*e}$4b
zL%N#g`KLH*-B5obNWp&zgjgm58JYakv57hZUzbUYV)=abX@HI0$BC?Q7ZlC$`s;jQ
zgVAp^W84cb?Ze@xFAYsOwYWD_VYsx?3tFW2q1qo&Te<I^qdkgFQ9hi?h;1%KpV|T7
z#RDgQpr^}~bi-tx!XY>cZ;s60>+WaNSXoeB(O@x9+#aag*~*=*ns$B>4u&K>)d$!L
z%uE%gGfOV0uXf_*uCHoJiW*16H_~9<5{xTWAA4h1J2wXhqXLYiYz9s(SnPwnR?@ap
zhim<1R^6`(uQ{7VgYxYin1mj0Wy|y#LQMRNEZ|M*Sx6~#^7%oX<>WBU{c>+PFVT5e
zL5>DopPuvYt5pJeaJ?<<cay)-J)sUF<O07D;6Po|6BU`yljo+ML%X+jl-#G#rw+c#
zQ9aNSb1o&fd#v1dU_XPwfk}UfsmAdWkGb>}66lmaZd@id1Z^QF+b#?!S_xJHql7cF
z&%ie>3Ir%7Wm|3uY*YddMJLZKGHFU6tNHQa_|UxQeTI_Yd2w!dW6+F0?*y}_BEEgv
zscFU}3DI9{{YYf_vtD9m)O7&U#R8sN)n3#{aFYqA{(f8Tl)?|2``}6ge^stnAn=nB
z%aUB&0mT*4;>pvB@KTZo3IP>y`YVa=H{FljaW(=_dd~eJhb(CJ3FTuwMgz;;AQt`~
zb&q3ygftV_3KyKJml*HfUg*?exBY?MWinMy%Uzk|ND%z%4I@H6nrA9B!oL61hoYjz
z2>HF<8P6CQRZQIRDSJ6B#@enC{(atG>{%-X;-$SaEi4R{4V&d|Xd(j3LoAC@i&eg6
z9a(+*%ncdF!)&+~d{VltWlbqsZuahR<pWqWT7$e^C;D_DKst_m=%G%Ss4rD~4Fqx2
zm7S=$)ygKX07;UThQXdt;c4O6eh!xAd*6IV@;+8nngsh)s`Tw&Ts?x}O7i>VEvInz
z<|tYcaW^58a%pS0ohfW8Yq44@p2dE_O3v_%7=8hr8V&&E^REXcC+e^ZX{K#uM=SUB
zFuN3I=Ric=lJD`F30%Fw&-W*g>Sp}y`9Itc@b0Akk#$;#*PP4=H7p;?D@9E?swo%u
zd1%Z)%n)VgP+5<PIX8U}Kd+Hv0){DtXP(zfe8+ib{*o9@mwQ~`NtbIYI|?KHIT8Q?
zD2XP$0Gt|LD&@RYZfQ3H#XKDj9r3aZ`WCMYD~MJs%{t)9yzc*j@~6|dyuWHo+~(Zf
z-!WzStr(Zj1r^%@y%wVK4R>yO8b(}L<6C*`odp>eD&qL8?u6IziDr@l%K4kdBY$G8
zGT;4ecl9ZvyFZq<j0P*rN9DY0O4FQ8=scOFG7gtNqu&xY1Au|JB)j@ip;!p&zmsbF
zNg)+0cs<VZ@umP=udFpzseV(yYMGKLl{;XCU|T#r1V^T%NB5`A!S&PLf}-syKp2U1
z{6>rrJ>RZGRrG?q7%?65Qa8zSlX5#IeM_k;GPf|tIWk;5>wHuR=e58I&~kUZf@FaR
zmA>aCmtqNNJc05+d+|>`brCXX4E?We?E0}3rFcV(ZB}8>v_Q&TygjcgSJ-78Ox*^+
z`+)nx5XzlnYi_nzez75k&@}Sje4HU@o>pa)rs?;~vC;<01;vA3LlmB99RII&p2m&I
zUrV3<-i=n3Hm7fs_R*DsY4~2@I~1P!`n*EwN4+3GNzPHpoA*hArE!oip&O-{-h#m~
zh%j%?2b&*%6IEDUaG5FNWoeI9KYiqjNw^U>9j+)6#5L9_nrl>2AJeyaa5wCq?&-4o
z4aJNY3MeX>PAD>B<7Nw=Geqr*%;*U|>yN}ti6mHxc5E-l!ex+`Bd{c2lE-nyRiGBg
zVNsbpP~?&7J0L-V5NN{MtWU+5vh)A?Is{l>5z~o(*4g>mhYen)ml`MN(4Ul2It0r|
z@6A8nYqZVNl#ih)-8I3X1Qy{I10w-4I!fk?c4W5Pqfaek8TtDA&^Rks7Ad~~W2n5?
zrAP?qEZiu)mp9jvGL3buq2PkWk|ag@D0I%>f;ga+1tE{HbX{@&|5|__P>HQy$C_-B
z6Q7_AusjIt?t!hRxAmn4{$ck{@bp&txxB!H?fhWShQhH`d0nJw6^1GAJt@z-_Nx3c
zv=DQLj`l1^<DB4}R6`MNdV&12RJ`r={M3bj*Pi;S6KF$}GlvDT+iL5^S#5$#+RONd
zhI4b?(w0Hx#F-`|(*%eV5Xhaj4?+qe@|on7`tkl})aHzwB=bXCgSf582`vxdeYK}?
zO^H%g@+lOm#R2UK%Jc){I48<zmZ_E33Zty@IPar59W?+YSS1qekdW_3k^7kVC!3X>
zu*SgvHT4ZI!O-z_s&X*&@Qyz5B~A87>FZBE$%Y1*mwjm>Hy<B{2r(u{M4n<R5J#xZ
z$5%Rs%q=NqA^E2TGIClhZNBio5BXFC06=ZZbaTdejj3T?hB=3^&`Ss5kQ@h0<feNT
z?H-=Kvem-ai0Tv{9dM79N<@UW6@T203<KS4M>ie!y=RW;-3%Ytrs;q<#jhJNw=o6r
z`uv-Qj>Ci{KUhMvG5sWBOyA5xODOC6(T%_``nCE*8n5tzJ?A&*KK);B`|Ia_BY7~}
zR-GKFcU>c~CF#Ho<-i|H<lSH25tlGIdh@?8Xl7ZinF-t~W}u%;5YuR`o+{UdJal02
zw!sk_evH@3gm{#7woDlg0t4f4mhYm|qf4I8PxF0Y<-75VzzGF%MtbEQMH}ViZ~&?P
zR>0<?xLt6EU(;wa7YDWIg$7YL))nddDChY=11VoVlI58-Y?jVJFEMo`2~L)1g;8{p
zzJJj$O;6p|XH7bp`S1)^y1$>W{YYWAFOpJ^&M8A|Y6uc)&x;6!N<8!0(%|`V6nTa*
zm79L)Y@sFR+gxY!3^cuFn2=wFtQVkYpwG81u6;;7z~;a%F*ap6ycLXO8ZNt>u-&b8
zi4Y@t!a)LF#am^;L6O3KZ(2%>_W$4KWbqVp+R96CX`aDB-<>_R#2OXZXvt5_edzX4
z3$sn7SZ%7%JD3z!i3X}jVK3MltHA4(Ixx4ePZ3aC5K-WqfE$vgF(3G_S~@NOQ?1J7
z|BPP4!Gtz$ko$-uu0)pN8DoNY|1-Nc9$bo{ty~AJkRcag>ooK}CXoN)c-3UT9r+p5
z&A$9%cuci>_7SdhdeNW{X|qcZcVL1$Hd+z7o7s1gBwEK^HnPA9@5O0t3h9@W#e}$@
z{YSk<?t_#IO{=luwoUA4F{Y(~jPcj{sq2~<ua%^Ej?x()rO$b&QTGSiG9x|>k0=t$
zXb!S|IJAa-+2S|DiS4Dd<Z%}_7sG%-JO&Inw4Bg&IIYz&3kKS^0>p*?LdNN{vciPh
zidIrH;|swNEN<ni)L@`S**Bz$iTpG;ND6_171xed&LxZFKlfHxI-^cyWyX6<)P*7O
z4sn?HM`&c6?7-)i^DvsFwD57Ncxg@6=2TvtRKY-vrcKO3qDv9IM(X=Gnn1f>%XZ~x
zn;K*~$C}dQsaipM{&!_ap6AA6_Rh%*_*1#g$!>g<GK|bZK)iJ1dC(Y7EGzW)LCVUW
z+JD)rucdwH7trZ>6WW*;4dSI^cC>RILj}~2_a{T!GygaTkCU?0k?w<AS=va$T)dQk
z)s)j6{-W(32L?FGQN;5s&J~#Bv@6FtHT1jS&CNmY>!C385+;vpscy;bgxlaC0TJ<3
zW_1n{lu?flcM<GOu-u+jhqXpkKa<!I8`w?Qluz-4JLM|$%8zj>$~gGHipye3A_lCG
z-2|FZbssEzZy<5S{_JDFN{QfygE#byvc*Abz#cWSWAPL>3I}wFrX|W5^v{~woh%MA
z_9dg=W!<GbZwV^1m16EShb?;(7>eb1$D)=E?a?A_qo?ml$5saX{4V<YK2aGK>5Gy7
zf@XR5mq#KZJEWcuO3JwkT?93`iN6<z3wRTp?ZVrYV&mfvIPd_lXPEK1ZP<c;ZPd>{
zSnow7;9&kF!5Xk&nO*klzUILZ&X~;qDx~c6j}ZTh_CNw3mh}Kzb@=JN7B}}O3zN7h
z{QQva+JX$bO?L52(&DFqa$^cZ?A=0?4ebT_-XeLQMP09#3HM*QMsk3(SVOl+hLxyA
z*&+byMY&u;Wc~H*(@gwq*@e>d^b;syCEpO%g`vAjZ?dvW<^>Se#5-%pFNrTwH7Py^
z2ofo%k%WBa{xpm(N*MJ~4ak_7iQs_t0fmPKZ!lI34O-&cRFrFcV(x{cKiC%myGo>K
zCf7-5>A+iI@9gQeF&tE+aU4&)1kV-gnUfaEnc%E_OrO8nzle{K@ghVhG9zebc^^|2
z>76ZVAo0eH#_bc<oY*rP8Gpr7$_jO(;*=KJi*5d@5&Kp1o*4g=262V!;kB@U&B%}M
zSnXK(&|*B5bVsqp-;8KFEfv`yyemblcw~}Bt10}M-=RM(E>=32kp@*J#|WY`0JMQQ
zE#Sv(kz3TMat{hkh@l$4cqIX3^Up#k+o;<*b2k|>cdQE9sOiyx(;YNx521A#>m{Ht
za)fq#`h*Bpj8ph7xUdX4DNDr`(f#JAwka`h?`!j0Bkmx4X~I2bk89PA*17>wzrsE=
zBLdGiPFIQY`&3jNa1|GI4SV{2-qiv7zIXL1>#yQNCh>2q=%xybSpH~CxSR@09+lW1
z<0SbNgrLVf+zBt%2mb>rN68+5x86meso@c}%)Y<URGyIjF}&KKgU!NNK-Mj8QAzhb
zr>AdOVq!{%)B>*D)?vPM2KvDODqRwZd10NfXB=O3V(0}ooMOFH`ZVH?E(tDb4$`oe
zHG#B{&xcHp9|X-F(kluz-dd0<zTVGOf1DQNJFJjH<=U+Lh-qK>chLG_1#~tse8lW`
ziXwC<`j)qcd=d+G$lL!*q-x{~?+f7^Oiu`1w)Zz|EgSixdj#zeRXXzY`_iOL5A6K+
zSKNqbZo7vgxi0!@vK4=#8K5OIMw=1`_r+VxF#w~ul;5p=Nxb7DqPjideJC_qQ)=f?
zG0695t6~fH6*pGoiCtVePqYJLE5&#qNye<Zmx?bH&qo=*Ko|jx5~extQhNF3o+;3Z
zWxkXSYdK}#<*3B2;A)|Lytw)8v#um_L{$&g`265F_=KIOu!V+8Dn+yR@C3R9<1uwr
z8c`rojSj2`V)mi&SsN~#&(N*H#ejxdJJ_-CN?l6By-xfQ)^qC(nd+ExA_zbB61oie
z)4RFUJf5-ZF<%eW;q#5PzV7EH;yPVQmYV)#KEzviC$Dn8y~cCA6IM56=g55aQW*9J
z#zUj9CXzIWK{y9I(z`_Gvk(W4LGsDJk9GM$fD}!?vW!^h&u{eF|Gew*Wt_70-(zM)
zZ)$dxa~@(}bHn|puoqKK-hLh4DM~8niaEJ4i0&l{Q!RT)&vox$Zb@EBB>!Bz{RFF?
zyQ1g0&a!h%g1KA>!c6VaII@Al#RTg#z+UbJ)pcTrH5?AU&(Y9O==CW)lQ$9mZQz75
z10wGjq#04qpD*Xytb|eEoIp>Fb`Qp1b<GS=LiM&BBz}J{s{tOmd75Oqy7ZBV!l#eW
z!CV+jQ1)cQcH=i5Ap$}`J|R6qT1N`qxnj>C_yfvRA|aqZ55qH~83|VKB7$_xL{ixA
z>oZI2OUS@itmk6#4QS#Dwdm$YloD{Fr8h->8d2FGBoCG6q7*Yi(Yle_KPHp`EFVy|
z^&VWwUP@bV`bN)<$=4E7|K9|~M;j?%kP^~TloE5j-o29W>+1aB>pRzhLI*GPr{ZR%
zmkwHD-gP?8{ij45?;y}Cg`BZgy07b@r3Ax#S~IQRaCUMxl&Mw{eG?Zf+4YFj8;97&
z-=iX@sg^;kqq|XuQA?GX`p;wq(Ufk?msoJVayD}<mKnCYuV}0Zw$+qb5!sYCSCK4K
zctH&GLA<iDYmb{<(nKq?^&C=09~HK#^*1o%Ja#Edk9`3|#k-V&LV7+#aVPO`n13ns
zPtk!)G-XEI<5>$YCwewHED@UIK=zF0beJ@360~5lw>Oaz`+?b0lRh?&!T{h3SUud9
z4jI@MB^FCr#zSdV@<4ex6`P1nwTm@`44YZT!#%Oy#h^Lbn7+~b?c7sUiT=kqD}g2s
zyCHL@<$oqGl$<2<jpT91*|LCJ&Xzsdi!QV$f^amUWjf<>wV{_{{6BdjrN2toulIQ8
zSPt|}&ZxO<UIRVhaLPR9{MOk7o_LQ>7ut3DI+)w>!Z3K)oJz9P)ltY>EUCUbX%How
z<Z+@YHBGn0d-$LW{Rp=7AZ329NUaI2Nim7C*)Dx<)nqilgBM?4>9XTpKT+K*_OG=_
zxq3Wm<sGI*=D7tke^Sc_mllt9!9s`oeWY`|2Q|@{yv;xr`)i8VC7QeF<tNV1?#MRG
zoyVUGLul~%)oiZniP{Q^j!E-@jK}M>RDv7J(aT^|#ju?yy{S@9+DYLP-k@gYaR@2)
z5F-fX<1IYzG?stwf^%s;d-T6}7Oz}VpYuKG%FbBv7`Ef2MNAL)Fw~!F;I0QmW1g1c
zalIj7<(ZeJYXc)az0>{ocx^k)nyyLJ0<M+2-&(5xOD&{u6fog+82R7_rXgU94Hs;5
zM5;!Bk>m(6xPrJ$iiQL!hkVl-Fwl^e(t(4Jvk~1{3Q(%NBQ0pA`T1o_yHq*(_R}#F
z#K;P3SC7)-eU0j5g)4KhhkWcL6Luu+sBH(fV`$uiDNwX~j6oZ*B~8iZJ(I>bksbS)
zFY1^#_UPY$kDm@kuE)*u5$ruovehPL(bq>(=hQA1^)qn|#1U-}gecB_qd(}A3X23;
z#x=fFo;+b1wv)sekGv`kjU@DzLCC{se%V6J|L<sKjrz|s1fkWi-+$gn4WoKl8$#K9
z)e0{ROnt_on=mto^Zd=0=$}w58|*c(<-?IEYC5pMrQ1K|Zs890weNh)%(&Lo<15pF
zn&<tzG;_kym@IS>tDj#!gC8?T>=Xv}ZS+?r{WkbcUo?e;yWma{Ti&7n2o2Jj?S7s$
zZb1{P18J$7mR%@28gFo!W|rKMNtE0!#jt4*TGE@NK>STAfFbG{D;$ch00UDf{yPme
z0~D@|Dx}-CfX~9Z{-7v!bWQcy`!F3JwJP8$jwkCXo^(!BRW`WE9fujKmx@b1MTMj)
zw-P<?K&5q%%&#5rF?1CR@X78TOJi7){#_9z9$D&a7r=|N+5*G~$V&30C(0wM;xN7(
znL)b8?;HN26WAJaO1!ex=Z>hz{1q{?v`WiIOr#MV<>n`Fc6$l<lGC6oye$qyvBS-_
zigL{>$p1(?AyW%7Fyjsctv@cI&yyECF7J$*-t%&!v-Ji)K3NPg2?-?VFmw9XW?W9a
z;s%qPo8x(a?OXX}b^$6yB|R;I5%XcZLMdZ9&jZl)2BWb38T0v^um4)9?}XQXqrR3|
z|2+OQ`}AnlBzFC^h$k2h%K5=`rTsg(=uT&N9CB-P!8n?5?^$mBT$KfU4#M?#;GpHV
zX5Z~ZZ_#?NqXC6%)E8M#TKKEEFnMiZ&XKEWR)(IJ3RVgUYiu&5D%7zRG>c}$6Kp5_
z)~f#sCMyaswmr{aUc@rd2op~tp$?&so0gm%(W?wC)&11UewxRRrksZ_s|)$94AT(C
zlpGpL+SpKG)9L8XRN$>J!z!THQ}#TgOHedOvHrtL30aJQmY9P<7b^s#o-Os*D>s(^
zYHm!KsHl0Bx!Ul#K&krL)Amt;jgaQQUH^ph=;q15Xxu6EEJSHoXE?4~%L+$HmhrwT
z?2nXgzG9$XYJU*)sPf8Fcs4`MBSAGW!DzotWA5=F#+GT1iB3A9Rs)gfZbQG@N$hz5
z)A{~^f;<a?#e|&f6{@EeqR{GyDF|rO`te^_?TKSJ_QfPLFE(;*ez4)49Y&Z(hT#kh
zOA-`<Cf`$XF1g+<TdA$M)Tty#aD>VwX|JTIFL|gM?))~{YV$h0Vd#?pMFKGY3C82A
zSFV&g*#ktaK?UoEhf_m--1DBIQk5g*zzJ8Stcb<gH$s`%*<t#Dp$>A#VEXMxG<SD_
zA7y_0a7i<&XzS;-73c9z)5S^Cq>wzUw|y#ePr#}+n7sKyoZ{VYtz2I9R{k-|BW0H{
z+f(kTy}i4fmR6VzW05zj(}?7lV%*oZ1t7d)mw(NS9RC;*zAQIW|1)PRQ1^On_uKdU
z%3cF4N_ZIYV5TcZim;N(EEi|jT;g)2SUK`c{AKT(qe#V!AmfKoJc9fMdW?*Jy?MOF
zW|jyqPjCP%aQMfE?a0|~EYvgDTB2E!u39cY5y0Hi1JEG6U*v@txio*nsXELmF}$wb
zMDpUoYFOemaj=B<!GrD0oL4eK{V(*R%VdX+RnOh3<6g1chxm@7s75rTTk08L(qx#q
zSYt&pg`KB_=5J$D1Ro!i2@5Sh_jBuRi{hr^BkRO-_!^c}ovLQpWL+jz&S^{)^GDR7
z>FcCmT}M71fJokjB#~Ikm?7(Jq)1YI{1Q5Tcmzt-|9T^JHQ;GGj<N=^_te(9*`9U&
z8*P-*#BE(LqB-`C%i5n0F$q8`TWY(o!p&_75a@4%(;C400G@>3h!(~uEc^R#>R>)w
zQ(0T!>XAnSAi8)%KtFE@7q6fA!BP9K{_9w9w_WpFOr~1-pohVm-s1MHhjKppc%tiS
zls)~5g)!yVdt8p&U!e5n>!uX#b&^;#S3MJgQ)Zt#en)Q&5J?ILd8=QgjZuk3ZL%+q
zL%sUUS@J0ODsZFh=yKWe?=V8O#9Q{fx~0_zwB%@gVTkw*E5}=G$G~3JUf@O$?59HQ
zx#f}R8jb>Kmwwk=6R?c{L4#<Db?LODi_0WVQ~teTR7?_~;b>s3)v$T(!5K)~NFM~z
zc=?4eMgh>vVSP4D*w&YqN905oXZM;>EJGo3(sr3+qz9MZ<Wo5Pfgo;Ut8v0DD+!?2
z-)B4!vLj7;qHHo!k`r`-vjRZUA#-7x*rHD*e)Rh=*ItM$<q4ARGQpKML>H<@#=TN0
zT-;}9oeCv0Ljf0V>x(Km@AKX0KE$&AjZk)2AhRWHKZ=Z6;<EG+@wP{D=%OmDy}hCR
zk?ceI;tX3vFd7VZ>4?;BKL~+G5+Yf|$%!uCS;6ViL}MUyxG6E+2cmnr!DcaxvN_24
z5g+T^CECaI<(PPj(3N5mzhH^s7~1!OnHjW7zZ6b2^FE?d5+eFE>E2lMJ*3)$NjT>V
zK!YgvK>HFi#hm><Cj(BFF6QvHSk<;6zS$Kmt`l|mi)aH-1w~*@I#>a>M*R@65cTPI
zZS6o^o;6yp58;H>Brj(fhunJ7_i)tfpn0aO-}PaqA5T4L+b3v`y>*-&)sEao>hmFr
z^e$mOqqRHnje?WMe0)*1<xCaM2YHE7_6KnfHp1qHPAlx|I%pAfu-q}6<G7RR(-^cS
zoN|<|CDZD^gpw|Phr-By$%sT!b`kcs5c%otOCt#fEe^O^>fqD#F0CHx6MAML_}Q2&
z#P2`y5EF=*`#Q2!)Da(b12sPG>1e#G!<X#DE%C#1G>(^>llj;0z|=~GV1RXQDF$|a
zyn#_e;|&|vXrVT^5&Q_8s(!qUzQoz+!Z4xz`sWHND_-3=g$(cXUQh#Uh|>}e63|SN
zPsCFi&wS@fsFx&&)R!PjXoGh_OMzsO&>o<tUoeOGOj2wra}V)~wh&AV%ke03xPNkV
z4<8mC-zTivs5C?#IT)z<_vtRKId(6=y6GURW4HPqPXv6F?L*s*ah|_x%P86cRz6+7
z{7v);<qeZxNoMQ>ad=r_7s1w1M5|&N+c*68U_w(ATh$1m=<@<zH=7;GRZ@gWWu+a*
zh*pJ`MJU;6T%Au);EBC?jEy8g`a#zqy{>actHe)m(!r8I4x=e&MTA+DIg_PbJKjy@
z$McR6Xh|?4H}=mfo9_l1_=!dlOvtrVQJ04RcpdmEyom$Hp6)|}&amD`CXplr*0sM`
zRzNRHQuA%v5%lB7%Q3%Us7*>a1fZ{!-C8J*T=`KbO`V9zF`9-ABQwI7_G^a$B{%{g
zbSTV7f=l#(?EVgucX>b`CH|qG)wm?qLMx6E!ba<zHFTsGPcbUemjg^498m^Amkc4(
zUtH^_W$vkf=}<b@le-7?8z1>a<!OzWQ7)4fjtihZtB5jyEh|w>G=3R?dL|wpLRy8#
zq$Wf+1(rc;vX8(dcoBY*h+wE6ti?3Uq<*O1U^<+M*ew-HVOzo={Oh(g<!afRw=P+S
zvNbaPo%MqTaYpK0?Da-pQR}?yq0)g`BTMKC%E_(Yz%I)Zy42U3rAz(cE)5FQbSaG&
z2FioTjCh-lqIWO(%)Hd*0S09PnS}{DI&XEcdho&+k+IM}urYI6I!rds6!Pm*dC7hK
zXBn)5G~z^+er#D*C-*4~*{jeH`YHJ*fm1Y}ixHhKnU2#Q8Dq^QCL#&`V^&|?5P%8P
z9V9#Ux$DsvbjR7zX&50^F~T6~04+*_VTGxa>Yr?3uJyr0e-)27r3{`EuN2w`D?#*D
z>|o|_VYaoYx4iGgvByhPd%bML*JKHvNB>hUB0>SZ&-oSjztAGd5P&loYsI<4jWE(l
zP#q22YDFvJ3hgC$Qy-3b!Xta3vYoR#6|e?QFc?ggW#jGc6li?-+C?DksBqU-Co!*w
zx8Y7!+I_`%2s`<D_rgG516V4=3xP=SpQH7Ha{+)8fHx095n#r_(>IKTO9E2BFlm$#
zY!{~89Z3u+essI$8D>+CmZ^D@eHCky9I`J^v#^_GMuw0dIQ{BNfj<NN!TX-65$Cxz
zq+w2;<Xs4YMO;|^6O@f31~U*_Us&e7853jWh6pjoPmDJuK_BO^h2xyvI<UcKG1HH4
z%*+heakj1*-o^mqGthk;z5O2qo$VNwzDocnaVZqVGC|=YqfnNZljxJaEAgC!4vHXB
zgwbmy|K8X?%3RP(Ei7K`$hm5my)ZcC68g`*EfUTmjeNvnR{X*MjAj3Ui4b5CABl}>
zVcP9%A$=qYHKBS%O#gQDy$Xr3???}evfK+OC_PU~j^m|V9Q!tj;A;O&M+2`Pnu>JI
z!-l4EnGlJCv>KN2<7odo&D(AC1NsBT@=NA&-&-t&0&_SKPEQNYS+;^?B!jH*BEY`3
zLb!Qfbe-zj{{g7>j*~<iNFhW$7?Q-nyx-<5aJzP`Hj&P2VE)c*Xv5@@UwS;4m(t6Y
zFrEakif_61iS{nb_}dpdIzbF*x`0j%jaS_hq%^__5t-#yzkU3rCE5aLO{SS-ZSu-&
zPpnNOtnJa#9CvN882>53<;D`rnB3FLnUlhCJtx(Sj;4m@hfJ=`uR@JiT}r3>y)x$D
z=3Cb}u1X3i4TLgZP9@_k<)Y_7q|?WSlz$2f_IkV?`1M}%2sH)UJ#GLf`O~dUoWH`#
z@QK;GrOvdHH^hI28Sz|YECzep_P*`&?*LnU^H_0Fb;Hj$Ic7T=VoC68;?<M4+W#W<
z*4a4mzcDUaS_fA=CPMNJ|5XI%o_)~hIXg-Ic$<ekgL#V}o-sW_kNJUd)XyhN*4#LC
z$I8JMMa`0A=N6T?TuVeMjx*}Tqw*v>fpd$p8QF=r!B^Ss%s5Lw`g}=S%WPlp5*zL<
z%|<a|i^<Dv!E*UYW<?qY5dp<sSG~1oBw%Yr{gsJR>=VFsNZ)Kyd*cqdGo&ZBl=UQW
zuh=4ybCm?xE7X5&>p-xZNop&D=+h!N5c#W|s@B+JW#sL```73X7D@{Ci@L`tpChw(
z<58W)tTUukZH%Qm-|i1Ov2JW}pgx-9v}1OzJ9^ObbSL@~dSrJEsH^0c{zHhT>{U(_
z{ksEi*_lVj#ydB~8``+3IqR2X-9&;r5M6cBO_7ZVJcPch-a?ocEVMF5c<Yh3&BMSh
z4aqZ$nn-NHv0I*1lAm&|A8+0H;LSfJ!uhX_H6cn+^=72UG9Hp3NaX88i=;GR5L@Hj
z)A~t~Buv+=^UTWd>MZC&Zke|03i9I(p|#cY#x}1*=!=)Ma3T>YJ@XxF6102b@}<2C
z&PhNP(J(9O=N2P%DSF6RR);0NdEIngid+UqzjXUe!xFUL|2GTtsCU{L6Y#&n*1$;c
z*)4?p()lKF$Y0g3z;3ItnY8SnnrYalmvWKXts>OQU6x=c?yzrNi~U!b1T2JU)QW`0
z9~L>KB^be2#me+H#(?Le8(>Plg$Rsy0$Q+p!eyNp>s+RttKj7wR^sJ*m^kS-Q676C
z&5kdV=y-_H(WhTK8oci5p<UWDXJWN6UAS7q0?b2bM~egpGYkL`%uAl!inG789!0OB
zA89GdFR5Mf=BR(;Z#*T(6@G`{%LAbER=Rxvh*0mkcaI!6HiPsogf!_9{e+B-Ajj|{
z0@U4yQGwrgtg!xJ<%Jn`XD)cSchhflu<55-9)QL?rt@strJB6c;9~UXb8c)a$N+fm
zzVAV@Lo_=j)>W3;pDhk~i`<Y**X8PA@y0!CpJ|2uc>w{^lgHRXiMGa$UcfS__fXZX
zs~FvTs<!Qw;wJVa)Cfw$6CcV|8$u6X7`RmsC$_((tn)c&dKrB9Ts)pn67U+p6+p}#
zF`1&=x8k&QM7nLx;_uew_S&LLweMpf`*H-<S@0tB9OFOW9BpYo)}76EwbjXH!THtH
zbq9Bcfou8t-if&XslM@bpU2W2F={8#FSz!TcJ9~V^p@rr)hhQ}FJRPbUy7Mcq6T~=
z9K6xv!H<9U`L>+xV?S1_Dool%{C_Qg(aY}{rckC&Pnr;((AKGLibgVohlbeKR048`
z$S{I7vfME4r5N^Go4OS>nwUXl7A&DsZ&oPiJZc$gI-?cH=`Rlyd78L{PTJJu*xfd<
zKv^je)N&`=)dly3CS0FCVf-$1(WQ`pBDIPs!HwQ6qH(4Fu@=?Hs(OXRg46m<10+g7
zcJWafZ2FBm*5pM-+H3hIPcTuWDs>5+v;Hlu11xhOPY^T#eD{t=aig>B+srMVXf4w!
zv3+P=a&XvoI%1#^byTHh`cYLrBTZ$~QmNMKCndCL`^onQm+>qK{08aD%EJybW1cV>
z*bvj~+T^tweiIM7;_o-`54DV|!C1*BA0|i6r;w0u2aAqc1>P^fIQQ2u1VE=juhs~R
z<Ff9yka@)*dNnQZB^KK4NVooFpb%4)J(1Y$UAL_>^WLL|X`6`3zPAL8lxfLAK2RtH
zmvk{B(o5|y<BS=NuJ?ev4A+r;@sx>D4<F<7i==RRGI<LlhB9{c_O~V@4N7sAysgvb
z3i&KQ)iVZRFFxy)t>__jV`Xuyl|j#H8rmxk^;^L11}KT*uoi%dwLcb#`bA_DHb)%y
z<Ml6JM*p!stl3S@V>x>nB@A8|SpTCGaKSLv5<z>8XvDbO%<56K+B<*YoP<T(k+vE(
zp8TK^W6^hZ*lkR($hcp}w2(3M=CEnWg=)BfFc>Wg=gE21TNX}H&;uMFnpD2_7`q|$
zkkef-qY=4WP=32!ZB2k%!4|3V^uUio<a_`12Z|>tH1&WeJF4h~jU77J9!#QqKxa&a
z@{J$GE&6!~O-mr}(`B(nKfyDtPt%F`SlW{z>ixpX+4Dk`{GSL%XAU*h9|F?C{_(?Y
zl=F34Gu&mVXTZ`d@9|OQuxsek*$_!0$V;D4L$UL`sB^=8zxP({9ls%G1R0p2kh_tT
zzQ{>eK^fL5{bNFpF>mg*$3}7X!^*JQ2dO`JF|r&vBv~8&c)TjX6VeB9b`JUe!#&)l
zV*kpRU3h{o*UZ=ca)p85)-VXR{!e9(Q$@`C?oOWsogj}hsrJiUoCkzUXU9-(Td}HU
zU#-uMw(zD%k#|$(k16Eyq_FEmq<zTRW}jrui%>|DcHu?vb<|y#`G?JotP(r+8E76*
zGeGN<t7yDH524g-;5~+nt?l3~2k=maABsf-7$*Vs`nd4=pOD<vx+{!{myYbt^GXxH
z<7YB5Jy2T!U`r=Rma|>*hWV-HH=iK3f(%dq&FJ4-rHwmvC7D#yLEM)FXV8AwRj!X9
zB+aJ)dH4OWxIJej97)3^Q6zMeEng2nI%i<!>D=hoS`Xj&k%+J$<$2GWzHAOU(#5R@
z07+vanct_-3=&!>Rx}2InLU*f!A8G;z3yM%EV(NO@Xa{lj&q!xB=`gf^w5qzNEceA
zG2HHbYa~l@71YL*dt5?sRBYTyP9JAy!i?)Kha`|u*QAwppo5P<*={^)TMGhtfd>z^
z#E)r<MWQ~0$JhINgHrX({LnFVXQYH3KD-CWNQTP`bCZ-Biz!tXI&G4pSboZnM=BJr
z7?N)k89varQX{M~N(1Q$EV;JdgqHHZ+sr6{lH?e+ks=wivi=c!clHG;AKvv1_bCFw
zmEX4tPk7H2FUXI8)}*!9I^OOCX-|PkirA*oW*ONyzteZ7H6A_(Kr3Iw9Cc6{J~1Tq
zP1eN~|ETrvGwd5kofsE#gqq%4e|8YK1xT_G!g=@Dg`f&oK`AV=6BnVHZytKezG`Fp
z5bi?1@JJS)<XI?!T(l!RGr*BrRO16~F+OKxEl?q$ht%j_>B!j^s5aaHScg?D+qj-#
zN4LcJQ9VMSajhxlhf-o26Z9yFq!uqGLV2MKrkdkl$zqsHZ(@C(U!6D!)W3pz2lvJn
z_~M@t{ncoOHSI>=;5i7NoWTV;Dx_#-C;D}%Y7W$EBh~X(mNl*8jW%>&!e!|!{lekD
z-}%+vn*+DPp~ulW|3?Lohw?Ob2HfVt(w$z?H#^z;J!LaP4CO*_v3%6m7dKtAd^H)i
z-A))JOLQg0?z~g;GzW;ai#CDza+>DkD41idcxCZwxMDQdOwq4*ZVdqfOUpCh+mE-C
zN`A52@Uj(nz-t4|I}-Vf*+{AaT1FLmw0p)e+VPQSpR)SKF>k&TcD;10-+I+cAOLa#
zPl_ptW5oXI`nUD`<zfO2m!0{llda}l!@=dvDUY?vc68`ysM1Ye<%nC(E|l6bh+ad>
z+lD5k+-!ZeFcasRWAD~mg5N52hS2e+uJT%isWQ%Y0$Yi;yLpB>J33d6%F)6^8aliG
z$KHDeMfElPqC?I(=Nu#sNkEd4hnxh-ARq#ga~x0vmFN&9iX;IQ1xZR6L2^!#VaOn9
z$b$pS+4wyF=Y2oiI#svshx690Q$^SXdwQ?k;kUYb^;$iHq(vlhqP>D<>yDoNjP)3;
z73<?iFo?=yo;Ce+5j4tiX0QUg5H8U)0=pvC0`+p`{N-kLp5l_%3eWjoKwd*-!(8aE
zE9yAk0I{AHWzXg1j}E6EcEh=$w_b+6%Og#CEZ?5K+&q1V(de-gs4Y{=`BSrVrq%c%
z*pNIxjj3O}=t38=!&)6Uk?}JeoBe|IRVuY~;xFHwXz!;O_P(!eq<np60`|{TO01E@
zQkd?drcgPR;2&tQ>2q@=F*ZS`JSHX4==@wqZ?W<*fCCeWZxa`A^w$~7ITO4X$Jp3(
znwh*y54Cwh<b?*U4`@Gndql{lE{P)y4GaeTLKqG)OJ0Z=$!ZtIkJQeYCwOP({KrFk
z1X0-6SY}i{L}OKYXM~LG&1FTzSs(9NF8d+$5u!esgFuoY`mWe!#KhYVBTtJ3tkMTg
zD&xX=l1S{T@y7Mn`U#(?y}{t6VK)omed6clqI4AYv6#i-z-vrx6tST}3in@ED}>cY
zYL7~D?k(pFB3v1P1UhT0F!F4BKzU#Am)(4ln$`v%ne=TVR8ug>1+~-)gY<_BSwlHL
z&-7l?=_6F+Ng(fo#C>}35$B(tSBi89vIT=02`<MUiz#FE635el)Ouw4#ndhQYE<4j
zOwp%jGyWMr!w!zIwjC{E3B56GVmnC=t>PzdM}G|aNbI!KOfipQhsgFlo_o%YKe(n7
z`~4kt;b@QV^yoQ~+(8UHyuDQF!*uAi0je(2hBBcJ1;;o7Gl#ml3p5ER<rb7<{}AjR
z!}_vq-;CX9U_wctz8iBy3y%bU<MMUD2{TVr^|MoWUHgOpH7O|Az%0XF15@dEFNNJy
zxbvFV-kMaiw699>@=Ix_H)S42eBVZj#_+fN$3s-gPMPVB$JReRiUMdZ3@-EG!+FkF
zX`3giE?4EeOXG9RAjSKY$8XP2_&`;wB$#tXeLS^V#k{yLnMRNks1gx%VFM;JGsdOp
z0(&hU=ub*s*)f;f<>R@L)dOqmkdA4^UCIUWmLG@S+=bn|&gEIyP;B;>{9<?>6|i5T
zKO!FE>%0(Ld!mG^h@)8eqXfi5)94B3IQ3p)xFGB*R-Yg3Ignu;3{=ZqqHS)E!kT+&
zak!^>JR^^=(*GnQ^b8M_Nd99d0ZMA+oWf&&uJB52^r%4am0ZnJwVC+5{i<F(s=Jnh
zj;C-W!avxbcP{&*cFOrj6DcB`?rj(N_REG}M)mt`2lx@V@~VIz7))?yC(hW7pR1TW
z=dm4O-YOksF4+@LXy(T`@_C|bF5bC7V`NiDPlKVYQIm%tTYzMaHtt~jrKI$z9r=ek
zJ(G@}+c7pnNk3igsxrsO8qw60s3C#M7WPM2hDFc2IZteu2=dfhd0HtB2QK@SKBykY
zcdpCLv|wjDjrgC492L6d0wqslXtJ-@)wjSy$YwO_2hl*iTg&c#^>}Mv!F!lHyxBbS
z>oW0M$$2J&Yt-SaHGAZP`7BRYnMV{O&DJJaTu9AZyyG1ix4C`d3Zs9K^Ni%k?d~*C
zJy`wvN&K89j;r@aFlV?7qf(U}7xm{Qt%I`iCI@2xE<8$>eWYxIiVCp2c`n90dxf*4
zi3jz&-Zu5qn#mZwn1tgPT$6OX$M&Ltty^c$ZL)oHCE$TBjtUBn$YLVuxw|onKJj-#
zK5OEVfEekzb&Jo0+{2+QFB4Bz-WprpC&cEEr55)j5Jet5>3$%l&Szn4*M8!clw3UW
zvq^#YNj~}Rxsa4?N@7T3Dn;&PAdkh-6q)GdIYBkf4`Cl{If@VNC632z5g{w%HsC$7
z<lNT6MK$uV;w8s#XMU(pwvYG6^s+XPf!Uw54vNpO_ef`-uRRSaKA6=+`pZ*O`e?{b
zNRk$x;SL?=_Ef){OJGHdD=0nt!-)=IK1@6k>)BS$pRavSYlEDjiGF66+>P_zN@8N}
zfId_sH|J?y1`mzXs~Xyjt?sYrxt>4X_ysrx6&uZEHK??h+?N=;%(oXg#^qm&{g8oj
z+*;EuqTBi%;NL9b1SnGHxnfJCl5x;=I&y^bMH}dTw#+-{pT>fR-W-+g2Ds;u6o_Yn
z{ZYIZMEF-d)zfnk(FF<5hG#%S1oh1XI^aZp&bXZtFk&Vp4``f+$csa`dLOyOrtPXJ
z5p9$i{~6hOBA3$25uW-9v&$(s!ZLG+{650kL(C)t#|CL?`A~H(SjUg5y;VA2X7@`9
zNs(UtsPPP3y>&^prX|l3vqUR&D1OwZHtNhfI}*Z%luF2EW_$ju&hCm&tn_n%i+H+Q
zxLzsHB2H8JM+yp8Wd)>UpF6mXW9RZGH8fB0(zA<nUahboe0y#cq{O25DHs1s8^c{n
zTt>#`t|Gj4FNg3Ee|V(}DGv_rVAN7dYqxPdXHy=}4@h}5{8^0p%sTw%Px*{glI<AY
z2%*kn_-b#fkX>?u@U6UCbwrI*y=ShpDy%5obS8Q~5DHH&O&Dy&>+|Hz@e=sXt(e5c
z8Cr1xda{-Hq@J|vxKX(64}pMtbk8}6mCGp!UaNDy??VJSeCd`c>T9n}jbu@c7+j>1
z#umpG=cOSc<*}7Z+4!xgv-JTh8oxv!oVBTDOmj`!?kW4$8CbN?79Rbx^-;QLGrg$8
z6pz-GV`0M6^Sd(<S3K2rJul+LQ`frhIM|~5&_Ze>(+8|(LfHdnb1{;bsjSez_&kiU
z7O+BL?c<#htluF>PT=3HgDxw6>4|ZV3^F8_E*zGc8{T}g2S1|I+m*>3ClG(s2VZAZ
z2oZjU-Y<g=FP!X5-gPSCe_%V}{}|*$(g%*LwOW-m0S;?Pr*^(5c$M?)9MarEB-1C#
zs2%)yrb9fH=1THtlM}*hT)&8&t~jD1=SulaMQ9sI;fBo~0HeF33MTkQf6A-9V*Cp;
zJ~Q|yp0nzRe#OsVP@5&xTtCj4QPUNuU+yAYLk{t&aJ{|%ORCaQY#hJ&$05#biRN?4
zdiKWW2i04{pLl+3s0FZ)=sS(L>Z#43dWoY(lxna}7jkh#a2w<ab!<R++UsF?Upto_
zNQcu4A3O#qm(}8d!mxp^X?J(l5x~ia(c;Fh<FfLTm(u!~?Gp05yc=kXf_&Xl*__p<
ze)Svsi1|IB{Z^QX6Wmbi_B&vW^reA_9S*yDw+vbF<EICE3HRdm%mUgN+^6S?^qf1d
zUw(K%@<B<x0W-Xq-wiSU72k%*Tg;k?pJ9w+GJnmcPKd3E$AD`#U%V1`cr?zt7?Pyx
z2c=p)bx$1Cen@?crov%ti1t-`VN6BkOykT%i66(r_$WKyGW=V4`S3x?CT(Elx61Fp
zw+Fh}zWVt0Cx3(d<GirS-{^_Z`Yo<xw&s+y4aL4usq2jMPki&uxmqZj4GY2(N~^U8
zSR<{87&7iSH6oS9^R)c#;xx(~;)ReUkd2h3h`tq(3T^y!`0b*H%4t*IAz_5nU44<#
z)jJsSc10yn!%xQ-(P>0#hiev^=}yAS;jW^{2WE|Q6A6M5$89s+e+++86$P(Zu4THg
z<9-U*g+gQ=U3#lZ<YDEh{18JEw!J*Kk{8eQr4d(hQ6k`I$ck~I7-<r0S+&XxzG68j
z2xnJWIH&F3s9t%`bnf$e{}{DBL$Iazxp4o=TJCZ89rBr)Tqb`)U#I|A5a#K~i^sNl
z6NZ9qHrj78sOer2Wl$LMyID$gqKRR%^8u6Qx1ZXn>Q)yM(iqshQCu`YQB7+!`UXfN
zZ>t7>nT+Ld8Nbp9kAT%DO7wBGncvUUlM4XPFb!K6MMNywW4HAjtk`cbQSi{Jrs#c~
z$B$rrrzNcNNm>S?yD)&OqDzRkF5xGDV$o-Jy$+%ZJ2`pH#v;{ceMS{_y34H)hhT3H
zM%t+g=M}77T*b0nvGA-owvVXv6RJ_?Yj2$?$jTqMBb+G7GD0g{saG_*_ImHgFswqK
zz34daL<?T!6yUeAKgY*jFYrh;l0ZEvRmonZT-|(qHT0V=BEB>Bpnd@M^7M`0DdXoa
z>GWaYj3Ikdq1~M<av=!Hg4ttkueku)s_UwR(cCnAYJHJ?Tstfq9Y(19Digx*MU+K~
z8H5jf-GILeO?#zQ{e=(jwkNh1G?B-Gy|c2I(0@&oj|Cd;TI+P}_lsS*P_$p+g)7<U
z(q#9@)@#Mq!1PbAslv|4#;tuC5nN&Cr{HgYr1zr``8HQQCIyQcs6AilA}qCA=fj#r
z2Zfi$q>^nJl5HuHZJ|N@&~vJ&9Z~^kgZ$4F={>7E*A1o=(JFkn7z-)io$c>2-jvuF
z@8}d9sO9lQ#ix~BXN5N`2{^HzqHBA}wZ5hviY6TI@dg}1onPI$LyGeE^S<T)wQdeV
zC#}cEf;?(I<ZX<*@>4Drq`gCiC=?~^pK}%Hzc1OrV*iNPXtRnxZ1WW<{mQ~z99!h=
z#ijzWcZ2;pGi45zuK2kmnc<|jQ93Uj2HwdEs%=9D4NYHcUSh0ezaT?YnXZf-Z2gDM
zX@<GKUScAF=N{mHI^1Y0adWuzwljQfjrBRl#}tS}6`Utu!&#(Q6c`imXZ+p8JNofw
zq#Cng_FGtW)SIglM!&4@Mm$pRPgvADc!Y^bYxG{NYV^H&Ip4<eAYG(%8h1YJa)Ed^
zz{C!c^gdq*B!<^&xUE*M2y8wJp`Zs@X^y>u5*4i<VYlmFJ&$F%HvN+CQ@heykuuIw
z4N`_QhQ6tXftz7K+Zu&v^zVFP8QfLqYDoG>j_14HP!8-U>gPo<R*U(0i=&-e)o7;Q
zi=6`lgD*p4i(>3>6ICyH9SNfY;ii-_Aqj94VY~WJIZR>uKoDt)=n5;*F`1LNHcXSd
z%(T;dEHVW_4KCb!{Ixb~*QZS))E`Z^ErFqFA3RY;ycRrR)wPyOvu)A(;(n-DKN#~x
z!$a=UA^OFa#==N6{<MPPvxN?+nL2<UzMEUFk1?+m>-R#~=cB|_=JE=7-l~UYP{Sss
zXDhHBsER;F$pV5c8x7lh9bwM=Yf7L*X>)Lb0-@|dcXU!(5=*YlFl+nZwM(xbcyiC6
z>5b1SCZ>YyJdFxqNC(O{l>@Jp^mXlfU6W8N<!gbAIRVNz$!Xok7p)2xm?p9H$(N=X
z2@bBTT@REqQ-8_#N<O1B^iFE0_#JeVYy7RCE#XZkGj{-$)hNen#YLRM@z^GEcK&7y
zS**F=OQxI!`q|KBm)K>IF*>CVpyk1RN|}ohO<0b4e>cXYlWRy!KpJdq5OPUawD~9f
zp_XLB@b~w-vS<5p1IM)zEs_w^QOJr)zrby!fn!6~cTmc=DUsnpQXbtRLF9uDQ$7Mt
z!wVg9cQ<3$1{UQtyT3)ADB*vI!c(!t7%R6GXDzd<Mpc8zZ!I(5hUqcq7e0cx2kc1`
z04E5J6RxUdjpZw1dvRU*C|AdA3zhIYf%u(fhIqer-FY#Bef@R~*!Iu`4bwy8ICn2D
z>OWp@P1-eLdBZ}2J{Bg1`3DN<8Xh$WKk2jVLC@W*pS<hpcj?N#!j905&BdoY@)2H+
z`ymzU`?N4j)?Nike|?oI)#&daHn$<=E8L;8%d5xPc>Ch}$2%Fkhalt~Rg13BP!=db
z1U<I^IHSx$Hw9~26Oc!@4;At$b#hzb6Pu;9Ke={6G+pl&Ob>{kxxeffiF#82eHP2Y
z@hp};L89pX!*9=5S|=oPc2&~AtO(Gu7G`fUJaPpKaRHZzz0e&CWyjhyiWn-oa_*+I
zBTzJ8yyHkSOr}M7Upmobd*vdQ1>;K13>F#GJ-FRY$zmVB)jM0JQAxZC^>}=NQhDR7
zoHs??^;vvv&iK7vL7%*xVK%hNghJED^UbDWxPrg^DQ-NOh+{`9-xJ0QdT%AvR9;i#
zGl`9w$Y8zhVwMLH2Na=$EjyMqjx`S$<I!~egy&j+dWO*Bkj5pXcDfX9MG5`H*kN0K
z7%1Wu`1YJ998ILAM*C-(o~n!ZlG}$Kxw5T;Fze*H+Y)V*u<%-hiYsK0FVWS0m%y1T
zKbE6e>P;ix&j+l;I+Tm1Z^WRJN9r7C^Mm4p>4|S%?I{<XvF0hHHRYJw2KGo6gW08y
zo-Mr~GEs-)Po*E`>YmmqBs-jbqEq~GZdAZ3G}7sv?z75sysF}F>P)b#b!yP=gL|KS
zTzsFOwr^2=c`Aet>GtC2yOIt}XZ+}}(8s@+b|JO~zDs?Uv`R@3O!d$)Y=-#B$AE}y
zcbzb_3p2|Ar$X@XllBx;XRICm<S&BD<~#AP&ZxWcugQM4#9ly?xz~1B+jnd}+*6Y+
zT81mCl62J)DJciJW`*E55*D4t%9GUtYDM^txL^8^w4JPUHAae1Nn~^5K>e|ZFA0)^
za<6jN24?m;=mjQC{@IEwrpjfUw$L6zaW1e~6N#~@KN$EBBD@rK@-bqkjfifULSuNL
zu<=6)4!=vz<X3y(C~VJn0HW^WEt1=)2@)n!d42lCHILlmnLrh2Hy$eCH)kP>OD%2g
z`bkCP1cFZ`P6Yg>SsM4hKsWcC^o+iMdkU^4?%avnI$(R(7EDk}Fy-5G@7%A?A06Q5
zAQBhp6riM_bFv#1R=@AK(Um_iFRiJy5eF@c@3`Hbl(L|=t>tGm582`mW#+0*<_q>~
z{)&z<RCU(HfmGTk{KUi)%0^Y);+{7uytl6&!Q{&-)YU-y;4IVHvS7aFfa~io4$iMu
zdJVrgc-QVjYn-h%1GB;%@7FZmj!%jsIvCEN)QZjh9SmZij`L)6IAU*Bau&wpoD28M
z_EB@rHiCR6Ss8Q|2wch&2`7?O)r7gw7oDq@ZElUm_Pu0^Z5B)~h~(|i1*!N?b3g(|
z$Pe30B#KrpMuk{+ZnyVf#Z%jLjIR0Wg(MO5bnOo`b;H&nh%!Gz#vf@CHX83yi;%5+
z`_h67*hJ2(8{ykTd0=O<b@X`e)pAHQl4r*X;ohE76!t!V5NL_IVR$(26w49*Ci61M
z0T+dSfwDn7Y){$u6<x-ko@<CDOL&KT#uP1ar%Q8-iK)od`Negl+6hI<*!|xrBiXwY
z2uIcBhYT~H&J-YVcXjp{sEF?@K$r;#Rp3%e;0?+!Grf};7XATYB;SjmtZZ}0<kXlg
zXwFEu8}kCJd1+E3JK#$B(1?PI{S+Lz-b&>c&oE0o&qfzw>m-*9?z)PoRM0XMO`?j^
z4{;7NWI3nu3UX$qWYr9M@UC@RU)Na8IUaI}rA{#^D{rEai1}FRvWjtraYZj`qpq3(
z1g$rGWZk0+qZVbu%zeS;w0|^@9!NUSG}<PM_K)@#nDG=Z3PhW$3bzdQE-l1DpoQP%
z3lg7YxjmOn@S#x{fA{3}OlHXQF?t$zj8@Mg=dd1_wv_(pI+<!L)lL3$g3mao%ylFy
zl^5*?&aZL(iL&1xr`nRB8Q&<W8cr$yIXfu-K18#U!6!8{dLLf-`$R5P8c);~WVG`W
zWAbx`3yZ*{-{H6Sw-pEe$%B~&8dSXCsg;1!F*JL&VTFSyJ@__t-`#C_3htVZG*UoU
z7~QfJUMX6WGA=lc@s`8`S6_l6#joi`?04>5C?q^tJd;pbzgmdR%sZ#-Lfc!Sn+QsB
zB(59k>tA7DC9t}HOYe00#nTDFdu_~7cptbE`3kqqvzn+dBA<DPKlbawfEQ;v2#)-6
zwSrXs&C}4o^kDBBLPKK45r5yHly|mc;D=IC_@8^RV)^}qt7Ag~W;V`oWbZdv@l0iO
zww{LasZVaj-Gb!U2zO#2d=m0fU-s&I&r&*);+dSg$A?cI=IXwv@E1Jj^4lm2_!Bss
zxTJFbja`VE^X7hUy8UhJ+;>lSV(F<}gl^x)X(QWNA+Pc05W%OXj&tcbA%{?Of+urz
zzZ1HA=_XY3yu^Q(WU5%kJ85Oz35CGm#L?T5Bq(f0lOhu@MTXohPu1QOv@6VImRF^c
z#o>L9Aiu))0<)hG0q3?n_-^CN*W0~dS<4N|BRBe)mnl+e0w+cb!+nHB8S@>n9J}Xo
zLv6fuC2lV<b*U$W1M})oA!vABenb#fkt*WtgvnOfHS>|xpw?&ExI!v^n_d2NfwQtq
zrP%BOw~s364Wu||wqsdQW*)ts#+TqMBkq09^kB*LJj+;o9umR}xp*Vx97qsj&(PWd
z;fzj(D-j(z1tq`LeENVjC#27P79|dG*Ea{z%BJlyMk|>r{wj}F$!kAmtE>=um%XbW
z`%a0uW<{)~gKL0mexm1V^uVR|kQsSE+LfY9oKl}&O+(VIk1k7$E9b0fz6Xr={+q)#
zT!{FOq~4>v2_v;<6)ub+Lc~gV$0`20tlwsXc)QK?)t#IL&>6f*K!<bQIuT(IzxhWA
zH-`J$ayjOsf-}ET45#EJU?tm5m(@m#MrwQuPYl9XShEAPATam-o8dx|q{+jIXY{w|
zTiz0P{}Cw?4k|q}=OTmgy7>W}3SqM;hxV|Enq4d-n@8_v+(V+lRC}+}&9@~v8u?DA
zmml&%a@hadeIG#C$^_9`I`gt^{SFk!F`Og6vwvy<e)9`-M<8h1ADpgsZF-v25>{K+
z5&I@Nuya8RJ_f5+@^TMUfz{RYxNYh3V?Ac?F<!r;JE7ianbygETT4eut*jdiJ`7&;
z9ZzGeYt&4}^b^SnSh?AtntGS^0`)Zl2L&Q9``J&r`A1(Jwmlnh4`KK??$<!qFRcUz
z`6$rXrybf)j`SR=EFBZl(SM{?Kogzg)Q>Q$k!r$Kt%SP<xPA_Ko*HOppb9)9YKbmt
z{hmhNgjODScPV^Gxkf~}XmPco$znNMU|>hv;%uDD=Eaw(ycNAn7(R%*3cv&tYmWxv
zXk~;Sf+yBov@;RQ0tCGm>HgiSl&+Pxj=C*3Hw$2<*{y#rDl-3gd6g4K+aaP5f{5D)
zo%ZR7xtOcisfS`dKzxnB-~)YB;3o{&^`NSlj`9m2AX)LM4#qa4T5aQX6H3Gk0`1)(
zIj=HbVT@a7z)8yF5%eK4K59hhLeM2XiWQk9qh}#8OmNLASJJg9UjS}(?P6DrdH-gR
zJGiDaUm7~NkT-#m7*rwm5dce_qCk!jO^-5}EN(l+Zg7ZK7a~`0Ilbum&2IaWY0W%p
zX?V@aunTS*OYf{!2g!8rLD^3y5H49WeWVvIy4W1p=1g~!>>|9F7_gGK4yM`2?433|
zHbl(m3gS4Ej$a;Di(5*X;}{wov93b>rLMC!_(D^1m4e7qty^bwQMSK6SV?pZ@Soe2
z95i)}X`lx4MqeE?*-IQ=#&u!%8;DVMZ-y^~5f#iOrLQBhkQ+lH)z#j9l@IBfrUJ|&
z)1+J?*X#_1^?n911AhpPOf#iB;5yG~HD6jY>tyI{DWjN@mA`*T<5#^+cdDOU7IWD4
zIdCnCz6>3wCi*i}w^lYlK6`%0(*EZD?KQjc{8N2{`!;&}GFPrQ#GvZHmD<f!A>Vn9
zvhmW|V`fH-u9W7!Yq8VH0?VRoPfI}qFq@g4S@~6OqHjJ;`Z`?yisoPud7U)B6B?{U
zV~SVJeIDlZ;bz<fO;gg}TYD=~9%1~%ySWixBnyw#j}#%WiM&J%jJJ*hkTK+CP$dW@
zk2F_hs5DWBm%_z8GPxlzGmrmRT#35EeNY7P1D$Np7VRthaP@4$t$4U)?=dgL#2;w$
z4x5O&p~m-(f2q`Eo}`(r5x+4W{A)qSfy|M#M%jVMhuiUKjRc|vE*hv^#S5t<zj^Fl
zH=tK0H*OxM9^tVQf*K(j`?AMC+fxNJk7Yw*L-r(pA6ExfQVBFo`u7t?qv$}u1}bQw
z2uLY*KiUbd-9pU=fDF(f3u1x>dgsAr?G(&GYf@YPeB$pLR_uKp=Y_*~A%DN>$b>=v
zzJ$2nGY37kSX|8yglGh=+nZjmK9VBd)m8H9R6NRIPMhMb`*h;Ug?~07%ymD#wQM0!
zSL1Y-YfWEMIKsYeCHV8y1gq6yPKkR@O1|$Gwp{h9$Y0ibVVvhVXH@&a(zWLO#pd7H
zds@n_;65$X+>_(|M5~-#aG|#ifJ|PgaN8J?+SqS?HS1SNYbhI^loy?(#A=}a<VqUy
zR~>3&$_b~Waq}#1)a5YNfe9x!n(s#D&6!3$i7^DxLx1?SDsxj(lZ2;!0UtgJcCOy;
z4>ZuNyCj(L`13e_+=PtEhh+wZN&drME)+)OvXyOA7Nv2cU?c#@c+8UJ+1$`RCQSR$
z1*dngj7;N^Nfd`ANm)z)^pFqRfd$(KIHbR1WK5@U6m=R4uHeEuxp|c+)|vBpi=Dho
z_YkV(b111I(LklQU)cwe2av`BW5m0g$0=)^DV^8dDc7j1T8*9>zqB`VbER&5Z{_lC
z=LO~Nu>Q3-UJHCDdi-~Os!)y=N`%e!Al7WuG1kM%=H>zd@07++V7pZg;fuWu6l&*I
z?TE`7zw@OgEmdoCLV@f4#H&LnKbJYX#m(KS?J{h3Hy1^*9(NiSiC!~n<K}63bPhQY
z7M{xA0Iu`Hfy0~M>al3%>XN`ke=VA$$963=CVbCgcjUh;5>>p*0&T8p%l)G#|I(KK
zQ$Op9Z`wOV!YMz~qa;5&Kiwn+L9rCcf%ByaSgjq}2pd6-Tr=J*NH)Y*qU1<w<bpG%
zJH`+c$!D*>BP3g>Ggz#1E?d|oER%94`ZaZC2HueI(2f^7-EFpu^ii@)a?}bgI@_=^
z%c_fR6&g5ElW4wuy)horh1qbp%p$H;9GPbRDB^IPXt9Gb+4t~bPqNTqzt401%mpO#
zp1nO+7T*i!nz#{P?rw+^hbcn-IOeohc?x9D6%qXqF)<NkT%&|vfUR7(`(4WNya>BH
z%D=`_35kdJwsyFB7usntbJ;goZgTv7>}s(5u&=k65H*S0@7e1c^(Hvk&o21*{panT
z@~`6jXd7ea?-wOu6)rXs*A`l${f{&Dy_Eq+#GAYeS^@6IN%}b2hrKZzQkN{g-ny|d
z8dADpEVH8W0N)5$p<7Nmt8YHl;@de8M&{XL6|emw<WNN+KJZ682H6#`t}?%ssyhJr
z7}j`(x?hR36Cl~!38Z=a^Tg9b^gq6zj}Vec5t><w5c&WcHj$|2ZSK-F-RTQm7%X0K
z&o^{)<?j}uzZ?{pG!<P1oPj&;mYP-0(Bld)F!caEK%ZN(KYo&AMh}J(SGh#oI+u}6
zA+9oK*Qdw)kc?)YR)Qx3HInKfIAtDh$qj#9S;kypa^^X&_tp;cu+as1PtZ`49WmWs
zbic_%q__l6=^-9f!UBZ=uXN623yFo#Yt07qRI9R82CVJ>O4ed0qK^@`9wcj`h{`pr
zEdMXb_Fiw8I{tUbRsfQvHShnIWa-S46mvxWO1AL;t5CRR&%K?+A6&Oj)B8;W@5~-#
zqrIhyUxe-h7HSJ}SvL<%cf53d1=3Grs9yzJdJwFynaYQn{e8xIROYKD=Ekf;QK|p1
ziD`-dYqq7{A;Md^qQITWA`vrb+%IXxht4&8&V<PY8n+1z9Nl&DD5;6QFnu7ps@qiE
z6l(TB=IC(^iXszhqT~^kd(ZxmHI$&}R~0Qu51E*oICt|GaovwB*z9687!ugklGRGn
zoJLWEkifshW%y;-gsYs|u1(cRKJ)lnuGV^$xiVJs?YhX+l^;tHCOnmzfHqB;%k(|Y
z*!5P<=teEE0xj7bGYOcjl>lur46UTH?~5E9EIvxl(Ms!tKJJjbtJOmD_Le2+Qlg0d
z`j!KZPt>BaIEbAyaV}W?a~vi36M<b`!@(m5kC<apY>X!3`EM0iplvKu^%OnCx@;Y;
z?&MjQ(0f#orzjK*=!kS_MCOqOTsGR9>QSPh@MfTFijKuz#8V{CrZv2fVWz-Qx#G2%
z!Y4*dRrRdPS3kg(b)5d2(iga39hXZ_CjM1=W4FG+;@%aNuWn3ls}_ky(m!xE#Qj!)
zw@$2jsmf-&fvfdmrud<lXt3R6c}!a#g4LBunK2&3kSa>){H<;G)=~55n)lLL*65MB
zkzyf0?C<kxpT1PleQtxRm;%1gZT)MN0mARSlr`W`>ZjNChJLKakR)fk$pqm}Mojr5
zyh+hwpmNFyuM6|&GxlnmSr4%Yg>((wL^Yso>A=4e{ykW8BRh6|+8Zm#v0*@l$Ig}m
zM=?aS#O@19ARE@sP+Qyzqn*BZ$)R8$wVFs#ADzXsedoDTjj@0SHCImSwmQ1P>4xhY
z#!Y|<t=~Gc>|>y^vc7!ZbM|+L5tDu86qHtyY%nW}59q>+5>2zDM%XaQ-gZ*xW_<ns
z8)3t4g#DTHExj+-^5FTmHDA}Gkl6MXrDW}TD1V(8wz4+rE|4vCH-210677h%=AJV!
zNWMxQOIx}<c46klejir8VG4AO*S#1rsK+egb_?DJR!Ax%!b%5K5N^Nt=Vcd0s&L|c
z8MTFEBBoAeQZnZd8pilvW{|soV+O4inK=K_KmByAv~@%4EUY}7w@m|B%Gj*-&gj0i
z1IOMa95}ry+4SR>t3t>%dydsP9RjH8iFpo}>GdUUN;3LxvlV&Lymr8&Q83rn*IVv)
z{2rz#1I!5Lwdn`NvDk<CG<n5&QwxehA2KfjXqQi~wN$z8)IKq+hk`?!B~Jn2o|^PG
z!W=IPbF8vP0kd`dZ~FId&=n&{N9W{ft?Kio@b`1{gWQ>GiRyGQ51i2nX?xx}AHP4P
z6bvKTg-SOq+#fPzhc^E_xiqf9e8H=al$G~}zU+4ATi^IMe2j-=L}W@z)^*tDtg)^)
z?{648)(w>N5ZbyPXJIlFz3(5%T}3`tuCQ}w13<*oV&`8Vvijda<SW2EDZ7vTSpS*}
zGB%8FT{tn_)n6x9^*mS(-9g8Mr+NCD>{w7Pez50i=&EEuxzvbCIh{U>{qj|3k4Z8G
z`Kayn$bInY^o0D}eff*8Psaeh0h3gF-<61qGjZ3Zp5NtPJ!`buG;MwP4fE#z71To5
z?=%0!|B0elF;N1-g+pb{khFcy<_j?)WS$C_b#9=%W7&>q@3#g4G%0UN1CC0zLobKx
z4s0#70s8=wtfIG-d0Q@n#M2L8+^taQjGGBm=t2a`CjUv{{dW1kxZ7UFQ#i&?eC?oI
z5#>&I1G%PElCx&OywN3fElU8{X7+ka*RRI_h`2Ueb!Yk6`yW4U);W4V_7`&D8&v#3
zq3K4)r?f@CDyHRJDET`Yu^71raw6$`^xV;s?JQw6>pXbLk@U!=*e^A|unM~<u+j?+
z)7D4L{(2p<Dcq`@vo>A-H5^HPl79_{>Sj3i6M*5|-1)y5j^EK^p+gQ49PR%Z4%z=H
zNAUl{b90sov3TGPEi^zlSC#~11xn6<o}oErz*NRrQ4xr>aB2_>q*l!T3H01e@=pz-
zOUgbIPLnATD`s1@F@%5`aTu%Az5$Wm9^=1L5EY#PlL}a{oLQw_fbeYme+Z|mf&Oc>
zqPx<F@w>QrWQwFVSeHtU?+O{^QZJ!Tj`CEvgQ?%9!JGoH8L)$K)ZicC6}`$@I4Yb0
zw`@ov@>V)tUq#g%kUQWY?>k(WiJQTAmHvi`e*?UUxJ5(4qfLHjFer=W>6J_GH{loy
zEGgJ#grH^NEszaN@tW1K8BT!%)fePt<-|oh0^QLHv8!NL7}uDS2#J|91w<`eg_@X5
z0{PqB<})wkZvY$jHykd~!*XEp30th!=&e`jv9|vJ@}X&}`@@jxdC$kpJ(A;dywUf<
z%v-USctskjT_7I@F!6x!UJWl~4EMjnX-`2@5bAD>=Q=&epm)wYsOGv7ySp`cM2K@Q
z$bqfthstX9+Jghp04f-7g~)$c&trST`Z#|ly5uj)2Zs-i&SS9A$*C+`-+`r-Rr~^5
z2MmycqA7#jg}~A~{)u-z;lv=wMbNK{%d6g{+Gu-&&rA2lQ|SSzJpV^3vp1d#CnQvI
z0Sl8nBKdaO4`OQ~00hlH{vk4cLnP@DV@NF6*NE$wXq8SZx1JUPE3x_~@*b1^jD6Qc
zo>;%nl%x<hnOjdX{TG=H$tBO}L8TzX<U8Q){*BO05`4%(kWxm-byT(yAxh$NOWY=!
z92ur63lFAxdJRI_IS?T=9kj@X#UZf29OD7nG(OzL;;f*Zi%FqzU&lNjA*;@Oi1-Ni
zAd0z8gS)^DW1OOQ+#{kOFrYO$Zw1>;BJBSn!vE#K|6C5ZJAo|$yhQ(F6+oWt4dFCV
zMExQW)&^D~XdnL!WF5@`^2Yt9qu2#Pk|A;{O4Hl*aU<9pvE{yfFf%-m==a42o3z(_
z?l~6jO{plT&312=<%L*wVc;m00>HGbvJO?i*C{|kzJxwVjuOXpLY~c)3R}&_*;3!_
z-f99~fkJicW`5Qzrde`aIlo(^J6G^JtSn0dufW<ZN^d={(`=`O3(_ZfP0)q00Z_=@
z^M-ceO}bumU$Q;xu{YZ|xbV;(nSzr`@*($q{W&2q!HZsEkT^k{DcXYP!PH9j`XA{G
zyw!>|t<p`AGOGhJh>6&0+u;&G0?6%oU&>zVny`E$Lsj1?Uruvw>j|c|;&X0Jo)`Z!
zMJKO`JO6sOR{22w0x#MUNIKf%crbG02@(COliM&`$j(EX*Q_eiTfJ-E)aO$XQgW4;
z4Dr^60saCDCJFNwY`UO87EvZ!HjPcnlVt=ef^`}z5jzE&gU32tB^R8z#N00D)`0C@
zgv#3ETuwO7?_GT{*5%EDWoMfqOd6$irhAz}@&x1Jj=ryv{F(?~A;v<|7Yo7IMX*(Y
zzhlMfUm3)^c|9jtD4TKl{v@wXJAqe5coph!S)B;*KPT;m{}x@qyXya@=!t3~!%}(Y
z<I}_;_ZlprQ?zG6FAreiUO%kc`1K}(mDqs7FB?A+!mLc+{O#2Oa|iT##a;v5^dbS^
zZ@0x9{*Y?}eNG3*=sjxU^2X>q4HQk5r@N}~Zz~_zG|%BMm3OTh=Y|Xppg3%Xel8t$
zVGaTGoM=wO{8-n!Wb?k{7)DjsuQD3B-;adjY1HR~=4F5QFOfu$KgejfeyCcsTru&@
zI)Qcd`CY@`hcV52R3rIe^YZt?j!H>Lsz{y~5+FVC>Vf7*IG4;QF=R7j=v*3^GqGx*
zn((UPeWyhEL*L(+1$cZk21|s19k*3ekdd4r#H;i|BEz3vYVtgeS4QE6HR3l;hSjqh
z)@1!<jmQEvC;x=VuueUj);_P41Xs^F_0z-7(78p(rx(cQmO2(6X^xWvDOQE&HrDr!
zugAoEO%~SbpP=q8p3zl!_NN7=)q#4Z-ZVgE+;=>D2mJDK@0`ape|640O;=_z>{7ol
zuJSjB(L>xH0qOuP{Wn*-Fgn0AQSow}0d#~jbHBn<fzS!J%u~-3lhzCnJ#{sN#mPH=
zpK=*86?}U}L(KVR{dpQ|11%d|)?{-L!q@=j!H&v!g4T0y!HZ5^+CKKi`Xuo6OuHJI
zH)idjm&s8l=UVFay5!L#K*=p`vd@G4fWlX&-)P2zZy=-h3Nv&Sd$P3CtwMEIYN2yt
z`oGQ%eXb1<Vt#GShN4Orrjnpbi{H&Wv6+N!47_u<)A9+4tcpP+hK-A={fX2Cg<`Bx
zFX!7Gqkc$26}we+dQ;pB{e}K{Bnw7Z7f?s3zt(1NX~kPnQQ9W0Mf}8&0TgBAfL6zg
z&W+9$kjJRJMwebeGjfo(?!?Vot?ngqOjT!%JeR<?vqM}A+JWR`4ldT&g>^)q&!^v6
zY%}!xU~*JEyP4+>kbJ`gSXvjRZ|tw)-I%jQCVgGerI-<{YX`>;+eYe#auz77lpV^3
zawi+d<?@MO3}KYxwLF(n4er!BI*a52%U)n~;Vl*G-J$5k%n#!9*;dBi;6^<uCYVAS
z=R&`uVk-bS=D+?m>;`4Po^KT%ww*FzCWDRy_+LvelEB(jL~E`>5Wgo@<D<`iNozQ&
zX6-~LnCvK%Ua~#M(Yce&A=~@qBNuzM!YRh+_1^lLmdud+=y7E}pc7hM&>J<lV#K%r
zIEimwB^F&M5Rflq#-p73Wa=B^G>*szkqf?r=!5bhd7W(IDDKTv>s6aRlMi@zd4jO=
zUUeJ~wQzce<IWe(UUy>8^$-4qVx1d~4594v4%_5zAV_u<>f41`NyND8o`F*N&&5=;
zS}$kE4G;p(-+y5>V^5p4OQ54w^p#q%P@XThL+NfzD+Rj&q=a|I#)BJUEyzBSf@faF
z&Vl<xR6>--62BSfG^L{x?pZdM7K`n)af&7`!t3@@RkkJ1OP$3m9TnI^ZZMpt<x@t9
z2Fxt`uubyckf4}|$&mQhyukckgCx4NcQuQ48iVcU;j0B(rMEizjyu#JzK<4YYu=oH
zeLnv(Q+4W1wRsw7z2Q#w<Bf~3dJv%S7u5?!F@ewdS1Y6e?A=ta_NZH8flZJm0{ZaR
zvO;fkBSZ4KYHu3nZq9T+tSy*s`j;)@5J(t4GirPv$BsA=V$-#>0BPZHXn)-4pWDoZ
zD;)p&H<q``@>YZ;V)kxaP#vl31|TAxHMnt8Wf~iBf~mXCJaE${hpHqckoOS0h(aRU
zlP*{rMm3x8w<kR2KQqGr1C@W)b^hl%2me19;RNryFoC-?zp*xcDNw;CHXb_oA6eu8
z`FKdB%4%C#E&zhRtNB*j=s-z3P>?+c|93tmgxU~@_9%dL8~4x~Z-cc9s?Ts^Jjs~}
z7;}(x8M}i5KJa(mDQxe`VD5>r8zc`&#8k@M+*AN8KC=Pw5phn!4|^E%AQy0Lb>jvx
z1BC;1j{xKD8aFv?19G26%e=Pv|I`9NBv%cA$k&hpT)z0y$Qp=d#!jTJs}=#;(pSK$
zyEWDb>usEb-}!_PcPUrwhDik~_nTGEoTx77<jrf8+3R*gg+XOnxOpUb8g3n8Wb`4e
z6K_`x5V)$k83Zu~i0QS14vA4PlDZxJM}TTq7)|<_^#S3f*i+m4I*q)OIfwQ%FBD=*
z;gWAX$cO#7&#SjwP%l~rrlAbtApxmM;o@sr!18s>7g?W$-yN1MX_Q(GIM(kd!}%ad
zov7+eIyG@{3bJ)i_nMs`5xWh)pmI$Xig)W}^c(mGXL(#X#%cT~d=Y2lpp<)GE6%oG
zGv~&*gxL^Z9^6udsuT&ons5M$K-u6$u?K04jM#s$9(~RoJ|3l6Y;Rx~a<Zn5xOhN|
z?dRr7`26L-PhlSTId{nWR}5TqX-n9MGv|CWq1CpdpGxpvI2HU=)00AfK|^*TCxV0W
z;A=vlK;(P;>(S>=!p`*eAB5w#EuK(EeF?m+;6?DMf%<Oc9Y7QMP}OC0@k^&9L~2C?
zdYLn^x!1?hOce_}2=#VFdP=WL6?{eF4>(bD_cFX&yPv_g;cl^g@m>fY2Z(#Qv8vO{
zk8pa#BLx2o&8l4by6ntk%o!+~y*h8@cu+v*Z6i&H)ba+#rhPDx2~Li<?o^cW5mN~t
z?NEgFmR}Se-kSf8K8=vg`2lpXM-CGHN)LI*V}dycc_(*d9Sy%TDi((%0V7NgAxC<_
zzG$Q@2NO_yJ|D;*$a%9U%$@v)>G#1SsG%Nh3UG)<N^9hf+}hiNTA#E0ytCbg(4C=#
zi~Xmvha5Z6O9q%aZ5Y49rG@D$M=T`1gYQd2`MljhD4`N|vSC3Ru^e|?wfg4<;)-0Y
zMy7D$WhrVZt4oe9)UY1IvK2kZ2{>wFn-b&w;<ka3sL8R!XT21b-g<{jtGldN)i`#b
z51g`F!v({cwe|zz;hSGPCE!?hioeTyZZMlCM&=Elo`U@#q$)2Et<(l*j6Uv*-x#oT
zof>~}#qi32SKFu$K|Vy_^7z+7QjM$G3j=aq0=4*K;eoYn-wTUk1b<p`ue{G%ZWYt~
z9>2Aupi1Fk+3G+;;=1Sj$esheNORoLCRFa4&s(tpOmBB26e&WK9A6Zgi13m+J?Oa?
z8;;wCGlUZbex6PS66cu@?@4?pNvK#jzev~_K)EpyB99EGgmR$x)x(7h(<UN-!X0*d
zoC45L*;fe=12%`YQoTn5v-UhrGg}B*lk7E?+zV0vjoQGWpp7$8hjAYQUs6h%B&GJ_
zGn;-EiYS()5avd?I3}w3JEJq)A%bx`F&0!CbUhEqn^oex)jt<mSqMPxJ6|$XG|EG{
zavwFPeowX^{t>_baN{mRF#$eM#GZA9{ywv$0$<#}c(QOOH{UjV`(VDqVG}r@`1Ti{
zT~NMKIMXx<SRH!c#M*?i4?Vc$KXcuhZ6Eo=&KEWT2AIIyo2s%25-<Jg*xpU(lKecD
zs0!E{Ils+C${?M}TAt>U7G*)&mW1e)H2rn2aT_BE<BFppz|n%StwISI4KkH_L&ISK
zV%!a?PW+C9t66t1)qp0L?k~FfbaN*nfHc1Z8--$On4;!>lA#-OCkf%|jwpBHMX+K~
zmHjLw_*<%ee^zb%aR~b^lY@7#N_aOu;~k{5?)fK|i{F5a&ULq5*>9)?54fNOi+s@H
zrStHVjuK^Wem!K`ZPhIz<QA>CS>M!{6R47l31e+APt;5&=T5}(=!(6eusMz{foCOz
zXeXEyEbSLPgac1ISoSD(_1)>>RURTp#@D=E>u`F%8T~dl^zhZkf&4GepG(CIWee`R
zgkX9vgZ2s!W$oqPps#C_8HDPn|1}u|QpIcjV4&Gp=s)KzieLWZJY$gU5R4nb(VH6g
z(pXg(GfqTyYQU$^bc&D2f?E_L7sdEobfc$W3be)yMOp3G?KU_&FeeVq`AZC*kH9i3
z0ZrUhD3%X?Q4RN!rYgnrJvbDH;-2%mb-2ei%9{cN(Eu<jfv988+3M2?RoApG%f7?z
zHGr@K%4esilFjc6i@c&8|Gbm+Td;-pWyM7ezA<F$tb8uuhmo*29$M#E=}(~%;u=m_
z_DIuz$oS(MW2+8h8rT+V6zB3DsY){9Q-ys4FJzKZ3$6h-KAyhqgWpitn7N+Nz4an_
z?aLrrVrMwxd`m7<RclA?xyVDOu!r9U;dSslyeG=*JePxK0Caq-3Yl7NTiNZX_Zg|7
z3e$|+UcZ-Y|7Y0dtd#)OcVK|{5x?VC0b~f7@`q8m7$-OsD|iE=X&6S*oIjt70jaz~
zT+5j`w`7luU9UOMJyb1)Gr_9`3{4^gi!!3ZxOZn_7jd279kDN;OSy%EOJsE@$SwmO
ztmG1%i;M9m3XbH@x>6Or+v93awXNd3mQbG)Ff5*k(AiGrulV?r>Ss5I!#Sd1QmuDS
zUV644Nv5*W9=NnTnEMG~$o^nxRv9SyC4b_yC^hBNyAdwG?jC)gwQif;!1lsp$mWln
z)=7DC=#yKw5&%!QTB;R6iN){}c{E+14=d^N!S*?N5_FQ^Au8bMbUl9+5X@dMe+kVg
z-I?-7M&fy0cf$jz=2?H7$9Q9@X(3S~>o%=NE|-^wGrRNC!$vA_c6{`~db*3n;U~<w
zm)(dhCmZ1H$v>2PJ))}YDej|>8%*o=HkjKHmH$<N2PJy<D9<{25-!zy_QD7YITjM(
zeSK-hruyEYXa4d)Zzvdhx$N3DHWMvTB+wp6X`35-titz-%L>>-g$>ny|5n>jJA$%^
zz(Am=8&1HbEU^9KRvj%+KvepZ3hCbNgI9Ze`wUB$@JA5g^n7oO_+GL1)x|uR!-0HQ
z?E4E2FOFiT0G8u|ncfes)oWvH=&f+ZX)GjeyywEe(>>r70)($Eem2z|YG{DzpVJem
zJc=k<6X{~Osd#+k%Yl4_brw_zaH>oF*<7~lF(LcXMCbH04QqW6%Q5?vM0GL=U=&`s
z!cxNTMwG4!qh;6<52;#_J#O<?>J&WIt0h%`%r}O=rH!XF<pj4zSc#j$cxCeCNQ?cA
z7k}@;F|XP$>7EyiQ#4C@pQgnwZ45)=E}GsL5vAT*72%)19EL{yA=M=m#Yj1w4ddBq
zzb9=Z33;`9fFXlhaD{PV1Uof6qT6tzu;+E(zh&eh^^d?VqJB#Pw8s?qdG$PnM@Ha=
zQ*4HWH5WnDPj~ZU57GM|uh2|+kDmI)LM#!v*XV{o@EJ<b7H#1-zlz>s;h9gzxNSJJ
zt*>=+Y+Mj5{9t;c6E_oE6sJS~{A)xSc-v^UYNg%j5(1O9wHxc`*<IS)pg}V-Kj`N2
zxA^?XUlvWFI5>8i`v5gy7^1OBeS2-`5@VOHKrRh3e$iopqq5_)G*Wo{2JQ5YMQdv&
zZ^jiz7eJ<4-p{7Fl=&bg_k0_Rt{?faAd;|`l2!TcN^DZ`qx4la!;KlAsVw2dq6w<f
zy5HD=x;sC=yIV<*bHFQoCz0^uwV_V(A*Ev?CQ%_;m}~aVJ*$Kl^*9(}MT5r<zOqOW
zN9|&o-l3K+tRWOXFOs-R%UtlHOVkygf$<$?86Bz9y(zuMC|*x|6W@4aKkN|v4Icw`
zFjS8?&~4g&xiBzYm($U}J7)mwy(`D6m+(F?6|n}+unSsyaOSbK8xm3djEIj06O}Z=
zyzHPFQ$iBrRG~x+l-Yd>F(z<wZGJJA{N~608*BN;p9_nwu1NQmezxmZ$wzd!Ui?Hz
z7W83$Ts|Y*7Y>>zA($71FNruDV}cpPFG+nY)LC(Us*lod6pqn_bpToSb?vzRmedg(
z&b@Bqpq>KNVe2kb!C6~kdkK8CBn;~bryeJ15yM>-AG`g}Er+ZKn6tG&k?{%_oY}$i
z_vg8oD$Z|M?*edMNopt%om1YXjKldOS|sTZZM2>0ru~Jok#<ERqfpnK)f}YVCs%kV
z^T$H<ga#4d*Ct;7(Xmsp#tAjZACMnT$k-IUBda__MNxKkz)#Cqu)`k0A@DI0rWOwE
z(MT{sH^0l1xLh@Q5n>5TE=~(5AY@g`Uekh=xY26CF9eR~=t!oU9aI+bft{+q9u~}e
zNeZ1a_nV)UJ<eBK>}#lBQ#y<j4A3ex!!4#FRr~5DD|H<zHW%z%zgO<+!#^kfJmir~
zwGNQhjZGtL%NpNzdMc^Sx{qU%-4I4nRD@7x*E_S$P#ikJmctFP^E(k!^^Uy<Y$z=U
z9TTaMeC*|qnTvrweUCt@uP2;QhM6npe+Y>Kx8K$C;)9DIS}3dE4~3Tg`GS3ynv#Qr
zHLKx>fl`U=&<3$~5=U||+GiIxto`v$BA9=PJ(^wf*QX8S&T(=u{^tT`o7KpZB;~$Q
z!-$kAw$@yxH=lyKvBdMG>%g*cQ#e^v-w3>T=N>w;7a93mDn{9;0gFJ}s1D1>St`XS
zMtGV?Zo`Wtg}AMyQ)yk<D`nY2q=3zrvS(jT{JmbO2A|b<x-Q}jZ!TNkkw@BiKNZtV
zA+VDwtxDyEi1LxkP7n86w=gxf_QbX^xq>b}BC1nrR?!UUzwth&8ByK!%Tq?ompojK
zP&K?}iZ#Z01ls8#XZwnSvUuSY``G#2afUcEHn7E~I~x2I83|Wb)lh#V_S6os*NfX}
zGG2py*F79?WbZBZojQA*40D~@j{cagKRAL;D;&su|MYtidVcb$;<w906<=__MrVuT
zLuVEi(xNE>#aCR-461}{X{or_;;EE0Pr0Jqy6~QT=WEig;hTerSlXH&Fza*ZFCC-=
zChyB{8npPMZI^~!m8D#l{10{&{D&>WtB<UO%fh;ZD0%iphr?TQ^QMp<r*u4{C!0P2
z*umgU6ht#vsV7_@bjaG>(XCMGIX1}@P9c|~zoRU-@!C}+W-gB4Hm!y+Eg$~$<rh|a
zU*+5l><Lq2=iWY{@}jz|4~xgslACOo+dTniVV;wBN*WA)Mm$ABm0osi2KWfSuPzw)
zVF-PBfUA2-2T;-IwfS_Hbyv(5en5J%^PmBojKpn`MSIZB6L`_Vl|$u8k8)(}5#hl+
zLRjnC?T)Sf`|O)dS}}#u^0X17TaL3a_S|#?oc9tuwe3N&oLKM7k#?UazvL#o`$<m*
z5`5)Hvmz3_jm3^LzQZVIe}=I@ZJIU(gcdKEQf+?EjclBPV|8b@vZHJh^1BqOK_@Vm
zDgW)W%|@?z<F3u{q3>N`IoQ@cH9itZOZ7RqT%X@mPJ8-QK<&!ykz5Oa4e5`KG!|gG
z)1T#X3wFi)YLX>e{pb}G{I~mrOPijM3gj$xi&KOb9N{AeI4y&p`N{9xj!ehQW*cg`
zDzjabN@uN)OZckPVJ6hwCVLh$-U+^XCs>jE+(Z%W5g({21o{(hz~>`ACOuus7V_wQ
z+f-~AaXt5aG{{M|_Wdr%bg@vwPzE-g^^Es;&u3@`>ylJYN}@Y}^Ir7TP_)J+ix#E5
z_H!va+-5@;`V2D-9MCJ+=rmER>sT9Mm(0-AN|Pb&`0R=3^`lpt^M1TeZ-=Dn$BvJy
zP;5t&4z+8+FKkBKDy{7|DjTBOwOa0jIIlph4$H!Dj79)K0?{TeQWz0FG}Q-RP57XG
zkdfG7fD_I#jc~OAw4-mjnTWONWAvG7bXp>lX@Bce)!Fagd$;@EO+`IL2XBmczK*^S
zLKyM{UgI!6{+<^9zi9dnf4I8u>oFJ^-RQlIE_(0X5Iv$LK^QfnCL(%`US<$PZxJF0
zDJ071oe(|3=mgOcoq4b4`~Loc`?=?wUDn!roqeY4x*ikNZw`a|%7vZ_k-WK%JPAXV
z)z;qZi2i_|G`joGIC1sX9#SZ0Z?D%*9E5KCE-|#&LVLitLlnr&FM95!;;TpPG+FA}
z$Z{b2S8n4OcYkDCV(aRtTDDPFMcpsS07@uWSSyes@>KN?;eE=sCIS7YeO;h%b*>zB
z-DL)BZFnItn}N6Yg>3@*Rt+%=JCW@x_FcGr%}Kl0;Xg_Fr$NcT0yGRmK>j%<+^iU>
zm$nq)`DWJk3X~f}pk9XQzm6)Q?%ebQWLy03@d8ab;4Nd!6~v!;t2uz-RCo2FNe{&S
z;hHcNeLUli6-CvL)7rc61!(9mn_kp*p%X{VvU3fMKX)hX@w)tFFp~LbHcwiYH%UZS
z_la=U;JOb((!<WdPBr^?=Tu=aury=cISr&nF*{4quv{Z)zUg9Pnqi=`+lC&qVn}wK
z3$lK4>D#h!#5#Q8m$N6b%}`|R9VQ3UydQonX<l%NSIFuYr$<dL(ex(ojc=4R#t6=T
zkXL%b{z-aV%z2LRl1=wL<`MfccMiURcTW%adDQNmOf^&VD+3acu$Bl`0q%ZQ+ZTxi
zmmTzeiJ;&l@4^pvHWYA6de4c)5)}~s;BI|dbE0CR^P_U~=<1-L*r5v%L{Dz((l#hq
z)-<w58TI=R;ytb`*Ad-way0nTpBN%}{1D@gjfKb%vhwTH6NSdBzVqd+^GKV%H=bD3
zX_qVGB;$_DWN`4)ceAza&ly)z7zBPFc+A(uhF_k{q_c90;^@aOOkp^Y!CNX@sag5_
z78hJ(@6-u;GT+dc9HY)_iKAz~_?S8R4KS%U#m9E0NV~m<L=&7rKEeznlXN3ryz^s%
z!i~8Vu51^O$od$sJbQzM=`-gu9T9$&zExrlA;8Psjbu4PBUGcd1`8>54QdHE0_G$e
zNzQVI2_8sCS4Q6v6wo(I*TW0<!8}%z-Y>*BV(7!$wDaO0Q@Xe^)`3tWXV4{bGi^Qk
z8d4T^{00Be0z(dxb=}}9&P3bV>qXbgoa0zj%u|qK`wn4)VI;{$PtXCg#e#!vJIvPI
zy}0aMdUH`j3H1!YB?i&cp~e#-1a&)&*rZ$VRR5GF8EYD`(hyQmjsydHmHnuA_5S5>
zMa{<SV?ICcasPFsw8_E)gWUUME?ygBx`&&WviP;ynuitazr!U9R~C=WO~U5w4MiR{
z(K)g$J(Qg`mFClFvjqIv8n98M65Sill*gh%5ds#0c%gKsq~3}p<YKY=Arqe152)%#
z?sAiVd0G{paMLIK>aG@!l78`YJ#&2CLVpksOC`q46>_!dIK|nYsN8oEN@B{9op5Q0
z_sZ?AJx}K59SiP<XH`cbOvTkJC!9I`6%ou(d74G=EAF}Flj0nZ{@f#7s+ucYz1;Q_
zI`90G2qq)eU{Q*xz?X-p?e3NDF2nougwCQf4>4R_@^*(z^jR4{p8C`LyBKr}Sfiy=
zeb`xPR21~q7Yb3htrRKjJYUp}#9ENF#TDc0=&=oedkvvEF=*1aSC$TuF>d**EgPV_
zuaG~|e`#5(D`%EZ=6Md4zx$G0GG&nv7J@%|EivUMZhw16xTxij`|{*#*wWyajq%K%
z+Md&HRk~L~Cm{=mVJftSvz$PVfa~6PrQBrLG`GTa4XFj6H=kxLsS@es!s*FXXTIK~
zTfnGbO$~p^dzQ_<64`eNUl+&^?G%TiacRhQWM?zsXZ%KH6!il5LBc6*JqpwJ#6FBS
zltx?6Qvw)NK)BDWgyzP0<?pHTD&;n^o-&RHn46B1eVI+U!&>}OmYyR=Lge^l{uy1{
z%TQV~ajY+7Y&S7FA2Bsds%2bTMs(Nj<-pvRoX_4;d>hQ0Nf@^ekeL;E5zR<860W=o
zv<-ozB@?v-2@j*mMXeUA(CpK_E6FcL?nv{9ZtrMlruE-B{olVoOFlo5oZlaEe}(h^
zPM^r8i1%knR+rcJG+YRuU>`0k^2>Oa$0sil@Vt^6Q&IrhIhbw)hX<*zGU?=58*PAh
z+Q0=~owa=}=$gLnFM%WR`xXAt;%}Z{U4Pr*s={aJ!swAO`UWB>lRt-9S54R>u$hw-
z;g);_zn(EKAxLmvQCM9BYkQdp7VU+41()y(q}=A3N4%%fW4n+4U^~Pf7VCd|C5Ew}
z*h6DFWpIuPNjpL)ga0JrfXQCPn`B13GsKizJ=romaCDyl>nxNUllj+~24-RtCp`3;
zT9K6}0x*t}gv`U!ps)S?Aqly>_<3)t%3{#fbo%}(oynqJPVFcWU-oBA1Ko+5m|5Y#
zGp@z^Kkrz$$SBW(hBN4{N$jpm$nJFuB$SjX0^XU_{o$mLZ=#V7OB1pe9~j<(Nkq_I
zh2bsA?59B&ODUc~q!B#BWEds;rxdd-?`gig8lRU*eWJU!qL_P)!li>@3hNv&77xv;
z=-+un&K#jJ+wG8dqqIC*ePtlbHbxQOkvgrd{ohhWl%&dz$VLKs^}$Qk+&0Dl>K#{X
z)M@_8FJr$_?YwIdr|2X$1UGh=J9|_d{?1-<Nnv*q@u{PteH~ILM=-^07&MRVaz$Rt
zYRlZ?k*;e3m=jCc{HLiiGYRTXIQ@w_Sazsv+7+p+`Se}W3C+(h5($q%cgfkEF5h_i
z21dNMC-3s}G<tE(7XHrqg)`ev|Bde<HN;rLR}bq725VC`@&F_Si<=AM;r>uRjUyKO
zr8=ZC>cc&L>r@t_%82TJ=w{y#!L1|cz?I)Fj7{(RXOReZ!uzf>7NPB-aw!|UolNqg
z6q@}+`!#>RcyAhq8XwoykEm0`I+zgi-agjGx|koznGOcwUdlzaCd<8igHOSI9`*%1
zO)%0%+sG6jDyaY3{)qCIg6K9S-xkp=hDAb;Us74S4N97j&7Q*DwgrTYVOh1tYrI;#
z)>tH4z*_WSxD#f&eVyXy({s)orHlL(kJWQKS00DjGCrq1m$`EMC!qZP8-t0=939D*
z+uoQIW~bN}%#toFc{Ayd!p*$yiESb<Jj(m~Yx-<mNay2YyiE?(J=L?BiciF9rWvH_
zK+N#5^3e&Sw1Nd}KEZNg@!;QZL6B=#d)-x-YRE_tLjZmK9@`y$=sS{`1&w|`yOq7d
z{9&Un^&HHNVu5X{ud>rI)O|L4Ph9F~+(?yvo|$Eo@c_F(fuD*O79&S+!eRY@gvzjX
zBcdYQi|gSSE?N$_R~pwZsGpQgK@*rV%7w;bhbtYnKBC@mS!y$01>Y`M5wY*=dxH#Y
zuf{*(6B`9BYS&Yx5E(rH2a|cUr`cJ9X=^$6#`3=i@jcB|P#=FKfjrnAkHGo8L(%~j
zPvzK6w~$nt8<ZV?c2RkIv8yCRHydj)!s^>vi~l1n+$}O=A73n%^}S`Cf5ktkH$j<$
z>n-i&HiSj|`n*NQcK!T$L%5)b?r>S38rsvBsIC@#?!iCo8M%~UnuMIOr*2&?b+uX5
zg~l#>uQ7fipF+DMkj|3hQkVS2Y<}tCzSsdRLw(bi8bG)ZhQ&8iomyhI5!@M$nAiw*
zh14rD*jX5*+bfvyu7!5I^afGNbNmy^n)tTXf&}T=zrw^xU%02ygX?|I@Gd%~z^ZuH
zx%<<$r1R24hhJq;@_SEPzl|)JM!Q_|Z?*Gx;Dh3Qdj#Tv&<`V9=c<1SQtBjgTtF_R
z{2CI}nEGMZ2l+-AATNHGfEyUnwf=AM@V%_?`UB9%^yc}n4eG+rp}pZitW+ogR?Sci
z=bukOmKq-`O@J?FqR})C@`*W^7YKlYB8@Ik&K*xMwe5N8$j{Bw)|Gyt#P-54VGped
zh}y!&1AD=l`1sd3C8RpQs@3T5MnQJqR_(WE#<Kg?a5MS#vry63;#Fm|pJ8q<f1_g=
zs_fTsf!KaurWy=;{2s58M4QT71C0Siwo61?PZ;Yuco3xm(G21dA0a~762;~(jg6cP
zR)<BDl@W#91G$1Y8i4SwmDEsa`<v|q-Ca7r(2zd+B)V$ANN&+Ga_d&BPM*9ap;$pD
z>&Az*Pqy_VdM7OdPDcx?8t=g-G(6Nml0gzra~S>Ai$1@S2G@c|d#<N{DrxDEz>idk
znhMA{rk6VH$?+IWId2`$S?yGRF$4M0xxJbP=lMX6sO$!pwTAUgA>2Q}ps2BwjKi*a
zxV|YYMmATK2)B$`yNA_F6ySW2SW6goj8mbu(dxa`Ky|9J-29x8VqR`&^PFzuA{5!a
z1ObtHySPiSnY~n(zuuYhmjV@$Ce4L8aX3@^{%t^!=udk)DLd$@udW+$sfQaL6xuQb
zH4`naka&GD3!YQ{!=ZW-8+>?oik<%+KytMvu7w0rV@N2Q<~1}Pr&}N?E;arN9bc@v
zBPlH6;f^na_6zYEZhK$9I}IUO(W~rIYM`*X!{1K1sw@mv6cqO}%ZTXL`%54ZEsQ1C
zqW0idr+;Emsm1vC79prUoG`kc_Mr@(vAok0%jGuJkDTQ%Y#lPL*n1S<rAuZmMlpTT
zzVtH}?W9iX4P;}U5P`}2(Q;S!QG^4mbfIpS!^xQwPNdgCPaI*$Rl%J2IUG%Izs~)2
z)(f7!Z8=RRQBJ)gNrb>f!}v7s5{m}Y`5K;qlbyf#d*i?8e_7Ulo$;#I2AY0aHL@%O
z3q~B71Hi-*Sg%ie!s_UgktN~Pqp5f~9CL)_vREQCT4M2W{$4B!-kKjW+^&|N8PfE{
z=Z`~~1IIn9WQ4G>ZZOF#f^x-d+3Z;{f2Srs=;|)<guH6%N7GFl`1%RKs4>cnpC`Uj
zo9mFHUbQ@qu5eHU>mxR$0rfV;o>d&^uI5rYvJn{w%}i73UV&Y_m6p9<(9aMZNXhLS
z(H2d1sC3CqbEDNtQR0-!-<>d349$2v*GW*8-bMhEO9KCG>kL|%kMbw34;>B)q-0)J
zc$UKb*~#nl$4Y7h8;KXb-~0#MlBnIo_D<QXF)2t<RCDv*QI4^VN~=Q8FbF7Wm=OXw
zQGdz&@K;PV@G5xn^B`16sN#M#%3iIyiScac>nv^;v8Igqjd%RJ7E1vd`e5=1X_yJl
z7O8t2o`cvY)H3rBpP|Qet$5Y)YwGZXLwK{i<9AKU8~#rVkb$t0=-Ai>t1$|A9xb{T
zL3}o!D>}x8FDCv`0cEY==4H7J@3nZu`F*~GAV)lpN(8l*op{B(;%Yy1A04T5;1>;6
zHy5DKCkRqf`H*z_f~CiHTap(z&OaV<?b&Sh)sho2a!E22bV(_at@d4+)r`M}-Y%Y`
z?al>G`F1z&4z=)J-@OZ*vkkCcm0g)?REIhFOG!_3(f^L|avfC(?-MqTH+#0VKUAZr
zqaONzSJ2FkQyCUZvJKPgc-*%*_ryIf3QjNb`-Cw8%^?N)u6jF;T3{2A5FM*?dWoDf
z5fq3W=39l=K5PB<@)*8fq*dFVP>MriNLVoJpiR7mxPJ4C4KXh3zcOY6u4#gx@OSOx
zYqp>9a}@iepM~!SVTqa9c1+BOb5oH8jg52j2@>YOpNjmr^~4?TG^lmzC1R}U^D_J}
zETNM($Z7}5V%!>O*ydV3@sosOFE;qxagIpWl!g$9MP-ymldVGEB*~y)7LXBVd}x@&
z!5J`#x7Cv4acQ7NvX33ZX8g!y<e?h9RhZ}(c<)oND0hq2175JDyu?rfTIwx<3C;?>
zBFA(|J~RAX8q|wtM>eEw%H^a|&+%5Bj@VPvoJ48$>4)c%tX}PQhjWog+JISiEk)H7
z1`>b#eey0;n1FBoj&2nbodjY<8-LfLM$-~jObz8X;K^{h8<!z=%{$<w7SCR~FZ+3i
z!)XAK2?wylvTw`@mNK1u2YC#?lyK~gh_XljR&gv^(07V_CzOB1k~m1>8um7O-Prj@
zI$2%HJLfNE%>*`|=z*-0lLA1oOsj;QgF)1&F&~#7$$km6-@2Pg78vIethyGwNVx8L
zy7kC|utxKehGPUrf>H^u><BD_9>P}xM+rwXCkXv{yloSf0slR4EIpE@p~KpN_y-Vj
z>^|Pp3cY-3+G`eK3qzaor_!ftDh=(hsQ_slSv_In!r7Tz{r5^8I>y0v&x1Y&vcF7z
zO1+a9eXPys3=4^*LBMetn&~es$QAMZ0S)2Abp_nwV^PJ6UX}K%3e=&$<#h!us-Ia5
zl&^M-^%9mZjk+w9H9h`A{QEKJwSd6aLgSX+!+dF-1%QYoz*qsJ5O3H2eAOM-FgsbG
z8gBlr5Ioi8#_JTf-jaJoQH<#fpT&tlzP|6fvfX90^?1|*y|>3F{|0|25wTpJ3zvX4
z5(8yMJuNs)ZItU_4+%t3JicF3;^iJ#C6?i*D0ol3Bw4`Rw~QbEkq~OVr4*WZ%B~9r
zZ6wwIP}_WkNvEKyiFnb4LCp+36Eh&%f8*y3Liy)lS+cmNfj)1?;;l;37lpCljb9$p
zYOkv@mcAH!b|*^s0{@-U1-@h&>MdAZcfVBcp+@9feU=-2@|GYMW=!N)k1`#7p1yL7
zw=C{?|GVa~1OTwlf!a_BoZgbVwOIi1DO8aY`nq>zQ|>wS_Jds<FZ|41Hhbxm)Rudq
zkP`V`I&AevqEIc@Z?H%chSpw}&s8M>rUNIhxf7wvHS)y{jEm6PS)QeoG9wQ60-fN^
z;^$41Tr%%ZtKHwJDCZaNd}cqP(3_00F0(HF()G^<{&TFHK7cqN4Eqn3d9_!Mw`KaB
zoJA81>PFD5o7LscheARg6%BS6mBB}yfS%&MkH$!Wpw``kwYP2fJX0Afi(PJnYJYFQ
z0gR-DsF(2@zQ7muZAKO74(JElXgpceRU9?uYRJ;&t<>t}*NDUIck&Mjy_0-0!`7#A
zg~NhpH?Pc0GrWO~(?{F$bY3bUDU@az7x*X+zC!y!-BC?_(iVSMz8$J+4qQJM|L$IH
z)AKWNpP1zS@<cBOadv}!p=O}y*GbBBO8I_WYMI;G2b6k*YCI)QCwH*6Wp294`?odn
zd(%<D@SxM$w9R?lwC*8;E_ns~rloljC1@}kiUj(*P)J{J=~Xf9jRz3c$X<lO%rrKg
z$d*_4ozFjly}CW#Xm;dj)#EtL*EgM6GGGgsM$yj%DI1(ddDHkSciy+<RHK=a+P!|h
z&ihT}v<lD*39!(i_e!(o^P``<8dp!%BDZ$oc1MeT43z<jlnl0@=2@-THzK~@K*Bvr
zm{>aS;EIwqkfHED;S$hF4-$8tIdk_ve&1*NnXz&zp8E5xrj%yJKNYA}L@ake`?J&h
zD@Q0o{5<rV6g1E6(Ic2NSty}it9L(9jo-CJ0@|Ak3i=Cr96`Zh+MA0Ko~`9f_B`gw
z`FQWDMJ@6a?6BjdPA8_SEkS?Gkz3k=*cyvMl#a#3Ihubu&7kSF;C|UK#i~|v$o$JW
zCS_>&$*B|h2uPHZ1S#sp5awoE63~6N5Z^~NNon)ej?ll35M$frvtkx41-4~m80DVu
zQ`2MZfutB8zjy_M0_?)z(IadCE>V~@nE*2Xdp58r#yck;>F9zz)krMPs@DBoDVMd|
zmfuAE`2$v;@3fddVOjgeCGaecs~f@(qG6K@uVMo_5Dps6-uRTB4q7DjSqn`4iHry8
z9u%Z@@|e3d8j;V1!h@b5TZj1_ZBQIZCpeQk0rll2@Mx4RoHfp%?|mQHCY#pDb}0P(
z1l(Ure@cZFhksS2`3wTy=cH+16wvp`iS=0hwOxxRt|J;a(!>@v!G<+PtgkSPW^2;{
zkCu{`rQ*f;hfx%B=u&x)sTuoAQAH&t9wlX5)R3n9G;u3OGXZJa7>=adN(I97?GN=i
z%A#VIyy%$qPg(yYJkbxqD~6Zt7ryoi`v#Kbv5NiM1e=_S7r2AjGmK)js(_j)O?Tp4
zn6@6sXnEm%-={U_9$7BK=U;O7ubx!rBm!-5iH-`V*WAAoKgxJ34WA21v<w|zDlt}i
z8$hPO3&e3SEN>mYUYdaZgE)%}VP-dwFCol{2*ucRYmCGLtioEpNGsr&=8E&4iCu(N
z@mF>QUyhi_UbF$ayRl~B{X9dR`dg8+IK?DkF*Io9iT!*G!wi##VTvewWCYm&X!rxy
z?!H1XHu6rDG{=B_wJb^ZniZEbS{*R=@#J{5n2p8@>iMu+l_I{<p0{D3TXX2LPK1oD
zN$Vb$SU%lR#usAjJ&|1hgx5KwFa*$~0kUC9SFTXnYHLpPx%G~Aq_1(;b3qw)J>Gp@
zojmYylw17Yd&xW7QR6X|fBnwKFbCuCO0XP$b<bkl3vzT`B|p>&PLB{VL;I<k{b^u3
zCG|aO!t3v1jmcPV245<UzgIG?){>y{jSXm3uZP&O>}YB0sZ_tS+j+J8^-wi8KS}~C
z3O24HD)42?WeyM?8xD&Kcn3=1U?%qx4YxmbPQ&jD8RTKlezf|m08^_vBegs$cu*2&
z!f^|}{m)!=Ll6bvGd-8cL)CAgL%qG6pSv@0Zt9WC?cqN4upJ-Cp92>!8e`Ds4KxLW
zuH}tb-C%cK?<g|qIsl^i0BnpOiqUuIFhhFn{n<<9x|ncD+QjvyGa@KEe$orrTmfZ}
zl#QPz;Fnp?PMJ}b=}Jb&lc&>hnQuSVp$W8LAS?1q3g3t}(qNhZ7V-0L%r~lE@5tf5
zKSA8g!6l=a{MY2!;TfwB)Np&hIKH~4a+Bn>B@XV;@O&N8B~8v{{T`k{#OBYnBD62t
z_?z$=y9u$4#+`eYob_v2(PG3e?}04kX0aiv_ZPgpBYX>vX4J8=ah``vXwlQ+zs8K$
z=p%A?UAb25+Rgbh2+2rV{aYKr1-t`~fV>nx^V7;jt1`SM%%fP(=&n}m-~4J@ABYEJ
zMR~&vTJWB^0Ef0I%PV02Okt6ZiC*ex$~Ov~4-uaR@UY1Q29HIS@9_Ji7;|52yY^A;
z@&*sr!voM;-(Y@jzSR$h9+e$5X5DEwGU)Q4>E?0SzuOT2ndML>y<~8sQ?P14d1{e(
zI1;pp+IVQaj=lsp>3t=~Tp_yZL2l;_lrO?sFtMnU3jRtRlj{w5-7g;tVa~Fgz4~#4
zH%IRNIPPChAyuJp<2{fg%C{YgpvB0PvH0);9ALw%TQovyP@^@1oO=XkzoNx+#;6?t
zDk?7Y%UrrKh39WpC{n((jc5k})58})YRX($Suln=iz<-kA*mg5$qB!0kLzSW(tOI&
z?`u^`)p+6lQl;JMg)-(0emr-NhDRPqr2O%Y<n^_zhJ8z|^iZ%8BQE6YusYaT{j4y=
zh<cuCsoHvK*;1&nufvo3B1}}2bGml>FMIN3%knRZV$lNi$coI70|vO*`H_|(!W2jZ
z^GRC&Vt+iEN3$r*QE>4j>=FIZc{<W!#;J@b`@;zzEEZXyHqSsgKBlY)_@o%Fwf04x
z>`4|;!M;qf*|)7|Iyw&WiDB{F-*IAeJ0yLd=2L`~+|$m!a0Eu+6M}$^c=V-AjA$!y
z-?`ekx+~8#E-yV10xAI_8P6~1KyS-4@!CwzTv<z41-(F%aSkZF(;2Iv@i^=YBQO{2
zJ!D{1w79%6{5Y~VV7fW=@JEP~)GyaZ#^5qSfX0a2zyfexTI_?cmnO$;Lh6m(aGeBi
zE(uXt(D#Vy#1c;2N<E>TP4(ZTtC*P59`4b{*pHcIDtpo`6Z*IgEk3b7TI-)D<Yh3Q
z>@B09$MPl_jg;{O-#Invxvor&HX)fFQXo~ekI*AZ7t`V+Wg>x#3pV#lfOfMj9jdyd
zNG#EK6J2k&(gt)na?lglK2D4nu`=?H4hX_y(IS2KGp^ddNJ4Q#*^7*nVoFg>&$DM4
z7c3jGg_uB&QvzgjZdod_kd<K}vJaQUmGN*2ln4I7(7TnKjrSVwKK6^+3A88g56`<3
zImZF^)t2<Mof+m8WlL>SH2Dn<V4&lMRXCzXtF>YCfmHctS|*8zuqN8{e_F=ZL?CWj
ztuRdOf!(e7G2PNd`|c4rjq7j%2VpLu^<70dyhiFB`;)uc@lX^2=<|3i#4`Lo^m#Rp
zP(DINQ$kUm)=4h9ABX(W;K-uu#ob)FSnBCqxN@0I9iY@->*yG9Z$zD>9>f?Uj8>}x
zLAXw@-`sDzj6&Nak+3|*$8G}EFFhBa=fs)itaV@ZEw0e)P!;d|F{Bx9=7+ZDb7fkx
z7Nm=ESN@R=e?NrI3e^Wxko;~Uo==ZH<u!4ai{&;cyAJGB=5-fgwfn&!P6`y%o|n8X
zD(+1XpdzLeW1`a%w4hh^dTg+Mb(0XoyD`YajN)jO68~W;kzcG2`n7f@&+N8-cW$%^
z7!PM=RKc(RK%NhGca0oW5r%PTKshVhPBFRFuSk%z9hE|TLg9+v@@lOSv}%OQb@F6G
z026-GMmP@LP_h&eAb}Qn#BC*IT8S%o2Xk>YAVhC)yoMO4wG$zERvAqMI_`dAG1mPA
zvcn5S;@xS#WZdD#WNlZv3Ko~2GOP%OjXtDS>B1+9$Q8bXLkG5Kwd2{{akM-igmof8
zD=7C&Po&oZS#w&pz$^<~fP(=P_vSJ+>r<bC$4eKFoS2ZFy@~)TQ+4kg4E74N621k#
z5R(X#RritdR-AcnZ|-0kFQiEZx)kxg6yJEvaUFR2!<5D;dPtpx_Rpt>ss*LLfDvnS
zgd>Xj0b0ri7jn`v7d??7#6IWK+?gPrK8HuL=nsEg3r90}fiCkq<xWeR{;JKC-R4XN
zN>UJ9gI@C24WU*xl7C@x?V?z}CBi+yTtV_dT!0be?XdvC(v9wS)5$OgFpl*WSm@ph
zQDJoiR_ym@{Yz}3Yi_vI?49ufwS<x`(2dn{RQLUQkV3HdMy@k9?P?2BOCFk+6J4E1
zoWJ5<#>F|4Co3qo{Yp2&7-o~mh&3X?xvf5T#5U3nhwRz>QK(2Y6^(bIhV=#Tak38w
zYdIVg1Et}9gO`GMQVU0ugu%RGbeC0-YtFg!do)fa>sS=7zvRoGmCJyq<iBk5o5-=8
zEB)g>`lL}dFDlJ?NU|m4j=*KapjBIQurOLMJajfm)v=PL5&rwGxkJ)$M`1Y$Y|~8V
zB@nHsvy<$ucZE663VlZdl)a_qoc{Q%lVOuKW<Yy&hXn<oz&U<*ukyx!V$@Zkz2Erc
zHXgcM-HQa8MzS|&?eic7d6lY(4E#oUR%oOVv;eqko46PLzD)L#)Gc;89hfK^lr)P=
z9aO98M|-{6J%%{ch!k%@P<G}jnStj)GS(Mw_}vCf@Ik#wb`%^uy1kIoafm_!?74>k
zEc5xAM|OKn-u4Tb8G?VtJ#A-1P?mrw0vCkx#@1k8voYr^2!bM%z|>zCAdyRlHueFP
z)HJ4Mv9L?>W)ZCgJx0bGB+kK)`Z5SbpYY()tIlzlrk=nsa?(`ozpj?tufOHJ`P!Pi
z+^icgy>oYBl`D5wXhj|fMzDncfl}44-6O?67k4P@MXWcRTlnH^fLi*u3l-@=hM7A7
zSVzGE!BFnZMK8!mQxEX=p#6|K*6orkq5>?0p*y}}(T-O^N#R`+2xKx=^D}k|!|#p>
z+-?|7oH6qe*k)GV@Q~noK@I1}HAqXZhBaq<V)c*<HN5tWVi38}t<>-AqFk9&A~A1E
zxhMHM**|Wn(Ut4p(9SYeqx43xtFXFRgo*NdFv#1Jn-!Sr=O~_!fINIo-}K8cLz>mW
zY;E+kD7V<ZJjT=~zq*9KKO9(7Ji8?a?Q<tJarD110#v=8eW5~bA_1s56Get8kTbOB
zsZ$mE+?7Dl<wScua4mtZSo$M*noGL@rUyM&d~Bp0FCRq+BF?YPQ|WnmSkAA1nVrmc
z5hhjfRb>6sXO^!IbwfoR<U*{htq2B3&xv7WAZOVU^#HVqzt(NvTIRRJMYLn<NA)wu
zi#rB-Dg!7p>nxy}Ee5-ZJsonFM!O<NNkzy9JWu-%_#bGA^e75+{-le%z!(CE1TXxD
zxvZ2)L0iJR$6xA;2%6X*Fz>|+CcthAy!w&e$7@;gvF$s!hdqVrY+k(Jdx&L%xChHr
zx(TxDayHi#c<nt+Le=TjOXkP56iTgKyxlGRQg~q<we{FztFL6$TSsmQFl`74NDBM}
zF`+6alOd{Yx*R!-|7HaW8gg)1;_=wmFY?%}c=TQA@+6X42l>=2vjRof5xUVx>9aaS
z#Owjvc4EEOz9&+r1Ub6v&?$FWxWX%@ekSC!c|(t5iY`alxYYra9Y*bqsg$|%f`zZ+
zC}QcwF3l>(aow=P4x`|9s>J8zjbMA7+Iff3h9$CTT5Ub{oAoU>vdl&~bAO&1z9yQ<
zeyb-x0TaPd*Y1R2=jYZqFOx$^(n3Ks_*XoaP;x>H`gF+KWQudVvqmFIeX5uqrQ~Nm
zo}H)8hI{EZ@~+@RTSqNz4Zr?!P6|?O1CPdc%4pepVXgYeS*><^qv_G1s(7{?i}*i#
zS+18K3CCX65*l%kUoHMS6W;t-;uZML<rKWV31w>2N<#-5kh8qI#`T>-0ACv_qBvvc
zPy&5NF!kXG0$AnP`_=F{3RB*ZYzEZiUut7sBqzC7R#bTik@Ns9C~4V)2>~s?lP}QM
z4JG^^)|i1^%4l>fN1r)t)Iplok2y#FI4>7{QoVH-MGVSn+{6)ydt*BMFk}TwPz=~U
zF+xV@3S4uf6j~W$Y1nsAz+pf?)0@_|5)&<0@djVnel%Rq$1VvyL`nxbjaBJsrB96K
zPmtgWK2%MY%T`W&SV7|YymaLvNow<O$@L7vTCrvS)K(*p=>pZQhBv<YJ(87VJwSCd
z2;c4Ns)%QOS~%vP#cSeS9WdS>%hH2wQo(pf2ka@L&_zHKVDIQ-s}H#g*J3Rgk&TW1
z(`TN!X~_X|_C?eBn%LB!m%WAM9MkrTYBCd<|IDRW93PyqH#(mCT1?Y^SDEf0C_S;(
z*3-Q?YKO+Ov>mKm8Piy1;w<vgde5h4oy+FJwrNs9BjjXxSizon8Q;g+bLsZR-e18I
z;8u##tEj&xL0mNrJJb#IJ4{F;dM;Y_71ZjdpY2l-(4(LH`ciRa(7HS1#S+Znj2NJM
zX1|Y#Ea5lpbmoLKF>%EH*%4qFY6(wXAl<yK$I%m}P%yjsW-n($x^lu^rx{^sRtuir
z4Py#QkyPc62bo#>i0`vuqY&1AH^MKf&FT4}-Ar2R{e#6?LvYQT07U}$;T2>gk~M}P
z*)RXV7Fn9>&n&?`60;kz^p3>UE+*snr?dXg5tE!}X?kmI*sJtBznb`MEx_6#H(Nkj
zhu&?Qkr&oL|DEq?OA9zF)H|i-9RdI3DOJ8^s6*U+=w(ty-1}t-AzwqTv<t52)sWa{
z@n+C#e!AG{QO1Pztke76pr)tGB|ZT;Rm9e!k_H{)j(jRM@(dAOw>f2_;c({DTd-#=
z`rpfg_wUlIv3P10IE3qu&}99XY0RMC&g(V)q5GnsH2ry~eSn`ZACrqMSOma@Bi!bM
z-xFPTpKNcu_z86)Bb`dBpbYpD!=AJMQcM4mU;885?L*aVoxil10AdU~eS2l>8F`MY
z;ST1y#%W8I;0iH2-?HnOaElH**2E-a{zB+XnGY#d=POCK!wMc(2)q2N%-JdMES`!!
z-{nt0x4yk%R{>mWNsQg+JPL_BfmPN2^x_an_9rTOMX(Kcgo(>qt6#B8Ko0HJ0$W$@
zhddKnH4)!b8`|mk<M=Iti(9^_Gx62!hY>k_%1;Srh}vfejLh^jZ&(uEB^fNn<5|x&
zn;cYFlgly{90t97PP3j(K2H=JL@+v?f*nvB>OobDTS=li<?qfcTB*1EeCwT0%|jsp
zZR6bIR9+2ZF`%BAju=eSzRP1q=*`lIxAUj9o-VI#GTowzS41oFRyqD<3736Lj2-;5
zlw>Jma-W}v_Vwt`b{2!3jB(f}T&M1C7}{a2RfSV*N1=h2YA!CywTHKANO_$>?$Ofi
zF1)4s=Cotg!g6~$u{X(FBPQC)wA#3SK9?&JyM+wxJ!v|&O_$_ie*N>J|2Ew6WpsMR
z^B6zZwCGPO(UERma0gpFXRIAPj~1?t!d9#P`WG9|THkd@R0NxgYFO#U<(=Er;^PnG
z+p|0E?ZB_N)Dxh61ylEYtaJisqe_zE5b$u7RF5<f9Tt(<z7n8Kzw)s+ycgyAitB;Y
z|5{&he&;wRDOyXD#Sd}AHA}ALxaw>BV;nh31tDasQ{R5VL7or{e^-+CUE`fGsT6}1
zG??AZzY%J^$pUX~-!X@G>gtht>bbKXs?MP?O=XhqxY*P=+^UlW_B0MPDh<H#*N7E>
zN0OYfo;l9(3#(Ug`QuNR-V4h($Rx^na82}wA@;9cHWjWCNQf>I*A(oZa04beFXgV(
zkN?U?ml^zoW^E-(%24Q0?w^yNMCm=f2i5^o)e^nTnoEmbVeVV}UQVg^=MJ5%nqbv2
zrMrm|HHzI9;_qC-`=*Q!;8P<+*J}JfK25#I&r;fG?(0joRX8}JQn8bKml0Yw8K)UU
zqXxaL?Ie}ol92U~=-;*u3-;P2-ZYP$!HDmSo#+Rt2WakArm2w@-K3?w!YF<l_%FkZ
zC)O0dO#*mWM|EE|N*GV&@e;QM9i4t~C6uL5zfX7^Sj3HS97-s9?;vqMAHd(<$G+Gh
zE!GKki$@tOd;xb3lE`0x4`*^3X#Dev>o4Ha<zcJ&KT^d-C>ndJO7%_0S^xV)n}m6@
zo||KumZYqJSE62XwT0a|&WqPoB;)`V8m21%tL{=Gs#Ed4rj;=lPqR0m*uN!#dxCI^
z{OLW*VWuWX@kT2du!7fAE?I+4BhPCgGyLb?3eI9-I$Yl@#Xc4dYGK)xdV%>A=Z-R5
zFC6N(`m0J(d{`dWIJGfv{V^Wzoy_rb<7`(56-VAKIr+ja@J>jpUdRY<8lQsfgplm7
z4Is*;84r4dw=Db>s5f}G>2q~ujzz4bLvysxOaB)}G^&vnh9smwTYx&8c=XiTo;};(
z^g&xVoT!ofYU~M3Ci4zmSI<M=mjf@UlUVQTg>0vuP7DEJVI$=2dTw<;3mTr|{~t`$
zKcLT4f5{K2AOFi$s_-BVGP}&tL2u!L#J_uV_wSLF34cJ}n$_DBpFWW%w%&&8QFi7L
ze1jd_@z8`PS3&jE--9k1r2c4@ccBnqU~cpY$WC44ttja6tu0q4K+QB4J~pay$=Hei
z0fYmqnVvg}q+A+{zmHjy%6o{h0Lrf^*OJ9RmRU5gy9)e4>MvPB)JWlDZlG~3LxO3A
zIOFQ<?$I%)f={?c;FLuD5?WqjWI}>ys>k1n!vl(sRV5~_`PhH#*G4ekA|#j8$1NsC
z+_iuajU18Mg>f6>F5Nuy$aWrA*y~U`xul8}b7j>mR5X1!*D0{|d{hBxrR8~Idcl4}
z4%)4D&YFm!pPPK6|G{<tEoofdLsg)<z9JpS(Es-S^e3i$8rp`~<T<y~2aU9i>p#=3
zuf9Kp=a*Nws1vCnnKp_++C$N0#r)h{Fldn%_w1_+lwn7AG}HW$hDd8ahN2W`IhdYz
zzv7&>0q#7#*ItwQczc~z7F5?FTI$^Wfodq&;N?!FRsG0WD2ugKCT9o;w6KLxA!Ww@
zmN#8<M4Afm{8&sF2fIJ8Z`tsa(nd~b>;`Fw0ELWzoi(_|G2Ut@TF<cC4pT@e_FKSH
zf!$5(Ie}|cX;pE=A(~GA5MXUCA5Ep$^bkNfLiwf~VcL^$_m@z=;%gwLVy3zj-#+17
zIu4=E)*6JJiH)C#@&8+yJh!`3G)kp6+1dA(D?a97p9_{0{dT#=GAklKqo`lNO$&|}
z{si+WF5O9HxU7-z|Fi)656im9%2@U#uy*lk7Mi9RkHndM%#h60l_etY(2`FQXXba5
z7VJ8UVov`BroR4|WguQb`GV-qTjZE@NhW5JJK$g=4V17g)~_R@Zkddg&Mf|+d)%h?
z$|L2eCa~;7xxjzz5KPT=n|ggsf>NsV@?u(rw2s7uEg7j@Q!Hap+V86T{MBclO~(DR
zq4x|k=q?k2`0xgAI>}|XZy4*jFvmo-483(wK>EOW`e8EPUDN<U79HBrHiXIpqBN!;
z@j-{>_fK%D9W2Y5jPY*lDXkikR*_1zTdvFN8`vXM)z~6~lS&T+vgaAU<RfZKHIdpi
zE!*nJLIiv?GaWBzLi~LVvmqXIb1m%s{8jisQ+7<Zzo<c8RaGqS5755_1SiP|I!D@)
zMG9~E87Qjx?GNPHGi;qpBg9Rt%rMBo*aT;{7UpnMOPtJ6+gzYn{_e{YTIIb{r72rS
zWT2b3{z|x7N9Igq_Q!0HrV6Vum$iTF_FH$?Qn@`TX`XkDfrq9&#^7}xhkBqq0mwfR
zU=N!ae@|kd8q?q1(A?@?bfMN=foKHaHK-;+o-dL2K(@|SxIi&Rt5kEFXRzOf(yXR*
zZ$X}VLmxAqa*l)4!lYIf#efzywFjpI>4Pg=3GwgywaDaeCYyJ!iQ>6}N>7#HUNow6
zEb!H`FlJXib3*&@HW0I(9Zeel6`qF8+j|J!-hWiB1rkpqX+iyURwFkWq)H8-@{&q*
z|5nH#ZYQ<33=QR0M^)f=O3n%NX-aI8KSfVZ>HBPp_uR`X7u&;Ak)o{x#~m+dm}acr
zbdO1?`6Pek2Foy^s-EeA&)7jC9bVHE9J7D>2YN6ETtbs)9Q~+;5C^*J0$siE4;aTd
z8WUFZ$87wer$$x;GBbmwrZ0yObXz<CEGz<;o<pZ_*Fg@MiF$|;?X&k;^{rzsnpMvV
z0S#~bE@aL@b{N@VrBzDPHQPX9?a@^Jv=jwVM0r#C(6sEbY;YKrozcC~I(<Fx{W}D6
zaHFvvhz8<mD7g%1MCvL`^gfg#bpB)Bq)<cyrd6Se2~h1Sbd?PZ3o~6n=#6@7U8?!@
z18;5tltjr&D$~yvehKPXNZ5ON-T}FQ*w|~L@4WZy#XT49us)H;%N&k4O8=hY`8#4G
zKdJ?>=0sMkyr+msG3*-3nBYT$)I|1);B#T_Y5y%t{;ZH|j#%)mS6g^8pWo)^IL$WW
z*qFOaC&K8;2KE36l^=#o{;&ZXxhN!_!C$3TI4sH?URR%PlK$ux;})|yy}X50lm)2k
zMqR7#SROnl#5_f^bIf9_&GJAb8R-hb%<`+M7jf<j*4XB0p15PJ=$n!Og|Li7Rb#|q
zhu!`)V+-=(W6jq+K-q}s#tHT17`JSD{m8C&aMZl=HRWz`f#G+!&Hfd9Wa!UZgfSdL
z?tjJ_zhrUgTtlWu{EPEd7d3k^n`n-=%5k#tjaB=WHX|LX?_n^NRYWj9j(JQM=puN|
zD8WG{P!oN>1ODH+Sq<mWejVg1i**-bC1Lavm_bSd(V0&e#k>e*K<ORT3*0aXB$*-N
z>hqi-f3VrSPKcnJ@GNWgQ_!o*%3jW(s=|U$?_K|FJovA(ZI%y!Id$&Yoq)1`^WpCa
z7a%{FzRdD4LX2$OZUr~FoGv@-ThjX;-!ie)oqX$HlSm=?p(<njn6L5Klwv?Iw1gfY
z+kt>+)U>4611|Zi931v96o$m%*HP{(<ZL6zth;3QWl!@T6%VPUNAL>KUi?r_71e;j
z({KGCIZnm!p@_pDVpxz1im1_{Z<uR^ej$krI5!%E;Y)i{oFy5aZfcs5Esnnh@dD|L
zWh=&X{v(<B{oCu94_A6CH@rOVXvS^7k4D3~C>^Eqvh!i^t|rU(WW>rF_l@UL;5)Z>
z3}Lr>pO&OTFkf#aZ~}SJ+{FSLkhQS^JzWd4f%boOl6Xgz8}P^LuG+<+ozLFaptH#t
zfMdwswXx8-J4?lDuGF#nBdfu^vfSz2Sb{Kcq2z|e{oQvi+u*#=-FucvaCKI#4*S{M
zO!$VIs)3Oll44!?mVudRIT)+G5g>UD!f23;Qe>#*FbwI{iQ)~Wo!o^a=~iiua6FzX
z$B;mCOVGYrTzFEDfbg<!YBW5K*-_lMyM42l`Xf(mhhC;z`|9~~q_~#b8ORnBg5nxf
zDERl2eu4b0U=W>1XpQ7n_|dOPjr^oza2%l^j@l0RJ%Pl`px`I=0b*A;e59O_Cf06g
zdSb0bsqs{XO%=TJzVx?&to$#Rx5CN;jrDeitu|Oz!LSN`x2T&k8rUMQQBp}0Xi$~%
z%D>*3!e;gF;>o~WMNs=X*i-Q6Yb24`*`8SfLXDSOlnNa+7gcSz0Xw`eOr+S5x*?!n
z2=p=o;=L;X>5oslGxu6RDM;T{0JydI&!*t{T9`lM?Lk&z<_O5>flAleiUe`nYP*Tg
zJJHvu?%WewMz^R(<zeS<K_(72LiNnb?zTV5Ps;LsT?FK8SmA@njd*cUr_=Q7vh$J$
zzEf)&dnkpf5vwHmuGQa}4<y&bs9lyvea&fjp55Sqv8CY4_5*O4Y~h{yKHLz&fO0R{
z={w%^{M)C~!nATmp+Mj72@r7V7{%n`6T`I32?9~0v4nISk8SYhp%%<_LYiEFC0a&|
zE-k{#shJ}#V9HSKt=LM0j68@(YGe3OZ>7!gx11N1$ZYDkt#&(+>6b{EfM?`c;jDi`
zJSmF1d+?B^8!{Pro>w_E1ASw};iosf7N<8vDJL&OKU&8PB|m={e9ORbry`Sy!R_T`
zkP;(C`(sYny3p*G{nZ)pi1Vkr-xt(F|ETlcyUL(21_tGr<kjg3CBSY2=R9!BN8)=+
z!~qyQ`3!Ch!otd|KMYctxfZVdeNH^UWw%lZo?!{_sQGu2d53F=QzIl(Aq{wbi_>~A
z7^^%-W5q}e&>tCmcC8Zk=Un9Pe}}f3q0Zu3zm~wc{d`u#d^bmG-qc)}O?{ii(LIq;
zWRY1Jy;h3D&5aV)u#S)@FcQ|>t)%LIwZA~}iI~IPN;3eYCItE~zh~yq$sH4b-nNIS
zZgD9o3a^NB<@{}Yz9VQ`(}w=Ux6zOco+AOiUs?LqgUr%s0mI`M!3>kUuWecaoVzxo
z{3itO*E_V{9e%-mr$Q6d(Zjq-+}EFIb{ambcq>sIl!r@69`n9f4P?8cW6MkP-BoR>
z&JlHekE5_$F)x}<dn*g^8Wcu-`ugZ-k4RoN=c6hvzPM%41N#{>lU1V1s?8`e1)^U;
z{>c1pX09Hx{Y+^?XN5w3%+m>xZCfV_63UD4gp-uss=NiVBJju0?(7U$+QoQmY~aU@
ztb}d^e2)3hQ~U3&Ru6yK^Qb$DTrFi?K$A~j-=RFIWk7O5bF-C{3vc80GcT=+t-tXR
z$rd%4%qTL+!e!kcfAP{1pjT{Le~3peUU|+J9B{1Nyx_X^%idnj-yDDd8UO@H$}1AK
z6%EaK4R^gJiMu;1)G(*l_$GAR5t~Naf|D;;+7{lC`tG=7c-j9VG0k~HFYbdqyEgh>
z6-xRZblJsKI}$7zz8&rYlfJ`q77qvQi&ZuW*hI7~0Ywjj)=l{-!f-UHL05Xam4@H-
znTLW^XQhaX>vM-Und2VXq;0upFGx}GGl}Nrpa&GkE}eaWGx5Tk_2GeCW2<C>!jJZc
z(<S6WkhqEm&iYkAZr8g9s3u~Q+wdOTUOLBJI4V_SbYmV{ME=Uc%X&3tzONVBe8o?#
zyZcBd6ndM-Mc|{C_ra=e(-oGOt#z!!*s!lgFad3IQzNCa*Z$*D;PUoX+hr=x3qcBc
zY}TG)RGYg*W&k;Az(;aLC?YZKcep*szAEN$^=-!{$q&3^SxhhGY~jPcjZLN17Adf0
zxH6Pi51&!L1Xy|9*qOfcI^H6o3W*X~aCU2h_I9`(s4#VLg2o9k)QQ3Y1)$e=kM*Yl
zB>Sa7Vje&;y@&ahkC!-)?+$(P^bJRDB<DX%G4H8_l|Hg57B#MaMENUcbW==YX69VF
z%%EO?$?I21q+L=(zpiXzfuxr)V^)DHgR}Z1kHZH=9(7LcVU54BV>O-L(`?X;S<GwY
zTtV3%Hb!Utc9)Y*T^*CLpm7SgSvtyxEp+idXQ;s374Hkd9D3klM7W+_CLVspdvnnT
zJB$FsYXHdL@d~w4>g|q2KCynWihTX>Z>Z?b_XssdPOix*f_cbiwxNiw;k$Yk5_;TK
zByP4J2^wjd>Uue4ZT9F96QXi&_@P})SbdiMbd8a9K)3Kup};S{VE9y;AhWoEwau@M
zO`!5<K>dNNfuXV&M3;5RFvhd4{~lL<0s*sNg@L0W&!4wfqxRL+&xme;W*p#~+b@W{
zdID{?A|uH00sT!#5^nBGoMCKG|L@;C?<%sV&><5k7czGFi+85}$<%jo12^>$KyZeW
z&_hy!z`c50ILE~OJL_tVuR^OnFojLJP%T?Vc=s1-_qXZF{zMHx(oR0n<{W7h(V(Qh
z5qaEO9JQ3l0piv`L>s|1V*rLRH!gS+os$*O-?#3`x9=n&L|IbCTYA8-0((tT>#!Ta
zH5Zs){L99MU7!b?ig&x9{R>J*t!d=IwIi34tc&~-m-+*-RWV4@L2o&b_5pIn94J3C
zouhAXQ|*PVq!}n?HN?&54r*QEEz;Q=v#e)YFa$`aj?lLaJzrqyjK9$XNn71j)7^4^
zH5`~h>qprSh+)j=0eT`~P+hIrZ1^Ve;AD53X96odHl&60ePDw)D7(eI_vnfJsjtv@
z>iq*EM?qApiC)SJMyoXzDel%-mM%NfgzBJ0lFxN}8$vnyWiR(2UB&<Wy23-hGqw@-
z`-g1AxzOFW30DmN7T%^?@_L0@fm@p~J0>oP6eOMZdHNt}I->YcmAm63*ZG^8AFZ^N
z1lhatTkdsD?Jmb>?!Ku9uYOJ9vqt=oVwUjL;htp{VRdy}0}2p!deVSs>>|nI^TRhK
z=eRM4H5UKoOHQ^KbG3_E5nsiIt$w$Py30_Nz-yH^3$9mRI{@fQegl0aB8S7i9qf^u
zyf&$A)bbO!zS<JUx<;<U45yZjj|TxECM<52O77&caoAlen?CI~Bu)P`q!0HexCs0f
z{0EYopSC5Srw7pQ{<xP5xtQRxrj$W$j1WR*x0-5H<X9;hdHX5HY$?0`&QqFlyPBRR
zTj2hGrBm2Xr{>?ksZTf<w+G_Hyto-PL@rRm;j?{<t)$j8WSRE7y7-|LcX_<~>nbYM
z0*-NbTN_1P41m`P!aP7M6F60hNn9$i5zq3<$>F<9`i4+Dz86pE97$Y7mAY2N7nfQ|
z6#MK7HeU!PNO;^}KLRTsDN&Zyk4PYPMBkx(Vi2qRimEbiGGghdS_(=5NbA0&?D!Jz
zgSk3yp@RchP#^`0x9V^auV>d^5@4nv>^nrXFSNujWbSB20BZv`JDuh(co-~4_<%4N
z?^J-ZpF&#?=0R9L11XAr)$!Qs%-^CfD;N{)k|l)tz$CYGIiDg0VT#onKZ@Y#<S!?I
zfOMz|9@Pz<OU`B%n=;ES08<-~TmaO~=$OS2-f2?C^UF1CWVx@O^_7nazvabK6@PZz
z`J7zJ)U&ij&pO_$nq|u-kKeV8Cl4(oegHb18@;wZa;RMtKR3}XwG~h-J}O~IY~%El
z-FFAhlVpLT+2N8E!+AN0cUG2UUi%@ACXJb3hER$tbm2uD`-;Y;VOl0BFMynYaX}m~
z#*ncXhVlE{`R3p{3`N=wFoNv@xag^U;3KfC1Sv{T@Z1>VF6g~?&F7xYhNXyp%1FVL
zSSW$zLO%M4MAX71V6ouNihZKT#Nkfd+1?NbvL(E5{E)o2TK|P!al-y+UwKC@^sp1D
z8Q1Q3L|gnf7#`&HM*UUl-bTCZ(U!+1=QqBrnr4dx-K&>DBNt>R^EsRHz;O!yw@?Df
z-}etr;(z9A#Rnf9oE*L7OTHvcxm;N}#a*xUErj}E|I(oFz))~@CPuO^Xg=Xog0uuF
z25n^AS?wA&)uRLY4H+O8h<pWYTY-C(_pPnBEka||rZg1Ojj9aQIivS){|NZ`Q;z=$
zsh_X?Z&Z)7OBz}7&j&ITZq-LBb()i;WD4jvu!Qg59ynYLzsiWH<a~g`SpkB?{|aL6
zq`bHvS=YGM>=AyyfL+_r57!#Di;J56%xau2i8m19k;VKGa8LYl5ooG3-(-?U%jVw<
z|B*UTiQG<yAF6_cb@F;dg4(5c+$BC>5YNKus+N2^PSAJX6A+o*YM=|CnPv9koR|+#
zFoW)U)>Q_iT9hE*Wc+qoPjRg;RRh=8$n77NrR!@RhKqtNB|aA2*}_)v=S1A@2A;*%
zy5D|VX({tnr7n|~fN6Ig!G+_hZB|T%<~ny`Nk8T}=326DR-k1&br+Ldu%Z`d@tY@_
zj@BCETL%3gA*X(Yi`ptO%;EZ9`Ih=l9iDfc-N^sr>8rz{dZTU!28QmCX6TlbkcOd4
z8YCQP1ZfGW0R)C_krE^YLAs;{1xZPzrQ?TmNHcSf-*@l*pXV8#bKdjrwbxpEZT34%
zc1ab6m6`bb>0_4LmAjtZXX_nV6(4ib)I9x3fA<IHD6~X=uSIl`>G~G$RPhcM{a^kU
zezvmOJgq#1^WGgxL`vkH{;8s^8Qak>Q;R|D&q}$>9a>?KlU^LE%XFqjf_=-<`D=f8
z#2Lzc^X$T-P7yp^P(;8_@bBmbL}a0!VPSp_Y!oV&iY-3^w$Xyi-Y|XQc!_-)9?`?x
zd)Dl#vE4u{T_K|wvmpaG!@s>B2*^V}SO;#Q19&)zpc*Ipb}$PQSqR-6@x}9s7tlMJ
zqTU!`ZYSvF6HjpSnP9c37yH5yJ4q!8(~$v+NsQ)yHG#;w^ti_hiqvm^D++sca+PNc
zix4w4x`(1k(NE3*HN;fo*PPtrGb>&`p|X=Zv0R0TH}|6N#(JOW8D`@AuP_JHXX&gc
zECf^OmnFxH6IPydSEc5iwk(N*IESKdU#U>Ra^E%xvNO9-^M#kGM(+yX;kcb-?G<{;
z3Mpr4XsT93Xh<rjI+tD?4BLgKIQudMxv_Ora1st#_F)O83lt<g(~)S0Rq#FWijuMR
z!%A(p-a)78n2)J-&Pc?od-h4@3`3^L?Rz^xA*|k;!8s8o;1KHlJ}g^KUa$n&bx*-L
zn!Pk4R}5N(DTM#8rW408hdMx}Z6A`hFdZ*<1V<#*7n*&q{{z0et7+d~GX7R68TIJe
zZ<EzI<cGv^H+CD?o%5Cg@G-<LU;w+OVq;E3PgM!NrE{A_y1+NIO6kJyEo#@!0kOJ_
zWBHrFyIC7t$~j%-Q3V*Oi{ghOr-xoFUi8K4MGnxkhi!_ngHQ8S8tk#U6D68LSiN{E
z3$vCYUG!#QYt3POgY2=D=vEW<uV(8*k86v%_x;gr1HNAkzvhkdi}vnAy5k*Ez9h7=
z!x%o&T|wvI^Q9vfC~qMn!{ZQm$lEML<dovC!Ii}_DxlB+?xI-WKklRG<T=7;8fUl0
zD9&*x(X&AwpY8@;7~kEt#4#zD>8qMQv<afknAVy3ioLJ{Zk55Z1&o9MsQ9LO5bi|f
zj&?f<XYtSLy?xO~;D*ovM|nCSbVoC;7)5s2fB1F@Hlk6QgwI=`=$L~y%jRs9IBmH5
zKhf_iONG-kLG5xIAzN(8b^akarhF;5`Wm~<W?1&AjC)Ta9~vXV$wt+N<UIO@lmKX)
zY`?9~a_aP<r|MMkq>l;ET#G5Bsl3-`Dbo*vq~@Fv=AB%xAb)OBvAz(-4YR7;V7FP{
zV&-L!L}Whmd-k*jpC*=0x@7v+aay%F4q-9U?;tWnAR9=MnC9#hPfH66_ybzv5pc)K
z*M~$zg1vBeudmWO?jvuxtSlQi1uSKBs7i;#LwgkUv%XY)aV~c*_bt2(`jfN!=g)hq
zX`5Gv@@oNtOEraVt$ov&@%o0*PNjqJbpW7(%&?;~usfo1Y%4Y}y8a*0;lmb>qAtf}
zQfAGfcj<0ecu04P+M06Wt3>w|f{2VCN6^*@ztH+p?C|IX2Iy46BkamtFZ{+o8$p?*
z-GQl|p90@b%whg%Hopk?dQlVtp*mw{`w!L9@0Z+KD0eqbv|npvOVC2Ih|4se@F#Ni
z=xT&K=)uk4is0p9KIZ|6Fz>$qy(#ArASwK7@PW|B=BI;9D#eL$UUo;Zytt_ALt#~y
zkGbDYe{(f;Wb|^EZklNc*FsI{KLN1TA(TJe@<nRVk0*8GxD3<X(GnZpG6aDnal_%?
z8&^UVJ<DP#9gS8`W{49%q#E9?C<60Zg@7d9j!j3hkHb)GV5M3uc{C-FRI{2IgX;7K
zU(UKa{dtH_Q?tG~EUB$ZtIu0(@5|HAsxJ1E*5P$0K&}^E^n`!B7uMD9Wj(L#14i=;
z_rK^nvCr1($Xib5OK%BckHPqJl(v(TV2n86e8=Z;Y4<WQ_3%QKu?nUWdpb_hdWtrd
zeH-|~>p&{5l>bt2CKy&Zx|n=%=WFzABmMq%)ABjraUYXua-<fabBY~6vZUTP;3s<b
zMSFo0d(^geMN}x2%IK4sX+aAunr+_WFK4S@n4%Xa+GtH2!H9Z&#IbkmOdX8~QfHb|
zcE&`Ty;9sLy1`fiRxmDeX)G<DoFo&=B$cXG>lblmju8$WT^TKtG~Az5V5H%16$L-9
zyW9#cyH%~b)JGJvNxYA?&U+9kFEkAX2v^8I<0EK{VJr_$&MtYYxL9=+Wy&wBY0IT+
zws9Wwly1&Ot6E~>dQ-5RC=j3EPaxP}I&FJ=)<wAdE88R?$aDJzT^l6am`WHTS~br*
z@&vL(5Rx><Sf7o5EWT=`<ohQUnBB5YK4Y#atpp~Stc2I1!DGG@Z~{cu-mzI<&ZNWI
z=lw~lQ7c*ws*W`zIz&JO99P_t5xs#}L@nc$_YQ?wgF=?h6nq6cBTTS#&-Pi9=r%8n
zgBfH7L`q9%%A)4Jk7}O+UDhVGSp%AfLqB}>gD<GP^~<Pwvad|yA4~;xD<ND?_A{Dy
zye2&InrF<(ae#fyfn4JSnXzwrzaZXc5F=Kj;IzG2CN@C%7oQgp4CeMbN~4>Di0b%j
zIQ~#rMVgcc2UVapuV^_;L5H1n_?64_yE5C<L<zbkH&O#4HE>oHeyl14h314a?fOU?
zU7&*RhVhhIF%}{gI|cK7{`I<5D676UpPEO@0GnZV2RVJp<GwVLH<?2Ecpf<2#G@J>
z$j1avrjU<dd2XU_Vw73ef9rj)?{W?rM^Wrtrm&BNr@H)&LmFFZ(U%he7T%q-Xc)7P
za1O}*oJpGUim{FiF(2okR$$Q7Cq^7Hhw?Ss%y$a14Q`1hv3$#%?rMTj6bImvPZIrL
z@%L9(Qf!4Kt<SGwv1=`Bx5izc(|7^ptmzcmyg_@q_9w&Q_m@ax?U)IX^7nseOU=K9
z<PgzkIT~e-{?j&Yl3TB9l^X_R<!;|HU^aBcal^TSiR#!@?T7Fo3)O>vZTIJGqBJ!~
z<5<3z(H{Qz{kMId=t65#)-oB#Z-j#2qki3A_-Xg0>W2(Kb+F>@BDo?w)+8Hf<uGh?
zUl%~0^yI&9zIqF~WQz&sGh(?a$V|~UEM_80#N{n%E{!j7l=4lH689EG(8Lv4M$ED(
zX@Q-{=FE}$)VZm__Vj2c7gLmTuY$dZj!-9o^==T_q(Tgig&y~jTR&dTEv6b&bAqsV
z{^)}e{xiCN_Qr1bSn0>YszmEH7C94x{Pi?UO3z-vC7&NL8<FyC<k<`p{HjuE-rK$U
zsMN(v{uc*wVfxdi{WKI&?0^q^K2vz^CicG_B__M+p{CsPk}L0`dVo>#IxgUWa2~N6
zr5!HS(}jaToEE2sMB`gt<Z#@%lfmouDd%P+L$4#85mi0DiJ$c|{o@QUcX)y4aN1oF
zE3}xXh}uC6XmwJIzgS&CjF-E^;wcM9K*styyLA334E@Vrj4ih%g_j)R6<MplX(DD1
z)E(`oh8C|1CJvBiio7~h>P+-!Jp%=9NHb3a-HJHLe*ZIhP?uW`EqM!}fdO&sDn>_(
z1n8aed$q~^3B%e8!J2AW!AH!onF<|#!2Po8@W@MAclKiO>b`r8>A!b<Gra@#^&fI?
zv|-(@k-qz;aD(xsCp2Y6|M=(74ZmW77BwXG^Mal0N9y^*mXVK;=$`Cf><^4B=qv|>
z@%LGDsyD}D5x&!6AAg<*P7%Q$8P2zOYu0udAei$$c=NdyDgIzNR4Ns$N_8<bu#yx_
zpL5FVI>8D%5LVL?ZK0EPBiG>nUKebF)cntVdFf9Y#yjulsH*mKau<2vFuD8_tw-G0
za`B2AtbJHC!X5U91tFJj`$Kp?-UgP_IGcJ+iT=82tR>9JI9~h|x>*ge*9RtFhWyJp
zl2~l;!o|}=TbF!CV~_Wgj+BphT)*X1$dNUfPc`lNhc))Zm*_qpDKfU0e^EzEu->M;
zB7l0oC(RE8F%ScUKp9>&=c}MCT%I-0(nYM!%rQ|eC80@#EHczK^FzG@Sm%W2Pjs5h
z<|N8{$8wiGwd9>XZ2hG$yXwA~xUNdNhzE}BW3*eAb%i1p!fv&8u;05VTPQCOw~d89
zoSlx-P3iB?EI+Z<+#4;9IYl5%i=x1$aLvOUy<-Efp&1AlkqwEPjQL29k^mXt`p-aL
z7pIVaI+5OHT7K7&*lv5>Y_|&GRCTagy=u&mp0jrHUp*PdcGJ%i0&awc0xEJLt4o$(
zkY!ttK#V?7rxVRMu!RW(1!=*QO|q5^rV`R#^h+TQIOb>3^oKzHn%j`ya&69FOBa&L
zPHt+v7kclz9d%F{Q2_)E0{VXO`6wT33!r(R=>adhQA+j;8o$H`INZ##AH!{8zNdTV
zFB8{7{0#trEV*`6Vy1A$)c+HFiQR!9dCyZWv$%uWKECGro#sLY62^e6qcU|?aQ(Pn
z!74X`QDU4?H~w(0lg2WVVw$T$&9{GLyps(xOYJ|LyVHifV&zD79M^SNjM$i)OVow`
zCO#QCdDfwa8rfy6Tpy^$lBJn^;+F&2#|ujkluI^sYce(pBf2!fWC|}P1P%#hvk+Uy
zp5q@&cE~TQ)SNh{Qq%c!8mR=k;~6!4EdAGk&#b!~>2sK&oy%m**4-132Rx_#kH1&+
z9rLe4bEGr=+I#Fs-5##{a*S>N7+TCZ_INWPxY;PE!bItd=!LUR#r%j*_y5<4OIc`D
z-PbwKN|b(?=+gf)1US!guBakgjVIjnm2T7u;-2yQkWO)IPG3dEoe4?MRWb|%aK0`a
z;-O4UY|Yx|$B8ukGl(URSN<ta%FrhfjRQJVdYITaDAb;|1sJVM-RQ+GF3CxbXg75f
zS)}RtF5^J06aph3c(n$;&OOQxbYuQy%Om#S^wepUfMdn7TH;|oZQ%78w<Lc}X?8_g
z^|bf<P*!yZuCOS<-0}{C&VlNvm960E3-RK8^~1%jeU?~><OPkU^EoI{+JLd_UG?@R
zYC{N~+$V3O1;EOz>b~6sUwKO1SBx#q(0R;=Tr9AeX`+UvM=mv;o7h4i%Vz^IDQ>xP
zqZU?wIlCl!YG<2&ESLQmwk`K94fm1(M*E*+-K7jo;6J(f-z_W8L;j<*R&ASpS@&i~
z&)qaIrk)GgSC!o#ZA;+_wOhkobs9g^(o7!$qmp2boT{j1+Aw?0!Hg)N<6~7ClS&*N
zN<5psf0zQf#d&w1!&s3nc*B7uLFtN<sq=khO94)I%*9D4OVvhrMJ>WAicUxqjHMV1
z;@2LoBX3~eIcrz+TBJY7WJ=&SpIDMhX+?kAxT{rZz;wB*HGXIWO%upzy1VX#^aUy?
zHHCvg3DUl2N9`}?#;-g0(*CQt;^h4HTfkRj(W$WfMQ$CN9T+4aE=u_?55j>>fBQF2
z+9;Q2^MnFC??kHn-043ocgS@9Ivf9QcoE4E_cE49T+P9=b?4xn2KSrxPK6faSQawl
zGXr0<u=1g5IN0Ah;OG;t_SC<d*t>4hK`8Zl(8K{fdAPrbT|JJPcx_Ke2@i1XEfg`V
zquXVq+-LR?T@SbxF#HjMAY6T&*EVWFYu5sm$C89b2uLmV5VtGZ5{5AEhtx%hiK}y2
z0;s+Yg!#cf`w=fJ*~KjzRHg~5r5=Lj;oUtz<CYY2GHZ-r8d1?n#1Cyz^B&)17x-Ki
zHGf2%S+sqXnE{zY)xVb!1uk4ArKXFHQgGt03MG(o<Y;j8b|Op&EfDx@x3p(7-6*c+
zo{h5%xmx-py2X=0`NufcTOoLaNkfQ$;z@mf{bHhjIVfgpOAl&Qm>$QtfQqpA9Lbz6
z@&4*&m-`{KYw4sDtpjX8CG*oKkHSKZiI?dE*o0Zl^+y;u+iY=1)jrSjuZqK?<`aNJ
z^$+<{7V^Q$@CG>ar<un#@Q<zEe*R^k_Wh3!B0sA^i$HiCv-Bdun53Gk#Q?<%Bjt^q
z{o^EW#MjJ{BikDIo+<f!{6_AWqwYbJkRFCE%VB~QFppu^f-%Y$kHmJ#4!T&2dE=h<
zuLVOy0C4O$xJg+$G;CHnxhKJy`;^#=bvC88;tOF>T5l2_UH7=%;YsIT!BWGNY;(Pf
zM~nPkqYwGgHW<dod{$}eE)hWw+~vK5o=j<5?pTdl*}!&|2y<mUw+&>|hQ2>6&G(aA
zXQC(jX%*OywEy%3UlL<Oq<%D?^UrSeNSdf70eSI^K1SL<H+8s;*5Si;cQMms;6(#D
z*9M)1$6AW?&6h_^hc~1vo;?||-?*9zuYlLVQtrgw04j_M>DE916`Hj3>3<El@#%K|
zuCI`#{cv;Dd+4eeZMRz~kM-N<B<r=u4V1cXho~J%!MD}YUw2;nKXZ95Oi9k1v?Aq8
zId#bvyOXsIuIs@bQ{@y((b%4O7)9awogQV%J}^kBL7+TDaYF?HEVYoFkZ>9c*yYVe
z^fC;_2bE*!Q%(xYrnqA4$0ld%v$qy+%qSiVQ)hB7ZixP<z-5+Ohf@~rpEeEQE)@NQ
znpyXRZbPup_)b8r1hry2z9N1Zls!EQYUFo0?2OU4V9)3`)XYNWP2Q-j`Uw=?KXqO#
zEVJa&LcOJAJU6;790g?SA;SEa9&?Ta@@&z;OK%#ZxaNhc0=b=a;mOY$!v$IUtIr+K
zehg0x`zMdduKe$NPKCGB{Kmq7pV@b?78k*)+_V7yWNY!4p3``wRL#rxuPox!iO^Pp
zK=seqisvzUeNOStM9oN{S8=HYdhy~vJ|XQ_3S{51llWq=<zl1<x_(3ox73YjHaI8S
zEkry(eTFE+NZ(bgnt$L`|4GP=brc!OWKk1G*RC9n4O+qLKDa|?+a~{56QfhpVj8#(
zXZ|gV>2PFoCbJnhq3NJ+?8qp77$nOAwhvB42aRr)lRAGm=07bs{-c)-ZjHRpFy=dU
zcAzxuOsFRTPGcb6#z5ADZ77#PdvfJ;+y$6drNMlWLSv>K#l|E1T#H3AQh}wX@iP_=
zZ@B|8@KJuRH=_8fb@wA=gxvDo7V@)P-|%hzgTPLv#Dz)0UZuvY^jmVHztT;KGonBl
zIl6mOt4~NF(;i>KI3w=MOnD_kKUwuVWiElC?lNj<QavO-{ntNs*J3OnYzC_T_jNIi
zn6T5yf;;V$)j%0(__Q6ah*#fmE}C-OLu&0zt8`%HV31)2+MRPv8|HYIj*>yng8{A7
zuk`7rd*9l#br}~CuL$2!MmQ_PIQuY-KePmZS{hG6Mbu<sIBtD~>Zh>dCeV&y3+)92
zvGTVUlu2I;SKVz*Zg}=YY4;PxqTuAKl6AFMo>-S0^4MPwGW~?Me!iSNfnV{2yt-XE
z?I}qMu95mE!89c4^+-FGjgpSF{`g_tAZjwRz-*xoA}wC$o8?67zyv%i^?D!0F611G
zSu@NrTuH3%BGA@SyBduVsH86w1JSY({V*Mj^zkD&$ms6vsm0Jdb0MS1vFx9GsU!Uo
zH$BTAyMgX^oCIGN^L`ZENBUOZumP~&$FKLsFQw%DJ-Pu|FE&Q)e}1Y!6kc=)ezoL3
zvZ$%DaQ!|M;tE|xj4B%VUI6XieGp1#m(i|DgrbxO^}5c0!|g1EL;hw7M=gqT+nPid
z=a#O{WOjdE2RL3MaWAXNd2!<gt`#|se`0NZUx|0%c*|ipt%Z`_d}W0$@a$~Klu;z<
zyx<afncUaw7_kF9IOqQ)@yQ_%Vfl<H>^UZdU%g!(8BhMj4SRIq9I`vvoIepSv6O?*
zR3`DaZ6(ZcXv<_zyA~G`RDO-S*+Bi(uXsuWYAPxqCPB*ujPY^VTkw7B<{UE)POTR3
zd;)!UA5MyP3ecGJxvc)C6r1r<iZzgx;6Ghbiur2*Q8?n092kg%Kjl%g5s-GIr^I;|
zGdGZ=*}N$PFT^S#c~8R?1RO7Fyb2YJ%FbG#!w=gOhbgPooc%19<}6Lzf^4-MI<1GR
zA<4B?Ia0Q9ie6$k^F54dAF6Sju%9(HNQMu=J{kEk`QisM?B{8cuiBN8Mao`&hGFBV
zA~t}~jZM{_cb&iy1F0EGJKfd7JkD)NvBMKH;_f5PzStF+yvMW}KpmqV>Y=8`jvjEN
zLx8yQ4gUKK9VQ}di(<WE>+j%aO4y(jjx^r@BMRGr6LZG>d)hPO(oT^#ON}<azlUl@
z!G+b3Ck4;f2r@F}pU$iv6@sSjE}wLk(5MZCGj#-c=U33OV<8LzFPhnIj%~-t<ZR3J
zMjGDIX!@^v!DWF*LRj6Iqj*qV{^8zWN#ac8kw%O7a~Ux_#w47<71dl}TU>vMW5IWy
z6Z4~iWTW(2U9}mzGEFWJY}WedMu5CAc_Xd4bi@#D31E70I*S+m24`A!p&zYMtL}_;
zyy?C0L`J@X)wjihB3*%<uD6n6mkL=fC9CY5md@nBKt2_G#>wHGb}Np~$EZwME5rQI
zIQRJU@lm#x$|JUQSxn0FBGvb;s^lE&j`Z)k%(OpkL*84#VS8pne2KVc?8ngccgIG=
z)=>`Uiu8e>@!NXgBNXjWa{1maq`o}|CCG*C?FUp1?dQ6%nr<MnaLQz4gfWLsGvZ&*
z$33RP)CoNDdx$M`#_aRCxzfCy37W^*E1k*qeE%^*+AIdkdMM)-L)K#*%AMdu{NV$g
z_l>7o!??a$DMCs6)Dw44uk;CZS-}5bd7K6e4+pwQ70IM4DTRI|^B_E>xnR=|$U%P&
zBi*sJyO(3u1eAqdq{Po8$;ZM&9HY0Wl!)8Zcs<XCe$LXAH9h!F)Q?BcV7Aj%xJqE3
zsH+~7F!n8_avq2;D2^s59UzU#zyXZX>8p!Ni4HaGP2d!noMdy{WO*Qtr?8%XKGG7;
zAMbSjHCIVI2<${ZCo*+Df4lS}OG5rPY|E&xeQhevB*k4xj7WcAg__EF#f0dRj!*B0
z&$<()7Ozp5s)X-fj*(?VmUdesxZ?q1;7Yu4pwDW1(fNoRjNyA(j>{9%Ar)fxx`qjV
z0YOFzyl{evFjCK-NpQ>`_QyO5eHD`aMol!{oujY@##pSpRN^0ty$U!^&I<-GLm$9%
z3gGtvj=f9m;$$1CTZ|)>qNspQo#WMPI>_@yne{r|oYPuQnnMRE?6952GAf+PGPvXB
z%9z{{nJ2WA#prg@4PWS0h}1M0*SHaD|4|mpw$dzCDuwzsWnnQS(S-{-^h>tfpT8t2
ze)d9*Pr7<(P$3#~v#X@tN$H1P)arYDoeeEJ+S$m)#SRp^>=+V&Cr5^YDw$;Pyq!8%
zCFZe4JK(BU!HRG4D&Kk5CFE8Hw&cInIWhOt(@5M`)tvtwdo0}j0+E*=r{NZ_FKDT&
z!X#-cajPi46bIZ*veK;KT*gF`w<=z29p8|0x!FY6Ddvb>&GoCypU1?MOWrzv^2lZ6
z!6F%d^?J6G^=Kp>V3NmM+2SaFXg)(wAP*KdJ1-;w$E2s5{bRM=EZKiJUev-eygfZK
zgCf`uG=#6Hba5$=7dNMo?bO6-R4BTBNl+#B)9Q-t*wNwT;YCP(vZnE}``RV983}Yi
zlooJu39qx}DL09g6Q2YTLtdb|<8ks$#vVtV15=dG^FuLN@e#b*4_7nqZS+F$LL*P*
zW<cQv?y6s(S?H3WW9p$#BZEqr&Htk#5(IZJ4c(E^DL2kl;^0em$Tdc83)T7_J{wz0
z`gH8vL<&}j8pQui5<-j#@RJ7e&K2py<7`_fM$s$yP66ub#lTG~mYfy+;>r5XBXEt`
zcAc_oAy-5IFT12>>cQ4EA-rfZ-elSl{aa#vP}$VGX>WFE@=-o`F|1E9Ni7m*Bvr8K
z1G3Sc(}*vCK!jl59)9YM^_f@eU(!;a^TUR6hRFZ^v$~td%pcClk@a9a@nq(OAIf@f
z3yrgnc}8h2ka95iwGYLc_yq?XrFFckfn$JllAmm6b%dMY-b`XVpFmECQ0mcM-K3_8
z-5NX~t}bkhy=C`^l7rze&PSoX3(SJuNW>>_ZkwHi(g|fLO9#^}hT<#<;-IzNe+^{%
z2HlrrpN%H;Nxt*BB^bp)KS(n;7NdXX+1Mhns1G$8R&4|*qPliEjQk$CArD6D3j9Gx
zscwGGQZuJpN8t6dA+BXL?U62OHXlyV8|1MnI%qx0x)pye$zi)jg^3ZGFgBe=*qDQ2
z)~6}|x6gs|4UN@0x3ivi*rBS}mjLMF|11-Ql-)mMw@NU|?g}b*ljR|~WA#;McI_h!
z9FCrb6unblB|{#);1lEVt;8KuihS}6JGi{$bx~=~GCwGD^GR~1@fb&ak&YUm(CmKK
z_MJ)qcU*$T|E}2#BZP7MJ@5_#!W5|k4_!GN?X`QKn?=dvSf}({Z>2i7@7rfd)Xa_k
zR4thcwp)#M#>z_reHB6lj{7!J-1iEOAez;#f>}W!ASVr=HgUy1xm_`+7OYDl+7BC2
z%Fb%|1Mi>Et`J<*k-KJXu9S(4Kt}Tb%$cnsjV4*_^ogcwXngFpP49FHa5Av3fgH++
z$eh8~&Xkic5aWnYyg)4pcOW4%J$V5m*INR566xSarp#Z}`Z%`Jp_Bod2PmcVF9qtL
z02ymVY~j3o+NvpLD@rMs3#nFBi1m$_6?^pkr`;FC4$4D=nt+r;Op^=tiCrCwDq=eS
z1p;0io30`ON@eLL3JC?(j!*MGKZL94Z#}&$R1-LUvc7O9NgP7M$9&%T2y-yWVVm&Z
zax&hsxMiVUQ7LUB<GyFA&ljpxr+2TEti)+55}gb-$gVDi!`C%<VvvE&mIB0n76gK!
zEImvU?X(IalUi<)UIb_@KGa?g9kQ5{z8Q1RGj|ohI5}KCKNvY&mO#gB#Rvg7x&b#W
z#)5QO;{C?cc|e=<fUw|ra8~Jf@Vnl;>z8w=>%sZ#cOk62zXR`MlVY5F@enT=e2YVu
zr38!*%>$Al>llpbZH&tWhP&@UUGLx>=b}uZ#dV=IUBKA$RYcaSeOwd+?D{#Uk*E)y
z^!Qg>Dyy18jd#uPOSV5aB%0)?zY)=Cd#+4gkhy^fXOx3C_~1@?<l)q}Qmk%U9uR$>
zHuwwUvD<~!YOQ*HeO#vw{*|ED^VK+9_}edJi4ypM1B1Y-M#CqR159cS%o^;LU{;kw
zJ~sGq>viwGtNV+v2%X35J~}Hg98qf$3^!EuB(af}&Gda^Fm3FvYFZ&>wV@zxlH<R)
zt{CP63}3`=;m0|q&pCsKN5nSLqP?Z@=r)6vVj#4>S+mGt*#nRn-OzD3e$TcKe%p*4
z_wvvsI??1SGtOJ8EF5ghH}4}%C_-2OI(x)8m@vMWnl1x!bracynUA3fUi-Y_MsC^r
z4M~W=wH=$7xIqZ1%%s>h(Yw6}BJ#cp!1_e-8?y_`pjGftqvc5v1rk%)2;w~1XvmlQ
z=0!xXuGo(i;p2vuML$b0tLMjsn)VIo^aCzO`fn<8U#W>oj6PIsckT?FI3Gl1-n`U3
zwjn+ya3=b5iwRjRF(E{_0{5Ow`b#kaZ`+^PA*fB|?9&i$lGp&oee5WC|8Ws9t*#W8
zZ)`;FS{rl0C{p!daMw>P14p<>d^+Jz^kvGb%Bzlug0Umt)YIVXhwIt(;q1qOWlj&a
zxFL@drEh!4?v|xzGmW{f@zHynz)+CQ;az6dC>n#ctlaR(9NNX6pDj*${n>`-j$@9h
zKqjx4cUZ~rws6D3vNb&4Wjlm^9pxu*Zcgug>Lfc&F;A224-_YDQ8B4d?DpivRL&Tq
z%|?;2BPomUNLLm+!FP|4<9lS9Dg$VgG8^p0t^TSr+(3T^pNJ$2q)m1NCEc9or1i_e
zUfK()rb0+tct?)$@O%28hSxo^A~>`H!=qf#czNj05L)i5hmc8047HKR;S}~A`DjE$
zvN0ci3p6J_37Ox^3FH0)>wAvInwUeSilTlfiO)4_&ae4L&AYBW7z_wXN$aMf{~nXt
zpmg+#0*l2g$%88exR*4l|AUT`?%Qt4nZv3tFh(4g>PH>AxK6mqw{EBEl@8b&;}-#!
zHTA{y!5r%oc$E-7JnpBX8Ibpo4(vAK99DgGKNphIg4|ViUxXU`2-{a=i>H)VTLExv
z#CmTtnuG1-Ixac7elDbThYCBwUGCRHigR|LG~QO?;+bE2Q4ck~9)BCWbfTK%zx%7H
z80RJ=M_v(Y`u_PBL_mJ~+%C#=uxoM;51*|&sW8$8`;JyOlAu+)7)ZztZCVwCQx(-;
zsuDinF?vl=ztH`SBKBU*thm}B$EDo-872lxLED`cSa%!O$8bL<Qmh=zg(i$NTlEl$
z3uDcU9H<80ikucyt6*c_rYR{26-S5C<%~XJCvBR}ufg3n<4}(Lf%-6NP6I4H&~^_x
zA6dZH98v|2SW^4dP6Y{!>tuV~Gk#g?ZXdIEyz-wv`xqUUq9s7)q$LxAOA78^qWEav
ztGos;MdQRgRJj3-qE-TZc_}&DK>GqHFv@=j4kXL%0t3^&<QnM9$984-=4__U%_ugi
zn$Kf)?tEyL#~{y28y&`4c7ni1sE4pzA8IiTn^=eL6_+*i(v<=U^)zWGurQpH+>7cI
za{~FOf}HAg)rYPIz$HsVKL*7^K$-bp<Jr0ydcaVmGHu`iYUBQjk*&{FQR>))ia^8V
zB_EKc^4P~dz!hu@6Q{^hu%AKv#>tPR>o3ua!NB{{Y&YCv$(=EVJJDOHlQBpg>%I=^
z`=~BD*|Z-A9B8H4)}U^f8PYd-Np^t~G@#Ts^!-+IwVkV;CcGtnjz5CuRej#93HvWE
z9dfIkF$$eDIO-*eDdD<4;t_n7W2Nyd)#r}jO59H;O$l{<`u@Ta&i8IcBL82kY&u&t
z=hNRXiPZfmB<BNvC_zN~SB>LB9ob{;nD0hjbpO>&r5u^Tx_!Xufg&pJorH|Cu3U-2
zCDVBq1d@AR@QPvhd|QnAW}EP&-&uTpQ7*!aQsE&NA-l%lfIU2*tA2GO2#tu=f*06j
zg)ShaA^K`6wtiMNzc_L$+j-+u=<;oc)YKUw1KXe#DzydA=nXuQpi6r$!|Xl!#S+Pk
z>Oirg{Jw|R6raCw#eEOp#Y)J#L|sjDwv2c~&E_Fj&1jUF9U~}3>2t{scp)Cu&owYg
zdnMv5HE#X&s=T;}kTKJI;Wdgcf+rWD@k_aa7k6CuuB8KPv3;ii4mV!WFRiplP1%?=
zd0EUl1kyf)uOhSf5%Ze0`6q*SBLxp)C_1hLh!pY8s?wr&C^?^RTK6rR;w^t6mR^)`
zYBZ*)!yjEkCr+pF=O)kXi8PQBaS&Yyp$_+L>jdJ9H-*h+G^L<ukZ>$QktBpU)59JA
ziW40__zKlBnphzs>NWTEH!`DxL2Ff(-m-CA`L%W;o`?4)BIK7+vRFSkY}<>N{kb#)
z@9x@oizNZhJ}i<{c4LUwiDjCglk`FeHnc%&7v%|&{x$yx^$_$PN3!y$=*ZzIo=V(!
z9CEBv=gpP=QDpBh$&s1E$>tXM6nqQ1-c$%qgfU%^$u#iDcX`q@wB;d}-fZCZ=OTh%
zd!E{ptarXkfeFM0y?0kunC|GUKDZ90<}%#p<W%)8cCKI?WQZ`qVN!2mNr5~|l4f`d
zoszA2U!aCkWY58QNCE5cdQ{ho{8lV%Pr{MhP6alQz`8^Xedm7ayjH7ivAJ)YX<|5W
zv7%jmfiwQOOr87MqW)8M@cZCJ)LBFTK0tw;fwC)5chX#7vj&KBgd7V&q5+H_%V{I~
z3ah=`7AtJ~nZx2FvShPFQ{6_6A&mvt&0^j`R5U=p2;aO7<`3&ygpweAotUSto<3;#
zAC}sKZ~^U1IS`f~Aq|b{2Ik~K+js8p>MB5rY&Y`pK-Osq%`81C{tRR0P+*&gg9RBf
z-iJ|Meo2l0UVtog<6OsMq`plKEhZq4s0K?q9^HMV<<MrD-6$LxU_#!7OG4$99%RT3
zeeZ0+Fn|X2{jgVi<L%Lsb`SIb${vp^sG4au3z?it^J7j+o@ojxjR|&zIP-91#LR1!
zMM~Y)QC!Ovnniu$WOYkDI=;SQc7|SF@jHB-jFvtiqjEcEq~Mu{NY6NL?D?%URTB|>
zu2GG|PU?+&Q&K2H{P9X*CFf3>Rkk)K{yD8ZBi&TUyWDMm?LQPd;VqKi#`Xvh=~>j(
z_pRx_{DR?^!bv9L?U!Za&@6q~H1|%W$|iZdt8QlvoU5h_T0S`@nM&h`9#w3xvOjhY
zi1zv*yGBJ!Nm$w4LQeL-g{jtv9z>h~-t(=$J76BaGX?p2#=r&SnU<C!pvw6f*IfXq
zIYS6@j;wKIN$l_;o0tB#M_0G>9#|gcT#=(|oH$GIqh#@^I5TWx(=~%PuN5m_O?awg
z3N=XJ1M#k_TgVs6b?imJ+_}uO*(Q_u&{61H&r!GAS}epqj0YH6D?OLA+X{LAh;o;4
zFqm<b@v7jj?H}@;aGQK{GL_-l@6dTCUveII(^XUy%4k(t#0ny2tF}w;mby9}?dQq+
zDHR%_EJm`Ird7jMrF!*v;yTzv){W9?oPgrPn+r|8muZxbImkc6DX$eIR?ZC!Hkb{^
zf5F$tt#t7yUw7eA>xWp`i$;-(SGzy3uS)nY(Zu_wxu^(|qh@7uWmzscDK7$7JzsQQ
z-PQn<SzQW0=*fx+9tie-De7ggN*CqeEY<(4ZLW}w-t^%J>Q41k<I^VLnB6TAQdKcf
zh*`NxlO^d@(_$LR6bRzHw&gpCgZ3T=q~h+N`P}Dwaw>VC7lBZpPqb;M*YkbD*L;=#
z#{4}9XVA`s8i(f$b^`rMhg}AYTCazd_i<Jq-u)p^gq)egKuV>hl`B4xyN-zX90WU7
zETqZ|dB6@EWOLIT-k3SNsYVjm@3t?Hug-GwytbSpPuKLOXJ=c=mdR4v4q7B<9Uy{B
z@*IEp|7ROjAB|q2v_hYKNlX4SbjA2!-IISV5oXhhl|a&hOF-VXWo{q$_BlPwiObgV
zt<~^f$T9gloY#1~+G2GO?_wL!Pk-MhLX?6?sb|R(Bmd6}AoHIJHkiazq6HAyk=n`0
zjc;Nkj13pmq^ZjVjI^zS`Cc+}HNB#po~J1nH_~-BXY5z|Ve(cI_!I4*xc+x`j~^R2
z{1C;%kTX0*M~B<UL>zd)y6LZux<?4h@;l#ycrCS*7s3USdOQgByT>(MJ&TXTz*jVA
z-968UxLF=0K4|{+<~+)r+*MB4qcCqv)6um^VmVp$HBNq#2_+7dd?ynF>{~BcqsvbV
z|15pD#s;;d)Yy90_YBKdrAYk*O_lDQyM3P)z=hE_Z2hGP+Tx;jS4z!=3lJ1jdddqJ
z-PO4_fT<t)9~m&`eE&H)=(+y$o(O3@n+pz_wE319ev4{Njb|~^m;tt`2d1!|3>!^y
zIHMTZT+&sxKL=K1OyH2n*k#8;(D?4)-PgQhVCUW3VDL7%Kd%vu7Z&^9xmlb9GY6Zx
zrtFQ=1+*I#oGXuBwfYQWv)YXADHBrDu6IK`5iR&YWzo?cU$T{#v+yM)%|=r-%T+V8
zK7DSl6o-zg+|3&IN7RTbANZrai#=XHuyhMU%c*%1>CbOIl~9j*TDM0L70zu`w!Q^T
z7Zpi!QA6m^@#!%6XZCgq_e<3{HmO6rOB*7$35(it;v1a!zkGUsGN~r*E=ng@(cm|}
zbN&16bEZ@608{8}mcBMk8Op2l=|iIZ<><_1o|}iySC39)F##KeC;{B~MkQOg8s=h8
z^H=$vg@AIE`gab}wjX9pig^)TkVdL2@wo?SP?Sb0?yEA4r-gXt7rtLV$LELI#?&FL
z`*!Ys8JwCGhD<GVLj@c?g*JX>2Z+CBrd#>C=t?d}l9rxNFCFdEXZY&F^iNMAy8+q!
z8fLyw&*9(hrodR$Py4O+7}3)9;MA9zY8IidHL~3I$xG|IvMYXG7uwKSc=kK{u{7AO
zB#+wTvHeOYLryP1!^-;O<^r9^jCb5W6$D*ymaZ2OTcih%g@cUo?kpu`>E7+RH>S#1
z8|bAekKNnB3QFLUAVm5PCUx3Jhy9(9G&DWZcAkf#f)`XkT|ZnwJ0a9hBay=Q;gvN-
z+Je~A5gk{7N9iv~Q23;|&7&crDlTlt4DW1jAiUcAq7;4eCkk_hNXRAu%#oNKIkt+`
zH$P!)#!-{m+-fvK`V)}btotY9wR8F7{}Uvfrpfr=Ali~qJceIh67!`C(sAOu^!Kn#
zE}Xhj=_A{pYAd}v?tTEgC^|ze4Q$`<mell6_Jok5zG=OHly4G1F08I3y~L;XXRWxT
zsCVf_C?+>wLP)I$U@x&7Cs7>mLM7U%2iOnj_mON8CM0E_7n-AAX1-;D2M%GykYO!a
zj!)F!Rmw+7Uuld4FA5=(mO)Gs?@03@C^P=^hZ>P5%IecyqlL|4ha5=AVdRp#btM2(
zL#nIgKXXX-POy{CNF`OL_!Jnd)t)L4%H@8j=apYQ&kpSUO38^Yf_F^{`Ni=NOTTfQ
zXiqhgSXw#i$Jkq*cj}ycSC4|+6H?o!Pmr}o(wsSM=`O`30`nFVz|sr<VlD)!zA<WE
z`soc=YWPCdZyH&H9Gl?1ApN|xy6W>(J7JesZ0_`)m~~O#wK<Jg6{G@lr>{BKYHBi_
zShCeJUSxWEWIJ5>EI3A>g~UBEKAqnkKj&N@#zbeIy4Ndx`<)|V4PVUCtG1Qghm4TT
zGXACJwi3~Vt*%B}{@Zon2T%)6{3()#^VDoVlO5K&u>=DmDa9y*dKh4fb=^|Vg~nsj
zzv(wf;i?R|54K(r6ai&D^FF}u35K+%`^E>Ha=B*n0<p4|?LUvR^noT&TdsPpfry-B
znc%89AgYPY@}fW!6(YM1Kl-m|4jt`g5E3&|aRj&46WHpn$;@gpGNWX307~8RI8wEM
zy8o7?g=6f-3C_x{+(bM_T;3lqq(CGcdsr6nx`ZIGx^@ivfA)<OMmAt}``UNH@!OB^
zjw*yu-?%;VOIm2J<k9H3l&I@kViL_=uo<>g1^dcuQD2iU{v|^t)05YIW^54%WAeg*
z$E%{Xip7lo?kr%I_7|Nu%c1_dl?QCN5cF{izGWhrQnlvwic8OQIWY{Or^`60W@t%Y
zV?!g+YBI58)<4-ImMg0Q*Ks+@A8F`f-8iu^p1*5Jn*vxIBVC*{vbiX_M$inDJ}m=?
z=5y&gsh^(rhi<)zD%gktn=(_I`#rIq99$*r(`pd7VW5&~Tp3cAo`~&b#amhp>RkDX
zz1k%=G-H^lvdMlohA~?$>wD%w;)NHs6OMPh^5@R{%eQ)ZVM|M5TO?<0qK5lZ5t*vY
z{{osA3}{?MnU@Abmr(aMdpSYJmJdH!{LpgLp!?aBD6<$KUQXjWXs#L}Geo%~KLE81
zZv~OSr5IiLI|Jp#%|}=7H_`~Z<GjDmQe~m^%r?cJiTJz(xY<%Bp2-S)7X#I=`UV?x
zxT6|}jmj5bGCD)%L2xxzcl65j3f59~&Ixa5f7*7?^jAAh{9}r{Gl!`2AKF16)bL6T
zmp^(6SWxtf$TCqTp3_`#{9LmkPK5-N2~B?>PL5}edy2apmATO!`@(QXAiSm={pMOu
zG=Nl~+}zllihasfcVO9Q;mfegflt$zY^?XkdySBdO-Rgl32o-Fi1MJ$=3AflLAALz
ze>Dj%@pggva4qid?E2SW<hLT(KLsz1AFFSPJUkA>IVM;Q5lBvNhA>P^vv5fPv=4he
ze}U-Wl8>i>UKjHK%AHeSV8ksnQ&1+_qF~XycQf!zfs2`rz-(nh;zKw2wG443{zvZK
z(cmZ&*6}aLC)I9l@WGbWce8CN;ZN_ejQQi<7Lxfdcf=fWUB>6z$FNOVw2nL6dz6Z;
zCRMbZo(KfKX0RjEDL845c*Bn!N+6d{s@stM=%c?8AFz9;0>8p;*P2u5>7o1QH=d1I
z{5%1Rxm8Zv`sEmyLz2kS{CBFJn{Xi`wyZ7vQ3Y?nigt0z^e!}qE^#V6A90^YYX&`A
zGLGX0xoh^XG>IL_kQfrtMR1Co``sCG2Kv((?s<jc|IN`I^SuetRXaQ>LtcdAUqQq9
zLvIW|Xo0e@0<fQz?4*AIeD0i)3;8$C`-`{`e5m0c?xa)uHSB4#a=mwHWM21)`td5N
zm2UaN+kXK!lpTv_H<`QEFC2A4YDF_zx5Ba=h@xz2;+TIiQ7U;|dm$x$kPWs490(i~
zrTxtPXQCCwRvPKay6z_zDScVNVW}4d_N-^UdbO^n`;z9EO3pNNkbTn=#q?)Zix)S6
z3Y8mgf@-Jnpo)9Ei8p!wg)cH>ww3A?ca8X5Y^vHh{&?^Vw!M6h`E-(mV}6e}9RZ}1
zP!M;+*IFi`S=-@rut)x-jpu3v4*G2|kk$9(w$QHHf$!N|Y__`wQ=VIr0~L=wF&AnF
z`RkHwim(%Lt|dd-z7v2gyyJiNOqS+~3+;ZwjM|B8Q5ll29#!SWl1&=@lyQ%yL1piD
zrFpQ!QsFNc#dsbQ3W`sSi<epCd<u;?3=O?xB)GbO*)C#KszAKBW^7V7(I^{aT8%2Z
zyPD?ralcjdP3(6DO%(6a?WGgipQ9Z5_(UZmvYFKrw`_}o6bzb7ruv@hyC{s-{a#tc
zRQ^N)Ll)p*!8?fei+s%xjnP17BA@b%Yl~w;47S=qczk_&oZMpLNEX$nBuspp+#=pi
zIPR3}s`U66{;~5bw7`9D+Q+#!6%9YUKa#04A-|;t_gU<C*Pk^qS*}91UR}tTzVev)
zgvY4DglX3NH9CaSDr?fC8|ohF_oEJ=IDAo2+7-2)EUCR{#p}6O)JE%y6G$+D<IikV
z(+Zlihh0=r1pbci>VWjfnW+p3$ERxmubZqF5$37u)iP2*yt8A^HSA8OMf*{ys#yUv
zZCnS3bje~a*?SinQ^)6ZNRAhA1_~YL>{g#0%VHKISU;Wt!Bpuc8N=5H(&-y_cv=cx
zR3}R5P;M(}mrVv;u7T+@YiIT@JkV*b3Lz*2R}`HM9>4HfqJ;4P%c%K|^;Tv1SPR79
z)w&Abr5B}loGr11e@8l)ynICYf%v$gkuN@7D>3r{`qx)C)I9Q=XxgY2c|{TgL<lIm
zl@(O&3bB_P&9s3-oPs6JBjmp^XD81vdl3~kCzmu_5A~IXYtENc^Y<U|%sY@5i^~&w
zbF)A_3JdIy_MOgH)wQpmf!%DkivcIl#yW1ktUt53x|B$1L&9-=bTuQjOG#7B`#)Yh
zY87l-HvxkyO^oF98O@N3<!?d(p}E1rO74!bMlwab(Qg(o=XG@NqE}9wcDv#~;J3#%
zb68&6e(n=?w(mVunK`hEV1{=pM+KT*vu(b&{v&s^T(M#Jve%*EnKGshauv)eyEM`T
zCtR4UJdgXWS@ya0q)yhH0HXRyv-9<2KQvvXk`%PB`gtudE|NA4&F?~}h3&#2#@GFY
ztoB-gU4}xqL?Y|F17{+Bx&wQ1C2284+{&BQa0l$0mi|F#tmT(0a$UX!TV}Q%R)9Wv
zTo#}~l<r;ES1OrX#j73f6&c3E{(~Y)9=p>I8?%Q)exqWW4WcE-eU1@N%8I+!NZrqU
z`nu5GADYxo%e9Oc|H6eni*urlm^%r_Z7lXIG0WWV{i+cmj<vui>ghW(;oR=NM5)i%
zlP9Vw%d7CVjfRbSzW3@kGj+iCpiiW*;bN>gQ{}U<hhJ|Ocz-8ozC6Gyng1aVP}PdS
zSI?avU-^9SzT($B@j8Na*F7U`6a#f5k)sYA5T?vCd6IkudP<=<pbuwU?_63`-)*B@
z*rCjN$ryP>{W^m_9%_n||EoBn;x2k@fWnR7F#|vIshXhXV#^P@bKtywuIM#5$nbc%
z{`uw=!{%$o!})c`)g$b~<eGCW1<8dy+|30~)sS#TDr?Wp=6m5%LPQ#BBQ)d(-np&z
z+ebKmFRNjI>ZyNj8Yi*gm48fwoMeJEozDj`Bv>mk>YnOwXn=O4d(7FSGN!TIwmj95
z>gQkFie{*2ub&BO_&wQJo&bFNB>7G;L=n+rrCT_q=1M$1<1WgF#*98f&??TX&Q{Dk
zg2H*j*h@DdK!P&Amiuu9-5cHSv+P(`3ZoqEGG5^Hq1{*LS*#vBGtzS=YOZgcan{1q
zIxb`dBYUyqTYtZys(vt;qeTkyj~}0LdpP^NWaQ#9WvT(1w$AeIe%OQc>D)>rG$&M}
zB=5IIlg5DU&dzg@$5Lje1kD)Nudsp`+odq2Sn>``6;14#yYnaY?LQ~{<!?EBoF6u8
zW<}&#!xR2<uPv?ol>6FP&*qB%z2x-Vgk*AohMUZsSbAuU{oG59`q>$ux{^e*T4Bz}
zbM`>iK*bZ-8~0vP+^{#&vRzEdrQtuXYtIRsSqh)9C;4`DWF_Wxj?3uv>ZA_m8Ewu*
z?N)sjHKOQLU=9~}zz4jAfmF)IAx_#LF=V-ufR`cR08L7mBgo>4GOWfx_=vHsCwtZ$
zJe&T`EDW4(xHz?!MqpRQdE}Y#H;lAKv(KB65@%>_TLHjGIh|u5Ug+sEdPmWp-;3D`
zB4+OG3PKa#sXge>?p_sq?-nOX?k`}ggT+uus@R*wfsSK602;^2v9r)F1B+ry@0zi3
z|31SV2rfgaQZSfO+AEmZ4;C&(aB(mw^Ao)T1e7&Wr-VEW4oE_bxa3%j^tNv`jlO^n
zw$q9ize`?cf1L2I%c6)AP^JXP3tbIof`fq;y9HSR7uP?Won6VfNqf>^_Fz>~s;+@w
zLvM+4ul>1?<?X<x2bzyIBhOQ`3{PVKx+U5K>sTJP6Uyc#o6bio%&`o88i64g<;YEY
zPqbfr3qtu3^>=0d4tdj;kAcp&*lQM-w?uR$w_IxcxALZfYSQKRfxB$PxAuyyihsI(
zno1oeC|(Hi5LY4jzs1)(5DkEq@2NU_HpG)p_^T_h$>(KL{Y6{vPxIwbV4rKQ;(_0L
z<`@^p!bpcGQq?L4?OlcW5hzV}-1#u}&}xFTRd(a6$X8pBa?{)z(mvizu<Zo2o)+cU
z{|e&X5(C9YE?}zrEemh?-&GF=C2Le=t1Df5<)d~0&$@hy&)<m>ljyvUVu082lj5?P
z^iQ?6(yWGb{rJaul`aa}rorMgkTgD~1PMC-sKP_}YTW9)U;7@f-WUGJ$@8>`TH5hc
zIsB{qOpE#0M&#3!yqKDyPb^S<1T{R~U9wX};1?6L(X^jrvq5$oHe(pDJX9M%KG8fB
zNqf7RyIr#)hnh;hpM)*tg!<lIlhA={EptlsaLT2vI>knjXO{tNV`v)`)^w}y6JQbD
zY2m*IhZF*%DTclf{E1`1dJjGv*&ucR+?^YOva9aB&SY4^Lr$uiY6pRwdqI!abw8cE
z)121?u?}|CD-Ov}++~58HCIu;j{Tz`NzcD9M0~e<%<%0E<Cp)cJj<#H+6`L&n!X<#
zmZ9Y^!r?+@T&8kEXsBOcrr%@*kHGqRs88L$&#Me#0(3gxTbo7R|8|;R?yXg%>tBAm
zHCp+5U~o9<r<tUo<qXWxmFRzZIX?<gtHhT8P>37ecH)2flrJHB@)^wJt%&9u`%9(;
zc}3Z+OM5JA^M(JQKF|Df%ecjr;3P)4k`JypMVj-#t=KD|M&Y1vD?o87P31&bS49|5
z32_Nu$jwIk)rZry#g0CTW%eQ{$~yDWiRnPls1sWl)K)4r?0<*Gj2C1|QT<w$(#*V*
zrwX7*4`C8ovn$SsC7;Ml=4J;JaEX5y6hZw4FbeEfAcR*7KRtyt2C#Y6N{X3gF4sYJ
zcODUHk@|!utKYb`Tj9o;gU{o%bBAM%q&Ri6k8GXlxZC;@S?oz>uqhH;75nN`|C-DK
zV(I`}K=qo7zyuxpR3J%;Zth4!uGRHMvgrQB=WqtZitN(?c_0?HXVB{Q$OJD*jj2DP
zBH8o+miqVZ!h~}~Cg-*_%>phOX&ogiUrUCyQcQMMpbm+Z5&@Dc_5ZUTAz*fsuHTb)
z$~tr!O(l&VN#*Bl9UfxD(dA-R%4lrw8eHO!y4O6e#BC2~FHBgp7u`(!#T-CFnO}va
zUv`W6(E%hlaO%CaG$hRb_i59C@QqX?*b9VmJ^z4!Q#K?QB#-svC)!!+LI5=&C;K#Y
z+aU#82i2G14Y(@U+$q+nYrI=}VPM?mF<*~nCHi#{5{eoeuka&pA^LIwiL{p+U#9;Y
zo_!Sc7s&AUqd2^{VeLzcPupk2dZHaXAis3JX>Ar*rcO&2*q*Ie20Vx%##_%dIR~9L
z_b5-Mochw|YyXG6_l|1n>)J(=(4<R|-g}fH5_(l>0RaIENbiK+q?belL3)?2bfp(5
z(jro%3nEAn0*Z)$gx-_8<L`aH`+nox``0<=j&a90B!6UNlU1HI=QGQid+n>e@O%-;
znU_(D$!Pmze}Un}(bl7(y{6s=`+&x^zsYwO;j>UmclO=aIoH$aOgy@GI&cWyI6clW
z?3FnsNr?$zTt}7c@oK=ig52?xa7cw>L*Fl86AVb+s1T+$4szntA0+yC>xksK9xG=?
z(GQX`2Jbh0+TOjkw?F38f1Q1PdCio>Ncnn9MP&6{BQ^PF`PQ6Mn|#J7PK!TYPw$G{
zs<6^;{&SV&U2#I{imJ~>Q(KkE&)KjN)d;h}YrBKE!l`F=T0A$D;Ty|qbuCA?L*3#U
zw^e!vA4D%?9r8DK*LyZt^wV!CHb@-?xh9`Y-*u|oaRWT?he6i9M8rgTA1i(jxc*Ll
zLw)4w3Bo}-is)mW7wuo629-Q}k4{P{yA^AXPA~J9Bh0C&1PqkW^mou6Zp1Xbcl^?P
zKmCr)g6O)6qTYPHVD}Ha`IWMb55P?;@;xt$_^HHIMm|k`pb~R^`K~GD_ay({<n^RZ
zkRhdmpnE#g<GV&vaJdB_-_6VTo6l|Iwa2fL3a#=PI#TI}pF&Gg%XDI{r%@}?2|r?A
zQ8Jq>#6Kz8Pw|_@Tf}@<pi@X5Ik9gfer@vNhti=At?Br<zI7Uk_KByp!1beB6#Y!+
zx=bf#T9@JPLCp|yXPR3oA8y8c?@D*Zyv%W?1=q$u@&rHWBGpm*G4$v56Ob;tYJDIC
zm?7vYc#0u1#s!o~h3&*q9`AxIw}1CQf!VL~*Xn{YJP4(P&r{W_?8hI_e`zQ6c77b(
z93%UA=1EY(UXLP+RdDj$Bw6cX#|Ksq>09Kf8D=-{z4=(JUf~}ejf%1^SRY9Kf@2nv
zcfwc8l^K9F724z1%3D}H+3O~?KbQ!vf4xk9-K&y(;UzL8!l#mcA;hE(-XJ(9@$$<0
zY#o_g@<KmwYSR8K%lU|E(68*3MOt`F3gN~WV1PQAgtIgm1>rB8Z;y0!^!K=Wc)f(5
zzkZ(T<1>U~_uiBMs#T4Gd(@G^PV=eD?}6X+YBUbLTN9bklq6c!ib=GHK<M|d00sc4
z9RQ99CoH)U;e0WINnir{<I=sAP)~0weVz1$rtQrv8#H9x&OMM|YDLxw&+r32FmeI+
zZvqMB4h<sEdNF07>U+CL!AL>$$snhG(g?&QrA(?V@n(ZuPD)jxdHJPmo@su;!Zv-@
zqR+vX+cbUhud1&d@^Ep^bCj`}HkB0Fm<X}ZR1x_L1)f(iCTc)AeBR?8f}D7o8^#k<
z8q;lJ{Ipd@9C+S%cNL!3NdR$3=+pB-zsHZ)_WB$%d8j35`A2W>q14yFz-FC&^44Jl
z+jZ1=n-glGQH?EPjq)T=Us~Ju*XbQvf!HaYnb58gMPB^RNuISP@~_jQdjy@92Hk~>
z5Icg3Q#9@?PCBoKVU!z&+a6JOq^{rmi+UpA(RBbq6tsU4)46Hv3`X9-=wn>9qxA#S
zk%izTJ}Fap3e~wD0Bot#{$gJHKbW4zE#~?X4*LSuE@H4U#Fv)@q=UMT_D0{rh^yz;
z(m-9sZVgqI5y?WVNPiLur{Lh|n<XLV=`g$np{WoihhUISH*n7zFc2!@xA-=odKS3g
z!1E*HLN=RAjqOAT%a8HF%n~Q6lABNfPk+s4OsrH)B*U??Wc7EO66b9NxedB^*7B*R
zeJf2{_(OE}`LE&jCa@SQo9>MYwIOc{9}D0S+qjV;{vZBUk-W$5>f)}jXyp5X0WtK7
zO2;B(n~&ALEp+tCRTq4SBj5Gez1A<YK35_ay3P62?0`yk(W@xi@$W)DW@())^1@j(
zKHbX>pB!b9=R4VkQi95LDQ5+fCLf+E(gIP^e8t6ffiG58c<dmADaI`|%iOZP^~ag~
z>bq`o+qL{){aLsd*C*1BZy$Dm!j`-Qv>gz{9HBU&EV|7%7*C4Y&1z(G<Z1CI9hrmg
z1yB;ps)dCP{&GY=1CGbn_Odz0kU?bllf0DKWvUl4=WE7ud&{PyROf4{&bMU*fiP`J
zYN<aW?3}0Z1MBtTox<^g5gK8rQUX4&=HsRp>4-GYN($jCj{XRJX+m=+wC8JkNrX=m
zL25QdhutALSSyAMW$pq7zZKPT^M~!1zN;3R>x=qA2O{)OFecT&FHbI)u%hsJhfe--
za2q9@@71Sgp;WIHX$RRg`MwpL^^lEi{MEogJ)ne~AR5brQv!gmI4GbXEQOacj{bMS
zb>L=pStA)ff~p!CCH!4>4XDyTzj%Hd5Tq`$cnvnt=&LhdS#AE8!l<DC*Jgbzn5!j3
z$vhx`NHTf#mKTYd`R42#V#KEgV-_C~w^ExVqw22`xFD6||6xQRFK#Rq7J*SHfv>>}
z*4mMM=nrUR^eaEwyM(c&m;872cz$b{1f=%P2fr{2(n<C~U;0$=`X+&1pQjrA_tv3-
z6q2CJ>Iq%{K`V@<*ngy*kt1q~*o4r*f4(V5sX+vt{Li-zR}mQ0r+e+wdeFj)*C>Hl
ze$+s)%4}4Df&X{qLZ%Y<qncFD!h|A$w9s~y96Af9EOS-vkKN-UOJii`Yg4f!uYI#a
z&*FEtLKz)i2G+*PyyQ0~g7Yo<y7Z_rf7B55;SxXbTB4X62$d_x9Hh~%F`A#PeUO*t
zok`bK$xADudJ&RL`2n&tiZvnCxnz=<r^}2T9HDVS#}gm&t@#qa$R~&+EpH+l<tFhP
zDaAbS9>Tr|fr>&75`?asp_BRi>IXYG={UV|p@e;)=<`&)I}jS)8BI<vb4{%@4laS<
z&Io?Oe`N8;=~7k^Jts6`OyQ<JR?gWf()a;RvBZ0t-!4q}7h#?_JIYP6en&v~mgW7p
zUW$mT3Ir>k%K#UdRE2`|@oCHT0&#^5%H@SXUxO<%>f`HLtX#T4#mq1HZ`l(m0!e2v
z1)o)&2p4OCi<PWgaYgndyqFu94yq)6({v(kvY+<BJaAiLFY4=ppNZh~1H5ZeqJ$6S
z;D`KU3FZSJW=K}**?X$X&t^k0Rd;CB4I7f{KXwL&mXW0j^q{YxwQeXU3;59Qvfg(f
zv~=A^rSb1K;y0d+Wl?OvKRUq_K77|^QlPz?E<)N0Iu<#R94C`cibLvtto40IEMFU{
zP_CP@ZPk@$+;#FvZEW2KlSd%=rzqg8aVh1?|6OkWQ-=OO_?(U(UkrPuSE_Z7<=?dc
zM$@s2_-O&_Xt~=gyC8Z2r@K$IuY*YaM%P80zI?f#)b;M&cGp_f*=a#z>rBn*@831=
zYQP_U8_qoCVS%R#6x*YCMJ7<Jq|w>0_XcWnYTo<_FBto>eAZUeRZ#xtO-)x}S9@Xm
zZ<=niQ~}`s|L5QLU5GGYXlb+yVCe8<&H2mnl&Eg72N4%Z7un{wm9NBdByx(1hc*~9
z^ed@6JP6Z1yVC-WpmzH@Z6y^%YN3+P<vGjv)>y?CDebQ0;p~IcU-BF-ulcsEFR_{?
z%iT(M{FQ3JM1}22@q?qYZ_cn^H-7IwdVj;^hcw!GPeJYK?<UKgf^N1*(Q=Q(mOM7<
z;_w}|mjRY!+VYe!wB8ivROj93Y_!4(QP;z?HBs=N(L{NMW7v*^4k}W2KxkGHTE}tx
zZ1Y&FvFQjVkH6L!b_Bs+3KgF3qW)S@+3gcb{Vr5Nd9eZN(h^FK>0#w!b^>x>e*YR}
z+O~ens_|j{W~)K-SI-mYd-0FHZ3h%jyd19h>`F3Rkun1+9^C-FuJ|0>7lL!4YRG#h
zzv$xvi$CpiMJ@($@?><R9iuy&)4M`4!}ZeE%r8}ye%0+D+7W!0f4n=Dm!7|N$J#?2
z$b&dEd#;(Emi034>a0+jsf?>O@5t(<-pn!(9x#(hcn2+XEHLP3nO~b!2<sLrEDFz{
zhDj7QI617j-(NOUVG}IAM?lZxlf%?teWszWB~iRgEm294emBG7-b1_O)v5KJOk>~3
z8mkV{U$nrL)N80KVuu$o$2>nniu^|>RX+VVPMS1ni;&Wt?-<-6b3DV<n$OKQHaQxT
zL}=W5z~1z6TY$~)#LuV0;d903U4$cpt=~vt)}-P;<4LD(5-;q~0E(03n`*FP+zF|M
zZ_X|}V6q=?{BBYbEB@I*%Hj^Jnnr%8aK~Eh7|-zFX}2n=MRy?eCvM<-!6l+G=lbny
zy<~od;#;+iP3}1y-B-M>N>*3>l;Oi!NtxNNo!%4QVm6&y5E|%4O&fm}`towPJng$!
zu4aCwLFM<FB5yy<@@-(frji<7bIw<V?smcBUzqA_P%Q0Eh_IHvTiVw2jT9}`AO3v9
zWy}>h@$)hA-qc_~PYCWeuyR$vp~RmK;$}b>%a4&$WqPrT4cBzmv*<gmw&tGO3Z@>p
zt=z^Pn$Njl0>lzk1r%{#I6H79zaRbsN!No_WB=<rUD9!@3+rd@Cr#zw8c^B8<v#`6
z?3gCr^|XY-Xx>x+i1?LpJ4_*h9%<(9uvs?&B63CgOh0g3HTcv}e;Ob8rIsK%?{^hm
zlKH+FL!LTb!&s<qpZj1`Fj1v;G~gQkm6I@**2oI=F-W4j&lnl<Sx$)rq)WXQeY#{i
zr(}Dup=kD_R_2X$S*ux^+vU~Ie|9oQH|L;*9mK~9Sngh(ouK@#rOfdW1u^y)ZJ{oi
z_GkEQ)<p2IR&n(Z0iuvo3(b5ESJt-!|AGiMPk?CR)3#3oZo4~tM#7SaDtZPScBA$;
zRfT7!;pXd8Cak(gh>E0z0|TaIc6(_oyem4!?0^mvF|w<mMQ4@Di8QXT5}hjR0UrI@
z@pg8c+DRT?@<=n{pY@v+v-`OV4WX+OuK%TLN?1avMhCwKGS_`i)XXoR3mZ;E_+|Q<
zs<1f>SJeEe?Q7d_muz&{uZ>p_Gr&i@bz%jiy}==nlV??|_~7qY>*H}5X$N>~(c*5x
zibmE`0U}@a6RbtwdH2W=f7wIrjxHP(Q6zU}$8|s=aft9<+(I@s+WjG>xw=(YxTRRx
zw-+#Y7*Thj>cDqU8LA+^*nqIA+#S@7p<`Bc@lO0pV#-_T)FYNEplu#asjXyPSLxO9
z-3hshqCZFN6WQVEhUKdW*2n$-7PL*pWo7)YbuD>QJne(--^{gO(qVu)qKso{mvtMr
z+GY>ga>mcUip}Fyb=S$U`YXdbBK2sE#wZD(_O-^nejB}cUh64w#!x%6yS^VdTqd=;
zJE*8IAMk@}e0%?*Ido3ecox06+NivP-|Wtubb5Z`bmDAI)lL{ro8?eN?ru}O!rH%R
zbtnQ*LvD-CbO0u={YTx%8PEsJW3?=EBI{6hnhx%-EqyrlhQfAx)fq4?y|RsKE(iYE
z=9+%#z5x-QDir*cO-@Sjp%gPi=h&#q!w^SRPnVqqxYVzIh6WJC;Muj6p5q4Au_b~!
zOdS}>Tn{<f)bAKC<P|S+P10((*dA&2{HQ#zJ5j;%mLPb=H!}@o0m!6x|D_m@f&1(W
zx)mZtg*Wgx7QYLKCDDH54A^MUQC;G*)1ysQnR)zk^Eu(zA-hPSf$f{$n=+Uk5!^0P
z1`*ku-2rJ~OQV({#WLg4f1(Xl@;LxI^w?PC?)DYju_*BDKmf7d@5I=hYeW6{%f>Rf
zfjA&Uw>cl*4WRa)JP)A87DRo0u7f7|xxxjYs#Xa%v(gSCwgK`2P`khYY|&e)PK&K6
z-A_PPHiYG_O=i(?&lwnRw#|X$ROhp-+D?P{njf@FX)rNxpZmrjo=S(TSq9`#0j7r_
zSW}1-wF}6}3)8W01gwBj%6tkxfk}N%R|T}vKK*xiH~T1cL%f1KlW6gS{~*%LW#x$b
zq^A0Sjg^;{{Y79792T&dv#~Gc8WLZW9o<~{@b1a+<*tfB-$lL7hZ`SypU==bB5H>@
zTdwPd5Ffv%O7dB#mbD*mzZB?u$YwzQ`3uwIvuBB~Z)O_E0)|K9Tl)AzU`EZ}|1yZV
z4XCl9a<?)HZc*-pM8iq~#k2WcEUmwN;xP5`-=e-wP(l}~H+k=r|C=?Gu76>&x$0QN
z>?S{PHg%PYskey5iKsI6N>ex0=6wZ6t(6bT^WQJ@#v&NKxaedzJTUxswwZ13u72Ly
z$<#9xUQN54@rj5tulw`S{O;?G{r1m#CqG6XEJv;uPe{5v?p{jJjI6;cc0Q>4AD1lM
zSYp6TKOx$XU~v>{iT)S(M*s8P)&DPBa9JWlN~4S3Nb$i_qx(;-QcKqDE+i6O%iniW
zn>5KK8>v-bNEJX-RN)2TIV#im$rm|hk4Xg*qTG=g=uz}R+x`G*1F4HTxbj0cRe(D2
zpj!a`dEnG4=lULC@k;9b9S)O$Z$A82%h21P{~IWV&AS+WH3$dwInA*E7Ha#dkkKUT
zqATDque+r6(kJk&{7*<Z<RPdM6^D#$v^xZatKETiFdSd5{*a<9tmsby#IdH<YnJ%C
z@YH?SMg%;yEQv4z<6qV4f)Te4eMzu3MGX-lnTI485_JEp?_yf%j5ug9*}d>)FJ&k<
z)^LT-hSP9XmHeC^Yl+F+JjO2ysB#Y<e!Fz8jAdOZ;w91lGr%nhPyIXpThsVd0limq
zBtW#hTpBiSe`aZ{<M|kjrF`CbT7H-NW?hJQ*lp`86YYgS7U6{5R{hrKEsT$~SY2zU
zDge(Y`rk0XCmG*ZfI5l?yYwYMG*>xWOfzw-LTfl0$*Vlu^caul84;tFvu%r$n@IqC
zob|XY0mP{ZUf0;(@(XJ+%x7%<gZv*c_uBDr%b6cvr5UfCo}lT}7tqDI9Nij+zTK8L
zMBu3u|Fvk0bwp}Lv&C1sahkK^#mGG;Iger99~!A82%8HD_~bs&ei=INjT8@D@a;OX
zkVliWDY)ZR>H<8q<YwK!`@U^WhAc7M6YyfF_qr>k8xt6%NF$~N2*;=Y>gV!TFwbj-
zQKnVar`;s}!c43~xwN6|$5dYo?xzbN9sv_bg@sk&OUaknK&f-}LT4u`D}0y%G(S;j
zrzco~zYN?JTd~g!@dP)M;HXR_(KA^B@V}t*Z$uz^<bYWya)DLU_lF`9Uqs7lL$NLx
zL3Is$3wj<^2nhi%v0^s>oc{_!3PBM5S42(!r;-1I!2cEmCapz>xB&eLCg=qqtFUEy
zj3H*ba_mwa%dsnl!ukr92TOz*M+UhU*zsZ?W7ylY6cgCVv9xGml(NFmf!FwK>;)kd
zZ}!+Z_ZTXknQ<%SE~JJV<sU@U;XLgD-3Hoauz$!yl2{?ftwkprjfflAY6|##TrV_k
zvCzM#Kf)Pd`z(n2r@Z10#iQ=RlK~mMf_e9#{oS1E{f$Om_?t!??KED4po72wZ5H(x
z)ZOpeaAkf9j0|AMs--n$YlduT?y>jIvVrs=1tXkG!1+5j^BRr0U$y7%SPm7>17orx
zj0w<3!i%7jdZ;*5;k~?P-KF9DM<S9}>wlIqdhT1!t#FWgA^b-=OV`rbK{6!zWGWO3
zH@}Y}WMwxZ<(z8A1})F{etQ?>DyL5F1w)fOtQs@ILuCZu303%N;3Wt&{o1y+giGqc
z#+nApYfVze)XI9yb1)H-OIIJd&9hQsN!;;#TGlKshP_<Yah`mu&(ne^Zy~Gu`hxa_
z&>suE!@|OO{3_F0#WyOTV!JS#9duZ<L)tm>cTNwlHO%13X$og_gZ$2NB0t6nBwX9T
zTVE`35(q~C&c;ytD%`h`?U-)VQ@obsP?$;6wHZ$-HWp#t95LXgh}b!IOfuG8n|oqQ
z^nRW5JpW{mYtA13%=b*fqaorpCdCaujObvMkOjFJi1IuP$+J-C%Kh6jGyAIL*(j$<
z%lz)IRq3;QW;mPEG=BNR$hqp`Y4<-&CsZKGBmG*2SRYaEsfx*AEjVMsP+>BSP7&(G
z?|04TG{nZdS?YLNI%}F#!9UlU*T#|y8$#{gZ|NPbd1AOd8t~i5QIr+Y63JoS>cJnd
zW!`*&ZhXaNAJ5oqTxr|VL&!E52l>j{%5}^)aed!DZOz||tFy!LT4%V70Aj%uPqu+S
z{6~@}Sq76pcVbf<avjeOU$uTT*ID@CskRVb_+aix^Ev%W^y~cBb7xs&j|CI*-Q#oz
zI^K4wdOz5MCc1@cA@B_gUHOeovznmHQ87TACW3*r7uNsSdHBxA(j_Sb5h2-?lRad2
zw{}*~QUH->iZ{at3<HK0kgR^~Ht@?({<GDEt}E*Cm=4vShu!T>o@=5Pzf5=Q<PX*l
z-p(NWZ+Fs7fDA)8WQK$0CCQG1|GecETw$6KKu67X;j?vHz&t~W_)}Vl6J4?L-g?Uf
z&Utip$nJ9Oto#4iUl;i0tlzG|&%?rR_IR<ji*4wZqO$IibM#ChrMdpj&n?Vh<5F4_
z+Q6r_J(?x<MSxE8HuyIsZ;w3;<feY!Im*S}vU48#{nG$ch9-kOb7&_o7g4TsdXdu!
z*ysN2ze*Dk5R?MUuB1M-8}aC~eMA=+L#B5)J_W*3Vjebj;JWX2AVSRAS+^%F8MP9W
zZ`qkIQO8^+g?bsKttEvJq2HIT1q9a!96A4FZrclU&-8t3%ykhyYPrV;|2}fM1iZT2
zJG22NqDU`S({*X$_2TgQNj1;`4h-=)t3*4Iy%3IJA2<Sc6ZwCf4A*>mt@X9vrVW~D
zcAJ~5W=J{QT;Xs<wDU{Cy1<Yu3v91=VVkU3t}Jc$*NkiPB*OBqf(jS@)qNq$Ae$TT
z1e(EcTvHz|3|0H1IfAmo;idd@l5br^9U@tWDi{V>_p6bmk!`8rHcsAY=kPyAU&Iar
zjJPfWN3#oNwmY;ZVGX#ff2brO`DPU8n9TOKUn8A&u3_P$LP#aDx_vl&o*r}7He*hi
zcYWA5@s-?4RlZgw9kQ4O@{+kd(=w!T5=Y!B(}K6tI}CWl)!F5E?Hd7afH+;J@ztgT
zLPlH#{EeJ%xiQd-)d^!?l<#YUYO+8suL$<Fh0B4S9%pWt-!gl95KLd&RyRZ(!~~P~
zSN3KI1Xp|hIwJM>=<Z3bH05aN*ualHT{vWR-tJ^=ZjT`?o;y22*b=Yl6y>Q<ZDuLr
zbb-||1~lpCoo!cHV4<a*2#kW){D*U=$Rb(5tzpe!=4syzUdcUo)$?C;O!Co2_LZ)6
zJw8<3Mr36cD1I)UT{4xa&;Sh=(;RJokSAMJhIIv0=9WmC#n%F)aVYH+Y;|@qn_S{;
zec^erB=(oF^pu=H+ER)tXF)+}VSBl5-4%o0mlG&2>OgWG25Uy&04swG_%SIX%DG34
zNhodlhtB)PqXiC<4)A^i&k#P@n```-w1n<R0ds?|d0XU#yI;Y(G{`nF4G5R<qNrQt
z>_py{_u~v?LuY)WROwyV=!0`NZ6mx1-hk{M+9W9M|9SuKMF62L2n2Lek3epQkA(yf
zL#1e<Spx(M;JCjICde6r?<@HFI~?R#%i87(Yl?t3v)5HGi!ZDby^Xl6k_keTklJ6s
z2`WZ_v6mLlbtgqmWU$ZQNWF@61TGZ=0bi;>!dpVnsOOH7#iznt{3{fQasq+2VZns(
z0Ei_YJ|>;Yd~^<nPhv$uGC*9J<3hL``nUfgpnondQ1fdb+=Mw4keHi47J_{Neu+5#
z`*ZbrB5OH%3^7Jzj<UU}VsNiCHmye=xOujPylSL$9GrQX+A0pYO1@yGCobJJX@Ob;
zb)$aFz40x6ZGTbwI<+UX$kK`}*E|=UBQ)TyrQ~6NSzi8WlqT_wM2_aK9HMEEHbInF
zPi3VgXhn`3bW=~XA`>(YUp7Xg&^YuoBVM&yciEMELFGKpMX6K~T*$AYRU5falx%+A
zyNnRlNbeYZrI$<nHMR8IcHm2+(XkIC>1U#9^e+}7lfs@=v;>2Ra+Q*6YN;Zm@GoNP
z2yuuK$N2WniT{*yQOQnrxXOX%6Ad&RO&LRnS{*+($NZSWo4PCwzm7-*5}x;7FzZQd
zMCes9GL?DVz|vsu7VUK@xjK+zf5J9KboiTw$lES_WANkK`paG$K?m>SpwsyAn(GpP
ze(}GL+gz3_|4pUM|C7hMalxNrQ)<|^S-|m9Q+iJ(G%vF!ZqhmC(|)ifB1A*5?Nd`D
zZl)J*<H45jW}B48>F_L8EJh#6;QS5ROgE8@?Vn1NIB=YwY?jgmIACfI5+c`i>6}Y0
z=TO0-SxD3POG)I%^m|dFD%kqc9~o$&(x&_B3m7pU#6YQ+1;pU3+{LvDi5Ed?6sknx
z@N*iyY#8q?!ygt8HS}Qz0Fs6cbf>ki>eHmRZHjP~IlA*{FIEGnvBy;77MYwOw+&do
zz~<jHg&7F80pT$luRn-Ni{hLwq{+_eIW=ffK>n$%T?B=wD5tMCWUWk67h^Gnf9S6B
zl%ln+uWhf9!o?=~P<uHfmLFX{c*v+~X8@MHpU$i&3t*eWfLJ`nU&8ugGKmYc`Tdp<
zy-vZ<iHbrH*=0$5A*d^ROOezhSuu;h`AeLb?)>nz!6*QMOOnLNC9eiG@!S?VmU=M8
zDJA$cx=?z~>VX387wi@Z>~oc!H8V!SNbYJBwc_~O{!h5ltq>lBw5xjRR5asZnPkB!
z1G)`ep7!F^5wRL%1Bn+1|DzG+j--xZeYA9^s1XCf7^_=E&|-~kr{kxh{J0Y0%ICp6
zK+2;)3M4@N*#_NT+O+!%AD3zyh@?bDzLs6|)b``cd#DlwVJ@3Mcv1BA<EUdbq6^b{
zGSH>)_y<$=bL~xGApnLY<SG3Xzq8U)xk*$C(RO-?g_L<Z=N(DG=7;GO<%I<qS<3hU
zqL0mG<Nl0f8C85ecXR+RUK+3-Y|mk3N8I;y!W)`8DOzEL$-Pvl-H$l_N!!-K^PL$m
z{?;DdQ%WR1Dxe&Di<vFYq&=hAvr2$D6$WlNJ-grg>muhdy5$iJCjqjg<VY9%c54JE
z((DCpGm?|99;Zk?br_gDyq(;d`CI^=6YLLkm|2X}Xxoy9MenxpAT}A1zZgHt-WcJx
zC6&#HqB*7Tq8VkT(vNvdtZ!hmvuQeWti3kz7339mMLwm3-1}+*WhG7k)o#WS!Pl4g
z^`IBOT!|uUY;r)WgpCvv*WE^(y)V-ugea_v0I22^?L-5asWbEm5b0~kYdOlQv6|<*
zd^uzzbxa`g5^{4$Xr9T7x>-i7JF2A=JCcouB&2YP_`{UJ3l9tdZ?UShjyhrEjk&Jq
zkNKg)(k(5d6tYrheOxy#4zJTWSJQMcpyR^}L{_$&pk&xxs3_P{-}?<dl+E)rB3r#&
z?b^)co#?$axr)Ae16|Ax@fev3I5esME1n%xtFwGOJ62j>r`k6@*O?5<z0U{Fu?R}X
zU7(XvV^)`j-nK1mgS(P@wKp4MKi!nhos^SP``sr~x^;H-ugJ#K6RAz2OD0-zF3%}<
zZ#~@`ls#NBfMHwG)r)Ti^)4?Xa`6N7Cb{azsBuWQtIx-JgxzLfLgB6(4KpaSMBHBT
zbeI#wELi85S#GI0_0|fE6B~5IIOK!6#W~Nank3!grmB4^-aN2*0~oim=bWHaQa)EX
zxrF}QdNb$E%g1mdDzd2|PZ56(H=4~^cvtEr@;xy&Ej|dxx4|}8syWCc_0*2tmdR>V
z0aG?xkRF9;mq5+ujxfp<uH>lrq8^slnU1%8N!Mkt9yTfhniqfr`6mQu^3wR8t#$2W
z4ZCw*L74HRWBeg8S2@cgkgeoWqJ08G7R2h>)|{YwJa=V%iidjD2G$R<lbt5;xJKy|
zKV8Mp={mcK;?7Ey6;c43)2elvXoqIHOoNXeq%#9fc#18M2J6zXamBoPu1)^$T7YU;
z|5`U~U`))3StFNR%ZIGirlw9&_sx6Gy7bF7e?$aT+jD&p0-NgIV9b5A#g$dS*?bU(
zRSjFetH(#R6jl1e9hFtqCv9G$FDmGmK2=4?48V;&*%U$a){=did@-cGv=BT^XpFOV
z>QotHDkBXq8-Ca(f{P5;I}|Oo$^oQhp8l9WA>oj79P+WPqN&q>!ffA5<-U{W^Mg5V
z-kqOT($jdOvrPJ>tem$^bhvUNRac*`P87puTmG7<nqB;#XDgHUfL=dOgX>hWAw}=K
z1+kJvH!(cc@?NA^N6a=#5%sxpuncC|Ko>GjW=TA9Z7clJ@r?=X!oo$o935pV`>HAE
ztX%S-oxO!V?(&QP%>SOVQu=NEw^ttN7{5QY`h6#SAOEv=vy=byN|~@3JEmQ|Jbcj*
z9bzP>5S4Jc^C*XHKC79#0^kS0Vmaq1fSO5ac$l7@KE#}%Lr~o$;qZ7!xR7^bqr?RI
z*wZwpB%4!nPjk-|IZmHIc9wB?!H4={(Y2_DF7%n-Lq0663n^bFYS4^{(;Xx``t<6h
zP6;_S-pE!}tI{PDaPm1nM&V7>;)hg|W2(cIuzBH~`Va%i5)Ucy4|j^ZOC1tXYo!T9
zlqHTsu5uZQfVZ1kHVw158zFMmcI{xItCnnxm`gxYqIPi76O_m=&asrl4=&9#`br&m
z6B<PA^#C$cQpfA%g1~d5mVlH2d6mKP4NOqmkNn1g+fZ{_rTgQ2*HG-Kmux<bJXZgp
z=^pmvM7_Mf_T6Z4w1=-X+#`h3%T6O`!WqcVDl{rjur^bA*VYvTw{{6u7fdA(<I;N~
zE-wx4Zaz8beWzqrimfiOtkoS&pNV`|o@x#RbO|pJIbiIK6;gig-`{g*$Hxu3Pili8
z$Mgx*S*bS=@fBeteRx4vWQ2k3zN-57Iia?T4b(R9XbI+r6QQK<>205`#Lj*fz`!v3
zO_GPQAUKT4%1B2=DWRV%;Tb1zL?qXT+z675h}U|v&SwEc?Z(_9`<D$3YJ?p>@8747
zdyA*^k{@M12_Unauf^WZF<Ct*?;r&%_Rt+;GBgu9thA{-CO7+xot6wk+saZgb&rW}
z?QQQRy!nkK{grm(Tu!gM>#v!G5|;WB9hM`zNEg2SG;StY1mgT1$MD-v`72D-<B?N=
zU*lGKQfML9t`5a%#vOD|bs5J<OYL&;bltYnw*8jZrHe@+9^;!{JBCig=C)8`qniRp
zc^xL(V7PIQqtm;9`@5nC49FA7oHf}jycn7GNNG*LHod+?vm6|!i&zu%l2T(^_8VGi
zBl{J;yhKXAkgs0#XPjt;o}baF;ps18ktF)cK0fgzH_=+NmlJQD5N9&llR7>{hQMCR
zTi&S1SXRFi%|mH*r?a&*-uMcA>&Fmg`VF!SyVmVX9fN(SFTFn*Ra*?`iNJt}_4+*q
zg{HFb29wh9`^aM2d;6B5>gU@X#>h2#>bO*&NfT%HBct6f>gT|T<$=Oj73;>8&&N0_
z@8kANv@$%jP!*h~2jeo=7X1_Cw2sw7$(AR?_1(0KToOl!hE}9x3ZWZ}Rz|Pbs}~Of
z8#NjagbS$L@^!;Gj)Ni02fS`pTB#=OYs2WyQj2*bbG+7*<D~-H`PS#hRDfNKU4KM;
z+>Sm%(K^z`#C`aNN`j~oh2Iht4TgE|MKE~p@xK1`Y4T86{n?7EW8heA8rWb!pU)m$
zdeQ{z>B-ZS%p1Rjc?>=GO*PZvL6LpFbWz__(7V`+CrPLgWW24YOhPc6gqWd})t>jl
z5)OMg?f9{ppQ*2=3I|v~#;#jW*%zQK8)%q<Ajfib4yW!stvn^LhOa#Ux-crD!wT3M
zPO8#<j@u?%yw&?0H|)lZo<X%tgi78(xH#+t48QXJ14_(UM$m$~S?(1$rlQ}$*g%CZ
zzk1hS3S`3WO&ZsfFM>fkEpy=ObIrxA=sg{_+gi@S09}ZcOiWzdC}MaFJU_v0Mi0%8
z7Q5m=jRgF}<w~>}N^E10gDkZDO(^1ro7G(S%YRg`0`VJ!bS9RQ4f#<`vPcq&j6vY8
zPnd|jp1{HM8yCw5y`hp)7=O%UNl$+h@pO58D@Fkmvrxlt)G82rolJU$Ncd-Sz#W@C
z0XAys9RCs)hEqch);D~B*0ni7#y^JFM&a9XfAO@}54~#_LP;a2_yRS=gO(V@wKq5K
ze{F8sTDz4h0Cmcl200Hfs8<<!;!1@YQl6}0Rb8RcR&dld$VmX}Nq(*rtN%ntFr~K8
z&~0-G`GO=_^^XC1ck*-YV=rgFbY@T&d+HE-HS7W4o>$Y1hYP767hajVU19MMo|tnc
zMpq<J+Y$b(Y-YN2Yj=*0V7%Pt96|RSXs|wf>N<+~;<}`{oUdIK?{*B%B_FG%T75m*
zwQpjhF33#a>quUKJkcttSGgI)q^44^8Q%M(hfN*<*)-CD;kq@W4!SiTuu<cN8%Fzj
zH@kA}n3EW!jx}#V5Ny0>mbto%Qc+uJFWf6$XG|;9?vlRynGSg5t@T7N-|8WsH7qgj
z2i62Y3OAQe;7PR-2f9J~ct^%^)7lcrGF(O56X*EKzIZEOOE63ECn<(xsyjs2`0FGI
znU+x_339ta=vvO@X?zaTi9(pG?S6!N|DtEwrCeDZ4_-F^fRCk|w0Wj_{xCinh|HLK
zEzzG!k-yQ_;AJ(bTm#AJpq9k-of8>Qg%4D;goVP1oz0G@FfBgtQUdYx<wnphje&36
zN1=(7owG^FGa3cY-$Tu<pt{rXQ!56sgt2O(hEe%mNc&03{tT_MapT8ZCmCd{o0-#2
zlFkb5H%GLqgCH~R0M=LQfXQa``c3YpPkMy553H4keT<x{B&rx^(?M><kVv_D@N14G
zMP7^94jc`&Wm2<j+p+7O>Tfk{;_RbA4#f0bEA=qNyRvQSc24+_OS-(18QwC)j9tAU
zoh#5opP~3GshUrMjTJbey!t?Hhc#0GZq$ESt-5qyVqu^q1~I(V4_ah8z6+Hwa%hO&
zzU>2#d;RPq(u4?8Wq(Jdl3XdHngsjAeAfnQo;RgP8cvo685wQP0_l1`I;oh=HW_Ji
zB<V|pzBUGsb~eoqHxWm$+oHZ5%pOwnTNdIu<dj=FT`9~2u-f-ojY|DR-BpPPPs_9j
zQ3zBK6utIlb0RSeDg)BKZ>Ii~tYS8G<x)pw+K#K=1@(kT3%psm@Yj#yD{7F+&o92z
zY5kEWvnE4df-_jR7_H<d$3I$dA;f+OIa@JfLNO_y<4|V4XSM9Xi3-W+-6n4_WF*Ul
zbPSATl?F-nkMDeGql^fnWP7uHk7dVXNy-GYNwNuF0Er2#JXs=JP=6{J)X=a1FJ6P%
z>Kxp%RX=~UqG&8NnLsjX-v|;flB2|YViX?{3k@WIPh@-?WOP1%KVMLNb1A2sWuop1
zLk~dARBtEnl&FiCl)bmx`yb-9w2m^;0fswvFe>bOxn6I9I4k%kRH^q+0MU?ztC?Fz
zTsfxllK<}gzLOVggSnXc5DaT9aQ<=3UNm8x<3g^K-6M<Sk0qT6bq4OKY`;2avRy-(
zSk9&B&roE8ZJ9mXt_TQKJExsZdZGOCcN+-n1(4qD%>K8h;p)nd?dWaBg5{Sqd+(%;
zHWUKcdR(_l?+5(Uf|?&@ZDZRBB>OOv|2~gi#kk{ZQ%enfz?5m4%Eo~tgnAr9uLJ@&
z;z|1a;_r<*s+a?{G9B}HLFp9ty{rSQSyPTa#B{H_VY=yxBba66vtW<uO$yL>AtuvN
z*<%Y;p#&1@RF<7UcjyUr0rwUrbIAS27R~>G1o;fe`?GbWp|U^{#EWS?2t@Cury0eb
z@mKk%Fyuo9&TnLw6YvIiX_o^Ou^U~hp~%!1uYJRyCzz|dMLO;|$}ZE8e|z;`^vKW(
z7+Z{<`iEi4*ghDZ+xAuaJhFxU6|3f8NNny!S}bXNgs`UgqR!L!R_a#Bv4YptI#oL3
zNs|sEA$EBak6+i8kV(12KuW;qt3nIlS<xiABlAfMuD$(0gIoqO*3lvf2mUlef}^}_
zfAD-!Gv#&n+1I|32GRn)^%b;k?0CQy7p6-tz7kpMz`68&WoYt{3+iSL*T0t`2KAp-
z(Z@4+13qXD%QrvJ4`uQ^qP$!2U4hbcG(Q@9`(|p9%yh=vmHrCYFk3$>#Tue99*-9S
zQBFowg*%tO(_L#V6}Z)({=J_&bG{qWTHl2Cx>pb5Rj~cJM$bLS#ij&ouwx7KHdvj;
zLHgqg<T*P0_Dk#)vd=x$p5g}2`0E4z(E4mF#m9w+_@9n}X{tmd-t^1#V$yT&13?!&
zb?W5|4Kk;#w<vgANOSh)(@XMCW0=oHu!$<24_caEv3zgh>@NR+n4Ww(U)vP+tu-1C
zZL4YuQ)8nRtUgqg64RHT!HSJ>NgIHNd9$C6f0qM%UQ(Wf7l&jcK3}8JuAhaL-!!CP
zPx>RA^CYptwo~|%?<QHo2icTCWb9?0Z$yg&RnM*qR9+j31IYj5?@}Ub70C=-V{cDf
z7C3_ZoF$OUTdIN83q~h={2A>VGU%2Rt{;ClPjw!Z{gg#I5Qp{=NJy!Yv^9kWEfAzy
z`)Sf^Q?>f#RT>Ox;gF*UW&8#D3=f2F%Y>0_Yd@Fz!v*ewe}|OEp&pjC*?s=S5~$p#
zZxr)x!?H-fo%Q%0y_9ADtTdyOxG1TAjfezWmyhX^1zen<7v{2@paHNIZ<&eZrSpfF
zXT5rOK1>EWNIekw7_EUOpY^qxHuDeJ4L~2NpRYtZU$h9Lxq6rU?j0(@>jSOB?|lm!
ze0ryGtcR<~d{dU)!K8r1%b8*?Iz?65;T*rL2&6O-m<H1G+q#3y!dOLH>bTvCBM0<r
zs9SdbtyWJ^W@cjBmDV`t<L<~`mqcbl?1W=qY_DDqTRqZGbA2}HFJDp&4D(mb&Tn(u
z^oan}IG7n^FtVHX7VU}XuwgB$M9zSn=;NZR$vk*Go;+ZXHB+RI!;g+0c1Inb3B2zr
z-GZ}TSnaW6OnxUsnnBtPX*qqwy!Z}2sl_2X8~v!4nt^<;8&FQ<#i1if_3bLgC5C35
zF<kw-_EDsWI7?o@H7~__54F*6@TZ_0;)nE+j4p?I%l-DdHOBXn#rdY&DonM$Q?UXL
z>4XItnljcftnu6aS}CK%NagNhKK1wP$cJ>4K|k$sYKg>2y7c0GLY=>|rfV8f<bIYL
zaw+ntVh9vXn8-N6-xn_NGNZpXrjo5mG(y*Q0pxq5CC%l$M@q>hiFmk==4g=bffP8Y
zXs)ThJ@jN!U|Jq*JN<}l8HvlYC32P;MtRZ|UUaYrk8WIQ%YTvb!s_x#A9SJ6MB+|E
zwTLBIRsSwvD_!2?URE21_6xW>`dVj+k110{KvXn(g)G(L>*hqx<da?;;=MagvUU$e
z+?O)Y&vkrmm@sEGw+L#I%$VzYzv>l4+eD6~vxy2#msC8qRn-XP;&wfuI*2^zh6R${
zHE#`|Haudqf92Vw|8&7+wPy20LX4?+Q1_NtZZSXQ=c7VK#qZ2O#wbn-3NvjMP??vm
z#Nt~38*U6xi8(EY)Uagvb(P}CnajnjGxzf85Kl7{K^|`en2CUh0*2dGP#I;S9i3JZ
zd3HlyTr5%TGe+7BbR4cd$uv-@3b^Xvq3L{R5)Q9N&!5XKP%n28aXBA+xo~wpCaSjk
zQmACepT*L0@xmtk^C}$98yZB8oos3iPZ1jMXH*P)Kp*D-Y`48G`F2^Y_Tz<0mdS``
z8q<Ifx?NHr8b&wCqVbSxLNSFClwAlOFHxBclH%JOkD5Ikk4+YOn>$7w+h;gTP0Pl?
z4C3Mf#6D0)xF=_-icCCvot*q!CCcD3=R}gwPF{l#OAVX?pB5lUS986asYeRs#IQUJ
zGtJCs#)D!Hk^nRqw){b%l}#6GvvxuepM(5?_}Ky)eG&MAyudae%?~vvRl1fHdWYhn
zVBJ+z211l3*wzF83Tj@E&_9K7*Y)d|0;F0FkZNqY;{3LoQ%N{IKR$QN9#J1X-uo0k
zILxGtldMax5Ozs4ST1v8DiqR^uliz<@dI$s5mZg{;5v&xM~bK;$Ws3DvGg4;qE|47
zLs9poJjU-BgQC4yx~Fe7e^MjuP>qdSy2#rV5?yj+fhyQ9yg$8lO35ec8n0%xLR#B$
z&hHS|JP?@K_OB1}!^QY*9n6GEUVwlg_8uE&mS?f$vvCrA{pAs8Yj`&OdNiZcaEUr6
zP(D#2G`9t6l{h+MMp<P{3h2EQ_4xCr^hd0xv!3;1wtUd|upoD4!fwQM)LUpCWyjSP
zJS%jw04?@mbZg4=mfyz(UD$6oP$(V!cN|SLAe+RdVeM7vp<3KyL0-@xJ}vy3gVf0w
z#IitlG2<5%YnHSQtWgYF)7`}81Oa@xw^O5hXwMIEPKZZtWNvx$mLK-x>)ufv8+(M%
zcs+VV`^*nWZ%K_C#4V>zMV!#B52Ji}>pwfl0TTFr7V;I=a=}<%=Pu#Tr=aIL4W2m&
z9P{>8_6mV+ksu^P^ZI8VFe-u`EN@By6U}to0DTn+8F5XW9=kmmIA7cp=6XYrGE}IJ
z=@<bG3c(z^MNs73mdrzE!V;VBMYx;Ib+x3L-%$a8Z5`UdXE`0*l^8mh(1S~c&h-xc
z{>5$F;W*0@=pij{ROqfzUlYVJ69{BFToBDP$Sb`hU*>a4-tHyC7Kb=N4BWW#AEcNv
zqd7yU5)ltd74$IcP<4hbn&_<sVjdtIU}H7f)f91-XsMt1<#tF`YVGD~-ghCh{i7nI
z-DQNGRzv@|KBJSl%9F<gvDIy?24j~uU+DTGxO*=%Pxk8aGrn)`6Y!`OQOwZh(OW)4
z&qJdf<;0>Q^e~c`HFfi4lG&~(v?My{_UPulz+f;&>H7q2|1-d_MHgwD<Cl&C=Pr;Z
zZCh7h(Y@KE%$K+Om}dPUc{x4Z2E+0F<ZpcuA@Ty^SfDO)s>h8Jgs+6p4UnbK!THNO
zj<=P`tEvQqujmkhb)xgEuf*1L6Xafwkfc({p6lEXFdyCCV^m*-Kf2XfOkz8sy7c|;
ze4WVa;2vkt;KFbBXz;!ckc|+fwXbl@YuGJtr+8L}vX>5p>Ye8{=P;(p6ST|;ahX%t
zRbU-!{2DKLjVk?4u(`}Dxk%FSbXJi`AfXFEi{EP!@DPnK)mgkE$9Hxy)9VI!C+OJ7
z@!2T^J#K(5;tGKBDFNjgOU-g8UXLNE<Duo>Wtxx{Q1>-KZYmDboBmVng1u_-Il@TQ
ze{#R|pGFgcA?KV}Su~CMag?%1|6muJTpr8&5@#wUPPym6=9dh&n}11mDV>b<jM46t
zKgeF3hY4YxsjH|Eb&cL6t!EsW_49Bmw?~kCL3!kkeya}|aycS;__W)18t*dDFzH{r
z95Gak_tfJf{scSYm2?U97T*ejDy0X8NqOXmfzI!05Q;(AC;{f~;5gm}PT_qcx`ikD
zLsmevNghfY$d8Ru`Qc69!7W$(Y)!b4oA2OD0izw6>QH)8jgEB~6U8r!sSlsrs5T8m
zs{*Md`FT`-@1B_E*=W5!(vrt@Q-Wx-4kFfoAitPt2fzJmIdm^J80z_k3GXgMxfZ~L
zW%|vTPL6eghVo|by5>nX3f3<gjehTaG7f~HF0TMPhfiSH3J{qPk;`&sJmy9x^ghW}
z70~Q}a!h{Ta|eC;WS>AcepPVEnP0{}4u>mC54}eLKg8vk&9(5?tJMW(j=iTrzA?oI
zjGbCF&ZhplQ)7DDvoZG&Ec4O5wukOdiKN-ObZ<zf#KHD71oIXVmSUaIwS}biov%bb
z`^Wn;`2|F0YPO7S?!5I5vq9n?Iek?+m-h8jC612IUwa0Axx1w%nw%97bcL1GFzc?l
zwl}%tZni*=%z}vmu6L^S*Tp9{Fa|-3kOW+md|)HgjX<TFGPsColvSc@#AAsK@;k}5
zW8WR8Y}T83EIFbh;Bm_+*t;R#tJ_ib?t=1!6=L`X(^1y+E-`Ya8!_`m<bo|@zT|6|
zSE=LzV&V_NJ@l9-1yLIV*$6RmrMqx_&AX6mMy)h*JXxWB6tWJDpfP?;w&|{oT!!s+
zk_u)$O?*2DoBS_eOdrg36+iRcq7WT*FK5A(|4dT<M&(6Yrzm+=#H8RTg&Bp0LM02%
zFX_q_?MQ_?X=D?I#105e9$5$5@ip6=-4Ik<!xHZ;NC#bO5jSfdPj~r+#<#=LuOJ?3
z4H*(#cIgkRhbdWf0g^zB${FLO8X$<hOOa*B5f${;0{Xnj&bz)!f~3Yv0bBUr>TwT^
zFB0*z92Rq36aAaz7F3&BVXwB#VrkMc&<v(?Lh8ckEY9QW4Vgyq_-=|BPk<bA?#uEb
z$3V?O9(QYm%ciE8TIVC3slJjQuc`bnnDpYjDfx4s8_~V&;$k01vIr!euO6M%#$1+`
zrbK=x%daJWNTd5jH{R_ru$lh5wi*R&3kt9AMwqeXm$av+N^IV+Vi1m1IGLZi{h}1T
zrCuC?d*@_FMFtznG&&!v>eUwq_`~$^XE4?bL%bs7V1Ox%f&RIi@JtKD%`e^{m=FtN
zx~JHh9vT~mZ}u}pgp4ybKPhKt3pbsMxcextxu&^v&zTDBXK+P6+XvWLiO68z+`jYN
zb;NI_V)BqP7B5Um-^zZB5QhqCo@OoWX`nYISG0W-qTQj<MP{+UrY$_x@yk|Ex@x)i
z&<*raObRTzExCXECL^<{1-$p&L*D@Ugli`$>*J(dg%{x@UMzHWEVfUVDH}gQzR6(k
zRF*PsR9;DL1$jvW1`r38a10KS6&}>A+_umW4=}oRo^#CItR*f6#;Mh(g_C37Fpk|Z
zpRrHh)ULPtDOt1$bPtR4a0HF~kOW-MetkAPWumB{)gL`i*V&O6AEII4Jnx(>D>6Vc
zyQRE*9#P4@oZT{+>uOImcNPNXOV$t^ux8}b!h2MpVRzhRX2zlTrQzLKh*}#%kYvJ!
z#3oh8SA~#!C3}cXWEVrKv>pv1Yoss131Vm9H**TsW}_V-%3W=qS-DlP)k+sA*P@(d
zt2HEsb}wDy&>3O8dj{A_!uGlgFug^0TDRnROOl19bzrmoq3>Pj3EzjKx0&egj59>Y
z7vMH%f=${XSt7gs7%Z-Okkr)v0YjrC;L^b2fb7Umq9tC<Gv({Uw<lwCm(?)8jCQ5r
zFimGtQ;|YoYv`H%26Rf34JVwyPi>%MQ^;%Y4Jje_2gOqWN!AF!+D9xo6kpK6o@#%O
zuJM%?Os&uwg9=KJ7QFmgPm9fC^yk*;ho!m4-u0Efh6?!?2dz6&BagrmMB+qUNZwl|
z^%+39N}SH%HH10(LA3?7QWi@Txn$AabqoYa{Gbdgjq!V4tZbmWJki^~IqRmm0|dzN
zM`l+DBM45!7e1EZHq_~oqr6!@YoR~V3$R2p`;MxI)${tFNxY5+uN8EG?Oec%zDBSM
z3C`GHCUJXcLN%idnWe_l>er@}q5QFQT`spFuEwpza=clg>J;5I)vuD#7tm0KK;?t%
zhkMAG5^QNnTpg{k{j<r#&g8GH_o;#IX(wOk85hh>>7C>GMf^ooY0SSk)BsyU2Q}*a
z#}9h%pxI+tNodt>7}`Y9Ya1Wa(LHjUU#R(jEQAQJPWtQEAZ*F}GW>~UhTPjhb$|rq
zz9R=WQDZd-Lid2J_>Glr*i`W7QELBjOU}0|YT<P65@J8;80gyit<ma^n_cRVj1r6-
zT{Mw8AOA(L*>=Lu^S6_LRG_f1cJVU)qtR|{nytmuSoTk~U=A!8Q;m*Q4=hTE-9Gn4
z27tDo?we>cH_45gh^OcFRyBCi@BHu`<hc3$;FHaVp{$`vdG&#Ni=4?y_is7^8qu#W
zD+Lng(IkPh@gra*e>l2*V10)7K;1f+(^pPz5P>>{1byWBO4qpl@{qucNe=ox)D6s$
ztLR~e(SwH8e<pniwIM_Q9cVExE`oVYZWCV@u;%S^Z><xr>f{f1)}o(_`buKE^xhiH
z(wNitwU?B=L%@<=_%MMJI7H)dXqwgLW_@k`kJDfxvfMBIZZeg#bh(!wsa@jb1>bVE
z(rPSN;x@>5_vT$lS$lcNyLV@2;hm>~j~_oCUw^%>a(`TYEhy&bdq~}sv<8+RMVpeJ
zD3TRc&${D;iXaDSP&u1yN-j1OQLAWOH2d2ITlD(}o*JmyX5f6RTs;~Wj`d|QGq-5+
zvLu~i1jS?Hxiw97b@dY7j0OnC70q1F@IAQc^6{RF?5&c#4icOyNhs4X(@vZS(m`Y!
ze^*&gX?vF#Q^#?pe8)f8k*b;grDgX+-5>LHdEJJd4~hjL6?kT77Vj$l=u>p*%Jbm4
zn-A$~Z?GQUNo<DMRZ?o&5QR6^cWnl`?V?S;9-%ef+oe4T%Y71&L{dZ2MT|4Yo-r<#
z{r0Y-4^*lTqH4CiF{NYsxR8qd3W$l0j*j(-5h;ktM))wU-t@Lqo$1XskGH?PR)W+t
ze!rJTdw*Sfn;i)9B*9*Dx<*u@PSb4H$W=UsjoqHb9ED0?Z`m&GeC1J7q|ebjKWZUg
zsBw|joWjRZWwfyF4R$x$I^E#bRf@7=u6r5tfP*wt!Fd2?t>VSTgi%eugF655?%%Zl
zkKQ?of>c<|s+q=O0!*;@uht>rc28!`IgaI{TB_fK8z^lh4J=yi2ZTKtw`AqXD2Jf=
zfWO;!btqnXf+IN3@tHRd^*)pD28e84$Ph6LH(w;}%4*pM5}&L}2!<~=vFqyXfik5`
z-s5)z>E(ocA}GB0niZ^TwU^&N5XI8ZF|${-mG^8_T^Zb0&AG6$_>I?y_Mq7r9|(5p
z^QT<d`oEZa^LHr2zi)iTn3?RdW|x#?EFt?^p`sMU6f%|wV=2+dHe-^lNLeab3Q5RJ
z_G~kjRI-E^`;gsW>|>qh(*1ny=Xmbt`v-jeFo!wjJg)0p-s|iAI?vO0tF&p#o!!#>
zZ$gtB-SW63W#ozA&a+-#N~GXTZ?X-$g?SC-F8V%skRomB8!x5B&d_4Ov3Xr@g}xhS
zMhy}orpHjb)ucoU`@1l7z$lpHb;HB;@0&<=P+W9$bW&jw7(1BX_BJjR0vD=WGA5-S
z+Y|&ne<Q%k{z6_i&<WRLcb#5QBaWtP=<0~}7!EeJwnr%Z4&Bk;W!E+Recf0NgQO+R
zOC~W5y$=?^L3Q#9F*DyKnrCM7`4QLlj!bY!GVSJ$uS)@V6*?V^v&B85Y~I%_pYyh*
zuLk3p^-sBSBx1SzqTXOPA!?*i@aO33CXKCRH=c(sqhhEfU(NFIqk1-S@4;B735kti
z@v>V`tO+3J?6yBaqz0;u*A$q^xywe^rM{%l5_2w-xJa2?j@{>fDkI;A$(Nru^gK$>
zY-fjE+8T~>_!!8#If&)Z+f}XVk~zX!DZvoeC{UsNIzAGAt@MKtewo|+G=DA_J7>jI
zR(<e;g#?{uZVFuoitela_+n{#HjW1pZnQfz>Vjk5J(Vm;+s)$OSn(yXFvSS{bdWH}
zOg8L-Zi*CemN%joexk#03d~P-%$$1Q!+m>2<1WUPFsn$bEouzYga;j2LYbEhP*C6a
zEEW5rU?^Zr=VPsYf{cph$ZlsxId@?g)=J*A-$|v(Lu7E`8(vmsM~hp&i95RweE%|;
z`STv3(atkBP9}|q>~_BHtqW6<2dJGzqca{)I)~-F@V;EdM*9AJPYD4!ZiQH9zF;@d
zecsPgbFIl-+%m+(Q90$2PgbT5gdyN}s(*S<*~Mhm76typMcwAO+fqaocA6I9=Sz6T
z9VSE}H3=XbgYUA_+31GmtJ0uyc9c^2EgAa-6=CF)beDA2-Ez+&FGSS}(+gSvh}DI|
zm*iQ*Md2ah3^-MeTrSdN+1WL750NloPP$K3;9S_0eqoyDOlHa_dROCg)7vIq{(O`{
z>z|vRR-n%#ij3~Nt&5DnSVV8c?=qp#4>>D<Y$dtf9^bcXhiSqE*Q9>*Gdn85@WNeA
zstzqgdVq;RVD<$4456E>mFvx+EnfvmJlH6Lak;&z6J7xBw09G^gHqKvq;jfb^^=m8
z5|61yZDg6QI9A)>NN3_tYcLA1Nr^{Ktv77$A$gbGDbTet@&*Ut_*yDw)t%%regLiE
zW7wfLtiKRmU%0#?gwoQi6ypXxa564-FX;LR<ImZy;!hhE5rx!@PAkVhl3G^hZbP2R
zQWvzAxbl;kc?HnyOQ{xVC56$`85SWou4`(1OpeOL6(Mv}FvL*_3TgP^69dTTdPJ}Q
z!$5;RobHmRPr6h%LuVs3p@Qe&V62HNds|&xx_RXK`CHhMPrpCH)YC9v>H=x|RM~+>
z<eW%W1*vefzT4sR(|Rp)69rYwge`)&CGXFlTCzd+ItvqEX~&OSmDKF%M`ivXBDqIy
z(^kK}6qmo7qxu%gk#PZSz+Fk1W-JDDvJItC^j=zUkDIFL-gAV2{^adD=A4)O2I=Nc
zOBG@aCAxOaJ<fPg{zbV@%?(xXfV(BkXNHS99q7NszMn$!hCY<D@%nq^OLxf{#GF9J
zZXr)en7Q#v-%QdHd|Re#vD(d3_zkw&_(RZai*d<ZUUv!G$R}o703sZS&%H|`Ackll
zd07%YWu>0}9Yz%|ywKYhHS~3i2yQLgJ5Fsu;?5+#`KzdCR!SNTD}Q~Q@@vJG{qh;<
zJF6%AO;`XQv{F^?mGUAzb@!J&1h(Fma7@>7BkhLgQ`ZBxHar%y<Mp=9RjJ9N$(r$i
z#d31wk0taoIfC~4k75DQV7B~cuC>1dH3PnR4uJgSBp%~IX+(0<F=B`P032-dGDvFa
zc|Nq$m1nhXuC&m}LN<e&C~Q7<;=ntYoIL*VfxMso@^!&$`jp7E$Xtz2?0R&4TEv7s
zwf@tfc;Yz5PIT|IYU^$qf#bs@@|rmzm!|skZK%|U#Rj4ckF#W8bfQ!OI%}J(;g?Bn
z78)<`e4Bkp)WuI0d-WG)U(AN52&~BZR&`ijLWdzXv2{|<F0;OZiSq`b&d_D(=H`QH
z;zM>+pN1}oV#9XXj5G<OYVu4D+y`5*%-+En(00ZH&?;^=bc(U$)b)IaXDFfNhQ7gT
z`{4&#@E#B&0#D$$!E=Wu%U;jsTU78DcWP`-O2zB0%o7IZJY~vR4SzXQbF`8b8h`aa
zosoJe63x~I`n=wmLJ8*(`WT~=O|}7xzp7;3x^e61#RJLb6?ibJT$fMXWxBt(*$+=?
zG&Iwf8fCYSgfgcS^QB5e(>VS9cKYw9Hh6<J6{va0!E8Kef%Kcg{w)+YPs;mD_*_!1
zbZF6srQ8WO>&MF+f_?6@_9%%Fwi_`GjFkuaeOTmtE4A7+NT79Rr;h@)P_ljg(Bu9&
zPe|QU#W5w>=+s`3t1nbX++H{u>zx5npc|jQ?=>qK6xcsW#6Ob^dtB5N4a1#qX|@he
z$NY3bUX|UgdllU#8Jgd4`7(Hvg_cRFd@rh)h?rx>0m>k)yVr5bJs=1sz$q@CM%leU
z_=@<N*U0WC6`kAb!7Yj5*A=dW4pld6R=*t*r8A)+htf{hv+mSdG}%-|76=qWnQ8?J
z&rc<=m^mpZ+B&yPN*4}y6F03oKXJ?dbZU~jgeIM^0hHmum;&11PCZIh#x6}@b`ywa
zQ&YL#Ie4gurcFX(M7|zpC{V+>#<t$vkUdSdDCs7=9sE(BaM>09CTTzKr#OwS0a-cL
z1Q{|5=71^2(yE&W_grLl(r2A6LO3UW$?T_BLlizD01dCl%rV96I~`M#_TlE{f(sp*
z5DtIZL!Fc>(<fEKOVWr&>%Qb%SHa{#w*5!6p-J_^#Wp?k1~bvsran~l2rDINC*VmY
zUVnt0#SxeLt{=<QX#wtqq+%%J>bkX0cMXuWk+ovj<H-e1M*%&ts=stmepw|&d2@+)
zr#V~<&4km=?+23>ND_DMI+m4LWxU6(K4FAFL9WWakcX+xk<%~2#W;u^ch<)57ImZZ
zulo_;45*M+^w-Wo#P^S8F;Xplf=0EiiTf>+Y2HeW=6m7GJ<L{tag(;4XyKN&AA?5A
z<d`-dt>CZ+9e5lMip-2h{q;dKcfKU4<*I+ZZ*N$+gWDk(_{jIll|O&&WC$~dt5<op
z{qknBfq>9=B3nEaOHmMI+hDtXxyjS~<~)J}F*dBRz$zu>^t|nIWNmM}Lg_eE9ty3d
zSBlz=*v*5}(kZ_(F5(xESf=!RMA6Bz{puuDw%BjC@eZ<P>n>{DVP=73N|M8_rUrbN
zRAf}j>muftn^5?Z-4iTp>!W^F5$g&@IneUrwwdD}nW+<`g-h^si2hOJ8Tx+HRGBN6
zIO8-anL^uAkqYG@94X-Ub9@B9I{cEf^rPn-Mp*2DF8dMOAMq;g-BQmol^-|lmKwH~
zhAHfE8miD#06^JPd8eaTs9O@l1419m7_mBtL)fna0d8j~xrnQJhQdEI7YWx$nJtt2
z=7vqb;6$$OCvEb3C&Gk(@iI^Lt2OK5?8Ivp$7ILl6#}g=?~FxGaQDLVS(ReZBRuAt
ze!i}QE=U#G-4TjrWu^l3yae4P2JW!)<5zjo+Eu2b);g05eP}j3p7q0t^1b%veaR4h
zp<Tlmi-ki8Z6CwIm1?zQX00SSVn5FbFBVGKF?eBNxXDJ>#0Hn>Tz=|L=pSUSEg_0>
z6j7!(BoIcmJQ|1SF7EjeIS<Tu>unsQh7LxQb0L=#B>r@=Vn}dkiQDJeU+nVT6w{s)
z<G5O!`q?);@(~p*{<bb9t?xC!nmQB#R?#aKdi(Z)UqKgRrilVPzsH};1tAwAZe|!~
zjU6wP_M6f3Rl6{Mn(+MXUcHdLJ~YKEeVf_!k?Cs>=gTAyaUi5-*hW}>PRnX2-xET|
z>+mb=>Yn)GE!bNzWFr>Tl@4cuHZm-m!Zi0N+LjLpJ8#svSWx-N@lCKvTc<wjRl;bf
zX5g>lb$YkXLO~SH7s`DZZGRntzjfeZ@~>Yy6?2>v4C|jh%dYHa!{tqh*-1v~;ETbI
zuzir0(B0zO-RH+^pnJtKx(p_aek@hqCJ40r2OLWBPxgJ6B=hP=kR0L6awohNdi7ia
zucdGg*q%18rilR}t`{<Z$z~VGd7x`Z+`0jG!0W9Gj2rJ6P{~aN9pwG67E=u$GE286
zZa&vdD}n2Ir2LW%$cu0bw2E`1`3G?^L~eLEN0b<Beg6JIkg{ZmRz`|ud4=vCxj6QL
z-!O0Cyuz)6c;%AK3_Dig7}FP~Bv&$|^}|W{xX0UoA#eyk!(UZyrfLn?;93_y$q>ko
zmHr-lxIeQZ6+2pqK{0Zg49wGWK`)Z-6IvP-Npe3z<N{FZpfFliA6dwkoVhAvsei~$
z@TyrpHbhNDlXz{5Szrk`PLC_cGdM<EJcd_RG_rS}Hcdf|FTe6D4#n0VGq^eg(PJF~
zDOH`ygK6>XYf`4ycoFM%hAMEo;}ws8Nt4#8BcE){&vDxXV??#IDW&3e0L=H=xYLi9
z>aG+_aD<f`E|d8unZgI2DB>&}<a1?+Y2hdKy)Vx9ib<J^aW|eIVrAG6o8wJZdV7ly
zmyag`wZA8dMHvYZAS-<{KFh-p0*K+QF6?(zO8T{dNprXt7GI$^#1kU@B}mi;w30#H
zXF+^aQTWmU*Q;5iy|0w~B~u+1d4JC)b2~gyCdeH8(q3`(Ej|FEm&aC_#Hb3AOYt=r
zDiCBa3H`sXC4dcCH7Rou0{)>?-{vlfc9;#E=NGv>vLWlyEAEeTco0%CiN33VaMR-v
z`^7!q_f+mB_XRvCGOpK<1tUhW&n-XjM?E#b;69k_15$OIx&hu;I`J|^R<Z7JdX^)J
zRE5ce_%tc)X0~)eLbAt;_%9w}IV3C#9E!H+-yeB2;mT0Bf>|BvVlfm+`b+p;&aYp1
zFJ-4{w6Pf1;Z*g49PuU<!w8D6kGhp~Y%k&d-Ulo&vhj4YscKQQ^z1w14F6NlE$*%A
zR#cG6eZh6H(8lZ(5ECi2Ze`EIa0F$NV!_4SlCSXT(+KXxa6PUwz$wIAHfX_7fpI*j
zn~-&evNhDAx0^SoD%hauRNr{=zZc94R2VlD7BQJ~Ta%>dEzAUEvMgrN&xXX|d?9@|
z3Qc5!Sk-`YFESFo3RmiPH2<w^3T@jjK57hNayxrSf;i|MoE!a5iv*%|Kw8yT=D^}_
zD~s0*XS0B~`OpYVjmI1vM(RiJhjhGVd7DhxWe;8*-*KZYF=$N&^Vnj1>&`$~%K|uz
z8+}Lvh>WE+!SP_z7Z>5`C8CN?**$fJE}WQj);grZE!K4Ca)Vo0EA@m^X3SNd@>uf&
z^LFX_rNdOi1~;(<{NyQ#nUoXOC+o5%PL0cD$uv%0N1I}LLY4fD->&Js#^G+{R3FNl
z;Qh5y^O^y&6r;2GYA1p-j;kG0U6U+4=_fxrEkBlr65f3mraH?ntACKzlLiZ!6F+M;
zMY3+kRlE^MJzRX5C|21DiwfnZA4VwNc#ye3>LdLz7p**0!NE99(4Y|5#aL%ZFz|^)
z5!P49tH+;Qdmg(ajLZ^6^HT2=!i*jW{&`(T|I_!31{#kfRAl?J9E6@n<&UF!*oR;=
zQbZzTxNG7l_0+ie7x*}#-zI^(Jo|!n&?}1#3sN6}>U*U<qJhc2Ylt}hH04!ambBrT
zFMpaHwm?r@_@?~xU#lSq`e8@Lrrm(7^{v9fVv(YNbASnuBws3w`=PWO+#&5T(9XJH
zj;?Mcw@Wj^wS{)21cO2}>E3?6oVAl4lRCJ0VlpoZ4=NybERT^6Q&1Css5~D{IkLrT
zB6A_^9rb#Fi-FRwjj==u=m{qcIj$hq*E4OD3aH3;y8IoXpQ*&SM0dj3N40mnuTAc9
z(OGOjbROcJKYQ96{>abNaPSpK`;pJESk{tC!wbNubYZQlKIDj6^UmBr{is~6Xl@p5
zbP7|ekn)!cN$g5Y4CMwJERKD{SzXRa@NKyNc?n+|<wkobJ)6DnC|hs+L-JS1$ek6E
zJ6jj@cpOwYkbt<b;}1>Qkkc^&5<1Z057kobNkC81h&KdVTp%Dxij<v)IyW+g3%Hyc
z^Wja<xCCOe{KZlu(y=#^ixwrnI09>@)8Dri3OzqEdOq-zY%cR(p#2`e?)DF14r@gY
z33qyw@iLAGX$6D5cC2X5@A>U)^!TeWUfhyQSs^sPh>{$TGY&4^)A(y6YNXP+8eQQM
zc6#}y-x4`Oy2C|{%P_j_Bxf6|fdnT@&S~Lbr4r(%pOy+sntl+#UcW(C4glnDf~rW`
zscV|za_#euF>q(09~~cE*2M)!P?=F!{+K?qeu!RlQIiMn=h1CQISPMt>{HYpasJgg
z!s>@m+#YEZ`q{HU8gGnEBa~ilLk8zL<KxDtyZ1=le>|M#+GH{`@`ZJ$4~CX_^yD27
zik<G;)7u<AV@j2ziqe{P&!eY2KJp;3(B98^fh>2e;CdFa)uho@WY#6or+8sKT0?}<
z@1Enwu-(M|4+_Mk@#Vx8IcGdbQ`h|U^FT9a9N{o|PdfBz1XDf-SSqre!L~v)d_@=M
zSV?=}J?1|^sBesCxcNL}_)ss=s+SL`&gyKU@<GM$K8Ie-#7FHAOKiBH7z4H=`6g~}
zWRq;rQGz>1VuI}d*qqnF{ApAsx^SjQ-aVnh!{L^>=p7Jk19sI-#(XV!Cg9k3;A_wG
zUCIa3k1RVsD|1tz>m#tePXrH(9o`*`P~2hKK3G<6Jo#N>*LWO%Kb<s;&)rYlo<wO2
zJsnE_J9YwX>ZIGRaNZQuhe^Bh%DGOEBO2HqQ6%?r^=6j|pi_@4&gO<=_kdEQe*wNN
zDB~~T?fws$ZXT^&F(f@)QfQaQv4Y?3$9;~<hpn63p3ercg77+F1@`L4m{VU)Bfn-b
zIrekLd>x<Q1(7<^JcVSKNaHrtg<yOkW;vdggjB}125`$Bggqv};{1cQU-mufeAf2K
zlhA+tg{mB?H6uH0KB>EYOdH9u3*xAo`nh1pcjQ@Ug2@@D07!fhnj>n07qMrFR4)<Z
z=$cZ*DBj2g)`6<1%*=}s-h(pIOa9#4dK&@V%udN@ZsZ^cIz~C;tH-7b`ORzBd9^0b
zwyfjrKp|2L{ghsXe$6Y8I<Ecj2r129Ttlto0#YNzf>{yKmQ5_Zk{7&F@A|93=HBCN
z!f57!zvR3o4*|tty>QKx$e)~uCAZIM(VPu~?vq1JQ-U97UbBkR9;9q;IbwB*$Cock
zP%Ij!nl`KzhHBnv@p1G#&R6SoG2vSkYD@YAfRYU(<F$Q>`bWxn&5tK91+9KxTpHce
zU#eZ*tl3nhs?hw;oWY<qg?1fd*TG*>2iX@jsF^rN)EO3MUC=bh5FGBKQe7h|*=SQ%
zbiCx2lOYvEZLBMOP`EyK9c^{)E~Iz#pv&KRlkazJDN>8^%u&_vb$@jFh4R@(BFb!e
zJC;YFL-{?1F9aqy!ba?jX2ieXcevBqhR#pzXy(pZ?4CpmP4s;KwoWPmK(AYK%-k_(
z!6zk|V7@j5&gbPFgTJO)W5P8y3S9=;+?VN*NY7WHM3M@naKnUxo$=OKb@)=X4#ZmF
zJRy==F^naQ9{mRO2%X91vqLE7u8DGVneFaH1#pmHiGeIqpheMkjog<HNg^@NI&kKn
zeFP3vm@i2c<fmJ94QZ7l^!BnIh&^T7JzDr<9VE>}OVaM_a2zxtm>?24R%Ax*9r#Oo
zH8QRE%1?gz?29(aXvZx?C4GioL6DAe(IzyKiybJa<`<vJF2DdAD{Aw+8n|`0R$KBm
zHh2GrNY3Vl!zR3`qk5<MZoRjtZ+g5SK>}aI>*w0FkA{bVm29^)WA1s8n8*Eu=q*jw
zYGo|b5O@SrsnTh|U7W@i%aQ>82KDx8*d|}$nSUz?k7wJT;)ZUtxHT%@tFi|7;%6y4
zoa<no8~KU{;vbL)p!JdD%jchv1|zTBhi$6T-q%xX{V9cC&Jn~vS)WA1e3ThJs(8pS
zQaaFVqMQv`EQ6l?OBLh+@P}6nBda5>mECvxU>D-HT!onqhpuc<y3P0M)pMkIgMl-<
z?ev8h;g7J>@lXTPkSrmprUdOZSo~S8{_6y2<R)qwoK7mD-p>tr`SvYbl7)$&=TT6X
zF-3m~H{3^A!ujb1V>Dl&Qb1QIq%-chvOn*De630MExO;yU&yV`r(hv@IX^I?BKd?e
z>rQVsYrI-axT!`K8~9*ocd~0<%G{tXI5Snpjkgz)lD1nCnz|1EQg|n?D6p$ftpG2f
zB55`(zFzA_>)Mp@u{_cr(y#C5x^Z55L(jX6vD!!!Voix|AMi#-=7k;a)7q!BhT2+^
zH}`ExYLz#VZg^9HY8>U;5Rn0VsZX|`*K<_z55<0|iFxuj5X;o1icm|MY#|mP12%D%
z{rU)jvyI)SkW*ua5%h4xM#ULPbNJddghQtdtp;I>qYBZM=!a0*QI1s>_r^q1)b=_`
z-tbBT1F<r|L-$QtnH$gL?a#<`)K&G&tKLX_0fmk#9hc;#lNOql7dGIm#mVz$PkTzy
z-anMxxViX^g>#rvSTeF}IqZe@(`7YxC>7hiM7@C$sjM3xP?{V^Z%{jkM%{%611`|F
zU!o!NE|xAf>yldE%)UAGPs{M@sV?AwkNTu%?Og<}iL<Et&Vy<av>>=SThCJ{(+rha
zW<vI2O1`xj<cwJ*g;E^>VC?dXL-QJ?F8L7~a3%!Nx9fNLE{1nL&CFRIOWIyoAeo|#
zvUqn`GS~bpKd%{|aT8r$Gu$KX-S9jp9$j#_B&1?K@MMyWnJqrZUKMlUanv<Z3-Qj(
zHA^j|Lz6-omI?CXs=@;TU*;s7QbGVn1tQM;dVqKyE>PaIdl$XI_}DN=(7ZzyBxObb
z^+~(K9jU7KHL_MmUEw)yAgSg2Tg#JXPT=uLAhor{xfcuoMQNroSF4>m<0K+AM3qqw
zVf;Xm!+rPWuEn&qo*<4jD-9GkMKD(h1~pR1r=iM9O>QyfH!J&a^(>1`R^p4HAP${y
zNTpE7X~^5$if{4jMWp#A<`eE?b*Gei5{gv&U;HJyQJr~Ga|#la2X0cfJZBsb$3Glc
z#iSo7xwuizjhnpv8^RQ}yl4S(qbxe+;r+nAK~kehP1i{vTAjUpHjNP1G)sBf-~;2f
zQg?yi+7bs(7GltH8mkXwQXK0bVn%n_HaM6IFb(*EHveuniA7LRJS-*|RC0_*0%X59
z)*E7SDylypNx$RZ7Y^4rB0al&xrb>ES?Qg2I3Hqq|E^m3d(V04;?rztds6F*CPM*C
z(~VkgjzG!iJwyUjw*pcOekfmiyQ*rzW;fNAaPi3RH`>|69|Ce%rlw!9P0blXcVio6
z?w3AglzaiP#N?yteHLzAXXfCGj3r|GNz1Nz**Zd$jz(V&WbzBuU~4B6eXh^Xlk*Q$
z)hg?3?BkzwU6wR9p*b_-)HhR#py1>ECLm2`?STqG#wBY)DfllR&(r1zcrv@Fx|48T
zHu3t|zd!V`lxJLL3!74W4Y@uSob`U8SHsm45`jr{AfV2&1)TW#BY|cADTY036c_yI
z5|`+AdD6?o9ZUG&`BNr);mH`g<C~JIRsAnVzIE->#3ur7;W~IN_qt2oXd4b>-k5eV
z9(qj)0wX(_m(fORd2?zCZA&~MV67*#ze8+8uw71j%Vs-glU(I!v?qXoje;#F%{gWl
zg2(YqJHI<c0lH<`z)3!uC8W+|3b_UeXYuh^A7BNl1%OM@v2JZOF|q0xY#s{6Zobf`
zma|t@(QBCdDC#-al#7}!;3sTIIx35!c9A+=#Gaukniix1P0QDM6^580@^e^KQqP=l
zQJZbxs3@oIUq@^21W}j38P`r%jPVz;GJL5z2ur^iB-(Av<%Y*A?MO_gv)rkD1sOKU
zjmC&leuDAJB<Ykbvh9s?So;x#;K;oR1LehLs};T3tmn5hrKEM2nEE><b0-lD05_vE
z+o<JSmDkA3F{%xY2`iUbwW8UdRA;Fa)$-QzglJ(-93dFWQF;>6`7`1#^9IRxtfpYh
zU#=cTuN(Cs(1KK<%!`BN90g@1S`O>j5PprDM|F5hMD6kqP+OThIbiJAX%_Q5KT&+2
z11QZy@l5-m$>bzyNSvo0G~|F(A?qUU{H&sQ#q;=#0fe9l?&vj?+C^cA*NX7$`d=)b
z@RhYBX4<K3@FUgT7d}E-`DH1$pRRiP?xr5<jtVbbKx+~%1>+73ssuA+munjGf-p5Y
zYLLB?=x;7P{Eer(t#U5)Folo4(l_KW_kUsG|4^@U!TZ~;4E2w5T}usABo9`s%Ql~V
zf{zCzSJbpU*9P5x>O#Oh{@A?mVBHCy=|UvmBakV{$U81$HUHrPu)ku$DzZrxGUuiO
z@yx>-j9qQouAI(AmvE{sNsU?$Uo46(W0hPLN-{M3QpHx;(`eUjb7Sh!NVQ~GbP=l-
zqD5(}%DKmwq}<v`yRQxLvi=S+><F(rbuekUT_6&j;YhNLqKD*qemZO%-@A;I-V^@<
z;gB?cwtC$O$1xf|S)k9KcI^mPdwnGI?2F=(!wMh29vaFZhd&t<r7_VeVqSV0tM^Z7
z1lL4a5CFsQvQVLN-MOTVyIIL-+L;XSw3qH7jV3O5mDCF}o_Vrf&%k?=rAYy<?0QRu
z&bP4E(M&F{jrtzlbtd!>3sVlImI)VS^=O10ngSO;;%J|?cqhEpGUi!&)0M}H$5nwY
zO~~n7GWRw7`@U(aO>X6v+Hh&8c(CItk-8{}SPED6<F9<kflf_&wn~blG6m8uh6`$-
zggsfVz~W)yaw!=ew!N)M*~baeM_4N#|8lQm(@p)Q4dT`FwY+MM(%9&~C{%dm=b59q
zmqPgc6qlDO1wVogAf^i@cR?u3FOl(8F&WTvH)Clhi*XJ19nQ#RKQkw_u~i_o1y|^%
zEeqF8xgzeWaBzHGVqJW|xI#rB81;^U{unMXP_$1l6azy#Y`+NTOK?E@G}s6UIOh+C
zsA;^MN260VImbs&7g5~D{7HAnwCdD_(2rE7TrJAXSPSKz$G3=EH>hUs`<Y#sDbD4s
zfMF)QWF?^)AVw>#%)%l5HKBR>aqF7i<azNYlNU3uPIw*M%y1i_GRy-SfqgYRtCxJS
zW7r<kSMt_RBrwd+wewq8;V6v^YqS;YH<eeYfHOoGW>{qxy!!7B2e?h+Y%MvcHI!Us
z6=VnI0P@sUVtw`%ghV^4W0f3MPzyi?5<5E`>%+tv`vpb(CO<Q5jg+9i;|L^nIjVEp
z4XP$rSS4lDAQi}r1}pdlC?@S%fI)wdP24h!?uAe}4BK_*%k=){W2&)-?N}AuKg}s5
z%TcrLjKeFE_ZCQTq*W?ALe}16#7Q_{8YTSx2M*iGLSuO-m-1mo+b_ud>&2qgrN({9
zv%0u&%PkN#7WDIKKC`3I4#W)k(w%g^Hhw+G8h7ry;RUMv-rI5>3P_I!fhhlmtuT(V
zbUG8T@hJ62g{9NpGE(Pt_K(s%L4?%KIjW>0vCZo6h!cJ?^Jwr2lrq|HMGdB&;<MoK
zwE6|^)v}AR`swumg||9JT*^qKY>mg>@9RAI%=ep(+?6=AJ!i;i6aKRRJa7<-{08((
z1T1diB)sAeFrTg(F5kSP${E4&anq%QKZ>rZmmpEUH34u)&Eu-^Jc|!?QzaQJEeS!9
z=lm7bMkD`lBcBw>U?eC>@8fdCcdsKr8(`YfSA7ghQ^Q5Ae<PW^VXJVu@ye&>1W27x
zh|&B*YQ`rkJbNw(n}pwHMoW`|460UB@NE%c+!%v5Og-coO^Nwb_hiQ33Wyi6p<0aC
zH#$e#v@gxNEJg03F8ncX%`RGd+7ci9;CRCzselqhc9>nP7kj-#ie+^)UL;pbqMIi>
zNVPg|6$^Owt*9OE#Ryd9_^C4|h1Z#YMX|C?V!6wB;4GIANxA*w$#<%rI9AX`%(|hw
znn-PUg#I>q&9!4<)a#7D&VIUbo89-_SzlGJwLuH>_J~b@P-cnEmCrn#PVEIh#MX;x
ziF0iKf{q-2czY5OF(^d01ao-J?_K(C=eM+G@McHpZ+jJoR{mdy6HA<uHjCjNvUAM-
zv}vXmqShfe-r*53YwmA5EXF0`x}5K$d5#1~d)uncwNT{k<y2xaX7HyyWHMAXf=T*v
z_<_zBJkt(eUI<#Wp&PahNm5(LRjtAVnUUYP%3jo<JT<dr4kGYIgGk<*MJ_TwPFoA@
z9$@PEx}qT77`n!}RrGlNB^jPo?$;5Ox<pjCvXJWxx$6*`l(LiANNVQX;!1tG2hPmg
zR$U#?B=~Mh>hx||T+I^ZK<=_&2<3R_XJMqPrXh=Y!GpVW-_M|S43z~jy~skn>30)+
z^rD<^%i1}kRR??ZT)ci%`j2C4oddP6+3c=3lZTlkTRF&lcVfKJ`BO)&qBoIfLHo~3
zJ7H+k-RySuHlCA*sD{(tC-!!TJw`>8hZlu=m3>Jzt+nHYRG<DJg<B!FoL6<NZ&rLw
zO^G8?!(WZu)g-VM4Ex8~RQ_SFkYvCZE<_v_!l$e56b_feI<gIM4e@hIUl1~6W7r||
zuYHta)c^5FI53faJ3-hBB)d0H6bM-lx=g9IO1CkZeO{%IR43v6DjLY9zO6K1X}S)Q
zdY@rmI75Anq_WVQXB5%jX1u8*^4o1WZ|0ls*DAqABZ@e+zyLd0oi9!Ckq#r0WKAuf
zpeR$*h@+wWHk?7&(A<at+y^cT&)R1S8(|r@W<7Kh-3RL1`lg%}jJIoYn~;YPff~^#
zWCBQybx7lG4+w2T$S)Ni&|Ftnf@9d>alw}^|FvfKg)S*yK#v5;bkwYI?>Ic>3Vdwb
z?gL=+I#g4?;l@8#<b^j`7={h)O9q@4KZ)I7k}%E^7tT^CjD&qwXsP4+Zpl&?`==@T
z(N&M615E3bB?Kgctr!orO>}tL>7s2UC5is<CF%zya;#2Ua)fPH5;;TPVX-v@@YdMq
zM5MuZE=ixHwr*&yQm%>;)!_A$?bj~0Bi>?>!-k7s5Eqn_9za!;{rh}feYfEiA82i<
zZ^hA@^cfz;uKQ*uQFHGGN9|tbq+zP&o$&{4$kVjfq^c25d%bB@lg}nOfex4Uso}zO
z>5U*MD?`n<xNwlKg6L=2+hzsR_KV?LH?LG45_`<u4c~B4V~&D8!&3|D5_(wsIEGmE
zHA#^a;PUUY<ZME+ygS5;Z1^m_PZ3s8)K$c3h0KWJ($gP9?A=J-`azO%E^x`d4c_-)
z3?2jQKo>?G*apxs$hpSD;vEwe$fBuxMBR_J5W&O)SGNE`CMe?phDRJP*yJ5!m~KKj
zg>Obu!@byibH7zp5Fdegz;fUgm&jKxx=!GI08`Yg>{H_jifa4$bU?E*-C1pAXiw0y
zMpkRrF1vqu+oIPe{Z}>rw~{-%TSL`TAe+Fcpr3~j>EH4tZ#P;+6#AvyHb3ua`B?x{
zP1s+|KijpJ^Q;29WES&Qd9Aw%IP7%;Jn~Q@P^mMcKRp#7wsKu^u~oJg_S0qP5{!1D
z2!h@h6)EV=*T`i`88sE$!J?q>s&`oxl}TkvEpG%>S6Jl}U{h&0Yuu)@2kHIvcV8B!
zor3IvG(3rdfpVr;JpxVp!_~Gn<1S0_<#1u70`I+8JK<-26da4vpiZ)S9FpmUXzGT*
z{my6{LvspEB*2U{St@!RKV@1~Nsu5^sYZ7F42!_iQ!0NfT|uNEx<YgjC|dgmO!zI!
zPr+#kKQg!JLq(U7uXldg^<C=N&h#7hC3du~Q8koWUN++1;;|CO^zq9y|1Zrq2eRB~
zzRrjZeL{Ob+U)Sf1o(H(&&4FG(AryZ_(#Ty_`gK#(un`G_fzjJS;LY0a49`#feCL0
ze9eW-6ed#Z?>F>G^hFi**1&h)QP`WhiDcSGAryDHs2ktL!!OX?mn05BExC&sagtEl
z(ik^fp>?+eMs)|Jq6%Ex4?8;TXwV(0FpC2DgBWx&8w|>SBOR?Nnf6X!l!Nbag$-><
zt)>jR^YrM$!^SnxL&7Y|I%#*bIy)pI6~{7P{vDep_FF{?$)%faK>On4R@baKv}V!h
zsXixHC3bfdnU_E%>HlPV8#m9A!5IsZ!g2Lj#L(z&@p8jYa&KvIR#Tr!re0>>=Sf?Z
zjon@E+@8OOwu2Y4Yd#uW!l2zys&`p7UMC1mEN<FZ_zNtri7jCmF|9qnc@t0h6A&kR
z1`UTi9{v0=c&XGpnp0`%UG}Wd=eHY7Mrm-8nj1^WG0WM;u6;e8p&XI+XQWzSw-+Ok
z)Jy?(gE!fdOThqyJ6i0tH~NDr1-dQ&Hx^q>AV=Rl)qI+CW~4n)>Ih9d=x&IvA9qUE
zX#|U;!_j38;En>!<iG)pBXdy{^gMPY@SVSb(8N;w&;T(m)0G->as%29ZQn?-U4x<m
z9+bbUz(od8nlIkB36cV!v(Mi%tNGG?8?wxVK7&YUMGOALbvQ#f{iERu!NgJWFjYu!
zt3>qqe4{68&0~l2?Jx}#IK_TPiIBELx<A%Vc8f9Ou4CiSuxzw>`oO8>J%2$x!C3oE
ztFr<@<Q4BwxU6m46X`zY-pHJbIPOo-D;?6jM&d<p-ZDpe&7WM&o9BNPMdQ|b*jSim
zE_jB^j>q^7Zpn2PAjWuA;-=&>G9{-S0MTgF>A@GR>`3^9@AWMwYE_6;vh5ZVfsC4_
zhdK4S&PzRo`#!%A=cq6evTsa%_KusC72=ylOwNDF&m#p9pBi(b_OEdN3X0<Lv&Yt&
z5jG`1xB;Q9!3fBw7{xAoP(}Gc=BkUkPCiV5x_->0z5jTIY{#o-n^-#b1zt@>9xnZR
z_kM@^Nc`kZ$C-kVtCAbDAG~|7Aw+lBqcXTHn%v$cT&8_J%n`WHH}uUaEQK@DkYCH|
zdHPoZ7`VMt74^)`FQp-Y#IEBDt*D{8_fvn}BZYjKzdzB<5z^N`UHSG3ITi~%sXXTT
zrQtPZ*-txW=!)rv{=Oi0I$g(qGw2zlrG`wqm;UtbLUxt!>DAO?YSu#{Y4Ir8dK0Bq
z=*zO{TsENWF+NEA6PM@m^7aQO<E@#A)GKId6eO^2oAnOzFxC4--%WByM?meSl7`dx
zA6%CMY_|BD^K|FSh(T9X`OxiARNx?UEP*HN%}cwCsR_Z_(lwU?2OiFiAb1+01g^_;
zFy7Ap2QGYlF9@>oNM?x5AM+=6ML^94e$>9nf^d8kKTVo1g82m4tX0Tz8?slcyb>v$
zy*my?KSXWKEtRAxe7cQGYFjgw^lkCve%$gIP66+8p_|Diq*k4<NWXNX-@8ipo6-~B
zzG`=;omvoxDLtJp^V}#gjqc@sc2=fIjuS%zH8(jFw!(DnpY&Dprg59f#dVg%UDM4R
zPs+7BBGtRxkA<3Qx<E8F@d_R?xSnHuP(9ADqvdDU{b=@n+!~k(!bN#FYehD3;CcbM
z9=i^3&Xe@B2}ZkGXlMbDS~d>apvN0T{f5bNX*;!o@9J3R&t;!*nT@5>g)|tv8H0}(
z-o);OPQTG=6#7=N`?kPczgb@Gk%z~dpE^?|m0m|%cxzg&Ah;+(w4uEZqHOyI&|{AT
z7wAInaGV21*1aLB+@&>3^X0{su4fL9*ZAaD;-E7Xkm~f@2mK1)XTt?F{4vfr?^TjD
z(xB%k0>8nJoBz)Gti^P{9$rnc<JuJ+D#&Gu%j1P{csM4+aU7yqgcMC<qjYU~VI8Qp
z&g6_dtnH1)ce@eq>y%-dDBnx*+$eX&JfXvhrF4Em(s6&>hoYxx^>5F8X+OH{F&7>;
z>ucDuTnsNsBe!1`0*!uyC1@%X#Y{|<vs7TiVFf&$&x*9t&Gl49anPa7#uLZZobkhh
z`0$6$)5%+JVQV<Yxb)!959l9S_TYYeekZEt?_PGNv+)HlxVd(>g<!VvZ`G>fHfsA`
z=pGr~=YBcu1QGIOt=OlA?(z+JMecJR=%9SDsRZZ{$32O&A1>Hx61dakO9TX8A~@g*
z`nzI**Z37z26@m*w!ho6vH(bp<%>U|=NQ}t0=bw@QQezP=Q7~@BR+2Y*MWj<P9xo-
z`Ob0s9WPdSe*a!8DO_0BZGkfCtgBHH!q*>g^u9>sUZ@0H2q%Sm{|w@4?GX%@{mA5+
zm5w_u0T&@9gTKafd@n6{0p2;UsupS8{t%@QV~SGCsbH<p2QDoMQT4b)`C(9f;)}-;
z$Vk)M!Rz(b#(nOa6>-;wDWl&+5rit)L$m>F6WHVkDC|&}j*aX?T@8;rLRE6brKtSC
zZ)W9@{LX<+;R#Ol0K#}lTXj9J9i}6(50HZ7^#b}HZjJvoBTQs*ZwQ=Oz9V5hC^@|r
z6Q|`lJ`dc1VBR{?<LXeVS?n^$Shw&PY|0TtZZ{D!ta#`ge#Ja%+XgS)KFU{WqdQ%^
z6@?EUWFi;iObfuB|7c3E_WVJf5MhQ6p?an5=X-I|9qIy$vyRZrYn;PLMQBR$!wo{_
z(?2?gta%k%U`IZ<;6cTnHmD+c8-?@hrnB+8P$J0bv<`9TYXRZTy023!V-qDjy+^LE
zrEIqPWGM`XAFC7D3AJFg!46)XgmDQ^f<8;lG0k}=4rk&8SM*+4NUMH*w4{jgsDiD<
z-L+HGkk(~aAc&5bqO@Ctb$82<V~iX>n-zef;3r4;^lUMG^`@RbYx;uNtJ1B2DW!#X
z>kHKNUP}Ebn!@C`nX^YY_qRFIHrNLh;7SSllCHhUU>tP38vFS-+Urg0o8JB!15eTU
z0bNRW-&x;q{=Jd0P~hY5kdJ9O&CPXQ$+YUS#ZU?UcXhd%Tc5_a#gPWaqR+s$;2{Bo
z;<bm$iV8$>&sDjU4fl0yJ=Cr?_+CboBiQHC=UoXHKsf`zjL+&_AS2ztYoWY>XE(E{
z`$DuFt+#jlRp{5KXVF`<kSB(Ro__)v4X~E*rV+y%>w)U@Mkfa0&}x9>4g5{FS7+oS
zu@5e*G@5&6_%`gPG}P3Xu`-ndjX*TD8V-PxBJ*<blhn=*t)<Srp=FbGN0|)cYv@5%
zlh<*EN3<l(uazob7L`<kfJ5c#KR?DPjYfoqnYxTd<)I2BPM|WYxk$p~TI*e}K5-~$
z4#@mvf_7@^;jK4sUpT$!Q#jmFg#i(U7IfG+gyN6Ztthn$eKXX;wM}!by!cBP4dU3S
z4bm^azjPUO&SUhQ3~&r2r71B8`IGz_={OX>zFzC<w)4<wlxr-=Z%tgwy>36`){KW_
zV|>ct2VLhKefjtF+txO9LOOg*1AW{x+Z~?HMSSpsUJa%%eWy-uIOAP>`7-nCc`K!c
zvPR!Y)?tepeogDP!}y=P+&-Xo=PNu*Vwo|gG&7;d+uI=-|3%e;*TD{`S_Owku7p{V
zl~fu?xh)a0X~!yxD14LY$hyTkzVO@pev*O4J@L@9v20Ih5WU+CmfSEQ_Vnb1-fQHn
zsO$K7xCX2u>s66&quXqHG#5L~(h0(H8EF){hSEKOp#`6Oal2vDDIc=^ofY!l80Xh3
zMQy}AxlJn>IC~-?40Cie0xCQZ;hinB=^xsA0+I)*5ZbzZ{u2QF<79Bm?C*u13g^Xy
zUKzFz#^#w&K4HJ)ep^mCF}ZtoQ>r6H1nJZbfm6H{0^>Vb*Ra*07oG{ozn&-0nH0em
z+1k&^J^#@mB{b2dDP3W}sK8qcWGy4ZGi;TMLepOJ>YMq>y!@<YE82F|maf*vXUVpq
z<R@Ht2C_Hwm}^1`M0uRv0=OKYC=SRbDG#fAgfVXDdD@1upd*R^Jx{1ulbW6k!mvUF
zM%Q2O=IFb+8ECqM<J%~phL`PDYv<cq+;)>a-7<tOH>BLGF!jlfaa0e$hqDf;onvfS
z+1hIgQEs_zh9Y^WtmqB81=`KC%Ene3&8r}KUaXZ`#;*5T-L_3Y_`8^evQQRpzTg}a
zPF3uD$O~LncLWmdxR&^%BfX;(VB+~I$fr2tS;V4Vh52ToT$SR8Ye&sc<Zgnmlv&`_
z2RwdE-hmTVZtX7QrOPDq@-|(Xy{<(y|H!u&B0Zaw(N!JQX<mN95B7z`y;yzAWy+~W
z!sQ?2+5pz|GIXx<wMF*Z{_KIFyd&jK>t8RuIscE5A@WQeqypst3Ao&#>zzKwCi6rs
zepQ6We$lC{Re&4cV8z2ZkxRS0UyF?2B$u)XW5e0rvL{^P_7S$hY|B?K*#T9}M5NNF
z-?xz?<B<yu>NaCk>$)3?!^k%;KBiMRtA9YwJT-RPOU?r7Q=z&Vy0>y{J~BJVc0SCP
z1d$a;s3Q$@8)}0B?*WAtt-}Sy{_0h>n%W5O0Wf(l%-6WCjO3!UQ|+xX){&VO#1Y(`
zxnMD5ql1)c(sJbV8S8FHk{@Gz?3QIc8z#H=R<1<#G>_B<bHSBme{8PqG+A+XH@?xJ
z`uCr*-)nF5UFYCyFVe02)igTGj|_plMJF8krx|3cOyz@&0q+GAMxw9<*m~Kz>u02*
z+C)Q2Up)JC9Eg!HtXd+~A5T`ONQ<t^XdJue;~Xp%c$Dw<Qw;5WI7guI7bvFfW|sQw
zW0`>-^B{Lia|eHYq|#7R*F!_mG@Ry>U!40Y;XeEu4@Yran8?|O;x>1?;hTf_X)Tn9
zP?m*(oKAIdl1Lb<p>m;iXAE>WgACKw$f-sWF96$)_yHEM;WyIg;WUClcud^%#uE$P
z`&Ch+E(%%kjS+iWw@cW}S5LFOnjC{`-_|Tf0rU%l627dGSJP&<f4EHZzPW)2_-?X8
zm%uMA*%AC-{B~vcOpv*4F0D5@g3E1VpW`H&PqQA(K-s%QfENbCfE8xeqLAzf_QLVp
zcI-@Lk_Sno@ek=1l{$vvm9|je8~KI;x6?J=SRDPWa##Dg0ae~zSnzjdAKrl;MgDa=
z%QreQwC|;U*7H_wD)}t?s~FB-3t3ql%Vfq0RpFUzVt=sHrMP<(cW>6$;KzP8j{F;;
zeV5&y(74f4Xyk)R4i1Jdw<J%j(ERvq&x~y*8Umfubl%(TI?F;ceiWaIOReSdlf)H8
zh~9-ZpQnkJ&J2yj+=!4-mHnJ*v8QwpJ>^>if=%E}TL^>*9hzxb>H7RF;<f~GfAJU*
zd()79kxD#OdghP)2~GDNr4<1(7yDNEOZij1vEK8?4$VNS-$JEV_kPtRBvS(s3uIO_
z_Grz*(YlB%YqI4<;l2du;zJeGZ;#QZp?TUHy(68YU(|sOdi^Obd#K6=Zv~+yM_A3x
zIqB+>n`BFnCC1!ZgQozl0u>WKR6BE>Vts&i0O*&iI9-nPm(rY<F!YGX(WiXc$av7n
z@BDj*TOX8URCli&-FMi%9*}nIT+Y5>Lt~g^{Dw+fYO&vzkzsM8-CXA0lfTn3(VVz{
zo*_}E+w??2;H|qLqw_i20-%q+2mV<0KTxx%JoKu;tq7$9xal9@egHuZcLSLF5y&pT
z&2@su^HB22<N*@zZO#0r4l*(~fQ^*9rjWD<CeImt(kLiXq+vB#!S?)lSIobvDlpB7
zU*CbTxU!VlfDcKt0l=XC?esgawg9Am(pCFW_5hm&yU)4xO;Yta%>JTksWH$92JF8%
zn1S=>=bqSrCZ=)E?mQ}LCaF`6*u@^aKBOm9RbODSzzaMc_b<cx&x^*o;Rt2FQP-bX
z4Qk<vhFozd&V@jYlKPN&^I{Q+6~)o%Fj=02Z;Zbtr(s?u`Lt!EZ^~0M$HEj^!PV2^
z{*g=9bbUkQRufgNgr_*fKL`AgS<Jo8oXy)e9R9Ca>4^b??z<3fC^z^6%P}JVU;xEr
z+bXNTv#3>r8}o7C8PKWxd;dZmp!YfN)CWLDaQbCx2W|?yF%Pff2uYut7+olC=nw!p
zKU-ZDcrbuOtJ&(CKzq+y$GAYH5M?kNY{_ESaDUYY{D?5xjG>~a*`!QxDri+?Uz%h}
z4L~ibFztafd}-EG)9BLsgV}m=cCQ|o)gE=*1sG>`F*uyZ9yDY}+F!*@0|e*MkErSu
zZBB*)>YK{`uPr3Fp;HN=WmPbMJDj-VfLt89^@sCm{LoVF#wsY`_`ZH=*)1;lm!Z)-
zg^Nw`)<tuNeBSAki)_kJEhkI4kVw+19vz{9uSuytCz`y?B<C|NHUF=9QjZoS6~~3N
zS76Ho@0#Blj!B<+QV~}$`1a@y`|{>$!XNMU=jp(y$%v0W{y*QxF@imdef*!$z~f8P
zp$ET(;u$+zHUVjBu1m-F!@?ZfzI`WLV;TZmNoGIn8}9Wm^9F;j{gXqk!u$cQBc|5a
zzSxlVPt+j5Oj)J<w+J_S{%@NNoG$qg@a<*_y)p;dul}#Z0<`TSh<bpizSvnLnB?-E
zIU-?mh<=4y`u{#hBqa4K5g@|)LFg=<JP4?bi%Gf;_BAOHNh}Maz9hNaxLW|se^S&x
zbpzhvfdniJ$nTgRj~^HPF!ektY+~7<kT6h0S|@!N^T|4Boc`a9FFex#fe3<(+kTL{
zja}e1$nYP}B%sOlr2lTB^-K%&JSQ#;5c_WcS?xeQ00p^yOb~pq-~NC9c^m<%DF68a
z4D=iTbA6PE*%>V8ie34=bFqFVc&>^4UhkJjx~7i|WuCnoF%WxX_)o&u!v5=tJzGXR
zRNSPI*TM96;~o+>OSoRy4o`8uhGaC1SQ^raa~gSweQ>a$BqQNeMXtCK^bv7$m9Ntt
zXw$z>a-;2`!Q8;YJ}G2FM%QK<P-Evb%w*31i<Y*{T(f3o`uuMNk(q*|Lb3~npR2x(
zt<BWF;bT@+>LUEQ<wo_<JL6b?>}mwF51)-~#U}pox$ycI$>5yYL1Ph|M5fE9Vcx`~
zYyY0rZMhfz^*@H-|BU^A{wEqH?f?RDA}*dczWzUd`~S&z5A*^l6=z~3M~BguNIl>T
zH9jEq<fCx!ioQy9F@)P68WN*bh*XMbg<<;)*emrZ;m7HzMrP!t5Bt|w7)f92<FDDj
zYx&4FqeLxNU#@WuAJs9qmuz4Fm_G$l{X~#`i&@v%f3sz2_)nMqi4p7F#w%|6bXmG0
z6_&-L`&(bV#s370FMMYRpv3CtZ;*(Id-n87crOc+msYJ$*~rtrz0*i`@r5v1hlGY4
z*`A)<a_IehNix%Fv+s1Q!qgFM>N42J5>~u=cS5^)zb+tT{8VVj{jF3~{n8Z1;{US~
zvUDhIVl_Kv0S+A$p2W2OHBw#{aaJR|ob4mNyo>e6_f>-L#?y59Fh?=yK-Zlkw47Zi
zEk|FVuH8_Okui7ic!H(;&is)(3X?}B{UtYc(j|wKH#Up@zASptBA03%(bpyj>Re7#
z|9^#stp<c{U8OisE_yt$`8k9dw7pMbN{kDyYeF|$CMlaUR3h_m|3*K`?(-Tb0VSNj
z5HxVB_*78HrGCejlXk?GlbOFMG)l+$a>1RQ-sK-QKmNW{XQv$6O&J1%B&06#{~xoJ
z_Fw9pNGitpP6yE*(Od(ryAC?S20u(t!uk9Tm*<G1FV0Wt73&*d(w9ui^~KR%?L_98
z3C9<=JN~aX&f_AR-0g+_`-KnqHM{~1@54b4LSz9p+5glv*lAeWnN@q|9Tv2Z#nBUo
z>oHGt1a<rO(5`lJ^!pim2^<^spGcVwOF!@Z8`ej^t@vw7Ezb4Tu2-i2$x)<2%W}*%
z$~E)a1UG1gaY6Cw|I|KtW9u?8={ss{41ToD$SstFUzl^|Uu-i{a@3clfXMN^8mGp$
z;{(!|WN2q&Go7nceA&v&KoTUepaG$FS@DybZnX-V%kqKY64_+I`%hnp`yHd3Q<J}M
z{ObT-VR|{{%@MtQTL>-t?=cU+698^GFLRsu;J;!L)$@vfe6UGv;^S&IKCq4Rixtl1
zwq}sl2<E8h?hYKu-OA(&alvNZ6AgT?2dR8+ft9xBt5)b(eS{C!g@ua+cyUka8#FH#
zyQo)cbA~*<Uv)%_Re@Fb25rf%#of>U<mfV+Hr19A$$SXINB|34qWzI+`R}9pKmQ#7
zNQYxQz*oMx<rE(CKRXbwcx|0z$nUuK`|Jn$pE2#*{-S=qbcPt|bWDpKzGbn!`@<f-
zh55d+!YcJ=KuGMcW6Qs}f6?gXFp~V2_}_ES7Al$6rg!X%qAfchRQb?3X7}a2Z~$XI
z@Vy7&xeS}IlUz<J?0<E4?cY%6VfZsyi`6zLl);4EW82Va6^o20aw#O&7&Du7X~no!
z<1#aG+G!(7ZM7k`tlMDRV%%m%m$5~rJrWvYrjXWUjA0CO`|fD>xArgS{R7_1^FGfx
z@A>@jd?dxWGVWC5dh$Wx8qdAWfv-KPB|2O8BB)_gC_poP@TN|V=NBm2I06?jyR>Wi
z{OoppOMHHjFWl6{fj#*_ua|9KPTHEt5@l(pGU>)lM^wP>k2rO?!U#NwHx|w?b@8Sc
zBY785h(k8&7V8cbM?vRuh_UW%*~m2)W+CY*^UL13DbZJU9IFxDOS4`^|MG!$(;(^E
zjg$lF`EM^fdw5}<Y>Hw};k!q9?FD`je-yG^M;E0Up*yX+6SPbrw<H(2gCnZQ;Nzh$
zG>zt8M^548`TV8b794sW`unz-Q3(6iCzK5r5bK$%^5jOS+qjTxDj(NXrw2P%uPp6~
zR#A)=P;mU4lCM{(YTbk;2F|36=8Y^<Yh2m!lmpM5Y0wmqc^Ce_rmlA57$$Wq=Lpcw
z?PIyOv3q^748JS`A17g)aC8OB(NjXJ0?Vo9<}LiSkkSjr_v7>`E<X6~bF+JwBC1=H
zEEKMn47@fM<z2mGzH6J?wup+IdSAG?YNNdqBX0ZaXy#m_9j`v4?`6G-{EK%?T!{g@
zLpm`cqc79x6CIL4**bH2JaaI$Y`H`hE+51TQt#Tn8yTNNoA)6snQZ&YEw?)e{U~^G
zG;2n2!|=Z3H7-9U#^gu)#Q+I`h+D!_qgk<*PRC$7QFU9_i9JuDIX@)^CvpMcC~A!M
zL>w`>D$g{stgk{?R$#(~rHc~5`<oiIWJG&W8I`{OR|mUnRFHF5f4pqG*s^M|Cp0^=
z7CJg<C#GykDM$Piw&dW6-+M<wI29;<*{ZNGtR^xT?P~+!XR{2(YShXflf`Bb)qb6*
zP*pyr)Bq&w0(;Yy)r_cp13V*(<7e*}wy+*G5zswBJz>sarK8P?*DNo=#YmTv{*o)h
zIHS~9y2^Rs4Qg+sN7j~;k@@cy>^*MnxABHL`lJ$BW*)mcx-Yhi>j=EFLXjdU<2p1_
zCTC1c5y7Vyv_@n|EQJ!x)dBo_*B}_c9*UsPehv@|p6O}BR{*g2b{z?T-D7=9B52fY
zs8OPOB754R6+}u1g%^UOmZJev^e3b`Ki?G8b;6h)nb8?AKqkxK03aT2JNPeTNm?>D
zF*ZFc{e-x0wrO^Wm8%>KnW>bF`0>^wm*Ft^d|;pca0ulVrM=dW!uE?sp`AIm1NhY8
zo`nM@W49`st0=@k7b+_Kg_=5;FoV<pV44X_S?)^fAyJz6;N&T&;WT#-C7c3RrR;4b
zyn-dANag-fbbc3wj41tV#>h|>TYIDy8BF}`8Lz$ExzLxR3{<pc+*AFiMwf~-l9_<C
z(KI0U)~WHI&*pL3(6$L(*&J!9BOr{(tLfAgf#$=xwmQgPT*i7&ua7C!aXFpuS>h0^
z_+5%g&%%+_!MvM_kbrau`wAxHs&iI3Fkwf+G|I7_nBvN1@AFVk6bFbsmJK0W%gRnc
zM4Ap@(l^jc(sF86!5%-e4lgfY;AHaE)~s~bv6knU_sIIhxB*1gP>u*$e6!5KQw>IP
z*brMpjvNNgNT~XY(6iI{DDgv^=?1!H0*v%%qX&}Di)q!$n076H1H!BAJ7`05Q{E>v
zfK~3skUo4kz;a7!l;h`<_~uC5KuFetd%P}aJhPG6O66orQiQJf)1zsO&W5F{z4Zmb
zjM}?#;uHFS`NIbK%%4S@0d~)zZ+{m;WhWd|b(6-ZqEPKCVYK_ne-Z0CTcn4@S7ulA
zByjM79ZRZXX8uXd>;$X28?!I*5vFo^*q*{MGL%X}&IK{*eFNE6c<8?sd;&Ua98<#0
W>*qatY*f#{s_=66J>1}i&G`pdh@3D0

diff --git a/articles/index.html b/articles/index.html
index 7a8f8ade7..eae4a6b0a 100644
--- a/articles/index.html
+++ b/articles/index.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/xcms-direct-injection.html b/articles/xcms-direct-injection.html
index 706216b57..8cca6e609 100644
--- a/articles/xcms-direct-injection.html
+++ b/articles/xcms-direct-injection.html
@@ -33,7 +33,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -251,8 +251,8 @@ <h2 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 <span><span class="co">##  Number of spectra: 4 </span></span>
 <span><span class="co">##  MSn retention times: -1:59 - -1:59 minutes</span></span>
 <span><span class="co">## - - - Processing information - - -</span></span>
-<span><span class="co">## Data loaded [Thu Dec  7 08:45:30 2023] </span></span>
-<span><span class="co">##  MSnbase version: 2.29.1 </span></span>
+<span><span class="co">## Data loaded [Fri Jan 26 20:08:58 2024] </span></span>
+<span><span class="co">##  MSnbase version: 2.29.3 </span></span>
 <span><span class="co">## - - - Meta data  - - -</span></span>
 <span><span class="co">## phenoData</span></span>
 <span><span class="co">##   rowNames: 1 2 3 4</span></span>
@@ -430,7 +430,7 @@ <h2 id="session-information">Session information<a class="anchor" aria-label="an
 </h2>
 <div class="sourceCode" id="cb20"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/utils/sessionInfo.html" class="external-link">sessionInfo</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
-<pre><code><span><span class="co">## R Under development (unstable) (2023-11-22 r85609)</span></span>
+<pre><code><span><span class="co">## R Under development (unstable) (2024-01-17 r85813)</span></span>
 <span><span class="co">## Platform: x86_64-pc-linux-gnu</span></span>
 <span><span class="co">## Running under: Ubuntu 22.04.3 LTS</span></span>
 <span><span class="co">## </span></span>
@@ -454,67 +454,68 @@ <h2 id="session-information">Session information<a class="anchor" aria-label="an
 <span><span class="co">## [8] base     </span></span>
 <span><span class="co">## </span></span>
 <span><span class="co">## other attached packages:</span></span>
-<span><span class="co">##  [1] MassSpecWavelet_1.69.0 xcms_4.1.4             MSnbase_2.29.1        </span></span>
-<span><span class="co">##  [4] ProtGenerics_1.35.0    S4Vectors_0.41.2       mzR_2.37.0            </span></span>
-<span><span class="co">##  [7] Rcpp_1.0.11            Biobase_2.63.0         BiocGenerics_0.49.1   </span></span>
+<span><span class="co">##  [1] MassSpecWavelet_1.69.0 xcms_4.1.6             MSnbase_2.29.3        </span></span>
+<span><span class="co">##  [4] ProtGenerics_1.35.2    S4Vectors_0.41.3       mzR_2.37.0            </span></span>
+<span><span class="co">##  [7] Rcpp_1.0.12            Biobase_2.63.0         BiocGenerics_0.49.1   </span></span>
 <span><span class="co">## [10] BiocParallel_1.37.0    BiocStyle_2.31.0      </span></span>
 <span><span class="co">## </span></span>
 <span><span class="co">## loaded via a namespace (and not attached):</span></span>
-<span><span class="co">##   [1] bitops_1.0-7                rlang_1.1.2                </span></span>
-<span><span class="co">##   [3] magrittr_2.0.3              clue_0.3-65                </span></span>
-<span><span class="co">##   [5] matrixStats_1.1.0           compiler_4.4.0             </span></span>
-<span><span class="co">##   [7] systemfonts_1.0.5           vctrs_0.6.5                </span></span>
-<span><span class="co">##   [9] stringr_1.5.1               MetaboCoreUtils_1.11.0     </span></span>
-<span><span class="co">##  [11] pkgconfig_2.0.3             crayon_1.5.2               </span></span>
-<span><span class="co">##  [13] fastmap_1.1.1               XVector_0.43.0             </span></span>
-<span><span class="co">##  [15] utf8_1.2.4                  rmarkdown_2.25             </span></span>
-<span><span class="co">##  [17] preprocessCore_1.65.0       ragg_1.2.6                 </span></span>
-<span><span class="co">##  [19] purrr_1.0.2                 MultiAssayExperiment_1.29.0</span></span>
-<span><span class="co">##  [21] xfun_0.41                   zlibbioc_1.49.0            </span></span>
-<span><span class="co">##  [23] cachem_1.0.8                GenomeInfoDb_1.39.1        </span></span>
-<span><span class="co">##  [25] jsonlite_1.8.8              progress_1.2.3             </span></span>
-<span><span class="co">##  [27] highr_0.10                  DelayedArray_0.29.0        </span></span>
-<span><span class="co">##  [29] prettyunits_1.2.0           parallel_4.4.0             </span></span>
-<span><span class="co">##  [31] cluster_2.1.6               R6_2.5.1                   </span></span>
-<span><span class="co">##  [33] bslib_0.6.1                 stringi_1.8.2              </span></span>
-<span><span class="co">##  [35] RColorBrewer_1.1-3          limma_3.59.1               </span></span>
-<span><span class="co">##  [37] GenomicRanges_1.55.1        jquerylib_0.1.4            </span></span>
-<span><span class="co">##  [39] bookdown_0.37               SummarizedExperiment_1.33.1</span></span>
-<span><span class="co">##  [41] iterators_1.0.14            knitr_1.45                 </span></span>
-<span><span class="co">##  [43] IRanges_2.37.0              igraph_1.5.1               </span></span>
-<span><span class="co">##  [45] splines_4.4.0               Matrix_1.6-4               </span></span>
-<span><span class="co">##  [47] tidyselect_1.2.0            abind_1.4-5                </span></span>
-<span><span class="co">##  [49] yaml_2.3.7                  doParallel_1.0.17          </span></span>
-<span><span class="co">##  [51] codetools_0.2-19            affy_1.81.0                </span></span>
-<span><span class="co">##  [53] lattice_0.22-5              tibble_3.2.1               </span></span>
-<span><span class="co">##  [55] plyr_1.8.9                  signal_1.8-0               </span></span>
-<span><span class="co">##  [57] evaluate_0.23               survival_3.5-7             </span></span>
-<span><span class="co">##  [59] desc_1.4.2                  Spectra_1.13.2             </span></span>
-<span><span class="co">##  [61] pillar_1.9.0                affyio_1.73.0              </span></span>
-<span><span class="co">##  [63] BiocManager_1.30.22         MatrixGenerics_1.15.0      </span></span>
-<span><span class="co">##  [65] foreach_1.5.2               MALDIquant_1.22.1          </span></span>
-<span><span class="co">##  [67] ncdf4_1.22                  generics_0.1.3             </span></span>
-<span><span class="co">##  [69] rprojroot_2.0.4             RCurl_1.98-1.13            </span></span>
+<span><span class="co">##   [1] DBI_1.2.1                   bitops_1.0-7               </span></span>
+<span><span class="co">##   [3] rlang_1.1.3                 magrittr_2.0.3             </span></span>
+<span><span class="co">##   [5] clue_0.3-65                 matrixStats_1.2.0          </span></span>
+<span><span class="co">##   [7] compiler_4.4.0              systemfonts_1.0.5          </span></span>
+<span><span class="co">##   [9] vctrs_0.6.5                 stringr_1.5.1              </span></span>
+<span><span class="co">##  [11] MetaboCoreUtils_1.11.2      pkgconfig_2.0.3            </span></span>
+<span><span class="co">##  [13] crayon_1.5.2                fastmap_1.1.1              </span></span>
+<span><span class="co">##  [15] XVector_0.43.1              utf8_1.2.4                 </span></span>
+<span><span class="co">##  [17] rmarkdown_2.25              preprocessCore_1.65.0      </span></span>
+<span><span class="co">##  [19] ragg_1.2.7                  purrr_1.0.2                </span></span>
+<span><span class="co">##  [21] xfun_0.41                   MultiAssayExperiment_1.29.0</span></span>
+<span><span class="co">##  [23] zlibbioc_1.49.0             cachem_1.0.8               </span></span>
+<span><span class="co">##  [25] GenomeInfoDb_1.39.5         jsonlite_1.8.8             </span></span>
+<span><span class="co">##  [27] progress_1.2.3              highr_0.10                 </span></span>
+<span><span class="co">##  [29] DelayedArray_0.29.0         prettyunits_1.2.0          </span></span>
+<span><span class="co">##  [31] parallel_4.4.0              cluster_2.1.6              </span></span>
+<span><span class="co">##  [33] R6_2.5.1                    RColorBrewer_1.1-3         </span></span>
+<span><span class="co">##  [35] bslib_0.6.1                 stringi_1.8.3              </span></span>
+<span><span class="co">##  [37] limma_3.59.1                GenomicRanges_1.55.2       </span></span>
+<span><span class="co">##  [39] jquerylib_0.1.4             bookdown_0.37              </span></span>
+<span><span class="co">##  [41] SummarizedExperiment_1.33.3 iterators_1.0.14           </span></span>
+<span><span class="co">##  [43] knitr_1.45                  IRanges_2.37.1             </span></span>
+<span><span class="co">##  [45] splines_4.4.0               Matrix_1.6-5               </span></span>
+<span><span class="co">##  [47] igraph_1.6.0                tidyselect_1.2.0           </span></span>
+<span><span class="co">##  [49] abind_1.4-5                 yaml_2.3.8                 </span></span>
+<span><span class="co">##  [51] doParallel_1.0.17           codetools_0.2-19           </span></span>
+<span><span class="co">##  [53] affy_1.81.0                 lattice_0.22-5             </span></span>
+<span><span class="co">##  [55] tibble_3.2.1                plyr_1.8.9                 </span></span>
+<span><span class="co">##  [57] signal_1.8-0                evaluate_0.23              </span></span>
+<span><span class="co">##  [59] survival_3.5-7              desc_1.4.3                 </span></span>
+<span><span class="co">##  [61] Spectra_1.13.3              pillar_1.9.0               </span></span>
+<span><span class="co">##  [63] affyio_1.73.0               BiocManager_1.30.22        </span></span>
+<span><span class="co">##  [65] MatrixGenerics_1.15.0       foreach_1.5.2              </span></span>
+<span><span class="co">##  [67] MALDIquant_1.22.2           ncdf4_1.22                 </span></span>
+<span><span class="co">##  [69] generics_0.1.3              RCurl_1.98-1.14            </span></span>
 <span><span class="co">##  [71] hms_1.1.3                   ggplot2_3.4.4              </span></span>
 <span><span class="co">##  [73] munsell_0.5.0               scales_1.3.0               </span></span>
-<span><span class="co">##  [75] MsExperiment_1.5.1          glue_1.6.2                 </span></span>
-<span><span class="co">##  [77] lazyeval_0.2.2              MsFeatures_1.11.0          </span></span>
-<span><span class="co">##  [79] tools_4.4.0                 mzID_1.41.0                </span></span>
-<span><span class="co">##  [81] robustbase_0.99-1           QFeatures_1.13.1           </span></span>
+<span><span class="co">##  [75] MsExperiment_1.5.4          glue_1.7.0                 </span></span>
+<span><span class="co">##  [77] MsFeatures_1.11.0           lazyeval_0.2.2             </span></span>
+<span><span class="co">##  [79] tools_4.4.0                 robustbase_0.99-1          </span></span>
+<span><span class="co">##  [81] mzID_1.41.0                 QFeatures_1.13.2           </span></span>
 <span><span class="co">##  [83] vsn_3.71.0                  RANN_2.6.1                 </span></span>
-<span><span class="co">##  [85] fs_1.6.3                    XML_3.99-0.16              </span></span>
+<span><span class="co">##  [85] fs_1.6.3                    XML_3.99-0.16.1            </span></span>
 <span><span class="co">##  [87] grid_4.4.0                  impute_1.77.0              </span></span>
-<span><span class="co">##  [89] MsCoreUtils_1.15.1          colorspace_2.1-0           </span></span>
-<span><span class="co">##  [91] GenomeInfoDbData_1.2.11     cli_3.6.1                  </span></span>
-<span><span class="co">##  [93] textshaping_0.3.7           fansi_1.0.5                </span></span>
-<span><span class="co">##  [95] S4Arrays_1.3.1              dplyr_1.1.4                </span></span>
-<span><span class="co">##  [97] AnnotationFilter_1.27.0     pcaMethods_1.95.0          </span></span>
-<span><span class="co">##  [99] gtable_0.3.4                DEoptimR_1.1-3             </span></span>
-<span><span class="co">## [101] sass_0.4.8                  digest_0.6.33              </span></span>
-<span><span class="co">## [103] SparseArray_1.3.1           multtest_2.59.0            </span></span>
-<span><span class="co">## [105] memoise_2.0.1               htmltools_0.5.7            </span></span>
-<span><span class="co">## [107] pkgdown_2.0.7.9000          lifecycle_1.0.4            </span></span>
-<span><span class="co">## [109] statmod_1.5.0               MASS_7.3-60.1</span></span></code></pre>
+<span><span class="co">##  [89] MsCoreUtils_1.15.3          colorspace_2.1-0           </span></span>
+<span><span class="co">##  [91] GenomeInfoDbData_1.2.11     PSMatch_1.7.1              </span></span>
+<span><span class="co">##  [93] cli_3.6.2                   textshaping_0.3.7          </span></span>
+<span><span class="co">##  [95] fansi_1.0.6                 S4Arrays_1.3.2             </span></span>
+<span><span class="co">##  [97] dplyr_1.1.4                 AnnotationFilter_1.27.0    </span></span>
+<span><span class="co">##  [99] DEoptimR_1.1-3              pcaMethods_1.95.0          </span></span>
+<span><span class="co">## [101] gtable_0.3.4                sass_0.4.8                 </span></span>
+<span><span class="co">## [103] digest_0.6.34               SparseArray_1.3.3          </span></span>
+<span><span class="co">## [105] multtest_2.59.0             memoise_2.0.1              </span></span>
+<span><span class="co">## [107] htmltools_0.5.7             pkgdown_2.0.7.9000         </span></span>
+<span><span class="co">## [109] lifecycle_1.0.4             statmod_1.5.0              </span></span>
+<span><span class="co">## [111] MASS_7.3-60.2</span></span></code></pre>
 </div>
   </main><aside class="col-md-3"><nav id="toc"><h2>On this page</h2>
     </nav></aside>
diff --git a/articles/xcms-direct-injection_files/figure-html/calibrationresult-1.png b/articles/xcms-direct-injection_files/figure-html/calibrationresult-1.png
index 3516bb38b78b172debb4d3642a394e547823bd35..3a215ab4a0a14627a94e9a52deb7af53b8e98c94 100644
GIT binary patch
literal 28255
zcmeFZd010d+cz2zMZ}0xRpwf!S`|<b5DcJb(ITKi5(Xh40s;X<408adpjAO@#Q^~k
z1xbJefdBy#G!;~4aYGm*Mgcdo5CREFzO}LK)8~7i_dDnP>zwOcC)d@+>^1DQ*1m_|
z{k!kY)uZ+{=p`CUP$(4o@Sy|8Q7EMZ6ly{B;zi(!9bT~vd@VX@YjXg6gU_sWMLhUY
zIdjM*429B8ME+Aq{ygD>LVb@qd|;nb<ip{fn96KFcfO?Q$gabln=f6oy0SB1HLdx;
z6@~4==rZgtw|>|Xb$-Rx@0km4{R(|NpN};u+jcy^aR1TeTOTPMRB?LZZ1v!mSE!aB
zw>aKv>AQL9=7v2xCCL&G_iX~LHp8Jk_0j1fiI7ET4=WoDl(s#%Q}4<zU$$_5egFUe
z{+~Jl16A`i?kHu_NudwqP8*_fa)VEJVvyBny}(fQsiS8v#YN=n<ptBHb<arcg`W-c
zR)EHY9c53~|1o=!J5^2~S<4M(_KcgwXWvl-O|}Cw2iIC*bDZK`d!k}xUtml8U6bv~
zF_A(H6bgmlGN`>4@?8p`r|It-8c@+uLx!;)@0U+$4PK*iG$gSF>(?REd{`G+h8ld*
zCYqbA+HgH=3{B>QJy?jeFm@-ZCdfjEisDB(R|`E%rpbc$)CHg+rLz&-IVU7?uElOj
z&Mm}*EGhw)QMQLr#gU~Uk+F4~#e_Y~h_rn>z}4!ND^WXoYfbj{*4H$Obh^lf&k$jB
z-mm!RmBg6@&CCqjO~t?CnJv-*mp`m~&QzI#peCF{u8vch>j*X7lr3j3i&L6=pj^uK
znzBswssjB`M^5iRnY|}{rqpuI-jpN<nog609+II6L%IfNO|X;n!WZD*;QnR0i?oL~
zGDKR<0}ngyE(LAdRf`Vw3tlcr9}DKy&)U-Bozk#mcVV@4Hz$x7@{IkIb#Bpa<Nyq|
zM3}t}v?P$**Y_g(Mn0Z*Mx4(n6w}PgSz#qbtSYbEb|Dm`nx_id5_XstiSNU|F5|49
zV#8}`Apwhl1pwz?lXgOCpQuJ&Mj}6so|&^njE85@Xoo?T-;U<;s%V0cubC2?UnRps
z*J$m#wN&Cgmrs3+TDuJNtG+dorx%F%BPx#cPGKzGhGc7PaBh?$3dbi0-BTzImxuWh
zb|H7UJ=504P2HlK&F9uwwBuuzMTX!bPFRE|=i*h6d5*;dS)ZPKVK-HEYFJ{e92J5m
zeV&o2+2@-fqnX5zR02z=U&|h|{6k{7lP&N4@;kE5?dCSunrb`G=95zgB@y0|1nbK9
zjupMips8$uLUU|vEU~Hk0AF%_WI$~~=6dhkP|8op;dxTya=kj*YR%6MGMUQFU7#`J
z`k4I1Hxcrz)8UD~xLZcui*w*Lk5#CBm8k_s%FT1!egI=pZbIe!M3a!bQ9_zinlm{{
zIH}poiH;3vkWwvs#%13@6O%t_fqn@M<+j#misXyiN1$@$ZpkouY>Z}~=I{oY<Vb*n
z&|KSWZdL*hU+fkQ!~O6pC~e^QR?x;U3g6C3_>SZ&9E3BWQCOc=iX9`~Q#kkNHZ7If
z)$gQ>%<zByR6p-598|<rH%>gb+~sSehxA@G^Cduj9r<U%h6hmn>b4QLbBGF72NWq0
zEvB6ua^uQ7ceBHTNG)q>$u(=nkaP5W+ZarqZHrlZ0+Axd58D+|>2=i7?X1N}W69!#
z{;^VEJdie5C96r1-?jOBIq#^|2YZ;h3JamyHSS2~|GS@R=5z0sg?R~Sg59d64W~~a
z<2H}CLNfLF+t?V<?r+WODjh@~WCG-?!iycwO+PL@TM`&M(g{D_ktDgkqgM?yR{x&X
zBmSB%(koj7et8~rRx~Eu#fG??LyT8-zzS3YKV%0B{QJ*+H2t-YB*Hm`ZG)?zAzJbd
z)^eo7zkSvO)IL`A{QMAysbEb$Rs~1s*)?-S_zuoKxc0ewhmx4bDC%@iGqB*~v%H9R
z8;WSU?(_E~n(I>I{8|nY@xZ+B95K(MiNYWVv#e#T6W#%((yq~l`j7Vz^Bd-7rEr>R
z9?<KAk-j<jVcy1$l|Y<Fs2)e9mgu6mq)&R#?b;z#NdJZ?J`T4~=jCDOV&{nlo3ER^
zB?Xi(F<`xCObPAFQH2W?AI)|+x*ty?RS#tp=;cov*Uxn9oW=wV$JFpb{DhToD4IWm
zZ(x6t3qC^;4pH?sSdPKxonTi97nh?NYLnpn{6x#USj!VtIbQS(?w-PEE8!}1<{NdW
zf>uQ9@;f%Xud>jr$}D(^cUtT6Gdl#bpTX^`mzdK#Ej{!{2OS78FDaoqhH3BmN8eGz
z{Nlm=pq;nIqjfyy)9f|9Q|75P>3UinhBeb)DVW%4-Or|LvGjGimVvZ`h<MWWp^S8s
z*LA6yi^u&MXf@NBZd!<BI<Ej*;~6sX4qkbIHPfHAUPnxK9ts>(Wjs;W_;R6+DN{H7
zr8Br%y#(3HP|frg=)T)JzX;)P6BC+DCqxh!yMWkEP~nt`2_JNAxkL0yUb&!3q>;fP
z{&WYqzwH?CWsDz!pXs<%WqGY>dWDXet$faU!a5ln7slP(vc)OK$<IlxARU*+_Nw&5
zN%|e<t=4T*y7+|k>oBAC21qdAAw1{aIbL4r-*Oqb>5j_mKGEyRweGLO5ecrMnUzG_
zmqc8USh!$HS$@JA_4i;m+v&Hd%1(S&%rn-|$vUr8o?}|;IC+wkpAp<(?4!b)>v3{7
z@Q9#~b}=f4rYN7i%&D51ijS+%Ib=J*Oe7-M-5l{%xCKfxsf>%^TH14-i9=MMUGwg`
zNYkb3v@kGl9=K6U@s^I#H;S+#Q{A>23}S{m$qUvGx^&rpw1#xCWAxyzadC)I{qhR}
zs=Qe&s##0roStFNPEO2r#^?vJSN*bG;bPDsloAS6BrM98KaShS?c1f<lfV9Zme=rP
z8-*ZqaW|*pjbbgkU8~AiA_ej&G+srr9qQ|QlH;bkLt(kCw;$;&DS5)|>1X%!Eht4)
zLPyjoZc|q&HraeXpPp}FI4z%0Z+}%d5~IOZ3(nJtyTs-@yIW4j_g^1!-GDmT)#->S
zj1D6CBw7fE!n`xKm?#~nvzlL0VUbsEqYSn&j%q5sE~ST<o})lv$GNg12`wk+Yx<am
zB>uVisQ&J4ir+KS+#%HpfopezN2y6(QTOOj*g`AdY&8dr=klZ?=<2F&5p#k)mU{4*
z)}#(JJ|JHkBzrQh4Jp@1*xo;rf;OTaT<~%CJ-5B{FP#fsd~h1qZakM*oMRR=T*35+
z07^&ke>mU-l<pKbZA}+9)?`22uQFQkOw5E+X?N`^eoTrA<eP*Ld4o}g)5%g!cqQL*
zJY)eP@GXqOPS=?*7a<pl(J=Nff4GXKY^24Xc}B4s?14JaMHhg4QQ471%UylsKq+U2
zQRR#aj33oa8)uh{rMBics~Q$}d~qi>nCvG0junQX&o_3Not<&CJ}d8qyELcKR2F{*
zI55wKAaC<8mQ!>gXJmekRjRT4?R$WF?e!+=(><CE+(%EJcz?!-02MJ^R|BDTlKB#w
z(!p(Hx)~l@i5ub#naaWVvm3!y;={`9837j1a`9aC>KZX6G_T3tEAJpf#R!F}ufdnZ
zJs9r+&nF3Ps*$Ac%3_%^$==)y`IgSPOzSO1N|q9uJjU3k=e*scoa&|jQ2TI~q|3k<
zWD7O`L5;|j@$GsirzqCG7B14>>+nb3w55hU%DhS(ky<&x#}Qn~&!XF}!xw_SV9@@b
z>2_XuleL9<884l$c`gEmzP@pR=JbjlLKVY{(v-P3*qT3fpX{&2Wk2mxVn`v5*?zuX
z@Sw!O??k(XVFJ%My=T0E#W4$06$}TH_jEV}SLIz^j#5BX6EP)1LbJTA&#15DDW(>y
zmhaIn9{W)qs<8X)@N8xqXG?D<qd2gArni71=;c>tHmUYIr=qR|c}EPr5ew1s1r{e(
z#v2Ua;(NsLMA3cU^I;q3l|nKNzio?;pLsbm*M_F|bdTH0pKwwcmFceT?PAgEAT-~k
zG$?XJ3=`_h^%RzOy)`M7>DG5{ZsZ)rWO_A>+=tfUbFc&)IS7D@98F_EXXKb+CtC>!
z`I*St-NUtKaijnn-R6hKrKg3mHb142keS5KcX58i^5ZEj7+QuuFZ`5XEHfTrl)kT8
zd-OI~^FSp<dsd|TnG`0RDaT{n6~70tvP^=f2b=Eo6^tpw<E1&UEz@ASfI{dtiFv!~
z{h7}8s@m{dN>)VMhydQI!D<&wV#PF;&u65{!{@c%ytBhVa8j^6&B3_7jJIx4j}v+9
zLcpbiD^YeYCa&puF8tf_8B~D>UUebd$pjZuJH)jcK<wTJUQrf{(Pui{MI+?L<@OyC
zm57>h<DNvH6n=$+A-Rd@7X?p6U*u7-{C3a9##(cabtg_<Z!<X;H*<aHg7;N*tL-<^
zpf?bvX6W~ni0~8?fG~?<s17D`v2*T-FeOX%8@{;bEdz(#mQMMSN*_tGG1Q492M<qv
zq9mMFUz`9`X#96d#tnW~Klh>e?qRiJj+g(sJVk}()x#$M!tlilUs5~=N;NN}#b{;A
z-D@o9`33%_)RqD%P$sRml-MeE-`%c;vRrZ1f>CAD$N!sQx%XXBDT}#&mhYHmJdN=l
zE-i}o5SCG%$f@(OM16B-lQ@3Qyb(C}YYEEM9{5pnYAI7kb^<3oBTT0>^e!9t16F_>
zoRR}(Y(FB_BL~=U6wRlo=rT&CW?S7aNTr5o>B>)7KbmAFX~~>=x*3~i{$V6|Y*{#w
z<NH!F8c{7pX>IW6x%p0h(2bufe;JxrN-TM{un<x4rESLNWw(in=tS>TgIf(2oVL)}
z$qejkz*Pniby=@M0oZsI9mdVAR?AH<eiIY-gcTt?4@(WHsz@mIFE9U1-4n1RcdkbU
zmV+Ztt`J-1_c`@RAvGWU?>!YkXJ^~~e{=rTC#&+rH{eD{7sqiTzoLjDjE2{XS@0df
zq%N&@Cx4d$@4BDKlSv89u+*S&6m}GtT9yl+uX(A1D1}WeR-ejl7LTFh`HaNDJDMJn
zSRq-Jx?;^69V?<{g{J*vC&8V6=Aq<C>b?v&DQQSP$!RiDeYimZ7!<8;7c^(kqgC7_
zmC(r0LP-7EHuRcbP(};q;c=+7@HNH%K?UuarYuMH7KDVqPRzO<I=7$n3s?R@x0}%>
znnDXOMSc1R$c&nc;_8>x0{x8$1kn}UD||Vs>yIdCfaU$b^37k`hX^&wK*ca2yLxl;
zq_6;f5AT6`RYoPJgxzpDr0I8bEqKI?KLC*Dq-wN{c^vJjVuf(O7PM-nB#t}_CN91*
zZTxV0_YHvnOqB55CX{O)jV4M$MpYkiRz%`1#wWyJ0BF>>k}=|AQxeFXbX$W(Kt6*E
zR#p}3!#y@0I6q>AN&?a|<sqU#D9<+?3r@$BKQ;?$Ifz{Dx`qNjt2#>uk<cn(<kd<8
z<a-lh0|3H1o06cd8IG%<t*RL)#hP5757|?|9oEHAx8Q)b5P4k#of8vxpxsbDG*DyV
zP+;$I=qVVAI&uopu4+H|B@lku(r)*E17NfN8Gy|V=j!D)ntc#Dm|XfHM$=beWSu1a
zA4&y3guf{js@iwtS#OV+n68FPKFt#i9J0^TIqqKBC#$0ge2;$_zOMi@sCr{PKqRJg
z;duWm^!etN$iD!vJ!-yH$;#1l)!=Mo4bQ`)>QSGj==2Gvs*|yEdHHJq>O<u%LY(YL
zqW<t8H)A(lqbo5u6p1K34}x&ZyNTz4xm;MF?pFH}3H(!kKs`tnC692k4cZpePIDdC
zA%U8+0?>rW9-c+jymu<_*kuvN{TBS<D5%pG9@w5;7VNxer`s7y;iHjbRW$?pslWJu
zk=5_a<3RXzMU!fS8#{gPRahw3!-Aftms_wdgWt_KJf#9{h18*ov>a?s$hxun5PSCb
zc`jgMBh8|H`B$xmza;Ulajag5>$&w{Qlcu|e+Kh)4g^SPE+7mWquS@EJC3A%U_0QN
z@<&~&LKl&Mw4rS4S2AWRa#^00qaHn_UQz8|&~;p3+KrVC#Aw<}9@I=sAWQreur_TF
z6jO_A)pPT91>_){&mxZJAHwthy`R8nps4Gwbp0J%B4@_@W9fV}IgfrxF{F{1nYZu*
zxD7^7Fn){v6F>}u2>LlK+ROSwQT)rHY4|C0IE)_6(hHi2?ftLe7+?e9rk(CO`Du@f
zb2RtgszMhfgkL8+XiCy|PItLr=+P%g_9STdPpT-~X)Ks*+@wsiF&dVHZQO!W%hl^W
zBmWGA8s?sxu17T4bEPk;Fz8`>bmZhm08>xz_2QZ)-?#S<o7Z#Cm5PtHGE}Di!j<DU
zVvO{t9y=}K+HltpJAk_gBo;V~wc%5JO%L!lMyyLs5s_p307ADyKO!9jdF|Kpz55V@
zWe4K=T1oA+m*cvj^WKuae};v@#0-zL^2?#&cztzUj+gCs$oXxv27ZxF4OZ7B;4suT
z4vOXxl_C$p<XyX`2zTZAMVkYdlt0kW23mws{m1e$FZ~QW?yeL5dQ(U+Z=?7U{5yOd
zegU;oTTQNu>I3+(zswUSD+jO{fs3P(`dQ`CbKSJ`0$WXsyC&ZV1?J^;qdA=@*<;m?
zMubs`z~<nZa;NRj@VC$gaH70k(~Af(k+2B_hyRyB9<U2432TO6xJrOn4}(>JUFgE5
zo#-j|I@~*gPW_oc=wViWn{PeJ^|~P~2k(%}o->dFlpM}wTaM+j4J7ZtP!#Xi+cf@{
zL3@<sfN*^F?p~5TqFSCS?G;iLa}pYwFtlW+WbKUjOmPvMhEBDh&mI+qAv{S?q;v{S
zrmk&b9YyBZTw27+6L#!vqT8E&kd8eI_UA1a>_(4`1TVf}QP?LNsAqq=zjFa_2?>Yq
zDg42l033sfgT)a};tg^EeOhb0w_vm^c38)ee%4~%)g0Lmd}c~QLoH`rR~@xoJDYD#
zlXn;l4G?t3d-I{&R9ejd-|&?f+&f{@W?&IRG`dtxT11ey#9cXRtPNgUTkFsna169p
z2OwO)|C}_Ox0|vv#EjNkoPGz*E|Jzpsh@Tq8M&o7GUc|~>Fe<2#4n;taK`my0{NyW
zo)P_eW~M#IQ3G^BwSHC3*X@(lz%uACbQQXbx8(FmX|<xZZo0S?Jw`aDZApKbQ@?W@
z^gZ%&=Qwd&g#Y5;Y27|nm6<3ZCZnoPN``VLw?=LLw<BpPfjXcuo$Mwyd>)c4``Wt$
zYQhqZk<Y@=yh#nL(A&^iP(asdKcm6Rq9p=%%E{}+RTX(HYCiAbUX{lR(!(NZW_u7-
zybkZgG4)6d?O7U#3{H8t$rF4b*bxDSfBXIa=BL&I%NX-i_vY}f>otfdrd(0CDiS$D
z`G$z}p^^@vjAA5zvTU6|?yCs79`2Z@;?Du?GvX5H;I*HSWR|lw@Lgr>=qTY9NV&G%
z!U^N;^-WW`z&o(=qU#=CL#D$4@F0r|Y(egXq8&sT#m6)MWCc&EJ(>j~EQY<W=^+bu
zMuRCU3ZlRMIWqD8ye(gKR2qIHy>AAYnfP-1f=8wJcKld=x;xpY#G9gvu!{*7@A;sn
zUcw82D<l+Hbe>TX*U>^~qr14tXwY1()sqWA<}9=clA+rrw0K9d@Fmr|7uejQ_||W%
zp<ARKckI@HK5O~ru>B;Q_4nt3hP;}ece-d>W(AtB{08Jev`}w&fagL!EEGan)Ujvw
zI{XS!5~;q+0c&ZIRXCE1Y*4EOEK)$FKVxS`pR*nedpQHdz)i#V{9G~|-^55d_%~r{
zT!$L$gc}8dCOii(jkEqRIRUF0&hcu0`}Q!r42RtXL+!Y=JE%V*+w?Dc>UE>nPc!K}
zCxn~#2pb+g6r)oW@uFn=ub2eT;oc=oLbhj0ZFRhxun5|s>C29NvVL`Rj8F=-<F+)8
z3v{pP1v-ltfvyH7MN1QUI~cPQ?M@XmWvra(rTa+kOn&6$&wC0bH!zaC8~$Wu6QWFW
z#qIeuJI9q-gp>pq1a?;Y0_d=C(ri(j4e4iLGn|p!-npnHqaZHKo}{H}ajc3h=FJ~0
z3daSfp`R<G%e+2JUZ#-Gco3>+;tKV)Guv7J5`tRLi4!b+J-LmF$F9Z4AL}vWCL@TV
zdJ$Nshodx@beIG!wM1MEDsd+`w=b!~qGmW0Qfpu!lybtI-9vVdBAoniIlvaALDw{<
zm$-gLo%SoKF!PFyb0cwih+YxjfE&3(=9<1UT5pE@oNEhDI|yIHg-|J0WO{%TsA*ge
zh(y`p_J*fspu1;o{Vr7W&(<@O6pD9m#!*a(cog{aDR_WJpr@N~&5|}%P}e%i9k_WJ
z8_I3DSG^p9O)(>PYb2JIG)1+(DY&Z(p@I*3o8KrzR=Kd^iP&j~GnnB#LT3k9XDB<F
zU?WmP{8bBcSb2Scj0v%mkU&$V-F4y|t*Z2@X{uFwCl_c(*EHu<%DyKNh2qxoIOO>e
z@&7_`4QCfMTNcF8z50KF8X!JSM6=k%o<0v_AT^VEq6-cLnmE53P5OZ|5BaT(_NKmv
zyyd@34+x(DUkL(DXr$=J>F?v%;v=N))iu#4x!WH(ICyPGw*SJyt?Fw+yWXhoB8=sG
z1DenMw03d*-L)o4`)vd&nl(}N=YC18u+@bBRwEEkF7Yi_tCLf8TC}-Iwwlwwbqv}=
zWM6R9AJn~LqO4I&`H!{b&!S(=%(%Gim^U(L@QmZ~*Yl!WNc4#kc?)CbRQvTNhk(Vq
zdXY0;_z%W8+%SiRSi^cF6ziE6{lQp%+^N_R4jzw--^mtzp{&aGFhYjJ)J(%$uag>h
z>3c8FpZs?QKxkAUuI4l+iaXP}11g7<X(3vK7@V+Qy)ATmq?~6#T;h#6L$sV9vF39y
zI`LxcL2lj|lxJ#ZOW%TQmvpP}yr}S@RFez@xR_{Zaya5Q`FBf}8=qg40F_jD`Of&{
z%ZbIDzG9xd^UTLDEjlR*)nA{hOzj*E)1JLDi7l+?Q%;paLhT+klX*r-?Uyc+fC%E#
zylgssnkg%)zkxP<VCjKXdV)fIs|Dya2#XyNUZYj$J^LAzWT*JZySqzMmN2cAEBJM$
ze{@FQJtIFZv+uMhOv85(Ih5oIQpfGM<bpVs_zwk<Mb?NdN(n_9NToKJ%q<;fQ{tDJ
zdvPVAymUT;@KHt?Q5=p~T&<zdymQn79e-8c{aAGm<zc&lVSxOYhY3D8BL-^Sgi=g^
zz8-}9G-R^)gC#=-WKN|Qu2}|`6zQWAg7%}%T{j0=ymx&3SO7V@t|PGw_bN?eTFLAT
zuBoqp42kd&E>=~oo9CxdMaPTeBDl<M3RtnPfXlP?bzY0=N<<-&vtm*WZ@Jro?ZAD{
z=epcM%Y&(U>*y6Gb2r1pkt4w~X5Bo-&v7AK&V_kaJt3+KE4?-`F-hw(32m|g)?XIH
zeJ&C;W*ecLcL1@{qsjs(0{x3L&A4Y17ttuj^+tbQ^XvwIqfLf{Xocp|`o}+_oDYJo
zva^-IJJ}I-G4DD=@;jSu=HSgW?Nt__n4T#L%Tr5>yk^@w{nvd9eyhVvpiMH`^MgPX
zyPqgDcZNPR+3jK)S_<b11cZm<-g;RWfKqF5d|Z!UxwIg^_jpw~tBe(_2av##T7J@q
zbaUlcU`=?4uN+mqzxy3ya=I6`-BLD2bF*v-xC8P_Jch!@CpOi5kE-Inb0s77TRl+>
zvv6!i&`gC~0g&GxC(`dWfyFbma}w-cwCE{FK2la`u&cz}FL#RG|9jM6827(L4P45v
z^PG3Wj`9+~$S70q8r8>IygTeka3RY9PV&QRfV5ShVS=%TW!tTGKe~A7BW<X{O|B8k
zh&}*-s?FJx!kd&$o`E8QhqlS>U;LVueh-dZlvh9`|J+o-oszzTt-bl}O>83+6Hp*)
zJa?d?1BC3h1y}iji+@_?11)$o=_;Qibhm_@p1Fh;?DaK!+K>jQ30xDGnyMMMXodh;
zjwJ1C#%TDO5q3|{PQrSVJ60rk-$R{NXj6O?rje3d^;~H6F$*Fu3*}Rlt?0os-SWrz
z8NQ)?rfqIa*Y_6y@#nc4Q!JLW#c!@;_mzpZu>`U!Awi~AN~pHQU`^hr;{J|&`DL=T
z5UpGu0@mcPJ9gC4;Bq0`OKYgeab4gFB`XwzHSMH-+PRS(YI`0@(IhvvjG^YNL8!#r
zb9txU>BXBY2tDP|4y`I?DR0sXtD={9yBjjMAVxGHip7pTMv}wVs6{k7)jpS95*<d9
zs8LV`j;)Im0!(}0ddk*Hth`XHuN`-rgZ(NiB=dk7{=kciT;i>O+Wr=cj*-R|v;1F_
z074)YhjJ>R_YuN+%`D=75&EH1#rdrE8zZfn7oKO8RP-%|s3yeR9}wg1bHoru^-|}7
zl#;oHB@O6ng#E05XhIKg?(!hH2%l{x@Z<gq->}Gglpa^vccqCRWcqS4$9FuGdHdh_
z2cok@#H*Ruk;wO}G}>Nf9K)1FyOP9|iF7~{Bj(%_UKYND*B6Tw-XD#8%l*k5<lZ9Y
zqK};IIf02A#X8(|tOmXAU2?3>_I$U=Q>}nqAa33)iGL#g6|Pzp_t59RC(1UvD^8!!
znpxM5Bk(`P^lk?^c-5MrQHZRW5|P%+*h{n_d0Sf*m-ZFY{k7(yVcs*BVTE%)(%5rD
zo&~}8w0$TbJAyJWoLrh<TAYS1inj+;N*g0|DoB-vuwu>{trm;Y8#?~XV3A_D$IG|X
z4uKe+g3cniVam;-)_*Ij>H7_S4sbw7zP&4Pk8RL5itF8~>R`G@=d^q>bjdm}-R(0=
zc&oJtJr!m}X0+%y?&eox0PjopaSwY;k9DvxXs<38<x1HX%F$CH!0&2%^IL1!e%Gbp
zxF(toKF!g6x8yhA?h<@L_Y242SLmErZPUq5A=JCr<a~RVBUih6li;5GcsomH%wsQp
z#s-iPK17<JNc8DRsfV8y`5~DU)NSCDe+DE^Cqq>4$MRlx&dvJfgf~>)VST!&F;E&f
z^WP8wdw3nor{*GswZKf)SPdFN@&1di+q)mV(e;9+$@YkEX-TvU)YGbsuESk(U5;?9
zsN2g?JF2&$W11`T0FXRi_o|m~!2Rr%uj_A5ruGd@E&xn`W=)}m`Q^ej`+!>l^qi8F
zDl<zAQ;rJqp&LCrm$C}PQkZ}z6h6{O{?5cG){Y&s#Jh(h{2bYJ3n4BO4lhnf+PP)7
z)&}`+l&AK}jmyN&!a68R)4tqw)0qE^#$4y>8Z(aY#-q}C-WOSuw!ogA?ja2)ks_8}
z49=B3iw^D*8PN{!tq?Xz>vvo)st|rYHP1Bf|7%fQuS21BlS^ylpm;}U^~r55zO8$>
zAUJp?mQL4!KK)5B6G|mVimLbgD?ZS&%t;Q55pEF(Z$mw_kX(CO@17iuHD4(3Ey!Hs
z<)dq~K<b~2Mf?88t2B(eEwZA|2#YD|*uZp`kyM8-;1p3uK32kEQm*utam-&zqi%&<
zf4LOcoN~Z{)K^d)X#-9%z+v1*OE&csMGKa~2*OR_Ri-2qj%LFbCRi~fKIOA}j>&el
zu46sC<;zF%ExH+3jX|&KV=Mk6ZD(V?;Is`_#Y05=y8O$#thsxjFG}ljmI^5F;_mDx
zME|ZER(q_scD$skq;Gb|wV#lHhi9=FoiOT1eq<KBsN~Ek;6YAt!@Q6<`^<sci>#VL
zFmvUVAocAHt<82p)DP8a=#QGZS16qXWG?ed3PKrH7u#u|(kz(zM`nr^<Bf=HBywZ;
zgYg*wyt$~cr2&1fLDAceZ^bS#_oo(%N`Vv2`Bu+Tv=dDDTOmjbt2h=QL1=>dJtLHO
z2rdWwXcx7@uIe#VTm*`-Y9rI$eX5%TJy6Tw)B7oMCrp_chxI%<Oz0OB=Qy}l`&I1;
z;}}4=Cg-#B*H7)a7f35LioKy<P>^a%0;8(mO4tGNH3e8Etxu_>D|K!tJO<R8JVCN;
zfMvD#4=k&l86aF>yx+0cOL+fGMgQ0G4`>gBEz(y9w{bqmfyCE;UyBzJ7vD-s{os1g
zoElLoE0K5#uLOuTkAejIArOL>Nc2M_aXT9RV)k}a-$4_m<j)89Gw)Bbyw5NNKh{S5
zSH|1m%T~`oNz)0_k^L#%NU;aJP+RJ8=-+ESyxyS(pBGj|bza_HZgYL)au9jL@IMJZ
z@X+JPM2-*Aj`5BWuC8JE5*_|3@<5LN1{{yK^hmRio=s`{uQXu7;RTP}K<Jr~C#w|u
z*~R@n_fzzLX31-w^fh!qA}8HvvM6E7gfQ9Ul67h*U&XS2S0q-b3aOwaKVovD18<Kg
zp9>bcLGujyNrVlmUYQEkVFRh@d=l!X?cY&1{7n&kCr)%T?lx_s*;2RF449Z)+W&Xy
zNSpTEZcI_9Nya;cgavb1@DR<BHl&&2BM)5$%EgkrfJDa7`1cR4-}WI{V;TjFl^&?n
zj5qX1CD{sphi|m%Ekl_;4xz_oi#sn?Mz{+zs2Q|^G(W(1X{C-jVJgH2xWtTx6B6oJ
z$l@d>h@LER=BY^D69J9jFCF_#qw<4!cgK9twu1Iv$sz92^(fD1P@*Fmo-ffUIzRa*
z4;jK+^<`?J(;Z0_-3uxy^>UpxN=jBnB{k|}<y5b$?a9V-+HoJw(&LXy1GsTl$4;9X
zanA+;x+Vpf6XKTS;F*}ANytPeb8M7Sl`w5GEK)`tNmRFTH28v0U$m1p+;mH<{^`XL
z%LXX$RkTC_L-oQfM^*1cjQ5!{gdWC|Au0uvU@QPuND+(iUjyodw9vp8RKLh?_z)Fy
zTYEy*1~R57XVdwMJuyVfoqc3o-7ZH^{eY@YL2|RpMk$3ohJEBM$s;7kyoD%FTL8LH
z%A#9^<ky!(WGOEWNt%M-y%+@V=%JqcC}kA0B4T&FqFGLK^X9RRW0<hmxfGYx3sGl`
zvb9kli*^cNd1es;4{AA}?sb)DZ>d>nC%!8DXA;2Q^&q#EU>$RebPzb*QYX>g0~~Eg
z&52`R5g$wv)YBHAz{6MzYTT-K4nF`TJLc3xtS&9?Y2PyuCx|ybJ471`yf^D8%z_S1
z>a9k3rjNRh=<1u^v(;o<&a?QQiaGd@K<U84i|!KY7CovJ%K0P+*ej;z(=JG2>ddJa
znvE0R8I$YP)|893sfvHXD)VZ}&|J0~-1Y|*a4g1S$snuZ={BFXMOFW{3Sps8pzH(j
zwEx4O=5cQYJx<d8-)j+z?Wqb*$=wY@dh3n&3<jWc|L0Se3vx``VrZ+FXXFcVu&-B|
z7K)$41rQ11JDKP)kRn}A%~W61xu31i6t;vy?#6rNvBJx6BS_ow(S0quDr`!!SWj8g
z2GUViX(IeZ%I$FBxM4VbLx=%CbA$&NaqABgywP>yhfdNW7(j$ifdd{ru|3B}=&}z6
z<WYQ&!L)fdNp@sgS`|>E<bN^x#abLS{r5(GDDA-tke(zQa}TYK$)NYTreUQ)!X(%O
zU33xGvK7i;>ikA#iJPF$=pw3)2ibvpoyyO%z*doEKFG;*w@9dR=GC~6(}%u}o_&(k
z#k=K;s;U&KTj&QYj$Sr(5F*?lTeMrA2rhNM_gA{{!3CFe#c(<0{&t(QKHU0@Az$aZ
z!r>>dx<J6EgB)+%%{V+xEpW!Q=V^5P-d_>#$K<#yvWmxIo_CV&S=3_r$K2a)_UAvQ
z6QxakHp$H*N3`T8pk2Q6+^elmibC^3bdIYuT3PLU5<TvzS&0`X6$?Okt!~c&xusUu
zr0-FLJ!4$*EmxVMQOAOuvqZ09JWb80ergX*j7Mu)a_H+fx?(l&#`HYDD#Is@xcmJz
zL><-RLG}hITXIm`aDpf#KaMKyfSG_L$&<`bYPaA_bTj3kXoOFEi%3mM1K`v&_NQk&
zlaWb$xR_#hU0xs_!kN#wOcugaSO?TH)V|Pd?CiBnI*+7djD1pKP(kMcg;yyzGUZRi
zXK3oOy=7iCt&UXX-NW(cPH@kbX-w}o<E<Fi9Fv1$z=MX=6=0m_59ZVylZPH9@s;U&
z*`gFUw?Y4s>k~H+MXH6nm-kK3wLZxN)Uj@ZMz*hIOdt*ryjuSusg_Eb`3l4B8ukhm
z3=q>lsK9cLNgL%&y=k>B(J2di<)p-oRuPNq7#bHIukg`)ed$Aj)rp!b`!=L5d2FDx
zVFT^r#jj=`#XmJ(Da+Kqw*1(>Q|gzMd{%3(yK>;X)kecd%ayNCHeK3yYW@1OIC*4x
zI`tg2V(47jQ~13|GP^fpV9!8zlRG}+eHt|kA`>F4AD_OD$y#4HX_D_gIJ=BiAn(zj
zbBdzL+bVL?+NK}wL)3w|3zPv`i>NqX+6n&KrH?jH3F=)BEkuKB0`a6KIF*`%*AH^m
zh;R=KSs3rZm7K#0(P6)a;{MM01)Yt@xwqKe{VCrb1sO~ZI6{Pmodpe}xWK)lZIDR!
zLrMP6iu3UL5X$bw_Mz8%pG4a#6R5?dUMjr%hS$Am<>5C5;)tWbkAWnFcZ$lRK#deL
z?YBWR{RV@ieE14X(+t=|E)X9m@ylF+2!`9c?boa_+%Z(M{_lf=v6uq!7?`JM4=Z!J
z$&k7QFLKUmh``^)>IOYSPfbirsFltwECK@)c6<U;M^=@xCTF@~4Xe^1UQa*>>0@`6
z+x4+$1*834)$%{6#tsAlC}`f(5GP`~9RHTTdZtjkFxFw))oZHwv?e#<+~k@yJ+WzR
zFIJi7;0YD*M2<#pYAw#P0|Z3;K#tJ>FX!-Idk$J5dKQ<H+8A5Tj&>)>7uM`{sBoPX
z5!|FlNpyH0`s&Rp!O%c*JAJ5Mct5E?_WX`$yfMgbr4`udk^PFNRarjPJsBgo+Wt5B
z8-quyYKG_pq4+6uOnGqN9?UB23Cq^)KXK?=B0=TR)_XoX?i*@^7rKtvq76N#3!+B}
zc)khO)c7J<&MtM52i|DU;K1TdXu~IMUthskf$&SK3WAB9!;6lfr1qq*W}RePYvYjn
znA3YXBPe?OSx6X_I5-raG;rH^HKH$!pkaca?}j95Trecq0Ln!TxC2&6c`~xFd}~di
zdQZS!0y&{wgflVT_^AMMfPlX-;O4L%sTudwyyvSp_!_Pi#f*9hLVv7dCyjZd7t+VV
z651n5c4-c6qlvVh(8+bM-h*FGfdPP+=wHd8(Yo@?*xg5;%@`+|2D?c6RttF7WwCqZ
zN39<d<C9%U<eK&3=+4Nv+Mh2X)|K~+e<bYzd9>%P9wFqk&tYdY>9+!Rp|}_ReZkaZ
z6|FN?aQ$T3?h3>?8K_hw?6{if#CbmWegB7Kx5K@|xxIMzO1Rtv-{dM3tKn;n>))}g
zg%_|v`b6f!O22<SS_N~BfVPK`34$jeQ>v2e6X9eXlu;F^r5obBB;o)`rqDTTUO;09
zG!{DCR+<lEZ=yOsB@I`rBukUqcl=lr)#XBt-*z~irkV0~#1-l{h8E^aEu^+2`koqd
z)xTxnc1$!k{*h$?yGUGxlU>9F#7e2KNhhoBzVXT0eIDkI&8jj@rwMrdkjSR&^1!ag
ze~W|NXg6$_zcC^yct&lygx0zDQ6}o^NWZ3r>c($gWh0W<F$@`XlZs$ac|ii%o$x<8
z2QGrb?|~)*BrTE<6a;h<1ED1tkqBxZ6Lojdwlw~uiIZ<Xl3fyP(R{Sdx5Nz_k>-UI
zvY@&;x399|blpVQ_ne7xk&~Lg&zj~PG((!GVSY5q_mrQok+R=TUnj$yDn0=?+XyT^
zi5@@jgW{?6tIvaUgy+g%tsGu#Fa3^FD9(m8X+yOG1$29=%uUFEub~YSdd;||$i0{)
zE}IjkAz4&=TvPmGv(ULt$=a(PKbxoa9IqG|mmx+&)2L3v>aI!s_@0|1zg5n3z;%X5
zCI+?4zpWynht-MU(hI*k@q7@umM{{OVp|=@U@^qGaIN1A$+&`F-~p%`aX$QdVs;zN
zMDM2`K|fE)Js(9t{+D7ya@m>U2Ke!^%Bi6X>NU?#<FWemUN`dT*wGD;y6QDx08u<+
z-}|WOIBRJtwF0MRc|f9_7)e$?aWu^_jX7vFe_O=*nc(Cx6ln;S?tqvO)V(aQHCcP)
z8OJfgIq(}*J-w?=sX3s6N)L2i!ps4QofhT*u0Mk+dj@i23y{6uqqv!@w(PB_0mWx8
z`|n;l88&0Zjh9=xockWM5>?`x$M<|ad~MmOZK*zAKR<&nLMl*>X1GGdBakm9Q~h?Y
zzG_jDX*HFuUW2Gp_0qjLz(;(Auk7~VyGX^nn()CcfaXCHLLAtQav`7mw7CvHKcPM}
z8+P?dgVkTw&+bK|`m_{+GDzc3-GMz&90jHbjQjS)ES!3Rv`0pN(=J3V`TBz0%U#os
z_Y8d>G<pofG7F@HA`QeqV413NqeHD5?ry=Ax(GXWLIN6xu9qL@{a26H?nN>xC7&;f
zI3-M^>F{}Ws~qr7Aq8ymuk(`|JOGBLCqE74n#`>V*~aS__I2$s$cjVU!lqpfN;lQ_
z21+_$%cf^z+J-Pv*5>>}QJ%sdAAUQCKnoNxr1hb~)E%_S(`(Z)l^$!gS#MINx}ZRf
zC0=rzwcQiy-$IeV4JDV9-g)wlT|0zfUX+x69m9mYDBS*5Swkj`e?I@Dc`Id)`occl
zUZedQXWXD})ubV*2#uX9-YViO)qB>MYwtGyq^hTPHYhdJO~=eV*g9Qx#3N(gh#1)X
zYSqM-tkp)=#i%+Gwy3$Pi^u%5)_&VfyHQ6_B(3|`aDxV=W@WdIKKKlAV^r~-*uv<S
zk1ry_n92T|6Vi`s?|dage%`wPetMX{V$c8ht>LJWcs)elZiWb8g3oTS!oZGWJw;~I
zDEe4lT&}nVuJzTb^JNxwVzoh1C6g7s7eSlE*FfcMaPlmqw0f>(DX`MuHL*M$cd{EP
zaY1p#a}ZVqlESSO{E>OD*6&gPApYwB8XChV)8ZK7^YB-=jPjH~r>2_&k9B(kZLO1K
zk2r<jlkHW@*aFIbIXiphPhA#9qU70`uSG^FOLa`rO`dxr28VI@p&L2-&XX4{P%j*a
zFNv}u1(7OdDa_>+G*yk>i%_i})sp6Mtg1{5!JpBFaTbL?OB~L9P`57nD&GB(H#iXy
zCY`UF!~!05;PsXrW;gSRDTWEaHu`hx?@))R4meSP%qw<u-a7u30>z3nbqUSmA#oAb
z`571Yu+|@&c7*()38;c9!N>Tn7}iL~OG%_ZR*S_G@F=v&aS%b!&J2Agdhb|yQAO0_
zfm#jUlI=w{#k(dyA4IEUawp2L(uWpl&HbYb&=c}E^GfY$`z-~Ep5x<k!dp+(DpPGn
z+DCDayWhH6-!nClY{q&8rpQ1N<t=<WA+Eo%Xdd3FAx=Hm`s9EQlvXp)+j?}Jc&N6v
zlz8Y=We6pdxF{!S0Do@gO$ND9qiLwa9RUKK*}w=Hzo*V-Qds)C-`^@P7LU(#`e}w=
z7C%)7{(V@d>8@QSy{>k0=eND^76HS?0v4JgHVIjmQNcFL6(1EcCdv(IrHuwj4Ikzv
zV&AS7M}LTv4Xzd=7Uf>9%cvEzV{g`-j}4-3E6WNE<_*4^w7!Pxy7=HDpp*A5{m|@a
zP^WGs89vRMjv?&??>)5dP78WfdZ9?e*AikZob$J1ui_mRR7v2iRm7ieA$HF;q^;<T
zj&oY;(WT-xs6;R+dM0CM>NIa~ll{<)H<>f%w-{U-7^b*d&<h&#^)z$|?Kh#lR5PX-
z+pIkL7wh~d<aC_-u8d}6vSq^s=pv8cXEIxpw2`nOI10i->6A;&;p11}UvViH?V?ZM
zG2x@f9PV!N1l^CkT*<H--Gr8(X9H9^z3HmQ&bwEP6Ww;dR|1)Nhq)8SF*H_#;>R>&
zmG43K&FdA678adQoGyc#Xvx}78aAQLFOeg-;S*giN+8=>!8Bl>>MD`~opqR5tZY`~
zUW?9iOu)ea-OU9aAsA1N)*jIddKOWn?T$+`LBe8`;m$hu>RtehFOE%>WSLb(k1Z7-
z*LiK=0g3Ox1T3AJ<RK8cIdR&KqQvb$a&U|yvuqYOQmbn?;`DsPc-3XIy6m|)4YqXR
zw;mlDu!io^=yW!n6(mf78vxKY?CfFen9HJt1RfISeSQn5)_k;EZG5O+3vz5n2F+>7
z5tC{8dO<(V?X7b;l-h6$Ql(++*bC><DQQsf9?`R$Iz3OM>q|XPM}uUapF^~hyRZ@d
zhh|Q=l9_#$AHoYO%t~2&;LDlox1J5<UeD0>?y9&El&sq|l9%p=$u=t#&kb7xfPnI>
z0=0<&KE_RnR7XuF+4A=+nzG4ug{}zsW$ntcX<y&JGZhL+>dA-wmh}`wH%1a|l231n
zntTi%XJ2GDHO@Z6{&d)uzYx)y4`Pam^dqIGqHSo%_3#Oz4WL*`<IA`Xvf;^J3+Z9T
z8t*Us4Y9VGb}QBsj!C__%Wztqxq>LdSKkdlo76fyEx`h-jSFvSmnG%(dPq-cW<bw}
z!;=T<vYXpK3uf<aotoFt1&OJs=(9rEgm|q+%A^+N&XttbS;6e@Tc?E86LT&9=p{ZQ
zMCWg}5f@C9SJ<!5_L*W$%)N)xH3$9h!jBVTwC1nhfQIpNB|u2>@l9H~@pi&OI8zfH
zppph1wW<IwB^Pfd7)QK|99!|>p%jJsRtu0;QwwT?|3uv9%ta-Y#a>ri&8cqqv4V6r
zsd%;6V+|gkOpHtp$NHJP{~ARi1(RB4W>}OTT5PbjxLW)cP;Z#z2Ju^|w*9LRWX0$i
z?ph@KW4A{|pl|}mjFEFG`&)c8XUICo+*{tp?EDhyF65QlVBDEV@b0;ic|PO94e;dn
zn;<f^^hR<9@92->WsY+iZawC-3Zw0*6_Mc8Id||vrnfL#{SV~iwio(uS8yA#1vOI{
zAv_&f9J(&mBbJfMydS8U(oj{*B9r!ms&|0~QuSWoe`vXTH>0cF=$jfj-@Uj4+${nd
z!<yZ-nQ3BM4Q9M|X%i61hE1N57+X+__0(`qqwb_1=?M31xw+Bmg!Jj*`TCXA2Sy6#
z&I&`Q$$3RmODBSRnn_=~OLzsX7Ecd4baN!PK#Q&)#9J|3hhB(e_cGV-L+CL^?#R02
zk>S_y<@_q6(WABL@Fr--fR6Cox7w$Tpj_@kl%8Sds4zYpKVZDN#$%)iefZY3`C|31
z-XLPk952`p)cEg2@4jbYDY=O#oHHHd&t2~~OlHFGDejzQp9evqY6oqU!!SF!`Br24
zUpVc1=5?Fm-zuYyC304ehSJ1)$Yz-hX@7+!4=$uZ1RSV(DTqEDp7G$aVa?=aNY!la
z^O=x5k3Ak4<t>lEy%I(Qjy=V^@z3b16y1QQvZb1n_bH<3J&uDzncf7#z5Rr5FIml}
z2A2dMBfFD+v9@~<LbrdWW3GdhSBNL!5j2TdkQVd18-H{nwqv!J6}z`<4Pw4bi3BXB
z%8OESr{`RFa|!y*Bi;Pbw4k2$p&s-Yf6P9UzNO^&;^=L=zQyGC6pWN@aBB3p!svJt
zyG0-e<x=%H({v(e?P>vT{5r0;C?@z(hI{_@Et2~;5=)l<9Ze~Wc8^o!aoet5A=jdz
z80uQwmQqsb){M*U5D7ijiiP~`EMnj%upqyZACcyp5cYSdw1((MI>{{n07rb*#BMQ}
zEzW|sLOp9%_Vs^l=$Vqv?ibR~MLG7wd+Jl)p~oORPR<fH!UGjinWKiCtc?j^m-j9y
zHE#|}1+`wKv}>%{vxTZu`Ejx>Y3jns6U8=8_;ZsY->O<t{_25V>c0jsv%E|S{cwJG
zd_?Pk8b~nd5P0u!aauPub91w-D0*2q;$GIjVM2%hCIZS{T^IMbNd=K}IAA~%jfsZw
z0VFxa%rplJ7`2l>hZ28o$LRyhYqIfel4i0y`b|+F^;)eU%g#>{2^N}A&T_L3dL+*o
zsn8-;{Oiyc7CiuoZ~5`Gl1Fp|E->=9|D>Q)-B0SGq^WYAn^DCBushl|TmkSchZIi(
z;HX^=PYACqQF8tzf)2{TWNot9@kV$HRn%zsBhY-o#sp>xJy+%h3AB-3X8hw-nJo8`
zob?@(nUn*QDo`md`B|@{G*bAHLgIm29_di&ITtgNe`v*NUaWlq`@Mm2Q|@T_&#B%t
z`MbCB`1GBngTU*Ju0EfBft#jKoaMEJZjLkGQYl)HIJ;Zn9y^QHeq(^2-<TJCBCF>+
zXp*L?`Rk0_6%e=^!mh3a52-?elLOM`!2M4j#<h?2kf6fv0J)x8*Q#0b=`wm@M!VoU
z0fwH6Jqq^!tGboynRb}6%yM>v!X>5UM7{VPgU#`t7h*<t_J-M!Xf8^?SkVyr>xpf3
zdb@$@QwnE|+(}s2=AnnG?gv?F%7i!wRNh>rWGIMoNAc2;-vt0BWfR{iHGdJzj?q?W
z`Yupl?ACVddG58jX#i~GHd^J8xq~@!e&Wq`C~cE;!o5#TMk9IQCCB3h6d10ottPs+
z8r+%2LARj0TgpWDCgxuGD}$9B;Xg5tB>V0q>>aUU-tPgSR6Fi2E`|!q5o~F7nif~w
zcQT{)Z_{O>vOiN`0owM3j?oKpX`YU$hg?vM5G>gtU}0>H6{7p$-*^3S3y6HdGg;aC
z!mz}P#heCSe+l|<=3S%Mam9NnMYWTyTgnEmqKnov(PdFELg1;2AavbAixzRIsxPC0
z^(b}h*XjFklLiwT7b>JdU<tpIPu`f$82oTJtD#ZPlTjXvC6TRn%0mb?5>T=g9~ff&
z18{V{4SJrcU-Kv?GCHkmhiOXwNgH7S?l)6is(sLmsyI{29aFY`eesw~)ve=35bYM~
zLc_piwBRchZc+A+7^7?=#LzgWzq&jDSpido2zmEIGnV6$);(TBaIGqdjvvx@p0iDy
z*j>s>#H(r2<IAJ>aZSbm?n}dCsL>~c6pETxGF5ht<)zs5Uf&6nzMZHl7mq<>yEg*J
z1Z<umJMJ2Yu0_W68sOMc-m67S`W(Kk)zuQHAsf*RzK1SeL0=~^^UK<|1C&uS_KZA>
zE6)60dR`{4%ITBu8@gWww%K}eQM2O1_cgy+cI^hQgn8720r+5D%Vrx~X0O?&xY)Q9
z5d5XUvTlG5Lh?MIVd90LG`o6w6M&evLHU*gtsuTU^Yp~pPWII_VJ9YCKXpIhZn9t3
z%LPSe{DpzYOJLit9#)NvtBZ^a60U*<Gy>$q_doEG8g-j;j%3o6lLnIQGxdSy0Rj(r
zzPgFIU;UMzfATO%&*(gQT6kt8;2sHUz4J6;Hmnif_#%=W+<AU`d67o=yWBp^-uyS$
z_GK9L#^g*))Mc+k648vQ&G`ybRG5$ByGk3TxGt^$U*z@v1E2@}_&hqw`eaZD$%#ZH
z#gfi~EGP(!F!`)g7MTcC?P|D%e)H=P%YXj`VblTA!9>o*P7>seIMj}XOTkhO^A-}9
zAhkW8;eAjJU<}w<Z~RsS`Hkg`9>)~1{sW#keA`4sPd`CshzsC{6SEFR8xz|EmsB|Y
z&&!$7Un0BNeNxllz2UpxFJfh{-*+yQ8%${BSs>nvmkrdghbB%A+#8>~^^fQOMCJa9
zgCF6BqQwXXNaz43IAszf{oxPr3)S`Z$$Y?r;+@=>fLus<IPa3Hf4lwX`*HJr3smRN
zvq7G3<_$bZ)~S3CI)k_95YP&W_BSSgum2Yg9$j;7Swg^}KSroY8oK<W?wE80vl!36
z&*Y9_gSYwpOLACeE7`@YO7U6JD_*|}v{UseQq7TYRm-J{vA(Y;;-$BP@HaRmEIFEG
z(L6pjd;6d=P^apYab<MIhK%AtE9Nfn9;;sjz@!{$x}0cz{L@E@i<Nl|GeWaV$C3$N
zXey090{-^VY)2xH3siM9c)<^P3tsTU<`}=1?m>bEydRw<e+gf?f50r}@AoTtyL4K9
z`P%>QLHYcXNH`4a`S~B3Cqt3Q8?g(ceUbkjd9SpZ422`(4kl*IbqwT9%-uv!BT`FT
z1sV!t_7aRQv-NbFn8>^BsJ4TOr!D?n+nhIx=pOHmo{~O?-%%boa!&LFo(~0w$FTcX
z*i3D@E}i=`YQE2OhiI<q-htXCPMi(~4G1Np<VVGZltLP-pigs}3V=Uf1*8iF`LU0x
zhKv?G+@SJ(_7@&3Me`lou^cR>lI2axfm1b?<nRrcQFs#V@&QnHr=_RzJ@faF(QLz&
z?4uyv-!{6%xPB?9sSExa-_RUJBF5B%+*ev1sRMsY1kHEQrHztrGUXs{B}yK2B!Woo
z$StMiZjBqL+stN;-?^yWj!jOBdEdH8;g8~2Kd$JVPkxC1XEKde8^aEzAln2@MhgMa
z)egYFQxoyt4TeWXL3$4(@3WoZ6;<Zrfuw;9nx3nTL=CU6L(ee0eFYq&ajx5TvxFe<
z^21X>s06+Qmrd{>FysM6spS>Af^6q4iFDpvp&1|@&W#JXAc-pl1aB8@LyZL7F(qB8
z`Q&^oT`lg(>XQ>d{wSx!BjcGx`!BZ!ZL@Vj<xzG228H4%?PFvTse+=Gm+_jokmuJP
z$rc-oR{3%)z!H#I=Cs{0PrPa$)2Q;6o82xW6`Cm}-M&y59@j+o=sh~Ogjqy3j2LY>
zS~pxn>Ww%@YM&6dwk`vQ4gRV_)*bgNI9<y^YC~&#Bv}}0?^Y36>Rx@1q7Art_FWyb
z7i3BJGV}N~`@otFz1A8qe{VI5JUK80NH2`c3H;%SAb$MwOpj)Edz}Bb%tPzf$CR<B
z93uH}i<?utN$=#GI`Zz$k-OlL)D+-+WheeaHmYtCBn=)_=&PI?h*GjxfIM4mjU`qf
zL?{F$pXZ13;$nyr{aiVaX}7D$M7P}zcX!9!XeV%Bp)D<IS##4)h~!-{aKy5{{3hz;
z19n!Xtf?19L-_ob{voQRX4eS<S<w9$Opv`#@yZTweqxg)rg}(pdXvJ`ZIE$_s0okX
zT{75jt9Zp<GD^7s$7uexZ!F)V=jb6RIsbaTz7wDIXdS?r+${=!q=fo&O@`ETKM1w+
zV-I28E5F~8G*a-;_1q+ovF)H}^a0NIb-qP`KMymuZOs}$o-$Uf1n=l-&dioU-~j?=
zGqs#il=*Y>ixa!kLok=iJYD7-g%_b@qv0jtBFawT8yKS<eF7Z%_T4IvW@ns1@vVh&
zBpJW0VjbCstK7C=7M_T0ar)un=d)Ph^g(6NlhKjqH+0H)w5fJd4Qo>{A<n|NXM|p1
zryHm93;&F$_5sb03)#!hj1Jmn`;uy3(C8|hNu@f1^Aj<jU*qZ8>+c1Efy(YncH8<(
zMUGd*<7;?~s2_HrdC2=<9%W;XHg#=YK#Z50NmR~aBhdBsQKbhvK0o_k-CcQDlH0mR
zYoF2%TDwwn$ab36ZZK2PoZFVxX|fb>2$f9DoYKTGM6->RrcGvM4rMzHL`VZQ2TThZ
zG&B1Y#i6o7K^!tP2jH&H>YUC#&%V$7>psuD{Nn?BtZ#j5z3Ux*@B4ctciGO2INz_B
zjv+h;Y1G+y_dBBxbuju{F>7F}QLhjCJk%GfJI?{;So<h)=HXaX5T)<f+Qrg8$k}!o
z#WER!cxVetT9oYw%jSmV)rrj==aI~K8KbN9!o%*qOlY2#vIA*9p+cw&%Gs7(uETG-
z2vjp-w}ax*@b-VmhRTbk+s42L2;Snk0Q-MzqWJYKy)XQ;j#lFxrE?<!lhFwf`tNC*
zU5ZC1Hn~T_L)T9L3gwVAN^)Iz779Z<rMGwkw($Y@e1P)*`7ci%Lf6*Bg-LFXRb{va
zY}l|GeD1?%%nz;uSi5`V;<+JKE0~h#-!m-mn}4QTZvE}$ERh02J>OIn*x2$`>VuA0
zsnAk2ojKb}{7#tHPXyook00=x0gq1Iv4DXb3XVF<((-=|`&f+aO^EBY`LsIgLfBrs
zxuBouKl0Bs1RzNEhGFS<^W3Bu!uv0?nUI$Ha@brlh&us65e+@3PGWR5^JjJvI1Q`e
zm17}+aTJ{r-k1|fM6eOCKa<nn(i(GbWZ&N$1O2j!D0URv?+Ga&oU%GM8+_|FFc4VG
zIjHkLoe{G{R7x&XDl8rwLqbqL?BDq~Gyxv!pQ8(^5f+;klIN^beqkrrp2UqUpwo+G
zS==^B8_j4cDu?{OWhq6A+NUuhe^sJroj5|tYm5AuY?vnS-22r5)d7qDFG~+PNc^Rx
zpCy6HfRF1YW8YtyffrK69-Oo&*RY^9Qzy1bW`#XSW`fGc0k}B^y2wobbT7JxjUVJ$
z-*B%jvplR!v?+`WCQN(<ta+v@4~TU#)WKyc+<uJY#B}6uIn`OFkK~1OLXo)ft*WN2
zlWkyYQm`MeTHF==Qw?NI+j!OTd~;OOqq}$voJ+0$Wf$0^yZuwkbrL=W<d<{C9Z)=&
zX~@_MZ{D8(5jS<9XDmq8`CERkTWOJJXZd>_xHK^Yeh%36_UJh=PdY^Lvi4%*9pJzU
z!6yT0zk$Ztd3nEN<j&l-O3M9{(;a$m0Kqy-rG6v+M>%ZF$6^H-_466@ptU~}|H9MW
zd=?-Rruwleg38jkq3*)G(k#<i&xFX<BzW#?X8)Nlfw}#00fa1JHuM+-w1dJ<NQY|P
zylO7_e-?22NMA7rtfA?K5zA71EH7R7qj3hq`Q%&z?I@m8#yRk&{fqhUj|fq*(z;5T
zOE}>L3@tUaCGg!%|5MJiSf}L?jp=wg?FUKQuXwaBy~K-mfj8Ape)K{|<&_NOo6~{P
z|K00gPXxz!PRVe<{5x^Eo;_h~ft@E{EpqAWL1L7T>-hnE1H@iIfG$iNXZ#&Vqpy>M
zg}G#>?%G1{?E!$fPkQM}XQ?Z}!{N6993<7cnhtM`_LVh0<71;je80rVTtBk6_vU3g
zbcb2+Qh71?!8Y{dN?rJIwJ*UNXq;zM;^i#1X-epItBt6!JyG~omnFkrTIPATzBMFG
zYz?xuJCY0#T(Io`QvyutZ)utGkhk#BB3uL%>2dD(Isnck^-mRk&((k#_Y?Z3?m{-^
zFI!55kNYKB-^)5+8FYsv{gR)M{Hl>_jEu{i_z_dvgihDh@Z_%cZdB8$M}-prvrz|G
z9`ho!_`|qR*e$#T3I~uoZe>u?Y3cNRr68|ZL17o6vziR&^s~Vj$e57zTi$Aa^ZF#0
z$fV~Pn=E0?%Q|g$^o#b??HvosGUwVx{8emb473<%qW9IY1LfyjGr0U5*LGnQsBf&G
z2L|i5ylIEb(cYdqZ5IyB<Omy)EBwF>+$?{#PbIh$dSf<wR!|elrY{OHK{YZqMSe)S
z>*sCYiVZ823Yzf&jYIi`9Im<Pvh+GUe#yqH8B7?hF5={DG;&vk5o%eNEE=k%nq%q3
zc?XEG!w2T%#V=a&MaYEy(5W04U+$X&q#En-xA9Gi#7VNBS3GqRZCKj+u~_I1WpSd|
zCPgrKWnsFbpfdtUqMJDIlTQ7Fss8@CoJdfwQUP0o-D~9$KiF|hHDry_XPiko$Q33^
zz)mXuTYw`;BA;r(qLwA~{+nPT%iUm9xS}Qn+vkDvB8E9A3@69#m-It)H2WnZc#q90
zFaaOj4=EF&O#4|Ypf{Iedq{Y~qdlv0Yhzy^;s(Dk6T(l$r-$w=71Qbx=6ub~OF+XJ
zAHl78JcE^9Ho-Nr4?uPDwUljipAnEhrRqY?=EdYwk29`Eq&*+d;*9eRDg_UOMw8WW
zFH4pwtjm0@z1?aH8oe=}IiNR>Bpak9v?|AAWUo(8ii1c+>*mT07U?ON>cV-+`(xL)
z1e_P71T*r*Z&LfGP)mC&=u&mnF4!4iEd&GdzGE2zzfQG;tG1SThPY=;y3}ruF-+cd
z)A)A<8ut?O+S#I7I7pMnOK&2p={R?3%N#m~7&ibnrE9h9rwvynB}Ab?Owq@<d4`n&
zurXIdObgCw#+M=wtC2~_r(S}g5u!5cFWHQW2NX*hKIqG?8?MBfOxl|f%Vvb&Xdb^F
zYvVr~XO6LJPr95JP3tvbNq1i5fnyo02Io2EqLg>SIqt@#Z9gPbV!SU66bpI6FT(c#
zSq3P5yhme`nu@p?%D}P@NjroyTIa&^JDlH74I`DROtfVDLx+f1WqbH|rVK8DmeZQv
zWq<yzq+?qUU4mBR)7<1*c37aF2!?f03&1YgBMZK%Y2?#0lIcjvkT4y}@n75#HQGm6
zs_E}id+$L3{v)WkWqa6E;y<<os4atcu_+BGvFH%VtFcO=!@+lObmgoy@S_4JPs%3e
z6gLY1l#d^j+*E6h{pvQkd3`Si{^t1vDDm3M3N0g`ev$g}g<y0B0kr*Lp4{7r8J(G{
zJaV}7CkzBxK>Kh@7@DdAA3mOkJ^+qwpf%s$e(w?U4KV?Yi@>R7QZ0{x-d_Ys44f|H
zEe5mV;Xqr?NYIv3N9E=4T>cV}mDL45_nfRLusqSatzMxUZ>^8VhgebcG>^53amP0>
z+<R_42)9nmvAQ)KAt9-0FYs){f>6_s;oGVZY~qB{w@rBPG^?-2P0ERpXWh~6u+DX_
z4W4tG7=LuTN%_Jsbx`a9f?B(3Dj@2KCuakws;=thO`ugL3)F8+e35pM%HYK2hUDCP
z@vaqKk&BNnX1fuiB|(t)CJQxQ+CHkxLBbMNsB13@b~JP~x@(KEiGFym-e1yAZn2$g
z-m|q`P$o11%c0^2kWg~qr+DfQ2*<ayDAL6ssnddrFR}5vhp)dQ(L?%f|Fobz>R_Wr
zY~QVUNqdJl<nY73tFJ9h4zc}@gAT=lWE(f5r6>L*jHDdj38Ne^1agzy@>Z?sBn*Lg
zD8PEx2M1f4*8nO+)p<U1@1OcbYY2uqcm}Y|oQ$apRwm+X^D?r3^9G;RqgH_fD98}F
z0C002cigGy<y<WW61GVq&bSM8E}H|Taz*CBYgfbL^Dh7cS56UuCaIJ*O%^%-5vT@;
z*3<<}Juku*clQXuo_umDAH*Gzq?6sPe>i6UZQc^(P?}r3&0Ztob`x@hEGQ$L8;A1-
z#F)1G$#Z59B>cxEF91EmgqjL?Qf*K<)yiNB{%X$?;LcWmMG7r%g5E~qbG?l^O!d!f
z3gwA|H2GZkGX9nCNCS8SVt6GD@aHcPzjcbp<0ta@BY*F{Vyw6zdN2m%_@>?#t8+k%
z164xx6QC2pnsWWO_8{HnLJ$FHEQe?A8V$sUaTef_559W+@QwcWUO8RaPz%;+!<>aj
zv`2A;=Y-S3;ibLtqGKa5+GcNSt0}s7d(k)?sOyyquN>o^CO+Jt^Fm|lx-bBEe)06u
zUaa(>gee+o)WB1(MW$)H=X%wG8emY<vx^WVc`dxARL2T$z%VZMF;QI^z<YCTEXh^9
zf9X*b_w=M4vvD^f_toekdG9>!k-#!l#OC#v(=UR$2YwQ}Vp0b%dKaHuRdghcVYbK8
zaTqP#y+BtNI47Q<70H_Z)6HRp?4Xd_`7T==5s9t49}F5PUY<P|kx%x1Ww;k%6UnAe
zQG^Sq0a*2tI9Y>=Oku@G{3T<c%L34REx@7d!`g(`)FnV7QfC*5Xy(ZKhrtTH=)pU7
z=9`LTzMv~(8hM31b;eS>inzisc)8XTE=W(Rp@*oKOP++vvI>%=Tf+AcMufW{&-Th_
z8>IqIjJF0KWSJR#A}flw8zg9<jpQqu7!PVp<)Ou!ibL6sIeZMMg3_~MrZzN<kqMp(
zV;GYclZZ0J#&b|u(<TZ(A1oL6MK;Ecvk);#>S1QG8xrnFq+7<WhZ2Ge0J??b!~Y{d
z_MCfLslWuww38x2qV8{b9J~LHLAgeXiKDGCtE$d&VwE_woHFv9l1ue6ape7Kprgkv
zJvLtE@L4)e?aHg<h&W1~58uvef4!44lC-y$@pgeMsCv*yX~B%&yCN;+u2S0xDIu^{
zsUVfZ!bY{J*3QgaMN4h*I?@JHAO}@a3W{BkPQMrTs~lSr*Q<eFZ*jvo;u8f`q<u@C
zLOV9|#mvimA8$xyGMW}B@l8V@(9(U4s?({I!s@_Z1iR$FGU;@Qeqv}fMI5&BmY9ZN
zO8b?3Po1{}!{nnw5`1*L`=3&~`l3(101^fdez39Bn@&x)+Ai;_p|G8XC$EDA+f3_Y
zsubz^&PNoJf6JM<Go1i^>8|<NUeyDdi0^2feC*3#Z;+;hdUu@r#D=1t!Id%9+CQD3
zPIp&i(v89I#rb+cQ~hYE*n9VL3w(J_yT;b;gtR5sPBE2VDg>{IruT04vb#H7w3Q^S
zrah=Tsv`z1GEG0Cke>O&**lyW&m1{V3@HLyx%dE-k=Bugm0Nph=><oXPj4+N#2Yk>
zLHk){L!wwDp$ucPkJog^%n|hpoClXx3iE^goX{j_G+AnfbQu+hhpDATAcIEoBGg)u
zvJbO9nmI;c=*7dEZDsqR^ZiWY27eUQI3nDfmW-B6k+W68+zV^N8~n{^#$~PWoE4Q;
z`K*uklb9h%O4B!gddvudBw6jPvQ%nmhl`IU>hMb6_bQ<Nu+P#q!AU|v)4@dj%Qf|!
zO2)d2C{&e~y<i2dw`+7B1tPfT;X_K%ykP&nKIbb@`z3pko~}k+y@dOn#^Y;!Hj9iS
zsKVc&0MA{r8s>C5#$P_vqJcNb{-6=N2lR#EYx+Nuo{%Jx=MRdkzKp!#DCIo9&v|m6
zd)-(u1r%uW&Vlas#7eT*#;&T~t<;M^{A}4Z0{%2!@l7K7Z*@#R$`gxDKrN`*O=Ew6
zv)ooaWi-KAnA2K9N}t_J2$iTqbkk+59&?l7c#peT%nYy&CN$$MwD0Rc?R1w)ViM5n
zAJ13!lW30fOADnq3{LmxIz;fJ4csSEDyjIzv1{q<qUjdo^ZEofQ~qmE8l!W=JHeAY
z;DMezBdwCcBD@Fe*qyEIX-g11a(bwr)=Mo-f2|V2&{q3u`Tcad@l({OA^q6Q0O7NA
z>>kbA@mi*P+C?2GgLaHv5qF?*_}7%fVA!U9!IPw#1av+%V6BE>utA%qUYm9hSmmS@
z_Vt*k?S%IV>A4Gznv35}${E7xhbc$nk2JatP%8FrRdIU`mX4&8K~Y5waXu}DGDT+|
zg0<!kXiR01RkL+H1RXQ86cHo^17q5N+<g@WP(JV3jD%K*jz-5xa3HKj&8{gH5?#^-
zpU8nvy<45<eK1scHK9N)?X&z~;gcVjanT9gqD2SfzWC>PO7DyvdBJzKI`;B6YrU;Q
zva!09_9w}l@$)^AS*w8#KH}kf<hpM&z;{&!?$Y)wuLF90U)lT;^5MLhuKbf1qDW<w
zp17+o>Nn5=(uRW9PSjLP+o#x_-~nCX(V{VxgbrCrI3pXbqloiqV-LVtE(f;a>sfYd
zvtDXnO%38sfGmGIfhVZON~Us_a^{sY@{B(fphN2X=6OH=I#4GXXU?iBd(ffK1=>A5
zp~&`m6rKm8nT*W8V63QhL`kEbgF_R&8@+D~ACS68bhZ5VX!99<6WP9eQfVGLQRD6L
zlOl>Jza$U!uy%itA6L)!v>$;07=j&GE&x?z_wk)i1E_qnw8D`vzrXdQdc<!n3s+na
z0F+wLb-~{Yh<P=_I%ax?YxIrHYS%I)X&222qa%aKIu+i9cD|=+M=s^-?GG-Fj$n+8
zh&-o`$881*_;F1DLJaFg^LuCJJ43nD{8HQQQT&^8!W-n(H=XwI)SF(#ah{}DeKG31
zY>@a&124MvS@7f@q>2rrgWA2TvMS+5q^_Ic-5rN-FqVTF4_9C|{?ES{+Mzfp8|pCn
z>YDiXcY<o+4mF*H2!<4N>;=!qe^iknP~Bh(2bl=}cUK`ZKr|RGd*cSbzm^mhvQu`0
zJz$UcBIj3EEN*^onb_Zl!<zrGI5T~9&Kh^}!NZ+_R=C^i+z-5ZnzKy#=W{#HEQ~a&
zS-je~FsDJ~AIi!HjxcuPZ(rNB?8uHCD;G8oB)&Qv_|6hHD{reAN10fS43!jG<*U}r
z3W#}q<FllZp3(lT^&gKc(g567bG9_7h3pTzxAoh`Mdao7Vf=yZ@XXvuT=Qk5m3Py5
zZw`SDiO<vk6E#^>G8VgA^2O0W5fSXAY!A4v9VZo@OxeLdNGWKfq-e4dYa^3Sz~foG
zzu}(iU3+XcA%<`O^rwimRz;g*=-HZ+mxRxuTBs1Rfij`g_6v<7&2x6qBZkAT$zpu@
z$#gFVIPZl>xuT_c4#2xy3`v`9B&mwg9zVPMa9;7g9OIPdoO|A#`?~|z*F;xkD8jH0
zW3X90F*MYzc|3*?2%5OjAm+wR#fQvoN9eX5xd&>ir=jglQt0Z6h)Q4W$x`Vf9rxV2
z{1zvJPcaPqj-_CP`a39Z#C7Tf#&E-u6S1kO_7xa@(P0*|a^d~g*1Naf>FRKfiJM5P
z+Rw-IbT<6s_EOd3@H`V88jdI&ty*ruWmY00ySIc3_CC6C@7ZL&9Y-FMIktYP>(NY$
zd&(5J6rrMY`KN*ptlc5GXR2a<uiBS=Cdbw^Zfi%{SU%2gHp13+Y#6t|Vy*}7FMf51
z&-Kd2;0L(%k7{c=mF+z_lP{RZ7j=5v=zOOLZ4js`AZA<I&+h|}{wjXOvr5B`vD_k8
zE|x?a_UP9(Dh^#CucOD29DZ4d@bQN?OEfYxR#n7YewBr-Ks%u`&{kYSME&)(pv0?$
zUCt5YnWgm{>~YOCO|S!nb#eMXecNP_s94yH<TDd0$zxf}<3XKkp#AOZj1WFcT|hkT
z<{cKuYmlT0M@-UfDsCcN3BH66zjAXA;kukp#w9STyZ8<AI8Uw8Kl*Sshinc?OCD4{
zHor||#?{1)y`Ht=PB>9RT!>j|#4vBaQh1p}{xUU&9c{5m>>}PD>mM~oEaIVkyFXB0
zL>@mVd1NtB0WC2}BerZ0m21dlSb|&Q)iO^Gi|yJ+DGjkGh%0?i{yYEiixvD$*RGSW
zi`eq_ckA*q#4Xvsz=2^R;ogCt3i{56TVZntovpKto($0dJAmi07GdK?y!ZI=HS;v0
zci5QOCr>~Rs>?bb!cFT7yMA3>O3rjR0qs~S=5txzFRmFOMwjJyH2`XOm-y&fHE{^<
zFw+B&oX(TH(K_Y{m`@eLn|E0=nye#xz0~E~8!=@@dzg0aI4rT)GA=m2a@v95<&ZS+
z6kvIGAw%kVPF)b^3)~-c*boa(Gx@L{20;u?n|iS)dsCCmbHSyvW8>=+Z8tYsRhTb}
zxI=0aa9B&Z;2$mEMI92PK-!eQ@$+3Z`YJ_4-W_d4&_OEUBc0S&yKRIWojsvPi8j6Q
z;lv4t14H+O4QmdQHrw`D*hZsD=07N!PH4Q@YQ?FWM9WKuZR=~`!<)B6g)Zx4toG+7
z9PZ%HDi^e64OrcLiz=%U79rQ4?Bz3V36iC9<$@3-pYyRrVfw%#Z0m?dzg_oeynqr`
z(mw(QLBx2(yEtrnKdCi3oT+Rdolc}5x-~L%wYPnuM`(u>$cL!5HmtC~N@dx!@JjfQ
zmFnhb#G7?Kux6U(4FC^!YY8amaB74;we55B??!@qZ3_NE3Qzdy<)KCtV_fr`)A<u{
z0357b!3<x27~`Sv`M6{!;kD~bAhIH}86B`sQp@;#{h)xNXE?p-+Q3Y6k^zhajqdq0
zC5%T_1ld5CUlBNq37eP4+d2a%Hb~E!(v>rzAr?ciuabTHa4up)*5|!P;D^!PzJ7Tk
z<pnJ?tg=BW0|6+HEXMVPT6~_`?c!UiA?P-P>X<iU?sr%H24DXF`jv7)t{C)6ROW7T
SwH|)ZT|4$|Fa6Eu)PDi`)W!+`

literal 23848
zcmeFZ2UJs8+dq0h6deuLQ2_zl=<7&xq$pic&=Em}B=jOmO8^x^H4sI^I3lr(h$6*C
zNkS4@kWhj$f(}9uJtPnVDq{3dB0>m+cK5-VdFQ_0x7NM?b?^G$b=Ue>YX;9g`|R@U
zr~aO2bMcU~Bj!7;?;r@m?BBQN2n4B7AZSkA{CS||AYSzm_%-k70mnVyH~1O1uSx>H
z7M$4U5d}dyDd>MHwBDiP5VRKBzh{?Q?45o|+;8DAKGG>g&2!9IZ#%n<=QIA5>_lCs
zb5*_k@k&iyyDNu{Rhu6XZ|u6VQgekkBG*fIh1y&-6NgPpmtOl<D+gMqdSK4^wM*8`
zS33)>+j!o|yi%Z-a1&ek=59+-6ylZ@U(?=65Ozo_lB8anH!~v0i1~`y)&0-^{)Ynp
zKcj$tUih#f#Xwqs;OT_sswH6gW|!G`BbMQ<n01-(-PxCTdiU*^rQj9OB8S!WlWe`&
zR+S}v;h`S{V!EQGSOfhiWjUi~`}Jc7)$Cq0C1_J+DiBmRJ(*91elZo<$W&3XWEiNR
z<JX;A4uQ9j{*M0#kB{u7KIoX+NwFjbLWSqT{Ip<EP8LjaGcRai+qja0S>G3LazJO&
z&lBS!3u-;cW%0rCtB5vS#i$anP0To!*hs?Io5`qp>BpRg#(cM!Y`@UDps(-4V&t#m
zR)IAy4D*s_!Z$E8qg_5^Uc8I_k}|?mxoza?b!sVYGx@!O1}yA>$;os{rQOU<IN4`;
z|I>#dl8bkvP_2C*qy|Bya$B$Iy;H~GZ3$LMHB`!{_r=Yj82)4_DV#A~_HOvZUimr1
zA)##iELG6nYmky6ZZEV^Swh^cx)#tYE5Up+AptM1n^RUpmdlH$N`t*uX^Zb`*{F9H
zB`TrtBzk~Bt_1E0@)j=p`>V?X7*|p>eujKQ@rF#C$)fJP*AcJ5ZV4Y-V-<fxG5KCJ
z&49~;T-)LKl)Sq2t$>P`-i;xS47OVm=Z}Qx$olns$J1|@+Uw`#6h<5&mso_}P*k&&
zvGpd!2`>r^^PJF~aaO124OZ9ii03A{tg>gCPB#TycUQidB-QU0Fp3k1{I7{-vj<#v
z^wINJn}jD5>7Cuj_(kq+2WBKQB=Bfs?&wmyGH8}goYm2nrTZ(SzF~X<+6SK&oRc|s
zGX&1SebJ0rcBJ_`@__-qam^9Yp6PHaJ@hdrGfC{X!At|?qPnB8(%9NaQVq7Nw5T`h
zM%Om^ucAfi(^a-*0gfDa2QywYCbv`wuUxkOw)hrSm#a^{(vMLs-G-@3A(tz~n()(x
z(Z>jrT7D)m3ed@cM3jF<$lK`)MIU_MaE?y8N`2-7ZpyVZ>xZ~i@e%d&bI5UTf929{
zlS7IAF_9t_VvM{=J!In1_u(?r9$xw#tJ<!M>S6PVYi5tL$dNm_P-*L?2#pI9tb={%
zLYhFvn)cT)Eg7<^=@tAQ3JcsU8>G=DM8)*z_u<)<Y2^zj{>zsECa&w3R45&q_)I@j
zo7$sK%bT<kKX-TkYaKV;5x1&*QnLg8QrS`dzB7hC!)qb^{nfNi8%MY4Bz#rpj9ruj
zsi4!iR_jAz<dxK<n|h@mfA-j8&P-%0k2U4olEC>|?diOiYr>X6k2h@uq^HzWg_Dku
z67e%nCz{{`3_WlAZcDAilYRq+E{EzP{LHlk!yBR7s#O#qWG{rH)TOI(3K9zg7An}L
z{hW~^1Sl2l=<)c;U9}D02j<*lmAi9)YtptczbU9%((a5Ngt4udj>ihZGV<glNIJ%q
zAK@d(wYaNfP+wZInc3TJ{;cA!ng_<OE7@!^IWhL0TJ%UEEjBRvH2G>fYoXaP*SfsA
z`9M}E<Dw2@IfJ_P(~NHJ6oWDI(m3qX-j14u859T{n!imVE1>Zq&5iBVYb}U$zuueH
zqr&+SJr>Hi`|YoCD)qAg<K%vMMZ%lKC+xnp?+CJ*X*#bkD7TINwH#VFyI@WwGrDX^
zUD6U%vAk$p)i_+VxmsrSkFVT2<$bF>r_y|dOkLx|DOMp~O<#YON6%E(C>`WIa6K+g
zki?E~BQuqsQW4daSH7*#ke<CgaSB2BYjUUVGo)rrm5~wVTs0@VZ-inOOjq-8@)qO=
zxEW`~FXGwDl}LF{PcT;I+x?S@|4&8{V+Wpo<5A+06gL5z6*gA!br9)hlxpSuV&GF%
z7XSWE*!eh%%6BtomrH$~1qG?0qI1T_!@875mfwUc7~1%XqKHG}^7uIU2FwO3v?=;i
z^+KQ^a_KH6sV8=nS<rdD_49MtCaD^#J3?m3bon@5{&Wb+_|CNsbc9ykL3acBKfa~f
z5N5@UkAJmbQBsIiH!<llXr_E#3KeT93Y;@e<utlf+M+%6)a;lI%NaLuDNc9U*BM?|
z>A|~3=!@^s7c3;u73sUSMLq{KF|X~-MNP-wlPsWsr~i5UljQ%msPP81=~MAxq<wM)
z>=DJ93kWEBpk}8nUdYga$7jc&Bp>^3HE6Fpw_P=a{N_Pt^b&=^+|%YoX<bdjQr**g
z(Rb<VAWyxV$c#TooNQCw1dEJA3(%pCI;nW+-E64>|AzegenP*mj>nU#hKAzcVY-j{
zBN<J5H_w$((af|!`{ccE1uQ){w4>0XA5wVRaLF%-2fXmXCTCnCvtrhP9>ni9n@{S-
zy{}%_@d6BPTSecDRWMjU)hOLa{Z^My=VE0J6f%`{Wa@<c7FAMXgSS4-`X1$!g$E%M
zJwH|&!`D^kNKJ7(i6Gvx2(<gN)$E3&_R2*_0I=_UlV?gS;<NXnUtDs8Op+q$5jb)!
zZ!gJNPN~EJMFr`vP-$)Wasg?F`?T}UP92p0I;lXr=h@Y9PI}|4@5Q^3u=4X%`B~5i
zweNw-hVHyJ*80>mY>0mDz9<EdWEAN~L<nbSs5go^OZN|e?!p`J55tD&CPb6yF}P~C
zt0*TB=1w8nMm(aLdwJjjrs`Z)&g?Po*tha58XUQwaDa+vnQmf@_q7ixgL&}Fgt+N9
z$t+a*L_ew7_4N#6Tq`x&OB(`R)R+H=uGK4bi+^8FTihz}Z;0qK`#oMpk1)5fVaNNH
z_d$xOllJ>!!sOgrSar~)&TWln&4W&B-Obn(owwOdja=HT@X$Ual`8vOG`la#!FM01
z-5Xr)0B7+Q7(5s+e}E*y-@``O9C}(-@L_b6_sgLozAIgiW%G(tC6v-rTFTdhcK^Ez
zC?9*8;FB<cZcjJIu_MAQFQCipzX63LvI=y)teL!$fUF!BoTH%hU+5P@^C=%?2jOPA
z0AEqiQmAVz{1r4)F06+-SJxhr@Z!n!CLDgD@6{O_&|bG<{%W(dCc2{=jn1NHw{mQN
zx`XO&&!?DWU)OS&LyqgZ02+Yz(;By`&Ns_P$4ke|ruIx)Iog_e2MX!omGF*rq1+B-
z=4|mte}6Lr5a+F$-GBGkV(4F5QPVLuyejB5s>C>UsnG2jDHi|Y6WV>pT9wv*^6@V8
z@FB0vE&ip7ugC*rDXfRJZ0K8si_FH)lB^w-t?oi!4|0KWEOy`(94N)14mjs{hAOwV
zh3T9%>yOv9uZKL7BJaeYWc&6oV>Rn+Q;MEy+4ac|z}j>{%dA`PPgnwu;SxPXP&KQd
z`;6z#3c}<6&>9|dLIOp;I)?!%y0AW0+33*izt7Jd7kOAei1(3SM*cva4E062t#ZS=
zb?b_)s(GW{KcX8lv)({y7p5V%;K&NzmjMgRJ3&2zLBE-*@Uvj3D&7UauQHweW1EJ~
z{Juho@-}W4RF)Wx%f~kkoSdpgQjs^*!*=2ZuH)r!nx;6MoqfcbV^OR8*nePp@i&`c
zng2!)e8^d>nkz3ZvMB2PB7K2q!s=S>%8nd@TTDT=t0aHjlK3Q26}2$1@7+b<w%F&j
z3ZKJ8boMv1Lc#Q~fVzp@Wxw{cv2OLfoC%lwGE$6#{nUS>ZuC`aY;-NNZ96zr!C-Q~
zv>DDx^Ro>on$;9u$q-O(QC56<^5UfQ$&T3CCanpST-^C8tx9*#FJqHiMo5oixdv9~
zntudq9(p-+rUEoJ4o%+LGJ+C~sDd^thbA+SJi;gGFUZ~=VN!YCEMFVgLuEG8Ox~GE
z<qPV-o6V91{pW|e^j7QfkN$WAooJz~RdbXqL~`htg2%KnpFV|a7%}Fl?{_{7Fvt-C
z!b($VMg;ebN)Ez<x4~I-N4z8Zlk6mnZ_hsszNe6@s@J_3J0af+H)o!C-E^;#ARnMA
zdi{aas23%`uSgE{im&anu~1#!iBt8mtD6BgjSiY?ue9~m^gojyf1E@^EQYs+vsfKV
zejyja>b8D`iA7Ii1+((uv>P&+TY5)z^xV*+AW|k(hokFeooGe3!DYEk`N%(OwwGM>
zK2k%pV);q#1;h(qVqVpAvE4;2Iy8Yv)<DlJ^EC$m9r233&`Y=XNxMJlvq0C4RYSfD
z+Nm1iB4+3@2j6fOw$GsxXab+&G!k)-SleM{4PalGb7F*5iJev7sUGv}QryfLiT8MZ
z_Mb{uobdL|>NNiDLozH4c5iejD-FPNaw6Rqf1WoH5))8AF)~WcsyFn!iGbN4+~@b{
z`7u66Y6=4$wJ$bLjqUg~+D%ZzB7W7WV=;Zbb!rE(E;nZtgwvB1oGlr>^K*52#?K)g
zuq7i(>s)G4NH8hBUg@|sOSt6MJ7VMa!D7*9LGi%0z^Vb*m9CC!#CP8zKS$@9h^asn
zbQhBRu)Zuz)HXQ^n|{Q;u@K{G_`QmEJ+QTQ`_#@W1UK6*<GIAo(=KLC#p2XdG*a?k
z<^YAS7t);z5^Li9@Hwg^2RU>62G?+~^Q&Uw9M$fVzg6sBTSbm@Xy2?eTenfoE>rXl
zP9pRX=Oi7eSH$Af^``aXV3DwpU=Zf~?<4Fh$4qjm<;1}wQ*UCu31<eY5u^6Pw75!X
ztnYV`?e7KjM$_t>=<?M~nxIeWjGX?2IXMh12X^({01MI0VE1~5;wY;e+Y7aZs|CXB
z-F~2J-AZH7H_AY!7VTXB?1rmW-Y%bQ^+L%nHl9;^>Ys=RIs>OsK`|XgeLqeRyKoRM
z@$b4og!pY=Jsr(`>}?(0{In;=v&wIpu4?~Af_BJrNlam4QrF%WqNVV0jGjma0<N!5
zb6rAKW6M`;l*Hbi%i1uDJX~J_9em7Ob+)VB_<b*LA~rEWK1nS<mu=JI`SJ66qzkUn
zev<2|)&4au930+?%7va97c~g6g0#XxH*20n{N5xQqqfB6A-cMLaT=0rKy&Zt9#Zav
zA9W3SlB?jF%?Xx<4Bax6IhNXiUQSQu=WY?%)Ys~XjcfYYTG1+&q1Dp|FoZc%r|O?&
z+{E#B)LWQcQ(zr&e9tV)$7H{vY&Qu<*{As$)fM_24`-ExV<cbsLW~!mA5}ei73nW-
zPN63;6RVW%iwGV23=4hDbcO7Qmtg!))XxE8y39;%kJPMx)b#SKI8hA3(p6e%xFqqC
z>DR{O&zInEPYsF<j+`6sAb$bm@o<(OITlq|P|(qISHRxU11ATT!$pcmqiM*}?M!sH
z{s%ITLnBv_Y#(yf@OJvd;LE|=;avvJsJNO{14&~?%#ZdxJXIGvkc?SqZ**DJqB?%J
z++{m6@G6lYTA%nC9Y2FHfT{8wVELN(S~8XK^&=bwepQ;FJ5qOU#O{hui1<mQ1NSU&
zMEsX=&*q%-FjOUD<*#5D+)bL$knLPBou@2B?opT8Mdk$P08MHoB&;?_^pN)=ZR6>4
zpiALkqivnGvhE(`Dtb?&uq*0n2=EkzOJO#S@2*qvUfByL&~53?gyFQ)^C%bs=s!1b
z#<9OA?w#O6OJUk9=%xOy1<qt-0!3O_`<k=q!(yC5i{`3L9rH6z(Ndu#e_>JiUH!lT
zBK`}oAf-rZYwk4EDdCJal6gw%N^@p5F3XlX9!nY^v6fOUH)+{vzqzcP_zV9;*9Feq
z`4q^XFg{H^Xcvu7@*EJkh=H55olBPVADH7m%b@!C6f%S7DCbhw@r}YWo|Lf*{eoLK
zD<SA(PleX*Tm{aBD8fYh+4xB+0@PaT8MaIts~#mpF^!MFcb81sHR8B60U5)Xh8|wC
z*`3VI8^Dx(qL*JBA4@Qq6lVL3t}L!s|5sK2SKCmfs`*oOrSkQniS8HCI0Y-QHsH6{
zn9-MmPpB7`bo*u<@hLNeRUoIvAy|P8HlJJ?Qr>lF%7LWtSTPqeCRl1YIQVoNyMI5L
zrl@`tke}N80z~H`3*9$gqGk-1cg2jwcK2d7JX8^N0>9|o-mIXqr2|Z7cAB_H8Hc(1
ztP-;!kqfN9mRNT?SL|#3Fy6o3?^W&2iEAoQ`!VqynL6ZfeV*L`@k!sd+Er0qnSEH=
z8vb<s0>7xGM20hA=j6^(z{h8i_jBGPTt+nDC@rp&bsp*ujPie357kwZPg->cJ(-BG
z&IbDXw(m};F3hLPxF30gh!zZ1VZ9QHrB&R?DSmG9457Wv7?HqJ`L3P3?Z8>$rro$?
z*Bx+f!uJXYL*0YE6RpvoYj+^3L{|LA1UDW!4K!#Xj-dC)B)ypJD)fAP=iBKfQ*)+d
zEDevn6Pf0zSGs&8aa58DCl2_s{>P!R(^iZ*DOz6EkkYeWG5ul7h<tgVU-#|Ze#2`~
zcIu_V=AODQO9O4gX+<{HTyp*2obh;&{-prPX=Ep{-`yN1I6b`gwCvwHwkvWgM7jK~
z5A#Z7#MbpAb9i|C;&TzLoQ|iOuz|M2LE8XwlNyBc)KWQpre3v;9UgZ~z{i<)HIlu^
zYOU&8wF<n;UH0<ETAQ60YCk?k&+#_U{|QYqn<ky*M(|FIwwJz-ZBf9SrI7nkwbnJv
zMU?c(pPOtsMiCkJ%JPeJp`tUi2p%;tEhNsOL&mIORsz_huD!f8KB!)=1ZTMAM4X(3
zBz7ouD6$`*+sQR^?Mx^4O0T6S{UWPuZ^kXAq!kzv#@RkVfhWDWbBN69Rh;y@Uu0$h
z^!K6qGIkH$%T0JE$^aq~RdcuOIQ{d~oi5{&-Z?$vi|LkF8jF?DMj!^Q!Yd*~Q}2I4
zegn{+ulP$>)^;1;?=EUESYA4>zD~rB{5q|RCm$#OtPW*dj7(F9I4T1KwbpC(+T6{4
zS`SYJ)%(S8mGg6dh$*U-C<Y&JJ9|oyU46dt)0!&4x2W5QvU0LyF451hXwYj)Ff?=$
zbKgtXqF30#?857KOofpK)f50!IZ|&%PqI$t0o_NLYgV+yf9n7XsL%(>OvC~n4*Ft|
z3I6~$HS{Pk!(B98+6JxmdCCZTyVDW!TY07AStoLDGw}GoazPlX&XDE2>mvq9@p`(8
z(GhQDoc>OfMW1ma$KMzbup-hav#L5`la9(radPbCg$zscajq2AyiSjSRv8Fo$RIS5
zG+y<pYc>|3U>qRSS>n9aC~{^;<7CXP!&CO8&)-69|9Fj}2scTY1uNc@v|rvuHU1>i
zj{aJbh|XTOi4;QWLpIYBa0+^apCrMGUfX(V^++}%?(^FOIzX>ZACo^rNU#t%XZku-
zbOl~OcYda|+i5VBDrobX#a5xrC(ki_$Lt-l-77mrskzb)U>XMy<ZP>u!6(j7O}pcI
znP{lG77Vv(9blL~++rp$GhX<`$<cVdBHu%jA`7D7_a!f9W15WfP)G|GTS3qE&ccx^
z7%96}VIvvSUZm2U@Uw;u-~0vDT+GKWNHAj+HdtPBKiS4>>M)5<4VQM@{>%F*d#vLF
zRER48LKi0+Mv4yi?}yh7|9AyuW<b0JDy^GN?ITH%7+{83#GQ+j9pg8it^uMCEF4(D
zcN#wV!pah5YwV;mX$l#H55oz<LX_?lgU}k)T<ffEV^cH|qiBl%FA^n|qP@8lru}b8
z^_v!7ZQ1cmh#uD4wQHQ9k^Ar#xLAGswpimU8u~&SPadeWUS`m+DKBz?sQ|{@zYY2n
zYLH@^!_0Oib9rI6XJZ&+YoJaQdV-vSG<NjCiZy80yzsWUcWAl-!ND<<gE(89v><Z0
zRgUGrZuE7=<^6MVUo{Me=c8~q$Z>27g+}fTO;*7HuCr(oMKi+$QhEa<AOq2X<$1nn
z<O;c(em3;S8Wjz?Wx12@$0x`IMTv($N%ejwprIfqE!%y-xBhW^v_rl0?SBOqIv%*z
z0gHcZ<pp#sV{Nc%e(`${G20#*2+8=F+lj7}LNW<mJG{MToCHTYSN^r_rFJPaDgNMW
z5DWzc0y4Z9I@3|M5Or4k%{E~uzT2~9Hvib0E!)P=y2>J6**iN|IG79XW{U$kf<5w?
zpkOqtr(nX{zI=>^@6%~mTYQ#<@62xd9c5Y-8DC7%VcnAfiM(+9+M9ii&UaTrl;*eT
za28_&L*|xKIK3|O7zAmHzyVdlUO|WFLv{0Fx`EFxmi1O%qTj?BF%7*1_dwKvu2qkr
z-h6L~Sy>9!h49t%{^-Z_p7CGM(6R3wiklhHXaY7!P8D`~OvT#&Ys3q-r%#xseh1L|
zgT1Ek;1<1^N33oYXifD^^T~c*S-_8zrlV~iX9{g>xRw>XAszsFF3CPnbBaM=ccuA-
zhA+TkwW#HKKV7JGj#xc+q#T{AG@4JnYI_rFE9SHdx+VijHxU^lhv^F&9|xSveXw2Y
znz#am8Fo;@4QNOl@}q2LNdP8Sg1LVVqLc#=@w>d@yoFHRdLEWkg=Au0)!}*=@YV*l
zp4la^Qw=JBw)|!34qvJn&$+R{)PXqqBO0S{VnCL`@Aj8kqg3gJe0MMefNvmZbnyL9
zb<~pymUXt#?V|<|L|6~RSyWZ3KXv^#2>MnF&^SzUiS5p69OgS0gEKI_6%Cc$52&r$
zDTC^Q6tw_rXvEuc|486sY1(`n(n=r@Yf?ki?SyP+y~W_n`mZ6!K`?&Ahw#0^st{x)
zewOnzVc_O$L!s``fpK60?^B$MA|BDA#q`bineMT%{cudMuMSlAqmBiN!<cav(kHMo
z)*z0|jgPCx$^3}!U&dSD<=PI@Zc{k<N4en&$lqF006;d5Ilw$^x-4Y7A%9gzyewqA
zduM`W?Gfu^zH?RUR4pst1rJ;tHaxa{yXt)CZCy*zpvRP(oI^cq>u5%lCq|QguD@-&
zO2(K2QR2{)5wpuSRUN9!JGvBjy};niX43#{IIBMY7awCj{WekcKzV74VgVSdPRq#>
z;&I|X{5FAVAb*UL3Mzj8i$kY47qb`RQ;ObATRJn)pdLy;xE->{<2C3CH40g!?}r}3
zShyVLJ#of97M-VVkyF%nqRWgv_KsS9yd&wbybk#VxxKYTQCN%B@u~6um4uGMd1B%S
z4SZh3K*Rl(j-JW`j92s?x)DysO3V#7Jv)=DNz>lkoHtAQ`!%PTBfW?LjG<?3pT%VT
zn{FS6PK%LhY}tfE{j;1)L4fPP<msj$Sak&ei_S!7)IK={c@d^qL_nooR1%~K)i9N{
zi0Oc}nnhWB9O;<6@$X2fcYSLT&brp3IACLIj+-7~_2(#Wc@t)#?ToAazS(_kD!Z@L
zgIUzK!M!Kr<dy?TzplO{O{LCE|2K#f06hwjUy+@#3|CQ*76!Pf36^S|TZisED5PDb
z_1adt5w@FEizUZo`Jza#e^Vw{0#;0Kbn{~vG6CP%+5n9A8L0qWJP74TW2cVD5h^JW
z$nzB8xtwmx?rK}IBK)|RLBN}@ruA_X@0X(EYjg_C^xH;mr`M-8F;ZxLdPk<FMIWcT
zqZCKQe}aRxF=zW9_J{*BQ_R)&T>6$>c6zeDRr^i1soLtTjwQfJ3|l=u{1CD3a~E0-
z6l8n(bky4Ckah4lBbJ7rf}Vc>h&^zs)a+c1n?fC83cK2;W0ai|<GK9{K>T=L<dNm8
zQ_(0N*4lkFO1pi3ry18ejF|936p<kR4LJ)B!REn!KY`V9=7VD)S6CeVexR1kJ541K
z)x`8yo*GFl%<RMDT;3R@@VyVRXm=1B%-PlnhIm$O0{fPUH+35Ktc-5X2H|hV5=!++
zoxa;CRa~nZeRFjt?+F+Cz7z!cm_J`(3})FB@SSL0%rJGvPLCd8;_C&b^pn>XHxKps
zN|Y9b?<TAm<+9k(e?$b|+=At?Ax#@_HkI!u-+eF4F{?n@1wyS_;b!1+=d7Mkwy#j$
zR{u9jF{-JU$wN};wweNLMd6@(&0UbNyQeASEbU(-kb-X(HVoe*E<`=C7b51IT)L4P
z{|#sDh$(|$s!yE!LVbFr;v6kaT6OR)+cHfmaQc>ETbKdDco7nm;9XV5dl2xc{hAoh
zHwF;IIU!Bj=r)V6lvy1`*zInp4X<0#U)dw9os2IhG?c#(qwvaOVT=|xD(CdJPs{0o
z^6Kce%lRH}ElTFif^?jDtt-}8JgwDWB*r}LvDeB=lqA|J`U|EBa`~3_;-x;`lB)O^
z`_-p^uJ^QI1DCT!Q_twao)oOG=)c2Q^eyl8zT9$Mm%k$=Uiyv&S6GfIKL1xwJd~w!
z`Xnjq`^N7v;%q-D)c6keb1qK1+)swDg2-R$e4-oA*o-x2PRGw9p(3!5c-ys(Zs<8A
z->Q{o?o&qNdj08#!v<P;F**|CUqKy+1XORlp-O7=if29Xx;JkgDyQOBJ?3pgeP_!4
z3{|tauXI()^25<-WAp@(>}<cOWAYKr**p&ChvuK4&ciho0p|s0s{;Tx2DyZb4i3l|
zr60Xf1;g17I>Zz0$A<B+pW9-VuE*4o`gEzIxzkkqLHTc*?ZvlhIP3OV+RkzXM<vUb
z3@RNOSTBu(WsbPAO8feI&xBsxZDyC7w7W_hUsPS$@{eP9aY_vZeY|`?`2bUSl$43|
z4|O)u7rO}xvUk>dejM0@>05U&Y>QP@X#FpR)(1;&{26Zi1o28b%NxsG+$si<`?eq;
z@Vul0k!u5f6h4k`kL~~LzxxNUdjCruiElTl4{U&~tK#>abRczGclwm_h8FdS<{M`H
znV&FiT588;b-CBvi;o5%)?);<pEg(>U+cTde6Yc#|Hu?7yOthFQ1(!-48BIfw=>Wh
zqql3r*<C`;p;wn02q{><$PC@lX^@H-pJ={cM~V)SU!%%mVn|2oCyH%nQ)K>(()lBs
zI5kQEoQ3zc<n}Ik`ve(^Dj;MJicfKdE2)Y=zgZ`rNCMGgM(+yF!}6pqexuyTt&zqi
z=WaQ4>C<mW3L^$zL7ab9_i#<<pBX|z`XxKRLq26W(u$*Fr62xGVQIDh{EzfvP|ZS7
zIj|{bE;sl8iWI~3TJ3w{o#fAvPjM?}vAk>mlhK&9FsE<T#AqH81WOrSZc?|!8HBxH
zuXVd`^=jF5(+6L2$m3azn`Y;=CvGcA9|6YqsTR$2h0DJnk#xM7n1YhVzd2j{HC@Ws
zx!S5E;AhbviUvb}&`iWet7M+@NQ2^|<b|4}`~qjXjayAV@A;is!?m(be@<?a=Hg8;
z(Ig^CG&c5iXFKYQ6Hk>cqTnp<0cWj;`uU(A9VG6SuS|$JPI_K_3A~5Q_#VJbqEF|>
zPx@w8l_(`hQn~EHT~wId@1sCFR*xX+oMd3CC|Mm#Cf{(C<wzWC&6u(BMe~aO%i~Vf
z|0rQIH9DGx#SGguVykGKzN5AMsvs<?+t{zleVm2O>h7w~`9aJ`V%zWxzb{^jx_#dQ
zDz1Pd%fu!Lr}LwI;3aBn%<E?p$=oszmdeNwQh+V9EK&9^>dUAYXf_NfjWq)wp0F!j
zqcW02ufkTmamfs1<gllSRMHFauokx}=P9~y&|?v0niqe#{<aaV0JtBO1pMMco5LVh
zd5(_7e(^1(P00MXf+I5_PSxh?z<#XQGg$oM>{_3ihdkj50KoeCX1!i=Zt8IT#9>l^
z{A1kr=wTCeUM-^7w06zOg?ET1%r42?*i4((9J3>EKBKXvXg%oS-;@yql@)A=&HMZU
zd5ze7OB;(Uf-2`NNl-NWnX@%`1$JSNIoq}~!OFKNO5@G3b(6!h6&esXJ#?O3u(|$;
z>ieDL?}x&|Gmjc*cig0#yXAaKqt~??H$$I<4bv<wJ==&sri_05M=~%Y$Zq&oYuy}^
zr?n_C{+(n76Q0g&=cWlZ=VNEXy{0^1;+W4@lDl5vJk;VNxp+snl*_aj>?&F?_!79;
zLD0x2<jR)Oif7*tM4lrlnOs)Iu_%vE08XlE8|VftN4xR1jFuOGObURn;UfO&EUHeu
zZWpJj^FYIdQ5N>>#Ftj&^TFWe0<SdbhP+nPnc;k=9gsO%<+hkOfAz|a7Gwl@WX;OK
zXC2P^&!mc07o&#}*;4yq<^&sQ%zGnk*`h$1ufHDH$8h){qi?Whd;lJ1<PeL|1~fJf
z#t8~+T0x96?frl946yi?d!m}y-SW1)?N33pblA=>^IHnz&Om_;2wb1eqW->UqD6U5
zBk^&cI{TA!`qh?jN!H6lGbp$*IWr)MdvqM{^L8o{yJ_uNyS5i+p>u0p=TPkq%sCvQ
zvg^v>$Nw@}drvg~XUpmz@3onkT`n+*ChT!Y{%Nf3%Gv^d`vb;5s;rM&dSUI7k1wvA
zKfh5vU92c~#?m7su8Ku5V`s91Z3#K0LSJc2qy($D)9nL%a?Y0OyYHetJa5uCPfm!g
zw{6X&%Zhx(gGn66%R+t^ht;g#p5F1ySNNqoi7A&wNB<#e|J;WO+rN<lNR>fVYn>P4
zP;ZiR<3r0*%#44x#bK#0$nGCpsonX6T=|YL^IWq`PAM+B6x7=O)4o|e1hXJlGei0;
zvkITo#Yw=zH{f7KmGkTIef1VQY_~V=Fgtf+F3sr9F0WOzUQBQ+8l`hEFSe_kK1h=L
z3a0Gk;R)7?L1ar*YT$;kFYehU$=pHf^qh=cUZ7{4nXy2vwHvpXATW(=33HzbZj&>-
zTDGNK_CJ2LFW<cX*Jw@})7LDUALp|aW#kui*R?)=no1;<4ppXH78Iqb@_a${oBSNN
zWP^5I-9VW8nDXjD{_2qhaDFFH*+B<yd&~iRP6>6N6Ot_UsA0d;3qAh&k9RO$J1?b%
z&pTZ`UDGuG#N%3q<ay?ey?2sI0|<%<lhj2#5-3EChWX)RwUeIq*N9)cg7wWtZe5#&
znY5PKW!AMIyWpwK*_O4E{S;75=ds(n>FZnM+Nje<rzmd^>%@151?jWd%~w~T&JP+s
z(a?cYQ^-!A7&k#!<0{6Cn<QkxP|NMVy4*gycVj@bTO(F#I++Y}43q<Onz@s@*qK$}
zyHmH(EhcPidX12SYfVGGQf0z8OclBL=tL|aiht(q7}ZZ*@Mp}zmEF1P#t(z4i6l@p
zt9y8A>Q0ZS00aqZW+BDO45`_d3gSNE8)kec3I4s@1#c9Rp+O3=-yLEp_K<KaqOH;*
zCY-OL;h)*no90gAyq+IP5gk`nTO41vG<=OC)`jUDzItvszdXrMv(Kt>-LA6l-TzJC
z32%8Yp+Mn?@v4thPr(`y-4!97Ag3C-kM0xsPkP)w8|S>8w_pCy^tB~F;Ez|h7&npw
ziB^+6^RxYt57kH;b#E?>9%0&EGvFQ>EU!jR!esiwk9uezPzO?Z#wF4#SOmD9594Rn
zt9stk!T9z35t2;<oFtO>*`*Z>;<L<7&g?9jCQF`u%cg39Z0d={;bZFY;r7b{BU5mE
z>-M-w)gDi(dWB`&@HOiB*2l3=_E^$x9Sm8xc@_ya*+?-v+Z{gU$h!wBC0@Z%j2RCf
z6W@t$oin?SmB#wi3zfy~Gv(PoaY@SfWY1*_)`I2bty3x6e<o61$f_zB8i#2n-9?|w
ze>E~c(`>slnX7BeCB0nK={fJOz9w^qz18BMS~m1-0`BQD8>zW|Vl_d9&0o64G<{V|
zZqcR2Q;a`DfsXZ#-QI5G>Hw9V8%wu|pQbc^eV|!Jk`+(>zNizm^xU(d%k9k9&_W$G
zBo6|Vuk`cm!W|;lH09%C+#M|)aYovL?rPRApeHoiL#JM~lwuoQPImir){F|GPah*4
zuNTh9k#7PtP?_Fl5UlxD_1^+CPa_(W>xG~3wx%nGJ1&$S_dyG#T<m}PRE9$FK0s|-
z+W78YrsJ#QBjH*U17P{ho&*82_EAPFkY36_&~dfa4?_!eVojQ&^;F8<KZbwsQ6xNS
z@A(56T1boM{(Vx-b&m~Ir(PrWJ>!?OZC>bYyZ#H8t1#_9Ke`<A>Qg&%0Gvf8E;7sB
z8*HU+Wi-dN&ITvpTRaSiZ@b?p)%~q-zxt`Y@;`=l<ndhVM_4^tc_{-0hLQW0WC=N9
zp+1*}+751E@<9me<1^I=@UNM}l49jCxdc{k*X!h-2q1k#Wea-o^s%!_=Ext23T#f-
zc1x0sN(Mo867sLVwO6%(n#CM!<z!;vNNkduj@kXuZ`QW-X9(tu3MBNHG+Is@=g>$o
za2y9B9P3vOZAB-pQ)&K5#SMt5BX;KanKLx%;ax7bhx&;47^3LA#9nmX9l0KxDfDP8
z-`zTo$K@TCZ;gsY=Y`t+br<A*AIYvQvPeo}qx03NI8|XLFJkuJPx20&N}|_3R1PDF
zuC*(LqzDq@xH{U0X!vRq#pcu!_Q_Dv40Ud*M-D-2r5J^+-Ae(_^c54%vQgKjE+m_P
z>V1$j(LbfOWTK+fqQv6NnKL9(hdat)%%;1rAI{NK|Mlb`@?xXz$AJP+?i9$7vC<?i
zH2|7n#JJ^S{k-;}^(b}ez$=%>;nFTKD%@<3NTM&E4BtDe7GA^&e`_>_n;3ltR!D~h
z<6U2zxMU8P*WWePzFzw5`c0i@LD3-qn0x>#`sx`V`=dvr)9|ve)jA{7$gdY1D^1s6
zvkprF*;0Pnl@&YL0ly2<iU*4tcw6^B$^w;%jaWsYVMfw3bQw6(1k!=oU}6|9IQ%cX
z$;RYtKZ&VV6&p^Ye(3l^F?1iZpkjmYgls81rIkl1dXwZ#dgi@8%XR|=e4PO@(hE<x
z%X1MSV=+k*x<9ALxiI^XBzFI{h<S@66pH>Ibk4l*U$A_@D#+}v!JWVKx#zh|^E35+
ze&kB)>$~uHy%6`FkH4GA=HG||ibL@!uqf&E>n(v^`E2uoAFlpoRSs}V0}O5hZ`MWL
zJnA&x=g`zic^Oqs<d)tz*5LxQfqH7V4=|zWy8vhitlB;(+}J!UBYmU_uJnoFi8|mU
zhHe}11ENoP)GOYm06Q+}w0<A9{cpRaElQ(Geowp?F!cV0+Ot1Sc8pDK*LGMIPD+w|
zA@RfJg=f7PypU_Y;V=DEc5%&NG9DBO)40JueX`Zx#@ZX_T*dFQWZuu(Ahyox-`lsC
zuw~;nN~yJOie^yD&rauX8&7jKFt%zZ`tyCQOFpkkwz;cxt6ILl;?8l(!i|4v*Pyqp
zKhzAaq6_sum*0F>0yrr_>)Nh->?%B))e_>q{qy_i%@;M;X{-n_=YA)E{Wf(b1j(O1
zJ&y)S#OS7<p>Ri8ku$xki{n3fnIWY$dir2Nmbs6aJSXEs4>h!^g0OF2Qf+*a{MHW*
z0Eglfdh-Jg<_buXthUjZGsQ|5%*Ff7Eh~&gk0M?fTMTCoOw8Pyag`+PFIIs#<Xfs{
za~vRpuJ7TkjXfDuNn4sWZG(3EJYKa$q?OHKpGUv#J3H`U9a>P}CUlb#+{<R@)R}x;
zPwPiYJ9#M#%K8@oTUtk2t)iwaxConTDL;N5NT_&nUpxAgw<c!f&V{`I(~S;RYhARa
z9~2p3EaN0)ojZGLf^|L+RE-lH-GKH674>wM#Ifc8J*TiGG-K1K7iOV&rf2s(u%VTM
zgD}5rR8APR5p|_P=PvhlV8~zZQ0L|d-~Tq0ZQio=Ux0=E!T3&=@zQkIk=`R+vJkqY
zm<k-fMwKVFzkeDyo<Y4ZpaPO05uIE7ZGCip%Fc0}IwF5QYV8A3&f$(#G4gon3I#du
zT}+*Xp5T^b*#qwJQ8Y8nR7Nt2CsPm`7%s)AQ|9slNHx?fb)$^0T>`U>ln9%?T+pv0
zGx|Hx$-wa^8<QZjswVA^NBsB}#6J=ucclAehVB96@pj6DlwaMm&HL;i_eTtdi>cGx
zD%lE!51Xfmofed8d<XVW1Ju-3!k#zb?Pnp*iu}vHrTLGfTL1=>!Vn0oK+NU;x%OcI
zuP)egBKooMbF(j;;y1Mc3O0d}!oE8luH}$C7`9IhTs^4nn3p&&4`fjbkxDomzyapB
z;`TuLZTZ?<QupDShw=SCl4PDW6+916L$4>W2GOYhraZ}9Ch|Re5p%0z7^pbCuz^dv
zXN1YI2ceJBr?3t^AD84g6({dR?qlvJgL<Nf>ckLmIcY-M0Ryy}*K3;Dq8bZP@<20S
zwG0p<u?@p#W|I6|v)VY*z&F$Xb&LOY<AM3P%jq+wG(ZKq4^WOm^U0&+SWx>Y@3Qfm
z$^mg?PtyX>+x>xUsKW1@-w{dKx{Ifutxy+deyc|UVyc-Dt=4Ma@bxt!4O)W&QBaie
zfT6!oyU8_32j~n48r~MJ`l-LT`V=4}_%YT4mSOsYzMci3KKRL?0YbBjY<#AUlHT;!
zqqe}mJ?|hyF6BkaA0m2mS*Y^D1GL#aH}tx>f|>0}eh`0}q#T<J*1Bd-IGMl2Z#j5N
zA!V4UoR%W_qeGKJ+kZig3uPzT<SWk~o)1_TT-5-NBae}xzaOJ#a~qJE=B)Oi7&xO^
z4uDbDQT+&s-fyIUUl4c4{}-=O2cGzy3*EPsxQrU3+qB`&U}xSkJLc1R#zf1b=KDc!
zbiw*KdT3KA9q$IRwMFJ8sjAAa-~6+7JJnyxAMuo5QagEP90t(M5RQGz=lpNmEBDs~
zoDti`)5kr_FAEE5eWX7$pob%F@qk%KI0!8bdY&PRwG&yRG2T8k_biU?-v+3!F2i7C
zeJ;MyYcp>xwW|w?J}sX{Y;kxjo2lz9Ve;bSR}gHVB&T5d9&h@6Jvd*=z|;9_gN{4O
z)2KQ5F1Q$<hl3S(>1C3i`AjQ#tND}ae6w{6)&}(^XeXu{G6y+SWenR;TQ;tUN(GFY
zVE`geP}D49W}Byj^0_LiQjY;b!I=-<>%n?Nxf>~z=1&D62s4EV`Hv!xTj!?WQ(jp7
z3ygjFXVv+l)HJq6xch|k-j=}$Cpk=2Yis_v1b_~8RA7II4vuNP$JI~DdK1b4FLQeL
zLn@gsS2cbe%bQ5*{{;bQ*_%OH041-Rw+Zbn#pW>5il!w%4aSgSfn@hZ{Q?&0Zx0Ns
z*?k|luxFfI#+_->DDvHr%|EHT3k*thTLuUvZ$sWA>)=Za9mZ>k$CNW^cl}5)z6Dh)
z6wDs5PNrc&A}7G4Re+DANAPL!F@?_TM5Z!G{y-yQC<OF6*w=a-RIzBtn7a6yyP)uH
z42fjW%r`!g;m8y2iucGx*w4tu37nzGw<T<K@eud8`jhli6QDZd2FBL!dSfpv!&mrv
z0+*~V9sme6Qvg7i%{z*|17dyx++mtS7rO~@IYl;ye2S~V=|Jf&U}Zno`qm_L>;jKO
zlW(^a{$CcsL9nqpn*k=cCUy^O_!e^Sp#!QJh@Pm`ZDxQ-BSudcjkEOxRg7Gv7^#OZ
zz}@)n;@s7Mz7&H!sv#$%-E6(cg~~`+N5M;s1x%TQHScJg>h=XCkk8iyqQi05y4J0f
z;u(H-V$Z<7LO(xie=B~qate7t{rzPO&V{WJx^F$172j)>j7oyPMGQ0ga$_aTm-H#E
zH?wB2qz^WoJGd6X%&3E;7uDyoN=vnBsTBs6Xj;VVg1O0=G^@Cdv&5i0ia`VIaxWHM
z06Qvp(=UoA9pnW_3Aya#E_p8XgP>*dEgD)AHma1Z?zT+li#fT&UAi{uNv*ZlZMPa0
zIBN>@Y%@{%Q;M8bL$*b1PuH`klk>goy^>7RH?y_^GY9#LuRxxuMyb8U6P44w#a3rj
zia{$RPE~6?S{<dI6k;oM(Q8s?-7S)>X^Lw99!>Rd!uLT24UC4bV#a1!k5ZP_yBKHg
z`!Fkzp{FV6w)n@CHf@84?ecMaH=oG&%AJ_~x%Tu|^gh}B%Yk$Sjl>FfH`W5<Q7|8#
zAvp}6o|7bC^VGGtR6c<;8awa=nF}W~X3!GC-rs8k#9`pt6Jr0<wV%&~F<N;M0~XJE
z5VM)Z)qlCnw^hJ7^?gN#U~9T<M8+TGD}`=+4@t>8b@Z0NI(MBR2wKBI|1N-kGpdfo
zuN$>iCfigh(*aUb#K<(A%=Gs1IRvs9ouYiLYj)rx$n`qp$s*8`w7wxAf8LS+7Nt`Z
zaY*v78E4R^&SIawe%CsfUsTHuxTMgQk;uXFexwOL4iZP^j0(KM(>ntlY)xtkb`?o7
zAjz>jETxT}B!3-)=Qda_{#Mqi1<G1gF4-BiD6liKMAlR^DMBtsw)8}G7fdrP_`Jv}
z$C=3^@(yJ2$CX>q;an~0yb{U%Sl=UAZsZtweovHNxZ+cW5SJo@8-_4mPh@mfn7ek!
zcAzQ&%|lnK{GK=TI(l=`w8Vy$V{vk7A*N4lpvgdso7Rn;`qE$TJ%83R0tmun&dRgC
zsg1PWD*EEV9wqzW9e(w?=cPt|Vf$8Mtpm(v!F{hRz%p-Q^t*XEeK}b>W}c~ZDlZjR
zck@sM3vmJT_td=unL0P+%|G*gvSdyf>81O0CTZ`ih-xphyEJs8xOys)1q#~-`tD76
z=bSDr_C7Fsk$XYz-0{rfGYCMBJat^NPExqF!TPq<UAbUj%D^tb+g&@-f$vb37F}B&
z|0cGUFFy}nL8EJdXaL~vEmpo6-U=$|D?EFqQru_-(-CLRgv%RC7ojpokyjB&*!p<e
zt&Q6Fn!qnH81WOAE`@#6uP2wbgpFx+ar^B+H_?i&)-OB1emJgPa!XUCY{!frs8$i=
zSIf>v&sdU6m5&j5=|a>;5+Sqf^kNk~1#4b7*aqMwJHrt7J<1~3i_UH~Kob)F2d+V$
zX_x4WS;d3=Tfu8VQ(b%Tw|jLfXYnG4^AuPS&`vzL%ntm5yuJm4b$`DGp67J58Gnx$
zjtWEJZ_#d0YsmL2tlgQ9>V>zIjo?Dmr6x4Gfka(SDw+K5B451qLr_i-&@+4!7wLgJ
z1YXsR0*q<3A3Ko`N)7M#M9sKc2Zqk#qhjh7|4XE3lCmTj9fP|H9O1-uU_HPO(FmNF
z;_hD$pt8PNfKBoypXxY^z7{8|wZ118#ee<t)5U02%zxu-mCkEz;QS{laA~Ps(@8h^
z@4!W!L~bK(Lz9!h>Vf*lz^u)9bMK!YfRXC*K$vq>2RC&_-c)2T9(^Au#On+Kqgzup
z(@?GYhPD}N6k@iZ-jddPb$bYUEbeo_@^{iL?(-Ub1xXJr7#8e+6|LXG{0g1E0cJMd
zn#FS=J)X3m#xG_MTK1U5G(WNiU;5i_q^t~dl(zt3X<;SiDf_79iMXgeBsOxB?r660
ze3zf3GyoB;w_?QaB|TfzsVt-}J(y&c%mp7qW~u7vr4Ve*gLUgp8)qB8YzYin-ua}a
zdfl|t=)Wj^RM3J`dv{_8V{7_H@+;hujI(VCHK+$LOVp!QM17agDMEL3kUl>!Yu4)%
zp9iG)8{li>egS>!ZlgN-zhZhSfC#ZY?lv{NdW+7TCGj1x(;xpSgSQd-&}2WjP?Z1e
zLQ&>rT2kbutLXGzU3Z%_{4*ah<M+t3sfVrdO_{yJ2zoPXXy^v!(jLA42B@pJqwUu1
zK_=8&{0uUM?qsIis1*aoYE5Pw&{~F3{|no0Z_=(!ml}<cB9BZdkguatHP-p@#8<Qz
zKzI1p0>P^s_)GBIWsHcIoRzJDa%n*}e5Oz-N6O)c!wbR>R3*sMkm#NsKBI+M+iv^^
zV(S)DG+k!#Fy2|-e8F)6m^W&@N?9!4jlG5zNVhcIcrJ~Qq07A=@Y^48Jwq3RO39Pa
zAp32^$i(Z>v$0v__^f({UXico)IQQddEmlS)cW}!@EBRa;ji|SS;O+hhZF$HYf#d%
zqfqdWc^O1r{;N2b1A^<%{oa_?_);^=bm3yUZYQU1RAHW4>)EVM&(RV1-H}s3chUNd
zl=Pkg;Bpnh;~3X8Lmh!2a-aK|(tyXGTp63qH~x~*qM-H85XWj+PiMk1PQ}m=!{PZL
zfQ42bR1GQCVEBrzh~ixBk4i;Yq&tKe6=r|rQ!gz4xTXhpYNq%M2(-OBI6<u+robnh
z!dpRQI~=jWCaL<jMs85|UK9{k&n*jNCNPQz!yDr*8B#2bG3e@JJafgoCX?1H&8v=l
zINcny8WkV?%Ye!HLsnlXxi-18N_h)mGnzC|l&gmBN7wDRg5IkWfO~cQ-NJ^i4z67l
zM`j2NG-FIT2yQ(+jXfl<8=5rNT4o^|tb3lLmA5g=!<rr-$)#7N(L5xZ8%}i=dT*7_
zgi?#aWOZ#Q_ik_adIynebJ>=wBMYbft&-Z~lRM`*Yc~Z~jONJIO;TU;%9Q+n(_@>i
zS$h8IFVAznXKOae+=DH}j|<^d+U-v9U0O&tCtwwa-W{=;jCy66+5aIj>d|rX<vrs`
zY~R-Dq>qsvH4I)t{p;eJv#vGdz66NPKN3#v<tw*Nmnl1_7asX}f@_#Vh?o&sq<ds4
zNG`zKr?)V4v*zXknW1d4&JRF@EJUL(MEVSA3bZuxOmaUk^p!f6!+$Fz6YWn03x8}B
zi)%<|NriU3ww6t4R&~GKTe~!(yGsyWE(f9*<J8fBrl7Kzk9zgzc;_DffkJ2bS(k&v
z{Y$D$stGsIm)~51x+LnGfUQn3w4{W;{~Q0Bl~StO@0C40H#WX}2fu{FwU;e-VWee-
z$ZlaSX>ZTkN*98{c0hG+t4F*CaQ?cY=s#0V?LRr9Hi$F6`kN2HRA4JYM7NM?5Ct8P
zzd`a~FDV)YL7WS>RJs2aEGL--D<B?#g9A+hZJLWo=C|ub;XkW}RQ@}G^l$eG^$<kw
zom9l$(QYXRt!l$O&Vm*wGi9k=W|NLK_nhBO^jTxye}wQ%m9ok^UXrlcumZ}wq}KZJ
z)Is^vzfkIBcKS^MyudSh2}chIBiy8+W>%mBufrr|h{`{>``@=H<3(Fh`YaxRc){pE
zT5|KdcA(<6UV)Cjw6RGy*I6ujrr{S2aUOddSc1!^)gc;%gC+fIMJBAG>Clc{#30Lb
z9SHvN@$IwdBvCbw$1Ji`z8yp6HEF2ghWfxQ7(RBY_gdiD%b%{oG^~u781uZi&f#=E
z(hF~9NDX$b6cu#6{1P`ZsJW%F)7zaK*RiXvERCM+p5+NDzt1H4HU4e$IMy|JcNg1)
zR|eEk<vkekp0Dbo1LvtWUlr)CI0S9w#jB?{Yc0gUD1us`;^92%?=8iQ*Zi06($~m)
zDu1I-WD6R5gPY!_l$GF*`lKSd3Dz?=5GI1A$O5i7ni3U>G=v*ilZCPWMGr5@=lM^s
zLtP24v$L}AzB%BB@u+C|cG%CHDKEXffu4Znzvg^=63$TI6-3djWcg#!xFwmco+jnL
zywVyMqwYNI=Hu}XX-91dILVN0+1t{kZnoVVQOD!h19J7A@zQqf72qYu8kPR~hOfVA
z*o>FHHw_%u=_8^HD%u1LaZf>U^@(}mK33`63ut3Ih}*W=Sfy`n&Miuy^==OOGi{8H
zq22iPGpy78vgIncIx%d{{zu3obi-ek2AdP5wSj+_Ij@%0bc~HDf|hBb^Y`lTz+Z8>
zh~5sDg|2=Y<%yP{v04gRboBf>IQvFsg^017k&kUWSf+S^T)MPC2VGu2Se~>7Lx5^(
zc!x<n$aqgNz%*P6yDa+Ng7=i7CZGx9{YaO<U5MelN8koFNIY#CM@j7Te17iJs(bOh
zv1?o<h2Xa9NpRaW+U+K=0WAd40AZ!L5$z`aO|A7509$sV&=3?A@TbQZthS_%L?*G9
ziLTqIGAxty<x53;HZH}a|J)*gZ#tUxcLx8qv0r=_j=`C`N3Z3amGF7vCCL;_e+%rB
zu7mpE|9gLkAp&}0k;SoZ?jk4ByvVSSMUU9tTVq}T!sU0aT?xkklrX_)qO);qwlh82
zsbl4fSC2=oTH=i8Eo)?;rgj41<GOhbW+O|@E`?4!3xET`Ukb=6WEIr9XW^`Cc<)xQ
zi*h)cOleJfy!fRf1d2<`qi|K%cQm@h&AhUFq}IA06FC+$ARV)NMFxe@x41O$Hx1}h
zGeq_pcEs3{YTL~ZX`G!-V#h&m@?ChM@dmuS{Y&TOhdG>~A_W+<bLSSa8nn8m+PZtb
z({SF02ux&K{z1svpDmoM2gPAhT(FB(k~1U8j=DD^kJew{oKuiMk|)56-CQ&U?Awo~
zZw^eb*~<Bu&rQf5Rt}6O^z<Y7^hTT@1Ss5&_6_*Z1GWLB?t+9OdD(p$1kE5zeH2E!
z7&W&Z1(1Zu2V`L4Q4D~#`mn1m%Z(U=WkR_Qd}(v%BwSA30<P|iUfq1~wPlq>kRB<*
zdrGpWCb*yEqZhCp?9^#fY<5YG#b%r^m1G#Y8e3sLb-wv%r_E1{=#iz+TR%yOB<aG`
zDMG^bAs%buV_>IRYw0G$HprF~U6%06u&L(l7pK-Z>5ac|m(r0>SVR(XZ(t!KYq{{x
z(|Ea@lqiG3XFB&tCi;7LiBEz~81o&2;4~Xm$gSSbRS-IMAt@}OcE9^T^Y2FH`XQEa
z%t=yc(SqElM&3=XG^MM?`q(|mX7*j5jEBjeH#9oEjvLVO4LtR)nWD|vLPyaejXc_&
z9fze4=9JCUUF5mEX(UaZzdY#QO=za;fMY2@lFTOW*;ZKGvtD&7dz5`-Yk1KQDyMrc
z^4L4<Bz-Qg$*+Ff_BdY0Nv~48Cri(>o?u0~H!W$Dq8k7cZZ_#x${UbJV0fS91J*~`
z1m%Lv$ILE*j+3QwK6D7PFr2kWUNbfJ=H4<pMOMMY6d`1yx_Iyh*9Nt|lN|DgF>phm
zSmC*F;wxi$IE&cJpm8)`C>76Yt3&VA)Ck+!TtuIOU`GuuSZ$EM-@;ZE;--p1@CUPU
zyvgy8*-4Cx!%@ORGxiwQz=+oz`Dat$>1<-=)6V26Kby8&S1X5xCf_5&J(V_39L(9>
z8^Ha_^`M&WLbS_-@-vlGvgz|?;;~X>Ltm()-dW8n09q`qXDG}jk0$xlFPYq1)%^ST
zQ$oSa_{HJ-UvF;tNH|PB7EJn?)IT<6ix;LzmGiYbs+G^;1}Zgh@LbFWovvD%Qn^4I
zW-6bd;^aSXBfw1i!DTwL$sayBrmv{Kx?I}rZa)R!0rv5;q3*Q3BUb|%wk%x>n?x6R
z%5tH6saZ*!{Hp!v&>Jb8Ez7S>eBxlruqo*R9G7tx#<{4Ao&oUka3_H-BZ}_DcSIN6
zZAq(iR7s9gT<yafF>BifKL{*A&Oq@gr{?np;>O|r4K#-CA#zOpFOLLxMcBS8)B72G
zb>aWf&9(m}nWb^mE;HkpTkhnhHq=sDla_fc4Vkr!N;kWBH$bw|0%c6m6mRX$*hP~r
zo0m+e+`Pd8=3piwmX2tad3!}AFfXa0NS1(#;GVbsirx1|KD_5W=Q-zn&U?O>$G?9Y
zi0nBTSd4oP!MJI{m8{_=SF&nhU~9|8f#!v^r1Fg3<^{+(cV*D4YPwA~;PC;ZcmC3@
z*&E=+6(`8tH6Q?8^%?Y3o=3@IwnF4?PZl^i%^X(ZcaeSrJ$Gh(eOC&MzTFg2JrJsg
zLJUNKrNybdw!P!-`2G5Dzxyi6PYAYnpRN_n%BaNmINB`VQ&@r$F~Wwl<-i~$u_Uxb
zb&FIP^;EZinma?qVTh>0KxE-rLJt-ZVeRoqCe6gl90wx^Vpm;-mMQsY)dS4<BL`$+
z&O@dD4XSVh5kKCAO$^voP%dJn9B!KVY4syIugiW!+iF;9J5Y27_2)SYx7(rWTcH|P
zu*^U`tO#e%e|wNI;tqXpk$ObC>=KtQGU-LLiQ5t&&nTPK)9qk_eZ?gTl$oGv=@j*Y
z)#*-be`}u<+tB%O^XC#$s?gN7!%o`e-(P-CO*<o0PNE!55{*pUidd{D_CtAOC6+!R
zl7Gy9P$^=@?Xz_M-nW6;WVH?R^oWcGprgo^xW?<D+FRH_44NUTBWm+VrB|^a7=>q%
z3q@I7-y~5k5z-#;56*bzNopo``hZ2(2Kr>Ly9aW|w4_Ih4w#?ct|~tFU5HtBGN%_4
zcQ=JnRQ@!t!r9TE^a15*y`ClyEQX%bZThhNIep#O^yo(&C2Kn6F}4cBfEQ=Vj_fv-
z%%wVlt|J<~9Gr|20;qq4EE5LfaaVpAa_<`7QJS-Qoam4?)?h*X@ky?v`)fP7Y>v&;
zsW4eGMjacZA3Z>`n=Z)(sjYf)#Z52cDNq=xaWPG}csWb7XCnbvya5hifpsSr>e*E@
z6D($^cg`=4KYGL&Od;p8T5X|HK6?mBWYw_&5Wx&^Cm;H>q{=#PN2rMH?xd&lWT;8t
zWG+KZ+3`F;LhCQ{$wh?C_FmIzsK92`vpw>1a;PsdnZ#Q5<0jfv(={M--8r95-}jg(
zyaS~-a4%@D^gxq#<n`G=d+v=iLTGq7D~>_0Sp<$(-$&xR+lsj?A8+%*Ck_Q={JC3n
z{$kNI;OoGqm+YsbjZK2Oe@VeWLzHKL6iPE>NLgt10SwVOB*?fN>=FO-R|Ki{Z)!p%
zB|LExJeF7Dcoq(#z(9P2Zs6H8_^IaF75e*7-y85}bLO*JJT)1RqT%*?zt{)I-#s*I
zW17PlP6W*f2>l}rL8Tc&TE%g=HAg^9IT}xT=6~f$tCOYxLM34}<pUm-_xD1eW_Bc)
zTeoY2f*qXU2f~_)0`A=vH`N*N?KJ=@H?=O!Ol!X#34bv6Uz`a>SJg@8=Xqai+KfKq
z0o&=2a7M6SeH}Q|uPaVMl%S^`SN}3IbaQE8%t#?C0Y@+wao6g#@X@d_?^8?GSi34f
z9I(94OD>tcKg&Q89(O<xx-4fnUrZw;XQpURy{BrEzGyWOQuz^MAjRt17aX{YZNO1a
zsOvO-<@M?Im|!4r`c<(T!rCcPIfJ`>zDL@qbaInkK{}ObVuRVUR%~f?&;{-_nY82E
z2RP_46r(vl@rSZ-1po<}@zgajZ?<$c3=ZT2p@Yj7M(#Z)jc{XF@xaWjLr)1&G;kv#
z{KXmL=nT{^xiLV~Sp|~c<TVdF_N<z@r*E0%+DCZ3-GaM}c8-RJ7U5gqhfq@LdONOp
zl>Mi>4`-)P3Zv%?^%vO7)g-nEgYlRh+!YoDi$oFOFTt?t0~LO@nfTw22n8-U<7n}G
zod5_t2cFjwXtw30p&7tKM^K(h1b>fDd~G#_^K6pTlYoV}9U53Y(Be?3&7EL!7@ar2
z0dodxE*q!OsrEy%=Uek~%MBFjKS3Q)6Y)C4y9aP6U&+ByMh4iyT*Yf>qgZUt^}=L7
z0Kl~`IL$`X#2!PJ6YKg2HqCm`+ZkLt{+N(*FwO`eEr1&)qic!ZNv>3Kk0z^703{Ek
zRHnz@(8`{r5chxjFb)YSeWiLlC90`rPF1<HvH=gOkKNb2-K^At>Pk{2WH1p=aLd=o
zrUp%0NJIRV5O`Y7ojYsEd&52vb<vV%$vHu}hignHv9%+vSdZFBs``>@FEo7etuD=?
zsu;r{1a0hgd^{}#d;%LEF60m|D{v4<UASpl%fwMROK-{7O^H6^B>+@6WSwk=!6HWs
zpr%PnWosKjlUw|Q);Z+|xfhSf2a6Gl-s&eBzm`<-C8fDE-g==Ju$_F$My8-`e*QI$
zpOM(q`Va>>+cGexQ)k%QM^_@&lU1AR0bNo*YJJ-$AC$BSV<iQ;TffO*0flMM=evp*
zLWVA(>0j(N<p?*n;E1MM$Pm?}q^ZXgle54y>mNXmJ@hABI1qqf0cNpvlplOUEsC)x
z#Akkq#`x(Cs7!`y-$GBtUu80li`dk5hOohCPJgm~e688F`$ADE6Uh|l=jY{0#v=6s
z%oLDJ^=D|gn~V`|+p>Q)LA>gCsZpuIx~G@|=7c%7K>$iD^;q3LJygzUpx2er-<!02
zx@?BJ>=h_1A$CoUib-_tPOjIpj#_v++<t4S^tC$JY?F~$S6$;|Jn5yYO)SciwI;$?
z-N$5d(^Up?vJREGCZ@ox=E<5)LJe)l;;4Ah|NMMPQXrQxEP<QhJ=E*6er?6<9Oei6
zB8vCo1yU~#x<1XHL<<LE%teeO;0=;F)W!Q@sRF`@X~?RMexuBlFbWglmWS4p2x}^i
zvc39w{3Us8HB%?=cZ`|-V5YJRK8R3WC|ISL%o;l$DFshLNM9_+B%@9TSsNuB2dva8
z8am@mmh5F>pjJ9H=pYqt)pHb!7LuckT1}OD70f{J3YWaqBI`s+*zeG~pZ$?sAf4<P
z@DEnS7L2GpNBLTJd(FNNrlb67ZHhsoq}jpZT*P=ZAM95smbnXg*zgQkns=M^2n^-K
zu8AS}6eB+;rXiX;k>s|8u6fFVsht1ot8x9OWs|kB%h>iI??rf|hkcG7;CM&;<G=Yc
B7@`0G

diff --git a/articles/xcms-lcms-ms.html b/articles/xcms-lcms-ms.html
index eacac9276..9b53769ca 100644
--- a/articles/xcms-lcms-ms.html
+++ b/articles/xcms-lcms-ms.html
@@ -33,7 +33,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -100,7 +100,7 @@
     if (links[i].hostname != window.location.hostname)
       links[i].target = '_blank';
 });
-</script><p><strong>Package</strong>: <em><a href="https://bioconductor.org/packages/3.19/xcms" class="external-link">xcms</a></em><br><strong>Authors</strong>: Johannes Rainer, Michael Witting<br><strong>Modified</strong>: 2023-12-07 07:39:24.923344<br><strong>Compiled</strong>: Thu Dec 7 08:45:39 2023</p>
+</script><p><strong>Package</strong>: <em><a href="https://bioconductor.org/packages/3.19/xcms" class="external-link">xcms</a></em><br><strong>Authors</strong>: Johannes Rainer, Michael Witting<br><strong>Modified</strong>: 2024-01-26 18:14:18.95093<br><strong>Compiled</strong>: Fri Jan 26 20:09:08 2024</p>
 <div class="section level2">
 <h2 id="introduction">Introduction<a class="anchor" aria-label="anchor" href="#introduction"></a>
 </h2>
@@ -264,9 +264,9 @@ <h2 id="analysis-of-dda-data">Analysis of DDA data<a class="anchor" aria-label="
 <span><span class="co">## file(s):</span></span>
 <span><span class="co">## PestMix1_DDA.mzML</span></span>
 <span><span class="co">## Processing:</span></span>
-<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023]</span></span>
-<span><span class="co">##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023]</span></span>
-<span><span class="co">##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023]</span></span></code></pre>
+<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024]</span></span>
+<span><span class="co">##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024]</span></span>
+<span><span class="co">##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024]</span></span></code></pre>
 <p>By default <code>chromPeakSpectra</code> returns all spectra
 associated with a MS1 chromatographic peak, but parameter
 <code>method</code> allows to choose and return only one spectrum per
@@ -340,9 +340,9 @@ <h2 id="analysis-of-dda-data">Analysis of DDA data<a class="anchor" aria-label="
 <span><span class="co">## file(s):</span></span>
 <span><span class="co">## PestMix1_DDA.mzML</span></span>
 <span><span class="co">## Processing:</span></span>
-<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023]</span></span>
-<span><span class="co">##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023]</span></span>
-<span><span class="co">##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023]</span></span></code></pre>
+<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024]</span></span>
+<span><span class="co">##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024]</span></span>
+<span><span class="co">##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024]</span></span></code></pre>
 <p>There are 5 MS2 spectra representing fragmentation of the ion(s)
 measured in our candidate chromatographic peak. We next reduce this to a
 single MS2 spectrum using the <code>combineSpectra</code> method
@@ -368,9 +368,9 @@ <h2 id="analysis-of-dda-data">Analysis of DDA data<a class="anchor" aria-label="
 <span><span class="co">## 1         2   418.926      3505</span></span>
 <span><span class="co">##  ... 34 more variables/columns.</span></span>
 <span><span class="co">## Processing:</span></span>
-<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023]</span></span>
-<span><span class="co">##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023]</span></span>
-<span><span class="co">##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023]</span></span>
+<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024]</span></span>
+<span><span class="co">##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024]</span></span>
+<span><span class="co">##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024]</span></span>
 <span><span class="co">##  ...1 more processings. Use 'processingLog' to list all.</span></span></code></pre>
 <p>Mass peaks from all input spectra with a difference in m/z smaller 20
 ppm (parameter <code>ppm</code>) were combined into one peak and the
@@ -742,9 +742,10 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <div class="sourceCode" id="cb63"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">rtr</span> <span class="op">&lt;-</span> <span class="va">fenamiphos_ms1_peak</span><span class="op">[</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"rtmin"</span>, <span class="st">"rtmax"</span><span class="op">)</span><span class="op">]</span></span>
 <span><span class="va">mzr</span> <span class="op">&lt;-</span> <span class="va">fenamiphos_ms1_peak</span><span class="op">[</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"mzmin"</span>, <span class="st">"mzmax"</span><span class="op">)</span><span class="op">]</span></span>
-<span><span class="va">fenamiphos_ms1_chr</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">swath_data</span>, rt <span class="op">=</span> <span class="va">rtr</span>, mz <span class="op">=</span> <span class="va">mzr</span><span class="op">)</span></span>
-<span></span>
-<span><span class="va">rtr</span> <span class="op">&lt;-</span> <span class="va">fenamiphos_ms2_peak</span><span class="op">[</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"rtmin"</span>, <span class="st">"rtmax"</span><span class="op">)</span><span class="op">]</span></span>
+<span><span class="va">fenamiphos_ms1_chr</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">swath_data</span>, rt <span class="op">=</span> <span class="va">rtr</span>, mz <span class="op">=</span> <span class="va">mzr</span><span class="op">)</span></span></code></pre></div>
+<pre><code><span><span class="co">## Processing chromatographic peaks</span></span></code></pre>
+<div class="sourceCode" id="cb65"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="va">rtr</span> <span class="op">&lt;-</span> <span class="va">fenamiphos_ms2_peak</span><span class="op">[</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"rtmin"</span>, <span class="st">"rtmax"</span><span class="op">)</span><span class="op">]</span></span>
 <span><span class="va">mzr</span> <span class="op">&lt;-</span> <span class="va">fenamiphos_ms2_peak</span><span class="op">[</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"mzmin"</span>, <span class="st">"mzmax"</span><span class="op">)</span><span class="op">]</span></span>
 <span><span class="co">## Get the isolationWindowTargetMz for spectra containing the m/z of the</span></span>
 <span><span class="co">## compound of interest</span></span>
@@ -759,13 +760,14 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <p>The target m/z of the isolation window containing the m/z of interest
 is thus 299.1 and we can use this in the <code>chromatogram</code> call
 below to extract the data from the correct (MS2) spectra.</p>
-<div class="sourceCode" id="cb65"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb67"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">fenamiphos_ms2_chr</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span></span>
 <span>    <span class="va">swath_data</span>, rt <span class="op">=</span> <span class="va">rtr</span>, mz <span class="op">=</span> <span class="va">mzr</span>, msLevel <span class="op">=</span> <span class="fl">2L</span>,</span>
 <span>    isolationWindowTargetMz <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/rep.html" class="external-link">rep</a></span><span class="op">(</span><span class="fl">299.1</span>, <span class="fu"><a href="https://rdrr.io/pkg/BiocGenerics/man/nrow.html" class="external-link">nrow</a></span><span class="op">(</span><span class="va">rtr</span><span class="op">)</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
+<pre><code><span><span class="co">## Processing chromatographic peaks</span></span></code></pre>
 <p>We can now plot the extracted ion chromatogram of the MS1 and the
 extracted MS2 data.</p>
-<div class="sourceCode" id="cb66"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb69"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">rtime</a></span><span class="op">(</span><span class="va">fenamiphos_ms1_chr</span><span class="op">[</span><span class="fl">1</span>, <span class="fl">1</span><span class="op">]</span><span class="op">)</span>,</span>
 <span>     <span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">intensity</a></span><span class="op">(</span><span class="va">fenamiphos_ms1_chr</span><span class="op">[</span><span class="fl">1</span>, <span class="fl">1</span><span class="op">]</span><span class="op">)</span>,</span>
 <span>     xlab <span class="op">=</span> <span class="st">"retention time [s]"</span>, ylab <span class="op">=</span> <span class="st">"intensity"</span>, pch <span class="op">=</span> <span class="fl">16</span>,</span>
@@ -790,7 +792,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 before performing the correlation analysis. This is done automatically
 by the <code>correlate</code> function. See the help for the
 <code>align</code> method for more information on alignment options.</p>
-<div class="sourceCode" id="cb67"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb70"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">compareChromatograms</a></span><span class="op">(</span><span class="va">fenamiphos_ms2_chr</span><span class="op">[</span><span class="fl">1</span>, <span class="fl">1</span><span class="op">]</span>,</span>
 <span>               <span class="va">fenamiphos_ms1_chr</span><span class="op">[</span><span class="fl">1</span>, <span class="fl">1</span><span class="op">]</span>,</span>
 <span>               ALIGNFUNARGS <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/list.html" class="external-link">list</a></span><span class="op">(</span>method <span class="op">=</span> <span class="st">"approx"</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
@@ -808,7 +810,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 data requiring a peak shape correlation higher than <code>0.9</code>
 between the candidate MS2 chromatographic peak and the target MS1
 chromatographic peak.</p>
-<div class="sourceCode" id="cb69"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb72"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">swath_spectra</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/reconstructChromPeakSpectra.html">reconstructChromPeakSpectra</a></span><span class="op">(</span><span class="va">swath_data</span>, minCor <span class="op">=</span> <span class="fl">0.9</span><span class="op">)</span></span>
 <span><span class="va">swath_spectra</span></span></code></pre></div>
 <pre><code><span><span class="co">## MSn data (Spectra) with 62 spectra in a MsBackendMemory backend:</span></span>
@@ -827,13 +829,13 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <span><span class="co">## CP62         2   574.942        NA</span></span>
 <span><span class="co">##  ... 20 more variables/columns.</span></span>
 <span><span class="co">## Processing:</span></span>
-<span><span class="co">##  Merge 1 Spectra into one [Thu Dec  7 08:46:06 2023]</span></span></code></pre>
+<span><span class="co">##  Merge 1 Spectra into one [Fri Jan 26 20:09:40 2024]</span></span></code></pre>
 <p>As a result we got a <code>Spectra</code> object of length equal to
 the number of MS1 peaks in our data. The length of a spectrum represents
 the number of peaks it contains. Thus, a length of 0 indicates that no
 matching peak (MS2 signal) could be found for the respective MS1
 chromatographic peak.</p>
-<div class="sourceCode" id="cb71"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb74"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/base/lengths.html" class="external-link">lengths</a></span><span class="op">(</span><span class="va">swath_spectra</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##  [1]  0  0  1  1  1  0  0  0  0  0  0  0  3  0  3  4  0  3  0  1  0  9 14  1  0</span></span>
 <span><span class="co">## [26]  0 15  4  1  1  2  4  6 15 12 11  2  4 13  0  0  0  0  1  2  0  1  0  0  0</span></span>
@@ -845,7 +847,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 shape (<code>"ms2_peak_cor"</code>). Metadata column
 <code>"peak_id"</code> contains the ID of the MS1 chromatographic
 peak:</p>
-<div class="sourceCode" id="cb73"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb76"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">spectraData</a></span><span class="op">(</span><span class="va">swath_spectra</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"peak_id"</span>, <span class="st">"ms2_peak_id"</span>, <span class="st">"ms2_peak_cor"</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## DataFrame with 62 rows and 3 columns</span></span>
 <span><span class="co">##          peak_id ms2_peak_id      ms2_peak_cor</span></span>
@@ -864,7 +866,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <p>We next extract the MS2 spectrum for our example peak most likely
 representing [M+H]+ ions of Fenamiphos using its chromatographic peak
 ID:</p>
-<div class="sourceCode" id="cb75"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb78"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">fenamiphos_swath_spectrum</span> <span class="op">&lt;-</span> <span class="va">swath_spectra</span><span class="op">[</span></span>
 <span>    <span class="va">swath_spectra</span><span class="op">$</span><span class="va">peak_id</span> <span class="op">==</span> <span class="fu"><a href="https://rdrr.io/pkg/BiocGenerics/man/row_colnames.html" class="external-link">rownames</a></span><span class="op">(</span><span class="va">fenamiphos_ms1_peak</span><span class="op">)</span><span class="op">]</span></span></code></pre></div>
 <p>We can now compare the reconstructed spectrum to the example
@@ -874,10 +876,10 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 visualization we <em>normalize</em> also the peak intensities of the
 reconstructed SWATH spectrum with the same function we used for the
 experimental DDA spectrum.</p>
-<div class="sourceCode" id="cb76"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb79"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">fenamiphos_swath_spectrum</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">addProcessing</a></span><span class="op">(</span><span class="va">fenamiphos_swath_spectrum</span>,</span>
 <span>                                           <span class="va">scale_fun</span><span class="op">)</span></span></code></pre></div>
-<div class="sourceCode" id="cb77"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb80"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">2</span><span class="op">)</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/pkg/Spectra/man/spectra-plotting.html" class="external-link">plotSpectraMirror</a></span><span class="op">(</span><span class="va">fenamiphos_swath_spectrum</span>, <span class="va">ex_spectrum</span>,</span>
 <span>     ppm <span class="op">=</span> <span class="fl">50</span>, main <span class="op">=</span> <span class="st">"against DDA"</span><span class="op">)</span></span>
@@ -893,7 +895,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <p>If we wanted to get the EICs for the MS2 chromatographic peaks used
 to generate this MS2 spectrum we can use the IDs of these peaks which
 are provided with <code>$ms2_peak_id</code> of the result spectrum.</p>
-<div class="sourceCode" id="cb78"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb81"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">pk_ids</span> <span class="op">&lt;-</span> <span class="va">fenamiphos_swath_spectrum</span><span class="op">$</span><span class="va">ms2_peak_id</span><span class="op">[[</span><span class="fl">1</span><span class="op">]</span><span class="op">]</span></span>
 <span><span class="va">pk_ids</span></span></code></pre></div>
 <pre><code><span><span class="co">##  [1] "CP199" "CP201" "CP211" "CP208" "CP200" "CP202" "CP217" "CP215" "CP205"</span></span>
@@ -910,7 +912,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <code>chromatogram</code> extracts by default only data from MS1 spectra
 and we need to specify the target m/z of the isolation window containing
 the fragment data from the compound of interest.</p>
-<div class="sourceCode" id="cb80"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb83"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">rt_range</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">swath_data</span><span class="op">)</span><span class="op">[</span><span class="va">pk_ids</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"rtmin"</span>, <span class="st">"rtmax"</span><span class="op">)</span><span class="op">]</span></span>
 <span><span class="va">mz_range</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">swath_data</span><span class="op">)</span><span class="op">[</span><span class="va">pk_ids</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"mzmin"</span>, <span class="st">"mzmax"</span><span class="op">)</span><span class="op">]</span></span>
 <span></span>
@@ -925,10 +927,11 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <span><span class="va">ms2_eics</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span></span>
 <span>    <span class="va">swath_data</span>, rt <span class="op">=</span> <span class="va">rt_range</span>, mz <span class="op">=</span> <span class="va">mz_range</span>, msLevel <span class="op">=</span> <span class="fl">2L</span>,</span>
 <span>    isolationWindowTargetMz <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/rep.html" class="external-link">rep</a></span><span class="op">(</span><span class="va">tmz</span>, <span class="fu"><a href="https://rdrr.io/pkg/BiocGenerics/man/nrow.html" class="external-link">nrow</a></span><span class="op">(</span><span class="va">rt_range</span><span class="op">)</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
+<pre><code><span><span class="co">## Processing chromatographic peaks</span></span></code></pre>
 <p>Each row of this <code>ms2_eics</code> contains now the EIC of one of
 the MS2 chromatographic peaks. We can also plot these in an <em>overlay
 plot</em>.</p>
-<div class="sourceCode" id="cb81"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb85"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/plotChromatogramsOverlay.html">plotChromatogramsOverlay</a></span><span class="op">(</span><span class="va">ms2_eics</span><span class="op">)</span></span></code></pre></div>
 <div class="figure">
 <img src="xcms-lcms-ms_files/figure-html/unnamed-chunk-14-1.png" alt="Overlay of EICs of chromatographic peaks used to reconstruct the MS2 spectrum for fenamiphos." width="700"><p class="caption">
@@ -940,7 +943,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 m/z of 376.0381 (which would match <a href="https://en.wikipedia.org/wiki/Prochloraz" class="external-link">Prochloraz</a>). We
 first identify the MS1 chromatographic peak for that m/z and retrieve
 the reconstructed MS2 spectrum for that peak.</p>
-<div class="sourceCode" id="cb82"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb86"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">prochloraz_mz</span> <span class="op">&lt;-</span> <span class="fl">376.0381</span></span>
 <span></span>
 <span><span class="va">prochloraz_ms1_peak</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">swath_data</span>, msLevel <span class="op">=</span> <span class="fl">1L</span>,</span>
@@ -950,7 +953,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <span><span class="co">## CP22 376.0373 376.037 376.0374 405.046 401.446 409.546 3664.051 3655.951</span></span>
 <span><span class="co">##          maxo  sn sample</span></span>
 <span><span class="co">## CP22 897.3923 278      1</span></span></code></pre>
-<div class="sourceCode" id="cb84"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb88"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">prochloraz_swath_spectrum</span> <span class="op">&lt;-</span> <span class="va">swath_spectra</span><span class="op">[</span></span>
 <span>    <span class="va">swath_spectra</span><span class="op">$</span><span class="va">peak_id</span> <span class="op">==</span> <span class="fu"><a href="https://rdrr.io/pkg/BiocGenerics/man/row_colnames.html" class="external-link">rownames</a></span><span class="op">(</span><span class="va">prochloraz_ms1_peak</span><span class="op">)</span><span class="op">]</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/base/lengths.html" class="external-link">lengths</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum</span><span class="op">)</span></span></code></pre></div>
@@ -960,7 +963,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <p>In addition we identify the corresponding MS1 peak in the DDA data
 set, extract all measured MS2 chromatographic peaks and build the
 consensus spectrum from these.</p>
-<div class="sourceCode" id="cb86"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb90"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">prochloraz_dda_peak</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">dda_data</span>, msLevel <span class="op">=</span> <span class="fl">1L</span>,</span>
 <span>                                  mz <span class="op">=</span> <span class="va">prochloraz_mz</span>, ppm <span class="op">=</span> <span class="fl">5</span><span class="op">)</span></span>
 <span><span class="va">prochloraz_dda_peak</span></span></code></pre></div>
@@ -971,7 +974,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <p>The retention times for the chromatographic peaks from the DDA and
 SWATH data match almost perfectly. Next we get the MS2 spectra for this
 peak.</p>
-<div class="sourceCode" id="cb88"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb92"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">prochloraz_dda_spectra</span> <span class="op">&lt;-</span> <span class="va">dda_spectra</span><span class="op">[</span></span>
 <span>    <span class="va">dda_spectra</span><span class="op">$</span><span class="va">peak_id</span> <span class="op">==</span> <span class="fu"><a href="https://rdrr.io/pkg/BiocGenerics/man/row_colnames.html" class="external-link">rownames</a></span><span class="op">(</span><span class="va">prochloraz_dda_peak</span><span class="op">)</span><span class="op">]</span></span>
 <span><span class="va">prochloraz_dda_spectra</span></span></code></pre></div>
@@ -988,12 +991,12 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <span><span class="co">## file(s):</span></span>
 <span><span class="co">## PestMix1_DDA.mzML</span></span>
 <span><span class="co">## Processing:</span></span>
-<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023]</span></span>
-<span><span class="co">##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023]</span></span>
-<span><span class="co">##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023]</span></span></code></pre>
+<span><span class="co">##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024]</span></span>
+<span><span class="co">##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024]</span></span>
+<span><span class="co">##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024]</span></span></code></pre>
 <p>In total 5 spectra were measured, some with a relatively high number
 of peaks. Next we combine them into a consensus spectrum.</p>
-<div class="sourceCode" id="cb90"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb94"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">prochloraz_dda_spectrum</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/Spectra/man/Spectra.html" class="external-link">combineSpectra</a></span><span class="op">(</span></span>
 <span>    <span class="va">prochloraz_dda_spectra</span>, FUN <span class="op">=</span> <span class="va">combinePeaks</span>, ppm <span class="op">=</span> <span class="fl">20</span>,</span>
 <span>    peaks <span class="op">=</span> <span class="st">"intersect"</span>, minProp <span class="op">=</span> <span class="fl">0.8</span>, intensityFun <span class="op">=</span> <span class="va">median</span>, mzFun <span class="op">=</span> <span class="va">median</span>,</span>
@@ -1003,7 +1006,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <span><span class="co">## 'combinePeaksData' instead.</span></span></code></pre>
 <p>At last we load also the Prochloraz MS2 spectra (for different
 collision energies) from Metlin.</p>
-<div class="sourceCode" id="cb93"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb97"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">prochloraz</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/Spectra/man/Spectra.html" class="external-link">Spectra</a></span><span class="op">(</span></span>
 <span>    <span class="fu"><a href="https://rdrr.io/r/base/system.file.html" class="external-link">system.file</a></span><span class="op">(</span><span class="st">"mgf"</span>, <span class="st">"metlin-68898.mgf"</span>, package <span class="op">=</span> <span class="st">"xcms"</span><span class="op">)</span>,</span>
 <span>    source <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/pkg/MsBackendMgf/man/MsBackendMgf.html" class="external-link">MsBackendMgf</a></span><span class="op">(</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
@@ -1011,7 +1014,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <p>To validate the reconstructed spectrum we plot it against the
 corresponding DDA spectrum and the MS2 spectrum for Prochloraz (for a
 collision energy of 10V) from Metlin.</p>
-<div class="sourceCode" id="cb95"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb99"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">prochloraz_swath_spectrum</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">addProcessing</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum</span>, <span class="va">scale_fun</span><span class="op">)</span></span>
 <span><span class="va">prochloraz_dda_spectrum</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">addProcessing</a></span><span class="op">(</span><span class="va">prochloraz_dda_spectrum</span>, <span class="va">scale_fun</span><span class="op">)</span></span>
 <span></span>
@@ -1032,7 +1035,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 value in the reconstructed spectrum. Also, by closer inspecting the
 spectrum two groups of peaks with small differences in m/z can be
 observed (see plot below).</p>
-<div class="sourceCode" id="cb96"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb100"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/Spectra/man/spectra-plotting.html" class="external-link">plotSpectra</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum</span><span class="op">)</span></span></code></pre></div>
 <div class="figure">
 <img src="xcms-lcms-ms_files/figure-html/unnamed-chunk-15-1.png" alt="SWATH-derived MS2 spectrum for prochloraz." width="700"><p class="caption">
@@ -1045,7 +1048,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <code>isotopologues</code> function from the <em><a href="https://bioconductor.org/packages/3.19/MetaboCoreUtils" class="external-link">MetaboCoreUtils</a></em>
 package to check for presence of potential isotope peaks in the
 reconstructed MS2 spectrum for prochloraz.</p>
-<div class="sourceCode" id="cb97"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb101"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">library</a></span><span class="op">(</span><span class="va"><a href="https://github.com/RforMassSpectrometry/MetaboCoreUtils" class="external-link">MetaboCoreUtils</a></span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/pkg/MetaboCoreUtils/man/isotopologues.html" class="external-link">isotopologues</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/Spectra/man/hidden_aliases.html" class="external-link">peaksData</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum</span><span class="op">)</span><span class="op">[[</span><span class="fl">1</span><span class="op">]</span><span class="op">]</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [[1]]</span></span>
@@ -1059,7 +1062,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 their difference in m/z and relative intensity differences). Below we
 thus define a function that keeps only the monoisotopic peak for each
 isotope group in a spectrum.</p>
-<div class="sourceCode" id="cb99"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb103"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Function to keep only the first (monoisotopic) peak for potential</span></span>
 <span><span class="co">## isotopologue peak groups.</span></span>
 <span><span class="va">rem_iso</span> <span class="op">&lt;-</span> <span class="kw">function</span><span class="op">(</span><span class="va">x</span>, <span class="va">...</span><span class="op">)</span> <span class="op">{</span></span>
@@ -1071,7 +1074,7 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 <span><span class="op">}</span></span>
 <span><span class="va">prochloraz_swath_spectrum2</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">addProcessing</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum</span>,</span>
 <span>                                            <span class="va">rem_iso</span><span class="op">)</span></span></code></pre></div>
-<div class="sourceCode" id="cb100"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb104"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">2</span><span class="op">)</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/pkg/Spectra/man/spectra-plotting.html" class="external-link">plotSpectra</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/pkg/Spectra/man/spectra-plotting.html" class="external-link">plotSpectra</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum2</span><span class="op">)</span></span></code></pre></div>
@@ -1085,10 +1088,10 @@ <h3 id="reconstruction-of-ms2-spectra">Reconstruction of MS2 spectra<a class="an
 the spectra similarity score (since reference spectra generally will
 contain only fragments of the ion of interest, but not of any of its
 isotopes).</p>
-<div class="sourceCode" id="cb101"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb105"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">compareSpectra</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum</span>, <span class="va">prochloraz_dda_spectrum</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [1] 0.4623719</span></span></code></pre>
-<div class="sourceCode" id="cb103"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb107"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">compareSpectra</a></span><span class="op">(</span><span class="va">prochloraz_swath_spectrum2</span>, <span class="va">prochloraz_dda_spectrum</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [1] 0.5932303</span></span></code></pre>
 <p>Similar to the DDA data, the reconstructed MS2 spectra from SWATH
@@ -1125,9 +1128,9 @@ <h2 id="outlook">Outlook<a class="anchor" aria-label="anchor" href="#outlook"></
 <div class="section level2">
 <h2 id="session-information">Session information<a class="anchor" aria-label="anchor" href="#session-information"></a>
 </h2>
-<div class="sourceCode" id="cb105"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb109"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/utils/sessionInfo.html" class="external-link">sessionInfo</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
-<pre><code><span><span class="co">## R Under development (unstable) (2023-11-22 r85609)</span></span>
+<pre><code><span><span class="co">## R Under development (unstable) (2024-01-17 r85813)</span></span>
 <span><span class="co">## Platform: x86_64-pc-linux-gnu</span></span>
 <span><span class="co">## Running under: Ubuntu 22.04.3 LTS</span></span>
 <span><span class="co">## </span></span>
@@ -1151,67 +1154,67 @@ <h2 id="session-information">Session information<a class="anchor" aria-label="an
 <span><span class="co">## [8] base     </span></span>
 <span><span class="co">## </span></span>
 <span><span class="co">## other attached packages:</span></span>
-<span><span class="co">##  [1] MetaboCoreUtils_1.11.0 MsBackendMgf_1.11.0    MsExperiment_1.5.1    </span></span>
-<span><span class="co">##  [4] pander_0.6.5           Spectra_1.13.2         xcms_4.1.4            </span></span>
-<span><span class="co">##  [7] MSnbase_2.29.1         ProtGenerics_1.35.0    S4Vectors_0.41.2      </span></span>
-<span><span class="co">## [10] mzR_2.37.0             Rcpp_1.0.11            Biobase_2.63.0        </span></span>
+<span><span class="co">##  [1] MetaboCoreUtils_1.11.2 MsBackendMgf_1.11.1    MsExperiment_1.5.4    </span></span>
+<span><span class="co">##  [4] pander_0.6.5           Spectra_1.13.3         xcms_4.1.6            </span></span>
+<span><span class="co">##  [7] MSnbase_2.29.3         ProtGenerics_1.35.2    S4Vectors_0.41.3      </span></span>
+<span><span class="co">## [10] mzR_2.37.0             Rcpp_1.0.12            Biobase_2.63.0        </span></span>
 <span><span class="co">## [13] BiocGenerics_0.49.1    BiocParallel_1.37.0    BiocStyle_2.31.0      </span></span>
 <span><span class="co">## </span></span>
 <span><span class="co">## loaded via a namespace (and not attached):</span></span>
-<span><span class="co">##   [1] bitops_1.0-7                rlang_1.1.2                </span></span>
-<span><span class="co">##   [3] magrittr_2.0.3              clue_0.3-65                </span></span>
-<span><span class="co">##   [5] MassSpecWavelet_1.69.0      matrixStats_1.1.0          </span></span>
-<span><span class="co">##   [7] compiler_4.4.0              systemfonts_1.0.5          </span></span>
-<span><span class="co">##   [9] vctrs_0.6.5                 stringr_1.5.1              </span></span>
-<span><span class="co">##  [11] pkgconfig_2.0.3             crayon_1.5.2               </span></span>
-<span><span class="co">##  [13] fastmap_1.1.1               XVector_0.43.0             </span></span>
-<span><span class="co">##  [15] utf8_1.2.4                  rmarkdown_2.25             </span></span>
-<span><span class="co">##  [17] preprocessCore_1.65.0       ragg_1.2.6                 </span></span>
-<span><span class="co">##  [19] purrr_1.0.2                 MultiAssayExperiment_1.29.0</span></span>
-<span><span class="co">##  [21] xfun_0.41                   zlibbioc_1.49.0            </span></span>
-<span><span class="co">##  [23] cachem_1.0.8                GenomeInfoDb_1.39.1        </span></span>
-<span><span class="co">##  [25] jsonlite_1.8.8              progress_1.2.3             </span></span>
-<span><span class="co">##  [27] highr_0.10                  DelayedArray_0.29.0        </span></span>
-<span><span class="co">##  [29] prettyunits_1.2.0           parallel_4.4.0             </span></span>
-<span><span class="co">##  [31] cluster_2.1.6               R6_2.5.1                   </span></span>
-<span><span class="co">##  [33] bslib_0.6.1                 stringi_1.8.2              </span></span>
-<span><span class="co">##  [35] RColorBrewer_1.1-3          limma_3.59.1               </span></span>
-<span><span class="co">##  [37] GenomicRanges_1.55.1        jquerylib_0.1.4            </span></span>
-<span><span class="co">##  [39] bookdown_0.37               SummarizedExperiment_1.33.1</span></span>
-<span><span class="co">##  [41] iterators_1.0.14            knitr_1.45                 </span></span>
-<span><span class="co">##  [43] IRanges_2.37.0              igraph_1.5.1               </span></span>
-<span><span class="co">##  [45] splines_4.4.0               Matrix_1.6-4               </span></span>
-<span><span class="co">##  [47] tidyselect_1.2.0            abind_1.4-5                </span></span>
-<span><span class="co">##  [49] yaml_2.3.7                  doParallel_1.0.17          </span></span>
-<span><span class="co">##  [51] codetools_0.2-19            affy_1.81.0                </span></span>
-<span><span class="co">##  [53] lattice_0.22-5              tibble_3.2.1               </span></span>
-<span><span class="co">##  [55] plyr_1.8.9                  evaluate_0.23              </span></span>
-<span><span class="co">##  [57] survival_3.5-7              desc_1.4.2                 </span></span>
-<span><span class="co">##  [59] pillar_1.9.0                affyio_1.73.0              </span></span>
-<span><span class="co">##  [61] BiocManager_1.30.22         MatrixGenerics_1.15.0      </span></span>
-<span><span class="co">##  [63] foreach_1.5.2               MALDIquant_1.22.1          </span></span>
-<span><span class="co">##  [65] ncdf4_1.22                  generics_0.1.3             </span></span>
-<span><span class="co">##  [67] rprojroot_2.0.4             RCurl_1.98-1.13            </span></span>
+<span><span class="co">##   [1] DBI_1.2.1                   bitops_1.0-7               </span></span>
+<span><span class="co">##   [3] rlang_1.1.3                 magrittr_2.0.3             </span></span>
+<span><span class="co">##   [5] clue_0.3-65                 MassSpecWavelet_1.69.0     </span></span>
+<span><span class="co">##   [7] matrixStats_1.2.0           compiler_4.4.0             </span></span>
+<span><span class="co">##   [9] systemfonts_1.0.5           vctrs_0.6.5                </span></span>
+<span><span class="co">##  [11] stringr_1.5.1               pkgconfig_2.0.3            </span></span>
+<span><span class="co">##  [13] crayon_1.5.2                fastmap_1.1.1              </span></span>
+<span><span class="co">##  [15] XVector_0.43.1              utf8_1.2.4                 </span></span>
+<span><span class="co">##  [17] rmarkdown_2.25              preprocessCore_1.65.0      </span></span>
+<span><span class="co">##  [19] ragg_1.2.7                  purrr_1.0.2                </span></span>
+<span><span class="co">##  [21] xfun_0.41                   MultiAssayExperiment_1.29.0</span></span>
+<span><span class="co">##  [23] zlibbioc_1.49.0             cachem_1.0.8               </span></span>
+<span><span class="co">##  [25] GenomeInfoDb_1.39.5         jsonlite_1.8.8             </span></span>
+<span><span class="co">##  [27] progress_1.2.3              highr_0.10                 </span></span>
+<span><span class="co">##  [29] DelayedArray_0.29.0         prettyunits_1.2.0          </span></span>
+<span><span class="co">##  [31] parallel_4.4.0              cluster_2.1.6              </span></span>
+<span><span class="co">##  [33] R6_2.5.1                    RColorBrewer_1.1-3         </span></span>
+<span><span class="co">##  [35] bslib_0.6.1                 stringi_1.8.3              </span></span>
+<span><span class="co">##  [37] limma_3.59.1                GenomicRanges_1.55.2       </span></span>
+<span><span class="co">##  [39] jquerylib_0.1.4             bookdown_0.37              </span></span>
+<span><span class="co">##  [41] SummarizedExperiment_1.33.3 iterators_1.0.14           </span></span>
+<span><span class="co">##  [43] knitr_1.45                  IRanges_2.37.1             </span></span>
+<span><span class="co">##  [45] splines_4.4.0               Matrix_1.6-5               </span></span>
+<span><span class="co">##  [47] igraph_1.6.0                tidyselect_1.2.0           </span></span>
+<span><span class="co">##  [49] abind_1.4-5                 yaml_2.3.8                 </span></span>
+<span><span class="co">##  [51] doParallel_1.0.17           codetools_0.2-19           </span></span>
+<span><span class="co">##  [53] affy_1.81.0                 lattice_0.22-5             </span></span>
+<span><span class="co">##  [55] tibble_3.2.1                plyr_1.8.9                 </span></span>
+<span><span class="co">##  [57] evaluate_0.23               survival_3.5-7             </span></span>
+<span><span class="co">##  [59] desc_1.4.3                  pillar_1.9.0               </span></span>
+<span><span class="co">##  [61] affyio_1.73.0               BiocManager_1.30.22        </span></span>
+<span><span class="co">##  [63] MatrixGenerics_1.15.0       foreach_1.5.2              </span></span>
+<span><span class="co">##  [65] MALDIquant_1.22.2           ncdf4_1.22                 </span></span>
+<span><span class="co">##  [67] generics_0.1.3              RCurl_1.98-1.14            </span></span>
 <span><span class="co">##  [69] hms_1.1.3                   ggplot2_3.4.4              </span></span>
 <span><span class="co">##  [71] munsell_0.5.0               scales_1.3.0               </span></span>
-<span><span class="co">##  [73] glue_1.6.2                  lazyeval_0.2.2             </span></span>
-<span><span class="co">##  [75] MsFeatures_1.11.0           tools_4.4.0                </span></span>
-<span><span class="co">##  [77] mzID_1.41.0                 robustbase_0.99-1          </span></span>
-<span><span class="co">##  [79] QFeatures_1.13.1            vsn_3.71.0                 </span></span>
+<span><span class="co">##  [73] glue_1.7.0                  MsFeatures_1.11.0          </span></span>
+<span><span class="co">##  [75] lazyeval_0.2.2              tools_4.4.0                </span></span>
+<span><span class="co">##  [77] robustbase_0.99-1           mzID_1.41.0                </span></span>
+<span><span class="co">##  [79] QFeatures_1.13.2            vsn_3.71.0                 </span></span>
 <span><span class="co">##  [81] RANN_2.6.1                  fs_1.6.3                   </span></span>
-<span><span class="co">##  [83] XML_3.99-0.16               grid_4.4.0                 </span></span>
-<span><span class="co">##  [85] impute_1.77.0               MsCoreUtils_1.15.1         </span></span>
+<span><span class="co">##  [83] XML_3.99-0.16.1             grid_4.4.0                 </span></span>
+<span><span class="co">##  [85] impute_1.77.0               MsCoreUtils_1.15.3         </span></span>
 <span><span class="co">##  [87] colorspace_2.1-0            GenomeInfoDbData_1.2.11    </span></span>
-<span><span class="co">##  [89] cli_3.6.1                   textshaping_0.3.7          </span></span>
-<span><span class="co">##  [91] fansi_1.0.5                 S4Arrays_1.3.1             </span></span>
-<span><span class="co">##  [93] dplyr_1.1.4                 AnnotationFilter_1.27.0    </span></span>
-<span><span class="co">##  [95] pcaMethods_1.95.0           gtable_0.3.4               </span></span>
-<span><span class="co">##  [97] DEoptimR_1.1-3              sass_0.4.8                 </span></span>
-<span><span class="co">##  [99] digest_0.6.33               SparseArray_1.3.1          </span></span>
-<span><span class="co">## [101] multtest_2.59.0             memoise_2.0.1              </span></span>
-<span><span class="co">## [103] htmltools_0.5.7             pkgdown_2.0.7.9000         </span></span>
-<span><span class="co">## [105] lifecycle_1.0.4             statmod_1.5.0              </span></span>
-<span><span class="co">## [107] MASS_7.3-60.1</span></span></code></pre>
+<span><span class="co">##  [89] PSMatch_1.7.1               cli_3.6.2                  </span></span>
+<span><span class="co">##  [91] textshaping_0.3.7           fansi_1.0.6                </span></span>
+<span><span class="co">##  [93] S4Arrays_1.3.2              dplyr_1.1.4                </span></span>
+<span><span class="co">##  [95] AnnotationFilter_1.27.0     DEoptimR_1.1-3             </span></span>
+<span><span class="co">##  [97] pcaMethods_1.95.0           gtable_0.3.4               </span></span>
+<span><span class="co">##  [99] sass_0.4.8                  digest_0.6.34              </span></span>
+<span><span class="co">## [101] SparseArray_1.3.3           multtest_2.59.0            </span></span>
+<span><span class="co">## [103] memoise_2.0.1               htmltools_0.5.7            </span></span>
+<span><span class="co">## [105] pkgdown_2.0.7.9000          lifecycle_1.0.4            </span></span>
+<span><span class="co">## [107] statmod_1.5.0               MASS_7.3-60.2</span></span></code></pre>
 </div>
 <div class="section level2">
 <h2 class="unnumbered" id="references">References<a class="anchor" aria-label="anchor" href="#references"></a>
diff --git a/articles/xcms.html b/articles/xcms.html
index 87e672e2a..4939f55d6 100644
--- a/articles/xcms.html
+++ b/articles/xcms.html
@@ -33,7 +33,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -100,7 +100,7 @@
     if (links[i].hostname != window.location.hostname)
       links[i].target = '_blank';
 });
-</script><p><strong>Package</strong>: <em><a href="https://bioconductor.org/packages/3.19/xcms" class="external-link">xcms</a></em><br><strong>Authors</strong>: Johannes Rainer<br><strong>Modified</strong>: 2023-12-07 07:39:24.923344<br><strong>Compiled</strong>: Thu Dec 7 08:46:13 2023</p>
+</script><p><strong>Package</strong>: <em><a href="https://bioconductor.org/packages/3.19/xcms" class="external-link">xcms</a></em><br><strong>Authors</strong>: Johannes Rainer<br><strong>Modified</strong>: 2024-01-26 18:14:18.95093<br><strong>Compiled</strong>: Fri Jan 26 20:09:48 2024</p>
 <div class="section level2">
 <h2 id="introduction">Introduction<a class="anchor" aria-label="anchor" href="#introduction"></a>
 </h2>
@@ -125,7 +125,8 @@ <h2 id="introduction">Introduction<a class="anchor" aria-label="anchor" href="#i
 &gt;= 4. The same functions can be applied to the older
 <em>MSnbase</em>-based workflows (xcms version 3). Additional documents
 and tutorials covering also other topics of untargeted metabolomics
-analysis are listed at the end of this document.</p>
+analysis are listed at the end of this document. There is also a <a href="https://jorainer.github.io/xcmsTutorials" class="external-link">xcms tutorial</a>
+available with more examples and details.</p>
 </div>
 <div class="section level2">
 <h2 id="pre-processing-of-lc-ms-data">Pre-processing of LC-MS data<a class="anchor" aria-label="anchor" href="#pre-processing-of-lc-ms-data"></a>
@@ -470,7 +471,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 the ranges of m/z values for several compounds (either internal
 standards or compounds known to be present in the sample) and define the
 <code>ppm</code> parameter for <em>centWave</em> according to these. See
-also this <a href="https://jorainer.github.io/metabolomics2018" class="external-link">tutorial</a> for
+also this <a href="https://jorainer.github.io/xcmsTutorials" class="external-link">tutorial</a> for
 additional information and examples on choosing and testing peak
 detection settings.</p>
 <p>Chromatographic peak detection can also be performed on extracted ion
@@ -651,9 +652,13 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 <p>An example for a merged peak is given below.</p>
 <div class="sourceCode" id="cb38"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">mzr_1</span> <span class="op">&lt;-</span> <span class="fl">305.1</span> <span class="op">+</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="op">-</span><span class="fl">0.01</span>, <span class="fl">0.01</span><span class="op">)</span></span>
-<span><span class="va">chr_1</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span>
-<span><span class="va">chr_2</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko_pp</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span>
-<span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">2</span><span class="op">)</span><span class="op">)</span></span>
+<span><span class="va">chr_1</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span></code></pre></div>
+<pre><code><span><span class="co">## Processing chromatographic peaks</span></span></code></pre>
+<div class="sourceCode" id="cb40"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="va">chr_2</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko_pp</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span></code></pre></div>
+<pre><code><span><span class="co">## Processing chromatographic peaks</span></span></code></pre>
+<div class="sourceCode" id="cb42"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">2</span><span class="op">)</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_1</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_2</span><span class="op">)</span></span></code></pre></div>
 <div class="figure">
@@ -668,11 +673,15 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 merged the first two peaks into a single one (right panel in the figure
 above). Other close peaks, with a lower intensity between them, were
 however not merged (see below).</p>
-<div class="sourceCode" id="cb39"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb43"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">mzr_1</span> <span class="op">&lt;-</span> <span class="fl">496.2</span> <span class="op">+</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="op">-</span><span class="fl">0.01</span>, <span class="fl">0.01</span><span class="op">)</span></span>
-<span><span class="va">chr_1</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span>
-<span><span class="va">chr_2</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko_pp</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span>
-<span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">2</span><span class="op">)</span><span class="op">)</span></span>
+<span><span class="va">chr_1</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span></code></pre></div>
+<pre><code><span><span class="co">## Processing chromatographic peaks</span></span></code></pre>
+<div class="sourceCode" id="cb45"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="va">chr_2</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko_pp</span><span class="op">[</span><span class="fl">1</span><span class="op">]</span>, mz <span class="op">=</span> <span class="va">mzr_1</span><span class="op">)</span></span></code></pre></div>
+<pre><code><span><span class="co">## Processing chromatographic peaks</span></span></code></pre>
+<div class="sourceCode" id="cb47"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">2</span><span class="op">)</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_1</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_2</span><span class="op">)</span></span></code></pre></div>
 <div class="figure">
@@ -686,25 +695,25 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 settings for the parameter. To illustrate this we perform below a peak
 refinement on the extracted ion chromatogram <code>chr_1</code> reducing
 the <code>minProp</code> parameter to force joining the two peaks.</p>
-<div class="sourceCode" id="cb40"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb48"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">res</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/refineChromPeaks.html">refineChromPeaks</a></span><span class="op">(</span><span class="va">chr_1</span>, <span class="fu"><a href="../reference/refineChromPeaks.html">MergeNeighboringPeaksParam</a></span><span class="op">(</span>minProp <span class="op">=</span> <span class="fl">0.05</span><span class="op">)</span><span class="op">)</span></span>
 <span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##         mz  mzmin  mzmax       rt    rtmin    rtmax     into intb    maxo  sn</span></span>
 <span><span class="co">## CPM1 496.2 496.19 496.21 3384.012 3294.809 3412.181 45940118   NA 1128960 177</span></span>
 <span><span class="co">##      sample row column</span></span>
 <span><span class="co">## CPM1      1   1      1</span></span></code></pre>
-<div class="sourceCode" id="cb42"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb50"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></code></pre></div>
 <p><img src="xcms_files/figure-html/peak-postprocessing-chr-1.png" width="700"></p>
 <p>Before proceeding we next replace the <code>faahko</code> object with
 the results from the peak refinement step.</p>
-<div class="sourceCode" id="cb43"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb51"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">faahko</span> <span class="op">&lt;-</span> <span class="va">faahko_pp</span></span></code></pre></div>
 <p>Below we use the data from the <code>chromPeaks</code> matrix to
 calculate per-file summaries of the peak detection results, such as the
 number of peaks per file as well as the distribution of the retention
 time widths.</p>
-<div class="sourceCode" id="cb44"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb52"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">summary_fun</span> <span class="op">&lt;-</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span></span>
 <span>    <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span>peak_count <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/pkg/BiocGenerics/man/nrow.html" class="external-link">nrow</a></span><span class="op">(</span><span class="va">z</span><span class="op">)</span>, rt <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span><span class="op">[</span>, <span class="st">"rtmax"</span><span class="op">]</span> <span class="op">-</span> <span class="va">z</span><span class="op">[</span>, <span class="st">"rtmin"</span><span class="op">]</span><span class="op">)</span><span class="op">)</span></span>
 <span></span>
@@ -819,7 +828,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 <p>While by default <code>chromPeaks</code> will return all identified
 chromatographic peaks in a result object it is also possible to extract
 only chromatographic peaks for a specified m/z and/or rt range:</p>
-<div class="sourceCode" id="cb45"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb53"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">faahko</span>, mz <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">334.9</span>, <span class="fl">335.1</span><span class="op">)</span>, rt <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">2700</span>, <span class="fl">2900</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##         mz mzmin mzmax       rt    rtmin    rtmax      into      intb  maxo sn</span></span>
 <span><span class="co">## CP0038 335   335   335 2781.505 2761.160 2809.674  412134.3  383167.4 16856 23</span></span>
@@ -837,7 +846,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 in the m/z - retention time space for one file using the
 <code>plotChromPeaks</code> function. Below we plot this information for
 the third sample.</p>
-<div class="sourceCode" id="cb47"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb55"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/plotChromPeaks.html">plotChromPeaks</a></span><span class="op">(</span><span class="va">faahko</span>, file <span class="op">=</span> <span class="fl">3</span><span class="op">)</span></span></code></pre></div>
 <div class="figure" style="text-align: center">
 <img src="xcms_files/figure-html/peak-detection-chrom-peaks-plot-1.png" alt="Identified chromatographic peaks in the m/z by retention time space for one sample." width="768"><p class="caption">
@@ -851,7 +860,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 allows to define the width of the bins in rt dimension in which peaks
 should be counted. This number of chromatographic peaks within each bin
 is then shown color-coded in the resulting plot.</p>
-<div class="sourceCode" id="cb48"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb56"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/plotChromPeaks.html">plotChromPeakImage</a></span><span class="op">(</span><span class="va">faahko</span>, binSize <span class="op">=</span> <span class="fl">10</span><span class="op">)</span></span></code></pre></div>
 <div class="figure" style="text-align: center">
 <img src="xcms_files/figure-html/peak-detection-chrom-peak-image-1.png" alt="Frequency of identified chromatographic peaks along the retention time axis. The frequency is color coded with higher frequency being represented by yellow-white. Each line shows the peak frequency for one file." width="960"><p class="caption">
@@ -869,7 +878,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 properly. Below we extract the ion chromatogram for the m/z - rt region
 above and access the detected peaks in that region using the
 <code>chromPeaks</code> function.</p>
-<div class="sourceCode" id="cb49"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb57"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">chr_ex</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko</span>, mz <span class="op">=</span> <span class="va">mzr</span>, rt <span class="op">=</span> <span class="va">rtr</span><span class="op">)</span></span>
 <span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">chr_ex</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##         mz mzmin mzmax       rt    rtmin    rtmax      into      intb  maxo sn</span></span>
@@ -886,7 +895,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 <span><span class="co">## CP2349      7   1      7</span></span></code></pre>
 <p>We can also plot this extracted ion chromatogram which will also
 visualize all identified chromatographic peaks in that region.</p>
-<div class="sourceCode" id="cb51"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb59"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">sample_colors</span> <span class="op">&lt;-</span> <span class="va">group_colors</span><span class="op">[</span><span class="va">chr_ex</span><span class="op">$</span><span class="va">sample_group</span><span class="op">]</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_ex</span>, col <span class="op">=</span> <span class="va">group_colors</span><span class="op">[</span><span class="va">chr_raw</span><span class="op">$</span><span class="va">sample_group</span><span class="op">]</span>, lwd <span class="op">=</span> <span class="fl">2</span>,</span>
 <span>     peakBg <span class="op">=</span> <span class="va">sample_colors</span><span class="op">[</span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">chr_ex</span><span class="op">)</span><span class="op">[</span>, <span class="st">"sample"</span><span class="op">]</span><span class="op">]</span><span class="op">)</span></span></code></pre></div>
@@ -905,7 +914,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 identified chromatographic peak with a single point
 (<code>peakType = "point"</code>). Below we plot the data again using
 <code>peakType = "rectangle"</code>.</p>
-<div class="sourceCode" id="cb52"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb60"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_ex</span>, col <span class="op">=</span> <span class="va">sample_colors</span>, peakType <span class="op">=</span> <span class="st">"rectangle"</span>,</span>
 <span>     peakCol <span class="op">=</span> <span class="va">sample_colors</span><span class="op">[</span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">chr_ex</span><span class="op">)</span><span class="op">[</span>, <span class="st">"sample"</span><span class="op">]</span><span class="op">]</span>,</span>
 <span>     peakBg <span class="op">=</span> <span class="cn">NA</span><span class="op">)</span></span></code></pre></div>
@@ -919,7 +928,7 @@ <h3 id="chromatographic-peak-detection">Chromatographic peak detection<a class="
 <p>Finally we plot also the distribution of peak intensity per sample.
 This allows to investigate whether systematic differences in peak
 signals between samples are present.</p>
-<div class="sourceCode" id="cb53"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb61"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Extract a list of per-sample peak intensities (in log2 scale)</span></span>
 <span><span class="va">ints</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/S4Vectors/man/splitAsList.html" class="external-link">split</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/Log.html" class="external-link">log2</a></span><span class="op">(</span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">[</span>, <span class="st">"into"</span><span class="op">]</span><span class="op">)</span>,</span>
 <span>              f <span class="op">=</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">[</span>, <span class="st">"sample"</span><span class="op">]</span><span class="op">)</span></span>
@@ -961,7 +970,7 @@ <h3 id="alignment">Alignment<a class="anchor" aria-label="anchor" href="#alignme
 <code>binSize = 0.6</code> which creates warping functions in m/z bins
 of 0.6. Also here it is advisable to modify and adapt the settings for
 each experiment.</p>
-<div class="sourceCode" id="cb54"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb62"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">faahko</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/adjustRtime.html">adjustRtime</a></span><span class="op">(</span><span class="va">faahko</span>, param <span class="op">=</span> <span class="fu"><a href="../reference/adjustRtime.html">ObiwarpParam</a></span><span class="op">(</span>binSize <span class="op">=</span> <span class="fl">0.6</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
 <p>Note that <code>adjustRtime</code>, besides calculating adjusted
 retention times for each spectrum, adjusts also the retention times of
@@ -970,15 +979,15 @@ <h3 id="alignment">Alignment<a class="anchor" aria-label="anchor" href="#alignme
 result object using either the <code>adjustedRtime</code> function or
 using <code>rtime</code> with parameter <code>adjusted = TRUE</code>
 (the default):</p>
-<div class="sourceCode" id="cb55"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb63"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Extract adjusted retention times</span></span>
 <span><span class="fu"><a href="../reference/XCMSnExp-class.html">adjustedRtime</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [1] 2551.457 2553.089 2554.720 2556.352 2557.983 2559.615</span></span></code></pre>
-<div class="sourceCode" id="cb57"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb65"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Or simply use the rtime method</span></span>
 <span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">rtime</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [1] 2551.457 2553.089 2554.720 2556.352 2557.983 2559.615</span></span></code></pre>
-<div class="sourceCode" id="cb59"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb67"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Get raw (unadjusted) retention times</span></span>
 <span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">rtime</a></span><span class="op">(</span><span class="va">faahko</span>, adjusted <span class="op">=</span> <span class="cn">FALSE</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [1] 2551.457 2553.022 2554.586 2556.151 2557.716 2559.281</span></span></code></pre>
@@ -989,7 +998,7 @@ <h3 id="alignment">Alignment<a class="anchor" aria-label="anchor" href="#alignme
 extraction of all identified chromatographic peaks by the
 <code>chromatogram</code> function (which would not make much sense for
 a BPC) we use <code>chromPeaks = "none"</code> below.</p>
-<div class="sourceCode" id="cb61"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb69"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Get the base peak chromatograms.</span></span>
 <span><span class="va">bpis_adj</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">faahko</span>, aggregationFun <span class="op">=</span> <span class="st">"max"</span>, chromPeaks <span class="op">=</span> <span class="st">"none"</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">3</span>, <span class="fl">1</span><span class="op">)</span>, mar <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">4.5</span>, <span class="fl">4.2</span>, <span class="fl">1</span>, <span class="fl">0.5</span><span class="op">)</span><span class="op">)</span></span>
@@ -1011,7 +1020,7 @@ <h3 id="alignment">Alignment<a class="anchor" aria-label="anchor" href="#alignme
 indicate poorly performing samples or alignment.</p>
 <p>At last we evaluate also the impact of the alignment on the test
 peak.</p>
-<div class="sourceCode" id="cb62"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb70"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">2</span>, <span class="fl">1</span><span class="op">)</span><span class="op">)</span></span>
 <span><span class="co">## Plot the raw data</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_raw</span>, col <span class="op">=</span> <span class="va">sample_colors</span><span class="op">)</span></span>
@@ -1101,7 +1110,7 @@ <h4 id="subset-based-alignment">Subset-based alignment<a class="anchor" aria-lab
 retention times for all spectra, this function will also
 <em>restore</em> the original retention times for identified
 chromatographic peaks.</p>
-<div class="sourceCode" id="cb63"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb71"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">faahko</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">dropAdjustedRtime</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span></span>
 <span></span>
 <span><span class="co">## Define the experimental layout</span></span>
@@ -1119,7 +1128,7 @@ <h4 id="subset-based-alignment">Subset-based alignment<a class="anchor" aria-lab
 <code>PeakDensityParam</code>. If there are no sample groups in the
 experiment, <code>sampleGroups</code> should be set to a single value
 for each file (e.g. <code>rep(1, length(fileNames(faahko))</code>).</p>
-<div class="sourceCode" id="cb64"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb72"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Initial peak grouping. Use sample_type as grouping variable</span></span>
 <span><span class="va">pdp_subs</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/groupChromPeaks.html">PeakDensityParam</a></span><span class="op">(</span>sampleGroups <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/pkg/MsExperiment/man/MsExperiment.html" class="external-link">sampleData</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">$</span><span class="va">sample_type</span>,</span>
 <span>                             minFraction <span class="op">=</span> <span class="fl">0.9</span><span class="op">)</span></span>
@@ -1139,7 +1148,7 @@ <h4 id="subset-based-alignment">Subset-based alignment<a class="anchor" aria-lab
 retention times of 2 being more similar to those of sample 1. Sample 3
 on the other hand gets adjusted giving more weight to the second subset
 sample (4).</p>
-<div class="sourceCode" id="cb65"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb73"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">clrs</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/rep.html" class="external-link">rep</a></span><span class="op">(</span><span class="st">"#00000040"</span>, <span class="fl">8</span><span class="op">)</span></span>
 <span><span class="va">clrs</span><span class="op">[</span><span class="fu"><a href="https://rdrr.io/pkg/MsExperiment/man/MsExperiment.html" class="external-link">sampleData</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">$</span><span class="va">sample_type</span> <span class="op">==</span> <span class="st">"QC"</span><span class="op">]</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"#00ce0080"</span><span class="op">)</span></span>
 <span><span class="fu"><a href="https://rdrr.io/r/graphics/par.html" class="external-link">par</a></span><span class="op">(</span>mfrow <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">2</span>, <span class="fl">1</span><span class="op">)</span>, mar <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">4</span>, <span class="fl">4.5</span>, <span class="fl">1</span>, <span class="fl">0.5</span><span class="op">)</span><span class="op">)</span></span>
@@ -1187,7 +1196,7 @@ <h3 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 40% of samples a chromatographic peak was present) - parameter
 <code>sampleGroups</code> is used to define to which sample group each
 sample belongs.</p>
-<div class="sourceCode" id="cb66"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb74"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Define the mz slice.</span></span>
 <span><span class="va">mzr</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">305.05</span>, <span class="fl">305.15</span><span class="op">)</span></span>
 <span></span>
@@ -1226,14 +1235,60 @@ <h3 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 to be OK and we are thus applying them to the full data set below.
 Especially the parameter <code>bw</code> will be very data set dependent
 (or more specifically LC-dependent) and should be adapted to each data
-set. See the <a href="https://jorainer.github.io/metabolomics2018" class="external-link">Metabolomics
-pre-processing with <code>xcms</code></a> tutorial for examples and more
-details.</p>
-<div class="sourceCode" id="cb67"><pre class="downlit sourceCode r">
-<code class="sourceCode R"><span><span class="co">## Perform the correspondence</span></span>
+set.</p>
+<p>Another important parameter is <code>binSize</code> that defines the
+size of the m/z slices (bins) within which peaks are being grouped. This
+parameter thus defines the required similarity in m/z values for the
+chromatographic peaks that are then assumed to represent signal from the
+same (type of ion of a) compound and hence evaluated for grouping. By
+default, a constant m/z bin size is used, but by changing parameter
+<code>ppm</code> to a value larger than 0, m/z-relative bin sizes would
+be used instead (i.e., the bin size will increase with the m/z value
+hence better representing the measurement error/precision of some MS
+instruments). The bin sizes (and subsequently the m/z width of the
+defined features) would then reach a maximal value of
+<code>binSize</code> plus <code>ppm</code> parts-per-million of the
+largest m/z value of any chromatographic peak in the data set.</p>
+<p>See also the <a href="https://jorainer.github.io/xcmsTutorials" class="external-link">xcms
+tutorial</a> for more examples and details.</p>
+<div class="sourceCode" id="cb75"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="co">## Perform the correspondence using fixed m/z bin sizes.</span></span>
 <span><span class="va">pdp</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/groupChromPeaks.html">PeakDensityParam</a></span><span class="op">(</span>sampleGroups <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/pkg/MsExperiment/man/MsExperiment.html" class="external-link">sampleData</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">$</span><span class="va">sample_group</span>,</span>
 <span>                        minFraction <span class="op">=</span> <span class="fl">0.4</span>, bw <span class="op">=</span> <span class="fl">30</span><span class="op">)</span></span>
 <span><span class="va">faahko</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/groupChromPeaks.html">groupChromPeaks</a></span><span class="op">(</span><span class="va">faahko</span>, param <span class="op">=</span> <span class="va">pdp</span><span class="op">)</span></span></code></pre></div>
+<p>As an alternative we perform the correspondence using m/z relative
+bin sizes.</p>
+<div class="sourceCode" id="cb76"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="co">## Drop feature definitions and re-perform the correspondence</span></span>
+<span><span class="co">## using m/z-relative bin sizes.</span></span>
+<span><span class="va">faahko_ppm</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/groupChromPeaks.html">groupChromPeaks</a></span><span class="op">(</span></span>
+<span>    <span class="fu"><a href="../reference/XCMSnExp-class.html">dropFeatureDefinitions</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span>,</span>
+<span>    <span class="fu"><a href="../reference/groupChromPeaks.html">PeakDensityParam</a></span><span class="op">(</span>sampleGroups <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/pkg/MsExperiment/man/MsExperiment.html" class="external-link">sampleData</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">$</span><span class="va">sample_group</span>,</span>
+<span>                     minFraction <span class="op">=</span> <span class="fl">0.4</span>, bw <span class="op">=</span> <span class="fl">30</span>, ppm <span class="op">=</span> <span class="fl">10</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
+<p>The results will be <em>mostly</em> similar, except for the higher
+m/z range (in which larger m/z bins will be used). Below we plot the m/z
+range for features against their median m/z. For the present data set
+(acquired with a triple quad instrument) no clear difference can be seen
+for the two approaches hence we proceed the analysis with the fixed bin
+size setting. A stronger relationship would be expected for example for
+data measured on TOF instruments.</p>
+<div class="sourceCode" id="cb77"><pre class="downlit sourceCode r">
+<code class="sourceCode R"><span><span class="co">## Calculate m/z width of features</span></span>
+<span><span class="va">mzw</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">$</span><span class="va">mzmax</span> <span class="op">-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">$</span><span class="va">mzmin</span></span>
+<span><span class="va">mzw_ppm</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">faahko_ppm</span><span class="op">)</span><span class="op">$</span><span class="va">mzmax</span> <span class="op">-</span></span>
+<span>                                        <span class="fu"><a href="../reference/XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">faahko_ppm</span><span class="op">)</span><span class="op">$</span><span class="va">mzmin</span></span>
+<span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="fu"><a href="../reference/XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">faahko_ppm</span><span class="op">)</span><span class="op">$</span><span class="va">mzmed</span>, <span class="va">mzw_ppm</span>,</span>
+<span>     xlab <span class="op">=</span> <span class="st">"m/z"</span>, ylab <span class="op">=</span> <span class="st">"m/z width"</span>, pch <span class="op">=</span> <span class="fl">21</span>,</span>
+<span>     col <span class="op">=</span> <span class="st">"#0000ff20"</span>, bg <span class="op">=</span> <span class="st">"#0000ff10"</span><span class="op">)</span></span>
+<span><span class="fu"><a href="https://rdrr.io/r/graphics/points.html" class="external-link">points</a></span><span class="op">(</span><span class="fu"><a href="../reference/XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">$</span><span class="va">mzmed</span>, <span class="va">mzw</span>, pch <span class="op">=</span> <span class="fl">21</span>,</span>
+<span>     col <span class="op">=</span> <span class="st">"#ff000020"</span>, bg <span class="op">=</span> <span class="st">"#ff000010"</span><span class="op">)</span></span></code></pre></div>
+<div class="figure">
+<img src="xcms_files/figure-html/unnamed-chunk-9-1.png" alt="Relationship between a feature's m/z and the m/z width (max - min m/z) of the feature. Red points represent the results with the fixed m/z bin size, blue with the m/z-relative bin size." width="700"><p class="caption">
+Relationship between a feature’s m/z and the m/z width (max - min m/z)
+of the feature. Red points represent the results with the fixed m/z bin
+size, blue with the m/z-relative bin size.
+</p>
+</div>
 <p>Results from the correspondence analysis can be accessed with the
 <code>featureDefinitions</code> and <code>featureValues</code> function.
 The former returns a data frame with general information on each of the
@@ -1242,7 +1297,7 @@ <h3 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 of the chromatographic peaks assigned to the feature in column
 <code>"peakidx"</code>. Below we show the information on the first 6
 features.</p>
-<div class="sourceCode" id="cb68"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb78"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##       mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks KO WT      peakidx</span></span>
 <span><span class="co">## FT001 200.1 200.1 200.1 2902.634 2882.603 2922.664      2  2  0    458, 1161</span></span>
@@ -1266,7 +1321,7 @@ <h3 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 peaks corresponding to a feature in a sample are returned. This is then
 generally used as the intensity matrix for downstream analysis. Below we
 extract the intensities for the first 6 features.</p>
-<div class="sourceCode" id="cb70"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb80"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/XCMSnExp-peak-grouping-results.html">featureValues</a></span><span class="op">(</span><span class="va">faahko</span>, value <span class="op">=</span> <span class="st">"into"</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF</span></span>
 <span><span class="co">## FT001        NA  506848.9        NA  169955.6        NA        NA        NA</span></span>
@@ -1306,7 +1361,7 @@ <h3 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 <strong>all</strong> features in a data set, which can take also a
 considerable amount of time. Below we extract the chromatograms for the
 first 4 features.</p>
-<div class="sourceCode" id="cb72"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb82"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">feature_chroms</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/featureChromatograms.html">featureChromatograms</a></span><span class="op">(</span><span class="va">faahko</span>, features <span class="op">=</span> <span class="fl">1</span><span class="op">:</span><span class="fl">4</span><span class="op">)</span></span>
 <span></span>
 <span><span class="va">feature_chroms</span></span></code></pre></div>
@@ -1333,7 +1388,7 @@ <h3 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 <span><span class="co">##  4 feature(s) identified.</span></span></code></pre>
 <p>And plot the extracted ion chromatograms. We again use the group
 color for each identified peak to fill the area.</p>
-<div class="sourceCode" id="cb74"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb84"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">feature_chroms</span>, col <span class="op">=</span> <span class="va">sample_colors</span>,</span>
 <span>     peakBg <span class="op">=</span> <span class="va">sample_colors</span><span class="op">[</span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">feature_chroms</span><span class="op">)</span><span class="op">[</span>, <span class="st">"sample"</span><span class="op">]</span><span class="op">]</span><span class="op">)</span></span></code></pre></div>
 <div class="figure" style="text-align: center">
@@ -1343,7 +1398,7 @@ <h3 id="correspondence">Correspondence<a class="anchor" aria-label="anchor" href
 </div>
 <p>To access the EICs of the second feature we can simply subset the
 <code>feature_chroms</code> object.</p>
-<div class="sourceCode" id="cb75"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb85"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">eic_2</span> <span class="op">&lt;-</span> <span class="va">feature_chroms</span><span class="op">[</span><span class="fl">2</span>, <span class="op">]</span></span>
 <span><span class="fu"><a href="../reference/XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">eic_2</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##         mz mzmin mzmax       rt    rtmin    rtmax    into    intb  maxo sn</span></span>
@@ -1390,7 +1445,7 @@ <h3 id="gap-filling">Gap filling<a class="anchor" aria-label="anchor" href="#gap
 we perform this gap filling on our test data and extract the feature
 values for the first 6 features after gap filling. An <code>NA</code> is
 reported if no signal is measured at all for a specific sample.</p>
-<div class="sourceCode" id="cb77"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb87"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">faahko</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/fillChromPeaks.html">fillChromPeaks</a></span><span class="op">(</span><span class="va">faahko</span>, param <span class="op">=</span> <span class="fu"><a href="../reference/fillChromPeaks.html">ChromPeakAreaParam</a></span><span class="op">(</span><span class="op">)</span><span class="op">)</span></span>
 <span></span>
 <span><span class="fu"><a href="../reference/XCMSnExp-peak-grouping-results.html">featureValues</a></span><span class="op">(</span><span class="va">faahko</span>, value <span class="op">=</span> <span class="st">"into"</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
@@ -1400,7 +1455,7 @@ <h3 id="gap-filling">Gap filling<a class="anchor" aria-label="anchor" href="#gap
 <span><span class="co">## FT003  213659.3  289500.7  164380.7  178285.7  253825.6  241844.4  240606.0</span></span>
 <span><span class="co">## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  226234.4</span></span>
 <span><span class="co">## FT005  286221.4  285857.6  164009.0  149097.6  255697.7  311296.8  366441.5</span></span>
-<span><span class="co">## FT006 1160580.5 1129426.4  380970.3  588986.4 1286883.0 1739516.6  639755.3</span></span>
+<span><span class="co">## FT006 1160580.5 1102832.6  380970.3  588986.4 1286883.0 1739516.6  639755.3</span></span>
 <span><span class="co">##        wt22.CDF</span></span>
 <span><span class="co">## FT001  138341.2</span></span>
 <span><span class="co">## FT002 1354004.9</span></span>
@@ -1430,7 +1485,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 chromatographic peak was assigned to a feature (for that option it is
 important to run <code>refineChromPeaks</code> like described above to
 merge overlapping peaks in each sample).</p>
-<div class="sourceCode" id="cb79"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb89"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">library</a></span><span class="op">(</span><span class="va"><a href="https://bioconductor.org/packages/SummarizedExperiment" class="external-link">SummarizedExperiment</a></span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## Loading required package: MatrixGenerics</span></span></code></pre>
 <pre><code><span><span class="co">## Loading required package: matrixStats</span></span></code></pre>
@@ -1469,7 +1524,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <span><span class="co">## </span></span>
 <span><span class="co">##     distance</span></span></code></pre>
 <pre><code><span><span class="co">## Loading required package: GenomeInfoDb</span></span></code></pre>
-<div class="sourceCode" id="cb92"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb102"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">res</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">quantify</a></span><span class="op">(</span><span class="va">faahko</span>, value <span class="op">=</span> <span class="st">"into"</span>, method <span class="op">=</span> <span class="st">"sum"</span><span class="op">)</span></span>
 <span><span class="va">res</span></span></code></pre></div>
 <pre><code><span><span class="co">## class: SummarizedExperiment </span></span>
@@ -1485,7 +1540,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 provides annotations and information for each <strong>row</strong> in
 the <code>SummarizedExperiment</code> (which in our case are the
 <strong>features</strong>).</p>
-<div class="sourceCode" id="cb94"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb104"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">rowData</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## DataFrame with 351 rows and 10 columns</span></span>
 <span><span class="co">##           mzmed     mzmin     mzmax     rtmed     rtmin     rtmax    npeaks</span></span>
@@ -1518,7 +1573,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <strong>samples</strong>) are stored as <code>colData</code>. In this
 data frame each row contains annotations for one sample (and hence one
 column in the feature values matrix).</p>
-<div class="sourceCode" id="cb96"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb106"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">colData</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## DataFrame with 8 rows and 4 columns</span></span>
 <span><span class="co">##          sample_name sample_group spectraOrigin sample_type</span></span>
@@ -1536,13 +1591,13 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 multiple assays which have to be numeric matrices with the number of
 rows and columns matching the number of features and samples,
 respectively. Below we list the names of the available assays.</p>
-<div class="sourceCode" id="cb98"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb108"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">assayNames</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [1] "raw"</span></span></code></pre>
 <p>And we can access the actual data using the <code>assay</code>
 function, optionally also providing the name of the assay we want to
 access. Below we show the first 6 lines of that matrix.</p>
-<div class="sourceCode" id="cb100"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb110"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">assay</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF</span></span>
 <span><span class="co">## FT001  135162.4  506848.9  111657.3  169955.6  209929.4  141607.9  226853.7</span></span>
@@ -1550,7 +1605,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <span><span class="co">## FT003  213659.3  289500.7  164380.7  178285.7  253825.6  241844.4  240606.0</span></span>
 <span><span class="co">## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  226234.4</span></span>
 <span><span class="co">## FT005  286221.4  285857.6  164009.0  149097.6  255697.7  311296.8  366441.5</span></span>
-<span><span class="co">## FT006 1923307.8 1129426.4  380970.3  588986.4 1286883.0 1739516.6  639755.3</span></span>
+<span><span class="co">## FT006 1923307.8 1102832.6  380970.3  588986.4 1286883.0 1739516.6  639755.3</span></span>
 <span><span class="co">##        wt22.CDF</span></span>
 <span><span class="co">## FT001  138341.2</span></span>
 <span><span class="co">## FT002 1354004.9</span></span>
@@ -1562,14 +1617,14 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 we in addition add also the feature value matrix
 <strong>without</strong> filled-in values (i.e. feature intensities that
 were added by the gap filling step).</p>
-<div class="sourceCode" id="cb102"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb112"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">assays</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span><span class="op">$</span><span class="va">raw_nofill</span> <span class="op">&lt;-</span> <span class="fu"><a href="../reference/XCMSnExp-peak-grouping-results.html">featureValues</a></span><span class="op">(</span><span class="va">faahko</span>, filled <span class="op">=</span> <span class="cn">FALSE</span>, method <span class="op">=</span> <span class="st">"sum"</span><span class="op">)</span></span></code></pre></div>
 <p>With that we have now two assays in our result object.</p>
-<div class="sourceCode" id="cb103"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb113"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">assayNames</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [1] "raw"        "raw_nofill"</span></span></code></pre>
 <p>And we can extract the feature values without gap-filling:</p>
-<div class="sourceCode" id="cb105"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb115"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">assay</a></span><span class="op">(</span><span class="va">res</span>, <span class="st">"raw_nofill"</span><span class="op">)</span> <span class="op">|&gt;</span> <span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF</span></span>
 <span><span class="co">## FT001        NA  506848.9        NA  169955.6        NA        NA        NA</span></span>
@@ -1588,12 +1643,12 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <p>Finally, a history of the full processing with <em>xcms</em> is
 available as <code>metadata</code> in the
 <code>SummarizedExperiment</code>.</p>
-<div class="sourceCode" id="cb107"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb117"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/S4Vectors/man/Annotated-class.html" class="external-link">metadata</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## [[1]]</span></span>
 <span><span class="co">## Object of class "XProcessHistory"</span></span>
 <span><span class="co">##  type: Peak detection </span></span>
-<span><span class="co">##  date: Thu Dec  7 08:46:45 2023 </span></span>
+<span><span class="co">##  date: Fri Jan 26 20:10:29 2024 </span></span>
 <span><span class="co">##  info:  </span></span>
 <span><span class="co">##  fileIndex: 1,2,3,4,5,6,7,8 </span></span>
 <span><span class="co">##  Parameter class: CentWaveParam </span></span>
@@ -1602,7 +1657,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <span><span class="co">## [[2]]</span></span>
 <span><span class="co">## Object of class "XProcessHistory"</span></span>
 <span><span class="co">##  type: Peak refinement </span></span>
-<span><span class="co">##  date: Thu Dec  7 08:46:48 2023 </span></span>
+<span><span class="co">##  date: Fri Jan 26 20:10:32 2024 </span></span>
 <span><span class="co">##  info:  </span></span>
 <span><span class="co">##  fileIndex: 1,2,3,4,5,6,7,8 </span></span>
 <span><span class="co">##  Parameter class: MergeNeighboringPeaksParam </span></span>
@@ -1611,7 +1666,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <span><span class="co">## [[3]]</span></span>
 <span><span class="co">## Object of class "XProcessHistory"</span></span>
 <span><span class="co">##  type: Peak grouping </span></span>
-<span><span class="co">##  date: Thu Dec  7 08:46:59 2023 </span></span>
+<span><span class="co">##  date: Fri Jan 26 20:10:46 2024 </span></span>
 <span><span class="co">##  info:  </span></span>
 <span><span class="co">##  fileIndex: 1,2,3,4,5,6,7,8 </span></span>
 <span><span class="co">##  Parameter class: PeakDensityParam </span></span>
@@ -1620,7 +1675,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <span><span class="co">## [[4]]</span></span>
 <span><span class="co">## Object of class "XProcessHistory"</span></span>
 <span><span class="co">##  type: Retention time correction </span></span>
-<span><span class="co">##  date: Thu Dec  7 08:47:00 2023 </span></span>
+<span><span class="co">##  date: Fri Jan 26 20:10:46 2024 </span></span>
 <span><span class="co">##  info:  </span></span>
 <span><span class="co">##  fileIndex: 1,2,3,4,5,6,7,8 </span></span>
 <span><span class="co">##  Parameter class: PeakGroupsParam </span></span>
@@ -1629,7 +1684,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <span><span class="co">## [[5]]</span></span>
 <span><span class="co">## Object of class "XProcessHistory"</span></span>
 <span><span class="co">##  type: Peak grouping </span></span>
-<span><span class="co">##  date: Thu Dec  7 08:47:04 2023 </span></span>
+<span><span class="co">##  date: Fri Jan 26 20:10:51 2024 </span></span>
 <span><span class="co">##  info:  </span></span>
 <span><span class="co">##  fileIndex: 1,2,3,4,5,6,7,8 </span></span>
 <span><span class="co">##  Parameter class: PeakDensityParam </span></span>
@@ -1638,7 +1693,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <span><span class="co">## [[6]]</span></span>
 <span><span class="co">## Object of class "XProcessHistory"</span></span>
 <span><span class="co">##  type: Missing peak filling </span></span>
-<span><span class="co">##  date: Thu Dec  7 08:47:08 2023 </span></span>
+<span><span class="co">##  date: Fri Jan 26 20:10:58 2024 </span></span>
 <span><span class="co">##  info:  </span></span>
 <span><span class="co">##  fileIndex: 1,2,3,4,5,6,7,8 </span></span>
 <span><span class="co">##  Parameter class: ChromPeakAreaParam </span></span>
@@ -1646,11 +1701,11 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <p>This same information can also be extracted from the <em>xcms</em>
 result object using the <code>processHistory</code> function. Below we
 extract the information for the first processing step.</p>
-<div class="sourceCode" id="cb109"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb119"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/XCMSnExp-class.html">processHistory</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">[[</span><span class="fl">1</span><span class="op">]</span><span class="op">]</span></span></code></pre></div>
 <pre><code><span><span class="co">## Object of class "XProcessHistory"</span></span>
 <span><span class="co">##  type: Peak detection </span></span>
-<span><span class="co">##  date: Thu Dec  7 08:46:45 2023 </span></span>
+<span><span class="co">##  date: Fri Jan 26 20:10:29 2024 </span></span>
 <span><span class="co">##  info:  </span></span>
 <span><span class="co">##  fileIndex: 1,2,3,4,5,6,7,8 </span></span>
 <span><span class="co">##  Parameter class: CentWaveParam </span></span>
@@ -1658,7 +1713,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 <p>These processing steps contain also the individual parameter objects
 used for the analysis, hence allowing to exactly reproduce the
 analysis.</p>
-<div class="sourceCode" id="cb111"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb121"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="../reference/XCMSnExp-class.html">processHistory</a></span><span class="op">(</span><span class="va">faahko</span><span class="op">)</span><span class="op">[[</span><span class="fl">1</span><span class="op">]</span><span class="op">]</span> <span class="op">|&gt;</span> <span class="fu"><a href="../reference/ProcessHistory-class.html">processParam</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
 <pre><code><span><span class="co">## Object of class:  CentWaveParam </span></span>
 <span><span class="co">##  Parameters:</span></span>
@@ -1681,7 +1736,7 @@ <h3 id="final-result">Final result<a class="anchor" aria-label="anchor" href="#f
 the grouping of the samples in this experiment. Note that we did not
 perform any data normalization hence the grouping might (and will) also
 be influenced by technical biases.</p>
-<div class="sourceCode" id="cb113"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb123"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co">## Extract the features and log2 transform them</span></span>
 <span><span class="va">ft_ints</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/Log.html" class="external-link">log2</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/SummarizedExperiment/man/SummarizedExperiment-class.html" class="external-link">assay</a></span><span class="op">(</span><span class="va">res</span>, <span class="st">"raw"</span><span class="op">)</span><span class="op">)</span></span>
 <span></span>
@@ -1759,10 +1814,10 @@ <h3 id="parallel-processing">Parallel processing<a class="anchor" aria-label="an
 globally we <code>register</code> the parameter class. Note also that
 <code>bpstart</code> is used below to initialize the parallel
 processes.</p>
-<div class="sourceCode" id="cb114"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb124"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/register.html" class="external-link">register</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/BiocParallelParam-class.html" class="external-link">bpstart</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/MulticoreParam-class.html" class="external-link">MulticoreParam</a></span><span class="op">(</span><span class="fl">2</span><span class="op">)</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
 <p>Windows supports only socket-based parallel processing:</p>
-<div class="sourceCode" id="cb115"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb125"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/register.html" class="external-link">register</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/BiocParallelParam-class.html" class="external-link">bpstart</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/SnowParam-class.html" class="external-link">SnowParam</a></span><span class="op">(</span><span class="fl">2</span><span class="op">)</span><span class="op">)</span><span class="op">)</span></span></code></pre></div>
 </div>
 <div class="section level3">
@@ -1803,9 +1858,9 @@ <h2 id="additional-documentation-resources">Additional documentation resources<a
 <h2 id="session-information">Session information<a class="anchor" aria-label="anchor" href="#session-information"></a>
 </h2>
 <p>R packages used for this document are listed below.</p>
-<div class="sourceCode" id="cb116"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb126"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/utils/sessionInfo.html" class="external-link">sessionInfo</a></span><span class="op">(</span><span class="op">)</span></span></code></pre></div>
-<pre><code><span><span class="co">## R Under development (unstable) (2023-11-22 r85609)</span></span>
+<pre><code><span><span class="co">## R Under development (unstable) (2024-01-17 r85813)</span></span>
 <span><span class="co">## Platform: x86_64-pc-linux-gnu</span></span>
 <span><span class="co">## Running under: Ubuntu 22.04.3 LTS</span></span>
 <span><span class="co">## </span></span>
@@ -1829,71 +1884,71 @@ <h2 id="session-information">Session information<a class="anchor" aria-label="an
 <span><span class="co">## [8] base     </span></span>
 <span><span class="co">## </span></span>
 <span><span class="co">## other attached packages:</span></span>
-<span><span class="co">##  [1] SummarizedExperiment_1.33.1 GenomicRanges_1.55.1       </span></span>
-<span><span class="co">##  [3] GenomeInfoDb_1.39.1         IRanges_2.37.0             </span></span>
-<span><span class="co">##  [5] MatrixGenerics_1.15.0       matrixStats_1.1.0          </span></span>
-<span><span class="co">##  [7] MsExperiment_1.5.1          pheatmap_1.0.12            </span></span>
+<span><span class="co">##  [1] SummarizedExperiment_1.33.3 GenomicRanges_1.55.2       </span></span>
+<span><span class="co">##  [3] GenomeInfoDb_1.39.5         IRanges_2.37.1             </span></span>
+<span><span class="co">##  [5] MatrixGenerics_1.15.0       matrixStats_1.2.0          </span></span>
+<span><span class="co">##  [7] MsExperiment_1.5.4          pheatmap_1.0.12            </span></span>
 <span><span class="co">##  [9] RColorBrewer_1.1-3          pander_0.6.5               </span></span>
-<span><span class="co">## [11] faahKO_1.43.0               xcms_4.1.4                 </span></span>
-<span><span class="co">## [13] MSnbase_2.29.1              ProtGenerics_1.35.0        </span></span>
-<span><span class="co">## [15] S4Vectors_0.41.2            mzR_2.37.0                 </span></span>
-<span><span class="co">## [17] Rcpp_1.0.11                 Biobase_2.63.0             </span></span>
+<span><span class="co">## [11] faahKO_1.43.0               xcms_4.1.6                 </span></span>
+<span><span class="co">## [13] MSnbase_2.29.3              ProtGenerics_1.35.2        </span></span>
+<span><span class="co">## [15] S4Vectors_0.41.3            mzR_2.37.0                 </span></span>
+<span><span class="co">## [17] Rcpp_1.0.12                 Biobase_2.63.0             </span></span>
 <span><span class="co">## [19] BiocGenerics_0.49.1         BiocParallel_1.37.0        </span></span>
 <span><span class="co">## [21] BiocStyle_2.31.0           </span></span>
 <span><span class="co">## </span></span>
 <span><span class="co">## loaded via a namespace (and not attached):</span></span>
-<span><span class="co">##   [1] bitops_1.0-7                rlang_1.1.2                </span></span>
-<span><span class="co">##   [3] magrittr_2.0.3              clue_0.3-65                </span></span>
-<span><span class="co">##   [5] MassSpecWavelet_1.69.0      compiler_4.4.0             </span></span>
-<span><span class="co">##   [7] systemfonts_1.0.5           vctrs_0.6.5                </span></span>
-<span><span class="co">##   [9] stringr_1.5.1               MetaboCoreUtils_1.11.0     </span></span>
-<span><span class="co">##  [11] pkgconfig_2.0.3             crayon_1.5.2               </span></span>
-<span><span class="co">##  [13] fastmap_1.1.1               XVector_0.43.0             </span></span>
-<span><span class="co">##  [15] utf8_1.2.4                  rmarkdown_2.25             </span></span>
-<span><span class="co">##  [17] preprocessCore_1.65.0       ragg_1.2.6                 </span></span>
-<span><span class="co">##  [19] purrr_1.0.2                 MultiAssayExperiment_1.29.0</span></span>
-<span><span class="co">##  [21] xfun_0.41                   zlibbioc_1.49.0            </span></span>
-<span><span class="co">##  [23] cachem_1.0.8                jsonlite_1.8.8             </span></span>
-<span><span class="co">##  [25] progress_1.2.3              highr_0.10                 </span></span>
-<span><span class="co">##  [27] DelayedArray_0.29.0         prettyunits_1.2.0          </span></span>
-<span><span class="co">##  [29] parallel_4.4.0              cluster_2.1.6              </span></span>
-<span><span class="co">##  [31] R6_2.5.1                    bslib_0.6.1                </span></span>
-<span><span class="co">##  [33] stringi_1.8.2               limma_3.59.1               </span></span>
-<span><span class="co">##  [35] jquerylib_0.1.4             bookdown_0.37              </span></span>
-<span><span class="co">##  [37] iterators_1.0.14            knitr_1.45                 </span></span>
-<span><span class="co">##  [39] igraph_1.5.1                splines_4.4.0              </span></span>
-<span><span class="co">##  [41] Matrix_1.6-4                tidyselect_1.2.0           </span></span>
-<span><span class="co">##  [43] abind_1.4-5                 yaml_2.3.7                 </span></span>
-<span><span class="co">##  [45] doParallel_1.0.17           codetools_0.2-19           </span></span>
-<span><span class="co">##  [47] affy_1.81.0                 lattice_0.22-5             </span></span>
-<span><span class="co">##  [49] tibble_3.2.1                plyr_1.8.9                 </span></span>
-<span><span class="co">##  [51] evaluate_0.23               survival_3.5-7             </span></span>
-<span><span class="co">##  [53] desc_1.4.2                  Spectra_1.13.2             </span></span>
-<span><span class="co">##  [55] pillar_1.9.0                affyio_1.73.0              </span></span>
-<span><span class="co">##  [57] BiocManager_1.30.22         foreach_1.5.2              </span></span>
-<span><span class="co">##  [59] MALDIquant_1.22.1           ncdf4_1.22                 </span></span>
-<span><span class="co">##  [61] generics_0.1.3              rprojroot_2.0.4            </span></span>
-<span><span class="co">##  [63] RCurl_1.98-1.13             hms_1.1.3                  </span></span>
+<span><span class="co">##   [1] DBI_1.2.1                   bitops_1.0-7               </span></span>
+<span><span class="co">##   [3] rlang_1.1.3                 magrittr_2.0.3             </span></span>
+<span><span class="co">##   [5] clue_0.3-65                 MassSpecWavelet_1.69.0     </span></span>
+<span><span class="co">##   [7] compiler_4.4.0              systemfonts_1.0.5          </span></span>
+<span><span class="co">##   [9] vctrs_0.6.5                 stringr_1.5.1              </span></span>
+<span><span class="co">##  [11] MetaboCoreUtils_1.11.2      pkgconfig_2.0.3            </span></span>
+<span><span class="co">##  [13] crayon_1.5.2                fastmap_1.1.1              </span></span>
+<span><span class="co">##  [15] XVector_0.43.1              utf8_1.2.4                 </span></span>
+<span><span class="co">##  [17] rmarkdown_2.25              preprocessCore_1.65.0      </span></span>
+<span><span class="co">##  [19] ragg_1.2.7                  purrr_1.0.2                </span></span>
+<span><span class="co">##  [21] xfun_0.41                   MultiAssayExperiment_1.29.0</span></span>
+<span><span class="co">##  [23] zlibbioc_1.49.0             cachem_1.0.8               </span></span>
+<span><span class="co">##  [25] jsonlite_1.8.8              progress_1.2.3             </span></span>
+<span><span class="co">##  [27] highr_0.10                  DelayedArray_0.29.0        </span></span>
+<span><span class="co">##  [29] prettyunits_1.2.0           parallel_4.4.0             </span></span>
+<span><span class="co">##  [31] cluster_2.1.6               R6_2.5.1                   </span></span>
+<span><span class="co">##  [33] bslib_0.6.1                 stringi_1.8.3              </span></span>
+<span><span class="co">##  [35] limma_3.59.1                jquerylib_0.1.4            </span></span>
+<span><span class="co">##  [37] bookdown_0.37               iterators_1.0.14           </span></span>
+<span><span class="co">##  [39] knitr_1.45                  splines_4.4.0              </span></span>
+<span><span class="co">##  [41] Matrix_1.6-5                igraph_1.6.0               </span></span>
+<span><span class="co">##  [43] tidyselect_1.2.0            abind_1.4-5                </span></span>
+<span><span class="co">##  [45] yaml_2.3.8                  doParallel_1.0.17          </span></span>
+<span><span class="co">##  [47] codetools_0.2-19            affy_1.81.0                </span></span>
+<span><span class="co">##  [49] lattice_0.22-5              tibble_3.2.1               </span></span>
+<span><span class="co">##  [51] plyr_1.8.9                  evaluate_0.23              </span></span>
+<span><span class="co">##  [53] survival_3.5-7              desc_1.4.3                 </span></span>
+<span><span class="co">##  [55] Spectra_1.13.3              pillar_1.9.0               </span></span>
+<span><span class="co">##  [57] affyio_1.73.0               BiocManager_1.30.22        </span></span>
+<span><span class="co">##  [59] foreach_1.5.2               MALDIquant_1.22.2          </span></span>
+<span><span class="co">##  [61] ncdf4_1.22                  generics_0.1.3             </span></span>
+<span><span class="co">##  [63] RCurl_1.98-1.14             hms_1.1.3                  </span></span>
 <span><span class="co">##  [65] ggplot2_3.4.4               munsell_0.5.0              </span></span>
-<span><span class="co">##  [67] scales_1.3.0                glue_1.6.2                 </span></span>
-<span><span class="co">##  [69] lazyeval_0.2.2              MsFeatures_1.11.0          </span></span>
-<span><span class="co">##  [71] tools_4.4.0                 mzID_1.41.0                </span></span>
-<span><span class="co">##  [73] robustbase_0.99-1           QFeatures_1.13.1           </span></span>
+<span><span class="co">##  [67] scales_1.3.0                glue_1.7.0                 </span></span>
+<span><span class="co">##  [69] MsFeatures_1.11.0           lazyeval_0.2.2             </span></span>
+<span><span class="co">##  [71] tools_4.4.0                 robustbase_0.99-1          </span></span>
+<span><span class="co">##  [73] mzID_1.41.0                 QFeatures_1.13.2           </span></span>
 <span><span class="co">##  [75] vsn_3.71.0                  RANN_2.6.1                 </span></span>
-<span><span class="co">##  [77] fs_1.6.3                    XML_3.99-0.16              </span></span>
+<span><span class="co">##  [77] fs_1.6.3                    XML_3.99-0.16.1            </span></span>
 <span><span class="co">##  [79] grid_4.4.0                  impute_1.77.0              </span></span>
-<span><span class="co">##  [81] MsCoreUtils_1.15.1          colorspace_2.1-0           </span></span>
-<span><span class="co">##  [83] GenomeInfoDbData_1.2.11     cli_3.6.1                  </span></span>
-<span><span class="co">##  [85] textshaping_0.3.7           fansi_1.0.5                </span></span>
-<span><span class="co">##  [87] S4Arrays_1.3.1              dplyr_1.1.4                </span></span>
-<span><span class="co">##  [89] AnnotationFilter_1.27.0     pcaMethods_1.95.0          </span></span>
-<span><span class="co">##  [91] gtable_0.3.4                DEoptimR_1.1-3             </span></span>
-<span><span class="co">##  [93] sass_0.4.8                  digest_0.6.33              </span></span>
-<span><span class="co">##  [95] SparseArray_1.3.1           farver_2.1.1               </span></span>
-<span><span class="co">##  [97] multtest_2.59.0             memoise_2.0.1              </span></span>
-<span><span class="co">##  [99] htmltools_0.5.7             pkgdown_2.0.7.9000         </span></span>
-<span><span class="co">## [101] lifecycle_1.0.4             statmod_1.5.0              </span></span>
-<span><span class="co">## [103] MASS_7.3-60.1</span></span></code></pre>
+<span><span class="co">##  [81] MsCoreUtils_1.15.3          colorspace_2.1-0           </span></span>
+<span><span class="co">##  [83] GenomeInfoDbData_1.2.11     PSMatch_1.7.1              </span></span>
+<span><span class="co">##  [85] cli_3.6.2                   textshaping_0.3.7          </span></span>
+<span><span class="co">##  [87] fansi_1.0.6                 S4Arrays_1.3.2             </span></span>
+<span><span class="co">##  [89] dplyr_1.1.4                 AnnotationFilter_1.27.0    </span></span>
+<span><span class="co">##  [91] DEoptimR_1.1-3              pcaMethods_1.95.0          </span></span>
+<span><span class="co">##  [93] gtable_0.3.4                sass_0.4.8                 </span></span>
+<span><span class="co">##  [95] digest_0.6.34               SparseArray_1.3.3          </span></span>
+<span><span class="co">##  [97] farver_2.1.1                multtest_2.59.0            </span></span>
+<span><span class="co">##  [99] memoise_2.0.1               htmltools_0.5.7            </span></span>
+<span><span class="co">## [101] pkgdown_2.0.7.9000          lifecycle_1.0.4            </span></span>
+<span><span class="co">## [103] statmod_1.5.0               MASS_7.3-60.2</span></span></code></pre>
 </div>
 <div class="section level2">
 <h2 class="unnumbered" id="references">References<a class="anchor" aria-label="anchor" href="#references"></a>
diff --git a/articles/xcms_files/figure-html/final-pca-1.png b/articles/xcms_files/figure-html/final-pca-1.png
index 2f75bd214cb48d3eb03e62062a735a30212e7537..f6cd6f24e4f847d9c1ff8fc3b068ccea6071df5e 100644
GIT binary patch
literal 102372
zcmeFZcUV(-w>G>Hm9c@bAW}7`fDI`EkrEvPC?H5x>53HT5a}g2juQHyNRa@F0s;ck
zdvRz25~PM6si7l;PC~xlj!rq}Jm-DRbDsB)?~gavHD-vCz4zK{-Rr*ZwUU6F>dO0e
zbMJ;AXy5f~7%d3e3I2B{^b`E_VeYXF{L?*EWej)*kM*miIPkQ~@!B0M1o8R9|FMLP
zEto^lG3YwxijF&BW&lSR(Df18S}m4fRr~qa<$ZMm$8;ty6R!WFd8Tw@$J^6RsmO-7
zT}T5%o%^Lmd42QtMs|-o=T;2P40O5fnJp|j`|R_RUv-0ewE}y)x}>6IR$pEh2^7}b
z{=)Rwg@(A6^{w!sGwO3EsAs4#LkbCD%KQlps06G+)+_I#^|E;i;V0&){=Yx|Uj+VN
z1pa>+0aFDPbGBEra-zsBAI#a)vdn_LkR^WrK~k)FZq9b+-K8e>%YEysh=q#Dy=6;n
zsv|cXJln6#R;~qAa)rq)e#&Dl|Mo;QILv!gkS(9=J`%D}+~Q;3OD$;Y>*JZnF*Z4@
zD;FEjn6<>qx4)!Lya@}9wJ(``wG!qtW0q}JAG(%7-pG)deEU54;%d8+vhv7d0h;Md
zZ@!HZ6^>Zdq1ixuhPnUkcZZ;88x0(lp^aErPu=TX%`^DzBVJ}D#x6`@W1>*NV>Eo)
z;=n{7dCOI9ee}7&#N3?1P|U1@FEzH34xf?|Hq31*H6<Q_xE11no+}zr_+Z%*FO)1n
zX-iTq6&$V}=`&exyyZI=cqY@mcDOc3gFq~6>1b_5kc_?T3~}^j#z@xIR2F~qq&IwK
zj$BcSH9dDvh9Aw|z8^bVf-gFS9F$$5zU}wjoK;<&?OU}9%5+;<U1cn&UERmd?t2WW
zAXDMflVff~iE}f?CR5QPmyTd+;zs(Em6a2)lSxs&xJ%eU=caQ)ZVw;INj9`61me-c
zfx`#+`1r`TD}Cu<NThxFa#v<(Xy{n81vMW<F;NwXyYz9bTqNm;ZhGfrr85hBYW{9m
zOxZ=`pwx;)``3LC=da=6;qBU&ii}@iw|=+%oS?PQM4(>T;it^+G3tR>$ml7m3vcV`
z>50KA%y^(EJe;D&wD_R&B~}hWb3TZL%NW`N;(Fc3j~{VSTGqO^??hlLP9lBnEmGm5
z`3YudxkAywDoM_s=_|ECN6<1nRO#4ehdk>RRHdWTctV=e8z+x)<4Uu>0y}@7+SE61
zMm{(!juOaX7cP9u+{M4e_%%X{pNnhld7`kf#7B`Mwu5EsWB##hrl+QsH&9ovUUi%8
z&K?QQ$%P{BQiyR))5{xi;MF6MaXVh^&DYO((`wbh9a4AE+>+(?r$f+f;dpYQ-9vGX
zJR!5mwxrS&$AMx;bIki$kP@4YLLhqlbJGkG``*2@1|3%0?hiv*BUM^A*shD%5vIH$
z?QpDdl-8xn++3_(&mE}o=AqhRgwvHIt>X7HXCd?3+4pK6peV~IN_rD<+lSvopby6m
zDJ~E=n^&8=Xf~uwz5v#yue4tY)LrM)(ZkL}5oBVIE&oa%JuWV8lcSZMVcrz!V^xBR
zJZG@7iD<RPJ<3m@jS5Anxf%OizcvyI*(C`FE`1ifTzVaC%+~F*-*hL)eQXNBxiO;t
zZG^tOt*`g^1K;1{7xukS87sO|QTKL~;4gUZ1ac1DU)*eg9o;`rDfRqDaDAQcTV3*C
zxtEF5fCtN|yE4c`;XosEHWuNlDwBhF;d?W_KYg)3{i5+IB3rhfjl}}P;)il;zDU9(
zX~j70Kqoz1i*rb^c<6nVSR8><_Vn_~;y!v5<uTUy!hc}oYk*LE^atpRtrYrV`&n>p
zg@xAh6B8deAjg`e<>dkJ7a9}O)0?(%48QdCm1gwV#&`7f$<7v{SU7Xy<KyLxjEqwD
za&VX!x^ntGQ3D5GrBTr6!G%z_GV8!<_KgCZ=An9`*4q@9Fo$a*@)w&((Gd|^*%r+&
z>rY|lL}|Z^s_8)P!x;`f({Ci27vBibCrtG7DTEfnWIyDe<GDN)uwuz#wI}^n)}YV!
ziluEIOqZ6HW>WRi*?jJB_%&!luMUA)9IQ1cw3mO7LrID7{gODl)YFJv9dL6~s=3m%
zcZ|SMGUqL^67{=7{si~SkoByh;lLDRVy8t@OKiyc_X4=e;-~dC;?l}InUV5b9!=@D
z79qT`FI%WHvk+`~6Q<zloxT#$wrXLvC3r4rX^gYkzn9lNP5&@-$->DOE?Cn@5$I-e
z0y*zNd+~Erws0UQ9+NWng*_I6*K&KjhqjY^O#f3M1)akB5{q?NOr<Jc<8Fo$sMWbL
z=qinOq-o0DeftLT0kldq61Uxa8Yk;^K8rN%#=dW#(QOcD@eb?!P^CATHj><=h`1f5
z$3YZyUZ|LBXXwdJIZ|96KFm7{VQO?%4@7@z$+tEHr*T($NdD){V0`|-z<{O3_-I61
zl)hx;+QxL&{L2Tw0X%=;lqU?mjo9)?#<HrNpp>0YdgyCuN$G_v_O);zR&e+i01tQI
zO!#oGHn829IE^LViu!SM+m>cjCBsn&Aeev>CVihBS&Zj+V4~H<7M~+V4>i-1972l-
z*g(9MRP2LiSEn+#qt4zDA8{b^V?T?D#=M%2k0I`J{OLhI;#I~W#qfBK?N-tvClf$E
zKiURSKe83r_Bc-ssW+m?-HEfE1_icv6rKl>UHk0hMUwP5iiOIRKiW2-IFxIyVfxWA
z)b!4&5yV36!YnARv^$>`GxcUgl5ot7y1<LjJXD*R^B9Uap7hWGb4bh;4GsqW;cI+1
z-`rzJA`z-81Bov!PtRBl&^i6W;1rkFtXii0rAwYh=_>$v`Ygof)s|LPj4z6dJ1;iA
zI3rZPT~lS!%^<IQGt$5w<xzqvupjW$%5!f*QfMum*2}LaiF+Jne0+SQC1aZpZftBQ
z7{GTCyobGwRIJeQ%8Kth7?&m{CwtmEJ6#&Tefw7X0|}$C8gdJm^zmDu|76N`6fIAa
za3_%!?TOQT`TDcw?G0M#or7Em2QG5F7Po$(z0m`k{*;Koi#A<yQZK~d#+sm5vxY0;
z1Ld92UF0+p!7yq;rO-E58Cm!VtMGYJ$|Zp+oebl0Pq&_<P-7k_xdD+RLsxdcoFA3;
z#+(Adpd5`A*zB&?FGz{7f8S|cuP$jnBZ_8Z>zWYPMN>)+eX%PX=z5N=@cCuV<Yl8F
z^<(MaawZ)`VR(6a5!drA<2}X&1m^YyE$n1tLmxmm*tRay0get_8X&+0RMmeIG9~(q
zoiPc3n_2~@Q4p4RE^5T17er&W)w8-U2>>H0`9`cUDrWEF$B%75&-F(p+IX9&Oj1FV
z_dJiWtUv+c;DE=^7fg6=W)u#3QiK#10uHiED}&M~&`Y<wE>iCHi?$TOZP26n=#Cxm
zfkS0vQzb~5L3kTu+VByR%^?n4(U`b6ilWQP%4!m4^~*Vf0z$8?R3-np$mFBc)I?!l
zDj*>nQ-^OEE_AeqBy26E9FgBz8_9a7bDE@7V{?kDv@kt=I-;;_>>N6JG{I5@T(Kw~
zFI!(278<HRX;T%YZ?#uCNODI}2%E}EN?loZaxLa{GBwjXjU%yxQdd^3i0f}aPL9;`
zd(DtYnMSe)5DPIa-s2PX0yvr?iqa%Fj8?4C%-BgGIB;=sSrv<1sIpB1z%fd|y%7#{
z%hJ!E8qA?WpYvht5a8o07opHJH(x`u@AT4L2rgZBVv)nDqCS(V9KBR_t3A@nH3ovi
zc4laaeS-i|0d|tj0AJiy;*im4eNetXO!a*6=p(Cg$>w(73awNcJy9_zo|l)``yq9_
z*+hVsm(~iJR00TqTYU`&BZxpv59_FAh?IyUyIkGe668A{fjbl*ZJPQPKXFZju0mY5
z@miVjuy%BGB!B(-^;5jOS9~LCK9H3Wo9DeIfJjK!IekmxB1hH3Q35F`f;c}jW5_I;
zLyAuJ7{i;XI_J;G?zwza@BM8jeWHgCv4?z>#o{uHU)x!Of|)|W;g1WT=A$q*r5qOv
zDk~W+BbO1cWImfmCP%1;;ZORb96;7uvhCbi;T%mJuW6rN27Kp1c~U!>%<vLPTGbft
z$c$cl#E_uU=0TDg-KU;3WlwTa_AAgjr06-0cs1}g!kd44^p4$7#CRE%Xv3-4N=_4{
z&`Lp7qWrnZ;C*5U)T1a0F8*!=j*E{^-rUsmtAM3U%ln2ukZJ&x);Z;mCE5rM2OPzC
zoKBLwaN&Yuu?yo7O6ou26z^J2lOvPKr0Vbo;s=Hle!3tC4sBiibe~BeUOy-kw2u&f
z9yFk1%SNTH%cF#Jv0DEzk)+oI>NLQ^7GuxJBywzHw_rwL;c6OSlh1Eu8hU);IiL+w
zxSohMthP!eKVmca!b1&}$y*>Rebe+a*oh}ts~-Nr1wPDO{E`VcMB&BQ=I7uUt;2*d
z(2pmlrY@PgyRQv{c5(K;@tb>@Mn@1NOpRB)Z~R3RCC_ELtNms3IJnS@B1xSnN;~qN
zCa?P|9~{xc08}qv$8ILh4JRc$iz>eK`>YPcamX)yIRki&%}ABPPx6RtIV8YRQ<X^@
zfa?Tm9H@=-$W6F)zrskdJl!!-g8zhAhzGC^Kd4>JPG3>O=0N;>I?J-1`C7}*apr7(
zPZIUg`7Qz|cUDQ2rO-wf>IE3zfKpKot;q-Ua|}h<0-;&9>3ZMebt04HLPF>CGS-<`
z33^89tqgq^haL}Tyq6d!8wDzPFao>vj(*e6RVTgk67kaX`x}Ubq^OFeq)G!<e!nLw
zEYPW&>7AjCs7eJ2Ejgks9ZM9=1RVvBkI<Uw$xZ1l^PrNnRd8|GEuAE-b=!RXxJIQ?
ze`j+{SxIT><_vx<_Rh<LrXN3{vVcY)_w*D?ArC=*1=@eGSq~ZJ&$)0v4oti)fJjIF
z%mSIdJOr5%cpS<Wq5+j>_!FpJg2TlkeBEbc6-V<N$a9kVGeMz^nzfqBd;%inCy>U^
zv^)Vq4*_)$mK@=mrJtdIdcGG$IfQ7nm9p#DmzOR%_tO`>z{G1JfY$!*v)@W}a<Vdb
zQ;wN~-(W1xYCw^ru?q6YMC|+^?#^N}TgAiQNv^)L#aEZv+7E_FCnyy53=FJ4)=fA5
zHW8SsC%2NQZPIC&H?fXbAOj)+x?4p#sMER0@lPm@ln7&f#A*voW5<VM`;;+<6tmN3
zjzQ+vL<V>eiopXMEWJ09v^=aHK78m_#Ifx&{`Bd;skbsQ8JV?s{Mt2)PD#A?qHGJj
zPqb5=-fz-@6?@4cEv0*@t<ZPwm&1(3td>cK^TxDa@SjAI7-hcnjj*>7?6SUP1ngM6
z*W8^Pn58B!D6_chbyxUowsof8nm;egcKJ?$1G(a1XPR!<4GQ%{Rqn@3qtZyg*k&{I
z(%*o#t>W~Tc6$X?8OBEK@N15hHi^~R$j9SH=1S*7B^aZ1eX^?|YU`{@mo_F;=z+M+
z<a(maNTSTQd6EMxIY~XlJ#gl>IPZpXUjcn4Y>=VnWvMbrLQ;lDE9s>?o{Vo{_OZlj
z1|Babu2r@aW?w{KH`!<tP$a$q;0=pCbW=D}PO_`A_!Sg*P{&`ODC97_%(}q*`dUz-
z?RsfX^r~W^9o>J-WXm7Tw>j$qvx;0XQTWhR_eG#D1yc6?w^Ko9Q?Ahrs-(}hRJv2g
zE90BjHL?aal1_4XZ)V~r8c`-Dpa@RM7PML!8fTH_y9@045_>*x|8TPdRM69>Pv?4p
zRd5tUN4Od;AJX))tpC%e=g-0*Zk?~M{HFEYMeYzAz;|Ii_NBnn({qr3^))wVY}NaY
zk4Sfh)?6gg$c*VON%2K1Oxamv*NwXqFA&)UaH$A}?=h`t`m;_7U^WG0cR#-9l~$^u
z9t>0OZ)XqZ5vUskYUdAXWn}<>UI|0^mY+qxvkFMQ$BE3=<>~=J^m2Y&LMZ+dim`aV
zevF`v!Q}SjX-a>|Y%3DHY#<Vdl&1pKrKO?(a&OxXsJ9L1K+siix*~z$M|>7rEL<iN
z7SuOKQy5Z8%D@4LB3b7r#C(n>P#rPBm>RmWI8T+_e?s_w$V4w2aP#r`_+V<p!n_&j
z^>PJXqt}enZCb)s&TVb27Aj*#2`1<mvGVfrTOq;0gfi!`Mz*l~Qhw`cSy}T>Epk16
zHZdVv0*!m~>EPg?9WWf;f@h8wu;`YTm-i|uDN#Sl&tFjl;<e|FWe>}%y^5M;r;`q;
zZjH%e7n|D@$)1j+gF*?kG`zbmhBi1v)DSt0Sm;J@?7R^6Cuba0_$jXxR)gl2(Cvvb
zmdgezFs=+o&wEo$#)myseZAu(z%!gv&~7L%9mGUtJY0Ev)L<nD_w8{#66s77YOY;o
z{)0o1l`4voiN}Xph+R5Dn&vwCKne3UBJ}0w^$?|zMbC2IdXQY4d_DIes&=_C+S1Bu
zB9l+n)g8OrN>$cvj+uYF+}tG};4FC8Y0Vd+^rp*7GE+o$Pu0)K5PWvC0r+m*Om_X}
z+U595E8TP10z##}Lw{z|p<SlN|8h3nA_wBTH_IsJVuONaJ+`Rv5-TID&rxU!0f~Is
zY;oz{YiPXi4<v1o+m;s_@vP-BlnY3Ut8N&+n@Sr?X|An}ml=D4=0k*IGd0F(e`U;(
z|6#_ImdGOS%`c9fp$y;iN?0wc0NtjIB82sIaG2K>6ciN7GAv&FD9F9nG=6G&y7V<r
z^j}D%)gGqKD=c&~$nB9!1r*^bT&h<!G+N>_Gc&cH`umHXhr`!(2LKW99_9#Cj@Y0h
zm0wG<JUfFT83SY=>NTV3p{PpQlTA)hg^NhvJh%BlS)rILJib0u=ak<Q)8(no(k3Ew
z^VI5<R|CsSOKPqH^3W;ZZuM&A>kk}3NGoBySNln{SV51Bc>->*or<HuPTtftV<r-Q
zeE@PNBT-m^8=;}FPN0;{d?c0I*M=9>6}HG`S<W{!86yyaB1vS_0TMHltP}z2<=?G-
ziVMJVC}QN?w~BAg`3&s|i)~&{xLrZ(o-dpG#Vwe=QVDdc!wpy4H?2iw##!WSyS@U&
z+?ZXaum5*iJOe0(8AbrU@~HgOyCjp+#aWV^$M3raa>eGi4ywypP;}?LTP9kS6QZ>?
z7Uu>^ao9j@!=nhB($dnF_z*d0&kdr7k;){ENF?SnZ-eF`ic-xEKzeeR23l?4;gm4~
zi9*O?-V->bW#`MNFGZ4;5l)PXld9@Gp7R$OW7$V4*J@%uufJ*6K;PZ;@|{G_8{~{G
z5vZ(h-n?<}Ae9d0<4$8ozXP0_M8mcZe*c)_=9OFLZ`@Rw%tttxU>An9D^`eS(C*}u
zQDvufFC{M>7X&?(BwD=oLN?2jlD`=yzXCLh1E$8NK9c+8vp<qpn{Uszvts{|*#0Bf
z3v4tjJ&()`z^{hV<sABfn~Lu2>G9AeuK$k5X9C7He%*vVMNwZ5DlKrEU2m-jsGBt?
zE1LdDV)ReVXBn<<&{ZP|{D^R^jUW)h!t$|m4j%58?l==)c+C}CHoUxO!yh}Cu+wy?
z?;p4IqzGrG&jFmklfb0(0{q#k=Z>FNKQF=|1Ya0J|12pCT_BCyu^dC*8>tI<kqg-0
zB7uqv!#@{9o>mlV%lBF>S7v<kCr=ILQrNN`7!@*U)9+Fu&!QfX1B5ERrH({xsm;CP
zm6xg&L{U~}d-JC?q(PsJSQTRWQ9Ew1m~uXh<b%eaXZx=RWOZ6UMl9?>QO*RajExBj
z#dpg(4+oT^U%F4oBb-Kn7_%>%NDNvbY8t|c@3%_?y1WMnt_J*C0%8FzZP&Y0rjoRn
zB{=+<K&5J_1UVDEv{lGAMD{7ykO6z0kBH@BR!pOd5oQ$cuF-R4Joj&wZxGXl0?14J
z^5u(>tm}--8yH)=i?I|Kk162QC(K06(3`fB<jx1bdIJQkt)G`IUeH+2^|vSQ0fBjy
zbVl%=otxWg57WT>Z*siW#Qjl0-hZ)}{5*ue18H(d$s^Ynb$l$nDw>BTW0d_|pKC38
z>^J2UhN|Quu>7I;aC}iEIK7icL-E3a_ipuEff|3)Dv?G8a+L}I)CGaKiJ4(vy#n37
zIm2>c7mCsd`bnrLV0JMuv&);=3z_%AnUO-?yTeMk)OA)Lm@JnXi7XLy{LlwTgc&Nx
z{|$6u&tLKap8A5vx3h170J-^hBw<G8Z;gte3G<Pe{$ea_sbfL=`kjfsK$%N&hv<V}
zO@{k%%r@w=#x13q-nqHC+$MI=!wJ-qh_-cL<k&D<dZI8f1i$C>i@=Kx|Hp_~oCYN8
z`ak9_(f-cFYJ}JU#451zUY)2$AQUxcz!2&3Q7sj~jSyz5SN&YI{}i3jo<<i$xZtq7
zIa^E%s3!oFd?vsu61O8)_&i|xM?RY?Ng9<$cYg_W1!?^jTvS9Ht8g={$P!N??y_ip
zj}yBx$ne|SzK@sVpB(!B>eoBhSfKIO|90v@oZ>s6=!8EE&nUj_uYjO&5lN!T%F2?M
zLkC4o0uSPd&gsK8gWK#L{1-?C_mGHE!~nw%7P-&f{sZX`1SP~CiRD)pDVIaeiNZmC
zDu@hB?ACSN?<f8?VlD)atef+PW|Os2<s>g%Dgh3COfpacjE5?duNaJ_)l~sqC4S)`
zX~>j+YIe5V`u_dV(HCcKDKJgmrRC#IXxBe>0seb8e)Bamds0#?HE!J^iBwltABl>4
z@+37kH}?R-=$!8rgN2>Y=c^(~NF@ImILY<0jLS>mBzG3r-fIN6_#BDLrGO~aA5nBt
z$I#M2g#a{DMGlY3e50-TizKl*6Ib6xRCKqtF4V$KQ0dqyi<}>~toO@!bF4|c)<!dd
zx<kXEJRHl9>&0S;Eg|?@<Af}GsaRlZ6#%(h;Rei#goKe#YT?_Mcl$C=YNxf(I0217
zI7AcI2-GRyFuP^wZBAR{u1BhhmM?AQ@`m5J#{otA!{U6|lO75Z?i`%XNB`Ba(j&`Q
z{R~$<;(DxDvp=IH0_*FdtX$6;EA61*HQpd|0BC04A8u8t$2L2Wf58EdzhRx<3(QJD
zyc$lZlO<NB9XXRCwzBooGhW`g6uKQ`BB(Ov3@I7DRVPAZx8=P^6mF9UgJw_3mva@j
zJicH9MykoRC`x4*K697)`W51O)FoM2lL)P;nfZBiCXmKO1Ep>Px`%2XAQlpk!)ynG
zGe!vv;AzIWV}6&gTY+GNlp|<)dSLoZ;FR6_&xG^u?@=2QZ*XX&`pZ|Fg|731^ha*x
z(?*;1Zo&CJ5=h?^y>z;-=Ap_+t&Oa6<F^5}2ICWAGd>F?csqN~dy!)^CVQX#U3yEL
z8zr1s?Z>t}gY9Qnb>-LsT$r5dOm~ZlV|{hDuh7Ay#JV%BUQ4Bz*+BAbNY7rseqG*a
z?UFZh_N-jtfNP&B7$MK|kq3c=7TI#uCcjq%MPbul9wU$yfIBv#Gu{V$lgla(+P727
z@4=p0%5w?i@OZ3)LM9kX&TImGjZ1K<J1ROcKaVrO^veyb_r}2`rUInP&=C4dmmH0)
zH*WLTmpLFZPvx%2z&*g^<*RA8l#Op_BzR9#>esKP8W`HfT17-#9o~K9{y0cgW_hTj
z3Lyc_%nHR*lraQw80`?nJuBbQ%RQ32&t@@=)5vReZeTOLJx)4li3hPj=QOebmy!Ma
zv#YVnqy%E270-pw_Gf(NcliE$pR<2h%qBbt#x0cxy2N!wgp(R(=DNs$$)Va?$l)L`
zXEP}peWuD!m~)YG7^Hz=OpZJ3t8-++)2w7!hkSj$0T7t1l!&cYif#@V+Q37K#R}XM
zJFFznx1d(Tf`i|i<XSX;_k}^*Rwu4Y2PV3&rs<}$>lYsU+G7t-W`V2j(1^n20R@?B
zPGO0*0n^!$_qoL0LkJVkNmY}LQgh1da=V_Ku_YidH88wO`{k3Cf&kli+CY4PRR?Ym
zS^!ZwU@K*^7`$Q`@2W*ye;d*E67O!A+Y<~-2~XlWdUbUbXP>so3Px4$xukIz46w!e
zlEpR7fac~F8d}hwe3ivC{wCzNKG}$OcY!$rVnN#6ADqr#Fs}6LHt36MV4{pP@;q>!
zIh2_{GaX^gL(_uybTPuv&6hV~rOMy+*s?25`!g7KPjfKfi~^H^X-h4YGs%b~aF1dP
z9LilbfruY;Z#Ws{?UutDlqe&g6zL_vXqR-R{VMOZ;_cO%sF;!I(RjZOK$$P#&m*e_
z%k*kAt$Nybty*B~^IpLMdAQQ+x-5b$a(g0W+gULfZny1LXB0KxMgZ3sc(!>-OCu>S
z<{n2_*apN0@Nb!bi5*(1rx?r!TAEPgqX<5Uh6t=Nh@Ajz4%+P8xpT|+S+#G8qCqyW
zrv#c$sA6h@jF%F}t2xl#H@#F+ZNx_aPPV<qM_WA}jurfw#q_D|w!NTt0dG(VIlKkq
z&HaBeH5O@rhbl5F-*63ZzKh;U?_`KEPs-#V0(aOSw}$3MgoW~PB%MiGDvXt~)0c>u
zPKk&oC9Xe8J%6&JQ7}x-`SW&?_rqkYge(Bk|Ip0Ty&fO&d-DE=A@Pe#)+5)>zedUb
zxOuH;r+3Dir0P6H4i|9_RBC}JT>|bWOk?qDO;-tvwGnMLtMg=>Hx7rh#NikLbUHn5
zb9ElAkhBCiQq(D~Tfnk1=`L}ZzV6@j8+I!fzxG7}H%7o=b8t_uK9Gt{(Qs_eYObSw
zCY<#94h;*pdr&{-U%<x@e*Oa>RwCjat>l6k0NB^a<wvL*JY+;|ZEc0Axn>fuCCgm~
z=oqYhJ}*&QBq<?FFWs$x+pEVsiY|<t3&zW`v9W=P5Rifk)U(!7v4xnLX^t0W0JIK1
zZ<f;qws%y4wP;|X@7v6V4tG{28d!M&2d>zRgP~#7O0NJ%BEKB=J~B=)IM<$A2o@S@
zF&vEg>G^qv8mLqg%!~jz3hO%ndN9uQbg@NKvozV;5JOAhG$IdtxLYkvkzqf0^E=G+
z_I70$M2!(Dxe;vVQlD#ig5WW5v5np+3UIRNdOjja`B8?>4bsWtWDq!4#hO@yW!=*+
z$Z?H^xk+H6#R%!z<o0&V%)KB6XZpY0lZSdhz~n6%TXd5Uy!ptcmP(~DxYcr<D4<w6
zVzsRE6}FOr@_aHm3Ys88O=VKFu8=}7DSptNd_>)nILQqrP$ro7Hm&)V4bM(W6U{l<
zZvg(7@;}iT|42)>*fZA%W?~sFcC9{kz>x1M>=DxE0gEKmTPYG`;Iglc?aEwDj`=Km
z2St$tM)MFp`kio&FQB@D*zhL}n#?x*-SPRywA;C+(>nhkNZv|8WMtvQ)=h8|>W-@6
zYS*4dv=uPtY)3yI1*69h{F+`+CQkObhmhHHSB4PywHD=?P5oss^8>AoK^2U9ocjHE
za=ZvZgcA;kRu%DoqOW<67m|H9+X&PZy!(})IW!#DDz))Y&Zb>s1c&H3!U*kpP%())
zgvpg;XcTR2OqT+F;|0!`QI2<S-kvg00fe!eK)|$w#BS&U2N+-fISLnG`nh1jZ4V~>
zLSSy>!?a>2W@iVwz%mISx%8W`wEoiGo|^*CNQZ`of|`iG=W2lYtndB%_d!n;|0r@M
z!8q>jB_$;#{S3MQit-?Xjs~4m)F2<d_(>36+DM4Rssrlg+YT1LDr2!*aaQF+4x<Ez
zJ76`y1t{`bTjPWtCrsMURiEER1mfM_&+tO#`sJR>b2$=FL<1OMxWWEsBe5#CWwsEw
z`G9%HN3^B-%gmY#cur?HgyQ>Bup#h56mk6(yb{&Yctk$7d0hI^C3auV1g!7fJY#;V
zMi7`L@Lz~X(qJ`V;|xHd%I8{_);K<w0b2`xygC;wPjgt`N4IQ5F|y!2iv>whDvi_b
z4cLol9ehAsJs5f3=u&Icc6}o^6(+)g7O62tC{(O@&A&w^wYEB0fDBQVqa~2PGj6|v
zFT+C&05@6rGA<3R$^!K8NkcI7rQ}&vI9$T!=K$Rmj_;PPkf(V6{*VGj^~!c&8|x~t
zk7N-}t;f}UPE)))bIhBFhK2TV@_n4p4!`x+2M~bozs)wO_aBStg*E7(M%e#KzRIP`
zzrZTUB8|~k7`gz^996&ao=D1jt4JNuj-y;)*GHynrSeM01%T+CaVC=0MChce1U*bm
zonjDY9GHVT(mg0~Bmyx%+G{s7CL@LHQm#1%xD=i%1TPRmr_p6TG|{(YKJ28OLu4Ok
zKlGE8!2d)mH=4tjQyqBtj5;uQTTmuOuAK4t@xZUP1+=e7Kf_lDIY^p%n<#9}BW+Ju
zkmwNnphlcn1KK;n*%)jP8UcR8kf6}U@V9;O+Kdl9wp%*j{^aN82P3fcg`X*Uw=gto
z49!z5bIsY1*L|(>IJ-`duY)a3Tlw@cYIVD8GdyXYm*MzO^@rJ}BBH2yXznp=Os~s*
z3eOWwTFSA`FQHC81#O4E0{40sOxf2Mq2vfpq8#<Jn^KLyjk{ZqD`o*Z%x+ij(_e+b
z<88uG`+<_>w?IAJLQyKN5?sM-^=0<9U3(&=p>bk1!bt;T$b(*Qp%3BLK##bN@s>n_
zy$5tCUUpc>Z8j5Aa}lwi;ba<fEWY-?+5)$=4_K*+BzcWgMfp8t8W6CP!h8GpN1xw%
z3Ph9gHexs!f=D8;A}urz(GLc%muDChs&9B0s!Xz->t!!p_;!B)p6#~Y%l{25FOuxQ
z)tyV3BUdsKggn<0w7B76#{&>dPY%Wrg&lsy;mBc$g^{>if8_8!r&q#`&KN^3#5^8v
z5{h3{s*x2OCSz(+VWLd4<rV%3gCQJ1m;~cPQH|=m{=VtP!PNWz^7_qCkNLxgOEU``
zjD!N<A&V+vG8Au-w&<IThdHEAs6x+pY)6Rnp!?_h6qyKR0iUHcc+F^@&)ycy>3nm0
z-YbJesK?4+xH$`~p5>${<6djC`<%yAq2>@U8c(nH-O7;p5ZrQ~MCT5`i`GB-r3!4I
zUPKbPH|au_(x<43Dyeu@ugk#m{yuqRy&Tie_Q;jg-j4lpD&>N<zt;z!4o&598=$cO
zV7;kxq0<19Bf@>3!r>N@C|seL%1bMs2JlsR3^^D3bAKcl!tW6(nq&3nUY*pydBR&8
zmAn~b{xu4LY9npi{cSd8+{|}jXI1~2n^}EUZ3q8Wn41Cb?bE`;RR&DiUUjy3Z8*qH
zxu%p!4sPV@+j-ypdJJB0jea$)E77pB7fniYS{SbNH@tptez06xsfKFskiPCz&qtab
zV&p#GR@LuL?_LNq+W?_mS;!tiNf6jzu;j7?D+D75O^4RyegbuqL%KpgW9tHv->!8z
z^r%u09-k<z-;5<I9x=*uOB4e{)YBZJQ(u@8F$e0`32wo;;UVk5ShANZZ(8NByqlR~
z!m2Q#%NH;IZ*!H`bld)->`y3-yyM?~?4yB6_22f<Y=L$)0IL%x&y}zjD}!Cr^*5Z&
zXgod$?08J5iHlW=K%YZq&FZn>v^ve|aW9_~M_2w&45}ARO!xKh89hQI);IxMQ8IHj
zFXhprN1%$!^?blmlr|B?XYCEPTg`vTCJ|4NCMG6aGJ0h%U9yk0QUiWU4F2{-&<$t>
zfJ=?4EnZ;2r*gxrzEHVlQX<;Y2}2`xw-geE5!Bvx1_F>{e_GhCs@&Uus259qFmP9C
zpj=Uj(%KMR80nOWpVS37IL&_1mA<iT5`dr2oA3lHR$e_17r|~KxZ^zfqr6}nZ2+wy
zd7eV+;%qK;o9$huv^RYGJrTMftpGMn>U_Z_i*#KFgSEiIb|xpAY=S=15FPbjf#)Cl
zk*plX18gE7{d~m1mBHkd2Nu~G8+99446R<z2Y4xjDXq)Q+d^Dl*;Rs@yim=s3o3|(
zC)#Q!S@&pziDz%kL4?y?O#flFidu%H(^vF8aQ0EiVIH$efdJOn?E;4p^~yCOh=oA>
z+7rC{M~U%Lab$(8bgRf!c%lZo9a-=1<c?=EO*mNP%=Q(MfW<a2)h7kDc!Jq%A8?g@
z+knaj3q!qHu-^tYbmGJ%B_)TFlP?^BMc#Y8bRPwZ_gqmXuz;qA_w#JBb8wJ-$73wJ
z1QXp1No^Ln#WwB9H?sObZAWoL-35D7v*0!ZWs$i6F1-b4Hc>T6;0E!55AP(^7HI?J
zLG~@O(SVa7<vfsx^8%^sF>(M^skfX=Is7yfYn%|-lE^`c6`PzD#hN5Uws}FdSS3uC
zzD3Eu%J5j@$dmmsshbgzWeqMqi{!U$^&JCJ11$x1hC1~g!-2E~PCz1PIm`G7E|QNS
z$ooy0!5mgxh}Do^?zqL=wvC^oEv2B#C&0c9P7x%k6w+AYN@h7w`~HT3?}P&rwiW{Q
z(UDkRZA{tZO*6|xt<*9kSZ*0-0ce>aim3tnr$^o_-!MaR)IPf95r2?>3sy^21h5E9
z27{s-Uwmvb^mrJK42Bz!@N)avB3TnYwcI-WqS7SKs`WWY)-ZcOf~H}2qPNs-{vcpO
zbJ1%<xNxlRvcceDqyX<}FO|u>tIO0G%09p1)%zIQ$L_3yg2N}BOg$2g)I1t{q=lIY
zVcqcyo8U0;-~5wKFt-QqflmWe(!W)6nDC*C)!g1j7#~2)hvLI-NHW*X80eKbd4;q^
zqQDx(%i`H@zs4rsw7t^$%cBd&Ke~%#a0-{4?BaP8vpu})oPlVc$@EGNaKHwNN}}vb
zxBBf0``TQ4a`GB@_gZIkcj;&7;-<OP2uGjc^#kwKo_S<?Z1gjgPQUWbNxWEyO1Q5u
zPdbzJ+#?<vPdWIddeoaDC&V3M%>GqjiScM=5-IMp-zhYG%0DeFEuO!oV$aTOjvJ<D
zq3&f)KcUh}n(b+I9UkS8(D2nt27SIB^eb*)NyeVv_2lO;7g0A>$jZv<Q$uqz-SOCd
z;cKU!+x7I$%Mi!XJi|w1qBxHn5m+!#SjkPG&j?I!T|Q>|tYI{xE;?34RdsdkoWN#J
zi|<O93l58=&fFI-?&P96Ae}oL*1N&Wxw!Y60={cQ*HhwLdqp9+lr#f+WEg#G9_>2Q
zofBo^dvtOy);p`9Kx$ZqIFdS&Iu(~WtyDeqZJN$onVXqu&*H~6ifJ`1+LGqlt4B;g
zS6Tu8?G}3k4t!#zDO3Qh@XO>=N?yBqsVR6pS9x*OYjd`+avo6U+DEKx;_*1-T^7Hc
zBWTCr9@aPanqOWlnk$*kWOJm0*(@osK1b>KsT&)4X@hrzAK{M~Z0a6{jv*6vd|UNx
zst0GEKsOt32wsoDzu&v_@nLA?udA&Tvq1UjhK>++o2HT%vq5Mq8Mjwk2q>yeU9>7>
zh(snVftYPpnqcNG3p|zeb#Ylf2tEk0W54=Q@mFaEkoWt_k@U1Q<9Dg4A|=56+se(z
zm>mILppW8=BIaIQw5Sp?oDIy`eVR<El(DgoN-U;R9jRpgN%B0YV&$UudXq^V9N}O}
z!nNDbtB-4<(2fcEhi5}Yqfd?)M;8DgM`{6tEkQPvifK_d`^<`TzPsvecc&jG;-@X&
z(l^`gJ!6U9>_I2f(>&EMeg{=Y*3bWOH!H=IN?IOgLHLb7Z*XOrLf}E0%*uVH{@{wq
z40>^K@sacTdGElC`q=)pHZZC*J{qzh7=G;pXAnID-$9rWXk5E^@!~1)WBX)Eg2H5X
zwpmJkVd0Sq;H`-Am<8&%7_W*sP{zky$HvF`&kGB?i|qf&d;FQ|c@J&~<nsQY+MRJY
zA^}ytSbHx!aceQm72tqxg2#9aNBPE7`uq+=$QcdRo6u@E`><v9ZCk7yKYHb}L`#{&
z*YW`TsrcJxS5n>OGQr>E&z{9#q-sVM7Tx9IgKO`Z<i<V0e|p;0)g`jvPBWF`XQYCE
zalcuLP!wA9(RDwPTaFXjc||u4EI2(V!0t!%baZ#iMEjc{5I1gzgoL!~>e^C{A#;-Q
z^IZa`OWtI2vmin=&O-Y`u7W=hKOU`@Z$n%=ymvx>yntwxp_I(shXnLG%^^<wrw3^-
zg^;s#(NVnh0X$~>0T`qhLbK09L<0Ptm-;H{_P07KkW4=w+-%l1Hcgn(J!zi%{AkZ-
z(>Cfe@HD6ClDR#tRu8dYM&t&Fo@2M3edAtEsBSO=s<@<NhL!)a>vZ@vTb<5pVC}zT
zadDBo&jXJaKgS;?s@>TaEGGRTGIB0E&L1z_{xUY!Qs{X@Mw+Mc_3I6+1=Rfr^Ow^l
zkAQUR+~$4{P(Qa`;?+^*<>lpE`Ib`F%bg7c{VOZv8=lLZk+fSTI27Z{Ybl|gF_*C;
zqXKbGQ<KV;%`e@SmfDZR5(rqxTV6<rqMAAF3zBD@4J1e1pvKQ;0XzJ%R|97u$@6yG
z9G!f9x3r+J9`>u4hzDsUdKkT(k3a5)KvXrm#OltneQ0zdkie|=GUiT7QIUJRm<6Ao
z7W~&q@6yvtKU2#DSY>VmqZB;mEBq<6sauto#(FHt6Z0|$KRd0M6j(1jT3ifsxsf1c
zqFY($IQ07&9Bo9v#BDa8#Q*uGcjGZQaBh(8jt~4*g)E~F3k7yMMrGU_bx1o6ef8j<
zUp$s*ds4&znfU%*Faq&N>?!36vQY*3`|4wikE7A*CEbB-$9{AB_H_<hNOFj2>)jYP
z4Rt9yFX2XE&6dDzHLX!eh8`bnfJYFwYx{YQosYx1NF)=zJ9ip5Bs4CZh3?lOpjZ|f
zho5ChpFWFsn?7>!qOKhxoXRqj52x%iPUu2iUN8I4nEQ|MGc2ZGd=yR~8>K==cwylF
z>harzQDW;vXWE7dL`ZEx<8TXZl_vAiqrVAEq@PNqbdc$zsR83h;`@q*L`xQ_N4(-!
z_mEsf+5M~imC&djaXpTdq@{b%Tp`~tG|@x52fBNDP&UAV+qAPhj?8)e_U-*#iU;`6
zmJPB|@Ab`hAE{+WU*DW??Yo1=+X|mPZQat)5YyMz<{Sdb$#<w^t{5s{aZE;+QsTe3
zRd|sc&u^_hY|L$`(4@`I)6Bk3w6a1!LtE6*#zT@~AE4hnH2YXSF)%d|QBk?QDwsPN
z@87>~cHN0Y#x^!K_G0d1DmxA$LM#;(6=#PWWgf_~g$s0abjS=O&_vI=k5xSns$^6r
zgm!g5$trWc->NY+=NODiOS3T~mAmHEo!Yz&eJV35VEwB=*EDr>boPBkrle2bY~}D>
z6XY8bVrbtODBoOekMCW6J{QH9DY4(O3e{MF&rX!(9qTvMzAN}Y9-qiMx|q>3kfW!z
zt@75#$IA8h;;{wfW%s#0hk4cu2D)ekPVCL`Z1B6YP#TMOv+MrFEDSPRj6aM_{_^Ff
zsq7v^_`U^q?>lEXy+AW~@Q4XilG4&Gq4gsib@#G63vvPad@Zp*7JjWL05i-9LTwAL
zw<toKC?g|(VhJT)sLq<d{={-ea`N+*yM2;+)SDLB(B)B$SV(7-LJ`g9`xXCbr>@Z%
z<eA4GO`kB2z-1Q6dV5TUxbyp+JAXbtC2d+~*#drdi8@18=!xhoPTYOzopdC&nLc(k
z<l3n)7MT{8fnqb%5%ZmUAdJGp#Rt>zl!er3Ohm>Bb5%@)Zm?Jz7ChTwew}9vC_#l{
zcR^}>xk3tE*s(kTg?Mt;%30{$)JZdWfe?942=9GO*8qCBt=k3;!pU=xY4C&Zr;6xd
zt$MZTeGR-7Z#@cnwNH*iD^gomv-2JleOQC0a1c9QzEIcT!D)U^f!hGo{`yVfLnYb3
zL+^TaQgx+|2&&^M05FRds=A1Xh==L-c5qX;<EzJD@J-=|rOX%{I(#^vGCKO%?!klN
zI#*6)ETHoWhpX;ie@VO!^5?`sth%)<Z6`S8O>r@Skc?Ne5T+v*&jtB2u|M_C+;rzZ
zk!X8MN2kMHmNne2;eO5+Ctmz@{-zGXqfDY1^i@XvZu0os>}px)kG#R-y@mJx#LEgD
zZm?N%3H@UiprL~hzug(C;(DH)$Qr^2V)W;OHo4z#gnKV1Z>*HO=)r@Y?I{Uu^x5f{
z=&(9Y_-;g8J0Vp!vard+&Ar<euGQlJ--BxJaR8J(_kr2%#>dBdK)3sN`)olkxLhQU
zA!*-<1G}Wdul3)P-4lKTRFnv^g{8{i<m=$~4P>#A2o&H$8GjsND{t3jMHrpo{I|W?
znUBm72ehmkj`rgpgvj>$k%otdJ<>|lbsY-go_vxr`#b_UzI$iY8r?Fs{VlaaI|#b0
zE5I|oZZ!j^=~X7Q#xvh1;H_-{J||F2$<rP@hv2`_k7-KRJI%5ARa+8^ZxX7ax6liV
z{-P+-=(zf=PM?g7jOITR@@?8V=+fVD@>`@f49+=hepgo4f5Qv+$%JB|%RJzAZSzZD
zb#Fc+>^`rWt{|T5#!|NOXpbl>gy_-N);6niJ`Z<LaF7051dND?ko#V|{|p5So>UKw
zKwF2Qt#7Z*r`0Z4s%1{jnH1Cd%~Im@c7DCBu1+khzsKHK_D6aV+#6*Pnq8;QoUzG`
zdxZaVDxNJsxbHAX);{9p(IY@wF<wSRxtWnh8!QWa?~P0;qSuD#Z&1D@krZ{$2V=^a
z^S<lLU{GHs^>XC7d>O0XpGowo9tgX&0=)Nbi2eLoed=b@naJR-?q^Yy8vBioeaXjg
zoPI@kZ=dg2dF&73fm|$hl%ZjD>9Tkk-0f^y%c_x+V$Zeuj^7(TrZ^Oj)xnpt`|B(G
z`#t{E%F4&juW$dqladJ-WH>iO)~s*v@TdjL9dkTeGuEQgq8A9+y_}*id%h)!(p>7_
zx3{o4-o~8g@ULzOwKmNOZGnz0rrdze6I)tLc7GM?y9?;f4P8>;r$*BYw^>ZXX5}!L
z7>nRL?4~b9WU^SkvbF{YPaisRq(!3cJ?qWgU)2FYqWz4~(M8Xe%(3jSx07A%Pj{s^
zn<U^iv^8dBrYqt!WKX5ChO~-V5<8A$7SVF5EfeTMj)YffbF7ZKCk}X=VnuH==Ypf~
zHf<XePYHk#C-PcjBwje3%M61t4Ic9N4WILjnc)H$3`_rb$=pa<NzWN!8OjR|+**4U
z7$d|!5v`<k<ds)C$d5fC&$f^Y=WnW07Vg*Qr9SVaw4mu<Hp%EL`e)Q@a0d!7ioZaw
z(im}Sr_f$7pUI<{wQ>Og^vPc*UlBuO`#^%LUR599OS`q~PoVMq$I>_!qgL>UawSm#
zWUPq;kiSlM?+%^PrF)Bu&cBB*u$B2oltNBU<<-MgK6TNRS(%x>;#Zv-4(Mr$w8!tf
zx$~<rbjyG^Y>+kue@}plhsXMHW`}uTYfCWJ)9bR3WkZCYgBi4{sw#&It}zTRJe^li
zpdL_rkB`~X_k@!>K}+9r<UDP^U)Z*_V}3gz_38i)Keu;NLC9y&H$1-oAiKXw0hil{
z0+u4qDwhHB!B&}XlIl7~=d(ep`|+?srl+Pl*;;ZvvXgP*b9y{`3?u0ddc_}H{5v@x
zDc*+p+r!5c4%dNznmTtmUmREw{oap(a4^rlJ@Yy_Ie7=OAAjlZmy_iWLvsby>M`5R
zW5ALz$e~jQ0t{gE<!-tCJqJ%OQmIkj=~`Om;qVt9O+Wn3#NJ@BCYJZky#9q{2Q~Fh
z|God>>m<H6#~qMd%RxALXTu}TAvE>G>aMkb=K=(xLf{0Fdk5V3eoHHWCbs{xJNq5<
zkVx)Z@pS<y;3I&t_=ME7_Ag&9G`q3`a>dQVlLEj6b6?xcY-H)V+tULG^Ea?K%ADQ_
z_d1(`p&A#=`G9MZp)m9Q<H35D#q^NluD@^snYatT1Nsv-e*7q)a^e8b$if<Z-@bj{
z!?!=IJ_<-W|CcXcax!uMDr_IP%WavS<utHkaw2f#0|CFrUOQo*4VssgoEf+=2~sTY
zK7pG4p1gmG17?C<_^u6jGd)x>cT!<xkd~?484J_a?>FN(Q1{>DG?={+qg8;x+^OXf
zlrO|!dXzA0zQ5tE4uMMp&8T<VtAka>%&<RTfy~s>LE>Pf@SpghodcAs$VA%_86pb=
zQ*Aun%Fu(`Z(C8l>wdJhZp!P|cVEu(f`5EbN=m&DhVJCLy1GfB`MoERjY^<ZX@iE_
zO}v8XvaJr7#v2|8x8eom`A5RycmBSuO8D!6zv7F*9NZ5e<gW+w$A7=oBU;_-wn5=&
zyKOha6A~&rfZj#&=(uEjH#(RJZu$cq4XKg%94#lO1vy=#lr(8L*Cpfs2VEP{D$Fb;
zo%|h_h5X9l_q@}#4XTUQW3q?+;quVlk`{ju{W>QCE7V&>J%&KE7A;CbI~>DikNGC2
zsb9TXE!l?xRKprz2Vho+@Eb60Xbc`z!9;}U8yKV%W%7K8dyIeI^YkDIQG5c~D)alc
zzqIzgUw8OvR-ODSzIgm8An*PR_m3aFg@HH@t{+*ImXtj7JsIWrO*n&U`K_(3^}qM{
z=H}*Hkid`sbP~Xkubntxj|?xW1=XCm5Dc;&*Sq}{5{<^CG85+Als@9l(Y-s*Fca=0
zS1xbW>3XN4@pB0Cq`%XnAF1uHqYG70<=|ZZbl`dqsK3Kz!At$%nFlenc?FrMSq1JG
zN2>K6KXu59DaPLJfP%#`3LXP_a17}MaXn1nfH~8M3>%aOU>uo<xO@18`0NA79GctV
zm_K&5A}NgsKGBv8#Fan}e4t1;J<ZL{?U)Wn8*n(v^YWH{!@pzF<Nt`tA(`MaMOe(+
zTTVg;0!Dj3U9HI9SCtQCPH)9OJ#HlnA(l@ecS53D0L?~famYqjIgSIG1!sUH1eH*y
zbb;&PvxfVjpgej2p8tgq3C>1tFU2g5`f*$e#a5&d*)Hr6%A5sXVX~>?5`~xJ{%Jcu
z(16}2V;65S0Hpr;i;FSfbf}7>0y-|84U~K7EBZ&j$)|1bLr2o4n&T-ST?Xpzi?0_?
zO-(7Ea^~meWwN>dt1d{y^~~w`v^4b2Z>>YTS!w2>BXTn(a}M$?2DxCPOZt=+KR0*1
zD}+?BIp`NiqtTL>cj{G|jpBXt3taD+M0uy#XJ;s#x4!3ln$_0UQgvM7(gtm1_h_Cf
z&MdvAiwnDJY&(BlSCoDGwoPy~PknqE96YQh;P&!=Ks5DCQD8|#C>oM{D$}G`qQ&t8
zeWjH<wT>Y*iZcx^4N|{+`&ZE}rv@w;?a^JvL8D9AO|fYkF{T&XBk|h>!#{OgJ0YH*
zk#Ua1CBad31S;#P&kjvfm@An}9l6M{PauH>;3{wl>cDiwC%deS_8T+50EM=-x3zuP
z1eE=E#d2u(MM)_sqVTzMV<6=lRbZJ9U=wSW&6N<OeCN2R=)3`BvD8XmE+{C-8gQ)M
zXOzk}U{}fZO%!-aO<YuI<Ra-oo$$?0VL#P+Gk$J`K_$hQ6-hsFXmwjruQOmj-ye-1
zO6N*!e8pq>pZ5aSEoV_bX}Scf>ta%(pi6HD*5kRWM*+3TRKZwpa?Sq&UP!6C9&ZO_
z?f@Ros^nh7ZGWN2rvtilH31_x_T_uBkdCfyCri;qqXc|MaG{)}_}};j;?AqIZC~po
zjle*a&xC)j2<gn@C*Af?w<uHAqipnb_=p6oIR2ia&Y5o4Fn5IhZq%)mT#UwBCED2A
z#{-^Mm^Q_C^5jX5ZX)cKftJa6PG4uw*8ngVCO?qnFi!vrI}-9On?Lz+1YJu_i{D{-
zz|R%YV-xJbUG)ZqLRA6{+mG(|0Esn70V>tHG;M2hnPh^qEgE@RJ#^p5Xn-8}sWN4f
zon;L-a-1WC7qlaw1FLG+*Vk29<Sb<OA~cgz|A4yhwp{%WtTSc>kcBXhgVC-o+67W!
z0V(GLKER0Xqm~&#o<@0`;TrzcL(Px*RCP~+dB(1<Jn@v~`PM;f;nB29S22w&-i=f_
z3|Lu7^A$M|Kr_Thy7X^t-}Y!YVu1w;9qGRq;ncQk=VOjj&l?^Z`4~H{8t`+Y&a+@f
zuMQL%0Bh>K_u4)L0{X9T%6-$$ZuU+*r4~|S`>ma-icye`2x4ST69+1MybLjOi12k*
zfrazO5k1A=Qx8V4-RbZAN^E>ap(LK~-c#e=zCM#FGh*?`$?qWnB>z82&wpsqL0@*p
zv!!x?x3jzydhqIo#uoViyMLv8cQ)|scI(pP(m2CTJWq@Ch%W@$H<x&v*H!(cR4&;4
zs1|J423f|rL`G#AVVvtU`MHlX5$$jKwA%P;C}<{UL_~UKJO)J!S6||DbcGLDf0J`i
z0`ABFlXV6L2KHi|oyQCE@(j&n_pEDxI1uJEwluEK9t>{*Hv`8vbGy}Q=>Sx@b3d0U
z>4e|ofxfL#{HKxp{QR-UF(w$|we~WHcdEORsNgGRG(Zp@GlRgaapXOSHwQH2Psuqs
zIZ}N`LGK=hRgRISre^8v-cfYsbj2-QGblNEa_gQ9Z-v<hv(B!rc$TlX=rfD&@co{&
zwS}kqc%*0x2fu;uqlyJ~wAXPzZe63uahZB20(k~Z5az&#yas1PAE#}pVd6C&^htLo
zyhg`U;`n}A%CvbmHMj80Un%r&9cHl{)Iblo)lT=C89KPb0hkI@ceai$Ji5(FOE&Nw
zkH#AwW%n0QbF+05`+5c9P<qq48(RHZlgs9L`kUOm5VZBSl2=z(7tL=w{(ATJZPV@f
zJZ8hptB(3xCo{nst;!D^VQ5@$r@y#%VsZI+XraDt`?>aZ#NM4jn<Z-rrw;_o?A^ZY
zBvhicsk5AOT!F*1!xmiUX+P1olb%T>B^#?pLq|&GdVU4x|L>FLnLhBrY&Bh<omIgd
z3W#;<69?AoUk9h8C8m*8NJ7ZNFJ8Q;<Y{yVlOnqvjxHq%z-HlXy#3}Ru;~4-F$^D9
zHa+C5!g=KE`d;(@8}Dbv*=ThM;HV|^Sq&su<v#yv9J)iy#>VC<(%H2n6SxYoL!-eM
zB=`b@sMOgs&%1Cw4h5qOK@OgQx{@cY6U5CMBh@Sx%;IgUnG7fm@}qbFzY17}&vF2<
zK3&H(Sqz?~U=#jV@NwbK6I-JUueOI|ZgzkF<;jJLu0-IooA49eg^rpjkn6o~h1~D|
zsHGdA^ajWADl73*38)I9+I{#Cwf`E)`uKhfG&wM~y8&6GOjgJr2<QklU7ObC!MuH)
zljBVOsSX1YXIoX!gAxUWXHIP3i=X%;Ku_LYb*pVTeQF$}Nl-R%h$Pno6G(yGPkxqm
zb#*1`=iG1bWvUwCAJzOF<Y8O$pH#H@AozwCc_UpOmNL;Bt8U2fPb4LAskXC1Z_{G^
zWi0{)-X1Nb4fzYOhqkTFFP6{yh}OFdw79sql<)rfOu@CZG(UeiIXjv>@9`7YbH+Ek
z)qPA2By%5D?KhZAQ3BA-8bH>~OA`|lUHgs-Qw>RRUEmAqP9U+H-R1#lf%pl>I`FM7
za+`67lLud?RqJn}U*Ksfx?bmQECI|&c(mlkwpVp^@zSlpPaL}Wr0Q_5nbqgui*kJA
z4g^HEzVwnCeeL#lnb`#|m$ie>cew?8ilNV}d<2UaCi}k54SgPxA@-z|+ps-(^5jj<
z?#C35%01h+mG|npuq&lPUrXjZGN<hk;a{LqUop@>njC{p0zy*G^Y^lyE)m^IkpJSF
zw($lmwQ;gzU#=!UP&Ia{J<M0lJ8zV%kuXT&^PU{%d)C|AD*<p?LpPu0GVVxU0>}1k
zdjR>7u3x-YN?yF_35+(Tb)cjh^3c^a^NgsdzzZBuA9c^&<to&49VF`db@cU1>RtJf
ziE4jq;NJYZ5$1v<7uF!ob7?!i)qt-zoX?B%7Z+@<Wtbrse@Wv^$<t_PYT7*IcRWkb
zAs8j!eBg2mhYO(s%q|`Rv>Jh&y$aH&M5_7PT?P!zK9ljEOgKJFi|Lk|cfG%ZHq5^F
zM7n>y&VYYP;NNWDzr}^7PH7=fdyfq)%1M`GZpJyv@Cr1(LW-ZfK@Goqa>hW`&n@FI
z*dFK@+B;e@w<FCn4F7rV|0C_Y<DyEJZFhr+iO`_rpbaBBpooCv;LwVJD71)@6v>iv
zZe^54l8BOnf`T-X1tf!k21x<}l2LL_l0(00oH=Lao_p>)_r3XNe&fv0^xj|9SGCrv
zRb~>@1GD?~?JF0#?s_^XQdaB~<?_0KqW~O}TiP6$q^ZOI!l^AZD!*~(saGRmhe|K-
zL+`kg61I;x3m6C8(9+V<)6x=6qWu(7l`B^JWjBty<;qn5M&F79ys>RtqQ@Y-bkbsJ
zmH|LvwEj=*2%8c9yF)2acBPF4Eyeraw)k<YKlB23ic>&}f5Kr^cbxRFjeu)52ymJ<
z#;pZ|<K(m{GCNt6OgMN#@9mrxQZDoRyr*|pKbxrgSBus<b!)~jV?Q}JO03H^ZR8Up
ze&luQV;n9^a{0Jl+v{Hm^}G2hqtV~v{gv$91cG@f&=vpr^=<OS9({b$eV!MA*_dza
z#u#s|MOzr+jcQgl6vix=#uT7<Wo551C~Vp>DrT8H(9Z8D9sknNk=WVRwxnms+I@~c
zLggr{GA7J9`t8>lC|I3#FeO41u*6C`O`-M<W=HA0Dba77&5wdE9Z0h;JoNsaCtm)>
ztaCbAFUDh@`~qr70pzVku>&+_+m6|euPlohZj9IfjM!pC8f&WZqfLEj#TuqVE8Xtg
zyW)|+O8aQ$Zq-FxSq_lFcR+1cq8#IR3f|HK|I7bu;=`P&^)ECwG?aZ?ogdVpu2CH?
z&Mhc#aR)&KA9cgcptbnHAKNNU090-uvI@VywQx_6NDXygS#qTa-E}_0U?YD^zua6L
z(dPwRurI?#IE5;PpMM+V{wJ<7<k_L;sLF|tZCR2E(V8r_vAp>#Qt|LyiF~pAsmFKZ
z^kOn|va_@MO%gzPSAe+L#$+0*#7m1m84sNHsZ)ASqFfrkFa-nyxv~v{0jizUO&|@{
z-OBl@q^hdAf2X~9UR31nBFp1@iX|`ucoCrFAm#(G?^OAj3ITW6ZfR**Fhssv6!#?)
z4Y|iw;c{fb`&krk-}${60p!GKP}cY3Aswe5&MI(A-bdc0O=A`w5%df+HJhb4I(OH>
zxw`!y!j5KsJExq2d26*f6Fu^#g@%NIDLRka@Hjlz0R|FHT;>OGy^m9-zf<ZbzFL<K
zeb++)HJf7kTk^@3_HnZ*7Cy|Cht4GUefy))0W5MK;+7z=gY{tki07~b;L7jtVZ0PB
zctaayiJG$=Uzcr4mvh&DDI7Am`6ATVmiuBR^y<E?`LtAbOV*{UoqHWoPf{SGIe)WT
z9lNk_02C*mYkq=G|2aChei-9>pP3x|VqH;;L%+K}2SzPCIV$Vmfz~_sE<Ol{i?0sl
zC8pZ=6_D!0$w^6E*%?*G8&+16!VkiYU%*M+DJ1`TD_^W&ARNKOyt;Jh(izVL*GoGh
z2?{^UxbHrPY*Q?H-+h}x%By_<m*a9=TwKwK6DK+-FZUq5e<*Rnd3+svJ8-f%d_(#w
z_F$!1>r&c8)&p?Z(|7H~y&$@-G;c+t=*ey|1sU)%Opmdj4}-71hsK)ra8lpjF5+?d
z259-j|H?eB-GJt^391V$E;7~W5OIQhPJv;WLT-9~dw1U+dHy7GJYh5+U|oZ&`ubZ#
zitI_v=I2gFm(nV#Stb%>p%w9M?a0ZTRMToJvIvn8<p|y$S9D~62;UrEUz+Tn^W@>+
z{!2&O>;^Ys@u7of+S=N@ONhjYXeU~L?xWzq2+B5jq7)Cm^dt^Sl9_|g`&tIA`9t+1
z=eV`h!7J10-tWC~h)c?0h>B!yLt}T5w@CV4l~KiIG`n%5`=~wZtVUZpqSw}+!v|M9
z4fKjTy&d;o`|j$O5Fn@VSwEV~dYyZ}1aA0h>*^LUR1R+)@O=rNs->Y}2UGVvuhY{#
zkHFt&#>U*b?c@BLS3deC6IsETZ=_*6H?XF3dtI~vLdR(b5DCEEGM443sDRnk%AJns
z`}}Gy=!i75v|4}0wK=rj+@U01P!Xl?;BbI7DPqFxm;*kXD9C$6ZU7Y+=LJVrJfWV}
zPRo}BLN3DrEZXRq_)KbGO(&9aAiLilSM~~?*EnjK8mZVSAA|_n_x};OzJga@&jpEu
zIe8O3QYVX3S6sdLnL5*m`?Z&f(C63sZ=aSbgpW%*JL9{&XOz<#l2$SWh2y<~Lm^Fz
z9>S&Pu}`@F&fyOMbL-CY;%pi)JXHAD6hV-!?UpWwA9Ej05C%s_{Z-~k(K-GUqAwgA
z;p<?5oUSD4)UGj7i$H?atn^2iDeMHHW{vH;f+RrM7xp7;#@DTBftx;yT^{&wWfwdz
zinc+r@$T)@ozNN!ynwM`d4G{(Gr;@p!M|~(I&fZ9H5iqxkDH~qOvyGa%2N$6p^s~|
zxl`<h0&5I#aM>zT-K~5N-OL8+8DyAO?pa(m?KuO5vRE;BIXXpD&wtEzY{^wVB^)iK
z@I_&lF-*aHe?9E~qUAz2@A_c&Pd*79e5Av<pJ5=GSbU(83MZ$Eb-*nV2&FHG_GtGP
z&z=oL>gd^@^Pesiv;Xs?B@@rXD0s)y$jHSP>J*@diosjtPc6Xzt2#@J3sOfT$ni3-
zv<VX@Xg!t#E;LPhp>_8T<N@-e3?*!V(W%v75Y+UW$cw9rQLg;fqX_FLNJVJoqNYy)
zIhyh(Oo{AZINX2Wl<ss0RFF@c9|i`FvKzOfj?uus4wlKBe4y=|M>N=lJQgGEGw>aZ
zeivzIbHdG{fBRlXzwE)WsUOk;>WvEP<tqv8@@9>tg-ucVE#fAhIuV;6va{h(*~P-I
z=IcsH4tbk{pCu{u1Vh??)9RtoY{rzN_nKcofu~``PmI1ObK@dz%N21~)9H8)3bxr;
z$0qMkRp~G7(WWbYgpb?E;oSp$5|=MGxy`qP>h+0uE}upWPYMZbntOY1iBrza>2;;g
zzgO*iq-7u>j=?BvT1afZDASWT>OL#8Ce3zW1_V+UW1|Vk8w4+s(zZjGU~e-Zelv?l
zfRVH38dW?UeTTZqPplE}sx0D!HFC<GysgZ7XH8R|q6?z++zd)dQK&b!T;CRq)LWRF
zB@eZzM{jPPZdyzL<r?H}+($C6dCj$RFI7A_6xs2fC_QVbK-)B89(3r(hlhv#@0dLg
z5Qe0@OL3{TqxSZ~53r1_2uoX=oS0}FH<+}0N{k@X^vq-`5H9fH4QZ(E8qT%Scc$j#
z*n8;e>c&#2sRF=P)r3!+kmJW#?$B%$R_a#HHClyd*uisR99xb^KZ~Mzf7@H<?NHy%
znhVvFvfid=r>fPxM^=xwHqm4oVo&b#e=i(*>FfMWPP`RLuVYJ&j(R`x!~)LI62$`C
zT{(p7_*5rId~+IuIbmAus$+3xY&ww}p%Ummt%;1yAAaGxdv3dmt5yHN>LO%ua`kD>
zpR{bV2HYk3{Fn+F#<;a1ErmGV;E+$E+$AY04c6;DhF;%R-yB~=$)0f>sfV$N*K_BM
zr|j`S^9+2Y2ii=4a-$3q*G-$UdU|@uY;{uMwL>4r?B^Aj^*ZES;TCKnPS77%_^yDw
zvWzCmqv2x8bsTLh6$!fVlVlS!&zu~2SmTS@qvn2dR`~oR{HH^pT{778<Mk4>+jQf^
zGnB#Lq_6ySV=Wh8xCqArLFZ(3bTkmq1R^nm?cIH{7%`MfD%T3?jr-KIm;U&{Bl0u@
zdZ~kB>CegQ&HBl5ZdG-{g-t$F^SmfLv2UVy=1p^g-_etQ@)Um|q?jWoi$_tbMkyok
z8HQAI@0-~!;{q2K7JL>#oxXT;nOXaPNmwOq1AnE_CQMcNI(6H%q}2PiO2nwtqS=bX
zFt_v<Dx(-;pCpUp=MNtsB>qAY#zaA@|C+dBVp$GeK)dJEy`3ud3@}$2*{R*Qal?k6
z360)60(z<|ZKv>~8<I1nI)>g$1tNP@+W&zN|BVbg^OJ4+E4lcE7pIRh%5?lOHfh#d
zxHz5@XJz_6)6l7DZ7*&(Pv1g=a!WoB7qD+ybUU4@siULg8bUln6%`+_2HMzB<kfWY
zh9E#ofOTn_m<(@o+Pr<fA9?r9&H=>VSrB8^wiXsrtd)kTGS{%%pD1#+gYi?3|Ak(E
zG5`G%5>@O=CQBYoGnQ(OuJAZ^GQky#eF*7uGbb+aDE-O!9kv+kOQ|UeHq`?Z#1i<b
zh}~3oZ)h(ORT{bX?3&OP*ROK-=4TpDKQ*TeiFtRW0wjn*q&%!Ei&%(SU9Aa!-1L7-
zDnq`k)Rk|{5_{#^nJvsxYdse;+Kux}BV<AxMeuUQast7=JK)1_&p!u)JSZ<hNH@3N
zg@wZx-q%$fokD3y8W7wWV3)T6=r6rg$f8|}@OB4Og0G2nEl%Fvr&!6J5Y}W2hR&k+
z%);GDX7#`5=YCc~zdhFgghdoSWP0b`)_^WBAv|2P@a4;44rqDSuV`rvBFCuk?7>Z?
zQ2|o>`=nPfAf9`tQm;J7J4+#~&>{$ZXXDwyaAGE7Sc&y0f{^UXqf^!tZ`|nacpx1?
zFbBq=0kcq%=I!q8&Pz2H`0X}7`M*~}eMwwf995@2b@6d+@)|7zPs4|}s)B;T;tFr|
zqa(@lp{{T>H%Gn7$dKD#Ndws8I;%j(Ixi857k#8f|G--Kw;<&|VpA_<aEKbu<|wP>
zIKp2dT~cwP=1ZJd`8+*>;V1;{=oQ@TC2NUlJv~a<%t%^)O<cZi$b(>qzSv26-99#W
zF1gu|dR4>F80G;*v&zc0^b<`r+Z1EAvLIq{7G0wtsEPD{e7u#5q#K)3WCcq8_vJF6
zCaL}c*=~=~^-p^K{xV1ZNP%{2#v#%n)VrvPqGR#|ABA^pSu2Bla6b4?R2U}?3n7X!
zOH1Pfe#r0gt0I*Y7du?(<|gUV8C{@3aQd{*I<ED4Uq)o*zvk$5G~n6}L95vE=*oo)
zGKL&w%Yc!!bTq;R;(Awc;086_ub3k^efgRL@9mW=0at$uy^CmXqaWo)^5-(S`&abp
zQ-veulQe}6hs=5vSLWBGbR_A{P_c~>xM<`f3D>}pJ|1M_FBf5aiu?LoyKkC<l56pz
zYIRMXi@_p>yI)zY|IjL_@HTlpniOysz7eqI1K%n1D8iG+U7ZDh)uO2whc{2<1_jgd
zK*asHM?L#D4CwI;%XxJ#*?vP+E|w-KKI`ybyBXAZ`m4zKTYw#iAc^ee73N~a`zjZ5
zYO=vhDDA?U=lDp-M=!DYe^BxN0Y6o)4f;R54r^RnuEb>}$0oP2Z)PNqMq^R6`0g$1
zx@%46n5lQTjvdjWL1N-Uio%4mrapGW)ez6~<hWLf&{&osj!$Xt!2`QJz1MjW2L}fq
z`6u?(c}Az+9@8%o85vhGdv1UEUll0TG`Ou3NNJJL+4*Zt!^0yqA@Lrcena-`qN?DY
zTr<A~w+%(f^@pBfRUg_i-Qs1%!`V*R#z2Y%ZcMU<VLF@z7>-<(RPFu-0>I7rSx4|@
zB;l6sB^V;*i6d`GS}R&@`(c8|N~y+JsaS!0+u1q4-!a_xzoZ~)pW;wL?vp>Y1PfB=
zB349tq;U(scO+;$zdcg;#-SFYY(_;#$g+!|z*Oae-mJ!-l7YZhZT;3y;n(I#D4njh
zoqX{+To`i*;b9*-rNi(>>2$OZYVL#LhI1CVLMUG!SNM>yT$73<T#_nK!I6cA-}b}V
zE3>MA?IxF6#Ympp?2JpPAU$_9Hcq~=14vO+(;&Qj4z<mB3yb6Zpf$MqD`*XDt*zrZ
z13#|n$u5-iBdcYglWgx3{IyYLbjFi3<mzTmwXoi@DBrc^u-~^LrRo;tkwRZRAnfVK
zNbgT6L~UMpO&DN_sLf$TQVHdj#8FSP?%8$qCL;}ET>s8>M%&w;*v{ZIwX1cK9>E2t
zT7*Nt>j>dsg2`=RV`d5yMfg!uv~vgW{?vCXm7u<CTBK8R`SvfO2=l|BcN;IBQAnR~
z?UN+m%XjoyOHb)3?4RM7c3<|HTF7l2*J`5q+}_rf@BW6A8W0G$So7OdqQb@2&2kFS
z!8Ue{WZ$^^(fWtJ5wVh#8|?zUZoC&ec=qf{;HsZzzK}KgE1~AHoMA^>TZW7N%6;gv
zs0IviOTVI}I4k*)`$*%*0EHDYOb&R$wYCT>fDZKUbNwfOvg8ZFezXkRh+>|^aodr+
zSb3>ele9|qw;SWqvb^S~c|m2`24o!<&$N{J_wVNpVmMYD936Qtcg(o=tx<EwsvPw^
z(Lp`Xnnp*wUJ5?p@9tJ>wdGsH@B#~@Oe^h@WTy}2O_yRa6`WSoVe19M!x}YYakoq+
zA?^fbMF>WVascUTJofRIpT7=A809v<j8i~8o1BYw)R8On+45Jmnlf%p8O^`5-r5us
z7ng=v0BN=Bo9dLQ9lgxoQMqbAiYhjZ*z^)o%OO9>P2=NP;h2SE_>?c)j*gB>TXb)r
z8k9wNFSITtpB;)i1>(KM^)cx;fLyTmG8FOHb-pZrWf?FiDg33qeR*Tfl%xNj_<&Yc
z$wdMo_w4^HWivkoFjxXpm^yW()$nD7?J4Ju6#XMc$3hxoaBLOBd5H@0&=(Plg&Fs}
z20?3WSnA6=plbTF4(qq#cD~ePA4gVc<r<%5!yWo@650Hfi(cRye@t-&L9eUd&7Yq8
z`jgCFwnZaflB!t3=<ss0cotZ497kq3?^`%<k=)-<$Yo8oK!D)?5XIiyX$GpJhLNjM
zA$e6lxngMY(k7&FWt}c8NmWoC#G<_R;!8EJ7JQ2q_VH=$hMBGbORDiHikLa2X*=;o
zsCw*a-pb3!$#jtHn20T2p;X8<{>8#7?-xt^)Wctv_avrRpNF<<s+|5S%rJ1pU9&Au
z^(-zNKiQQvDU2VK6l@i%(1dIDO3d=!$+6~S?_SiL>sZ-PlL$h-6{PzsruF@|Y><C)
zqigr?_A^7Rz97y_TAf6@*EsjKDho)h1^ugnJos|hg`vcAxB2kN%BD@*MpOF}H%Jlc
zUYp6v>QcY>o+4GaRM9wcR{V&Rn1($V+n-co#HcIDFn44EdC0T+jFhVO*=`i!mBc?L
zaZI=@c7LWhoN8ta_}QMy-f?=3HiJYS^`qexH2B(Z0Rd15Xq6~|>CA~9Sk2KrUd<fl
zOpK8xMO1Ag%;!$A9Z1=hdXY?w-M@c7`8#;wj$LT==BkuO$Z5nI5Q_rmhBhpg1;>2$
zI90fseSZ7p-%96T+5pBQS|kBrB$oiydKO!5<=}AAJ)Qrc)Z4eKF&Ax1*<%+z@m?1w
zdrKaEC(&ol+;rJRH9<CQFi6yt$9;Cn-L@%#dtu-HQ5S~Ry#tBZAzmoBLc2HJ&Alk1
z%m13ws{`#M<ATm7yP{srbY>EDl0&Hwe4Q32eB%{F5hOc+K&b;tBu6PVS*YDkFTI6T
z6(cbCXBgErBdKQSsDACPq9QOqV0yQe2hqTLz*H8ki6}s==G|@|r`$S~ON0};Ns-+=
zKu3BJ|2L_>#oLHbvjYF?b!}&NF^;kx`?#j`K_La{#UBgFqUyq=$G>cGckMPxCu~v(
z84)FW3#y&3`STi3;>S1>ugACOr4bXy$_bmw#9uQf4dQcW2PJy!lypKnPU2nlzX0`@
z{wM{pTb=v%-E~I8n>17#p2+j5=@gnjy!NjAw(q2o@b-!}oa1hDv2`iL^#y=1bS(#+
z%)S0Dj+=14ai-a=#rwo}GsNerp`mZesCy)8NAqm@$}(Y4QW+^V54uRG;RAf6q8vKU
z(9U{Q=KX-eh=s{Uq-11!W{_QR4X#sP=Xnq<&w_xOkc6yY5r&sVuXuOQt)vBG9@}h@
z6Ja>Ti5|1^JJW&PP%IX&Px#o6)JGRGxK2fjn)yuTcNlD3MrJRL6W^}%d8X<)x8XKv
z?*7y`d8GoA&5NIP^+lT;;H6T*GM|U%w!bnDIPE801MUXyma!ruVF_8>Bq$7C5F>hA
z`@Lx_M#h%#M(DNK?FBXMgwZ9PAFHv@vHz%$*H9=9-7+mMhlM+f4`fdMA;)gUQku=b
zvM5fhF=@VKJjLn6ju)M?$!T=Tp*+?zYtmfwysTmUDV?_SB}aR5O_=ceudF~t%fN7S
z4!gMAZM&iiGT+%$K5hjFGVjda4l_#nzAz5={r}c*pYD4#v1JjHMNCV(DA%8%>fZar
zCRon6ep$@%^~72+Fhr6KaT|hyf`;tJX0bm4)V-3js4BZDd|_2sh%cNKg%?&jZTI=O
zV>gxl)5QCSw)9V=Rsk2W-d1*}Y5vJb9EiLU9Tz@^aLbq?QJs#o=pw;hx9TXvR0}oz
z8G+t)a@h>FhNYK5%0om|=Z;wQ?8z@qFUF;f#i|ARAN{T44RhtGsRmBJdCv|q`40(n
zbu5IscAEj~B526wYG>i-d}`r}nfXn@v?*ar2nlb|n}F%KT2^vmnP+Gyv3!w4_DGDb
z4GuiK@79TXtXzATnN;)hyB^-ku(!Tn8O?vC_{U^sSFd$%!FGAKjccrpbxC)3leKF}
zo55Gi<LLM2GM*Z|I+sz9zx#Q#NPyc%!Fv}DGoG%EZra$Ot*WlA-Eue@-LW`dV_Y-p
z*toGsKPMS2mAL5ct1TYmo<QuA8F(RZ813GZT+|VN*T2$=>hP(1^R>RJl*zJX88il?
zK&Y|d@?~f-R>xqH^o(@d4UiK9&Y=CiCW%sfav#(Zs%hGm?ZChr&dWs3U%=ktv9^ue
z)NEsLkBgXl*p@$jC$tu2T#p%;UMEM#Hl@IHt2jozBvgkuA&xrrSbc2eE3Xo6CD_Po
zK^Gcm7-v@#K7M@det!`odhH0vLrOAb_?QBEMqE!~u|9kJ-)B76Fjv#k3hzj;!D7V}
z<B-Fje#6}QR9DBa7Ef{mbxAT$gN{vqx!l(!vl35#qP%hUa#-JY_l#qVVJ{I44UO!Y
z<k|Af0Kd3jFcmB2@}nU#GkfSHmci!jXq7RQA~u2D=FX>+K;EA48CAp%K``e=WVU!<
zu~Nsm=nK0F;+JI@D2I$0GRil~CPiADc-|et5=E&21=e~R#=&%j9zCAZZ0lNOH21Ck
z3?Ru!go%DwXDX5N!J}pDU{TVoPfw07`g6Vol)J~E*yVu4nSX@!yS)e|=8;g{dVq5G
z@LfeCd(#pT(;J34K_X3v8%axmczr}t?K$DcD{`V*s;3IuP)z2bL!q||ZbUe@+8L)$
z1RDU$`N7VZ`H!Ed%D?lBYHvF@W5V2?m6w+*M>BI|)1KzU@=q*%@Hu<7Dzlu*`GnLK
z&a{Mh^_AtwgLh5sU%r$|eZ4J!K^OHVPme&=TF+rw58j;h$ecv<w7yB)GQIdo6BswL
zw+Gp|AG9zW%S1FhmsikXxM0G*wtxO?m%sBCtL)<d>jgFQ5Ij+e8s44?e@vzqd6tx&
zJ?B-y%uY(!{W+|8*$45BZC;)LJFfK%(UX`PvN7b=<P6d$KRxHUfvQAm$(UxYJ-hu}
z0fnJoUkN2%g6V8^)UMMT*rCfxD&e_LC>Ti{WnlB?$Fe!H<6lOYWGu+A!a@GY+&(3)
zU#xJXOgQYEUy6p@S)X!piIZvGq+~0X>J{&H@@#eb>!y;^oZ3s<sOs1`%JMrk+!pzC
zH*V=6&3=~6xew&UyAMf&{Zw|r${n$yW{q596@0<GkLN*XuLVP$TTcncItY89%R&bu
zY6zpr5u7#EBdWQ3{Y61$7$(k25xGknZZP8G#iMAZ;O~4bqD}0P8}-wooAYj>uyRKR
zRXWqW6aZkbO$oTTrmnEQw~wn08N9<1*7U>S!V(N^_T2pCQs=9^plFZg^k7C0??QLd
zRX&^afEiFLp}Ma9&uB{Nnb36m1rS2620v5L@u!y>t^?R0=rejc;r?&WNxSX7hTQ!?
z%*YT&eEOPhs&#zwJEDnOGB3X^DZaQ>aJ!aC2b;Mxq17gMuWDc$7(F_YJqsu+#mU%y
z5H>Gzlca=sLVa;KhF22{2Df_H0UH**z83BaT<?sNzN-g>v+!mkaJvj)7hd!^_p(D#
zd4}@v7nCHqZ0PR)<_Oa)5o_`LVY-6lhYSicaS?j}ujDf-(gz3a^)Ec35>(f|)*Yd}
z)7_`=*$v|mN?cwaaXc(Yzy58>)&}5u*1`F=)B`lQ2G+2mai{eN_<0f!r23+=tv!{&
zp+Ky3MP(1628Q!NKPUKo-c10pwzh8IRD{Kf5g$FCF<%3ezhc9Y5v_U~9+qY%h*mNv
zK`+*4PlKRUoxH6-P3Gs=8uCo1JK_cDJ$}KcFSp<jBXPWNK#c<G)u*N=20F9XIV{2F
zD+DoyT$Rs{AX|N#I&CE&d~u()yQ=r+@RX_(1Yy!%RqXn&yt&S&a$tUF1u>s}DU$=H
zF*oI2nsqPdAq5Nq=g;I>Z=@5iqXx!Ky?YaQ0m|t%%l0jSKV;YwoI9@*6eRH%_<3Gs
zWeG39+0D$!$zd<+dBYkeu$wvEiWi@_)Kcd1D9xQIEEI&Od}u*V_&Hm@Vbz$!gi&K`
zAJQ@~3v1fHIzP1To^SpXnhScCw{M?6|5gQh^M;k3aG}V`)%cmZIXE~nmv{a|GK3lW
zUNtD+KhEkAcEf=T2lE6ChwpXvark~IO`Aag_&UdTBatjDED51LpBgp#i*34CL$pf)
z{H9^#PIHUC(=DGwoP{A<-Zw&%9=NS%ggBbYR})qdCUP0X04Hg=shCRr*wIgOc|2U*
z1)<1lAt={;t-(5r^?HOe4W0qnLmfqj5Z50PWN8M_xL#A;_C(zofx=f4`_M(QFkFj#
zopzPSVvQUAi2iBnN;();9Q|WhaTg3Lx>PJap8^i`Vs%s?s<c1%^~$S>BE<CDs$63s
z_I4<7<|jeQ2S7wtsB;K71Kd@7w*FxSrWf`NG~_h;jqUmw(uWW`eb;*1#>ZqBN0j}t
zB!1KR>D-!k2dn1MS~n0_9<OHGVc)p<lj8V6&>`7)hY~(DVSM4mg;3t7f;W;}-05rt
z57EVgUW*}~l$$qH!ih4-sN&SA?ZSvYEzbIqA$g+lYQ8B@gg*Z^a*7e4*EFIWi!qw+
zAmuPTBAyAT@kB}oC>Nx>1r-cn<@E=^<#`tCLVaqin1=~H%u$!S?2_R%;%pZu=`d6S
z@K(r-HQ>drsB3axqQI5qHPMxNbxxyEr|ooIZUH!;$aFYIc6|8~G!ecEfpw)W_u7Jg
z?!G^Bi3KRN3L{`B22$%M1F*7Y;E}W&;OcN5MM+-71`)VGwA%zo0&B@9UUcwf&|<pI
z^c1xo7Zl`lDH4QB;uu`)vtR}HClsXy&2k2Sq>gahx*%t4f{5Wa+!i1H`1|0O#7H0r
z<Y1bx1)3BA<->C75hmX5Z*DQu9ol&SFkilztt=-QqV<aS6d^9dvf{HyI63$9@+#AX
z4(A-Bs(aM%e@jG|SW7N_(JFL#+I@sW$89l@;J}^bhkY^$Y)5*()FTBLoma-WqV~`Y
z%rTSI*JF<zV+s@CA{A8@OzXd|%N8KnpVVku33bv-`j)$?213PkR}#1D?*M2h5GMkW
z87ezEd@Z*Hm-+OQ>OyHM_0l$*HeUM-(~bc~JUfO7*R}z_TL)JtNhkj6je9?v1z2ST
zJ@1u{aB^AirouB~;=bK}j_3H<g3qUdP%#bEFfa<>j>yvvC<>2WC$2{gN(zMDZVVYz
zx!Sg&eEMkfg>U;1Op;JX{*=h>!&Jz$g;i5R@^gRkrm1;Hx^;V1{oo07v_#S3m+#};
z!)7WD!gUM-cMw+qEmELmGzRuCD|o5tj}baCMi^7*r#_V&IYTzLuvm+P-Dx?ZdCBS1
z=a7CEu2IkGDBnY2zRLtnAn6gE5NtU-*gLqbU+|rL=uq<)l$(A3FX6RqL=IR>CRkTj
zr#pMf|Ks5NK3wmfXUCpaZ-1yNr3)?7Ts2E3mc<)`+@gJ;eDu2CjYX<<;mCUV_3br*
zB0;~T$Quz$w}n{J*&1T}v3y6uF6g}`8j4i+=z$)zd)5myqSs;Ya>_}%Obu6|b>+&!
z{))xlzOfwrQwsn$yj95HE|!ye*JuY;Kqp$<JL_eBM>0fI8Q0+kA|e4eOYQXh;G)ls
zl*3T!W6YIN`<fOAL>N&b3Osl`3Id)~ICzVINuD6CH$VSym_v|KNvTO7Z{JWV7bJ8&
z{=ga^PpIMk0UqS8gjSV8Pf6-nY6kZOR9WA~>~lWr`YeOvo4uLTBTc?bGszOF4Xreo
z1q~sVlAv(pNrSfYeS#hjOuQj>RCt0zh{(uk<noESE?PX)mQNgiY6h-Sfo5PYp6fX=
zapL9V^4-48*-RqwSC&xHd!+nHrvN5j(!iil9fJ6FP|-E)JLKdpyW@CQ=V6(zPYBgq
zE0p?Jz%suivo2;@3antRAoTdkbmiBEeOps6czG(;6vS+~^62}wdS1*@4lH#P26IEg
zwy%CT;7J4jAQUpGY{oA(1kswAlrh?R5&f0lY!=3VJha-_)EInNql1)=H{&2+hfqWL
zK$ugl(&K-1>fIFJ5dK+X<?mJ8SGgZbKVw@kCZy39C+}Y_yb)^CCHiqZeW*z>KlOw}
zx!EUfWxaIvN}&!IaSpzFJWw13x`EDFEP=3j7*m2`bJB-lJ%?ZuT~E{-d`DFrI`Cs7
za_V@VK91}OJ;k(AuvNFqb)4sGWQ>X3QZH%#9t<+>R6c*rRveyyw<J>A_2q~ebZ@>J
zDmnjsguHh;exi@;(9AGZq-?W)PbL6R$#jS1ugnLWrwG;njG1u)@yGRl6f-_R+KYmn
z*BVU0I$VKb#u_9}l;Pm#D<2)XM)`=v$MbsG94$a|Ip`O(WP#gC(Mcr4-GNS|>O&xD
zu!6=eK<71}Ual~fRp4T)(}pMI@WQ{q@k}q?>GsL@1c#_&%j>IFx>rLh600C5bhnJZ
z(b3se$@_pa@mTCmoCOe2l?<LzHE<DYFTur^BTUXTwr~pi!R8ry-QT5f$ol%JvcW9e
zQXLeIxz^o_@r+@Ke@}Z>WaVI6Xi{=uIgo~FiHWWbHQSe6cw*+R&wN*P`fnkjbYc;M
zl3-22pQn$Vu<n-61zdWwuX^G0;9JGRpN++fMMhp%X#05hr_xHQWx<T8;mKcrooKcR
zif-+v)MRTQ3oY9YVGbh@4bHqd5OAIWayB8CK=k3N{KA*nw)x@irlz3xLU+eL9yw1-
zQocJXj&|S@0#=+kKQm-a_rh=;1Aq}`sJPZLeVi5Yg<h<Y+qDNxHMcEKU&Q1W1^2K7
z*Pb74&v>h&&f0#E*<<GUCK|Sf@bK6Sz+#?a$h(uxCox-Ux4`>G3>mX_{9Xu2=_PWK
z^O6rXIe-#w&LBPQ#Iz(g8jEo>Eb=9aT%|*rc=p-wHk_3{INs4k#n2T0*uW+z4Mq6^
zl)pA}St@l(BnUM{lO0)!lw}U6-pvG<zSt~Z$B92Yt8xse#$ODE-GJkH<?V7XtW_>%
zg;0}S_kE_nqT*Sk2m`&u8O?&hqm@|QW6>&^P;N1R)_6rPF@pMD)wBD9mv@O)92Fzk
zt5MlKry5t7ZrWJ18;rkVbo73zbLqwh9me$#E1w0Uxr+$2wmk=;a#jCQ$1Sa$eE@TB
zlu=-)6$`}0#Pp_jdT2W?fk!Gw<8PmK4}@-{z)5<`F$+jByKO}ZDRy~df;B9AP%@l0
zEMmIZ#cge2SQx2n(=qIQPGof<Q_U%bI`Y)Xs^NrHQd{MCNIGqa)NeC3B8fiMj8ZG0
z56CO+!Y!qR5!9bFy#3kd4Ya&__YO2D6RxMlOfR9XS!XV7YIa<rlla<Z_Qrea36*1W
z!0fCv%eudWmYZj5>ge_T2A8c!&G@mYYeHu@E3NzjvWVX$)nxFR(XtwC6_B6Tm&`w;
zOr;V9u|o?x4i$y{YmDe%apQUuB;tBRK0~FjmzqDpy#CNnkBSfTO&1F{;+xDNlGM{I
z1y1vLF^BC#80b#bJ8yA6QcJyzY?fCH7Eh23;{c+aRFe(vUs}gjyPFW#CD=!04RKI1
z4SHzdaLy}=A)F5t4{Hpz*#un-YLCKp-V=l}>~lbBj(*3mR%g*L7a}=wYu1kX&@L6b
z^LO3V4M}V`)tCJ|b?NQ55@*U*`ofW+D$o4t><Xc}@3XIZj<RO~xB#>Dy<i{e5WHz3
z@pZ2+_Zw>&x`Bjx7jzm=%^F-~h|u&bCFEvB0G!bsXv-wV(mMDQ=z@$;8%bOo^C*J?
z@uP^Ezj^)q$UaS$D{V(%J(KWi{Qh8WTs%E;oKJaOra&PqTxPTm$(-jNS$&9CJbcPK
z+ZbjqJlf3oIgV3cANKr)Y#54;CQR)%`s@s<r%&P$Cet*Z?{O8;6DM|uj)=kr^3%x`
zjwkA*_bZNslm+wndpup5f9QH&$MZ0wo7dq{vsjH|Ka78DZlLhKoRfNl&C0z<lF-2b
ztj(pDSMped4K^S47k&P!^)XvQ^4Hkr&e?^g!4NVB+o#t1L}}OTQARrLe<DFQJNIKg
zUHqRL$hvDab#;qEWDue!3YMS_$E&Rz10HQwlv6w3t+&`Yd_543ifjyTj`hYh-%-<8
z@J*mW3>4HSfqrfeahAvW4D3QQS(wmT1_nMgv$E#(+*J5Y{PM<w*TTODxYoxX6dR=<
z`ufP(OQ<gQ`fiyJ!tb=6^vy<3o<5~pC;x(O#$ExE)gh_<G+L7)_R7#7wm(`rTo=i!
z(Pl<L4Z2Wq+dAz1spp2!NXco2`ntNnABW6A@3e;WjW#c}qK3l6wQ*2L>Y%*xBbBaR
z)z&`XIV+bmkB>!|(I$J*?rJ#AgC8uEf<=B|SmTX~U;m*Pl8=3H*hq~U4-vd+L^@rK
z-M$jBXkEb+hK6Y`#yvwHt{DDo8<A|GNdbtNppj$vipgm%+#!<vX#vE?xZ9=6KJG0s
z!ey2*Oc1r1Q@BHgjpcL40^(GmpcwHLz?cIh`vPPYuTw$n(}gfd7f}nWk$^%7^NFVf
zG1siDZ(y)IHhgK+NG(C)W4!wkJ4v3CDgMl(!KRq+jR*D|rT$8Zo=eV4_~l5CSQFjR
z^zy~$So9if8XFu5KC**SX|lK@iX*!!gJ308Qzmd>y_3IFlyIDiXU{K|>1TDze$Uu7
z_q2}*zKKKVh(vmNy5WrYq!1V>yZ%m@Uts_hPe0zZ!54X(3Ql<=IrV`xwn8ZElluTW
zO<*zeieP#jMmctH22LO{=NP@3f<n$wOPN`o(qBzrPn(NbQ<it`jkYF-@qUh`=C=Lz
zCS(z%i1U^Cx<Ezzm3V!s`D1Xp1(u&FAP^j*czQs1>-7`NpcJCuP?iA=wB^p9_zg;e
zK6f0LjT@VdCI`f+K)`oRIv(~;3cRumEr<6dgG9-?JG2ZwxnQ_~5Na-kisTkwEhk<Q
zV%ZpbW6^4$(`K6-8A~P>-vrXZ#o9W9eehs_*T+MDX|Ne89r+H&L}5WsJeuhG*t0X)
zc4&60PeM+8|NGXiCb@hhRkcpSOqK5CsK1a3?<dK#ubiuIuTTqAb4{%Q;jr>V<Jizr
zish`v!jVre{(UAzB(MTWaawwMvQ2DxOp3?x-CK=jLSGBw!EP-UDK&O4AR5pPBp9fb
z9L;+U=8h5w3|YMZ&QuR4JkbZU04yjs*~L=maz+XZ3Jjif+UQ=jHQs2nkJ}I89zl%v
zj5RU<J4x5AXEKzE2$Wn(8H;$E*_~V|t$HVOGLz=>8F69#TsM!=(i!i-lvI{dSJP32
z1iAb?kMln3NzeN>Y4LSxvu{uJx3nUqUtuVu(x?dr7!5WKI(;qhzWquHZSp8c2a^<c
zD=2IRa&6K1?c8tI)bwdR-4)XpK6u}oDelUwGxwRQ=CqwEe^^;73tY0$6352exAz9=
zr6TSNLjs;nUwMJ<PbCe`C#G4-;md8d6*%|oVgYTM?fi(s@UNMaLYvsdkC{c)rQObe
zgPaIbg+qw^G0aP%G%IQHWWmFl{nb$dC|Ra`_jxFX+~k?Pm8E4LC>Q&w=Y0ER(!HIv
zE{_$^pYz~%RCrmbvQK@M;NrdUG)Tgq`j&2cbo3G#HO1!Qfc0(O1>-HI70iI8fvMqN
zGqq}J>xr7O$!9u;@E7&-w+$!y-h#J$E21*pcNBA@*w;45eNNc%xNRQFOFT~#TRK4|
z=O!g##POx1-=Wa^+275Kes=j|7S15|c7V*$TC~Y=3g+?r0>48m@FzAWgoPD`Tu%;C
zwyiXQBLyv<N1XoL(5!WLx#Ke!7KT)&jCjFxonuM9tHPZJ#74$22gG->_$0t^zv0NX
zpqw36Ops&@>(<UMKK9BnGT!-1h=N3#%`>TGfj#+|@vSaO9p|fj%`Gk0`;R=3Zc5JZ
z8H+}jY5lJX^3OmU7}pj|7sBt;I<^))$%JqgOT5jzk>Fk~GV*+;|3*aHleM`B;NX18
zTjzjzvL5NT4qYc>96BoVc=baD^<EKI9@b3B(~Bb88LwWA#qucdYN~zzd0kNBd6yr6
z-rpv3vY2kJe2vWDSF{GRd~p=u_7J-1m#{4Yp{5=FX{HC37*V9tSu4&Uwo-<tK!^8N
z6ISsZD=s3@_ix<#M!liczFkJ5su;J;u00}W?C|+_(QaxDL67B}Q1v>|m&kSO(E4*5
zpW8tJZnEbdvt1HO>YE!;Ko3hBfZbUKDCVwKIL_5xN639;XK|57jH!@He^GT=m#J&U
zEAV1w?6^i<i5?c!_JPRe_%x~`^8{jd)3g-yP~M+AL$hUO-tviAcfFY<PN;5it$TMk
zSGZB+<y_$(2-SPx?Kl7a1(%ds7)vhvTCinE9efBpiC#^dK@4L1=%a`c;JzWi66g&R
z1rwp(+?>JZDGiMk(j#u`F#3>~_v3}<%0bD76Yj1n5{>6lU4%;Au6l;<#Xl4c^|^ID
z;pA?B$2GeP{Pk{q%6+{kcL|Y``{MRfZs-(8ls}9iCp2$1cXeGC96XYFh@CWbJ<n9t
z(^S^D{yf^91U<{ibA@~H7sHPp^7N|qd|5&J^?Wz%u`+#|xp4z`Vtg=LRX=gjkUI~)
z4BCc5%Bq2KvZTajTfKG)Ci%Mxwi0)~k^(YNI7S}v<*2O6>?r>oJCp<TFVAIhQP3<d
z5sUZXWlCzwU|+i2y6{VZ)JArV?>xZh@v?-ag9?F9DVQ1|7E9)-Vhy;|RDNE+(gp>z
zipu95%|Y>JmbhvPs~$v=`Y9hg;m*QCv!JyMWk=H@-mG=9)kc%ck2~PaT@US;nMcW*
zFotg2$wL#Sv1((<*s{TWP)9vcIjWOxDfaYIhLEtZSwrzzc&+-xE__h%qv9UR?q%2A
z9O(e;I3eCod)%?u@}A~q%?jvE-V1PaWEm=mgCz(C$tQZ`klZ6obQvhFWsQ^u>&H_i
zz?yr(l*KkEP4K|?`d?C0w~IEvgjA2x`}1TfXv~ea(;qMuO~D>!d<7G-A4UBnQ)x-|
zN7F5j<!yoRWJ7JUX!3Y)dXNb3!La>E9L?cpMB9foHt02Ke7{<P>{`H`E{%sfSo1Yp
z#jT*CqBn1xe%9^zV5`#!-!z`+1FE$lR-aKNtXMl2X|Isp+id#ES=JNRol5%Gf_~{P
zXaz^1aKaZSTUY%RKO_N#hevffRIAM9hw<`Mr}}JnVbef7168DSh?aTxvIwHlW<KIh
z^qAG~eelfzo>b=K(Sdm*Ui=&4wmwWKG5B9^k2n?y_zQ)R`b(!W_z^OjnQPhEsFCI^
zb&$E;pavu|*M6{iqtwQU+MG2k^lL#`!(>lpE#4X_pg>$;nNky^1w*{M5uK&TK<jSb
zYlSxY_67!hqvwWNwSFtkb~;ab_fb^3g`_w6)Yzi4dpzb$j!<o$)56fB#-)dR7~bpf
zT>G}GC)E!#MvWKSm@ZO8iElu)975dI0lEvPZ)`ep>QJRFas4J@Xk%@y3ipu=_^B3H
zGssB9NdCu&(jzTqfXXl;h&a2oAhNj!(Oy!b?v1nt<60YFPEO!o_I3x01R}D{I5PQO
zP-_cp%d=!#S3l*n((mi53F_m)xwmUs#HLphhX6tp-gv~({PqJlap_Wql=V8M=ey%d
zLFa$Hqk%lm(`!gLpG+31anngB>O?2NG|jc1?gp|7m@r!b(1H3K4b&*_!;iOSxi7@Q
zh7aBhXz`K=acM;Io`F!R3!0zfr1%uPn=QhfYG1$mVbZ#&>R!{LReI|!kL$GmJLg_e
z(QfR}Fo1X7;NCzT^=0fRG7wH&FNUeM3CYInW!-0c&z&}a?h1@Diz_II_*z}>bX+NY
zapvUGsez{eVJpm^%IYms!X-k1c)@mFJ73xTFj{vXLPJ9nD7!v$0~`hjUJtll17MSn
z^bI#JrKf47o(DL^-0n}9izH#my>K+6=r$$7aZ62z>Fwo!8rJ0Iu?|{hb=)~z#c114
z8;K@{qkXV=CkCZA+GeKP{w(1(d@6itVmJK98!+OOO8s_MB?;LqBX}_>1{1g68{MLh
zAU3X;3$ld8`m?XpAv}&;W(#)TKmY#kYXM5vRz^=*)dLjuVp;ZIDC&g}+C1!SW5api
z_TEQZV{(}gp}JyBo;ptAz>(>jP+ynqVGrkoBLX5raiJrZ#oQ>^=WjsarD*>uZUx5H
zS*!7v3jpo<0uiXdXS~P<D1>ll!B$&UHB1l&5^8h}^b4$4+*-vL;XQ+TNE)y@VIm9{
zu4vrKyq-S+!-n2M9Ko5Pa5k*Fm&s5{Nu$}PIaA~k=-H2hnLe>S3$XDP^Wd|C6mwdN
zHo%PHqm28X+rdD?zmw-FO}L5bC^O_mNnx60^q}N@KzX)(rQ>;8%=*9O3Qf}TG|(b?
zfF5=whZ5GSD9tKACWYd<)bD<~gu;1`c`T5{LT7}#3$|>P0-)@(Eba2E8W{Cwn%!kw
zZ|2E9oBTdNrd_B`x*}1D{?$aP_EH?2liy%NYj;6LL_~xvS}J++c(}rn{kA(G>??`K
z?w>0H&EM;fvs?fqQD(+5m<d-NkT~px_EE~Gb>(jTZghZ$T#g|q7Sgj6^?gGtKHbr2
z+lOoD(OB)IFrBoqQ+5zyJR!4dbM(aZ9Ol?V%*_M#%tjavut%3F0(W0ld?!zDA>3SD
zH-9$_2iLPDC_y|QpqYUWtc5O^2`|O2&r3UJ_&z``LOY+tnQLQd1~dC@Gx?Z~*X|ox
zdCZmR$QjgEZI~MgGtphC+~~Kftq$f?Kn;-ip8OOoG`lT|;pq|g-mX4|!;T-<)8Y6}
zu6$!hcP0@Ju|xdZn}<H7H9m{WS~0h=u(6pDu|JEYCP6bJHUpw}xmImw0lcVeaRqPi
zd9sOQ&Nof93%FS>%lZQf3k3Z>YrfRipm;cf$Aq4Q6!{QFdU(dpYeLou4Xp7(Tg*-9
zw88#E@|J<|PsM|(o4)$A43B;^gXWShxq>q*S6=Q+H?S|XY_>1Fdy=j7DGk-43gPG@
zd;9lLIzglMPr{;mvr!-x%nYvl`n2m8OtgA)V)iwh$*a0yt+X$?7&$@6c!Vzs2gCZI
zm21IL7A<mDUgi_2HV2~{QT@WvEpK#DOTlzM{h0))^V+iveJze1^gvhs2%0GcRjfDF
zpt<v|;5ZOY<Pd#6vN<|hSd80LEK&#yFzd@}9sePX+oE5ADFMA9Ni`!F!E#1?Dxh|N
zzKbM`VbW^DZE=Y{6By#;<V-ZoHaaVS*Wjfn2*hu33Y*zn?coH#@Jbs`B6z*k!z&I5
z`y`+x9lssTI~+o&W}+}=t#ruA6R={Iw=&sBApFEk(P)ANsCbkWY!CNUzzsKv^oTwY
zfzy0Bf3&CS#l`URx0OsgoGJbtascl2ve7al0PP+^N4J-qP*CXi`jO0Q&23qaOTWtP
z!UfEVK`g)TSsXcbDbz9aPf-SVl7b_Bqe!*hZlg34-Ho4|hZ<#;$+-R!SJvLdJZ1zu
z&vHQKwoTtH^5_~EC>`YqC37hnatg{Q^}3!dlUJdRyz@2t=2cbI22SsIt01Y8NdI~H
zeMVoF8}fuEe?#JsXM2MI&D9<IFp`=%>A?KO28uJYcIm0JINxFrFBrnY=k+nh2je-L
zXvOs|t1RxYr);oI#>&EasZ?i1+@^ipgHQ51r*@WJ(SigNL;DBff-*aS<vM@}hyI9q
zvh*h<f~7dpf)dK%@#e6NothzS<cS;YOra!Bl+e`hBkBU{SzdElbdfU7GaeG-S0l%%
z>dCyMLwj}^K)*+iXVn%r#Ezu2j^8L&-$$q^y%}^F>kTv#<pPaGUhMF>ogW5Ad`k$j
z!}eoOVBo`+&g6h`eHu?g6a6m{kzm1dFvRxz4|-yZ?)K4CC%JEAx~y^lCgrQgHC)X9
zEe}j(j>_^=?2lcrVZZ%aKK;P&6e$S%tTkrgwy-`8a+P}ZF=3M#PlOMwKaE}Vmp+MJ
z+IghF+4elS^wRW%L8%+|Q=^IoxDUzW(>=y3YVG4{jQk$f91I=d^9ybb<+%)(NsRP6
z;~*K}7S3iZXZo6b5}~lb#hEhk&RGG+bk1SIJUX1sU}_oSjUo~GX7^V7;JyY6OUsFL
zJykWvu<|};TW*!tP#MZ0U7q7=LLdn-IW=Dl`UqX5+uzNjB7G~s+<fQm#Zv6=L*1gE
z?tP(;Jqd&2mbum&IK*&Hx`u5Kgy-}d(h~cYMm@i1X}85y@bJoet%_&4;D$M?xNujm
zo*7~QKVjyxAO_jLQU!6xLhpaW|9Yqo1_U~H&I?OSpBoCO`a>zxetD(qY|6;fCS~l<
z{E;|mtGC9?VG=VaMeLxeKS9r^+hs7|0YxI}2&>2vciEj^p+3i73KVn&&79xa53}=c
zU=_;3qr9c7QIL+y<nD%v!URt%PbjI|-phTg8+$uN{rJy3ww5`W;vL!)rgR9#(2yj)
zLs>GQi2!ePU@B-&)D;`x6u#mbJN_b{*b4FYhAu?;JYQnYL6GD4arNij%Vma`H<uk}
z49lI7Aft>ZB_0?seaRKdx_-aE7ZAQ`3R1~dgJ2vKaiZPlr}q8;)dlSeaMj86kBIcg
zD!A-+$q67q>Y&M^{MCHeIm5v7)np*SD`N2%0Ya|k-vudm9;>A`HI=UU{ssitEJ~6`
z|2fnM<!4kpXaNd!J+eas@=W`=fUGOLwCmhZa2P=W^&=|kOm*qFGG++;{{NvW`)+E4
zeG(3G88dLvv<5$1oedBSh4Dgjz#o25AX=sGJB8I4WY+<BB9#--yMFDOn>=25OYP_l
zAbVunbbO%z4+b%C2dbuSZ2baseI{^sFf2(t5E*f=n~;)_wrF`Fipa_koaA6Mdd8r-
zSG4ErVEmvY3Ku`py!0+gK&f1-?GVPn9b71p?4R^=%s<Im&Ln2d#^Yc}Y!;-u_u0Y9
zX=q=P2VwVif{kvwS6!~@!wd2uS^In~7KUn@Z~%^NHinvGb`WQZoj`}Pu8a*Mw5pWs
zVY_>7uwK3{cLFZ()`IDf@>{^kjEY`HoZWfP8$n0^)F2pE@GM++e}n`0Y>$!J3NSdm
z@{a<ZqNF?1n{!|5^BqjAX8rH=CuhMz_#=e)E5@+7faRIQk1&O*1s*J+nB3!R2ioc0
zycR|;{ZYk35~>tPq|Hc68}~)=%NB#jh48X#xplWKN^-1uN#eZ<9nb~y4;QCE3*CK&
zX<j*8WaJiZ>)eX)j)X)M6U)zO{Hd%9`15c+bYgY<QCP$wwh<3D+Pv7?5{O^sBus*5
z<_K8{<5}jUC2xMXm9q$*v03MCM-sfwW8dv_dXt(V=P&*GK?J>(vQ?^>%L@};ycfc&
z-8Sp&;!=Jv@MF1mliZ412Ae>Vj~xvl_z$rdiRrI7gY9|P*O`B60VviyAbRVAL=baL
zh~*GEtBh*fWO1ZH0YJE^yq*w>&veu6H~PiyizA&BjD|tH5O!tXhr76lN1wVcxe^p7
zTa8xEzf(;6o!<uwV-lCmZS>2%qy*kMX#B4bsYgo;mo@vjNVQsRSyZb+h{(fb@9DcF
z7v4rs*?H5kHTrYXi*Z^D&I{kQq#ur|O}xErnQ)RRtjP)1gM;GG3c|ZRhkTq+ExZFb
zVs5!D`i?>45QBJiq{Fs}V&-2WJdAbQ9~d-lE!ZtwN^K=qP>7kYVsUScU*_b@vm<hs
zjs?_oP*yj*-a8E6Gk*WTuc4W+OwvsCXw2#kL99>7g}yK&c9J5AgDU>5l*%CKr%@X3
zt~n{1A5W9-8ul)!G>=Xs#J=$%ZGRRufaEgJ@Px~Ku5FX#Tk83#u(0f^hUW{d1k@?!
zN@_S}i(B5A0!bBc^3kdnFbnFE<$7YR8vxBcEN|2=au^Or43@^$p?*2Zc1a$fH|l}+
zJnm;+U(5L({rm%$71RT}D???A`|^Fs+f8)hGoYsN8+YoOpI{E-Js|#vFn$eu55u9W
zu|?^%SlG=y@7PySK7apWPNDQYWm;x|wX1yzyt4^EmRLh}NuV?kLujL}F^r*ynRXDa
z+khG1wxWQoe?6UU<J}SNVlZs}j9AeL@ZD6^P%f-n&L;M?US3Jv&0J@T`CgR!8?FLo
zG(z`=Y=c1VYlSa)H}aB4giI=!E>pS?{*x*8f=W&iCqZ>g_2qoEx93Bf?xm|^ozJU9
zEyv4*S>&=O75f-|tnx^aVXK#6=(TrPW;RNlrt$Z*I{zA=wNY>|Qtsf#EeVkve**wR
zZmv(o4K@xn065J`$M5eZYY?UedGL3NEcQru0WZ&mxH~tyms^pE(sk`6R(K6s@F1zz
zd*QRV5MS-Kr#!tUAUzh+^Wiu~!N~n1Bw2<TaQG?Tk<n5n2LC5LuU{%Lk;d+2ddvD^
z<i3ifh30`9d}!B=Ng7{E`KZE7tDsrZEfF7s!xOX;sH1-u&(up6?yD5!nU;XaCsSD6
zypD_TyO~00c7RnH7hduFigsP@Hd`pk*EDGDx~JzxVh3E7K1qw)8?(aFAZ5D`QZV*P
zT94ZF6Kn^TbsrwP5m3`loH!(D;49MVd-7WLKpuSKy6Pxve+J$@&%s~`7b-<4|MOvy
z5iCo(hP9a)u2z8U8d=w1CGxX~`z;@m{yi7+FRXIgJ7AR==;8(?9kugWkckyq-ybc_
zpJVyh$&pzsvR;<YJ^PN+jCjLu$<Vh^i`Hb_iCTd(dG`8|v0Vqkt-B!${3hJX$xW&)
zEg~0y%tQ{!9Ov6qh91@!e5!K5Rv-MY<XA&V42Ajm-1$OQG9DFcw%SWAJGb^%^)P)<
zRRWy0N6}VdY!q3|G0_!<sLlXkk}IetjQ6t$T}YomAr;e;BNYU%H*kn*n``2$5B+&Z
zjBkT#*^07)hIAaaMTLJ?c%?mYJ3A};rdajru;6{?%p_Hu!jNR+FUcaio3T3$@(ZFW
zas@G%lZgy40*b2tchav28f*d@yyF4+^}RnBj-({sObMvbMN)n^>iHAPT}%51VA{$@
zgzW&!dr#+cDUA4gD6Wp1LAtlPmzy3$a;q}_*Rca~Gf)iB5@-K;MPx!>#)PlfjaQgz
zsIYxg-PzA$?!9C_Cjna=5<FMY&G!Elq@XS;y0mo-I1H<x01uFTO@mIVt7DA<NDx8o
zX!aobQ)0{bk2pAe0uR^)1fdhBc-bQ53O(0AsRf;zE7UMb@0)^--h2G`v5bC4yup<#
z?{Zn8D8YnXD4Q%6g&7EIimrXg-%hfJW8eJ|G1wooD8tI}!~|KmAw#WNaziOActmlD
zb{`MbGbB?V82c~HT!H4YK<fX6bUykR4B`5z?PQCoh{XjJi?Ho30Zb3+(}}@nGwT_{
zcH_!E6-=*Cl=&i2j(PFc-6XgS`f2>Ciiqn9NQ<A8A(ltGZKboI>^7)2HqIYc<w*T0
zTLNQ!b;>{kCdp^?EOsaoeo;V*pUE3Pst6L%p{krDQOTV5d>wf`!_PWXmf^6%!F@Rg
z7&V2jDF;_~g(IJk9@+uvM8_i>Ot_~MJOqqe_aVHDVW)BNw9ob^rV~n^5tra`!8K!`
zs~asUC<PZBwH*q7S$pXUjxMk!TC#k6^DNr^D$X~H$N=m4<fIK&&i*tfPfDPWGsp;e
z;^{qK>pQicFuI#J$`ep-mAjTr@B;S6mvegFygq~0P7WXHycPZ6@R%#?pcqu`2zP68
z_1^kC>XZvgW^QEr^U{)PV(XC|*gZ+)%8eUVD9QbZfq4UGc-yRhZ{_6dy}KeKaXD<E
z=CZf;9{vfYLtRM=_HBVb`w&<d^&K4PXkkaIct7Zz37en6<y{6$rmf7)-*)Z%39Kgq
zt3zq<17N9ee0~>Xt+Bffn;%T0h_60IDNAQug^tg{ScIq9?E|1mQ%=<QhZG6SmfG*t
z3B*s3TK7YBSbkhzU4wC$3#Wes6G;mz+kSG#f7F%Nhj)(tv7J2^g1Dt{IAYY{mtIzH
zE~YO<ejnMKQu(q!0>J?12z%CwPp*vDpg<YSZk)*y#z5_sH7GgCa_Eq%rL8UL9?gC+
zmcTN8%tfRdP^%P5j`0~)5!?N*tt>288CZ4Ph0odLPJ?2;qkdz?oFU3K8Urx8QnDON
zv)ji`ktCsi1V1Z>9K_SXP?8mHNgM{=D^<a>T7?P6I_^XpQZXyXTq%x}!$?jb<X~3H
z1+@_^)6DjN=}iY8Sa(|>o=8nx(vSKaHb=|yXmCAP*t`~`EXD)6{J&Ir3kwd7M(P^#
zZar7qW~riR5O-^9VfDVn1?JX--d<Vp&eg6LTN|O^#udqHE;ajq$a~MQrnapObO9nN
zBB&?|2&e&6x}fx;sIVxaA|M?_14x%%gS}D=(yJInDWdcqY@kvN21KeNhR{L@1QJ5-
z42Wm%bM`*^?Q@^!{<!&rhLEhRIp-Mfc*i@&NH*ra%GcnW;@9@II3e@ilXD@T@^iQW
zgFX$Kp$~!j9nHaEZf%7%%mM}=PR=w2wA^BJ5qjg$j%lC*Og?U5@g5AWyB7^4!_J^=
zIT>sfP||S!;>NovSt_KQ=q@kcl4~|#zaDLB_o`SZbuO&i!JqddNQL^YSFLsgVzd`v
zR(5{{<hC-bQaiV~A=IDtNM0!LY)SjXi-?odd#{ayj9|4Qz8c}r4$Z0a4L-ti$A&u`
zP^YAvO+<J~y<c795`F?kpKYm6yX+|WA34DE7jm^~yLQw&@I9=D)b=LYr=atq#n~In
zARORV?k}1idoY6}LoOfm>h5D!6|XPbTI=2Pxy-Bks%_cmFpwfsPK;BBb#^ek>SD)@
zLMPu;wzcsc`qlu{d^h@)_#%*FOR2d-bh%oWl~EeIn-6{GeduuK`zgL=72lf>3)C4a
zV;gFx0KPu6)aH=F5vDF%6N<e#T--?SXs-5bvi5T~i*CFBTBYoHK<`)g;Njt`w3&7`
zCD^0{jBzW_*>As;Kg+lX=F$Xcn<%d4DLv~;+xK8!(>=9s%?ar0H6U$))o~WgxXDz^
z{!v872g*vq?*RJC6H>e(x6O-$>Qf~Je*55XT|k~R_@N0+>RPybiH(CFYT#K95M$*R
z2mNo@Kg6Ahq{Z{+_}}GoEqz<;t64?Vvs<i8125E>Kai`v-s|ISuItSoCJwAK+fR@A
zM<Z|zl%pH7bghFjHJ<lBZADQk2uJq#KHC{%<4uSpttu=bad=c`E%fWY5u03W{PJ5x
zP4{1bJrS^`W;E>mn}W~oDx$VPC_b-`ERxZ%KSA9C0Trt1`mMP4ff-gmTsSKjM!}R~
zjXxP8uoy_F`O7M!`XALDQ3+jG2yrbC4)nKfFR`s;79d5Ew;a;=HxUIG0^=dblmTbl
z7-`_{`}jUU$1Nc>^Xks^luO2+3AP@3bvNPrD3I0&JcIbU!SL*&bI$s_LWI4K9UPi3
zpa~cPQe>^!iv3*d`FJ0L=Jh?VM3mFQf}h1;-O9@-RepP(@aMk>8>Ms23m$C)gEegS
ztQqPsaS)!HuiQP>@a1)7@XYhu(kb`}W&J3QzRY4*RLCK;Xky>uVf=I9A|JnyN$!ec
zCp`v_O*LrJi4Q*rI@d;sqq?W-@DG&NeK-ZWN|ymK0}*go*zv_`7{FZp9f4&5L<b5A
z#jaS;!ciowo1G<uQbf|A4mkIOxH#F=@PDMgZRKoj)$lz!9Dduuy9aTJCrtR(TgGXk
zaK^>e)D$oo1hHdf&35cm`ugK6ka23zg3$x@maUya2Guh!xC@<ROP>MFfg1oOw;!mB
zeQ;#9`<YPX07Y&SD!PZtq(PY?HHD4Ki1`BMRPMt7-fONsm49XdPHAsWx-gg(7jaFf
zYudD@U#7@{qpGRkweCAIxq~y)uM&kkK;5u@1q)~u=wSNPYL2Z32)UI&Y_vZ7DTS@(
z9ja&FP`&BpZBM*I_3OuGjF0Z-TEWW4=I>cphgdqUL-l}CvN(wc`XJ%;cipc@rl4l-
zMGSTrt7JS;=uwI-wt=?}A9Ju!)n<{}wPx_i;RDHAc~ZQ0T)W}2dIf8Vsi~<3iO_ry
z*m@v}cf6Nliw)fq*7`cWDLU<ZsJ&!ugQov@4D*Bc)g~)$?qA}{)=y^+-=xz7>*mVj
zzUN8STjXX5zvs3y4XtTVsU9`tmSO!QxkK~0YlWx5MzKV##XSI&VdqW4ADws-1A4^)
z2&b*RaqUVY_Ao66!gl9-`x%YAAi_BGB+R((j9SJAv@Sy06rZ=n-c4Fcs(Y!(D3=)<
zsifFsX7zpv-F4{R?f*zww}93<K*y;8QCvYuYAKjceB%btH=kVpj>BU`jXt19pD^8V
zwV1xbf&WZGD=5UC8_3@Caf@G0EpNLwt5&;!c!sCGiP?>}17I!|D+Hp%U4$+6(rnK@
z=I4Y!Ohc>uW5}y081kA7#fO2`R$o~^bxa>V^x*8V3b&0PJ{?;moj?(n<0&vlk@e!G
z=lsVwCy(Co@y@A5@QeOV5o61<5q3LVK!CW9K_V~zZ5CtWN2#p9Nl!AJ^sPkZytEO&
z9RRxX-0fDf;VFB!h?{;<KFr6j(pjtusc~tov+{V6U@wQPdnew^2APDMm#p2%FZH_w
zGV}Z~8NtQU2j+x@NmXwC_1tWyJnT*@?WNt-N!`|Xk?C2J6ZZxPg?T$Th*KicJ@Y_1
z?jhp=7;+^22SJU<4t^5QG2d|Xc3l~^)9ZU>^9t<;fsG*(?<IOT??Pd&*`~QFYKcEm
zPwUqop8=>I%yQar7c|fcE>G|K$d^|!qyAuDgypF%YK3+$Ossns91ClgKq@Gzp1AxM
zu8S)YV4(GLe$Xes7wi(9Mv1YS{QV=<)!+9hYBG3DTsPkh(y?Eg@f^&sly-A;^!I3g
zYCT0?B+{8lB+yy47$B4ErDuS682t_tZ6M}80Zi9Y<ubEQZxd(&PXK%zSK5vFLx+4H
zO-(r{FdD@*#7wo0>|GEb1@(5tvVR$dYVvc8I*L<--pw8gJ+{n)E?L3lt!J_n04<Ga
zI$@$oTv`DpiLoT+Q_WDk^-j3hH8Wf;o25`#old4GE~M>izjJcrG9kt9{giOkOzWNv
zE5<`Qi%Y^!y6i}UYeC7$Q82-c0+f4Ji~W!fDX=ac+><{mN_IW~r7%yjvRMW#IbX)S
zOLEUqcxAQv8A+WK;5U^RiF)yEz>sa+1IQH|bi^z(a$5dYH^<(>3ReBbwW|e>71~l)
zfVq4cQ2Si6e%lEXXHe~Cyi{P$87^7pxnN;}S|JjJ8n^48m!d3?K4zwV){NhRPIFhc
zHuiaw^yZBOKL6m3NmZ!*b8#a>M&bSlK$8}LQoG$701ETYf$EV~P(1?bS*VL#HkjIx
zsi2@x_`G5F@g@uIzCcKLm@3Z&3Jn}{fDyc41N!T(sY7|Du-bRQGF9{AUDu1<liH$!
zALLD`WEwOE%?)M0r^Fq4>UsOtQPTN^n>XgGKy-LS-E+wz@5p%mYIWFpAj9T&20oY-
z)IK_8QKbYKhgU+Ygw&}gK2$3vP4xy(;pf|4UB?>=7)bANOxjQXIFpc>8sb7Fv#(mQ
z25=;xctHv%VQhLsS1rHshd;jY5E39$Q9zYGxRC?8c|w`<@S(9}wSY+~NLGDxB>d<e
zcj(*3qtKnO^xoAx$BjT)dQsL9kXwBZXxjatGV=rIRNo5CYtD>mevT-7Iz=ZuDWKJ@
z6TI-vkJs~3)Qg^OxfCal^_imAl%ie@%KDyB%=j>T=;LDp<k`)B-r>$qyC#OJ{Gzsc
zTRryk5Mgs|O#dL4er@*hi_M!4pSLs_DNvW>T6NGRl)YR)x1f2d3iVMVx_>({I{Bk=
zZ<SG%tusDwro+f&l-pUcXwP(a^VA?pzm;5y;nNMgm@v9#yV`MFBXTAR!>zwZHKj?o
z7bLKSJ@0V3B<4#}5;~mi?`Y-&({h-k20iNl4RInKK76Q{)G_Q!>Y*Q9d=jUSQZ`!L
z*&{~0*y)a?Pxk~sb{5AWNUT*3_VcUgu0r?pZ(Awgu6CUv9HE##-7w;@>WfxpYeL*1
z7D#LO^%jc0N<8A-W%Moxa!j$wE0M4DKZFy4c_GLyWdjL<s?iDTFn{<FgI4zR+?QFK
zAq&ykEvCg#%l_>v#yD8gS$Lta)n}2MXlYo5l)?@rOX{Au<2kr(1*CR`XS3+x@Vx8b
zwJ_Q$%KNw#2MZLxhk!`H8220yOGpqkk9DOqu|TPlYH3qV(Be-^p{;8mR|Ua!4y_6)
z6HP2NjWE7>J+@ADRhY953$*m8vDgJEgzXOd@n7iEC2;va9?#gBU;mhiiB5gKTFSYO
z{Uk}K@4jSx$l>G3qZC?2WdRg+9r(KklJ)Y3j|)yVu|h)NE<VZHlbbbk#cwkdj|5-o
zoworlxTLdtnE6HGJHGI|ym&NIp7{mUe%!OwUOlXi55XO4ZHS2P@C2VC+X${=<J*Qy
z2#{jyW8RA^CQ?wL-W|Cd7nXF_gR8o@S+&@fUb7itjrQ(Q<%E>XqcB_gTU1v;{;Rf2
z>fPeE<p!@sBg#aG8Kvu3Az70qcrTn;dD&pEyneo9=U56Gq&5(sK}ca){8^&5mlyi%
zs3h3*$VsI?g)J;8IxB7GUeO)0hkxFaQ=IH!D`nYvmT!?IXr{LOC>B`*)vN$Co}+ne
zW`Shqcji~TcdjgN*R2A2C@K!zka5s2$@kv5fJeJ-Y4?~zHQI@f<V{Q-nwX@FgPYfS
z2V7)A1e{3ntq7{R425yXs|p@eY9V@muVUWnE&ZV|M>OJelO>e<ad_G{v17fw>KOQx
zM_bt;JJtm7{XaPJGFK(~Sr&Vk=}oz7G2pHluDVD&069XSpHq}>VhkBGQEYvpZr^&V
zgpNxXeRNZh0}CY1iRQHz3<SH1d;68c@bUiP?cfj52tI^0Z~FnLhBvShvlHF3hiHr*
z+0#h?Vmm9Ji*^!9nN(EgZjwlJLjzi}zBmZ<!88g;Z4Pi)`Y9E;mOX4gm|j!hl-T4?
zfcm~;h6pzdZr*s}8I4JcEyQGNl0l4tAo3x)%ZX2t4*z*qIwAKr_lv5F5F<=TnvbdX
zXX_9lHq%|ck#S^ve@L2gkIY@~gE&9E&l1E|0pYF7>!21ZLhLS|cV5tpBcFjIE3VnQ
z-z6KOL;c=N2;Wo4)$ZG}{4sV?cc7Y&{`aLdD;dhUD0bQ_iKZ!|#GA#$U69q~tgMLG
za{(&#`L2sgWgbhrm7&y0qWA@X8^#U#S?HL@I)JCXG1nR6q+)DhVqt<`4YqIBG00Kt
z>$Y^!g^p5sItG<X)$E#{C80xI=czG@{A)Ij#$)E(N-El!dsVjJ)%GiFDMuZcd)WVm
z+*q8qEWBai1!sCpU{%jidDSz>I<WjT3ie!KjT2uzY1tNxSB!Jmr_?Xw)dFhG)BfQY
zR5o(W5ydJhKCex}X5toU&|3v1L44@f<*8HLWga^KpedyoF8|cFu%&;7=MqppRUg0w
ziAf35$*o3=Bb20$4o|cA*TLX_nsk9~Vf(e)S0r?_-#a(g0oUtjf>&Wb(2+B_{c+9G
z^=JYIxwZG@0<fUs^WTgXFg+6yeyfJ-R)Nc#^`s@NZ1VB!YTEDd)VatImy33Dv*U!(
zk6VdXCIZ}T<_>JXVrf>{`+0%q<)p&#wB#Qjn$*HL4&RbnC$6w`P1k0wGi@H6_OGjQ
zN7`ET_-vT^z`Vv@O?mlf#OlaAxotfUsBsL2EgOL<WNov`U%oph@W5Ar?#YCNt;4<*
z*eM+pStmsz<XG%E#gIX5;qr%#bYc3SP6Yr`v{Z#BbAy>_p;W~On$1&XaMSXa1-m%G
zJpSYI9bvrS4jsP<jBL{4Q;gaf;1Np1;FtVR{2I-y`r8zZ;gwBON-#=QqZwN@aZ38s
ziY-eUvc}@*t;1pNTkLo3uglX9^qwz|BA>u(e|lPH-EW_!vOo(DN3i|ayDqSq;F*iH
z>N|*0n1U?S8%9MIVG<RM;Xu&wvn2hjYzsyG&(g#MbgB%ICaP}#W-j7MDktzTOAmuE
zK0Ru(v=N7QRi`LSaS0qk%CWcXd;E`0QVo|ev=4ZP{q1SGf#n$CX!7%6rLjvGGdS=T
zro>%K+91^){k@`l%!VWa^}T3Vcb#XQw@OL?(@|`%Ew#nHK)|ycGGV4e(}AjuBR3MZ
zQ5NkSc*kBoQ98~)C<8-yoZ3U!nw81123?(At~D+PqqY;}rQt8NOXh#rT7nfSzO`)8
zcrKl}1H511iDjz>*{0nkS5K^*mq8Zu9Q04cOfiq^yGw{STdwy2)1|%Y^zjoPX9HJ3
zuF;5hSHZ5m{Z(ZZRCDq)SD3RM5{^CBS>*+VWvVq3D&?1c#_C*S)e->R9Xk{Z9V)#b
zk5~M_1ZFJXW;-cfh`(k>&GP?U4*Xn!CI7h*NL?PBGf!>GcS&Me{bR8M)fu~*W@vP4
z=C!@Dll^o&b|4{R*SbYh0*vEHlNeJIpO?S_7{~)Ja#lsmsk(D?Y#O*Np`Sp5dGCbI
z+@XnXGeBQab_e$$uf5X=snal0y6kKLemH<W^58hrogDrD3}Ym0SRcRR=RL`}3Vd}u
z>#85OD@^JJ>*7D*H3db>S^(w~OkZlp3Q9Q(CF@(kQJZ3Z-S_!jX1s4YLb2F-rg_}5
zcBSh3d^Ka<H`^EYm|<%|Ru<?#an#>|CCupv07V9Mh(;z_V=n#Wgl+4#3j$-fxFxjj
zcw#Y^VYDnk>fbkHPQeQT_Asthv=J^h!EqqM_s_~-eB$sI_wwMXTXGk+t!UBvf!PH1
zvD%ej<k8g#;wv;2-IHC;BLZuR5Dn2=+PJMQCP?X)BZ!q?c|7t(V7c1+emu}P&%smo
zc=uHcv1)Z9;0p(EhdsBg2)j9j)z8m%F%lw8z28Kl-_t%atDH#<(j3@!7M);@KG)$(
z5`_A810z*LVGYZb0RI@cdmm~!^6HDne#6$S;J01LC+UKBcdr+Jd)zwo8RIdg9j30s
zw=U#JLL<Ty84}GtX&bNGpRmm8sTJ0tE!bT|Sb`cWJfj)#Cddn&KS9&ao~dj5vpY_I
zw-2(Bb~-Yx79(?yOx!mE{>suMZ)4OAnGMTJTdQCWmHKs6o8ZKsPQ+h7X|xdm>+m<Q
z8WjC=XDZv62t2B?8;YS48AQtJXVLxj3ZLNDl9&#xN<sR*ZioQ*Cd{erEKA=bPujq+
zn)Fi@AxCw2PBfUQa*4Sx8%&F!mOXN{&+^&BidS#<&|C#^Ip#wx!X=D8l}H-+s~0N^
zAmhe{zd@Q=3(r3?-)KAv@s1rotOK@3*y;&>eD+?H4K6pEDzuHF9)PEn+@UDK47Uet
zXLckT(?ilG*X@1RAo(uerNVC=G=7gTm+ZH4@n_fMTYw)PfS$TZkBE1`FWru2g+{!+
zcVpw%Cs|mY=zDq=0T-)Ju!w9Wl>|4myB{@BdhlZbpheTv@{L!ETaVh_->!q8mtpwy
zsacz7U3;8Naou2TL($}V^g{j%$>qOEKX3pi;5}zXZ_lSkU?Ox*mD;UXUeF?0)^rxb
z2q<bu=3AW!W0JAad4zTYLy9Kvg&$n=f{X!+{EsbNK{$1QF04!ChkHec>}a<*1##kO
zSoowm-ALNW>9U4!yl)+%yL;4^q8AwHw*~s-g7j6o;*s7;Y|ytLH?cuUQa^l*&^_`|
zf)ezM2oO`#mj3l{5TdF_l^Zh14o?oOLvYm!gbmyM#%VS*V)2%`gsD>vmX_Vu;Bo*b
z3KN*nJDQz|O_-4UI?u}t%;Sk^9_X_ej0UmVtzi+!Ao<XA<pK#8lPTTtG|;=|I-=^J
z#^F`7r`PxWwujkKvvJTAMrBcN%fY0Iml~aB<RlD5|IGc?ne^qAT3TIl9?2M{)#hpD
zNoalxIfTQa$%_40656}`i{SDr)3;anfE9H|d$;n_nsuoZiHHu$rwaSq|GBiY4*fGb
z(Gf6plAk662Nq$@;3u?(f0|)A_03_Xd+IOP9_MwS(HEqdX7{x0mXTfclLp6muP?@c
zz)NOj>1AiBd|?J7=F#k70|KqKC+Qvxi+Bt}DU_Z84p8gZFF|kU#L9mMz0Ud2cv1h%
zypIs&gD36I(n0+1VIeRVYD#Wtp(%BvTT0U0WI1K=i%V&ZI=}^h#4ncej;)FM!{#(w
zf+uOKFr~P5<LM30SB>wJEkM-gkM<}-HO8p~ub!<-KkIjDv!sWE0mCjJkORLg1uMRK
ziWP5r1Ezj(R{K665fL!4a^U6J`^yXS^MI0K$YT@3w)e<laC7DRYNJ>(N<VOTmV}KW
zKLy|N$D`$Z%F5<*=xS@=l+-P8l*p21y+G{HK31e?_Y<t;meO6s!{}IGVVD1azb%HG
z<`Zx}t?u)}A)ImWnU|}LA?yDomXT#Y#+I1O>evAQTNuVcmrBUWoC$F3D!~L^scD9(
zM_}L%-f_K!3@ZK!7R-6KJIs_O05d!*SxVO-HzS-tQ!?fzL*$%iNjEujXZw}(EjvLB
zn#+5zFO&!3f3|)n1a*nXTSj@qxJ9IS_5k|K9c?B8-G1fSuHEN@)mMZS!{~39hJ1FG
z0R$y`_!L_ofBYYvs4{8ok~94eANp*i7Tf+aYtk(dAq5x*B2{P0yxx>&7;)pl8)!fk
z0Lq{{fRMsh1)!q%)63B3v9jinF>PlOJLE5E$Op~n29{tte1wdaV@xP)C2)%I(PRx+
zQPKbC@;W_RXDH1&zq5RPoS`T>de0k7y}bC;hfU<k|6)(YI3b<dQ4YkmSn_E&4xt{n
zO(5*E96Z(S=9k9=9(Ks<+N(HJQks~6WSqyoONMBNMo0#7{MD(W)QvJ82KpeO;a_1&
z!_AP8P>Hb9c(RrR;1q_|EBuzxymN9|hQWC5$74|FmBmn1{tu+{_lvX)bY7P{UO3o8
z+_qw4`0@z|D`E!U(Z*1fC611^*+Hc7tH~_i_0pr@S`oxe=ukr(h{d8Y4!7S}EdTP8
zH#loWmf)g<W64n%lCV3P+>REiy5^;p`9iJcBVcwOCv*o9aSwO;b^EmQRZc-QNQwFj
zX$e*PYiD-`&5W2YIwe<N9Zy+U1QsMzxlzWt4GPwz`}`%jxX&D#rJ)xFOsJ_HeFVtW
z!vNRKTq+F3|2j)pBl72JA=W^Uh4`|s;~manalHR;1S2qd!3H?@{r(#1MGca=I7fvu
zYos98O+T;=&%tAN?hcEDnXdVXZ^j+aV@~72TFwa_2UmRw_yrFHD1${;2N@3}Bci?E
ztG-3Z|DH)yenM6LO7Z7-5xyGj7W<mLYTKU>=t@Kn3qsu*cv^0MKtC-#?6Ms%aD`1L
zs}4?<v-r8$ZC+j@i-*MZkUZLEC$Rus9n97F0No?2G1JfG!1BoSf8i;?$pz#V0g>zQ
zp}JimnLQr)YmoUDVFa*czWUUs6T?QjRBm+jE@GW1u-X;g0bkmaM7PBh>YR`!c8bys
zFPn!wLH-UN*$mmpxnU7R0Yr$~JjETRI^#L%3%FZ!#$^mcjHoIa5)SSr<93476`oY_
z{~l<Z1{F<ABDCi9TSS6%GZBjuH;G(%%DSjq48Sf_cHl{<xO(v<7PUpVBL5f}x63~i
zfn2r%QsnO_J*Dj$c~s7TH=C%T?p=NlQE#!nT@-7x!B=hC%Zs{5HK)G=TnK-OZRx?@
zNsru2>-!|@{u#&1=F1E0SOkoi@YWL5r!5q|{uy(#LE5)cdh8!PBGcS#DY>54QliG%
z*9l&C#WTp6c}Go6@E`IejU3F28=0}<uKz1&q1BC+=#{&h^nNW>f0So6;#X!j&V>HI
zfn2^<vbX3}*|=vWY?CYK2SL7FBJ*@>!zlGo`hHj#H*gpwbEl-kv$qwYe$?lFpLi{8
zzT`L2%pFA4A>4z0Y?{;))lk>+`E-V=2_Kt~srA|@jEp{YVCilCCOXl(Rj!t@2XN;<
z$=rA)qq}4hA|;yqaTAFH#%7god3GV_5N=9QdFMO`O{)GgH0k;r#8T51Lzw&9*u)rF
zGV}q;iy>QQWwIU&I<~x?&o-}txCWJik#arDPt~k+y1=#kA4qmP*uei2VNx({z(U>+
zD@NM(N;r96rL+&H3aANv+7khnR{#zrhAc#q1xcTuG)i3IU<wDc-|`0?ptg&c(&P!*
zwK4-nx!Rp*;KHl?grIm+FleG{C-Kq5@V7lAwo;bg!6vD!HP^l8;uea((ybIVz&rtj
zyydFxYDWJZsJsLdbD1OnWS5m}Y^|j7fAoYCL6w)GQr|xlE^YGcEqFD<X?`dU5t9F6
z>ZgAkzL!%~zt8d>`8bU2?Nk1FI#mEnneFtXO>?DgbbR<f#_RL*r6HFvd9LG?S_0dm
z6?#Dox0qb`8Ne}<4%~9Jbo4qvq9&I>7EjuTM->WoB7M7<Bv=8k0t|DUdxR|(-jH3$
z_B$BF0CyS&8*m2<qEoef6)mzYYmmk1fd@waD<Q}Z^S$xmfy8QE(V>BrA-X;X>;gc9
zJI}x0C4_L2q3HR0uC76rrLdiuUrX5Fc~I~}Z^$4vmCI82XLiULP&Hv<nFNrGq6nx7
zFK_=-nanS0jML};({p?*HhCgMB-}t;dFN*&_74cL{j1r-GW7ui#X=IPS)u$-NJYb=
zEQP6F95CgMZEx8{U}ewzj^B!e;_s6A;j|d?fb%Dz81l7==fdtvoU%nA5}SS-|642a
z-d#7}$Bm-$?ag=9Oe#&|w@!s|+np7khnLb!O51&hCtuA+W(%zz-vbNtWBrccA|pyr
z+jk3NQAcs@O5d(>17x-IXgkPa_j0}x@Eo_%rVAJ5oa;!8u`K#V*j<>Sj*@3Rv*XV;
zjR{cr!aHrF$z_{JsFS$NTDrvbyd!Y01aUpg(8e`#A(xL<Iv$|25&jZV)xlm4M0XZe
zb*~n>S_EXh)J<^p|D7Jotx|+s)c;$uX^9_GaKn0BtsdCandW=7+AqBG0r>bw-M_^C
zYXdm`%5S|b?e@G{jonhxTz;@u9Job=c=q`7O9t?2(F}W-)Se&F+39nfkn6^!$gBl$
zY!^L)X4iz&?}VyH*Mt~tDJj{%_?N{ZMD$+~^V8){oX(@1kmd1zxu-1*-RZ0bF=Dkz
zh-xhfWn)|RiNsJHt5;MJ5qqhO@1^`qIGW}qut2%!-=WSj9t85I59|Qcj1=RQx-Rwg
zEWIS#hO^Hzi<7;c#`q>i__w{l8ghBaNk3M2D9?(!ufnVn8yIvFXXbkTg#Z2JE%PIm
z=Z8+H%XG>k=s82XLt;(IRUo#o{t_>vOTCWP==AuBU7y_q*l;z|L@oJ9S=K)OlmD3Q
z?z^`Br(?AO3;i#N0<=HrC-05kT*Zu1ZZBmO(Hx>xt;*{lm-2r_4|x<oq_t)-k!!Zi
z2wx4?-U1Q$!G{|RBNoT(<)q6W(#TqX-7&!R>ITMOG8O!7(tVgG`qJm$i}-X*n%@;x
z((hy9bj#n#^NOu=hOF3rFBvb;2F1&2N)hF$6^9{Xy2N+eC^A>*2iXIEjwTe8*;|@*
z1JOf8(1I|`9o~yI$E}9hV3a74Lv-S{#`4-dI=O==tQ&Y4BQqqSvvNKHLq~6UkC(o)
z{QGB{k3b6P@76=vhhgC;az7d=3itp;*n*#es=#QX$;vPz6XW1&x7jZ1pD$vVQ}JE)
zu$Uugp(rvNx)@dwA%@})U#f$tyJ1urY#!zWV!~gDC$Ro(|9*sh1U06P!>Ap^42&g$
zZhU=h%^~^4Row;BB#`)N6HX|112f0_W((j)93S$qf}TAsVU7GOBKSAg2-!9p`?lk~
zUgQPOGM*R$_Fm!d$d<=}ugU+SzBPpgKw@p<h#2grPc^}F%bUrpm=Az8WuA!dDR+Ib
zqo%#)_hbUs4u^?<l*%H^qHYhBG0cES072A*%k`5~Sy`el{zNiMk-jrjc?u^XT4i86
zXQ!EEL#BkI7YuN50`pcs*HRDcS;qaq_i1<%9gWd6rPrf={%$oOds6-MRU`pD?x#x)
znHU!?l{PWiCRIP_4N`-P{3ehPZ942(+)E-YYFwD|p?t|^-ou`6nQ?}0tM{X<EKt9-
z@h9#Ng^li!55n$oWRJ^)m7l-=lKK3r_yO!>O93^$mLGQa7o?am#8P^IiFbx_e#g6p
zJ-Kic4gX(}CP*$_+<kD_ul@f=J5=*AAQ8g{2j0hsD>=OQt}6Zxk@A1f-x!}s!t@@a
zKg{h~2;XwEMeNUjT_uUEtOy6@VN_3Y+tbzY>=)_i&?C5Zy}$vnbIZ5fuD%m=2v0jk
zA7@k!?GCfqL$nl~MufmPZoR;TKfUY@tJX2~#z@4ryW}UELfBwN@H`ct6vZCD=6~7T
zu18qxj^ajUeHd%#SWP$E{MEoldH&Ky{e{4*%@UfW9x?vt!q&9fK=c>sxKr*db|6V5
z>WaM<BBMv0pPf!I8$=OH=~X^RmSVw0^E6$)6=XyIPRDdo)0mm{Ul$7ay&2&c@&mMr
z^9VzxmWDcoGd8Vcgl7p~NW>8KrEZ8GOdcZ9?)gQ*ixf!v2|KnhDUfdcB^#NdW{b-h
zlKEnfUZ+-Dy$$a0?-j*kXHcYB_h@pxKq7DSl|P!Uc<APkOy6m|Z7tANw3*CQnc$Lq
z=I%2t$o1_zqfxT|PcTF5?S5D@ubz9RSHx*%4^87RE)<rDB9}I<x_<Ld=g}bD0q7}!
zQ0h3aiE?HFrWue2S5}saMeG0(D6|&!2ItJjByN3m!S?nJFr`atS6<!*#0tA_eT`bo
zVm=}on3y=3n1}=0=K=B?EbuD=EgraRCZ3Td0`sbe;}lpBw98igWk7)%{|{wfqBJB)
zGlrY*0*{v}NsJV&(#N&;jr7_sYXMwa2rQPjmb8*Nkt~0$l@u_6qA<Hanmu(7z|mE@
zxa*=-M;qweYGQxutEjyvIhpbc?VnK7V-q6e_7KNBRsy&e>VE68XX8FuIxO)Ros*I4
z>_Wn3eorD2;Jc!b&lytclqS-%uvhS%X&~d8El4>C3vVS+;?UIrucFVO{~SJ?BJpZ%
z1CgqC+|ZX(nqovZJZGH7E~gTV0nX4&X~qblu8-q}1FvByy40!(=Vf=*HyC7MW~*^O
z5K*HSX{_Fkg{d$0HOWg5xyscp*zIgb)SUgd4Bv1_--34k4A>EFCcPlv(8rkNCj{u?
zBK4_(Fv-SYvt9j@f2<FJT3xEx!&K{BLt_@sca4?RPSInAidep?OhJ~7zYIR+zjFO6
zP2WcK+zs`~<&;3<A3Q7i_t5J}D3=rhhehKrRG>^llGi`ukmnI08)}PQs8?y`E=>G8
z;l)YygMlv=y*9`XVHTA@(y1cE)A-x6^G`7Z-66feSWHP|FCzmZ9?1Us^_k@qZZ_Yi
zm0_murh=PP*&zS-Q^9TlXLmt`AHZNa!&dKFRSrl|l6AzTWa#?5@72-caP!pdYpa1E
zW$Cf#D($L_A^600qBElED2`A^0x~bYIue#qo3Z?|u<4<$z6qdb+SorT2HRGDIrMYk
zuXVp&6^8ep<yOG_k0~N3(d(ZWY$i?j|2svlO93a;iBNy>#p|NeKe}s){ydsHi2?wg
z;!5DeikA-a&oI~+@(m2goaSSC?-piWY~}zS?7z!VN<XmH+!hVdJ7J3hIRG{$uL2|;
zm4|UM`hi1P&2(S543k#&Vr@;V{cr2XYZ~hj9riAaQ@Djqpmik9U5~m6&%Iza7Ubgf
z;ME5ZXwUr&w3n5=c|s<u76b0T<$>1;<x$(;)pL)dj!wT?Kw+rcfr-LMYmU%m&~7$9
z5qZ(%`<{#kS?w!p+H{t;J@a%ZQ8lbXXBn3y$XK7lj5;7J@(?wOD^%zqIZnyRUHo&D
zm*dA37F0V*f~f`I^}cV0ZI98T7<$r=r8n#1z5q<Hd~C9r_N3C2Hj_<eOy6p!?`!fw
zc7?X86hM|o|CfBxuoNUI^PWP?CZkTj8P;i2YImp`jc!k1|8(ou`t7%>|E@n{T2YR=
z(WY#m`7k0*zM|Jne)J5IqOs=wm1^aWH3Q&6D*d`bd+kT2c$q;*4L`wR;hb%CljQ$t
zFTANg$#~z32x;d^c4)sfHSfVzCOg;kWTDz9fFt27GWM)_aW^n=i*|3c3bGsiO9Low
zqN#IM?+}L~OuYg$)u_vu3ek}1eL~;_L}8l4BW-a5KJ3qmKvluggP9y?a*HNEmgZwI
zz;!<TQcB4V?ZZx%lDgCH#H;-oAGcMk3`1}s=;|=aBc-mp=&Ki25@d3FF0d&tzhBLs
z6alq`ezc7wF&m+7HZ4PIpLD|CUWK*Bq2kCbTu7)Gbd-ol?5K*9wpO{&ZtWNH0;q&s
zivOn?k#0v8yAB6YddI#o+P7;O8$Z4O+swmnREoHrT_XYsy#9sLTQbIfoAUX$(ubOp
zoZ#?1KRv9P88|)mPbx;*NXl8**8~5-Xg}J_vhv5>SG%TQSW>rz0#m;u;8X`B0C)=c
zW+%;!2WM|2gos=jA4^C`@MG5Hq&aGDb`d0f!8G43EdA<EQv2jjKy>dWmWWn0D042o
zod_IQr`qFL|B734zBGmuLe#e{_a+a1&~YPofmXMSse`LDC5U~~f_w+x#>K`i4Ak2$
zOlnqtSQ{OGj!Rs$>L891VQaswQU{uk!Q}Zi0gcwLSN=^=Bovl=?$7yq(vs2)C`~)v
zDn?z=LZ$SXPFqq6$LRR@D{d6X{(BAtDS|33OT;#7l5(p;NJawP43;r(o<RCKpb@Bh
zkBAJx@FTr*-M#egw9jkIp9O3nG~1%dH#hx`Jfk7t-zouyV;&J*ku#@rr!La;kiL|L
zh6bH6EMvh(T_O=9`N^sWBpu9_?8f@mGk?Y685AiPrnw?R;DYvNH0^#@sm_C|NJg#R
z&4tJp-+`9U;J4H@O!RbPH~G6FcFq$yV*@K%pNlg}>c|HAl3vd3-pq+A;$1<cDx*^W
ziLZ@vwO~w-6tUGzdB0qNreQ1jW@{|F#{M9#FxUR(;8W9$l3^01{Sm66vuwh%HQ4gu
zrxieM1HzM$o+_Hzq?!?_a;oX}qI??67`i(Uuz*nyngL?X^q5py7eNEmIi;ENvJZbf
zJ=ArfUdla+jLG(9%!Ap&j7DyH3wKC>Pu*-4w?38PlvusJcD8l<m9+Ag%EbvC?z+@s
zr}wPFhHD=r-u14)q@acFl0)&EKA*dZG1H@JBGi56ZSHwn;EeQ#6r)&L`;#%MqV!@+
zoH8KO79R*0yyu1K9W}_ASWuNuO5|fX+DcA9&fJxlPUzrf&XIO6Pc>lf+mP*7N~{Nh
zo3W&7T8}1|c^eUfG1H~?OgQJf3O~vM3Rcoi(0Ldo2>L|~kb2>7h4kzo#`(;BVN4Q2
zD==2L<}SOX6F1H%@ME05U^SgW(#$N^@vTSiD~7g*ZaH*mufN&@&^K_6;MwgP6tpgd
zD^OW(>|6h0ebS+_?SAh2J49GJs?;i2F1Xv}8HT;yReZAch}>PS<Gia_3bg9fiF*Q9
zCWx%s!YN?iF`0L_yqU7_LEW*7-|2{YuClr@seP)!81IvHGz1Yo?9^T|BnWAFAZI!q
z9I0>1dOYGVpb^c83H`-Ts}jMtIaYF&g9>{Xa0r!{vDKd!tZ)K|+?c}L=cBvvw^s<a
zx291Ud1zmQog6=`$cl5wqqymf3hEQDHyX_!$fm0hdh)2+pBh(wRkCRzn)8@uvxl*x
zXToXo2Vk@8g(Ec%0-YM7MD6hDB!m>&J9p?6OD-C4Z@o^1C<ELZDJ0AHWMG)Y;S1@f
zB%8RwjykA|K5JNJrNJ!iRP){pt-62WiKHR=n_Zbz5045bbcF|WIx}olGHkWJ4|Kmp
zpg5cWmp=}uNC+W=Z5fn>59Q$Ca7NDb$SY8M0ykg54<TyYV-@nAR|C4HwuoMZNfLF9
zA+Log`hnLm7RJK8ao#Z)+R3!5RKqO3HPFZ2TV+|tcICv9V_Hr_c8?M|ju5)agEfF`
z++$D;l5MS|IE&^+W(=~xe(Vu!ko?n<62AmWXf!4Ut+7FcyGb&l-*#U?vw}ON$F>Lz
z{>n?Ux5e5c8TJ|b30)5_U%nifqK^XoOlUfAUgG0CFd8>mEp&c%VxqHAQt?H#J7}xj
z8@sw#MnF)o(tW<zO1j+{nK^JHW3UjL<`vUS;BGtLSYjrMcP~ff2H1<AJ8&wunq~ZZ
z?c`MAtnf6wzfCiz)qNpS8FZ`HLhmmI86GeGJf|lXxyAMwZcu!<RqF<5@0%b}tqhBZ
zbFx>kaI)y++`O3ys+8iSd7L+8%F+c{?37WB@JZ;n<wcZ;=nu$>JT+Mun(W0;c^cIc
zoGRZfp)O@49j7|#*BZpgkQZc(dcD)Q+g}&Su*(?jq187%tFq4*$uZ@GWSk}4Rkdl@
zNK0XrR~q)~_Q&Qq%5lmP;ya#Uj>9?!f%;qPn&`VA%?*J6xzd-cD)}zk6DS08Ze|QV
z#@K&V`84sgYUcBG@AL7bG`RxTjU%B)6uUgM<1l8a=i=H^^-=iqMGnjl^-7(e5uEQj
zl-UJl!8~XkoxLg(eA*{tw5jZ=c@#TKkp;9RX_j(dzED%njXnWW152L=AGvKFrkt6V
zN1}fC@Ih3<G-vkopd`d|5QSod<Ihi{7ZyUVRY6fO7=!cXd{KERcqJRyby};`)z$03
zm~v?WA)#t9@6879oD}gXUNSue4$b!n^H!xEU(5-Ny#<Yg4ZG5sZPh=_fF@qeY5JOk
zm&%mn)%#Js%6`;SYJ|ZT9Kx03YYD-#YdclNIA5OkSr^8EuoiRb*!tOUx+C34K+&1C
z`Ml&DI}3lZmkn;`eJ@B$msvc&S%nR{^4f<-t8JQ07EBA4@xEk{JL6YoHD;7lLby<d
zM^`JvXj~}kGI4PBYnp!TS=lG}on=e1UFQSlr1<sl=Vzy1-2|O%sw#<sq<iG1NA}g{
zk=V|z<GH3m=>bUz^c1f#Rm09|%{P?}Cb_1z_v%wZv+)hp`#vnB++#ltZs%-o7SLr$
z6SW;a5fY3?oYS967J+U;iw`Z4x!?s$r0<F8i3wD#vSRvfpzN!1%HMXr$O4P05bu-$
zy9!zS2prpPg_#w>_H&Jo`b`r$RFfj+#z0rk)Z*R*FynW?{ZYQH1o#$LwD$y%qK`IT
zdS<bWbuhLz-*%%}>!b5!7-xZDheV?v=1J>1zo1|aOm&&^C3Lkt5flbZopIE6GxY-(
zqQ`c^J+wH3!tMs-0KKmEm>wx1owxu7?1ai>o6GLY-3+bMRr!3XSDeU*0ruc#ZET|M
zQQXRuto#5t$g+X8LD?zM%?SNX>-ni<VWu4d^#q9*%7Pso9S3vz!&qyqOrD&K6@xE4
zvyygS#WZA~<xkurr{qV(r!C+EkiJtZlG6NWqa`w_BcFENvca@o-;ihUZ1|qgp@@vZ
zeH&ey2QJvAy`M;Vl$|}iG133=wC1}Dn+*(sMU6(=w4Tz#q0r=WFlzr*d+cuF72xH<
z4OdJGo1}GY)9PLavt2!S+GkJ%;<=RI>fMxEPncqhCUHStk(AI}%=B^Is(p*blxeFH
zT=y2F+%K>@4~8{RVr~|jW#i)veO+=*Rb5Mk@FYcz@Pz3e?^Z)T!?v^p*JQh?IABrJ
zl1DxrEpYFxX6$gv&}^Q#%E_0yf9`^5zTJ?86r?-7r;AgRPFEK48z`Mhy+_^*r==cQ
zXV)YrFAn3+<FA5hp^K%DM4_{{C_Uk}tXgeKgjo%xdaIz(BJPBas>aU7k#J+Bc`m~=
z=cY0leg5-3o;{TTAr8ZXO{%0yEgUL7otT#y^}s$pfoUJp^i*6g3|up0V`F<@UVq6;
zF@-e@LD$9Q>6V|w@x#KYFXgzOXk2MgMM%Xed>B5U1<d&07@QpHu_vz<5A|{0Z6+aw
ztDasi(B(ax2EO<Af^Dt-tMbdBac*F?_Yu0Pzk_3ur*4UgRGWM|u+f#$BdpfxneuI>
z)7@6EZXymrkGuJ~aikW^F+y#3Hg8EJZiZ^rB#F-GYPqW+Y4@q9k7;P{N<#w2_j_QZ
z{IHIWP8_pS=k>+Ur_<?*rrCIXDvd^3!`wlG!^0Bbi8sLf2rq7_K_XGP2aosZ7#|<s
zcfX)7SuozN^D98UuEd0OrPFPeX)?r@j~ZzL^*7x>Gb78eUANi{pUsJbfdKRu9TCGT
zn5NTAu#Qm&#%Q7>iE&NOK#zwbM?&TG&1Kc+T3DduCS`)C`UVva=t|*v{GRU|tWi#&
zpzo};kr6j{M3crQ{Pp=Ue(ZD8Y<dQ!`KWhT0{zTXF)%&NXwatP5_xu_HI$3$R!)3$
z8}v$Dm_0;SChSizpHfx{45rrdsA7kbX;%_bJaK2`D*`}MW%u67fP&_On+tU+dLR#U
ztX{bl&db}azo0R$@bSct%1gV=OXu2P%2z(9)qGu0$ka99Y=mPm7cX8sPUvb+>IBC4
zY5d~{V6RNE$_E{un%Q+rOj5)9l2U{%f87LDg)oz?VMR@tUjTc)H*gwnW>yfzJ}TeM
z8Wv6Vg}EPV2-lP#y2XhvYG<DCvTEF?#=EI?|40ZloM|en+UOGMWuL>ybaHW)UKc!V
zO$uy(=b6_3y&qU7>dhwl;DgASqlF<aF0oURl|8qV0GR6qns@nKPuL`y;(Df^^zA}f
z9CJ(!N`){YNo%|*>4+t+Ww~b-VwGo`n(q1>gGCpb-~>!i(;PLt51q=p8}>&=M7Wht
zzL!W_1tf7u3pB#HVT9g8>$L)HU%#|~4hWA-_QX6avKSSUr}0ePhu&s@R*!MEp>tB`
zvFfk4Bo^VwJ2SCQzQ@6(qlB>&V56dnF}Lg7y_&i;778c}QC^w&bt^^bPE!Eo9O`GF
zLhbDAyw{XD91^UXs-&Psl4-AC-;nnWzEbA--!>A4VcOC6m!=$4+@Y>?V?~CM9+eXY
z{sT4qv2`B1gJ_!48KnhbwxL(9x%zUb<p3nILcVh_*)pz~z`=C(Jj>`QcX7*CRaMnD
zxI`z6u1#SlAU!<BGtt%R2U@ce6;Ed4C6Y=4NLYmhFW$qi8fo$741xxwY;I$6f8pv&
zDPCzBmOTl)brbcCBSFz!9u%~1bDAb<0yqyhrAS2vaHmaqroGT9Y2vo<X*4E#n!H*c
zWWF&NASKIvKA*I?nf+|jCudLm`IxU;!*VVd>O*gNNmX7=?}D0vUll7ejtutNmYkyT
zkgEMs(;nHbTVCdbZQmz6`z_kYv=${KD5$o+ZbCDu!^Dtgy`3nasrn^T!w0A*6e)4_
z1J(Xb#Lkgx0&astV`B@bh7l_4OFva)L4QzEIUvRDu$>cmB-<UTta_dUTjXW-&c^{f
z<R0z)cw+7dUYy1*%Z}WiLq+NQ1w_iB8vA9cPp^W^c@BR#j@OCrQ&Hkf=(xCgKjZUQ
zcSREEu$v22%^RrT)$zgUl!dm{O%sIykaYQA#?4v=<4aQS{CApoeUL_XC~fq%OxZVd
z2Bx6fgK@`eEFekIJ(Mz6`K1Rom-PaMeO}Lg4u&<=UuoFjVrwr{V|tw?thBz9r>Cto
zRaDB)zPYtY6RaJuRU{Jjxkawc%hGlzKPxXU?^LcSA7y$vL}bzLWM&fl&CJ@*y?XVk
z{Yuj%ul6m)ghQ?0pGrI}{Yt3S=we=psJip~x~8&bgO6#H1!3_|U20v;%q@9-vXId1
zTGBmpDO%j9#~DfASfomL54Qa+U}xQdX{eNuFqbU=S4&oiwM{&W1t~j}Jv{`yx!7Mr
z#5mb&;k`tjOK8DLXzw|EmPwshFDLZ<DRb_ESCjdmAjG3*WYj$I;>8P#K|v)yrlprx
z9Yux~S%ge~q5?x{X6P|^NRe9PO~kopFGNlFOY|#A-wBZ|Xl$%Y=vdf&-K{_=BF!?_
zbR!&R&|)QDFf0-CKFhVUI79-N)hTdYvyTQkf{)?)(d7s_VRQ43aLO*$qiHn7uCw{t
zyj+28YTvW78P)vJ&8en><2|<9=`Qn(xc&RuK`W{g8gUJylc1EvtvVsZdZXG}S0p8n
z6z2*6(zWgiKcj)3q2za$WWfjBY5YbXDi|}NU@*OIIgI-lLgCM<mI5uX``)ievZcL|
z36<+TB`nx$hor1qw@$<N(SlgugBs-$8?<!g_)7^fM_T8RFLt%eFfR>F?pmE#?J@C&
zkK)>0KFwj?u=Gjcb+n^~^Fz7?*tRF{3`p^{0T&H`*dhpx;JEc>N3;XmTMPnjyH`J;
z4IcH{%(K5I<~<GpuZyi8;95NWpp_j?A-q+gdy@*M#Z)Xf7^me0oIVS7gSjNwXP_a>
zZGfq_^n$3sHKTU|1Sy*PfZ=jQV*q?=?Ls0IOL&4lyOj;Znmv9}x*cvR({3tj=U%(^
z4pNTurPK70`AmWO<e|J?E46nE`{4HdZMJ)e*PN0C-d_4A_9^#v#$fbD7fW!sO_{-+
zu$rZ94=u<9TQ@6@EZ!5*f5~gCaDw4n^x291asoJ-48<4UL9wCdBSPUIMsigqK53ky
zSN`OA^@-_sp_<Z8*%-c*FY()9RLiN}?Zjbpn_eLD_SP`v`i5kLNwx_eK$$DPCdNWh
zyL*BZLsvuQ{mK2bzAlgSfXiNPQ-_R3Z2iAn)pi(u?D>ky2@sS^O+%tM??Mxp5gK~o
z>FG=mxdG<ghS(b4(YX>JxQ6qauk1_79QV_u(rN?E`llMvJt4978oP))!>1)gK~ma@
z@q*3V0r?7j#OFK+gx)5eh9`Rzv3u7-34alqDRqWe;%+{2&AUeyFzg(w_@3w-Vc3@7
zH8uk*(cFs8qd`-4V(}f8VxNLt&Yt~N`0{Wp2nvb3&{4g>6kvwOlD^7`5&cA~VBD)#
zB}LiH>A$rqxZYk=`?nyrny%LNcC#GQ?AtUdRb!D4XlZTrssxZZ|I3#z&IcV14D%a}
z)@@Xfa-{~hO)%;b98aaSij{(Z-0(TpH;A^P)9rIrP)XVZA|-~MWk9@lAjbeF6`uAc
zi^1U+yrzzxJEYC1w8A-?r1ABMwgTY4wMkiqc(+9v62SnQVMtxU12vrcfr*+=j>+AB
zrA(Mf>@X)3dNvsnpOEz)PIkvteiIcp5f1-maP-jekHv~QiQ*C((Ml0~2_5G)Zo>i?
z?N=yaRA7<ot25=H0|Kfqh9!_}k9K(4lPs8O(f`9j)CjiX@zeN?;)yyeDvel!hk4+*
z(?HZlPr(#aZQW3!@*+2(0~~3esYXRGM#{A_ve`F>vOqX$XgyJ=fdX2`BDBMvj9r0c
zLsJF5XE$GI2SEJCW4>WuyOoy<n7FYCj!QcocOC2z;pFouV1W}mwsJxM^bis~lW!ie
zzX31~1%MeoeU&r4`7=NJ0wDR62sWp(@xI|QYm^AFhW(YUc-Wjrh#g92`+pC?JQ|Hg
zFA%nV7EAm@ajCvgTnl39yi|LXW`A;p!`C>-`#xqgJOlyrf&O<S#nFOF{8TCjxTO~j
z9X8lOgyz4DjTY@s53LB@rgPl8O)i?n?&;9zs5KIq)ZpSUC>gIo=<2xD;g_m!8>2Ts
zL#L`}dtGSU*IuZ0y{WIS&kqsu{LL%dk3#{?mjN8A0fhgyw~MD1-^$^_7Utsd15F5E
z0hQsIxl)#(qQr4>Pt5w1FORCc#*zrQcHpV2Fse;0^H03)>_47r>(dlcbyYjgfYRmq
z5!}Qr+7JtP&uhQ}MtdWeKJ@G<bR}iy{mX!UXexV!uH<jm1}Ajat%$bsMSv(#5?65h
z%=a1J3lOfN{^zcz3$=hj3T|jvc-2!CWV`L~2g`;Ld{Y@8Jc*onu<KT}D-!Ofcm+SQ
zRTmDh_yC(jXPed02DpWm11bp#hu@s>JkIouC!U-fS_gVC9o!S6*KO|(T5oI)zjH_O
z4(MmA!VzXS^Eg!a?)LQ;&HtsIc?U}T*sei8_F35GuHdOgePFzAS5bgH50F=tCI;!p
zO3sae`RLYMDctGH8xk_ML%NZAs`Ab^aF44Hx;PsQ?59j4rG)+aF4lcjT5urIbC2_O
zt0*bWo_gE$R=PI&wLygrZVwE)=@?!t@Xi5DVp7L<K;;;k*#r!1ybr{+2Kwl;;R#i}
zL5w5`v!`l;D%?;*q29y%LeuQ-vq<3~fLbKHK{<&N@I}(t`AJJecgUMfX2A$=AuTl6
z@+CdWl-|aX_s?Bz#@>l%k1a&^Y>PWtue?e6?VLm7ZI_8sZPXrObE^N>o7~o*@X)By
zf4svv5>uiZXlcJqs*ejwmmiJtHo1T7loV0o{wr`r@&(~dWj=Gkp;{XNTawUUWh>zn
z!h4uoHv?#FyaI!Q;sQuO%DVd|D|G2IAe7#Kxidf$Wh!rs@0y?q``HQwc=sskgnpjB
zoR#dZ@m=|<g&*UgxLKv$>FK!d1o~|Roj5-?mDgF}M=Az<Z?M;Q2JhP}-_4MQjFXLh
z$aM_q=$Ckc9+h0$Jp4JHBfhAi0X|V=F(9=KG9UQiZwSfJlvGfXXmG*4|M7ST&&6)v
zPFEQvFw~bII_=+jlcd?>Q>WQwrZoErPEqRg)C}C~H=;usSlE4GtX?Lh)>tN>L*B@(
zLS#~tluR1}J20;kbU5v8Xc1=F%2)I}&j<a~ZC*0|tK3xGHZLE60;nJ{flHr1<`;Wi
zPettM<yE^P`s=gsLuScd>JK!^Nl%5FP^1OYrga9sQz$ieWN+~JrT5|_>r&cabeT4~
zJlQJ<Y@bMijIxIwP?tcTF@H<>ZK$nU8D;V-zVB?Sp<=i9UBJPplP1oSQ$iTUeb{ZG
z2vi|{Ayvf*Z@Dm<q0X3bG?RppGjV|H?Z4=l>^CaN>X?0W6EEP@?8o91KGm+RN$yLY
zLJO@1h06ymiq7Mow`_$(5TI~5|0xqN53YUt6iinMdD{?}mQ_x3SIzHx#T`C$)0Z4n
zhvgglT-YGv-*dQ#1y-!{>{12CqScqoWq?Ux59_p0`Ys2T6wAGG*#wvCW_~~RSTC^h
z`H#*Kvf49a-L%_HWxcXxLvf*2v!}pB*T}h%yK^Ph6wb&h{{wMVpMn@Huxe+ArmT=Q
zboIFwP7GGmr{2UQVsr$CHIIkfsff<fWC<-8ayOG)i~f-ee#=AXO4G%03)hF7p8XQs
zdEMdZ`#G6zO;VoWWK)@6t!J=#MKjBt`;cq_K*G<56ld)5?jtf9^$UE8GiBz}88z2M
zt^HHwGuSXr<?1k+4OpSjMMf6HzV8Pn?}Nv{qUny&JLd$DX8pkAwH9+o<O&FUof2!E
z62|~>(y^g#VgPM3Y-mL9kNloeqWo?wgnWr66!<<w`$SJePe^qW+c79!nG-?3jiC^U
zC=$0W$bew)k+<XBCs9^77MP*ohSkMM!@X@Cw^zms`0DcS;$4%i%>h7cgUbUiH*1Zw
zB}QhvCoiT@<5ot=HQ-pclV@h6<qHgvwJp2ik0Xms(4mtj#{@Lh@1o{|Q?oDR4BrmN
z=OVv15z3;hM-$m<;J{*x{z=pcdMLdAjXfcI;cTUXpSZeDRx^K3`>Yk{I-p0rHDpgu
zYh*{8>QVF1HZsJ$ux;$eJZj5<_{+7@R1m%1*8Gd;RT<E0H$;-X^7hGxQ2fa<1e(>u
z>)*R8zMG2+HR=Q|<d*c5TTOGIE*2+_2|!12ohGPTbRBfHC{Y9E-d}7qu!FuDe|u%!
zqlAaiK$V@%KP-Hber<tIVXhGc^T&{VVE1DoJrG=8IkDC2UIJJ4f;;NJd#S9@yc@$J
zNCPmVN#0Tv?xhkdgPK98d3fg3zH&2k!^k-TE%@yt01qd#xUHwm(5W9j7MDJ!OQ!^u
zYu1NWJ)e)oXtsRpjc}aViod=Fjp>bd+vWZA4>VfF=*C8<#-#;dMf)O)Dhw9Os0=S#
z4W;D#DjXeoj6dHTr$EBAzlt>VGHQCQ9GP|_V!tWMO|0qqNJ0~jmZC0opqBcsmVh#$
z$4v*q-RRK6pnclY<HL$ogiE$glA2G64E6b#q5fBS2}sRByC&tlGF8N8$lzU;N%r^2
z&Wn)j?mr&vSdiCoMP?-ixjbqLc-i%d6hJXZ+1P;_5`gMOs~$YJL6Z$&>_ciR5tU92
zh7=V)12k9cbT*lDvqG$<tG2B@W8<@DTp>S_oKfGZ(#AZ9gr?A$XX6cN0tqccGHsm#
z^@*5i6*@fdM{1!l>x6w-F;__LdC3Ic9~QweO$V`?fE?36YUJOpO%re|0l6D{?a#p{
z9740Y^*`fIHN}xtrBa=xOCLTDzK-EDe2@nGjJZ9%%}Mm2j{b7ocP5t80op4GqxTJ5
zm}>!#$;HzN9IgV4zC3UtKzs2t3)Yn>81^d|`dw|hGy*bP9O&~?J1PpJS}PNj2$!TS
zcYs)q)bx5TN*(a>f&BF{hWWilh$`&!1&wolpAxZX?~La*vE)E_OpsQZ?>SPypXTnU
zzt=nKpTu<?a!N}62T$yIH*G!nF(B=p;)N#bt^wPwStal4=Ibu!i=e)wScj<taHRea
znw(~-*6Ttk#XJ}HlSY&MvatUTd+!+*RkpQ@E--+Ng59VnK@0_G3!+4cr6S@I0TBhk
z0E*-!IaAmMS`pAfL9z;hN|qc86Nn@M$)QwoD2iO9s=8y9+TDAfy}!Np_dU<O=bmHz
z=vFLNm~+fA$9Tt>@2fDe6%`Ok)u-3^4;=LhPT)Igv~?)xIp1-=EkjX?GqYvfD6Q_8
z%QX{oZ6fSh1siT5CTV{gAJ$Izu08OeZ7XV&`!sjbCob8de$edP-E+SViQPZ?#OK>n
z?2%(EB`0N2Ly3MxnT&wv<WIGa2`7EI{WS#QibUvX!)g#?B@jZV)leF$_}Ht7%8e3m
zL!LpV+E35#nR|2W`r(ZjrgWD7A7V<&4E{jD@4Fm3%dvNA_Gkl<%f@DdOAPsfzy4V%
zMK=do*vM%oH)pS`HH5lE5r38Y(ZWkVEi!YoyWSO>@(;vN7-&P52wkknVOFO}w<&N(
z-948M&v)0kA#Zc14-O6z*YDDguA;6&syF==?~e3wKpuE6EG*n=vUYVeHYR^X3%$I&
zf<a5>oSd?<kJh00yD*4t`9yrT4`iEkIma5gm)1JBEwz6UnWFrPK5w!XRx%!7Gx7}@
zeoN$*Zw@#~;QdrDP2CN_)U%JLnVKl|1pQ{%nUiti9*Umk$(pB_MLy?y2DUXZxnqx)
z>>pQaBIoXFJXDocuR_#E7tWv#KSh5lXn91!i`!qW?J}I2{m7t7%*7K3+i1s?7DJ+)
z9S9Q><l)!>1N4@>Lw9jO483kEP#mc@bnuT@9!S>t0;diW%NLFUtVhpJNL7^8_Rabu
z>wnI3bE<lSboC1SDLVDD97@<d1U-3rsIb1T<->F<`>v&Lxu9Y@K$~O4Y&>59m$7>r
zZ;SDryfn92(&zb*bcfC*{8$M=M`>|oq?W~wSRo>#G_$*D&L=CuJ+AKo&(Vb=ru3sI
z@YzIKUHTSo`{%-d0do5~aB_C4<NWko(eCPbAEkwJwCJSsrAuFi^--tV66cZTRJ^4I
zsVw4H5_0)O!wAk!&-Y9Fvpr2>rM{sF0ZESbCfa{2|1V3d<^G@GJ};~~a_#WXnyk~q
zr;pH-nGJdLC0^2EHJ1(78H%J+Eq&Nbr`Z1O6R|V{U(Rtnpst-Qxn*UWj=4*n*YjP?
z9KhPJ*Ap2RzoqUecc;%ciLESkk2LfzxD=3Z<=~?XMVGY*=gVq2#ewF&srf6|GcKpn
zjnI+!9bPsphveWum6GnbWUi!Ualj3byHs2A5u8fK!z$y_KJ^e1Q?^dXSG~7Wp~ke(
zM|Y;JMFP+8#=ASZW(~_TnU?{#UN#^i^C3YZJxxaW<6Ok^$YIk?LTV6zIWNDag(+IM
z*C3|fOYhOnlikij%kQ9LAJ^zZJZlZ&zqC#)!cJ|nmLJ$&^qfho9|<@1`4U7{ie&~k
zsCrzO5Q`tnycrxdaPJJgx<0m#FK+3{TfX)xob&UL-zS4%VNfl)jjQiLoWIAt<Y<F~
z<$}ZQBMn^L<2U37XVMAus`K;j4H)VGYm$eXl7$6r#KWs3iPHpyf&Qqr%#>VBuiOYI
zrnkrUFeep7gelpL+Kcn{fOA#$IZbwlxLj0iw@Qkco$4CHs`Q$QM!UluMzek3Yf6-X
zqAmsZM+v;<`a8m_I`I!xJQkr3C^HwLSfSbinndv)W@mC7?RFk1QQm8@GLo=Vg!dZk
zsEB*6YFpb>7+piU#8^@1%OH+bJ+{1%rk^|nmGKlvY;_a{7})r})Lxo$y|^e}IuT^x
z2g<onZ2iRuz>Kof>dmzuN^>1cA19dFDLW(uevB0u;C&XG<QD}TGhuG7Zu5QNBd5~6
znw)VZN^y0}(1bG;?a|IHN4zXd&t(sUYc8~K*J>!i>-wE<kKe%$;8J)B2yLGh>yM9d
zYq!FaDOfsi;#u~-xH{^cX}qGu_pwRWPbP_5#X64_d||Lrgg%sd)5gw-(EDQCyXS|-
z#>R9cpcBBxlolB4^<|<dMAd6}Q|;UE>K2jkvy)kM*AITqgSNPEX1X>5dpF>Q{fR;*
z#j%&k<isQ6K4^<(`%bGT3Dmd?q?#Z)4iG{=;|6@*=cSl2Um{~rd4^J}N2>ADoblv$
zvKt(sb8riKd=4Qj^>}wr#nOstj{-KsVVrWPlo4z}5m^IQ6Yg>w-UudfG}qX7rm1PF
z4UbK;cT!$Dym2c?71HaWu<Mt{J5>*}iv81u4vv*84?{oHm~r@Y1T>Kh&Q0`$t1b-2
z<L_o4IT7wWHivZq#%oHJOdCR`DuxsCUYe*-5@p(2Gsgq)OWpwg+FUbnf^&Z}VvGvE
z^OwWuOoPVKL@%>}S5pGg`7w(0x~{aiw!zI!SyUe2SzvwkSG313_`b=z!7UY)v#@eM
zZitI`UfTYQRzqc$;FKc@g(p{(Pd?Y~-n3v-0Ec2nW>y>U(c!G35=eJGHnlRICe&T~
z>6`aP$_{lW>*+kg@37jRWN)iejusX=cAge4>JZM{J~Q$$ZhquS*0omxh~AyVp;>^a
zK_zpZm1m&Dl%3&K0iX*I7Jx<gfZ;qriGr|n1X$%P&;xv8#h+wVxC<bfm*Xq<3uH?v
z*ySbQ?Ni8-Lj%AA_6+~h0VA7vOWbT}>+KLvE=Us}+CRAV!TNuhRcvDf@a-3!`Ic1V
zO2aK4)EMnViZ66Jc8I7nd?^S)+hxY4M$U{{<hJoCB8qVl3g~nFWNF!YX~c!UC40I5
zv5D@Y2`Fvx4+xTh#(WL->zSF#j}}6g8Yk)$XEu}xB(}BKPDnml_HgQ-7RmqGl{3r}
zf{`e0nss>2)8niWwNUh^p6uT7(KzYMj-d@GOd)P<Kk|TGI`G!pFOONc>4|6eu774@
z`*i#0lM}Sos7J^h_3hiYbM@@2y<W?<Wn(|W*^4$Yzc9U{7QA)hN!C#dY8HR8CSt_j
z5ASdF*Om#3w|l=wzjDTInCJU0R&DbH*&}zdZc@(dblYh^k{11Yd-)yD)OAQ{!)f~u
z(-s#)a3yJ(3{p}X-yVy9o?E)9Tt<*Acu2!YT;)rrg5s?h@U|6HH6s%pBNL0TDQ`=;
z&tBHrO6ek=m#vvfcbMLe#zmPMJEPi3Hx3_^quh}J0X+QxZZMVy-&@C6K=r3@K+kPY
zt}US$%oEWjD&4UOQ2uIj>`BFk>dO~uh(AfhdSB@AKoxfm?GL0^tLr>h-Ra~oxP|le
z*PUdj)NC1&b>GH_eSe|{wQCM~DuSGYsc=MiC0qrdbpunnC&qaC#d);n1eEIyJLi@*
zqak^4x-hib_Vmon+>mQv%2!rv5@~)o*4C)tewt{|;{_E(UZ^Od70_6O{C1gW57-$b
z1|@2tq&0isaKlr-EW&8(GV9c}d&6Un)2ZgWOW7`m3=?wcp_B(^<{Q=rzEx*+C_Opx
z%zF|V(Z~KCG<>j?cx4n@*oST$T8$&R#6#RDyg09c5*{z$t7VC~N8yO&Ro+qWfP}ON
zZdI;7j%D33TZbsVktJ?!9FnKi&FbBsw%;4v5%D+w-u_TXjq0O6lev_qc~rEGrizl1
z76cLDQ*`0A-^>NcHZ0Ym-#)tF(qnN2QzvjUrSC)`{Vz?3%N_D5iw;X1a+B^*hSKH`
z6b*g0v9LjK0(-M)-c*8f-~0OkY6EQk5&F>V1KU$n^{5(zv5?0=BcMcZPiG&L-vbkh
z((}LEf>DNvI#u)&(xW0ebS6kR+dN>HDM!1`{hNu~k-~r6V+f$1fAz#SP=Nyn&G>f|
zup)UeEee-<4?5HlBNx@5kO!qv2RZhiHt)wlU!RoD%720Y0ITwzF7`5I)Bs&jheLfp
z_Af6&FTvfAU^BX8{<^jx998@UD>FgL*1dR_RjwcCynvn$0*3iZTI<v`pqjN8BfvH3
z-ZJFaBsOWsP}n-8_4)Cmt6lL6)OY_mSUi#X=<22VDA|c1W`CUBBlgFpe;hd2gDE1c
z&AD8C5a@yyl@k_%7zRh%w>{gJMrkC*wmB(&pTrMD{hyGQ=c=<L9Q}Sm_|Zaj;7*-I
zy$TXSf<_1f)7!q+IQ*TPtl9Pxt%=>K3vW*LL?n=#o}vC6X98DVGKfs<2*oKJ7vC_z
za_l(tMdI1r!u<UFMw_wruH*kjHy2_e2VQ6VMfs$^c2}>1+rb74{*my!%iLQ={<Tjp
zFJ5s6{oePm_r-GLnA)xt{%?cKJc)GThJ}?|D*vnU-xehM0DL`@i&}}!Mx_XqZThF3
zlqn0Blm9lQV>2m_Iqn<H#K&GDk*!8822_4F5A{CDjA<2CJR&BY_4Ks(?w>F$5?i*9
zq@mRiTgam|$Dy<2?%liB0V6*+fUK>CCb8{OxGyrjY8=RA#eryu+_6$Y10Ws2*J}}E
zD_G1OSFA@;%>wanO9S8ied_iH{NdM;GC{EC!TtZVug=nw6`TLp-q=>r8Cl9rXo-75
zzuwbKQhlpxT<^qhV&ArP=sAS`MhylB17N+Hn+wD^k3bc|KjAg2E->}S8tYoWj~l7Z
z@9yq4>t58taJC~@pYIv1BqSq$LVCSqT)C%z^*?H4QyAzi8BpZh5{h-vF87x+;oS0Y
z*}t`iVx6s=i1gTRQ*&1fsGCtyE5hKGnCj}s7c=vBxNsw!&rbg88h{NY003i|@@i~b
z0uLWPyrJANnk#16{yp9&Ng;F1uwIV|eGhSvP#`f{xti@;{PcBnzYV*;0<e02xY8ig
z6^pC9axrrsp0%(22QUpgk_9prSS{nfv%P`P2=^rx{_&g(nnl<h|E4^K!5>JwKlX&M
zKQ%l}*p7u>TV4};ON3WJJb3VOtx#BRQ$A2~Hy_e<IoW`DB*a9&K!b?3|932R+n&Hb
zHV&54DO<b$yMb2!g)aWVRjjbLL#vSjA>WP-fm5t51|rT-Z+AEk2tG@&8rGkl{8bY?
z$?4tU^4Pw*^^Ie+W}}RDql*EU@17`dZ;FlovIhC)K^`LY>@VG!0A#}X)WQ8<U3?*y
zv#$L|@!Z$a;|-CZqyNH_iwN-oPb*@6-y3vk=#Q?tNc~NK_E8AyD2DL1u%a7uLB#nt
zZ8F>lt@dDyFGqFhs@v4Iv1f7qFt-2LP5vKjY*Plgso!}4{&@)T!|+&506)B(hVBn9
z_m|4Yk1h4bmil8${Sm)mtz$ppw?DD4AIZW$I?VOghQ<GI8(m_Kb`kCfTKNvl*M**n
zjQC~y#}UJ2FAUwT2k9%xO5hA1sHCovF}je_EiCtNEMjX$u}9ysb=-0NAwf!GuWWvA
zE6VTSl0KB;Aa6q35EWW;jNO#d#$PPPwFqYj2M!rt5^cI3rp|XE!%Vn!jr3}n=YFL$
z4FffuXBI}%ja}rFTH3D1UxsSMq!+gzQVxCZFd)w+1G0xB2ar|-CYxU>%M%SGreW+?
zVpEvXAx1S1N{7kB$A<xjk^!Z|77>Zx0h9E<>7-6X3qJ`Z)Uw2R^1g}l-1{cZ^YJ~F
z1RnxIR)BXTSfWcUlfH+Hwz2d6*`GLb9PXR_y+08#5rDqCZ=W(bKqQ_%0>!FuGta$M
zFJf4XREA+1fsgY$M`mVO1YMvOsbd%x4%&i4kjsL$U^~d^f~ICP@N+%51TW;JxZNtT
zaY%3$FksLJJIVRkqo^x(0&+vp2dluv93ftd%|rN~0sXaXmZ7xL2wX-K%dT%=+5ngL
zt;X((ZWLg9sC)7D1vuxS1r)AcYY{nw^&Q*u4UZFSu<ukBgNz!-R~fa>&l8^31A?ms
zGHR|U*F`Kteth=sCE{)mj%U&=7EcO5JD8+qU(8oY&74W#NRp`sy|ep?R#g~B(m;-n
z@ZGTA%a%X-J|ZS6>)^f;{@?%JdIJ*9Z=Z}cklR{=np&fo;rZ$pwn-R48K|k%Ki;vM
z=2ha$8a&`J(+lf+@_2`~7tAk@Rb=j0F&d0Lfc4U`wMQa{ToePoNouY?xUbtj&w|D5
zxUpw1Dn`UT0xti<!=SoZW7}L^3zzXnAvJ!w6{F=5<h}}bz5Gl0VZ`#@M$^+O?E4$9
zt<B%O+<(n})_0=9vv&+dV&9RH1c|RJn9|1DgZaPiw9W=Rp%Em$&NcjY8Vv5i4g*Mx
zxw;Pwv*a?}zR6|ke3Q%6NmT}TQcDITLZ4W*Phx_to)Tn@1P~zo3&|<Z3b-%VmhU5!
zyY&wSW5iYsvvrWEa4`(y0|8Q&@qpTB_KsUsz)c^A4c-STjRrq@w&PYPYw#iyLBR<G
z)M~--zP@ylG|?O=94=5<ihkPH*9PZxP&R#4^z%5@a1)0P0wcxULJ<9vj*`vFg3Fi?
zv1!7{p+Gb5t6Xd5&j>#zZySPqklDICjf2eu8sf0;^ovLwEySczqpLMR-u4q%5y;!7
zfm~}U$lF2>&)M+oEzJNUh`wKsiVpqeen*idG|}}<dq?k^_Kw~z9q1J1U9<A)Ihb)w
zq|f<$742uMo1Dki{xKvVK47x!!=KV~cHgAuu+i%Pxw(fofEw>ySl?UmDmRw#m>F)C
zxb?*eboE%=6Vr=y6|lTtZ+%I?2JcRyRWgg|)A{>L4-ef=h|o@QZvCp{skil755>*9
z93vR}eedQ_mu=`TULbCpfQfxpo7aq{z-3q~7+lg8^!4XGUkO=lm-`<EwbrKX;EH~@
ztsiddhuiw$wtl#+f%zZN){kiGN3``L+WHn^{)o1IB#B=WpC3u$k0kL&lK8(Nw)JrH
zFK7|2OLk#2?4z63a)VQ+Vs=+eUJNr+!gQ1aUWkWmP<Jc<tuL0k`vRR2`$Ak2#5Eee
zDrsb&I(5qERFeY04SOngAanV84JiraJJY97O&~a0SV%VVtD>(#jfE*w{$X6pErb7i
zU`4Bu5{~XHL2lK_@iAqlS+p_V#{2rIRjFSI5b4US+@!k<(>mHGxRNDAdpijEumkkL
zy_WkYf<CwkCnDf4Cw*o>8;!dwqmoOP#GSi4Z6<+VvG(vHqs)?QaC=r0lh-px14Ml7
zrnbKhjHn4%3!0VxB}n9JWo30B1tg|dt;P8sB?hU+g-(OEZsB!Phwy15lvFZ2SEp%h
zceJ6#UA>BM<KKgk6ra)XNb%)wsN2($mJx4xD$_!RS<#}Ms#vFEe=J)<16uxApE9o<
zh%M61mLba+vRo=oaJQLf=GRqJ;OIs5EzB9NHQ#$PT1(-JLGr=ngyaFj6ya@paCyAs
zxbnBl_}krPFKf<smSOa*jJCNTGuH0!{0J0M{wF|7EC0NuU0H?H_g(<yT($v48Gcd*
z+rUBJn7cBKLoVFj7^Z9P_UpCF#a;Gw7_b5YbEZUX9D1<s3=;GaH0UDRQX*nz045E^
z9&zrS<&q(q?!z|Vc{jm$voF1#?;d9krppP6&Qrb@WRPQpd*s;n$G_S5diZA~QuJEc
z49H+2KOtv&3`Be_rd_uV2@e)9IY!?Nwuav>WtQE*sYyh%6(^{pEf~BQ>WRsQgE${1
z8-5Dajh|TyS3!v*M)p^$p*!mVXGP8u`=|SL#;8|Zf*z6EX1>y2_#Ox&Fv1cHG#508
zn~Bq6rWK6F)S%KwX_oVKmtQI!iHX3JjO_t?{<O1hYUej4<7fB9K&#;<Xw#(0Xq~~d
zs1?%t7Jtt^^yZC6NkaO_ko;6!m4$A@nOx<W@;~tC@8D~0GgD&7jOcg1Aw2H2V~Ys=
zw*+eV^n}*CKc?Kgy_=1H@C=cWS3j_bgVP^X8gYD9^IeFJRl)voGZf^Q!gr(JO^G*<
z-9H_UyK2PqjGqG^AeXhcfsfB3L{MS!9%w7JS_9A~n-y^JHdWDM{L?9_GLq4RIfExp
zp72DqEpiL)8(+oxVS|iamoKeL`0jLs_gKLIy&E{GPSmr%)3}qZ(a>e&H$-*z%xuv{
zi!_97jV*9BE{ZX)F}V6_M9fcH1s(D_q**ZM?TH5oyeRR*)Q0`rU5mO(%;jv%Ex;p{
z;o^~uoZk|Fa-WaLIN2~D;;Y`klyCRjb={wNuHn_-RlE0JVa<UPg6#ej+&}UKU`ke+
z3qmOA*P--~nl9OP#wt`PS7yC+<-dV;woFRmUw|kpSD6x$Tg!o?&kc?U#8dHNLqP1J
z*KjXE{^-I}5T23>xLw!#mUuorXy-<XUsLk=C(Agvi$}x#hQiO|RaEF4Td(CyMVPe(
zcQ}yf4<ab{vOj#9uc#S)b!8a`!)1o)A^W~Ud8%5P)HcOTYZ-YGz%69-)JnBK!XkEF
zD<&>BwsumPG?Lq4@BLi7O8%pSFnqUe$-E9f-y2FEb5X=Q_ml_e>#pm>o5aqqti?}%
zP?KbSOj3O^(vr?qC%nQxHB+adF1WN-q-)pENOC`U`w&^vD>)%fFo;xoeZG0hAv@Yi
zt$0Ywp+Sc^twYjppb{8Uad?rTo?XuFosJcCFz>GhQ9SlUp8ZiuguxGwj0Q~M&k^L&
zd9?s;iIb;h9wr#|fB(_Oo<*ZneY>f>b;?swO3XmgT%Z@Nh@47hP{<o%@z^T0INU=_
z;PqOlAy^eORq1GYJxd@-4qYYs={4w`S$|nbnE3JTM{rL<3t#w1bMjz;gaRP&#mC?(
zE%bHDH{yRjIHwC8^aoFyV9$X)*x&spN}dU<Nn;g^r8?$PAG_%O#J(_9;j<PbQd-}(
zE2tMudUWb;y-$qMauq%v8->NIV4Pn+ADc0B-zN%$X391F^U4|ZbBiORv{N8jR_d3Y
za4u>+=L{&?4#ECtZkXGM?T<UECM|W>g?pEDt703|BM_w8Sv7e=cbyn(bA@$(>eLfh
zZP?z*@^HVK_GHZ)<9ziRxKSo0QRHCPzx0)_9E#cgSG+p6obPbYh{FXk$PZkH&ERT+
zaO=H8Pi6mQfyBhd=b{WFI)UWIwaAQ2u!&D+>P*l+?lYH5?+aHmKI34Fv?lh8>@43m
zj@7abavJ0;EuUcNlLP(bw=&Z)e}f@!$*~%dlDQ4R3i-ni2+7p-Y#ysFFJp<i|7%zH
zS)}a24pJhuE1wx0<}eIIyZ7systI|+5Ab2ou};19M>{tKgv3SP!)WP}i#$n_^L-Wy
zqJ!S@)h!Fhw*6gM;f&u%=bdR@A_zzC%JZmxA2bO}Rt8b`zA>018EbM%mdV_<DhOa|
z;JPsVITc5)3OWZ0$96PCN)9xH99d9`y7#l}@l#D!(nSuAj*heM?rzudgzo=T{jfe=
zY4087b7@yOc+Ss2pZ}Rum~KNOqlJu!5^C9%WpU%!4<)o3o3TSxKcjI41F_!Sz)OyZ
zu3v%RjRkz|n}`WanFW(YrR_tmG1>ze3HWJl_??>S?^PWXSQTWnzd6UH>zK|se-ac8
zgZ9=QRonq51rTK&sAK*crJbiq&4X8O{QGrQI9IMwMdcdw;%HmWGvX~NUoO1|(-Z$M
z%RG}@#Qn}OP;Y&c=gGk)$NznELbO1Z>i?&89y=;>x^a^Pv&-kL5~t6DHXqK<vj3t|
z4Se?--az<JcF6NI5%v1;ABP0XMgMHG9eIRguz7r^-~L(_`{tn#thl&;4G7Q{idO|1
zhEb-m0`vP9%0}zn_JM+A9p;k;X^Y<PUlgBQ_ZDT<2(=nPh#l0U1|R^!c30$M<0e!Z
zvp%dYYGR&rkh7wLGJ6}Vr>!+r79Bc7#7%>G_K_}o_w<MwV7dAh`y;ZKSdrL;ak#7Y
zt5!}V2s8_mH4w|lwd^fd;i;ic4Bwd|ZV)+#c|7gEaU&w?LCtFLT^yRGX|#4^3#ixh
zU4Vk(0qPc4matb*i8JD#q+FL&EdBNW`mxZJEXu!nWf24&o&(6I{^Bs|_J&`6ez4^m
z`3Qof1hAK_e7NF|!#-FM@%!h;X#E(iAGY;lCI0YRKfL_^b$g<vHx@#)1~<siOFv(p
zo4ikjO@3gYzduY_ngt3(UdLkWGSNL*##T8-JSjeSL2i<9gMs{D=T@yo2b<9Iw=B$=
z<BUZrlO)&EiEXO>scPSP<)Yer0|NuIG=c>^-(!MJIe;TZwyEtccNbFj7$2Tesp|>B
z`z?m*sLnrb>L{gC93*2^eHWr6mHVVM8>i)@qdHQijPoyF$V=#u8(Bh$xF}b9Broyb
zkEogX%l>iossDQPK&4pR(kAJ}_F)!>u!G#M1JV$v0|Tj+_l2$?0eL@@tlBjX|Gn0{
z&U?+=$R4&=t@}U{-IHn_ORi=#o5u?Klm6&E6J3B-z2V>3k>hISa*|fM4YF9|b?H3u
z;_YSBEp?|N6Y<-IsCNBzoLfi9)l<vH`TjvS+NTwtg&?@o($OJJWyR4)@k&m;?^;-S
zBmfIOd+Mg9KRi+c!OYk|87-<z7P);ij@LAr(JlGbw}r5JtL>4vasA}C_-r<d#>Pg&
z*a2)R?m)+7P&R?d5Ca@)*R98&N<5HpcC7Jh$F;R06Co-{kern~Q`ocjuCP=V`EIsF
z7D1ng3#b{j0tq6Lvb?;zRTe?cKULZaA~xG>`XUYg8F~ptg|INoe*j}0l(f$=3OJM6
ztRTHunK~C?zM(;elD1d9V;^QRWsR(3clj%Tr!9EzW}z7}Qx^rzxvNAOM#o&#Pz!<H
z;!bJJjL#|mI3vNuvdjmZ-`eJ~Bf9eP^Y?l8zWaGWh}|;!Z;b4@`({)s%Ug|;G!zrx
zWlypVsv$>2MR_|_y+&GPpkuSHp`ig@#1}X+TJO-jA^1fbvt4Oj?u0`P764tv(!3$D
zL+?YZossxYU~5-stt6i{owCSN=JeNE|1CUrU}5Ed+i*9r9Gm?=K<Y>_tww^axv&6v
zv#C$lG9Kk|jk<}m%%BK*x=qdXwWU-;KO2?XLBs?NVc%(>cd|`&?(!Unk;hmilS+2f
zR@+<O2cfPZ?4KF3xzw1bsKxivz0X2PCFQxcReDJ4$-U)k4<5gL7n|XA2TxCd366Ju
zYw1(tO#fNZW%;LvUG{{9@*n}Fa!haMM1%9M$b+_&gs58CNbj5k9Xy<i#hPzZY>!}T
zvdG=1V(mdWd3jndYxPRgQ)D}{jgy<anh9rjOGx+vWe+`hMqBXU1Gy~IyazoBfqS(H
zRPt-`Df{NSWuLZRyL{@zM@x`$Nkoag3o!gxm*uAwk09sxFg-o3gkNl{V`OdD2=?hx
z;_~nJoB~?%9PAAPiGCh4K=hxSK<ga#R2{#<GkRy525K+zgV2Eq6F6rQJ23P}EntIw
zE_5`~Q=4n3r9pfB2KA*&);jo(oZ&)A8Qp#%z)`Xat&rQL0VdqyF*MF6H!{FBzE?#h
zrq9!_<>gh5g3Zf699g7M2^aTlEPlM>r|v)wL=vHXT@MBGJY{WK@=Y*$iUh_|pu^1a
zA~#Aq&34M%jht1JUOF#{Qrs33$&By>Atkq^+Env)?!!9T%37SCnXbRHn}iA}x>#<#
zc3JN&4-$Ei$)M|0*VITr=Hic*^TJ~IB&w;YN$RrBVN^^|QuRkN63C?c&e=w(Eh-&8
zoEHd@iSDXZ-bi^3$T0~x_)SHnqYXetPMQLx0mfi!-JAl7mv^g9su_Jb!7CvlL4PaJ
ze>pU9D8}4PZ}Bnq3*;I`t7%A=tROdeqp6QC=VzdDIB}w?x|IK5KL5cv9il22nLA?w
zB+9;QC}$;^wtF!c<&!A98N8>OY00D-r*z?4cK?JpBZmOMf0q$ev%3A_8BC1|;93nZ
z@KbV=VNA3-22_s@ln2i@&}n35HeFO5?4<JYK0)$QP#%d+A%ox+tUd$#&iskPC_h1G
zKj&vaORdcDTZc+?lqcRa#KUx5S^G-O+1a@zulFmvN9zd;-T2%7;u;yRj|=h@c>!xT
z=il#Xl(9n<GrRp#^=)c=|4e;#z}FHiQt8#BJ$XcnqK(`E1ZST!oS$_zgKHz$nKKO=
z1ma!~84+tE+v1*BE!1fcED-C*{{pZ>j_dphVatR^(98Lwj9DX`)18SDi%(R^KQI_c
z&m1wb^S$&vldc~0ArMmOMybuhso`cyd$XdtlTFOK*1jq_|H%`0!5Xb1yXOzqE;H$Q
zvoQo!yo;;+Dr)~vTT)Z<BoW&%Oze)QrCn8!+c>K`6AVc{XXNhmRF6P)%Hn4p!koIz
zsDOTmxtn#6uc|Dp;nbNIw$e1)=9@L6Ip$$1ugM(#5*xnd>*FPUH=wfhbtm|<1$e)6
z$$X6iqy1@WH4up73$MRser}~{ZoQ_14hK;^I_&ba(jAvr9q1u=#vFynNS2+Q&pGoh
zGZnZLx<fA*OJ^DH$rAP(ewL&N(^}ctdG<v{#$}j?zOB~`QD$A1ATtl(@L96a&O^(_
zdH>-y_=mgT<Kr`47&ag+R6)r=<_r<X^b4w?sdVg(kE-Zd;ZFvic;=_A%8I1~&d+Ol
zI<OkWO9*vZus_@c_Nx*1^cm-ix5Hy4D%>uabHte~s7QKNYMysv(rL`(MG*QW6+>jg
zbmM$^kaITJy||~mv$NBnzrSDAjqFP+kt?5b2Ng-_=z;DKAfUn*t(ss{=77z$?ee~2
z8!xW~V^&Rz<xVPsh0-6y#SudP=qLVZ>zl8hjsrPTm|vb-q{^5Gp?W4RQ(dSJVXQ&j
zZ`+mb;cqFl!WT7ratkSDZ$aV6xv}X&GHe!Qamso(C~yotI3edex3C%~-Vb{sTDLb<
zsQZ1vW}rvL|Fm_KYwJ3jny5AsBbB@$asCj`J|&mI4+lA4Lie?QIJQ^sm6y-!g);zH
z|0;rv+b$68Z#$d?%(AEsbX>Y*^0Kq1IfyRc6ZCBarRcW40~&TV<%IHpSSuz4$l^*a
zrBb$DlLt4ZGLW<tMkaQt*1dE}5hdi5Pur0fn)5s`?*Yx>f~*_dK1_0)$(*2=&+64V
zwc<CD$3Zpj?EM5#ntZ$Vm2glUwKdbUCcHzHG1mjm#HIrFrOBweFmELzcqk{%3)V9O
z!AfHu6EG2Caa0BJY;K<K_G=mDnsAha>dD8t*RkAOcLb47em?wXfN(ddcnY!1UHfXg
z`D|Hhe;%0cCI_tx<`BwGd?`yJPXxQC_}$uae>yvA0X*-u1nP#B;61F(d3^hJ8PDI(
zk;1lLgV-tIORs`%?G@<WZ13ha>)t2>nm59mquQt5Z4W6vdm`RQg3YxaTK2_ZJ8Z4)
zcK=j%MaKifhr<%BMnGwh0Ybu^AgM{E0T>*5OYCoth7cK-gX|>vXBG&N6`Q?B&4}lV
zy<l-<HjyhfN~SxO7BrC`v^`t<YGUmxyrC(X48vx4R$kV=^6BqWk}Z)oIYt_}`?&q@
zemf=M8HIs%AO5b63USCT3BDVzx2EzT`B4xzXrjiaO@}MW<yQy1j<Vlav5`QSp`pa;
z5UQy;NBMb00o)L#5At%tSB;ZdvG8o-Y<34F=a!U|^iTEvp?^fV`ptOg@bLJ1lyAFy
ziAzg$cXqiSesAgLGvoz*xH>F;@~W*%FZmZ!{dmfJZtK4OJifkXF~=yCT$J-J{}g$y
zT%O^Q3U=~7wU*t)kQ2_b>=;UXyc3Z*@W;`g^Kfq9Ye4Mm=s58t-EStDGzte5aFx}&
z(=zA9$Uu2^+gLN2f3cJ{3wusS4+m6;3d`&EWlqjOl}&Z-iKOpPx2#`(IC(NGZ>cTy
zQhBNXU8rlCw(g+EO*KJ;n3$B5_4%?#3q&U*4A$H{OC>Yk1Pdvh-;2kSnk8cp#QtYV
zwv|}8oU=+28V&|_p+ryV#o&YveY<YQIu+W|odl*h8lp1WdQRTH)1!S&&)QcN&5Fji
zCqiM5@|j9WU|RJ>3CP<Qn+K^mc>TgA(2t_-DcyS(s6QiOSk-SFNIzn+8()JnO%Gg@
zu^_k^TzK9kbNQ}nVeWu*<9RHK3uxXq9cmNfP#3FW{K3g4u!%d-s@-tDyc+n1YpZv6
zI$0_k$kQ(#xjm7*`2cIB2wcT-ynK<Z*EE`2XS>4i=CeIEg8FH$X&?jY{#5qXvc2Wr
zxFI#+g%3hRze~1WLoi~X_9E5O3}>8(GCAezs4OjX|5b;%wy+LofDVbo?%lPU6`Xo^
zLO~ciGcnH7LBTT~>3#17h}tdGJ?>4Sc~k1w&<mP)XPJ5KRNb&sGbMv08-r7gHCaF3
zLnaQKw6!YqY_&eb`!o7c?k3x#hirR!e6IE4ET<ntcOJ5z#zpo%MaORPg<4)bIhZ=J
zan0UMzdzLtxhc_=^gLzEWZln)t}W;Hbkoaep_@_)OLh*QE3$6tnc8F;RAN(8(7!|(
z?s85sBIWdpS2=)AzzUj9^LfKC;a#`BjMn;6;z$dx@k_7yuwwQmgIRZqpMH#amgfng
zhQ#l{1+_2^$_lD2Pj+vXudh8EpP8&lS(YUOl%%6TXv$Ib!lwl|_wnc4m65>IHrv@~
zmj={8Y<?0;JYx%c*iNUxENV8Ip}A>i^2VkMm$7Yh200b%TvV)MkpOwIn-)9qh{tzu
z{PB+1PQ&Bt-kNS$pnUA;Se&0&gkxrS^st5o!OzDK9ymg+b=Se<CdGL1Djj$Gpu;^k
z;5jYB+sRV(jM&&%fuefVJ>KBcs&N+b{6U(-!7CnegSja`60e%7dQn_Fb{YwjUWJ_g
z?fnaJ;-yQM5}qn;@Yjax(0NCdl7X)A-ozrLZ?1YdKm}QP_fodAv0+$}Y*dXakIM9i
zn>{0-Nu?@!A;U<9KZp=(Nq|Sm6QJ;^PuTE|mQuml$*x%YF`>zNeM~x%tfjlov?tDS
zt;uE_VzTPjDkpcM?_Qcz#`IYwKMmb=d~glE5HbnE>YYg;LsY7;)}&4kE!?cY6$_U)
z<}H#XL9^&o(H0Q%9hWDL3X@1XZjp;V-w^1RebQ5*I-8j!=2^L1o%-vo$I&_+Wu4Hf
zr1$)7gL{h**M1@XpmHH}wtOGw9?4&jjOBLi*+Vlx#jezUPfxob6$YYwK2Le};Pg~z
zv0G;;HP*RS!8^$&Wyh@$os7j0u~ymxF&_QUUFpRy?)OTH?oec;6^(f{p6w`s`+I?M
zTNLm8_39JM+MIJH3^};EDQKKL)ryiUDS>Mb!_mkapi#0An+>Fp&M*%q3ZaO<$Wu^c
z!+2a%mAL&U+TAHSE-&X=9q~E}OhUlT#2E#@rQjFhc-hyp`YDfhFfKi&n;A0vvf6xx
zY&&e^qEtX(phbfit#}}cN&)EtQ|=4w=Ujl%kurnh&WN;Y&G=k8J1>Sf<G^iHAJtvg
zQvI;jG;ucK9SmR|QCUhWhaPk|-X_E^#5l#H05qFG%ho{WB-AsLKvj>CjOixauBAlQ
z^oJ6Co`B7W?dUE7#6Z4o`$=#fVVtWFAl`x3DwVlvZ5)N3p**u=_(fXR6m-v@3hO)M
zUR(NUyYT$Gz0OwY_UgLDa+NWm>zEVLar4zt*bTT0fOnENy>R$>Y36U((AL|^3Fm>J
z2_Kl4xnWq|J+lI{$A{m)@R@iLetx0s;2TH<%DzOlJa#O5wUM8{Fk8He3zJo2z#wi;
zQQGXUJrM0&+YLeuJYot~#9x_hm{QEK4p`WP`(_G5ooUjZUrc`D)q)D>(9rxW{B9I#
z4^3|mVXa9iDtf(D=Q}q|_ngl8gvsua7KW4KIxJ;w5Cx?U+!qs>uR0umcV?u;0t_bv
zw@K@Tn>e48qjtiHb#Hg=&j0kA0N)lDldu~B0U<&{LUa?uFlu|T=Ty>2Fo#JdyishZ
zw&Kt*&R7*|)6fI|P<3AnF|jHXuWSZk*qHO~TaeHP*1Fv{H_XPo=S%uE`ZF&F@2|Hm
zz6fly2|XLxo2st6bePD+qd54m6dm5Z6R2_*-GHmM94BqKqg1lVBBqw1pFH!J!C>gr
zN-91)19?UST5B)EAN~<r0&=hi9W)FZfYYp=U73iu1-20Gn;GIN-@QAc>n@J)?!0yJ
zd66#jf}GtN9v(8W_w&v7Kut(;2N}T87%lG@7!;;T<r-%dzh_<$8+mF%JA>$a)_m@%
zb#DhS^Ueph+`1aP6n1a@`uU$S+SjS?m1At;VCz4?j)XmYA?_~L7T()mXv+2gv%=V8
zPW7cbZl%{Irp_uNyxpFq<~5N+%^VM)dj+qz?k%Z0FXl-iP`5wgf&mWq_V$iD5!UCe
zr!^yOs;+z1IIM39p2^q+FdOaMG~{^b;~n-#8;&;QsDEG!3q2g~s6i~*%gtULDE@08
zr3+0vGFd+BEqXl6tjET9vHilCT-ylCF>C)*7Y1r$Ms$GVRDnz<KNGL|5^0m>c_F?&
zKVQ6xZ-e^8cGSqsu)Et*hLI{>_44tKv{6`ZBmsKUqc+<#bTS1e3H=@w?_YF+yAIV$
zop;gEUB^5h))!v&VreN-)~3`0o{(Wt^7FFBwgTsY1w9>=neN9sM2(A=nXKplS&s{#
zUCA>`qty)`<}i6Xcbpg)TKt^8`Si-}n1o}KH(-~*cI~hfo+pLDmRbp(d%o2*Q}+J?
zKj+N{T|j{vbLXzsSEKOvjLzw1A!PhCkgbM?nR&h}^<AtbgW$$h*NSBst)T2v2ajt9
z8$-CZ%FM^84TYo^;(@G0#we^$N5u00aSa*^qZC<VUI5FnBXqUcJXlU#{`f1dfDMYG
zCaxfLtLo<BGJVy|u+(O%za}mXR^QuZTRCsA@r2}1AFy#9Kb@`#MBD((Oy*ozZ$)$J
z8Fw$DWV6ufvWvT_CIMur1(lty$V)GOK`<#wDlp}GT4l8@rK7PEt^1@-tQ*pD^6_50
z`FKZnE<8p$OvK_q9DyMwUb{;CntrmYIWbEhs5}P6A^xW<pyx>p0F$ZE=*UCth)FQE
z%kW3pU%YhbO*U_t^{Lz5ltS~Y`IMru!Vx+uCsw)LER&YT5t!~rd)b{%b|O$y$-Wtm
zI?1;8Y`pk`dU!S|3|Dec3GXKYZpZq3Y;Vr;x;5EmX6UyRFH9dCNGs@%tilA+9Cn|K
zE*t)6opbB)jt1+TXdd=tX!GInQ{~LyKZ2C~ax$!M4BSf$%j%z2j^O}P_ulY$0=yI{
zQM`&7kwCqxKOtO1r7X_j$&TZsQDM8B@|G@qyLH#n)$()lOcTmEwA9PeSas=@u-9Nk
zoY&m;^g-0qTz8%PdFL8gK1}(GKw6v0o~#Ey60v%svY9<a4z}@NP`~+V|LXo1R<(l^
z@(2ywW%bUaXs0va0zO`VsiO*P+N9vLwZEhr%)~*|)0kMQw0XFIkzVT9>Tr(PmBaLY
z9LVG?VBQ@i%{IHn&A*f9p$C;L3K(lTdFZZ-vy_YCb5+tenZd3f49tM>u>VcZ!V-V_
z`o<c8v5}Dx?*`B}s(S{(MPIe)N%35Fv4U-t0v%K~lX{t`8e{RQNlD_pBC6q33v;e3
zCoWFMI|z)!^F4h)dFW1koHm%k+h%ZUBZa&Yume1weT3r^^E|bv;`|BQf==aw*qFAJ
z&SI#F_oC`Pav&3J__b=%!9lgR2d0!JOq>ogo4=c!Ub>L)$$|C!oZABHDa8XeI5QAG
z!s|GFpb6XiFdFkx7thLfFk5SE@UvGn_>Fa(T;l7;uL;1fX#Tbw5kG*fC=kgzUu0Sl
ze02nir$8alV)g3PXGu0KX}m@C-KbcMiyzSf-|$|grQ2&<w#zb04@|11t8i?Y*$X^M
zfc;xLZxPWC2Z=IM(9%s8*tpTd5StBfCbO&8tdRwR;WR7~>pKQMGqH+jdoItuYsi&@
z{Wvsb*TH6%0`!Eb=d7tD(Moy32U?{!O?LCx;iw+pftRt&V*1POx@>|^<H2dRl+FJ^
z^!Mgq7i*csWZ8gm_u`U5?s<-9xu?(@d^)1uhmW0JjKG;0hNaH|k+V<l!Pyllpc0%s
z=(FOq-L1!i?ekm8tVJQTJ#H9Q6gsNu<cz)2C3hmMNtG!(2z63Xr@GSepvA+tH2OQb
zPnIs4^u#W1-f4gIMPgnaE#>+s*Xg!fkMnn;yC@TDO>(Sc_}y`B0zF&Gim_z`#+BGY
z?Z~4((9>4v%Vn7tCf_PACFRtMMciH2`(xv9Wbdc$`CN8dBV)cQri27l!rp~t`iT&T
zQ8t2^TQ??ZIi$nTD%xVv5q!ib-6f0BaJAF5UL)Ly2LxJ1C@UaS#Q46wi#~poCiWKb
z$ZQaKwjj(SZ|^<%2-RIDupu6A(l`8NA7_R*E6^%5MNW@l`B{pK&JiNh<0&=v?cgZK
zqJSw~*$<upN+Oj<z>Icv9Y9~CpKYCIR*X9I-No&!a-v+E#^Bb_yATm#0@4j2K564V
z+Uj_jwM2pM{5B}tj<{3WT`%%QdT&9whioL?ub98;#b9zVX_AKtW$#(}2WXMG!agFI
zjJ+T%tXL4p=?B@3^{y4{8PO<G*5G1DOa%^qrINAz$S{yIc&h70H+6Ir<#5$>MdI)-
z9|j*EtDMTu>8nR!$)yI?3uj0{W-c8X-DN4W8(YANwt(7R12dGYnw(?}+t+CR#StPb
z{l^gOTA#S>9h3D!oD_?z5SE&}H=irZtC7A*;Zc^ce}QF=k@~={GHbnc>MgY#Z2I4_
zkItA&#>6OWJ&6{DLc&0aa8LvPHHA7=7uIV{+1a$0-g@lO+3u7QCn@d%?+XEk7HVV_
z5&(GxY}4B-yEj6Np^n8ESkUbUI(8M&PVaZ-HL)Tehwi0Pc)abP$gngtp9`<N_zV{x
zkZxtCY~MAwu=|?)vl@*tr3-dC^J?f=J7{I`@x72VG3)quTpG4YA%HZ3Xrdr*Y#%br
zioq#|m7vLQECzx$IC4<@=|RBxh9EgH=l1%=JW<pzj4)H!QZjZIv-)|AUA;tI_nvD7
z+wt%<0&<-#WHR2!T@a~?00Pg9FG0b<3_qfvyQSxBBe$@M`PK!G+Ik+mGd;37i>Q1S
zBBBkCF_iql8q6KOg}|@_5(nn<BW;UcdN%uXwU<85i=qgZ<hs7giW`3?nM+tIfk>^Z
zyJi+F)?_zptAu#k1#GPy>KB4%AsTeu2%5I0CMKU2&K0@Ew5G~ImP0-r)+fP13lp2h
z&zH7LCyu`N{P4)v(785LcrHgaX!0qJ2@PjTMfJNg3L%6P1{o@kuClVSDpu-QW1eMU
zY57z^M#e=QOq&&eLS%4hq|kLlJ6&n5zaktZKd-ya0!Ys7Df&@-#y%6k=@1**TN>Y-
z8|z%zC_MK>ZSjS<s@-Pa<DP@+Iwj1SfZ6*xI!<RG^-L#`2n6?SEA_Mf2n|HQ@{b2V
zSr?+2zImXBI1ev|4XXG7Gs6w2l$6v7EHu`-<(Pgh6GjAP%T~*YdY2<7_w2&-B4Oa(
zc437JyX|5es~K8+sxzfzm4v*RY+T;#^z_mj7-oBld2!h=WXHE^4g8xDcidW`AiVsv
z-JcmQYM%zHCUR7psGppZ)Uk#}qY)e-aDs;!Q+Qgh$VllCUZ5DLP#7%^=Sj?+LCXl1
z+D~<}ZwdRpVpuS~+@#SLoQ0X=0*m|#FLvCLgygz-s#Ym|1a^CazuWC`-ll@fAMhET
z+|W@JdiHK_5QTQ^p?Fn!OX+OR>s;G5CkM=-n0=3ROSqaC9Qe7HYTS@fkkQE-IA8RI
zgAAEmM6#G0qX(U#lF81rQvLaN?XgZZpMCr4H8#sE^v>8Au^y)D4IONIOoi4t6$_T^
z8X}IxK{?I?wKGYvUc+DQ8Ae~ldsa|x&z^A;0)c4>50LXcx{X!1gP3HZVBiVhti<k{
zd8!jNiOPP33lM9A`{96$f)NLZDGkr3dJr7wDU|&o7v`mrhuXli4Ub%&oK`q20d4IB
zZ6Y6Ji;wCp1w${LGYsRoXExt;G9-?00U)uWwBpJ4`l$!N3B$xwwNE5Q+Oy(mlWoFB
zU4<ryh06?U#o1SDVj4wHU_);iYt8KW#M)xOr)~nD+Cpv58F|C{38XEV*mu^fUR|7m
zCHEg7>IF%FNQey#z?#aOuwC%40MAo3$hrWH(1&EL8M}rWSgUkpThQUxLJnomvmJZN
zc>!@IiC0yD3;YHRh*xbxf#g^G<C+xxxA_MCr=I+Yqk!iu+>d#kzvNTjBg<j#<qx^l
zbvy4R+@JN}%xDK#MTs~X<hT|ITOWwj?Kcby!dAF0co)}pEEj>D>uF<cj8aR8%BG((
z_$h5!0~zL7ZF=s4`|V&*s_uN{5Z)uS%76FYrX4Vx8SJpF{2~J@dI&};N6Bq;Up_V_
zdtl+0!@{3|GO1!F>n9jO_CMI6SR-wnQ+v$GExvxwfJDZxcK^;!wBq!sKU2}qK5HLx
z{j;^n+7kcKwqj!F=}wfY6d2~dv~yS0i%bapwhsp{d#J7CK6cjFef76-%!ETd`tg{x
zV`&h4;b~34iE|n<BHP5bV@VnWt}ZyFux#Ca)?cuZV|~hg0(|EG1JjGn`xmBn3o})8
z@7Q{iQ;@QJ+qSPy9q`|M2g9z!tGduxk%9u480!~GD1Z&IkfuIW`$ta6->k0-eUEc^
zs&wZ`3V0-d(Sx4?=%q;MEO7Si+KYU&!n{%iwDq12njLL(KAqJz`b>%Yp&YhA>+k)Z
zaT^Ps;L~^)8D!T(Sj(3+a;%^30wevG!8t2j2g&gYEP$5>kMx4I7U52t`1!55*Q_~*
zfgr5iK`UIOt<b)!uyF)iO|S(Stb1z+H0@_t*Z3Y&uW{~XPi_GCT?f+XB{=+nMV5Uj
zd1J`w96#wwn-T7d&)RcrS{fl>y>}K6^f;}Y)>UQCC!r{tPa|sXt`uf`7G~%Y-J6AY
z$}W0Wo#}4ys8H&@OPC8F<nO8UWTorpK^2BCZsB8+(@1Kbs%+&LfXLR4K3!}n%hkKs
zGKK+GJLSre`4J>`91V-y?Chb4s=PG*+BINV#$d4G@KZs@H$P)XWwP9<-G#5&PP+tb
z_Ogr}zb4iT9b9#gbQ`+wLju#%(<OC$zVJ#*NfD_&14{7gACvC)dZLV+vHfAYRw2XV
zy}16|xb^^IbZoB#T6*3YVgwH@q5*C`5Eu7U0b;}Gh+ZvJHLjOuWX+0)8!AK~m!#r6
zt!grnnF6Is8jOd6^oq5v`A^1>uqzxL=}e_ce9I=z{aUs5ByO>S+}iz!wbP-+nc0mT
zSDJ=~C@u88E7q1p8!q3-nq~bh5Z+*)(^qg`|6sC$`1KF|>yKdd4EX%N_ocAJqVf_*
zw95~e<VbU~CwHl!hhYOo^dH8DM(aWgy{u7quR01F*7Ap<K{&}{PN%!BARh+}(cpC8
zp!(nQ<8L;SgLa1{B!N3bPG7F2BmV0tR%qI>NlA1}YjQtP2=&8P-psNbxPuD^RH!JZ
zh`6Lhlnz~YE^vd%1)=B!TL&9+&lJbCr+41AD6{_D$%CBE^a@=v<BlFZmg~C(CCk{w
zMD#w|**<rI__`1TS~iqPUa}`FkKNR>B||EDQ_rRhuYI&-h<?Jh${dBuPEy3)=Quvm
zf~U~9r7>?IH7~Tft}b53r8};-0Y{3w^w5;qBJI14?V+|n=xEg_lT(J9eb@Dg)u*+h
z7uk?q`ULLid#iDtoy%RD#pbOwN{pi)*0$ql*;T?vCZ`?y)QNfS*%L+QeiwRphFjC=
z1cV?yQXwVIHll>({2W*Azw%g_;(`sssh+TtV%rx(1^23#ymtN+{CiQTO!<A19EY~;
z`Ms!sI#Hm*^>EMLBPK(0CW*1^cK^%oe|A*3{A#jA`YCfPc%dR>*P^~RlV`HEcYhL}
zdv>N>*aW-yy}U#tq{LaEDBdRV?{9y>lso$OV(XPN#$CG1-eCHO(kydG#x7;#X-BkK
zF%mdb5-~4)1&r~@o>;cz`_6cO^$%y<k&O49alPj#G~<W^^dz@pQXZfLX9J9*6N3VX
zKJrbx%OxWY#35BzzD^j@D6P+4eHXlG+8N@~!s?MU)X8&^-hEPRk7@9T2Qzh3_Ds%6
zF?@j75Gu>Jt<1?>aJl|ZJ)6p=Z{J6pWtyU`_WSaq4CP8)N*p;lp-GqVCYa`J@!`D1
z<fjkw*H>oXU}<Axx<|!Sq)gBgf|%HN;@QyN(b>8dF;8Z`Vt`9A!DFRecguvIXy7h3
zao2Bebgb_5sy!O9i2oDQb3`)gWWjxQiFuNk>N(xT?71&(S@giz648wO>uUgFQ%zP!
zXLT~QXbD<6L<6ycU6st3VP@ZJbhp;zEjNK4l@Ak)hnwEC&cEw<w|aP4b-7Q)%Yr>$
z4amf+{HVWz)QYv$#yZ4Oh0R3@mCU1)+bMAxp~bS5MWM?3-a5uiZAyzffX9EFU@+R2
zbM9F-wTWZk{XWi=b>-6jLz{N7^@CsPr-q{1v5$RRe)Cl)8tY632rYI;^wNk+CM$j2
zVwaS6R<<NQBOG0vxq=I+e(3Ap_MmFtKI6Z<HzK}^aL>B?l~iRd4zFC*)8a1347M7c
zb^T~p`$E|N?R2-`#`MH9DK%ejY?9+N#(|_9!QWpdww<aY>so%cv)R69)@oR#D4t|c
zlPT%{c25+Zofj6I)5bdgrgsIxkY&o8keEFqTCjKO%<qeng8QC%)~=gTt=FZFdros?
zd@x_sc=&x$_hO6MQIUGcV?wz@QKnLjl3Uuv@GOp_3{rIl+JwsvLbmybDc*I)LsB|l
zBe-wVye%==B2!4}fOq<!B{l72h`jwVr^Cc##rO7_FYK?V^6t;8o((B`1MeYNE^U(I
zh`q-?Lew%bTvpd+c{D0Lwm)A5-AvjNxAI|8|D=Y=MX*0^a2b&-ml6HQ1<!T|SAp}i
zDWc!S(L@X6DitCYx4lLJ?a~O#MPW?)u+?!L9_h#De>?*Ir{X0m8Jlbui9V8*Vxdbu
z#Mk{F>I_i^4h)qh>bav*K2#>O_}9w2Q;R!1?TC5nZlWg;gnJr2bl{}OP=tljo|u?t
z4C5s`*ch?TF>@`V6L&``jxIggvxK8nKdh<8k)jIDmErT_V`QE8`8`rrJ@vwrKorx~
zqwz^)U8-)Kz2^DG)q}_kyQD+pAo?g&CWlVfj?abUn(lMlyjR@9SZC(k;w9_`8O1fz
z%LNgne<T)1jYvq5tW3n=?+>%nBXJ#i=aw_x;3{N@OJ>Blk=8u<O1$WNUa%ZtTV=Z&
zovb}+iq>l1U;E%0mxZo-<}_|T=_FrA=<t+Q-@TKh(1>6Ae>q7E4fPSJ+$yvf+tTf~
ztJ3)dT}jJalQ>26*;1+He47=r`mbo7UtU(b>7#M@@MA{nULFwzOJX6RS(loNldOKj
zm+E~&i^t`R1|1(eA$CMA(QDg<XDV#tW6BGw0>pImm|PSu_oSo5V|l4@YR5+Ox+aWs
z7rhS<h0)0@)V}Tt_2s3?TZf5V&wjx>4-uXWYR;%Qzuth2Y{sP;Ikr0|`6k?51Zl)o
z!%N3B=7Ue_O&=Q=KRL4cnET7644jxq0j_DCcu&$VMnA3c5AVx#)Cxy77VTlyuDCrS
zL{Vkz4Jmm?fT6YwCs`7E@^Y#Ou8-JR>ET1Aej;+@IiQ-s-XqKHx;u6k;VXa6vZRR!
zQMBE*E?xB;eLM5y9<oo4BlgIgpLrNEc=q+DQ(PN6x+to#d(nBDjy{6ofwn>%{>V&%
zzY>`ZdB~L*B?hPYhQ}$vw2Giz-e%m@-pm^(EBzx1RV%ksXry~LlU6U76A40jTCx`O
zOMapp8$QxU1%u;`I1uX+jvhQ4W36>H!o#Ox(ZOYB*o0T8V4EMI1LwN3J)Ku#Y?1K9
zYvIcP@wUs(L@uXKWRsEA!%TAxx}Mk$s#cEQ8{<TyPjin>qJcsv!-Qz6sDf{L+$7L}
z=X<t6qr1_5pmmlZnQJ=Nv9mqWLA=hcqQQ__aU!*3f__rx_UN&k2U9(XRyLzq%JP+q
z#LkA`0RdPRO$)rz6LUBR?U{M1&J1Dk1}?pB^cP=jg37i%7~w_INB52Kl;3yN=5aY<
zPHjEuUWHeXHXfWUiJ<c$_zuTvK8sB~E{;CCn%X%@cOt+vUIlyg@c!B*GkdALQFJH4
zwP>lXf`U&gVQQWW9{62wu9G0I)wQkioWl5}w{z6;o6mmH(^C;TAv$mQ+*M3eO76+W
zj&_QJFyPIk&!lVuqZdbU*&deM;R;V1bb%nPuCz#0wjsI^e|4N{v$|Ly*QB?ZcPFvf
zDzZo8gfGk!05c*N(`*o*f}Z>-rT2<SFXj)@UPU*bID$ks+S|xfuA=x#Yazv=A;F}(
zy+f$qwo1)V+oLaXx%LK299wet99=Pe)!Ah0r|{rii;J_2B=@a1CKM%)>A0~eK5O>f
zm`p2oG}Cxe81I|1G==&pG?6ObKHL2m|3Q~=CUm+!0-2)(FMy}tTe+9wr6+<EuCF@H
z{z=FA(;{65^;4`2bMrQJysecdTkn3!ju@plyaqRae|CZtt0DGlDe-39LF(BL3k)3L
z-tbV6<_WDz*>+vk#-#JuocOPh>sY?un1~<SXO0Ugts<ytv=OM*33^eQ?4QPQQ5_w6
z18q!RJ^JqVC%O89Bdpj>M%}#}o+s@Z8{TJcE1)GCQ@!B%#`7e1iq5#svoU_WT+>Ne
zvMp8keY_+)KCguN+;dv#zCmx9#nR24!*oB564Ubg=h}4LDgir>wH}U_YC5URyGaxA
ziKlz15yiBQJ3L>$Kks5eVSMqi$w1zZo0#`r>1ZDxcpf}?bAy9@g4Y+*lv>Z8_r{&N
zvw0EwDlZ=+b_vaBNA-R^*%SCwnr&Pd^|K>VAW^%cK`v0PfBQ;W>HYbVPxZok!r|H}
zl~=gR2`Xu}iL3l2`|C+9&vGH-Qx5TnyRSih)Lr1aD<oE@^zLvLrRs{SP$A_`Fe5!S
zL3=w#qGaS?B{4m_s#HbG*kN=q+2OZI9k+Yb>fS&ho%_h3zPN$=J!Z{1igIRr_ejNk
zA*I)H3&hUQ3Wi@GE-#lxt?g}GFVaLa7cZEe-#N^3mw^e=Y<r^qpLVYOE9q>Dli3*8
znYu;SrW@u=p<H{|gABz7J#HD2;s~wGM>=Uxrey*us7SROr&7tUW`YUj1H)=cGc62h
z%_<p6d_<N?#1t0h0|Zo5UKgFU?vM8`xaa4y*4b;nKA(Ng-Y0dZ^p4j_-Dh{|+-kq+
zaH<z5xL<3kgiHp>op+Otf9@b;Dm~u#jADJQGD*NlU>SRh1;yTg4<nPYZ6bT=5S2ET
zKYZlTtF63{c*OSUfddG=SS9eE7>EzvGpokOjd8VKe|?3RGC7W$sKGZQ4m;h9X{308
zyrRo!hvFSI1Gkz5+3^TjKNlOX?Kc$`t4U1<jR658INz%I_I=QUrNl~5e^v73Quv|Q
z9|qj1@X)_+E)%W<t!_0#rO)v%I$?{_A{yGp#3jkXa)Gvq5fhT_Q4#Y8A|tI~_QP4(
z^UU8@3s!<dh+nSw;`qjF@)zgNiR2z#_j`LT3)ck4=iyKg5JY(mO%(2DwynUDoD!;O
zvy#b-H*0yI8A`TY#8!xV?*t46J4%EA2vhFqK)OGMST~$NyIaR!S~S}BhiWS!bIW~a
zG%%vI*K$nu0CEZ`p4k3a8-tfE$x8uvSZaZ3A9*Yjb%f91YU=TxejSsw!X8Ttw8Cxp
zJnSsafOhl^6|~#@`;_(|fe$M0Ol;o$RIzyD;gtzH_5&WFQ|3VLzRU@>z)coJchl!=
zC7YqHD_*%{qRE9HpNzj-i;pYg-3xQqiIErQsxDpLQ2IWwAb25*_^`*<Ka(F-^?r!e
zEJFMCxmz8181(hf=4lA12wD-W*;;=aDfea))vc((Kmj(j7`f}d<vaDXN7>>`F|Ii`
zh;?igeU}a8qBZEAu1HQ<@o9&&15>;Oh^-i!#(WyQJ~T*h_2?b~Gw<UdqXZm?vZ=fg
z4m!j3Di?4;-z(4qgMX9L;=Syx+%wrWsCfa|gtZDVvo_j7xhvbJNwauRpmt!*x`+K>
z=0()pS`4+9UY$w-=x4bhDzu{a%NpZJ(#FKq^%9)gV1qvXmvxMgsWH{k(}ya@n^@hA
z-5Sj^4gHwJ)_A3-HOX))1!cE7DEf(>x3IO5o^>6?*{OURm$G3|&%G#N3F!NnGSlFx
z!HUlQH`qk*-5OG3m~Z~VZQ_HaZtE9K+;LCs%Fy?(mnz*Q-IQAjhmVj4802-iXGc$x
z<U6UF<70a*ykbQjA=`cuNc8g4o)ZAj6Aj*%iFc5FlmjOK^l9BF_S!b@(x(hS9scsX
zN}~Gw?2{C7bl7<}GV$vHr$0B1TU&ARj;Mf;=GI2oi_@sV;2c2hcx`eyJ@5JN9rVVd
zPMBpz11UHI4-q&miIll>VF8hzi_5D^2xL7@^A8MzZ?%~sm7Y8TFSRoOczRp;#nm`G
zI}J}KI^bFM!!HBF7y}ZsF5IB%xK<Q~oScQJcz^b{m0EHex}SO@DiQ7Sg8&+5hzBe=
z<|&$g8k|N3)Lq&v`y3%tgd^T6-G}uqhduG>TNNtdM?W@qd5_qx!*t1F^m^%XAm>qy
zWipVS`pc!K+oXs{&r5vHKNo)gh;nw-JMT6Ow^f>8{wd!`pd};Kju948SGD*B!zqDJ
zZ2IQe5_hT(hcD`C8-sHg#P*I)xiaSy-KVZA{a)x3@h(-q5*x1jyy`_cLHg?sDLVXI
zr;Y+e^qKCKj9i%B8@#AT7A15XqJe;QFNaX?nrEgTTTLw{JC6XamO@Jw<Co=TaVi#q
zVE=;%Fmt^Wjv8zakYe$15jSpYjmF_C_J>s=3?DFa9}p-KStmA|D+O#A=q+EUpr=$Z
zHDQrDu)yaK>yfGXw*!u;p^1nF*JqHp;UGU3V0ky^rsa8Ye{n_Pw^gfepg;7g2v_!q
zn`%^8F9A8N0lpMf|1;+#ojdCV60X4_SAsAGcCq0|Si~_HjTBOam0X*L6Fug5&`#Dd
z=p0-A{M-4QxA{VTC<xY<FKu*fNK}Xfd}lfqK-LM<?4onjb|6W)+04N@-6U^UZtZFk
zvpE>29|Uf1#$&h5AOzaq?OzG<k2y)Pv)x;F-%WIFx6Q1gGDLBtJRlxpD1+fb(~mCe
zRN32%g=(>;S5;83W9rRqm-^_GR%5tHQztZAB?^7vHC6vlx_6ObMe-*PCUPJkWno<m
zm|0^Sruw#*jhKz40)FG*z)vj*R+w#-z{kSso81GndeUeBFVrv%TTUqxQ#o0GX};;I
zBhZCeGn@P$Tpgt-6zK_FrhzCem<jXu>^puBZ)Gs9j$kgI$xd&xj`9UZYNMf9O*0>O
zmw1MXB2w{bV->=Wh+TFFjp)rd(%`C>gh#xkg_!><70lm<#h@oukGl%Em)Un&c7K^Z
z@lxS&VAs4hnfqIMz|8F~aDTMarSyH-#viu<$tx7&0zTG1{dG-!>q>`RsJEu-1a1)`
z&pzF-mXMhce>GWnGk;<rqY%gW*}6r*4SkF}9QmmR!)V4fuA)uif5h=Vo*ON-9B9z#
zZ5gre_;VsE1Kx%8mG#;P#)-_}xSpbULh+U8FphT2E+yE{sBlwCq37Y(UzV^3^35I?
zs>EJ2Ug2?YDo^trW*z#XdI6fFl8QR*T?^E8Ui|;B|D}Nnv3+^VtBpxIYRkU)eVF~>
L|Nd=n!iE0;sXgZ~

literal 103834
zcmeFZc|4ST-#>hysE8(&Lg^e8N+|m}IvbQV%Y<Yp*|KjT%jj$|2q{Y#gd`(`kX_lb
zCs_w0+t>$VA7*~vQRj7?*LC0b@42t*dH#6*xQ<sDbC}n$e7>LM{eFKw#{^%~R@=Xi
zdmjWr`>$R>>Ol}Q_>~#jLw|Wc|InWP?zV;+61;+K<8m1qyzF(kV(bn<$Ajqq86w6P
ztsv+mbQO8Yz#BU|=pSWj=+C!Bs9bB}KGl4X`HsT~*B<zpZx%vqu2*>^&&r$%u@$W6
zZ@f2q)WlUU!(S{rxyX+3j812Uo!!8}%~Pk0FI=nZej4d>&~iFn3jX@H{UxeT!@s^g
zE|AX<WYO3A`|206Tl+kpc(|3%BeMy0;?oJTv3>W=Irlat^F9sV#4<E-mXl%Ug8%d5
zKNk3p1^$0&f$zFFIf&f08|JEk6voNzp<m=RdkX(ZI7Y_6^H~WF<q1tBn!rc6v?r>w
zax3KmmITmCr{@qfGJ=n?MnX&}bD4SH@>{Rr3%E*2^cp(Gp|DVWbxr=o(9qC&o%vQ>
z><5BJr<Uu&@TV8k?ms?1UTo9}00-dPo>Ut>XTG7nHJwYnTRw6)S0=MExo65_Bo@69
zYe^caD41rUUyb(HA<W0(=>zW3j8F!i>_w$rLi$mW>|1$!lsxlrOaX_p?{sEHWhI(M
zmAB@coH3`}=auuAj5S-$kFD5RSoNxyF*QdIKVY%$T8k8^Tpy7}*0!JM3uu!iZhiao
zkR@3$PAT@h$!b1kJ(?dcTD7&r;`H&y6TX|}-isvyounV}t=4{%m-#{>{AVp`iyX8C
z4!%T3dIpAE*m_iJIUYqLt?LddW1YN^wG{Zsc4u-#a|U92%4&3URHya*l#P_rz&gHQ
zzoeMhf(C4{S*=LaywdMesprfY8$9h6(Z?6*H~(c-JA;e^7TyegF-6WPo0rS&bD*xc
z&Qot9w=(jD(e$_DC_~egL=47vzQpEIL7xKJ&27n1mdNosN<^t?F-<64W0OMBp+>UO
zFHWW?8|JEo-1@$3qy2{f^pdl_zTPbMo5l=EmWY^Er3t0eV%!6+GFZ^4+g!-no>JFv
z{q(N8g>G#P1ERz&lTa?v%9TWtA90c`p@Dy-1G7GeB_}d5F-gCPvOfDxxJj097-rZa
zBDrG;oFtANYDnv_Eq8jeS{KTNPUG3u(blGpYtG3MS#>_-rV2XY3+;PMWr>|jkF>zW
zPx!dCnMgVIdysX{(9BPzM@c*OFUwmQ=O{LbI1`E-@g~|bf?~P*_wPS18Zz=JRZz+x
zarZ0jkkpTPf*}-DRn;Tc!F!SZxz*HE!FfW3@7fB=hT-MRA?W2fGx5tNMDH^nA{z3o
znv;|skWVz97fNsSU7eNvG4iQ#(nGGHbencLLCSG6k#o+AJkg5g)z^@$K&nrza?WwQ
zB;3gE?#FzDr+b)o8AH>jh~=l@Qcoiy%&&VM2`rtQnlk2u&A7KoqY`&p<mf{d+JenX
zC~*nIKsL1}*Bo&~zj?U)O54V`_-uf%5FFn6!Ma1}>)paLmCX~m=glj(%o805&9S1i
z70k=2ZN?ciZ4HfXY+Pk+K_qGms34=PrnWCbL9(;)9qW6<T6_0=Ron+*?{S%WbnNAO
zp%-tLzY*7$;S>2#O-h10%w0P*`3J_^&BJ5)L%E=ro15EgnFF=j!R)qz|7?&=zm%NZ
z=ERFRu9xWe)`=@f5ALvfm!Ef~zzYj%%J+LcEev^HxNsp)6kdOOnj#4myn?Fj)+*O3
zM8Rj@WvvUPTUKqX)_R0He34CFsEQOw^1fa3$8J_Ld{0kL^muUic%MzO;#~#{;Ex>E
zcyw0R=CbBot97R@6gXPsHr|x0sjXcJUZeTm$+0RYF0ZaudO0}EZh<7>3~cyH<a(~z
zb}!g`vaV;BmVoPVPXysUXm3W2?*&U}qO<Tmc3{ptX1vcdWVt-{c;7R<`}0V7<6@`5
zvGiE8&F2|=wWE`%Q^`s<7>4`ob4n&M`&Mam@6ENo+b&z>TIA0}ZjP6<Pg2@lXA&rT
zKHXq-W##x=ITq!?W$?=tp=cBlqC3yqS`e7jPZeuJl#tOuMEdd=n`7?G^`ztS_ddW!
zPB*iSW6j-$O5>6=DgA5-_dL^RJRn_^fQ7Vs!dlva*olU_g3z_#C_VSvw@17pyl7am
zw6(3$Oq)Duc1krVBkBI#Da|>dblFA0d+Jl|*0xk9Q5Gd#@xxHO=`0`p0FxJYL%8KN
zQYB@cPd4dzbN*IyTtAnn*_Ptqo{O@wK0kEZD3igfs6%y*W%uSv&k+@7)_8eyv1HBN
zflfmyRklyum4@o555xz)e*L<XDtJOT<o0Fyn!gqd;R#Ls=zQnSom=$xse?|UXG2D)
zgJoXk=D@d0v>5``CAq`OSXfxv<+=k}t=;Pxm66!VvR&a5A#vQ`8WcKTur#=7sN0c~
z6n(9Q&tmrprHi(G-P)w*M^<b6ngC*>vF^fUhLh<M7x9%3HqyH8W`|+U>8bb7ms%S{
zJ^Xgf`ue&R2-mlvNl8hPuskKApFX&r2{~}ztJ7Gr=n4A_DwXQ2R;wxcVenhn0W0IS
z@9i~UJ>_j~Y)%;IAwj5qP-Hj&tJ4hAh!OUwagd(}Vd)yB#nD>iY(WJfE(A{%Ztj$o
zV*EaItLJIGTA*FMYT&#s^ol*ej<Pz(#b=)Z5}Npj?S<7je0y?hBVr{}$21;<*K^88
zg09P>pF!n86TY4J3|x#2e1ui4NSZq=Inj|IOH8o2t0T0%;Xq-Yh_Eh@JqSxwO}lUt
z*if!fmP4&JC102?6F9r6rl#g>C?@aaX2AYxVc-|6FJ8Rx;<InvQcZL1(GB1J>d-bg
z<uU&<v!c%-VEWFc??|wKl(5nJ;X;3Ze*yAx^;Hd_?Vw}*0rT{#K)ORIUmNl@hlwj#
zMY^AuB9Uk_JF$H|=jKLqV`HOX&drjghji1@OYmBFSXd$fkEdkg3ZU<G(_LA7JUl$|
z&RsZ^_|P-v>P3BM_=E3ipN-AkfSKy*YTrt{tD&i9S1kWX-Z3f1W4ye6<i<PTWh)r7
zp2v^ZS1%hPC#_+>bFa;GXD5^|f79SdYRM~D{I0L#e+C}D6@hmsyiw&vnJWf~Vg>6S
z|MKNJ5-AZe5TkeT0SorMq<yzSqFxgSwzn^$^!j>wyk1T7LY|%Amy_uR*AtoO`E>XL
zWl{|xXX&dx)%NMrr-F!k1Iw>FXp)^pSTb6^BRx=QXQkY!%eqpuQO>C>cBjgXiC`?5
z=6}bxZNt&oZ<Gg3G14*OwX?7ou7NFbiaXrDy|FaDhIP=U$Jg-kHTp@GVyk!oh(UWl
z(F%2M6BlV9m;d<jL#LrdwA@rmR<>kqzKm44<$aNjjqM%DL#_;d`%@%dPd%9gte!SH
zmEi;;x||S#ac+A*Rk1eU-cAf4tp$+A8>TH?Toz`*rGqG6<Fkrwr5g*2y0&xfD6*8N
zcCkEs#6K6kJ%d<2SG9CbT!jCHyIF-@=bMx?&-eb6)ldNe0UF%eDBilCFUo2oHyPCY
zl$}gN<RO8pQfXHpw{S>thjFOX_NVd6ms))KBXnqc-V5P_MJLA!hKvAP#RCtp55GBm
z8(3iuwphX+G36nbTe>-&l`PZee!>Y{WUDN3RaeT(t*xrtRzfKNcml0%b7iIh4p*Y|
zwTt*o#My{6%9?Y*XcF+Y_^Dwm+1auGLnHUo>b*SN+&6}7C5lvQ)$aTHBJ@2^1eSvH
z)?>U4b#&T&ufH?8a2@QEvn|Gl9uc|et)p(|0{kxdtsKX-xPQPVZ)<Bf5Pm1&0Y3y2
zNWj??`o{up`iw_M)VFAKDv(GZ=4d6h9jTWA`p|L27tM(>Gc#9;gSzIB*3k=7OWqQ7
zZ$l^~G}7#fUq?xi`U;4jjaS|E02<;8NfVAf1Z$1@27prT2?g(kk_QhSEQ0@e4_P>r
zV%>*$^l3WQ*sTZ*3n+F7h0`ey#Lcq0Zt%biUq|I+(aY5swisFaWN~q^8P~>mjZ0Bc
z(H?%=zis0Uh^ZFOeUO1Qa4nh$ywdfKuezNIa%;d#{@!SP7)|m`2-I-@x3Bq8-)7*q
z8!xQ#w9C@6wGyQKH;z({M;wAh>fvVhCWP@w2b@Y@4sFPr`9o&&5smOgmPQ<Jj;$)F
zs;UyFcvSm+pYi}1&p0MYX_rjPKPIz_bk`DYb{o~iyS347fh9!|=GN(llp=0D!IHlS
zr}<@ePTl+fd<&6mg=psq#|WsYsjYcsR=xN}Akfaq5?5f0v%;l5BI)!GY4sSd@x2WD
zS5jJLYl1ZT0-V{V_j`*h!ROs59m;X}uFY3S5C`~R6z#DAp>%-!yq4G2XxVg-Ya-w_
z>qV~R)NwF+Z}G{Vvbm{R)<taW2N!_8DpVntzWfPu6!G9P$D1f>9l!<WC=?2dHcH8Y
zYE<GjZg@sOz#P0b9V5^f#mra!0w;1zBN$sevkY?s900&5M`d02VL4tc>=PBzITw!c
zHb<7DH{E;2u$}Y*-w*O-F^He2Z}C{NJ^yW8O)Xnevuv@xda|=rf_zT-Qd6uj!s84+
z&r)qBm$sRUeh&cd`8PukUF?upK3JqXRaJc|3%2+eOC|#Ec{UzEkMOt8Vxbyd4>jZ%
z=40XXN%}5jN5`;CpIZAY_KUH`j5{<7YE#a}F;(BZAS2^_J)=`rG<`WAOHN7Gh*hc#
z*xt<JUpx2rAl6ea33&&3ic3g*DDSX!2*X4G3>ENZS^@IZJyaJ4Tg(A<bnTr{-pJ{m
z7>KM}HfZ+2<;Q1VJ@tnGQEei-e4q({DP%(E>C~-_1U&4H`c#m6Tdp%ekQhvq9<PG;
z`vHKu&0j}RWr$lWVXC85V7VgHY5{T%U7PJAxATwKcTvwn(<z{!pKo3YiTc(FOMxJ<
z^8g;Fdu-54I&3i=M~1#%0to<^bQ4RK-T4ITKFH+}$dv^)iz9V3LE;t_<L#-gPioSJ
z$%aKB66MCdC3tM2Z4wJ5^lrdl04tDMWf4k%Fxg+I_;CF`RT)a{oO&Il7iDGV@b?f9
zN{@067n5RaiD!g(tt~7p9NRY9u;dvy{6<<Dx^OH~t!jNKcYb3!%j+=VO1qNy##ZGE
zQD5Vk#;&X?VrDd<YN`4(SCyw*M`{nUw#o-&KaTS^K1|A|rKM5MQ9_~bngxBlLe2S|
zwh~XWK<o?15`RSLIUC1B(7A}nnKNfrB@<fnFfMa@#5B%k{)GbCTY{Y1jahd=N*v*p
zTkg%GjfFnsLM&KBHn&~qS=0%$-t0f@Y`$TLCI(4qivVO7rpiym()#EW?-q#xdYSJ|
zY_kH6tYmR>m-nT%lEtr=InGbb%*bD(V+^}uyar)wLYhw#AXgiPvjvjyoM23HJhkQO
z!Y&9|3szF;5R4Clks<yFBB4h~0~=G_>%Dn4)Qs``7ouCaG|x4aWZG8J$deiLQZ>)e
zLK|ki_~FBc&F~TPB-{Yx-bhEwIdV+OgPJ-Haa$}he|#2H_L2ZJ55)uUKwQ)}7C3RK
z9I??!nBJxZkoB_{^V%AhqZ{g*P*ocy$pNigRqF|n?gH?DTfi1EB5HIw6OuWz8=5Et
zN!&EXow|a-kf~cK5mf<{Rm7ZXRjl9ikZ2;!#(Z;d3NZvCV|~JDd56uqlmlK@>{PYA
zs+!iZZiK`Zt^=mOHB;<J8FgBw%(5UBMp**L(gCZ!2v1V7h|ls2O%Z-%42zJGk?EaP
zr9ti;{{|NJnfaz6>TlS#u~c7M`+!7Q7hoytBCT6F-#77Ttc%qtpVA^$E+Wt)NfJ`4
z2FP+*_)urIJJceXecj#>c{5V4G>1;Ep1Thngz?ljG>Fm*is|z5^1EQn3c!${posMd
zL1ND7QA2#N#iiVU6>(9ZXIwU&m4?H)TUCYES350px2BUQa5%{CuT20;v6oBK4#5aE
zU);gd-^g4Mu)m{Q3$y-8f65Ge#96w@ZT2~VrfC>M^)Zuk8BVxxVNuhPz|X=DH=|J2
zwy%1~wX4*cFte~!<t7jrU!MD4&_np=LjG!DaKWOUa*19@<nk=~_FH0gr{&<RZ(l3I
ztD)I>z|F*u(zuAh>}IJFuUronxTlI7!xB1L-5FXApEJzaY8o3G1FttXKG8AhRcB|C
zZ$o{38)<HC`#PPI++LaPiU<R^wfJrqE&+tyj)g>{tv0$bnA7N$-XW*8MBruKTZ=??
zXOrbZX9aXp@jZ$k<X%VjS#6<loI5zMdgkm-w6F)3(!=JN-r1}+hM&hukGc-7U17D?
zJTaVn-sgvcv&?ZK6Wu+pK^B$wt&qre>UPGsQ&ydhlFvyvoVXRz>8eEkI|RHp9ZUWL
zgPFL~H+7HD9<acK2pE<|kJ{*`#U(l0hlgpYB40<fO|Nc16Mv+nq>QWRDi|HAd$M!0
zx`(8XsJIfoeEIU3{$2&-X@f&`+hty)%F17?m#n8gWr3`%?Xy|)C?hR44=oohJr5O4
z_U=q*^RJvMDQFMqTW)wbsKi%&6KD$Es<nrO(r4(%1Z1c4u)JG-ep~rF2d5`tj8U?C
zW#*rU?&^WVRYgBq_!P@)+fCQgRLPIgW-R9JpR(Q{l=|n$3@y8V2@rANg`(Ad&y%XW
zpNc6lYKluSdPFcnWE2eMx*$ocD#9PnU}C8}#>v?(v8uQv?L|TOc04BVfRZRu>aP@7
z@~Y1v9Er^6uw^s+N&0GFu5X2#)pigRCnqNbep?uRQ=YNqIXG%pI0n-}ZYOP~bvHXv
zOXST0Mq}_Jc%5(rLF)de-OeGHD?<L2cp}qF)_<AiNQ3k5Xf$T(LUS05vkpDi68vUN
zM5n9?J`V4mHt?b6HxLMFOz};bO7F!w*kWQwM@PFS`R1=2_`mR{m7PMF2!j!}A`5x6
z$4gF7d}oqKv{C5+ml}71rw!e60!4Hoi=ng4OFi(BuPKHV)2)Q{4n9+#vu%qjo--oH
z6`eOYRBMgYrwGX*ps@qC79>NUk3C(TIt-(og^%!{Nor2q6$_u;OFKvfd|TBKeCyyK
zeH}gJ){T~k&DHartu$ven$`n^i0YdXbhIEC!V;P)ldfaR&?2d#I{-Y<pP!$fm0r!b
z=I;6G>K<AHVk>16cnlrOw%F5ksbbh7LJm~p*(~XbHUA(p>QZcLQTyL~kw&0R#-RI+
zKQ0VJeM?r{Uhbrovbr)%yt|%NC<V{CB4ld#lLtEoWA+0TG<HYhxqttDH?kIB0<AxF
zx}NTA6X`g;(zEc@Fj3>Bep3>ffIuv4_9V+^Y4e1sB5T)2>bS<Sd~+UhB*5?z1WF^h
zEOMrQaeD*$Nv;YhB1E|P7~UM?VJKAt_R6iQqUGwVJBMO|1b~D&nT2rg6h-)rM>}hU
zP3$$beXT0KT^CIoUa}Doi1m<5e2K*laKW4t%<k^7Fte9R;D>eCNpRQfFla2<1<4Cm
zERr5ib+xs%m6)TUdwRm5g4fTRrtYaB%SGWsi0u-o`O=vj!=fA6Iuls50-%ZF?FvgF
zDJd)>_ZO;2H(eIwN7dw={1!)c6TBM^+b4aT&G*t2HeM>lV6JDa1OaK16tGsthcMk>
zn06wzB{l;+%>QJ1)Xpl&66-H>D1LaIn@dvO(Qtrf<4;Z?_=(X8^X|7T-I$MO-~o3x
z#RV=g1a@W@?GD6X@Xl2gi`xX+cF#DLriH|E!8!z+^LFbXwyU@TD0p+<zE2Nz>ep+%
zv4z9$h(JK8TQ4D+tq**&BE%+y5zYIH5X3E30O7X>!J!6K=*>w8%ey;J#1?fNn!a=Y
z8+8<*1t!>Hr!0}DNp<wRB{2yve{a5|f*`w@0$6Ec=#lfK<Q)OoF63toH`oV>Jp@Y1
zW!NGQ_~%9NU)v3ECJ9-~<KD*gI_k(cR!v`>peeLpt(G=PBB9^)*V4}miN1lL2C?mf
zgioV`Re~Xa>DU9xnVZ)pSsg`Bv0<u~_DH|i#l>slC?XJ*fY|&L<F&H9Ty9era@dl-
z@8P|&3R}>BSbH7CY?ysxB^cn;7OOCTz>iJFM17l3^j*Da<ar>l)VA=3J^u&*x1PUT
zAO_Pbrlb^LyrY+8WY}mS-?hsU=*|b!GN@Ek1*H4(J3JOQGCHrqU{=Q5JPhpay^?8*
zFpKy+*el;d&@0Ll{*gwIRkl$q;OhRD)Cy|(4OG~0j7b)|8*a8YFRjpvG<2G-oFpX5
zdx<p5mW8qoi^&pIK@qv(391|&6S}uK-F&>^Xs_AIZ)W0oV0b-oarcf1ExHdurOKxd
z@4RVK-dRu3!C=+{d5KOmOv~;KS|!>Ca<TpEg8M6q(^VnIp#q!s$2&Lpucd)k6uO$%
z$O%E6zktoR-aRfaYjTc-ML?ZT!=S-52pW{;h3SNchtKEA?NK(BNNA13OakWegC3Yy
z+k9l;>;NBJ*?|wYZ{IdYKCSz{W)E4gm+-^gkOMb7pXpmQgdfiaT>_Czy68JQP~sl9
zmIjp@wUu-@6~_YQ>6H`cC49pc@EG0-z;o-9F+JCh?~)0y@%!1ueklo@|Jv0ZYX*ux
zW85($gna;(w;41g9D8z1y^M+6!@DnEzT6y|>iu3YWdDu_0cR8r8L5nnjC@Rg9||1t
zE$Co?#q{?F7fjbYCo4_#)Dr;7a=+De38=FFnqoRxNmmk<c0tu2v@>47;VE5-?0?Pz
zH04W3G4f8lw^tqIua>anS6&l@l-EFfy`XcPEOt-@fW~_YE(b#B<4>r1A{=ws%2-Jc
zIC$yHmoIy{V3rqFwz;8~m*+7UURd$<<6{PWfDzk+x}kw<W4I<Q?=^pOVp|Xrjd6GS
zfS*g>@lkM_62#eh)JsB9{da|tLf{$Ugk$DE&1S8|@QT$yZ^yF3Y_EUrgltX%Q{?~I
zDe5j^_Wty*V|J*N^#4Ad6&N6~Y`{1%n0(=|uzzmFeDw#K;@ITmBo+4%fH=SrgE87^
z9Y=G1szpKkBiH^)5mjo9H2tfXp^0bmaJX7+Hi|_Sb#Jz}qaVP&=X9vr+LkCE>x0-=
z9XgKX0Y%xaOBR}b1JZn}Y?<rfiQ?=Fk`dmHU6%q!z3RDkY!zs%4^1*g8|=K1VG{aj
z!0KaHKtg&466C$Pe$QCX90oJ7zhQvh9o{Ge1Uk?z<2iQB*oeNsba4)NFM4f$u;&+2
z3Y;%Ho&Le6JxP%Z1i&kif2|i5IIjVN37yt~AkCSdby~xczxoz?nYF~<aij;q1*sq_
zCs$ep<Ppyf2Prjp>R&lFXg<e`^Y27UaB#3c?7|U!LM}s~>{vMF3(VEN>ymafy&tnW
zKUnsfE*}@rx$;m1<HGI#q{IKD!+%PL|4D)W7Nq%43j9wB{0}1he?)|IIz-=IB>>5F
zwL&O8R+zugb+o<!_k{_i`?D^$31(l<>X`C4Vvxv$160QJI%wR>5+y4tDtcZ=xdCy0
z+QQb>)(14RxM6%#KnT;M?@I;Edj1hp&~G9c=iRO|qVIbJP{EClE<Y57(m#RbhJWg`
zf*2ngJG%%F<b&v<A9q;s7}gM3>vKU8y)_s0Elrjfq^E9fE0tjWajEHCm9gXP@K{F3
zgYZ{b9~A@$5<u-m!p%T?*HL@K2(ToU3~<yqoxZEo#^TzTuCdTNUv6z+@P&v3pTPc(
zFLLzNP@gg<VhKrlAs7spj6lp}fu=$@rrEuX=Hljd5vYq^_v8}SoQKLc$QiO11?UKc
zPML^mTK|KlFQLd_=kQ4X4rpZ!V96wv+K9xKMPMy}P#cDahxx!*Q6*(MD>opu88-D?
zwToexP2#^L2eIXkSS^&A|EF%eM;|!a8WGDE`~tg$xY@Q5l+Zen2JVUSpif}b;2I!C
z|Imw1+rcBz=L`d^z?=(s&DpZp;p2dIR#;kE64yXWCX0@t^X{)q14*;K&7|;dS1Zcq
zfp{x86|kFmFy?ea9_7<ivZZm9-X<AzW4a178Zuu&*EqsnZmO+?OeLq^@_8d`KUnJd
zBG&Bw)FN9rL~f$XnyYRefm|9w(O@coWe0WBf6-I7>XIGSTmnONZmw+`Xw}+3(q@<L
zv~BPe-^kX`Q_lvCz@!4;j_yeTn^Of~*vApM)rs+55^RuiYeQcIy{CDhz3o@U8*hdc
z@Q;8wFFs^#ME&>IG|fs**qU0c|Am#$$QMm?Kd#<u0vs>5p@aAvdSxmjnVO#9T<J6N
zWq+q*1n8ddfu_JpzA)mM-e#=$a-JYj;Q-9g;~d$Cn8-!gjA1!#bD3dy`Njbl4fK|e
zHmohz$-?q{HZz`LA~gAE1DX4Fx?)&k;4z|LwoHuP?DqwYL=WQPawma+I038r2-*gN
zLg{5Tz2kFq-)nxOl|bFPQ{XgMx`MSX*kIy7Y%LJ{Yj}nZ@JKr;^yTc?9oRWFi&?*o
zTt3l3^+x*1!Yg6=URxPc{-Q63)>5aoxnYZy7;i7+mNM_QE6>@#1g6n6qeT;)eyFD9
zNy3%Bfgy@GdT&bwXt=7iWkAguD!5%2lDE_E`mg=AVa;dm)CRS_WxjB)C7KHuHS3oW
z6LYcC9Gu*O3#H@vM+Cw%l7R!HTNzWYsZV{wtbf9+gOL+0poR=BJmCzhV=HGGUcLfb
z?DM|0gf`Tyu%gUjWJKS@F!djV(N4iZ!vMhv<AINiU?zhxRfqL87*sS5@2WfiYwe6m
z<n;PAq9gK-&X_;k$S#Drs;Ezu{$k7;NQa|*d{yu0{@T&mS#Sp4O~g9m<XEwL-P(lU
z0r#G00fv}H8@WT#652+<7HioOx;fa{ll-xL-RV(EJg}-Gu!P(9my+ifcU4d2Nxi`t
z!eCY0`pxUSq{JVVutl2-r~}?Q$N^3m0kKUakeoD2Ptmm!y&a1CKezMTp5Q?TqYjM8
z01dAyLGMPQd*!J&PRvR#_4*BC#=&R`E9(*A^|W)j<x4yQzDd9-{B_bgjQr@cABaJh
zn*9P5-@Vajv<>hP&>!2*I3~5GTAQ5YKmZzc^_1)AGAXw{Sc}Rxr4L(_hq>x&DxL-r
zs8a*W^Y-xYK<`{1(66Jc`u?f}s%IAeRlo+_5<0DW(&R{}-1Y=l3p6i{GadTs2m<J;
zn>64J+hq5%v2AT**1f=A&#~Jh{E7|7u;!6^)1jE;I4t%lm_h)pg?9RUXmzlN?4W~`
z;fW437?<G=(@DI75VtG5kqk(bxT7BQcCQAMc!Knjx{Y^TdO>TatS^)J;;j%^xQ@bN
z?UUWCN6L$GbDclzq@TY|+=&|V9kfuInVa(`fj;^MC^-bM$jYYg%qlkXz(&5FeZ$ke
z%XpYR>v9JS4;mtcLovI9Xs@G&(C`;DAo6hyy0<F=*P-bMWy`F>M875IZO$sL4UoUm
z2d5B`OwIXzgSt}NxtuKv2U$2yY;Eec`E+Q`sR5xDeGFEm)Y%?extva0sb=D6ks2u{
z?hWi5-H^<!9DH1vltis~&Fb&E@104p(?vM@jt<lP&X~NMx5oI=9OO47@1Ik8$9OUa
zmJceH-JX!A5qf};J4Wa1LD$!`ZNvRS^V<ChY!8?-R4N4yQAi){Om6V$U2FS#-ORHK
zMTpai!C*8*O71Kz0>%F-7}WT*$ctX-G9-my(kfng1O1TDn8E*<6Uu0An<5~-J+AE5
zq$PPi22u(Hd1nIaP(DvJ(qac5f!i0@-rlAydQ-TM9yQ{Dp^r8sZ2=Wizq6_;VYg5^
zqJ%PEMhPcw38k}I+XgHnV-LnRQnx9-;cjgi)_Cc`A9M@MW7hGo#UU_uj9G7fF>B+B
z+@egiuX+5oL-Ac+g4_TG6}(hxWyKvTvw^DJ6JDH4zl=XH>&^tLSkc!}N)k#+LzS!j
z_Oyxct!$IxB=>G(8`Bf8Vj|Y&#y&79AVHlR@+DPF4g!7o1>yYVwheKi^q>e^-(_za
zJE0YH;%4GoW4k5Y+RW!vX)Q@0JQf!*>xCGZbMSjAwe^^F2W&3|+JxUe36IgCZCIMN
zsBs-uaW!f4(E^P-$-%d;8L@kxJb4lWfJZiLF%nE1wHqX|yb=MT!UgzHS7#?V>F~Mh
zPf=hWJAGF7F)P>53q9_epfU>J+G@7)!LSpV#^Jlh$A|Kt*iUm$6#wlq9C2ums>&nv
zy;`czFJ{fHir_zJl<{1;=5ppiNY<6Y^s+&xSR2P;%F(jH`9VkD;t2d)x6}D(tE&^r
z?~hnH7kcS>;mZmy?O_qOHMsW4eAm@o0xh}jis(CUFK7m|k6}+Gugydo)|9ku1Rzot
z#oTjb2`|*s9XxV32>8SwF6;P*$2CxN!H@%Yw`km@E#!-5%C*TnhX(6-h6o04;e)~7
zq4i^>$%<d~b<alkhkGR>;!%!{b7(dx^0l}(B~ti%4X(3(tyeGCoDNiNkcX%eWd93+
zL0`8O0%^M~_6Kq>zT_CzgJPkNjMcZHbI_Jzpdqv+QosQXFb@}Y1Z&s4zlxl`VPZo3
zc>EFO%Q-(Kf*#b2yi<WU#$d{kC}~Ecq${V6lqEg}f7}W{9}eV3RkY3tOXL~Fc)CuE
z&>kB%nePp%A@GiPqj9*tn5p0mzsjXb5$f6z_Mo+wZ+s31-XX=~0`l|n7B63Jj6Mf)
zWx~GU+=3{xtgI|~xhB`Q91hUaB=c0tF^K3i9>+c$I%(Eq!xzW}@iMvaR>L3`Xp6Vn
zkl|*G?3oAiPd@pl9gcsZ0x2VF=T0J#W|7(%yCL3&A0?5Pw?QXzkNE_gf;SD9L=L>S
z@5woca%2!o^U^|Aa<d1~Tm?rnac`#)Hd(<}TEU27&<Er?Ks@YBM$y^UN9)&~T)duh
z=+L2opxJiG^?qBMl*&8?vC<^S<w9V48s#?4q9S}g<5oN4*QlVIQ6{>OfMs6q-r|P`
zU_OuS;vJ!HPi`Rx+kQumoyKwR{yHwxrdwP9{-$y<PJ!`c`21QA(#OWQ``w+3eSdhx
z4p9|hCFc(9+pf9R;9zUJ2lHjiA^n=*a;=NolbRoj@FiQ)9PYl><;yj1bC9rLWaS_8
z9B0?=-EWE#GEo(0U5RLij{2KI)YD{7Sm5qS>LZ%fq}hopQ0>>jlQPW1#vY<vNK2%z
zFJ<Z5w{MLPluu6FI|7I23W|y%-lJ*MxajEUL?@fy;7iYyPrA&2I#_vtjd{m*Z{NLB
z7#bK*@NIm4Is@F#C?UtR%=(T9)f}7d=OuWEyE~(ga)XCJB>Vfwj3$oCw8bLF%2!sj
zI&bbBrtly06*E8yH*d-X<}T+ohe=9G`sG9hEnGj#{d5Y#MJC9rN(tZmxS#`ucXga9
z=8Et|N|f1xh@jt7mP>A%n3&u$H0){EUn8iV(Rut`L7UvnfBB*GeUrzzd~s^DA%b?a
z;!({lSua0$q@d4Y9amL4uHZgN+GlC($p)Dyz26HY!?mg$9LeX|dyE7)2jb{dJ6AB$
zj9E+o-@CCzX_e5p$6(Y?A+bURD{6&>g~i0XRObM4f;n|u6ur^WW9B<|2EymJUS%3R
z4oMtEK-XfkK7r<NaDr2*onqh$-+DDm_p63{RZ~;b1mu`JZeNX9^?!5G;6^TI>m(^`
z$WLAdcX$T1-PSBx#fDPMHM@CpE33qv*qwQI&}9M*CR%AXd`-7lN=2U$M~De>p3~ya
zU>pT&X2}A?)0wQC9PRN<l2=En%B|*n5sfo1Xpd&?sZ7;OK^%Nk2a$kE_Z2V!#hI9x
z=slSdwE6Nm^TYiR_fx(!idT9rtvYCAv>nvcWiTB+9Y!CKy9L`owWp5CaKyba-u-Hi
zGRuFpIDHj33g>0o^q_N*DmFGm&fF^MrhNG5A)Bm*+j)(X)Y^^r6KHa)v`Il&7EWhu
zgs9(@%((tyXf#>Ib>tu<N?anXeb3a_H}_xZ;tTomxG&%gB*tdh(-J3M^TVaLv98@K
zhq~OUL+h9UhfY#u*?&A1O6^#c2s{gMq(%tDMtBxEI$!U?DW80=jcjw~=Y(k~pFe;8
z^pjf-OfEVn4-7X*V=xyT@I#PAG=XGZVDX7DCsqTw>s3if$=mnuU4k%1hYufKD@zQj
ze?#wTUCHQdUl~%e`}Uf3ba7p&ks3He;Ed$?J>@G?O&mV|og}U_Z4Er}kvc-JD(i8P
z_+sZ|$M}gEV|5xoFFWmQ^XhK}!uG*9Cx!z@EqYo(+LrezWiHxZEeG9XwB6;F_eLI8
z(Q^itY47N074#e<SVJVA-~C<_*}aoYH4HrY1IL4rxAS%~?Pt=OA;VkUF8lxpxONK>
zDpMeCsLFGu`<>Sww$JItMd|LYi+s=<o#delk9>k&r=<LNe&y7XJIYa&#zjk()^gXC
z;iXC;pczL8=)iaq6o^@1=Ygj_^^iLc4a=~%har)Q%igVD$ow!}=Jv2C9b*w{^5ICE
z{8_p5_wN_*HE}hrSn`RpY{MHro<?hdjaNB!?H+2*Juc~iQH*3J$lo(hZt=3bPfa#$
z$4wtUefo4Lz)$YpAITk)x0nJ2*HR(v`rCP&PAnj+?CX@kL$CMOU`pwgL&HFF->YU<
zY}!u73AuC8XO}YXASpL^Y3dy=;6Qc8=3Bxo<8Eb*VXPYR&7D3!>$jt9!ycslPi|=*
zZ&Qx3GR?-Rsf;;O`dS&TnHv~Lq!ksd4oy!_e|MQk?Py?DrXHIN2y{}a{PM6LkB{X#
zbf|@u`Jrq{b|}VHfu8c7si#}V#KscF=}Ay@?=HsT`q9O2$|o0|>{vfFKYz(v;Y6~j
zuC8kt-q`l0{5iPBU3>d9u&OeFxp(l5uDn^5l>xOq*U0ZRwM*BSQOEV#0NyD+c`$e@
z3r9VoY=<bLo{Uy+WeckDBHlg5&#hRF!J@sG*6$|OPh`8Ch?)va>Un$`9zTi0;Wlq%
zJjd96J_ut`&cMrpLVM;Y+tR?5ji>{Wr&F%IT^V@t{P}a|3NVO}%;aL+eq80X0|@tI
zhJ~8no}z~`wvRWISoGQzBn<dI#9)ulege;C#7C;92V~+5R|@dju#INThmYz7hZvrr
zb4!kcF8!|!@-W}Zo%UJeo1`jdjLQ5_3NNt%?L={A`3D$mOTJ?@3kOH{P=&8Q)MjFR
z1=XG)o$%((o07&m`PsRlG`O2sqPF3qnjbtqi|AR94mZ}AAQOgnyBGYS-EniyOubXr
z1>wK8)ZjCT+Gv-#{=)3OyPlq&RpChZk-%%YdY3N7ZS4(=ei(Nd6=Bt<VAmXTo@(M$
zV)Zro8>u-{KjV!RUIfl0B4lgaot|mXq#DD_82AC_6r!3&yKZhydMY8{!^8unjj{#u
zF|6+ojA<#VfuughV3z8O=2~$#4)593J$}ny8lHL~Gd(@MR_v)GG!GV>qa-m1K$yQK
zh(A8!eSq!%n+y>P9w1ZdH244Z`f9Pj!C(}0X63AD&)2UDd8MVL4J`2>RJit!Zap>(
z2@MrxTG%=;VjZoXKCyg(q_KUx7z{I{8VRqO+^O<?rG-N18XB%OT_HynyX<|{;d!9C
z#wqAkIsO#fGdnxmSf?I|<cw@5n(VWP!@Tp7XN5(^zMWQunmL1OZ|}f|#{^O?<Fzzg
zaY^FiGf5yjRx(xF6m+rG*uEFe^0FLOcw<-Dt2R}`-`HJ~@4yop8d@;uv*4R&<iHxJ
zQ4n<^PR{dgAQ!DcKnhYdubqRUyci+)=O77AnAPh}I|iu1)!XS5d^)AUUNwm?7;_Q4
zD^^MCNM?Lr5G8WGE0^8FPw>!v*6K{*^f3OX%#aJO0A8j&*C-F(-E#pvG}kk{l^%8y
zPE$kXWM^jnNW_dD0?Xr^jAsxe{TmrU)zsg=A`R!Scwn6rA0I!#F7X8C)R;e)1~O7Q
ze`677z+HqHKTBVJK@AsY8x1m+O5XJzerR91+VkM@u>FA@Yx@FKp@`DskCxIZD*Sye
zUHVjRA=Yct6OFM=75cxV0<q!C6b$A}NK;eO=%TmZJr$L@IVYQx;H;t|seQvq4uuA1
zyI$e8hf;1ItzT6Nukbxp@@;C_tH2TDUK{fUqnN@2%2ssDrZ0GCn)~XBB_&{Rf*s@u
zQQPKDM)7rZb!Fz~=XW|0Pr{etK@wnB;!VU%Rv>HlGCxfGxffih5Ih_&YkQkBQl7!%
zj*!31!QgjOq+UASv4a{4qNHxb6yA~X5pFI$Qk{IK(Z{7VUeeM3cm{4J-(g<HZ+(Hk
zrty*`D-VSH+g9=D_Y@ISg^-CBvIv0oOXdyMN?Xg19VAiPb=P7U`%Bzy;&4h|)iy+#
zWws`S@kQM|Jq3+Y4C@2z%6~h^m<qr~9L6y#A8`Ti3^R>TRiObVX2IwbpV0@~$o(h3
zei#=d|2YegC7~Fte%#5{cKwd2>7X-9(}7^*U0|$s=+JUb@Rw1LRf8+T3x)0Dab`*5
zA4?*yjCDVlIFaKaw@n6-S46b;<5rM@Fm5e>3u)I(R(Pt@UkkqXmpKIh|8L<}nW3z$
z09rhdCTa)UE+fZwx3;#bJbwIG9$AZf`}QLNd=66DKWr2;S1c(d6;M@FRCF4DX=P<a
zTVKC3#Ng}-BUE7MgS&OKDi{ezc<~Z>Y}%6?OU7eN+LF%A4^@;+0+eFN^Tbs;Fx`JF
z1n`@HShyQ6(H~Ak<S3&I8IAe~dHlDI)^|a@&;lgFfQ<4@K)%i?Ec6V=UjiBC^Z&N;
z?}C07oMqjiTgKN(p_Hm(pzy{*H}UNYJ;$CbCICE=bLEs94iw|Zp#0wxKH}RT_OxP>
zGE$X><E!}bghzWd))YChVwkZQu^&nu;Dnwz%%R|^d(S3nvqCcAHSVDPb10gLFKt~x
zMtySU_tCw2wXyiwxCjI@hK38akMpnx@m=!bw0I@CmJ#~J57KAu6-~kM;j**+?+0!}
z!`lyggX%BPS4VW;@EzeJ@;~k!0K`H}t6;x!zu42azldWRRLa{RJH10XuAF(H2Gw?K
zQ{%sI;<B>z!$%(#bQ!Gp##psZlh$2i%GFSWT%05S(<KAB#L99|+r1tRps9O^X&1u5
zP4L<~JDZzbyCx=jl&v1%BjniO@$qpMz0B_|OG`^-J9|z)P#*5~Rxsn<vuntOo<$7}
zu3tY9(J)$Y?(A77kXLne=tYRx!b&I(S8DZY)ZX6S9K?%D9PSCkd#q^y?sg4@IK1z%
z(g!fokW%jSQZdTMcI!vBfkb~2^R?Z>m!Pbh;MBi1iD?5A_?)C@w7NfIwH_=}5A;kz
zAncJyeq6sWJR;#ui5HkI?mO|j!jL1=d-iIrIFj8_KPMJ2<NIrizk)<El)v2Al#U(!
zDpP;@?^POussC+=K{?9uZ-`f<vlSqRg&Q|+sI%4dG7l#K<nnB%czrIK+Yepe*rrMX
zsS*RwE#?brd_qF#K2}yQ3<d@8OcnTC{n_UHGl!2HIWkYEq9tx_-fx&w7-Ofeq4Nt&
zi=(4s1!r}$QxL#yjLGfA-(9g*@NW!(`yqHL8){Gx-gD|;a2=H7CeHzj<XYdRo<(Kr
zc4Y!*Kgs=w4uXb1vx;s#!gXW51fe`raN4rItHMXl-*^N(RTJre^1q2n0AR}8MP58?
z=cGD-ystlV{E1HCjQ?DLkVu2WfWX@-1Eub~h_LXMLSU6aS17|tGJ)f}E3dBBCkX+D
zkD4e@c{AWktso9GAajnO8edaZowPsbULVRR7+!Z<V1S+OD{t(IWW!%bvGMOXqo{7z
zTg<M1a81q5r)iXJN!ZG#{CD=K6?29b2B;uwrjXSvU?bD6T@k&WS`!pux=^1$di&zK
zkg$aId2d2XaZ}TTXInz?`s{2Up9kCMD5P{E7#Y8B-##U;n*Gw{Q}a5PH$H2h4J*oE
zFktXLP<=MGV62V~t^h1~(cILe{gVf*k_d<jP#9jctNjMQAB*D#kBlZ9WMh{9DJ6hh
z))^yg>~;hB(MeWU6lAUXRF%+5hGA#u)o@E(S^<q>sAae{65Iav#fgRO>wfhujiLiC
z+k9RkH(jzVk3NRXI3rOh*S^>{!<_!+iM`OP4ypOSREdpi=H|{((b0nfa5Qt*a*Iu>
z7Hbie!moN(cFfwC7;R|wy~t^9ZqZJkz;SXyrmYY}Qv^|I<$WH*Go~kahCdM|24ibs
zWo1RvM^al{u<HYACn)oulefkYLGE<=qDb2^iKeZ$qDzMhdjfX9`Xcs}1w1Eu&W$(j
zW<4_m`*VwI0GT@HP-JUrd=mMK<=V~F$+Y&T9-8LxBzC{Gq~t>tixc|L8^+!|K#@iw
zGfwAP00z<k(48On0*E<W)9E6>2I=wQfDiayG&jCOJPzWEq<f1h$^h{CIwuT@N5{ax
z%{%zjnjrS$rF~$JSssvdGvQ0Qkl1LP&Apqey_VS}-1sY}6SL;GJNzq+W>GMqbHF3<
zH^a2%E$O63D)@@2z5NXDPZ{Wc3*^3>1AbDu=at8|sZLRm^{FaE{LEr&cz7)q80_jq
zg!x@t+mEMu+@HyVAZt=&fM7q4gOYx1%Hsx9?aG_P6O6e(xfmZw1oKR6;N6ftfYAXE
znjMN&+IVdxW7pcC5WCfbt`OU3<!pt{1b;~;TC#Kd)48G9q3T8>;d`HV)U#Xh);xh9
zZXNw8RR!Zz89Zde-M{ZxCsOlF1caf7M#m$%q{0n}?;G+VMr*7oZQN8Mp52~vKCv5u
zm1I^{CiR3jFi?}PL|M+w%{8;`+c&ZM;is2Bxm9`jmh#!?cw;lOfY_Zqrw;5H;Erg}
zV0uu^^OF$%Bj);T2?K@zH@`Jkbt(97(y)CxTWjQl5y^}$cE-lWCzIVfCKmv!u`SO8
zQV4_0+&y+se=>2nX%~4WSY!_d!}O|iju$L1Cs4GUPTlWOgDmbZy9|*+E?_@s2h3#K
zzWcDuAx<tD^VMz=1)28Gtd1PJ1hD%`@Rw*vY$nboJggq8BA|1e36w?|ZnADb-amXa
zLpw2^xmqC8pk;RQ%F}wRwGFYPtk@>3-Y?3WZs(Vd_ymp<3Li~Tbw8c$+c&-2_|>!`
zN70`+AGqSDqPankejood!^Fe{9ACtz<T3fR73^^{Q8*#G3)0mJ>&7Z+34oKYZ{Fdh
zJNm!tb&gXWn`}_E1Z$m`+IY>I#y@^X08Bt`vu#n^sAGn7=l4RGJtXKPya~u7j8ORK
z!oU^e_sbDIPZbacVOHSXat;W%L}r)%3r|0Lsf$)xNK8D{NtWY;kYmvuFC&&hf2LoV
zW1)50KNV%oP6wuEf62arfg%mA;XAUxx#x{zFv!)Ts_IPqS5OTVm>_3nMzX49TP<kJ
zTy#f>|FZi3M{dQ4^5=!LMg7nx3GRuBQWauK{+qu;Vp`-zzjn0^qPslQ;jM3Kn#kzV
zzvDmpC{)*vd(6PQVE@F_mvc;^c?JKgQ(eI$x9z{llTMK(JF=2*6xCjM`a5i64v5PU
z<Mt<j2*C9By8b2Y^A8vCz$51cU*2EdMA7)o$ugCUqqV39+<;XP8#dyDRJs0=laVK)
zeb}p|q2W8PMmBT$3PQ;j+R23}2=&8<P4^A=0y&MQ$mjlblAeeDp8C%I^re`X7?AGx
z3LSk*c7m=J0EHq)l9RptIgcHDBq1%$)mOB=ka9W5>7G1mq#>XQ$Dt&Tef!=Zowo~2
zYQ@~|4^Dj_sY@3)#}y=w-qHgMOd)pkg)coNRcOKWdiwk28g{bC5AsF~qtO}o%rE)7
z6jr^2gy;3{`2CMHN*46}?X)Afu%&%4-Q;#J*1+O2>((1?lkUa2jVR9HmcxdcXeqx@
zuH!X=yS_fNllm(2>3`*pQG>wJ<_29ipv>?Mvi%JuT8W&HFmK1o;HvUQUAv$4Xl!Qy
zO&Df(B(nWZ3~#FIFL>}T`)BG2fM96)s2;{ENtL(f^9j^enmaMGE>t5(F5L3vlAr0|
zH)1!z&#=kdh;i5sk_13rv+eSWtlq2Xzf?>{qj9&G^4>D9+}sZzJlu?p`|NfIKD}X=
zVR-%ey`|1mNF+%QWl5;?*Z<>_MaIs;y)G&7>3@D&Lpu~R3edD@2okQrAOU?u8g!~O
zgm3p?D|8g3MynqfQslKmMHed88+8m4-{DH7d_L`V7f(1}ML^@5qR&;_J8#bu=m|MB
z`RUYDGecoTt^^S1zqY30Iec-ugcg~P!9jUKE|-+(6|SY!aU#)gDcexM^Pb1iMuW4u
zW68W#3#rDY1=g(^E!ey-7g|SJ4&-J|PXGu%n8x@0Hq%KKI<ojnCD#oH1MmgHiQ2To
zfd>NZ9?HB{K?W#*Vv-D=A%@2gz4<>eE3^Lt)kM~S<Hy%PN?^=C34w5*XbCb{;fIDe
zEY1Y}4%ry2oQ8tfYP=N=?NcOKn{k`|nQKvnXAkUoN1mM1M<Ua5{$1^7DHm`nf4JMw
z<>Z<l3#X))o}xet!xhrGQ;C5fI?}06`L}#Or2|9>8?@WF5@Y+wo8?!sP0qW7i~=nz
z734pQv`;V?9HWBt!Ij4RcXr{UTfYTH9DTCLf-8lp9X!lQ{u)R#y)J+F(rdeTogC0(
zFARS6xUGR6H7X2;fGGcW9hSG@Dj=^zjuk*^io>a@oErXXYwC8q^^O$#HV6z7Ar3@3
zzn$hwrtn^RbE~_v4tM`TpuNG0d#1s@7^^v*j7~HPs9EdYEx-F)3Ph{oJ0TpTu*V|0
z*?8$Tn9{GL_nH9wqqmJf8w`BqMY0+2^G0IY0a%3wDh#w>wRY5woyHR=cXr?_Rb<Zt
z$-`KXc5i;5<Yb~YqQ}=)hIgW{nGJL|^tx!!f&l+9H6-<1wEEzCz`0g0JhS?{Bin#N
zYXq=p2p26F(`E<~_NP!9h@#VQc2O!PED=z~drKn-UFqh0rB^t$3!ho%p;tIQNiA|Z
zxLuC$6=4Y(;6Q)kPMFU#D`P|4Gy|Z$d6cP^9_0|^m{*JyD!fZr!uuaS*aDs@3q`F+
zL;E&+$6w>9kK2i5za#yAxOm@{;Z5rA=(y16iuHl0+|`Xr;gSX`hCs!Z=*|Zs;&0VR
zS!?{NH7a}GF2<RU-*>}wi|OgQMy!b8qribZ3S;+9fSU8((U&O@6;Oq41>b-);zR`O
z><?hF)X$$kABW#Y85j(Zz~_n&9XUc34?N}{Rn^zgvFKGy2=LAK`mJD2Fpy*vv(gSc
zrwYDQY+Ns>*R?0m%2J4R|9(*6|J?-c7HVZ2%k=TIi5ALvKajeu%Z7(c1P`fy@9qvj
zp=N+7A^@JJS-iWGlCp!3!878IbZ`RyGosk_cG`Sdchd}XwPRAk4NXk&dxqy7Bw?9+
z=ZtM9cWRAb;EOz?*+)OM6p2<T$gV~#b*Ic7M1?U<0x9OF>Hv?^Ji%~4ZH;1|x>pUg
zyZX4XdL9sB$A5&MgQGcKDQaZa=f7huPkQ_$s2!LfZ*_HboKD7=9w@i$lqy|-QR!b(
zCVuYRxnKN?t}6hy_5Teh^mNHY*Vx#n9Hit`dP)WiZW2pZ$TJP5IU(v5<XP0pJ}3an
zP#MFF%IGk8=f8S29Nd*<$j1Zb-onCSk3G>H@GZkpo6i>yxZV3W`sDW%4@=8YQ}~jo
zaT5bX>uEN&^`_r*9%ybHf-0~42A@s`T4Bu`8^E*taCXRSO8`FoI82v22b^~!)8H;&
zAX(%_=ojMUl&=xY@-aB|QwK}ftBZj;^2Z6h@4VDylnv&}fodDQ%tP=<Og0Y92|D4$
z7+aDn3kZFmTv?FSZq4#e8&v>dtKr*}N0DB4=)76e?=@hg;BU0fw#6PW5;&0cc8MKV
zzSJ%X+sI0}xm7GYvjiTH=!Gwd!;HliU?kXqNJV+@%t_Dp?*?+UmOo_Lwv^P{**)J4
zX(uaftMj>=&4Z9r_)8>pyzKt|U7|<q>BBG7Mp_ly<KGE<vqq6uUf)20!oC!FN3RP9
zG&((Crl2R@lazc1cc=2j`eL=`EKttJkr0ETIk=t}bN*Mc_rEj~CGUmd)t1Y87|b4J
z4GoQ8M|?zoQN_T{3IhPnj4$#kE-XC!6N1vk62KAHzouN42FqCwMogZ_#_3RXrc5~_
zMIZm3LUVp^sREBohXVFi^IG&s1nS4p=5!ulws0vazsdBn!QHOLw2JPxkxd!W#>)I4
zS+?n$wFBePGt<x+SRB5;|Gq$A>>xW`nN&)4&K#9#M!sgMzPuCB`~ENb%Dk66hfe~C
zH{A3kNWUOT8uYJelD&D{+}t7z3!wq`eDsj>+NTb-CR3L)(6qhlfV^rYO4xmqgr-p|
z90j~KpkUO3=DkQ2T^MAKx{TsG7)kVX^tF4BP(uR`@#|EmDo(+*dn>Z;=K#pFL;yen
zI%E~(PXu`5zc5W$WXokWAE(y=pd?1W(G%hsw5P7!vjLsQRf`5a6nI#x;UO;GaU-q(
z?q&+Q2EzEfIO}Rl=tL+IE(rxZnYZCsHi%MBFJfB%u53(4RF>~cf|?{+3;T-X!!3R3
zmk|rWn=1Z|h<^t(aT)hT+GpYMT+pL(W_K+|pt>BImuvl$%zezcXuhmKBGT+E->01K
z1==#QPl*g;Ent@V4deQZaC#32xNz<7!lTE59b~(QhAPWny|OS9JNgv#T|ncgskxcs
zwO(c%P#(Q@_GmCZsLo#WE|)m4r~Jo`%t;@d0FtAPAxO>h_l_*R0gpOHzrD*Ec&x%6
zzSQL+vlVgqm<~z>MUw?u{9XE2(T!ZIHvwqp7L=6_Iy&n^*Xo<n|M=q%%Cqf0X7hwE
z#`W@ak?vh-sT>Lb#J{tW$WRs)7%WTxKC<iKC+%W)FlM49c69a-tYMCKOcCg+>Stb@
zht%))AjfjRPzlH7;|0K5Pr}Is(V)Z4S<`q74Ci3Jq*WAWO-q6%j{6gF@Qr=&$Xv+i
zMY_k%kja_5F~M{R@Xu|j9NbRp7r!Oii12tB$Tshl6o-0lEGs3Gnrn`ccpF?Rzc=`u
zQ@c2S?JW)*3JrZVv>dAkJ)Cu9^TkLIoPVR)N!G^QDvfuRpulfnv4U!TY{J;k8l?U3
zZ{>8kTUG1~a19OyZ&{70iO8q&M)dY3Xb+D<mCv(taEN_DH4&O0ggN)qR|*u;iH9s4
zPKGGg9a{6@!-uHeqV2bi_~xG=1c<1hl6ry3l9Zo%1Q-bTQZyIDaa>N~EZ5aW$_ek`
z2UECGzG^GuCd8hg8npEDyhtnA-%#5fE)BEe4aW8C;PJ1X*tErVcT&aL5l~eP!Nu=~
zG$_4u4{Cxq&{F2d%3oTgHsM9NJ9ZUaMNWVY!%*<7sTt7S8-Vuj$!imb@qm7HEK+bh
zDr5109v!XL-^l@tRRr^N@V%CDhpV^24lm;v$@3>7_zpxiQ+loPa7`}B(nlg1h@n{8
zF>TcG?tWTD@E1{omGbmpkZ@f0y7>(X7Sy(R!rhW_No#>)Cn%1|;34DIA1*S^?R^V+
zF~M{ZeWw#ZyMF+tbq$C}$X)-;?J=0QXRF;uAFYAeNll=eQ;>I@5M};%z-(F;q#%Zs
zWvh9ayy5Q~F)}h@S#vD6806$syLh6bVVCTeTTIpY?@q^9Il>~Vp)x0SFb{Ujspif*
ztKaC&5RQL+QcB{1l?b(+&lZ8IK!fj{s;j|Hd=tihJ9qsZX1{y++Dj$>`wa&sj^~49
ztpShx(vB)Tri|tq_;U{8UZu!KUm?~Cx*GA<?!m+u-#&Qi;;VmMlJc@DSw~+-CN<{I
zL;(Qi%wfac;A(Bvx3%R`-vlc08TOMMps<_CJa;|c9RIzycipo_ES$-xqdmt4wGs~6
z`6~9^?_?F8%QpOeM2?T1<aSDY;NSa!l}a1GBPIS1YhMBm_1?GtLujQT5u&C-63M<-
zrc#N-Bw0pDl%4F1kt1VIvSnAoBtns0i#=PV$*!!~_iZr$&#3dB^IY%q{NLw!uh(_X
z)s=Id8Z*D|=W~DV`~KYD>;C^pUp?(^B#khi?zckPB8Lj!=VtQ%acaN2kWWC(Bazsq
z?MJTnXfU`m$uH5I5);N7P6-wEuupg=2gKt=dit<_VNvYcQ9U7AA^G#GrXGxj4{c?6
z83{n1b09s#rQv<DSHS1*>gl;g<P(bZGrj$(R3?4of$ptBNCDgiz7u>Xsaw1vOGPd9
zGR_d<Ff)=H#dqgjq5@j^vqYe?&l77NUK0A^X01w7mAXA<8=C0eWDbR(FXuz7S`)JL
zeP@-*<mdH}K7(Upui7q7l%B*ivX~NRr3L<Ap!uJ#v{|JQS51X>$t(B&VIsM0M-UCG
zi34@Zujpc1ObYmQifQv*{`gs|4kIYF;OWS*a`&<7D28skb>rM~&_uv|D;1C7=HyuT
zPSyn?7#aepCl+VwL4Tsi4kpn4(%~mgd?>jX(+&bFA73VHZLc741s@-f5@c*}H2GN|
zyfz;mrofuFcjXOHEPgg07a5(!Oq1VkI9~pD<Zow$Mw-eu$`7jCpOO^E!SLF|?M^J%
zh6P3MbNcy~)2vQG>v5pmUjuYAQ)L_iK<kr^uqi{$LUZpcQg=&x`?c3F3|X?CZ9GhS
zIely={^#zE!ZNy8$4xO6Jx~PSQwLZKgpqK!h%A71OIJau?PF*Fz!rGlZr$uD_z3=T
zHss<@fG8BjB<}<cw9|ma3u^GpCu2$Yl&q|Yh_G9{+u=1N(<tgfPDb0~y948G6*hwg
z7yYb_X`dC>nO}Tz3P}qw4C&)_+$hxWBOgG14CB2neo6axT@8Hy;A$|I6r-K;9UFb%
zKkL}r@9Y6_%RlKKeLppnl$K#lx^H^tIiHW{GRrIP+FHuks}uRD3U!77h|+MYxFOxl
zt1Xj`PoVHvAb+^oaE4J2pviz>KRATh94mR`dw)Z*q?z<+4ofEmW?-asuvkYTIH7#N
z3H8t9{u;{2=;&VT_^5Y<7XW8T=Ol3*#|x2G{8o4F2=6_Y4}P^s|C)OusdR~2##~jb
zd6vpsj8DVV6?%|-$rauK8fN+POY&Ch*RN;jI>TQ?o=@f0dEGN~j_ZC-X{gC0-#MDX
z-euWw@pETek6onc(M^}yplEgP+Ui?$=Hcyh+ROL*Tg&JEjOo`vV2a`g^Nz)e!a@~~
zjR$wdG&n)vf8L|PSD%~3I3=iO&u5T#>!Q9_HEE<sBvygtDYIPfpI!id<<%P7+@L?A
za+|0A0ib&q)dL@K?mq_TAy8jZNiwF-{cCmuLohu0?c?QCT+J2jR@`>S!%4LFDr1Df
z^AYvNPSu-~uAXaP8mh{@N*ic;X$-IItKSWJORs6cdjL+}Ali(`f~({w@>^6rg_Q}Y
zkpgd?I3cZb!%=`QFF%j9&~X$dP{3kfau$YF)kssNEE4}HR`^QmWJ)441@gj}LG`@J
zuz61SqGsx4Kgg|A)&shv!}~bkp~vvvOCXFiDCh#vIHkM~8pShGl$o59HH>VINy~mm
zfF|<*$JDq{!>RWhRL<Y|cr@CU*Kg<yI)%@z<W~Q37*R+SgTc}z@Z5I!KX*6KgDt2#
zYW(3tQr4!I_DOwy^~Zvq)0GUR6w^LOyDj@@8tZ>dqH;rTxC>V96`@J+E-hBB_2`$n
z&0j#W#?6*S0sdap($dmB$R^hc?eo+@Hdlq+fu$x~RYr53(>+?Pq208N8|FGO^ej`m
z*((=#?)m?(usV0_`Sa(?d_|n;h&A*`;`rn@K<k@}ViATHb}HbfV2D0mf!~5%y96Kn
zXNEE7MhbIbkNMQ)-P{8I0-(FT*$o*jL?UQVz$Bc+^h>=peyI(zw-$FKj2e%?eMfWP
zB$(>uttcq-vP^2VE<V50vH6K2bO*8Q7cbVjc80{LypwaNy`X0v{kJZgVwVb4myBeJ
z9MZ-XY`^OPuW1{2e?_%hcM=Y$#^_^MNF!+g^ZmYN87h1!UcJR?NA;MYy;fy)-@kKq
zf8qh+N-8NA-14p)G)82AS_Ewib5rjtBCl5KEtG4gWJ4(g1U4@0yun>$g?y!Zb9h#1
zhfmWug6RtemFpMt+IN`~gX#KTdlZh7osB+vti{&b5%2$T0YTA*XPS35zHh@{GPUe+
zflh=yZ2mXz)3w>y6&-eUY-948qO4(&J38~yG(tVtk;Vwq9I&?jNB^}0RuZKbX?iVE
zN)Nm2t|?auTn89hl4Sqhy9Q-7{_bc&vw=$U7t<tuQDVc?-=_hNwHye+Ouc{I0Enh^
z(l)Fk6LK|j08YIL!b0bA>bL(vMm*SbOsK`2XOWez`~RZ@EDQaQIzWm+Tb`NbcZf_{
z+k@cP!U{`AQyyOqn^k^!6dn3}L<J^bT+sPTlf9}02bkixt@_IT8o<72!&*T#95@+7
zm!2TTXH(w0G@bS=zjSB(!dPClsLOJM#`|*?cV=$lsv~C25{#}oovto9k~Jb6W0?Bv
zJ2QfE-C4006b?rjuSl&AK;T1Ve-ivx=XktxV+G(iA)8?*PGn(ZEbVcusMqlKwfo>v
zm>6CPiV4TP!4Fhx9`&HwM?d4Qh()Yn@+z-$c49UD?;=95J4qNkV#VkgMnt~!#dxys
z;@aFv@B;9ST(!a3RnD1_y;dare}3kqlig;FmF2}B97-<r-9g|z1|BxaZV+0Rd*AwW
zEPruAr#Qd<4<^LBd4M+0yvxlL3#{S9NMTn#uv0IuXD?20H&%KfAmf==KeX|f?cn59
zSlEQ95DcL&)>_@*4&C4_|5rM}er|AhT@Icw^_qJqNTvAA@^_ZnJT>9Yc&fcz>QzR)
zM>e8x5YX5zn22AY{8H`UeWy)#J5e+xo2DiF({-4tqNDoxOLVI|NX$Z~Lvgmt)LRs8
zRpo5_E2DWIcdu)<gC8#-gBc@adnxH)P1*_i-}nTgB|zFyKuWTco{ze;X--gGv#O(<
z8=q!<TbuHv7b08)YNCFgR}#Mg1^zF*0^DG2Qb#QxL88}U(;nY19;BW)4-sd&0XX+@
zGNO)MyP-UoH0@0r1phNoM!bI>`<(8hcu|qYM^~37Z;wXo%u;x$X_+Bdo)&laJBV@@
z|E(&c>w>|wV86PgeLuW$6!y~E7U$}3>5@m0Tk61kj(9YlHu(o}w{RMTM0YQovM3Y0
zEcjwwrlU#eNZ*0wgwU#umO`^W!ir;O7U{=Ei8H@_`*4mGMXMt}>X&Z_ye`Ha3ll&!
zocgt9<lx{CX4S#U4|kH>g755je+qdsJoVk=iFg~M&Fk`q%lVd)gHLh*;H7l_wjVqE
za!3Iyd6-~|&h6fyNK;5WbRwDu*#dvfQ2>3mo(Lv&!~Rf8@=Fm5Hp1;ojsc<N^1@Ai
zk_z1u2!e`{;+}2!_Ws{eY)!D&uV3GfaFNb6%&t&}me~9cBIzWr|9WNr6jsRhHUd^8
z{wsG%G%B>(1nK3fI#{iHOqgu=xiY-TaEay~`1PtnUKbRIi3tfsb8crc`6Z8oagYJX
z^INX2%Z3}@p{+q#fZt-rCVHDU8;qaukt4u^Ot-<>NCCX%=0c~EFL75qN1i&ojgSD3
zLf&W5xR~XI2)y&tE%ifANXqmx;OQ8Rb-ubDKyzOJ)?U7|@dPw}^bchC?#3Xn$>$`%
zwM;jribIfE*I0UtyMZ#YGe=Ki$h6F=VbMTRdYa`g#x14Og$<D(Y$z7)VNVs{csv~=
z0nwnO6-?RClN|rQsx3N&cW0O(va@*Uhw)6xPv7OZ(WNwH*NDDZLvZM^_^}TCepdgJ
zDSF1un?N)dxovGNnrlV*uIk6q1-&*Ud8_ZK7w8!asfOAnTtTipZkBkx&i@d9X6@><
zGOER+;j{@4fU?4Z|J7C@=Kb6bL>9a5J=zGCTPc2aMXxnS-Rd-rLbi%?<vr9V9v@Hk
zf&2d-N8NwwYYldk>2~SkL#X-}9cCP4dS1O1{}{I}>MZI)4U$orOO8G6wK@a3xaN?(
z2RCk`TyK!hB9oeKf@tm||DJ7y+W~{wTfMxUP*5aVG)z)Wq#DHcr*CPNa(9}}^K$yk
zx%aaO(@R0_;O?kEyW6WbXikm4ZK#O<I=S!=dg};F<pC`vy78TJ`*yyO;tP#pqW6~h
zWq5^Q5}w7Sv4DVp4mg*R20$GES<A4`ZY7|w{#x18@_fcQ-$AviF~|}RWLaN+CMGDT
zS+?th4#t|4o$YMT7F~6;`~2xA<F^qk!nf&V-GB(j1kfjE>gA03#<jMgSta(ZzMrgb
z{U&o#tbt2PBkd-<JC~+eE=%=ho{Y8Yt$41wM|Hb)NaxELR?wVB=0|2Z7>Fw#Wg8Oo
zRs3c5DrsaC>3p=DaQ`iw35XU}#;%p!e(X539tS%(bmsgomO&uw5k&JdPgRrLB8UbF
zhf9Ug6uQ-Sk_~5RocJem)8DUQCIGl=hx%L+R_bB&antqhv0pvM?}d7e>2gwMrq7!3
z%i5Ndo20&MmYu0i++0$4v-WlE&(FI7`K_jb8vFA;kF~6vyb4{gBWN0DcqWk~yW5a9
z=4;j%cX*<3e5@^RdVZ8#9-b`+qu5iDI8t8j3o9;7*NUC2jCc^5*OFOz$~%IzqKRbj
zeiR2S;~Bgnuh|$#GN*mj1cK<GhApS-m8dZYpD&$l6sN`tri6EB^36Is-QAa<CRbUx
zo)!44DX1yI^YmCsVBPB^<2$#A1(^PNa3QKBoTy))Vj=MM0(J5?O&D-=qw14BI0G`z
zM@p3nE<Tq0*+ZAvM2-Uk*OU9#gUr_|;NB05WRmhzl26f;R|x3}-IC)$w%hPF)g1+O
zW?J#|!O%O<GsL%>@DSWVtN5)q0gr}U9ayB0;pPtlRhbP%XAVbPSVW|AKl%#nJkS{E
zI$!uHcM}v~i!+OaN1d=Dy;;hYm<<q9keCcuXYtg);N#Q|POQzr?NbZpH&ds5fA<sz
zVA&@(P33l6H(x!zy>Fr=PK?dJlSEn!wv8^5Fl#h;4el|E8CT~`uAzTT-QY$+X~CAq
zt}qKW*m^-UH&%7F!M1S(>1+Ck4q}NkzhT?zo4{yW4OhXGHMg}bU;iwr@<X=q6e@->
z=@O>fj(3Oy<q8`|Dn_teS;rdvNsWdu+-a*Whw_DNMCz}lkofn!SO&7V2U$ui<hp>Z
z-?nXABG_>z{b-)_Lay)<v41XlZO~><?L-{RR{LVIm0tjFDEXsV_SPHlAo2$M*BD$S
z2z11y)~+W~xPxq4k|zVHVhl?XiA^$TPo9`umgU`zn&yDTv@0;B4N+vRYiT<ji>_PL
z6K?)n9!!vXkl*9(IyKLEa$2j^d3^~t>2T|zw=Mh54+>XyT|C2n&O-mW73md6j&M!P
zma7J@Bdxo-CK}d?x+Wf57j#`HrgT^>b{MvI4ID|mIf!UoNHw$yY;TV}`Nvxf&ylNv
zGD!yX-i2{ZltpHr;PCL957q4jYDbq{`uEQ+$IU8F(kXuQP#&6H7XgajVRUq~_+F!L
zRRRV*&4TS>F>X{d$Sy2g=0zGx?|PlVh!QJru!<NK&B?&k0Cn_)KG|*%&AWrD`*@%R
ziZM;;;spb1JhMg<6x1`cW5CUcmyW7$$jr(rQyDfO66?-Ma+gZLdGtT?#AZ|cHk4Lj
z`imD`u^MS+z>;NdmFGmKw1mLmc3uSc>8Zr9qWRcWAf|V9<b!vD5{jLF7FGiK-WKe9
ztcVd`V5M6um&*lh?QpuO%A3by=%!NN;jyqM-{F^VIG73Qso%XuZ(@2Ukw|ANJFEzQ
z|3yzvrAb(vJXEm-*uvvKiY9bSKDwX2sZ4ODd&6C!hdjvkmb(E8a`u(h#g)5jCgMKS
zF1e7|CLU=^V#XcMljJ&BXTb;X(6j$Gm_+cwWQ&SGVXbzC9Yxj|Pw+0krS1bo$ux10
z42skonIZTP`cBaDBOK1ubb!JX$NQ%>Wj+v0mWP1To9KdNVK;0<lEZ%POfZDZ+SxO{
z>qs9-j3dpfTxwFGu1X{LL&nOA(Jqoj5a()gML5DuGUEJE7TOzT{}BF!^Qo6(73AfG
zeHwj<=?1Uz^EK2{FNcn`zYZKY<xjj%q`jw;z>5P7F|Lf1v&hHT4-`$D#X6c?FBTsh
z0q0C2>#?tC*w<%0U6Gw(KY`TUt5$OJ__8XZlyF{!gXTt#fCXfmj-iuStHeH)NOgfe
z<#C6-)olt7;?6ua)hk$$G)q&Dndvn)k(QE*!JzF95i%Ib^2fhLOS*v*(IQBE+v)zD
zgvm|utnj&CYArt9U#BkMAV<J$d`FImK|7M5_NE~&`-VM;#4{2a$j*q3sh{x#Kd>L#
zZI;po->mh%&Ld2cbV;OoWR4gyS<G9ZU0G#sSa1r<v$f;&)7Qr6w~KDdZoFn4PT5{E
zicRjd`QzCOoI02k(uz0Qfh!2>yMwgy;&kBdR6jOJzEG;Y(9q}M0oMfL<{<GB88(#T
z=W&Sxlc5-lD;mA5)a+t&5Y3IIy%r+{0x2si%k!k*<wlx=e_xDx+2-DN6XT>J-V@q6
zcF^ngGTCDJ{X{+>0*HNKr$@)c^f8qnaf^Up@m-LP-Ls`PbVWflM%{?Hm*IWbd$=#`
zSyFG)p{Bhdqb*Cu9Zr7YRe2)Ml{U_bKPf*@?F(0);KuJxOUHIubvSVgmgCq_Ejw*l
zkrcL$axdx+TUo_|va+%x!v<%(`Fuvu$Hi0-!$-B@g19fb<)kVmVCZ&@OwXwxKp;%+
zIM`^ud#kVLyoF4aqQs}d=5JbthHI9hQL%c4hMr!{g*#CZ;pUQVX7cd~?7?SxHg7y&
zTKf9MGH`0MGBSSV9FHLqneNk`+^@~e4`Nlk@MhGr4~JaK94o9k%2K{@<y+ypbjyh!
zoeiBoA)o9$+CxP2a&InXaFTIeo@cN<t!vB`G9HcYq$ydfXVj3Z4a>g7eXxV#4tAP>
z1eu6J*vM2xr<@!*iP6!(eE32p^Rvja!gH>T!+BlBs5iR;BjhAPhrxJn@jca)ac4K9
z<IL%sC_;7Mb_WM9W|su#J%q$LXM?hJa0ql@9jz8m)~{wKipmyfU?)XoCsSP%*Yo?%
zFrYH#Z)@1~Lb%Jum-*fAn-COZl^w5GjJh)MUfRUzr>F|toO9V|Q9j9t++znnnIBpJ
z2@;Q58t=*^#jYqg2mQ>UKY`U&w#tS3ZvvH?C=ZAp&0OppVG?vM2%f>1neZatifa7x
zQ>vIfz4xDg)6JYsJGJ})r@1?WMa~H89&%8dl~mN7wmb&5u{~??7J6S9k*&^SR<m+n
zvNJ<*B^GP#RvqiE;zObPl@_o&!*q+v!6!m4&Y~P?nr)|BjU{y-L~|lUB%R(?Tw}eA
z>Nn(Ct}YNRCcx$&OH=G88)aXcREOvRA=siL{95)`ITns60vDdu#PD$X_KlAkWGH#@
zi1SJ`^00$eOw3>PZADT(6`#d4!h%&XYelw?#<NcFs;$b#HY}Jo4cJd^Y96X8P8gaO
zA~W=|BC9XmeMncb4U6;dz^#4ND*<UFC=Dt<WgDl<8hI|vJQo(A+%0`PnfR2%N<Zv8
zaN2pf*4t0Cbna6;2z-G6C2>C&4{~e6ftRX^_C?!oCQPpLV=#STr0%0=-CM_F>giU|
zv;t9g@*rD=cI#=9JYxmJ$stwAOIf2D@Pf<|-Sz^uaNw=pc3hCVT$b*HjvGDlL-)dc
zx2w4~YrQe(mi_42A}~q>51Y{=C%%V&NFXX^kecXS^!HK#j|&kJ=Xaa`u6UEvKe=`J
zr3KZaFSWI{4v-z+q^6plk&gA;UD1YT-ld8a%M%&Zg_lm2mX=;OZ}$I`-uKUI32Ez8
zx0<p6IsQ(Ns~F+ANWEJcu(HQ^1gqt|j-bV!k4U3jKjn`LMu{4(w5tuJlc?4mPJJ2V
zLW7CB2<0rC0RkMm_sM%<Ls6Gpmc+j2DBwAM>>B0XAQ!>iP2nuI(y{B^7Q!UtI22~Z
z>a-SIM2Iz~=EUi2m!`g7IQrbok=x{>$y%AW_`#@p?A#|Qnb-#9;Qiazo!Rq6Mr)z#
zC%7nfk0x#jKALak%w?ijpWD9CQR33?6caJQ;_iSBpugPAYwOClO0%1xNCl%o+{{n}
z-VD2#&n~F(l%gMo3xYohcGgo!M$UKE6;-spbnPLzF%gbn=d}wBSh7Z~7ZoQp;gja)
zzr#{#ODo-h!l%_G*{%l0q<?_*sd8CfmZlfk*^8b*k3E!re3mwW%cUIL6t2I&zhkKP
zX6f<-UmcZf(~Q@{ZGl#MGI4ON?}bjobYHdmbCuNsWN`PZcx0-nD*6cK=HouuNce}S
z1J1lzy3SOR*rL->)Ux+zQCgIMgRnDf@iq<G7skM^i_A1gA_w5a`Pt{6axLa_!O)V?
z!kn-v>;C!Y+!!1r?8%{13=VU{EgjhT6N`!Pv{1RutCJnm1jB1518T&MRoK4*KdK{G
zCmanIYhfB`;yo`~Ii)yxXRb5*fzUAncpPtlK}QaWKRF~`_lmBDezUX9oZ@5`!Z)XI
z65C_=ZP2wl6fuTX^<sGi_f-^z($0JW6Um`0mqfA%@MJInp>XT6;>i{t6226cWHiA&
zcc$DwYdax|ZZ(vYbxt7M*}q1{BP2GuXOr9a=G<p3G*$Ofvm8`+9+Wu3!W_>GEjS}P
zd#E(5w`d}jv?77v-CK7MNk))FS?`beF#<JGeaoj=Q@yW4_m4=o7~8^L5Tj^0s<G;w
z6SlN4^s}{)<?z_p>LEnr+AE=>Xce(BSlE-od*@rWU53<x-TC^F0}9K>BHQx2Y=SRh
zOBMJ+vqwy)s8(>4r<c6;m}LDNcrXg%fdX@!&@V_7d^>B@D7vRFJNhkEW7p!bK@7^Z
z-Lb2EoprA)*^v`@b4T_D`;Og$2r*rRsP-<+HG{Ok^`NnULxV*TLZL{O1!U{*u8_hg
z%AsiAul5>8d3rAV{<uHEe!$hpE0pRb3f-Q3qHj=Ubxpfm8Fic6T=Can%3Tb~ihhLG
zj7n$yerBEdP^do3XtsdL8`?K=zC8l21j}VINWbKOFuTc-FSn+_DK@H$GbG19CB<IY
z6P~_{R%LELaqR}VGHa3N%`@e-Z)q=taidf#5KtPCW2x@3#8-E{tb&%kH5Gd<x?nzJ
zuTvL6UVJp6)Rl91*H%-?@cMkFUf#O(%N(<Y$MaYun{xN4+tPPGzmmA{x{i>&3b&wT
zoYl_I`99bM@8Y$;s`s!JKOZ{9E)5oxZvZyD2@wzl@Gvtz0Xh6Ra%=9>S}1J9J!TKB
zE-$+K^HbC;7Q*Oh7#?cYgn#7?stDSQG{WiTFRq0RPpy*)Q!7!$7BCDuim9d~+0qlz
z=O$2LRlrR0F_6C~RNLXV^)e=AEpkZr+ZV39lQ#L-*M-xMY&CKTA{pho&JOx3lZSIo
zz9f3&WqVYU3`EJ~r`0|Ei*L9dtrp@IMn^|Rt{waKMpB-J3lWiBVG(eupxf)rTO)xI
zF&_suek+O&Q`<{8O<I{wa!3vvDw-%nCZ<NUTNNOr$&---@qi^z|LVR(Y-^Pf%bv^U
z+F=yE2=vxEVYG<Vza|h7<+T(^`TpS|!T&sMLG=wi+^dBw`AzH~xosoNGgf;(o%#Tg
zLj{gr<FXo6j$Qm2Bdv_pt;kZPKBqyR0DZ5a?=+CI^1Wl#1n8Zqf*NzN2j~FT75H(h
zz_LPg{eYV~J$TbJkWf}B<p{B+Sq#QF=zA#ySH)T<?s4^k;3{^M7M8GBWp@rsO^?d*
z3a;+6VB!|z3l)S_DGy|kb7A1ZC)i4S`Czn2J6L>@BWzh#?!K)QA#Y}xMe<Mzs*(f2
zq>@>BPyDX=Ryy;j#6en5-#Upc*j<8O2vL}j)pAJ@zva^;)t%3J0*j5ntx<bubK2|m
znZ1Fun<cXgIFM?%SJBQkeL9p^Zia*dYt!*B-*mbyJ0_6c<rj_n(cDcsulH8inS4z6
zRrT5|on#1-8u;>Ob_a19%i+oph-7H!H5Pz6{-+l}?%Nymj4ML%EZSX|rI0}^;l|wD
zM(_g5HhwD6T(TE--TcRU%DwEsK(CL(^>Ex3eR2_Rd`!EU7r>-zbp|H{07xJoNALuq
z<y@dy&~mb-!}UV`@)arrKE*h5CyH^(A~7pIm6({AK27oJ`*`_7+6?wD%-YqL0hu_{
zBEZMRXrv8cp0vpPYEfWcpSYOzInUzGohyv-+k!J;Adx1Jpm?p7<5|}^>g{+nvF;R6
zw!M}HDJ_IwGv`nV%$#QqRNIw+nNJH7>1;#8H_Xj+NSkGT<B(3dHy#+AAzzsukI%j1
z;qk8laDaj+i=^MloKnPkn!lX9F0wN$@%cz0W5e24U0#AH;_G;SjIAo|IP`31p<!Nt
zEx|-C#LuVE2PRVhx1q7_PezOQv8{Zp`A`(CEiG4<5fjgA*z$g|Ao}y(YB~7_F5ph|
zn`(>I#5~h{EP1#C=#u<tR_yq6mj*`puSu+<v>|`|PWeOmNYiEIC7GR|=fQs2YEcS;
zm;bVa83;H6l{1r8Imcf-R3WYGPyBotpEDCKt|Y)5Bra8<KTq8zlQ@5?DZ%hzTGv@<
z*A#+Yo#33i<?$qvP2@4<jWXSmjskQ=zf*KIM`U!xyXtbQDLx1J1M7H#r?EIopDro+
zSO{@!SUT*IOMG_~w|l=yhf}B$UywNGoc;PXhT1Moh?_csAFqmj$K2}%jqwnI8=pLj
z;p4xPfJv=_$6e<NoUu^VWWIuDzOXUAV}N-seqt}F9R(A==e?6OFGt77`VdZKFQiax
z4do2;n<ei70!~m_)4<>j7*3~1*ys0Ou^EH&0y`B0^K3}+heI2cp8LiBe{kD0rIR!L
zYjP{R+QB6eXX?=N{qwEJqxw!a)4ge<cxSB5g2#e7=G|I+vB&aEJmS_7w{#RfAbEtJ
zOC67S?+fqpc@fxcDTZW|`Q=~>a{}$)UBV`cndd|J*y}(t++B1U(>U6_AFU+4GA;_G
zMC+RXx*h?~rNuUij^o<CPka_0x;pBRE|!wa1YVt4x1a6OwBrg+d~ty6N{`Q{?*0W=
z&ZS;<hQ83SckkX`WmFWx{{){wrvN*dqkD5<P}0U+Xam8(+5tP7dz_`q+1_}<?6d<v
z4vB^C4E_>3(LV}SRtHw`j;gsfJ!NWQWJTF8-!KC~060^fLy404uV2Tr-$g%(Fx8{>
zeO@@(k?$^|D_N!ZRaeABotaR;v8>%HPj(a}s)e$uz$_4bjLT-`f?T4ACHUOJRfAuU
z%3@voYd)(6UmQ5PYpWrHaWXOU*T|7Va%Q+1S9_C5C(;}A2{4vBBD*t2s(l(wpSmP0
z?k#r$`bS9mzEy|rVk=lhpy^5yUhyX8AAnV(wjIbnj+L%|j+IJ$)w->d$WrkHJn5Bm
z{hiMZM2GZ^)Ub(1PJ5)-SAxARm1d4NpgZOgiY%o~>_D@zhwbrYK7$T3qipS$4T(}0
zpmJ8E(8l}CC%)f*Cnkz3^Pu)#=F`XY<FNCO@8$R4%JFMGv}@Ue-+r*XeLE(s;nt+6
zWmA-$o9D~IPs6?cd=Oa9ecCtJ%>}1OyH!NvYqeJ(QG|ZVCG5KO#3a%*E6OkhQ8*4X
zSlpxkEsJEKsT7<@o&3x=)H)zoP@Fh>8C0#2<ZOUwFK&eJYzG06@|wrQL-#FvFb54@
zx{V_&cO4zax5<ZTjWHrI_wt`39w|vyle8U`^m1eLryrkJZIrE)uJ8rv(o$Dh2qugT
zPC~m@u0VbYfjWK(TYC!vC5N?6z&-;1&Bu}f-}1ZYD*jGQFvwa5wXxq7#b5@|u?-VU
z52|V>$$j_wWcdip+kl42iG~mAHJ!o6*u3bz9pN$&owm?zy;vECvz|$PwBDIE`Wk2_
zr2}y+EyN%HtWjXseL_b+OQrQXCeBlfr;m&}T)FN;vKBqw9ZkF6FO#%>`Z{zm_u4~I
z9wtfd2H-ywV|ep6&4LOjDqzjwsTSDmSK16jW%7!O9<wV02M_&00f2m*xukqLhZ6>4
zq6Jo)?z{3z#m}Bq75p)LwY5Isu35pWS94c);A&vCsCm}=xpboHTHVT0<!Fu>9|GNe
z_C}+a3>=*D@3SYk3FUi#8W7KL11LS8366Xjul3c&q*<<U%t5UW;#7V0Oh8Rh$W4lz
zi1vtJ_%<04J}W4&YmP|v^4+dNm}siY=3m&vf5mNw<5QSbd=9hzDoT|btY-GA=O3L3
zrm-QZ-%23&Oaz6ZQ6T@tWUgIDiNgr-P*_+PKSoZdALwrDDg1K|-^!!&+KpEK)+)7I
z^*tXa-sK%~@fwPZ1x{}df}c!tnCY)olEw*eTL5&$m>%Y3QNez=-Vxy6`sK?_WXU@H
zOK>dC^_b9Ou}gZro}=XKM5S-hC&5@+<6jeu_awPl3GHf+QaiX_>)3Un06@Ph$%ZyE
z1B8!L7bs66G3z*<6B923nzl^EVEGqc8Z1Ct4@@HM^&KaTP)ai^e%vF&?7bzay37?+
z*_JZ;f%2Zx>a<aB^o$S>Nf^rNHl(b%et$g;DD&0E5%1BMTj2}K?d$k@4o$gUS&bG7
z3<IIp0vOlUvSf$-dcxf!tvcKRT`@q)+2Js?RnOxwmV6h(=|lP~8K!<36<MJQb+L(~
zV`GYthagdV^LSTVo9iiTBJ<Q9grhW$q6C|TQX=Pmzv9c#Ovf*Bj*iB8_<#L9T>Go#
zPPw4;zUh|0vH)&`Y#rp9nLS$v*2jAI6#1)bCsURB`PGh)NJxu$^%kJ&E~H*w1gghj
z?S#eC@a!__f%%Sa1!83v4r?B^uRNKJ>5nlgLL*n`-{3G43_G(kGq<x2Tsv@dccABK
zjI#cR<m+?Fz0}><wLy5at$FJLwsCv@trKi5j|NW60>+Ol;C2cjd^?x=#GhE@_x0;n
zfB0?~2pnB}cSY%3KBQSyLG=07#SiZF_jm9#zO9{1=u&z_&F)EB%PnJjDKhG#+(6kN
zVMa^rUqUZ^Ia=iPxZze0VtY}{Pibj&wu!o~Hh6m1K^i04t%Uyc?dw-3jb{HKC-h90
za#6wdgI8eVuO86=LZ>cUmP@kQ&S=w1M<MaLm_AwDrPk9($d3IVx1a#IPe)`@RM!9P
z8DPr^9*>2WvoI-_#wU@KRur2kC`A3S<n5@Y!!2bCbUVZN*f)``eSJNH#ll>95jX?)
zi&*NOb)NN8SaQ>2sruTetpePfzX>7qI}St@VB&(J+n!gPVQI{4N2sBc0{5Sywaj}^
z%jv|`l3h7<T8ezsEMnY#^ddve_TDeFBt9q?pY7uK>NRhBOetz26<qJ#A%aP!OIczD
z76u>5BuoY`M||P|>T!NhS)(}lWpZBLY!vbm{PM4Q7Qfia8uW2)uuCL2)9b~idY+N3
zJ06Y}Z8IQu?MJiHMYmKZFIGW{<kAW5V4373_I=zw?Q}XIXK)xMCaexHj|)B$HWJw#
zn4H+|uL|~VEv%;|I?ScqRPTt+{2JPx4tYY#z7{!ukMvBjiRaTD2BgL&LE#7geZZ1i
zdu>2aS*e7kxdzrI!@%%*VYnQV0Za%yD99!VC#^)A>M2~E@PoPFSd~kp>FZx8L#x#Z
z<<D)h@JqF+xs_LJ)uGYXzc^}l7Q;qrzuH`)gZ0wHymEg+3K%W&4QJ4NaqTsUB23R&
z(ZMti%rMJ|P=D_OXG3%-N=>q0R`0IsW_kwNl6rmadLHrOL)<GXN|nA%>HKo|dwNXY
zHd;Zogw4&sA%NWbxr3c(D86LXFQ<A|#pKGmzkg!C7jcC)3XDOPxEVa@D^}4LF_P90
zb*E4OG{#RzMe2t9yBSzLT7)9pX!O$fq*-&}rVlo(#UzgxRGhO7^lUkXiboSx=$U+_
z*&uP1sXbqu5s&1TTA0-ls+SF9H2y@-@S(CsuW*^HpTVq?5>+)}of<S}O8N2^m0&aA
zp*QH9LP)0#C9@vJU`j!JRM1)az=mpZ5k%kjf#qeTRv@u=huP_+b=k%B(7U`%8};<v
zQD~4=_OgL%)@6~5e{PGN&xO0tHRlS?gsxB_1gUIh@42kevJ2S7If*43MDikR2Axsl
z!!^A1;+}j%4_1yp3bQQS6P#mA^&FAGAN`-X&tO>ST~HNbfL0fdusmYhfa}5f7Rah9
z<Xe_9R;=9XQ@-r-19-BIqH-r*!M>u2P@VaadwqA1)gLdmW$CfgO=Z1vfd#9*whM>y
zX-FEbRWC~+C5{a>XAjsHHf6y*;l0FzXC$`*D*!J6a-`}{_b*}{;Z=OzM!|Z?bWavd
z(DcKj@VP>&V(Ed$zu6te^H{j|+)K#3EF&YMwr(=Yx$XsHwW53VUgtQWAhV#heQ29l
z<WlQsvlV?3cx{R(i=Rq<O|jnPKA&#W;8pzqKAjXbZ};sCYS{$uB{1Uur-X)WEi#uZ
zx~FN&jzA7ZqK)3ya^%f`j(*j9?*n?Hl_q+wN+y}2_4jm$GOqTFjd8rvvzBj{(KEWF
z6-W>MVG(KCH7Uzd&(NH{L6b+CzF?S&Fqd43PnxAMfKWt55t*~-F#Vy1SY7G)2!G<e
zFTj=O<mTG@5Z{14&$=mIbEioC9o^RLgaGS??=YY{Bg2RWOORP!n_gEFdsr}j8lJFM
zR>dz}_YlOK+{bmPUJWN4#8NNDO2<bv+$vy@ZOF!9oPdoS0I|r+`VZ{`+)^9W3O4&(
zAysBE+1Mc=M}Wr8VBEg21y$bQz*mQ!1z&qAag~|izx8IfthLJWn9Y$ya}}<Yi#9`7
zZ`^!4H&z+dSmSBy=Bq1)PQ22Z3M~)1asv0;Isfwl0LD#%^n}oU2<;OU(!@^g4m(2Q
zV1^fV3BV(~+aEKc;}}NLpgSfi2(jGLP+HHR^twxxqUuZem$wisrS*p#Mp)N=H2gIh
zQ@p<xWhuP1i|Yd}H0Rx{&V2IQGQVVd!^KPk(J1<3ZD}3at?r;1mutO5-y_l<rt~>j
zEXUl_xcSES)c*R}65D3iM+dMRWXE!(SB$Q=U=QVIwUK|O*ZdpLn#rXYl(v+_L-*$*
z%?k`1<sK_124-8+Oq)vYIIS#@9uy)KSICYk;P$n@+0Cf+(<MO=O>gosx*XKK9hicq
zpb`kb8Sp3eUuX#$b$>!Uh=_=p7jsm2(Gf`a8nfa31DwwR%fSHPwHf$P7*)9u&7o1M
zn2>BkENtR6zAg88+fm{=`vK&@#P?=XIIM!sd^^BdU3#@(^DiJIY5jc7kPf!+cFw!#
zUA$yRDoX`U*f>GuYR*E7uAt&=kATAo_G@j1{IPq)bramy8v4GgbS74{46)qz>ar7B
ztvuvVe8WLta5w3W!||Apl{a!ORt`Q)43hP}zA}|b3Iq=I$$M066-!6HLXbGvew424
z07HQpsa+JU^>+#fO5yFeRlJfl9AAFFQ%d(IcEhD7QeQy4V}L;3ORXUg9rdcfg;b=4
z>6V6k*|obMs~lD|!tgcX%sxQdc|m6zsY#}4#2g=Z2C8X}C7)JU<~r;C6z^~UC_oBv
z0pdbcnC&{&G?u-(I5{4y*h230`XHUW2Q}N+y&|!-smri+)jn<Qm<G#_YzIS`YnIBa
zN^{LRahW`O_(J&WbC+zk%s>ANwXdviP*H+F#WzCR1>02_YGdX|$<KGP0g=*BTRTs9
z*x=qPxKG80wBm6xZ@%3k+pq+F#KUNNqeT`}TeEHnm|rm0V~HXR2<*2U9n)@|_fd1N
zp~?K1xBX0@k+5R=nON4@cWh_2`Ue=FWc((}eK(;MnW{e!F`D2}YVYnfgh?|Kdx)Q?
zK7-|7nSbiH82=K#eP0S`R5#nmP-i*SQri8Ncr|KA0S&o#(h@<(|GSD=<!R*%1+qGA
z+y_)eqi-7s@@iEy!~rZyV_!HJI^QGCrBf$ejY7Bhgc<97^;(hSM{J+;$tGy>qvpM#
z>|qW)B~dpU&m;%EUUSdJ@QDq^742z@4U}t4A5j(4T)9;hE<^Vv&Z+%&qVVyw*+4nC
zDx3CI*Lc7eX5Cr$8!Bx6iXFO1tG&zA-T6Z@2BMvpmMWq2O!AxOs)9YB61hVIF(zr!
zM+8#5Y3it!KNS2xL`1iO^7c8lXXN=<)8VL#O>sDm<Ff&!RV^aN74Z6D2!dTxp~*sH
zlf0-HoTbMvl3n<S7*ezAl_6s<JEN7m4a9M(t}(KQ?M{T8@!O+0GsXwgN46#osshX6
z5$eAUEzBZACm@`oBkY(>`U+j0L&Q<Mj9Al!-0ZQVWZ?KC7}oZx6w*hKKc4e)OCm4V
z#)g?aNi)T>Z~57FHk7XvRZ&$IB2$Gq?KVcGKx9Ujw^2u^RY#wjiKuWRjZN2!e9cm!
z3man2_Zs~`)`D&DVx*JTA7dRaW5Z{`%??&7SXa2Jr&o9q(+LHg`Cg--T1X$zai17%
z^-q%H?vXveQYWjjGc`jSoUo1#VbkZwof7k3L$Fd`8SkzZW4qtn`D}Z8`vk+(Xi|7`
zSTgCMx<Hiw<^%<<6)msoruev6$F;U<s)&fZRTB9)X{9cwIdt%1gY=L7Dm70H1R`Ef
zf4{K_wQN>xx01<vPUk!jdm!YULdk8SDS{YRl3M}MAg1=FDDza{M!#Z)v;|RI<gKj)
zJBXS?&<`)Q!sznOB*5)d88|>v;Wm?l+mf=C298IXMvR@ze%BD-4q!kBt3;k$g)nck
zjG4pzuH6MbLe3b;%fA8hkJ%cjU_&5m9ZvU`#J$alXMSHkqj2kdO|0$omrhqHiCq52
zSFUdzJ;I_!NuJ7^eRhin`}!m{wV+jiBn$|Oa(A4sZr-1%ZdMV6;xeK6Dn_1ln4Q@G
zX&B55tN3*H=6x`SM%3=QLN{_<Upxoib9G3UW+CXW$bQ^AcB^;<w{>x27%c_vO=d-&
z%r(nWp1ccNx~1ZxPi-1kq-p6kL=Z9mNMjWKR)Qt%6`P8Aks|wr?4WCI*L*diRmMM*
zd_Ev+axJyc`24qGA<OTtE*sh`woH}5kM_Vf2AV1hL3b^L^0lB$N{*MML)kDpN|-a{
z22df_gQm|fA4H*At|pPj(QHyxV_P=VbD)2mJM!Z<sE}?N*&-C94hb+Y&$GyeXwPx|
zK~zj9m1-bFwHrxtP7dGF8gi}NHOn5gSYu!83Y!J{K;J|-J^HxwX7ZsFBU^Uk|K{;9
z?@u3DMr=jWC`i)atGPpy2@z}I&g!+&jp=h)A_KiDwz=CcwO_SC2&Q_acs20$nk`A1
zK{)2MIuGIsM4|m_vj@dP7$VFSGycJl6?Oa*`B@2`hq5qG?db^jSfnv=O&&$hsEZk*
zkRx=RV(3LWB1B#7j@(~l{t{kww;dyJ{!)^X{m5OWc;<M(kI!a-P<@Sbi=kgHd#`LJ
zXrS7xf?Sy@!hu8aPrVSee(NO}6;?|M$gCq?Ty+WPOUvr&5RWvCAIhW8OHWP4q+a<u
zcfO=@^c+Sq3n2Au_qSl8hc33f^TJ}J_XoiF?#mA^h{}0~_XKZCwT1$7(#smsf59A9
z$Z_KRK8<qYUoR1WPTe~ggGFlw>^l%0js^c3en`!mzhN>U)4wJ}5QARn+SOjn|0`=n
ziJ;kUF9-C}Bhq>z#Ltv=-x*o7Yv4t$5^#m_*Z1$-hA1}GFW7ynRJMBYKK!@8Ylj~U
z9YgLszh}#G)`|i;S>-MZrkERx9?DyLDI~W6L^EFLiv_!uW*o9Fr9ujTYhwL?(>t#z
zDl7BxB|y@JU_ASVM&o&U76gUS`tX~Vo-o(SY=}EIG<8}T(D%;kry&((uU=2i_bn_^
z6l^YEZA#d6e3m)vi0}-VeyXE8M7#>)DXn#W$DrG<ozh)Fu;0#Ld@ADy;0i9n)r=pG
zL;qxs)VPx8*_xZlsUtPu+TC%5`@@Z#P?odG2Mz-*!QWZ(xNbXbNTX5R&q<Cg=65Sk
zB%=94XAZf!@vKr!s_oWGzg-1Ha1zV^erMu4-PR%l(M5Z7*c$<mVKbv0MCwN_VRd|M
z1rSBA^dAvJtXPld<+E9eN)A8W%@Vn$Hk~*qlDzeT8AHVfe|Fs1f$Sq5=aJH5&7-It
zt^Z^BHj@8k5rHJn4-iJV1N{S5dKGuxRhG0}o_+u`Bxct>YkGAZw?(3w2@n=?VCtFx
zM9@^ry1tlxT@397nW*I)JV(H+#n9_F>^0THV0(<(Q})N{{tWQwt)R8PF;%E<j+uRS
zg}Yon;1E`6eu<HOhkuQOOxIoQY<G?HkpW*TbgtRe@B=fTm}Jqtyo9;JwBaz&YAvP{
zX}oz{X}v)J&7J)>ofA~JaTxAsVE9AY#MZTt(IT^q5e-D!qiE8RcL)~{+}hdcAwVdr
z8&^DxPOQIq{6#^RQ%9I~FoY_^UWRnwLj(b=XLgfoq^X`JL}$bm->iMe;5h&ZVc&uJ
z)27QJCB|e>-geh$1~zyd6MZ~V$_^K$v~-8q89F~C-QU&`x>sRgFarl9=tnpn{L)n=
zILz+e{%05RNfVr5xPTf>x>Z{+CuoXYdqvbd^eM@816^;Y`6|2aQa;vDfDK6?s^Jew
zOa_}6fIZXkXKIv@aZn80m2(@hH2H|C&ou(t=9Ad6(j)WXjqWmzkN_?%FvhDDwnNE?
z+n5p>(kNxLw5J;;86<e{zh5u+wnhXo{ZB&r6bGHH@yC0`k<dfowWkq2`b%79H$<cK
zS$p4~ewxQpf#uX_G%CgltbSbj(tX(X@P3<jKT(syFkU*A?ltRJfH6e(n@2xQI{jWW
z;Y(VXY_nW=k6e<)9YSj(1`sGlH&xi!r~y{6EJzwu2$^*|*RwNt+rMp#5ziWJN34%k
ziExJblH?6Bwj)R8GH}jLP6?2mt-{m`Dgy@hyvIM?d4oZKU^N^si{B6=wL3G$Uyy@I
zDQLBqdRw_GktWZ}R=AL#lW;K!ko}_5)zRyfd{rdBGJm`txe0pFyedqvU2ByP@!7M<
z*7$~6k6r6n-~0R^ukUg{N3Gs&LNq`0O`EwXh(^`N>w?mh4Ja-<;|_B`>=)Term4eN
z&j!b?@>(zB1H#uF9q0234Xkg$uWP?lc>qZ9KkvF}NOF4~_~k|Xi!{H5Z7#g@Go1u~
z;2V;1Hi<P*ZF|LBW?k|Vz31a6+SixY*=O?J@3AmzuNg(qNEXA?JeY(^fFV|ayvWDb
z#R@M-eg$vI@0MYErTr|BsF{}1#S9fe@03x?nSHWeDi2&XsNm=L)=13FmQSe{aNKGo
z8(a}!6$!%GtrRdq=^~yQ^qB^QGtr|(iMp*`rDLxk2%Gspx#EK}=LGK!YSza&Xhb-B
zt`Fem9WzNrI!nM+wW2R8J>rEtJalf2W7jA&O8G}uCCk3Fw(i|3&kUx+18Am-?@6=!
z(84mXJy0SOvLD>dIY8SpWBOH*2jaPt5l<HSGJH5*i1F{3oYg<6l6P6qP+Y>$eZ9Yv
z^8D1`DI}40W6n4@#K7Pf{Nlj60P)P>EU5P=c=CRNTUgH*(h19f9OW|q2$dzBXwXZB
zyQbKEY9)3DLI4Y&<2-i04*m;X#z}c??lU!;`0C2H=;@YOc@?ee`iF{}8Z*C~<5!ee
zJM4X5XMPmY95V148HZb6Z_5~oAhj2P4oLc5#Pr)_g9k3ZTpe4GG2*9K*|OlnL7n6g
zQ!kzDC{K7}>TEXL&|Nn)p;_N@cJT~;b$&?D^Lv`jnf0G<{NTbaxWYq|%^&0OM-Nj=
zdgT@b6Lf>}rxyUt?L-svbsU)q5*|e=&`^3<1hOEtXRRHwg>Wna-G{59RYnKgWagc%
zRaI3Rx!9R)ZLj$n@x{6?O>BMppnf2ZZ(2Ys-hHk~_^6U9#_0+cyLQfNh=zYv{CF?5
zs$6c}R$NphARA&L7h<wk{%-myOylk_Qv}~qz03(H#QF->`s3+eJx=LO7SC%d$`-_o
zuc58HbLny`&dbK8-#^Ig<B1=uV+{cdQhUQ=jZ_5m>Hau<(-St2#TE6~gO5Td-k3Qi
z8{-wo4wvQ*g0~A895cF)(?8dXuZnz?{A0>_mo!mrl=d9*ol=LRWr*O%KRddI4N=cy
ztPxwwTbLfWhVm?R_2ZU=p8M6OGh!0tnqoTU+x+({@mE+k&1RKPSMgAe9(_uB?Ri%Q
zZhROG0yVD{{(Y`6@uu$&VmfgcH0<1vzA=azG7;Hb{E2JU8p)n#OQU(<oLPQYwaDiE
zhknkKG9N8hC@goeUG&E7$wBjD?VN)6<<`E1r22c$93Ebrh?!~VI;$EBxiYKeaI&yJ
zj|*_e{wAPkT~fb%;6_yuGrx!JW0EzqZNBCr8SGIw)<e=Bbz5N5l$j^5!fW|MR!@<m
zg_e67Kq_sl6*5)z^g&Gd+&vTAo9}UbWbA!O150{|BaROpPU?;oNocM*`o74++;kV(
zRYbc6p>@5{(a@Qm;IC|W2BW+yuu?gyCp!fPXGeuIED&=u4OdRWTg?tO;K^vW3-CWN
zRh&ZU?b)COFxW|^m$Kot1|W9Olu5X4kM8PI5<javCpRya1G{E}nDN)w29Dz3#vXaW
zw&W@)(n@oJe~;^Y^gNfwgjM?nbD*jbZN9yaHbHE$wHKBj#@y>Kn)m{74<ZByce(zV
z<jjXqYw@qC$s`}r8cq#UHc2i6-CQCt7#0^3cZXd=EnBM{n@-7S<4AmyK3m_&XqdAS
z^D~h0c>A06IFo`GZ(hzOHa(7}5RpjIIt5bJx{JY=)t&IciwC52^$^&25&^|7=SI)_
z!{EhlxOrTcL3a4Fq220|T^4<kGW1-;IA{-f5z-@0{i!@uyY~$j!1KXF*J-}5x|~w?
z2UV!D*UOo8BIXfX?aNAVc=vs(=h1QLmaFeUl=@`BY@77_xdpD-f78HQX8yQaL9S=O
zmeBX|Ip9p&S>x>MdXbZxz)#d_!^D$U4Cw74*oIilP5t_0v&HoD$ADDU!{qyksLNU0
zm^h%D7*;l^JsX;ZC;zP0U5=NLxRLd069<ydWzx+NT1}nWFe;5X!k80GR3BYu_s2Yg
zN94VCe@~YcpcoA12Iz__dm>9y%KkmQS+cG_ah<9UcWg?m_8zG|Rep3{>iE{lG#ee|
z2VSmCEAyjfQ7haSlhazMBE)nTT9Z|TRk1BC=C69V$4#*E@!bOGGKRS>$QJ>_v*%r!
zW0}b=CeNe2<i?#oP7}Z)e)0+mfu+`ov^48XQV+PUCv!xOF!lNnjJR_@IA^4g@Q~+9
zb>bcJ^)_Oq5u^w&-^W5vJupHp8j?Lx6~2sy4nUf=Ax4MirEXiM!&ZzFTnQine(DSB
zi!gIZ27=?%x+8!5Gp)tYs7VNXnE<{wz7#G!>k%{6={6vja#sw<`MeLEc;VskCM$2Y
zfWtj2I;;(UEj1(0ieim;$rzB)+zmAF_yO_+?PrIY_{sd2J(#GP#2N=>KaH%+%+(p=
ztm{E?U1`geZ$HfBfgmMGPzMb?LHb)XI)!$^0Bj_}MGGykKcKm3${k<LZL)s*r4RM}
zVU?I^O}uBm6*<&aUb2qXAnT}m(dZOKlsGRf!4fj#*IyE&lV+1n-rhtY7iIEC70qcT
zbFy@VI+qKe=0AXL%WW;F196)C6RFhjtkE;f+&UNmiX{JD^EmumbWM)P^P92T1_bvk
zL<)DSfj>oM%A$!X_e+OuS#sKol(;<o5%ayQIp32#K_Wnsa3L159=r2G%@t|n)Q&WE
zbnZUB&;o^!JkAQURhHvbZ-9tVE#u$uKD^QhyaOQ=%3#=mvqtk^M8MbovNNGtzRmag
zuxc>aCJX0U`5kb-PjTDhOOD9HGa|72&owcM3O>_^-xrwv1@aY36`Uswns#V~p~4tL
z`d-D~iI%eX8imY#d#Z~W#M###O5`$L<<q3_m<a4pY*4-^WVtgeMdrue=QFXW#U;<3
zVS02=RiAwL%gg!Y$1jHSZ`$qgUm%M%xj-aa4a7ZK)9P3BU3pJB<Y{n8%=Aa%BU|}F
zy=;cE0iyqA=vdq=e~mG0NkPOfj{K|48WW+$LH_ghjk@4Ky|v4o4kg$7`AV+W7$|_;
z^vzEo6l>wlglww`{r!2l3^5w1gls!~({y7CgdJPTQ}F}#b6hum8wd#o%*J6PyyCEd
zxv@z7V<tpTdxovFR6b0>^(Z!8`ZD<-_r>2t5&F2bE&^*9r@!}757VA1y55z$zIkn9
z)Lrg7u;V!okrS10ODDRrYPkNb4|Cee>%bbGRH6#gA6t2X#3?>%f1F`%n%xA87JUod
zYMB1CKS?4N8{W11$rzhS9ZkbPFC5WhLpNxr;~oA*w4c-~^H_H9oQJ%|O!^1YKUd5h
z9A=pV3h#>#9x7h|O(`!>u3V+ipwXGpPkeCT@(2C~12Cy9De2|7KXpiMd7nSdf&iC3
z{@afFFYJ&l1ILg*l2S!mI-YN(ys+5ImtnycETELodoU%rjYf&fISi^3&pa_aIdnOT
zMnkBAk71E+)%kj&sa^x%$06B}o=tD{A*J-EbNH&wEFU5kxv#+&AB@Ff@kdzeKtYY7
zmo0FQ9dZ@q2AjNC%b{l!DUft6{eOJs()pX_Iio5E(ld=7hp14`BDl9<5iXD@0k*Le
zwo&sqMXokOXOuh!(+%83`o~s%GvA~uEbno``NBaG$O)_`|E_}@a^im#jUm5B_*Vr5
zR2AfKVwQx?eBN68q->)N9q-)FB6ng29egTyJ0r+z<*CQtsOFkA!me2_0&|~Vl%@Yr
z@K})eJboNf+Oz|fB%Q!hzn8luwF2V12}CX+*irL3;9!ZS1O+QY@}KNx<AOu!Z|=G+
zbmBF+Qnq&Fk}PLg5XcSbfJ2CRf&zL$XeS)uGg$YIolDuYE8UH}AW17&v!fNRX?`s6
zs1t6$mc?NPm+x|vo;~>h#(*$fvG}cXhkpdtq<KD8h!W9z2TrezuQ>)JT1Lvb(iS%y
zu%}uq=O_j#c?n%N7nh^CYZKTRw2XgDyC@K{m*#>0xdj&`GQiLEb<uue=>9`kRH7#0
zK=kJEV{?A`3XIkjW}%Fz<sT3l?WeP1)mIffi+FrX8^tZ%l=+3U{naFpI8<$^*p1lh
zM^*6CkH(m1%GI|IIsRU4y9)ko=b^^<Q3_AVvv5qmGUDe^)E%xaU^712X<8qt#9*AQ
zN$$!T73ES_&Km7TdJibOPL#}l)mLC*cX$e#;~s=M@Y(Sz-1WNd@zRcVtY;LP=11AX
zsJxa2-28uYi2S==!R89sbx;yTH(|fYj{BIIX7go@eoTy%GjsbO0@F^Fhv%CtRmt{0
zX>FP%+yOK@NpQi!Jz^iTH8Ef3um#H04}~~6?(~sG=4)HcX>Sd7{;g3F)qG#hJ!rT`
zz^5>2NFEn|+rm$F9EULG7tEJN*=JRFt6e>J-u@&DtHJ+_E{WQ(A{5dV1%EpPeVIh5
zm#(oRB_@~Me#Wqu+)gb!<7=fyjtkS<astT&Gp)z#FXVmPNB*o#;Q2TmP~F#q3aj=k
zBEg0XZ_{B2En$7<CGq_>EZQu?m~|wuc7Kez=!~L`Ai?%}B2f{pNoBA6aF{csuLMbI
z%htq}$AUF59c1iuF9rj$5@$t|j$@4GVltG=$(@%|?cU8Q(jg}Q$cXSw20!V^HLaKS
zuX(b68&)RH*#nz#fzER-?#TrbZA{FDs<nFYunBvx6m146bwZebQ6e=$4*Fu_{1D(#
zs*Hj?*_*JO5R51$P8^OjT@!&_wN9Xf251GfOF_n_TQLV*72tiE*@A>tm`(AY(1iuS
zyj@y_Y+;&+OJ3B3S8@Nt3r?`Fq5$o)Ne=?M^e0O}lD9W>b_1)FtsxoP8ijq`Ee7i#
z#c(0K$@?&sjfEg&3iNGgMKBhpkj9Y^p1<2c=$>w@1D#stpcF8;m|U*;7;4z;=<Hh|
zdU}kfNg-z{?pz@c!iCg>@Uy!-_Z{X@b&L-8?It9hVd@wmmZ8z~swL=>k~Dk<*ky9N
zeAM@RyS^&dgrT+EmrA<>cZ_s{ZbD_N-~>_XHRNVq0IbE+ui`l2N9U1EtmnPl+7nRo
z6A@T+$b?O5xDM1TfT$p};mi%?$Oosy%FIChBC8Z0n!+-K-_u{he)z5z7pLYw=e!R@
zz>6V$9_~WRElaeE#c#to@GHD7iE5clC%uN^S_{|=T-&hz6h~XHH3TzB8o(Wg<hLN!
zd6Se`j6a2C&Ue0dC#^epT?=6mxZ+z;=4wrTVRrGO=O$UbGvN8BjC-9msg{RZyFEmx
z7TlYF1rW}G;K5E8u2p^(hz(w0BlPUb53Er{%6rI#Fvxq+#ktWZi9u{1{hZ2DBK%ye
zL%mcrb4$xk2zMZTpQVf(2>QW9m8gma0rT?-TE+d*tAc9j@F#wUU$r!8FI#R&WkoI|
z4>FQB8&e;>%m~4`%*~P$BFa8J^p0REwR9W2ryx49YKfoff%OWxB~2Y(S~+Yp2e;oI
zlMPg!4Q;ENybN)Tfh41+q`j)@1Dpv%@jNi@@ZE^ZrPa}>)BR`!$8$#n<*LZblK2@i
z-<Fys4s3{jDLJrvqhb|QNM8~9KIOBh!qK%_O`}iq%RG%rc2+o^O`0zx!&v_xYhNA@
z_5S_;M$%1Dw?(p6gOW&;ElZ2#O;NPjD<av_i0r!U+268nVIm~il3lkV8G}(|SC(nA
z4B3}4zcX6C-TS+r&-dQ@`@R3@VVTBz-mll|oacF-=Q$_V>0_Kw`Skk)Q;+^T<?$QH
zV=N_`Hw)c{nhA};m8$|SiFEj0yM6v4_(UJr3#N$fzKG0ps2~|0m}kb$%m(VPcg)2%
zLt@k+yLo*`TxFyOl#5tc3~f6HLu%NbSp$2{xL0e%(DU~3O&XB8`<mNdIw9cISyjF=
zi93r4((@7*z-~N%Vpn>EnL8DP=f`ZKa;2iN!l9&XQEg<wPWfTOtaE#*$BwR9nl5_X
z`2>z&T|*NzQ(4-*ee)OfDyOB#_uoEW@=*|xC0EW1B7p78j%NbM-(J}&{N;RdIt*3V
zB^Oa+%GtefHB>MbBBHFH)`oPXw~B8LYI|CG8OlXp-K*wZW73?7JV7de;k$4=PHHpd
zL+{IxeXHG$9`{K~*E<I0;tEXk=J>5PQll7pwiBU$gn(}rXk5S5SVR9+C~H@Mh7F32
zoh8ooGf@9^jVTF)-XnYmU&3%tH>k9Jm<N9bh{|jUTlKQN^)Jo1CXo642Qj1!YBn7;
zC3VEtw6on`vdBt970~}=GI{I1tl25m=To-&qT;z%PO&R|A(LCS7QzFr|Ec#zU?Z*W
zOZ(ih`O-Pj&_ezSsl{9?KXg)4JZ?%zHvrkmY5!^;p~HyQBePP)y_Y|~)lu)=k2E@6
zdN6y&f~W(0p<OXGCFWVx=7v>^j)yscV7uMz+u-chtiCGE);0+0*hd4CQ6Wt51Ib0`
zit5Neb|TX#jQz<L#({$`m}ycdqPY*!EM~ZSg!+|G-fTg9he)mtQGjx7`N!#5hT-q5
z{AZ3ztRFc2%hsf>Z2RID8+zb}U)h9%$J^9#7k*YfKebm12<^(cbNl=<`B3dG_Cy-X
z6DUHek5#~n5))JtYTi}`GuaZOJu!4bdM8VF(q>&ZmWqw5XY{I^xVZ6vaefE!S}KHl
zg8jW4{$Ok7OVkd&Jq$s>TNs1seD^jp!?=9K6ta~?7pd3c8v~uwQ>MWjhZdgmS8scA
zRjfsxTq;ZsVdH4q&OPX(aH;q8K7t<MC~w*!y*R~E+u89kvP?={^O5uVDlgBhE@Uu=
zbC*LUVqtgZFN=t!fl=34daK$A+E6fJhr?5p&+h#_!pVQfc9VMu`P9Z$)4wda8nJcL
z+4og^d~LtyyxhDEchteI{^Ov-vO3K2!2Z+CCOIa10)137{rh%11;6@@)p=wi$Dza&
z%8CxQIN>v%EIO<+N#Uu+rq+yyWL=%tTG44y_#z73J>EM#Rqyh<D!sO<j)_b>7vj8b
zIl_B-xiE5NJ@g7M8XT<>DsS8{b^-nD@$pV;eRFc%4*JK4%8uUN+|C=9JkOXfG~{QQ
zJ6)Wp=^7z+cl*WR857VTuDR2xN<i+hdHF&^;hY$<cvW~-E<=dBHIVz;JwF>HK+T&~
z7NKUWwm(FN#&|QNL+co;#Vg#b?yJ2j_t<#ilT_ISgObzF?-OqMOP71nCz^dcT=dx%
z-)ek?24X2pQ9|l%ALX}IRaUwvQno(1oK{RL^GK8K_roO3r(YUOY*!r*aZMO_<%3>w
zYTMCKT3qtwg^<_O?_vq_*62xjO4&0YAGYV{t-PM8pSHr#cFM?B)61I0H~5sToUyHI
z^T32kmmwm&_UPIPW7{eJpdiODaqLKu06QY=(dI8bkn!&GuU-2t4y(+y)J^8M4V8yF
zQ-2o{nwd=J_8g(pgZA!9W{;aIXVAsWC(Rw{HDzM)^MQmKTt_MV-5XI$mKGgMJaTG_
zO{TO}GED+m^(&?Q1e#oHg@3^O?(W%a`OP+Qcn3fDu6}4Mg0#qBmi}j!sPVXePD||^
zish>Y>of9(u`p`Pvv<N$#;>P8h~K8HH`&L__ROU(!FY#!{IKddCzJced2RMvP$&LI
zIaak>fi(Z|AS5`+af$EX5d_IO8S4D3;XMDiz!HiT6fo9B<e6gLlVgmL(Xwknayfe!
zlCE(CnPt1usv1=p^bXgqQrVgwwoNsZvCHTX<a5U0q3r({n}I5>uNq}pd}5g=Y|9C#
zNO=|uBBREFynYX}u{0nk8kvU?9@g%o8*-%rdxo&}<EoSIRZZQyycm1=D9`f7``j58
zs?*VgvJ)8-q!xLORY@R=@fn*Q?crfodju`6&I)esa{Eh{zK}nB_#iY(Vnq^9K^5wj
z(9qBb7=`q4s4N#zD@aBV%fZhXSt<)3dCIS7;43YLhhPA0*M{7ESo{>XdAm#u#pufe
zaw$C0!kMRS4%<ong}%Z5{{H;ZmGWH*h%l})@(N5I)te@<Ac;)0wx%-jy(<^;$z?5q
zsF53zHJi0Jo|=AZFef$BFDDdLi#Ht)aNJwcO-(Hqvl%HFBh5W%*~{~F^Y=?F2fn^~
z!<Z5J*=M{VyHLK%El7D!c>L65yc6^yV_rnnwTQ33p>07>K){eQiMPw}qV7!*o9+*q
zlXb7ENj}pJYbb^)JQOz1jAP;lnp#VnT8}cHHm<3)wyCva!o9T4bv?37eXTmZsa3hB
zrzge}HnW2!JpWE86`j41U-Z&-Rcz>XoK3@)Ai-F_&>imMj5uqmNi6qB=8>(BLh_n6
z<Tfzh;||^-gV11=TS#7f$Lv%5A$cT)f>-bz^W*yt@}j#cce36u-m8f%*HfhvO=2Zf
zFUM1euV#?0)}f~%1o=QJAItp%RjwMdnut~d>^m3c4u(zAeqOe}7X*SS+t8FODODaZ
zTPcaP<kV1j4iyESD&`SyNqv`gFMcpmVKl+QiC<`hxx*SR<WFv&$$aUmi9~K%j;K{~
z!Q6qO)6ajSWSNx~eNf^gRqsRHKPPdnho=>N;J`I%<AKHZ{I^?Zd#oD|LixztM3U5P
zT?!XTWm~B0G^NTv?m{P8R9Xz`JUaUXDD3?=LPA2eltEa3-d{SKAm&LyeCxrdMwhKd
zc2z7xkl@}}4>c^lY|u?A%5O}dbnpw0Yy7v3@p>av1PABA$EZWe*?v++?}|=4b=k#d
zOtf>}tut}{5Ee&)S1_*}4Oq#@e3!ZR@89cEVn^fHt!W^-qdz9hh1E&T565)oS?#Ht
z=&<d~tL(ePDdIK5{B5UQJm)g?2l~7@O~d(3fz0!=DZ9zxY18oI=7C&yHde$rbnD&O
z%zTf^sw%k|_;p7hD-v94Ul}>^?!x6eG+}vb9})DH`7F7ino~;AagXj!x6vbvAhY?8
zN6|~9QlC8}mg^7y)%K||@u5%nNKv;tJpXOX*0I!$ZWo*U)aHfpM|RGoxDMv@*TJX&
zJhJIRL7N8lQ9cDs7MDO5FI9*|SiNf*(vk?LJWG^%F}XqhlhQI$fOMq-#rptRMpLSI
zaCcYB#m0Arf{Zq~uJPGqSgZ^Y0h!AA6zR%yZl1!HFr-H^Ev}~T0@Ge-#o~9PRUhS5
zx|KMQ>q+k}3`NMM^-k?ZPA`RR%c}%~BtOCMsQ7ZwX5XYq$h$Io;yuF;7vQM%H5v{J
zl!*ZuL+<+v;RPj=^{-za#(HDt0+`L%qhC2)b;t(Ml%t@k_dL^_F|wJ~9RuT4a9CH@
zOmXkt`2&ka6Y9d<Ey1o~>$aaL=~z5!D<1rA(k5nekldwHREb~TZP?*PiSyY?t;10j
z=D6oy=q&U<EnY9E+^X2zBg?bHXk=(dOqS0j>bmlKW#f<e9G-oU${~W~fvf(Si&haz
zuBGA+v9W_6UXFa)=n6wf9kphI3Hv`pRaZ&fDf4b#m(Y?c<yx{Xfm-TxX5Vl{xp&WJ
zp|bhD=2w(B*K9&K!CtQY;tMe=CAr0SRpY*o1hXthB5l_q^~WKF7MKB|BYVtjGkT-C
z`hZ^<gPygrsZ3c#b|LfDESJgxBgH+s<jO<F=!bPW30FqzcBmFvRv)jLo0Y_H7Q}*&
z@(TsbN87ASw^;lGh%+A`lKY5nVbd5pVmstQX@5O3GIChrODmtsWOmSw4gCD=01hRo
z)_NE03KFE9+K)+gQN9$G8FUtPZ5WGCT^J2gotZ2sr^lvpv${joChwcROG>y7c<`QC
zw7tlx<+ejsXfq_d20*~ShpNPlTo=BZv;fV0iv6XFnKh)v@UoIbB>3uYG(?TN)ym({
zDD#A4$apYqXJp~{N4lo#Q0qjtOEI0Pw8;vP&T=A^o|>6=lT46GDc*xSEKga`lZ=@2
zCrCm0$ArC4_0@~eDSo#Tf7b3w{*Z&96x1gbW>!XKMA!CfqujG3He}P*^?qAOg_lW%
zV<JZp#1|d`Vp*q!N#9((M_IHqlJROhajJx(To_R{{(ueka(Rb?T>87u1~j)i57ILg
zWT#Mmo4tkqQU0p&0Hi*#M`;Uv3qtUR6CwuIxWXUa6g+?4!>^t{cYEo3#=%)+#GrR<
z#+|?FXTQ#CakJ;AgHcBr3)S{EXG44?bM&xK=OCJJ?t^3E>qJFGmoA#Vdo_KpND9Lr
zS$(-kg(hp&ua)7!G$w7`=uGN^7j(bl5@}v7H_gRLJf=(bC2%%Y-FJ6&)~Net5y25|
zVE0{d1fA$nz&a&tQV9pJRxJKdf)M6>P+zAyZ?lp}=1G0wuiFFuWib+_c4$jZ7vnwi
zM7)mZJCO{m{Z@bfHR<|lg!j`s8>{`}{3y59MwPyasxXtaKBua;BB8@`CC**j$@EK@
zw>dcrb&n#v$Q9ykJd@cm91F^5!LsTNJ)RU6-&M!N!xBv-3l5MoU%YrR#ay7oCru9F
z&WE#4l{mEIoC_@*L)4ZfZT(DxR&F&La&2GTT~{y%FYEAUt!Uqgsbr-2F)Xl!Qo@{w
zfbt8EIb0?-u8DO%BnV$jAZ!owU#@lq%Fx44WV|!RDfcR^0HPaA>Ybh|nbQ5Yad3o#
zFOlqbSDL)D+jCOez1~$2S+)<{vJ)w@NK=OMyA}XXij%s-3Qw@YE{mmJ%qEQxhCL?F
z?~s2KG9EoXGk4FVo>4;b95u&8MZ6A^A7Yo;aMTCXu4SPiI1B$#fhFK(9SQ<)ubd8n
zc@<DBlOe0Js@(-gmLp@#)0=f~{3Pa0z1J*tBIor`+dY$23@NNkMu&HQjzWjG4p~S%
zs`!H9LIgoiyLB_QS%gj}CdY;O;pjOz+cqe>&V1=J0vv_8mW{<FLtl*9_ID@Ez2*=0
zuyOCb*jZ6gk){1P!kNI=CPLRDv%~3-4ATy}9v%czYAB2MP)YE4t-zxre#+J(hmjh`
z6j<vPV_gR_T8Abt*bco)>`P|;aH^pbkOF0F21TnL<@4>Y(Nr4Bt2;c$p*sDXLhC3#
z$*ns7WT#i|QRJC=^$ogA+*BsDM-`oIG<>}pGe-*&>c^!P5=IKNmm^uxADl?9Zs9s2
z-#zb~u9W-SALHm9Md;u$yYgWI8&+)CvvcQ&G1lqxktS(xE`pK@8SBogz}?18DmVu9
zO!I`T#j--251*eo&%9r+e4TWy*3<qo7jJ+2L0*<o{^t<w2VXkDaENI;O0L`{G?dq<
zA~!_4Ql&IL-4kShL?%AsMEG`J9xWH4zQ7ial2b+clqTnNTg%kQ30m%(I|@!Ji*+n)
zy=EbR7O?U157>Df)He|Gx!E|8@$SOJZ*#PAM#IVS*>^s!^V6d~gLcj&dK@7^u5w(7
z5?HSb52x|#KOR8^K1Sp^_SmlvTn8G|aZ%dIN5&f9_qkmQkga>{)f=MeePUVV3>K(!
zNn3Kn{Cz{yA@lNwJ^8P&r16Q#ZF5u6b7AxNg`gFI`J0rR3U<kdbI;xDEw6H2_3cDH
zZ%yn~{J6AhUy)(BAF!{NZe5X}=3`t5XBb<s$Km6LqLU+3Un`M^+Hj?(msS}PEbZB;
zy=Pv0KhUwb8rF)HRF_mj73qtz9y+TwJ73Fd^-FD+I4}Dq)B+9_-n`oT@@!|~I4$TL
zwjwIG$w3GrXKJicbKM5S{ltF&LtacEuocd5LDMj=kWg1IKmq4siDqO2Yzoz?DX)op
zQ&UOFnD=oqPgwgapDb+ksshwqjqKW3E*7@J)^MxM5%N0BEX^{zw?-+jK;7Eh@_BpP
z&)x!A_1pf$C!&Ep`IV7dv=>yDF|lFW1^>eI7Q4sy#_u!T(Qgp<eT+4x#44GShnQvs
zCT=26ggc&+M!iP2(P+obv2jH-;b@N8?`Kh(mXLMEHtG&+n54+R>m06n!wUeMTL0bj
zpHpD3munoVsT=h=`Jm0g^zEdHry;pcYt|m+PIxirElhP4F_n;9aJ?2ycmYY+`!rz)
zNB=iy9wl;>7Kp40cHI_L){mmE3|gaY+SF4^IwOW@I6BpiV?{l?_HyB7?OA1%v5)L?
zY?pX<E#|$cBQ;{s;EP*VkJ4+qZDHCs?%cV<!i>LU`!2aj4<sWM)v@@(t#_AtF=6$=
zqq3i3Yra$-M>?!;VFS1MJURBA)vD(Bh#SK0(atNGvIs{EyM9!bij9dJc|-xj`1P{#
zXH{F8^N0XQp1YCa7=A2dd^;B3maTUbJW#XNT{0a<%@U#O+%)|DKh$pe6S!?YKVHan
z^VDdO)bM{I4Tg2`s)*XjD$HhB_n;p~55$xhLP{^nG)3$&=Kw+vF9j(9c+K2z)q9(v
z{m?&Q7CP)Y+{#aq&tYasmYdqvVe56=jmQP}4lxYE-^M=}V^gu0sg%e(kx`)K{@Qf|
zcmSXe)>MrefwU9->Xa`22PeUF3{;mcH{^ayZU0*(UH#o&$k-h)`YX18?|26`Us~gj
z^uVsl4V2h};00vDC)TQ2ztD0Iyfwy(fA9lsBiA=4)HRFb;|on{nNi{${9{q-DgG)B
zX4H8k&nT0QiH6OWEk%}yrdpaYq_m0{vP0LjH8|dibl_CgmrXa0`5_V74}O37@})10
zA6XTBsjwt+;wVh%{xfWFs$CY@b!v&cnmTHJ^viR3)Lf}n<(j~dF1cA`qVBlbm1e+(
z-!2&q7Fe5;`mU(^@!5GaZecU+DB32X48GTh9Pf|+{yJZOleDzrS;x@4?iF(s5&H2e
z53`9zSIwZ)OTN9tR|dG2r8)d8Pr$k9k=LV{GJ{>=<GQFtYKhRN3r(#<78VvtQIw5H
z3c!O+wnJvNLk|~u8~~KIL&NT+ofKv?vH5P2x9w2;=;-L-MQQ?$^GF7D@`bkKaKcJ4
zYSZ}4fDDm!I<Iy^ZkbW0<f0cox^-=nmXWQ0%O~bLyi_PSPClR8VnSVvPEJV#Y8g>t
z+dh6ZP*8Xw=`zeTP^ukf@n_aoh;2bDH)sG!k>26Jhu-ap1plq*>}gH#S>n6aYYSp2
z1{So3I%y3SB_gHm1+Nv0XZm7b&Na&gbjs_$$Lw64zK(745~oa;%{5MSgt%?K{*c1M
z0zUh1AK_c-zHDQh+A6EWY}yRVti7r{yw8PS+9aWD{(Y5pTC7EO1)md{$iRkm5`0dP
z?XkyK!o8%haURSi#mC2(%(IL5<6M@Ucs+7TmmGuhIzaZrmG1VMKSgd1d$f9|u+2d+
zo~40W?xh5mpN@DXDRf*eI=YpTs$3>PJ*z7H0j<+xvCQmH*|X<qsM}+f)6n1bF9YYl
z@)cQ;tXtV~vm@U~qmH4@g@i7HUHNqeHpa%)%dV<M<mr=bMFe7Rtnk(QAMOyy#T)z2
z1Q5U-1bhoajYR_X#MFi@`l7q9KHerk2(aVpzPYsgzJc;)*f1<aF%8UL+8^SAW5peN
zDmVNQuf5q|5sUv77Ml3p4qnq=ylBR+HS9ow-~7F$RpX(~>_*_jpE-N>Yz2^B%MKTs
zm=R~UA{Td2OeF6n?OU|aN7t@Yq&#z42ZxhQm@QYD{YV1gq>lSllGpX;Rk=ZI#LSGW
zve{=TO{EAz(9qBboF*YiqIBg;EPf=84cVXTK5=1I19`31#O`|>8xEE2C{Eb+ed!_G
zlHf%Am^k(K6+ypz|8dHun!>l<1_nF~)g9Cs?5=jo62&%-`iXBu@UFVu$f2~q=Evf?
zKskIj>0y^qv9+hCXXqlYnj?Xf!1+EvzKn5gHOww`<Kn*0V<%JjDC<Oqjwc70GG?fT
zHEp!mv&fw|k?g}{TLFbHGLOtC4Qo=MJe_6|Cz5ck>QdVnNtTzbPM7UkWNYD*|0hBW
z{`B5Bhw_LI(pMdh{UBkH{ULRf0jw3u)ce04ejsQi0(;mNw}eq<leb8&_DIKSgf&N}
zI*vgc@=LHvz<l(EM+GdO!bsS^VNcJw^@#fp=N_oO{BmX4HQ{WHYEVdcaqO#iFZ=$O
zPgiaQioHhrvlpf!LS2cSdLZ!qS3x;~+LuEC<IftSfPX?xJ^5~YeHBi^0X!*QIL5F+
z^-y<PpcZc7cGrE}P_v83cIE$PSS7)H)H<`Sbz~Ryn-rG)1z<hF=X53!rrD7(v!5Iz
zu(iT{{{>|I3U?pL+-lr9v(7iG*owlKZn4pvWzl?E`2Uf1I_ED<Qn&*K)rxrss%X)k
z7d<Q;>7q7Azj9Dty<(5N1Laa-^rADMnAC2Nft2|VX52?5A>r7Uxs3?o9O8w=5<EOP
zqILjI*eNY7-L1rdTu(8PlyV_C@$ewd7fF2(IFB<A%Nq5D`*=wTydcYEYhmEC=dP=W
z{vmls^gH2;JMmKAmnTvh2$N-r*sKf$dCJ{NfCs-`rJHMZBPn>h<h09d8dlOju+f+{
z9KRBRt9Fq(`6J$wmdtC-QUcMfD=7e|_AX0*r^Vn_JnYR8$%wX1)Fz;=JiH((J_30X
zouZUe5@yhcCI<y{A-n`~B~ud<9q0U8@}V@ijrlSMCV)^9p%;?qAIov2+j}H@rT(=E
zTv7oMtzLqvD773_6s5Z2O0~#5viTze`%ylTOzc?rxF{NA8Oc6U4O~ejkPJxT`GU5}
z4at8bq<Jm;@w&_m2rqc!iv~qr8e|=ua>xSR6%*9Ysi$cwPBI*^VQkzMlU&1&j0Z*&
z)PF|8K|RlYrVcHL7k>smGoIf1)ihf~O?We^OTMMs=yIg`ysyFb0y-8SK7K$XnYDgy
za@)*QZ@kRV`^-O^*}>QzNlD~WcWjM1pk3yM5CwxvF=;Y5$@WQ7>7I#ZJFM!jdg3!`
z4vL4{DCDP&nJY&+^M+;{^Ae}VJ%LVIyqFx%YqjvOx9rrXPSH=-9C#|JaJY?!yu*+6
zyT=R;fqmblQxFlKGSo($WKXqGlGdAzUXlClEs&jA=M14Qc!!xQyulWv30^nvY)stA
z*&XnWi0&_3>I<;_+-sxaW+SD~ix$hMxW%Zr*eq*`Nhc@X{=Y4*r+A0i4r--AYgwp6
z!6z}EIpw1}r%!-qP5=_5>14(`k+N$Li!HzL`$p6@K>E-O2O<!Q|1Yi)Ae9Kp{h*E(
ze60CxQn{*@5SvY#2)2IPwzc#7it0?65a;jMWB!QorD*LqjR#ef@*0-CoXBXtx0Wsb
zg#L<P;6BSq@WBlZlRs!AJB|2zuvYB-VX9p!<5*4$I+5Pt1pIoYeVHf!?um?iXs4f&
zat|<Y?^)DG+$MpL7}y*@(A&Cgsk2I;wtFAH^>x*`d!EBn(_FiZ|GW|Y+TZ4sIIUBg
zYWWsKUvzkUY}dx_Qta``UF%l$2mQ$oY`r2+IW_xzHIdxwiO0TwGcL?bqWlfF`PXSB
zxwvh<!@+qwy}YjxdQ#oIiXO*(n7g_PJm>#&gmO3)xtK{{r;fGfmT(cR|0dLNUXENo
z+(D%&nn~IAZ$W&`j{IOZzq(W>{(_%k6)S(<z@HXMHK7g;l-EiGDSE9UawGXKjk1;3
zT0}kyFUbR@crjvuurv<;$n+PY<M(6foBQWjq%n&1Q;SKG<w{=^J>>QM<oRkEay^(0
zB+ztbJ#SWh4>^#=;Bze9x2Tghcm>ydQv81kG0L1h$_OI7f;^x0e6_aX3;_rU{(hpy
za6Pi1K~8O?6rgXcuh`v>*|3=O1G0L^rA<r!kHqpPJ{Dmm_#d%^0bJ=x@&FUIZNE3P
z|Am^|!^P0z)CxXC|L^5ogjY$0jSr6>J*mG+esnsX@WpWNqP=weDA={nN(a#0p+kpO
zEaqjmMb!8i3b7-){eQNW;c|~zKdF3QmR}v4s(I+CV`{X4nIB8Qo6N+Rsg#+kggWah
zd%quyorjs9JcuA?`SZ^<5g%CLcHFx+C9vf6Z~xl%?ca*b`(t4pLL=}tum2}23u)WG
zF9pbnWj(K0wFi5AI3b95cKvC|ui4nI3$wvuZIv`&f9FrrBrV1=WSfP*6A6VJz)lOp
zAmp`<djvs9=R(}8<ZaXgR1cwAdCqQ~uNJKyD7t0pDhXLS7k&aShsoLUf!s`Zn|MCo
zXfBMn{yAh|ml5Nzc3C00E3f?u3-qcfUVoVdcof8nh~fVXVXmiTuuT5|pH2@_B>Pp<
zEpZhz?ex?v`Za<*wMPJZLmimInf7mzn*a2N1zQz4wQWW%5dB<o?HoEgTbeqU5bEsO
zI{Bt>*==D^Gkm=+iZa8-z9Z?)scm}N?nylpG#$lzPhGj@f<!tbs1FGp*m|qTAF$PJ
z2NO*`Xv+fk#)sWibhGru?Z$Z>AqNv|A-ud|@h>AnU=KMFPO^QEti9L?`|@5{dvmP<
zzW*5Pe_%`@_0Rv8SpSdSI?|!}JJ`7iDUaa)lqjtR2dP$6It(|Wfjm+^+%`jf8VOWE
z>Q{!Zn|Er7uiFnkXUJcPtzz$ANjeUj>aikAx+B6Ky&1%HT+wcOKpoKg3=Z7Or}aeq
zknghJRYVe|l*_IW!}ZgSMMg$$SPb+GBqMloRhA>MM1=^yq)%`2$;jzWzohf^3K4w)
z8Awsxx0KLt2lrLWs)^l2E~9>pd$v%s$0v%z9<e{sLKC}RE-=mnj{W{8569wu2is&M
zCi&U*F~>-FV6&c=tcbF!Zs_kkU(=qjA4u1k+@GNKab{4iYOw2pUUBje7n-oQ)?*{)
zeUuy|1DpHX)n6ZmEvS(_L`<_R#f6DN%BrjtM#Wxsc6MT8CeX1=`@t@t2W&G1CYX|h
zvzZ-snH{EC&2x*K*TJK29Wy&*R8-2Ufe<s329l7So5HUnq?YxEoZ%l>U~G6Ijsw~M
z;lD&E-)Ufo|F?%pOmY%knJ#gD3PI9bitTaR0xckj{!~927qYM#(TAfR90&U@grJE7
zZGfN5Udej<3mBK&#(-;!sOrm%r>VYDM9n+qO$W<UE?|MLoIER-9G-f{yH+WiZ`d_j
zyWdiAbQ&YbX&H6Pto^NV%oh14@#Tw+$W^xjf)KTLv#)`EHvK*!V%qPf?_U5^h6HuL
zh%L+f141gfnMZrp3q6i@h2<S5e_T=bBTVT(qD!t3@zNqo(|ce0)@ixN5lT`$N3Q)b
z(-&D=`1UN?6+EtlEyV%jXOFeo%*EisaLSvgdXP|9pNrrt|K*B)a>3_EzHx#qqI(R%
z8)aydQ}3<)Wj#6@ge{y4%t3PPW2CLT@{LSCLeO<R>F;}5*d=z_vpIw<*es6D$Ia{8
zV)^%&{MvIY_l2qFFo&kMTE4os`gd~fjJbfV-~H24D+l(tNJeBZk@zNnr~c9}eiqa)
z(`1V>Ef!;8>TXPBla0P=;i{2PtW(UU@Zo0?k8=64i+{`EwwoL_T`7D>f|dWFiU@w5
z!O-qsP@PC(5C}oqMwx}i*N_i?wB$_Ji_ES=%O$BB>Ag4Tr>c|6g+j;mFeBdY;ur*p
z-3*r5LBwDocD5C^b26~)Kf)XS$N+U;TBvgdp-#u$eSIMJl3JtU4h6Ygkk}8EuAE_J
z(H{a61XcC72*$T=-J&g?G+-elsL^M$uXI3Fz%!fW2u}Uz+b*d8kC3)%zPoO_!xWrH
z$W7r#%8!%TF|+Md2d7b?vF3m4tbsq4d11lQLHxS`>7^xz@4<iKk{^jwgG)w=h$;|m
z{-y0zqCE5CL=@#Imw>XHWeVEvFpY0jKt{#<({D5oBJc4h-c?Ox-8Zx7E65Zne7zRG
z{-Hzq=iwjdX<C>`G4sq55sUFTOx3W(W$v!D8~-XKY!kQ1gLK&9cc2N@#AV`1_Mmq9
zro8+)9KimBb(mx#7gL_O8@<IBP`R&bA26(c4|LwIG1*L=U@o_3o#D=@N%Ru-4;VG4
z9`T*!Lq{$NI9&t5loA~qKK3Q?vdV&%|G`MTKe1}}cWP1$g{c|XO_5iaN-9BkL=j~e
zl7B_ZRWyI3pFX+aFH`8W?S<3j=GAM}9Pc^j6Z)|D5HO0{DO-OSMP*0H<%mYLtSUBc
zh2DZOsW7es5k}C1`;N7yW8P9;3nYcttVBiIp~>v*>^LHRC1L<XPH?_UR=&&I#k5Vn
zOK!f4YpEG+QOCJj{b+f<OWpMJv?z#@?zh#dvz*Q*tl>=Z9+MqP&?|6^ye3g$^8Vkg
zUxza%B1M!{|GoBiX%-=W9IwR&*jE41iBZ=h7cdo3n-PggpuMZdXI|U1c11d<%GFA!
zo&V<^&6mR`t2dCILa@h{Tfb)rP!~9GrtPD%+9Co)q2qUn2AdB;DBk&VDBewkAb(!}
zkLbnD;SsB^n+TJWX*W3pM$hD1l=0egLL2Qm`eoBz3932$o);&U<lOZXE2fu6FAU-6
z6OTmIwf?vlmI$ZzvRyH%zBPIdm4dv6AGV-+*mPX==!z>5k&X5$iQf-h_o1!G5`sB-
zPWJK{#sMtee)&B=ky2f<l$Mb!&xciO{{VJ;B~R`u4GMKWM9$KA<8k<UJ-FO|jbfM_
zQj=dAe5933u#^l7LM_J3Vn*G3gCQOHZ)kFdrzZ>;cM0dPPEqP+l(Ch{iG`EMGrqz9
zYk--190?{clR!6_=zKs#Ab6ZbdP>#Z_#c={k;98|{>W=BqvT#zMk%HuZj-5Ne(F@Z
z<yzlR=VR5MkS2~uA9wGSzF#5im8a@YSWb6}|EEy1-LCW5a$(O$EZr^l30d1`R_S^j
z!zoW0c1{_gT{=^9JZM(@1YsHaCxO$U0?a6Qd^egQA`mj(hM8@#@gr1;m~PvK<#o%_
z=s5%)E%#zVjYx+S6;J(<w(M@G)lQcR+Q$1!y-f&KxDZFkw3a9NLaO>gFLI#RT?q2I
zU$59-wjjH9c5n_m>J^mlpx!ZP4ZQIAcZOr!2YC=p2%u8;oV#Rmk$wL@TMegq3-#%|
z#83~_1%#HgV(q4-w^O4ymR0{tUzuWzOnVoVjvKb>l`M?N25~#f{BNmD>mqhvMwA0%
zP_}9(bpUgoWB#w)f1q2?iC*O5yPwZs@h5VYy(q=PQAIs6C$1x|%=-|Ty^MA9slN|`
zoE=Tr>-kPCd7ryI%%s;f)lF8U^mlXFC4ZJ8Sm~nzq0UhRfReksh8_~QQ?q=$tQx+r
zU4`>HtLX(HboSbKwwk_e<((hD@IR*zv5|+tg8pl6$U=JE&*cmMC<U4PLxlKWVofkw
zKZ$bxwoCvqwM@W?QU5;0!w(G4Slf3oVfWy~ADM3?><889zUq6Zk08vig)RQ|(~HSq
zRnrAuofVv?O~z9`mH2Z1+gwHxa`6sf2=ksY3}m>Vw{Xfh>OS-ws*0@z>B3S9$9{{S
zf`GrQ-XIc?)3{O{GB=KXgTNC<E;~Z@!WN2keX1A`DXEha>V(-=CH;%TQ0*a77+5u@
zXH%Y=%-o&YMMuf*s8ED+NwQa4`#}u^vKK_!-%mJZyNLV`JQ{28lf68;P>kuuWqo-Q
zIEfL|Qxz(2U}a&$jp5GE?QW<a`J<P#JY5v_w(C&AVX&3Qb!}9$wQ6KAjWB;i{@oeW
z_MXXn-5u^<znu_Y;-La$B*l5|s0in5{O6U_$xuc5E5;fwo!Nu4^Q|^k064y0sgVvW
zr_myMC2woF_xdXF2W=;HWjI%@{nt?GF#O^3MWrK#tv-<sJi24kYI5IkZmio$BvR;m
zI)iVqj76UERC}ZSo|x3f8B39|6W?nr`ha^uywd^usKhupdWTOcQzv&@J3Y#194bhO
zVqX^UvuNipg_<3#6%4HcsffG<*&(6B1#+m-yg?v7k&sCpD^7)XIQ%c1rDYH<0?@4f
z=)K7xg8h_^B2AW7u2yZz(A764zse0?yQw9$R|Zsm!7cjYVawO7S;Mxd`hC;)<!qBC
zi*KmHpF$704!)Bt<2MDgA8&CO5wI*X!lgY<+g#y<7U;lUyMRiC$%a3rYmt>iN>7lD
zP?~{$weqN0zjKW0ANR(R`{vbel^vpkdOotnIznvm^^#u$Nzh2@k@TAcTalGIv-b$o
z5?#&V<1A=|_>A!PJ)m}TIj?}VN%h$ZK3wS$a-DY3A@ULTz*FQiXwcTEJ)g|)t!43z
z2lns(n;KXmgSFxpd$bD`s_@MJma^VV-PCmkQm3E<Im}-L=`Ky&QBZc=gBl#^{&Xn9
z(I;P9%)m0`Ab5v3{II9Uz+qM-<<7geR~IXBHY~<D|B01n*6To8%0YG`hZhHAZ9(B(
zMEIE#_`gzCSXVQsyMgW=ymEhP$RNl4uNl3+6IGs`4c|7iN)HJ9KdzglueSL2#Ztcg
zgMUMmA9l~ocZH4bt>g7Q1hr5~6pfp=ZaG4=QT{qaEB)All*g2iP}T9SIfj?j$tI0#
zi)+^c5?NSxUk<*Wsj*7?`CH$aGl#16T1a&|{A-K~2tzzm<Vmx$-3{J4#tt4fQX+Xb
zunkqMY8x-|KlHfsdseuR&RXhL?BP}yR-5Hdp9ii{XnjYbEM!~cgr74u%XHQt!_@9U
z3u_87)~!wc#jvLpiE8+(dIJ|K1f7CP<(vf(ElwKU3XOpN@!KP$%vF9><VFY}%{vd}
z%q2An5so%CGtQs(swwx4&D?O}FE=7*X&fD2E@LcdlP2D5<MTPz<0Y|dAa?G8Xz0x6
zbtg?mr%m2mo8>e`w(5KoT4@!NNh;)-oSG`vBPS=mK-4zE7|?g1j=oG12xo@nzgj(g
zrQnKYpEB7-ZA%2f=th_hxJSm=+KPfEA-*>FF0ZNGDl@&IloMGVv(!}0$K?yNsiJLD
z(Ba1HL+sod8INSVE@tUM_DZhXolDU!dP(%gJ3AkPz!{oPF*>1Rt6-$vO<BtGx4)Po
zh}9M`ba=XD4RZfd;b1Ht^kw)h$g#9p((9Sk$gd=Y2$S2r?YC_vdPDU(Pg@ln&x(QW
z2m;vSto%Ap6^5M*)Ik)y%_)z{2i_&Dj#9oyQPF*dsRV!0Bw<6mmR7}&sL@wyDarzm
zBka5i(ei_FF}O5-^#-+1&<sP=YL75eXE=8N=*zsVVcK)JfQKNbh|K;nGhairE5A_k
z-mI2r7E^60*aB0-rRu`>MhZ5XBIhw@Gv6N8#(pSJ9e_DMl6H=cLR`QrkdvTuLt;Zi
z!x%JXa9qj!2_Zj!q7urw&OtX9)y3DGeV2YbD8S!e!rsQFP{Dcq{Sz~u3p}-hn@~>C
zCo>Jbmqibg3Tbe*j>{->r|2n5zO=?=-g8bP<YVgt@cynkv)P1HoIuDiY@8jP@2PkM
zZ3m(opw#D3FYXrvuMNFW&xmSr_^wf)e4Lz|+>(e#)LNk-j1VZ$;p_Zo^vO0@9Y^g8
z@~=d15eJo_wTwEcaF97ortQg@7jCz>h2nXtMsHx9+o|>2CIg6rd~k;?GP}7`nk^~N
zfr4PFIFrQKfvH4GbZw%3mL3Wr6o_SUqwWmc_wYjPBD4+i*<mxfEn3wRX4f*@#%kpv
zD)-<9CUs`DR1z2i&_Rb5JD2P<dwy50?s138cqg}=rJ^!XZysO^RM{cNh;ahpaqn0o
zR<@OTpK$Wi%VI`>*6bW=`J6a<pRfl_IN8_rO$olr1Qh+v?@`~{!N34MIo&J2cEkQ)
zmGQ7F*&|a7EhqvK6k`rS8FK5V*y6FZ^wLZ4qpm2_jYQ@`o3)B04K-F?5MVuDMj)lv
z35Dlrgo&{R^00$k$w#E7K*wk+^%6?c?_(YD!+ELNFe1hAY=}(!l((^?BTZK(=FruZ
zh*em8e7s2KTxsXrL?Woaqp#59JZP@mD_8Em352N)$S|WzraWe6XC0QsvCA7<O~xC@
z(%-e1*+mfGmO&qKb+k1W-)3zjc&OZn6FI3B9U%b+d=959Y?CF<Gef)QTqe#l3&<&%
z@G(+FFl5{a%@*&`bCQ#{@`H1Im`<njT<3y61wE`!>KF~g#*D<tAUk;?Th^lkey={h
zsfP;B9hVACVhES5Tns)*RJ3S0EcKd_KMaK^FMN*JmXiv(({`o5IW+*=e?AK1ICCfk
z5k^5I?%0#(^YFrqlt`FfL!A%W8YN+FXt7?{J{G4yxdImV*S<?FD5P;?p-@x%ovQ{3
zih{e_Z%18#BH(giYJ2uT?I5FMERN<`)Z`6A?%!Cs3~Y-ww$>KBeslm=k)_qC>Uwul
zGb3C`@qjsFGt~%<cpJ&C6~ls3xVX86mO(WMd2?v>idRl1fl23uo!28;Yh?H?lJYB^
zN>rKO`RS+SNrhJXGf(ffh&qZ-2T?`I#f!fxs{?w`7LanXjERGFRknDNa&dBUPjxOe
z3nuz_qglLJS<r7HB2ZA-J<q)~>Re^ZWD$u(3g~%M{sdl_{rPg#p_CS)?wi%>4XBT*
z_4b(B1)ajZ!^%yW8yQQ7-uCU=kgvk>WUHI1^WDi0>3ylB`sO-pYRs0%N_YdEv6|&%
zWli39cB**xXR@4jIPc)#ecsu5d>)$A7<7oRB6xl14-l|r%a$4#`RBv0$cYeMLEA_{
zK3-nKRx<e#{*#1!uHBxwmB(*RseD+c-e6q!R7$4!LP;5JcVvs<$_V(dl(b#e)L-cG
z6!-MyzgPs}(`_G!^}&P?G3CTxt496wvKdn((YY(|8E0RNHdQ&z|7I{0`gHHI<R;b%
zJ#30oufZYRpKT)7An&9xR-$KVEIiTl$WW#TR9DH?oxdXe;m{-K1*tI9QI*A4fvCV!
z0Jh{85R#EoN=xat@seC|Y}PI_eGx*~IGwrFpL3~{r)>_l!Q&aJEtW$q<8q3MMQ2MU
zocI$aUKUt?5a|*4lAiI{{cMgv`6qR$jRKCP_{V3K_U%LR*CS<7B?%6DCMjC(PYFKC
z2h0bsso{;^RYc1u1i#Ou!agOJL3zg{0f|#jx*w$4%H^)KUb8i(_A#%c*MS2sB1kC-
zxgTOZI^{Ej(y9WEaB%Uzj{6vyT9T${@3u-K6OXiSOFKZW37zr`pqCsY_f>yM>tFK?
zLImsgU7G1UBsx}Q(QLgd$~=6$;^6FS%UxN@rp0&%$Z03|tRF+1{d0%!)JT<kCg4h8
z_S+5@7nib2MCqqmt-Gue93xw3fkO|$-&qbS$`m&%OwpwW1fx!OHf|IU@NyExlh?BE
z%DttRuh;rT*dNZi2aX;=plE--H*R$NF2dMERTX>UqXFXMA#Iha<24oj{iR)2R`%dP
zns>fsM~;n;k4pgSev1$U3)9KR1PM2$ymJXU_`>!um^%ZD!-0^Y>PvR_o;{s?u{F>M
z+EdY4FxC);p1&t*r~3yn@K6yLeVRp${J^^PzKLYHiM62pK~S5;N0qt{y2}*Pr}LwG
z=}H!pwlcXpEGB23R|fFuy&paX-MXfu?%17VdeY%wZ72RYrlEMNxa0^*=EGdk*&V2P
z*m%e<gL49?Iy!|aE^Mgx0Yz%1gj%cJWs?188m;`RD~1liMr0#O`&V;`<Kb=M@q3jU
z{eBf<l?(HRMQ!4uI@T(3il)FYGzn@#<zrWpee~;v8Bdih)Bx*N$<XRpC#EKSI%%{i
zrp2HCDvnp1TG}J4Tm}v6>5cqim)$P9!-m48nS+FTNEy1=W&|2DODmK_L)f(V#GuG3
z)an|CZy}5~R$_|Ve+D11eAc2eZ3?8%=FjGai+K`ky0+dmgZB2D-py9EIG3F9svMxL
z?8&UN+|>m)Wz$$+Kf<jzL({}m<U0%r<9*MmhrxIrzd7fYD;`QOGa{Soo=!NBzG4%Y
z?^jr*k^&;r;!8LHmX~8cK<PZ#Vh6A{kW;zN^cH8W*qEu*|1oZQ<lExqS<(lHF><tl
zo=hRwNUx$McSO{Pt!z$yb9;-}&fNYs$8vhR%!IYo$-vy%3>YR_Mapw*b#`%CNF<Cd
z?A3&>2^Yz|HTWIeB)c4T-yUyKY8TGSgS9PQ%BEC2_nJ^<<LK45@>fQ>XQ87JXIop_
ztwmt;v(YkU{zTRW;Hs0QrKKSlub{=(F#9H0fEm*<HZpPml(WW_X%BYgCNk~8(3<yN
z^vJq%l=hP%%CasmV7zpG{NrSJu31OPS^kZ($^D@_Wtw3oZWX2*ZKB}xuITbWiv7a<
zyuRlz-(^4=BKsS{InvJRD-EsjxMUO*kyxnV`zuo<b~i~cFxUMBby8X;HIx%!6u=G|
zj2+-FF2y?xQD>E)uS!j&u~5}$d8|XZY$0=fA$!Szqkp(EWHotbCEuR;F_YSBxdvIR
zzM*-C$Tc`+J!?4)<50ql>Tmgy3Syph>}5aIslvJ)O7FN_J1#e~W0x#WbcL1#xi&1M
zH?rSO`ZUq=NaY!9Mxl84GtY*ewrvE*N(<9|Af#jDb+>lG=HkT>x&E_)Z>@6J5i#mg
zRQ2dN72jJT%A!=Y9PiCHe0_qQ<=wu8n2}MlB4L$}B_e9F6yI@Kx#$}gnk+=)Y7<n;
z9b-+6^yHv9>@b1#YSaUFJ9im8fF1X$By~SQIcTmq|BK5xD`qD>ayB%^+UBJwYVWpF
zA#@vJDdL5aRnvDp_Q`)$Ju9VsA@ZfPFK?Acd1gU@tM!*<e2AL9o?hLrG`bm${$A6A
z<+Sctr{iQNO>pDJ3o8}@FIX8j+SE8Gbyq2Jl5{mSH4VwuRgdLzb@X!E+A1H*yN`b=
z*CE4-YQ*BZHnw%v*Vped)12yXHBXK*-R0u7R-NHo)V>Kq8o#j_&mjD;;PJ<4@o%2O
zGMF$sj5ieC;|P_DVy&>pI`2tD;OGH_9oWLLy7pRz73xYIyt#zcs^~8PP_fEaq|mn6
z#tUK^UZ!_dmyECt%_9*VmP6BnI)wt;*mo!WU6U<y9PFIB`Cu$Ge!^hMO!nNlT~Ts@
zu7hInSCZUHCZHvi&&b<5FP@E&loFU5m0f1LmSV@t$}TVfJoeDpUMl61Zhl_hKDwoC
zJM97Ha45W3Bs1igveP>kG=Z6NbH&jT(ne2J9O}VUD`<Y;Tm(%D6i7Q#G?0h-6LBG9
zTM%3<o<ziv0%YDbR~foG>}p~pzuQhLn;WKRD`vSvygvN7&WYd0EmlWD09&{+s7-0V
z(!=nvfx*CA*1T5U0sE`u8?KhIo6U=&lc2>l<BT{hN6%RBt$C&%!tb*rLxLKhI0^#l
z0Iecjaztod!jW@1=BFh$1UsL<a%EcWlQ0szKI)K_mcGK>q^nHSc$jFX#uRT&-CBgQ
z;$f3%m+bG*0%eY9YF6oqcsa(VH2a%MD=N0CxR~-+aUb&$q{&^6D)pQi?&vYK-D@-N
zLunUCtx<}D8{A%N$KaKGT7^B8+r4{7=QF1sdjro=)#GkEcVE~66%=le=ue108*_?W
z*;|T|ckCAAj6BLExc6CEV&aG2A)xuN2xMW>gAn*r<#LF)nig*o5R@Z7;HIY5*4duL
z*FgWMiTAHxzb1!=hsPYS-dC#|?c7Y1<g+d=@HcfDe4oAWxen~4I;Fk2vSs+X8=^9i
zrq_ony)@_ou+Mh$)l@RHHEybh1uw9g+=15U=2ELPG&SqGAWY)uT$n7-Yq|UYtkm1|
z%ggSstDa-Cem{Jlz>O9iCvzN&66-!6P_@MBI^owh<#=R6GR*B6i7H_AEqtZGGteuR
z&T^V$>*z>nm%kR(B1maZ2sESz4z(3%wiP%AWi**8SjF7xoFTcYC_gQi52QhZ3>Ttw
z=T>T4m1=^=aA9qA{n2;c&ac*;-0MB*^V{sA9SaH!>~yq}QmwU0xJ;tMX{@pUd!J&K
z$?kiFI7jt71%jlw?bLFeS)tv6-!PBb8&HRzwo*sO(oe|2z=@KB<ltk4t4zhi!jI7?
zv2%l0=%DKH5e?-no)t%Z^e5a(Mj3kUzn)R_8pbDhoPp+O;p2}+TjTJhU}z!wo_NPy
z_X;BAd@dGePyVdt!>8MzmNry^5kjE6yK?9o(%RPC&~S85Zm48G-1Ne{(b=?P7x=83
z=c)!Ap$xvOv083(78Ds#_!+0I_$r63t)>piJI8^Upup?h1AMYe?W}0SG@4_~W4|Gr
znrl=8Tb6FW>bdoCNoQp%Q$>b!!pnEw>0wQyly)_>wG}rKb)%C6k^c055~$_lNkbd0
zJL+!DOXtc3<i-ePWm!J!3%$z*5@0P=AqL~4bZs9@z?qCc-0$8}>fvgzt8C%7Y<sIc
zYaxK<lXEDsCtlRTi*l!&%;fnPR2Lt2v`@8cjP%6Ikcm5AVGc1bv#;j0eU5E=+tuZ>
zZ}IH4*tz5r@=1jQi_TU$cL&CVMq$jb(uhgOpaaCglzPJ>#XvM<1?9TipFG(-x~W<z
zosVeukQv?>OvrVvVJ8*JX+y%<mN#{14E=DrbtN~qk2ZO%%4U`VW-RM6^wCp|t%0EX
zLg_M99CgytEexOv<N*(3W}bJw1hH|B>Mw_}GLfn~&4b^y0Murja%$xYH6sP+*fI)g
zW89m*oKR@eqaeqs7M4!)kA;pZ&kLPgp7lrDdRtjtvpAs{q`ELIHS|8aXD_9lGlMS{
z?;7{G9D<s#?15PPNmPHWy5oj6D=Q`2+sZb~Mj|2`4cdcol=g@?ydzAH?|5xOx<-|Y
zDjspf^5hI1v{Y#K@+M^GU^APG)C^%oxVUkltvg@*bD~G))LQife(R0_n3jm@mA!JB
zmr5Pj9@w+#xCZZxf{ZORrlQDGbT=}}O(YYJ?Tq9)y4K4x_T4R9DRZaq)oqJ?!@Mz8
zHI!_OGu=VW{$<z#^FB#_fO+q^W5Jeod8GUK1oUU)d0J4wy=Xfd+S}*rKYjW%1H&W=
z7Ks6JYpZ(?*r*(6>nF7Mnrs-2T?I96;ZO7O+94zqwXb`>j#Rj^rz0C?3kfQkYF5Ac
z>?2s*No<O0dH>6ea-WHJCx!O5JqZO4%m)o?(APa})qL66nLlO#;+;uiOl&w>B;&T@
z#5swy<TdYUW9cjCW7??g@hG0|B-DyUe-p2{!)DNQSHzNG6tKMWQbdc7Vj@A+`oxa}
z?}6^BfQE0;P;u?(llj9ZFMnC}G04=V<(&0=X?JXK8?h<CkX~NZGENZVPPrCkDnt7G
z;@Y)qV(TPR4lpAeciopBTKZW8uQlyZ8wrr}2w4eNniu$8p|a#&dwzmYs!dB|U8(@1
z{LPgYv$3*;nNdUWr+p?|UdRtPbd`B|>>WbI$>AI<hbDg%@NI~|J48g8dfc=Zb=GwE
z@nLXF7JW*>#K%J7wX@^#K>=}(rjLQg*)QYq{K_0mZs|Ogwbvr7rccUWuKg(78b8!6
z#JSN5q701z$?hg8a4rT;sj1fJT6HCy7c>c;=&Xlgq#Y$wF(RhXHGPjwpg*<Nr@ct<
z(i{5|(_q+vB*moc!GpfW#Ie(tplf>>=#1*}T3p4*&CnOH2e<@KfVxE2e#zH$u(j$j
zt#Ye{u@n(zU?fxHvcv*=BIGHR17iZ|{@%MplD4`(AcRUjOWnc!dm@6o?K%MmW^Qh-
zON`o355)e;oR-1Q@`25Y6fT`4Fr&Z7w0(>%CY+LhivJ<zeXaz0$9PxvfE%_IAzqAt
zMfw1D;EhTt0n?|bsK|9FPnFWH8r8Cm6Co9vTE}F+xq{a_2+rQe+S=NTtRiFUOe$rD
zuDX|xRw%fBzy%t+f^EGB>5vHut~%hP6}ztjGhU*^ew6QTUxpk{FO#|Cy~OuAo+Ki7
z7Vi+Ntg?+NrQ6^#WjHx8(YSk$tgL5`#N3<$rTtou>XQ4Q>JPQYp+F9{RoaC<ojKeX
z&T;iE{@2sUIyQ_8OqAX;_mxPlDR>*~*kj76SY#((Bt%Nkyo<)Y2sn?44~5Dmy`cfc
zLMoJsPnjvprDl|uGghN|2}v#HurF`brIcR)zupz`_IS^*)Z>X~1~`cPm$4)FN&gT?
zxVJjpU^y*m_^0q@qM@rx^`U`Do-4_VVmI3g`%{9-nR<TVt)T{=R9%)5rR?~N$2wvj
zgNi!Ff=hl+J{IY~t$iVM+zpLb3K(kfimf2FGl_Gk0oiResYq$}6M1DdnE*DXj#z?s
z@Brr>nwU$14nw9}s702%yH9g{bDfBmJmnCXt(r5kVFT;;sytuA+hT~XBH(kG+^1kn
zY_Yr22+>VhSy?A(YwIQ|z>mzSn;>Ph`B2&=K_#0=y7oakUCsRzi0D8$cNo^hRe|ya
zM!+O+#c{~5k-<;3<LiEx53IYB^yw6Y0c)yF*wrpV5@;6umB$%4aKKJ(R9pc$iajx=
z0?<By(tZYIR!N?*GLFSZKqI6utW01ri2;5{CKq_ec<4Z*6~{z5Jf_YK;K&3>h0XnA
zGwZPEzyfpU2K6d9u=k_9Kx|qE{0j$c+BYm)EVwePS-N$wevukEboTj@Nw+eBQo2fc
zS)iNm5he>6Y039r0ku|Cn?rHt8JGqgBSJ4DaAru+?^!6@>r3A?Ie=5DBDUfnhcMn|
zHF+MWX*`+$CVK1AyFm)+@$A0MIv2FUDS8(ZL{Zz%kt|x*{uM2$bwi0(lB*1{DS9D<
zaMtjax2MEht!9G1;6hMxp(a<r`RVDWethr7G4=<XlBbT?c@Ha$X7sC1vdYZa%Cr?h
zqeoYV_GhCp&liLv@t+OR9iJg4(nd?yytghS_#`Cl)k@_M$wy~P2s&)mQNNm+nzld-
z^0-_kjkm~1JG;6{8No(g-LJ{wYaMAg-a+$r-RI#k7qSSSA*cMipH$iOLwumq+tnu4
z)|B-qLsS2SmLbPduDZFwW<f^@|Ib8*$`+pDW|abGr?I4u#{+ncH`}}=I<x?cuS(fF
zO@ziqyRpYQ;YY8zn_Vp5uYXU17A6}a^<+$>H2?FM>1E<I=je&=8VVv8FitU1^eCLN
z3cuIPSP-#JX#2pUTcg#xpABZ2kMtg|a}+N}u~<9<{B2LrK%8rMB65VI-6@Ew@jdmL
z2Jx6%rubMyYxjqnZROL+ct|lNIJz%1X4!b37}YrbxW=Iq@Vb~-8%08c<vv%EjN|t#
zA-)O@pd>c_3pzWsWu?-96ZA8hR0*G(WxiEtkPrZX%!yAt^3cn4(NRoHP7ce;#6My2
zJ@_Lmo?U!;;2=t;if`S#xp9#XVV<YXIb|;|FS+Za-wLaVvH0O0jHEE_Jf7v@eTBrx
zv}TykHz;t=_nzv{z&6$nK7})Z4_I$-)5k=OZvS7Hvj52EN5%{n4LrV#orpNBH|!=`
zfEN$Uo(pMz+;9r?&TneWCUaa5y^==KU@V51O%x#lb&n@GI9wG)IYW7N^2z^KJhwVA
zs)Zqj&IMUb&F4jWT@Vi(eR;nBJ$HO;fzr%$XPW#{;lpJ}H?BjNDuyi>kY8Yeq0YjF
zsv&D8RJ9aiOMF{StC16-+?Op8$}x>BcSUVN`=I7+Ka)<9&ShbC%<B|d4A9e#wL+h|
zu&(BcJb?dF!FiR39&jw=j&3(_d&IGT?`*bcY_yvEY+CG6YAGIcqtJF2yYC}{6gsPk
zQi5!ZV;?b+2scrFwn(RcXzTwvXnfvoVdWT3*z(}v)SU)mcG})1gXD(Sb$6o#(Yr~X
zG7a1$h^8&*phpZrPNe2RFFuG6q2*pcP=|(NR7o@f%|ZD&Bc+*yx;XXxBA$X|@7^H}
z#6q5O76t2z_~zL#qrlt3#KCtY1$r-aFmLbGOWmhrpFPX=MER@AH&Y!`dLko@FL<$b
zf1?(ZHOY=TvmU6AFtrd1)C(TvLQm>#{;t?N1YK^-#8FkZH_Ceisysa9XG6e7dfO-l
z-YQH5)_-NShas8o@)v>a&?hN$ysvjm>{!$w=zSM2eVA87(ISU55cBpvpU*uDUBdup
zZfd$(^e`V8>F=e7V&3j|?a!iop;Mx0!P_emr!_`99S4SnI&7@gL@nV%`mg`(7)ySv
zNv)$cln|2hrqSt1gZ#!Y9EHD&eWwU;0gtpgh2&qjaEN$qHC*r|u<sV+jzpg-Qua(t
zO_{G!Tn20o_?;a#nfT$ahmdIJN0@kZ*6AEGySpCR9vmhE0l&?8JyH4STIFph#j2;>
zncs1tBteK$O_dkyj|cXBG;*qO<_Tm}cWcek7XQTmlGdsa3?sY>2ut)$Y0agC#cu%~
zSu>^QN~8zevZs`rPsfs{z~CuYe-T36&6othT^95fNDsKbX6LFcZ{os|w{C<qAVPjg
zT*}l;T=7}F;|KaM>3K}&OULR)67m1A_a0zPrrX-^iwzxn#RgbH5fBtqdQk@<ARwS1
zg1|^qiXb4J1jjalf+hme6{CoN2ndLj;3#0|hAt&4B~l|bKuGefH>l&@XU{%+=KR;W
zzW@JyuW<}9A#a{%J!{>o-KzkzQ#C2z1Kh6}@uWCV?lSethIcXRPau=0Mv644S)xzb
z-+8&-|Bv_@J*0GP>QvrjH*5wSQwD}#vhJKprE0e2Z{+R1HQFz%sJ;%(0OLn&r%qvW
z^*%Od8IsCp7KRX;4q#jEUOSQn({!#s0rK2$vVJwrW1SK|Lb~Q5eb&`geHHpx53J5g
zdfU%e;urKeLd*$=y;?(gw>UcCnBU8zLj4aG>(n@vm&!zpbt%$gM{9V8EZg!D@b>v<
z7U+aia9n+=C{{+TL~S`#Hrh^!VrgiD{X;vIwA)cKy%JdEV^jH(x_;Zt)WmzAk_e>j
zlzlO;=atV2c=2sNx-xC*8KFO3Kt{;76+ZEpZb(0&q^5v9=<0x3bod&j?J}N7Dx|m{
z9Pv#er6sk{fiO+N>}+H`P*@P(?5f+N?MIdnm>7L*(QFkEJy@fg(~_)X_+CatuVP8p
zmyfXk>dxOtw@UIR!w2bMGt{G-ZMr{x)Llm}Crc!w4ED|vp?umzN#r)BP?Op~5Q=%P
z21EoRCnb6vhzb*j_gGQAMFKU%IECDwsSx-qByk^<YjViMXIhlh&THCK#v@t2(ca(%
z3oB<Mwn<TWyttR2pI?LzgXza;;yIKb3uyy|G}S$O>eur$EEJTCbTE2>UvMiN4Lv~0
zVLH>s*cZ0ga}}`nZGIEnR9~-Zr-<tXx`fJs?WPvzN|yJvs2_33?5vC-DHEGcVQV7y
z+{&m?PcN@kPrv=lB;1!zI@&g3j|XtC)JF#lg^#faMbf>=as#c8%6I8<jj64eLJg{^
zNTEgey^A;1%;P*{1{e%{g1*#8a67{xWnFfzpi$BLIJFQhJUIbYj@~8$XZZoi^(qQl
zVYuMCO6Rz!II&*T*E;K7&At|u79-FiDUE;od1t~Du1+g{<Bdb2tqjEtRrw<&289lm
zr4ibGpPDmlgdZKWcz?T3aV2BxF==N;%=-yDELXJmBB5hz?}PJ|gncO@h5O7nPbCD#
zxMmJTlQP)tk1HdeT88`1#q`=y97seWs3ij+LQ4FXxuRPOU=d3cLep=K-S@~E9&y4C
zC%fp@T&syWoKQmv7qDQ3Xwj6_H^IOAY%Yrw0e*Djhw<}~*ms~#e>o<4AZH_MDplg(
zd@MCzbkk49_J6@xCY3L}72y5z5g4a4eJ`qTKlQ$znLoDdvH|XMwr&Ea`_YvcHLI+#
z-tbM9vNJEImgZPc!(u&L+aNK{d2w|?=_3b`aUZLs7th&FkC!MOJz{sDHLt6T5b;R%
z3YQyi3$yylYxhG^!if-g`7;z=9J_Mi#U8ha9d8>N!cW48G)0&LbmxuRab%anH-t&I
zNW1KY+&Nz<H9h1_&c6$JVr^c<4OFImC=sv<s8<vK(F}n2p*dQ#lSFF5Qi%^f`Om$Y
zWi>Nl;2x-{JielRRMv1TxTNfaiIhZ9cY&{0z~rbalq}sq1v2d6cGi;b(B>LM`sUbK
zEIGNnc3)WRseH0t4pHLZJ7?lpw$f(2sE>{jk<>5JzME1MEi#@gdx>gGcU$;N<5#1%
zD<2AOu9IlHvX)~{V6b*|M6E-KFU54Cw8BCUUDe45UV5vi%2iz5|AUJE%iphzXl_WF
z!U`DNX-U(KlxD}ET5kdCVx`m?q|aJb;HDgbbC}9#c*+oYeAt#{c8tRxJ7{zAe2P1r
zlYVAoeo#&q+n+eV|3y*E^Yrw*EyC7QOCTSl=2s5HZAc1<j|*u{fIkLV{Y=2W&5&Rm
zu8_`mesUxD*EL{>S0u|`8{G7`F<c~EKdXj@ReD(NGg&59`P_&)dNgkr?nR)(3*}=w
zkT&t&FcJ3MQd~8pWo@UPD0O`Szq|H>%mOmZi(+d6*mtM5IbjMbD<c%?(Us1}sd(&0
zNMSjdo;=?`mRllWpJ;w;Lq828c=MUQ(W`mP^MpnWOORpX&))v9Jn!*gOB+J`?Xi7v
z?9aM`1kS;lnYf)P4#j82?n+EC775xu_F?4mz;l*gbp}wmpZE*#dCrzcf{!h<3if&4
zj7p6VL)Kya@R>8uNg<fHs!PX2lXa=?$F5|UCF<<|kiN%Xj31{%H5Umq3qMs{_;|)l
zU8L9QnD4gtPvdv2sw6%;0`>TXMpQg;Z1FL6JE@$eN<EX}F?1QKXBfn>)1P)hWC!A}
z=b?Nu&g1H_DeitG2=lMPEHP1Oa#^F8K**~jcPNfKkW^E<g<FzklLuQYX41WmQ;!jf
z2;kk2YU-Pt4!e;h4hjbD58rwdnmpUP_8MwpO?9U&3SZrO@;%!4vwRS;>&)9Ic|}Sq
zk(nsrJz`MWgkfXM;ujyB6QD`;5Mf7<gyTwZT#;T?Li~1VPR+d8Sjr!jGgt{N!#$9o
z12Y^2aZ;ey=kP3r&I792u|wGR`P-3sZULL8W$Kg3tptu^+<I?O{~VGo=H%Aiby(nu
z4m;{A@m>N;COFH*R8A`*FxV&)f_#g4__cRv$ny0JP<m~X3qjq;T18}w3p_X=m(mRc
zi&^|0$UOJM;p9m!S)W}f-|PBJpS`u+9){t4IRGR><k7_w63Hc9fh7G6vJ`)dB+i$A
zL+*^i;%*jxOn-2`{HcU5o}|zI`o3NC1|wwBZ-ir_hub%yO*JQUyXt4kY*|~tZ1wZ^
z_m-{Ri>Ta(aNad0E}e-B<f)(zR#gY99mdz*;04rylS`Drjt7qBO{g;xC2jsnjj0V~
z3_$yKh}tB2Zbn`ds)vWLV;7UhEwKGZ&0n5QEs?=bRg6D2E$q|GZF8@HC`T<!a0)oS
zLv`l!tI>+?k#sG?`mR`ImDR1C3Y<WC+Z4!`Rp8`@tLjlX0X&-2t?yu*Ud%@Y%l=sZ
z##hVLV|C4E`CgAe4*D$Jf>j4>_?)Z4xaZ0)vmCDnp8=TyRAWMEo-WjpZKU$}Y!l)q
z!tgVlWFJN$v+u2z@LESxUB34zg!r4D@uaxNOXho~Ucs%yuyl>E)cB3l#bc9_*iC&6
zE#@5~>rO)6TC!82sG?E{(|*2SXK1IlGM&>_dAq>gBA+x$>?ZHcI(UXhds`>;h^<Gk
zwI8(KCt1VgeS_?0x+kIwxW8;JOaAhb#CAn`<>Bf)E7>ga?AHnOdx!6o#~ln)n--rR
zgnj$~IYRKvA_0ffuOxm<+OY@Xy9{egx+xb*D0HTd-{UJ%a{EcsZHnsgZ+G;0r;=}b
z{#<joS!F|{A47nK=d#uP*!@<tmfBi3{COLgMw0@iD;l8LU&%dV=ZfsS(K(p|)ddfB
zqujUptY^7ehMo?a%OGGXB%8&mn<W5NJ(}+^)JlkmGV4%e`n(%KNok*5JuL(EpR!Sh
z-&S6MJoRDTX29K(C$iJiSRa<-Cs%D<DAwzRjVK7!hPrY^hvsxJ)<^T2@c}f|=ogE-
z&V1ggI_i+dbQ1MD5<bAZv`qHce>o(!l_-zJt}GzAms-NBpn?Z`vRVQP7Zyhi9_d#4
zn0@bJ<*{2NgHK%{RHp9qT?y_h3(+$~lDP=GLZKD6-{_Sme}X-E$4^MZe>p!_5G84o
z?z0>jofikGd6enBlzz{0!r5D6yQ|t3x2zuDeJnoW`3WlRQ`E0zs|jkGBp_5bd#!#e
z-m8QyRALMEL;t7$<HK821PU!`W=;l3x<wp8OZ<3Oan$E|8iI#=76mzOf2=a!?J8-C
zU@6|4M;WOj;I{Ry#HvFHiTz{CV?8Q9=AX9QQR-m1*f&Bx*6)VL*&Ou`bCA{4y1Fb(
zaH1eV4azwDT+vASIyjL8>8fsbQ)#QB01R3NeGvwv)i*c$M0!ZmsML{Oh;3VorkL{G
zeMck+>02t8B|&`K>l=t9X&t-@afCPaSt*_9fl`)2sjD&I4tfj`PGx{e%sog)U(oID
zVxLo8SxrQ%okcIS)F;ur5Aqk^_5hZB4_;*Ac7>f7_*x!k=LJa|qis+G(&}HKE~vi7
zHUYUK2JFI?BE7x@=m7~O6%ww9wJ@E>4`TD7EYx$w;i$hA7(`aQ@xvONnW5ld+RRdc
zr91ZZi}4Nh^_4GMwoG~<Pe}6f{clB~EN(}4|C8v*9eRTF>=Wf%_UWe^_z(Dj{=Or8
zw@zgJLB2;7N9y<Q1m-S+mq{Q_JA5&lEOF*O>A<m8K|;K-htijwMUD_p{DefN$WD-<
zH=YzC(!QhDO60r_<Rik4jdNp&Az#DHNVQ4Yd;}|@`%YqqP~BNbchwx8(`IP?OmQc%
zVAov8rl5A-Hr5I8IrLYIJR)~;=lmd348wn{Qj?PrOi+uFBBx%lsXSkXxuN5@OmN=b
zmcNrFMXVpDJV2f*_%c*qc@<qLy3kULC3u9wD4{Ke(vfk{_fna+5_x<y(4lK)A+o6Y
z2699_dkqqr5TFam{KJFA&U?4?*218G3$@S!zEd`;@+QfBjpkflNcB%L6G-tM$;YzQ
zTvpA~_MH^3ecjjW6FTu_S?3ns<v%u+)xKKk&C%X}-r<#_z)*axQc7aYPm2$tQK231
zpei*2x?3uRl#k|x`nY`BB4@VG6S2-QgOcw@r2Dg#-PzfK%i{PDYAH~fD}8mZq1yrG
z@<}9r-QUUs-wr!+=koeF;&Eq<y74h(2l!vM{E;$<^{BkShY(Qz@n^$l2dP>xF`Tg!
zO^W^QPd1}Ifcc8$FnA*D66%p1fs$sJ;|7HZ-B9(VhQ|3TZK0^BK#SU{U&l#&-0iCO
zv_{uaUoO8*NC^?O`*WPd8%%I1%H>@3Nh7IsraIGJyc8QmaH`NGCz{SeiZM`*4Ew0}
zT2k!XX6uLNzCW|{SmS>zKX$8Dlp`v@cmmRHe&h)Luizo@{ZO*u!&#*}0)CnBKSIF1
z5c9m5Oa*^OaPwx16}$k3w>NW}9Llcu=#;%75H79fJqr&`+qXUcsnX@^h9RP7zxf!*
z;f?cyl<pF~wpbTSeaD4h2CXYc96(_f78WXR04m`4x1cIt`;(je$+LA#&>F&E58b=B
zwdd!#K``&>J0~8w(~D;4avKzLlOYxMD^FA*S?Gub6gwLPIAw2+Nt0O@BFT$SqneNa
zE2{@$TznV?`#e&IdPv21=ob1p&%z63L-@Wd2^La=4V*}Qu2vb1AXBrv0Dr+rosH-s
zFa*W-&({;pORSDEJ1V|r-4$S#IAaVCqBlTqrbp|tV^6kQZc<tpbS>sv_5dNG(fmiG
zuhwbTVrX{ya#^oSg#2mLmlFL_U>0;ht7eBg7~>u3ygVYTY>PkH|MclcyLlDlU~#$%
zX?4KfaS&kQt~&T#eXc1UXN@7vHGei}4b3uyMkV@_+4ooGXrHhGVdZ7<Z$Wed0}eq9
z4?~8+R^s&`ZJ`2Y<{I?2pv*d`W_i%3yqykHjNHJyOy5H@ynnd{ln*IhGIAF~^fdUi
z#rQ!=ilnlF!t*Vj%Mk6Dp`js79>9?OXDe~mVt!A~7bj{3vRc0X8|{PWjX4^EN3dTr
zGQ4&BvW7hId?d$U=X7**#9ar%hni2ly=D+i>Myk7#p(f8XEAZi+S(&Fy!a)g#3}yd
zr1W&g74R;sA3~EaFNgh*!{IEREpr7EEs}P=`sD2Uqg-x1D}wYjuat0GWJNaigoC~O
zWyp}q2|~7(^Sy8MUmFY-BI5@^a>;)syh>Vmj5l#)L)U(+Xm5P&{)alIrlwwQZf^T|
z!TI*~_K-U$4V>-&_OZTK%Jay6x_p4h#2Hvw(UyKYo8-S$-iX*MLE4J_?+bNSB7Ij;
zY5f}t5|Mjzk#Lx*dhg=l@F!2647`8;{#Tyo_aZ-EV7EGObiJuegdqjHc)|$}YAaEO
z=T@&r?t;VbKt&yHOZ|gub`Ov)en%67s>82W7BpJlt^ObQbiDHjXfbA*#gdurCV_5&
zh-VOQob?Iqb2O>b02F0;s^GiobEMJ}&3B{tuoLP|wEqkqE=Q#mLirnogw$p;6TAY3
z@48iPqRh)|p34c;SAm2>M|m8P_BZ)OqDcM+_@gAyjP1*pFX!{DIvPqlg#(hvTTz-2
zH2Z}oobcdY6iN`2`_)%K?mRd9gW~)QM4!IvhD6ux>t&A)Izww+{w_M>oyWhGeEDLW
zpnOEJ3`!1fibq!H!w9bLt4HRc#zq*G*1ulF1v6^4EZ336{%eoxUxy_+OOIKfZM|21
z{<YIVe!gAj{s}Pi{n7A`uG`lO#ZbHg3pCY5QD{OwU;mHuTxhmu_B{Tr-0l~HRP*0L
zS5#Uz{e?na>_)?G7*2-V*)Me4mN0~d>O9Y%x8vvM7r2Dl;JWtq!r!J~l)}&U>(>5s
z`7%#9A!c&fz7!~*+Hwwt_y5*(W!W6-kN@P3{wekZpmH7>TF{bA;<+b$?>C?oD$p8N
z$egWrgjC|67npxJkAJIP@{2)A`1hfFzPmUSO3D9q(2wl8s3K0m{1tS6eUs^hJeR8y
zO)Jxv3eGEjT{Znx>jh!&L!rY_BLEP*>&|7kt1lN1==6Qb`G|D_o1ol(mhoZX2ioyB
zH}?k8m^F2(-pX^9Tl{=`i?L|xZ`*dh!~AMKvTetydDm91jDAyURE_!l!Tv4V)|D@R
zj<I_1`RZdMiR9R$B~ylxnS%MBgfm}jtZ+SJtgUfG*64$R1a`-v<-4POd+ZNwi1adE
zxxGX|oaD2yt@<u2M%`X|{rloqLb=`v4)(P4$7$_x4iA|gms65nu<F_kl04MB)**Mi
z*B~0>(u(xP7wtnC&ERRR%|0<mA{kfbd1OG{t4`%WBB|l+Y0qu&Vr|VrORHO+?<yN@
zAY$hW==CV?U?>B=d@TNx7}m~A79C6$6?muzMm(MZu+<0PwV@xXAU7{_En7ARNq#dL
zuqF9TW~u<tOp{($`VodA?DE#4Kb%q#a@Bd)*fj~X>Xo<5!NO%X@1M}Q|Mt;1i~TCY
z1JQ7V;c6>UE&~22vz|h$Q1H=>It#7#$w^zxUzQ^o9jm3XNqJ#VocZL}-7J1$h#U<P
z{YHW|{}13_JO_}f#JU5|+s@{ukAU0!JiWnDaOu{zC0kQV0SN6_wAEtaRtw=J3v|(i
zkSJhYJ+M$-DXw~|86Z{%z|@BPu2}Paul%x8{Cq)2>m9R+1bsz%bqV_B-B}?76RrDi
zufdDAtR43QDvvXA=3$_WdO)EcQAI)j5P<&VMDbWmxE%#5`sC~<Hbe>YSqGzTG2pAC
z>0-UrZVCm0Kt)K}Rx^5mw5=l3asV>ytK?us%_F)^&>w<_eF}D*F(-C9A;>PRvshVd
z(<%NI$s336H!_&Z%RrYTpV=3qbsVsjkgYu+b44$}tT&jrmdfXDYZaQGoOTx)b{^P*
zc%6mjh%Z<EA-*Gg(cjot)_rHt76U+dbbR3l_x$#vau3iWFzNODo=a|Wm!57!w-Iy6
zZR;hsz+LO5(1l3;+#&qa*0xWcO7w3s!ytH^O=}$#3W4@gBg2#ieI2FlSmiGQA+2*l
z@WvsrQy^eR>-)#oVw0+zRlf-;ic6o*MUaQViEh9YF9H#w$~!F%wonMUqX$eGFvlG!
zj3+6-`taEczS7NLl^T9RuvCTLtSqOB^`140Fz4+jOt}XhF}+Y2a9g}NPlSRw86s$#
zzGvf&6J(nDX5efmJ#29n`UUu+eQrkCb+EddA-KM`gJmbz`TMc=*7=DP-`80g7m?<?
z%iqA%z{dF+^G|=%9mh5P@cpHLtNZS$t(PhWy;u@8;1{uA4#e9PzgMmJ-4WeJ!iwK7
zulRlZrS;P_=qmp1j~Bn>cE7GS_M6NU>Q?7vpUw4`@`8qX%`_9`9lf_bhAd_UB47N>
z`8^xseggqBGE8qW#l-E)8viD!JQ%eMjEzL_?^q+Vh)R@8*goMFbo)QN1HHpas1ksq
zxa8`qR`^O95+f?Y+7g*k{a<7`o|JEAc9yrF>%+=o2swKd!BIy%x+1_I+Ua$(ka~KL
zBOAwc8-TjyX!|VROXv`w{q_;Qj7S>@snq*0BCD=1Hob6hFj$nnvZfU57Jc2fVA)d`
zJ-PPoAHERZAO7UiAaNup`0@(r+}6Pq74Xi7oU%`79b71XaA68>PY)~<IIysZ|Kdv{
zbRm9Xu<xz-{f7Pz<@Q6l{ZMW{l-m#G_5->7KyF_^-4EpU1G)V`Za>`8A8zRnxAd!{
z|HCc);g<f0x&6N^=C-*P9SS>J3-BZU_9OoG|Ks=@X)Dq5IMw{2RRzJ(aP=?pe>sS0
zJ)|Pid!O(`#}7*gyqKP@N6UJM-T!pq*)t<FuRlVYctcJha#yy>esp0!l#4;b!DxpJ
zFStJ3ArrGf;K>~R*0p-o+v9Ht&TCE<!?06H2WXQduKctFMpY)ch>@I<{rPKFIOybk
z<Pj1U)uIkG{w~-uVL!H&qr1@zkbqwkt;zM*p?d4nSoDXC3Jm)^wnoP<ouuq9yy2_O
zE)WvymG|U2aw357#;<OfT0W+iB@&418I{o~XJUFgIw90MdxgCcMDA=-Pi2WQxVnL(
zDb09Wdu4SKy=Q^BpXY1*y&(RIwGiY!e%KCgaR-Mtq}H4sQ6G};OwgXbT9<%#&%O85
z;qRuCo~o<uVRM#`RVB>e8)Wp!-1pMC7Yk9|rZN~^yzAJS%S2M9RMaLu>(WZK?Lgaf
zcsJBXofk(Y%8ct6c#D@gMM`Yk^Iu(laUH|7Z&-FU*9OhZbor{?JZfv!7=i}JPuZ(t
zy(==?g3%v#3x&yxxH!27EJZX%aK=PQF4u5=t5OPUF<;I7%T{ymzCB$(%~qSg6j34l
zyJUdPQl#>R?hB{|Z}rZIgHCz;!`7+1_Ojqp|C~~(XAg%E4%H<<t-oO5Fuhq-jy>?U
zmyz%o^L2Zzw>F&OSJ=WiO1&z=mWj3~;ri;ucXOeaB!@&x;1s1{Q^rXoJ1kzG%EV4D
z#}vBQ`}Y$#ZumsVi}0?XvZ-INR(0$$zQVWZE?B%B&DhlGgD~HkznAO!xnz5gMDg~Z
zi6*W@Z;ke}6M^%7X~0kzhJBj^WU&U!Edr0vQxtO>2fe~d&=lXJ21hZGIQQzfFAFXp
zk`{XC{%T&<rEh#Xm{ez<BLqo~N@XFwXWSv3X-1M2O`<n0TBU>T60IWi2D{PGXvNG0
z5{)=kW=QxL)BT+UcvL4#=|ZaK&hD(5H&8fgT&?I}zY6j|Xzs-VFV(ACCnFUaPm*8y
zT`uFco!NK$Pj4H}l*;90n68&~R;kqF^3oaBLkd{e@G$|Az@(1j_XD+w^M8$GZfU|D
zpWKT;e_I-jA<WA-);aGBdkT~ARNhp|GS~$}t#gaxc}gJc$%1A38j<3*bA)Q$7is(g
z%)m3#)B5_sGv$@^p0~e$ZFxexZp^tmXHPOKNwl{~gq=uo^*=qZv#trRM_tW|{#y3%
z^ycxo*4zLcDp{57tHLf6=>@t<EkdX^OysW7nP%Q@T+1oLu<ec+uMcGT(^!^;TsDIn
zpD;s7n0ak%pgP{d;0!Tv-3>2aT7T{_we;=2AFa~&DdoCcwL<5A{omxvP29^?lb0EV
zUW3v5<mE5B3e7L(OpX%N6xn$@dgl^~_Q`QP2)O;!G7)z37)yp>Q&cwPiit~hv=N?p
z&p&?6AmF1aL5)39A>*q1&gc9<xT)pTauK#^I%oC+Cr@2>b#>(?0Y^_m0PhAeN*Nu6
zENF@o#_knH@5$%-IY{83_o_?i2tTD(y1^j;U#dPuF>u_q{s=O#3T{b3oo%T-eOE|j
zN(S2s<IZ}OG3-YqvP`e3$Wo@{dz%?OXq(xQZf(vr$!eJiVfd6O<LLaj_xv0)!!_~_
z{1urqYw2ydc4l6Wq$<*@JAESTxKSkem=>D;?%A6l4Rc!+#prEh@rn_-MLjmL=Snw(
z$v&Y7zI3k+%#o^~>F)e;x5IvgMW^f;K-g5B=ak%yzLM!M6Jj;<F$|i`1Ji>P-|zJ$
z)zI*jtRg)kp%}xil=rDgKV2lihqELO)RZjCG(E78>#*=8WZ%7c`861IExdH=@%(m<
z^Ov+><bj0~A46ntc=YFB=;0ftE!heQSb0dOu2>s_m<DMEG0ReNQ`w_s1q=}^-e!8V
zMpvo8^up7m|J)o#tL(fGsD%_;Os>{@_c(0ueIadIZ+9<?JF0cvgJ0w0><KXLK#C8d
zS$H`QG$F6^FU;U3Z^ZiSOsFU~i&H3A#3jDK6S&I48=)j|_ImnB9Ffo<zEH=xOoQy>
z?f4m75p24&D_i{<9;(B6SMjFT*Sgum>W9g6l!m%0%oo44x23lRTE%t@yMDE!K64e4
z3)B}7)KuA1XY|&w*!<8OSIFKJ(Bmz_t{kJEM>&+Y6N&<Bj1pJ<{nTqQ$1Kv*V|sLT
z9W7nw#h6mUR1A#h=&h;IcGRAJTlmv&<rePMY3K9XZFsp;$Lx1lY*MTT9n~<m@3d%Q
zQ|`b0^fPbwKwh@>K8s2#gVgwF=p1YAbrH4*Oyw907sT~6l(^$9sN8(5y#KyFz&J$k
zeb8<!S(fb=LQ<FP?N^>`<36I?#}^q;cjz_FpIq#Jdu4tTaqQs-Hy>qQ#yFmSF;gn!
zk6buCR@w5q8iFYLRE}s_6O?j<s-&~L_eqfs-8bxKvaa$HxO-x%wxhFPm&;~B{R=NW
zUg}SFwpYoU^KVNC>i2ZleM^rQ?O21Hc_G;wPx3}<L1wQz29SI%Ij}IYb#UTXJQTY!
zvrnYm-+C!#>!k*?ctT|BrNvt>^<R8xy#p;lxy<AoS@_{lQk|kO&xAl$-7hlM!>Ro&
z^EfZ#e<IDxh`NK`mdAUuy_X@i@AMv}m1t5u(K0<<Tgk<~A52kk0Z8OyUTxgJ#}+|-
z<9`)P6#ow6e0TP2E3R=Gp-or}Z1I1<0}`AMyc&Z42t+>G_*Ke-E?x(7k=@MqS-U@a
zrZqqWc60VB4oF0_>ym7~w6j9ljC}lq1xKYWx#hK=wdz)ELfouU8k9C3c)sDl@f#@M
zS#{v~iUZHp)-Q?RnLi)K^2;mEdMd?@ep?amzh}wOZ9ogJ+n{dBSkq7eD^Wp!Iu`rN
z0XbP1z`6zu^iimHePn4sjr&d_=VCP@><VXhs*&iKD^-z({;;-l*vFXqO&nz?7*zq~
zH<oAiY|O@y@d{`Gi(XcX%J0+S$mcC`6$Q)+UU?AwZC`phrTC?nCUNvP2LOLhJr;kf
z(rn)QQ_xSf->TNp%7U*KTBz~=F&RX9-i$&Cl}E<Kf$;z2AX_~DX|uEG9>p-9eIQ@T
z;2wNgoLOf@PvYxZp;@G_|AdxP{Sdbw;<gm|A#Ojk+yA14{UL7uH;7yPDHM5VQ%Xxq
zb35|A87}D@%ajI76_1|7E9GWMj%iC!e>P}?{e<_3ik1I3)1NsMm`{q7Xe(-kK4VS)
zXOH5d(r^Pd!ufdutx!`4AO_~X^C~6nbKGYphsk}hS<Sx4micKW{ih@BOfl@V&ayB(
z=NyCkyFk&?G4MCZXC|`2X|!Xt<$2&=IBXShaduWYWA89A;%bra<zhx9`e(DHDtt_&
zcWFVrAcpNzXvuv8Ohna#yx-ixI;g9w3&hPX1$0%$)B||ZI{sYH!lVDXQZ<*zB+rag
z(5nWj2bzcmYE#S!1t>_J+=`vD<lGqcnThky3q(dsD*yXh)`_mS%aC1Rv(>D2R1cd7
zJ9Tms{#7NXp^yW7US*YPHdF|74T$|W)vW>BH$g`zag0;QQLfJO9vr73yVNWHq~0~Z
zcq70aOTXR+4Y@_*X*6`1f8qHj-egugxz%|ed9XJrfpcO&Dl8MIR&Ph0q=k`>MRX5q
zwP7Y|Vm-(vB@?46bd3#nwwHwobGstRGu;Weo+t+b%0hf4DFHVv*}-D?&{!SJZ&j#k
zS;hQnZU(EAcOai{hKGlzmsWokPnfE7C@QPOouryGI~jIn6P8*0O}}|pvvVD)W2X|}
zxw+aQ9#3w+rr;5ug(c$1(Tgbj*sNgs#KPAjp3bNmok@>r2^(yp?Ul=gdmYkR;xc5)
z6Qtso`_U@?b?-nv-OVD@6#^$X5Hj0}#hVa>idNy%*&{OCkq<yMzYYJoR(Q~fhpUS(
zXv~6|mP(UQVRepaoCnL_ei5>&!7)43-posrN>jqkOwyU7t}!kAuof>8@S~fO8yxXH
zD`mo}V--h7imvkAytLx?`B-(!^d|L@YVRTsWiRNizrf_wQmIsjIFC&X-x`114tVm8
zgV_L5Jm<xql#;J9j<BPR{}xFuVC+XQul|il0rKWm<5xor?!9$ObMs_{{e*!Ku28B%
z7#P}F)J<)T1~Dyo_M_bp{wR7Y*~nsZYSwDPa&%u7uv}ktwm&&3ej^@u9`M$~o^t+G
zHtw+=2|gb#3#d)})<kAiM#)+EZui|b-Jw;`!&IFgWDDRvlfetY$p;?Yj>h0t&DLLk
z4ZbNH2WQyzMp>tr5KgrsbGfeH!TCW^fk&xMIeP__xAx|Gy+{N0Q!z$+KC|!l+?Fxj
zYDIicWmF{j^JN*S!29I!h3URl@$__m>*H%m7<6Awu5j)z^tN+ytXqOTlMip(`Od6l
zyERvk7i|H;1CMG!nF@DF)J-h$QR%204d9i<7fVAE!dm~{2azVU>Bd}K#c*fYRd9F|
zj>iX1v?$9IeEyw$b1bx_scN8;fIFMiI#r*fHhzi3Bv+g3_by$70GqzQX*}7hy){#c
zJFK6Zu}PRaS=LIx^_^q*qw6Eyt2H638)3%?eEs_?a&9DkwZ=PCb|@%ai8{Mdt|nk7
z@2*g8k+lrAcL5`i!J6)*Ek%4(Q(@x<<_b?17kU@=pBx||zjf~wYl!n;6u5Vn$9TWE
z@apPv?o`tz`7Av5#=OT3j>10m{RS$-uhYr52Bm)OZC}0It=dKNw>eu@R_76jT;%hK
z8kNz?rdSU<D97H#K<nj$<{Pakp^vI?Wnd#}MwvR(bmf*q@pSbuzfMpWsJ<Lu(8#AS
z`dbsHtHmLq{Wb}hH(uQru=M%B5}Mv5AaXL`lN{O3zDBj+UZOasdzr`2**wiq*t7Bw
zemV(1&82hO=xWReRKY`JP1vit(16-1(rrIM@EGE{RkK|QMU(Z$2JGtIx};p@ZZ%Zb
zTQ5UPy`O4;n1tBxYdly}?p+<%xNI?kRp&gHU!yskj-4JXD|EH7>BP#hUS745&oWS^
zJ4aVx*wqIU%%RI-d0-DyeIg@F^>dLgfj;#XK3;WZ1VW$&Gp{94owzHDEGHDEj=?TZ
zU;RGTKZ}$b(*j|kQ7i6rQwaL~pw)$Z4unqqT_hL1CkWtvSgG;N9hCEL*WAHpEfcHg
z@el?pEhi7fnzvHu^85C=V}%3SSR*R<0e?)gR0a9-Baonkp{~)8u8u-~+W2Ru>O5}0
z9H=<j-mD9x<%Jijx(S@sa;#S%^9gaBe7-ve|59*cnW$X;^ly>0iM*P?p`BqY=LkF3
zrmn7$)S>Mi(~rugRUG}*ryn)JHdFQ0$l8?ydz~cjwB*NmaEGOHS)&A;6Rcx$yEDJW
z=DPeiv^~YCaUoyO#mE0DE}}X!vjAxp1M}A9s@69S>&otjk2GA9?rRZ7B2Ro?Y(Ft%
zfS;~&^=q2|vbr#zDS6O-9?WwICp8!?DNAEom@!c_3VBh`v#*K*`VNi=qM-uF!CZOu
zVG#O_HXp80V-3XhZnqP%w&xBlQ!ARTWnA_clE<)zUbzO0P{hK>9aoA_eSn>o7ur3h
ztNbRzEmO*$VbkFbhNHZ)WB2A|<ZjSTEm3j~nd!Lk6ayuCM^7y45sAw_uf{C|+ur0m
zPjHvjpV~uI;}1OGrWg8$W%WPlzsxoBw~s#KCnF+aoqV^1YI}Oy_TQ|x>fB12SlBJQ
z)#c1(qrMEf&wiHQYmYObv7LiHUB9Jk36Jsp?9_Ug@A3(mg+D*@(css~)HPbtyiWWG
z^+262HE(cyCWEP<OjIT24NyAde92bndK0O%!F*zGSay>)<QMK;l^*Qx_iu!T-k17i
zb2gJD!pr^5H+LwR#*|*}j9L^F2AuYK>ohZL?AeLp((x(ex%|iU!XJFC1z2WL+vk&J
z!fMTXmu3PD8`1ZDSTO%3e&ULJ!3S^J=PU97Q+lp`bwo|W+wfl#&=5_+RG7KI$t2ti
zFI?=D{cgGFRH+o(IS;1hbQCQ4Ft@_i+WLJJumJb-FIXTn;3<W)pMn7f%ddxq#uvhb
zzkVgb#l<>ee27a9upVC<tx@M@=$G+rW)7V`H?RZE3^dg=zK5w~%3@gV^;h_g$*cxV
zR#VMdjR!_kx7J-1ewX9YYD#y^-3T1y33qH=LK07ajG?!KZ_Tu;cpWf}zQQ3m(MR8K
zO}=?}q}`N%E)D0Ok`W|*bd9D9%<|Dl>O?;616C~l;KWDgm6ipC>jOwINm9j@Itaza
z#?HKK*$AfM=a#OnE_DorB9;)~9U?l{BTbQZJk!2sJtHQK<JbtS4^J9~VO?Ry&WX>f
zSHh}<h26Npb|WyJiK~66y>~Dkg&rE{ufBip-aQIN_>OOZRE25H+kr9%A)kSaGL8sV
zI<JfHHc?znE~_9tJ-xd$2R{{w&wD6UAyBPinS7Mv{N%Z53WY*RU#oe}ZxrQEuZQ1d
zAeB=inO-mg?>r&he>9?7rWNhQ3hL13M^f~%PZZvhs;HVfcdqFgO+zmwq4$%=)|8~d
z4#%9E%(gS*bX87)0y^d6<74ms-Wc}v<f!qySJ-9wk7#vuqazqC%mSgnvA0V#&vlK{
z=_+(Z1M@cy$9HaA6qO@0JUrb*=GeeMEGwOnLb>2ZnqPk!ExhZw`PoG#(T>@EdIVD@
zj4mB27_Htk=&6FVTf+pZ!RWnCXC;GA`M^L&1_Pdzj1oicvY*HVpK6U)^{?*r+){jw
zQN6BA-!XL@e`k5%r-KH%?vA4wur&2E9w4MXA)G5GmRhVSS=rA`fOFnEI2ado(|Xay
zYq2!vW3U@?KG$FIt%EP_M6^)ax6kXBBpbd@>wwQIhG=0F(QW=!@CHpW*tdO+ra?BR
zPoMtOiY;nR{3If`{P5w!o6luDH%Vci9IXl0v13#!*_o-iMl5v*JI>{f)fpJ1naIiK
zm_be|<@Mq6%o&c2;Nquldi3i(T0sjuwrjAo5e?R$-pFSdC&=_xS`mcNaz?xE>~{7Q
zy7yc-cy-y}8ICV}W~z8F{<I<EX&}<UEMks}`M#|56N~et+NNOU=#GB6ABHPzfQ(~8
zV5$w@x+hW<h1g|)C0TI)8oBF*g~{@i(|xpbzT<NiA*Zvh3i*Aa0x6oq(q+qd$IZ^H
z<L$@KDLJk*wb9M*&&%OBGcg@w55*kzr8YR`X0Pw;u2N6%q+&OnxIUGgDOK|hD7SBD
zg+yIGGovgx><WVNV%ep+Qnkg!#W64>s%iO>B`VFm$B`E$=up`8Fv-Us_OR)^(z>AV
zYSH3ad-ELUDD9EsaHh2TDkg5*DMoZ%diW|3W~$7@V_1Z)gTm9b#-(Lt%kz)J?(v@(
zH%)0!1I`q=(a-(F8OORBtyFG8S0Idx#o68)mU9gg>vdH-vPLuQU<R_dcG5L_Fy_+S
zR8KgrJ%@t!eTGpcsi)wq5$()`&h5sZI`#2^F{7-kOg{JY^feQ9niZK7g0;6i-hmF2
z{aC*oq&}_%R#82{yH=`nrTlqlyu7Nkav>a-_Jioz%u5tM>vC>@7BRD<Vr1>k=ws&N
z4L;?)zjbM(nY7-SX`SZQBp9mrPP8~MEVU~uX`es484&8s*R=b?{vgQR{Nr;LzDqrk
z)^Q}FyXV4%3zET!;>v={i(kKf9amVWHDkpoQT84haIvTrc9a&caHln<nF}YRku#;#
zkE|*ArJo~@<IsV4<Az+nfadTAITVb+F)}82FI}=^rX*3l@+b7G>?b(v6OM_E4RCMG
zaltn|+*<8?{;5%NK#?AMZpEV$X_myU1`zgiV6PuN1fQ=VzaC~Rp@X?vgA>=I<C|FM
zUFr0I)R6l_5}R`zYT`YV1Q%)$fTeRS4ZeK*hO3ga$x`_u#>1GiO4;JU9x;o96BSO~
zr`ZG9O+u_MNN$rD78LQ@C_9j|w$@gELqbSM;&aCj3D$Ng4UX!dimdjccBmRS#EGr-
zT<erh=o?RP(&846NK}lx!$gMOxN(vvP@2Hl(u0Y-Z8sGG*6S%13+HDoI%Vc<uLkn)
zs6;q3M`2<+yj#n{4PY!@+0-iggo_n{(qjXgDg+D;8REg)8@UOfu|ZSYzCG>VR&^s>
zr(5`hAoS*@pV8uK8qPR}yb3Gq*_)m*Z$&Bsr?sqoX5p?)|LVModRMBzjx?Hh4}b!z
z$@8oPV`obsN1<9<w>&d8@80J~X>jGQLKpXhNL1=k34ek0%-~#D$UtnVy^u`sM^o>i
zwmeezdiBp2md8eRm7Q3_&awMsZ3p}AN;n9u(nau9Yj?tNVCcz4Pcxm62rv-u2JT%I
zdhrbSSj<qXqzosWDAV&Mq*g|Mq;W%;(DQ-J3L(W44w{Amk4iq!&YT-4#Ag2T>Pn5l
z;lr`n@L&^t^?+k0%#1rh;HGQ7m78GDxFI2f-t_9YLL!m)Dy<yi4S0c888>GTssdl&
zF=fL6zZe%f|E{_7yJx)@U_?=t{Mk02lb|(^_hOK{RX`aldyX?jS8wc9K%(~!y>mUE
z;lsY3K)2wMDJxkB#<#R`E=YmeIBvA3Dr|GQ_XCjMCnjZ(G0%qGH<f-yhQH$3EG8}1
z);L5Bj|Y7l6S8pir=Ub}NA!y-3G!>?ZrHpTmR&Qg?i`Zpxe7ttV#R|$Hu}~po;s6Y
zGBMCHnvN|EGjd6d)8$mQrTU0$*tl`}6v6yNc(~n^DGV5$i~>gfn*IW3;HQqls-2AJ
z?l)hfE(;dm)v5!Km9SoW3oK*j&%o4@)MGf9HQJk!ky>T7`N2E4PRhmkk`<W^#5po~
zQ#~;?mWT1}cg2HqMW-62O30(Af6^hS>{T!oVSZMTj*elV9eoMoVkE)(ISc_`aRu;=
zxtT>UBkdHM9}_dB@MFArsfr&V&}H3iMlJOyzV*1FZz(FT>(@X3=yS4f&L@3z-ph8l
z^JZe#uV#-`RQBxI!&q7wk~o=_DJ2tge})#X&<ta&U+6q=pVb1Jcb>b6NsD;(2)L2j
za&5{&KVX>92LDhzm=O=cp9dkWg8rsz03v=faI3z*JTS^rreJqIR*|FlON8s_tFy;o
zikgiI#g@QH4$$r|&^`&E=^cP3dc;;w>kY?J8k`6PF92Pfr^k7BoKEWG`?LVYL^?I4
znt7NTu6YI8JOB^R9;X0R(x7*s;hI+18V~)XxqPQT6nfmwe%I_R?OlDWdFJ(g7`Cj5
zm6i_v_;|__G`_PyCW~kh-c_0?j@v=m1PuFi^1AI$vRqnoSU~<MkC}R>@|gL6;~Qq5
zYLeoDZtojv;#{Tw%#s_f(S&=rlPn%gj=n!BBS38eI5A`}(j`^W05rS<w~~cL#l%7|
z5O5id-UM?Wif-G|4}3J+Vaa;e)M2%z#Dg%?VZ@RM{>T8X&gvGxj<lVsvUgWUS8s?@
zp0fu!0^e!D33YWPTwN}tqQJ1*gM@~w`cJeG&7|Lfz}GAWd0JlV>>Wo=pSlV>BY{{^
zfy^ClhHKOf*OWM{TUDpCd(R%v)BTNUP0Wc2oVQs7IbqsJebi)42^^Jxx1eMqnhJXM
zW(`UrsfeA~ke>S^la%%*p3R+{oGip1GUnoz7O#3ka!)OkqtuA_ns>4vXL-=8JI4E)
z5`1RQC|GC;$`d*bW_p>FE&etn@6BBc#~*+~49=9YI<clCi&zi1#SirNDYJ-%MDZM$
zdaYV?{hmxdoYU0=6?F<|xtR|D3P6*!)ia%@0YY1W))s~%-L*(8){X9DIU0RcdDHG%
zX*)9$HC9VaysUykvm;}5P-i0y$2wDk=Zp%21#Ij+9u%6lgR&Oe(d@H46=57U5}lgT
z=SLAS;+Ym>L=N?_OPXs&k>Ue`4p+)e`1|Q&sJvPNxc+${%ZsiEPj7DWQx-&0_cb_<
zU&)k`2QXrhar68WK9ek%E0LCRR6<_Fw+S?v)w3)HeB1LjAlCZshJ?ky<DU>u@&V%N
z7&yTJk`z_!kW|IUM4EXnuC#55C-E;J)p^J>+PJ1+Y(6M@0KMV*{BvuUg6PVCRT2OZ
z&?bn8A?2A;XY&VhOu^rnz^J>~`icf`E5tVLwG>#ZIb4Cl<+co)GS?^OxkF*OpjtEP
z6_AL>fL22ieLgfdH($nblIw;M&x-&Bs_e`x0kS4zG03caNY^GLiU-5;)$T;4IZ@mw
zr6EE^9{2g8aKHqNKAeb$9fsz5yW!Wb3}9c1SF3u{@SiWSgnSa2A2KK>(v0eZ9a7Wc
z&J9sk(^YH+UtdXjvJ-zdU7RdK@kgyi7TlpLD3Iv^BRJ*34?PaiLy@JLhHLE1L;-h+
z5RssX)|f8RMgHYO49_h{dm_;FTBVu9pjI%FS_6tO46IRGY*#%+HVi-YwrDwiBu0$G
z?(EzU^W<0x*VSwXdHUz>R;$27W-Hb!uGTnj{Az;N=gag2rHOZ|{mSVgQ|h1;pX1=9
z;Gf^VgG3|(_SH080)Xug@S6)CyJwuygh*Ye8opZv==Yv3YkE3<uY!Wl)4B}AGX&<5
z2H*e|row}^=gn#$2{JMjBd@Uffwrk5Eiq)z@>a%V-!)<7nUCCoQc<VuF!fq4zJvR~
z7(W$r)lV&+BtsDfb=f%>a$}Ko@`42mI#ggVHVzzH)XlElxzFX@STRf<YC-1)K7kKS
z^%Q#W@wmSOAwhQ}r_>)&cnk+B1P)CKM%BtUkv^fG&*)N5*&DyuD%Di2lM<=Myq=L_
ztS<b*9=<HYMfWL)`m<Mh*eBW}rKSGsUP3ZN!!*A;!aN>IZmO_;LJG~SNZ|q|Y5C6a
zk9fk&_*`~E0sU5y`q`OD|J)A0{>!?9XaeQMH8@4J4S-KL-h?b6!SsL+bD0zPyK#Zs
zh9>=)eEx*g09i^jMC7bl6NA8XhakuI22d{VOEF22AAsVzo=mBo8S`tO!W5(PaOUpA
zAL`|2n}mdL`^_TK)5ABgTGQhvi%yuCjl^-~+J_smb3-Fbwqq^bUd|UNC~9ZVt$YAj
z)Huz=m6q#Di{`Bt%*bmrNoi0Xf&r;10B(5x|34a$PwN0-3W(%A7Y|NZM+LaPiO|*0
z@2@ryeo=L~jFvP1AxQ(=lm_E!Oxauolk&`heUt|_HZ~Dk=brw=V9;W_2$@neDyU2o
zB^2LogPAYjDQLU`Sa=HqsIkMNGlA2a5Pz#njBuMf`Tm-)`IBZs*J|1TxBlm-i`aPl
zXqCL{@#+jfI&pGJ-t$iHlArS|3_bZBnB~GMK5lq0GoRK08f6g<n@EHgj=BH`A@e>~
z-^aONGLnCpNl{k7Tao*#OD4S!NhUx&Vn>H=gojj+{lC%Z;4#Lij@2HFgb{42P7s-}
z=J~9f_Bl_^fcszu51yYZaw*RYI7bNZe@4F(Fw82w*P(hNS4M4JCm)9LpEzkxa_jht
z9kbAA3nmPWcAU)u>Se8UMr42wk`=P7<;RL%SnS@*DqS6(uFAMhlByUDEAc^4B)*e!
z0fIBNE(`eit8O4J?}1wFu4K4fo-|fyQ@WJ+xu?fzO7-&Dm9@s`J3xWAfyN3tn=Ohh
zb=;u7?&6K4*wY1r>Ts_RCUplL^1Yyi@Pxx_@Ll$k6KS|-AT~<0eVT?Uw=hnA<u>m7
zIizy(jnUT`>z8lA0*KlRvN!<{H?5Z!RDz&_3j{hkhw>RkFL-Bg6;g|9641<Wsfkym
zdx8Ug_U#`I3c89jj)d42tN3i^3wzDHKfbP4uu?=<5o;FFB%0^x>^$&0?~Z_cPF^ez
zmLb;@b#fs#^p2;00r}MKQdBg<VAD*@-?U9$#we!XNAe@LKa7e$XXN4-E3-Ld7$Lm+
zm%!mUx$Xier&)X;8D8Eoe-FzRZpU=fqs<pdeJx`8%P?5OHF$EH4d3A%f7M&aZ3Zu%
zE+#5k&2#5RT}Iz*4{M9+y2t5oXOC8|4{VgZ5}A~s;5(fxR=-a)>hQ>|?#+wuw)`jZ
zgMNSKiin-7=?Qp_O<d^QyRFV01yi>Y#mlD}$uP>QlsIEZa8`2~Aan;yVl6XqV`ah9
zzHbu%UHJaGSiv5GY-n)eljmln-5w3U8Lr6&sX`q-1UU(s2S5!mDdC7eXK}UPNM6P6
zQf92|@Ux`!xBy?90xPS~-SE1*t5L~%k`1UaU2LiWGj-v7XH#s0m2z?%o5@^TD8qen
zuxafMe^9#j;Wam&W#QpnpD<{Q1Wf&_q4s=N=?eE2`P6yGcYH5%FKTnZOFTrtE8R0v
z>(2B@m~0`k+z%+9>eDmZpv25@YMQuBi|Z<a7+h}leXF3$0M43@rZluN!R>g+vpBbI
z-O2{x3<XJK^??Y@y0Rw$T;>2!jI9HA_J!sjsQ7Y@APk@cZ#i9eB%8=8AC>0Dm*Iy~
z_jXfsuhF}TV%@rZ4Q5zLV@;MfyC482o&PadvMB4e8sED6|50?ya1HRop1u<sD@7og
zr0jG&+lIzW{r&u;^+y_reopkl5B@@|aA75XXY1X8E=nE)Ev7ieefju|hx_hy!S+Nh
zS-Lb3w>ywe>R6rPeUCqhZ=FSnZ)GU94~$y!o7F6tFPhqo*OY{Sr&(5{Qw@odgn-t)
zHNFD^IJfp@56P0YYnOEl*VHHn4&>7&3&Xhe5lQM(2?hg{x{AyK2Y2_85(w*fOYaj6
zdLZ#{4uEL5&0riBCrxSH$S{qIQ}&*C0x_T0Un0qKYW2|RYIX81?>xbkbKk)(V6`)b
zl%ZT42WEvE-PMKJklw8vcKuk<=eW1=EMY+zAwT!b_2$MqruEf+8L5;rcogR`y@)7~
z+DmD<AuURnnE+d#ZgIZhn1R%=4WAU~i|`uUL{`!go>W7G8+1|EGrBr=Dr~K$VZW!8
zMp@hKu3(I-bAiK345H9z5;zA5vgz!JaaBu6N}^ALtvb4<ByaR|raDCrq>lk1-E=AM
z0FYRzYx6WVjd=mzcTmZ5s2UK~8KFTej}r>t4{uBeA*kev2Rp3MJZIFW>!<=*PQbMR
zUAMG2z3{C!F;3UQ??6(Tm5H~pxFS$F=ykIHVSV~`%a&dZ6WJK#U<`PBo4gnufgWZN
z5zTW>^L$Om?CG~~PR^hPNECz_jpVY50$Giy4m`VCRyH#SLR{z2aq#Wm%<Mmek#au`
z4i<`2N;C~E8c^Tm@2FKd`8j*eC^_s=|1XZ$f6<?)s@5ydY>SzG*v&+vk!N?Sso`7+
zZ|lica}mimk7@IK(C`O38*>LwYqgzkI+@feJpa8L_%47+^!s@l3EjI)srKhTmC45-
zuYP*gH(QBPOu!<~E*ze%V@y2&feG|_`V&N{zAa!Dcp?wMemR}YKEXZ>*610k4|uKx
z{16D^aGhqOwh)OkPcvaWXAR}24}tlTjQiqKIJaiObo;nv^X{6DO-xwc?)sQ0{-h5z
zaA1G_gFyB_9(8DSiry0^7cI=U5mIM^+rd@Om8ZPoUoO-}<0&B^dN0BdA-e6#+vh)p
zCP?a*uO3^bh1XD9i+_zOGnNX0pchrBllg*eh%7UWu(LxKkB1P@^^~lj-bir^$lo#E
z|C_0o(h6%1D){JiTy{~=zMHd*T|vjRXP-n&wtlajbE@|iKjV3ICMm*cCMl6#=%yN_
znWQJ1q~tPS*t#AM9id@~l!iV*Z(1O;0RmXo4^ZDEzeqgT#eZTzF^Okq0eyuKEFk)D
z3=MI`KsLvPpj%d2D$UDCC6dW_4+1#50lM%jdEqX0FlI;a@38etAzr3QsCV28AwnX!
zL8IF>F6OPg8wepW_JpYK<olNwA&|OIBNsT9TOw^UjgkijzwFna1nPs<cEw@KG3u}d
z5fw00*q@O{k2mJUzViB?<m>0E|5W^;JTvwB_3N={tnEb=>Pg#~y*Rfv)K&fgPH?y!
zJikiXV*}n=BD$h{wF_i{-5V}c9(h->1O{?yCV;z}3;mf@jd=_1<a`g}dB^2{ilQF#
z#5+?eUC+H2by<1g!!ah)Ro@*>yDJLJBu}~~&3*rTd;;S-=((ceLkw-EM=oLFSyK1f
zbf1z)Bif^R@1HYsatqT&%*o^CQ=hcdr(a<1$A$9WH4pf(*y>a3^K&+6kX5;}ZRHlM
zJB-X0z>jI+Z@1@rbFB%5@WM}b96XhR8j4qmY<6<h8yV8)&xoy7Wwt?ZtV_~rN=H;w
zR7r4|hN;3-7cuK>`wJ+2Snd1yC4^O%X*Bti%EjM;DNtP(_=VnldNd=hL4Yqn?=`^t
z#Y0vp0Lobyo(dWc4TJ)5iu!q_o8dZhg7g*SDevH8uB*k{1`_1OrcAD)xg{Cwh%Qzw
zysJrR^I{G8sX<H)c4RT+lRkj<E18eQ$x-M8i<S?^gQ};!R(apzJK4>t7pAG@q7r(m
zDM?MfqY*uol@3M;!dA`b!#BD{37Z%L;9pC|=>?(HND#gx%<k|%o*UkC$1ejYLa+b5
z7r`MA+hE{+m-ijma-4a0t7jIvVW60&HoF=-!j4-u=WfMOLctzNGOE=oQgufX_4)Z~
zFXO~1#>&9p_<!f6Ve0Ed@W%8)9uneHYw6d(|7LYyT8ROolnWRG%m(a;6E<G1MonL5
zqz=m|dW*vo)UhL(7)|s0^A#o<MR#mUU%jrmab{tveUHL>AG*p)<kN{lYzMKI7~W#Z
z(51?2smtiPpAMZ!&~Yz!4*)#SRb|#DkRMqB@3E8#QCwE+yA7!g55{=gPIsN!M#(1@
zY^8MHVA|+)3+whPuds}LWnk11)AbmmPPFx|8<w#h!pzageRtD#&tmQ=2i!apIgiaj
zqdy~eLzdc2lTLXP<jm_8RVnK74f*zmeDMUvp7MBY_m1l$+bDQfN3NvH$PHp4f~<Lr
z{dxb<U_ooeI6~L27*o+H?a+cvPHI*Im0`t>_Mvk_O{-V)1+DGm)EkJF5@Pph6&lM0
zYQ4%O9*r*Vi0&m4Osl9x+9P>~;wV0sR(`_aWAh?4V`>DHN4GzXj_M5R42PSt)>M$I
z6Wh#4&|-V<Yrr3>8^*hA_TO_=Y=Qms+lLda>sHDXXL#E5#U#p;9}ZEY8UOtD=9|5%
zugO>6=QQYz%kK+~>+l@#4x@5jKYZ~j<Tru4bvX6WZ>v!JAy#b8=Iw=X%Mg>BWyBe;
zPIeOAkn11aA$)qKr9l5foYon<IAw0Ky6R_}(M2n)=gX<jNr_TiW<6Ya57FJihm3mT
z308L;+E){1JUiLjlM{k4l)9L{nSPb&Dd#>hX=ug71_~YwYJi?nbjO2A<gU(<dTG)B
zkT1j-J8~5pPo12)g2^src|`ZA%7zsQo%X|zP!G=ibfQPgdaqm=pH#f!8tX-9G=2Ow
zCRfzgx04<w=Id%ah1Fvu#A@Ng_7*18lxicA;ZrM*S5=-vLL8ROKR(9*(eRJG(?u4Y
zGI=P&p~ZI*4%67`#)@IMkOx;6JpFh__(DVFX8xx>Jx7A(6Oi{l-oAtkMN2JzorhwS
z2JD2`lxs3qe%zLjLMV9rRx3*^x+6VGE1)V&Qor)Pnnu3Vd7Op#?8YJMYBvA$bY2AV
z(Wi+>a=qn}5ZP5jP{|)2wIysK7RXUvn!VeTuIV$aQtES}j#VbGaZB+kDLb6Sl7D#r
zDr7T5jbPikxtq8%bS)OQmBMvan%IQJ?WNq;+K@i=x*M`K{<o(E_HhEXQ6>%h<~*xA
z)l>Fh*LPb|ynEa{IbHd<Xr3g+d*|K3UOg8)Tl_pLx^J0`%Hg0;7W;6GjaF+=)tB9P
zrD!k@c|m;;r55Gm+NWzPRo}F6kc_+Sr(V0Kag)IAWg1C4yFyN$f<q(kvAK8;dfuCo
zNkuKLQC%KpGotU*<9b&U1S$;~H8Fj0FWzJzN@wnUI}mqz(9^9{<-Ka=)2eAjMasQH
z>dd{k_}&IR8}`tu(m9_<mvI?iMJgISlKlm$bz-X%wB0M_Sc+AB>OZCzIB39r`+6xB
z^kRNYc~L~yYrC1lNCo;at6=?JM0Fy4AjQZ@`o$=<d&fWZSFU5947|OMiSg~r4BQyh
zoqI>d>l(;LJIDrlK&3ss-^C!>ge+oP^C$mpptig*c~T***6Z)aNKJhEvTg+)@o4h<
z$Mrv4fv2a&MK;8gIqjjy69O=l7(#63Z|86rN<N`2yi0>{5L*;$t3cUAjLmo!Mt}pZ
zMTi%fnA_g<q@KHJ#u}oD=;xb6tR~zJ@wp+oA)3BE-C!incT6#=%QW-(;z_5NX~kbw
z4n|fO8yk3C7g0>Q7`VsZqbG7K$}i3Q_v1;TQ{5O4hHVse(L5zAL#370_}d3X-QRwB
zz&g-JZW5j~Wml)-56?w3vKW{tTGnm2V{P{{eFC2hv@^Glhh*W^bYpY~Ow1JJp-mWO
zsvl$J+&B6W!@e=i6`?E_^Y!gpbYaTx)to*dNmVRIbUfremOXktF)T{@Fq?rZ3f2tG
z+D%C!#@2fdXx?bT>x}$bj~BL=yVG?Z+Y#AiwQd8CW1#q4%G9Eu{2oK@>w1|!gKWjy
zwpCK+@#)DOO#l?qWJH;yB9+>v=ss?Bf%^W;t>a>=UJNb@^6vR8wz+!nh_D<x@P4{$
zQv7wrcl~3U*U$O&dCx@N7`7yIwO}pdy85FyfiXMp8DcZ#$LH0ZDz(y^HsZ#4-RJes
zKi#re55bIwOm5d3#HDJ+C7hNJ#s*%CzpuD=Ja9p%R|e}gMgWs1J8oxaFD+8PbWspj
zQmZt(HiS9&>3(?f;!*P3D@FFB{Wo@Vl(a0n)b~>UPkUGT*3_BB1GO_lsdcO!X%R}R
zV6cJ=J2Zjxu>uAX@~~79kWf+C3{j9(lF(YKmQBKD2oNcfG_68MM9UIlUBYIJLLvr`
zC`%%jB_t$-+;DT};yllMoBv?$$NSt5=RLpkd(S!Vd+s@3mug+xZ!4ZV!>rl&-j;wT
zcEk}ZKjXK-IN6eDJMGhys({9s+wba3Nkr&L&UavW*31P>MkwlrM7Paq%gHk~>hWhG
zd&TFRdd^&?iD$=4-FvAu+Pgl3jJrYd%lJ=;5Z)qJ@D4ISW--PFX$h^&B#)SQ66bRi
z?&uBJD6O<u(Q6#1OAT<~&D!_DOeUjFQ<v^)l&>G)`gTDNcaGg1X3r$?(Vu7FX{7Vv
zI(IhDN1zxL(b@Xl{>>V5db?Eq5-F@x_QL*p9+vJgaDaVmtZt4mABKRmh&b2|XEl+=
z`3V8$Pfe4wa{DekJ*&n?oKO!cU+l2O+3=}lg`{SVo|vsZ^-%MtKNy#WbvyNNBwl?n
zN}-zmQ_8XJ+^gT*Ai;)`gPMi7gPDw7H8lKWHE7<pqLDiKUN_4fNVE^+tQp>|-+$P-
z!3Dd~?WRrl(%EB>vgDR>_Pg-G`!`8YxFO3E-kid~Su-qO$%vGt_{5Ts;%Rb}V#jOo
zk+qkm#WzVlZWJK><P^7o+lVqw9vN)QsDfXYOp~Z~a5Q_`-*tpzdJw{obCh%Dcy1@n
z9|WTHn_k3|nyT;_VKheZA#P1;K}a)f?>XiJV;I~s3%lBNva-C4RB8TUz_$t-u!Bl=
z=bK%idS-zKrPXigAhf6a&`*xSysFi0h&z2tWxGem45+m}ZcZEZi}ZwJO7*ynG9vqH
z%U8F8$(8s<nctnaMwPU0RJknVg)&!|@%1y+_`fi9Zf8o47pzQeQT1EDN%s_NUEZ5q
z1qDs7<x?(w%T-#Bv>z=B2TWlE*}3tOTfFP^5L2`fT^Y{-h2#I=lfPp8%wNZeJ%Y0=
zUkSZrxGOVR_$L{^Vh&qgJE|c#q3FINuq!eXl2Lp7_h*V|yAwd}tYkf4j-+)XX6K-d
zTT1T;uRN&yh!Z}K9%cIq=-2^ji&-bC=DR`O^61GY=%fvF62FN}vApIsp!Vuu^}o(F
zaum_4xG~$giS3mrKktDRTcmhpYE5<BoEyRI{cf@9LVOQ3#395}PtE;I&-3Fhou$%=
z+}^6kb73E=_OK=&yrM=5lL~3TkhHM#%=(fYee2qV(rI<V{Xequzv1GhA7x2R9o&Ls
ziSzhmbk3~HB8f<wYetBg7t0DoEu^n^dJcFd^Z&EXiQ(vk-JRT^1{Udyo@2pA#HAva
zKJj?yaykFYHg<%$nRvp1Fn>TQ$FKUx#<4@Hx0q$-S-xs=#9@Z!Z7O@*E-Hu<z|P5`
zy1~|j)i*p>B+uT5J`V0*G9_1u4O27>c(ojpwI)==w|$NG+4t=Ym1?<rgiVI)9*i>E
zuA7Y^o&v^!?gpus*|lBDKVD$``4H*%!mfxMMTDpzZX|~0I*1#NrXN;LF66at`l@@X
zvxUD-Qod2ua|>{?vxN|L+~8ykFIzXcG7CG-6jD!;e|6@=lLV*QK>v_{4H*2EyphSe
z+B$+(gw!_XMKv^?C=(n(Z$mLAJ~j-wiW=3AOgTDt2eDPXq=h~T)de3R14D|!U-(a&
z>}VsSo%I!<hp)jG9Iu|~vdiBJmX>Nc3HM`G-1YL^BcS1U6c=YIzw69M18lm>SI<5q
zB`+!Zl$-8t`Lw02^F!3)PdtB%r)f#bJv5@emD2(zcchFi{3ykZ;C$>&z1-Cm`E{dj
z9U$qbe|wIGkh%`_f(_o<HGgtWJxk1vbNX9N81F;WVJ}hr!Fi{6WtOJbp7yw}PO848
zs0A}!vtFKn%DC=|yFa+Ak|@BufZ=^p{exR1$+5n|NPRDWtyD;AJDfSO=jkrRa6myY
z^PGo4os?yHO=fQtm))^CjhRXfN8OHPGeW6R0z`DOcbR?vl)1Du12F7a`9)Emf{ndz
zfl8FI<~7Lw1XsnXcAa;JP^P1|VlE#(pUk7H=#k`hc&|N>)En#xi?gW&n*Ev{PxsJc
zdbYTdV&sjW77OpwJN8#KO5Hq!yGDjA11e^OXd@XfvT=WoyG}PSNPtZPoFU4iQhFG<
z0KQ265x%hoRH)wN!>u@9{pKu5>{nPt|5RXK=A#v{V@FU4d|$-jgh4;A@89etfJIoq
z7rp1n!A$i|(MH}}^bNI%jdBpr5h2$c+7gkkcm|vJ)c3!1=mV{aLb<ga`4T(5vg@4C
zp60@XrCv+#4ta^Bby`fa9l3H2qlgpqUU}flNw{Q42KT{AHjnG``M%0iS6e6*Gn?_f
zfno8@_@#ri>1iX@V);V;L(J$l1n*?Ps>oFS;m|1(ieS&CP%kta&mWoMNai0E+r5SO
zkRS(*VuC+jkD@j9Tv`>%rRF}RTp5E#J@+d0tpFd*9BLal0=#_64XhdYjV*$jPS_>^
z_jQj?#80y4tcP(|z?a4VKCVeS8!W%L-CtSk33Vn&&lUnZbr6g82?BU~UliHn4bxG@
zYh15OGdzC`nO?jl(2+FD#I&eMOc|IO%Z@Eb{bC^dxn)8~+<xSy5``*@8>&F9vABUY
zUHtTlnQ`Kxs@sBskfk_ZARr=2q7aid+qEnmBOJX1zx<?+Gwc`0npxPs_=FI!HImzV
z;MsmXx3@lmHDj>t10zYlB8Tc}H90oNLG|<l)+DyGI;jGcl8g-_q6)jtEX;?b;$(fP
zrK8c`PmOWuq=puxRmA31M38HhNH}yiQuqrfWG)=kE@Q{dr-AeW^uR%2peVzu0<<-T
ztRRW~vpo1WzCzF_#l>O3m`h~RtrqvCu&lKL)6Je6bC7V~=^M>+cr6;=NYD;>4svUD
zfXA7Cjk^uyFog#p^@-0`;yioh2beg5)}gY8uTd3y(9p(Zm&JrhotYOYoTvHp?X;2+
zHR}{0%=spXc7B;EjxgYM8y7{U5hk{7hQl5d;x4MiPY1&)3&{Zf=gu0*#&K=Y^jId=
z;))%;PZNL-XN9mwB868mvZWK&AFE?<7XKtK#t4N+&!J9H0F|~`O6eWyl?{7_dFxrN
zl!(>ugXJn^3a;bq({h*0b+tsNc$5eXT=EzoOP(o(J{_!iLnjk_cX5_(d96X?qPHqi
z^0v~ugSwS5+*QFJRFUSL(&_F<(%;3(+XOd%9WUha+vu(=eW)7c%v{fpQYgZ;gHktL
z$FHU-JGzF^G~eyE+v0R2bHg#kzRFw4cG-psV41P1+5YSWQ<QMLivUfTi-*U$q;3RK
zsZmf|a?S_M>)j-wrP7*mDcs5}5^99Gp8Fy^?)dv+n|hqoYcif4F1(XEu{W9D?xRQl
zt*$mLt;36iLAJ$`WWH6AqcJ8*z9a6oD|0<Ks^PbjRpe*Ut4&9yu>KdRA3zh$53M(>
zvae${bR#6a5M}cvFX69|<PYD#;o4$8`43oB@+BucbIP(WmSA};zj7R7<2PAoKeXwd
z+#kF!>Cpe0{T{<b-@ObU1%_TvV4K&8#U=UsqIUEqc!e~OQ>34SX&sn_xy<C<`C;5q
zHC*i^(%>@J6r)X#1@*v~)OU>1I|2pq6%ajQS0i65@s^q7(pph!c>nEP7Ehn#ja8uh
z_FhygG<+$8kx&6f9?tcyep$V%5*Gy6n0WwVA@+DALTJI8&oR+;gH&wjjg2*(o``$0
z=_|y*4GtcHKY?cn>V949g7xPX{}gUQ3XHFUCvRp!7;GK5ZO39xMl&o72o>zqi@!@Q
zRtr)guYvv8++dE-z1-(KD}m0Hr85AV*Q0lx)6IsIkPVYSq`TrslDsCB-!6b?07>!q
zISgl2tw!NXSfoETBtUt(<XR>UjyWP(R3F~#r2paG0f0S(>}OV$o+l|Zq_lK}gs9#6
zW*dR_j?5GV-(AZAaucaUTIaRd<;%gRB=yM<61ax?T`7p1y5op@!HxOJ&a4~AUpbXq
zz$Ax4!rFfY$LC%*8=kLs8@i;m{<dMX&1NOfdaN3iau~a~fhxvCu0Z0l^L#ktC8uCm
z>BLf8Fd-hL4_4V~0xLj9GmbOEUrE`rA@V)<Lj%L=iU0Ov*?9N==YMFR=m7+yvjqc0
Sc4+R#C-|W71FicLzyCMi0Q{r?

diff --git a/articles/xcms_files/figure-html/unnamed-chunk-9-1.png b/articles/xcms_files/figure-html/unnamed-chunk-9-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..ba2262f305c9af1d405f0b441ebcc6137a53394a
GIT binary patch
literal 85614
zcmeEsbx_-1*C(_ETA)y*NYO&DVg-t8iv}$Y#Vr&m?q0!4u>{xR4#nMyH3fpZLm;>W
z_YM4>cjno7c6RsA-I+Ht@SWV8d%pLcd(P+B4SlaH`xN^HHX0h*Q=pub8X6jg2O8Q#
zlgF5-5rYRKhNuUose-H&>WTWCOE&tW9#0(Pbe+-AsDkhQ&?8q5EYQ$s(STBKH9S(b
z=X|38R=zZ6_8?8y0fm7}*lq<|uW(A6oZ+|j&%37Wao>nLF=_y9a;yrUqg+Xize&0a
zp;zeFd<*n_N#H)NKx_Nen4RsV3o8crxs&k|^D^nl>E~q6SdBfWbU{XI*ExAcg|GG^
z`{p-NQ{#8jHVa{i159-EMTO};NbR|?F`fm#i})w^%l+?v|1E+4mcV~X;5BAf6%H*>
zjd3c`Jhxt4&Z%bV;T_elz&-}gPv8YPCjihjbjO@$?J3YrDF=lLs1F#>3=oHyCny9r
z#7_qS=ON}Zoke#rfmQ=xf5B+U0Tl%ry0ouEGS>4$GXQyN{z_}$+2J3}J6{frRFKO6
zS^i3$J90E@X|T%LmG!I&lc(lpOQSY-GTZ_vR<n6Fz_4K}pL0(m(oQ8v18WhcFcQ4>
zEn4&J_mIC|;-ZsSf>6x0ZD|X8_hnMO->r4ngS!cff2n~~e{asn)-Gy5r7LC$ICrBT
z%blh%^qJO@uFJXlTr;QkUUcZ+jZ1{N<{7lPF1L)b+8af~CA9@7cKUB_ic%~;*a4Ai
z0O4GN$9Ht<r#j$MfFS?QyJ`70VR9N}s{Ix&Xd_tUSG%Vh<9XoPiEQ`#i8oDxP}Gb^
zcQf?5Nx@gJs<4l&=zHcqv`|jQL?AK|P^N;!D)8PY9a8{bfxr~WmwQxZzL2U4r_)pL
z6_j}EZmMy^t|D^zOkuBxZZY*oC9p*p+M{0f`&wtyuwhTdNoDhSGZUCC7thxcybW~H
z1III-F9B*xllGK=cq+L^N{37Z@e6lCa`3$vA4TV^>ZSrJ=i)i?SFkgl&xdM@8twtF
z^>2o%aM17GR=6vlN;D^k+FEgn8~Q||3ZPyPst2b0(Bp{D<Y@!`3KVO2;(AAl#-1(R
z0Z&yx_V(9^L!H%Yjp}$^FN5fyKAUJ~@4=T4%H)4etu2mmF}2hTC^lw$Hw|qIwUG!q
z*P<Zb0?tx50oT^(>dENriCHnho8i0P18N4px>tM$h{wQ(01I-+dH_CyXR&9KP6y>J
zqbswIDa86-s2|?72@B2v{1f00fp&F^W>`r*x00|j7{(owIyUq@coRR=ov$_?a-{~Q
zQ9|yf7=>RJ0%)MqTF86tm=F~8U|R76&QH#^a6(xRJ8$Z*0k?e6QzlQP%3F7$Vtf=M
zK56mmPjhTEY&_GNISD}S7dayQil-i6v+E<-ky1z20&pT=M%|K68f*gY0wBI?PV5zI
zN_z`yg)4#If}MeDVdgIFlRH2V#6`{Dww|1S(l~W)?P?}TuxRb-kaS!K@A~M}om=2N
zs7e&5oBaXBUsP`Iqxki?W10>4Vq<R?<Y|;RWDCX}(Y0&gS3aLNE_u`hIgsr4OMuTD
z!o#vq=hzP5XE>h$`msZA<1ds5P|N<1m+n^!Y3K14JdNSb`XpUAHTrfXU!JSD!o@BI
zsUdz{b~GZ+x8s#p^2F^l4^OGFKyoA-&kPF~R|q$cPALS>0Z($pa^nyEpXle{*~J%L
zlm)6VO^w(#r{ISdjgoi$l!pNXTgHFZ7Dg__Hv}1`O<FrK2pATF&AP~(!u>A1Ei3o8
zxept|G-kpyQd9h?*xh(^&JdoL9SVuQKhtl5es5ZROq<~8W!qt|5SbAvbKxP=B4>}_
z78+5xwYff`Np&l+_5JWhvAZbgx>d=kfaXIMz`ty2PY$_4lz(T(-KFU_eYPZqZy?SC
z>mUcoSW^MH(?K;*OH+(biM(TR-$z;guX=9g2h!s4?qaUfyejIN2(;R4JZ1N6JR=n*
z!C3Q`t|1M=(Dqm7xd48?(_YZ*5J-4Vq(&Ip4>U};5Rzz2>Omw64vgpZ!$vcZ>V;me
zd$F5r$)5OcoV5x_%c+!7-TU(JQr(&A>kGfGKK0g~cO0onxm>futc@Y>4_8amk#DJq
zN#w*I7S+hn*#-S-O>XA_RqnH2AJrQ4pEIr^kg&X+Kg+c)JayH5^>bM<hhJT2sIo8j
znO76J6`bdA?dGGq>@TLSjw<sEIQ7y4+_`V^K(|cVWbOdy0AKB`)h53JN){hc61U(~
z;9NKNP){@ku{5^)N|}!j+Mp~_ODW~{b`osvz?nKZHxZHzW7|uIt_xltaW>JOyJi&X
zQO+g^*&4_r!$&`FB#yq_h63>O@ZQ_YBV$HQ_J;tGd@Cf$JBo*s<p32Ws%~mUq|V`)
ze4V$DE9QZN!8<pG`ho^z;DWtv_X${JxX-ah@bn$e?NV+;DiBZO*l%t#4z~m0&Mop<
z5pnje#@8)sS5gzCqQ#<&KO$I@GXGFv>NwDRa>v}mfe&Y4RK;Jx(qQkeuw4>#T~|8p
z6h+h!m6}eTeOYZH6p1$?$Srr;9%y^H?c^01r#sCL$(ixv*z>ul+5*|k40B%C{8|5U
zy?PZkY9G-IxShz=yhDoQpUJCvE<BUk*_-U<Ge+SRJ}0It-!-z{HsFV=QFBx@c{7F$
z%x^Bu*75G(*~v9qqqOCtcfq!CLc?|Y?&#;uRTuTs<y4z)!$O)}TG#!?*-?i@?rK;L
zo{e`umqNVYZ2_Q1{q$P&eF=~MW}XoYv^159dx|^aX&$xvRk|?D&rn8JtTut_l(fLh
zLgunjcfO*i{c>Z#h1<t2diDCN3;xq#-*wZ0B01g{IX)gyyV^qJyIY|BBfkshk^Drr
zy3%=py!9K9f6t5VGzH}N+KI6Gtrp&$nP)(j*ezrlgExGd<y`BZy#`jK%rCvlGGXed
zH;l7Ok_y-R0~;+NC>#}a=CpLL^v|9$D!hI>^n+ck40E?w=tM0d7cbDAkM%nzNJV`s
zw7A4*FLAW}z;@HYL)4C$CrN?(gYw>fEGB6IrUG>0?+>@rimTC*i?A=sm@4c%(_HeN
zpaj@+?=8GxQLevzL)=;QiEQ&ip)aiR*w4STVqOP>ZnFw5H}s)pOy}~>m#52Hs1uRE
zIT}f4RfjQj;0`yd8K2jgfT&q`8n&^AZAK$TMQv0N!$r0bq2jr?!y?PxHNR6jGs|Sh
zj?wl{W#oGL=G)&@HzjIjMvopTZ3V@K!AlUott^uc#jRkyiO}U6?kuX_4XXOlzjI6z
z0^ZzEwc)umR;+3Dq5j1ny0MckV*Ik>gts94c=AtM=5_8xm=lT%UGtC0TJagqdkXw*
zPSVxN;G(XH2OWR~2)qTbkOXfA`mx#A9Q-10;f3N`h`c>yO}@V2>O*;&Hj?4qVVHTj
zu>4-1S42_8j;&L!9;nK<(t0Knq-KesZCyVIOkMh+R+eTUf9-!{nAVi!cAWBcEy3<^
za#d@d1MaivlVRxG(p&9am(aD=q`kRw`AE7j71#jX-tZ#WpZ=i+V+xxyoO@x~-)fxj
z&F^B44pBo|AY{=H-}^1iIA*o$RhLsV%y#pqpD4QU#FM)SgA0lUxw;1Ui@FjajZ~;k
z{9ZvHk_xVIf}2F415BRk&a=0P4T2Zh*XH1fa8H=M3Y8ZsL~z()4vUc<*6mH-7q?4j
zM=BO(_T^XV27Xda9c8r%)+W>DThaPGrg}KpC1%8G7F^U#Qjg3m=?&LQbZ*8O828!@
zOD*f0{_s|$RK8q4cUzOC#&mv+U7bWzM*HG}IP~?{&p*EusakM6Ji#SrUs|=AS&qk8
zTX~M{OXo*zTNB_+9FBdha=w9mwG5s>u=xBHH|Hi3_zK8z^VO``j~hCoX4FoE3J`4m
zIDvxZaD$)@Akjoarn#wC(!BEZC#y*lL8_aBHGQicBPOum2CI`Xh6vq6=hfSN)*{a`
zVz$ype{2$+ON9E%x1I)J0cAAfK;UKhqGTHF?4%Oz!tvotPi)MrABA>*T`gHtiC6O@
z_G^|8O>v&>E5n??d-6zBShfHhchc;61r?X4Ds2(B*;;5M$747j<LW=|9!1})npQ$t
z<u6a==Nq_x*>?2)D%OwR6hN0t0ScPmE0aIWjaSB8-1bz~Us;bW6-Ex^8bUoeNcW6>
zjs8ggRP(0CkY#aheAKC2Q2W91PQ4SE?@~&KemzDP5$>?)2-Kk-vb6iQA#cJ-F*cK}
zQbgNgCUmvWwthlY{JMPrkq6Uc36d@}(EMI4R+cDKM@_9VC-X2(%6Sc12=)<xmH_SK
zknBJ|tEWBJcAMk4oGM5aH6xFmuw)H-)T(Xe#IFNQ3e`2gqtJum{vF#E=iRD*&vW{y
zr?y|zCY+Cc-Y(^hFi=j#|DhI~8;@-kuU?iWIyl=hoPHb+T;t%a$(NiC9KEdP9i66(
zGbl)3Xqd*v9TEMI<6a8M`D*j2cCg>rf0+EKdg?}AYbo_01cm21v8Lt8?MgPHx$&#X
zGAyJrEGkDMwK@@Dg!{?^Lx6N3GL|=jhVi@%c%uuR0C>(bj@^ZL6g3H-@FTdLL5vD2
zENTW^bt%x{O7M>)v@I_jkAZAF^Im{=>FX1XP20up(GnU?*NoCv&<)fXwT~9Lm$|5x
z`4&j7oM#%6`*5+c%DE%1{w}ii>%QqJwvqlMer@&@?h(6-RE$g0%&D{W)6A`{j9<+Q
zN4<_~hFX81p3y2{S28Qc5`~EF+)c|}2wC#&T~rV<EZP_|B?-xob|c)s3o^}6(afLM
zW0D;%$^)PS!BkxPy2(23=BY&7bR$reSpp2&W!t!WqgI{YF-cz_IcOkA^W9Lfb_x}u
z+k4j^VGL;*MHN95PGRiow}B{FHa=aqur`o$$|QGxeJnp8Hr7DMW^V)Y&ld1^tmD;#
z^%-fhM8u+y&EnkC!i)52tT6d*I&=f^6jIrt5q!BJ-Z1a1B#%1iw+%+*7Hd+Ry(*kM
z4d-fRD7uB{_R6ija9D=lg*Ns^eZ)rlU6}P_Obyif2GYg7A(oSzqt`$_%dLojGq0K^
zs^-a6S3&|)`_t3JYod0HCJX(wTBp9l^<zbngg(v<*;=aC^5v^p^3SMa)%2(b7iX=N
z*Nc>s+Lw9yB*3@&%R5F%9gIG9<86(A^qfoII=9G%u^M%5dBB4IHYR_{5WJPoZyHwd
zuz<N9FBlA|__w6rAS^4+l;&_xT%9j4vX0hd3t&Ig@OpoMP|KGerpk48N`TQTBwiPV
zr`Uk~g!a7FEe}U{DiiBV;#Vmkv#21%BX(Io+FB!f+R*%0tK1X?=md^Bo?8L&(^ad9
z{&Ba01?+O(=LHMJ+EoXep{I=XqOL|utvstAP@<TWx(iKS@ideu3Tm6z#|t@7YLY)K
zav_yvp)vTot$n#+bV?doxM_SOu^CX4d{|^xG>X1S>ssf{?HbkKKQ<XV=sNmCI9}+7
zfqe5T5A98njAS9be5ae=pALZ$B&2%??}88Re1|YT*SP)^R)lym{LE{uDaP-Dhey{(
zCUK}=y3jD%+F}*?Q{LV)!Ma8M;MjdFou*~lSxHU&ngSJaAE7+3&}fMeP%Vc{i-jb3
zBgJava5|t7d?f*<v9+s2e=zUnX$Kx_4YyvIP-#OQumjabo)P2JTtBk(RhJ@IHF4<5
z2^4WUV!!e`M@e&x)kYhumX%Z>nR64(^>}i=upCIVYe-hCh0sPXdsLD2hEzkZqj0Ww
zOencwUfK8b;1Hir2$6$w@EaCZ3(Tt#0E=jAtEl-6K;E?1LHU}unLX22-&9;nRM7y2
zd_A`UjHo{LEP({XZ^4|^TTM!BKe^a%PVc3;y+lbIO>xYiz&TSm&p<A@{?CiM5ebF*
zKOY)JqTq2>r1yEK?NijA$QkR$lWp#c0XEsv_fH9hCm#0%ALPS=N-`8Hc%X{T{!wt-
zN~}9Rb<C_Dd1f3?8_1Omn7+V!E6v)D2Ixcwbpk{B6ZipBys`54)8)WN;48&ViP{ec
zK1F02V207~*kdmF;(S(g=iMErgu+HOgp|8-V-(*=ril}(z5#(tlZxDY+#Ec7zp2?I
z&2th5+WypE#|@nU`>23PnFJfo8SdBk9&m0BE9<^?0c+>%HhMEu0uAHfW<?RV`e0bD
zM#FjW;vLk{mVb#Ssi#MDOu6Ub`S34aYV{u`9mMcCKDevIq7kC+>D2w1HouaqM9r(s
z^INj(v}Zb>=-=^osFuFp!wxD1|Dnx$d1|q;wEG?Z&1JmNs>t13iQXq3OlGB1G!z~B
zOi9<J_dyLBT2%rn)7Z5+DI@F9ujRkI?0HRkM?GlzU><trQ0Z?5wlwGx5r$m7bH1e4
z*1XF&(FVQlev|;`tesebn}mSnFhRp%N^djg+h(%@+&hYoCA(EPb;>D9$T^@`n%DE)
z9<|e9Uw^PSSY&Y4NYR&|CK2Ebo{fX&0R2$zs%Xty9ozw7p}LzJ0H~*jLPuZj%G7iL
z7K9H9d$#6B2Ud5jNbg{_O=B)rh^fJOI<P8{<SFDh6u{5qsQ{MT64+e=*ef}`hRX71
zaNP0v8Cdl(p~)HOS3h;a<Y~8gf>WDox9}d2s6vHCBk$x7Hd8Mkz9XC*e$+JL$_C9#
zdV2TO*r4j;-4pF23p(fB^WW6#&Wup|SNrEU0^<IGezdBXfP1O^%z4ywM}Wr8aDSZt
z&Eo$(Srni&b1@SH5uu091M(m~ywEG+t$Wj{p=Cu?i6p=j;O&;*b<M!}c%i=QcVKY^
zRAI^za-|Hmv@;FAJ6(z1$0$Mva1Z!u+@iaQ7Agsxj|uZPxUmCcfoBCzHt&c6w1zWH
zY$hS_R)(peS^qcCvRuPv;beF$(2p6&1Pum@fZ^iGEsFQFTLWs~S5O+m<(1ZPpC)<a
zVZql~oVh4CXEGO)I#%l)P$C>o|Gjc|9(G*xQ@Z2mlLC^ye`i4fc>?GN`ymX5KL^Sb
zc%Qd`2)ph#r57LIevt+_GzxatH`%5aT<AGHPS*OzjIkfgZ#24&?QoWwT+#4e3iD-<
z+^y!PG|C=o8>l*lUv$%0op|&A+wf_%X6=O2QdM%&^^#KleF5$x0s7vNFAH>N?S#L~
z<Vt>T?zT>F^^OtRr<aSV#qo@u+TcAEBtLjIB^v_ww->u9>c8%#y4^vodp`Tc{W5hd
zSb(mHJAVpiTAiX-gn@>&)@dEUhGfIzwL1P|=Q{7{z&nPoHQjSyp}Q$=FF%csNB#OV
zp8_vqsuzXUGA7n9u3%}+-9&^@-i87eYBq0*Ah_MSw0E-x#ZY+Vv4C;vp4*}`JpvIC
zhlZ(PwbmLfpBYw)CpSoUG$+5m3yjeQxltk!NQQp~U9RAW)m+o9+`fU{Cd1vqDKoSG
zs#!kUw-Y7Bakz~P)xKZ4t><dhJuILZoBbcnzfgGRLFq#aK2A9r+u?hxkigMtZ#=)T
z0<S7f-z_0v3x&vsRB3Ma%-gDgVL|<;tU$2@_~l)B)9~=D0MII0l6g;iLbR{kod{@+
zgAagx8a%ggfPUuI&+liJ<Vkm!`!YiJR2uJV*^AT07<=$*5LjEKF~s#1SVVN?Jg}{g
zZ0(K_8mdwoB%bbuPO)DGZu0DNrlvgfxf^$wBiL7-{|dbE1#>DjM!DAUyEdnC^&O9n
z7I~0%|Klsrh+OPZ+B-MGaaV&0C}?1HRdzvC;aKsI>fZal{|%7NS&_Ci5Cgr6b~%-b
zp549C_@a36Q#=Bot6zP;j|r~-Kw*4TWJ3#OAK|!j2xyZp9yRT7D>`{hOW%D()B8$t
zkK_La!t5_<g1DhD08jt*TX1DR^&Qz_W-+QdT@NTzLQ()DvUf|o3+sIW&`>}xP?^tc
zKFWMT<JG->(*lPTkURws=d4p_UPHaXw#go2E3PPmbpD?V^1hU=7#0LgRB%FheuMq9
ztOCga$pY;IlKx4(r1DzcmA?Rc1th9Vk^)~}i@*1i647voVC#$15ujmg!`^OeAET!(
zII=|O#u+G-thTXM`Q^^91ax8>et<`$;9D-F3hmE@?CJ3i*mSE;cm5Oh$USN@TcWdp
z9MJcp=FO*F1U3@1roxT*Yj4}fNbdJgm5lQ8hGW|c@!`RZtYU-PMt-q}cTzxOXF|mb
z^&o#+T<6q}CyoC)(tGz`S2opZit9W#zr9EEUwso)46)MDe-jQ&bA5iet%$_w--*=M
zycfL0*4>W}fpJMQs7!7X&?SWM06K4n_g|kz<Z$1;>t}Vt2*;w3BOlbA>D*&evyMh#
zPT}+#xC=x|IQw%QOHYFSSFH_~ySZp!cm#k+J-CiyiUi`b*I7he=B+Ls-<bEWPK~Sf
zX39FuigXx+sw9syp|A$d4#X>2sSAcQ@j}}HG;`k??$`v7pkNsZV>n+0&7{NY3uZWH
zlNx*_qs_%!(R`Sp*T9H>HN1=!P_{Ku4oJ7=yPxMelzFd@|Nr^;Ii&<=KInp?lg&5{
zt^dcX)w}(m=_hxix>;<{9?16W;=#WH8r~O{ZQf?!TW~f{x8VJD>%KW+Hi@SlF739e
zAk~4?3Toj8l=lqYbd&G5luq4BgK6ZrU#!gD4F;I!S^U>s05pB$X41EfDTp_h{v=EH
zAz524g=Jgg=yvOyf9IZVdxF&?z;w?v{xvM%R6FT9GTc%5et^Ry1$}Ri@qA>IRL{Gg
z3~{f{i*%@WM}U1SQ32&_DqIw3N40WwPhLDc*T$7D&+VKTxo34^Z&Q;4FjA~&3aYnX
zkPyA6n8*o#)Ktr~QDSzb2qv{#6uBSO@0NKmAGwBP{U?z~2tX0plX*Y?hv|#}XT^Q1
zv+Yqh&tmGQn()!r$;ZdX_lv39MU9E3n!kiT-4fX?<L$XVC$srSc59z<NS+R){sZM`
zpDLomiWyM4sM5Gt9{cOeABe1WO}WiWn#p=Peh+012GbwgE*k)?E}@avFigbN8D``?
zg4S_-iY0lP2WfKs7sypXWwiI_W8sk;9^0G{7qE})_<ZI|U=WZRwcGZ)6c!DwRDt5D
zhYH_}brz}?VgwL$;{7K@aIip)bFR_w9j216q}-G0{}y;Se;VO<e@bYj#`GweGbEC`
zzYO1-&i1_DOsOEn6OUF_p#5y%{ghu~sPDJV9IanYt$=o@9bth*Qq29q2nH5yU)uzD
zDv)|cg8iQK2UHy*0?w94TB4F3rr<xRfLf;W70~S8$$sn0ws9x>&9d|V@E#fC3Qo+&
zq^r%I%dh6bSio2BV8Km$3!|Hpum16}fCuyJ2-v7mlx)SxIKBnAY|ZCSqmBA2GggB8
zf=tw7j~-Ob&vs%3E9trnKwf;1pbEaV9J`(~E56SQ7nAr<$~Kd99Y5|l8?QYl#+~~w
z!FjIWdK7{DXM`g4$i>sm9eq~+k<tuG?dC_e{ofp8jE-l@AzOKCZ(V*=S1Bbn-0pt{
z%|NP8x$gC(R2{`Ze0DoM5zaL3(HF_!Y2tjz-=ov3#UBr^2Kv#v>U}o>tM~8F1C;@~
z%1#yJhe35{_vz2CP!NPtVXiPk9$BVVCe@r!54d$2n_L0bDE;iV+G03Q9Q|xp4f~V;
z7qz-YY@kvj#DAtnQvx$+Y>cQz;f?oZY`eWxjY_UvtHT>;3xLzUyM&D4`~-kwR&2qV
zq%haL{GS=DeB|<fWw3cSKKH8mV+=<vIMzlt_QIIw=7V-NUW0bpDM}$<L1Da<sId0m
zJNU^1*#D%ftRA&h%HUQA93O&<?0oOe<yv>TS;e2NUncBybKWX!)B1QWu*H5%q>Nm}
ziiqhxXQCz8uj!WW{oD+6Q>>4(Iyc@V-LkAaaB%PXSHC}r@SlJnG{Bwh?~{})c_&7c
z*FGc?SkO8u1ho>v48#AYv{#94F~_hy1<=#|dLk6mYJ=;1ug~n`D5RiK<{p{#^WCIw
zBo%8>UGR?WvDZ9hIIjVe$ssd}QnpS&{<oU2dlS?Tj&r(sXT%fl;gdB}z1c5#OPT(U
z50v9kx!vqd!WO;94!NU6s<80l#cNX-R%4Om)TPrgt4En8jT|zpKwLZ>*o-{KuV{IH
z$pG}@f!;gEa<Mz@`WeVaxZko*7}>gQja}cJ`oB({`_5yB{`ZNqZBt;E>GVRLPT0LB
z60*PO&ip?%_^YBxp2j0vbM5#YTz)tkhW%gpM--M{4gs1+Kf^E8PZO_u0a;vWD<7Qy
zEs^HSlI}Q!Gzmj3)s6JGC)ACMZ3h@r?t|sU240l$9pA;ZM}<8#4Cj}?8(P#kqNE>O
zFhfXeCc=Mjb7t?ecbP7yM?a&wRoWem4#Kf;ZOB#0XtQO#*lhB}mj8bQVRcP(&=GJG
z2Q*l%_EVkC;-yipCN0rF2{}r+BDz3Fi_Pz2;q8FdSU47>kD;C&s=}!Ifp>S6=%0*c
zF=;4uu^t4uss@xP&b2J%5Q&_=iJfYq_5B4r{GGdzx3B{`&!rn{;m(=@tTL}6nOC7E
z4Uwoy*ur8H59$hm?b;3lFxJ#D1_`Ha?z!0DG_Z<L=xao^>D%ovo%s<<^>cz_bFm`|
z{A)u*3t$x*r3zHFDw++IkWeWZcsGz!qAnpA<T4k<9QLwAXs{@ASe)pHU(-_*>~Xv9
z?U_6LOm5#*4#Gm!LbpMKQ^`cVSWZY6PQ-T;p|01J9Yl!~)cv2G52qX(<Q-iEc_NQT
zXu|Wrg^-4)11F%F#O%x8K7J>{NLj43I4r~oqH}&n(3^I{&PcA7C!Ol>D{)ws3rYNl
zs9LuE3sj*sY8Tb*GmWWu#q&z-7MB8-XVXQEq+o%l5s0bz-C6Vd%8D;XvRTRk(e+U;
zsulT4Z6~=mbdDe188Mg0=?)*Qr$7!4k*=0SFbT%P6{CyK4~JF!OV7n7XMlKa+_ReI
zRyFcm>N%A<APoj6^Qv)-lL_jTuevr~bwS^HMFOVA3yf|XllBINFpRR=0}-!yrk!)j
z<JHIf-&WB>BZ8d|VVrA;{AtknddO7ewT9cC07R5L=L1=;vr7!J^Y0Z9Bt$-|0jkQ2
zUlm2QYc*IvGAszJ4QlpbgBPH3^=x6C0{a*z&jgb(B|$c|e$d#K150HOn6iFy>z6!8
z&0od#(#NoFkHYisjq?<*^BZp&S?9$gH4v<{&}K&7pa|BelOrG|w^t7EwydaYg)Ggs
zq_kLAD(~)r23dJFH8=1`w)RS>aHh00Sb~6_)(N|zCP9d}VbNg)H<<ZGBC5(AsTMA?
zk&2SGZ2~OAdFmk0?;=mjH#}{>MXQzi@BC{)VY$<fU`V|8yp2b*sF?quD3`n!aIH6+
zzfi63uH&7XfZDNq15><uih(CZT5o<Of8>r>6eKmVk8`g-6vEkupq`69KOZf-?lF?o
zthf*@OwJaFyRK*AOuUp&6ZdAEIkD?qBu8CD$mcG^+(;`O78R2_rNEiGL9L~ybYK6k
zVH_Sw!O8Q28m0}tg$2vt4Y6^R+fFZWaT|e&&ujW9LvX6~OO5Hk^=|%QSKMl-S@t(y
z52ZrcWNurR9p-hS!0$&SLD2fuqX>YYSADuV>TU$@o=K#CvkXgo^28c_k^jxfNUI5B
z*kzc8u=C9&eEaW}=ArX)ujk3UWc~4CV&9HP{LX5@cA7sMmWo(PvdXQNq5m3%r#uQZ
z_V`8E@pn^u6H_f@=2ZW-zpG)VgsCgfCX<SvBwg+9=OK4DlMfS>H_M!3FQmECFA#37
z(Fo+(rDOD}Xfv+edNxlZyJeBDpjayt)4&d_K%Mgsf@Us$1(8(0roCo;!(MUWtjf2L
z$CNzK^}Eoy<i@=>(2Pd!dcO5yfsQr3maci|bq5{r?++p1U%Ihj^~1K`tj-Hld20$;
zqqkV(47#?PBDHA~8e>htxh<uC^*2w8Umdsl!mf00*0%Ew`Q0c7U_A?`YZU?lQNake
z)g#;3r4w!o-JDtnj^?bqipwqgzpeG#r+o~Y%xeDWyK>i{_W*Lffh}9ry>#=j3A5FO
zAXEi@BhJTl!+v!@#b{sfgY}w_d(Om$*H-J#xUgnE>McY%TlNj5_>JMHNKYWDs<jNf
zjW9U&oNBp1%!Y2yd2)N#3|MN%;##7h1%*K^^kH2meVl#fOWP5BG=7FRJr4(<UzsXC
zPjuH6`{4=RHet~~ZHm$dW}J=eIMw}cBhsSF)`Smbh0~+UU|K97usL6t!T=0GdW*QK
zJwLxqZ{`oHW^3NP^joOP<5{Q=n&CZ?Q_GeLeZdwQ_gqy&r0vNU37Mb4B99VN+9d+e
z!1-v}gwlw1Pe;>7g9;BiG1<sIztrQS?aG|VG!LW8vBR4)cXxH|*K`*<P4NzDKILCE
zN!RmjY$$9@vMpq(Bz187>^(4bnLK_meo+>hbA@dgmv@BZQs|<lS^?%rtxLE>8$Se5
zMg`+sE-t=NX(9Sl&hR<sn>7m^H}E~W6$UV5RE@CD)c(nzzZsF#vXOd!1~q}}k4;}h
zsXkL6v2;G*Ac)ce#}Jrjb3EeG^dN{D&*-6UepnZ%zd;Ep(?v2w%G-dg<!uCVco$5r
z*KvH1`|xwBh#kaP^{$1X0p2(2$IqXwCj8#15aQg>0lr4xAa$Q;_x5z>C3gN9%QGMb
zC7m8X$~UyvcqFkN(yT;A^4Ig`EDgAiqSsDta==%??WJ;3Onh7Xj_{ed1})|cf<V)Y
z?3cx)F(u!IZ3<0-=4>2Y4p)4xyI(Y}oF?(#yt?J~{XU_E19}*)fKHcLDksbDX?mKz
ze|Q*3#UJ%;94h#*v29mT;C!awnyhtBSp4T6)rB&jQ13^yxped!^+lq{vsnTqmpyST
zKZLMA+G&^{)miM`VC5|xU*RRs;!2zS!`p*PkvRPIfkd%hOBne!<5`51dD9Y_BZ0|7
zCz7Zc-i@tDoMqc`23|My6+R3O@nijfXCW3+DpVKaW&0Gh)b7Q4w$Fnu$e(_t#`?<Z
z=8c)m=<%{VpiHgsfJnd9>vN;*o2QKUDiW3KHQB#Kt$tY&`$ef45T3=+F$VmQBIT;Y
zb{@#sdjzkeYeeoC4u(03=a}g781Q^!!matpV^_<f{+9hIF#%WnH&0p(f^v$%O8Scx
zgK<2zwK~S;i=D5qa;&3|F(;ZoQ+Wn%U$uIl6KYR*;f6h}8Ne5Kh0D>KkVlw_we$Ko
zX1p(>5WF1JMfup&kY4tIrfg@qG`00rP>=|rv{cM<FaAuX0KJqvptRd@szkzR*t|m8
zq^S*iMnjC5@Ybwuf1;aH3PD}IBl<ht_tbtgWi8`t4i9j<k}0ypwiAUi8rYstyr=~+
zrw^VfSD+mOR9S4%|9;RQkJb#II?NAcm^(=De$a@kLG8*G64V`l^M-b?FN=|PgLm<D
zhL?PYP~BBYdN@PAS#Zk<xmfL?+lpa;RIkJZA&sYLdh8)@p=NX{&bl(1CZv&GNL@jw
z8i17s59&M>(fIkr#O*E7g?+$--#9)b&%}Oq6D-+`a=a)1-9QT>AP_ZOm%}4j5%a|>
zOCgggpIWi_$zwj~6^clE1;s><#f~W+Dt|W-%trN8mt5f;KdnNWXdOp@)|uFA_WTM9
z)V-TggFrp5vj{ru&@HTNy!^#MH$}uC{j-qwSxhV~IEzuvZwJF2Nm*ruY7+FGU*u<=
zeE{C{V7o(&2OissXSDzv4aUf()t=czve+Eebj|AjsV<7Nd7KuP432tY`ho!Xt~v9G
z?`KlN?{cFOEFnU)lATze!L+@6E5*j|-g+|czih_{Wq+>7KiP7ETedrV^qK5|!B3Tk
zAIK$Ix}AG#b%eFS=DewSh-?bpf*@mwkjL|2$}xbu2<Q!yua)vaVDJHgL{EWb6%ua7
zX&&Y)%r*I`vyX~W2dusD^v`}swH`1(P2=nY+DP|~qR>Km`=dU`ljeXFpuUMi?~4qK
znvhPiXe|P#;Nk!w(?aY@ioqUUuUbGO9xf@u!^ygpgZ-pPV0AV?CP)*f=MQ@@d2Wmf
z=Xj=v13R{LBa~OHulG_?Yl6w#hI;5bGx=cVyZXkIi85s0QGf+&NP%Qez&4h--ZKjT
zrcwgqXqKy+c`_8njny-7&_5M<XrK4w_{o6I3yFP+H)!npK_XfqX(oQaC9Ef|dA-ak
z1=YXxfC8|@g^{|0fH`5B)#O?$w7TnsMIH+Lbvk=poUnSt(8QaQl=d`XhD(|&+xoL@
zQAuCjj*7Omr>_^bix`hc7s~OP(@lz=@77?imA6f!UwYFlQ_MW6&O^smY1_G$70glj
zxa9L8BRWUaSc0fH3k^iPiJ9CxlM>D&r1)o!50R^|T~d>f#-l?YGGrVXApdMDpjgD(
zOVQ6clE?o-9N8*P9OL^(2b%<$f{{KvDX}!#lpE%Gu72tHc7;zRy`Rsyvb&kiOy+a7
zK@iqL&%|C@@z5t&viJ%Xf2m_=QYRaHhb8U#VSvboKh-_s6DXMC<K6w81?xudU8eB)
zc++?_t1Mkeawsg?lN6ZtgpO@|^+xpc0ZpR-$YI=cngKc$uq!3?Lbd2Q@TDWLn{sH~
zPmC~<Zdu}`i~*kE)g#WP6t&dirJ1zwQ?Tj7ZH|NP`R@i7YYPYJa}@e=CpzJr2hrzy
z2OAqbo+djF_{W{r8hR7ZJEjD8^U*u>c}QOjVKo7y6dFu29w)ddfniVs0H&;)u_}G!
zi!s22sI@tlyU0biupm7t@(K)a!Qdj5Vu)hQFeLq^#gi^jop2qpA9iT^(8=ldDSpce
z0k;OWzHob~%r{JON-uueNM1W#(@WZHS!L4|#nbQ+EexeGQaP>+(K1yPmMbTz>@Uo0
z-&s=YdM+HKZ1mMAn_;W*>C3#wkgFx0he6HMJ%zn~sm7YLR5}`)Tz2og-<WT>hOs4|
zCf0w=VDQ|5i=FW)f6go?_#oaFFC_k#G+v0(LR3B{UoJ6-Qct{>6)PQ%Bbju0xvz@X
zKqRX6zBI2+Dct6>3Pnq_Jkj?6HVi|{I^NG&#<NPl3ILDLs{=X$W|jB(C+aVkd(JkM
zk}j9^ut|h?wGZ6SE|Yt=?XhiHyO{7wFjPNgI)`AlWe@>H`8ze9JwMa>J`)x7?kBn{
z<Wq1y-S01H!jpQF*b8wS2Vx`QkNiF}Zr#$f449qcDC{vvBphP?ygYnEJKwQRcIIz|
z*24MVbiRYjhktkh^QPT>eToLHiDuA<?1HlO=*PPh$mn%Qn%88QxpTlD09V52Y-*1L
z%sxNqli_h;sX30!Sh3>d{2Dt6<p)mUU`@0y+}gj`97+O2kol=eN6C>HoyR<DG+Jlv
z*p1DV+hbICO`RIDRa-12Rh?`U@3y&1)0QdR)8K2=;h)Q|i);&Gko1I&nU0pY88if{
zw3)K_RPD(!<b4=im@%&T-qgJMQr~=whF;F4Oa)^&vSf&5(PEUv3Cew-6#n`t);yWF
zFp6v`Dj0K@k#b<1Y!@cCgKl%K1nYkQ{Uy_KV<&OIudbiLg{>F1;27F*C62iDJYnO<
z5*#BB7v540ZMXRTUctJyPgU45S5#gHd=UbR?uj8}j;0Tddfqk1m{bdU=vBDCBSCA*
zNaCqOXD_E_|BE%?M?$%D(*?QMse$Xll~60v<(+{4cP2c#DdKri6`!~PTgGp_NlqIN
z_@_nj5Te$?OsPLuesf7P8DD&|UNCLFcx2Q%Z|47)@`-@QBv0+6yxQN$S63C?sUzR;
z9$D=xpgow8cVNYim(Q$7uu`g^B7kK?K5$Ai<rh&6sA8BJV_9QOYI;!h)Aq^sck9g8
zmW0t#-(tW0&cGMFksB;UXJK}JtFL;-7WKmh-hUf7VCXwpXGVlxjSctz9ceUeQ7xLY
zE^q|?jRQ<|Cy)PZZU_z$)v{1pdsA8YK>b}5wW{9HXQ_|D)Y#d*iK0aX+88en)*XRP
zIK!UV!w+KG@kKJ-0F5M-RDbLq>0iPrFS_%6R^!iS#eShz)l9AX2ktG1WTt;sOR}Cc
z724t&{*r)44!<ncryk*Zu!Mjw<UcE8AIIoA04ZN{g&SD&sLcFul@WD({R=)%_x+JK
zO9t%CICDZ=!o$dDb!^i2xm{-|n40uEDJLZ_n>iH#*fS4eI-k>iQmVmkhe2C2_F0n#
ziOvS4f(lJ&$FUo}Oh!m?s<z0?2_~Kje@5!rZp0w7Pzv^QOlqgc<lXG(H%(x3%(6B>
zlZh^sek$Fh^u;=&k5l2xhJ!wQGCCva!|ybyRxd@yxA??j8{DOwj-L>^hV{UQ%%gdJ
zDt?izerqASAX*O6(S^Lkl1E^dcWw@U{)UGjUraogdbz(7;=l{FQ}R1b7#A1sPj7AK
zOy}yfd{l$;jggQ~6f?yvjd5_KE;gW;>f$G(?}i?4jLMmjabxtNDAZ5_Mzj|H#!ocM
zk2}lIb@8oU@QWv3<jN5YuBA^30NEtYlmwd6USxx^Y3mQYNTQq{S1@|}I;u3|Qj%}(
z=}9@9idYj9TxKlUKeFI>!U`TF`NQn66F`an2%aXX@(4&(0I|S}8?QUH{jBi7)bq<U
zqzV)#u+2j*{9dxbGVjQn=zDd@M`7@JR9fTj5&lzM9EMQ2{0<aytL(2<Kdq5eA}n>>
z7>{Z{wuq2z{ZmF&lknmN`H(*A)1SOz(LLX-byvpQAS=TbCsN%2GAX{?*KxCg`#~O2
zJgF%wurnf`<F&oRc-Z_R(D#p;n>=aaL%FpAS`L=0wk(<O7D6V%A-@E%?C4%>@sP)5
z`@Skp%npAu+(rGsZ6J9a(iocn>_5O{C#N9cERK%Xd2=))BY3W8+t*M15c2J8x_vVv
z0Zqb@-@Odbg16qZUN)h3PqbZ8C#MP*`~AuErcMQPe!>?fsq|B{$GB?f$S0ptmDqiJ
zb&S`O(_X@=<Gj8?=zRKWz-P&SZBJ=o*S1H-U&ruA*T|c_QmX)!>ND?T{<f_hBa2=p
zjRBUc2EdWZ$qJjdt&=m0*)Y4+)1ad&S!$Wi-GxQ>(pghy*JS4fBhwE&*scA4izK!1
zm0<<l)eI)j7w6V((pTcs)X#YGdYmq7w}1H!WeiXcxvPH=PDhBJ{Ux|2>iN1H&b*28
z&g29bzdk^S{eR;d3Yj(w)FS}OIcILL1B_<|Z`U7xbA289Ub70%S|jSGgqM;i<?E&!
z`YUMsv|<FQ<%%#KVOO`+>&5Wq)EfeE7Qf`ydTx5FGN;j&wqx3m9!botiTZpxQwK%y
zA68GrJNY{!4_9BU=8I<>DoTHliaB9sa?JYSCEA|&qQC~5E8G21MO8?B4=zP;v5?cc
zH(oQU<KY)We_iRbrvJ`kQz{J${+T{b@PItj0O+8QM)q(ct;05+xU54eZ68U8^U3lc
z0wVEj?$4mL=G&eu8FWwPC#;y>hAUdcHT(IA^ix>bfc0-)qJ+}Y6)Vs^U8MFK*XY-`
z9`DYjXrX!8sV~32lp_zl#3OhJ>Uvp`BvJqbJ>R&|-3?-@FC?_)VHn7D*#&mGe4UFk
z8SX8JQ!6h2lLa21vq`xW`bd1t&5TgZ0$;7#i-yp>c;TCWMtp0%KXDen#$5(}KP3u-
zl{%(*iwq$<uXIb1j)uR)JNE2H_S<1r!7!^-O;w}UQIe0*MNdPr<o<lnpiySw=PrEu
zm7OxPFCiY&Q)#hC(27ts@vGuemIS4%z`Own5va1i{M`7oN`2n|rg=r^e>%mnQYpb{
zDHi?B_bs<}oz6=nu{#&7*fzm~1VyW6<|kZT22qnOzuoD$&ZLRGl$?kLmDXa}sJza0
z)(ICdj?mjKsiVB`(|I4>P%NsqC1Ty&s00i=xFy=}ee<-Y1?UXCy;L@OaKpcdxle=a
zPKx~O|EOvI)74d_cp0!M_{eA^B~m-U|4J-!_2GqcrZoJwbl{HX5pP%vQS_mlL-eAA
zHnk)1AK9>~Ox!<W1EljL&(7ZXZ&-zkvzG4&u32}^8i-8ri94-DW1)-leB+TIBv>{&
z#Ojb0bq*_Gvpn(r{%hE*vr=DB$6`Y15B{m{ocnqKAmWXQyG(S%#@c1EbHp6C86*0&
zwBJ!CT``L(R`#>U57zPlo@4aE<*G*~^qItGB<p4&Au<^suL0uTXCecwtegiVdcb$b
zs^LcJ-`IYBp<REDmus_41=83J;N{8E^sYKye;6iI992ViN;G{#=`JH0f%gz>HZ2<3
z=l4ZLDIIa{UeMPm7CvnFORk7V=LYL0vcI>fX4oXfgODh?fl*amNSL>w3NRJ}<1*$Q
zV>8|L9PL}=r~NDECPLo|3WX4R%dA{RuQR#alrS{2M9VJ=y(oCuY>>74c#yCb8!iqV
zpcWrE+*80_{yK`reUkb6gmNuNaxUV})XcsVdN%$_D0|H~_bktM>(vMRy<xKg+wZkz
zifvVxBOW?|_GM}Dn21xkeCs<TQW2#}znpM>mx&p#r#q$D=7UnR(8@y+*xm*)erZ$e
zpvxSJ-m7EezMlL#kUxMuNLz$mjpKw_c;ZmEIAA5QCk<d{6i->?6FEzo*A}4+C?(l4
zF(blg#Q$xz$s79DFG)o!9f4y%Om7Mit-`Ci2}##!_`7QQZX5}qlgL*ckM<h@&r-dv
z!WdB_ieNwYe4j{5hW_F0@(|?vxV9)#!uax;ro`ePy+Xqdfvb!s*GpVBkiYy^Z=B1*
zBIcKT5rN2tmDiVFJ`(#$6C&{-E~%;Je`&8p#4EkzYIo!cQ>Iv7xUUb5u#}idYbta~
zKlz(+lrqjt2Z}hNC1hN5{wmHHqP9}R5Tru(_ar4#2;@-4_=F%@;TdiJx08M|p^2x}
zTxW4@3X##YW>TF=wi*y>!9Ek<MWHOsTSdAJ)o_vB*Phdhe1P)3!I=ly(uy_}tkkY#
zgQ$Ad-k|zddyGzVf-i9or&N%YvIB=#^z<iwaWuET?C3NI6%gNpMI{AT%UKE2&vJ1C
zu?j?nJcqACs5wviTm*KXYhXR8JV@B}csJ0`{4V)C{F_%Epoqf!;J@wySW30>m_NhX
zt`$~P8E&?Jn6<Dp(tByEUmRs1hs9e#0sN8M6t_}pb2aqxt2dr289^>EIF|k^jhqjK
z&n44YebK<7&#b8pONXzHalDB209Zdf1iEa8e4vW~pENz;J$hcs6jGvsENDh}&Wl@z
zQ_JYtmYCOBVz{x;y0dL;5eV|uG2yZ)=ugng2PevB6HjUo5=Bz$_Ni9Nis2+A%EmC1
z2$MALy;mP1+jmB<?ofQx!thLWNe_qKOKDk;cglvknks5F6gTXTrv#P9=3wrPs!}s9
zh@1YlGC%}VRrbrC6Qy4O>EUy{UV6-Uy!C4KK$3k^n9B@3!U*1jWEG)csILFBQ`PUt
zoiR3V=IbKmZ#^_2r09n0c=Bih>k}hidk`TH&iaK=@-8*S4Tc!Cf5d`lX1VyIr>{S9
z9M)<faAB;4hD0wZzG6D2QDC7i!qY~c=+>)7Q|Xvv78Xa9W<CpMn`$ghS3TnmCljcM
zkF=T-D_MDDPX4|_LQFZ)=S_x~rSW@6)7fiZ=AS{>yr{eL4l5d9MV1z_XnIKzrt+di
zh+v5}+eR#3tUhSZM|lb_yEc!w>;rP?hf5cI-&2k*{KIhXH$UZMIv^mN+|~;TVtw08
z<_W8m${<X+*tVArbv@Q+>S8CI-^LVI<_Nv{c?vVy?WC41zV3zV+-$}yODlZ9FrR^Z
z44uNXLnKfEU6p?}urroK?jP2sA<lz^=18L(ZNkcbkeKCD2s(&YxT?1NjOGbAClqgD
z3_(0^5tcOyPiV9hJQ_e&`vrFZjAepxOLdV;F6A<1NPWNL{`L?zt=F~5vn`c<LstQ-
zs-fkyn*m7zeQEE^32To8E~d*4S>ti-)R(+@mt9qyPLKTjzv3sdim*rMII1|X>n^n!
ztbMrR@pEwJl9h!}i-`-C`3EBOZpmB6#_N_N&q8x-M06E_VGwT)T7z&HRj&H$iBD^%
z#AEX*^X7{FWps&HN23tRQ3p6stLF$F;?~r6d{9t%K*ehRJVTP`rQxs18S&NN-?vcz
zMwisE^KXXDf9LxdKv`l~1MtxzQO5J9e64&LuD*fSrimTYle6GZuDQZ@R-tx1(j+>-
za?7ENE^{LSBP_wxsQQ4!N0~X$?*+tfCFKBs)&ZJbQGxNmAHKU%X%pEKNgt;@REfL@
z62LzF(evmFB;OPl!|5A|qJpV;hcqZM9Zie;`)MfOIU~rogTC21B6j3V?S3JiXIOCY
zD||1*&`5mUn)1HvO9+t==Z9gNFt9d@2mQXmV^5_ukhx_*P4~=`bDB8f8FxXNQ{6<y
zZ(fPbp+U_Y9n0MO;=`sDo2uHUy4NdI<3)0lxFHeWB!fM<Xzds^L<9SAtSblp@>4-0
z0L0&4zdt*vHLaD1d9!GTDBJfrSvW{ul5EPB!HsC?#;1(f&?HnnCZkc==)3(v^Tu7!
zCY}tGJo`Oh-WtXdc6{~couxr6GtOJVqtqlgs4~ceDpTW|qUtFw7c{@Ad*R6+Xw(fW
z!&LJNUhI!Fa*W|wg`ebyiG8YnXUu$=DocA1M-%TCz3miv0Fob5{bs}S!Tr^+XP}!i
zA;PL4z(`vo^mm@~aB2V^<_u)gjaD!T;@b~di0;&X#{Ddlv3c#aH&|e>t`q=IhpS)i
zX~my$WZXVFBimQ`T=q{<VWD997tf2=r*!=mM>uaR$m|ZI94_|Ilh9hB$-NprL}#2W
z{ptAY4=+mda!X?xX`ftUA2djcb8GS&BN(QN9>m+sBJKBkx9ksRTY`A77s`)H<&Nmp
zSLrtL>3o%zgjsH`Nc47OefcsPW~HV>HtCyO2G{;<k5JH*Uog@F3E#yZxB?Sk%pY#Q
zr*EKbVu*Pi;GUxD?f40aZFvzo<N3)VKiRxT2Q%%%a&xH(M77V-GFq<k51RYyIIU;f
zvh*c{l$Z%Ovz`Z4#=M#`G0+1OKc+rclc0VjRhUFvp%cAU<_LxNLBjOYpZA?Ss!4w@
zTAwd?l0f`asZ-^xnFmnfR`$xxEp4Vq@X)+(lZ{RYJ5ply*;yxlbTZz%@}AAzgxIa5
z2Id6!hX*>#6@$zg98AT)*U`+PAjT)3UhMC>Nhv)lWzFY9r@V-e-C^<J%xOB>T|=g3
zjEF1^PKazqN8{&t(D{<7uniA!JhAl;J!sdpQ+^)X(o=t(1Xqe?e&RZ3>cvaAOz`!p
z*XQ>jS=ZOIfR2gxW!AG5Z8<*?*om0gL~=~YZ+0ag7qnS2blE)Lpvrgt>=aGLa|Ncg
zc3@ZFbIEqcXqm}=nRHdcqsbGv7m(yVOCPHiIbEBe2W}{HB(0g0r#}0`9ksCd#Xq%|
zDi{@F1Nhpi<U`9SY@bIhQCrGTZ#Zunb_3)#=_D|QKljpQ;wgPvGW=qs)tSk7DI@J<
zu>66=PLz)BTQ)EffY!r%{`9jw#dgD6;xA;B7jd#X?9L9LKYw`wE=0srm7qdDeIkc6
z!@WoRb}AEhc-|-JddGfc#oh(l_86y!Jhoy^f8S{*L47Kyi0irm-ArO5Iy0nw6uu|*
zPAK#GokS#khQUziGmS6WR3P@R?71#>iW`Q&{29SR9Ppr3=36iYy8RomA|T$=u{Prg
z+%fXVe9Dcd^B?f-HNbN*-|%QLpxG?y>SE{lzH_EFN*#r5t8pi>0jXiR#cfH9Byr2R
zzug(C<SVwGckyGiK-7KZh~kL-($t%CS;8c>GILyopTz=>*5E~GpBPVZ-4;xRYU_UO
zs`|d@k!kAD_3fMn(Yn2jC8*7CxaOLgh}bZS)tNJmBcZfF*VXG&daFV3+>*HLmCwR^
z(JY7fnn^IrD#Pa;uh9s}1<%nYj3Lc+*0l9ErFjmd)W=h@#mS2KX~dIu67!V_-u=!+
zY277{^$Apu=?JfX$?WkZQppAGz48I663O!6hUh?cB_)Wkqe#AK8$WO=-7k=cS+w{f
zp04Vr|F(D7n}Lp6cT<`^{vj^LD@;?l7teOZe7|XXlS8gT{2ZnHWnaDnL{NH>FXAC8
z2%4MdG+9cWq!=<%qITu<FCR=Ca;%{3KlGW>U-J@V;`i^1oJ-{<t4&y=S>5XTLD$@W
zej>hNpBttf`T2nQ+AtDH|J`~~wI`aEflJrYvaDZ!^kZZ1G&5h@$t-QDbS9r1z+{`C
zXeOmuATPbF()$$Ha^u#{^D?vWQSEjpPo!7260t)sHIK$)0s^9`fPqubOhCOd<T*La
z!F%3RD0z68UqC!0B;j?kONcgoL&AvRt%&%1R<!En8xg%h0aKc6BE)v6q5`V&&(hZG
zv+bia@A@duffLHO+A5moS_(9<6t)l{=`{W&{D3Rw+sU`>(5}n@$_<?F^(+agqP<)A
zW9}rbx+-+;Du0bdzl;3B>J(1Vn>W4`4BXSmE=yexN^TlgNwnaVuJ7+&HxK_G0E<9$
zzq<G<5nWi!M{C46hniNTUjc1fOtWWrwRieOf*5M#^?)3|lfKrJ%(9ybKbWDfM$*yg
z|4(SEZ8mxpX__#<JEs5jggWbUwEEZi7fTO3&H}p$wfRxLc=lyJwDcNJ#T#5`d+2~V
zKDp-t(kHleKv~Q+&3q;aRe)U}5k%2A9YQXe8lJC8Sia5X=F2?YS*-cP&c=H}3w`5c
zZx2}neElPSp?}Cyp>h5Pd0{`J40q||x7bvYs@LUkca2@Q#;*J%U+~N9s#{#jZgYNB
zAhf_7IK&8o7Kpwe_U0sa#w^`3RJz4vp5TrfP*nS{INuIESI|<3^)0qmPLK6L8SBt@
z1TiT5QbIidn_BYM1*(jQxg~uag4VcxLcQM1R~%b<R~9zN*)6uZzrioBYuw45_e#J{
zBSLGcCnT=pQTH5r7r?=>K)D3J+&Iz2JR;_v{Hi4Al<H|ibx9!99vf~(>saHXChX<Z
z`4Rn&LmNr34taG>>o8~3dYY@thDr|^#G9NG2J0?zD=#SXl0xpXHq<z4F|#?{uPLl(
zGh02&-KPZ6S<LGJIr&!Y7Mc=CMVpEEulUWd&h6wj=a0V0b8iU%Jjb2#64wr|b5Z@@
z{FipfAfs{ov={rdrg6)<spDIL%`4QsMGbijc2!{8l8^=@pgj1gB<uu4azvMiSlEX0
z*u#F~Ea4~+?1ZJ2n_S+Q&~7iz?z^JD%|B?jnVe6ldNsU0Cx44AEIGa)2^~voucY|n
z1fO;|{xJA6kz{gWM+Sne(ANUdNkT87J}-#1#k}f?PT>ddR+spI{{5owl%ZPye-HS-
zfd8HbiBI458{ZrBRp4K|--GyoKA=BOQn)jc&<@D)gUap0NDymBdRr3Ol5qA>pG&CD
z3T(TnOJ(0S#Sa<<%b^d!wGyg_K6i_)Lc^*gWM{awmtzx8>PD<Q^7$>MHw7^%NxN71
zNPWy;v%~S4Mei%>o+74!{Pv^1A23G|Jp)>SSPG&s#K9I@-FKSPsc|Qgu(7DSc-*AI
z5?0us8OmftyR1li5c2@L0vUjVMcvvW!Z;H13GKm%{@hR|PjYn@30C3ThI;jJ$3hd8
zk<vK&3{4%sqaWbK&77(;qI);u&nDFEIqf?VCk@tlQY1bN$XP||#uRl$T5D`S!-tw|
zD5={Jw89P)b+>Ai+f-rKPF`M<+?5Ee@B`DNFk&%xfFKE7np$t~D=m1U2*~l<IXq72
z?RdgauPS`o6YC3nbDlF#jv0(f)I%e0G|shfw>?Sg4(ECmX-%_7Z#r+1a9$GHj&vlj
zg=o%-hgewDfx@>HF?Gmc1BA*+0|Lb)Ag#hKD_qx+-2`hCz7<K11hLRe-vm+{?3zc|
zn&h4!776v+D$gFZAJJ3r9gV&JtoTMFAYR}{3Wq2(@h~E?MZ>9W@cxa@i+IHH*b`PV
zW^GG)!{C<%wq;4T6){^h=vWntc?V=2(UX9vvnDlN)##l?6O2AUAWi~$*hF2|8z=9x
zm^)pDElXVe0zaxp6w47&B-A$~;Vt`BqSq?gN@DJ{nb7B88c^t_?8Bxeb_G!sRCPhw
z1?nlCvA}mDF&g}e!nZBSwm{50a%c!`f$bjC+fJ$a1=HS?d^;dL#8{J#6t0spztqe{
zI*Zvc_%(rV&1qjvsRk+4Q;zHokB;y<4#-inZoxKZO7^$ddMN*;l;<RM+mYT4*a|@7
z&In@R&<Bb-6Zj%Fbv%c$N$;Pg*tLXut<7YVFta7uzDH?~UzUh_0a`W5efiL0dauhg
zuQ8o{1}D+P-7oV)>jBwb)Fxj|a4nDCmT*vE!i4tJ;LHlsd~tSsjoKFYOcIic^tFVz
zlr#T+U}+(1!U^E|b|i=)(77fFZG&CkV(YORAfNDmB>2<sGU$zF?L>LfbaVmIW)3q{
zU^LZgL^ug5Abd%7SAc1N-C!iHTOsx%E(h8kbts61B(s`oBcWbvQi|;(vmyksV~Aak
zuoc;rCOhX1p6D{xEshF>H+MOkMMM!$Ux%n1EP2}5EjBwPS+d5#JZIjW(DTQ%rYXT0
zs%cK8S2<i!c(Kk-vBbXEV|BMAGj~}ZN1|v^7MFQl%*m7A<I|@ByYUS!EgQ;ypQB0=
zGL0x8?Dsj=8CLDFI)3w<oQ1dzuJEWVqEdzJjp)u=jOJZt`5L<qwRt4-HSn?|y9qKF
z?8d`Bw`d}`BJgHP@=_D6&)Nx3G<%_D00V|;3!9`=2O~PABDRjn`v98pOyl-Ew1zyf
zasTlRsAho9Y3JS|dbhp!-%0rr#{%5lV>OwOtps9GFuNo1VN7=%kY^25zodPnDMxK)
z#mO@jem8(PJ+2Zp7s%HXu{U7$C?|4j@qW)c)QUefui*~oHai?IH*%tJJgl*sb{Xe_
zAh+3glOC#=lma&rh}8tS)}y)waF5lI!s|XqMVIkgc(NwBc}ib<XwZPdoon>p<ltdJ
z2sgOUu1S)Fy6!SAE7Izi&UQlGy1<Qxoelpp{Qf+p>b%6W?XU7fMO@r-%^l}k0l6>m
z=L4ZF2>Xh<s>#YgVsqv)5=THTK}uw;ka<8JTjEd=m%hR$)>xc_2OYofA$4f<j*3OP
z`=9bd`YvZ%7Nei#m7&H)Xk?i~XA=fD4AZ#Dwaz83wG(EhMY&sI1~s~0kxwLK0enD@
zG`1t~Mk2;@Qa7bF&#=1Bcped6;&YJ4qG>?asY&)Eb=y#%KRwV#$ykBi5Qw5i??(LT
z%oJCX-4^&l67K6K0Xuc*{e=y(Aqi&}I@axzJFh8yB-;^XPk>2)gs==)hdI)yXs}Do
zGlQ*3?iuWv6no}~{=UR63u0bjz6y1!uxoSj+@VdE@d)6U{{9k^cENN>;l{hH3<GKz
zSk_d7KF9Wm?%tGEGNokzYLuN*ToKrIhuP*ygS(pK`nO`YoH_^T-{QCGWe&`Y!oA8f
zkF>YWai`2E>m8Q#@9=3$qwn<=`?OfkP&Lh0bsgVWL#gmf3RwWbQ1v5fDhUS~pG@f<
z)7Ve<bx(mDfvg2GM9d`O)ixXb+niqkiuiz|@||dny|WtElE?!`@fKMih@GS<d!=gB
zU5lAZ#6)3Na@=x_y8FOaf+M{naaq#<RwNJT5@#L~0Pz8RkI;Jw@csnwOTd>;J^QDD
zzYlzcCX(w*z%K$nLUVC{FVb(l-vju7KA=BO(zqo-Oe@6h<9^4X51Zx%v9rb2TQ1Qp
zwxlHV1u?Zv+U{dfAm3edoP8+Bps|BUop{t}nF9r-nNhqgc&u?w!BHfnFY?j8Ll!B)
z^eJVdWSCJ{0345^3cW9}x=~K_A-G&)A3J)&P!9yrG!erwm59W{VZkhwSaryfiijkD
z6^KQhKM_4*3?vd^0IL!D9_NK2sHXe1-d{92dN<zcljb=UjRMbyd0DlY^c0nt(Q^ql
z3rOps7Q_U?Q4^snQeYGjqQ;n-tTebpV2vferwMsNeO@AcM8<%eN1|55;zZd5T)>Qg
zG%nV#QF`QoAm$0x`ws9fcq0E+ctRn%r+H%08N4V62QBJEfEY4(+uC^vvx&e}9+iKE
z=b$m>?nl*bv6U%ATM(5*ib%{HaU=<?h)*KPSP)x+U?RzVV>ytHA~sRjj)VY0Sk$O7
zvO>$F4<#WHO$iSGB&M}!xI|5Iz4^cF^9~e<JF)?8PoK?vF_-C-q%RSvWIhv!youJT
z-=b6>(@}v$Yn)1`k0mY<2yIDsAf$@eZ{{5}nER2|0$qbkXd<NcB*AFhedES0<|^N4
zg2Wml*`+*==;RJhtp;Q&h_e>cw*X|0JoTs+1UVtu@pw}(P6B!eq$zH**%vv?881mf
zIw#$WP)l4Yk*aC7h4837fCkKqQWFw~t|Dj%i9(D7K{k;@dQTG5gfI}qMV)^+SNL4v
zi#f@T#^#!kETWqqNsjbT;Eisgkq!hlSGc#xwJo+1LF@>kX$)Y92-!683yCKk0(}Xo
z#hiACSepG3ElQCRl%?1OOn{ccPk|sI^R!#gWL|VQm@3F`@U*1?FtU<tAc$#3dB~!E
zb0EDRc{`z9fm11!NpYeXqn<S73Sw5Hb|k?x(arUFLChC2=dn_p19tAv(?x@w$Cgoo
zxGV{oC%JzQMue12q*5ZA$nvzQ`|4T{OpV^B;mH(&iG(B|rvW)?z~Epasuo1u+{1b<
zo6ppOXau$mK?-84Y3}l9YMTG~G@wTox%VJXoaJto5Y!<{cudwj*Bwg41W3l1{){5=
zD7nn8iNwTWW``_U2}y_XP*Eir<IV<m(v(VBjE#t?N2wWwIpLXe+%YwYKBAwF=-uyO
zvB!o{Pz)M?w+%t|7|o$pQ?e>F4$(J}CLRZ*62uk|1QeENXB1^Zt=8FjsPV7xJxNST
zvTKX7Exo3R(u2=;-AG70YU)V0B|$e)a$z6|i6gr&i+YSQL1m_7vxpi-VhZ}cEJ~;k
zA|fqOy)oSryU)=8xF<DiPbjCkzfT;rngzHEu+3)IBNB<2wVB-4B`j$!h^-2<FCiv4
zv#9w?!vpf@!Ess4uBHwJRX?SCQ<Ut2ip3b-O*y}6NYR$qe3|imS#gGYxg@BP_N>Xe
zcr!_aYXJKYBQ65eB+io83a>ZVehXkPh%S%}!KJ!sB>EOoO(sj~NFS=FPH3kdrP@sH
zm%%!sf2i;}qwJsM-doPb&+`9?ggUv(bMAm*DI!vhn+iluQzKj?gf>W@;*K>z%*e_b
z>11=yRgxehu>diKm<U{3pl&ySWwLh2QlQ!U;=%X7M|7em-Usgg4L_Avq`F{M^%yyT
zh|m(-m4sxU!Cg%hU1ny3y_Uh5Bl@AkoPcIdnhAU&5F?-#ge)NLXP1<8=3up=N<3<)
zh)JV+m?;ZK#)2oir-89Y$XIQNZGkuOlwNR(jsr3%e6ErA!T&(MFNkTAeJHwus2ZKW
zp++{w)T7>1*I5v|0+H0@H)!gmciwoWjEa~DgmUPCKopuVnbX>nNCc{*aYaI13Sw-&
z6JyR)C~SL5HmjSau$d%WjKsvl%wT5@t){f@uSHEhmV`2)YW0{cFUICh>=uQ47q&u7
zX>5e>m-z!(Fw@&Cy<@P>8SYISTKzkI>O?Llzssjv3g2DWK@nmakf(b~bR;1U*dT~4
zMNGlXBw=h(*f#<qlIC7KS=+EEQx#o6RuQqZ!?_c=_>fGrN0Y`bDSTFuKU5YprO-kW
z&?6REQ8fxh$RqKgb)h3-4rUIdg1Xm?+0FxI(tz&ucPtA31N!c!_Y&a!67atQKTPA*
z_yOSW0MGMAgvTd<zf03xZ34Um{2K6=S$JT+2k1}V?*V)Z^eLLgc>nZ0_&G`8cvqgU
z59lv0EjrB_fq85teUZ{UZ#suQRKd(7(nN$6#K*4rv`G7n5s^I1vl$VEK$?IY7tFLI
z%974wJinR-q^^+tiZleuT2i+nL0iJ()UTZjqye8t#Bl?xg~wJN0<vfnn>Z1;RN$2$
zl!hP$G6<AXO|7dp0e^xpZ_<oIDp2zPNd<{T1cg^c0|x5_=<pa61&!jPXrgW;P@m^^
zTjO*@%`K_`OpTs2=OW5Q#HU;|{HP*AMvSni;6X`32!xp=>IBzv1RoJj5EHRDCu#vE
z)4X^paB2~?2|$IIHggs4yX+B7SKM!jCn8>QiYFFx)ESdn5TzxnA7;QlNV=)_tH-P=
zntd{-{#``8ZX5=|D1?d#tME!Rbu%>t8Ig531z<%)sSu)RlEz|><Oz^sTw_29*_1TU
z&G^eS60}Foo36%DHDw{*o^0dHFb|wf1IR;``~o}`8oY0Ss)zDn1AHV=Bdo!jhzybF
zA~7~#lW(8Q+n<V7ZBjdZ8X*XD8Hp+qlT*)KG|&n9WUdMOCgHfONrJ`nJeUOt`4;jG
zxU2%QvSgJeiiBE4^x|y0wnR#Yo*R5Fk=l~aJp_qR;4%owqU6M>sAYU4XpK-F?FEsB
zzl5VHIZ#843n2bvZ4KZf+MR&D0wGTG+{378B86-~+Aw<?xmGlWF7p#WRG}B$>!pkM
zssY}mil^l18+#C=#Hywkrzrt15IQ1^gn1<D2{v!W9Ysc23xpasYNrDfO#?8c!7F0w
zi2=g>y@`kzNvvox4)hy;S7mcgsSuzN#7x7oAgT)Sj>hT(G!;anv5%ecw%8+m(wuRz
zcIg6;bqzp*_Au9oL?E;$TEJ_h5Jag!Hq8O0i;Kop0jwe_g-Qc)u0RQ*@(3$%#bQmo
zB)n~eU*m98Iv{5c@I(TTGdGb+m1Hpzbwtb;4!M~iikiITQPN;t0}dlSsye0>6jrpL
zlps@lQcx*FkTVKfFsmX$IJ9bHg;Oam(Kug|sIT&|`+GR0r2?3YxrhYmQB^@LHBmaU
zhXUAw7$mAxpdvyGf(l46C8+~JWt8T@??i-b_A44-)U0Jx$wT;PaYy|B?7dm6CP{YR
z_lpw|nfa}Et-ZRsySjTgGo)ukSsD=#Iig5O4>l~=mTcIN4H$k9VHkeZ1BVhmnxrhl
z@UtP?FkskzkvCYDA1vEcgS0G)%Rn5>baRH&%XD|`b+>QL%-H-8`Q7@e>eklm=5RQD
zfQzoa_xmy<BO^|n^FQbOk4~s15IxdDC>61vNMSQ$P!qHvN=@k$X{iW0<zOk0sd)gG
zF!q3Jo8xN8XZ7aqEcrZWHvWBT1I;PRfHju-biD~t99k!o^l4TB>oh?H{OQ^_Yg`T+
zvR)2m%=D}b*vIbwEP035ia;+Jlx%VT6cp*>w9x9VV&ySO7z+{=#w)7RwXwpsLMuI1
zkG)TTK$<I(Jxi}mm>40Jilm-YW!6yIh)AF1>9_*gJ8UCF{W`z@*md_fkP*k>F6Z!6
zJ;+Aa#yMvru`(nl_#g!F7#}fSh)VD(A})ixF(i@*$_Q#6^SRx>!v!I)jpwc=ctNq<
z^-tIFJ<i#HmJzd!wQWivnuxR!aSr1w&QzR~f=V@&(KM;eRYbjnG6Cv@N&*6E-H4AC
zl_Dlo1f^+xB51&R)cR~HyCB;`en)3x6^ZH;j22=~<@0&n?C&vAsAWQ#ggd>~+6MtY
z0$+AX8K>XxDpho(c*tJKIx`su$BATwc&alj&H^DPaHMe}X+@X;By<~4MHgs&2qZK_
zRS<NbXcT&$^^G1Yl!>saXxfNKL%DI@sU+xWzTRrM@%vc^j=u=#dD^&fz`|;RKH%~b
zIr`VK#fkG6y`*o*x<Nc<w6-BMZP5Ymr}nkN*+fzb9W<c;f+97F7(6BlQL7xR)F~&t
zbn{M|NRmiVP+p_TqkP{-_2yG$!o(PSp$WJr*rJNm6SR4F+)1ZAR<E&V+FKqqS0rsV
ze4E(mrI<68Ka(Fq`Lq;wL(ba$U2+Dm0p0+968Klh&BiS7Vs|R0001BWNkl<Z#qQ^C
zcHdKvUFnaXd`NCC_TBIIKwd!peKP6wCn?|eEV|F+)0bR!w-O*eZzj`Py3Y$|2ck?u
z3B(sBkk3x3tbxdfjQtGcBt<Ugx7qrXXM$xK<$dlXUnEEt2?ecL(npZ=9E0?Vq}He?
zAZKOuJVgYMhy;yjg;E7Bb=OM~r4&h}M>3HFlOjQ75H1D<3bYRSeW{W*)@b4R&x)?6
z(rHr`1;mm@30`s;I3_n%P+pPb!F=a7bFs6PGkwH^%jWfvqyugeq?Es_y5|qFUMc|{
z4Wi@$xQL+_U@p%BBtC8F_JrXR+7s(_(j%nq|DTKJJ`s~b)QaBvJ*fmr<uX^Z{!Ait
zK&*7Yo<u>ZZU7IBKayewF%%I{s0cxm*PRHlqj`Af8APK=s1VWq(dT=Lcz{RV=V?!`
z93)V|WAxMJB|G1ch%sMFcTS879U!BDq#oNl>+2~ZNg0xo$HyJ>^BEm+-r^;pQr8(s
z>x{RasU9LC0VNPZLL~)CNTPCPv;bD+td6KiCK1$n+XW=59GH=~p}%BfY7;>clCrZI
zb$(Zo)a|K6Fbe7BHi$we2r+{+k^)J^JU363Zr(TN<t$wyfCzSTF3kGf<XKiK;+0UH
zPS_f3Hbf-PNw)`5%8Od-hzi*u)}5k=N&_vTe0I`OzLqu#C7lPX0RjQFG4%CV5>9O-
z??gc7fJh8^-8BU$G4w*@sR^zRN%uZqRQIo)<c&Z}yZ(NBkcF*dq*ex{G$v+l!2~(5
zm>;(r#e?raA_^!MiK^QRYaIl{^sKS2*O2Nmq~e1;p&|}NQn`7f?y{m|dN3DpJr}Ly
z-<yPzh>*}pky5}XC<T(_07z8py}r&?Q>S|%h$h6C^_~KlGL`^nl5+VNl0*zTe_#2h
z^bN^Cv^Rp4l(~v=ozK5HhZ`dKdDqJ<zcWwNXShG6+vgEL&>`WvJ^e6muf?hL91ALW
zl-P~4Z#~w8k}0Egb<O(z>ofawjNSjIjlZ3)Er&IVq>2sR?YvKoSVDrN0xB6$2Ctga
zi=I8Mkk_3yxf<e8dF=krl7>ngqJ$8IczUmpGtO0_IP-maq8LKzY?M>I7PBr@ItNzI
z7+`wZnI*Xb=4?CFV^zxh%F5s$>Ccd#xuAV)VOX5#?!N)8f#~viCy<;%3WZXEL=IFY
z70^kOQp)EP3Y8*B2`VL$M9`WPyMV3UunRtMD#!Hjnmc_T4W%TMvbfW^@T_#qdnQ0$
z<BC>Gwjon~$0;QSDF<4T5+o_2gl_*!HbA=XOU%~`F=5u{yU%CbL;7tK)HI5u0zRw<
z9SEwMzbCHE&ekb}sKN%*MIt5v6OovR%_a**)|XMz!*+orq9|0#Ni9lsdQpIwL1fi^
zJ{F)QJqYTq&j9@>4U{6Nd|sI(L?1y?&VXJibWmBptB`$wQwiWBQ5%|5fHQV|6F!FD
z0+0+|6iL;T&)n;#%dq#DwSJW9)K)Q3;dR1G+}L-9wN6%$6G@O$U*o|$%sL{;gLfiv
zt$s4!Em~2mm%(%Ija?3O%}|A>=(5h<RDR<#c})rl9TNUY_wx`Q@@*m#B%f!W7S8vX
z{1KNwtQYTrx4WPJiX0jHkI31(KWzLUGvMF;eh=aW<v$>U(w_t{@}C2*0?*{7`%HfD
zWx^!>r~Hi;GN;^&WZNpdO;jt+&faJhDdr5X=a=pPwIMr;k6kyQoxvv|+6IYzKocZ9
zfzNa;J~yq17Pr)}?@6|$Xi^3tpSf=osm#ES#4fWbzSvx)7VttW1Ib$?270;$h9t?Q
z*{TDA9{3B9pfxJS93d_;$QFG?(hDTcva#kP@|-Aziev|2?I1=#AG6(hOzczlK+hBs
z45@69S<aeCx-)=N=jbPas8FY)jZmZvq_2b|LQ-@_k3yVaL+mgLO(+s7B!XM3Ey)=;
z@%f+S(~&>;o-hyXiSGMxh8Pc&IV<Ivk~677+TvvG&_wmX*=m89jeXr^vj#HA8EqoE
z3_3o%|0IKsF`%49^%TKm2RRt>+^@gmyB{LnbwFAHsm$4F>X|VSjVV*-@NE1#0IA0z
zLi7P`yNtc0NN3tRLi8OFpdH$G9<5FpjEMnh4M}lEhF0f1(*+meBjY}G7z5rZprH(c
z6-^RNs3TH9QFIi94~fqg4Ib&0EVYLekq}7MATejUlQh8uXcH0zFAA*$#3KzDukeHH
z5T+E#twj<-B8B{Ur93)j3JO(|fjmIFM06l1#7BZa(sWt83Y84n`=>XOHsti-C}<6y
zC#-j$9kaGt@7I8L88pklN<^YTf+rST=X1KiL5v8V6pJokq)?JFSm71AP(-hC)+zxV
zTeQ*y#)K|=7ne|tB1M-$5)VpL{@aGL6-6)sijCi4BSeMv1}R(O6wD05j<B-v1<Y)P
zF2Va3(FvVC^PZKJtQ$}iZK6(s*8yp}bI5`UiI`%pco2xFT*7Q}gCeCerH4w{(N!rS
zUJ*-$M2`w(*42&1M3n*9r|Q$JD^dzLOJ4ihZ|n(&4G|j+o(wRV0$LKP%)unr0(gw>
zDt2PdSZ&2&C7{>3CA1ViBFW%&KubhKk&;kr(5RdlX)0>B$3X-bGmSO_8PVsp7X?+d
z=os_;pQuj*(MqmjSg&&iv<g)6n8!AT%FU710WPCSIv`0=`1D91K#^KOhuG=ZNU(x!
z5{Td*szPZgBB<`U^O=4|K+8gy54pQ}Bg2Dp`oUv82=cA}eYy`ynidExdZUb4qQwP>
zctjPnItP|*d~ngT$EBc3s8vgG7LW6~$y*1)JzK8wVfz7>Riu=f>hxX;6A}^)D}7c`
zm?1e3$ywyVKlr^jf{O0p;M1u@q|$^K&}Z!fABn-_YBPB>plfhR6QUxen#xqvr}x!_
zqNQjK*n8}`=L};&qewv$9568WAQS<xk+f2%pmOFb<mP~)36fCBXZ>$LwW&j}6v;=D
zQ7BQwc8gUc&;1K0Q9Mf&IjVU2sNFpVEP^7XhVpdJ8G24=laOqlV6P)Zgcg(S3LmkU
zd~UT7%G3q>6iRhKwts*pJDuo-L?IAEJ|mI^9TUn4u6LG=)m1B$3DvI0rEpQ<pXh@q
zl1-@S9tyrG(^UyB6RJssVlxr6)2A^9SoLWjD4)Te+%z+#b#NCT<Y&-|7`qCHz$vy8
zyJQ0G1L`4Nmd~ab!Rvss1w@6dVHH4`fEGncd2c@ugDN80wb;`+5{w1^LVgPy`o&{n
zMB*+BeUI@|_xhBv5n}4GTA$xhm^V4V<%&cMDp&p#3SH-Cm5`Tul8}RkWQrZUKDC*J
zY|uRbd?TR%h0X!8DxreHxzj&CryR29DptWD)?>AKWKQ8U2$wXWSO*01`ixqLqMo1w
z*XuWKRBCm+`!D)FK9fI2@`u$;T!7C3e;xS0$ROw6;fD|0{Xvxz;P3OvP&BMLv7Rf-
z>@)en7mt}Il-uN|e3l5d6jarfAwE{dK4DUOs6mThpPSD+;f5LbQ;P|c#}3PYnPsC(
zZF$B`L_(jFXO{FQ?3Tj25wrU}Ryc}c#b}<u3GtcqGeR6DluMX}A#KHMP(Dvn5db`P
zmU3C-2+Sf7T_g<y#wT<dF>?{+6RDpwvex&NK|=Y2T1Dt3l;36Q5-K^YbJ$TrxrjXX
zSjgEyYbQ1W#oK&gv7)LYRIpkmQfp|&5%C#ZrtdI5k-WkU6d?&dmbk7oUeKg*Lc2h5
z(1CR}l3Ivm1m_8-4x%BA0oQJnNDH#vz1tUgmQO`K&Ynn__pnN+hxWw!T1&PDxyy2E
zswZR{ttBFoP<_F)*ZIhk-`8eoTsJ;?*+5oKFqJ2@<V+!xP+k$W+Mp#+gnkZE*pr+B
z5p|cG9YQXH>_t)=(2KRR8>v<bX*Ho|apRO7<>z=u%@*+JK4}2oV^wOxT!<|>gVzCr
zoHYBe2ys<W6A@iwL+q#-GN3Cyp-$G%-8C@;))`@@kqnj%vqPaak=j8d`ORlN<YGIc
z8ueKw%j!Tf?ZcUCU_i5g*b2%DZlI9R(sKo26cO)H_sHe)V@EX&&=j;=i69-<c1sS1
zm_*np5e*t>BRb|RCzU`<&Sufhp!(S<m1S3cn!naL%bH-<4$$VFJ4*yN%pkFz32{52
znkVRzc_K85R0_6)>{Lgj&LBD?(2EFO5l+iO1KuP|6G1l=E}>#Yy_F+Osp&E-dlB&^
z;q=`s8TA#Z+2ZCplgyFr90#h3$TS0q>Xg5(qSb=eD@<HccG+M_Cwwmh1F^`NaK+4F
zZzWVC&%q_VDGE1ESm&|Nt)n6`ODI=tbi9waCO7O<>kPeC@Pn9v3ulPEND?qUq1p&R
z@O^_XBPm71<)CBokXD+|PvE-Xo#cB4l|l5USJ6adp7(bgaps;Fv@H_%I0+G4B9$Iv
zA&wNuS!UCU+6A;UlubefgYWGzQwi-?^!foE8p`^ZZ5Jp)pJmZdmYNVO&KuHpMEo(s
z7GTJ{{TKP2)AN6aBkz&6W;=mygjg&Xqp)FsJJU}|7@JU~N3|d>m%J+(&{TmUWX|Tt
z#bXu;?F`NNZZ7Myn$$I@MIt(t8!48WWIbx>QPT%|BFmD(IV2RUN<ohz;`IjKcuRkn
zfmL6>#ygu1e?<ljHmRGM?spX#<O*z1s%>$yute9QH#tCKN#g<6Qj8V8h}huhor0cc
zjcZ6#d5WIL)CcUJ)ogTaOFVbk_n2wgVbKP}9kP20LZ8rYq4Xuqnf6?_9)lto!)k*f
zU|F>kVaoW-_nGZAsfG9$zx#wq4Cr}6H5PZKo-b&f5aWv8={l>Zd%B|{d56~?<I{ha
z>;8akeU*=NkJ(g+t;1SQ&pB$BfvqB7ZA1r)vj%S!I<ypBrCMww(JGWz=v6}Pl1tQ2
zF7P4bb@0rR`Hv}&t*uviud%pr!q&z)>zjzw5KT$jFKJCn;il}aJStTzd`l$-zSOvd
zM)@_TKiU44bx@(A)d?kS4#rj>9`s5P#=v?HR~^WDdR`>dqO;dFxOClWtz$RO*9{RN
z4)PfKr}kRr)>chG-&xyOn^~_hpSwPDC_j))BcWT3J12-ssEu`DR$UcNLGDT76Kq+c
z?SQ^jOR;c|l=wH;7%^?1)@s6*Lb*gZ%sMs=R`g22#e()!R|-)gVauWYg8s?I@3&(<
zvp<fnP~PUE3+S+O2G{kYa=>0?asDE=b4~+<pXSY@&R;Yg*y&|s4m-*VCtwqHm1!4r
znb4&oNamLhWy%51)45ehPoaFwbVeo-X6Lxq@`8I~!XyfDp6&PcNp}AXl+zFOsM71C
z8T&{hzf<YfZxe{8>r8Otglc@|m;OwC2<6iPcopz6@b3Ws3wg7A8~D3qKG*eq?*soC
zxk20aofq*l`Aj~OPetxUWNz>myU6-8#n4>Jj*lEpn4B3U4e0v`wa_fDJnujXxEx8#
zM;jPy$YDaQG@<_<7mTNLTTF*fD7OT_Gww%ZD5%v}_|2xEJ$97y8dH+Hv=(eQV`qp6
z{#*u!!b;%>3A1v!>A~~LG-O^U14Ph`CTyRu;~m=XacpAdDoldj0_Ym-3=*kPC?h$G
zX>o-+x}^yBx#SIgAIvIXp1YA9xgkqV9)=a*@3UXtXaAn3T3+FBXGQNgV7#T?$()~R
ztWdQk)gU_tVnOXoR)dILd9>@Xy5f;lgT0@!QRs39a#@2$T`8*?Dtwi}+}r>x2MpRB
zFpYnKrhK2GyzrhV375M)alE-F^1Cq$VD}z7p+)>Q_nx-in9)0zvm@L2Ro=B^L-o`U
zoG|ebIm`fCs3NHr+$dt|NLU&CUO-PHsxh>?kSQDtekVemFs%odu&WYxQK9REo?4m<
z5!)uzsWY!gt;0?Q?KI2J0gb~RJ#G96Wi0i@NU>}wH}~?A?Bw>YaPK6cRt<W;d$!gg
z`;T5LbNR(ZgRKR1IbjzHy(+1%CUosFhY>SN#99;1bz@eVFm%{?#59)XvPZRv>h!rR
z&86;HPaCK>hUJ`b5s~Fz<hKts?OcedrT-XZkq~2mnGo9^?Mp3%KjK_awA(-vDc_Hz
z+K|SF^iIGu6@JH(`W{&Z6gB!JB29t2*p0g%P;J1hdeplhOHZ+|v=<X51f(fhUJU5k
zQQq1Z%YHzgc#2t{_H~Dz1?-`DhF;Ue^9fVOJa=0Wsn6Pf6DD!J-Zu4lKp%l^Ebd&3
zJ^>m_+R<I9yCw_+dLEGFl2H}VzR&WqAZ<cT1A0$^(}ZcqLyD(YPV+lhT;qe1tV>qA
z9NibxDq?p9bt@tZ!N0r?h>ckG5`}a0>NQ}H8z`wZA`46NdcZDP%=ZH34wy!0ch-RQ
zvqTg20U~ieVe%MQ^d#x_|H|NZ(}PBn1d5}CseuxtF~rfD$I?<?1XDX?kw___TY<vj
z&Zpc+EhT~us5(<PKtL@E)F^}g+JkLc^oc<~HNfgI#{n8c8uoFoJh~@t^KzZAt+0|l
z%SMr!M0y#p0bP%9CnM%fN#kyES><uG<rW7+p8?)|k6Fzb9v-mgrVJctN}64ZbNk$n
zT|=VPyIgk#t-HdFO+Zz@#H}Wwrw42eBX%B8&79uO6615TtNCN^(4)=?)D|_51f57T
zLsME-rN_7%Ts^bVWkMYWV$-AjksQ=jX91fi>^#6q5mZU@xwY;O=!0%gTv&hhnC-fy
z2nB9S(d+<ql^aa~P*YyggnCH5X;buhz;C+=9IUe!E%{7!<g;7=Ug7<@qFD(_<_xz1
zdbmQ<9y`pkqQ2;{s=>7DEargj`3yolG%lfe)g$ea{=Es4vv?^N8~1<S3c_K*?4&_Y
z=TxV2bd$?<gZ4dE<yrj5Pz*p%6}nl`8vzVCt}2ppSoaZE9|KnuX_W0X)i{bX&#Z*H
z7m=mluAFt&pXctPplzp&N3o-?*x~4s!!}dK2UEt6*}s2}e^Z05KgV0@68Dbs{p~KW
z$N~G<Po%?~)o%7Zl7`ItnjkGEL}UQsRBq0+qfkvG){1Zym~U}-Z^(RI<@VHut$zSn
z+leE(^xD6~Kdeg{-%{9HTv`YEVnCk+SnY85a*GX9#?2irEF9KtG4CztdBM9O%Mf#Q
z-%yeI3B8KM;(-od^cBW=dQK5`vhA3>CfR^qX;gp1?>GfDTgv+hyAs@mO)z$?V+-QN
zM;}fT_7X7!^kN;rNhr~@mm;zZ=-UB(*n$6-*5AwLV5&9g91l@(nd3-wecD$O;x}z1
zAr2#E8PLl<GZ)ZHj}#@#>s_<VBSr8YV?D+I7?LdM$A)rQP!D^wyC91K`(8v=E`y9)
zmgvV!PuKo{#f5+}4fdpEaFp+Fd%0PpX*~7?{1z3+Hk<#S`73tL(4FJvSQDP;Z`D+C
zz}~Q+^?#NBNzt9(3PMY9kWdSi1C@2TF^7rJc+4^)4G?o;gsvi-H+WajmbuzYEjy4G
zl{u}qa<G;PC>8PORo+{X_0se7S2-0;H~>~9@uJ5*_Kbo4@!Hl_U4=wKRS`2+gp|(a
z;2m}Aw~3S|dECvGLQ1kOKdS-YXYxZJpCrJ03HYnPe@D(PdcXVkzsAEv)c+6g?*Z4y
zor?bk@c;4YMF~HX&*U@tK8mBb7m+C-1??-(`n)FWE8K2G<}Kw1Px@TIe3%XA*q1X5
ztVhM*uV_LAdgie=&ic%RiT9Wr5t*K_fAN^@kdSzZTia)Wye&mBM_v}({5*G$&T+52
zz@7K+aqf=A?Rre(QOlCmOKPK}aZ?J~t2x!x){3+nNppw2F=6uDr8N^KF`+()$TWkp
ztO7I#?CmY-dm)$|j^+Wig1AS{+`ULHFKq&T3(WEgH>R4#Ea|uJbM1UYYKJ;_5x{}0
z|7#xb(zU?w|Ku-!v|upR`29Z1gMdC6vf2;mvZjg^jm>3{ksgcDF30;Cms-jLM>*1j
z%3~G~O+$IxVW%baRfWVrc^g0wYgoM$k%hzF=}PO2VfuPRmLBdr!IAt!m5;F}cAvB-
zCQObU>L4QR`@FtaGxQaU{w{}SW<9OwjS}{PCC-1G-<*@HDQY$VUVxVJgNV$t^N@VP
z)Pk>arg*Bd<7I1&>KXj5VRaFR0qqj)z*F4HfL*&6NJ&tI<V=ZKOL6CsVck^tT}$eR
z%qe%m6C|OepuG;-N8<5>$tl3Ave7*Q+MnRpg2gR4P1}_vu-8(YW>553*<FsthIn(y
z&>P&Y!^|DZ8~g~gi^w!k-u9SUlPZJXvxG~6#2S4wze7I)S^2JEn#En&c-DK}xQ$_Y
z{fM1=7I$-x_w8%Ea{zj=q_{d~_!v;uqrJVyfezT=F?)j(dKV+6_L%!E>L^g&O_-&k
zI=5hOFQCdX)18E#0*ykQ1Y~ZBqk`~~!S5u@vB12~e5a+D%^6M&ZeVG~C4L*wo?_}T
zALcm?V~rcO6t^|e7@BQI|8PnF+lKmTBAgkVF9?@2_n}>C)K-9*!`_-OITJAG*4tEw
z+iS4XVQ*!S+gAqP^O#u#uc@yZ;$=bNisFXH-dWNguNdtqQqku)CDbaB_7i$l;co$W
zjG19yUD7|=VR|xRmdK9V$<OkC9~zn?MRI{qC+soAN=R>HaHOu5SbxmUGS7=2N)B|?
znlLQzI}tfp(f^$ZlevdG5m_3hucv455(;);aW6%Rc|@Lg=A9)ICev>J*A{nu4I*`O
zx9S>ONraW5*;N^|EF)1I)m=J2K!{aIv)5%m7YV)cn1!J}pU}%pMCf{v)C#U@Fz*Md
zj}*T5wDVVyT93Ia#GYk&CFE?fQ;<KP(;Y6%mh_sJxVaL1GNji*bV+mD;Qgz-JFBSM
z6+Js;d$GldE@^r~8a0&ZfF16$e**A6ugpH=N_n2Uqw^f7eeRwQ=+$j5u8!GGBWBHC
z<9}G6X9DbB<J~(Bdv{6S<&tYz2yt}4o^@2sIgZ9=qhXjswjgfhHKxR3jtuesIS%a&
zF7DrFf4-pq1m0D1Uvkr<N{A!R;7qzQMT(yCrY6~f`qhLAf&Q(HJy96^MSz;2KW4jG
zG1}W@YAXDpW?lo#8D0udX~OhNd}DntOYVbF1IHWGJak0fC*9xrJjk@%0aQgD&vECb
zB`#JBuQ-OoK1bW2Zz|+o$;x_)D@W|C03u={T*(dT!ohl60~VJX%8Lnoc%S{Z&YpwQ
zz@r29ID@G85h;PKFY^8eN9^CyC}Z$10ZbYDLv|OAUUP}NJDXK5Ck!rlVi`#{72(5%
z@?}8}&U3g`Q?<9aG!2jN9w$^$(7qax>C@~rkG&PpbA@P&d+E`?U**QC&$3pebcYMu
zI~<)WX@(9v|4shP%`-09f0yg!J<c6|nQz_yv;5;=adS=3nKxR^6|xi1cN1zVu;Vd1
zmS8H{qNZOcQW;Q9#8)a;b}bd^s$?$YTR*{X9h2YxiTd=aEjR7N?K~GB#UJHcCw-R9
zd%U_cXV`I}_iy!B`7zUTOLhj|<@MQj_}pT~w!6>H?wILuNpoazqYU81N)bj4c9!?D
z+8;76HBGs~ghRHr6e<MNQgR0NLZNF-vjb*vhK+UT;fP!z^{9?rbm!Bl4%Dc=p}mkW
zO9vmpsonC5LUcr?9(&7Ce2}nngFmlwMG^sh7?G0*vB-cs-S3YysR9)h?Vz(C5!BGo
zo~tQ8a+DtqneS`DT|>aqUiU~`(YrHXu`qZyV>AZz4u>yI**-pDbi1J#U*P6mK{#k}
z)0X1!gk81G@y<5KO-T~~49&2he#KLroUrp=M>%5*)9VqLN96XC&iALtF;M?w{;YhH
zA5#XEzRd6Lob^GERo~!Cdlu)v%y0QG^Sed2hK}<bneS!l3lq9b?6f(DEyZn6ONhy1
z7M5l!A_Gq;17-su))3bKwHCOnP)?Hy@Oz0g{XFm8xybEfVAk22&n++U;Z0yx&=zD2
z>>-J`mB-vqnA);>B_bCsjd9pJj_Q_Wb&U;bRYfW*>Z=i1MC6ttMZuN@VN)-?P@j*_
z<d2VhlFeogfqx755)T8q_ksU{4CMZYz<)z-zV;%3!DsTvM}CeR4EQ#g*Y`8|V=iy<
zX8I}q!b}skHE~O$wiD$t5g(#R$XR**iXmK%6vrO@-M_^zJu&+DCU1s6&!1}q8EV{l
z1w}6Xe3%z08{+O|P25e`;}yMcPnbML`>>X;^0PBV8Vj*HX0$tG6$hNC5zDHkYS(<c
zD|(|9{p*$#_BmMn4FBj*6P7z1-~EUDrI|wa3~o!IEa+sY_jAd2=8sgQvLw8sa8<_*
zl_E0p6!$##J8$yl3mMkG%CFCVhOhP^R0=f^q(~%bC`T2o-(ngdT8|l6+#3Kfm*#m(
zT%P0Z;xad92ke=*`H}6KDthdpqx{yJy!iq=Y`+Lh8ZvNFbU*P9=WL%@yugj#faO*}
zv{S~V!Cf!}1*R3e?NjRu+}^HOjatfj#bDiV$_12zVg_cF@Y0p@w=HTWsG^{LIig*_
z+;1t~OZZU__ZkQReb7?9^(JqA-wmFRu_vx<?1{%RB)-beP8DgKQ2iCf&Jgd$9F`?c
zYI-gqDIzkZf8JAE?Xg<C%Dc-y!9PDDNBQrlN4WOh<jwY{`ROR=zJ|UK26-K)egY+^
zfg&bFvVEFsn$!!qv^y$f38EsY0f`#71*r@acN)s?uo2J|>_Aap>za<#Aaz2eZZAKy
z2LwAXG}kq;jL7|(-dk_-=0isIBpdvtCKNjybDQ&JgY6}x4=F0xO$Y;~YMc98(DV1$
z@1HQXfD%G4H%7Gch^isEfGj~Z7Pczt&uPLSAPdloM9>k_fUXlk8REVWMiD#hp0$qn
zaS!(fNLtam^E15p?GZD5l^gahJH?WruLV$7001BWNkl<Za_EZ@6AYnNq%@#4SNL$0
zh-${TKV^I+pjDu})lhs#LG;L62m_5DdCZl9dZdVLpVcy;UFOh3Y5b^9eN~aPB{V&1
z?J>zwu%v1=O7~ct*SOJ=>h7HJZzt5s;?Eh}zF=L>Mh$(BoiFMAUM_7l!Q##(R7^<6
z|C-8@a81$n6FNkq$^fA`a1_7yr2Y3MZ@NFvpK}U5)VK=@iin;mSPG0aemg-hP*%8W
z2Dclq2aeIVAQ+3g90?Z;vF*{cnr2^794g|CK8u4+U%x6K7x}39GT*pib7PnVT_DP1
zr}sHGj;MixB*gxh`K5yPJQVluuzR#%;0$5U;4cZfHn?4b+i?`PS8Tue3;f3)0{j$z
zVJgHefq})Jb5N_CJ+!&M#SqRFtX>W1SwP+V8vn`no~Ap_lCSc!$BMLV@Iy_zA9LpJ
zDxqp2jf7YVY-!x6puL#TjmJ(Ss#YXh5%vuAc0!9K^gMQ+&?`kWAWhdOEo7fpQa~SO
zaKpT{K7Tb*dSv;i9?8`cZqLwOOxQLM7YdoiQyyhFWY_I;WFui*5UzVn3YZA$z9uaD
ztbD$1JKSNq6_G(iZ8=P}Vz7LhFMc>@gplG#TpbvEd7i`Ki+uB<fd7DN_3!b;JAIbR
zf0_T|-p}x>Eg1~l2SgKQzs^_RThbp3sV}HrPG2mBnouZezsGUkVbzM>E^uCxn%w<M
z`4K0JALH9MmQ?nL?dl!g@QL2y(dJqSRg{FQg6}6JdCXB>($)No&l*kG)r2d#d>$8a
zh0&Z`jWgExZH@E<qgE6b3mV<0SzO`0gO(~Cv3uELb`0rcmz&@H3SXa-GxPSzI(Vw!
zvCPv+26YRvFISR%MMDSpUg7raJznZ*+M%U>T~WFoi#i~)KvmXA5$FvaRdT3HlJ&#U
zkokKSulp=t^O)-%b8yJsKmR%Y%lDtOwkKJiSKZHm45~b8k0^2`=q|7_v`7DdzkEj%
zwlrZ^NJFR=hiol(xKmiX-R5vuQ^^tME_&=aOFG`?qi=YsP_o(?F@4?BTg~Yoev=<N
zNgMf42EScUzmY3E?ER<NYu#DD(1dX}mhtBM79s_$n==}m=iaLY?uy6U|A3c&|JV5H
zW}s7mt?p-amkT#1{9-I<%?`)h;zF;*R_k-lkjhLP-}DRG%Moo_i~==?nAV|^&OGbU
zpvRH0Q&Jy&p6`6;4ZeGm3}Ag~*?+&t={hUP@3jP)FY}H1H~5KS&ZxM}rC~+w{}lh!
zp5u5hWlQd}TO6~~-r}-<k5`BLoY<Fmuf53qrlF^Wa8<)7qHBo98r}4m?NzLHJa&1H
za|4GJz=KH4*_k9zf%4EH@BJ-)>9H!4uky1~gC7fVOOdvrT#!rVSL?l~h<!z@4DHo~
zo;&RAKyhosZs`T|QA6*o36uIw-fX_gPm7Q$OS}{jBC>?uLU3D}I8wxMLfVpWHG#^3
z9&;?D5rl!Iy^zpTk9pstr$QQRb8=N+yQXYcR60-{6f`FdrPZXtkoolj#|j(naw+XH
z?TwkY+uWU<u)lbR>rzoy1LpA@N4<!0V`f`@>MuB|`3e0GZ*k>qAq`8K*E{?AUQ6}F
z^Zh9+A;2UBwEYgBE6;JSca0C^C2ki}hOX(FuHE8N`FnhRS4jFR{3bugf1L&_7vuod
z46;x7ll;Znl7<OiS^QoiSVin7$`zz#CKOl09`#kh^#v(4z9*<fpL*ue2Od=^(&d;N
zLT_o(P!sFR-2UF5;p=Z{+BtCd7hf>=@jvob`j`01A|hKQ&2~dInloB<^fo}I35=*6
zjk`8xel1ZP2<o5JSS6%^ChU4xDWs}sUe0!&x!Y2HAT;BG)f)-3^4Pnc;++@uBmaoV
zzfQ&y|5xA}-{-OXz{<0tp)Vrqk{*D61pK$auR<pB`Aj|$Q9KNu{~(fQe;xR7;C}=D
z)$j9IK9fI0nK0SzvHZLuZ40rVvoSdZ1vFuhOBU_CMc(}de&xAz$FK17mo(uGO&H|{
zYI>1S>+AM(=B%u9CZ7KNukp7Z%^o~!tc9ihyut4cneScV)?m!S8<yfJ<BYzI6h*~q
zwaxMJi@f#SZH{MTW8)Cum-vNOOTsln+%_y<PNYJkbna4r+C(7&-eGS#>@DAcATN@=
z`^dPZp!u9eUD7P~Ebf{@bdS|wm!rzk78<`7(M3S3h#VRGaY1{eNOxBB<Syq|M~n|5
z_OPLP`{QT4rXIOUqesrZ=>hs@_|=#HG{5!A7OU|NcV5!8qmrlxO#7A)D`utuaUe@Z
z&5FT+M=x9Kam?PZt|@Lh?BSaGQsZ_(be=!C7wO#uUjjwUV8Z%YKCN=Yk0npp6R}&n
z&%GxmO!_7Dj|y&kpQG~^xKs9NyryMEj}^U2kz$`^v&X^FkMYg#7PJfCo_b0d<}dTL
zmkQ!_q1`EHUsEJ)@LLwYqY0(O?c^Ya7mrzZ%4tLvHN}n0cO8^q1R^c^wx_(&@ssLE
z+c(gQ$ULBr6m}{2eT5r$jrSG_wR-fk9&^%CzWt=k@|3}d^FaR{Uc2xC*Ir&QC|mR;
zg^V;!Pl583tD1g1Vrv>PK6CF%?3$QF5UmK7eD(%S*tWziA=rSPH`v>b;z&WzxIKg4
zN@(xU_Z!N0z$_GQ$QkP}<*cRMmZJA{{^V;jw#PNL1f?t5wqza~dU456ZgalxDf@vK
z0@W>t9CcjG2=%za{fI`arYUwfDN7nG%^v6-kC=k6t?^ryrU*#msOAm*<($fCQc=*3
zx~=6`l!uPqy+{lZiJ5axVuQWqJAgjZdxWim_Ihrtm**(+GAGlO35P&&x23rGBKz-`
z`PvnOf34H!WkAmizGsLpY21#%?`uL4a|YHlXD}-*_@NMc6;2J9B}+T4a0d|XuILp3
zUG!M2wwNtn<L&R9=lFzNVp)-OQvo#Z^7-<+Tzf5`cLiSa$jdFJZ0V;Z)hw5BiYI4i
z#|C>VP#(<b|N7VX+YdP-6DC!+CwB!k%pKFqMI?sIH|&O)FR+*g(yf;AJ0CaWrrZBt
zc=TS1dR*dPGx+lrs|y))*BxhLEG+F#3H^jOU2iQHNXbzgMa;@0wWqk1P^m@TbyV+r
z%)RcM^1s5*U(kd%gg6#r6|wV3s`J`ZmEiiB>cuvZm?Srh8IqfKE>mutW=s6&JhyhX
zIrf&OFW^@UB%-yUsST?*WoLMoi(50cihJx?i}N!E%Om#e5j&}&CnYPl%ORiR9X;kK
z0N#+^kLbfc!EfLDDSo}~-nAivoJ-)Y0ykl@UC{hU)^E)#2DXJLyBv&rERE2Vx!F<C
z)X02{yIFBuzrly^U*Y}xzr~*}|6hJ`*JB1ktbFD?UTw_9aBYi3<WLjFPqBwanM2Vm
zbS}>op&jORXQu`~u!L7EZV!ClVGjiku5fQ!Fdr0zEuR}M9qn-A-LLS?qi(II?zv}V
z-G3$kkTRfnj!ciR22X{qsgcLXJG^pr&enE7ms7^Gmfk^wi3xSRz@OK+QA=?Wk-HV{
zLrc&JzvD4&M2-&G{pP#8_SUo3e9C^$1@b;E$^4fsIhb#{->skP<h^B7T+g;IO5?5p
zLU4C?X<UPZ1PCNpaCdhN5FkiHf;+(-f?IHRf<p&)hsGLS@BJtH-gDj=<GzpYjW-4#
z7^`Yk&6+jkx2k8&`BF$TjzQ&`^?laKZ*#nA<tT^Bvtu>6wv6GS3$l!zz7~lw!B`?p
z<TbEDC1tH(GS6?pCg1g{=_`?=-xZr|BRdUq3@Yw|ediO4>iIP}%Pr&nUIP^YQJkxH
zA_3>}kmxTclJ}{sPq;H@+rkizIZprBq)PUeiP%0QFVT-WyO4rb;DFTVO*KX&>b$XE
zsY(^owd(qJgdXp-Zx|gsY&IVA2Jy4P^2zvgmc9@boj6lJ03W}u2zOV5`;dMKA3jj*
z49U2AVEK#YM1Ei2MQ=Ee=#|j&`<@r*@AkQzN4=yMUz>i^UbYV{Z%5O55KNb&&F=iN
z>O%HB=G-TKy+|tHRPc5SlF9cQEx}3Z0x4u|&vk_dw2j;WyD2}-ci%Kw4YaTF5bF&`
zH+;SAX1K(@mdWK6@qG(sl;>V|X?-S@iIa6pN0n2$i2QYearC}N1=uSkxi%H59CjT%
zwzJV?Bw@=8%LO!Pa@9<YP1c;udpN5}Ep3<(Ow&YEt2M37|Lp@;Y6X)+Z1Sjov_mp}
zgW0J`Yzq3LV_M#?K-j&FF#zJ>ib<cKz4I-Euwtf&1+<+}d0H-!?3Ym=2)p{+vShe5
zH`X=+nMp2Yn|E#yN_zErVaMMy<D@TZT|2hI(POO7AYc=iQK{oMhR^wp4R!??i8^~R
z6rFz5OO7n5uI0OXwj@L+H=;Cn6Ic}ChNjZFT+7$5JF<35d}Gnk)IEu};&tYfb$N0=
z2T8hIQ!FYEJf78z3M`hf$NhMfd=g{HAq&DTN}!)D4|S&hL$5kt%_mC5txjIwgD?&b
z%RP1*z5ngKS6+KJ@(?%>fb+l)UORe(rubss(g9XG5FSoCPfquot{<v~Njp6R9-k6?
z0|Io;NjmRk*kE==%0(5H53BgFc_bl%*(fzo-2KH#2NgXW0fUUPk3K;?Vf!IZjAGA(
zKBTJUn1I(~ys_ZrGCj2cIc}KG@;nJQ>C0EuvJR)C0{smE=?(Uh8VCyecxnY`2`ivG
zGpTy4t6TVhxA7%<$fd#kWFnUe6fXGjvA!Hej|il*|FS%|>--3`Ic?TXEl73_J!RU}
znql*SOcAJOOgY;wVZ$=+vD=u$`tHBO0~`^kNqFM;=`5^l5oU?KY#@v`Uy~}%{+x<I
z!gR?DtNGeM1rt5FzkajcanKb1XkJqH1;>BOEiyS?+t%OyWS>p`Z2%k5X+Wtb6RzDR
zv3jVp33?i9T|5=4^md~4T<P!&8bRI}=ZEDV>}v8pWVYHfK;^G;!lTwsncZoIo4Ou@
zYRhMNBpz=XkL!Q5G4!l}wrl9q2_7aUj&~|L5hLE_Xi2`&HBGHj;J6qW6<OuwR;OFc
zUCOH7$hIZBR_pY2=<4x+c=v+pGtU=rvbHn}eZ^yf%9jqY*ESR?*!NCmHFdGt+5$N&
zFjMKfzbU){-pOzKn#HTj$0LVrx6&Kn?*4c%@TDu443|xe$UwTJ1|tdnE?QBU&5oJ2
z60|+&tR%Y*3&=O+FzYl!*@cCYWJr8EaqhYPm}p;|%&WJ$@!%Ad;WMH)@qRK!guL*a
zUvn3p9Nk#~7V+L8{>goC9a(yjC{dN9ih{m~=+Z*@?jo-C^-}0NitoXeGGtr$kBK~u
z7AGh}A;`C@pfd-R<mD_b65T~@J|ofb_zpV76h&{1O8S}0vK{}JD|c_`RV{<G8QM;b
zhEks5Yo4bcfC2hacYu@n`uutoIk(I-i}GcvdTjCv6o2ieC_g3P%55o#Zpjo^8Ns<n
z?M%Q>D4^^2F<SQ@w53YxsZ1K8l*>2!l1|_%1@`Tbu$6#w{&shBup{hF!Sc5li|Qm%
z1KOr(k4faI)|@-XuFd0%v2l&<QOgHqq@&NHKFpv|KF&m3u8EL&)>GXktBg?D=-}k!
z5K<OcAh#OAbq{TS(7AjR8N&=p%=*~&{o&Y}_Bjybm)(877(K93aoA*#Rs3pJ8y3+V
zIHj=ufYt8xpnZc<dUAf(w4~QMrEc$~iN^#gbQNq;hZ*Q5h?kX!$qNq0oz`9~pn1Cb
z58S2T>mDO*+?a}swq9b-CFqNe_QpPxgNmad6ju($G|AtgN;zkoy33!(Qrg+L$ap9*
zb#!v<eM0Z)M+YOepmVp^C_ZFt19ix?hNkcAcQ5+4ji-scYJjeH3mPjDM$YZ|{vY!0
zY}%4hKlh$6zxyUUc@%^@Ee;tbj~OD5DS*9y#JXHQIR0`V`%kq<p4ZU_(0Thj4T6&i
z#UTM4<f+zUXA-pP;kUCk{!z=Lqpfrn7O-uK;qWPB>l0%&8k#s83Qkl)(AzF!%Iy)O
zQ67uFmm{h%s^D}kHLwtgu^Jc2*OV_olejo&rebVguy}O(`gjaX2SFeAkFJkev}bF_
zD92q^1dr^CTeKkWlP>);G<J%cp)-~3Z)R&}3;m%!6dQ7rJHo;pW5%AD-Q8H`Q5+$g
za61ke47#C49m)#^p2aLI>ue_b%Pa~McYw#O6dV1917)Djj%EFVxooiV@K4_5o~@lv
z8ZE0EblEY|C1Bj)45atdA!T;$f#X@gL5(Fl-Pu5c$CF##z3$e-PqJJ{-8WNCOoPh%
zN2J1rHWDYL6{NZ<l~C+8=d%spBel5p3-LL<=9jB-?)slffn%cxE{0y|%GkuN8+KF0
zfnY_&Z?dLlu|~MR1T;&jZIMmiq;JkJ8jcl?JCYyt71aSAX-0*OF+)kLZtx5wu83P}
zATJs;JH`cV{i1!sIJYlqD>pS1h;=eDEq;g%?0<@Ji*fY!{(h95AKfzAq|vPE+(={a
za_S4RiW^35KHc(|(tDfW^5C(i=rIk#+5Na<6J<7_N$i&Xj!pmKc-2byH>Yst90o<@
z+}Rpfj`_EfnGNX<<4zTxt2I4SIqvtFR-%kO{ATIz$kB@nY;upT(*<$R4kC;JLuyLR
zAM0ggepb$AjuR%jf&eR`@cT>f`v$5nUWAtM*80Vm9Qb)Jm9mLlSs6U`cBkWXMBSeV
zjS=({eb|PxF;m3F`eMP`wU?^qAJjBW%9frd+(A)4%%6LL4(EoVBlN3yLGY{v#eU;#
zFY%~IfnucQZEGN-z@rFGnY+%(bXlQxTj2bdJEc11Nfk6WbRRB`ufLwE^b5);cjIxN
zrVHAwbnI6dUd$GLFT!8ml5P0vS-PJtaw@li!do!kHywI9<vo%+>}7224pL{xh~7WO
zu@s>V*EH>(dQ0gg%DNnNc))kj+4^s0q7<QOFI6Yk>aTfZzMO>&aPxr4gOo5)v&Jz!
zuhOe}-eI!!Y?7yPW`9s?zP9zz;i#Tb9oV)B=8%*G4k&T1TE@=TkK!gr4HjzD9G8pX
zQ-Gx(MJV*LOn@k1zj!YT42`Wy%31Byeg{`IMZDo3P`Xcg%9UU_B0gEjnEM*BpIH8<
zJiX*(v)HIN+dp(;h0*J6Td%7DAMuI~Ido8pd@A$<_voN6=8Lty@aD>mybI3Fc*nHj
zru5Yjq6~j2GNPUjF4S#eU(trm)9cKEfT>X1W~}wdbXSa`MYEM#T}qv-M1qXB_5?>F
zEw~Ag?GMVa7nY8KJN1S`S@zTZZ(HLNW>%K8eLh>h&gIE}4L_BeKR^rHe?H726eJ~V
z-`imNBX_PP1I!Zd=Bp$03+yGY%7ZA`d!$ALx;M@?cS`Nv+srZHJHqjuT&r3KRpmdM
zZ9h!M<<E^RIrP0yz2lSW(!2MnzVOZVX*)0?<#7+%rET@U<BoBA<KiQnpRkoLIt(Rr
z(|>F~JS@epF;JZSu(|xe7LwgvwP;?_FqSjQza0|#$9Psx(@BS*%*-_tzM0#LkuC*m
zj}pCjAH-6s)W9PWYq-~6UD~D~djHen<FSSFH?tLFbNBj$f_}|Jlo`2}QCm18X@Mo9
z<V<kX8JfUF(nTTZMVT4j!K3y=(Im2>=zS3!lSKD$4bJ)ZrsP%)f)UxoaO23T{8wTt
z_ps5>f<A|q)!MAgH)c|jSI!<EWhvi-GNN9w_iVNWdJ|>SkjbmYUXn3Mm+S@B6%D!v
zU-&Ej%t){YPRFV!v;ELp46n9HW8p|U89vxM_PaU0J}^&Hl^7i^q}WE{EhWPmcgkP0
zu!bw+IOCldA?V1ARp|&F_t^Ph^0^nRWIy_pIA>eNodpO<rOMs=Ny@L9FkObC4LvDs
zf%M}kRni>15A9A!XEqns>ImzYZ~VqsTVMW%&^x}jGpG@(+QY8U(5b=^c>k#1f4Op>
z1V}oH!ge$|Qn>w*@pLtZc@coM!=hb#;Wg13E=qh)GIh-Y?Pqi2s63#c*p)ePR1(*~
z4<peE2<aYn*jlq)R>ZO+2aLMkZ9RRM<d~bjE^3)-_WA~4!f?l;o_U<-^ZR*qDKB=>
z{5GJs9rfje!NcBb4v8_~K<UnybK**WQ|KoaQ%ATW^Lf+w(MNJIr|q8)zZSQKf!J=f
z_E+;imvLNnPC43fovl__gB~c)#)FaG6T5S;k-Wn?UdJGH_!9!a4M^`0xfcpA%v!&8
zOyJ`vN=Q~SNC+o1)1O?hjJ>xJtif6RDSmX=U6Z?~SEX9(Z~&*}^U%J%0mfJ6<-YY*
zjT7kBpABhk!rgaWiDo6Ouly-R&~wuIzVYB^yDtZB!Z`M>S~)k+a7sxqWzahK2xPfo
zvg~qZbd2fVQQ0wsqp6@Hk*YEFO;$;AWo`X*pIA7vVK^0eMoy}x7b`thKTQN)#IQT7
z5Jqf9(uLLP0Oh7j88P3|9e-uvQM~|h!uYYs_SMq3e~A0qLJ$(u4nfTx$@>qVctv|s
zzBz1m?^tpQG!g?BpQ0;O#d}$aU+1dHzsfgVC>LwGV}A($Q*lRZe7>iDuz(A$adX(Y
zNRBuW9q$a(9vAhPOqV@gW|$1&`?)4@EKR2bg8lKmhfHqDLSkCtl&<TvuV+<v@W3~X
zQE~sYy1H<IGuY@Iy_e+_|J9pLsaZI5gG2J4|3R)Mn4a|i9)`|JT3EJwlp3+1$<=HT
ze%`w9Zx^|1FnKHqn7tDv&@8puZEop)xRWIg!RAP84Phka$Gd-CSHg!wJ5@`SY-k_T
zMQcUbJ@%Q+yP_IBU;dv66&Rv^<?fw)-PE3gL=@C9xB5g(Tef4)i{Sghv)3#@LV>_L
zGk33-wC;k!VW+J;gA>so{zeTv+xfYT^TR)MO<l;zCp2cks~PlP9B;qloZTsbyP@S^
z-(d_AoQ!U{R}ty73Boz@ew2}{WBhgu^f`U~?-$VyfB8%=pm`o#KMLw^Tx8+^PDmT;
zqG40stVW!y&W@!iXe_q#g1`NnS@<tcV|onqP1ynd#GWXE6z1Ydtv@rw5nD4zacTn)
zZoXYV;sBR&|KT;AYJD53X9v!Wr_Yl<ACO<v=TszNQ}yW5PwYN(=)|G0(&zip^Dd-1
z`rw`Vzq(=7L#OFzgD$cAoPJC&F36~e&Mu#I7rC!<d1kBk*itF=D|9eQAf*-M8k!hp
ztADJ)i8<tp`&}Pi-BpSET@zm2-z^M2(&}v4XyeN{hD(Nj#5mpXc}Nw^Gju4FKy5w!
zP{)h)Ano%MGG0RSJUD?vC4xa<>{MuKY+dY=@M*`iUc}MuI#lxa>eGwUa=oYCo%Ra9
z<M`_{&e(jOzb~<_7S0J{wPN0=f^`E=kPHC6-QsPJjdp-FOyA@J;^T13zA30yD%=C?
zy|`@}TV^htW#nZ9UkOKJZ6Aj`uQz|mrQ=C;JfD5T`v$JHG@}yvcc90y_k$h;In}Jf
z;Zy*z-Da(nq8!EQ`_`iq@joQzS2xEpFgANEe=c%j2b{z>puBG9>GT!f1xR3u{rZUl
z%gsTFwB4tEQbh81zcD=(j61?C8<!mr43T^hh4>0y_5FC`a@GWaM+t1P3^Npx;`Nn*
zvqdOeT-pizR;|hM&pP)C1cm}36sb}5e%G&dp2SXRz>vVXn5r118hEjZ)#okWYmVjt
zLoY}@$3#xGExpVC5i%uEg^VY{_$}Y*MkhIK*~Tp&a^>?2UwdFh!b>|PP$+VmZsx|}
zI9q<zKQ}1}?#fFqbfB}rtpR+=baMA*ThLp>&DmC+^jlT`Y9qOKryzU>vMHYoyXnzJ
z2f(#oJJntqXR3@A65E3N59wcL{rH3@+cr2%L0^q^K8v3kf$syE!9P2Fth{gh8=;Uj
z<W<a(ui&e{6G!_mFT=xf5JRlOoIFWJU$#}|>{T}*`y$@7bjIU#{3Q@rdARj?k_F7$
zDFLtt9K-pO386mdg4s@SPiZ>`fn_q!L}n&j^MKoylL8G9kC9_0G54yFAKu+N6b3$7
zQ69wqGim;p`YgbSy@i$cS$m8*Q}_woBbjG^^8SCC31VB|QVm)m4Gm+-Hvf6bT|v>R
z%}eG1c%F)n`8pTH0gem>+XV_Z56R<H_)9~YJgibZ*$uAaIdT?dbQRIF^m?n2nsxeM
zjDDWV#j9t2e%KuO7P<8)PWjKVOy?T}9E4d^!@p~pd1cmlzwLfU+*ulA9BQZhJUoXB
ztW3R+2;<_ODEIO|wgQw76og5PttgPcOW;|=m9)8Gk31IlhUMoAp1HlpB2RMqC@l(Q
zloaW5|4&CQSjVW&SU>}y1A{H5%I=x8DYJFF1_-PaUT0e;EP6gD*5!y;|Gtp6;_pIT
zzdrh8`x6NQAwb285C5ZI_?NkD$G$OCn9ux6gVn#pmeOP8$qS4cH-U-ZR4({;LjG4~
zMR#A1E)hDP#QqC?6s;@nckTa;YgyKX-T&>;Q$fb#=zn|Gfh>mp@WRnJ^yoht@+lGE
zfZd8Lgz$iL${u}9WB<(Be0_P6&UXA8aqxJM1IlC8^-1hxXvf@o1>SGkf756k36lOB
z3qYqdZk31WX$b!i!ukN+{IC1wgML$vSI_voFz;8ej!!RdD*nr()`8S#o>)=-wHZoN
zp*+tH0F;kr*2(%j0>MzA_A`+TiGaTvt;&R8@Z2F)9l7<n17FVn=2`#WHgJ*WGSY6O
zNas-}d=<bKZ5|fx`%JdI^fh>e5+z8cbD4Lf0j@QcVFb!{%GfE>Npz~)N$#VQ_pH8X
zCY>!3Y6v?)0T8~6))$`mo}}83dhH0Fh>aebvw!)ah&}1(l8o~BBH5ZU3*nJa^<)M6
z6sr>gz&q^#SOh)QJ()er!_>ta<eAQZ)}fB4gX=XLaMl!JQ<5m+gPADo=lWC8FBJJ7
z!tVR^BmSAbZ(lq$ec-A3kxfXp`^GJ(p&k-I>M5!YOV(!w{wgWQ`~4&hk|*>`h9D=x
z>RC&Ia#r^n@$OcOCcc=HwT0Tq=Y3a2gtdxRA3PFQ7(K4p6gtd@eEi#l3u{^elF7JZ
zU2rx`e+CxG{lljzL~by7N5A_@$9IfEa=XGpNcLX?L(miG@j!vhQ3o;f!n!z#ZoALA
zE+BiRB%v$NBVYxgmGs5{%-_jP1^*a+z4oMaf3uhQB)D|nQLVuLS3leb1MjTthg*)H
zhFQfV;2*Cm+|>}yGPT<T{|8p=t~wF(3^MT3rSnN(`eL23fNg><YXT%14<6T=yDpw^
zJ)sr-Yv?cX*wmjFu*nVEXILttGrrN@DK&UbfPoajL;F6<_?yK&qETEdoMWl4&oaRf
z4LI?%1;A1ku5}!)d21&>Ng#BfJ`hc;ElvGp6gt9c+W0s4<$$^qt-nBXkK}ZfOghzw
z)V*(;-O~G3WUS(!?)jzY{ra63Gnl7DV1!N{b@+oLsM#mNfuNu6nI{1A{&&UC!=hPo
ze82vG3M#j8vF<I9Z<tOahadf87%s)+rlSNccYv#7KDwfYs1Jyz8891;;bU0yJUTPn
zV}x>>lLT7p7xU#6iz$EshOgz^1tokgu-NrvA>>)tKj9dt#tAwyr$dEPJ0bGEkP7!z
zvDB&SjK3D3Gth&sbzt$)(?-+})l+md(udan{{{^O5LTgQ1@3p^_F~H`XOHf8%@zwW
z&+I_6fa#L?;@%MI;zzr}T3X9zFEfS63)`eD%R&b-Dcwrz!YooL-M@R13^;kbxpKb?
z>BPvA6P@nltb<70l(9TF^PR0b5ub{k`GW%h?_gDp)&TzD2c)r(gTN;Op(*jlExg|l
z6_KNZ*mabk6pPHiQvH()HlJ>%AOdB<-sgMj)N0oNLKs68?>!X>?g~1^97!E<PNAJ&
z^pD!GgxVy2Ae>ePbqKBt#D0a<>HhcBdtUv49#0Ped?u%l0$eC;Z~+}{`lXmKaTiq3
zU*?4+QBIKL3$8biBXF+#n$7;K7K_?IaCi`-rT2ls@Dbl>9^bFie4jY;lw(Q{>VH=B
zs`TE{IoGFOe}JwQ<>c`I;l$42FMOM><%g>e(l|PD=^WBwd3MIG8oXfgN<whZHN*2H
zbM*fUHsq6<{BKsB`W1sR99;{A#RiM)KhsU!EQEN}ETdt1tUL<~F!p~o;#)6U6Sn?O
zUtB`6A<x777iaM=<A@q$MP)}x-1%PgFN$ETK{@dpfpY+e!?8_Am$$608X~_BBGa+i
zxuy8z&i3uEfq+vC^q4|yzH(cA4F*bjJY(gZ{{btj@Pu2=OYn9o1ZnGCLtOZBc9>zv
zN=o$P%zbrf@m3nl7<htcfTcHo5Bdv70H+Fi`(N*sz^Z&3$Knatrys#GEtyZaxJODq
ze8wrBy&SyHsdaE`kn*3y@kzvX!Sxo8pBLMciS;LceIFunEA1b--`D-ROM5b%OIGif
zNzX_-E9)Quy9-2rKLct=PLE>+e@;W3HvSyPKpLj!fWz*8`6UF?8}4Z}ZTX4c)}k5Y
z|4jj8Mg2BV7rnJaq06X?e`Ws{rYEGovSFGM(cp${F$LK_r|s#Nh1P#C<iE6S702`l
z`UbFujySi_H}`WK_%Ef1*L;>0E92e`4zQOo#*I0+o3G7LLYDXYaV0z@5``5djul7O
zcle7=I~XK`ATYZBgb08y9}_vAr3~H^p-P@$eRaSN)pxF48~71tiftIW0sCd|-zze4
zqf}LMJ!!;9F~y#&>UhKzLViGX+!x98K)&GybDgI7s3-li^nA2Um_e(=X>DjDhda(C
z@&AK3dWr?8QYC=Pf)__Ai~i4-w;(^vOML`TSUNm&z*V0NR*~VKVgA1jRQ|sQt^WT`
zv8LEDJR+aeJzrdXEXJqOjKA=X5N1%t%mW+XDB->b;p;e@a3YjSVwe5A@WtU0{r>u|
z-GQU)6t#k1xf5}qKc&cSInj#!9l|&4^3;g&kNQsOtSBT9tcV^<Go;Ay9cGb$^Ez2M
z?(c<DCoW#7a;|1)Ks-fH?2de?PYF6b;46_i4*dq7RU!Xh2SitWP`XowR40!tB@|Eg
z(;WTr0Mv-<{HZnzoM#jt1oY&WTJ8Mdc)eo@zeakt(z57vVflK=<1e;{M_7$NrLno+
zC_jSzGJC|wdi?bU3#`3(FL7xb^*k5f!oQ4A+Tz}Q<C&vJ|NYY5L8h7LU!6>cIqzhv
zLRMN!o1YnDG5_&{1S2qb=$o6-?u2O9R+8awkNEd9F!d91>;1;u(jE>xr!Ucw3PQ)k
z>JzxO<QZ#C_52GdoS@WPKNbc$AO#=TZX7J`z^v8Nxgtx#KbC;6`rnwK8p)UU)F{=U
zkusM=8KNB&3lBT{3Fjj=QV#M03xFxW_dx_Li$0`|-T|+lO%WZS6`+%c=H#&Xe>jx+
zx1%wH)!s$$M-*}+9~VV<c<K1^W)uhug8erUQ;5zHVbD30)umNzIFCup4#!~gn=rWf
z0<^$4fGmXfeSe`}ep)IFbe6;6)cyS~ZRS936%3CnfY>jJb69eWypJybdyFLr?Eha!
zl-Y>C|2d(Y){8qr=!C%4niOk3GagYAzBPeiqP+6rJ^Trv`QG5l5~x@!_CKDSz)nLS
zQvl-Vt?M-%S$Sx{H<Op^Sc_h<4Hv7w$B5x20KSZ;*5IzyB(`VEj~*zo{?1|f4ZK4x
zi3cRbHE<4|a`EaRJkatD%ZktS-xJ085D&V;Uo8@kBBygOk-=%rFyCq;k^uw2OMxfI
z;;5~4eh&e$^(ySuFBnV}f<gdSu8M?Op=PR8NO}WG;~dqR`hUC>-i4tA_J%$v$_ir{
zz0=6Yy$WijSsqdP9`rSO>>O5k@7n7uQ*d#7LtPR`uFgnkMDiAyUrLVvQ>LVd0o5kh
zNUGHUKTF@(3c9CTPMca?<-^xqj>sodh#AO^dgs%sL#tlM_au|u|AX?&>uelb={Nve
z^$SK*&J(UyV6GUhIwQZ70e(5=o-Z>a^b%ir&J}N<*{jmsz$I)uynGk4KN!W!4Z7p2
zdiCLW>@iUAS}>WtJ!mb<4=w9}&Dn`MM+DQXvS5f66oqt_672bdPQ0yW3Vc5I(U>`A
z&;hLo2YDvr+GPy(3+Tbh_II8AzD)(%Ap~lzLVd&Y)KBGhHfD3JRx19=vR+K|Exbma
zy5Z;QQ)MAaN_81{71$~MjWkU<UCvVJSGxgPQ?2rxRJ*e?GJ(O4wb`}!Zxslo-p3(L
zCpIV~UU2um&}8lj3Fcyr7&pIE3&+G!n6XlvMJonHI?fkP>u$jflYLw=7SS~`qR(p(
z86Vn4Q!6LiIjoL&B<J2@3w?n|ij=|V0%AnI6tC@>+9zC1s`HB_Hi<kQyPvYP(89$v
z6J64~fUu!OM5R)#UU~IMzLZI?x&nosE6UDd6mNh&#U+%@x@I22N}A(V8ra?I8Cx;r
z3cb=`qnio;<Vi(yD7hNKRH+ylxL*;s?<l#T%;ZBjnN2l6vL_}$ccZA-*y^)l84CzU
z*bm`m3%fL&&k=g{j!kyPN=_)$(MR%5vu}pjdWfvZI#`HW0)ERC(MYaQA2+L%X*iCl
z1T`e|K~j;Q^6hmXwrsc7_7E?bs+v=77J)p!(}bCNCRo!unsL}lAOdYFk_$naDIwL9
zG}lPYlbQ;XriPYUi%gvnPFU%Q+hFF&#Gv+EsNtK(>Fy*#nQM%r{T-27I7-vR8|GZ}
zk~mv|95GyiK6N!2b?vJ61T%zAc<KrOIpW~le2&*5JI)|Z)BGLFm@lv&OJEZJD|A<J
z<SqkkD?ZMeAwQCJ-CwHFV8Y8-@u4?koM88QzyOb1j41vM3gTO(IxW0RlA6;biQYHB
zrVFop%#&#5W0iqe#?5e<5_?r#O---CnUvs91nb|@KY0~G*S6Az2-2Ht6`po!4j73)
zcu9pnsz_%lI97);qOUWeT*5?g-1hJ=&tNnpT+MMG40!L&Yy116%ucns&456UO>xY?
ztm9(Kj)qP&?j1RH@f4^i(3754@67;?lyJs7ifCr+(74dVTJ<FU!S^ceMYpygxl|uX
zr|#cL`MN9a{7JuzPkuPvTGCUka^a}<{b5Iw|GHvJ$Ya54+_vY`QI;PIkN)btY7H4a
zgSvVGwoPQ!O8Y(kug}+hO~+C24fpd&9wf^YZ^wS#e>INU+lR;K6~G9QN$+SI-p)y+
z`NAYfaL)Ly=8=#10^za&BUqPZy%@xE_zugs#)Z4LH?K_3m!4AMKHzjMtYaWnD_~>}
zg|ELgavQ#HsS_5<?MZFsNf#EvB}Sw%j)*8V;Y)-NQIg71fwr$|o{P18;X?LN-^}QM
zTP^dR&d|@OkR5g3y{X-o41JJj@hV3bpOF8Tw_45eVcxJ<cKXTNm6SNjXl5dUp#}8)
zNKa8(abWhnr_{sz>~()qTUPkHuyFMCTmJ1Gfn-&|6y`A}(<6=6^9L!87Xe?N#5_R#
zAndI_<^f821UN2T7)WQI-GLHVDnBq|HA68<*+%TR((Jj0H<sV~g(X$IsU=+LB{o-c
zWDyy<UGU;40M-X5m+0b|`%O82y*(G}|IOM}wcx>>Y|@r$VOKi*$z-lMLEq@Z_TrT%
zNA@w|j8<0hI^OZXV`$|X?8-J_-OeY-Y4s+BxPS}mn*#3M>Of--n#0f+E0J8KXvl<r
z+?60Il5SM_5{3jG;jY^#`IB)5@d>5}Dp@HOydO;ctW(|aOAO?661mFI>RG#Lk=%&z
z<wW~i1Di`w!&|Ox{#2Mu#mCu-Q|%&eW~4Hqpu%*_8%~Q%S_gj%&((SWp~kp<;iOFa
zMMmmaP9zt_)R+2{SN6QK4%U%lnxSa`H~xH$%#F@N0T(CL<utP`%kWlW4O8-(DCR8_
z^|~E)iCUG~%CtN0$D32-+mV$hy~Zsn;@I(;>Up!5VKXm659kMpUMR_lU=E6mii<3a
z75{$S9BYx^5<9*mSOpCiy&p|?J*L5^O&jT<r{i~e@phZhi>>Kv|F8^O61CPVTyx?S
z12P^r+ks0>*{4skE6ivzs6@^no;+T%I9uDAQ%m^=S;fnRtQ|^NDDQ#Vk*_@9Dn8K}
zZ)lGKDoQidn`m31e*zy-5=k|#s4ZPmms^3*g^#KXiV~LU$RB%MFeXWY9O@>wQmKec
za&;T&YG#q<UEsHguZ+RmP1|YQb80CK)X?xiQ5l7#(3>_2b@#E8pBscfYDx=ZqW(q?
zE~C?<1SJyg*{Xf)-@E|0>EPLp*ZfW~F5Q1yF*`$S9I0S3RLE0O__~1Ne#y5A(USz{
z2j^Cfs78&h`Mem<aNn=c)|SpYy;%QO30O;Rup)SRb1$6a8fVF34{GQZSoYHQ5*~Yx
zO{Q9I2X&1jljIZoQ!liOP@}d*<-98jKCTHDusb{12Vq#LtSo$TuOC-w?R9OHD%9D&
zSrlW9LTPfW8a5ZC+0}}hlMP9C6{fwFR&u>NH(BQ|3wKiSucHx5Oz_o8`lU}zRu-6U
zY0^2fS0hXeo%gvxVI7|bW9b<0Ne-PVN&;JI4VfcQM0d!UyO)e3MNJ~}sI@*4BuWyh
zd$8{ND^5f)?WB&-tdfDm;^Q-8n5QDEDqsRQ)*;9v!fK8?ZuhGZ!!z@yp>yrWp4UBk
z1#r0y1OYf8_@mtXwlG|5_rdjq<y?TSMK^tadfgJz#+sXsR$LG$lpV83ZuJ(=UOV(a
zrrB^_c+vnC4$}>$r=FFT_E&y=9US2;LYH=;&M|@9BRps=<y*aS=3h+Ltd3VvZ)X?d
zF_<O#T_<&qtC2PTqoqVQUw))XBrV>El$xq0QO_{_q_wL0Z$#l9U8MBUT`}LSgeq?e
z9mIF3lzr}9lNo2#yq_?1!obZPyT@%JaM<uJd_guNKcZ}pegBLGU3r{oOl#P7a2E)H
z)W@VJ(a`cDOF})93G5POiIH8Oz()_uwAy~X+cz~6tKOYPpGxRnQ*6cDx{&R`?1>x^
z3M@=vF>r5;p{>cTpsPDwLcD)}26M?wmW1yJ2^#Eaqqq|f+VxdC-+XgXq2OU~vhuJw
znY@n8WFZr5PwM{p=sNA^(g`eRYbSGWc%a3ze=vC+iF#J^ev8M-rU(YVRHryo9S>@`
zZ~r8($H3ZnZCgu}{p<Ss1}PSdRq*B#rQ|-E=93ai@#nkbP`zZ(<P{!-DGYZ`GB)#c
z1P%Ka=ZJ8AEwOwLafY;mYCl}mcP04+=+?m^xr*qZ+yV}n2$V~XFij;zRW<dU9$uM(
zA?lUfHw{AL@wGBj{;E3CK(V2P?rB?`f(5dA#%MRB4|1(Q)4MOxrz1quSp&Piz7}bo
z#3zmX1&t<<d;E898Z{;Wsl_stNk>IT28QIw0b+#UqJj#H^{v!TaUgPpX0+}N4O0bH
zCXJfgP-pc0Y&s_Asjt+7ucaVeL~RnRyO|tdU-_L2Wi~=(w$FIIQTXS&@R10>nkAXY
zrch^UdQseW^nlqQ4yZ0dFjI~qLP=3*GXo2+2sGN&MAE`7BSoT2@v)3XuRIw@P~#+)
zpQ`+FS-cxe$b9>UBLsna0W(wz$xOc40M`Kp@mJt+aj+w8?jipl2NIM}L_{q~VSPON
zIYhS$GoF{y{NLin--qrIX=vDt%%zPuguQ69D+uE@n=qPSR&3bTIU1|M8?6?IW-hR1
zLSH{h<>K5eV=%?wzI9dNDJnhySyuhJ1k<2whmFk8OEnK_x@XD^hc`Mprk2+Yzpoki
z=nymO8}u^-PwpMwZ)URwhB6i=5zK9>vNXv^5k~#WFD22Yc~r|RiURkS6ciKC;bh|+
z8oelHHMLqd4;82EZbxeBtl%oBSi%+cVDI&XZ`A_j4vKX!O|Y6GNyCBK+iot#52RDK
zT>{ftsb<hCv|D!l8sdD+m{3g(=I*9ImY}K<RFowZAb<Xar&O!B>HehI_I9|+s6DUN
z$`FAeaQHetv6VQUQBq680=Gh|2d9#|1IoquFm-aKn&n5B<@ra<CW`q3BZIO@2}=Rf
zyP8oMwk(_?L5%f&|40vsY8yi*8x<#>u-n8xM|cZ4up}_4l-wn8)&<?Cj;iWCrWkl`
zIs~E9wiGaIl$@Iea&rbrDJ2rpb*G!gCD_{YUMb2q<Ex|inTu<5^&R`Ytu<6`z-UZP
z@E6Z*<fra1o4Y3Yd0}uR)X*~}?K6_DTol8Asm3RvKIb$+M%NWsqm66Nckm(Zo2<p_
z-11ba$|4l{J#YD)$%EUjo7z@rFYqpaKSzx0ty=htjL53A1TqY&ugu=$$QMiby40}y
z5n<G3{NwTsG6Bi&FlF{}8z<;7@WYf;^-*NILEw?Nv7|AlGO_-y6v7xxiQyIKQ>UPV
zEw?`W7WWvU0~p7_k2<y|T-b&0M>{>uNtosZTKDH<suJb`_+Xa8G{|1nuaUbrKFX;h
zLohf+)tp1m^sEb^^*V}`Y#xrIHO*tB?}f)q;)Q<AQr>odsr>|*11pt1e6cCN(ZO$y
zSzlEWwQ4^aGa<q{LmVEFG-0QXHx6EDd*NZO9G7374jFGPaP%Gx3`B6Si0gOBbS&(-
zX^-BA8p&%~<oB|L7^r3-%`ChfBR2#~$MHm`HFwp}kbRdOmMO?YAKbACK4LGp&?3S%
z%U@w{=G7g6?TlmOQB#SRntS3j%Fe>4%-Lg1=pKizoSNWjw%{S}?3)1}zh+(G!EBnb
z`V{XH8nVmXO3oTzaXKB&2udFm37=;D??Prb>o2ioxL3l+Cq-P^w1~X@yl07~dzsbP
z0zj4GkdG9NTc|rku-I4Oq1$eZGDI}!XO3B8NfKDb9$8YPJ^AqR4S*(z6IVX02D)9i
zrUsr_Ap@$&=AJ;tEXdO&to)V(%9CGY4LYJ8be3GfKn;`0{TataiJw{iMbM+ZGS0gC
zHagG0>o*2LNlJuowTu^4_Um<XKXm6VDV0gqG`kv1=7QI`qa4s8<#>e`q=4S|WEh~0
zN)=&>5VXk^qK#}wSs43*A|qnK=?wltJ!uCpVk}6;Dr!M`OZ4_Lz>N?eTRHe^;)9FL
zNV-pSf>@``_-c2*fqObHXNtwqJoQ+6kwQSpaQ3v>{1@huM6qJ(kv&_K&9_n(qO`|z
zvbS-r?6d|j6S!o&okud+o4z>+x|w8k3v2>i_eT)RCGrw~5Uc6E)Klf>9|bF_aL?vY
za@G)qCmsNi2VX3;ZToUbV7tEV5~7#T!<7e&<iA+S6{eUdoLe)s2IjT=><jAGPPu<4
zcc~+PzT{H&hH<k2c@smw1_rWjvAdUunr_yyw^U-wKZlAHqeYOo)4;P7nG6+;&tzM7
ztyn91h-Ra^<oZflu+yQ3_1KW?^+q}-$7gaz2=eA`1&`6X)n-cEVf$Yn60Sz^QTsSy
zhMORVzpF{0o>ftGRF-r^tyzUnYjAyeO;aLd7WY0>OHscNGa;3#6g4+mt!eP^W9(5%
zWRcoH0<%C~D$}syJ$G+**cQmMobJY0<thdKJ@Saysg3Vi7Y(uDg3<j1^=~!zUC!%W
zN*&W%*;ux(MF>n6uYQYcYK-1$im(-3MDH=W<bI1RVnVy8pH$*GR{t9d@KOFQNq6L7
z@P%&W0jRA}n7GUbPm3&JTwz8ZmB}c15|}^$bV|NxD8MvFpE6Jy3Zx5n2IS@!9J>x|
zqg=vpXr^&#+PnAa#4@0-Ak#J~{iTBY&+N@v#foEF`B?!4F3^lCE-GQ{kPdFSCmluZ
z4BUKrLjJdz3APU0_cMW)n5UgBH+xTzkH61sma&`B7FJX(f^E?i(<q6up?Jm9c~3P}
zw__is4rSIUpt|_y2mmeIxAgRrRE!IlMIpgen5ZR@Tqn~$bqoX6>yL3@6;&~?NC(c>
zUw*K(RS9V;g4-V%zQaxqg)qo+NMs;}V#D<YI1j|1zA-qa@(IGiRP6oFEv5ZDB*nEg
zzR~Y3q#nyF#VhnEb$${k5R$*Jx$gzEjU$9`K$8>_qwsF4i*DBmkq}qn%81`VXm%54
zkV2Q}rQYLJzE>x#!rqf#`ptViGQq2BY?Y(@^aF2;26vloC%YDJN+JbB{)Gt#Iq|KW
z5auo$<cr!jSysHQpyYLRxyBFp`Y)%Tn*rqKzrzE)QsbOaZ6`u=2vv0!f9x&jPKCBw
zPz^xb{$#x;hgY!jwsW<#UM~26Z28>?J^9hyf2U{e)S^<jE5%+i11tH?Sj~}QUHnB`
z=$&KI5i!z?liJQQ18>I@Od6*R&mu>ww@^p#d6cJZ8dxH_p9UJ-D0V#G#jC9K<^3@l
zgV=~*470K6eR6WoXbO4be8816!n@8GTXk-#n9_%d6E;y=90IZgGJ3VriKI>5+R8fd
z{AdQ!rImiUC%jrnHzj4FS;xlQNn8D<c_aYYE4`Y3{IKfpA?u~A_!9<Vi8XgvoBgsQ
zBSfj8pOe=W<%Cd+&8b!+7TVkTPmhyJmP`(=Q>$W_yP{2L8AoVK@?n0rpx;`6T*Eco
zbq;;qHe-QP68Dm6hx$P7g_R1o7QVTL=><Rz|6MpC*C*>A3@kFK?XJK(l!Ap!Z0Zu!
zC@rFKOr<ctW?q?QDR~uX@jyplE+!AiUzN;lZ!I<99q#3v&cmhH!IOl~$+d~q_e?M7
z?=#Uu6IoN)&73hakE=;{pF<L(;VDZl#qLflD*HwmMa2gIZSt8Mwh1pj>T%NsI<938
z7Cd$5H0Dby+u=<2mc<o)eXSI&6cQ3Vl5oF=b_en%9nspF64uhb<M$BH0t4osv@{R{
zpYp5njSKOU3x85cu)U6SX~YF;NTL7y<ip_i$p&}72T;!_fpZ3)MSQ-sr^cUaLMPB5
zO-?1RG)AaKelgVyVAA5UL}&z*UUN&Y>@gjTH87!(K_{yYyJ=)v^S+10n24g9;w7{b
zc7bJkXK71P43<9wOLcWP-&RpWsU+}q{M!7+Icw^~tKh>b<=A81D{BpoA^CU<5EH`e
zY~ZbXXDd4@3W-xr#6{#Ee${mA2Qg|vw0K?~^LP09C1yhJP%{R2vVQ|d^8hY;MQh0K
zRG-{Fi&~;i%l=0DiKHTF=vm^}_Z12NetB6t?v-c6Z)^fDeHoqJKch{CiA9RtRK}o(
z33`FxPR^BtwPHggh!on-R$s@sLuL>#P<q9G@xXA>>Xy)NB6A9;^&R=@6SIhT`Sz=W
z!Pe@q3x}Ls5`l<!i5Wv7dP5;<eV-~mnnKI&!~xc{2hWs}^Vo#>X`~)ja(F@DwPI*C
zmv%#NI<GY9K*}fjDfG+5s-}c^grb?ZYlah#Mg{)wh-d`A-H7r-ZAkEJx{P9lXKzvz
zT&dE7<>P-)#bA5_x%Sv2_iTMXYsuXI7D%q}j@LvO;d?5bKr{tJg+s)({8Rxe<Qr^H
zaBTWml@{%j+wJz|xkNh59sI_y_|<<ByuYKT;Nx_Xc-><roAE&Y_zUtV-f>kx-<ErJ
z-yGOTqK@YVu+CfXw~F1W{J7q3{!;kRP1aMcf|q5s4CiCLi_&|vjqfUy?(>Z3IW){x
z&a9{m#=n7A8~n9THZsta!h=*}(WIM==os{dEH3TfGW_cjM*YYdWE!PW`>dtdDy3vb
zge#9gXzKBBq!2-9v3xyACGR5rIU$$zy#Fk2VKzu61Le}HM-N96g1CV8p!&Ir#N{YX
z;(^_N=g-cOV#nwj^cz~KVZ%+H?6K%Rr2+4C(4;PrhiA`1Qml|D2oX4;m+AcQhSQ3}
zU}JEufDMQAyAOy7e^!fuP4{7meORj}n%B75cyqkbP8@vQ{Yr3o^X#7RK%?LCtl^e3
zQ{{td`yeIs+xMfdCTg(-Y%SbN;_)DP6*~Ln7q~@V6kCk&IfOkAaSdoWF;~%yDT?Z~
z582-cA3xzd{`BzwCXxJv9v~zFZWg#yc+)S_rDwkRej~=Z-mretsrODx(h?M=>2>)o
zKC#~wZxTZtlucWoL|5G_{;G%)yKe1%=^^RqV7k3^c*Uywx)kK<IO^*iJ9p=T9n$T7
z&(=aJR>gU?o5p`EB)MVP8@Bx}Aq)?CkF|nsoaY#hUd*?mOhV|LdP&Wh<sVzQ((%cF
zm4<+5y7Cm7)$(f#v-NA89*@4JE3K;W$g9|&=#==8BvJ7_C`vKFt1Hm$VDQ2=MP&wa
zuFn^gBXSCn32p3KDcJy>$p9HV96Bqq2@5<osDNQ9hfp&5!vpME4fPHKEfK9i0%5|}
zz7_lknV<-)Qtwzfi&ObHvk#T=1kII+O>~FBx97om{i$q7-PR-C0%lMUjR1i3w@N)m
zUn~nUOMM2l9s3zo0wP_?#2oo3G_>iE`^2J)s|#8}-%1B7CU}~x5Gh1t$~yS7my*?&
zRp}$BB=jZXt_lcmHg@54)EZ0caFHm59o+|<c{#~7?MCy^D0pU{x>YWaa$k36qsRuh
ztm;+6Pgz&e!(M*D_O%?sqK(KLHA|i$b`nne^P8Vxs;O1Nw!1q>@uMOWn~3Luj1A^&
z49JX#A3UOo5X@VCh@TWwqH{&s!du64L&NUuV!d!D3MuYa#u2;%!aH1s<2n+*)b=mi
zQaaZf@@iB%o%D$(ZOEuWu63f>sF5{YhJuDa#Ws1j8JV%()zg${2981l1@MX`y<%4p
zXNWX)Chch$+~()GeX{pL7@+NRrmCw^vLElVvL@Sp>}|8BA;d8!)KVWP7W|sU@*@~j
ztXbyz2#Nk4Oogy$R+HEpU-M;GRl%=07n&paIYP(GHVc(T9XOV(9Nr4ieHer8Y&$?*
zM{K&PsTxw*MD|62_1}7KOT})^Bev>{ihx6Ie9GZ?L=rqgE9(WII9e7KkPcU~ty3Ql
zozhH<d82W?0k5|nLEVp~R}Up!l~~=Pl&7yRWZLW6Oc%*qxCJrwhus5YJ<6OmLiu#9
zvv7~&4$^1^n(GX`6<k&)#TI!U-6ah^bSbAE9YuVY&2Iy^6`v*qZPVpapr7jzY^Od>
zNV0w@muzI7)9Q7s|B+!((+_%EH1KY{g3%VcP_N_t9|k}<hp-NNih+`GRpU3BD|jm&
zB|2D2s*B9)H?;ds3im6-8CL;m9X|aa{|egnN&H7bTmm_e?xGK#TfxAGKRc0)I98OV
z2HZ_aaRJ7tPp6#tKhIlNw(M<4+H=rQf<NJJ^G$Kk<{;D+1h~a+t(B12!#j=Mv}B$&
zu&;3P+ZqU9Hdu!-%N|W0#4PyC&74G&7?9Wc>Su2SesqssJBE41J06oYoX!2LEAdO-
z8cw*?tU##QtP^zDJRlZ6_raqx>mk`pHR#Dxdo|$N<oi`^)e~wXk}u^ArGK9rU<k0V
z-p+z1h|XK(ElAst@?v<4uQz&G{xStlSHRdj^7uQ$IAL<=2uM%v@^!*Ds->X2Uy@g9
zl6G{(YV)dLsKT-%?{j^m(ovl+NFTOu8*bh|e1P;CzHb@Uhpn&zEgWqDOLGmSdJ2<M
zyN6b)NaPTLaOFd?fVPZ<#t=u0<^r<h&_Wxo2;BC#4RANO7DKpZhTBD|%C6GSsyU@a
zx?E9*1HYPZ^gswW={MScLOeRArsVb*XL9@Mo4-dM^yqNj@wH=U>8USybd<lgHuO#b
z%}ovs>kXB@w7&2@1d)#rS_*&d9};jed2-%T8tZ4WS1bHOB9rGc4D@wSJ|;UPGm<$c
zB=oyfaI)FAoBH<WF@$lIIO@tR`o?kwG@35jS$0Lkia`4@<KpLn=9vankzY8i#o-Xz
zQ*hzH?{|kDB8g>mP}BYfxNs;D$fcq=-&?uZ_b>a&;aOmx8T4cZQn;J_pdqBzrf{)3
zlUU<wt=Kcv+GIvjNhhY_{6e45AXo>Xz3y7KN97C7N7vv7;Kwd76_H&#F;VBU%KZ*1
zX}D49_8!MCAxyZyI!cioh=>4e;SX7Xz9chA%F9L$`2Lb|)oji?mRzPRqGYbIoIjk!
zw9^WNCG&uyaq2r*J_O$;e7{nX1~MBhSSR3(fv#WA`_E7Be&im3-}aZ32xQT17UZ|$
z?DWpQt>_mo8Ejw`s8s4|?S9;oywK-rH@m=6UQXKlLS+@CNwH#<uT0luc5VT@Mu=Th
z?9Ag-eks}<s<e+~YT{&whnV+?4)-;c83W4~C90p@pZ3<KZmi_7Nf#3$9Aoa9^O{tW
z?PP6{OHe2htm_p>W-ayd(8^__37ty5@CE`dgp37&qdwU8T+}9`j_DKJ*#eRC=%A70
zU(;fW5gf$kwV=&-Dn1f^(@QGF*KvoUtxc`{qz5sBrwk_xuP9lNHzq=)O=Yl=1^F<&
zeo5(=nDn`Zc@l9Y5v~jq9>dA5a`}%>is?<>DPP0V_IQrD+3$&u^yC<|SfKH5zT{%m
z5Rus0Ck&qikkWa7m_M`DNamlE&p)uLKyb%3_1<R${DJpwijr?iu7cvzl@FSHCM8|3
zEA%Iu*g@mJoPm&C8wE;HDQUi8ixO1043C_tNbWe|ij};?^8yI>(b$~L({f@r-p$$6
z3lc{Ej&8>qvXFfdk*uDR=hcws#hHr|vNmqHrg6kL^TOFBq|`~-0f{-w<sMSc7fWMI
zI}jBiY{;{;#*<?~X7FeASK{nE?iA4f5%Rlkj)a@fHG|rgQ%K!{%mHP@F={#m?xOOX
z6y;USi*A)TzM^!g%Uw!_%O5?e@|cDi`CkyXiYkCL4WY%<(L7x~C2pOEFAv-a;$(Wt
zL`PC&nPdW}CjGO$;h6z>aFphk3<=WG<p|eG$`H3GA31!+m^&Qcz9tollvKWCW2hco
zYOmUU4pAeN<|{sg);cbmuKG)|OKcM{UJlJ_A`^biMUh#jLEr+abm@SAMh4<#UbO6=
z*cUaoZvHM;?Cn?5O}PP`0p~CkcKCR=1e|Tr?d#|;U-XeTB*+gzJ1F=OMe;{%v?~u8
z_cmFX-%O4sjnK!c+j4f|HWJWTLDB*Iv_|>T(1WiMQKb$sV;~kddy0<J!cBTR{1Bf~
z<GaUDn1^<-ms%Wwnh+POn@*@+NRv<<InervLVF|$nfTSR77Zu<6@2UKUAvH2g71H@
zRz>R@P2`qK_j0Ly05vilN$PL#yEH>d0pD1ctNR+zO&Kth@X@xixS-I?$&hGNl0EXP
zZ)}nulent6YE595@JZL7HX-8689wbfUldpwZgFmBhk0X6WCGkq7QX$$5^NNyp=15x
z-Q%wJXU|d);st#SR??m{WL<abVPmOxU^XR>Tjy5L9}kTgs!U-cfA9CjBNgp{3bpvx
zSJpc=#bS#?zRdn*J>ce;+nWu8KacZZr6RlW-I*j@kF|*%=F%JS%1X7ywgFU+o{eJJ
zS;9vxsdQ=0%smJVEWv{tl$vu~^rL95M8gdZfwfl}O2<6d{^eafgQ2e8%PIk5Q+b^U
zh}d0t@R@m6ui?+0YK4A+7|L>?M(g=SVQtlrQnJyl`oPonA8WOj&dNSl6b`-t!#C@3
zr6!+CnI}nO$!19+s1)mEM<a@5Sx}`DvR>yEg&LUG5DwtRrA4EsPS9y8s-xEIYW``C
z^tV%5irJ>h?ThP`gjVF;Np{LoEdp3Ep>1KF0bf_lH#uKE_{&fNXK@f0l5o9SaEn&o
z1|-KbDsRKmks5G~S6&=Xs!~wC*}JAK373_x^hsq5=7V#S9}103NhF&Upxp@6R~TV2
zA_Dw$-er9C)}jWH8sAw1&L-y(@9E~V(V<*er=J!+w)p1Ov8eBc^)HGw@X|D?-fwn&
z@$GZhp1UC<-EeYkx`g<IP<8lx>Mw!#XvME#;T5Bsw%s>>0`*_lxF(OP#ju->eC}H-
zicEmM+(f~vp+5OK->x+|PC(<zM@+FHNhd|L<ya;i>js}fD<bQvG61d)QGjGi%ZCN-
zlaCJ33TEqdu_-@o$4!t*#RShNnqD0|m_yjxe>M&4f0zu=$BU&e##f|9^&YHS;i<qc
z5#Ds|_~Gzk(3LNWeYA`FqkwBUHIu@Riut8$P5Zvw@3L<RJ}c<c*XZ(!9#oGJGrAsH
zasY>HK!+rD^>rFj-fcN8<!4EeuEi1NJ=X(V(a47j{60^v*<M^m_ZNgqenGf4eG>Ci
z?TlP%8|J}$&YF~%^OUBCTj>WKlU4p&AO&Ufr~eCoK!CryqxzvD%}kLes(PAtd+K+<
zu8(;5S6ccTJrn;Oe(4XMaJW3??0CesRb-jap)j?jsBu{n{DS>Q54in7#Pkgv9(m<Z
z3r%Q>td39_>0fl#zsueIn(&oKzA2I4jL4}%^lK*nc%V9os7*j`1J#ux_)J<Qs?L&b
z2V?^}|KI%X`!5*gpKEuyt1J3nFoZ+UVW9qgLM}7O3&;vnk68YOChV1nDf3~*E;QRW
zE$)^ij(SG#_f+QzDFj&;880y;)Hb1f1|3n`h}jINg(B7op*`kaK%afyeduRs|2FUo
z!2isTkD2{TYx48o@`e2b?2mV^fBh#=);D*r-~S1e^@RZy{hNxo7f}}h{h92Ofc`LJ
zt}8Y_SK+=KQ6WOoq&A>WC7~9iNx+^Z<P|_hU3<*Az_%F@6DDZ#IFeIF51O!N8D9IT
zH1=Omvkb3k!d^xX4*Q-UCqbTQ!jT}fJ@%d`3_2p}Dqv3}VPeTIXM{}n!83kG5E=~!
zce#7C)7;&o0Jl#lI<+sd1e{X<)0P6L{&Qftp)hDGO8s8QEdFc!;a@%CtXeR+ewR1i
zT{5j+^xmHOJC6E6jXQ3+{`HT2W0A3#QNZ1Svd@U`@Qw95yw+?O*{|~5%R|mxPo3`b
zMw?(FAv@}gVW<zdKD)`|tAN&9+WMM#4J3zpkWt%25Q*$G;YLIED;j@XLIl+db+-gL
zIqZ88<vwlB6+OuCvLww1#Cw@)@j0D4NW!GTA6WWZk}M-BDB@_t^m`t&0#JmBNA)}H
zHUW{0_MU2M8Kw%~8iu2Y8a$~Fh+Yt+!COtz9o4pkcc&^r5Re5?8hk4VGDBK0eH;*8
z5yv(Ce84(QIL#z)h?O9Uj0_srRCH58P#Y%iTl}ej35eG>jH-;t4*dv}2eS>BGef#D
zqx&_@_EnF$A5oX3UE4d7G>eEKP~E4@0X)GAQk79o?qnMT!DK`y38N9+OO`b6Xzn}g
zDTQ&kk%UnhKWgRB7m$pjKFg2{!@eL?n!H~zylRPe5@stAHj}f&cnFP0UTeaa8{EF9
zzF07RSL5r5?RuJpAXFucTRTOp68ilmlarR^D;2|+6J}%C9yN5Yrm~hi5d}1Uly~~n
zMKe(*K@zd(%Wfsuh}ug0*b?qkggXK;ikORtUW20OcuKcs925E|q3tmb7ZJTuBptAW
zCfbY$u>b)hgLem<1W?rn+*w?)cWvo!8~hE4ltTejuM?;>^Mei3yvL<}mHXqGe&QIN
zt(iS2;Lb3cvcA=_-mmGmmJrW4SYI%Yk(?yCQba9@Hej{^)kkE%Wqe|Y6GNV4M3kgf
z6ZQw16NQ`9bjO<1demAIYM^?)zY_M^V-}MBhC)b5GClT5B6$dnB<)MY`1y%t@vk_l
z3rW%tbH(s-M6Gvg$;Zlf*hf9}Jxw@h@Gnd9bjk`t$P(Z5OdsT3$m3H>g^6?0qjkf|
zIPBUoj1|6~a5;Pay)}I}=JYZl<q0qQ4|wH7llL3CUomut9<zL^Q|xKp6Q~}LHN)2p
zVLzg_YwCN694qojkxhbFq#1^p!(LxA|1*zW8KTw1R+Gm?!`E&yq=?=|s-?m0H-uk)
zzJ3|s2J}{gNtkQO+WV=N@SSJ;jwH1${ueaE%NcVm@k)`W5p^BVj{ywfmL!cMV%yQY
zozT~sFv=vEu}eXi3Wk?={rNZ{PXvBi(Z6|@yPtZuDB<_+*DT>+pn0!2f1smzRTS2I
z8nZT*VX~n;eayihd+cV$^45&iTar+17(e!?B}fLUjUvY_jqPdQ0C>z{e8lks%eFn_
z)ERsg(33yq>vqJ(Oj*8u%;|9=8%11AxZ<4S%bxl;!z995kcA@F=Nz71vv+RkRmHFo
zWNC;a2QE=<6~0c$p(5B<dH<z5e0Xrho?9}V2}0YU-rt=&q{ZD91?=}3qADTfzAylj
z?r6?5QfG2@*o%Z31X&K469Gn$RKPyA_~VM<n|HZ8SA>@}VHVK-kND-wHDf(twch7)
z12ASiZs~6b+-RUaA22J4uUghC!w?3l>YT#~fF_J0W-SPI!Ohi{y-mZWKH&UNB1W(B
z9(!E03U7Pb>jAS61T$u9$1Kei2iqmnA)<84Ryx{2l2oL+jHs=`H>r3j(ultOI)7>d
zTEEZh{*;4rL-&eh`!z{w2ikWMe(0&?174attevq~&RBH2x~f34HViXMs1;$7k;9nv
zqQX}TW}6G<L)oX6cQPVDQWqSa^i2DPjUBU5lB7V$nqgld#DG074Ekuttg10v3CtIa
zlOR_P`y^6b3F4$^hU%>$-fHQ;Qs#g4Bw{Wl7(*Cq!aSo^kfb8k0ksI&D^+BC;!d?A
zSzKp`tZ65}#1N~F<^zYh62ybzM1+J|Uvoo5k^*zX(OlGJ&P+#aZf|KHF4_BmBvqDf
zZt!D(HRG!b4wid7*_+VaGQ>6`uyp$w)Ag7OO)@~5u+^5nxn^>`;Kmb0&;x1{F>8gN
zDB^xb^$MYT?88hH3O|)3tB4bWzjc?pV_skt{p*T2+c3WRGrqFzsq44+u9~uGB}wnG
zyl7d;j>ccH*Jp%QxO$)Ku^?AlMw^I95<yR8GP0F$A<2&mqe7%DqbCO6Dzfz0tD4Ph
z#6}3DOys#{_$7dr?sZGIH=s8GyV0b^QC)zHl2Aot+C`y=6{9PM3X&)w7=^czWCFH>
zBm-vSG0TcH)ueF(Q*@QZcBeb5BkD@v=a%91r~1Q^v@gh_fOD}4*nUepf;1hm2~)bc
zVQ3djJ{YJL85u48$PkPG)l;oAS!psG#%oEE5~oIVqewHXm|XRY&MJH}VM{_=$0&4E
zi%8sGF?wad*85z@Tb%AmlD4>sC5#5_V!`a)1gT+jqoR)>#0C539ZgW=W<ah3^@Yam
zwQPTJ_n4cAURC&MLT&;=W@1~;Eon4h9%jTAV9M28z-5Ab&b$VH`+&zrz!iw72;-XV
z%NZ4yj2>=iHVPk)cx((&EE&5iW?M<BL3+>Vkt{Ncl_E|wVWtQ#e_|h!75(c1KkXRZ
z%!pA%5F2LypGOZL-!pUGFggX@DdMuj+^^|xLin}+l>hwjMdSSQOc7p|_<1Jf0sW33
ze~b-mS^c8HAA=k^#{aLU`9L$=ny~p=#W3$_zYj@CQnRJ`LC^Gu2ER99^9@6s_t=MP
zroS)ATH_8CDrKZl#Cef6``CSGrU<Y8MceX){qt#G0K7jP`|o$J{{isEk5SGD_^)<9
z|4%<wjlM9Tc#|b^4pDl{hoAom7*LNj>98Uk8k7@+Hc@o$6uC{r5O(11Q%mS8iSm*(
z0yctJ741TEB1x?z-Mq`)FOnYq1?(<&wItn?q&850Ajk<o6OJJ@5q0?#+<j_04dkGR
zBY`^rNcbU8J(9Sw!QG|+qm}|v{<9=d%fJ-{Or7!xP~m4}Q%W0_d<MAt@9+-~EBa%N
zH%pHG$Yaa}2QNIN19UXs&SY)q@BB8uedoFVdI~6ScHeD*>wm&u8?71JV@}umTzUYH
z83$CmWU|#bEpelWIjQL7El%u&)!Je06?>zng$L?~iddDzZ%G^Co*;}P>T~rUOG1?9
z{%+14JwNAE0W{j;MgMGY*WwNoR339#z+yiW=%`D``bA-JfItXEnrO1}=%uEgWOQ)Y
zjmB4=dZifJ26w|FHj2<BWDrDS7{)#IHe;N@%>t@dWGhOdgjC_iI~M7n@zs*anIeo@
zw$q3{RRjeBNw5Yta@0#ps4TAOX%_+^6>e421wq<dGCz9SK94-f1nZH79!Ga<@bjnO
zZboQLxS^m4)X$t^0zX&e+Ebk-=oN7eXhqx)$N*Vt+&q&*Mht>%Jk5z93r(7^iz-U!
zr9e4Js2tTJ@Dq)DIT1v0<Rl^SV?k&^T|NbeJ*^(m-IRE<q5c4}8Vb`XNV3RCr%072
z2SwHrKkgZQ*AR8ZaOBCt5{?ya5)sRCfs<GHnXYzJnluxbE@0LgKMUxs!z>hjk`dA1
zk1g&flQ*(3tk;Pc4SwR$s|9l-$m2uKhY6dJAj&Q6xutIvDg(;WO*%$TEN(PrGYObS
zj#^)`uPt#>F^oIRd0}3rdc(*C#7yC<3G1~bj3gvUY%(G)nLnu6?vGgAQN+gLnt;5o
zh&rI^6|*M|-3^0pdg=>V=9hF-i=EbHukn>BNh@`Q+fNWPdQ;KOODKREG+~dTIsCE9
zO2%GRxC4Wqu9)4IpbY*-z%CTAR=A^&#*nIN+)I*Zmdt*j2~$ltkR+|h)yJPlF-P;r
zVXei_HL0)IP7K}thP`()QPm9d=ilGa)CozTzSq#rBWk=*X;KLQ03ZNKL_t(ya?ju=
z2P}W?Q|5b|bG>v}dym&5M6C(?d8hR)5N$-AX5wzX#DkK=Doc;K5?~z7siEIfxYjXx
zqHz;Ns2$B?fS&p^?SNuM7%TiJV6Q!@Hw?2x^+b|t!*EBCpUHDo<eQRIceFncWTl9Q
z8S%oieQ{=rYXb+Cc<c!aT_O;qFUhqKsLB$j5~`<wZ)smvnluUMr9@~&+%t%_JOg{m
zPIH(N!SN$j(Z8Y@jv&!f|5<5wqXqevB(#>kS0yAkZ5W>bteJc;VL7ka?r&)Vpe1o3
ziCc+pH;gU>!U)oFU_4$jZI;Y>NqST<)KjiyiJ+2uAMi@A$+vFs<RDRZnow!d6G09V
zW)5utB%$(D-Ij6gse6mRmS8f%34GnK*;~_`C}Jh?lZ1&Dw>M)YH9;igY{_)B1N+f1
zDd2dHf;f@5cFXibgBw@)cFpX432L_n0ua!bimU}fJM2S^8!P-cP@fxyX@mQw!5>S~
z_>|+bEv*6A<7)G?j2VkVg|7q6l_a*HL(k|`lj>tmWkBnWrs`?5A|Fc7OD6XNVO$a`
z%8e#Xj(Ie?$@#?K+ky7V;Ku?L2iiqVKW<seD`wtf>Ith=!!Q}p^@j07NEOHcGHFun
zsF#wV9kbq>bKQH4`95D8Bx$tM^WF-w19-@trl+nHG465o><l+%ag+&a%lLuD?br0j
z9=)-Ix}sYx8LzjrbwV18dr6b4E9UQKas!!`jkdT+(N6WPrJo1vMv-bqwc1i&82n_P
zvzf*1MdWDB_ya+-5E{_yl7KBcjoUBlo*fi%6v@(|&jevA3De?Gj5eZHlGLOE%AJ?Q
zl1WvBjiw(t>{^g+D8jg-dRWGxRH`^4Mjm~+lMiT5IIa|Nzo8Qw+6R_yX6O%k>T8c2
z9D1RM&8Ym%I;`!fAI-U*S^E7g&ADV4H}umDqsNk5<+AoziLWzqkx?YFUNL>Fiw?2`
z<t(8sQcF=OV(n?}TjJajUJmF>Ng6A{txuc(AKka$d%U&q7*o^tH+V8r_%<P*j9F=k
zP?zi*O{g68Bh9d98RiP{*pQ|kS*@rCO&EEk52$O6Qvo_jG>~Oay(mrvi6l0bZq(5{
zwDfaLc*7v8ivFlbO|(}M-#Y3mO=twsbTrqNU=7|3*sUU(j&`X?x+XNmDd-waKiM!k
zQN&j5<Tf;JQvBV%66A47N)}p)RGH)zacpphig;U+#vYz1;zW}4in%J|M6Rb?9vS*c
zM{^aBF(94Cu4}e4K~f#{povzH*M^}|#A?OpvZ1dfsp+U)*<<7)BAv#yhQVc0-7!j$
z$|(GPPqnc)Ic0TAkYmgCK;r8y^B;aR=eEa&q1`aL7fX~^1G-nFszY6aq@0D_ONRb6
zK~&d_e<;Yh!M$lo2ZCHj)bc5~tB9k9&7LKOCmh|Agvp%iqY>MYAgcj$YUz&@z75C?
zWVvQ?;j!Ba-`?RwC$w5K?J0@0YmZ!H*jU`*&LKTE_~TDF9p|!)BL(SL6J{ON+n)Tw
zJ5on;2C1qTRz2-?+57x2zv%s+Wp}wNE$&z%#{RiG>r<1Ms>Z)&2&10*dl6DXt_)${
z;wB{z0zj-1vJ2?0!<+-z;OCMsjhID5^gZ=%pXpxnR1X!gQN%;|*nQ|}AANyA{0n4X
z0K7jP`)?^pn4r+j{}z9K{o`NRefH@^=)d`i)a+k!Q-nQ%uOjy9lZ)(&%;4)x?n=0*
zJCKMvBhCe3ED$zPeX5?WLo8(Rk~d&Fpb89CDB(MX7hH<Iu+O!9K{f%sN%&_#tRhYd
zAW(m5xL878K&;GjjHe<4kc}cv_qm)<l9By+lHd&#$pb?HHh+ee1<xcS*nHMIU*Ut-
z1%BL7pK7+Nh*ATUNyyJyr-H|v2=Yj?dh@xzx4Z9jeAEa1HFN7R>J}%@p3y)x%jC*q
z7l5tsW=(UcNNU8=j@a}GDSK)g&=Me#{Ep3MN-{M+;xoYOPg=rvyE#|!+??BGRIUQ)
z{PW7lh?yYO8NDiDPoNmanp}D6XF#(gR3)n|Nff4*^yoeyq9#@uq6RHUx`MF)lhC~+
zRvK3g*i}G!LsBK#QC1m|6hdpFO@-YTO1n%-=p>0WY!Q%k2_Kc3WFuKbWZoqPaV6PB
zGMaoPh}D2OOgpDUVUw7lFcWhyh*lOhir)3rG#9Pa?kT7Rphy#eumO9TQC^U1MR!w@
zDv9bv(KKrVdLv1-AnO5fRlwe;HMxFTmPf5b32#jqGZ^AjlV_6DWU^Hyftkys8ZZ&H
zkPNR&eC@H<0w}<Gbp=^Td>v4OAZbI^0o_Z4s_4c8)e^F*aX&AJ5|qotMvzP{V{`XB
z#VC_?AiInj6rmER;0p_~F6|5>MI4t{w`4qOwM!;eM@(KZ4vJ`Q@I)7lO_2h!6NJhm
z0tf`L3P>*rRuXiByG}?IQS*$DqOixhfaoNc4Na4fs%4V|5eSx27ab9)qq<gPS(OXk
zNRVX2ZY?f4Oc<!N#@CW8J=JQbCr?1FB}qB-R+DRun<ns*SZA^*p`O0YXrB>Tl57F;
zKRI7gR$w%-F3D$PR~TQ}Jbjkz&ZC<Ch9I^9**om1CRz!m0P9&j|2&>AzoY3SZdB1}
zO{xT0uBgrgSqpqy^nr^xtW8KYVzmW9Mwk_&vyw1yb5y{LkK2S*pAn*?sT?XPVj~Nn
zKNUpf3ww4q58^FJsEe#fl9`+$NtAw<UJ|q-nuO{z>FJshB6h8a)({#6HkX8h7lK#|
z`gy=SQ6)*=OV54RM3StddQkLzW3_W6zt99KNrpB-lBjtm*@RxOW8p(T$z&T5n_ZGh
zN)krBl4L9KZNe;o3f!!enNvbulten=Q}aVe<g11>9twaJ0Jb!9O{fgf)wo8H>H&L^
z$+?u9f&{OKCZnsugf|<5-%7A6T4G_**ITNa0j(^-2kdEKV)ku>PLQhyyfn+CrluD=
zjn>LxulKl83F;-&L6RCtt~QLih)|l~J$faGMj(VlW(~DOm`J_SWF^^76ww0Vfa{SU
zSJ%v+of`yxN?D^;7Pns<fpRH`8nSe#l^{t;s?~0s7AYRt@0?6)fz+BbbBHrZYAZHB
zSN1Eqhuld4t;ei4lJra($`r<B6;b^zaa1VsAIXmB#}?-b$ZIC5=$EtR#(6^8maSee
z+6rQQ#97rav=XF4CPmaGDn<8Vmz;5z_1!jfTH%$)#Gc7Ql9eQk9D1q9#u9DBc8X|k
z^MQ;gMM=Ku1+kIHXZ63}pbcm}V)G2#RYYCmCxUD`+DlEgl7u2!g*O>>rN|+oR3;xv
zVjGce&FnIlb_vKqk*pwRLC%ns!nYtyKy@AMnIhOR+rye(LDFl+SH)qQ>p)&5q7j5~
zxd+5H7wMx)qWX;PB)*P_s7aF|-=>@(>P$`v)tCL0C~!_g)nkn$8%?mGuz-au5fNFC
zZApYjC&6?ydkJAQ-bKV(m!zV`l=}sq>ROPsq966tO-9IuZd2ie!M7eQ42Xmdl2Cy#
z5!qS%L=vpStPEZxgz6YCAj<-l>J@}a5%)4?CCixJ7Cm-9fsjOr1o_mvj3OMA`<A*|
zGd2;WZu2PE-M-(l?IljHsGLNM0OLr5L1{^}k{|@W@|YoGq#+oPTTkURNi;={D=Cr+
zR7*`VhG7;l8;fsi`WUbwlWW8FEk$e`^|{8iif96Qq42FB>xk_nQAfxTIZz^S{DAff
zr^<ZIN|BVqUYFnfxSR!YW$=^T8OR3kCECc%LWxYNCFv!FXd$AILLzKLEdydMp%(aM
zi<>K==^0(^NCMmPETYK>9M%g`We8CcWklzU?G-^K1PPHOS<2!_zZN80G1!Ps5!pCQ
zu!N?;iHPxvRLxl2gxvI0SDqXJEJ#C`Do`HtIHHD3wjM!|X-btxJ(0w^!oOBD+XX%2
zcO<zg4AMLn$U35)k)JAtsm4_Sxw&F?mC$m+YF^HEddR3~Nwq*^L6)NE+pl|?wIrAk
z%T|;GzIsdDNwA7!J=K*USy42w!|tp<|M)n}&626IxQQf$p7zm;cg!Nyk&qin(gArQ
z@#Bnm{j)yv88!vo04^g|pZrhB>ZZbtN`#ocDs@H{QbMc)c4LUGMyQC`0PGSOuPwm}
z;<M<>fIZJhsYx?{1W6GajU1QXKXxCQf3hxL*k4}z0^t4W*arBUyPrqE{{mbB|26RM
z0>4aQ_TB+*1Fr$U2>e^X-v>Sf{(z6zzyBd{^3$n00RJBFPj+?qasR!wEAuliAU`ep
zhza<ki|mU`kedV{lY=D9WJ%~X6tF3)h!{TobGpxD1xZKP2(r>8gn1*#M&n-?690w0
z*mUud8=eC(Gs5ovmYxTop0}cjiP~*!00?qO<X)iEA?GcHQRqJH^Pgd!vbLH(cVm9C
zrg)ZlmCw{x(hL(rG#;}s_$Wy!A!SGNnI7GMz7k};V{JW6uBIrjdH3qhI945vR>U-8
z@eBZps5&EbM8$v>fJ>Mp2vQdiBq?ItfxP;;&us)*6&9KJiva0enSUS6xuSuVO``rx
zFdU%h@g)gy$M`Z%Au2%<PuaTwf~<EKuG8h5OhS1OITNG+kr5$5miQX7Oh~6dNe}_m
z=aNBQ2|@)jW=P<bASw_kLCOU{t{o-`$Q47A&=C-sASFqaWIie+<POBiQV>lduPNYh
z9kI{O1QE6hR7T`TL?DFS0rtDTQhdU3Aj`Z{2|AF21gX(2BpJaAq!whI(N6B>M@0FE
za1tpq;n_15SeJ<+m1HW*a(PaZgzOYr31q!vf|#f4ZgJ{h0%ifpNRkK$pCL*RI}LPM
z04*^<QW{qW>`DNZxL2Ny4TfwpGNz9?{*sYInUO;(+NK(Eis+3*X-R4`Q2=O!Et=8N
z3uK0<0#bTRQba7SQDjvvGD0ZGg_)Wu$$xUrC5$&3e1?o9RZrQOnVd4pK~e%$MQkz*
zxqwi`A)v@Apa+ntu(^{iY%<+Ll!g#JI%jg55h<b~L{*X#$qXW9vdjpdNqV<dN=O<=
zqA;RmiD1F+7&AYz?n$Jf09T>_nYt^NkWjI6i0#&E1bJM51GA>cVZ=<75XL|HJa$h7
zDFW6hf-1nBNT7p2q+BX11woa56CxoMKuw<^`-n;r6U*F=wjun;nj%U0=~@)AvOJ5V
z%XpWG9bmaDqb<%F*_ZLI21Be0n5-Q{TO>Y8i2g_I(7hr{xpVqtatg?G(W1_EK(7T^
z8~kk77F*6QIcDM@5Pe2yNoqb>|L5&#oh4?_gpow)gnR~~X)pqz(<dTVBBIYbAhLj_
zK~W@(00^XQS(iWk`-@(&(2n>F49|!r7Y*F4A=a8C1JyHd)G#y&d`2ZfstXf8Btdk$
zFnL{aCS(;+<4hI?9~^dF`ZiSp=`uoR#5kaZB)gVvRUB^hwxUy#BtvnQ84$vuLq-Tq
z2pKh$5cy<^o_1CeZ#;673C0qd-CT(kTOE*WX`gSpk)R{O7vLcYfruq(qw;`Ah}}5<
zX#S*f_D}+$GqQthG<C1YRu+e(k=X4(r4v#b{IfA=iw?L+$laJn2V^h)fA-$($F?j>
z@A{2-S(m+cTu#KfWS-2d%DT0?TbQ=o!a@kO+gM0`w!{m608faE-%uMk#(;RjLWutX
zgbb1eLdXKr2w`MH({?vhcU5&&R$eOe+)qU8z1L;VIVKNd?KqLAPi0m$O;wl4Aw_1K
z*lX`S*Ie^5<~P3a4Z2BDsm!runic23uPYi9ohC#IA+==9G~&P?an;y7cXPCL)g}B~
zKs_}iTy>_tK%|67Arte6F9PE2*(AupiZHs-n~X_xsCrmcd<B>EGxzTtK*x9;!0i|t
z4^HrtQj4R6ic}+BNGOCMkh9{P;Ez4(GJZefRG|j42qHob9>LhfWMj5fR|vDxTm|rv
zdEGLVOp^+}nk<Z3O;hk5HrTpR9kf<}%?%dNa!xGhb$vgFwczkCpafJ+;y*4mailsZ
zCdVXkSzC2khbLR_Wd)u{9>W6Y8&Iu{%~PKXwVHlCo#I@>TpdL$K;pdxox&K%t=+eq
zJtcaID^^Ss#j7G!@iSC!Qa*S`?h3LL7^79f-U_2iC#ESKUnkRY<Gi(-Xk)?q6?<pk
zX&722-;gEIC8$wQ?+9d63zCfmo0(~nn%^mkq#0Koh$s1q%R+U9nk&$tLGUx++#(eP
zu6o9MnEb>jln5>tl%>zq)~4zzp$NEu0@Y>QEPz|~tHVgGrrYZ?GCS~|+Uw#2Ef-}E
zp^_&cIP|zuLZuymzM>(+P^oR9#-1rJ#rWn*uFJh<?7Kt3sal2@9(7_swA%VC6?}z?
z5euqMFgU72N>h+oa4yk^P@*TtnJ_6?JI0ehb(!X7P^Gtf_sP~QnZUVBaLz~oArr0y
zjgEW_xvfYs`6#I#TE~1;ZGMb5*7i=&knlqxsv;W5(dwfxnYMiupn60cIK|BqaTH^a
zw?bKDG{Q1BwY51IN`StYZyW>ZdQW=kzxrU~(f>r}e+#(b6t`Pg?MGkyu0Lv#xrHg*
z?2DV`ks})iFWY7FR=q&gYZ8%qU9#-Jh#7)<e<PS}d#8-n2e}u*2)M#{6K^=2hd$SD
z_|kr{>`OZK&(8iwOoR9DU}@8*z`wct{G>Jm|1$8m|D+rD1@>2fzwjsgnSV+OP}%1D
zld^XKULLTens!o=<^9*6RbNBUu4omih*kPL+<)NZ!SKs~%OB4U{dWBfy@7@9zyf-m
z$;<0M)9;E@ArmNvB9uSvSUmunKW~YKLmR%Z{SNT}X<^?PxFk|<I0~<`!kiAw(${?F
z;dh;AUi&#V9|T3JF?SRw%ioPnD$6r`g-xu6*@cB!=BhCpf8u$ZvFCpM?3a1%xG|1?
zoSBxcF7I^|1J%T|{{G0-QmOR@kW&E>T&?yVS*oo^d$BIB>vE$iLPZcvOH=^o?Dtv^
z4zRuMuAO>I4Xxslx(tD(pwIFT4y{L*Lz#cIkCiIk-b1BIHqAl~Ox~s4H_eDgi`Jq?
zVtqM}v|8bkYSqZyXWwRhaqW{>kxLOl#18+-d8$-ID@7`J2nsj@HGL6s5GwWM!$FXG
zXs5gJo%QAGUZECnLJ2k=RhRqtOkm{YyS+LJ4yA$^*jfYwa<C#TvaCV6Xf4{DVqvIx
zyN9LR9!_xx;tzVUe)+$Rk17T{`yhw9S)b{D4aT4ap@``_7FY4!;j;BXtp+SPr!>0E
zv09hIG1j*gbeXude)jo_Vga#bUOUuSHR902Tz#!JfHFIK<|xY;I9z||4~G-0RDS*F
z4t-w*RfnQC*nKx`(Mh}>`%p#g%jB>OJH(e8AneOWDpjbpfLr>|y^g`Vw6|Kr{NdlS
z?De0S_e$|lQB)5TsxHIF{03{+()VsTuu*NS9=?A8#H&qklV&Ma(8$)?WuHS#+5-cg
z)HzfRZTvCj43wyq<s2XeTHScv=FThcQ=5-+bDg{q6!LYsl^^HJgh~-dQ(k_VTdpmt
zsJHQ<*Do(Gs+#W<mSwvb`%<c1B4v5z-d?w%b6AfZS`{Y-pi%{`CWnDj1H!QR>XtEi
z?KdoQq!tX!!nD*``WPA;172`dan4h9S#xf#r4Aqxan^ro%x+!Q?t0iW-OWX(Ua#GH
z*c>U$4xIPPo~>v(e3#ttcZAu=viIoq8t$vfM92biUSGeAO@IA<wfhp5{asP~vaZ_r
zb$Zy~7ik$MNX^T5QnWf7tB3nr=3_YwdA9_zijtfz_jkC~u6I&D+fEL1)3nzQd#6)b
z`lImDe-VneHXN3Fhfu`fsJL2fQCrrm!@8pwC|=!x^}qDPVb79ke#^3$=emqVFNj!#
zz=@*E^Kh~RX>W`NwRqn0`3^zBR~y5J!LPP|+r2m-U$fdJRPzUxGleWn<l+v;dWWmU
zK3ggxt+kQm+Uv!^K+G!keTcD?g9>wH*|RM;$YFliqPfgHI|GS2%Pu<jV6XM14rixy
z!%y=}hjDy8PL_Sh+3VLn4sPjp^8uWIh};;n_B^FJYqPUGpd9wpr4SV=xayIx1Yk?0
zI;Bc@c`iHnrZ)MOJ><rJWeFJHm_Jfq?y1${8eCuMUM`kiAmeae(OM0lcZX+Y<G2FB
zVf%HJW#1D6OYO;<uXmXD*S@7toq_UI4ttLElPt@Mx@A7O!#K9*DC#Xx<87^#Yu_hP
z!tx#0OOspt1Qx=8;LEcA2EjQf2lqzJXK`L$Ig5Bi#JsX%V~gr7DCZs5&aNmGYwPv1
zQO8pL)AokLnsB(J2wLlnan5yLy!^eQ*S;w~fmt9oJ6V{`p5h+xAUxo~AIVz!0$cW1
z|JwJtB>)XtKJ#a)nv_a4P6vlIdr_Z8YF+kr!O1eVsQBe9K(M0m5?+yUx;baRw0}bE
z%ZK^1wZ8|v0sO1LM?b~;J_Y_Y;0`}~a2McT0e&3=dH>)3{O8MO{&zpS=KpvrOJ#1w
zbzgk7Sw#}&;D-Ve1a)Y${GCdDEu*m7;Y2G^yy>}Bp^idr4a}%NEmQPM1B5y+{dXOa
zEaYT|;Trid>s+Bm3x<A{p=+soUN1SCu#NAh4TO1Y$NhO%BrUvb{}%wY%^=j$80$b3
z=bRP)_}qJlxT9*t0S34R&Gp9jDG=`PxHz;Hyjz&M%ZoKvD~KnMWjP>7!B=AwLcx2X
z)`oQ57Kb(lW)!W<!RDt4(g?MGb<P<}QHm#gA(P&&tp$9kWO2+=aj7ECf7Yoi#wH}{
z0xk*F8yJgJ+#%@WOm0A`Imn`>9ap^vvK;{P;^c}T+3sJQftKQ+=0FxeG9ewbdduP?
zD^le8pd+=|!Ld|mE50gB@kSp5tr+X6Si7YN&Ai<I<OtQh!k)G@uxdpc53N#Fja68I
zP^rZcmI~FnfQ%x>Y(nLRwyGH5tqUu(t+-iX5=twC>}XO&n~Ek&lhsCXqk*kyZ3ksn
zt1*2R5VtG<GsVDM>XJI!ixrsLV`DHFoA$FTd`qtdp)^XtfF^2yZaL_Tc3{#_ZCz4F
z4V5e)1hRS?=Q0tCP0s3B4nYQ5cm=7+&X3Mr1FuM>&Rho(L=Nn=N=-s>g=TT)@de4t
zeMVgl%NeH-XJ-r>&3GRuhp~^K4qP>`r5=FB1po#h*?l^dg;7&*1v`v6L8QEP9)^W^
zgs(V^%`R04xl#k5WhpXJjNw+SXbQ#Vi>sA5zxq60j+I&=hf1g!mlbga267EXl$Z0b
z3Z+&=3O+*(jw*p74s`{|(`#fVmbQ6-hIu*cQ4AE;&+HKu8SNg#gDarE;u3&IeI+EQ
z%kOfN`5Tq0#rimd;Bl=n_+<9fPSFB>u=eDHWnBggY-xp>^D6~gUbTuO3npj~+}?ob
z6mPx`THK9*i?y%DMQ&!c{nqQj?*WL{MHOG~%P4MNjn%n%`8SdSb+|_HbHx=$x%S6q
zno_CJ)<HLkF)4#v_63I~p?1y+BaK3>igUr%QMYJFFW`6>0<$xwXEWKoq$Uy77|q!o
z01Iaac`b-{&@InE9fXRMge#U3@6@C=O7KXXh&Zwj<Qu?~tx?(Zc}rb@C1=&o{lfkW
zAVqKZ4epw;<+43ns!~PKVyW?3h1z=zFV6&jqu)}cmSsO$=4^?WG=VM*JHLRTvpu&Q
zmigf;zB-Zud7w`a3TTsTFVQ7UTnZ@U11KsEX2lVxK9X^kU9o`7t_mS}Xu#(QpNyp&
z=CA~HZ;mHVQL&<P*W)k|liGUNEMuQa_1PqV)O5u|SjNF<`<}LjUZ`XonNnrJ18vR2
zsbytg?90+rC?-SUaAYr(YV6sn#>&)EsH1@;NF~fBd7^;@<Od<!vkTxBFh;#Sd%HmG
z`Bq2CAl23k1!~kKI4Gz^#*hF`ixpy};==*xt9TL1+F=2vz3mM;IaK$AstW+=si}fE
zidXYnsvsfTXID&`z&T@&SBE+aN>o>%RwP+Ah-70L=gydd@sPciUk7?E>w*SccX+lF
z>*A0SOlGZQLiH4H<G^b%pi(?|#mxm*JvlgPS^OvGmrb$a)Gd1!#%%S9&kNA01wTcI
zl_J@)HnXR+c0Y(NOk7wZoy#0NIJK6tQ^W!pQkGz{Sb)j=Er<6NQKilv=as60%P;4l
zEqjH!2XrYwn-;J>0<(iW(}?-U8tbx$*<LAHsMRm+P=sW_-a?9yVnNahwzpIZV$}xo
zce0Hma-?Xn^)W+k6mb(_pOGr08)xD$DYYyzGzhYsUv&YqDURF=P78i_U{PaoH@+ZQ
zal;byxL%`d>`@kJD_@NvoxmmY!D_Cs1d7b>sHVjqGt^AT%lfQJ@q({{QwQw=-cDk=
z$Mxo(bQp)W7W?9EWTAYXMPquaP&>tGf%%UC?viK{cvbUp{)lBePbJ68G!Pr30a*h1
z3zNFK8(D7$U&{8#1GC?Wqcn;u$^7F^sM#3L*ZuJ&2JugzeF^aX>}>%4GvFQI9QbE3
znTCG@_$}bC1Ah&ZMfm5IYu*I@Q>;+Z&%-kC+x*E$G<^CqZu{q&LoW@i(4Sa)rqW$m
zDw2ducnFe&d|Z*OqLZh*wm6lCW5H!RoTVcJCfmW0j)i7UG(WLS#FzHc63vy+%#L(?
zh)N1oGwg&gd&+C8iZ{$BO3aG8p_NyUz`G3NhrDxv9kza&v=xUH7%RYQ({FxuCI|3Z
zu{Ql!d%@X{CgQ$l_r^%q2(?C1{vv=kp!YIiNW`1^RU>x)hlqdd7~L(NPoAO=xp&=`
zmVBwE-?k2DHvPoZIYQ2?<~^(S(3Xzjh9jjXz*LdZin4sJR`6%6BF+kQHFLV5=UL6+
zpFR68Tnj33yBxa3Qio|W?y4dwFyFqqqmX_803ZNKL_t(;C2p!ncI2ku_r_2t-2u>*
zg}o(6P57q**^&E%pDJOF<dvhuicd<&i}poOlQqc3@F?Ec!;FC`RzybDdjl3-5u_AC
zRx9$jFxYAUR|<Zv$gN@roagRkx~Dkux|);eD=rnJC~30$?{-R^g?Yng{fJ$q&VhPT
z$$dqV5+6H!@|2asrvj6wtSWA-NDkzUH|c>~m!d=fj<hm{gG-TgT=2V0cwz@-O$EIS
z)QuzT6-`3fD9wixVeUz{AkK=@hIv!%5MAPO-%1z*+7+ZE+N+NF7^Eat->(!LHCDpT
z4(aKJlO6Yhn;oU|NGM2x60Mkyp9}G-;HH8b3Q{0u2P;{sWDBYE3^5b*lFiNvJ?3@+
zN^?McMQg^-f{5C|jU9alm5OYk76bX!J8G}g+>kFDCfU*ZisXb(m5>ydEd@6<_MF;r
z#tj|~4w0GYB6D<1do1wi9OA%dW9F6C4z1cNk}~nq+Uh4L;wdCtvZ8ky7m)UKftCW4
zjLe?tD3DhLH)ZNhkX%e^BLgJj(}JHo>B#!d)rzYL|Kio>@p7zaZJFDGn<n~EjKM>s
ztQ2a-O=>_i00B)7EtecOm>hyMjD1GROn8xR9Dc3`HtVJ$(lDn;Dkf>*)j-QyvEq{0
zJ%ndg0IzHoR&Y~sgY{`{Gj&pYUV!%NIcM|9&memX2o)EYPYu|J&pgAOiXRg3;qo3W
zumhJP?$nW%IoE!!FQI5%f+aEGo>h|-x?cA(VE~tf{FxL~51Wc)lZq(Ez*Nv{zto{;
zN*xvV+9H*S_@jy+TGDk9rk?a7;b%qG6Jc9$bIW|K;j1)*0Sc;$Y%NV(n~|`Bnp;a{
z-Y7Ch=0~2Q18v(doGCtK!gi62NGDu8S6rHCn~{E~_|%h*E;%j=^O{|w=)U4o&(u25
zM2upErwX`PC{0C@>BvsU1wYTk7Y8=+1*dyK=5wCjNX>UUU5YCu<iY@U3rJg9vj)rZ
zaJk%I&iO?S%-CurWZXVbkCiY?xE~75>?!M0Ud(}<2l{KkpdvaF2gT(`@(MYS`;KV?
zAs4#sNY{Bv3FHe)^_BLRZST>#r(wh36_=iJKFov!&Q*M_NRFi5vJBkpO+LUmC?nmN
z@HEUZGIj>AhoL8LDn3nw$uq7qQl4`s8GGJedgf@^3NP<}!*=qNGSFZ93={1%BV(o3
zz;L9vNl>2%Q^6-sI?9CLDW#xK6gM}d^@@Ef)I8Ivw<1w#wt4FfxQv71lB4t;(~08K
zO!IW4@sUzb7#kyX$ZSBI%ce=K&Vr9+x9oG9LTOmirGb4#=4yakwzZ(TLQM{QCPeVb
zktc^}H2`_-5DIa!wZ$FSkt?O<!1UCSwcu9?35ruS4dlVb&A0(A2|swsXh3S)Wg1Pi
zqXl$4p4=(TzT&edudI%R&yn00VigJvQ)~C_=3qgcITJ1-bzMxBBzww62~(lYHaGlh
z&#D!<u!5cSq+^^tr99-G7MN{LccM7$*|rNHxM;gORitLZR0yRd^@1h?5=$Q$eStdD
zsHen&Xn;23Q>NVoa-RwFOg9HoZ{u|JokAEp#?wL^Y@aKAF*%mH_)qGLZB4Zd2er?V
zqPVnx_-e&(p%6z;+AQn!Rb^h=T+d!j7kcN=(~3-q_=%l8r8D_HCX@V=%Zj9g#6-7s
zP#aQL@ER?^QY&!|R!~z6oQ+w4KnfrveDwzEx`HV96q!#7E@hMDXoNZkimNbX`nhFl
z6{(lp`V{=s-B^P&ZdBqRsCe>`SSo)CCXb-iG`?>uewqoN6l7bW>6qUs)TD%qg1;0b
zd+PDwbN8&T6rVk5vu4)@)D85zf*%~Zf}qt%SfWrewMMk<m^Ol+2l~AdyhF8P?!l#m
z8~h?+lyDaoKnv3=zW#+$=QYz%kU`LOOa8)ho)uA)B*>Ws*uuwO^qwDMmv&aj>-tI%
z)%sf>f*T{{SjbCepOiw)#Q^b@F2!s;a^3Dg>P@%zW`H*xS6jCRp&dQ7H6Ot(+P&gD
z<>=M((3kd$Z(jnuKVN%>Y1{s7;6DNWL*PFE{uc0G0)KzG=1cqk$P&#{VCN|-B_Ds`
z8vw`97IQpyl{hJNG^e|c;3g}opv`CBqn%YimJx7*)=b!{Re{$+y?nrfpQf1Cm$ujg
z9&8Ks61+Cd_YQwonrE3hddgj1S?h$_F9*~?>96NlMDJH*SMZOp;%W&yxPNZG>kT_u
z-m`znwEMZ*76W-t_#77R-{+tCIMH4i7&Lw}kn<5&!;iZjATZqz%*O?JKGS~x#!EY_
z*xV5Y_9o!?8~owEXGkCN&e4Iz+fgnYjK^$GD>BS9+n(Lq6Fz^+$#})lI*{{*?aktO
zx7-4dDsG+^{cK14iGX+;BiD28T4!vgeWWydN4obqr|YJT=reGGXTCQRpI7``@tO!j
z$M}X@#+%~QoMXw6yefn_kUOjP8;4*8Plimhvy?qQJ8D$iG!ci6{b@y;Ofy;$5I2M5
zib}$d!SwK&jL#kWQ;>NkP6>CS#+oRVsxx6~$sI^7xZE)w6@1PFH_>dAy6+jTi_3Y+
z>5|qjpmC;oQSoD47<x+DSzsh+BwZI&nQ<3(I2hmdh!)r?ZfjbV;*=U^!Y)uZ5Q-hX
zhx<!kjcVft2hmfG3Ni=A^J0g}#|Ef{3&HthAp5)&h?+gD1pTzozHs0p<C!3-Wj+Ci
z66Roq*^-e`t5f`xaYvrqC}HoAN15ibkg!xoH?`;uu!%|WQ$QL!$c4Qiv1K~R_$yGC
ziO)0Lb5HH8{rSW}lL<phT_0PVcK4V!L#Epmd}`U9XPN=i%&aqEzh-~C;P*4_Ri(Za
zDAO^!OU0E;92(|xPYH^jj<^!R?I+qg)2@s~BT{Jh04H30b)wl{(LTvE`%2h*#y2x<
z4Wx3+xSD9M5@GBa&JDnIHBfG02mYgs-)H0!bkV?eNytZuW*5m_1<ClmzrntIIi6T$
znyW<EZ`gj#*u(KMAZu__#osJUdX@0cz?Th|-<pwa!HpHafAx7BEKt`gb}`Xh6`FRU
zz09<`GhTcH+>~C?*y|a_K+c))w)GVkx48Odp}D%^?8Dd1`4^lX1#&*+MZyYKT`eim
zJBULhUGF`K@JYdMJ*D$#T!K8OiLkFYZ76F4F7aUr2z$zLK%)|-8Nbi?(Noq9^Kr)S
zs}*<(N^_Zr*9Xf?`!rL>$mOpTq!j$c7cp@f+(v0GD}Gu3{LWipmP0Ufr-WhAbG`*2
z<PD^@8W_K1F!@PoF8~~I?I}lA{4YLz-S;M%?<M@!lTMqZlJ|<sp(4)`u_SuQR_tjV
zEqFv^cQ?_@nQqrJvjAtuiDsP%(~7HOlOy3`!|<$SN~V{|R}<_8jwWMp`>U_>eG2Z`
zCC8VJIab0D$#?b~rHYIl`(WDNGW85cfvgS}9rfJSzY;5c&a_)A&Vw^ht<jlgn+S2w
zX0FKoh^sS}eW1`@C}9@Lx)5I^!qBsQH`DHB!hXf>tKd_)5o`{oYwr|4^^EUi{7~s0
zH_TsE+(oAS-32Tz@AAW2k(4IdcA&oomM2!vGGU0M(?l~U(6E1}Ak$M$lZ}Nr9dkKk
z<Wi8~go}3t*-!Li&u~(a-6z~lTQ-*kKX!~q*@~<Amf<wf4jJbgW+G|4;;7Gry-@NA
zS8pjY73$<EtAY%k5-PqV;?>8z)dX_B%OhPfcP`uwR@)sO4-0VJ-_+orSUm?f_Y7Z2
z^cRV6X&R+&uEev3IkZf3L%m3Z=NX@ldHRitj3Z6Sh|Bn)r)(-=P~y(^2$x!N*Ry*o
z6ZTIze?HOgDl)9N_=;(sw}T^}R{U5XE2K==f{O_^fYXM2o`~CJ+)av05sobwE!vW9
z7viW0nP!!6<BFl3@!OfO>zUqJfb?k@1A9l_EYc3+n>@Ly_}!k3>zUt9#J%DNAxDQu
zPr9WBXkM<_jh>Q~aJFYPXZ+O>)9FOmC&KJW>tfjfKGAG5ev0Jtge!uZPq@6F@v{X*
z=S?AwnQ&FD$m`gW-ctN75kHNjvxPA6+CsI7W+|we*YEHHA4%zglhvNpcE(SM?qts@
zH5BO>r;49DhBs#1F5$MBaGHr*Pu7+pCj4XoZ)q#BfOLv0LYz9bca$(DWG9p+)9gIu
zZo&_V@TAgzI@4Zyv=@q38{4;M{I=jy%l<UuC&ibJX<cxsV_rL8CeDG}&%`;=Y<sqE
z7yOivU1I^cK`onSw!&QXRwayzZaw8habqDoRNOPuVNFNi2ch_m-6+(-Q}6Crf0~Fx
z&;Cw@63I=2HZ#qTkkT`>1(`E$Gt<n9PiuB(GtDGuwG1CW3o4QMND0%P-U)T~Omk%K
zt^Ppi!Iez=VsC5Gm7spjbSUc1NcS`o#x=XIJ1Rn%9HlKVRcaH+>xA29j^EF;pC~dO
z^ZcJu{5TOlQrtyFMo&7oeG>})a-!W;e7wz*cfk!$xcw{<uK+yhbf&qq045P?YMIV@
z_UoB2J><^p$YsM`q7^lq9U464I1?_MN##rhfBAq1<Bf6nv=WA%eV7SPM&i)2|CQHn
zmWxG(6Sd-fv<&wX{?eoW_vgLm$JqlO%mx3fB2y&2_3G<Nca?~bGNgv`O#p|6giP7a
zX(t646=qL5vHBR=d$do4NiCb95u9`k?}8f&t4}=Rnfckxg)AU&!+dW6k3VxB`qF;!
z?Ms07OZ)$|J>Wr3#7`7=*)xCri$EerTnF;kGVAY+bRT%86N{vUZJ@rHh*t_jB)xq|
zMR8~d<ToKs6&W4*q!RW8zipV`ti;QS?zM-?FYPC@iSCD$c<Cu`2-68b!9S{mOWXX)
z+q~4dih=p7N*FWYiIN6@Kz>JzQFU1eA7fzr4g)C9`Qr6*z5GnT^DnseXJuCy$m)UD
zRyGItzxdVh5qJObGtS2QJo?php0Cb%`eQfyKt5S9{rQ>raG?3!o6Zx~XCsS{7}kIN
zU*dn_oTp<ZxZmT?o&tErbekvJ$S{v|!E{M|`-1i6hkUCkc;7P3cX+%5aL6W-?nB&}
zE}+|I`X6S(Cz0(p>K7DT6Q&bSd1IM#KfL+c10Fof@U#+ABz?2;S~*dy9pTbGuX@E_
z4m2NnH2Ot0bH)Cf2aq=rUsxKU9%Y&>P#dO|TRy)Kriu0Yo^-1cW{>vZ>PUa(seaA0
z8u2@>WsQx5JF$LIslFixhc+Abx0QHt#qpD#;r2w=dNfS5L#EjV#^Z@@JJMe@<TjFy
zGtK=>92NiZ8ILdQp7O?%?l^pkOm{P3yQlkunf8Oo@XZ7GUGN{N1;_5O07HV`7f2QD
zJ>?WuPXH?2<48JEm}gcG6S8d?-WDtTuo1LR#OKR1c9H3gik}_14W!$daA6Lw>=XWJ
zMjm^LtH|C{&Wy=jcMZesOV)od(r$b9?*_(Of~c|0=C)Y+e)i;5!A}!W&UyN4d)5zk
zbl(;BcV@Z|CpI5@%K1_tCq~i@R<^%#%)LY6gx|NMlO?5B6*ncik7wME1d#wSkXIfj
zg?QnZk571^h4^$&=Re{;4ca%xcFr8~cMZgnN?RWDW@*UnIZvBAe7c(HKADKm8q!fi
zJ`<z{=C)z~v}H<9xE(Jz+9_cVIrdCvEqOJvesaahk0#pfNGsrPuNeJ?-P|&sjC7AG
zJ_WQlZBUhlx$hY77)TTM6)%}EOtc>c_V)$H%<6F>j4k;j(>?>|Z}yji7?|!(^dD5>
z&@df&rlUf0VTEPf!py!lo#?+ENUM(Ntl-Cy)%Rop%wK&T4%(jWFDvqFPxmlE*>OBg
zxY02kd4}_u^=H=OmTC5s;}zTE3r;Rp44osl2|s?yogch9-=A{6$%L?CuYZ=`xueAA
zGyVIC_!x3;tn?ZWhc&^EBi$b+{6#}L59D=AIvZ)9CH&AbpN@p>Nb`Kf_U*uY(zAOr
z<Huy|ugikJYRTsgnnJkj7~ah^&-Qe`D>oQA1^-BJ5-6vUx>35%WfcYTJCzb9;twXm
z`;qic!Cz#XlW$c6=r!RVDeM~NyAAa;lHXLqPH|V}xA?8%KDLzg*haSBC~%dCpS-dk
zT-yU4Y$w{^O|(yXw!aSf$f4Dd-Y@v=M7x<-?Gtf0Vf(d?`TU68{b!thvg7EZmgGCe
zO(mRqs&|yqP>wt13HTwDa+@c^JAC-oOw;T+Wy=P|?cU)BSCM@vxF3GPn~#MWXS};$
z)7|399pDpBy}%fF)ACs-r#wH~a};$c5Sek~M0>I3>J6pcn)JsIt~mbmBi^}MbJ_L`
z&jMNx<3Z7+xF_JYrYq~_KyC}U3hmgCPmDci45~=&BI%YQqh*`qRiW7m<0&vsv>*Ok
z{72tUH1Aj9<%X-<U+4S2VV)mx_v`?MKI81kjNewO_hjEN9!F}E=;R5veZxGT^YjWJ
z5gs~Nd8XqvTkY7#fnG+|&-Wa?_%7d?N4kAjSmzD8_M~=CclC_3`AvSbjwDUQ_64W=
zW401WM->eYJu3Jy<8s4zl8F01<QrFtb4P6RFY%pw6RQs<!Y6<W<hKCs@@PKcVi(By
z5qHlr!7caoxFK%>>N9RPaq>xIIvwcsiZ$mvJwN981TeDsj)7JAc+D>MOz{P$^GLIY
zxvxmx(ml03qSWoZg;yL+@F<E)2o3>)Td?2+2pWP1hu|8VKyY1Pkq`*(?he6%ySpy#
z?kvHb1@>;_yTAMXf_LtDk27b^(lcGv<yAd1-Gob?8F;8oB@D}oh)21E5b!|ihXWR;
zubgH5-zm)Qu`N;}U)<5|mFgdwfdKnV{m-_<<Q8)E-)U+{{Ltui;nqj6(hcm7cSml#
zkar`mW$Gac`6YBUFII0p@AgZ)6FQ)}e&w9izU*+`{iYcroOn~|;vzUMdPbmMIj1WG
zJXBi+q*+ojJ0THkiRxAQh!3nHe3%)nfsnBG|C+PDF0%lysA?&oZ^^r}<<)LpV>QXN
zrCYJjT-&KHo8@@0c0hblonCnrH4ImrZlvxBsb|mHU9gyBFNg~O`?UEc3YvvnURmWy
zhb-B?1>hF7GRy_$kr%5kuf}puLxShENWCM%7?M8DOh+lc(uk}t?F&56=rklu@%fmX
z5_}!ReH6O-fJsQDQzz*XSES+2<N3jPdYqGNv`5cUI_Tc`TeZ~y3mU!ZEIJAPMh;E+
zx^(Pm7nAMffUIBorigdP*GGAX-zFP5pme%%-IcU1U29|Mk$bEl`e`!zIO%caoi=j8
zOvGqxh^4rrhWNWKiC96DIgWDkMUTAJ<)6A3c@Zmu$b2VX(qyTW`t>qA!PWkdqP<B&
zdol_K2yLV*;?b0ZQ{~3yA-PBONG|1A4L#S&B8RO`!E?*r_=vMIiI$}Oc0(CR@TrRU
zqQu3?Z15-q87hQd-eD!N=D+!D-^un-#zES;uKs1;Q4e^le>pi|(Y>HL_|`PfV_W7B
zMOa`LjFi{9?(8;b^+on7Ae8pdI$3ZRl4Y7=VYJ{Hy!)uxzkx^B`aLXDV^MBBJYyrZ
zg0vpfY=%3XXz}?D>3eeK^Cv`#ht<8K+Dwb!-LL-b=W$L-Z=3H@lwG7D&EIZwEE?8g
zR;?KAEHo$;oiiD2kCWEQZnJ3`Er2B!H)R29SeKNhU9o|82B2eWOtbj!M6^QAd9fkC
zTs~3wta|LrL3-;m<KnOLy{&vyuQfTeL!cD5YrwHD@aufH!<z3`&=#e3AqAP+L8<<y
z34K{NF9Y2$o|rd8ldMGSr&qK4{mTQgB9>Ea^?TtEtPt<Y(GNA@HJLwQ)bf9ReH)s5
zyU{+}e$%gTbvx;M;|_Zy1?1-cY%%t=l7gV#C5|?d0v$jTN1U{Vr^ggv!F@ytFeXiE
zJD!O3_u;I%zn0^@`U=0FkcpWf_3&dirhig5k`q0Ffq3{p$?q}z8Me86PPQQss|u5k
z^$GFm{S&v>%^0WVnFwXuM%b0Z*-$Mtm$%dC>EVZ#B&&8l`^%fvt=2>9CzR6;xK*sF
z!BF3exZ^J(v)Fe6N{urH_EmQ%qVp}Cg;t;f7I(wbqXhiBs0ug2Sqg85$RH(8l&iR<
z=(8`KCCZEM-Lo4QJZGUD*{#YD8Q2OX2P`vIEO<0`qbQFbI@`5&mld?p;ecP;y{M;6
zVlk8K7))bG)jubKKI>ePSJte>Sc-79&QA(_UW$qgcuBjZ-Af<TCnJ2mEfO>9-{&iZ
z`svlLIx=WB^~NGkm|HqLQ5l1@c1X$_4F!)0>N2^?G#ae;-OkQEGafV%w2v17lo)LM
zm!?Mg0x@nQ>D@CVvfPKl1>G+7G(gkG4OW8(S2-GGThwz>ip9V?g<d=SM`@zvdpojT
z4=w3)@_hLZPM;>-^Q{N!zbzs)m^FP;eRhmIEk)v~^tOV(?3Nmz!t>lpu4$&K9eXq?
zn5ck6BUPq}_9JDIo#*F7^hEBnPt+|%J$u2UP^&C4eGe@YJcKf3&#0H<3mZy01yh(y
z&FgwnuOfLg-Z6~Co8t@}y=k`dTMOr*6u)hy-*-}vCXcJ5UqktVJ=2AJM0OEOh{9Bo
zagb4&^O(i+b+h6XZ+qq`QO23U@MVFtW`enXPSWhzdJ#L`7knl=#@ds4h<jPBZZ9?6
z%Nz8Y&wagbH1A(fx{-r1xCpyusTE8GF5+tO>M^@rP3bjl`cBS<6ASgoR<<_g0N4G9
zk2@F<A}RYpED_BTo^|F+nHPC$_Jq_U%J1nT2;))orFX;0^;nWQ=rt%=fOb|X2O}FP
zA#$Gg(yu`4=F%<&DzC>>JjI$joclMdsWREVm6|eL;#o9F2Vb2WS`ZiBS@vF*WH>LG
zQm2GI``Go15qC=O7n#eBOuu!YAs6|WdD=mn;8d?o{L0t<9bW?JMS>=bDTFWuskSO}
zAHkTNfKZo$QDmrB|E?#lvw1E1_Xy_=TsI}@`(xhBKGFUYvWjFL{7kG%=y*0urNcbS
zOY(doXSJxxaiMGbS&lXBN>5Z4ByDDyPWtT846>J#_i=lM>^$05*R(4P8H=7%@nwYJ
z(N)P-8OQrRJ~_~!O#QjZB|k^gEe=}@v>I{AsZp0<Tu3Tpx#J$nk=-}x0hUERL3<N(
z^5V04*F{LrJCTzGD2ewV<7LRwi^YEK4Me54p}*R+&rAwD(t>?XDNb7Jh$%WmLG8B=
z5+{O_D$@7#y*q&q?$U)=8r)x&xX8cEXRhoHvYVwGAzO&LkDb-ic9UWrltp~nQKn4t
zjBF0uj<h@bmHS?=Y{6T99B9-V<mEzURngXkN@RC_wv(XS?bF+SIYHlFEv|Iqd?V1;
zW|0K)5J`~;56LY@04!Gv8WPVs4yuguH2dq-J!kmLMNzpv$-SBcwE$k^5-vIST{s3Y
zNRF?68Rci=VBF~HR2g;_j8YQj$ynr9H93FlE9_SJ{MqNP8LW%R+Bl;DEAo&HLI&uO
zXA56jzvq_SY3kTUGk3=Fi_f0DWq1Bt@xFEPvgi=nI=j-w`yb*dEVCE%{g=<1jNJi=
z{fY0t=w7zvq)Wf=pV#csx4++R_PWSTo@JmB6lm<eh}9MH5#+eijo8~;PH7z^7(fR>
zP|Uj3x5jZ~dGf-$<8APWJWN|XGu_-M&&IArL_a?XwR__YHm*~c+df9*+K}BXtMH8<
zDS9+{j%lY^rOGMtO4~F;ye{kbSl0!r0L+%$KW!~Gt@>BX7EU4;%7sJ(h88bF<?4+E
zF4m<VTK4D!KWlU;+Q(W5h-_}^VxjBuE)^~}+Jse7&e_D)m&Vkt3lDtECg@_D63SS|
z7Fs?gZhKLjDnC!y#Vw!P=>FNl=E3VYDN%UXg9RWp+Ay0-ulmCxBW~Qo`RJ2c6ce>(
z#!v`#UD%t6IlpU-zRa-F>&V6%TBVqbjZ%}A^aI0|^#@3B)T;+wX`A+4dwLF#pa#3f
zwsN?_3cBjc^GfNZ;5PHc`6|Oq2bZRe^Z85v-qbwFmJc-`-(}@7L^FMRP5cd&PEpod
z3;f3uVUEm;TGxvfm&pEl3muQ4qZ$i(hF=JA3IU2q5d6BmkVLuTJl4Zv6D@KlhiO&+
z!h(oKUpp=lfy>fOo|s(16+HMveBcM3l3-W1207F<rUoU`GYpy?`SE$~48ae-2CS)T
z{Y(772~=O4dS)DY=lof?N8zYh;=k90l&tQar3FKxZ^x&8m7y7?D-@=F)@Ix|kiUVP
zoMpf?=Yai;D7L#S0I*$L5)zluYPwl49+2D-vNDL<tcy8SEUPU$8t}XpGp?K-l2Ck*
zq}!VqeO0$vrF!{d2|@zOP*`<vo{8e$_{?g*z?I0I0lfFuy)Zd$y$#@V%Ay*Wj^gug
zc(YVbaN<^~L6u6VR=sOH^ZMI?&L!PeJ^qTcrv}thLP;=Cl);iQKxj(XWt(<bXd_sf
zk3wIX-?3p|*3NT?b^KHJ<?|0Ot>y@QUNV~W8EQP#7`p*OL+gEa{iin6R?`r31-B&U
zqRkn@3)3`s7=gOqmGDX;@s3fUSUwjUC0nQL=j?jVt{P~q#1s+C$eC_v4;nTzi^Dyw
z=2t!^pAbI0ifPAk;KLJD-t;+ZKNWqb1bLjCtzM=+6bL*F#jfI5hHPNTb*lR?;n3Xo
z1r8gQOxx0U@i;w1lHCc#C<+0{l3;nRp~JLmfoILmFQ2Qttv3)mmtSo$`vxPICm2}`
zCGuWAeRe)E7vrFbC+1TnMA%m^kv?Nz?!MJ#YSo|0LDW9U`JOETagi&bFADe`<`UMR
zA7Rf$0s;>bg&aYgc45gxH-rlQp%FDdWK>eVWCOXTfP3)Rmv-lK?X#+Fc;jlIQ_LU~
z>0{kXiV+rUTsr#YBui10cBKYtT+y1MWa#?Z?i!RfESvo((jsV&Ixk##Nq*kr!xTsR
z5=W+=Yw}Ol9%<SY)QIi$e3v6ax3`l%Fz^{(*J{>C$HwcHL0j%w4gb9h0G}b}Kq`b{
zY#!DKA=uu_kAHlt;aMtU66Dt!ocPYAV>bREAoQW)f)ct&p3zD=uA5c4&d$OWLarLY
z@`h-$I{MgCkk{v?zq0m+H)v%?5X1qeXJFrS&)e?kHP}SXA3cy%!QblDlRm{-lDn5?
zH~VW0!9>8APjV8iL6Wch#Y3+IWW^sgq3R8R$F9S6AD7rQ#N5`!5)SmoyLneGQ*EiQ
z5Z~~a^Y+FUwItz8lk8xGnv^Pa^%dauwVadVlEec?6i8^ttNUq=G_ay}q@LGi&B1bO
zV1j3oPIq)exiaLKBrT#~gQ#PIqe3{JGis#xC6xf6U@O4`nNXT=YN`zOO1MgHF25jM
zUz(*WVJ1T!iFWTZ2h%s8nr8Bq3t?$pX)DBh(akEi{b??dvjL67XDboExJPYsynS*W
z-M1f3o?f(Hox)=a91;mTC5V)UkkTF~QneE%DVD`}d5KH_H<Kc0cFJ>IlLeR`#q+T;
z&ZUfiCf(PhpC&SDvXTVkNi|p$r59mOHUuZUep*~b+KzyvvEA(#tD$Du6w(E5v{wnr
z;=}faMJ(B+yw44-!_RgR8krh|CXQGr@;~IPVFsj6de!7Y1(T1_$(VUSdo0)OuuzRI
zUk{l4!nq>R!QS+pQ<LZ8rfr94*=^b>sO(4m3zMuk{A`$x`65s+WViFvXXx2af}E=g
zL70PYRdJ~{9%V9>=I=UIS2iTAl7$$#SIoq&-($Y`&d4I>oaezcB?QH_JCqQ@+*X@1
zb!rw)B_gaKnM=2KsWB!6c`UB&9Ow+lHF*!9Pkg^murgWvs><E=OI!H0hJBG~%CP*r
zMc2%`hXH5$8RXpxo_zt1OrAXaqSd4h(^G*-;jsZ_%Zz7}Z>Qdn<DaRb>ymmpQI0Fn
z7?^co(2G2ix05(;O*s3a%@ay{Hyy2!u^H=ko~Rqq_C>w|TG;A|9s9ZrLkKE7_kF<A
zY__a~okzhfbQ$SLN^7>(ZJ`r`D@-JWHYALH2<xz^d%u*WDTSYdt(rz~YwMWHz+A)J
z2z54n0y$nKiA3Y0CA?Up%ap=Y;tz1swlFu1V%qSNgJOz`?COu^AJtXo5RZ(Dp26>Y
zt6to+tA=T}={GHQ&GkK(TIlJ%#?C4Eq^kE;@(VAgFYtrV#S)a`2zSoOruO#Ua`}k5
zd3mJm%<>Og$*&$y9+)`~CAGXXV-o567LPSdBp1{8>bMm1CODuo_=?4?3!oFooD-rp
zYpXD5ju#RQ0QO`NsnLHyF=2YF<SkK~&K<21IdBx#G_`0HvBG)IHQ9L2HFvrEvP>zF
zjzzb(7QV9=z~l8s1thC`#k>i8%ox+v^&T7LBIWHD=A$FP$6e6{$C`<esn&t9gS=j6
z5u-D8Q=I8ZT9+$1w`F+?WUX24QX32gvt!AK)+b6@m?wXUtb8a#Z(YJWryT{LgnH{p
z8~;dMEKN@`I%E7Orr6XpGOs7~F>Kt9(gN7h>&Rjjh5dc*L)SMIfD@;7T1-A-jDb{%
zPlO3JJI4wLJ(11+@|W+}IjYCOooAyU;v~`e8NT80p3;bYr4t4=WoU7i?`s8OvT!D8
zn)9;&FEW3}ZxgnWijDFXwf)<w9Y)G{<!O8Kv>5W3$6I4TBE1+C&C&h$6?^IZI+um2
zA+el~sp4p<9-b_WxdC1RIX2>}$I9}v7_pdAg3n)pW8ay5#JhR6*`ibh_h}~I#XP(c
za!(h@%6*2K8mY3{1rlP-Z>h~oYv`ER3YC8$EQLO>#i^0=;xQU)WiX5ftLTTytPZ(G
zl}eMOWONX0J?*zWW{M3uAat^id;Al^{gqzf4jTLN43|r^rQ{?gUs?<vSW57QIOro4
z?QQ7_Nb5<EkkN%>yxv*kMEBw%>r1`RAkojTIQL4NxY&9v&8??rGnLAMJ_q?|m$0z)
zrO7QY&g7HuWUXM9n$TGBV4BenD&j#)H}1($B7VRI|DIys_Ei5{H^;E}S;paB0v2v~
z$&w6Ly4TOTY~PXemy(BP&6w3_nHhT|k2!u=Ao*lfD()jh^?Y{!63{bz#RIRedTt<_
zgo)O4B{R!47)*phw}&z5<gF7qqedGit$Lmepf08=QbN;k_;6AXRh1qU{_3O275d_j
zXK4OEG6dV7p_TYn*-U|qL%68fvjq^$SQrMn0QZ@-{@-KlP2aH?7@Jo14>X37!3qf}
z<MZ4v0Q`6DO4ig2tQh_rRE$~s^rpiXBSKvL&UFLM9vFsgP9gb&xk5mT)Z>rxIQ3ej
zvcwYTt<Gt-&pcG|c96>qG(a8#no4W<1o{GdsU<AOCzJQ`iY-YK_YbeDn<-$a!`e_$
zgx>b;B_6(qJaC)QiJ?!MB;6SJ`V|Pn9~cTGc*oUw-ot3kt=xm&D}!t$8#QZf@sU&c
z%4!GMp>AY*`6cYdGgdYZCqt>j=cmQIt+1WyOy5Hj0zRXh&*Hu?<azH_AbB#tSdx3?
z?}~)(L8jKn`4#R52Lp+-Y_)&UYb%%3x8F&kU>~x!(~W2xq>WjzVt?O7y>bxar1Efn
z=lF^t!_Ih{er;uKDAo2mV+O9OV(q@DaL!AnA}MMB;2kGd*J3cqu4104gFFQhU2hh}
zK2F%U9PRl}e~w&z=Yvuyf=@EIVO0sU4AQwdno5IH>_rJQ{FKvE1zuHDfCr&|3J7gm
zKH-$;+p+V}sn_HfO#UQeE0fuLu~9By7h0VX+=VRhdYB*yB)+m8cqLYHdkYvAF)niM
z4ES1pdUtpRpN^_}@78x2*&BzCELKnRGVfAxwxZKWh{1ql93WIsFEj|Y>@jMRD{1X>
zcMCTiFOZkk&X5^2r?Ro=+3>9?E$T}0!clXU>k^{XXFj}ox2s8f7wSJtO9ER<U|6@F
zJXv7aCM`;sJ5R^}Z&g_8#)I%qMt-3jtx5diy54cqCi#qguLs*Aul-T!wwNUb#~VY`
z$wdAs?A2<m)8>qMNIrianbm}kdC+S3l{4|+kgp|MfO|mvUd_Q+&o*oW$QcIxLr%x0
z92z>iZyz}&rWtZ|BWHq_FDKv!CwIIk7a5?>Zh`A>M%yyr@LN!a$veEoMJapcPEzDg
z%Tvv+-9uNyaXTd>ap#<nXILaPk%qk&BW*+)GR$0QsGHy9r9~W2Dh)&i1(fBr)!5ch
zXcMq~L@GgysYagg606~$@^(*H_;$3I!(g7UP#o|p0YD%h{|*-=Kt=uyc~m_{PJWDi
zF$Jeq<OgPScGDV)RW@yH$`j5B8zz^v5~YyYCFY;+s$X8!7VA)22}z|->u36B<tbj5
zD?>4RMRyBN<OkcRKiLaqeMrsk9S)%)jY=Dq5q*3GbeOv<OqmH~DBjy5w*vZj#!uYB
zXH0km=ze8CyF4bb-S>&E&`#fz0pLfDrLCWh$&M-3X_J<u#OEG9O^N8Mrh{ZCVH=x~
zD83ruB85HweLs{+;~zcxc;k4#{JPp<WW6uy+*E=C!DD_po)3dy<jqN9yta9F!LQ;M
z*7bEHLfrI%69t|op--l*n~u|?GKr9yF+#e~w;QTrHIoHyauOoZQA(0=R@xP=it2`7
z(M&Z{Tr1paK<VhZL@!lD3VZUj(;$jRI#PNGKF|!~yolzx6HSxzVo6Q%-hLN?no~&_
z`)*Gwu11on?T(b9lCxH71h^QD5#96cQ9t))wZ4Gih^6=E1|=S!L(D_y))iDvW;}w)
z;s>BJtq~yN_}z%BlwDK5IHfn~y}zpjCplj4Pu`gVeiD_i{B3-g+oaPYAJ-ZC&VFg-
z%|`28bLS&kQ_N~5@l+FM^C~Y*mQhL4)a1$^_iJ!e%>WaF+I@X*)NE)>yp<F`qg5H!
zcOs1gubdngJ1nBPHPY%4MF!LQfH*a#1y<TpZ-p)+W@8R@b`+utvmTbWXI@fXybSu~
zj1b%jDymE?saVvj_bd}$k_3TQIj_VdXCkukV@7-8j>7nT((q@5xJx-^kz;(BON-9Y
z6Bebl1|ppN;(#SoLq)NY4vH;Wc<k{3X74%iIFp#sLqgmesfrSa<;I*!ol;U?j5!!r
zN>&JWPVRHEDMnTiQ+{d_?A0R>f61K`HYm;PUnHW96R5lcufBMBGxI%1U-?Q-r$kbC
z?xee<rX6nL4#0MAL>EZQGz|I`wL%f=W{jdnZKi`;fhm&!q1*GU5zWN+$=BAql;~Py
z<=k-}moHNVeW9$l{!R(ze5%~NYOsonZ&s^HH9j|CT5y(GjbM~UK)BD_yLYA-MJGB3
z7<69@hp_{LmzL1KroPe=aEQY&6zm+D2oZaUmw-EC+`Dcg>D)BD|9G<b=EggtLcw;c
zbI{^qAvNNsJ)Jqu^hL<^eSc#Ej>_xA01Ky>e%}7{D-qln1$tAyypOT;>PWcs6Ccn%
zeSk^`54G6LUD)r%lA}n2>D)Qc7?<ePwIh+Ro2_d=msX-5qeL=d(+0EaC0{c?%W2$g
zZJxdiPKE9MiQ|^FHlo{fu?MpQY28cP2((v5N0#sHg$h24iz$w?1UI?laG^PN-614U
zGb84-Wa%)WQ$CVBooR;0ERO1WOgSyWUg?VkmHLcb1`8mU9i-I|;}Ejm<1;-fYE=wn
zFg)KaOu|^){`$QW4`*h|*;#D>aot$#+`bF!CFAlGD#QRKUef|b*=Y-p7RT)Bcu;ww
zQ?j;y!yc<)jCkd&6tLYdE`t|R8PchxtW=^dMqw6q_2MAFI&51m%uvhxvuy@#sL!Ai
zrALzPi(*kh%^=Yr<D}<<iu43(>aC|5+IVH}W{_i|YJB7Izf_T`4#5nX!Q_XA;y624
z4Bp8)ObE8~+fw1xH;QznV~)pO38!W5)`9sRnyD=kt2q2beSy}{k&)Vc)Ixq?m~)Ye
zEhUQ_=o=D9R41U5dwXW`_Kc=xm!*m)CrQWH*eiMvmPTotx3mtJdPOJCBnb(1dF`Q$
z(MdOQhzxz^8uXep9%q)RqVEF?hs8!;Gv~k=#gBPO{#-P#^$csgi*n4T5=ft$0>p{?
z$O-2A$jNxr7=)F7m+La}OWr2y-{M~Wd8=aJ=$d@CNoNt;+<3Fiq!^j8=`1j~IT7Xf
zx*+X4aTshzbLk8EPxWm><)Zj1Vday9fUw@T>T$}7NbeG(LlJWff@a*q;w|HB3rXGz
z8!FM`vuVbu3-2h_e7r&Q)xh%Lqty>%{1N3Ski#Ft(ZG~P7RuO1i?W4-4c@km=4F7q
zr}}}$H4q!sU^#UdQKk1jqc<Q{SyBCCYg&yc726y!rqg?_MIuPe@i2Omb(qqXRXSOe
z1l$*Z8mIPioY;yUq)58$$meLg&^RJPFVVG5n6GXN6ZLxH99wk_xWkv$YAIHjwW@rX
zAE$yv(&N$Ut6!t@9`~JAROv3=Fv-Ln1+9-dPWmd?-G)8oM3*ly0N+VoshmqdN7Y4Y
zU#WLnR-(=|4x^WjrBj8ihts35vB{`_PJoMw9G%e|&!0c2GAbt>-$Ga_5kj{@*f++;
z^<*h~Z<KikJ#_{CVyPbKQwS5nRaRhluGqW1#oAOKKH&+F_j#SY-*e@)B_@HJfV1PA
zkW%|~IIM^W4+VKcNBBpk5Uq9*TGQK4+IS!$i4x?A_nG^g--+b9q*3!CvD-m>kL*&a
z_oFuhU`q*C@zx>(_Z*s(A*OCl0E4tvY6#e>u`7yMDT{Pc2}9tKUMhjR3oYlTf?JLx
z2~nKn`wE%C>KDCU@WX^N-MC&%R6?5Lcgrky?TX@rKnCdbwTi|Gakh_hOe4OmH{tXp
ze+h#QY*}bI!kG_Zagqe-QOx8~V>4=mVv_MY0mCFF*v_&iUm_w41xmfeL!XXtd(lP-
zy&Z(F#8q*GSMx7+Vm_ON**Jz=6fSNLGM{A)YpF&#zaI*QFZ&!^3D3bh1fD0BD~oGU
z!zta#V?mzy9JxL0t#h+iGgj*J>UhWdR%6wfCa;Al*U7qTQJ-}vmhdYUlDceK*+g5^
zgorL4ut@<*yX2$fWo({IE;nEs&RB3*Odd*{gST0grGrs46mY&9@C#)?5LXPzEM`)a
zVrFXX^(@A9d@j70GR`*1**EwyX+#+W|5ac6HW9Y=hZAu~2hs_lJq~_AP%`>z;lS%^
z+76lbFEC{=jLhn@M22mi$<vvtNG!gLlBV9#kysoeA%It3ovy8!qL**t3N(y_w?Zy`
zvn26;7J)eGy4WysSPQ4@vkHUE;WwvjpYYd3b<*nZgqvKwCt}Pm9ZuB~XU4b$<0nN8
zW^(DO!e*B>Bkfna?Mk7VRu_V~yS$>H%JZWKiA!GK=SO)DtV;_ICsg8KlA0WV*_N2%
zWPZl|C$suF>x2bP@Dr)ff(YO6FQjo>@Jk+-FL#pWk2iHjrKT=FXfSgDy=yI0I?tFi
z?`tS+LzL)DW#uh?80YKbsB?Um(;2>^A(m9dDPw0NLiwnjF7b>30foNDEOXyRnVoK)
z`Fj)kQDvGfY8HuAP0i5AI&MUG#-7$puKsHdmosjmlZV%K6v$FE=M%m*uc*clf0P>A
z06UD9*lCs?B2lLq&}O0K?Z-}|Vo#c_6bcKK0i}ev3YyO+0X;obudP@E<E#~0F$8PP
z5*B{mrI&?Jg)`>F{G^P2_4X@O7MRuH()}G22`9)@<%7#BYEPTOo4(i8M+M$0G6Diw
zf)u#wqx&PalJ*Y5K;`yK2E3}GWK@CYMF53I>*<rFU>L3#*DRtJ5|7{Mn1~7F<OzH_
zFLX1v4<T^u+gpPejrD7R&W}!XbVmfwpJmw?oB$QMEWu*}_BBdwT(3$~<j%%M&M!DQ
zM7neyyq=7=T`7o{2Z=(z(=ee!Mun!RuLzj?#dD~r22o143ClcueZgq&-oM1w8l)C(
zpRw3Hg|ZZI>sKvO0E838hvG-iToV==FbhB2n_Th89H?R|+Re1XRM&b!P#dofU&fQ*
zG0HvrfI;$Iytj)`g}>(MQ?iOi-{R$B|69H{eM4|n1_QV66@lgruNA_Y;)7y{&Mau_
z&y3<{p%<&Ld9ku~P;OLJZKHk%K<G<9l9kj?Vtnim8Vpo}Cg+v@ML*CqQr$gmZ!zXy
zc-szZ@hmeKon$(7SA4n<VGD}TorT=Yk2_2({vy;$f(av+>G2>Nl<WkJwmT-@dXVqR
zdec3*{hTN^ptd%uYOc`xTHA})8kc-j+tw}lGdq{|l@;qsk7ULuLlbDTnq+iO_8p<B
z&+fNr^wrMO!Fg@-IfxxzOvLWC*ac7lF9yCeklL!mNLkOcz$-4bA+cGLa&xw{*!t(_
z+M5U9KCI|cI0@n%-P4m7(Z`V_6eCj!Ba~o)8QtLiCg{A{Go!2I?J<emKn(XX&Psp!
zUHhJyHIM9(_NrbLF8kJ%7RM!^{Q1n)95nHQ4C7>L419BYO$eq_cr&VZrz*nZh<bxc
zTl!7K+rPY-!qX|{>8p_NJ(8OXyg>(pZv90E!2dC3t#)(`4?zuGd(tbzv_lnt4v#mY
zvkHd6;b8BbDxRzOq{aoH9Mk(cfE7^qp(|QFBA~F*>cGuoVL9}+I$E%ioqy<?y>K7G
zC-q26!zBVxz2%%?f9+_d?BFRh|L*k(&sFu)ONqQeKuddg^~9G4b)T>qF#BC#Zlt{s
zc~l(z#%INI{wvNw`bUV8yMskcZ*qW`(3m3x?P7s{cv#nGOdUOHc}V;~?rwY{)uSbS
zk~ExA3I%DHbAu{t$Z@Fg`=-0twEO`-8p~v8qLWZv>eY$vy9W-wAfH~R`#pYx(*AW)
zeta)hY$Epc$Y!6uP^mFTUH8C2sw`1NtN9#qOfAxiy_c8;PErRkuluh6jM$I(z|S6`
zYb2`Ju9p4JW3J|!NB2!ThtLPmnfjSyzQK1ob{N&rXk!M)_?D{Sl+ktUg0i#cs+-j_
z7rUXG6%akisPgukF$b?78Tifm8Jm~I4hgw09_`<8F@A4*KAt8Yc^?+e#ubB2Od{{(
zP}=Qi7XsuN?Cm7BAfoL<CF%X9dDg~CqzZg{A_}Yj-GgjBL%uot=g+8wpaHh7Xk69V
zOA5hSmfgkvHaSntZVwNNg7#ihoI@U9(6U@ah26R{Aa2G=O9pgaa|&#;y~4f&K|p(I
zv`Sknz5HJlWqDPVKcb6yzMH`aCQ7;49*X6krG%H#AW06`HXWlXCDsmK-{dEvSOT9k
z#RM8cUX|-PE^~T4Cs5uYDMMa=FYlFO>cLF&j3$JNA;Vgv9`(8Rm|pOSxDqaYi-7ZS
zmI%N88di>*a{XgCWCceyh7NWdmH0@eGu-<<+75`lI5Cizi$e8NgRG$ay(&A00;|2{
z;L{dPq-6fwFWbZ$&`cElyDRL_50yXt@V9*ueun^aE)a(}-#`Hm9Pgu<s{0SdVhpYO
zO+WFea7VKr#<bHR+ZzaxFR~HbYhJT9Y~r=Vgntm4%#+u>Kht-BR2Xd6<kB0`7qTBD
z)OJ^<W!K!`j!KrL{=q{HP~#DD>GKtZ&uBh)a_V64lP;D%Ls3{zQ&`YVail)I@jF3i
zqiAryN`3ve%Ylx$|4lJ2TzkKi;|+HQ6z+lZO0QYa3Lb>|#IU|IjlLEM2g=P@+{n7O
zT9>|Q#%*r!1AKnI_dcOUlT(>DmG0=34R7Qdx}*61moaS8GU>4ZUu!<w<%TuP1_~s|
zim&))svmUzS9`cXqqeXgppDesuo_zLyXG;;>Ir^iJRN(q4fxBla&-Dr<Ml1bO4QdO
zq!YERdfz#?%zTPoh(t{DArSMnd79SW@M$}>jy(e<?3Wwu^}2UQwI3hC&54sPoj&9k
z1xRu0WaR3TN^{`rj%treVJ)+m+FTULDI(@QLM_{@EjoK^u+Dzd$SRnONU;VJTeC-+
zKrv*^(67m0a{7%AGI(<ADed0Jdk)AaG3j^e2u?bKN&kBu;+6rp!Qtj_>(Mml*3ES@
zbqrQmq*{MC(XpGjv&*;Nlp2>drZ*$Fu=3PzX=btgCz2#sy3$vevgBF*PM_V<w%r{W
z{8uevoPB>wwWYApo7{dWw#b6O-Lo3nmj3KNp+Eh;{2)hVeJv_3thBYmX78qXvAi`3
zI%%O{js43sXQSRfs0nX**TtuE?bnoq6OY^Bo{-#s-Zd%--?==KN2pg;3V3w$O$`J?
znmdvU3l9F_FIRu%o+&_Ckn#1#%@5t^nY&5#kvdAt%iANC75@K*9|_Z_+9!N_yj#xo
ziDCCu?e%^?lceSNzv!p!VUnuAc=%kcfas>I)NYS|b%_c8Qz{(}NN`KTQP0l$??z_@
z?#|}+UEjzzsegOY>AxtnjQaaK4{->o;$Jhl@+)`xf4b?*)yeYTIO5O^{{0ll`adko
z%wO{#HCT-JEtA({eDy@vwS+NPAs4B(^j3T@C1@i1ebR%S1TGcf*cV}f7vMlxPxYn!
zZR)6jzZEy=_i4^SX!hTT{tf0I#(x<8uWelHJ?eOU%4vkFh5STzE4e-<Jj?m41^!m%
z;^CsB+1Jx|oomrC9~|xMLf66KcLQe#-_${}{@8Kodp2A&+(1ok+g}tZ$hE#J9LHxa
zGuJ++(jU-j+wWEkf?Z<+AwH4Vi(cKtLHyi|48$j0*ie9oMXn1^$S$fDf;Y;jprZ_^
z%5}nJqw1ARlRz5^|LE9R4z||~1-J=yE6_PfvV&n^zeH&%|3R3TJFj~^vkKxCyK&Tm
z<+AeCL0sUql<%6^p8J`Gt77@;S5EOscmg;|Bie!O4ne<m%$BAFP#^`Awhk+hePhz-
zDzyU}K@`i&eE$hl<rH)KVTSeq%rBeXqJI*a3I7Ctn%;q}7P(G-+|9n0&9)Vo=VosO
zU19399*M$=j8wkqe!Gr+{ZrwKS|kj!kU$5~{C;F6(kj{m+^H+uBi04?()6d1C}*SA
z5IOueQ@U}<2*>2iH{Gcec8C|FCM6x@sTA10_0NT&M0Hlf@js*|XSmZkysvo84A2M^
zk|n_<NW9#(GWWb~0G?r2bOVz>=Y~?%o6dUSV1Y2w$&GWn`Td}Zh0Qajb-!#bJ{!Pt
z67|75$TJt^LEl00A%&j~eb!iLK@<=bUjHeUdJkYnAAJ(Cu3?D%AMwdoO-Qdb$}RZt
ze7ruacVGT3I=-I{dT0F24DR71lN+2}W3nl&T(&Yn{nhijU6*ggcYuH?lfLJ&sIxqz
zetXbm@w<&-Mmw@5Xy<MlJ)6RV?e%2#p3mBgNW)?Pojy6cY3|8NwcLW;OJ<0$u8{3=
z6jnvjeN4|YcfLd*?jqY?Miy31Kn*VGz?{DkH5Wk)6ou8h6<4KN;g&?_o!enD^t-ZQ
zJNb2M9W*3b*l728zs+!A=Po_F>SX0|{tw%nh)y%io2?S7poFB4mrK0D=F3e<xA1Uq
zm<h5HVjrWh>Y?dP>>_A2zH^mM{&Pt%MO+(^f(l+C9df|1N=u~Am-+fU!u~Ut5g5;q
z5?;7nx&wQBsyoN9LCQ)^g&b2fop|0T_vb1s^eWu^p^kKQSSoW>sJ<R1`<AxY@L}A{
z)#ZBKztBayqzn~`r8`#OA28Q-{q|Hr4vqLQ^>r8e8V<aFRQic!66i{1%c_6ahVUR!
zz0xkXK;SN3-3})-`wKf$-Nzog{Q`gV2<a-h&PDSFg)D?ND}(zac{)+1OlkNOfBM1|
z8>SgSQ==V_lt)UqwwC@q5|}lZ!QQE)UniWta9%@w@LazsCJHn_27U$9ENlE$7>C30
zCxF40_B5>}B~sHg=eDMA#Zm!mDDtOZWi&rSMDvtPD77*#gbzDjcV1jS67vXl{H9j;
ziXuM(y`*EW7@v=*n27LphjoaQFRz;cSw#MEXQ#fX5U{}VTR*+Todn?@);cCwhXST^
zp`f49I27hExkql~fnS~~74O|cIl(307N<4gABazhrWsOoQ^6eO14tF+p4=U%ev~%A
zC!Et%@yJN5qnm5xwexWxfDjygIlTn7xuESB;@t^{5pr?ZaWtK)4cz9+{B4VgM9+_W
zLIBydyjL^tnXUwApmiQCn9IYON~mVPac_x3PEdep0Ole$6-uX;1i<C7j@oXsAG9O?
zPUsKyO*8O7EPm97xlwN@`{|KQ*y>8@qIhd3#gN`_>ff|sxTBwPfpMQA;X#EW)R%Vi
zch!>()cVl2OaSK`!Y#YsFWG6ck*IPz)^(m!`t?jTki{Km)|!wAoQSmdud{^RHr}fg
znkATy&()h+LK{0W5&DZziysB9;>AT*mv6EQmP7s8>lM8F!o*?mt(*_{ilqFaBei&K
zOBlR<`=RY#ZUH?0v^z{z$K|yk2K>Vlh}yc6uP72Q7!aLz75!9UZ1QyM(ua9)+7Nyp
zO4xA&k1gsa5C*g4iNdzHd#v@VX@<bMi;HPB9=Hc`Fg)|&Z>8W|o_kS*S~O~>VvBj3
zYIAv5|1)}u`;SCeV9A`<gQ2RSjxwC6oUlT7^Zfnu>C-gMC00j{U)$qPr#nl3w6g;4
ziIf-19{6c$3x?hvZJ`0_oNqzP=*W?To9;_K63}ITuMqp_x5k+8CqsC73ykN7UY@qU
zfvDMxB*R3AZlF(uu+HG8q!wA@C%~bIXlJ$u_m-@~cA@)3$FSBh{vNM{ML#mnx*P8H
zUA0eu97ypuFzQdac5Mmw%voi1hNOToMDZJF_pXRZVBV3JtECTrvhg3y1Kytr$F;jz
z`gr^C{+0G=I}hn}$%obF50M>{6d9gt<4=6Ys=S|Q{T@6uiz#F^K~~`Chs;%zyHIh+
z*?1ew#Jy<1z=oL@wraM*w!|=^EZlVhH+n;UbMP!b`>|C%ltbv`2%tIx^T8buEO3B=
zr1zcaB@)=Z#^dQge`rPcvf03i@DLU^A4I1g%hx9Zj$C2kp#5zfF<;)HAySJi_R`7E
z3DGqHXXr&B7`9$+P&ZV*wP;a#VX3lD5Ek@uQz?EMy||3`juP_2ImHl^nbh0Gd>;rS
z7?CB<K)wt?WDuV+`U!;>VG6^I@#Z9ONI|dnw<oUnlzZac&t_&v!3=@?@x(R9E;g5-
z!Y^ulFmH&pHY~N~dBtW{SWHC!!*^|E)8V(XY6)*TG1E(_S+Pk8H$5$cxz*p%3^S7x
zZ~)6dpYQWt3H9VVeJO|NinkwXmOBc^lt-w5XUbT>OnPnRsgC`A>^~<oPZo2+9K<J^
zTIv2omGJ_RzM`VodiK?VWe4z`+Z+UZq#O6g`9CLR@M?o2Tgn;XIVIt}2m4eWIQ*vV
zuy6s5nG$8yiDYS4F3aHU$2kvhG!TUH#Ld>~3|3)FXR)qkQIE_k*;2Q;><+mfxxhxO
zfCJ*(vVxgNW+|b)<X~77^KYeh?}QQi!p-TRvWF60^u2fKnRrJtpYW5o;Pf9y(LO3=
zIO6A{=&jySQdDR*0LvBfzrfs+j6f7H6&Ax2s}<~2)~ww;##}8X%Qb$}t6A+3=+(@M
z2z17FA>pGtdee!~_x#}JtpM)}w-sAmgf^Go?Rni>wR!*q%L@%S=nBOjh{{hPEUD|r
z{O}K6YHnDs5nx_Cc38LL=uZ-ogk*s2h12lP`=@gT0IOxr(c?nqH<TksycQ#v#|upo
z@Pm+cXcj>Z$gXlK*5riKyAA%JVLVlv#t?B!J)-R(UGM;|eu9PT|Jzg2|L=}J?{G56
z4|L*2`(HVcXhBxwn>l|H_-`5^h7H70gx>=Cf3&nCjv;o!3ebOit8rt^f_s|yp9W`-
z<Lr!#4G+jN)_-NvcpL65#roYS2>rCXL)wqf2+N(_{-$7W1?Az{Db+YK1ssrszcy&?
zj#Ob~W852rTs6G>>gGZ7zgY{Hec|a`D<}yE=JMJ@%gpr}FUX@C%goW0BAFD?8@zq`
zbMHub&o%Xb7iK|=0RNlE>0Afy^*vuAHKqJhUA>f#-kMGjv{Ug8eOPIsgdOh@a73ik
zT*7p*sc_@wTd`&adCFdMNCDJT{#O9%9HOs2(?YOa%^A^c;2DJAg1=1$_Uff<ul%6!
zHO!&d@oXGiAigCg%?-!|CW8|H_RBjQv-iItT#8HHm_#}m;oW)KRZecMa~Fc!^@F8>
zi+9%IE<O-^_qKm%^xXpSTb9CqU<M7f_yjd&y!mK*kz)e~vY~*GCNA{NVb2J9CWBzr
z>m)qUvB7V?3tUnlqIfdi><J2K9G@?re71$UE3y9YrakrHfl$J&dXfCO-KFDCEnCxU
z9J%8vw|Tsc*+B`nBAg(K?$xjWTRC@SI`?;h>-;@w{daAAy~f%R-cfN<>4k9%^~KFX
z^-^<!6}@@leSg6~XNl|~;}k<Eu<zBDQZw7*0ngB35+;BDr%cX!yt(m>M8xo9gk0Fr
z1^D3IDruIezy{lN4VxE$gWY%#C5R$K>b^oT)RrdDiE(PMo>-xg3q$P8!T+arUNmI4
zw)OEzbgV~gpZev4<&Mer;}z`ke)8Y(%^mtd{Hg`xNKi$dNpx`8V@K&<CHw}QvPXak
z!!-Fsg4%|D3u%snAg;StL)#I!N9$spI+kNUr~l(MxzR~>2FeDDU>tfX7vZUsV4UZ!
za$|EU#9-p#;ZlE5McD6%!L1m!=(p?~{NGuDYY?`I(&*_9Z2ILN5UUeaBV6~X|55pW
o?f=vr*7yHE{@>yNWFLwu@n)FA{nP#x9PCe4Qc<G(y^;U_0=8#quK)l5

literal 0
HcmV?d00001

diff --git a/authors.html b/authors.html
index 4cc0e537d..ac7daf444 100644
--- a/authors.html
+++ b/authors.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/favicon-16x16.png b/favicon-16x16.png
index b789a0564a4b28f861f86d4c7754d27f92f5f345..240d1467f9259fc476dd2812f89ecf68988c9421 100644
GIT binary patch
delta 103
zcmZ3?v6y3n2NTx|Mkx^%vzC9~Og9HHi8Fee=o%R68kvO{8CV&ZSecq=8vq#wArXf(
m85kH;OI#yLQW8s2t#b2IGSey<j0}vhYfxpanmTz4b3OpYvm4L=

delta 103
zcmZ3?v6y3n2NTzG9(E2b&UUdvvCTnD;*8$Lx`sx&2Ie6K7FH%kRtA>Z1_o9J28$}K
n3mF&~R7+eVN>UO_Qmu0HQ!>*k8H@~!uxq%i(tLLE7Up~aN~Ifl

diff --git a/favicon-32x32.png b/favicon-32x32.png
index db24b35f1359588715f3be712ce9881dccbf2845..3e1b3194d1e8881a95f643b2ffac1275210dcd03 100644
GIT binary patch
delta 101
zcmbOwFiT)U1=kBkDG?TnWVhQ!8=LO2d7J1O80s3Cg%}xF8JJj^nrRyV83u=sUuFOT
j)e_f;l9a@fRIA+ll+3hB1|tI_>>4)B%HA`1HG4h)@)I52

delta 101
zcmbOwFiT)U1=n*Pb`C8rIeCRO8=LO2c^m5*8tEFChZtB`nHX6a8fY6BSQ!}n_O7#G
lU|>)!ag8WRNi0dV%FR#7OsixtGBCofp-b#o!{pWM`2dtS93}t&

diff --git a/index.html b/index.html
index 36f895229..42e97f174 100644
--- a/index.html
+++ b/index.html
@@ -33,7 +33,7 @@
     
     <a class="navbar-brand me-2" href="index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/pkgdown.yml b/pkgdown.yml
index 097ae200a..ebc901b78 100644
--- a/pkgdown.yml
+++ b/pkgdown.yml
@@ -1,12 +1,12 @@
 pandoc: 3.1.1
 pkgdown: 2.0.7.9000
-pkgdown_sha: 13d8d621105a3a527a05e46c6c805a6564c39791
+pkgdown_sha: 548a9493b72ff93d3ed8392d4ad30b77d8b15fa5
 articles:
   LC-MS-feature-grouping: LC-MS-feature-grouping.html
   xcms-direct-injection: xcms-direct-injection.html
   xcms-lcms-ms: xcms-lcms-ms.html
   xcms: xcms.html
-last_built: 2023-12-07T08:42Z
+last_built: 2024-01-26T20:05Z
 urls:
   reference: https://sneumann.github.io/xcms/reference
   article: https://sneumann.github.io/xcms/articles
diff --git a/reference/AutoLockMass-methods.html b/reference/AutoLockMass-methods.html
index 0d05a94b6..ae1057f93 100644
--- a/reference/AutoLockMass-methods.html
+++ b/reference/AutoLockMass-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/GenericParam.html b/reference/GenericParam.html
index 4b738d783..640863774 100644
--- a/reference/GenericParam.html
+++ b/reference/GenericParam.html
@@ -22,7 +22,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/PlainTextParam.html b/reference/PlainTextParam.html
index e88573325..48aab2dd8 100644
--- a/reference/PlainTextParam.html
+++ b/reference/PlainTextParam.html
@@ -76,7 +76,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -231,7 +231,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## Save as a collection of plain text files</span></span></span>
 <span class="r-in"><span><span class="fu"><a href="storeResults.html">storeResults</a></span><span class="op">(</span>object <span class="op">=</span> <span class="va">faahko_sub</span>, param <span class="op">=</span> <span class="va">param</span><span class="op">)</span></span></span>
-<span class="r-wrn co"><span class="r-pr">#&gt;</span> <span class="warning">Warning: </span>'/tmp/RtmpZ5Cxxr/test' already exists</span>
+<span class="r-wrn co"><span class="r-pr">#&gt;</span> <span class="warning">Warning: </span>'/tmp/RtmpPglK9Y/test' already exists</span>
 <span class="r-in"><span></span></span>
 </code></pre></div>
     </div>
diff --git a/reference/ProcessHistory-class.html b/reference/ProcessHistory-class.html
index 7b7c57d5c..51e0ecf50 100644
--- a/reference/ProcessHistory-class.html
+++ b/reference/ProcessHistory-class.html
@@ -48,7 +48,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/RDataParam.html b/reference/RDataParam.html
index 8421b2485..79c90706f 100644
--- a/reference/RDataParam.html
+++ b/reference/RDataParam.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/SSgauss.html b/reference/SSgauss.html
index 52510c1fa..2c925686a 100644
--- a/reference/SSgauss.html
+++ b/reference/SSgauss.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/XCMSnExp-class.html b/reference/XCMSnExp-class.html
index 8a0b06c4b..7518c4494 100644
--- a/reference/XCMSnExp-class.html
+++ b/reference/XCMSnExp-class.html
@@ -342,7 +342,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -684,7 +684,7 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
 <span><span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="va">...</span><span class="op">)</span></span>
 <span></span>
 <span><span class="co"># S4 method for XCMSnExp</span></span>
-<span><span class="fu">chromPeakData</span><span class="op">(</span><span class="va">object</span><span class="op">)</span></span>
+<span><span class="fu">chromPeakData</span><span class="op">(</span><span class="va">object</span>, <span class="va">...</span><span class="op">)</span></span>
 <span></span>
 <span><span class="co"># S4 method for XCMSnExp</span></span>
 <span><span class="fu">chromPeakData</span><span class="op">(</span><span class="va">object</span><span class="op">)</span> <span class="op">&lt;-</span> <span class="va">value</span></span>
@@ -1226,7 +1226,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> - - - Processing information - - -</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Data loaded [Tue Aug  1 15:54:19 2023] </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023] </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> Filter: select file(s) 2. [Thu Dec  7 08:42:31 2023] </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> Filter: select file(s) 2. [Fri Jan 26 20:05:21 2024] </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  MSnbase version: 2.26.0 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> - - - Meta data  - - -</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> phenoData</span>
diff --git a/reference/XCMSnExp-filter-methods.html b/reference/XCMSnExp-filter-methods.html
index b6e18a34f..03693b0e0 100644
--- a/reference/XCMSnExp-filter-methods.html
+++ b/reference/XCMSnExp-filter-methods.html
@@ -116,7 +116,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/XCMSnExp-inherited-methods.html b/reference/XCMSnExp-inherited-methods.html
index 8ba1fefe0..ff3d7f2f0 100644
--- a/reference/XCMSnExp-inherited-methods.html
+++ b/reference/XCMSnExp-inherited-methods.html
@@ -70,7 +70,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/XCMSnExp-peak-grouping-results.html b/reference/XCMSnExp-peak-grouping-results.html
index 4f062a51b..6be3117c2 100644
--- a/reference/XCMSnExp-peak-grouping-results.html
+++ b/reference/XCMSnExp-peak-grouping-results.html
@@ -44,7 +44,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/XChromatogram.html b/reference/XChromatogram.html
index ed8865e3b..54fb6699c 100644
--- a/reference/XChromatogram.html
+++ b/reference/XChromatogram.html
@@ -28,7 +28,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -808,7 +808,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> [[1]]</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Object of class "XProcessHistory"</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  type: Peak detection </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  date: Thu Dec  7 08:42:34 2023 </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  date: Fri Jan 26 20:05:24 2024 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  info:  </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  fileIndex: 1,2 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  Parameter class: CentWaveParam </span>
diff --git a/reference/XcmsExperiment.html b/reference/XcmsExperiment.html
index 19a40f3d0..da316910f 100644
--- a/reference/XcmsExperiment.html
+++ b/reference/XcmsExperiment.html
@@ -56,7 +56,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -179,7 +179,6 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
   mzmax = max,
   rtmin = min,
   rtmax = max,
-  msLevel = integer(),
   features = character()
 )
 
@@ -234,7 +233,7 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
 # S4 method for XcmsExperiment
 filterChromPeaks(
   object,
-  keep = rep(TRUE, nrow(chromPeaks(object))),
+  keep = rep(TRUE, nrow(.chromPeaks(object))),
   method = "keep",
   ...
 )
@@ -444,7 +443,7 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 <dt>features</dt>
 <dd><p>For <code>filterFeatureDefinitions</code> and <code>featureArea</code>: <code>logical</code>,
 <code>integer</code> or <code>character</code> defining the features to keep or from which
-to extract the feature are, respectively. See function description
+to extract the feature area, respectively. See function description
 for more information.</p></dd>
 
 
@@ -779,9 +778,7 @@ <h2 id="functionality-related-to-correspondence-analysis">Functionality related
 <code>"rtmin"</code> and <code>"rtmax"</code> with the m/z and retention time range for each
 feature (row) in <code>object</code>. By default these represent the minimal m/z
 and retention times as well as maximal m/z and retention times for
-the chromatographi peaks assigned to that feature. Note that if in
-one sample more than one chromatographic peak is assigned to a feature
-only the one with the higher intensity is considered. Parameter
+all chromatographic peaks assigned to that feature. Parameter
 <code>features</code> allows to extract these values for selected features only.
 Parameters <code>mzmin</code>, <code>mzmax</code>, <code>rtmin</code> and <code>rtmax</code> allow to define
 the function to calculate the reported <code>"mzmin"</code>, <code>"mzmax"</code>, <code>"rtmin"</code>
@@ -1036,6 +1033,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span><span class="va">chrs</span> <span class="op">&lt;-</span> <span class="fu"><a href="chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">xmse</span>,</span></span>
 <span class="r-in"><span>    mz <span class="op">=</span> <span class="fu"><a href="XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">xmse</span><span class="op">)</span><span class="op">[</span><span class="fl">1</span><span class="op">:</span><span class="fl">2</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"mzmin"</span>, <span class="st">"mzmax"</span><span class="op">)</span><span class="op">]</span>,</span></span>
 <span class="r-in"><span>    rt <span class="op">=</span> <span class="fu"><a href="XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">xmse</span><span class="op">)</span><span class="op">[</span><span class="fl">1</span><span class="op">:</span><span class="fl">2</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"rtmin"</span>, <span class="st">"rtmax"</span><span class="op">)</span><span class="op">]</span><span class="op">)</span></span></span>
+<span class="r-msg co"><span class="r-pr">#&gt;</span> Processing chromatographic peaks</span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## An EIC for each sample and each of the two regions was extracted.</span></span></span>
 <span class="r-in"><span><span class="co">## Identified chromatographic peaks in the defined regions are extracted</span></span></span>
diff --git a/reference/absent-methods.html b/reference/absent-methods.html
index b2e8c6aab..ee544f846 100644
--- a/reference/absent-methods.html
+++ b/reference/absent-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/adjustRtime.html b/reference/adjustRtime.html
index 72d5f2747..bfaf0b2cb 100644
--- a/reference/adjustRtime.html
+++ b/reference/adjustRtime.html
@@ -86,7 +86,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/applyAdjustedRtime.html b/reference/applyAdjustedRtime.html
index bad9421d3..62ae042c9 100644
--- a/reference/applyAdjustedRtime.html
+++ b/reference/applyAdjustedRtime.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -158,7 +158,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> [[2]]</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Object of class "XProcessHistory"</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  type: Retention time correction </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  date: Thu Dec  7 08:42:45 2023 </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  date: Fri Jan 26 20:05:36 2024 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  info:  </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  fileIndex: 1,2,3 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  Parameter class: ObiwarpParam </span>
diff --git a/reference/binYonX.html b/reference/binYonX.html
index 555ea1f28..899774e85 100644
--- a/reference/binYonX.html
+++ b/reference/binYonX.html
@@ -24,7 +24,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/breaks_on_binSize.html b/reference/breaks_on_binSize.html
index 42fc6e695..db9b2044d 100644
--- a/reference/breaks_on_binSize.html
+++ b/reference/breaks_on_binSize.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/breaks_on_nBins.html b/reference/breaks_on_nBins.html
index f40c8f0a4..6803da2b0 100644
--- a/reference/breaks_on_nBins.html
+++ b/reference/breaks_on_nBins.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/c.xcmsSet.html b/reference/c.xcmsSet.html
index 4327ac5a6..d13a3dfef 100644
--- a/reference/c.xcmsSet.html
+++ b/reference/c.xcmsSet.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/calibrate-calibrant-mass.html b/reference/calibrate-calibrant-mass.html
index b3dbb7a4b..a71556d7d 100644
--- a/reference/calibrate-calibrant-mass.html
+++ b/reference/calibrate-calibrant-mass.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/calibrate.html b/reference/calibrate.html
index 2397d0584..241b3a567 100644
--- a/reference/calibrate.html
+++ b/reference/calibrate.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/chromPeakChromatograms.html b/reference/chromPeakChromatograms.html
index bdc21e609..8d8cdf248 100644
--- a/reference/chromPeakChromatograms.html
+++ b/reference/chromPeakChromatograms.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/chromPeakSpectra.html b/reference/chromPeakSpectra.html
index 9b838b36a..ac822d04a 100644
--- a/reference/chromPeakSpectra.html
+++ b/reference/chromPeakSpectra.html
@@ -82,7 +82,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -313,8 +313,8 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> file(s):</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> PestMix1_DDA.mzML</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Processing:</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  Filter: select MS level(s) 2 [Thu Dec  7 08:42:51 2023]</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  Merge 1 Spectra into one [Thu Dec  7 08:42:51 2023] </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  Filter: select MS level(s) 2 [Fri Jan 26 20:05:44 2024]</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  Merge 1 Spectra into one [Fri Jan 26 20:05:44 2024] </span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## spectra variable *peak_id* contain the row names of the peaks in the</span></span></span>
 <span class="r-in"><span><span class="co">## chromPeak matrix and allow thus to map chromatographic peaks to the</span></span></span>
@@ -413,8 +413,8 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> file(s):</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> PestMix1_DDA.mzML</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Processing:</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  Filter: select MS level(s) 2 [Thu Dec  7 08:42:51 2023]</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  Merge 1 Spectra into one [Thu Dec  7 08:42:51 2023] </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  Filter: select MS level(s) 2 [Fri Jan 26 20:05:44 2024]</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  Merge 1 Spectra into one [Fri Jan 26 20:05:44 2024] </span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## Parameter `msLevel` allows to define from which MS level spectra should</span></span></span>
 <span class="r-in"><span><span class="co">## be returned. By default `msLevel = 2L` but with `msLevel = 1L` all</span></span></span>
@@ -447,8 +447,8 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> file(s):</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> PestMix1_DDA.mzML</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Processing:</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  Filter: select MS level(s) 1 [Thu Dec  7 08:42:51 2023]</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  Merge 1 Spectra into one [Thu Dec  7 08:42:51 2023] </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  Filter: select MS level(s) 1 [Fri Jan 26 20:05:44 2024]</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  Merge 1 Spectra into one [Fri Jan 26 20:05:44 2024] </span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## Parameter peaks would allow to extract spectra for specific peaks only.</span></span></span>
 <span class="r-in"><span><span class="co">## Peaks can be defined with parameter `peaks` which can be either an</span></span></span>
diff --git a/reference/chromatogram-method.html b/reference/chromatogram-method.html
index 404fa1924..ccc7af982 100644
--- a/reference/chromatogram-method.html
+++ b/reference/chromatogram-method.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/colMax.html b/reference/colMax.html
index 264b328a8..7c40ca047 100644
--- a/reference/colMax.html
+++ b/reference/colMax.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/collect-methods.html b/reference/collect-methods.html
index f449c20eb..7eb97d979 100644
--- a/reference/collect-methods.html
+++ b/reference/collect-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/correlate-Chromatogram.html b/reference/correlate-Chromatogram.html
index 0423f9eec..21ff6bd15 100644
--- a/reference/correlate-Chromatogram.html
+++ b/reference/correlate-Chromatogram.html
@@ -36,7 +36,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -189,15 +189,15 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## Using `compareChromatograms` instead of `correlate`.</span></span></span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">compareChromatograms</a></span><span class="op">(</span><span class="va">chr1</span>, <span class="va">chr2</span><span class="op">)</span></span></span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [1] -0.02314823</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [1] -0.09306005</span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">compareChromatograms</a></span><span class="op">(</span><span class="va">chr2</span>, <span class="va">chr1</span><span class="op">)</span></span></span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [1] -0.02314823</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [1] -0.09306005</span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">compareChromatograms</a></span><span class="op">(</span><span class="va">chrs</span>, <span class="va">chrs</span><span class="op">)</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>             [,1]        [,2]      [,3]</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [1,]  1.00000000 -0.02314823 0.9940933</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [2,] -0.02314823  1.00000000 0.7977335</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [3,]  0.99409327  0.79773352 1.0000000</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [1,]  1.00000000 -0.09306005 0.9887213</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [2,] -0.09306005  1.00000000 0.5018044</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [3,]  0.99355819  0.50180442 1.0000000</span>
 </code></pre></div>
     </div>
   </main><aside class="col-md-3"><nav id="toc"><h2>On this page</h2>
diff --git a/reference/descendZero.html b/reference/descendZero.html
index 0c911c122..f25d80b0f 100644
--- a/reference/descendZero.html
+++ b/reference/descendZero.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/diffreport-methods.html b/reference/diffreport-methods.html
index 326ffcbb0..19fdc166d 100644
--- a/reference/diffreport-methods.html
+++ b/reference/diffreport-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/dirname.html b/reference/dirname.html
index 87005278e..ba474c89e 100644
--- a/reference/dirname.html
+++ b/reference/dirname.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/do_adjustRtime_peakGroups.html b/reference/do_adjustRtime_peakGroups.html
index f5c69a91f..796373c05 100644
--- a/reference/do_adjustRtime_peakGroups.html
+++ b/reference/do_adjustRtime_peakGroups.html
@@ -24,7 +24,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/do_findChromPeaks_centWave.html b/reference/do_findChromPeaks_centWave.html
index 342dfa908..1c0dd6722 100644
--- a/reference/do_findChromPeaks_centWave.html
+++ b/reference/do_findChromPeaks_centWave.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -226,7 +226,7 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 <code>beta_snr</code> to the <code>chromPeaks</code> output, corresponding to a Pearson
 correlation coefficient to a bell curve with several degrees of skew as well
 as an estimate of signal-to-noise using the residuals from the best-fitting
-bell curve. See https://github.com/sneumann/xcms/pull/685 and 
+bell curve. See https://github.com/sneumann/xcms/pull/685 and
 https://doi.org/10.1186/s12859-023-05533-4 for more information.</p></dd>
 
 </dl></div>
diff --git a/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html b/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html
index ad5d1409a..8b549905d 100644
--- a/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html
+++ b/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html
@@ -34,7 +34,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -306,7 +306,7 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 <code>beta_snr</code> to the <code>chromPeaks</code> output, corresponding to a Pearson
 correlation coefficient to a bell curve with several degrees of skew as well
 as an estimate of signal-to-noise using the residuals from the best-fitting
-bell curve. See https://github.com/sneumann/xcms/pull/685 and 
+bell curve. See https://github.com/sneumann/xcms/pull/685 and
 https://doi.org/10.1186/s12859-023-05533-4 for more information.</p></dd>
 
 
diff --git a/reference/do_findChromPeaks_massifquant.html b/reference/do_findChromPeaks_massifquant.html
index 801dbc124..d3a15daf5 100644
--- a/reference/do_findChromPeaks_massifquant.html
+++ b/reference/do_findChromPeaks_massifquant.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -300,8 +300,8 @@ <h2 id="references">References<a class="anchor" aria-label="anchor" href="#refer
     <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"></a></h2>
     <div class="dont-index"><p><code><a href="findChromPeaks-massifquant.html">massifquant</a></code> for the standard user interface method.</p>
 <p>Other core peak detection functions: 
-<code><a href="do_findChromPeaks_centWaveWithPredIsoROIs.html">do_findChromPeaks_centWaveWithPredIsoROIs</a>()</code>,
 <code><a href="do_findChromPeaks_centWave.html">do_findChromPeaks_centWave</a>()</code>,
+<code><a href="do_findChromPeaks_centWaveWithPredIsoROIs.html">do_findChromPeaks_centWaveWithPredIsoROIs</a>()</code>,
 <code><a href="do_findChromPeaks_matchedFilter.html">do_findChromPeaks_matchedFilter</a>()</code>,
 <code><a href="do_findPeaks_MSW.html">do_findPeaks_MSW</a>()</code></p></div>
     </div>
diff --git a/reference/do_findChromPeaks_matchedFilter.html b/reference/do_findChromPeaks_matchedFilter.html
index 862e9217c..700a86133 100644
--- a/reference/do_findChromPeaks_matchedFilter.html
+++ b/reference/do_findChromPeaks_matchedFilter.html
@@ -32,7 +32,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -282,8 +282,8 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
     <code><a href="imputeLinInterpol.html">imputeLinInterpol</a></code> for the interpolation of missing values.
     <code><a href="findChromPeaks-matchedFilter.html">matchedFilter</a></code> for the standard user interface method.</p>
 <p>Other core peak detection functions: 
-<code><a href="do_findChromPeaks_centWaveWithPredIsoROIs.html">do_findChromPeaks_centWaveWithPredIsoROIs</a>()</code>,
 <code><a href="do_findChromPeaks_centWave.html">do_findChromPeaks_centWave</a>()</code>,
+<code><a href="do_findChromPeaks_centWaveWithPredIsoROIs.html">do_findChromPeaks_centWaveWithPredIsoROIs</a>()</code>,
 <code><a href="do_findChromPeaks_massifquant.html">do_findChromPeaks_massifquant</a>()</code>,
 <code><a href="do_findPeaks_MSW.html">do_findPeaks_MSW</a>()</code></p></div>
     </div>
diff --git a/reference/do_findPeaks_MSW.html b/reference/do_findPeaks_MSW.html
index 4c3b3d2d3..79d6aec53 100644
--- a/reference/do_findPeaks_MSW.html
+++ b/reference/do_findPeaks_MSW.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -162,8 +162,8 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
     method. <code>peakDetectionCWT</code> from the
     <code>MassSpecWavelet</code> package.</p>
 <p>Other core peak detection functions: 
-<code><a href="do_findChromPeaks_centWaveWithPredIsoROIs.html">do_findChromPeaks_centWaveWithPredIsoROIs</a>()</code>,
 <code><a href="do_findChromPeaks_centWave.html">do_findChromPeaks_centWave</a>()</code>,
+<code><a href="do_findChromPeaks_centWaveWithPredIsoROIs.html">do_findChromPeaks_centWaveWithPredIsoROIs</a>()</code>,
 <code><a href="do_findChromPeaks_massifquant.html">do_findChromPeaks_massifquant</a>()</code>,
 <code><a href="do_findChromPeaks_matchedFilter.html">do_findChromPeaks_matchedFilter</a>()</code></p></div>
     </div>
diff --git a/reference/do_groupChromPeaks_density.html b/reference/do_groupChromPeaks_density.html
index c7fdcb93c..314701f9f 100644
--- a/reference/do_groupChromPeaks_density.html
+++ b/reference/do_groupChromPeaks_density.html
@@ -1,9 +1,19 @@
 <!DOCTYPE html>
 <!-- Generated by pkgdown: do not edit by hand --><html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="utf-8"><meta http-equiv="X-UA-Compatible" content="IE=edge"><meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"><meta name="description" content="The do_groupChromPeaks_density function performs chromatographic peak
 grouping based on the density (distribution) of peaks, found in different
-samples, along the retention time axis in slices of overlapping mz ranges."><title>Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density • xcms</title><!-- favicons --><link rel="icon" type="image/png" sizes="16x16" href="../favicon-16x16.png"><link rel="icon" type="image/png" sizes="32x32" href="../favicon-32x32.png"><link rel="apple-touch-icon" type="image/png" sizes="180x180" href="../apple-touch-icon.png"><link rel="apple-touch-icon" type="image/png" sizes="120x120" href="../apple-touch-icon-120x120.png"><link rel="apple-touch-icon" type="image/png" sizes="76x76" href="../apple-touch-icon-76x76.png"><link rel="apple-touch-icon" type="image/png" sizes="60x60" href="../apple-touch-icon-60x60.png"><script src="../deps/jquery-3.6.0/jquery-3.6.0.min.js"></script><meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"><link href="../deps/bootstrap-5.3.1/bootstrap.min.css" rel="stylesheet"><script src="../deps/bootstrap-5.3.1/bootstrap.bundle.min.js"></script><!-- Font Awesome icons --><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/all.min.css" integrity="sha256-mmgLkCYLUQbXn0B1SRqzHar6dCnv9oZFPEC1g1cwlkk=" crossorigin="anonymous"><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/v4-shims.min.css" integrity="sha256-wZjR52fzng1pJHwx4aV2AO3yyTOXrcDW7jBpJtTwVxw=" crossorigin="anonymous"><!-- bootstrap-toc --><script src="https://cdn.jsdelivr.net/gh/afeld/bootstrap-toc@v1.0.1/dist/bootstrap-toc.min.js" integrity="sha256-4veVQbu7//Lk5TSmc7YV48MxtMy98e26cf5MrgZYnwo=" crossorigin="anonymous"></script><!-- headroom.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/headroom.min.js" integrity="sha256-AsUX4SJE1+yuDu5+mAVzJbuYNPHj/WroHuZ8Ir/CkE0=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/jQuery.headroom.min.js" integrity="sha256-ZX/yNShbjqsohH1k95liqY9Gd8uOiE1S4vZc+9KQ1K4=" crossorigin="anonymous"></script><!-- clipboard.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/2.0.6/clipboard.min.js" integrity="sha256-inc5kl9MA1hkeYUt+EC3BhlIgyp/2jDIyBLS6k3UxPI=" crossorigin="anonymous"></script><!-- search --><script src="https://cdnjs.cloudflare.com/ajax/libs/fuse.js/6.4.6/fuse.js" integrity="sha512-zv6Ywkjyktsohkbp9bb45V6tEMoWhzFzXis+LrMehmJZZSys19Yxf1dopHx7WzIKxr5tK2dVcYmaCk2uqdjF4A==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/autocomplete.js/0.38.0/autocomplete.jquery.min.js" integrity="sha512-GU9ayf+66Xx2TmpxqJpliWbT5PiGYxpaG8rfnBEk1LL8l1KGkRShhngwdXK1UgqhAzWpZHSiYPc09/NwDQIGyg==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mark.js/8.11.1/mark.min.js" integrity="sha512-5CYOlHXGh6QpOFA/TeTylKLWfB3ftPsde7AnmhuitiTX4K5SqCLBeKro6sPS8ilsz1Q4NRx3v8Ko2IBiszzdww==" crossorigin="anonymous"></script><!-- pkgdown --><script src="../pkgdown.js"></script><meta property="og:title" content="Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density"><meta property="og:description" content="The do_groupChromPeaks_density function performs chromatographic peak
+samples, along the retention time axis in slices of overlapping m/z ranges.
+By default (with parameter ppm = 0) these m/z ranges have all the same
+(constant) size (depending on parameter binSize). For values of ppm
+larger than 0 the m/z bins (ranges or slices) will have increasing sizes
+depending on the m/z value. This better models the m/z-dependent
+measurement error/precision seen on some MS instruments."><title>Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density • xcms</title><!-- favicons --><link rel="icon" type="image/png" sizes="16x16" href="../favicon-16x16.png"><link rel="icon" type="image/png" sizes="32x32" href="../favicon-32x32.png"><link rel="apple-touch-icon" type="image/png" sizes="180x180" href="../apple-touch-icon.png"><link rel="apple-touch-icon" type="image/png" sizes="120x120" href="../apple-touch-icon-120x120.png"><link rel="apple-touch-icon" type="image/png" sizes="76x76" href="../apple-touch-icon-76x76.png"><link rel="apple-touch-icon" type="image/png" sizes="60x60" href="../apple-touch-icon-60x60.png"><script src="../deps/jquery-3.6.0/jquery-3.6.0.min.js"></script><meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"><link href="../deps/bootstrap-5.3.1/bootstrap.min.css" rel="stylesheet"><script src="../deps/bootstrap-5.3.1/bootstrap.bundle.min.js"></script><!-- Font Awesome icons --><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/all.min.css" integrity="sha256-mmgLkCYLUQbXn0B1SRqzHar6dCnv9oZFPEC1g1cwlkk=" crossorigin="anonymous"><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/v4-shims.min.css" integrity="sha256-wZjR52fzng1pJHwx4aV2AO3yyTOXrcDW7jBpJtTwVxw=" crossorigin="anonymous"><!-- bootstrap-toc --><script src="https://cdn.jsdelivr.net/gh/afeld/bootstrap-toc@v1.0.1/dist/bootstrap-toc.min.js" integrity="sha256-4veVQbu7//Lk5TSmc7YV48MxtMy98e26cf5MrgZYnwo=" crossorigin="anonymous"></script><!-- headroom.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/headroom.min.js" integrity="sha256-AsUX4SJE1+yuDu5+mAVzJbuYNPHj/WroHuZ8Ir/CkE0=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/jQuery.headroom.min.js" integrity="sha256-ZX/yNShbjqsohH1k95liqY9Gd8uOiE1S4vZc+9KQ1K4=" crossorigin="anonymous"></script><!-- clipboard.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/2.0.6/clipboard.min.js" integrity="sha256-inc5kl9MA1hkeYUt+EC3BhlIgyp/2jDIyBLS6k3UxPI=" crossorigin="anonymous"></script><!-- search --><script src="https://cdnjs.cloudflare.com/ajax/libs/fuse.js/6.4.6/fuse.js" integrity="sha512-zv6Ywkjyktsohkbp9bb45V6tEMoWhzFzXis+LrMehmJZZSys19Yxf1dopHx7WzIKxr5tK2dVcYmaCk2uqdjF4A==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/autocomplete.js/0.38.0/autocomplete.jquery.min.js" integrity="sha512-GU9ayf+66Xx2TmpxqJpliWbT5PiGYxpaG8rfnBEk1LL8l1KGkRShhngwdXK1UgqhAzWpZHSiYPc09/NwDQIGyg==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mark.js/8.11.1/mark.min.js" integrity="sha512-5CYOlHXGh6QpOFA/TeTylKLWfB3ftPsde7AnmhuitiTX4K5SqCLBeKro6sPS8ilsz1Q4NRx3v8Ko2IBiszzdww==" crossorigin="anonymous"></script><!-- pkgdown --><script src="../pkgdown.js"></script><meta property="og:title" content="Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density"><meta property="og:description" content="The do_groupChromPeaks_density function performs chromatographic peak
 grouping based on the density (distribution) of peaks, found in different
-samples, along the retention time axis in slices of overlapping mz ranges."><meta property="og:image" content="https://sneumann.github.io/xcms/logo.png"><!-- mathjax --><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js" integrity="sha256-nvJJv9wWKEm88qvoQl9ekL2J+k/RWIsaSScxxlsrv8k=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/config/TeX-AMS-MML_HTMLorMML.js" integrity="sha256-84DKXVJXs0/F8OTMzX4UR909+jtl4G7SPypPavF+GfA=" crossorigin="anonymous"></script><!--[if lt IE 9]>
+samples, along the retention time axis in slices of overlapping m/z ranges.
+By default (with parameter ppm = 0) these m/z ranges have all the same
+(constant) size (depending on parameter binSize). For values of ppm
+larger than 0 the m/z bins (ranges or slices) will have increasing sizes
+depending on the m/z value. This better models the m/z-dependent
+measurement error/precision seen on some MS instruments."><meta property="og:image" content="https://sneumann.github.io/xcms/logo.png"><!-- mathjax --><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js" integrity="sha256-nvJJv9wWKEm88qvoQl9ekL2J+k/RWIsaSScxxlsrv8k=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/config/TeX-AMS-MML_HTMLorMML.js" integrity="sha256-84DKXVJXs0/F8OTMzX4UR909+jtl4G7SPypPavF+GfA=" crossorigin="anonymous"></script><!--[if lt IE 9]>
 <script src="https://oss.maxcdn.com/html5shiv/3.7.3/html5shiv.min.js"></script>
 <script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
 <![endif]--></head><body>
@@ -14,7 +24,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -60,7 +70,12 @@
     <div class="ref-description section level2">
     <p>The <code>do_groupChromPeaks_density</code> function performs chromatographic peak
 grouping based on the density (distribution) of peaks, found in different
-samples, along the retention time axis in slices of overlapping mz ranges.</p>
+samples, along the retention time axis in slices of overlapping m/z ranges.
+By default (with parameter <code>ppm = 0</code>) these m/z ranges have all the same
+(constant) size (depending on parameter <code>binSize</code>). For values of <code>ppm</code>
+larger than 0 the m/z bins (ranges or slices) will have increasing sizes
+depending on the m/z value. This better models the m/z-dependent
+measurement error/precision seen on some MS instruments.</p>
     </div>
 
     <div class="section level2">
@@ -74,7 +89,8 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
 <span>  binSize <span class="op">=</span> <span class="fl">0.25</span>,</span>
 <span>  maxFeatures <span class="op">=</span> <span class="fl">50</span>,</span>
 <span>  sleep <span class="op">=</span> <span class="fl">0</span>,</span>
-<span>  index <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/seq.html" class="external-link">seq_len</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/nrow.html" class="external-link">nrow</a></span><span class="op">(</span><span class="va">peaks</span><span class="op">)</span><span class="op">)</span></span>
+<span>  index <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/seq.html" class="external-link">seq_len</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/nrow.html" class="external-link">nrow</a></span><span class="op">(</span><span class="va">peaks</span><span class="op">)</span><span class="op">)</span>,</span>
+<span>  ppm <span class="op">=</span> <span class="fl">0</span></span>
 <span><span class="op">)</span></span></code></pre></div>
     </div>
 
@@ -82,8 +98,8 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
     <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#arguments"></a></h2>
     <dl><dt>peaks</dt>
 <dd><p>A <code>matrix</code> or <code>data.frame</code> with the mz values and
-retention times of the identified chromatographic peaks in all samples of an
-experiment. Required columns are <code>"mz"</code>, <code>"rt"</code> and
+retention times of the identified chromatographic peaks in all samples
+of an experiment. Required columns are <code>"mz"</code>, <code>"rt"</code> and
 <code>"sample"</code>. The latter should contain <code>numeric</code> values representing
 the index of the sample in which the peak was found.</p></dd>
 
@@ -135,6 +151,19 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 <dd><p>An optional <code>integer</code> providing the indices of the peaks in the
 original peak matrix.</p></dd>
 
+
+<dt>ppm</dt>
+<dd><p>For <code>MzClustParam</code>: <code>numeric(1)</code> representing the relative m/z
+error for the clustering/grouping (in parts per million).
+For <code>PeakDensityParam</code>: <code>numeric(1)</code> to define m/z-dependent, increasing
+m/z bin sizes. If <code>ppm = 0</code> (the default) m/z bins are defined by the
+sequence of values from the smallest to the larges m/z value with a
+constant bin size of <code>binSize</code>. For <code>ppm</code> &gt; 0 the size of each bin is
+increased in addition by the <code>ppm</code> of the (upper) m/z boundary of the
+bin. The maximal bin size (used for the largest m/z values) would then
+be <code>binSize</code> plus <code>ppm</code> parts-per-million of the largest m/z value of
+all peaks in the data set.</p></dd>
+
 </dl></div>
     <div class="section level2">
     <h2 id="value">Value<a class="anchor" aria-label="anchor" href="#value"></a></h2>
diff --git a/reference/do_groupChromPeaks_nearest.html b/reference/do_groupChromPeaks_nearest.html
index f3d8ae363..ff4ed64eb 100644
--- a/reference/do_groupChromPeaks_nearest.html
+++ b/reference/do_groupChromPeaks_nearest.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -82,8 +82,8 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
     <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#arguments"></a></h2>
     <dl><dt>peaks</dt>
 <dd><p>A <code>matrix</code> or <code>data.frame</code> with the mz values and
-retention times of the identified chromatographic peaks in all samples of an
-experiment. Required columns are <code>"mz"</code>, <code>"rt"</code> and
+retention times of the identified chromatographic peaks in all samples
+of an experiment. Required columns are <code>"mz"</code>, <code>"rt"</code> and
 <code>"sample"</code>. The latter should contain <code>numeric</code> values representing
 the index of the sample in which the peak was found.</p></dd>
 
diff --git a/reference/do_groupPeaks_mzClust.html b/reference/do_groupPeaks_mzClust.html
index 3bb67c04a..f43036090 100644
--- a/reference/do_groupPeaks_mzClust.html
+++ b/reference/do_groupPeaks_mzClust.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -76,8 +76,8 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
     <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#arguments"></a></h2>
     <dl><dt>peaks</dt>
 <dd><p>A <code>matrix</code> or <code>data.frame</code> with the mz values and
-retention times of the identified chromatographic peaks in all samples of an
-experiment. Required columns are <code>"mz"</code>, <code>"rt"</code> and
+retention times of the identified chromatographic peaks in all samples
+of an experiment. Required columns are <code>"mz"</code>, <code>"rt"</code> and
 <code>"sample"</code>. The latter should contain <code>numeric</code> values representing
 the index of the sample in which the peak was found.</p></dd>
 
@@ -94,7 +94,15 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 
 <dt>ppm</dt>
 <dd><p>For <code>MzClustParam</code>: <code>numeric(1)</code> representing the relative m/z
-error for the clustering/grouping (in parts per million).</p></dd>
+error for the clustering/grouping (in parts per million).
+For <code>PeakDensityParam</code>: <code>numeric(1)</code> to define m/z-dependent, increasing
+m/z bin sizes. If <code>ppm = 0</code> (the default) m/z bins are defined by the
+sequence of values from the smallest to the larges m/z value with a
+constant bin size of <code>binSize</code>. For <code>ppm</code> &gt; 0 the size of each bin is
+increased in addition by the <code>ppm</code> of the (upper) m/z boundary of the
+bin. The maximal bin size (used for the largest m/z values) would then
+be <code>binSize</code> plus <code>ppm</code> parts-per-million of the largest m/z value of
+all peaks in the data set.</p></dd>
 
 
 <dt>absMz</dt>
diff --git a/reference/doubleMatrix.html b/reference/doubleMatrix.html
index edb10d12e..6cd51dda4 100644
--- a/reference/doubleMatrix.html
+++ b/reference/doubleMatrix.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/estimatePrecursorIntensity.html b/reference/estimatePrecursorIntensity.html
index 5bac6869f..2f10487fd 100644
--- a/reference/estimatePrecursorIntensity.html
+++ b/reference/estimatePrecursorIntensity.html
@@ -26,7 +26,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/etg.html b/reference/etg.html
index a2929c5f7..6bad3e7c7 100644
--- a/reference/etg.html
+++ b/reference/etg.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/exportMetaboAnalyst.html b/reference/exportMetaboAnalyst.html
index 45e50a5a2..26a6aa731 100644
--- a/reference/exportMetaboAnalyst.html
+++ b/reference/exportMetaboAnalyst.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/extractMsData-method.html b/reference/extractMsData-method.html
index dd371db8e..0985c2f79 100644
--- a/reference/extractMsData-method.html
+++ b/reference/extractMsData-method.html
@@ -26,7 +26,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/feature-grouping.html b/reference/feature-grouping.html
index 2612977ae..e110ed390 100644
--- a/reference/feature-grouping.html
+++ b/reference/feature-grouping.html
@@ -60,7 +60,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/featureChromatograms-2.png b/reference/featureChromatograms-2.png
new file mode 100644
index 0000000000000000000000000000000000000000..d62ec8dad189809cef9cb8503fec010fb51c0d45
GIT binary patch
literal 54998
zcmeFZc|6qZ`#1cN<kG@bu2Lpcrd%nhWDl*FY|+@)qU>85gE1&qX}KmDSIC+WW-#_`
z3`&uN(AZ~?eK2DigE2hk^!<IG-}ip~?&rVf{_p<0UXfm(Igj%^&f|C=?_<%ko5p&3
zc1rGqAZQOl|C$*D2?arr;GN%gfPWc?OP>edcHA-0y9U04XYDFE9DEb;)3*+QAgL(+
zAA#ucg}V@R1VUW9j0{Sir3F`JqOFH{zGw|Zc9->+EBp3s+x}b9rHH3rzPyXbdmfQ<
z_R-F#8aYn~G^b`ZUp{}SR<QfsX?e9`HQDpq-Z}mL`|p<zT{>oa<l**x#ds`cYnA$@
zYdV>!O`Akgt+mr;LJDl!Zq-%aVkaA?keQRUyMFEL|9<&Drh(p!$O$R1I(NK~kxhs)
zmmw7-`>OpCce+QGkR&&-qTW<&W=kaHIv{zsP}Z`ydCwSBDjKOLU0<z9bVY_+r#RZ%
zV+kSG*;5_^0*#vasCZ`sR5W9fSYat)lDaj>Kv6lCaz+bl+3q$?myIU10uJ*9_|=Ei
z#SCJ(HG<aZN|K18Bu){{if7oOx#VTQrchB_tECOb2kzdP(r){LSR$G=S|w3#X{l#)
z^ojeMM*o{#!R?2I`3vQJiDNA4Zkx0YczO4DYUd&zN%nCq<9z+1KyBd!w7Q;_?RL*6
z+hwcDU*JrxDH4qBKOcO5SMoo$Qg|o3ETP7l^em}sU@MdUam5y6&B+$?YbDY3YPxLD
z>Y~Im>h+UbKgSk4siiL?6WcHozm(+kxt0`DT2*OkS}9E-+RCNZnrYSfS8rk^vp+bN
ze+pBzY*H+mkm3}wG>Li^oV--Je&Aef``TR9u?IFYXL{P_M%IQ*wbDZ_vGb|IoWdbe
z!8}mYzxI~Yn8<aOmHBnX0Oc9@%$2Ryq2z_7g1FIY)Z+A4#$U%*v~sI&KjYOoKf^kf
zeG2oP>Snd?r3M?!auqO7(Lh2|_kjWb*=kDAv&o_Usj6yjZI5|W!4aUr?*Gu>%r^rV
zBF4VL)i^BhzU)YV_lsw_V#uwkEaFa<8PA5@(OfHG?pshx<Y}Tp^j?T0Ma|vc>WY7g
z3NQEVjXr3>7@=vbEgN9^w)Pv$ZaJa{CcigA1vC!1Rk}O$+dgh8*c(t|RiJlMQK7xP
zRB;q9+V0L)66|%MXKRBkZ}jc^rBroM{$+Nj>oB@S@UF_nO%C#XC51FAE$`ySZryzz
zCS3ZU8gHYv<?rd+<cmH{XE-{}GVMjAxfo1vo3L^B5DK$?e$@V}S&_DnB6n+QAD2ZB
zXV)6cF2zvts<OBDGP65OJZa-0KlXW2LS&-X>i0$W-O8G|Xf&#pGhO5QQV?=#wRP9|
z6N+g4$(0mor}WuS-d7jKd<vPMqgh`VX~bKET`NVbSh;vsZb5%jEpZ$}U=7XH&cL>C
z!aqx>Eod@p6euCkV35;6^HEg(#7MzdFD>dGXOEC+;fwHEcM{poz&w(*IaEtFQC=@%
zdQ1vj<dpvB4q}IeAgfgDaWtA;&g<p8W%R&AMzs2hTiPAVWIRdAEH74bk-gMlP9mKK
zLzA)V`?#SVGM?czwsQ5%9}41rd_}kU?(XQP%5fGFoZ~84#ezXh_oV;VW$F{<KJ0=D
z!`pn8w^K?(uCae$XSX(WKhNx;-Xf43@1$j`;;njEw{f>@XV=y$$h=uaB4$?hiLh$F
z@hr1(#X((NQ4bm3na@q$w5lk0P}{ICdh4Th_m)?Mas8abNp*k#Gg35-q6$@87;A~S
z$%E3u6?)VDeajKbCXXZ_r!@y8rLd!`xbbYcuMgZhezAtZL09oGRFex&3g$dKQ3H%6
zjMe1c*p?Pz`|L_di`+n}7ilv@Ob26zqB}@y8VuQCjbzWQ<TQE~6J@+=!pOGr0u>~W
z&j#_8ta2qAv`uQK0!FFFZGR!DYpo|yP)pQ9gY{+<f~=O{K>|?R<ReD5jUn?)kTZLn
z(Mw~7hgX1~R#SXZhVwPN6x%7^Q%9%<7q{qMxtp$|%kwmz8QYRu)Lp8TuO;?qy=o=-
zho;9f?g}*{DEF_`7$8y7*mwGatqwisu%n6S@_(Dv01xI5L2)Y##(S-e*}>yXTf*@x
zRAo$Mc#AZ1$U!UgK97y*bu|gg4f(c@Q<oKUdLUjV;Zxa6;Fqw*Ldmf7y-Q|@;z)3p
zjo7@3AKXgr<WGQU{cU-hWgWyo%Wy~Dhkw#A-7A5Ie>|WoH3>k46Nve3(CKzF#2|Gr
z+$3;rd~K5#G$>9?bNIR0jx_Pvl7Hwr(n>XuoI7*I`sZY{&+;zqkP2@6jq<q}C8n9q
zkk(1ne3WHNobIky;U1wPg+wuAwEakcW5qY`+@mh1{NMfp(1BULFj0LV%&^Jh_{}u6
zs*OwR*LSrimKln1m0O97qf{#K_Z|=KC_~45sM8I-fTLCSuVwn!f3895{GgPPbFxgx
zL%gc@WlRsXnvkn(vcbkI4tLb{6qX(tUzNWyBrY7nKUPU=7zi9Q5!)4^`u-LB*cMBL
zfUOJyx?*{<rm1xDvX5tTD|LY2YxUDLdMYlGq+Rhzo6A5~W9r5TinD<jo0)<ikud6r
ziOo+}oo@ZV9qPWet1EH?{#TXd*~?j0q}YYeoL?ET;2EI?_vvtp7<uN;JuNniG%(M&
zbs_OH24x@B++}9ycb{?FhUQjEyR~i(5Qco<EZ>q(0fX5g{+=XuqlZ=7zcIExvBH}!
z06~R%yI#Mrv`DjQ+*Hh!uKdK?%2;ou?(`6n?o`VUwFFTsZZE5VSw4W3W8UY<8eE)U
zVyt})P(8<)LzLoRn;Ge+;J_+)G#TTa@fD+$@>~b^Z@2mvczPh~=c={B6}jPV&v-H=
z@jUbpXClscC*|&9NxXvW#3b`JE&4)8R)G+U*}b3@BoiHMP!aYmAaHFB8@IZqj<IGX
z9ABxcvX#&_^rZ1n4>F!V$StdLeeB6<-P<xKI`((H`(a__6pmNPzarC2r;Yx+G=L{*
zN_3D+Wt6g_?{wJPrA1GgMx#z|%O>VVoR(WWRy5zo?oQeO>vpT_nIhKEnWCZ6Zkmj6
z*>VxenJr^cx$++{3_%enHwAbQ+NhuDozo?PJ0NP1AG}4m=%X%erR7{PH+{;AHKHIj
zIU3)~dCH&|%wn7?v?*IugIUf)53UcFmQ@}244YON6_j1XoTjZSV|8VpRghV>cUqHd
z`!IzEF4j!{I>a~Nm3i*Ad<s-AxvWPm2_&=h2t)l1n=*v9-nrT)72{KmUt0|Z(IM<-
z_C8zjfFy}oa%c@21BB?Z2bd_Gg0Mib=mDxZ^)Y6(&)~np0rj)+&$o;h3ijq3VpN!<
zc@xJsIjjz|EuVoV$k69%yKF;#>xq}+h8J{f8@8rZXVnYgJifu=f;KL6V;z74<W*(Z
zuDsRCsjke5<%W0aGZdRx7hcAcx%?-A%2V=8f5fS2<|+2JXf(~;&gb>f3**_2#h*5N
z=I(Bd2Tkp~!@DCZQ2vjjdc87-%0-p+VCm8;I$jk#COV>>tkX1d(P##s)Ztyff}h5#
z^xq)Jum?eHj-9~}wmFdKefh*k@uTk;MA-f&gSBY}id!{NLgdUkrvdm;6tf-~?WuN`
z;@iV7Ra)<>fgsQJh7<8UrOUY^Gc*HOLwcL-#3m_+IjFWj%*2OKOX;{V3Q!D<$Pk-t
ziC-85@yapKcdcmee@8*S0l3f=B<X5ZtZ2^O$q=uOLSiIA*j%w-Gq(e?{Dd*cD8p?M
zE3_g|(TNvVT5-?YVRD|e@7Lcmh6VpdD|g#7bJtjd4Pmf_9~-~+=llU&_5XeTPiP?T
zQ(Ulc{rdniBbfR!Sa=(dQ9o#J#k+@M18z?$Z`FiUQYOuHlvzjk;;odl9eAH>BCSa9
zpf_cYztU7K!gZQ{(dh&dE039`M6tZ~C_Z?NtGlMf|3FfgprnX{h0{!pApb{bxSk(v
zz>{~2Z*Jh3`3J-V@YHJ@2h0XMg~t5o2cG}GU-?lU%;)1X-+GWHIimRvwpgdy5g_jk
zUtT?ZF0VE#<Qj^^Nbul}xtb8#h;0N12ds~@z&5Bb_}4r6DINNZFe6kHqfpn3Gwfl2
z>T{`cCT0*cElr5p_LQs;bJVog)g6$N5*|>&2-7UC5l)_lsj$kzxXI>Xj8Oh4ANFI$
z`{q&60%MYKfsd_}JuM@|hh4-_B=(p{i$M_47!a(8RtAwEXDOD)M9#UvABef{maVN>
zH$)kuK1+93)~-*v3|Q03G;%eHO=*~WY;meEvBx}I>-E+?2vSiHaR6Um%u!V>hi=j;
zGzwzA+v_vHajQ(zF%GEFu!61EjH|J83A)9czjSv+jS{1fI+m1*_1N96fJg7=-%w!;
ztt8|QyA13;g;sroQv&|dV!05kSU^{;3L`U0b@7}j#y<>y+_gnbf{xEJ6C9U!;saU5
zD__)iBi$5r6*WG#4~iHcaL{@m@&J{l>p1vWo)!sH*ng%#duS?~TxVV_{Bu{Vw<oHp
z@q78>5ihidzuLfnWHbH6z2C(l2=>=@R_JV=w5?X|dSj#TMwXoekHsJnRLdsmzQD5j
z+^M&S!sck7q8ps=x@PQCI2WcmHET(vNHVfdTdPmhr_@g^%U0AC6X>iqkVMdF_fvtO
zmCKLN<)Q@+y;GEO)FG}+U(QsJ&8XO;q120XVJzfSf*Euno1pGw+()P1;5^ZN&FsT{
zVzL9C`}-F7!td*J|MV*thY9u{cv8Z1V=LmWkw&~A=rrOHIxx_JmCK|I@Eko1&9zCU
zMdZ81thp21RE<7gPqr)$(a!PR4GBQKUdH)hb=4S@RYhK>7OBI8^RwzY+6gRZPJ92^
zh4s;|eCsAJG$WgpR^B@x0YUH}R~Lf}#+{>8+!2Nx5v?3D$81ff9-_7*qkT6_m)dm8
zzMc+g6oB&f4_WhqWT%6zP`BC7bj{ET7SqyC4G0{Qg0vrQ>P2Uh=f-ryM-(q;i`5=$
z=mSQgB|`5Da$~=0W=kmdpB}9F*5ojd(;0LHWVADqU2oHF^2QlYM4-YGb9rIe3&D&j
zT{59UbttU5#BbAO0IN38fgAAxRO|X^$#HNcKkl#B7vFDBwF@id88DelHdV&BKnI<5
z8SG6^ET3+p^-klxDP>*Ndt;1Q08UKFI2X#9ZJcZ#C6b8JDs|r>C_$CKr{%H5yHY{-
z(H~C9#fv$NXn@GnSI&qh=;Sav558NW>vA6@O&sj0v3JDQgxHRjslDr{{@J#)dnXjJ
z7BifJONniv9$YQDnScM2e4Y?gcp$}s$I|_|SdP0~>g$7K6;IVPUW~$6>fDr4AMyiI
zg@IJ7gy$Rc=2kOdu`MN*w-um>vI%wSmIquR+N-ougNReQ=3SOzFV%VQD>w2@5unb~
z`txm#X+~$#s5$!TH=``;&Q)zz@4vnsDl9*2PpMvSr_M=wAKX-N^kB6ZQJV<jL}809
ze?I7(_1srevhbBr-OLuVj3$-OPD;8OOJfF_=i&=)xIoYn=S$;}vZxC?a#IElW(Iz~
z7-^$d;12)D>QCS~O-VxaCw~4>&{tU;;>AYe<OfxiL#Aurg_t<YFJ@88a7VqOkb36X
zD)Q%Q$r^X#*{QcubdtUcK=1F_vRkOCs{6>Qs@gF7v+ZxYa}6Y)EnrkaB+@Lg9tC>^
zptIs&XG<%(VsurN^=2u1ic_&d4~p7FRLu(nic$(}XH*b1Rcm`U&Dgce{r1n6BqM_)
zo)qxzI9~!(43=<UPYKE*L#JO%MLR;VM~EtB*(6uB3(5=ELGpr8q(+G$nF9LjF{a1a
z2YVo=*oB9_D+L8%zQs~qQ?CQRx@gvPBykA3+g58Dqn2Mx?mG$9Gc*bo4owN*O5PNB
zgbHlF4&G7r;q-2M&5Y0Ko}@%iPy;o54Y0x<u#^StlIV8G>8rFQ?+VJSEaP0{u;YYe
zM`PL92XjnZy65DDebW%+b%`H*47*ks*KrR^^i<^yx6-TqQZIXNdr}|(dF>0CFW6%X
z6Fm1a#Y^w_nkd-2E7G?hwb!kEe!SVcnbxO9sxAh?_1aWHR`VV}iguudNxs7^dnBzl
z!7sGKg<N{hWvsYa+r$$dApmv1kjZzTxrrB+tUIxpjg!*%n!YPT5qpe(VNqClOeNM^
z)*mzLu%dInxZP6{F*M|rJn<V;Xpv$?NslqSm|k7Jo}?>db<g+qRR1#tQG0;Xf(v^#
zXJ_;1U24~TEFC?NrO7=xKWXRtIDxTPla$$tvh^57GnEI+R@(;U9Urox6vp(@OtJ6Q
zbpmit&+~PEH5Pb^0@bl1VUXv?j-`?%u^lTD-y2V&bZ%+Y#>`_~QEACPe}nS&48U9^
zwTjOf3_od{eEnkL&>t1QjBrQR^#-qtr6FI&)?Y}93{;e}#(Z#Hl0%kS=^YgnjlxTE
zdH4T@AmN{$k_rc<2i&w!w?_kBjxqb4`6j#r{y`J=3oQKmf3^x&w#^WAlI_BoHzKz|
zR_PKJq#R?CC`tP!&$;cqP0Ma@=3Xa&g`*v@E%B(~W||Nw*-ks?@eBkV&bdWYwHO+h
zs=%0MJm&-~?S6ju9RGxo3pk$FY_gq0%}!5yjjD@x_IGcCdL3OY6usfUiFqRb3VTqn
zx2W#d+Lw^PnWIj`i#a|nW2#nN#Qb_!hT{cCoA~UG!Rq13Gt`2sJ*lR&V7?3cz2$GF
z{X~2_!PxeL#(#Q$xUuK*0L6av!d9IpT*SKWz>@;3Ip4JMZ2!~keUX&eXeGI5%G}tB
zf#vb@Py|-9EChw(9yo<2rFdS;J8-oJRMpgFhVPAjWfb9V#5WNC?dk04<~dK;)n3uu
zmmVAE*!2uD@m8QUOP#t_A|eI#jv41|mM&C{z16+XVN~gU@~}Br9jN=_YLA0*&r=BM
zRs3%!HxAdV4*B3>TQEb&w)C}{8wSM=KRb4!F6K!jteHmorhM%E2fn$Y3|~Cy(;%7(
zEk)u|U0YC>*$vbM1aliyQdAN%s72{_;9W(1wv)Y+$k(S;Hqa-XN@W?9I}^w%{ffz*
zZbOAx)r1%b1V<zP1-oNHkP2~*>IvJQ5~YkM?;v=GPMYhLsy&#KAht|O?gf=?U?GkY
zmaHTYVm&{sFxm+i<=W6NG^ztgi32sTQTv0{E1uLn0l43w2ukNW0VqOfKvJ-r*^@IC
z>d4LlC(ly=NKpiw6%fkX`)WKJx*KIon~cjd%<T6--{>;R_z*n{nVj0QLw*UdkJCNT
z-n;EFy`c>PkdrSwXwuo0{iAUh|AGL6@mkL$dc|AxSjHQz1l!08?1s)NY*9koMqf;X
z=W45h&nx}9FA)$*G%DBobbqdt;Xch3fvVz{W^Ix?92;KJAt44iDf2AYINTU>D_2`<
z$j?f)jacC2Kh+<zv_}|<$Z``rtEoxkAUX88O+|AVuh59vn*RUc)<i`U9T&(+eDyVB
zf!cwi6VnoW)$-2$GBST~xkXv)GY4YHG6iA9A(k2PcYH;?L*H(`X_Rrywg5!E%WaNt
z!C~bcDvSii5HY%4^>FuFaC}$3{>^{s)Aswj+bYsFS}yl_*KuNBe7%R7X#`K0cwXA~
zRlz)6#nNBiT|coXPE`ViVqz)t25jN>uPp#gmn_USkLwX7EL3HNHAtjQzx%wZQ+swo
zy`^Dq*Xh76190gcg{r?^H|4)R)jZy!F+aw9WNCaxrsg9Gu;6JjVNw-OP9tf9>#@@N
zWml2F?VKma7Jq_B-Pt_8msofJDy#~%Wv3BG3A<q4Z=J%EL6H9WU0DgFAQOg<&??#&
zszO)g^H6oo<9)wg4-UP>UZO7G6jO$@dfco?z(G@b!6(4``r!Rduls9xjMDvi0YPk7
z><r1)^cqxuZ}AJm684%LZ_Y{t=X{h;EyAgY*xuV1{B?w*{3HB3p!;s=JWW#%p7oZi
zaIBV(5rAN{t3(M^c{GVY@5}k<g4HOlwq?I%&~X+bj+#|92BRfE)0Lh!@ZC)LN_*b_
zjO_p6?*wD-<n<LA^&~lulg!xY`wIjpXIG0l`;_lF)uDQ*Q9Ie6mktlQJ@XrAbZx$R
zf<x>5Q&KJGT-r!wGgeha+tR_>-qc>Z&w@+Q@m;Lhb+sqVfAJy&eS5W?HCsdJh~(gK
zUteSpB@D~Qs|!Lx*iWdUqEN42Ey2IB9YJLym<4)G%KONAs`%=t3;U`qk&a&QInU7d
zI$p)zX<kvPJ9qfx`6Spm*kkW5uC_>u48PVOn`+g75ZrU3T^OqW8_=e1Yn<40yN6++
z_FWcv%TWSZXGf7Y-wV~(rZ`Y)G^%2}xMW>?e3`GF?8$IDSyPPus1mn~dL1`5%T!SD
zNf-?OJhl?oJQ`S<Ft1hFZw;uk{Ll={Tl0W4#`=Z#=QvLgDGz~2iJ^u8;H}>YWd0W#
zNj97`egn3t{mPN5Ha=IQYyRnTz_SM(hp0-nDXFc|$#(R6eC+p``=R@2ZUxtE5}r23
zv_ap@;C&RqDN#gy;BLMiMU;=5a0gKR{)<E#ug#|we?{w!225V@j7kFW+f&lK*3lgv
z8h_(YmJiqo4R*Ds&HWFMfSeWFSF?ojXV*5D$59H`%gr5@OsVY!i-}NSABgtvW5#Ix
zIfIlC5SLW-lnJP4_uaahl`*pv!uow;*Ol8JjAGe=_DjPDVr!$yd;3SGeERxF2)ZB9
zNBcsl`({|$59VqZw}=OIAP94WpO1O_JL%%Yx-82B*W-ED*aeL;hflGcaUSEl_#D@%
z!=IXFj`n~v=^dTV)Fx)W9}s>o)%~-+bOv@6oqcY%H%K15b}7`C8~z;VxbqR6Ue<7q
zR$SKb5vni%UjK_oocR|}_A(GuRf{I(?Yb7?mOam${r&)#(Aqyq7_!bPOYPI2XU-sX
zK=UBbVk@I#efm}7^up5mV12%~yT4PvDJ{gQUx-$m^08)c(5=F+^$bVqM}AZVRewx+
zNPi`)s^i_r6FvYWsQ<#(4(9Nrjf2jBEHT@tJc$7Ex6MR~)^AWRvfPs7x!XcBFJ_R!
zkM6_sancEaD0Y_aKxES70ovkV?;R7mP*iDe|83V7=Cxb&oC#$<S$G0~#nqLa%Ak{J
zTUm@q+^zmIzR3Htc|NLroFC0$wnUlSR6AOo=}Ev5eFgv&a6I7RL$r;k5}&{U$%am2
zw{vU3`m&}W*PZ?PLS#bFG#7H-=QwdaRc&%u?fL^gPJH`&JIf37B&t7;F{6lSUdv^)
zp83D8!uJ3E-<<1SPud>H3RA8uFWc@0-?vq?@cG=5f0FLIB^BCzr~cQ}>)wN5UwTyf
zgQbSlxU)ZWVb7yo`Haf&bhZ(6iu0XuG<HVOW8>&bcbw{>wSEScD(M;f@ZYll;)HBt
zCV6s91(Y=?HU2v{VyIl=(j=_y{R1~t6DSut)}{^W`}la5%C9~c#rEX@s->!0Yo}Ic
znjh)~K$K@$4uZs&U?IRzV-SEfmB_`*crSm~`tIQQA3$$avH5mke8E(qJmgMbxk{~H
zxD#qp_sUYCt}iVBwm(0<eC^3s2WxD3NA1WL#u?m=i3Xd))f2m~KoRCA05yzGwX4m0
zE!G({KDPMbl$|hSmA2{0{sJOL18(#Ex3+TKGNuT1FAmO(m!wuZdL(1Z-q5VD>D7h(
z9d0x~H&ph4D4_Ae=b@3>Fip95;<;s)+wv>_3K+Gap6piKLkk(AHb?+fQm(?3_P_K%
zzE8EaSL%7%UC<)Azg13o2Sm3@=?kK?HS>|#JmY^COTv=vcu3hQOqlc2{UrfG8Ljmo
zDR#$`xq<c>_Mi7Fh^BDYb#>^P?dQ$_%76X;E4+j(b?Vwi0H>oSq-02XA9|TMxM?RF
z++0f=bnBUa%y^>vjx$4aZ2z^9qd+|*uZ8G{|6TG}A2Pp$OLvGeXL%<x+q@3<xV8AX
z$X12<IBW^}n>#}T@vx?a9t)E9LWu$(=lWvw{~2za|7^;_r5{$cjJhCy&DeW)#wwHX
zk9{}Bc=U%M<TT1%XK32Yk}=Jp4Sh%-OIp^s)3XG|BMa@bXeHZ&^?^ptQ~g+irK(Nz
zM%Up{K<9gaZ%xkOq-%V=kP#p^{$G;H`osqh*r%x8lCPuJnQd)SFvoHk2YUysdqIcS
zX$ebJWrtA`0kqe!?CfdJ-B96BsaryR3CN~p^4bQXo9u7${7|8IGN|NjggIB*;7Y<0
zaAXU#8yu}@d><`&#(ViZy*p<NYnHM9*$Ci)T!av7*4~~+chP*mApGfM?A*MrFP#Vi
z)e@lu7H$P987)h&{6tX_ojSGk^?`Rjvt!kaSJ>!R*uqaH1O3Y=9jUWS>SgLK*xWaI
zZ_c6+<Rk+?aONii8w;T1!y#$_OhHon_0CS~UJDv=L1<&8@etDSjpwzb1|giq!5)<z
z{ZeLHW#^FdVuTkH62N)akMKLY<zA#PVvO6y4K`_MkLxY9(vDp#Q;gw8Yh=iWwnih|
z2hR%M<mtfc8IRV9u&Ydm*cCVc`Yd5qKXBd(hNmQTm*xaa6gqo#EB{=mhb^1A_N4ga
zYz}kapbPpYp1kB|pMmH6)x1)qYk((P3|V5swZ#OX*%$yzhX>TgXTM-)(GS?gApeiH
zWaVqy*jw|iqq0vn{sw&;LR&=Vz0BFHGVS=Xe7H}aDNRj?^z+Fuhf2_QQ3#Y7s*aH+
z`HX&x;On;ANfYbv0c1RP0Wb0obJlP2<P^7S5!)@>-KRO@2x>Mgd=>Dbc6ZP-uAR3(
zSpIeS@BpyXYZ*rMReVPO^aYGsvkO5hAy!=2L8lSd!wJ5vLuVnVVz{BB-w&B_<Mzmp
z5=Ha6n^;@+-;8l8E;fyaVCPs1{{|US%NfUUN?`@LeB!85BTmFzkPjkx!_mHjGqMS<
zcS+WXWvdYS>g;%TGfY@1pw_BGa~p>sP#Ez1OyVcR)W%EC(nt=LJa_hNqs%dxwArO~
z-82Wt$$53Wae8t9CYa>|cp$7l7uU1?wDEWH5-|xREo6S8g_KR_mf^lK(3Y;swcCR)
zjg~AnOo$%?t<Pg0G{XmAB7+1~3q`N#ogl9OJY-qgTFd_cP}VI@CBs%yBtIG%->#ax
zXYLFxb7X{?MU{vhw0{vORUOD`oF5^yTWk?c$HOKW)>JWb&B@<oq54Ds2;pe=1?*}r
zc9#430CzQv`{UR)=-&7zUEB8=Foi_zb3hMy3b<h5><ev))o#T_fpZ!U<~~ptV!0%b
z<#C?hC@<8Fy=Bq#?Wiswp`fO*Y1}uyet$EEo79|js}pkKVGyzpLB}r|%<5!l<d_PD
zTyeJQk$_Rjd04WT{XlI?yS5o6cOjLYzrN6D?E~EC<#yJ_T>s0F>btXbH@)vK|8YkE
zx;OcliiPdZVzs;?xvOO^NXqmiI0Ars4wALae(Y3#L3o{=8Kr7LdnkYXSt;;Wm7^l`
zHSX-0o_HlS@7s6g*KR!(f)a4TmhE24G#;mPSHhD9$!6LBQrL1**_h>uwe|q36nMdd
z(MXOPa*_EPNR&WTu+GH7G1acF{rrmK6op$p{g)tQ*w+SzG_<ZNdM10cF*M$RJeB~6
z#vr2~?B>R-`&9oU(gl1?lqH7{mS<ZIhU<F211Ot$NDa?Tb-<#W3%_2biW!D&;+y9v
zTAMK-Sx7Vc8>)Z6b3kz$he_ym^*G#Ag6_WgFN_IB0F@2wnp&8h?z!s>iQXSH#ImQn
z?WhL1W&^0j<PRWq%qi6&;Tf$TQMgY&R6)R;e)<cW4_Qz=?{gm=h{_j$W|}zWTj{2s
zxet-A3(5y9Kz(AZz_;}7_kp0%-huCBk_}7r{seP>T~Kn~q=154BJk{sq;+27gW0Sd
zxl-g7ww}VmQVBuFcR89E5{hc@*Z~}drUk|@;Az?eQ>VT%s<0@I?&1c4v4q>q_d1P%
zRa?l4Y?SLB=kSP=a?QISFF)<sZ;dU1WNc7IRIikz)+Z({hfW!imaco%ODSHsyb^iq
z>0T%yb&kriXi~;UZ(RiDm)s?3o)V?HF6pQSlI3?G+pO>fc`q~H_>5Qb9DuoR3?1dL
zE%=5&#2!#k*DlBdWqf+kK)a~zovrYkF!K5f24nik4#@BWNfrM}SheljZIqXLsCC#C
zQ0c2&!Gk+H%`XfVoy+*y$e@tt0&cU>R9S+sa$6Xeg;?AM!9b4FUxM7mX>yeYWnr7z
zvJk@z#aUq@=^Jo3XHOJK*L-<}@q)S^T$DB6mu)g=ed@SORzaKZAGboUW^mYS+WI#Q
zP!2H>bATeI)yvK~v2%2br<1F6OZ>fv%&t54IU8`aYQ_s#<W4bCYIPJ=dnhMSrzj+d
zt)=vA_b;M33C8TqoO{2{ov4WS)EkZdl=1#|38=6-tgb_Qc%8umE+$)Lnc=bYol&E!
zESJ>fDw%I$><$VZUneCM+Ks)COJS2<R{$H{1K4oC8P<}W)#&#$T7|IlO!8n|@->!>
zdAzX-<TSUQrF(3#kRb!ZUa_v~A~;JTC;>!I5F!Es%`@<5J+Sxd9J97ude@%?K@Zq-
zCBKP6P6zl@*p?l~c#Z24L8E5JHdR$Otw15OLQlo`kPcztae?AJ%O-!^jn>Dh<epYg
zJb2o!@+4cTZz`A~m*%nCnQYHqWi)q*fnd1KkNsIxRpSCjN1hKD4<wbZ>J4QOd&<HU
zV4%o8UdgDtKrmD!w50+(ZM;pL2Fn^K<-NSFs!XV=P$f*TD&w&_rr5U&*FT68eG~b0
z=q^4(%(iS#wA;Fg3T0oV$^bxJ7R0oUO`7DTYR^Cd5#~!6YBRMJ*S$M$u)A^gkrXc7
z#u13fyX@$*d`CMk<g4}yuLAIVIO`Un$x!tIxJ^8$h_Vvx=)G!pmx^g&VcsCJuufEL
zy<HFn#G`h@4RvpHYn0`gT-NY4g|5cie9PF-T|69!HqKD!7r}+qV3uMS4PZqyGCIKn
z6!LHJi=2EWTXSyvos+7{+II9Gs~PiaN>6t|R$8tW?LGj)kj^>sr_f|ZIj)OU_v$uk
zwoe&1L~bCEJU7lTTk9-jJmJAx&g{ZmU18|#aehGMG>;Floa<glidPWB(lzeb2naw#
z4_Efj#^E2N3;Xo)24kr1)DhgRc@n=UBRFZHdO=%{<vra0NK9qjWvJG_n--bPryURY
zJq#nVJ+t+NB(FiVU^6WjG%)hsOIV`aN`DXHBs=&0S*~0QAL>~3dI0#|ZK*Xcls>9Q
zmo<T1%hsjxPU5N#Icib9Ly$!pfSS{W?GA13j+T}>7ppeAa9#TZBFbjSfDs^fZw8ER
zjpq$&*<4JTezCOsFVti)82%(a@%0I{E>vWB!=Ly%YE@JZ{dfn;hkV|!Fws2Dv8b?a
z+$0gSm8v!yYfShxar(i$zy~P0i}N_sN~$a8<HnU1&0RYKQO2W6%SF`D__YMIWf}=|
z+1^&SBai)BE2?BcF195SG_Tg1=UQ>pSh*gVo0JQz{ZO7VnEuIs>AU=dR}a|i*_ZC{
z-oOi>Qj-GocB{v7Qk4+Y?zNG6rio10Wn;KV$Fj(R(6_YA5wFm9>!ZzVCCP>#4i;MZ
z=hR$IZkbU)nM<9@B}7$hh-LHRfJa8tWIFDvj`{mjXu!~P&6taf=W9<IG|*wAyB?)b
zD-@Q_shYbgYm<hGLc&HlhruKxu)>r{WVljQ*!<XvCt-4K{G1|Tb6{-!4ONq%XL#xo
z<kVN2+cB_|!`OoFt&%Liu;Y%D1MQ|gz}4O(+Fh%&SPynRnB#iiXpN4)@I+3#uDT1!
z*QyZEY5<m<z?jC-9a{VY*pC>t=hY>l?&n~-%omvA9d<Jx#|E<&5`8?EX%tZEF1&2l
zw|W?B1!jb~=Vj2h*cCWMU*?Zh!!WM<c8~&^fi+P9Y**?@sy~SA6z<BnA}s%qv>ngR
zdA?un{U5Twlq^ml5v03Uw?@td{5@G`^0TTagmrkdSKm0t*aJS)=8K&F=&?3CwwS{}
za2G0*2o=isPub==4k<O&>3~NLg6@6cGi)&EM&YPB0NQ$iS-u~X(W$C2k`2ZrHQwyZ
zkmM;<C>xj1yY;K>?6JB2>+B|Cc+y~M`J^SseK*AAHg2U^zi9yND~GNZL?Qq|*moSY
z3KzJG<Fov2bMU+j=4wW3zcUO$m+>q$<^E(ui9mj!=^zl{uKsFljydoFr}VPmjhCf4
z$cwa1a+#)Btn3KJl%3mXeMZKeEq=Did`RgT?-tGleySvYL7M4RQ|EB&{rJ1ay6aKq
zpa~PCBaOMr?gs7iGgK2~UAI>)5_BZwK+AsklHsXfOz`M~sZ&JKg>^><YED0m%6~I3
zw&fN~WQeScr&y}uAMX+wiZ0)fUd1eK8Ex8<e0N(oj~n|QRy3FRru4hY;{eIX4dXm%
zdd^43f*DB!y$SFa9S5FA?LVNG00BWxjxW8I>nNDfKPF$peMug0A5L72{mUdticNdk
zAiZB-WvC#gM*VIvEvut~Ltu0=x~O%k`)VhdO-o>C^XP*h_NL(9Z)~fvqgtA!3l47R
z52Xsob(w*jjjc<gL+{3`--j3g^e@2Opi3xkdJ8}aFuqu{CQO8exqv??r}D&c&Mlc<
zlRP$eMTnWVFI`pH=KSf>yS@GTKTpF$O?=>?EIX{dY)61~Mii^gQpR)lo&H-jb-nst
z;Z+*G@>@blEv7idU4cjmiyqK}Q;BJ}(6iLLV53}fUq>d~h&=%Y6O?uwkPqg_nF8D<
z@NY1MN|D@OjrN9LN1IGJm|@dR&ta@fjOfx&lO&%t>>`Gz45GbcJ20Sc2(HwI4JvFu
zCXs1g@K7CP>q)L8sU@9iU9v9LTQBM!8eEuR`Wf2?;81+sZWO)6jGY_huB{UJ`c{28
z{`iK2;5LDXs#n885v{I0e=+SPMQ9Bt243oJW8!s9ezb$qsp_}momBRCbNR4}!ii85
zx6<8e`5`wmZm^y=dd`1qu1zz3P4}qO%b<QXI%M`a)UI=tM>o|fTF7eK+=QX;X4p44
z!?4)!(sMy`)zn?$kW(-?rPJs@hk|&uT!mE%+4BnL1sK~<EIqfcntQ+5h5Kb}B?i|s
z@rh9$I}@E`H`EzKVGiV|@$O`_3UVjgaJNH+RSAFkf)MF-CoPwGORDE!^2&Qbt5Glo
z%@QYoZaI>7&yza~=34~8d`kq*RWIYl{q!hiv%-l0iG9`aD)@f@Ts}~}dIjxxCe3S0
zrHW^c`-^!wmK*JZ)RE~4GHOrp!vE6}5ITw)psO(p(txP@J*qDoVkNKsF*I<nzPt|b
zVhgQmFjP;W8d#nLsmR?JAd2mac_BjUvd#N$-z(~UE}i^cugCop7t9YAmbZX<Yb1;`
zxfka#q^XLJOEZ+|!}fKo7T?wzwW_&aJ90^RQ|n@R$N8~bs)49q$}S=rIdoGivtx5a
znYi}|RDT~#ywtCk0DTqUZr>)C+o_et_-&ZIP>_H>;Iv<re2tAlrX>ubv7;zTUd)(Q
z?dP%Y60oa%$SXtr-3Ouw_5%vsBMmhvU8{FyUs8EQG3jBAG!oq&YoC$GdCnkNNK1pd
zHkcr)UoR}_kXw|--FAwqP&??&Khxv5UHhtqeK2e`u4m<49J35_J!2=kQdgOE5_j2a
z^1_aiqN(LZ&j&YA4mLK0<#ru8@;dK20=NxQs}jm-KA|51@3H^U<vuB~;OA`={=JlW
z>&Z^fvvJQmWF?<9wb;(4b00GjgM}&uv~pu!)#RT0&CA5c15Q;W8M`e{QEZ+69FQXt
z-@_)V{qB#J_0jd2bP!0K5I}f=6m69kPTSeNUeUM^Mq(VIS`xw?Zij-J1Ako0vumZ%
zy*Ann{^g}N5{!<Trk_gJl#;XL_|$BsdBgoV<B7V*R;L=d5By{=htgeetu{p=E~xCr
z_&+4Z6YkDc7cH<Db3`Qq$IwS6M@+f)DfucJOMO9%CY=SR(dyS=fq6n(wak!E_B)0<
zE^c>}1<PwWU<0}DZrAcZx_2fRx@PyQd<xFEBeyvW<&@UmVn^uCJZZgW3o@AdoUe^1
z=L_SoxeLBCFBt<Cw<(+p{bhr->*>11V2WLy?=arI7FcICk%}%?`-h<g#ywd1+pMI-
zJdpF!o%%ndpTZ;+l+J)8Qo1YWw_5Jl%02&RT|9R}_tai1F%W5aI=ScE{Il9TO8u&?
zvL7|}9sI8mt_e@ij5;O>IYgTps4NL_WJ}=w66~tT4?$*xa9Z+f!kLzmbB7ANc7M{W
zIu|sZxxaucJy_I1`1wk=7^Cie#z~=)U|JP!m7&$PvcCXC-SWooEt=*!-EQMxJd2NL
zPTJ)bV)<cJA;!-2?WSM^Y^%@hbN%ea+=cZX{ci<N9v}6ZH1?4(PoYmO1h2|#HQ5As
zp*&<SuYA_Mv#UF{T{&bC6dq9fyUFS6NsV=DXm^k*TRlbzK%0tJyCJ6)X(Tp4c5mS9
zSZ#Q2n{nCb#bhg5g=s$8hqxteH;^AyIOm0|NL6|~nH1k3DKaaC>*^$dE~MWjACHaN
zR0~>3%s;%bojo=DaPU`iqK+A_Cln0R<?Y*F1D(|r%Lz*lamaYSRLRJo&Je;qN8#(;
zN!b_VJxK1zLF1MzfE*~?W+~e3uy_-5{M^jvINc#y1MKR=8SraPg~#GA^ykgyq}JT<
zbg-tBnKp-~F?7a#&p}SFS8obHU1(j9YZLO7V3C1s1rCl__%m9tMu{l`>nE&j<sCj~
zm7>!pTHw<)MGkG(ZSQA;aP^5=u6uW}6|`Y3X1l1Daf&~sOTg;=Q_UDV&|$YGmY#tM
zo83D=_7Tvo%kHs<nI}f6$fI-yr&_**dK0B-ZmoWZjE_DRp!V=jw0EhGTG0KqKE{&L
z-_FX(YJKIPmHE<J<<Sfbo9Sz-RT<(ddzow|Vc}{nMw0G!ySq3^ECdps@sWa|NjX<K
zBaPcmzfEm|J+s%6=y~buf&A{{&fY<si%ufGkQ7N3m%N(7?9~4gTg>ig2y{hFHs;Ps
zSXzQ~%*uRHc@x)bjm5A#&22b?3@zNJg;B}6nQ+PtPeCZ@6&SkAOVw1C4`kU(b?Jv-
zi_cw}OH^<CoG&#r6;7s@p5tznpjo%^J>w~yA+4^iF0DzWfQci&-QTfy?|A00$g7N}
zQ9+lSyQv2=p1mXYxujtiXV<;nCE-qZUg(f(_E1T*OZx3kNt1&DV!z?v9^0vuEal|X
zHkmpZZ9?rn`z^hi+r#6TgSlpVlsVc2MK$U$XLD&oT+9%8XJhB#k19!CH0dK0FPWp}
zwa1+`UxnFV(G{<@lYV$x2<XM8d&=P-SIcias@Y23W_LYglaB1b8sIafogZ1^GmC65
z??9y(5-9h!ai9w4I~4^5@Mq25^%;+y-EH*n7hilC`u>NRK$v3ga(<GtKK0y>N%emS
z<qu2mB?#DU-$-EezZ$%buhA3q^}5(Sx1Ux4M=*|xTF1iOm*XJT_D$L8>`MHy<%Z_)
z0Yhdt^;!MxOtbQ9o=*hbMQ>vgx_>WsemIH{o2<4%;M)b^7Q#Y=r2J#~b`t{e$k3sR
z7*6{B`nsbEVQ>Uq_tQ=X@YQGUbP+~FeI)IAd2hIsmE+knQg%9iM{^S~&(~L~Xy25g
z3R{m+G-dvjIPi+`^L3_BvvcN78^%4&<|~c&JIzMo8ad`wk89-eA^M)sghf&S0{`Q3
z-tcR9E_PQ$g+QHH&ELy|_F@}?Z=U<yuh?|Bj9-k^-&WR^c(s=)x4*cg`BKR(_cGl-
zNB%so<|poc>L~k>Bi{Z{7s4P_evjbLkzK}70(FQ3`;H^<{df2pUa{M`p##6XdfM#v
z+V%TpK6fgVz6qr2;bWuC>}6UP%B6ySx4KH$H~`5B80`8~{0JT|r}{G{q#2u;7^Xh|
z=s62>a8<rAa`yv_;$4aRaD<s~(e=pwbB}k667F1fvJv7OMTooan?F^*R)5~?GfHdz
zTJzfr@}v`y`0`tGD~xBwsmRtd$F`*O@Xl3Sxv7P0{C3vw?yZHMkEAULFIwwuI$_*x
z1ewe#@yeLp0h-T2`fTty`X~Zl{Yu24Y_vT>)iN~g#i?2ygrpd+o+RLk=j7W)V4mw{
zc4)-GuVrsTOT4^$8mMO?7T$H+<>-op_<XM(evE11)%uoRmRM@=M#xTN1NE6xw}18g
z^!~8Z^RRlEPIXODa2A0RLDZKs4>OtrzZGe`Ts{+^S&~@|JJR4|EF>!6b4+f9agkU5
zIg|21gQ02V4NpDNOc!z?2kDF^RXSY4e-$*e`&@|BYjA=3Z;w^3{5DK~F5@gazrS<e
z)E~oA^-Zv71q6P1H(!Y2g)`=>N7ZlA>^gXe&Y*<LPX-U0?6}Vp&rL+ShMYtg8^7O-
zr0V`Wi67JYYk71|ILr`CnYy1x+}Wcn9f0}FQA8!mMLZr;_E(=59-TjEvdvL$?qS=J
zIqS_kho$xKaFjq)9uW3mDZ;b)ccQurF|V1{rkn)Vd-7$gN$PGx?%<BaA!YrLrPNEx
znm#KJ4)$S7U>FkvBjT5LUypT0qc2{(5g2`XN1Lp|QDE+=AGVX=2!*IuV@!XSKQkMC
zXbip1Jcf}=QMbyMN{Wcm4<8Y#e}=3u@H|oY7yMN?f=&HkmfgQI@8}CX#bYlJqa$I%
z7U##7?i9;j#>1!G4d-v*|A-Ql+5YlqbJ0wpV^FqcqTxH4w_L8AuYfdsD}N~>7V#0C
z_wtxXo>8v@qJKcpWx?&lJi|XU851#pz;is#uigSHZJqsa3Sow?t`j<pFzzNbxmc)H
z{sos)eKq1i^MN7Mj84p5H$L`yDx>Uh0=2SE+4-e0!<g9o@vCX(^{+u|66f{()&2J5
z{)mSow(&Q$Y<K?V+;3)CL4tiD#I`^IcgwSfDoK8)#oaX#i&x73PSYps(({ln#KXVH
z$TbYYwYoHl2b^D(-aOcTw0lU7#M)o4sc&YDhc^R-<z}ZC8RNwSr3HQ9`)e3pO8O?>
zj+zDcdq+}5@5#CMoHdm9GdZ%KBI^$i-YoBgQQJGM2KCyA69N*rC1-5Z>FT`gVJW5F
zuiyx+HvW2+6ueh}J>p4kXe|8GkF|m$=0r_*^OBWBdw4TB@}fjj%=3xB)6hh$_hk9K
zSE2>#pZtGU+hcX{Wwi3{P~0+2J9+k!JEshoVf8;E4hS>F5aoHv&X#%C%wu22e9&r-
zmXXqL(XnKshWe|SJJBbC9Yo`T5)D%ujt-yfT)f=*LF<NtXV;;jhn}x`RxH0<X^W}H
zG`9%>(8}N|FQgehtWZzX7i-Xrkog0rZ%+OjT!nziftOjLfr8Nx46$vh&*iprxTvfM
zfmSa*DwX`>E>rORtDfFnjcY~av2fNO!*0XazrAzb0pxZ%5p*0ud6%|4pe19qb8y>+
zXU9`yS-``qU$$n<ugm7f)9MuwinQs>E1W{?pbxM-FQu)!qa1jduafn_H#=(nsflak
zF8IJnGoueV&c}fTYyG+wiQN$EXk^6!K~(_`f15D42>s~!2sh6TH^#KTlLBIg)Xu|Y
zJ#{kIiR-~jpPrmR;kbxhkKfe47M~BAd8nu=%?cV8bgATT@+g0kG@l*nBU&lD->u0P
zR=Zu36tl!{Z#uKFg*oE+uu1>%$oXOQOEP_C+Ao%eQH6;YLK@)Bn)***(&q%C*Z%Mm
zw60MXtvryRDiD1Nf%o{xzv-k_S-EOCffB^)u8ixlok41U!<FA_373OkgZS6~3$RTr
zZcmnMo~**bC;Hd9i255g)OPaM$0tya%QE+cN&0=>t~C)d9I~jrdbqRT_J{3S<#2@j
zH~!MLQdjWgrkL2-8GkoqMfKmv7G<s@V_4<Yh-W1Ha_wKcy=^Sdp2$ra)RX@yu}v}Q
zZPTE>_RRW2=P%6HS5Hh89Y>g5{&nanzu~K6+tKI!q;h@CIO;Fpa>9Ov{rcY9QsJ-3
zw;rB^pM$&4mUj^T2|s$)^^r7M39)FrE!7BLWALw4{-wH!gToaOt^%6+_j?|i(%gSn
zi?;B5sBpJoS}j(820xdn(13IjLc|$Jl7*WmzwaJa7GJHjMVQcO^)bNc?`)HC#KVow
zir3&XeHQJ8Rhw@n;p*e{7jMv}&!0b-r~X4<wj=_MLdVd5#OiCt?#HEpmCVQ354QU)
zZrAov0)7Jz4o{l(iGd^Rzx>iq@}8Q=WpV88)t!YUE|(-NTO!2u_p>75&BT>Gc!kkh
z`Z{#`(sc>37&z<Zo9~$g#Scrb9NK=y<<7ucFS%VQw#@}`905I};ZeYAMEFtn#~ud}
zJHf*@<kjn5iq!d_g%Fo~j?NwWQLBAE$US0Ngx0;fcD67MjuA%e!-hPAvd)w|uVGH$
zQ-sCc<>v1e6UTomZW&Jyc?qx+r^G+1F~_5Gz)yYt*0AQ?hK}uOk#HHQJQa-q#QmCR
zS5cY6C#X*}VHMW8{Z9Il6#M$CE+b>`#p*<T&4;jus7#{<Ur5f_Z#N$9e)tU_8z=&z
z-ZsKAW{fFS?+vW&+LZ8_Bh5{6ei3wiBY}ySu~pp_`{`O(&~q<V>dw<g=MKYhbjeQn
z+Zul651V9m)ZU=0iV4$&lFg2YBJdYP`B<P~sxw`I(JuU>QfRg3-Cp;1c00cLgi@+q
zj!1dzMKGDe_00?S@aR*BMd@wdWH?iq%s`o7o`OKz$@$fW%*P!WnOPm04<{v_Ud6-B
zf7xvp@LB<f?WcG6YzuR^aLnkR$1a$qBYjM-wDFiXBHxbeg3t7Cq_FEx;K%GQ?fZn+
zbyYgFO=Ekqo<p>_cS5L-_O&M`H|72AJyZjs{_(FpHt!x5te=MWEt}{4oIM^7b$z*6
zsQlP@`a-P!qtFQWn?T$!T6G&tACH#K)&A##2C;d@_rRe8)>nYpNLq>1;Lh*=jECnQ
z<GVx39ce#>+~H)Q&}Uvs=Dup%)bQ{>jTq5`+wgCtPKXS|^cxa_g@Ntdb<t1Cj3KAh
z#H9)zwm<ta=^+NhGlk5$DgY;!el6uBFn@Vi&Fn+4U}3r2wM4i0SKIHFBqHZva75q9
zv8n;BuiU;X>x;%mUz~1Puriax$u)cW|5o`=snN-Ci{b1s_D0C?YawS~(ivlytKhlJ
z69T?+)TG#lGy2EVLz>((hZdpS&|BQ~FA=2zo&9$P4zhe<2xE3^+{sg4X3RRj4C*;(
z?uv+)csVZYa&!9ru(Hc#BS0#C=i6R2W`FKris+zz`Ew$AS#UB|{zqBnbL3o-0bx5g
z$LF?-W<H$tpqPSr`;HqI@>z6v@8Wu>HexyMfv}&Pb!pY%AXJ|p9MP$t08EeCXIg(f
zGi2$2)2%)~WPNbv)}l)uT`Cux5_TkC;mN5QmQ=m?nALqr#7?99`OPec`Lot%-w4>5
zs~ymP#8Fuc3EMwkT|A<X&kTGsoF9d#H*^XcJ)cxesf6c>?s(`Z6bBK%1X{4;E)N@;
zjg;@`wN^=fS@IN)MxWJw%_8gLB^I~bgb*t?Jci_p4+i|{u6L{7z8>9t<g+E@D-U}K
zM|ebu(9hp{<!zECBrPYWKLc~rG2W3d3w0~|EZN|pxJ-w|6T@fCP(cQLH<4dHk1hEg
z40`K#I_2I8b^4TCllbV}gwnQNke0N}CtR0&wkLDE`FBK)Xxtv*2=c?Ehv3&x>j&m*
zO!1jEOOYXWj083X)jQU0@s%mz5d-ZWgA*n2X|5^5|GOMwbnaW@U2Q*c3!jyD>h;gY
z0$3v-iykB*>aUnl$Zz0|O<iXZH|LIQ)y2J{>EruLs{OT+)_1FM)#uyC*rC#g0vS~q
zaD?(*Sj01>$h^PO3VYgnAf4^faPt~0Zs3gfVBp=+z}N`Rt90S)&i;KT5wv@I=R-Qt
z9OL1l;*p$JEzJ;VZmh&?&i2y!h-fUGEPh)RvDkJ3LI0fTyVx%h5}F%3cWiRosfV#&
z3SMh0AFhk8iB1+wHhONp`G<u4aRfn1yhc1jtn#77eXu6a#eBm_F!KF2Il($XJAKg&
zpJV1mqxLbf2yyRyx%+Hx1fJQ*@lul?JA2;AHv$=xi;c`&58fHPXD)TP$x(0Ll`j4L
z{yHb7vH~ya9&mjUaq408KYJ#(toDTBkPEdDHHuq%PXq&@`z!hk|1No0BIy{O0Qd6E
zq`1!ye0dxBLVw61!WlMXU{+D}N04Y=^L+ltqMYOxh<EqJM+bDu2bKk*UTyzY-k9~*
z9$X93NxJVR&GOF!xv_(K2V4)t{^enR@fyD7MZ}~4>o`jZz`zND;r#m1x^Q1_yCI>_
zR~?pJp#rPO(V2>9<v$UGH?D#txLo<rQOZ>+#&~;NEL=~j{7*E#Ds!C_U5b@=S3<Oj
zy?Xxhu9ejhsc#nN{epi}i$5LM{aH5i(p~L&yJ4&v>>56^KWFh07ct!aMYvXY-h@-!
z$d)_AwP<d?P|$vOl&5c&Vkmp7I{wwp_vY6#OcAXo7fo*C!{P5=dGrZ%qS2LJYM1cl
zCNaJD@Sk{N@Ae|t6^W%f2_iM`{Z3;rAmUjYY6z`*FRlTKxRZE(ucpwd`LM9d_!?SY
zOsmXdh<NbAB62k3o)G6a0=1yK_~zX1K^v7n_CGk2E1L8-l`LDv_pF<wPpc~TH@Zw-
zMp?Xa)<tm9I>A~p*KJCu&Cd1l&6hrlH`~Q%{69>+hhJ0M^92foBGOc(cL<<VrFX2D
zL)B0Ok)ogwx*)v;xq^z6oCrw2YEVNJq)Lkohzcl0q=hIbQWJWIw{ySmz0dC-IOpuW
zXV0uzvu2LHs0G}Q2Ya?RCh6A0wx^E1yJ}0#ZSRR1m`|MzQEsTT6&(z}9OMSuaj45C
zI{!h8e|W3fQc|>1DkaPH(eP7J-=&){tA*+6oKrIh;)=|D5FGCF?H(a2`azA82#f3a
zU3QO3w$^i$aBHRt`Ep0#_hsxABHS!^ht=g3|5{x@#9Q;SwwiA#uMMBxoe~`!WV-P1
zXWVl{{Ev;TWse6ac;?#S%^gRUodXe1fg%2tMIWvc`|oml9(k2vx9!g!mdWGQ7S=mV
z);c$=&7H}0p`-<X3*(|F-pg?yc}+%TE^b44!VqkCYM>=p7qNfpd26?va{z60V!Vf=
zLeaKrKi~4&{N1>gBIY{X)>i(844e(#q<pa{MRYv=Dcp%9=0l`P+)oRerIjA8)vO^B
zk+8lX*Z(Zu?}g(j!R};;HfAsayh{^=_7-t!&RVfF?A-o&NT1K~s!o3F6`St)=PW%k
z?+KRy(gi$=b|HB{F`<pcnk9>crf?P^-LhK2+<Kv;HZHr$zV7>&f`%4*uxyBg{@cw1
zLGt=U<2(6(Rpeuj*|4aw@9`5ow@k{X{`ya1ZD50;-=veCr#{yo?08zOChr_uk4Ohd
zAI}Z@f3wB%JdDklJtwTtaj(f+KGI7d6;)Ds0VOp%J(>R`O$R_H&!wC_X6gp8LAXCr
zp(JYUFf|?ewS<Dd`Uku(K<r4;xu{;q$f4Ac=vHgc*A*CJ%CTY9>+@4*X?$OrerVby
zhJ&6?YEv8$mQifI49|E$GoBq8cPr&d;V%a-9<wUDK`7{B+7!2o-ol>kT&^p>HGQ-T
zNee@7*8f8(9FUgijP*S1#WsD}?Lr<o^7Boq6Db<|I>iiFN&4|`l0St=9i1}^Yb|^8
z$EbMZ=C%z<gkNm9VOwy>bydwQ?8x>pBG$FDFL?+;E9SB{XEsai-TA)2yZYP$lAjWU
zS8SvIz*FVv-+$dw9I}dv2#|p&=Luu~5MGR&3){0os7YsTht>d^<LFu$vPoNAT@-|C
zV^3MdYRNO~WxLM37kmOu6-HLAy<_&__TM(BGd28m5_KM2^u_iYl8}#j;MM1N{@N}i
z^Nw_oZut`=fSDA_u-n{-aaGa#=7w2F7Isv%+kDQMG#Pf3Z<6(b)2N^7$pT~qaJMR=
zI6Ko)sB+24Q+M1g8`yACLLSz(Q-ACR;neRQh{X{gP5h!YT}>|5bDT}QK0SW_B>e(O
z2($Xg_@P}`OLm#vzB9Fyv@bd3XyIr9iz(<KS0B7in8Hu?uSUEZoK-^(c+a9TbBAAN
z0FUrhV%|yz{_xcjonw?EE)@3pSF4u8dryUqYm@mifGB~PB|gl1F8*O7=@Ycar1~Ui
zmto{UZ$yo^-ouqF|FuWEwuObGyC-vUTWc1VO(&)i?;V)i&Tds{$M3zpb<Y;os^VpK
zJxdv<k;XV%7F6Ju`IaWz1BIvI1M`aa7XzwtcGkHw%s>8qgppSm7~uDm@jKA)G)QHB
zLm$&vkDa6J9ICs9Lt;klBB?vf@Yfc;jRQ_j0X_<$*{^7Xh1N0-lYxz`MuB#x3;0g*
z@J{;;9sUuMvv9k!DLt(EwTCKQ{l@!C*LFSWf2E--bH_xjyqr*lZRLbod)*=Pt)(@D
zX6HpB`MexF^V?);%ke;S8F+aF>&6kx+h8MrfdQZeNx5AnCuRSf^=RILtm1OCLa3C)
z&sZJ48Jc*=?Ntp#gf`uYdMo%&UF+iN#N^~n4x;C#3jIF_VjU_%7w?UldJZC3^W}_>
zme_3z_TFnZGX}qUR_`{7`OCdu+k((9hn_j1U8Q_MI>Uu&N&h~CCxsGze|nv;j+bl!
z5$ue8VC;w{G1`a@nHBbr^>uV6kk~x<$_1Y@?I$1u_Op-yQ&QBm*P`{CD$rKkqb=+1
zOv0{y5^_ZlG5%IErX_%#Zse{c@_c}dc&9KN6Rc8|vk*rK%lNbaA)mC|*=a$Ut5XUl
zTUJWmRn_=vPB*m`&Mn0#iaZDLMhOjoiGZEy>V~_&xB-CXH#IJW+4y3m;5EV7gw}Td
zY@L9n>RU|^xTc>IY{~c2kpp8|RDZ($LJt`kv|49Do*K2LG9VE=MDVFhrwF6!Q^aS#
z$nB3QcaEoUokMSBnTqKh(Zhi$BMtbKNn;Q-#t*2lm|WI426q2E*!>r%(2sS8_b5no
zs)ptI8g&*C@Ned53*OIW%ZoP-))COl;Vx{-K*j{ha)gY1kG4UtM{)%ni%f5;4N_4G
zJc$`Ni3P;n({!Xk^pmJ?oA(5-b#ai8n1v;U`bHlly3CRr2f3-&HE5n*-|h`-S{`)O
zVNt||<VBJc|32IRXr)}qS<6rQ{Yf(eiPepE+L*&NmCz8K96?_CHTYEvZcQnuj(MC-
zO*S@~In6&t9SRNO4KmP=+-7gVKkD6cF)w*j|I7Lj?y*Jbd0TZm+5+bB3Vrl+VWRwP
zv++xq=5pB-(~ty<Pe~fr!{Ia8@{{7zSKzWNkSW-O@=>bvN^Ubv{4GzNZNl=wGKgo3
zY*gFS%5jtL|AJgmg5|1}H6=jNK4QlI0)BcLp$Aw{1w5;@DKScH{^U8u5s=)EfNbfc
zlO7!9^t!7{TI5Pyjh~)em=s-L>57TSM)>3pMiTqZ+gVEm+46O*-%_yX{I;>+Hu-_-
z?eH6epG60YJm(Z|7Rl_Y{0C@yBc?g_ctGb8+g`1oUdeqfNcOsMK$X<}NaS&ohr3>b
zxT$NR`hxR{uek}Xs}|MK!3H7?YvZ4iG_SX@Rv=u>!;Zz>;~r{#;dv1xFi*cJL{!R3
zwQeu1KWpx(OkG+xI34o&l7N7pjdq2X1Hc+uk`>Yexo1h)TpdrN^b+k{tlMYCaDNr&
z4?O?6wL=RP>*T(ZgWMdm*;)z<d4jzyy^fpwi$;4d&sYC^_#+CQY{7D;ZmUw&Q7Hx>
zvORxO<NbrFx#{+yiN9;@tBnGqgU7+<DmSP@CBs#ogYT|W4GDuKQNWKJ0U%y*U%~9U
z36V1Pwln!h*y}DrZa5Z6x}1S*=ugo>q~p&I2=&bjjzM)SgZQ3%@3C$OS0@c=2sY^Q
zR*u-uU9lYQvq9bHW~&*0%8b6d$fO@R?NEepJ#ZDVV-iF=M)NYDC#P1Lx}ClX+ZBsU
z#s-`ma>W!sXX@tolX30To>fg{PV-!c<oE6BO!2eb@-$;*$h22gJ`q0HSk&zHMfPmY
zeu<ynGYt@D3VDshB6yr8w~8E}s6)M{Q+3`pRoof^DF~f%+5tDPBtn&&^qjX2CIutp
zW;>!q>228asWj6|*J5D1yY)=T@LTnO@}mTJjxBZe0(8cD-sw~cy1s8Lh3Zj-h`0m5
z6ZZEc?Tcj3jP`qi`=6j{{2vr`>b#t23k^3&$B;gQtl6eB`Uw&mP6x-j{(BH(74f~#
zpT+gGL-7l|iLDKD5qp4wK#L7-KD!BHKVOC{{Sm03sA?66kj#1fRi3Ci#oQxje6sSr
zkJv}~$&f%HVuf8aJCr^|CgJaT5Sv^G?;>>~DOoU~@+13<wlA6f7DM_?Xkc7jGnyh{
z^@EcqE=Y%d;YZ3<n6O)-@z(eJbV6*w3U&8Fb*v9BJulUwsSZ(z*sGYKA~D=q89(z<
z5~Qx5F|RtLmw+5I>wN=7E^J|HWxgix;MWUskd3uil`O*;tCl~m)N@Os5yyv4{vbL4
zUep4M$1(_~v*0X{#EKaW&+^MLDacb4HY2?|<I#_}pE%zmyFjbGHc3xZ<OM>)GkGp(
z2XjJo{_lN0;T~HT){`5QeF!>Cvyof$TBQ^b!kdxPVVgY6UKw{_>owsa0WCH?_6=#w
zX;5of)5uMv^xzO)DD|O&?Tt-umy_+|G*)M9x~|$X_8DsX?(a!LMS7Wo>5qod)c5wt
zfx%MpiAO|IHh_X#fJMfNGsQXRWm)&lwKR#8hF|dD($^`lNGr{KKUkN=ZJv9I58qm=
zi7idozoe6Ya^ye^&Th$sD}s-nO)|~qHU}Joh}E6-gTcR;Z7P{4%Tqo*S2XT*UONXH
zd|M>3d3wFmJ(2J<%L3IKFUQ*~dVh$pRFS?y8~i(`ALB3TbZ%dMPqiexj3rIs?4Y&$
zJl<d^jB3`4Uq7Ld?rB2jWAgx8=wODYwj?_gFDw|x#Tj~3GH9?Q5t25taa&WP#rck<
ziJN5_=fZJu3&m+U!|2?BrnNMQJmcg_o}YB>)hwQxEp*$F27I?f!*?yq;jw~);v83@
z;lW+DRyT84tAv$yYkPBZx#I{bJ?~uIVIBTSL+H2l3#oh1jO$=VdCy1!O(5<3>Nmc8
z8}`z2`H|lPg^iPuh|=)VU*Pf8_MX&XajFJ&`TR3tHy=5Aa+;Jvjvwy`62c*g9r~W-
z4@C%pzr7%NvYN)x$;I~H$E#a{TqblX<rNhf=@KaE9YK%3J9qy|x19Y;e8M(YM=zv+
z8?-gW_*QL`bO#U28^`$T^SartkFXh?6HK#a&&uR^feuHwDGD^d!2k4Je!u|2*AyZY
zzaOQVW_*|*hd<2kAV%XHCUlDAq4D(Ik_8;$NEXf?m&{};J#g#D<npv-wL~1x!Wj|Z
zoHz9>hc#bDId^W`n>u;B-S4oQiStF>nAM_!&~G;{qcIw)K(yOC6y*Kw9$==upOy-3
zVTH8S3_CyhzR=wb1VQp9^`Dlw)+8eiK6nQHKk?twXYMy)y`O(Vsv2o*1pwI6-ijU_
z4;YG@Y6^<O?|<MUky~WJiQ}FpFHd+Gf<@9A`Sb)GS@O?@QEsQyl7tenSvr`cq~}HB
zt}-IOxo|ow?Y?K;mmH&5f!cJta@WSe`8Lbv{((}-jx$gHf4+ib8bUXTwk6k<j5Dht
zImh!lmz#Z!?f?kXa8U0-vwAEsNTnoOe=-Df2;U^9T?C>pQ=LFsv5+3P(Kj{e!jYBw
zd=05JccumN02Q~t{yt%x$;7&wQKlBLulJ6Sqsvi!3WuVoN`*P2Y!UkO=?4=p^ZqY&
zA)_>crLhR%T3a&gH&T?+F2TH$=g=}C2k8FdH=n*}%ID@9vkrUy(zvVe)vn`t103VZ
zWw4d95#D_Jeq+X=l>Dn*h5F!K)x+ch5~$#GJs~w=Z5e0X%UZg0D(obEe|~YdXKibJ
zzy4ON?k2T>V+S+gRf#Bc0XSs&d;6*j;Z0=Cd)ro~n3BcSkq*&E$Xbjj{U$uxlDyJ_
z&vk%5Eij9h>7cnl&M{B8bD7`w?1aXtlCKi2&jZ3tH=W~ogKqzHY=3`(&O^NvT(N0Q
z-2|{0iD9wNp~ilebStvn`k<rwY`|hIbdqTTs($n4&Ej&Y^26c)*Y6E`NROh2yE*Nv
zIUdbYkk6n1a<K}r;ri3j&NMsYJ9l2^p-Z)187nG<mQnQ5hh2<){|WWxUIX0d!qqRr
z?5)tvJWx~1{1ON5(JXjlOT(?*Y|{GrR{fNAMLkRS+~V@J-UCw9Oo4W^uBPaqFQ6g*
zFlFTkjAkPxzWgLL5_;)3ms(Y|3sCUB?W?G8f*{QL*<p_8b|qxvid6YeUr8^b*qDW2
zreKCn<k-vzNeMuLq!r05B(W5{&$!~&Zf5Mxh=u5&?$VdZQy~Yp7a6ar*RAL&r?%$q
z5*8|`t;p$93OLFP7x{kgGeF}=wA`^XtdPsh@aee?qw+1oc?@SfyMFo_VtCv}q6`1&
zpAKV6&uClhgPztqk8$th16X_R&jA$NUkm6Jo%+o=UzU|p@W#1`@m$XFCkH)<c|F<{
zmCR|v?{_UZ_FgTnPIVl`AtgpFP=N!dL8aWwAR9?i=<}&3tZ7Bl0ssUu7>eE)KBEy+
zUg>!e{ZUfEepb_)CuqT8T<+t!&rlvIfO1j^nz;q&mAtb9Ym!+y$qxHt&TEU&J(#Pe
zXol^G1WU68DF(^NnL6qHAASQfeiKJ2Nx_bE>I4&sQhF&W-6m;>cV^cGBPm;`@#_OU
zl=p)UUki542bvskZM63hTceKxZi~??Cqs8cd*dBpLWS+`w==0EGSCA^J-5dN+fPIZ
zm4|U##gazE-1h%{=Qwkdf2e&+8+|ugV!C)6Qyu%4(!M^@d8m%rT?cwtWc_?leSBh%
zVivCWK|XWc!2P7io~MQI{JiXve(nGG9zwdEv2JPjJjTVXYi4GsCaKc+VT#KPuUKbe
zhYETiN)cnzTfQJ&Ic1Ce+!_6<Nk$%WDgMMtb7PT`<{%Xz;)L{$ZcMbd?b{T_QBMdo
z2<p9AjIEFFj4SZ-+Px5bLYrfwtG(D|1nT`wE8(?<*Md}n*iQ&Gv;vP(G(`C2dnn=n
z{*m0tY@<N;)d0mGW=WM@#o96R*+_kb&^Pv;7ypP;gQ<ZjW}16@ck@#U&4xLwnN6G-
zS*4QQjg%xQM10@I)<7z)?Oe$v=D7Iuk;1b|a&eW5yHQjwV1Sp$Wd&~0SR}>Y(u@x_
z=dIk)Y*fY=ts(-YX}W+9KHFR9dR|GieuCP9dwlN5-CfUD!wbOSS^+C;3fjnh^I5GL
zrHcx?rlu^`{m%yH^yj9Ba}_Zb45GipdSLeh1!e8%+BDcR4wYU|pJZyxxk_QpC$}cQ
z4t=Hylo8pHp4e|o?KcdDS{rkaSSh6QGYdsW%E6cnA3@e!j&w~1U><0?w=$V`F1)_`
z+g}yAmSql9v`lTFWF#etJk@N_aYA1(4^oO0g7>r2uRnDy@aGZP(mMxws6hh4|0@7Z
zj#dy+7~$K|UogWVLmNu?n+;K13BFMJvum*Qk0!S{2WlQ^Um@0e@Ur-Y7mi#p^|y$x
zZLwbVF-3_SZTAVBbq69MC$@^F7vf?>3oIa6cP_f;Pe%jDbco%U&GE9HU@C<jy{d=*
zzTcC)*n%#YH|fpAa1PzrEU*4)dCx6laLI{qmvWRE?+$~S;M;azS^943=e?NMw=P-A
z4m5Rz2^vB-LR_qc<0Li^B}4!sse06^g9{eu+Bi17f9zJul*AtR{5{v}6@O4)-(mX?
zzGp*1k5JmDA7eFklZ&nwS!BQnFSzNjQq-Asaq4-JC)I}7vkAlvavH9jt`a}Ed)KEm
zsKs+_*&|UJ)W0dD&_`l3Re3Y8@rd#pDQF<d1fq7awmRqu4%nD(j*21$Uh*FB9cO~r
z*KAvV!zESg^jZPWjVqN4DBCG#tCC(X=F2Qa$3?q7G!}8cGfA)U6K2{wfCp6~UM)#Z
z3X#N>ASWF8ajM8(%DUWAro}Hld#y0fiBPoYOFp<9AGp(MFAJAwp^`k*SQegwF&kPM
zQ@D?vWaQz0*HooDoJSzG$~R|2HirF{B;OEjwLJH{*eg70(><Psxf*vxC!cfq5Zwl7
zjh3<NEc-kuXe1=@e1?@qCRsg<!#e0HMtCvlA{6IVm$gZJ<Wse)o}I(o8pWHe)vJ8)
zp*X%E{U9ZxIF6un)mrIm7>lWrc*=}dhRP@}u15v`2$b78&N<M+UR>)<iDJjpYNsDH
zFtR*CLMukXbAtKoS6O!QDMZiMqYZ}}m?G5Z&ZG;kUYY>`7jU}WGVkB9^M<{b?B^Ig
zrQ8^;NRQL_@Yj9h>I?Y8CXGo5{)cAesa&hb2e<pM(XVR=?3L=dpX<3T0=dt#NFEcn
zc!6=C1$_Jvp*w;)a_u$wPRwg{S{-~EdKM(5-{E*FiI99V?*G#QRJdLSTd^t(Eww((
zkKWaxClhF(++Tb7<JYq1;a|-c_Kbt{h*xtA?pV|t<XK|aHcse!H{jNygahR6FQ;@0
z`g9UT7+?Y&7{!n23Nodm7!lQi?Ed9+b>F%9JW7LQd!1s7yE3sifVbP_?ceYksCZ7g
zgXJticmEEs?gmAyI!-525x$cz!ieTXD@Gt7cZ{Gy_Uw`Rjj^oR0jvLkhvbEYOOzfV
ztc`YS+_qyP+`rt7NK;S@t``Jw`1Wzg!T;m;Q*yP~70UQS!vIh|TrwCsPq~~8j~zxU
zVjPNkm@1USEjVr9f^5ysh0nfZYkQ!XC<m_gTk7oom%sdv$7XJ%&e~|KP5seKoz?7!
zFomN=RH|3^4)t=3nuH9GaFU*QhIf}#Ke?UngnT=AA-XBF-#7`9rwLXY`kTDUw+pr?
zxUTK=ufb8h@*oxYGR4g1vz}9v+fceccvI;o*1(YMCw;X=#Kzhe515q!aWJ-~h*5Dp
z`@A)fH4Y7~PM<R5n-uz(Ab#Hj?)P<|4G+H2^_Q+JV20^#P<XwZZTQeBC8uQU<&2&?
zuCFB<3Y(LP&H~Q`NOG8fc&YJryS#n%*FoBA=TS<7N&TH|gW_AU;h^S1)RjOqXV^W9
zc>%)0i)9E7)^MPu^^0F2AL5^A@$YGE?t7(PX<+f%+e;@1BA+@q45|_<h#rN^=H&nA
z|6RtztJ+r$m79A478~1FwMmCP8j7YpbCpR92taTh;A>7;;n>w*vmempVNsXOboNEL
z0Z?_sNpD(lV6m<Y4Gi%_d24fvPJ?*?Mo6SW|3Ln^lCRO<ZlcLt&{7<eql20_$RQa=
zVnrHfQdhJc0%X9Ivy%&~YDj0j3%_dGX*%`nv5exzYVPHXyU)p_bq3q==ARnDVK$?S
zAaT@8_nVSXfS5U#*6B#_GA@)GMJfG#@gl&`>SD5IYoZ4n(BvjAy$3+weT5O_*3hd7
zzh@I{aR2;yksv1$Df?Q?UltNT2`q$qex_XLeA`Pz{;USnwYZtXCv{ND(0{vUQa&?Y
z&zx>)`4@iE$<li{^l+rjkdu!1z}M><SN#t~ykt~yH84FV?tZ_@Ld57!`@ShuVfjMP
z{&2D6wf$tQ|H&ZNp^FcOt=TL6ISSdDD{rYnXw{X=5pIg`r?J#YCNLwHi?my#<j6Pq
z^56t9uM#iWypKxq|H2&i`W9v|j2cQBx0lE#6w=vy4Hs@vDS*B$RFmVvIg`il57%Rn
zvtjU*&2YBrWS}Q6h^a>+jta5axjA=Ey^RT0TaJxdr`6_6#vX?jT7@>`s9p?mfU~5W
zYMdL>*^=zVo8btRMDm4AhmFgYj?-j~%ZFEw(F5FN6vE|GyF<ts?L(xlyB?Q}B`m@+
zxHIlL)dpgl>$ae~&u>ddum5&2gT2r$<qBN<nhB)`JJ7?{RAs`y(?RI65I;uJ_H_6l
zg(IP@lD_w5K#)3-`q>Cw+uVFB$NI_@KZTyDMA8_cAQ74Rohe8gI29p<e`^ATaNoIS
zJMWi*wTavMcRac6V*|{|5)h=Qt!0zp4P9cuf}(wC4|^Ij!@WPvg(%q>cfXBGPlbgG
zJ_LN)1n{jCmaCV3?ylacC54L#L89uHnUp&@$r6l(6kLEG;i0zkwSgJP>8)F*9pql;
zsWfOS%X#718j~@VwK-b2LfQDrz66T|(|_xuH~A25az#bSIEmaMrT}GOJd&8K-*Op#
zGonHLNQhc@DC6z}*T4&}^sb|dC6oYY0O!TVzQ2A7DTrNUWL@<w)qyUh>O*=`4X^x;
z)t=yvN#t%bIU2!1-(eKCqOv$CQMu+(qNYVFw{snI#36|}E5I-?(JulAYf~vZkoca^
z^K@9Z<yMX*jaLG#9hSkCHCD5=)U7Ov1C>0~-OWQmq8QcQ*`$O(_<NKx5Hxfj6Ujgl
zbyLIdgey698Qn)Jm-=yjgZ|(T7CCV=j)_z81bE^FVR%8!oqswoW<#(27;!rvVU(<L
zWayyTHuzYa-QpXj6Tw{A<C7uLmw!6;Z!+AU&Qk7GR+YDRxCcJVO$Od(=e;`8R}bA?
zHmCYOp3FpAy;?MIH#)qbG*r>>U?rwG6M2HUg4x6>cV+#~1#Abl&6wn$Ca*Xc=rO+U
zriNHxqg;%<4Zr!uO$~;n8xoQK5?WkND_yL|^&GDa6eM1smE)>H3xxHF-qt&nRsh@p
zD^YDhE*`I((8tw*R@7ZyLDF*0Kf@tcA1$Kju3S3@^!AzRr#UFxqdYrJx9&aZ`@-I8
zQ!4OYlxhmwFS;@B-Ep|_)s0!m@-`MW=pG=U;UBvYW1BlN3S<yJa<37#Iv4V9u2!^^
z+t+rH@RN@t*&*SfYKzdS`#Z6`%<6IiQxEbWmKb@ugXNMrtbiKer0BBjH!iO7QlF9J
zjY39#^?38hZnly8zqzitVI*qyh2Te<#?)n!8$L1dBf%d*5<+crLBrWz2z6P`mw6{Q
z39(I&^B?J!vJJHA*O(-iQwAc2c#Te%s?&nq13;-d%RY5>$n(XMD~$4>fFoHM%9RG8
z(+g*Ai;YJ0yaKSMQxnfMjvPq}^o`lSfIR#yg&krMTDlsW5xc$^Ev}bsl>Hj@>ngma
zbKh=HohdG1WT7^23f^nNV+shR#`sQp>GA&;t$&`Ejg72dUTJNP1)ic)pNyEKEwNHx
zJ^j|-qe^I@1jCcDml%+JtO7W1U-JE5Cq(9!Kb#KOKN=)C-Dh~8GOrqqEZ^soGU)V=
zhf33FzU`{AP{G`9(qT_$+J8=5l(kzyAR)bP_|mhrFLM5>Nmcc=3hjllkuR=;#>kF|
z$=INKE4K&x#>|z;(L4-&#>7hyQZ;{WIBD3GNr++7wOZps?}8f6@93CV8?pL*^NFqe
zTldbRz4AD&vWf%k;>hQuXSr`4&#_`0+=#JwD7f6IMcInbML6|T>M>|Ewdd<f@J$Y)
zim0@FLRKO%XY4@U4I=2ofq2^YmbkXTxuLWt1)g6fQfKui4y6#f*P?nWw>f(YPY;ir
zp(4>Aw@<8jNfOxzQQ?Y12}r9N|C%te_V!Y@no!yYhu}C3*8IEi>D$%e@cJA>dGzKI
zC>;IKslo-U`W0SXo6&h~te9P|sW+d|9@=Ez3rD>_P6{Fa>IrACqi^!<cKi|FkERv(
zac6uuZ^~P2>jYFt;T%q7PL^RosNymOF1P>r+smL@eURG4xP10y;6Nl&XQ$HW>xJVD
zt)=fA4ahwh`|J5X+UN7iPqZ4C{6x*1LJ;`2WT_pbEs&Z$Ph;j)j8`w^3wAc{V}VD<
z(&laKC;>nWs(Jj7<BtaLCfWFu3oI{yj<ha7-mr{e^GcU*+yT9Zv+-w3IEPOPMA%Hs
ziYYxTuCV5alZud_8=-HV6p6SnF}`23xSJ5yHZ(HBHF5MAp*w_g`eTFG#x-!=+=TF|
zHDa%Tvt);ETrpjv*tcl8_1GqCl3?&v$>kh{NLsc~@7|_+Wn<<S7mPw=L8&ZMat`dF
zjc$b(OFTJ&KeMZ2ue5!hdG9I6-F?*PGSqmMaXEj%I{XJU!q#RwFnl~Y>_vP)%>J?R
zYrX*BX;4|vjiJhGb_1;iXZOq$JXROHwMxV5HI^LgdLu@-%j5<&Lr_IMO6iNmVdX4a
zD8szZpOVnN<8lxslazM?T)dJ~nH=%HSU#QBIBmeg-9QqD`pdBTTpOn+!ZKLBbop4L
z6zPg2)wyWrsU73iR(GlWJymM_w;9_QnduODC{P--PO}5-o-#u$@<T|-CI;03PIbaE
z@GAJ6iWy&J%+QIu;=bxr)Q^JbfiRp0?sRX|!oBLB{ln|nqH7>2w8QN^pwZqFa_3w~
zw@BIRTkEm%QmqG?n||rd{J}@Znig<d7#`IV)l0CW6;>CG2P%Z=;skjD-e$4VWvLPF
z><+8G`)V=+c1^bqyK4z%^!)5XMuJ~~=(wvMuXudA4r_YvJaevNI)1%NhI)|Nmc{ep
zd+P<0ymTPpf=Idy8gRzZK5%@PXUBjz8apXnKX%TfQ>}QN*GLz4ykwkdkAp}zPF%4W
z8T_|45U<qoN6-(-WxooQgvXzQZF(u7Zv^jClKkp!2dYZ}dSJ3Nq&hIh+shHEv6=69
z`id@!f7%3%0@Dd@{B)dI*aipzUcZc8{*)vK&Dhyrrt8Gs1(g7f2xj7myt8#Wl=H)y
zqiLBZwySSQnQoVJrk_ENwIl)PDnl#nZM|w9mg*;0{s`!3-6hLsT<{Y!M-8!xo#TO;
z<n2S432Cc;{0Tq2IQ4~gB#Cnat+7Q%%_Ytr{Fh|z3iTdHy?jS#NoIAT{XA?};DvMj
zvz(xQ?H9BirWJ%FQknq5F7Q?>KTda`mGsTYLX<|r!C3w2frS7*t4GP?Tc5bAkL@mk
zv!n2d4uXq{r277T2Y7{CP`<VFziP(+-tJ25O?hZr_rBjxXVjGzBM|-vew;_1ZanPh
z#(=mv$o4Jra($DZUhhk8d>GGvIvK%KRLe~P{SIfqWQ4k#GR2v@mbS4p6a4gm=ofwC
zPB>y{tJeLWw(yvV^?=e7g3yeEI$N5oea^Ttu)Ep2o{zmuU%g>`yJTc4kH^Pn2F^gM
zT5!}(MW&^=`2MQ|?!7#-G&gp>m=I7+vH{W@^<zNfNVK&x_ViatQ<A_U2S&v~ecExJ
zWAY!DVNjR6j;>l~T^kc8#=%T^XJCHxixD29^IK1OSyITdNm_>;@OP7(2=<A*3p!UI
z%Af~>+(`PiMs-24>&y{zR7_IVO7CtZ_=>rWoJGUDbE}gdcAz<8$!^nQ<gswXx=>m%
zzpC^=lOgJi<&5BL=i-`Hu!^b3$d3084?g6*Tqo0l2TF`Z;)nla$c|(14hMZ87<k>8
zfErFpDK>q6YJ4y5(K?1d5*lvfPng0fQDzyAHW!g^Ua}(j?qeS=*SzG*7P;;^al}vB
zIMNGdcHAY&otHzaN5DJtK#9Y)$MnNV>uERreS0kgY~%C*7>~?O9V<>0R>XI{U5>bl
zTuq9jh7x|wyqG)(!n-ndnh@Ioy3(f4ErV2#L37WT<w4psCa0rUtYf=5`&$5%@meMy
zx>!4vOiIVp`bJ<Ffi+Zn82Qr;;WjkuWh&#`qUP?2a0z4P``V(2-aovtK5=_IcH(v$
z5Q(6v7ngF|T0ojtDTm>qZbt&5>6<jsC;wr}v6(n$@)}Klk`|{=+#0vnI2KwfLAVlg
zJV;dJ=yuSzdj=hR<k?ldYARokkx({rwE<K|2es9rMHLj!hcl(8%fD2csMCc!ZXF<g
zcRUayR<Gr=?|L47Bo_I{tA;n=u)f;x14wN($)6xVzUt1NQ@JP|*X_MybcTJ$N)a(=
z0)Tm1FGFgt3=O|fs^8p?n%L8Eh$tGP+F;v|QXiz`oK2a(BixP&JR+nmvcsb;?>oaT
zysU4=a#QnIMqPmMYaxVXxduRNP#-$2^c|p;m&z!4+8TVCSUPqg>XRCCH(RTd#$(>s
z;c)m7`tSeFn*c(fqEd5Z!0W>snp6ksh18<mqY=JU_mH7?rf@%fG;bX$UaOSS17#Nz
zYo&_=z$-9AX@Tu;Z_gM5w}86wyAX06x6qOVg>;;2^AwIy++;mm4?4(Z+6NgxgxBt|
zdpOHrHt!kq1!1NN*^zeIyRpN=Z7JBH)JmGV2r4=EJUkIg0wNn!F~cQ{5prYofpmD{
znl^?_gFhmaaY*x|Hh6G)fTa8xIJJ^QCFO{`=Pc^1Y3`GplcuF6`NNxTq`Sq58Ftax
zIHi(;8m0Z>mM&&%dCpJCRa>_STPT5?fhP3cUJ;U}4E%~5DGBU~Z~1d=@s)Gv1JBmR
zzBQN>9%W@3y?9IzLi5-aP>n?z?w7TvnW-Zq0+l7Ogkfj)=6~#HaXfhI;kM7v{*QUR
z{t&G{@ZIS##}?(&%4Ea5W}U2zJ3RGYOMSYYjA>#)QF;Q~hI^A36IXx*H((~xzRvqG
zA}0HrVtDjN1ZkYwZwqa&gQJXnc?-Ap`v7`VD>J6f+fxBw_{<i?^=RzIdz_DlF8%2A
zTSFAP9wU2sGZ5>k`0*;E)^xi>ih2Kx(n96CFdn%7bx9dDmSb9klfF*cMrq>|Ep6(V
zBmAeOkptgzJe$V2sih#5YjHUPtBXMzW^#RgU9Y%)0A7Wd_8tF#GEF)YOnLR(``nC9
z6P6<(lLQeQz3s_Z1H0FmgZDvW4ckUC!(Pl^R^O=~lHS*$nlJrqPW2P>Uzdei{4$#3
z)nr@#j%rlw>L=O_@MM@IiRK*lY#2T|r>-Q?X|^PqV_p)G{fOs+%ioi|h}e7L2+Zxc
z{)RGS&qO8lR46oa+`e^MYOG<e$FWu!dUi{G=z4vs<%O)ZVqkjt+@=WwFP>FlksaX~
zgL#DA5zDAz-i$22&nIsYe8w&ajtVlXD}$PH6W;E$I-9F?*pIZ1i=p5g?IVNgs&4tq
z<B+MN2dP)#h&OB;rI>v+BqaEsmg~uLIFfl6oL?=#5p0@wo8z+ZCg0Hlb?}<hSrQky
zD#L6lfI8xt-IIbE`S6yJeJU_8RiJi-(B+6CxL=uaE=rtYvz2YJtsxenrtcK}Xhprh
zQw0$84F83Wu4DEk|LWwfK7k4Sw`7Sa(x&kAR;SO315TfJnSAnnEkPC3%u+xNYu9mN
zt3ZJA0=ar81eKSEA(j7kUS`&D;nXuEZ%aAEGX*>|8aBJhK}SaNSOJKYfY?jwZ?&JK
zO<$z00OqavLV`N}!8U&aA5KshCaBp1*r`AbM70&^+DEj({aw$Qv+;=whMhX+F>Yx{
zE@lLw7m};T+5t{<#tTyuma#XE2f^zFh?is65l*Kbn{^-eS3RNQn=nhBqJ*H}bvk1U
z+N}l<M_l=04sLMMF~J(cmlr)yXx&I2k)$fN4MuY4wo4M2B!u(pGk@>=g6tVb&b26~
z4yeJCV^&+wD@dOCB~=f;$UHuu)Z^o)10!i_!2x?Cq(uhDnVO=+->k^8pBZz{j|pU@
zXHEXStlMe=l>SlVpPo$ZF;qi6N2~4d!$gw;?j0aDEO?o26u(Ql8B-3~3oE2py+<}S
z<s+kutArJn53napbao4*b~#YlgA3#I=mGyLi`rLwdVAmkf;QIER$fNGg(QkuvQSX$
zCSx4`!xI`*Em$v7m?JT@-LLv?bGqFrixM+UvSBiEG7Gm(?2=DJcQ{+5rGx!si9ax>
zB6?|RgfoUsPfU+{LmMr>J8<XZe-)xT>!+zS<z`c8=B7Gxnyli7adm(v=6y7T{ginf
z#8M%WQEy_9oCD3=-d7=V3ht>XemptwW6UD#o*I)H>)sPTJ(p|38wx?}m_a&wHPhDJ
z{~inw%er8%)J6EV`)h0p*}Zn;)~nn#+6|HnQXixI>|2gSNqF+fpUBkgu^PDfERwu^
zz~PJUN?69-JH>$1r?I39roN0GH8<h2VONvsC@y--b`hHKw&w#q-4a(rUJbL0mp$Kg
zW@^qrT71tq9V41UM*i$CqhxBu0qRHsJhLyw40KB!c$K;%;bO0s6%e9${43}z$ZX*f
zJ_Z!G7)#Iy)cYYy2O92T3sC!S&F%_N)_*Ldb`g-SUMp_aIvj?u!L66yw4nfY8~vgw
z8!4HMm;yQ6$S5KvW$vf}e}vR@w95b5cQC&!SzK+DG~=0#E-@gV6X_tF-&_yQ-{;`8
zs@`3@&lb7D3?Xu_<c;G_Ww4otrpP%%*8F5HcKar&2HZ-CslbM`NbWIc=ESq<x7+$x
zMs!`z`aYd5G?h|~-Y^FzB!5=*Ig=?P)9}>^G9f;VWQlf6`_8g-W#n|-hIy`DQ$pzE
zuAkB?*1w4FKu;s#s|^`+%T5)=yXxduEDHxArnlpbKvwlk?}t5X@X7lK(2VjcL`1RC
z3r(|;#$)1F5(io@DCH4i8z0y;&9gdW@)Ubm$@g3{eN`8vg0A#&Skl2arvSOV3;Eyv
zDhEW5Y<Yz>(_S&Kf~n(F11eC4lWIxviiX{wBIB;bNs0EhpIM+xOx1gMgh9harDXEF
z!IFv6{AUc|Q?$}v!NZSdz@J@G?dax8UJ}G+ab!_qR}mz)c$COgWG(n5{*138v80M~
zL$+MPtBq~vxJai0RMPkj8IeBH(^F)cbqcs939kIpB@vYjAPmZ-Y}1+>NQF~o;HMl+
zmI^)32ur!;B1zu%OD6Ve;-N3MQL76^422EoG&CPcP)De<^KwvlcG!5?vk1c>ow>-=
z$~Y<y;VE6xQV+B5YE`E}YyZye@eN3t*nWWSBS>|YOe!OiWJseRuSLwWO3HFQeZd66
zeqU`E%l^5-*-X+1G|LX5Q$dd%VR7{!^~6%#th@Gn&q>eZFVZ>Q{?~(V--Hi>%&Q$B
zim2y2z+6*M^3S!7_;#{T4mk-AYyiaq&knaEZ*vxR10FfRXxbGu<-%Tu@jqxDm3s5e
zrb9jk=U^#bHn9SerQBJ?Sk`2KJ20j?8H0cAs+(@a4F&ZlEXW&b^D?V}s#5(na}Wo#
z<jT6E^yd(gm=zO}MOa+E61=p?sxJZpy9P}CE4Rz0Z4iQb_cGn?fO$z}D)=fcQ6Zv>
zGvEjh1qnv(4{mH;?oA9o9wbQ@)4S>d`*g5YB2&}x2~#jWjdm2&aYBWs<@8WdWAi`=
zJ(YW|F3LvczFcW|@z%(X221&rt6YG=;az6h{F2Xn6~IAPAoZX1ygV#I+9rnpVI(JL
zGqY=n3$Cs@w82?E{j;}l3)&}P2)}My4Fwqt8=<YVK(7f}ap>F?9pHt$0g+*tjCcWz
zRMCia6!k#Jn!Vzq$uMju2D<5pQI)c`$=lcwP-57M(uZ!YcyOgxJ+`#g$s1^b4L;w5
zCncnTG#}`<IWTMGQnF|LIlvk6OCl;1dp2*8lCd`!-kzE9D4!H;woXM`O5#9GU<z5V
z14EJ{)m8tzUBiYvS%Y7>@u)v?&2p(*Zg;jgtSsPa4||1*2#U|i25yceeIs#Ty@Wu8
z#}jmfDy>BYUw@PI9ZIgLFHM+6R&56X>^Q`vPoyc5WuO9IO-P)FDI`H2d{y!fwo#js
z@N_xO3#n~gBp#9=6-ABpe2zsD8U1*61ygXL=WDDpV|@ncmQN)~z)Tsgt6o<PHW(3f
zmni=zhHv)i;3u5#(L#UpEls*#C9$CqxyOf~aOT?r{pidM#_B{71f`I}K|;I^eXU(!
za;r>lM|XnzGu;DUkesg3H1jbJWgD9%HC3@RgAXl)8>FkjM$hk`1eCYkC8>kWYz+<G
z=l<6Sj@DNDx$i~mPMk9;jUJD08c2-F=!eT9>Cz)V+I0$FrM^W7j|iVVa^TR-d;Zc0
zMIrcT<DmmkH(5X1AoFQJz@^5clkcDd{Wr8cSpGTDPR%$W{P1M;4L>rh-Q!_VzTH8K
zr3G;EsID&TF8tgkM)`)1l&@7sty6lM3Rlb6K6hSvD`hX9k~^r3z1FM#T~HDq!Y)!)
zCna1CuI*|q-;Z`2M5!>BL%$Z2r<L|4a6v`U+U>&hKvJTo>W&7rE#rX?gic(slfYhi
zXA7a9N{X@E5J>7VuEn1xsgjJSVxCWwCFZXL|5L@calZCRP4kmqx$J(7r`vj%LQ82G
z$Ut2>Yf!_dfBXCEGNwKs2r{IHEQzB%c9?$fT;&;MQemd>8C|(?5$->3S&fn1dpP&u
zJH8J$VC{PLie?#`!)5UGeJ6s%h&PAsAl@pJmQlvemi&>geX$5pM3C2N1a;A&F{M}w
zve7?-Eb+GfdJ<DIy?17YNFPZf2{ue_<v`Xp!>SOWeK_7c%=5*-&*+vMJ3|wvWSYEo
zqA!^xOjxMWpfwwJu;wfv#;b3_lI<=8GzuMU=m4*CS}x6ONXqz&X0h~ewAz7(dvSrz
zgD9#v>}@KwVb~jY9@b*Br2EK9vQund&{mJY)Y@O>^X>-FXctG^9kPPNNHNJ4=)deA
zKYIKyz+J6&#${ivxcgNRGWAae&p<IrG=}GE)x5$-Bo}rTTg2Q}rtvTD?F??Wrb7`1
zL2}bQ+w~3KR3Y1UGQRNbEmupW`)1Ic%LZ5Y=)Rts?i_0=1!awsv?HSzaRyz-pry48
z$x(|%SF?_K_8%5HtgdNyt?VNYZOCpG#46EwsejDnV1mx?rzqIE2Xx{CGQ7p{Ji#Mc
zMBicbd`Wod)>gt0;Cc;tuK<1Nv7uvPMU&+hxK7r-$ei4~#SSf%h~_2qWH=~KF;ybs
zl82MZL1!rgHB3h`bm{Eh8K&0PNRppqzo}jKJg1Be*%*x#AI$2{X;+N=90o}4P?cvS
zz<h69r&jyCpZ4ZRsAM*keKB5$XwExo38)MbaG(Wm#85U>r>U(EJT8UIFvsDIE`%ZS
zN_bxU?vjnq?n9MwK?UWOOV)Ckppo=IYnnfQ<hU>x58-$4<!#xE;J<qRn+ssT$Fy%5
z7pTro;WTRMy$34spv{(GtBH!Db%QA)?8eZAXlDH*2C0_I->37;N`6Nt6gYCK9{*I>
z+}9bstfBD*Z7?iH$o#HCtrh5y{c#{BtH#+tEW((su(O*<fKGyW8!!G8ldfwtOn3=&
zJpEI{+Hg>qsZbg^P|&(L`8<$Y*A6K$Ej>*evQj0FxM9xun0XNR;-rEE>7(Y)4im<D
z+nCnzI?6rFd*aSIs9uaFcp`te4k)t<ZKwg?Q`M3$Zam6i<aA4=JmA1%D9tsn65On=
z=|f-whtBsbf%M)8L#Qc<nQB(;R5`QzsK)Q{=_@+Ti90b!=mn#N-)gj!d}*-*yUO^!
z+hVoX+)I_xiU&e*Gh?pkMql|F!i0$2yeKa&;fh5kJ5Gz@K?uGr93imgA0Q}}5REz#
z2ZU;LcE2HM=*0)d^S*SEHVI1;<H~90yQaCRIg|RJ=YR3*be_c#Z5Pjr1;~O<VUgyr
zd(FJ+GWH3*5q$X6_nRc3s5r09fvsdQl~ENd<k}J0G|oa=VL=QxXuPHo8OcmX1h2Rn
z41b0+t{ruB0n4qBdVDtEUAVj9QHVd}F*1@bV-WYz^HrAJFh`k89@AweA$zg3v?c!D
z>T3!?$LZEor%XWEsxl@<6pVgOI8RcezGaxto5GdY8tt>OkIF@OsYwqAwjshl%v=q(
z67G7wM=6yn`o1|0=o`#o?etWVS7(nm8)h&_x6Utvz^?%gLotk5Kg^BS0iULmA}%kE
zOpNnWJfZ&@Mf-vwL|ulhJ9vBBnu=C>iZ1)>`DpOr1y8*uKV;L+MgT$Sp}OiD@0(C>
z?;@v#8FysSN-<IL4?-C_pK!Y-8&rQe65Jfb%s=OuMTxA#=Qf}CyvOzb3WJqMan+Za
zUJVEJpW|gtho{;jCy{BnJ-f(nhFlS%)1RYViVs6*Jmd2^*3GGw!K2ID|DCQtkXqyE
zog2Uj-`$*~1wTaIKw_%H@IDk1_kUJ?&zES)YNz;5PIgHLaXCclmadY#$h4({R4W40
z`z#(<D~)j)bruO?DD9v1Q0Q|l%57?x#>*SZb^MRZPHWf|bNq0<r2-Fsc3pbYdqIgg
zn;M((>yn0eD0`ic_swmrI)X1Fi>IC?-ke?90uH(+2X$y1XwFg*m6FXw^nP<SKf;z=
z6%}6nrjzEZ-gsZE9MpjQy`_jh=H$uE#RS!dD+eN7RaoL$emUO0ZVDzTe0R&nW$_F;
zfZRlzfrC~`86n{z%l@TXDn8aiOg8fN&FLPMKghXf_tctDJ(f`J!R-XqNdoD_tFjHp
z0)M_&X?}D!L_P=#2KN=A&q*x=tM4?X%J<_pAPL448Hs6gstEGuid8R{ooFRT#9=Ug
z36JEuy|vYOblJ7{&%1Gb$QJXcIlxAav*pH*&$eJPMN~32+oSr`&^%&KWJ7;bqE#Y7
z5!g%rTX4qcE=!=Wm6=|wZ3M?ha6R|D3QXIG&CL&fHGHh`8lFH5%V_Qhx3ySWyDW;!
zefxg0C6<?7^1$QUhn+(kLK>&4{Dy+87{`15zo*T1QZrjf>EQBoO{-9Yd#$OQa{`7=
zn&gwy7?4n-JtUsi;S5i~&4_`!&lFCW535;msIgR(=!jc=HK0oN-uZo%8lEuT{wW5M
zXb0>480slwDz+gg$v;>*Ba=A5PCnVqGmzA`^qhGa#Hg_!9ggUUR9qNOG725YssFUH
zyEcb=LX6e*R8k~ya~{!ZnIxyq5rEhe*K2zB7)gA>_{y}<mi(HV8z*1YGS&>#_OLMt
zED6Kml`)fh(*5$sD%8wXkIC45X(H8y<;zPgS^N1*OEJfs!-?7tnN68m&%}jgvjlRu
z-izdeBuGLA(4D4Eo-u3ci~p=zfv$!A_Bu@*(LXIFZGWldNA79#DlfjI3qcT~U9a9O
z?bb2_ppp>U;Xjb{%ytEd8G6O3mwfcr4(E_#7NPD2Aaq_d;ka)+zFRJG+m8#I*O^L*
z(!e;ss7%=@Si{L>uAR*4=l+DGnJ}hhu=|u~&LhhovnR|#GT18}%v3Qj0%5we6qwI!
z2^R!E&jR<PV2+!5V^U!unUv=9{t~|^bD_mnqXg4;J27kL$5u)aU;#BqZM+0QJ;KyU
z@k^<WxV^VFw84j{R8Y<Bzaq(OVxfcYK~PwCB%x6Yv@-}LMEuzKv}*xuk}vfHwTn(u
z9eN9&q}1H}I=R9e0?tOl@uWIYe_p8~4*ikg9PFMXL)XJ4MQTdEzZTm#;xwzyRtQqI
zpi7yjA#@8!v-*hvY-$RLcCGf{wwM@C>IVxpzIC%mqxl5?!MF8XLJP=4T|T-)_h#f?
z=nLYzOhU7E32?Z<^v8Hi?N9F`3ySvv!DyRc%I=zwIRp>?!&tUAGhNpGFg-SyUorDd
zf;09t`RE^Q(i3wzA3n{_lY{>mrSA_5*UBwa<(_=Re*b-)3{TLaEH>F<S&q@9i&WPE
zCniI;o~u$J@-&>n^<&;7=z>vN>{Q_XF7dy?tzf|Lj4;5nSUdk!7-(woJp1vVMuJ0C
zI=9=byqLWO*(}rGvr;hZ1!jTS&x!(ZMXU9&nG0`oO8zKJ$olWJ8xakMgjrf`84DkT
z3;noIJ!1%9<~o0H2OCb7J<jfrT0=N@URt>3E+RR*pp}e*>(le;jl^DWJHAp7_1fk`
z5(b$-9d85f4+b_qg9JdjjE%I8H3l}O5W*QGGX1(ivPf!Gqd?qWjiYqvvm;trR34%P
zR71@Vk=y<HVN7|b(xfRWy(R1w--Z%qsO~^R8n=5>^{0Q4)3*T-=|3KYZR)%pBSA;i
z{(MO;17#nJPM)1^pZ6Fs7tgcB>7(bI4yzI~C6KhkV5U|5L+&)YsfdTo*<DPocQZ%l
zf5}ioCd0Rnlj4~dK3IuEsvtZQtQUPro0Dd9;|!N^gA=eDr45Or&7~$c5s1koGNF+W
z7Icr>$H3BcJPYSpY0x2utZF*6u?iyP?i^DqQyH>tupZ|_5wK^Mhxi$rsi-FwVM0VU
zXWJby&p7R)i}(zk?|u$GQn|%50+N~z)OGOJgY9%ZNDlB^f%TPbZs6HOG3vlmZd9_U
zy#%z+H(`1ah=;Tld4R~rNu&1(3R5YM2w|7&ZaOj<S*Lx6GVX4d6U;px)+}7y{h~-7
zyENyeLfm}>Q4*2g>i_ZcA!-^yNlCmc>N5A^YSTgB!txL<P8O34JBQ8STo_p9{6g_0
z;KJx*U?t0st<xw4kdoElPkLok?&%|cjY(?{Ja7bJFo^xu?5p_69@*Y3vw7OJrPq=&
zyAcDfye#9g<zs*FKJ(#-o)VBT7)XMEGd#3O3V~J^!h=ZoryLD;?L4_zY%)CRO40+|
zB|&FmVthS3JaVnKDMT@HOiV`cta$6d9xqE9GOv9K(eqME)GE`SWK4>q8c}&jG!L3T
zWYrENlfbU6LsEPN)eaVuXXN?<g)*)Q+t`Vv<(|`}i=zjO&lC!AD5t*M%|ER))Q+OL
zfb=CAr5(tp|Mr!5`h57E@{igFjMAGqN_OYi_c%8=he(Z_8y3`G2R&ed++%8a`w4xW
zy-ONjrw$Emj@w-wLG`?w|2Wuu2Z@!p8VzLj!aWOFw-<Ym+ZjsubW>6!<GOu#td-*Z
zWmk09%@F{tJxyQblo_)p6gMQ%SW7!^E?QEx9bfi!$FamD_<6dq?#R&1cXBLzSQH|}
zN`*Ew;f7s3Xtdk-?O=~e>z`*}MvH41YlgzF>(raQ>a!z`Plye7EYG^0OL{g}u9IKa
zcV6ia81v&1U;7h<mgi-5e#ztJ{LsFj08Gc(t7mO!`sJuqTL+*-I{rE=#6pubjfP4d
z8d{}la*tZTToe9GaZF`0_RTII2i4SC!y*?$$H2%YAw2TqA0nTV50prnMFp=5A#sW>
zKuoQ|taE)9OFVG~78#q|wS)}@g!Su86U!{RrOTfcs_YdE_XPB&ud7hSs|9!61lpew
zwb^6%9IcWnT&xc^ny|$bziU8-p5Sjz{RYMvT?t-~qxyUP?OVX_48>73s+A)+pp7`L
zhNB#VsE#Fo#+J4hX^glPqyVtQqx^q=xcck)6SMPC({kk^mhLFDE8pj5+pjx)e_!c+
zU^VZhc_MYi8{}ph!9KS0vz*k|Bu#Qm@?ggH3Lnhh0QaA$^Y(ryhMi4GA3uAf&h5`6
z*`@RF;5Jq$1N1$`nX11Gd5AV_$e(-KBQGYSkW+)@e&=Y$PW3C%%3ae=M~3skZC((1
zPX5B14n8+A5PYqQc}NE6##nj%*2=Xyz70`&Y%lwq0aYFATgq*Yo;vw<L^xa^osTJS
zY9+Zl2A*SQX}(o`8uGvScrI6+NfOyX9~d&jPQiK8m)v7(f7X;=3+8U*aMnciRy{=f
zR_dIxe)OeGP28%g*7I!Sy`z#wO#l~2^D!66t#1i2wQ=n*JEfcVKGFp5sTW9>a$n;A
zXnNJiRhLYYtIfNNSL)g&RMr}dNCzondcK}7!Pv!*5u)m|?G~cu9N{$o9eV9Fp*PUA
z_%KGv*lg<rA8xf%vxMgYPc>pgmabWi3z4J$KkdC&R8!s8FC0KYlwzYdkpL=P1*Aqq
zN~lVa4k}GT6Qp+@6cnWGNN*|uP3XNt6cHj)q<11jk=}bdD?b1Clrg@WZ;W&CUK}n=
zgk<lv=bCHQ-<&fE{zD9@53p3$bS$xEb1MzsEER8n7mSx1y~(YvPXlxs{Idl%-ADu$
zld|wTHD-|x`Us;LMD`HgG!Je;+aGFdc0O#|EIKeDcj(bueo4UDV7H;$kyzQCTK^E&
zTjlx*Wr@^pYq@KEQ0!BWWtf~_n_7J*8z6?|`VJ54XQo8tC~{myS&V}q$uOy=gqlWA
zCn3l^yGPUIq4s;to-Ye<D#U)9f4}>hDQ>F_N`IO{TmKv=4z@}lezO3cz&!I<73ar~
zT>@oGUt1$v<wBnctOkaMG>Ix6eZgFYMq1|y=RS_xRGDWisuJb(m?}~^U-b7G-#&ky
zM&0ouvggZUwp4)gOSmr(ChGdXSV)2H)K@i(0Y9SNN)QDx(#xz*6`xA4sF6tT;Qi6G
z6JK}4YbdCzd8)%Up35^CF$H@Ct?us6pHJx<ak+FMQmt4r?0Uh@Wyr@|ZPC+cQ~J*N
zbK{uP+K|WGx1Wv?SbD28Lx;Pmr$HlU!mIb^h*bo_4bK>0(6t>NTqdUAgIg#9pt+A9
z=3ug4T!>yoN1KSFGo-e!5h4D_;9O@L`)zTVLdmFwB}?ssJ0%{{40O<Mm>Ay~63_j4
z-GWXJ+F9o+R%6{5WfW>7ZlbOMjI-eo0x=PA^D_1Mv==SY;n%|S#<@T?ItnDcM2Ugb
zL5nvLpi>nXTgfLFki~F-z~`QxhWe-hNjsY$?)oe~g=E!`D(1nvT$ft?6b+Y&7-kPa
z<nKkeT2^1pgSfeIev*K%Jz(;NT^qi&vjRQ*SZ)UN-AVDQ)G5lA_JfajcD=4^{QhXJ
zE~PPf;vfrS5h2^kK-47pKIPwSKwPM%*+IQ<L!#WkDb9C=EOw>QpAZ)|;VU$Fd95W4
zP?4d_m&QMdyTxqiNbAsF@<uqBt5muoT%GX~_u94y5|?FhFt<|C2m*8siy(Q(-J>_R
z9Xgw!iowYPv>lDQ<g#NZ;@tprAJTcnR3BZH#M$c0Gn|2lg$pD?95bSN=PGimxj#RD
zadRTz`U@Y5m8@i&Ih$w5(qI`APXBUe)MmW)H^ntYNmHQ3`x3YA&-cr!PfQO?y?W(U
zN@I*%qGF+Ng&i@|G<&_bz#}Mhd2e`!hn9tOQ68IH7(=G<v=b<2iX|aY;H9%5uRHU~
zP0ndG;|uWDd#DB)N&0zdW6yO2Uqx8-0|<a~0cWFB+Q>=;{ZLUw#sd7AaVLN4`toPq
zk&Fdiy)jP(<UtdAHRX=nb+bE=&NpUy$ds@i`T-(MRDp0qe+t8xs>@41r2F>MoO2aT
zCoxcd8l3fDKM{kKRFx{*MsGKqY{}Uwq4#-n5MTr=!UC(?Ng6wHs3=nc8iuftj#}%Z
zG3d1y?`1sDnc-v={J=<4&Flm~cugF7m!8s_nWb={=oFv@{r+Gb25HI@Nq(UPJ76E8
z-IF&+<0m)JgVVP7a=>O5&H@Q(VR(MLW*$B0Z^+w|=FNy*0S(v~+{NI?R@?I>?9KQ^
zd{o=H>6+HvT5t+NyQr-2gq1n-ar`*IJ;P_sayxtJ<s%GgUc>TAti}yEE~6WPM2HLY
z=Etsjp||VK>0Q*@`ZaE4dk>ek33q6Ex$$V6J^<STI)5tQjX=vc_EK&$oi$1;IMWg2
zaYE={ZCrZc*p0}<iKsb~lDU7Hm`ys><?r?lC);}_^o%-B?Xm%96z&<<Q$_LEu_5k{
zKpzjPxk&93x_M|0)b%{K(24ZpAnC(~-214I&)@twB7`F(?OL>I$J8U@4NT-QCrD`3
zh42IW5G!NyY4lbzzJc_o!-pe;CoWq?E~IeEk%h&ou70Q11@u3SLN;%9=<}5JT}-6#
zj%$2&ta@I1PTnYP@QVs2(%+SQu3Xj>`*cmOF91UkdA{POBcDyLKP~dK3-<bz{tj#4
zkUam+K135x)bp~BZVPsNN2l0A+5P0QfZ;?|GZ4Wi$%<LUGNhffbci2S=bs|%z1*To
ziqEK(Q~$hehR+$Ixu5Yf2P4^h!RG7>*{&C)wyE<E`LK_>`%Kj#nsK-<@t!0S#bAx$
z+0$|_bycjf@;PsnH~Q(CyQxER`omG>C9WK`w!`il0MgX1o=IF*AH5EO2h8-*Zd6YL
zS6BVp&a5V-48?ss#0TQfe3Tq-`nY3`@(#2KBZ>dDMkPDp1#F1`lRTP7|AFs2@2fEP
zkEsc~P@g#vTyg|chv(@&b0A3F-yzn9%t^s(s`EUc)$2c$bjP`g%KY3Cb^kUvrGRZt
zAv(XVi)^I{sd~0oH=lJv(Kz7XD}kE7A|$6}TEnTM`xSE>w@17B$@W~0>{&VnLo{Zf
zB;xz*AO7aSPGRo;J?_h8)OHj_pEC1gd$NJ_f6!C=lOO)W-wSw*IDpRoT3z6C07D~W
zH-Ap8)*^#=uuQ^&{m-`&=jHkE9V8iX4FN4jl#GEkWZD&kD)69IKi+uZlQlT^HG~GK
z*?5kqbKT~_H4r3V2uK?fv!FP*DzoV*S=1LAFzx#K=Tr$d66f5KUAK<%?p6Qwo0Id}
zkOOt7S!kJhRE%eWYA;(1H_lFo4vFH5Xbb#mxhAWaX0WbPbp49<2rqF4?u(%Dp~C)2
z_`1)(;!yhInG`^Ib|GzxFbt3XsgejXmY*<)SoZW?-MPornAG>3Ilg7#XiCri&%4C2
zMVf)B?qpR~gn_do2u46U<Bxhw?Ky_wJny~EYZN%X8Qa$86L}Lvb76bWkS4^HdC-aI
zfR_s~LSF!3@DQ?MG+y-)O!4up{a@-gb<+L1JbrGx^9TLd9UP+_rK9?0WbS}DCr$vl
z_!wa##|3|d-?hp81f)d|W&y}<%j;RhZF0DJ(G5roDihr%*p4echkTMG$T~iPP{@uy
z#B-5`R<YH%owD_)F3^vv4fDJ2#i6&{k7&Fv{AYFQSUL(tk8W~}z#;YaS@}4;<OuWo
zTJG2FK7LNVDdXSP{nN5XZ&n-R0!z3q1^PB=4EVB81*_wlB6L?~;1#zfFS1)pgB*-x
zY4IMVjN&FQ(ek0;>TZyGJuY^`$dtjHzCLrIK56M|<{Omjx_S@o+J5$yx5(3TSS=Gx
zXMHXG53%KTe*^<LxX2G`%jk>X;yqkUg1#)PItBMddQrwg6FB>L|1XRWSYuO1WqZ;<
zj1pk38{*PW9O(dEm<n2ia?DsE89`eOi;Z(c;LY5p2?3>(IPa%==&O0&&RyMXq7<Th
zOjqY6ehi%&$DfDB6i)pW5Iq-V)CWKSzi6|jWg2~J`Yn?)zj8(AiB3U<2M9ge+bEPZ
zemhhx#6JG+G6(We%Ov^%1BhIRu(QvnVNB;bFjb_x3bY3lSk%p$Je}fnI!OG}=+T2X
zmEY%kVyuCl_jvsRBW*1$qjVXGvWg(6PDtqyZ_o+zofJeVq_sJGQ0g(;Q~@CX=HeIW
zmyfPq{pcgG!h&Cy5QJCI1NrMJw$eD2s=TKyDpJVX<L|+Ugcb49Z!Rj~bgti#y74XZ
zEy_wWgrjYKk|6-_-KuhbC0YE;UED{yKYebn3Ls9OH*fWo_viD4tux0(uQlKi<Urv#
z>&k_m-_s^=#?@=YSsS}<>jV`6tf<z?1IIH@t=y22zu*)VC-<8x%4)z#oe2yXdy1MP
zpH$gVRugOb`0eNC+}}?lo#OjY19AJiH5ewMrTb$8*?0?^8$|Q#%l6dRKgZWr*wVvz
zD}*o9cf3mzF5QdlVuU{F>BJcc0&tJ60oTji@8VB|OtfNtcDJdrfy2Z&(Wa>f8XcVn
z0dc0~>(I`)vF$xs+a`EIeHo}>@7hG*=+s4~AC)4Xiacd-*mRSb?bt5^2AD_cLX}?h
zBVrC-4cR}!>n`uB@8-;BXgC^G<a=`11U)i-7I~ZA+N#@2GGi<x9cg+Zzgc5Z^_Jhq
z)J0v;o(Yci52;6_D@l#)>uf~yd2sW-8fU}1!i<0uZQ7;?B6fqw{X$JJHNYDk%|nii
zxLR*g2CJU0;t_l`S|hl#m(nsA@CIHtp$YNn)tRdL>AN;JzFquCQ+@JQnlO77J~KGF
zlocE?D5#&cMtWFh;)8u0eEB;#%2)kg!sqSc?_8!Snl;F$R<w{8<!fO*=BmyGft?I?
zvzs?N#h;1|GjX_5ft-f?V;Z(nDJ_?;V!UUB+q8GdeyYXrUzPaO)}K0Sa`6L)4@2=k
z8~bL_&QI8X(XBO30fy*VUEpJCC~JdNs*r5@{Q@SRFd+oz^HcCQ+6Wnp=*+wx0_O>U
zO8J~cYkum0+gJ2Zfk8PjH4})cKa`DkMQZ#JGJ7jr?&-hp?uE98cmw5oVyG%avt&(s
zF(+2|u3AU}@KN{b7;f<-t6lTvB6hRh_(*R!8min6j2G{haJO)zyY~UCvVrMq&MP)P
z$3H4I+8|Q>fXEYS2!1tCdTxi4AZ>nU0GgrLj#F}iQs00>Hcd_3o0{H^d%l<?<Ll1y
z5b=~}h0=YDn)rU|88$=sq#gPim5|(MJCjXl^(oi$99SOX+fbeV%CZG?Q=(4&y*E2@
zLv;F!^RGVh0TB7`ZM&K&7itd<fN2LSC_^YCH7>{#EAVoJ-=P=*8zA<Vu5t6gCOT&s
zFYdutK21$a>quR}5{5G0DOj8`S78q0SWSMIt;JUmM3&03T-i3-QUK976Mxs$H9^VY
zvlrywMTJ@Xx*jq(O>3Ev$^|+@2SZ9jx-=8oBdUAmQ0nWZH_>ipt2-0|2$7>^8{98X
z^nv1s_i1p4xe{LSdlAd;wr&8VPA3I7$jkM2rI0F6S1Tw)QLj}<;q_<845ZV=>dVM>
zDbS*h@b#zP*e$1%+4!zDKM+5DOk@O@KW%S^_uQA)Zvcd$eADyv?tOnE3)5GA2Towv
zNSMnHWJ~z^W-DehOgg?lawXDx9%Y`HJ2>ncVFoR#DS8s`7PwM8y@1^srXqX%on+?G
zQ&{A8EI$V=SZFOFrKbF~gnVA7zm<*3ubI?}nE*W?l0o<rX2r)rJqT5zqe5LNkHy&f
zu3+x~xjY?VbN-h$9-+17Z(%8~c$2XF>{=tOxi(H%aNb<cP%8eYd5*Kv{)rgbBAAiY
zxE9#QIvs%n<%=i?v&V`EagKBP&KMU|<kBhK)B#uW7Tq0gcmtnP)VIBdLdM325{%F@
zz2Yeqs}eoml*m~Q;M)RAR)<)0>+ou>GTw@XC-^;sn|2?9*+}64;5rtYv3e{OCziti
z!xA`7xbPcZwjh1gjdt6(Cn<Clj@<M}F1Nd~U&{~=|Ec=W@r9ei<1MJMUP!$?+CEOo
zl;^Ymjf~VVd;@@|l!bht42B)Z5jcjpwQwQI1MJhyHoPaEA^;93AU!fK5+T>Lb8zVS
zN((n3k3y#uEVwu;^g(rpcvFE%Z}SV&Wa4Ho2o@wU#G`%BsJ}1f{&0A+;%i8PO|;Yr
z&?=qj)&~G{C>=mwTS>r(j6Mr#I^mLk{R&ne7|N`5p`*yIR9@m48>4wpMD0jcx)9wg
zU)@wY;YY+ET19`>B|lPs-&iLmvOim`0@n)+3zv&_(6m{^wdjv!!k5)-$vPMHTtcrx
zPJM)`15TLeFzy&VJrIu@XTo?NXOJv6=MAZI5^t*N%qnW=h*;fTV;ZF7AvY<PB%3j0
ztkU)~8;YBEyfi2MPESt5Ry2tEn+!U*d&8MOf8eNoV8%|Rs|6+IA(^o3*_^n6FtR|1
zu81&g52w)UkVBU{59vQLm#xXb*f#r;X13*f+^6}fV|uZp8mEUvPZYvA5Hl2$REn=o
zp6Yp1rlM>ec)?g*AgNI11!5b_y12oJu0x5HwXN6k41pw`Y~Ud1-;-AB*=O4G<gdOm
zs0hQjd$V+FPNXHq_wqAel&rc>v?t9-pD?F8;5Et8{4lN5Q~5~F;m}QP1=04fDHN!a
zkxbX>=qeT;E4L2bI#bOk$CsQ%<I3mpRbOSx{tFA>7!tAar*>p81uj!@(!Zh-ry0_&
zemuBoT+piyOl6R(S!;L#0Gr}zW0Ovw_EYS(S&`jYD(BZ-sTDP2CNw{wq#d>;lFHNN
zV@ZmcCj;WsT`D!qai9Zz{|M!nk-47Wwf*H0{aVR_zj6Hel&}W&51cqDF0u?rZDqxd
zjVhZ;(n0>6DR}C!Q!aLCgsvI$QpI4llma5HUPfUD9uXgTB!t`#_qAaiRltA-&$8sK
z(6um(ErHJcrd9ztoG%ktOeOS;9UC=rPqO$!3LLxp8X|SVOy)P^(1LQvU`w;pk{=o{
zV#}wYH%;MASAt&}>}Y(j_H@gjF;~LvC&f{8WWj<gEuQG@1IvfAmav-BIJF)|G7f8H
z?(6(EH|~7vBCh}YNf!Y`GZh|$`DbsDsZ$ZzfDc-0l)SC;mJZr3Dj0g7#}-W?M6g>s
z4@w08r&g#s@AJIer!JW;(~9DNb?4T~UC(#Og02&;lHtg*RThdW$c9j|a7h8Bs4Jl7
zcfx4VM#gu^wNUnK2J~pHw&L{|Y{)XZ#4axS9GPmSEvz*~!<rMV)CBQL6O427(%hY1
z+I&94I-$1TYZ>@Kbyve#kLdX4OJSJQ?EvUc1M80Hii14cm5jtC9MvE327Fe0+FC4C
z5pf3uO-quEU_AVWG0Xgi%sCGR)nfdk<L2lq!e{Z$Hn%LrOm@tCdv7V<0!C9`ttxWL
zPVVYW_R9sC%xKRSsa-lfr9WZFHUy)O0&}Fp8bch29SPLLlp)WJSy71bb0|SNp<>_v
zMyO->URQ-`30f$7Rn|ooS(pbbu1(F?hE~DE1mr9pDBu!A_zhpy6;qZNZp0~?fMV=H
z6!BipInTrtjJ+pfH)Hi~1RQI)$VbTFEzP|$D&@pa`%pSIt^G<$lA8dQ1RhekC&}CT
zg&S63fwEEQ^kktxpx+Fv>oV2I(MdBSjSFOds5V6GzUI;dV?F&2gCGLxV>G9(KLaX{
z1DZ~nM-rGSL`QFwC!Ql3!ao6wcAaQByS<cnmcbG?o-D@)q*bMawju5BrOwS-&`poG
zM(e(>56n9+ANo$x1teY+LZT~D;2udZT^i97zbU}vkT{eTQt)9^_}6NR>c*WN1^A<V
zTY&*r)V~IR0mWYffJr<1!6(Oms&*B^=mw0H&NnzBq5IJQRkXB)PJ4*118wWw<4i}x
z9WXd=S`(qI3}P3Som@%us5CK{Z6yr^y}Z^>ApZnqU6hfh-w7m+!)JwNZ{`1QesODz
zB5lCgY7QU1VTlCOjyt87<Znxf_oas;8CRYm4Lg*s?|K2v?-?fw4P|J%JoUDg1ZD1l
zT;rARXjm^fb4Xc_hTD}UjSUuFVj3Kb+xmy+b4BEwJ^#unnUjGMCKs}_smlZZDX)G5
zz!LxEFI%MtB*T4*(n#C!IClY{8@-aK#NbSIha~&WC**w<b>DCk0An8DWLb)`SF3Vn
zy|kToqu5!wjE#@BS_Lz;=j1*$|F31WbzSv*Eh8jzcLu!=bn`Mq;cRv-N(XfvC}-<x
zpZwM+U0-=`RkDx!!~zd6s)5B*#v6ILR81-)c)ELL^{z$NKQi=cySxcUihW+hfOCY=
zxnJp=th3}B{xQRDIHcfxDnS<Vu~Hh*7^e-mu!Bla!Y*=NQt$~rZLT79Q8hBQi6MYH
ziWly`BGtwi)tUz4=AvTUJh<j<h<)+()IYB_-#(Z&haH@gp_aUR?Fj(6jjN`rO5F}8
zP%k_HK}|?@Q1PMu>Ja1i%T$fOe+D|;7)G9VrD^N>!%U1NjOgJ$Sm32Y#GvT6&q7(K
zj;VaOnZ<F<yI+S`d4#x^Pjsey*=D0Ws@>VSfR0qhf!ScegV{GbvLg%@8)G&C7Wo!g
zxdoW&QLh!(_9cL+>-YlnSm_00CBAgvdG-21)5;CM_a(lZ@a91&H21gmkV)#Gr0<h)
zb0Gb!3la%gbXNW%D@OPVU}YJSQv?CYBhh<gzx(6f|KWFkHH|s+P9vCLP}{R40S7Vc
zh$O|N@gYg|NnE{{E43<!szaFH?r}~*E_{NP;cX9#a!7#&9X}=XI_lEW)n$2nZmghC
z&S!G`d?`U3SP@jcU0vD7*CE?wu-V3+C&(5-Xc1`f!_qcIQpPu6&K*iUL=5LF@wJ*I
zQW;Sd2}ULLjAJI_`!ww0=_7+4s<o7O)N&I~k~VE$j$J7b2wdzl4kJ-}%K9EozE0?c
z;9OlnKsmk~!6qp|2b&7zG7Ew?IPK#0htFNVG%rt7BnrVFiS@T@5YP3XmL5Npn(f_J
zChEh_$qzr;*6RfcnOnyiek(ySS}(YG(W?JtA*@~IuShNR^FGt~>4Ag9%ISv`HM}Fr
zBNE0EBR)6UN?L;47d$v7k~JWY>k&w?&IVE=&S_K}ZJJQD{Gor8R{r+Dv8H-jr9C?3
zx)q;t8OD!|3V`o?lQ>TEkv1Q5)l8ZoE#)wkz@9H@M7sQ5K5^t7lmAU2UNH2S;Xo`u
zz+fJzR?|Ssj8Wj%^SL+RO%HqxY?+az+i23B)YuMaZ(+cBnRYw~8smZs^3KHZmW)27
zs$+8!Eo`w_EQ9%+d&+e+$gjG#yDyv}t|N?xDoQ;qjtTzLpMT{RazhaSO)-^^{;68s
zqFfd#lp(#ab8w@NpQ!f_k!;yKyVYzk&jEKWzR?DwmK<Qi>;+#Hzs;<O;44x?)yvpW
z`XEy18XHDHK_kk+b~JpJ6ok{-=oB!-m_-)duNa`5$7amU3l`AuoK&QD$``#^@vZ^G
z^gG*z>Iv^t0pynDUDMtK!v-^<$on`UvN3UI@9emd=cS&DW~$7?qpNM^*%pjrccA$|
zn@0_`JuhWP10doeUz)ERh&w}=%BRV2k1goS&$Gy#Eno?_KwpL}3_~dOBzCXJD0vCU
z7<y5xG{q}5W$~v?>NRi*Y`+~v9yuz`cKM5fo9mpOe*n$o8Xlue0P0!tPK|z3&D?F-
zHZ8phe<`dD8@~d!z67hvg5^vGYC;YGPeCj4!K0};{#O%0?vrwNp@`9y4XysIcr@+?
z@ZU}V$SUuOMNjsik`t~f3<Cv-44WIPpRX5cQ*8zf!RpRO=;oaSkxZRKya_4VMQ}w^
z>%>RT&G=8J%LpF&+6{Rjav%9GNTUbO8D^(#WTYv%NL5*TM#7Q$1q*J7Xt*~`+x`md
zs{Bwm)iAE|!p>C(!OG_c<}L!trH!W#)I8r|I~FgCus^u(e*~zS1j(p3UD_@x9PK|(
z_B71(8azWTFd{i9js)304+Qo#?bCtu*HB2G#y8TEi~QR4{#-O}{6idWT-oN)V~&tH
zExslyWm2v?#}>sKI5wUk-oMzwCh(^E1bQDiis1rQ+I-)f)3rid(y2hiir#h39d2QR
z?mjlY{#uT<-(8dtQK6_>3gY~(p|?AK&qh)Ncy%Q)!j~KA<d<(*Jp)LZQsS*d7N$YR
zkk~%C+YYCPGv7&wQUrxLUED<9oys;&(95wzg)QdBH_lJUYd{9k5}+Qdqlm#qNLNQT
zuD~>p(6H_f1t2c?T)>te2qP&q^=-w5wR~R7A2Y*d;&2mD1YtA)?^frfp3RvCN1W5n
z&BooEYNZNK1;4-UMD&C!tuT^a{K|`F)R(pGZPPh0C$=)R_7wVsB%xJWVKO4cr$2mp
zPynn0hPVrCX~>hHH$N_sYB+UNsWef9rlbRwsbR`G!5F|D``r=6CE&Q-0hpU&_J`Sn
zO&1s#6T@#b&2FXr82}%4vVb>rOtYdB6A-_13N#R?BAjU#B>Z6VPU>G^hQ_-9ZYS;Z
z3^eWdj1Wy=D81wrf4^WBUL)7AGe#<eBRd=}Q-S`qB@5RUP6p{jzDtZxbRCRC7fn(&
z0+V*`5{>$KZ+c%*(MueRLT%67sIYSuNJ}xyH-L!%SjNJs=`IRaMCYJ0rOV|&D?3V{
z=K^qHXaTIeWkt^1?GY_TU;1ijEW{%|`!2Ux-o)ABm-Usg@ux41+JOF%V_z9<I3_=M
zAKb27tbSznVzEUsi>^R{s+y~seMeBQUr4*6xsUqBR@}_=a36J<1!!%8^9z3EJZwM#
z&b3_9$n6U6bduvB8ne_MItg&9sb(z5l}rkGV>jVN8(KDuRMwcsf<Ui@#a#ne0j%k=
zVC$KT@1Tp4qs(2u)g{oDfP%AaKv?P?#}_Z^-IV&%pgd|OupUUYM-0&Cx(d9w$+4A^
zyBtg0%-l6-VchusYw`%Cyef~eRw2QaO)F5w!0|zV6YJ7>iHXr-s5##<lhIrach{dF
zx{3zIRgV+ovCOAdBmW+ZHJeXuyIol@^iTeeuWi_h3QU~MbGlD^7DQf<1R60yy|J+Q
zGjh%SxDK@9D};&iH(-pQe>kngZsvfPr0ao17S)BFy|q(YZ5O{>?(cPnh~ws-mz*EU
ztfZ0t9A7fwJfLKu?RY=-{SVDnT$&<iX0)dx73a+8pkgPXY#j|fAHGzf<dg*2EHj3P
z#a+|H-v6rpYl&KKw<$zku#2h_^dXCd{Hr~O2F>*GCtNqVIkw#pf;{P!Z+*<AC4Z`W
z>TgX3F4$OooUT3P!+_mb7L>j)uipO;Zcqj1Wsi=`9JgTsv(hTErs-`sgpYO~0Lnh`
zR~M=twrP;3@nMz`ZHlC`HZwHwm^1K4@=Ub0J#(iR4!A{Aot%yjNH*B2mBG|Xd@^f@
zK)#qZFKu2wL<bzYGR0bDe0z&Qrm7UE<+oyV#w%>xWMfSqdrSK^klkk7>}96@8z8VI
zc;p1SOPpxsXp_oFpBq*=-{7Bjx|qsYr0_+HWQ4({rTVpe9#~^2$0{nn89rE+GS=H6
zsNmO*o1ll&0vv3dmsf;93ou<Q=nt5D!%S%N1)3I`G^ukBMV`{PW_5y5hGSFHx;hS+
z9?~1orl~mc(R)9DdtBZ<Re_6J#0(Y^nt%V$-gBf!bJQ*li7Epd@kY?4QCJ0c!=a4`
z=(B#Y>PE?ffYf@yC`SflE<HM{LtgK83H1*5N;Zztpmjud4dJLR(MQ>1X>(Ghq9)YT
zu&7Ml!xRYF0Vg6?>&}dAY>!ar@tK{5?4!e*J;CdIEi<~%BS*b5vYV_yBIQzbQDv$!
zzdMA7RwoEkruU4Q=f?Aegpm|HgeJ-XLlC}#PB79Yt=eLRijaA_veZA`&^Jr}W}OC+
zmVRX&bI?;`o7Y|I=mowDX}J6qb;lq=R_lEAXxsWTMPHpvn)EZ~IuKq6nMSjGQWErT
z=9u0NZEq8sPat&!)z?fbHvVD_K1##iXblXa5vA*^r*6B0qFyNjxK%qoJSg(bP;ys(
znk=k3`1C*{$nZlfS^V8yeqina4C>TU-DQT-N{+V!E{<{KPp{CFE`8w})?d7IHQ88u
zNt+qkGNppg`tx7S+|wo2`NS4{l~uRP%jbWgcm{d==NZyJeiIY1p)vNd`$RJEUKlwy
zch6Qp2OnTbaneccvFkJbQLW<`m1f4=OR8AQMD)pCFY|Kl_Vaor{!$g$0W`>pGN!H$
zY?Q&|cL8Y!3lixC(gz872uC3MmlWNvJIbm(*y_BL_QhZQ`p`tNevuZ=8;v8>kY>bh
z56pD&tC5<tuL<U;j;Le?#{x5)9~0PIL&D6MLVqk?6#Vqy1X80U?ci}CWn)lZ6UVKu
zgTWj-9K_zCe5;Nkr>^2%8>^v5qA(Zg#QS-qAzk))SDd(L#?aXs+GL#|yI&8Pq(KV2
z=kDwHyGz{eluttPg}z-dzd_jh{L?8~LbbqX8HP>eysDT<Yp%6qy(B2Ysg+Ast>4Bs
z`!chVpu5ZZ>O~W3$Wm{v&blABK=Z_|l8Nr)>Yw-3KYU19kcxrD8gOcsRkQmrV@H6+
zHVz<+4WFw$9yj!bKihzC77<vmM}%8<&)3wo?L7`KENUKRcNIe>DxEg?eV^mIR95~^
zBlH&iw$Zt_vZlKS2mCVH)o~m17D`f=2hewyn@LA{WOO*j&r~C(Wcgj0KY9x`NHrKI
zJ$*ElwRq{3>tslp5sL{iHZHAzdo5rh+OQ6|AVI28j)DyKj?`&Uc}eh|81C;|j-ZH2
zG3LW^pLzsV0kKc`-TVTi1aFN}6Wxd`Hp-v<6B)=5O9heave6q_f??;y1Qi<gkXc0q
z_zM{Ru+%|u>=^50%%cliBMHC+YX9lfVg1IM(GJYM+s<qIv2%Q4srgGCyCcHarI5ld
zerCo5gWJBW*S{~0UDzyt5<!k1S=M4VClB3PN^Xvvddjh4fI94BpfXc`Jov)uSs7Ot
zFk7j?qD7}^%0w=0%h;BJT1FuxCNRq2o(2Ftd_Os?x?}~Y_RUN5=}9Ezar*ZlJsL1*
z7Hi22Su9d{=sb3kFd53wWcCY$r<gIg6xY758IA#SEhL2^$|Pe!sIo5R;7|<Sj}C*+
z2GPoFdLkTtc3<7TgJQ_Wk4&X1N%0lJw?EobZpY0Oj-&%5=Voa-+^5QbOPxyoWOYh8
z=1S~3&n3BS4x!hY7wqMzfoQm<b2HspR*k`a^aWu8TOMJmLNZ5vz~`dBjBJul9d?EC
z{FLyOd|={vbfuy<ClIK%a55#hYd==l`+S@-&h25sN81cG7FBP>eOanJJ>BzVGycU6
z7<nBBM;6#$ZXRa1pLaB_ZO@3v)F2Z^;MJD9_^2;_k;MIzw&&YVwd&sn)|{Uw_hZ1Q
zZaszXB84s^MJz)eL^gas>9{1T-fIcQ%^PJW*|fUIx<|OLHYx#J#Rpy+Pb;tg7&Qz~
zjLLt;@YLq>V1;gAG-WK)pz!sJt}o$Z5$d+fIphgf?T(<GL)KJ21JV-=&D2v67Ce*E
zVMwDV$OlS^Pm)-J5hlXk`lBzrw;Mb;^SGKHIJJVw7bDF(Kb_(>b!vh{cxVVCGiQ;f
zlL(fhGJS(yC}%c5brsLyBJY<Q>K1h7I_H-G5A0A7A8GAq$Z7-QzrQ)fW_8WDKUC}U
zW4#=IZtTVaZ`3q5baw2{`8v^B(}+b@&;(!b+UTv@r!RgL3~0wJi&s1^TI9_As`OCV
zP3dT2CXT{)`a+{-sAC(uC*O)DQLrT&uKf~hDOxhhK#if>Px(YA{{_`9Uc)-uJKpSA
zry&hy6Mx+4dmnPrgfpGVk03kFTJeA(&hy8P4Y)KrXy8*=4JuM)Krf4X0g7h;7MU4a
zXsH`sd&ufj`e{`IMIM~0J#bxbwe|%kqobVyUWKII=dM@NZDu}hbFLy@7?W@LMEmc~
z$~LYt)jgf^0$2}RpBIC^^+6W}^e01kpiH?%lEaunKH|HqiJGa*!Kny&O?FVoiy_FO
z;XZ>I3ulCSsLW5AYe7G#`mTA_wY^7KyQQ3PvX~x~*og0-ChsmhV`#^V!Gbg8kBHc)
z%+0TzQ`e#|4nB2DU}|E4RiP`EF5tz~H6y$@yhLU%0`&JyHY)t(($Ry5EhVp|wV+zv
zlgy9`$^`45EF4sMF{!oF)NDBB5yS3o;v%cpj{~M(5HNp9;%b!uJ^bPQ*Q#g9F@Qd{
z92M8bT|i(y>f+5eB3!@Y?<BG?Th&qT$gJED_K1^>Bon{nweZbK{lLw6HXHR3{m9sM
z%+k&CqdHLcxT#?Xu8xglZ&7J|D}=ni(46gSeb<>{Dl;0b7ZMMjxhc8JfbwVW;%d9N
zskHXCkgC#;wp=S$up5%ILKRm5L|P#GUxjc>B{2@ORHMSF&^IE+dFj=RHm7eNi2Ztj
z$*Ra`>3A;&Dx1Z{@clm7Q%+Yz+Z4Ll$a8;nZg#?en?ltEN1!b*mmZ4%hkS2jL<vO&
zwl$;B5_*_HFW-YdUeXT{3Mo_9#a(|V7tJX+s=+MF5c&i$4v$B0{H4C(tnni7Y)IYW
zlATwueIhgOHz;p_n*0#>l)w6vx8l_}8TEpqHD)4`DDe>J<+#2(0M;*SVe9UVqY?H%
zH|ix(wG-ZB7CwM<cFjB??k0^w`G=s=TN%iL&G5gZ0YQ8uD|f!^S&AAl_Q@U!=vYQ}
zc2u)qariXRbggVmdthW9nFF+k4F)3s#Cp#xXn!<%gARJWdTRJEO9iT#xp<sWs#g^y
zd`pTsiMDa8PmMXP8PmE{J|(NQSG$Nf;sH_~@^};0N3ETd8;@DmD*Y{LZ$<~4&a%)C
zXw0%@DT5fg5qPxLP%620?n8DOopBn~0qNqAIb+3ak?m#gfb&J>_ukYI8d{o!Vy9dM
z<On){oONwaexi%fn)8Z}90%Ickt{{wAc`HopSK$Ju@0M&$#H3ZO%dA2(EeR)ZgMaj
z2%oJxXRJLUm+yRU0s0E0=d6b?oG>27SvR<-1>w2gRr1M)joiy$ZD##;SF~kY?w<L6
zWb0vax3j%%KYj<}VuDrB*qdK!o(Tbu)OX4nL$mxN(5L;*Yu-{9p0|k`77XU9P~(0S
z!mLyEDoyS1lu|0lKk{e9{Ia<$Wb?;l7|h$KkUqE`Zzk|`Uwv)qEOIOc`ZDfSdaBsU
zXeL?JIrP@;0=+5@bS>a7y&EGY=F?mm8yPwxvvqIM>*ekaZ0-|egXNH0iLWL%eZ;P4
zrviEj_JGZeigHwPvryZUSQKis3*dJV^W{URL!LD13Cl-Q*0F=$xoaN>>3Upu!5m%}
ztvL;F{Q{CPa#Bg}KwluB-M{7fYTa|wp6#<Wu4D8Imqd0m;e^T7w!=^iEzpa!Ju_Zc
zeal49)$2%(@JDL`3hFo!`-{gearG&&%!AyhMW(S(Cb+MhRza{4V+F=#c*0tY;TsxI
zX{04{N_ON@jWp}zpto#|bTV>D$>^!g8UPmVjo;ZjDCB)LnLVne&ee6Is87Z(L<Rn;
zuDJR)lhLs0qkzlavJNHpq`D?G3qF@d8B8rb`Ewj<s>EVaosdx5&<E8<h+~Dou6w?p
z<_~IDYPU^`2tt*~fZ$;NuRfhu+ydJ<b?RA@+L*a*H_nw@2D^WZP>5$AD-4Y9PMmE=
zjeW{4TpQAoc6fJBz+L*jKb2xF`?l7aU7dgAF``elw(XyM^JH)*+pb@Mq#7GN5GO+R
z_+(@5c(>m_4q(5^$u>FZ9Z(k`gYm)UZ*gTyby?FOk6cbx#y@i9hzu62;J;9>=Qzjb
zYG*6y0`^{1?ZLDT&-Re+h~R=Ca?FC1tU0{zaoG>cRWWImb<9-pQFLwWT9&Jtpl3UM
z*bf}#8#0@fGQ>ZiYelt+e~I6&x1Fkr+Bmkjd%iIO%OCR23sE02Mzaj3B^Hn-e5T!w
z{K*kdA;}~~NIVTG;_H=sGh6IJ&9mKp)L7`$oS-Xe|D)lw%iAq#sbPu-%`L&rB$pX>
zctUM65O7+2UA?8z5l9O_QKY`DbgSZh*$jA41fjuG*KIT2pSdkiGi8TM?rUpyAB6Y6
zb)B}J*4QdQRFdgJea{!I4^qN7t@_R@g=^Y2FX4#x=Z=VqBc(%<c22+oe^W5nE}LI5
z*50eemJy+9AGtKyt~7eRF-7!dV3xbNbES4o!H6fFpTWu<*-0bom09xJ-&*dc^o{F!
z)oB0aSDc>7pYq*X$3c75d08WeJ6M5G#+ZHAMFA0BFG%V_8Zy2=xArItDfkFfk~b#7
zf;B7;ViSD0f<0M+a|7fj2HpBvr<)oBIFG6I9;h&&NsCe_xmOYMn+RDYg}!h*U#Fag
z4L!}NM2@q+DD4lP1qbMP3qEky1jImb+3IcG^&<qjnh>uyisxxj^8`)?)x%e|L;dFt
z5xiVcw3f3wq5^@=G*Mlx%Trvr18oiU4AJ}c5izw4(COBascEVf(A`Cmn`>LL1N9Pq
z$RNJ^fD!j~=i1uZHKDdvLHBmGTWd1rTlwI2AZK**8{tf<A2y><L@2a9@$6yWX=`R(
zi{zeoAmCfKpSnfIGt*`iIw`rmoxAOE$D$9anig_zzq|AUIkx(0ENr^)KC1Z?xDp&G
zVQMZHyS^MjSV-@uk0PHlP^**sfQuGcxFE`HSWqfk)tk`uBZyqmc;(yOrIKszbuwO;
ztMcU$cdxm%e{;J~Ltc^rBH9J}k-`3-s(&HCA!dZm&Js{vbt-^+hq85&3(7Se4r{Y{
zC|TP!|J4KaYK&VavrQz}Qox;`ybP2+@K*C!eB@SKP|X!OHUcY=Q|wenxjXBEj;BTY
zN5NFFk>CpUXCb}^s5?~)v+|1ZW?Oz>Y#sP67l?*T(@||xCUI_dNy8AZfw;)=Qax9<
z)r$70qIv2G%6isn$|3npUd`jCQibM39M$Z-h>!F;qnaWqj+e>D7R)OEOaDJIh_e;`
zmk;r_0QdVJP_56HY4rU{`zx`Bd8@#2J8U*q0aFXXcfY2nr@^!*=<%B^t;@W$qhin2
zNTTqRX8;F8%xOW?k+HGQOSkh1l>ROk0Bp#4v^<}Hkd1DIafitN1#8ugJM{ibj{iW5
zDyLSd(u3TqHXadOhyb0MrN4^m`P+^{D6;jZY(SB!3G91;?90dvSXl_g!yu3=Ddm34
zDE$lEsBg?Ja(aOs24faQ-c>SyC_bSZmJ>`(SRBf#lE2@>-9e=6Vi-UQK;oN?odUIi
zvk<p^-%o$-KKg^Xr_}`Zg7xjR)AMaBbvi-!7x-K?8Xp+sW!1+3Gmfl#<kXi;v{!Jw
z(G+<h5!YTn;wdiseB8+1hS9b@IO`+u^uh`w`oJ=YB{mvFJ-LVt_;Rq?DvBbCM!kq-
z$^J&_gEw?+^EQVVnj)O!%c5a23%K~JLq>BRG{_%A#xEomsJ0&LOd3(82tRop8-HRs
z18)^?k=`qD;=prtRdDX?;N55n#(`OX=fd7?hc}ysG{BN3)0Zo@HIQSv86P?|y&o#4
zuA8XuX1HqKuK01eNYm!;CqR}x5tI#-a8ze-?($GJ-rhSOc&a+dV8exK3vH<qXj9zp
zPDBwm@jY&n%Px9zV*g$kiHhS0U4*~#mmLfuN^@{~1~w!eG}Twk-Trr54Y09s@;~zO
z7%&OL@86X)(h8rLpv(UY0wGsE5cT!Xf<YYQ6=0ZcK-zfpOnp)HZ`e%y))~N;TT2Rd
ziKUHHJKG8@OaGjHX?m5Z*cAJc`JQIET7mORO1&tSi_vlKPv#|R<U4df5O2Wrn-&xx
zEAH#xzkkLcN(wtT^87Kc9@88X)r)&)-OomHcK%%CxsevwS(-Gwz0kRkxxJ1QSgx1r
zl3%YETvw3WAFaXE1~zNp1Y;)*bODYgMo0BD4EzY+JUQVq9#8t_M@jx~a-`W!@+(aE
z>PbuTYY2Z1{tWp|G3KHV1^G?l8%i<qD<0A#o+Q69);nEKeuZgjgtCxdA0n@={Jokm
zR@;Ix#B*K**}3F*w%CR{ZSHuH%YyMlVt2UT$i(LDhse!|;0JZwr-a`{Z|b%_6>p$N
zK7hvJNJlOapVJSt2C0U)uB`M_&d+7h3|<g#h(@Qu_!fR_!XCb=*<<vIw!URt$5+b@
zd$xHSTJ}(QX<ON|2{TS&&@9y*3=82evxgk@0v!WUt=6Ho8w2oG!5VQw0$1$G#Ev2X
zUu#GV?rTr}Q<)K@D^Mdwh?3<#Bb;3&=!=2|;k)2lkAjxe-5m$RU?IMvLjATSh(*6k
z#h9b3hT%GM;N5e}glh<%zwfrR6SDkk!GmCh^lblpDWsu{`fmGxOpX5!vix76?f;*P
zqyZYiqMdY0B?-EJ7dc8DMA<=L%pb3Wd-K<n&4=QlHSQCzg$J3*q_#pD3eCB9$jlU*
z9Po$pBXKsG(4j0hjc{Hz#Q9_M(nF+zaLmuY&-UNspAZ#KFX#-~X~FLj3_mrRPI?1p
z5!}5G`J=}m5*W3vwHC*`C!;{9l+>Ato%NoC&%k#zAb)5qk;I3rgqt|*cz=RN#l7nh
zgYTl@gP4(JS6z3N@zx;veRWcVm#}FNL?R%;EZe=>thBkHU<E>Wcwif&bMmhs#P?6g
z%3RqA+47&VWh2|RFQvZUP<ABfI%L7~Hp^Sz%ztBq<==?ab$_-;!`Pg=4;}nY-<Pre
zTsxgx_4>~<O<Y*{{rNzVx}#4t%>2qo&z0cS=ZZ_(?(Shtj+&&1T{Fx9yk9Y}<#|wJ
z0x6F6(CO8IcyQUsvf6MmY3_YdGkbm+f7$T1eLv&t8(nk-{D`nl;eo7Nd>#K96>R<+
zpFUUl#>7IAw!0;yk4{gROPgPNNZft4Fe8Ihly33R{xNw~c~9=q!!#bPHPAr4sMRRw
z-lY7R)v2#$7&>S5wO{W$xUij5fhyog)Ofs&byKn2eMcYlCJQ4wI=in@2bxXE5Tc-Z
zQMsn&u|%8WbBAQQ^amk2>!|mV4c?~qH;O5jKbl}ml|qEYu(MkAQWhP+@UjNn)#-}w
zJ64fiea@8v!#7)la(~Eau4XGIE+&vV{dN_Q+|tDX`~;Pj<jKY;>qm#C7`le?KhwQz
zM@mFm?Ddk=x$iLz;pOo>Bc<xQNJ#hMp?UJps>eQVKYp+Aw*QK#{JuXl|K<I$z)76R
z;{Nr!X1`P3Ijo}%?E4fY`dQYK7)_dxp+-Tlhg`Ck4yCuZb)Wb@`mu$5HfPT+mcL)k
zPnb|n%(8H@9~y2zdWQ7YIib|dbH8jbizplAG$EGda$}8H0_9!X+{xrf%xaFbP161|
zv#B!riLq-<nd5ih!sf?~F3$Q2<6!ep)Pwy-x*y@bc27jWS{YfKBeb<o*OPasdv%rv
z=9JRCm7+4*`@2t9pHcTXQ!ksLH0YjKlh8Cz`1;smc`>tTALAx)>xRLijPBNWl~HLr
z6H!F@qaSO}TuK)H&>k+N*!Yic>Cz+;0QF}|26GLPT#9mQHE_bL(L(n+P?Jf#BjQDL
z1(Kj+RVCbrL93z6EsfddSH!qZuiz+UzC9y+*H|IY_0H(pk(PHCHcQ}vp2eLyBiz2J
zg#xBS87*I#d3;Zy@%~XmGQyL7`ca@+Io%=VNM`n@?ynT~F!E_IDL;^JNY3i(3Pd(q
zb|ockeq(e_0c-Ksvk*9HHZTfsaxL!6-wI@VFS7ov+jOmWT@9HjB_8M$x%rC*mi)0^
z>}Rvw!i>shZI%5_gEE>_{g^|{p?gN#bF{#bmGNf3YWTMZUwLaSo}`P53_QjH2;uiq
zYwNR*hd*v7^~|gj%W$L!m7Cf@I@-sji<a;He1Cj+-~Li3`Qe%G_mUre`aqle!16}q
z0Sv}i3fW{CFxu>?@6nu@@$A!u7Lw-7*8Aa2Qn}+-TfAZ%yO(WM_M;tHS?GFI!ncrX
z<9@q=SkapK@hES8!aJ-e%fhDHr#MA$&u*2VqiW+I((@o`OzlC%;1Pb$bNA7?d`Uz^
zZt#5k8npNyF8p%YSTo#<A2F?^PfXx}Jrrzkebzp^1sC-0)74#u&EO1ZEVJtihbD;4
zuHn8`KaN5qW9T(3?4A@Y1cbYQ*(qz={`wj8h}Q?*6@_bMIZw6=X#=xvcDpVs*MVPE
z{I-wXXu5RJjW7K#wjUgT-dS{X-Riwa-fYN!g))D<*!8LF_l@oE8HNa?XG*W*{BD^Z
z*UQ(Xt`f(+)PcvCWY@J7%4AGC-FOyBld6cGn<Y@1Agz`!hVUrJI|z{Mr`d-C^_dj&
zWIG?8#pzS9os<3d?la4J88t^?@TQ9J#lK{)1ruNwQ9*VpK|><WbACV>E=cPI8jnj8
z(E5Sh3Ic?t+*+Fc_UlKvhh<BG$fL@m-5mI!l(`_A)u|;mZz=J76(bPWg~SAXH4gWa
zpnyvRdgPYYW^v3qB!88#i3}1X__sP)1^u~Wv^yM6`U0O+-^^)iK?X?^XwYNnq#wvm
zb@w~tN26G74Eo%3JWp+L;z7N9e`5n5p$zRb+9L6t%pSGk&f#_$Stjv3Frt!F?zQuq
zqCb5t+7IR58P5z+FSfU%W1e(3*2~X6qHjIX`-D%K12RV+iLhnLy^rlc9=-cBT^{;t
zt@*vfzQUiYSkYFl`ny%q<7m<joQ0`qle?A=A)JcbKm0j6`PF{q@U8ub+0BJ-KrjlY
zy0N7i27NV|7hxySz*tew_84!~W&a8iEf}?$EDl_zt;|CPVU_mR2(xJ%E>yXa%XAHw
z4i9(6Xak+J*0O6CErxf$2?AcLIOHz3$CZ^TX3b^x=V3b+6cGI>YlKe5iP^>dwdTw0
z6_c*&#|=a(>j782f2FeU`$uf4wn1N0a`8e4*lj)@-Q!CSn-ceY@BY~OMZb?|G@j|B
zUTm_evS@DzeZ;>0y?c_SNK)p^^rp(bAahYWV-zlG`>k7};Wj8BNIfY27{0je&s^lJ
z>*=st?9lqsYlh{p>+j*~e;96;lI50rb>$ml9Y3Q9Rr12`jV=w;RBKCs@7KxvJ+|NZ
zkB@r74}(7>2ai0jOw+fpOn@Jc6$k$@!O$t>`V<5_`FchA;yR<qZy?!)^@4IMy!WhL
z2fX*O2zt_<zr7!g-6A~d4L`T|Ey%87x&5qjify=;B%<sx0#>ZEDcU|9Z%4ys#I)Eh
z_`MUCr6hzE4KvkV8>T)?biobQ`QE9U?m5oge|@!zBI?b%HANJS^q=YNs4X=HVKmQ3
zF;FnL8>O8EAFR^5TamNN`)1UC1#L@ElSKcoX>@LT&!o}h{raE{ywtz9AL^&;kVRx`
zJ)?FdFF`aT{)l7$j6k_U`4j(^^gXr>%zI~a`)iGvi}T!w-0>Why2B4+)WUk<I>MGs
z5AX2!V`}s3y-g;FG#vj?w|m7Cb(2f|Y=<7J+StyJ<-ic`diQ%lPnzq28g%eu{P9=I
zag2Psm$psR2Fx^&6Nny5nu|UxgvavySUQL0emWp`>Xw(zAzf|p6My3K%CIfX<FGw?
zwxFyQ7Ushy`)^0QCi!&ETF3d({qPU=KB1^4c$Uxzz5k`OgYIyZZ@Rj>jqAU8+oc6-
zFi+O&d7ViSQs3u?9FueZqhrZ;A3bTgOA+EZ*D~AmsVH|(YCGQ`DXZ~1A);)(nc~Q&
z^YD#J{^*44VO9cMCTD2E8QX|OTqA%39$d1JYp|{Vyuui4x#%IIP#&^S6PPV%eWrj(
z>EeTRw!<;X{}5v&$yId3!eslIJA_a_{=VH050?174DpPJgLYYua<BeJ<8`ZskBF`a
zkoD}qEH-nlR@QDLk+hJwhlG?PJtw65t;))bv{}109lV<KSncDldl*BmI)AJs30bxW
z3dI+8-*J&G9lzsQ1(Sfjjd1GVBtQrgZvW`(&i_-V+kcFr{JY66TyZ2xZ0Dakh5oAk
zD*hVzeu_K2{x>>sS!)zhVvg0FS0doPX1Mo*TW~Xdh4RHdEzze+`hedO;DVSvcha80
z{^2;XF>dblgqkF$=4C{=b^IFIV1LU2>tgS2gu5|Wq_LI@r;%LX16Gej-eJ(2ii;or
z<+o(}@Y-vCpY*?pN&eq_8TjuOXd5#~*Hhsv3^tGj{8X@D=CLDW=f@c_<XRI|`?h*%
zO(qQX){yMoiccsZC(OEz)ws=p(;L!x1vt3)LKy*q21ElKDi~}&mh2TTr0&-NPYX6b
z7+y&8Uli!>qebdyx*HtsYd}W)HHu(zJ9h#pA3h2Jhs;!Wbbhdsf4I(advb0PA^F%$
zH?PW5>u~>f?Q#)+jU0gssTfbH4&7wXUCV?c)ZJa7WkE32LQQ&SpFzhxNTIS0?WDcd
z0tXiU?Xb<i9y;^DFYM_`mfUMgLGp)}mrzhMl&-KcLhKV-u{1dj+I{J4VJ>!@2WQk>
z8<z;oG%I@6nJhJm=2UoWCWM^0`Dknuoi1~f_4ffDx+R>PIMX8*<f85F2`NJh38c61
zF44rS2Tfl0aF0<v_4xf11N&(g&9wn|mwG8bIOoqOw)i%?fMXy_C_(BV*P`EdM7L~d
zE?*lN%OG{3>)`~Uxy6MmiB{sfS+ry-JzU*5UEfgV(3DKFS1VVLKYag(=zJ0SeR87T
zpep#5hk9C6>9hE4A><b4ld`LFEWR}RzQF9&$DZyBV?erF{8+lm_mz%(rQ%{g@Fn2E
zUCWHQ-cGrU<Qw5k&Nx#~6Z6E|{m1X2sm%lf`_01KYeW~$k|4ctfvLvY#iZ79D%e9>
zr$4|xfpwQm>Udg>NCmijnfGR1&(dC-KDXFt^gXHS$<8i37CRFx)?dJ?TEj=M8d&er
zIJW!@f+f7<V97w;v6F+`F`9Hlz1XN|H%!H68SEz!Y#Zhrkdqv26Sf6D_|YxtJp~NE
zw3K*LsJ8QSbs0)CJF<FO9-G{h1C;#T*a-gJpX`;iFg)cwz-SK@>YllybBco{ntZ=`
zj)Sj0d_|6};MeLtBp=s=P6`goj=lqn79cOcIA~Jz_p+9|kr5RI?5zcNDK!k%zKnZ(
zgry%j)_}^;>5^H%ZSQ9O_rrs^!2?TxdpLWyLg2>`=@6NDW5$gFCT>sW`0lZCMsde>
zDr6^iCRWEC2`NOBkBS+(dB^T;NBu^CDA8{uIo$Jg^#1oI8AFQAKf)6w8*1nSYlUDG
zGzys~lXj7oakylEM`(IpDma-`>lFbG3fQB0QQgv1k{SJoi^e9gv_I5g*>5*tb2%y2
z_;|{1H}<QOwepW;v(|F|J3p4H&siVN^ICJIPsG{Se)j?oc|OW(0I<1m>LM}*>23lY
zYp(Uej}tPw7A!6T#!Vx=gfFYRYeW0`(ff|bPR+HK2HUe1i%(vY@BRON{x=E~bL<3P
Xk!;S$x}7{m{xwoX=WgMhhyVN!{B8$I

literal 0
HcmV?d00001

diff --git a/reference/featureChromatograms.html b/reference/featureChromatograms.html
index 1494801ab..42d7248d5 100644
--- a/reference/featureChromatograms.html
+++ b/reference/featureChromatograms.html
@@ -1,29 +1,37 @@
 <!DOCTYPE html>
-<!-- Generated by pkgdown: do not edit by hand --><html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="utf-8"><meta http-equiv="X-UA-Compatible" content="IE=edge"><meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"><meta name="description" content='Extract ion chromatograms for features in an XcmsExperiment or
+<!-- Generated by pkgdown: do not edit by hand --><html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="utf-8"><meta http-equiv="X-UA-Compatible" content="IE=edge"><meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"><meta name="description" content="Extract ion chromatograms for features in an XcmsExperiment or
 XCMSnExp object. The function returns for each feature the
 extracted ion chromatograms (along with all associated chromatographic
 peaks) in each sample. The chromatogram is extracted from the m/z - rt
-region including all chromatographic peaks of that features (i.e. based on
-the ranges of "mzmin", "mzmax", "rtmin", "rtmax" of all
-chromatographic peaks of the feature).
+region that includes all chromatographic peaks of a feature. By default,
+this region is defined using the range of the chromatographic peaks' m/z
+and retention times (with mzmin = min, mzmax = max, rtmin = min and
+rtmax = max). For some features, and depending on the data, the m/z and
+rt range can thus be relatively large. The boundaries of the m/z - rt
+region can also be restricted by changing parameters mzmin, mzmax,
+rtmin and rtmax to a different functions, such as median.
 By default only chromatographic peaks associated with a feature are
-included. For object being a XCMSnExp object parameter include
-allows also to return all chromatographic peaks with their apex
-position within the selected region (include = "apex_within") or any
-chromatographic peak overlapping the m/z and retention time range
-(include = "any").'><title>Extract ion chromatograms for each feature — featureChromatograms • xcms</title><!-- favicons --><link rel="icon" type="image/png" sizes="16x16" href="../favicon-16x16.png"><link rel="icon" type="image/png" sizes="32x32" href="../favicon-32x32.png"><link rel="apple-touch-icon" type="image/png" sizes="180x180" href="../apple-touch-icon.png"><link rel="apple-touch-icon" type="image/png" sizes="120x120" href="../apple-touch-icon-120x120.png"><link rel="apple-touch-icon" type="image/png" sizes="76x76" href="../apple-touch-icon-76x76.png"><link rel="apple-touch-icon" type="image/png" sizes="60x60" href="../apple-touch-icon-60x60.png"><script src="../deps/jquery-3.6.0/jquery-3.6.0.min.js"></script><meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"><link href="../deps/bootstrap-5.3.1/bootstrap.min.css" rel="stylesheet"><script src="../deps/bootstrap-5.3.1/bootstrap.bundle.min.js"></script><!-- Font Awesome icons --><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/all.min.css" integrity="sha256-mmgLkCYLUQbXn0B1SRqzHar6dCnv9oZFPEC1g1cwlkk=" crossorigin="anonymous"><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/v4-shims.min.css" integrity="sha256-wZjR52fzng1pJHwx4aV2AO3yyTOXrcDW7jBpJtTwVxw=" crossorigin="anonymous"><!-- bootstrap-toc --><script src="https://cdn.jsdelivr.net/gh/afeld/bootstrap-toc@v1.0.1/dist/bootstrap-toc.min.js" integrity="sha256-4veVQbu7//Lk5TSmc7YV48MxtMy98e26cf5MrgZYnwo=" crossorigin="anonymous"></script><!-- headroom.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/headroom.min.js" integrity="sha256-AsUX4SJE1+yuDu5+mAVzJbuYNPHj/WroHuZ8Ir/CkE0=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/jQuery.headroom.min.js" integrity="sha256-ZX/yNShbjqsohH1k95liqY9Gd8uOiE1S4vZc+9KQ1K4=" crossorigin="anonymous"></script><!-- clipboard.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/2.0.6/clipboard.min.js" integrity="sha256-inc5kl9MA1hkeYUt+EC3BhlIgyp/2jDIyBLS6k3UxPI=" crossorigin="anonymous"></script><!-- search --><script src="https://cdnjs.cloudflare.com/ajax/libs/fuse.js/6.4.6/fuse.js" integrity="sha512-zv6Ywkjyktsohkbp9bb45V6tEMoWhzFzXis+LrMehmJZZSys19Yxf1dopHx7WzIKxr5tK2dVcYmaCk2uqdjF4A==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/autocomplete.js/0.38.0/autocomplete.jquery.min.js" integrity="sha512-GU9ayf+66Xx2TmpxqJpliWbT5PiGYxpaG8rfnBEk1LL8l1KGkRShhngwdXK1UgqhAzWpZHSiYPc09/NwDQIGyg==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mark.js/8.11.1/mark.min.js" integrity="sha512-5CYOlHXGh6QpOFA/TeTylKLWfB3ftPsde7AnmhuitiTX4K5SqCLBeKro6sPS8ilsz1Q4NRx3v8Ko2IBiszzdww==" crossorigin="anonymous"></script><!-- pkgdown --><script src="../pkgdown.js"></script><meta property="og:title" content="Extract ion chromatograms for each feature — featureChromatograms"><meta property="og:description" content='Extract ion chromatograms for features in an XcmsExperiment or
+included in the returned XChromatograms object. For object being an
+XCMSnExp object parameter include allows also to return all
+chromatographic peaks with their apex position within the selected
+region (include = &quot;apex_within&quot;) or any chromatographic peak overlapping
+the m/z and retention time range (include = &quot;any&quot;)."><title>Extract ion chromatograms for each feature — featureChromatograms • xcms</title><!-- favicons --><link rel="icon" type="image/png" sizes="16x16" href="../favicon-16x16.png"><link rel="icon" type="image/png" sizes="32x32" href="../favicon-32x32.png"><link rel="apple-touch-icon" type="image/png" sizes="180x180" href="../apple-touch-icon.png"><link rel="apple-touch-icon" type="image/png" sizes="120x120" href="../apple-touch-icon-120x120.png"><link rel="apple-touch-icon" type="image/png" sizes="76x76" href="../apple-touch-icon-76x76.png"><link rel="apple-touch-icon" type="image/png" sizes="60x60" href="../apple-touch-icon-60x60.png"><script src="../deps/jquery-3.6.0/jquery-3.6.0.min.js"></script><meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"><link href="../deps/bootstrap-5.3.1/bootstrap.min.css" rel="stylesheet"><script src="../deps/bootstrap-5.3.1/bootstrap.bundle.min.js"></script><!-- Font Awesome icons --><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/all.min.css" integrity="sha256-mmgLkCYLUQbXn0B1SRqzHar6dCnv9oZFPEC1g1cwlkk=" crossorigin="anonymous"><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/v4-shims.min.css" integrity="sha256-wZjR52fzng1pJHwx4aV2AO3yyTOXrcDW7jBpJtTwVxw=" crossorigin="anonymous"><!-- bootstrap-toc --><script src="https://cdn.jsdelivr.net/gh/afeld/bootstrap-toc@v1.0.1/dist/bootstrap-toc.min.js" integrity="sha256-4veVQbu7//Lk5TSmc7YV48MxtMy98e26cf5MrgZYnwo=" crossorigin="anonymous"></script><!-- headroom.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/headroom.min.js" integrity="sha256-AsUX4SJE1+yuDu5+mAVzJbuYNPHj/WroHuZ8Ir/CkE0=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/jQuery.headroom.min.js" integrity="sha256-ZX/yNShbjqsohH1k95liqY9Gd8uOiE1S4vZc+9KQ1K4=" crossorigin="anonymous"></script><!-- clipboard.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/2.0.6/clipboard.min.js" integrity="sha256-inc5kl9MA1hkeYUt+EC3BhlIgyp/2jDIyBLS6k3UxPI=" crossorigin="anonymous"></script><!-- search --><script src="https://cdnjs.cloudflare.com/ajax/libs/fuse.js/6.4.6/fuse.js" integrity="sha512-zv6Ywkjyktsohkbp9bb45V6tEMoWhzFzXis+LrMehmJZZSys19Yxf1dopHx7WzIKxr5tK2dVcYmaCk2uqdjF4A==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/autocomplete.js/0.38.0/autocomplete.jquery.min.js" integrity="sha512-GU9ayf+66Xx2TmpxqJpliWbT5PiGYxpaG8rfnBEk1LL8l1KGkRShhngwdXK1UgqhAzWpZHSiYPc09/NwDQIGyg==" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mark.js/8.11.1/mark.min.js" integrity="sha512-5CYOlHXGh6QpOFA/TeTylKLWfB3ftPsde7AnmhuitiTX4K5SqCLBeKro6sPS8ilsz1Q4NRx3v8Ko2IBiszzdww==" crossorigin="anonymous"></script><!-- pkgdown --><script src="../pkgdown.js"></script><meta property="og:title" content="Extract ion chromatograms for each feature — featureChromatograms"><meta property="og:description" content="Extract ion chromatograms for features in an XcmsExperiment or
 XCMSnExp object. The function returns for each feature the
 extracted ion chromatograms (along with all associated chromatographic
 peaks) in each sample. The chromatogram is extracted from the m/z - rt
-region including all chromatographic peaks of that features (i.e. based on
-the ranges of "mzmin", "mzmax", "rtmin", "rtmax" of all
-chromatographic peaks of the feature).
+region that includes all chromatographic peaks of a feature. By default,
+this region is defined using the range of the chromatographic peaks' m/z
+and retention times (with mzmin = min, mzmax = max, rtmin = min and
+rtmax = max). For some features, and depending on the data, the m/z and
+rt range can thus be relatively large. The boundaries of the m/z - rt
+region can also be restricted by changing parameters mzmin, mzmax,
+rtmin and rtmax to a different functions, such as median.
 By default only chromatographic peaks associated with a feature are
-included. For object being a XCMSnExp object parameter include
-allows also to return all chromatographic peaks with their apex
-position within the selected region (include = "apex_within") or any
-chromatographic peak overlapping the m/z and retention time range
-(include = "any").'><meta property="og:image" content="https://sneumann.github.io/xcms/logo.png"><!-- mathjax --><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js" integrity="sha256-nvJJv9wWKEm88qvoQl9ekL2J+k/RWIsaSScxxlsrv8k=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/config/TeX-AMS-MML_HTMLorMML.js" integrity="sha256-84DKXVJXs0/F8OTMzX4UR909+jtl4G7SPypPavF+GfA=" crossorigin="anonymous"></script><!--[if lt IE 9]>
+included in the returned XChromatograms object. For object being an
+XCMSnExp object parameter include allows also to return all
+chromatographic peaks with their apex position within the selected
+region (include = &quot;apex_within&quot;) or any chromatographic peak overlapping
+the m/z and retention time range (include = &quot;any&quot;)."><meta property="og:image" content="https://sneumann.github.io/xcms/logo.png"><!-- mathjax --><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js" integrity="sha256-nvJJv9wWKEm88qvoQl9ekL2J+k/RWIsaSScxxlsrv8k=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/config/TeX-AMS-MML_HTMLorMML.js" integrity="sha256-84DKXVJXs0/F8OTMzX4UR909+jtl4G7SPypPavF+GfA=" crossorigin="anonymous"></script><!--[if lt IE 9]>
 <script src="https://oss.maxcdn.com/html5shiv/3.7.3/html5shiv.min.js"></script>
 <script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
 <![endif]--></head><body>
@@ -34,7 +42,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -82,15 +90,19 @@
 <a href="XCMSnExp-class.html">XCMSnExp</a> object. The function returns for each feature the
 extracted ion chromatograms (along with all associated chromatographic
 peaks) in each sample. The chromatogram is extracted from the m/z - rt
-region including all chromatographic peaks of that features (i.e. based on
-the ranges of <code>"mzmin"</code>, <code>"mzmax"</code>, <code>"rtmin"</code>, <code>"rtmax"</code> of all
-chromatographic peaks of the feature).</p>
+region that includes <strong>all</strong> chromatographic peaks of a feature. By default,
+this region is defined using the range of the chromatographic peaks' m/z
+and retention times (with <code>mzmin = min</code>, <code>mzmax = max</code>, <code>rtmin = min</code> and
+<code>rtmax = max</code>). For some features, and depending on the data, the m/z and
+rt range can thus be relatively large. The boundaries of the m/z - rt
+region can also be restricted by changing parameters <code>mzmin</code>, <code>mzmax</code>,
+<code>rtmin</code> and <code>rtmax</code> to a different functions, such as <code>median</code>.</p>
 <p>By default only chromatographic peaks associated with a feature are
-included. For <code>object</code> being a <code>XCMSnExp</code> object parameter <code>include</code>
-allows also to return all chromatographic peaks with their apex
-position within the selected region (<code>include = "apex_within"</code>) or any
-chromatographic peak overlapping the m/z and retention time range
-(<code>include = "any"</code>).</p>
+included in the returned <a href="XChromatogram.html">XChromatograms</a> object. For <code>object</code> being an
+<code>XCMSnExp</code> object parameter <code>include</code> allows also to return all
+chromatographic peaks with their apex position within the selected
+region (<code>include = "apex_within"</code>) or any chromatographic peak overlapping
+the m/z and retention time range (<code>include = "any"</code>).</p>
     </div>
 
     <div class="section level2">
@@ -106,6 +118,10 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
 <span>  features <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/character.html" class="external-link">character</a></span><span class="op">(</span><span class="op">)</span>,</span>
 <span>  return.type <span class="op">=</span> <span class="st">"XChromatograms"</span>,</span>
 <span>  chunkSize <span class="op">=</span> <span class="fl">2L</span>,</span>
+<span>  mzmin <span class="op">=</span> <span class="va">min</span>,</span>
+<span>  mzmax <span class="op">=</span> <span class="va">max</span>,</span>
+<span>  rtmin <span class="op">=</span> <span class="va">min</span>,</span>
+<span>  rtmax <span class="op">=</span> <span class="va">max</span>,</span>
 <span>  <span class="va">...</span>,</span>
 <span>  progressbar <span class="op">=</span> <span class="cn">TRUE</span>,</span>
 <span>  BPPARAM <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/register.html" class="external-link">bpparam</a></span><span class="op">(</span><span class="op">)</span></span>
@@ -179,6 +195,34 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 a time into memory. Defaults to <code>chunkSize = 2L</code>.</p></dd>
 
 
+<dt>mzmin</dt>
+<dd><p><code>function</code> defining how the lower boundary of the m/z region
+from which the EIC is integrated should be defined. Defaults to
+<code>mzmin = min</code> thus the smallest <code>"mzmin"</code> value for all chromatographic
+peaks of a feature will be used.</p></dd>
+
+
+<dt>mzmax</dt>
+<dd><p><code>function</code> defining how the upper boundary of the m/z region
+from which the EIC is integrated should be defined. Defaults to
+<code>mzmax = max</code> thus the largest <code>"mzmax"</code> value for all chromatographic
+peaks of a feature will be used.</p></dd>
+
+
+<dt>rtmin</dt>
+<dd><p><code>function</code> defining how the lower boundary of the rt region
+from which the EIC is integrated should be defined. Defaults to
+<code>rtmin = min</code> thus the smallest <code>"rtmin"</code> value for all chromatographic
+peaks of a feature will be used.</p></dd>
+
+
+<dt>rtmax</dt>
+<dd><p><code>function</code> defining how the upper boundary of the rt region
+from which the EIC is integrated should be defined. Defaults to
+<code>rtmax = max</code> thus the largest <code>"rtmax"</code> value for all chromatographic
+peaks of a feature will be used.</p></dd>
+
+
 <dt>progressbar</dt>
 <dd><p><code>logical(1)</code> defining whether a progress bar is shown.</p></dd>
 
@@ -225,7 +269,13 @@ <h2 id="value">Value<a class="anchor" aria-label="anchor" href="#value"></a></h2
     </div>
     <div class="section level2">
     <h2 id="note">Note<a class="anchor" aria-label="anchor" href="#note"></a></h2>
-    <p>Parameters <code>include</code>, <code>filled</code>, <code>n</code> and <code>value</code> are only supported
+    <p>The EIC data of a feature is extracted from every sample using the same
+m/z - rt area. The EIC in a sample does thus not exactly represent the
+signal of the actually identified chromatographic peak in that sample.
+The <code><a href="chromPeakChromatograms.html">chromPeakChromatograms()</a></code> function would allow to extract the actual
+EIC of the chromatographic peak in a specific sample. See also examples
+below.</p>
+<p>Parameters <code>include</code>, <code>filled</code>, <code>n</code> and <code>value</code> are only supported
 for <code>object</code> being an <code>XCMSnExp</code>.</p>
 <p>When extracting EICs from only the top <code>n</code> samples it can happen that one
 or more of the features specified with <code>features</code> are dropped because they
@@ -253,27 +303,34 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span><span class="co">## Disable parallel processing for this example</span></span></span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/register.html" class="external-link">register</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/pkg/BiocParallel/man/SerialParam-class.html" class="external-link">SerialParam</a></span><span class="op">(</span><span class="op">)</span><span class="op">)</span></span></span>
 <span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co">## Subset the object to a smaller retention time range</span></span></span>
-<span class="r-in"><span><span class="va">xdata</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/pkg/ProtGenerics/man/protgenerics.html" class="external-link">filterRt</a></span><span class="op">(</span><span class="va">faahko_sub</span>, <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">2500</span>, <span class="fl">3500</span><span class="op">)</span><span class="op">)</span></span></span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> Filter spectra</span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="va">xdata</span> <span class="op">&lt;-</span> <span class="fu"><a href="groupChromPeaks.html">groupChromPeaks</a></span><span class="op">(</span><span class="va">xdata</span>,</span></span>
+<span class="r-in"><span><span class="co">## Perform correspondence analysis</span></span></span>
+<span class="r-in"><span><span class="va">xdata</span> <span class="op">&lt;-</span> <span class="fu"><a href="groupChromPeaks.html">groupChromPeaks</a></span><span class="op">(</span><span class="va">faahko_sub</span>,</span></span>
 <span class="r-in"><span>    param <span class="op">=</span> <span class="fu"><a href="groupChromPeaks.html">PeakDensityParam</a></span><span class="op">(</span>minFraction <span class="op">=</span> <span class="fl">0.8</span>, sampleGroups <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/rep.html" class="external-link">rep</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">3</span><span class="op">)</span><span class="op">)</span><span class="op">)</span></span></span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## Get the feature definitions</span></span></span>
 <span class="r-in"><span><span class="fu"><a href="XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">xdata</span><span class="op">)</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>      mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks 1      peakidx</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT01 300.2 300.2 300.2 3387.143 3379.317 3390.271      4 3 35, 91, ....</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT02 301.0 301.0 301.0 2787.766 2786.200 2792.459      3 3  10, 63, 117</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT03 326.2 326.2 326.2 3416.877 3415.311 3424.700      3 3  37, 94, 132</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT04 343.0 343.0 343.0 2684.479 2678.218 2686.042      3 3   6, 58, 114</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT05 344.0 344.0 344.0 2682.914 2679.783 2686.042      3 3   3, 56, 112</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT06 365.0 365.0 365.0 2684.479 2679.783 2686.042      3 3   7, 59, 115</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT07 496.2 496.2 496.2 3335.498 3316.719 3340.194      3 3 47, 103, 138</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT08 496.2 496.2 496.2 3402.791 3384.012 3410.617      3 3 48, 104, 139</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT09 522.2 522.2 522.2 3387.142 3344.888 3434.091      6 3 52, 53, ....</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT10 523.2 523.2 523.2 3387.142 3344.888 3434.091      6 3 38, 39, ....</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> FT11 532.2 532.2 532.2 3482.605 3476.344 3488.863      3 3 49, 102, 137</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT01 300.2 300.2 300.2 3387.143 3379.317 3390.271      4 3 35, 125,....</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT02 301.0 301.0 301.0 2787.766 2786.200 2792.459      3 3  10, 97, 198</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT03 326.2 326.2 326.2 3416.877 3415.311 3424.700      3 3 37, 128, 213</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT04 328.2 328.2 328.2 3626.581 3618.755 3631.274      3 3 64, 159, 231</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT05 329.2 329.2 329.2 3626.581 3618.755 3631.274      3 3 63, 156, 230</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT06 343.0 343.0 343.0 2684.479 2678.218 2686.042      3 3   6, 92, 195</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT07 344.0 344.0 344.0 2682.914 2679.783 2686.042      3 3   3, 90, 193</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT08 354.2 354.2 354.2 3614.061 3610.930 3621.884      3 3 56, 153, 228</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT09 356.2 356.2 356.2 3830.025 3819.069 3833.153      3 3 78, 180, 241</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT10 365.0 365.0 365.0 2684.479 2679.783 2686.042      3 3   7, 93, 196</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT11 438.3 438.3 438.3 4052.248 4047.552 4060.071      3 3 81, 186, 248</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT12 496.2 496.2 496.2 3335.498 3316.719 3340.194      3 3 47, 143, 220</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT13 496.2 496.2 496.2 3402.791 3384.012 3410.617      3 3 48, 144, 221</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT14 508.2 508.2 508.2 3524.859 3515.468 3529.552      3 3 67, 163, 232</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT15 509.2 509.2 509.2 3523.294 3512.338 3531.117      3 3 54, 160, 229</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT16 522.2 522.2 522.2 3387.142 3344.888 3434.091      6 3 52, 53, ....</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT17 523.2 523.2 523.2 3387.142 3344.888 3434.091      6 3 38, 39, ....</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT18 524.2 524.2 524.2 3676.658 3598.410 3704.828      4 3 74, 75, ....</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT19 525.2 525.2 525.2 3690.742 3662.574 3706.393      3 3 72, 173, 238</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT20 532.2 532.2 532.2 3482.605 3476.344 3488.863      3 3 49, 137, 219</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT21 536.2 536.2 536.2 3712.653 3703.262 3718.912      3 3 73, 174, 237</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>      ms_level</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> FT01        1</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> FT02        1</span>
@@ -286,15 +343,50 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> FT09        1</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> FT10        1</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> FT11        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT12        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT13        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT14        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT15        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT16        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT17        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT18        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT19        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT20        1</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT21        1</span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## Extract ion chromatograms for the first 3 features. Parameter</span></span></span>
 <span class="r-in"><span><span class="co">## `features` can be either the feature IDs or feature indices.</span></span></span>
 <span class="r-in"><span><span class="va">chrs</span> <span class="op">&lt;-</span> <span class="fu">featureChromatograms</span><span class="op">(</span><span class="va">xdata</span>,</span></span>
 <span class="r-in"><span>    features <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/colnames.html" class="external-link">rownames</a></span><span class="op">(</span><span class="va">featureDefinitions</span><span class="op">)</span><span class="op">[</span><span class="fl">1</span><span class="op">:</span><span class="fl">3</span><span class="op">]</span><span class="op">)</span></span></span>
 <span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co">## Plot the XIC for the first feature using different colors for each file</span></span></span>
+<span class="r-in"><span><span class="co">## Plot the EIC for the first feature using different colors for each file.</span></span></span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chrs</span><span class="op">[</span><span class="fl">1</span>, <span class="op">]</span>, col <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"red"</span>, <span class="st">"green"</span>, <span class="st">"blue"</span><span class="op">)</span><span class="op">)</span></span></span>
 <span class="r-plt img"><img src="featureChromatograms-1.png" alt="" width="700" height="433"></span>
+<span class="r-in"><span></span></span>
+<span class="r-in"><span><span class="co">## The EICs for all 3 samples use the same m/z and retention time range,</span></span></span>
+<span class="r-in"><span><span class="co">## which was defined using the `featureArea` function:</span></span></span>
+<span class="r-in"><span><span class="fu"><a href="XcmsExperiment.html">featureArea</a></span><span class="op">(</span><span class="va">xdata</span>, features <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/colnames.html" class="external-link">rownames</a></span><span class="op">(</span><span class="fu"><a href="XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">xdata</span><span class="op">)</span><span class="op">)</span><span class="op">[</span><span class="fl">1</span><span class="op">:</span><span class="fl">3</span><span class="op">]</span>,</span></span>
+<span class="r-in"><span>    mzmin <span class="op">=</span> <span class="va">min</span>, mzmax <span class="op">=</span> <span class="va">max</span>, rtmin <span class="op">=</span> <span class="va">min</span>, rtmax <span class="op">=</span> <span class="va">max</span><span class="op">)</span></span></span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>      mzmin mzmax    rtmin    rtmax</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT01 300.2 300.2 3362.104 3413.746</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT02 301.0 301.0 2765.855 2820.630</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> FT03 326.2 326.2 3391.837 3449.740</span>
+<span class="r-in"><span></span></span>
+<span class="r-in"><span><span class="co">## To extract the actual (exact) EICs for each chromatographic peak of</span></span></span>
+<span class="r-in"><span><span class="co">## a feature in each sample, the `chromPeakChromatograms` function would</span></span></span>
+<span class="r-in"><span><span class="co">## need to be used instead. Below we extract the EICs for all</span></span></span>
+<span class="r-in"><span><span class="co">## chromatographic peaks of the first feature. We need to first get the</span></span></span>
+<span class="r-in"><span><span class="co">## IDs of all chromatographic peaks assigned to the first feature:</span></span></span>
+<span class="r-in"><span><span class="va">peak_ids</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/colnames.html" class="external-link">rownames</a></span><span class="op">(</span><span class="fu"><a href="XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">xdata</span><span class="op">)</span><span class="op">)</span><span class="op">[</span><span class="fu"><a href="XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">xdata</span><span class="op">)</span><span class="op">$</span><span class="va">peakidx</span><span class="op">[[</span><span class="fl">1L</span><span class="op">]</span><span class="op">]</span><span class="op">]</span></span></span>
+<span class="r-in"><span></span></span>
+<span class="r-in"><span><span class="co">## We can now pass these to the `chromPeakChromatograms` function with</span></span></span>
+<span class="r-in"><span><span class="co">## parameter `peaks`:</span></span></span>
+<span class="r-in"><span><span class="va">eic_1</span> <span class="op">&lt;-</span> <span class="fu"><a href="chromPeakChromatograms.html">chromPeakChromatograms</a></span><span class="op">(</span><span class="va">xdata</span>, peaks <span class="op">=</span> <span class="va">peak_ids</span><span class="op">)</span></span></span>
+<span class="r-in"><span></span></span>
+<span class="r-in"><span><span class="co">## To plot these into a single plot we need to use the</span></span></span>
+<span class="r-in"><span><span class="co">## `plotChromatogramsOverlay` function:</span></span></span>
+<span class="r-in"><span><span class="fu"><a href="plotChromatogramsOverlay.html">plotChromatogramsOverlay</a></span><span class="op">(</span><span class="va">eic_1</span><span class="op">)</span></span></span>
+<span class="r-plt img"><img src="featureChromatograms-2.png" alt="" width="700" height="433"></span>
 </code></pre></div>
     </div>
   </main><aside class="col-md-3"><nav id="toc"><h2>On this page</h2>
diff --git a/reference/featureSpectra.html b/reference/featureSpectra.html
index 1647ca5e9..d969367a9 100644
--- a/reference/featureSpectra.html
+++ b/reference/featureSpectra.html
@@ -54,7 +54,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/featureSummary.html b/reference/featureSummary.html
index 192d77038..2c8683cb6 100644
--- a/reference/featureSummary.html
+++ b/reference/featureSummary.html
@@ -22,7 +22,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/fillChromPeaks.html b/reference/fillChromPeaks.html
index 013f8feca..e4c427144 100644
--- a/reference/fillChromPeaks.html
+++ b/reference/fillChromPeaks.html
@@ -23,8 +23,8 @@
 ChromPeakAreaParam, the median &quot;mzmin&quot;, &quot;mzmax&quot;, &quot;rtmin&quot; and
 &quot;rtmax&quot; values from all detected chromatographic peaks of a feature
 would be used instead.
-In contrast to the  FillChromPeaksParam approach this method uses the
-actual identified chromatographic peaks of a feature to define the area
+In contrast to the  FillChromPeaksParam approach this method uses (all)
+identified chromatographic peaks of a feature to define the area
 from which the signal should be integrated.
 
 
@@ -57,8 +57,8 @@
 ChromPeakAreaParam, the median &quot;mzmin&quot;, &quot;mzmax&quot;, &quot;rtmin&quot; and
 &quot;rtmax&quot; values from all detected chromatographic peaks of a feature
 would be used instead.
-In contrast to the  FillChromPeaksParam approach this method uses the
-actual identified chromatographic peaks of a feature to define the area
+In contrast to the  FillChromPeaksParam approach this method uses (all)
+identified chromatographic peaks of a feature to define the area
 from which the signal should be integrated.
 
 
@@ -78,7 +78,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -145,8 +145,8 @@
 <code>ChromPeakAreaParam</code>, the median <code>"mzmin"</code>, <code>"mzmax"</code>, <code>"rtmin"</code> and
 <code>"rtmax"</code> values from all detected chromatographic peaks of a feature
 would be used instead.
-In contrast to the  <code>FillChromPeaksParam</code> approach this method uses the
-actual identified chromatographic peaks of a feature to define the area
+In contrast to the  <code>FillChromPeaksParam</code> approach this method uses (all)
+identified chromatographic peaks of a feature to define the area
 from which the signal should be integrated.</p></li>
 </ul><p><code>expandMz</code>,<code>expandMz&lt;-</code>: getter and setter
     for the <code>expandMz</code> slot of the object.</p>
@@ -182,10 +182,10 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
 <span><span class="fu">fixedMz</span><span class="op">(</span><span class="va">object</span><span class="op">)</span></span>
 <span></span>
 <span><span class="fu">ChromPeakAreaParam</span><span class="op">(</span></span>
-<span>  mzmin <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.25</span><span class="op">)</span>,</span>
-<span>  mzmax <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.75</span><span class="op">)</span>,</span>
-<span>  rtmin <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.25</span><span class="op">)</span>,</span>
-<span>  rtmax <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.75</span><span class="op">)</span></span>
+<span>  mzmin <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.25</span>, names <span class="op">=</span> <span class="cn">FALSE</span><span class="op">)</span>,</span>
+<span>  mzmax <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.75</span>, names <span class="op">=</span> <span class="cn">FALSE</span><span class="op">)</span>,</span>
+<span>  rtmin <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.25</span>, names <span class="op">=</span> <span class="cn">FALSE</span><span class="op">)</span>,</span>
+<span>  rtmax <span class="op">=</span> <span class="kw">function</span><span class="op">(</span><span class="va">z</span><span class="op">)</span> <span class="fu"><a href="https://rdrr.io/r/stats/quantile.html" class="external-link">quantile</a></span><span class="op">(</span><span class="va">z</span>, probs <span class="op">=</span> <span class="fl">0.75</span>, names <span class="op">=</span> <span class="cn">FALSE</span><span class="op">)</span></span>
 <span><span class="op">)</span></span>
 <span></span>
 <span><span class="co"># S4 method for FillChromPeaksParam</span></span>
@@ -434,18 +434,18 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## How many missing values do we have after peak filling?</span></span></span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/base/sum.html" class="external-link">sum</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/NA.html" class="external-link">is.na</a></span><span class="op">(</span><span class="fu"><a href="XCMSnExp-peak-grouping-results.html">featureValues</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span><span class="op">)</span><span class="op">)</span></span></span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [1] 3</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [1] 2</span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## Get the peaks that have been filled in:</span></span></span>
 <span class="r-in"><span><span class="va">fp</span> <span class="op">&lt;-</span> <span class="fu"><a href="XCMSnExp-class.html">chromPeaks</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span><span class="op">[</span><span class="fu"><a href="XCMSnExp-class.html">chromPeakData</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span><span class="op">$</span><span class="va">is_filled</span>, <span class="op">]</span></span></span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="va">fp</span><span class="op">)</span></span></span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>          mz mzmin mzmax       rt    rtmin    rtmax      into intb   maxo sn</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> CP249 286.2 286.2 286.2 3252.556 3236.907 3274.857 1288521.3   NA  78048 NA</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> CP250 380.1 380.1 380.1 3172.743 3129.707 3173.526  677133.8   NA  27040 NA</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> CP251 447.2 447.2 447.2 3858.193 3828.068 3914.140 2156784.3   NA  61256 NA</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> CP252 497.2 497.2 497.2 3382.447 3379.317 3438.786 7672169.4   NA 310912 NA</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> CP253 510.2 510.2 510.2 3750.211 3741.213 3799.898 1927179.7   NA  61368 NA</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> CP254 531.2 531.2 531.2 3340.193 3328.848 3383.620 7589092.3   NA 254464 NA</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>          mz mzmin mzmax       rt    rtmin    rtmax     into intb   maxo sn</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> CP249 286.2 286.2 286.2 3252.556 3236.907 3274.857  1288521   NA  78048 NA</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> CP250 380.1 380.1 380.1 3193.087 3132.054 3216.561  2094879   NA  35928 NA</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> CP251 447.2 447.2 447.2 3858.193 3828.068 3914.140  2156784   NA  61256 NA</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> CP252 497.2 497.2 497.2 3382.447 3347.236 3438.785 10344783   NA 310912 NA</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> CP253 510.2 510.2 510.2 3750.211 3741.213 3799.898  1927180   NA  61368 NA</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> CP254 531.2 531.2 531.2 3340.193 3328.848 3383.620  7589092   NA 254464 NA</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>       sample</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> CP249      1</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> CP250      1</span>
@@ -460,7 +460,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span><span class="va">ph</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Object of class "XProcessHistory"</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  type: Missing peak filling </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  date: Thu Dec  7 08:43:03 2023 </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  date: Fri Jan 26 20:05:57 2024 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  info:  </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  fileIndex: 1,2,3 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  Parameter class: ChromPeakAreaParam </span>
@@ -471,17 +471,17 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> Object of class:  ChromPeakAreaParam </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  Parameters:</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  - rtmin: function (z) </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.25)</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55feb8ba0e40&gt;</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.25, names = FALSE)</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55c9fa2bec18&gt;</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  - rtmax: function (z) </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.75)</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55feb8ba0e40&gt;</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.75, names = FALSE)</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55c9fa2bec18&gt;</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  - mzmin: function (z) </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.25)</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55feb8ba0e40&gt;</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.25, names = FALSE)</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55c9fa2bec18&gt;</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  - mzmax: function (z) </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.75)</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55feb8ba0e40&gt;</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> quantile(z, probs = 0.75, names = FALSE)</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> &lt;environment: 0x55c9fa2bec18&gt;</span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co">## It is also possible to remove filled-in peaks:</span></span></span>
 <span class="r-in"><span><span class="va">res</span> <span class="op">&lt;-</span> <span class="fu"><a href="XCMSnExp-class.html">dropFilledChromPeaks</a></span><span class="op">(</span><span class="va">res</span><span class="op">)</span></span></span>
diff --git a/reference/fillPeaks-methods.html b/reference/fillPeaks-methods.html
index 20007ee33..25f185325 100644
--- a/reference/fillPeaks-methods.html
+++ b/reference/fillPeaks-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/fillPeaks.MSW-methods.html b/reference/fillPeaks.MSW-methods.html
index 37add06f6..aab1bf5eb 100644
--- a/reference/fillPeaks.MSW-methods.html
+++ b/reference/fillPeaks.MSW-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/fillPeaks.chrom-methods.html b/reference/fillPeaks.chrom-methods.html
index c14d05169..02f2ae5d8 100644
--- a/reference/fillPeaks.chrom-methods.html
+++ b/reference/fillPeaks.chrom-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/filter-MChromatograms.html b/reference/filter-MChromatograms.html
index 4920d8dfa..618fa20fa 100644
--- a/reference/filter-MChromatograms.html
+++ b/reference/filter-MChromatograms.html
@@ -62,7 +62,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/filtfft.html b/reference/filtfft.html
index b4c74f672..d03d608ed 100644
--- a/reference/filtfft.html
+++ b/reference/filtfft.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findChromPeaks-Chromatogram-CentWaveParam.html b/reference/findChromPeaks-Chromatogram-CentWaveParam.html
index f5753f12a..1bd6cfd83 100644
--- a/reference/findChromPeaks-Chromatogram-CentWaveParam.html
+++ b/reference/findChromPeaks-Chromatogram-CentWaveParam.html
@@ -28,7 +28,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findChromPeaks-Chromatogram-MatchedFilter.html b/reference/findChromPeaks-Chromatogram-MatchedFilter.html
index b79d277ae..91bb79398 100644
--- a/reference/findChromPeaks-Chromatogram-MatchedFilter.html
+++ b/reference/findChromPeaks-Chromatogram-MatchedFilter.html
@@ -22,7 +22,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findChromPeaks-centWave.html b/reference/findChromPeaks-centWave.html
index 83d159837..09342a738 100644
--- a/reference/findChromPeaks-centWave.html
+++ b/reference/findChromPeaks-centWave.html
@@ -86,7 +86,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -389,7 +389,7 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 <code>beta_snr</code> to the <code>chromPeaks</code> output, corresponding to a Pearson
 correlation coefficient to a bell curve with several degrees of skew as well
 as an estimate of signal-to-noise using the residuals from the best-fitting
-bell curve. See https://github.com/sneumann/xcms/pull/685 and 
+bell curve. See https://github.com/sneumann/xcms/pull/685 and
 https://doi.org/10.1186/s12859-023-05533-4 for more information.</p></dd>
 
 
@@ -510,10 +510,10 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
 <p><code><a href="XCMSnExp-class.html">XCMSnExp</a></code> for the object containing the results of
     the peak detection.</p>
 <p>Other peak detection methods: 
+<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
 <code><a href="findChromPeaks-massifquant.html">findChromPeaks-massifquant</a></code>,
 <code><a href="findChromPeaks-matchedFilter.html">findChromPeaks-matchedFilter</a></code>,
-<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findPeaks-MSW.html">findPeaks-MSW</a></code></p></div>
     </div>
     <div class="section level2">
diff --git a/reference/findChromPeaks-centWaveWithPredIsoROIs.html b/reference/findChromPeaks-centWaveWithPredIsoROIs.html
index d77df0a30..eb950e446 100644
--- a/reference/findChromPeaks-centWaveWithPredIsoROIs.html
+++ b/reference/findChromPeaks-centWaveWithPredIsoROIs.html
@@ -68,7 +68,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -319,7 +319,7 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 <code>beta_snr</code> to the <code>chromPeaks</code> output, corresponding to a Pearson
 correlation coefficient to a bell curve with several degrees of skew as well
 as an estimate of signal-to-noise using the residuals from the best-fitting
-bell curve. See https://github.com/sneumann/xcms/pull/685 and 
+bell curve. See https://github.com/sneumann/xcms/pull/685 and
 https://doi.org/10.1186/s12859-023-05533-4 for more information.</p></dd>
 
 
@@ -446,10 +446,10 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
 <p><code><a href="XCMSnExp-class.html">XCMSnExp</a></code> for the object containing the results of
     the peak detection.</p>
 <p>Other peak detection methods: 
+<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findChromPeaks-centWave.html">findChromPeaks-centWave</a></code>,
 <code><a href="findChromPeaks-massifquant.html">findChromPeaks-massifquant</a></code>,
 <code><a href="findChromPeaks-matchedFilter.html">findChromPeaks-matchedFilter</a></code>,
-<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findPeaks-MSW.html">findPeaks-MSW</a></code></p></div>
     </div>
     <div class="section level2">
diff --git a/reference/findChromPeaks-massifquant.html b/reference/findChromPeaks-massifquant.html
index 3c605e4f2..41815903f 100644
--- a/reference/findChromPeaks-massifquant.html
+++ b/reference/findChromPeaks-massifquant.html
@@ -102,7 +102,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -561,10 +561,10 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
 <p><code><a href="XCMSnExp-class.html">XCMSnExp</a></code> for the object containing the results of
     the peak detection.</p>
 <p>Other peak detection methods: 
-<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
+<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findChromPeaks-centWave.html">findChromPeaks-centWave</a></code>,
+<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
 <code><a href="findChromPeaks-matchedFilter.html">findChromPeaks-matchedFilter</a></code>,
-<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findPeaks-MSW.html">findPeaks-MSW</a></code></p></div>
     </div>
     <div class="section level2">
diff --git a/reference/findChromPeaks-matchedFilter.html b/reference/findChromPeaks-matchedFilter.html
index 9d237eafb..47664b312 100644
--- a/reference/findChromPeaks-matchedFilter.html
+++ b/reference/findChromPeaks-matchedFilter.html
@@ -96,7 +96,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -459,10 +459,10 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
 <p><code><a href="XCMSnExp-class.html">XCMSnExp</a></code> for the object containing the results of
     the chromatographic peak detection.</p>
 <p>Other peak detection methods: 
-<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
+<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findChromPeaks-centWave.html">findChromPeaks-centWave</a></code>,
+<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
 <code><a href="findChromPeaks-massifquant.html">findChromPeaks-massifquant</a></code>,
-<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findPeaks-MSW.html">findPeaks-MSW</a></code></p></div>
     </div>
     <div class="section level2">
diff --git a/reference/findChromPeaks.html b/reference/findChromPeaks.html
index ebd671be3..fdb0f8984 100644
--- a/reference/findChromPeaks.html
+++ b/reference/findChromPeaks.html
@@ -44,7 +44,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -186,8 +186,8 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
 chromatographic peaks.</p>
 <p><code><a href="manualChromPeaks.html">manualChromPeaks()</a></code> to manually add/define chromatographic peaks.</p>
 <p>Other peak detection methods: 
-<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
 <code><a href="findChromPeaks-centWave.html">findChromPeaks-centWave</a></code>,
+<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
 <code><a href="findChromPeaks-massifquant.html">findChromPeaks-massifquant</a></code>,
 <code><a href="findChromPeaks-matchedFilter.html">findChromPeaks-matchedFilter</a></code>,
 <code><a href="findPeaks-MSW.html">findPeaks-MSW</a></code></p></div>
diff --git a/reference/findChromPeaksIsolationWindow.html b/reference/findChromPeaksIsolationWindow.html
index 60e3b9972..a885a59a0 100644
--- a/reference/findChromPeaksIsolationWindow.html
+++ b/reference/findChromPeaksIsolationWindow.html
@@ -42,7 +42,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findEqualGreater.html b/reference/findEqualGreater.html
index 5dc47262f..16f12b3df 100644
--- a/reference/findEqualGreater.html
+++ b/reference/findEqualGreater.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findMZ.html b/reference/findMZ.html
index 0f2e6f336..06fd95f46 100644
--- a/reference/findMZ.html
+++ b/reference/findMZ.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks-MSW.html b/reference/findPeaks-MSW.html
index 2446108cc..66aebb585 100644
--- a/reference/findPeaks-MSW.html
+++ b/reference/findPeaks-MSW.html
@@ -86,7 +86,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -416,11 +416,11 @@ <h2 id="see-also">See also<a class="anchor" aria-label="anchor" href="#see-also"
 <p><code><a href="XCMSnExp-class.html">XCMSnExp</a></code> for the object containing the results of
     the peak detection.</p>
 <p>Other peak detection methods: 
-<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
+<code><a href="findChromPeaks.html">findChromPeaks</a>()</code>,
 <code><a href="findChromPeaks-centWave.html">findChromPeaks-centWave</a></code>,
+<code><a href="findChromPeaks-centWaveWithPredIsoROIs.html">findChromPeaks-centWaveWithPredIsoROIs</a></code>,
 <code><a href="findChromPeaks-massifquant.html">findChromPeaks-massifquant</a></code>,
-<code><a href="findChromPeaks-matchedFilter.html">findChromPeaks-matchedFilter</a></code>,
-<code><a href="findChromPeaks.html">findChromPeaks</a>()</code></p></div>
+<code><a href="findChromPeaks-matchedFilter.html">findChromPeaks-matchedFilter</a></code></p></div>
     </div>
     <div class="section level2">
     <h2 id="author">Author<a class="anchor" aria-label="anchor" href="#author"></a></h2>
diff --git a/reference/findPeaks-methods.html b/reference/findPeaks-methods.html
index 4b5456d7c..e194901d9 100644
--- a/reference/findPeaks-methods.html
+++ b/reference/findPeaks-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks.MS1-methods.html b/reference/findPeaks.MS1-methods.html
index 8bc04e50e..51b401f9d 100644
--- a/reference/findPeaks.MS1-methods.html
+++ b/reference/findPeaks.MS1-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks.MSW-xcmsRaw-method.html b/reference/findPeaks.MSW-xcmsRaw-method.html
index 35d902252..69f18c135 100644
--- a/reference/findPeaks.MSW-xcmsRaw-method.html
+++ b/reference/findPeaks.MSW-xcmsRaw-method.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks.addPredictedIsotopeFeatures-methods.html b/reference/findPeaks.addPredictedIsotopeFeatures-methods.html
index a2c5a400a..74a795726 100644
--- a/reference/findPeaks.addPredictedIsotopeFeatures-methods.html
+++ b/reference/findPeaks.addPredictedIsotopeFeatures-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks.centWave-methods.html b/reference/findPeaks.centWave-methods.html
index b717e418f..dde59eaad 100644
--- a/reference/findPeaks.centWave-methods.html
+++ b/reference/findPeaks.centWave-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html b/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html
index acbe170c5..f314ccb81 100644
--- a/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html
+++ b/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks.massifquant-methods.html b/reference/findPeaks.massifquant-methods.html
index e573fab2f..02df74ef4 100644
--- a/reference/findPeaks.massifquant-methods.html
+++ b/reference/findPeaks.massifquant-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findPeaks.matchedFilter-xcmsRaw-method.html b/reference/findPeaks.matchedFilter-xcmsRaw-method.html
index 81a9c7ba7..ec13665eb 100644
--- a/reference/findPeaks.matchedFilter-xcmsRaw-method.html
+++ b/reference/findPeaks.matchedFilter-xcmsRaw-method.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/findneutral.html b/reference/findneutral.html
index df23a98be..0ea0b3328 100644
--- a/reference/findneutral.html
+++ b/reference/findneutral.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getEIC-methods.html b/reference/getEIC-methods.html
index 4e288c138..22a7862f9 100644
--- a/reference/getEIC-methods.html
+++ b/reference/getEIC-methods.html
@@ -20,7 +20,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getPeaks-methods.html b/reference/getPeaks-methods.html
index a9bacd946..799f10c43 100644
--- a/reference/getPeaks-methods.html
+++ b/reference/getPeaks-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getScan-methods.html b/reference/getScan-methods.html
index dbfcd68cc..4a377634c 100644
--- a/reference/getScan-methods.html
+++ b/reference/getScan-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getSpec-methods.html b/reference/getSpec-methods.html
index 19769fc8c..f1ba85634 100644
--- a/reference/getSpec-methods.html
+++ b/reference/getSpec-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getXcmsRaw-methods.html b/reference/getXcmsRaw-methods.html
index 75f0480f3..d48b82853 100644
--- a/reference/getXcmsRaw-methods.html
+++ b/reference/getXcmsRaw-methods.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/group-methods.html b/reference/group-methods.html
index 3031d3bee..818ee49f9 100644
--- a/reference/group-methods.html
+++ b/reference/group-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/group.density.html b/reference/group.density.html
index 84321ba85..576431a86 100644
--- a/reference/group.density.html
+++ b/reference/group.density.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/group.mzClust.html b/reference/group.mzClust.html
index 8756e415d..806f2bcaa 100644
--- a/reference/group.mzClust.html
+++ b/reference/group.mzClust.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/group.nearest.html b/reference/group.nearest.html
index f19f611db..4713e8b53 100644
--- a/reference/group.nearest.html
+++ b/reference/group.nearest.html
@@ -24,7 +24,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/groupChromPeaks.html b/reference/groupChromPeaks.html
index 6b49bc1d6..addcca803 100644
--- a/reference/groupChromPeaks.html
+++ b/reference/groupChromPeaks.html
@@ -14,10 +14,14 @@
 Supported param objects are:
 PeakDensityParam: correspondence using the peak density method
 (Smith 2006) that groups chromatographic peaks along the retention time
-axis within slices of (partially overlapping) m/z ranges. All peaks (from
-the same or from different samples) with their apex position being close
-on the retention time axis are grouped into a LC-MS feature. See in
-addition do_groupChromPeaks_density() for the core API function.
+axis within slices of (partially overlapping) m/z ranges. By default,
+these m/z ranges (bins) have a constant size. By setting ppm to a value
+larger than 0, m/z dependent bin sizes can be used instead (better
+representing the m/z dependent measurement error of some MS instruments).
+All peaks (from the same or from different samples) with their apex
+position being close on the retention time axis are grouped into a LC-MS
+feature. See in addition do_groupChromPeaks_density() for the core API
+function.
 NearestPeaksParam: performs peak grouping based on the proximity of
 chromatographic peaks from different samples in the m/z - rt space similar
 to the correspondence method of mzMine (Katajamaa 2006). The method
@@ -50,10 +54,14 @@
 Supported param objects are:
 PeakDensityParam: correspondence using the peak density method
 (Smith 2006) that groups chromatographic peaks along the retention time
-axis within slices of (partially overlapping) m/z ranges. All peaks (from
-the same or from different samples) with their apex position being close
-on the retention time axis are grouped into a LC-MS feature. See in
-addition do_groupChromPeaks_density() for the core API function.
+axis within slices of (partially overlapping) m/z ranges. By default,
+these m/z ranges (bins) have a constant size. By setting ppm to a value
+larger than 0, m/z dependent bin sizes can be used instead (better
+representing the m/z dependent measurement error of some MS instruments).
+All peaks (from the same or from different samples) with their apex
+position being close on the retention time axis are grouped into a LC-MS
+feature. See in addition do_groupChromPeaks_density() for the core API
+function.
 NearestPeaksParam: performs peak grouping based on the proximity of
 chromatographic peaks from different samples in the m/z - rt space similar
 to the correspondence method of mzMine (Katajamaa 2006). The method
@@ -82,7 +90,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -140,10 +148,14 @@
 already present feature definitions.</p>
 <p>Supported <code>param</code> objects are:</p><ul><li><p><code>PeakDensityParam</code>: correspondence using the <em>peak density</em> method
 (Smith 2006) that groups chromatographic peaks along the retention time
-axis within slices of (partially overlapping) m/z ranges. All peaks (from
-the same or from different samples) with their apex position being close
-on the retention time axis are grouped into a LC-MS feature. See in
-addition <code><a href="do_groupChromPeaks_density.html">do_groupChromPeaks_density()</a></code> for the core API function.</p></li>
+axis within slices of (partially overlapping) m/z ranges. By default,
+these m/z ranges (bins) have a constant size. By setting <code>ppm</code> to a value
+larger than 0, m/z dependent bin sizes can be used instead (better
+representing the m/z dependent measurement error of some MS instruments).
+All peaks (from the same or from different samples) with their apex
+position being close on the retention time axis are grouped into a LC-MS
+feature. See in addition <code><a href="do_groupChromPeaks_density.html">do_groupChromPeaks_density()</a></code> for the core API
+function.</p></li>
 <li><p><code>NearestPeaksParam</code>: performs peak grouping based on the proximity of
 chromatographic peaks from different samples in the m/z - rt space similar
 to the correspondence method of <em>mzMine</em> (Katajamaa 2006). The method
@@ -175,6 +187,7 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
 <span>  minFraction <span class="op">=</span> <span class="fl">0.5</span>,</span>
 <span>  minSamples <span class="op">=</span> <span class="fl">1</span>,</span>
 <span>  binSize <span class="op">=</span> <span class="fl">0.25</span>,</span>
+<span>  ppm <span class="op">=</span> <span class="fl">0</span>,</span>
 <span>  maxFeatures <span class="op">=</span> <span class="fl">50</span></span>
 <span><span class="op">)</span></span>
 <span></span>
@@ -230,6 +243,9 @@ <h2 id="ref-usage">Usage<a class="anchor" aria-label="anchor" href="#ref-usage">
 <span><span class="co"># S4 method for PeakDensityParam</span></span>
 <span><span class="fu">maxFeatures</span><span class="op">(</span><span class="va">object</span><span class="op">)</span> <span class="op">&lt;-</span> <span class="va">value</span></span>
 <span></span>
+<span><span class="co"># S4 method for PeakDensityParam</span></span>
+<span><span class="fu"><a href="findChromPeaks-centWave.html">ppm</a></span><span class="op">(</span><span class="va">object</span><span class="op">)</span></span>
+<span></span>
 <span><span class="co"># S4 method for MzClustParam</span></span>
 <span><span class="fu">sampleGroups</span><span class="op">(</span><span class="va">object</span><span class="op">)</span></span>
 <span></span>
@@ -361,16 +377,24 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 overlapping slices in m/z dimension.</p></dd>
 
 
+<dt>ppm</dt>
+<dd><p>For <code>MzClustParam</code>: <code>numeric(1)</code> representing the relative m/z
+error for the clustering/grouping (in parts per million).
+For <code>PeakDensityParam</code>: <code>numeric(1)</code> to define m/z-dependent, increasing
+m/z bin sizes. If <code>ppm = 0</code> (the default) m/z bins are defined by the
+sequence of values from the smallest to the larges m/z value with a
+constant bin size of <code>binSize</code>. For <code>ppm</code> &gt; 0 the size of each bin is
+increased in addition by the <code>ppm</code> of the (upper) m/z boundary of the
+bin. The maximal bin size (used for the largest m/z values) would then
+be <code>binSize</code> plus <code>ppm</code> parts-per-million of the largest m/z value of
+all peaks in the data set.</p></dd>
+
+
 <dt>maxFeatures</dt>
 <dd><p>For <code>PeakDensityParam</code>: <code>numeric(1)</code> with the maximum
 number of peak groups to be identified in a single mz slice.</p></dd>
 
 
-<dt>ppm</dt>
-<dd><p>For <code>MzClustParam</code>: <code>numeric(1)</code> representing the relative m/z
-error for the clustering/grouping (in parts per million).</p></dd>
-
-
 <dt>absMz</dt>
 <dd><p>For <code>NearestPeaksParam</code> and <code>MzClustParam</code>: <code>numeric(1)</code>
 maximum tolerated distance for m/z values.</p></dd>
diff --git a/reference/groupFeatures-abundance-correlation.html b/reference/groupFeatures-abundance-correlation.html
index d0937f3ff..66f90101f 100644
--- a/reference/groupFeatures-abundance-correlation.html
+++ b/reference/groupFeatures-abundance-correlation.html
@@ -28,7 +28,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/groupFeatures-eic-similarity.html b/reference/groupFeatures-eic-similarity.html
index 621bf0fc6..4466355bc 100644
--- a/reference/groupFeatures-eic-similarity.html
+++ b/reference/groupFeatures-eic-similarity.html
@@ -78,7 +78,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -262,7 +262,12 @@ <h2 id="value">Value<a class="anchor" aria-label="anchor" href="#value"></a></h2
     </div>
     <div class="section level2">
     <h2 id="note">Note<a class="anchor" aria-label="anchor" href="#note"></a></h2>
-    <p>While being possible to be performed on the full data set without prior
+    <p>At present the <code><a href="featureChromatograms.html">featureChromatograms()</a></code> function is used to extract the
+EICs for each feature, which does however use one m/z and rt range for
+each feature and the EICs do thus not exactly represent the identified
+chromatographic peaks of each sample (i.e. their specific m/z and
+retention time ranges).</p>
+<p>While being possible to be performed on the full data set without prior
 feature grouping, this is not suggested for the following reasons: I) the
 selection of the top <code>n</code> samples with the highest signal for the
 <em>feature group</em> will be biased by very abundant compounds as this is
@@ -312,7 +317,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/pkg/BiocGenerics/man/table.html" class="external-link">table</a></span><span class="op">(</span><span class="fu"><a href="XCMSnExp-class.html">featureDefinitions</a></span><span class="op">(</span><span class="va">xodg_grp</span><span class="op">)</span><span class="op">$</span><span class="va">feature_group</span><span class="op">)</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011 </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>      2      2      2      4      3      2      3      2      1      1      1 </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>      2      2      2      3      4      2      3      2      1      1      1 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>      1      1      1      1      1      1      1      1      1      1      1 </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033 </span>
diff --git a/reference/groupFeatures-similar-rtime.html b/reference/groupFeatures-similar-rtime.html
index 397f4b3e8..367a4a1ab 100644
--- a/reference/groupFeatures-similar-rtime.html
+++ b/reference/groupFeatures-similar-rtime.html
@@ -22,7 +22,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/groupOverlaps.html b/reference/groupOverlaps.html
index bd66c1179..273b159ff 100644
--- a/reference/groupOverlaps.html
+++ b/reference/groupOverlaps.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/groupnames-XCMSnExp.html b/reference/groupnames-XCMSnExp.html
index 7ffb6fecc..d7df741e6 100644
--- a/reference/groupnames-XCMSnExp.html
+++ b/reference/groupnames-XCMSnExp.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/groupnames-methods.html b/reference/groupnames-methods.html
index cd9b70820..b04810b8c 100644
--- a/reference/groupnames-methods.html
+++ b/reference/groupnames-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/groupval-methods.html b/reference/groupval-methods.html
index b59e1daad..ebfbb4f75 100644
--- a/reference/groupval-methods.html
+++ b/reference/groupval-methods.html
@@ -20,7 +20,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/highlightChromPeaks.html b/reference/highlightChromPeaks.html
index 6641216bd..7242263a8 100644
--- a/reference/highlightChromPeaks.html
+++ b/reference/highlightChromPeaks.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/image.xcmsRaw.html b/reference/image.xcmsRaw.html
index ceb7a209f..67030688d 100644
--- a/reference/image.xcmsRaw.html
+++ b/reference/image.xcmsRaw.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/imputeLinInterpol.html b/reference/imputeLinInterpol.html
index 1c73fdecc..2526fed25 100644
--- a/reference/imputeLinInterpol.html
+++ b/reference/imputeLinInterpol.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/imputeRowMin.html b/reference/imputeRowMin.html
index ce387fd23..69d6a9981 100644
--- a/reference/imputeRowMin.html
+++ b/reference/imputeRowMin.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/imputeRowMinRand.html b/reference/imputeRowMinRand.html
index cb34a7548..575fcfc20 100644
--- a/reference/imputeRowMinRand.html
+++ b/reference/imputeRowMinRand.html
@@ -26,7 +26,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -191,35 +191,35 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  244584.5  161184.05   72029.38         NA  238194.4  173829.95</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517          NA         NA         NA  240261.21  201316.2  179437.72</span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="va">mat_imp_meansd</span><span class="op">)</span></span></span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>                 ko15      ko16       ko18       ko19       ko21       ko22</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  147887.5  451600.7   65290.38   40787.93   91635.45  162012.44</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 205/2791   1778568.9 1567038.1 1482796.38 1039129.82 1223132.35 1072037.70</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 206/2790    237993.6  269714.0  201393.42  150107.31  176989.65  156797.04</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 207.1/2719  380873.0  460629.7  351750.14  219287.97  286848.56  235022.63</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  235544.9  173623.4   35882.08   48149.57  185792.43  174458.77</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517    140725.6   84084.1  222609.07  286232.15  435094.49   81878.46</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>                 ko15       ko16       ko18       ko19       ko21       ko22</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  147887.5  451600.71   65290.38   49246.37   91635.45  162012.44</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 205/2791   1778568.9 1567038.14 1482796.38 1039129.82 1223132.35 1072037.70</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 206/2790    237993.6  269713.98  201393.42  150107.31  176989.65  156797.04</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 207.1/2719  380873.0  460629.74  351750.14  219287.97  286848.56  235022.63</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  235544.9  173623.38   19850.45   46023.31  185792.43  174458.77</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517    149786.4   79591.63  222609.07  286232.15  435094.49   59448.34</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>                 wt15       wt16       wt18       wt19      wt21       wt22</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  175177.1   82619.48   34637.57   69198.22  153273.5   98144.28</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  175177.1   82619.48   16968.88   69198.22  153273.5   98144.28</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> 205/2791   1950287.5 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> 206/2790    276541.8  222366.15  211717.71  186850.88  188285.9  172348.76</span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> 207.1/2719  417169.6  324892.46  277990.70  220972.35  252874.0  236728.16</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  244584.5  161184.05   72029.38   28146.13  238194.4  173829.95</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517    122405.0   86251.20   48322.54  240261.21  201316.2  179437.72</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  244584.5  161184.05   72029.38   43139.31  238194.4  173829.95</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517    120349.2  123319.72   95360.60  240261.21  201316.2  179437.72</span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/utils/head.html" class="external-link">head</a></span><span class="op">(</span><span class="va">mat_imp_fromto</span><span class="op">)</span></span></span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>                  ko15       ko16       ko18        ko19       ko21       ko22</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  147887.53  451600.71   65290.38    2778.231   91635.45  162012.44</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 205/2791   1778568.94 1567038.14 1482796.38 1039129.818 1223132.35 1072037.70</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 206/2790    237993.62  269713.98  201393.42  150107.310  176989.65  156797.04</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 207.1/2719  380873.05  460629.74  351750.14  219287.968  286848.56  235022.63</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  235544.92  173623.38   32793.36    2387.531  185792.43  174458.77</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517     38276.79   37741.04  222609.07  286232.146  435094.49   46964.75</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>                  wt15       wt16       wt18        wt19      wt21       wt22</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  175177.08   82619.48   31192.57   69198.222  153273.5   98144.28</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 205/2791   1950287.49 1466780.60 1572679.16 1275312.758 1356014.3 1231442.16</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 206/2790    276541.85  222366.15  211717.71  186850.878  188285.9  172348.76</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 207.1/2719  417169.58  324892.46  277990.70  220972.352  252874.0  236728.16</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  244584.47  161184.05   72029.38    4217.932  238194.4  173829.95</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517     44101.44   78784.93   14870.97  240261.212  201316.2  179437.72</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>                  ko15       ko16       ko18       ko19       ko21       ko22</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  147887.53  451600.71   65290.38   19617.55   91635.45  162012.44</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 205/2791   1778568.94 1567038.14 1482796.38 1039129.82 1223132.35 1072037.70</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 206/2790    237993.62  269713.98  201393.42  150107.31  176989.65  156797.04</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 207.1/2719  380873.05  460629.74  351750.14  219287.97  286848.56  235022.63</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  235544.92  173623.38   14992.43   21335.01  185792.43  174458.77</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517     42550.31   45065.53  222609.07  286232.15  435094.49   80160.63</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>                 wt15       wt16       wt18         wt19      wt21       wt22</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 200.1/2926  175177.1   82619.48   12629.94   69198.2221  153273.5   98144.28</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 205/2791   1950287.5 1466780.60 1572679.16 1275312.7582 1356014.3 1231442.16</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 206/2790    276541.8  222366.15  211717.71  186850.8779  188285.9  172348.76</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 207.1/2719  417169.6  324892.46  277990.70  220972.3520  252874.0  236728.16</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 219.1/2525  244584.5  161184.05   72029.38     391.5481  238194.4  173829.95</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> 231/2517     67198.8   34599.03   52982.49  240261.2115  201316.2  179437.72</span>
 </code></pre></div>
     </div>
   </main><aside class="col-md-3"><nav id="toc"><h2>On this page</h2>
diff --git a/reference/index.html b/reference/index.html
index c123e2755..7175341d3 100644
--- a/reference/index.html
+++ b/reference/index.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -454,7 +454,7 @@ <h2 id="all-functions">All functions<a class="anchor" aria-label="anchor" href="
         <dd>Group peaks from different samples together</dd>
       </dl><dl><dt>
           
-          <code><a href="groupChromPeaks.html">groupChromPeaks()</a></code> <code><a href="groupChromPeaks.html">PeakDensityParam()</a></code> <code><a href="groupChromPeaks.html">MzClustParam()</a></code> <code><a href="groupChromPeaks.html">NearestPeaksParam()</a></code> <code><a href="groupChromPeaks.html">sampleGroups(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`sampleGroups&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">bw(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`bw&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minFraction(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minFraction&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minSamples(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minSamples&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">binSize(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`binSize&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">maxFeatures(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`maxFeatures&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">sampleGroups(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`sampleGroups&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">ppm(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`ppm&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">absMz(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`absMz&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minFraction(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minFraction&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minSamples(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minSamples&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">sampleGroups(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`sampleGroups&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">mzVsRtBalance(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`mzVsRtBalance&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">absMz(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`absMz&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">absRt(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`absRt&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">kNN(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`kNN&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> 
+          <code><a href="groupChromPeaks.html">groupChromPeaks()</a></code> <code><a href="groupChromPeaks.html">PeakDensityParam()</a></code> <code><a href="groupChromPeaks.html">MzClustParam()</a></code> <code><a href="groupChromPeaks.html">NearestPeaksParam()</a></code> <code><a href="groupChromPeaks.html">sampleGroups(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`sampleGroups&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">bw(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`bw&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minFraction(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minFraction&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minSamples(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minSamples&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">binSize(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`binSize&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">maxFeatures(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`maxFeatures&lt;-`(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">ppm(<i>&lt;PeakDensityParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">sampleGroups(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`sampleGroups&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">ppm(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`ppm&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">absMz(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`absMz&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minFraction(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minFraction&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">minSamples(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`minSamples&lt;-`(<i>&lt;MzClustParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">sampleGroups(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`sampleGroups&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">mzVsRtBalance(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`mzVsRtBalance&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">absMz(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`absMz&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">absRt(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`absRt&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">kNN(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> <code><a href="groupChromPeaks.html">`kNN&lt;-`(<i>&lt;NearestPeaksParam&gt;</i>)</a></code> 
         </dt>
         <dd>Correspondence: group chromatographic peaks across samples</dd>
       </dl><dl><dt>
diff --git a/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html b/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html
index 54086055e..b6562f380 100644
--- a/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html
+++ b/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/levelplot.xcmsRaw.html b/reference/levelplot.xcmsRaw.html
index 2bc6a9eb2..939c9e659 100644
--- a/reference/levelplot.xcmsRaw.html
+++ b/reference/levelplot.xcmsRaw.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/loadRaw-methods.html b/reference/loadRaw-methods.html
index 803295be5..9bc880404 100644
--- a/reference/loadRaw-methods.html
+++ b/reference/loadRaw-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/loadXcmsData.html b/reference/loadXcmsData.html
index 0a089d308..d912536e3 100644
--- a/reference/loadXcmsData.html
+++ b/reference/loadXcmsData.html
@@ -50,7 +50,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/manualChromPeaks.html b/reference/manualChromPeaks.html
index c748194e6..d3b4d915f 100644
--- a/reference/manualChromPeaks.html
+++ b/reference/manualChromPeaks.html
@@ -30,7 +30,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/medianFilter.html b/reference/medianFilter.html
index e63b5e8d5..3bb90d2ce 100644
--- a/reference/medianFilter.html
+++ b/reference/medianFilter.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/msn2xcmsRaw.html b/reference/msn2xcmsRaw.html
index 4ee41ebb4..6d34e0545 100644
--- a/reference/msn2xcmsRaw.html
+++ b/reference/msn2xcmsRaw.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/na.flatfill.html b/reference/na.flatfill.html
index 7701b2321..1802e51d6 100644
--- a/reference/na.flatfill.html
+++ b/reference/na.flatfill.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/overlappingFeatures.html b/reference/overlappingFeatures.html
index fd49312ec..dd4f875a7 100644
--- a/reference/overlappingFeatures.html
+++ b/reference/overlappingFeatures.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/panel.cor.html b/reference/panel.cor.html
index f23607da7..df3f6dc76 100644
--- a/reference/panel.cor.html
+++ b/reference/panel.cor.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/peakPlots-methods.html b/reference/peakPlots-methods.html
index 2249db276..7bee20a9b 100644
--- a/reference/peakPlots-methods.html
+++ b/reference/peakPlots-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/peakTable-methods.html b/reference/peakTable-methods.html
index 05993163a..d72365f78 100644
--- a/reference/peakTable-methods.html
+++ b/reference/peakTable-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/peaksWithCentWave.html b/reference/peaksWithCentWave.html
index e864905e9..7b8a8b684 100644
--- a/reference/peaksWithCentWave.html
+++ b/reference/peaksWithCentWave.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/peaksWithMatchedFilter.html b/reference/peaksWithMatchedFilter.html
index 4c82816ad..b938c0482 100644
--- a/reference/peaksWithMatchedFilter.html
+++ b/reference/peaksWithMatchedFilter.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/phenoDataFromPaths.html b/reference/phenoDataFromPaths.html
index b2237b4fa..e44f54f4d 100644
--- a/reference/phenoDataFromPaths.html
+++ b/reference/phenoDataFromPaths.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plot.xcmsEIC.html b/reference/plot.xcmsEIC.html
index 98711ccbd..5ee8c1a4f 100644
--- a/reference/plot.xcmsEIC.html
+++ b/reference/plot.xcmsEIC.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotAdjustedRtime.html b/reference/plotAdjustedRtime.html
index 7e036de8f..a6728d35d 100644
--- a/reference/plotAdjustedRtime.html
+++ b/reference/plotAdjustedRtime.html
@@ -20,7 +20,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotChrom-methods.html b/reference/plotChrom-methods.html
index 37914c4f1..4a1c5a255 100644
--- a/reference/plotChrom-methods.html
+++ b/reference/plotChrom-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotChromPeakDensity.html b/reference/plotChromPeakDensity.html
index bbeafc153..d38507a5f 100644
--- a/reference/plotChromPeakDensity.html
+++ b/reference/plotChromPeakDensity.html
@@ -24,7 +24,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotChromPeaks.html b/reference/plotChromPeaks.html
index 94a9cdb56..e55bc5120 100644
--- a/reference/plotChromPeaks.html
+++ b/reference/plotChromPeaks.html
@@ -24,7 +24,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotChromatogramsOverlay.html b/reference/plotChromatogramsOverlay.html
index 9cbefcae7..8ce108dce 100644
--- a/reference/plotChromatogramsOverlay.html
+++ b/reference/plotChromatogramsOverlay.html
@@ -28,7 +28,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotEIC-methods.html b/reference/plotEIC-methods.html
index 723375e93..17d4f2e26 100644
--- a/reference/plotEIC-methods.html
+++ b/reference/plotEIC-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotFeatureGroups.html b/reference/plotFeatureGroups.html
index c535c0777..df47e9cfd 100644
--- a/reference/plotFeatureGroups.html
+++ b/reference/plotFeatureGroups.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotMsData.html b/reference/plotMsData.html
index fc45af292..a374c6dab 100644
--- a/reference/plotMsData.html
+++ b/reference/plotMsData.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotPeaks-methods.html b/reference/plotPeaks-methods.html
index a175594b6..462ee413a 100644
--- a/reference/plotPeaks-methods.html
+++ b/reference/plotPeaks-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotQC.html b/reference/plotQC.html
index e89f2805e..1cf163897 100644
--- a/reference/plotQC.html
+++ b/reference/plotQC.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotRaw-methods.html b/reference/plotRaw-methods.html
index 041a637d4..d8d4246ae 100644
--- a/reference/plotRaw-methods.html
+++ b/reference/plotRaw-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotScan-methods.html b/reference/plotScan-methods.html
index a05dee7c4..00b813aee 100644
--- a/reference/plotScan-methods.html
+++ b/reference/plotScan-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotSpec-methods.html b/reference/plotSpec-methods.html
index 7c8983dad..e1ee40558 100644
--- a/reference/plotSpec-methods.html
+++ b/reference/plotSpec-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotSurf-methods.html b/reference/plotSurf-methods.html
index 4ece67f21..a3c25f5c5 100644
--- a/reference/plotSurf-methods.html
+++ b/reference/plotSurf-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotTIC-methods.html b/reference/plotTIC-methods.html
index 28df2d196..eb1ba89be 100644
--- a/reference/plotTIC-methods.html
+++ b/reference/plotTIC-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plotrt-methods.html b/reference/plotrt-methods.html
index d8af23bf2..31bfa4aa3 100644
--- a/reference/plotrt-methods.html
+++ b/reference/plotrt-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/profGenerate.html b/reference/profGenerate.html
index e7537ba49..fb10af669 100644
--- a/reference/profGenerate.html
+++ b/reference/profGenerate.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/profMat-xcmsSet.html b/reference/profMat-xcmsSet.html
index c75777e48..1a22793ed 100644
--- a/reference/profMat-xcmsSet.html
+++ b/reference/profMat-xcmsSet.html
@@ -44,7 +44,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/profMedFilt-methods.html b/reference/profMedFilt-methods.html
index e68e83366..25cbdc1ed 100644
--- a/reference/profMedFilt-methods.html
+++ b/reference/profMedFilt-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/profMethod-methods.html b/reference/profMethod-methods.html
index 05d482c6b..0305c7c94 100644
--- a/reference/profMethod-methods.html
+++ b/reference/profMethod-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/profRange-methods.html b/reference/profRange-methods.html
index 16987b245..f65ef39e6 100644
--- a/reference/profRange-methods.html
+++ b/reference/profRange-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/profStep-methods.html b/reference/profStep-methods.html
index 95d508933..3a3d83ee2 100644
--- a/reference/profStep-methods.html
+++ b/reference/profStep-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/pval.html b/reference/pval.html
index 1daa073ae..9a0e3cf58 100644
--- a/reference/pval.html
+++ b/reference/pval.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/rawEIC-methods.html b/reference/rawEIC-methods.html
index 97a607b94..f97417bf9 100644
--- a/reference/rawEIC-methods.html
+++ b/reference/rawEIC-methods.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/rawMat-methods.html b/reference/rawMat-methods.html
index 858bffab8..db76bd81d 100644
--- a/reference/rawMat-methods.html
+++ b/reference/rawMat-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/reconstructChromPeakSpectra.html b/reference/reconstructChromPeakSpectra.html
index a5bc44042..cb68be40b 100644
--- a/reference/reconstructChromPeakSpectra.html
+++ b/reference/reconstructChromPeakSpectra.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/rectUnique.html b/reference/rectUnique.html
index a130f66e2..1c65484bb 100644
--- a/reference/rectUnique.html
+++ b/reference/rectUnique.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/refineChromPeaks.html b/reference/refineChromPeaks.html
index 6f8d8f8dd..132cdd9b5 100644
--- a/reference/refineChromPeaks.html
+++ b/reference/refineChromPeaks.html
@@ -68,7 +68,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -258,7 +258,8 @@ <h2 id="arguments">Arguments<a class="anchor" aria-label="anchor" href="#argumen
 <dt>ppm</dt>
 <dd><p>For <code>MergeNeighboringPeaksParam</code>: <code>numeric(1)</code> defining a m/z
 relative value (in parts per million) by which the m/z range of each
-chromatographic peak is expanded to check for overlapping peaks.</p></dd>
+chromatographic peak is expanded (on each side) to check for overlapping
+peaks.</p></dd>
 
 
 <dt>minProp</dt>
@@ -380,6 +381,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span><span class="co">## Example of a split peak that will be merged</span></span></span>
 <span class="r-in"><span><span class="va">mzr</span> <span class="op">&lt;-</span> <span class="fl">305.1</span> <span class="op">+</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="op">-</span><span class="fl">0.01</span>, <span class="fl">0.01</span><span class="op">)</span></span></span>
 <span class="r-in"><span><span class="va">chr</span> <span class="op">&lt;-</span> <span class="fu"><a href="chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">xd</span>, mz <span class="op">=</span> <span class="va">mzr</span>, rt <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">2700</span>, <span class="fl">3700</span><span class="op">)</span><span class="op">)</span></span></span>
+<span class="r-msg co"><span class="r-pr">#&gt;</span> Processing chromatographic peaks</span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr</span><span class="op">)</span></span></span>
 <span class="r-plt img"><img src="refineChromPeaks-1.png" alt="" width="700" height="433"></span>
 <span class="r-in"><span></span></span>
@@ -387,6 +389,7 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span><span class="va">res</span> <span class="op">&lt;-</span> <span class="fu">refineChromPeaks</span><span class="op">(</span><span class="va">xd</span>, param <span class="op">=</span> <span class="fu">MergeNeighboringPeaksParam</span><span class="op">(</span>expandRt <span class="op">=</span> <span class="fl">4</span><span class="op">)</span><span class="op">)</span></span></span>
 <span class="r-msg co"><span class="r-pr">#&gt;</span> Reduced from 87 to 70 chromatographic peaks.</span>
 <span class="r-in"><span><span class="va">chr_res</span> <span class="op">&lt;-</span> <span class="fu"><a href="chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">res</span>, mz <span class="op">=</span> <span class="va">mzr</span>, rt <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">2700</span>, <span class="fl">3700</span><span class="op">)</span><span class="op">)</span></span></span>
+<span class="r-msg co"><span class="r-pr">#&gt;</span> Processing chromatographic peaks</span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_res</span><span class="op">)</span></span></span>
 <span class="r-plt img"><img src="refineChromPeaks-2.png" alt="" width="700" height="433"></span>
 <span class="r-in"><span></span></span>
@@ -394,8 +397,10 @@ <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-e
 <span class="r-in"><span><span class="co">## is lower than the cut-off minProp</span></span></span>
 <span class="r-in"><span><span class="va">mzr</span> <span class="op">&lt;-</span> <span class="fl">496.2</span> <span class="op">+</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="op">-</span><span class="fl">0.01</span>, <span class="fl">0.01</span><span class="op">)</span></span></span>
 <span class="r-in"><span><span class="va">chr</span> <span class="op">&lt;-</span> <span class="fu"><a href="chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">xd</span>, mz <span class="op">=</span> <span class="va">mzr</span>, rt <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">3200</span>, <span class="fl">3500</span><span class="op">)</span><span class="op">)</span></span></span>
+<span class="r-msg co"><span class="r-pr">#&gt;</span> Processing chromatographic peaks</span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr</span><span class="op">)</span></span></span>
 <span class="r-in"><span><span class="va">chr_res</span> <span class="op">&lt;-</span> <span class="fu"><a href="chromatogram-method.html">chromatogram</a></span><span class="op">(</span><span class="va">res</span>, mz <span class="op">=</span> <span class="va">mzr</span>, rt <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">3200</span>, <span class="fl">3500</span><span class="op">)</span><span class="op">)</span></span></span>
+<span class="r-msg co"><span class="r-pr">#&gt;</span> Processing chromatographic peaks</span>
 <span class="r-in"><span><span class="fu"><a href="https://rdrr.io/r/base/plot.html" class="external-link">plot</a></span><span class="op">(</span><span class="va">chr_res</span><span class="op">)</span></span></span>
 <span class="r-plt img"><img src="refineChromPeaks-3.png" alt="" width="700" height="433"></span>
 </code></pre></div>
diff --git a/reference/removeIntensity-Chromatogram.html b/reference/removeIntensity-Chromatogram.html
index 1c93f797b..dcc1d2a0b 100644
--- a/reference/removeIntensity-Chromatogram.html
+++ b/reference/removeIntensity-Chromatogram.html
@@ -36,7 +36,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/retcor-methods.html b/reference/retcor-methods.html
index 35b62ddf9..c7ad8945e 100644
--- a/reference/retcor-methods.html
+++ b/reference/retcor-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/retcor.obiwarp-methods.html b/reference/retcor.obiwarp-methods.html
index e364d36a0..a91676410 100644
--- a/reference/retcor.obiwarp-methods.html
+++ b/reference/retcor.obiwarp-methods.html
@@ -24,7 +24,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/retcor.peakgroups-methods.html b/reference/retcor.peakgroups-methods.html
index 08fd76538..2cd2a3a95 100644
--- a/reference/retcor.peakgroups-methods.html
+++ b/reference/retcor.peakgroups-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/retexp.html b/reference/retexp.html
index f1361de07..4b140b9b8 100644
--- a/reference/retexp.html
+++ b/reference/retexp.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/rla.html b/reference/rla.html
index b07baadd0..dfcdcf688 100644
--- a/reference/rla.html
+++ b/reference/rla.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/sampnames-methods.html b/reference/sampnames-methods.html
index 7a1da9ef2..580a0efe8 100644
--- a/reference/sampnames-methods.html
+++ b/reference/sampnames-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/showError-xcmsSet-method.html b/reference/showError-xcmsSet-method.html
index a98cc6a3d..121cd52a2 100644
--- a/reference/showError-xcmsSet-method.html
+++ b/reference/showError-xcmsSet-method.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/specDist-methods.html b/reference/specDist-methods.html
index c36c01361..78518a99b 100644
--- a/reference/specDist-methods.html
+++ b/reference/specDist-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/specDist.cosine-methods.html b/reference/specDist.cosine-methods.html
index 4a84f921e..7c3b3c76b 100644
--- a/reference/specDist.cosine-methods.html
+++ b/reference/specDist.cosine-methods.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/specDist.meanMZmatch-methods.html b/reference/specDist.meanMZmatch-methods.html
index e0292b03f..7d8e81691 100644
--- a/reference/specDist.meanMZmatch-methods.html
+++ b/reference/specDist.meanMZmatch-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/specDist.peakCount-methods.html b/reference/specDist.peakCount-methods.html
index 45ba856e7..58ca525b2 100644
--- a/reference/specDist.peakCount-methods.html
+++ b/reference/specDist.peakCount-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/specNoise.html b/reference/specNoise.html
index 54205659e..56169ac66 100644
--- a/reference/specNoise.html
+++ b/reference/specNoise.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/specPeaks.html b/reference/specPeaks.html
index ddeedaab0..6f85e8111 100644
--- a/reference/specPeaks.html
+++ b/reference/specPeaks.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/split.xcmsRaw.html b/reference/split.xcmsRaw.html
index 085d61745..d5c2e919b 100644
--- a/reference/split.xcmsRaw.html
+++ b/reference/split.xcmsRaw.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/split.xcmsSet.html b/reference/split.xcmsSet.html
index 67cbc6794..d2100fd42 100644
--- a/reference/split.xcmsSet.html
+++ b/reference/split.xcmsSet.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/stitch-methods.html b/reference/stitch-methods.html
index 96792b258..59db97c49 100644
--- a/reference/stitch-methods.html
+++ b/reference/stitch-methods.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/storeResults.html b/reference/storeResults.html
index 7846ea6d0..1d973fbfc 100644
--- a/reference/storeResults.html
+++ b/reference/storeResults.html
@@ -34,7 +34,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html b/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html
index 81953ee2a..c5eb0ba88 100644
--- a/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html
+++ b/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/updateObject-xcmsSet-method.html b/reference/updateObject-xcmsSet-method.html
index 1ff26e434..ca3041af3 100644
--- a/reference/updateObject-xcmsSet-method.html
+++ b/reference/updateObject-xcmsSet-method.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/useOriginalCode.html b/reference/useOriginalCode.html
index 59a474164..c30a324b2 100644
--- a/reference/useOriginalCode.html
+++ b/reference/useOriginalCode.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/verify.mzQuantML.html b/reference/verify.mzQuantML.html
index 5517929d0..82236a571 100644
--- a/reference/verify.mzQuantML.html
+++ b/reference/verify.mzQuantML.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/write.cdf-methods.html b/reference/write.cdf-methods.html
index 108a8d27e..56514d498 100644
--- a/reference/write.cdf-methods.html
+++ b/reference/write.cdf-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/write.mzQuantML.html b/reference/write.mzQuantML.html
index c016e76e8..905c52e39 100644
--- a/reference/write.mzQuantML.html
+++ b/reference/write.mzQuantML.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/write.mzdata-methods.html b/reference/write.mzdata-methods.html
index 1ebde993f..c3a6b06ac 100644
--- a/reference/write.mzdata-methods.html
+++ b/reference/write.mzdata-methods.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/writeMSData-XCMSnExp-character-method.html b/reference/writeMSData-XCMSnExp-character-method.html
index 26bb83e13..1b440b675 100644
--- a/reference/writeMSData-XCMSnExp-character-method.html
+++ b/reference/writeMSData-XCMSnExp-character-method.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/writeMzTab.html b/reference/writeMzTab.html
index ed078610b..6783e5105 100644
--- a/reference/writeMzTab.html
+++ b/reference/writeMzTab.html
@@ -10,7 +10,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcms-deprecated.html b/reference/xcms-deprecated.html
index e9fd0f2b6..32b9ed428 100644
--- a/reference/xcms-deprecated.html
+++ b/reference/xcms-deprecated.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsEIC-class.html b/reference/xcmsEIC-class.html
index c09f5743c..ce9eb36c0 100644
--- a/reference/xcmsEIC-class.html
+++ b/reference/xcmsEIC-class.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsFileSource-class.html b/reference/xcmsFileSource-class.html
index ff0d32a74..5e913522a 100644
--- a/reference/xcmsFileSource-class.html
+++ b/reference/xcmsFileSource-class.html
@@ -28,7 +28,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsFragments-class.html b/reference/xcmsFragments-class.html
index 68f633916..ad96f884c 100644
--- a/reference/xcmsFragments-class.html
+++ b/reference/xcmsFragments-class.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsFragments.html b/reference/xcmsFragments.html
index 23bea38ad..0afdaf556 100644
--- a/reference/xcmsFragments.html
+++ b/reference/xcmsFragments.html
@@ -20,7 +20,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsPeaks-class.html b/reference/xcmsPeaks-class.html
index 2bcabde1c..3eed3bfad 100644
--- a/reference/xcmsPeaks-class.html
+++ b/reference/xcmsPeaks-class.html
@@ -12,7 +12,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsRaw-class.html b/reference/xcmsRaw-class.html
index 4d1da7ffb..5077a3903 100644
--- a/reference/xcmsRaw-class.html
+++ b/reference/xcmsRaw-class.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsRaw.html b/reference/xcmsRaw.html
index 341df66cd..572e46d78 100644
--- a/reference/xcmsRaw.html
+++ b/reference/xcmsRaw.html
@@ -16,7 +16,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsSet-class.html b/reference/xcmsSet-class.html
index 06e07ab5a..679ade83e 100644
--- a/reference/xcmsSet-class.html
+++ b/reference/xcmsSet-class.html
@@ -18,7 +18,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsSet.html b/reference/xcmsSet.html
index 51aff15f2..edfa6df2c 100644
--- a/reference/xcmsSet.html
+++ b/reference/xcmsSet.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsSource-class.html b/reference/xcmsSource-class.html
index fb6928f57..ee960af2f 100644
--- a/reference/xcmsSource-class.html
+++ b/reference/xcmsSource-class.html
@@ -28,7 +28,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/xcmsSource-methods.html b/reference/xcmsSource-methods.html
index c74e49be8..31a5ab246 100644
--- a/reference/xcmsSource-methods.html
+++ b/reference/xcmsSource-methods.html
@@ -14,7 +14,7 @@
     
     <a class="navbar-brand me-2" href="../index.html">xcms</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.4</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">4.1.6</small>
 
     
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/search.json b/search.json
index 1360557d3..472615edc 100644
--- a/search.json
+++ b/search.json
@@ -1 +1 @@
-[{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"Compounding (grouping) of LC-MS features","text":"typical LC-MS-based metabolomics experiment compounds eluting chromatography first ionized measured mass spectrometry (MS). ionization different (multiple) ions can generated compound measured MS. general, resulting data pre-processed identify chromatographic peaks data group across samples correspondence analysis. result distinct LC-MS features, characterized specific m/z retention time range. Different ions generated ionization detected different features. Compounding aims now grouping features presumably representing signal originating compound reduce data set complexity (aid subsequent annotation steps). General MS feature grouping functionality defined MsFeatures package additional functionality implemented xcms package enable compounding LC-MS data. document provides simple compounding workflow using xcms. Note present functionality (yet) aggregate combine actual features per values, define feature groups (one per compound).","code":""},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"compounding-of-lc-ms-data","dir":"Articles","previous_headings":"","what":"Compounding of LC-MS data","title":"Compounding (grouping) of LC-MS features","text":"demonstrate compounding (feature grouping) functionality simple toy data set used also xcms package provided faahKO package. data set consists samples 4 mice knock-fatty acid amide hydrolase (FAAH) 4 wild type mice. Pre-processing data set described detail main vignette xcms package. load required packages result pre-processing updating also location respective raw data files current machine. performing feature grouping inspect result object. featureDefinitions can extract results correspondence analysis. row data frame represents definition one feature, average range m/z retention time. Column \"peakidx\" provides index chromatographic peak assigned feature chromPeaks matrix result object. featureValues function allows extract feature values, .e. matrix feature abundances, one row per feature columns representing samples present data set. extract feature values without filled-peak data. Without gap-filled data abundances detected chromatographic peaks reported. gap-filled data, samples chromatographic peak feature detected, signal m/z - retention time range defined based detected chromatographic peaks integrated. total 351 features defined present data set, many likely represent signal different ions (adducts isotopes) compound. aim grouping functions now define features likely come original compound. feature grouping functions base following assumptions/properties LC-MS data: Features (ions) compound similar retention time. abundance features (ions) compound similar pattern across samples, .e. compound highly concentrated one sample low another, ions follow pattern. peak shape extracted ion chromatograms (EIC) features compound similar follow elution pattern original compound LC. main method perform feature grouping called groupFeatures takes xcms result object (.e., XcmsExperiment XCMSnExp) input well parameter object chose grouping algorithm specify settings. xcms provides supports following grouping approaches: SimilarRtimeParam: perform initial grouping based similar retention time. AbundanceSimilarityParam: perform feature grouping based correlation feature abundances (values) across samples. EicSimilarityParam: perform feature grouping based correlation EICs. Calling groupFeatures xcms result object perform feature grouping assigning feature data set feature group. feature groups stored additional column called \"feature_group\" featureDefinition data frame result object can accessed featureGroups function. subsequent groupFeature call sub-group (refine) identified feature groups . thus possible use single grouping approach, combine multiple generate desired feature grouping incremental fashion. individual feature grouping algorithms can called order, advisable use EicSimilarityParam last refinement step, computationally expensive, especially applied result object without pre-defined feature groups feature groups large.","code":"library(xcms) library(faahKO) library(MsFeatures)  xmse <- loadXcmsData(\"xmse\") featureDefinitions(xmse) |> head() ##       mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks KO WT      peakidx ## FT001 200.1 200.1 200.1 2902.177 2882.133 2922.222      2  2  0 458, 116.... ## FT002 205.0 205.0 205.0 2789.457 2782.376 2795.823      8  4  4 44, 443,.... ## FT003 206.0 206.0 206.0 2788.770 2780.807 2793.606      7  3  4 29, 430,.... ## FT004 207.1 207.1 207.1 2718.476 2713.314 2726.487      7  4  3 16, 420,.... ## FT005 233.0 233.0 233.1 3023.353 3014.971 3043.942      7  3  4 69, 959,.... ## FT006 241.1 241.1 241.2 3680.200 3661.067 3695.533      8  3  4 276, 284.... ##       ms_level ## FT001        1 ## FT002        1 ## FT003        1 ## FT004        1 ## FT005        1 ## FT006        1 head(featureValues(xmse, filled = FALSE)) ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001        NA  506848.9        NA  169955.6        NA        NA        NA ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7        NA  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9        NA ## FT005  286221.4        NA  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5        NA  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001        NA ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 head(featureValues(xmse, filled = TRUE)) ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001  135422.4  506848.9  111872.0  169955.6  210333.1  141880.3  233271.3 ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7  167516.7  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  219142.3 ## FT005  286221.4  286594.4  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5 1145753.0  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001  142254.8 ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-features-by-similar-retention-time","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of features by similar retention time","title":"Compounding (grouping) of LC-MS features","text":"intuitive simple way group features based retention time. perform initial grouping evaluate retention times m/z features present data set. Plot retention times m/z features data set. Several features retention time (different m/z) can spotted, especially beginning LC. thus group features within retention time window 10 seconds feature groups. results feature grouping can accessed featureGroups function. determine size feature groups (.e. many features grouped together). addition visualize feature groups plotFeatureGroups function shows features m/z - retention time space grouped features connected line. Feature groups defined rt window 10 seconds Let’s assume don’t agree feature grouping, also knowing quite large shifts retention times runs. thus first remove defined feature groups assigning value NULL re-perform feature grouping using larger rt window. Feature groups defined rt window 20 seconds Grouping similar retention time grouped total 351 features 117 feature groups.","code":"plot(featureDefinitions(xmse)$rtmed, featureDefinitions(xmse)$mzmed,      xlab = \"retention time\", ylab = \"m/z\", main = \"features\",      col = \"#00000080\", pch = 21, bg = \"#00000040\") grid() xmse <- groupFeatures(xmse, param = SimilarRtimeParam(10)) table(featureGroups(xmse)) ##  ## FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  ##      4      5      3      4      2      4      4      4      4      3      3  ## FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  ##      5      6      2      4      3      2      2      3      2      3      4  ## FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  ##      3      8      6      3      3      3      3      2      3      3      4  ## FG.034 FG.035 FG.036 FG.037 FG.038 FG.039 FG.040 FG.041 FG.042 FG.043 FG.044  ##      2      3      5      3      3      4      5      2      4      2      4  ## FG.045 FG.046 FG.047 FG.048 FG.049 FG.050 FG.051 FG.052 FG.053 FG.054 FG.055  ##      6      3      3      3      2      2      2      4      3      3      3  ## FG.056 FG.057 FG.058 FG.059 FG.060 FG.061 FG.062 FG.063 FG.064 FG.065 FG.066  ##      2      3      5      4      2      4      2      3      2      3      3  ## FG.067 FG.068 FG.069 FG.070 FG.071 FG.072 FG.073 FG.074 FG.075 FG.076 FG.077  ##      3      3      3      3      3      2      4      2      2      2      2  ## FG.078 FG.079 FG.080 FG.081 FG.082 FG.083 FG.084 FG.085 FG.086 FG.087 FG.088  ##      2      3      3      3      3      3      2      3      4      2      3  ## FG.089 FG.090 FG.091 FG.092 FG.093 FG.094 FG.095 FG.096 FG.097 FG.098 FG.099  ##      2      3      2      2      2      3      2      3      3      2      2  ## FG.100 FG.101 FG.102 FG.103 FG.104 FG.105 FG.106 FG.107 FG.108 FG.109 FG.110  ##      2      2      2      2      2      2      2      2      2      2      2  ## FG.111 FG.112 FG.113 FG.114 FG.115 FG.116 FG.117 FG.118 FG.119 FG.120 FG.121  ##      1      1      1      1      1      1      1      1      1      1      1  ## FG.122 FG.123 FG.124 FG.125 FG.126 FG.127 FG.128 FG.129 FG.130 FG.131 FG.132  ##      1      1      1      1      1      1      1      1      1      1      1 plotFeatureGroups(xmse, pch = 21, lwd = 2, col = \"#00000040\",                   bg = \"#00000020\") grid() ## Remove previous feature grouping results to repeat the rtime-based ## feature grouping with different setting featureDefinitions(xmse)$feature_group <- NULL  ## Repeat the grouping xmse <- groupFeatures(xmse, SimilarRtimeParam(20)) table(featureGroups(xmse)) ##  ## FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  ##      4      5      3      4      2      4      4      4      4      3      3  ## FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  ##      5      6      2      4      3      2      2      3      2      3      4  ## FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  ##      3      8      6      3      3      3      3      2      3      3      4  ## FG.034 FG.035 FG.036 FG.037 FG.038 FG.039 FG.040 FG.041 FG.042 FG.043 FG.044  ##      2      3      5      3      3      4      6      2      4      2      5  ## FG.045 FG.046 FG.047 FG.048 FG.049 FG.050 FG.051 FG.052 FG.053 FG.054 FG.055  ##      6      3      4      3      2      2      2      4      3      3      3  ## FG.056 FG.057 FG.058 FG.059 FG.060 FG.061 FG.062 FG.063 FG.064 FG.065 FG.066  ##      3      3      5      4      2      4      2      3      2      3      3  ## FG.067 FG.068 FG.069 FG.070 FG.071 FG.072 FG.073 FG.074 FG.075 FG.076 FG.077  ##      4      3      3      3      4      2      5      2      2      2      2  ## FG.078 FG.079 FG.080 FG.081 FG.082 FG.083 FG.084 FG.085 FG.086 FG.087 FG.088  ##      2      3      3      3      3      3      2      3      4      3      4  ## FG.089 FG.090 FG.091 FG.092 FG.093 FG.094 FG.095 FG.096 FG.097 FG.098 FG.099  ##      3      4      2      2      2      4      2      3      3      2      2  ## FG.100 FG.101 FG.102 FG.103 FG.104 FG.105 FG.106 FG.107 FG.108 FG.109 FG.110  ##      2      3      2      2      2      3      3      2      2      2      2  ## FG.111 FG.112 FG.113 FG.114 FG.115 FG.116 FG.117  ##      1      1      1      1      1      1      1 plotFeatureGroups(xmse, pch = 21, lwd = 2, col = \"#00000040\", bg = \"#00000020\") grid()"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-features-by-abundance-correlation-across-samples","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of features by abundance correlation across samples","title":"Compounding (grouping) of LC-MS features","text":"Assuming OK crude initial feature grouping previous section, can next refine feature groups considering also feature abundances across samples. can use groupFeatures method AbundanceSimilarityParam object. approach performs pairwise correlation (non-missing) feature values (abundances; across samples) features predefined feature group (defined previous section). Features correlation >= threshold grouped together. Feature grouping based approach works best features higher variability concentration across samples. Parameter subset allows restrict analysis subset samples allows thus e.g. exclude QC sample pools correlation artificially increase correlation. parameters passed directly internal featureValues call extracts feature values correlation performed. performing grouping also evaluate correlation features based (log2 transformed) abundances across samples heatmap. Correlation features based feature abundances. large correlations can observed several groups features, many within feature group defined previous section (.e. eluting time). might thus represent correlated metabolites, ions adducts metabolite. use groupFeatures AbundanceSimilarityParam group features correlation coefficient higher 0.7 including detected filled-signal. Whether filled-detected signal used correlation analysis evaluated data set data set. specifying transform = log2 tell function log2 transform abundance prior correlation analysis. See help page groupFeatures AbundanceSimilarityParam xcms package details options. Many larger retention time-based feature groups splitted two sub-groups based correlation feature abundances (e.g. feature group FG.004 split feature groups FG.004.001, FG.004.002 FG.004.003). evaluate refinement feature group \"FG.040\" plotting pairwise correlation. better visualize feature grouping addition define custom panel plot pairs function plots data points blue features correlation coefficient selected threshold (0.7) red otherwise. Pairwise correlation plot features initially grouped feature group FG.040. Indeed, correlation seems present features retention time feature group, e.g. clearly FT273 FT274 also FT143 FT273. Note however abundance correlation suffers relatively samples (8 total), relatively small variance abundances across samples. feature grouping abundance correlation, 351 features grouped 253 feature groups.","code":"library(pheatmap) fvals <- log2(featureValues(xmse, filled = TRUE))  cormat <- cor(t(fvals), use = \"pairwise.complete.obs\") ann <- data.frame(fgroup = featureGroups(xmse)) rownames(ann) <- rownames(cormat)  res <- pheatmap(cormat, annotation_row = ann, cluster_rows = TRUE,                 cluster_cols = TRUE) xmse <- groupFeatures(     xmse, AbundanceSimilarityParam(threshold = 0.7, transform = log2),     filled = TRUE) table(featureGroups(xmse)) ##  ## FG.001.001 FG.001.002 FG.002.001 FG.002.002 FG.003.001 FG.003.002 FG.003.003  ##          3          1          3          2          1          1          1  ## FG.004.001 FG.004.002 FG.004.003 FG.005.001 FG.005.002 FG.006.001 FG.006.002  ##          2          1          1          1          1          2          1  ## FG.006.003 FG.007.001 FG.007.002 FG.008.001 FG.008.002 FG.008.003 FG.008.004  ##          1          3          1          1          1          1          1  ## FG.009.001 FG.009.002 FG.010.001 FG.010.002 FG.011.001 FG.011.002 FG.012.001  ##          3          1          2          1          2          1          2  ## FG.012.002 FG.012.003 FG.012.004 FG.013.001 FG.013.002 FG.013.003 FG.013.004  ##          1          1          1          3          1          1          1  ## FG.014.001 FG.015.001 FG.015.002 FG.016.001 FG.017.001 FG.017.002 FG.018.001  ##          2          3          1          3          1          1          1  ## FG.018.002 FG.019.001 FG.019.002 FG.020.001 FG.020.002 FG.021.001 FG.021.002  ##          1          2          1          1          1          1          1  ## FG.021.003 FG.022.001 FG.022.002 FG.022.003 FG.022.004 FG.023.001 FG.023.002  ##          1          1          1          1          1          1          1  ## FG.023.003 FG.024.001 FG.024.002 FG.024.003 FG.024.004 FG.024.005 FG.024.006  ##          1          2          2          1          1          1          1  ## FG.025.001 FG.026.001 FG.026.002 FG.027.001 FG.027.002 FG.028.001 FG.029.001  ##          6          2          1          2          1          3          2  ## FG.029.002 FG.030.001 FG.030.002 FG.031.001 FG.031.002 FG.032.001 FG.032.002  ##          1          1          1          2          1          1          1  ## FG.032.003 FG.033.001 FG.033.002 FG.033.003 FG.034.001 FG.034.002 FG.035.001  ##          1          2          1          1          1          1          2  ## FG.035.002 FG.036.001 FG.036.002 FG.036.003 FG.037.001 FG.037.002 FG.037.003  ##          1          2          2          1          1          1          1  ## FG.038.001 FG.038.002 FG.038.003 FG.039.001 FG.039.002 FG.039.003 FG.039.004  ##          1          1          1          1          1          1          1  ## FG.040.001 FG.040.002 FG.040.003 FG.041.001 FG.041.002 FG.042.001 FG.042.002  ##          4          1          1          1          1          2          1  ## FG.042.003 FG.043.001 FG.044.001 FG.044.002 FG.045.001 FG.045.002 FG.046.001  ##          1          2          3          2          4          2          2  ## FG.046.002 FG.047.001 FG.047.002 FG.047.003 FG.047.004 FG.048.001 FG.048.002  ##          1          1          1          1          1          2          1  ## FG.049.001 FG.049.002 FG.050.001 FG.050.002 FG.051.001 FG.051.002 FG.052.001  ##          1          1          1          1          1          1          1  ## FG.052.002 FG.052.003 FG.052.004 FG.053.001 FG.053.002 FG.053.003 FG.054.001  ##          1          1          1          1          1          1          1  ## FG.054.002 FG.054.003 FG.055.001 FG.055.002 FG.056.001 FG.056.002 FG.057.001  ##          1          1          2          1          2          1          2  ## FG.057.002 FG.058.001 FG.058.002 FG.059.001 FG.059.002 FG.059.003 FG.060.001  ##          1          4          1          2          1          1          1  ## FG.060.002 FG.061.001 FG.061.002 FG.062.001 FG.062.002 FG.063.001 FG.063.002  ##          1          2          2          1          1          1          1  ## FG.063.003 FG.064.001 FG.065.001 FG.065.002 FG.065.003 FG.066.001 FG.066.002  ##          1          2          1          1          1          2          1  ## FG.067.001 FG.067.002 FG.067.003 FG.068.001 FG.068.002 FG.069.001 FG.069.002  ##          2          1          1          2          1          2          1  ## FG.070.001 FG.070.002 FG.071.001 FG.071.002 FG.072.001 FG.073.001 FG.073.002  ##          2          1          3          1          2          2          2  ## FG.073.003 FG.074.001 FG.074.002 FG.075.001 FG.076.001 FG.076.002 FG.077.001  ##          1          1          1          2          1          1          2  ## FG.078.001 FG.078.002 FG.079.001 FG.079.002 FG.079.003 FG.080.001 FG.080.002  ##          1          1          1          1          1          2          1  ## FG.081.001 FG.081.002 FG.082.001 FG.082.002 FG.083.001 FG.083.002 FG.083.003  ##          2          1          2          1          1          1          1  ## FG.084.001 FG.084.002 FG.085.001 FG.085.002 FG.085.003 FG.086.001 FG.086.002  ##          1          1          1          1          1          2          1  ## FG.086.003 FG.087.001 FG.088.001 FG.088.002 FG.088.003 FG.089.001 FG.089.002  ##          1          3          2          1          1          2          1  ## FG.090.001 FG.090.002 FG.091.001 FG.091.002 FG.092.001 FG.092.002 FG.093.001  ##          2          2          1          1          1          1          1  ## FG.093.002 FG.094.001 FG.094.002 FG.094.003 FG.095.001 FG.096.001 FG.096.002  ##          1          2          1          1          2          2          1  ## FG.097.001 FG.098.001 FG.098.002 FG.099.001 FG.100.001 FG.100.002 FG.101.001  ##          3          1          1          2          1          1          2  ## FG.101.002 FG.102.001 FG.102.002 FG.103.001 FG.104.001 FG.105.001 FG.105.002  ##          1          1          1          2          2          2          1  ## FG.106.001 FG.106.002 FG.107.001 FG.107.002 FG.108.001 FG.109.001 FG.109.002  ##          2          1          1          1          2          1          1  ## FG.110.001 FG.111.001 FG.112.001 FG.113.001 FG.114.001 FG.115.001 FG.116.001  ##          2          1          1          1          1          1          1  ## FG.117.001  ##          1 cor_plot <- function(x, y) {     C <- cor(x, y, use = \"pairwise.complete.obs\")     col <- ifelse(C >= 0.7, yes = \"#0000ff80\", no = \"#ff000080\")     points(x, y, pch = 16, col = col)     grid() } fts <- grep(\"FG.040\", featureGroups(xmse)) pairs(t(fvals[fts, ]), gap = 0.1, main = \"FG.040\", panel = cor_plot)"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-features-by-similarity-of-their-eics","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of features by similarity of their EICs","title":"Compounding (grouping) of LC-MS features","text":"chromatographic peak shape ion compound highly similar elution pattern compound. Thus, features compound assumed similar peak shapes EICs within sample. grouping features based similarity EICs can performed groupFeatures EicSimilarityParam object. advisable perform peak shape correlation subset samples (peak shape correlation computationally intense chromatographic peaks low intensity features notoriously noisy). EicSimilarityParam approach allows select number top n samples (ordered total intensity feature abundances per feature group) correlation performed parameter n. n =3, 3 samples highest signal features respective feature group first identified pairwise similarity calculation performed EICs samples. resulting similarity score 3 samples aggregated single score taking 75% quantile across 3 samples. value subsequently compared cut-similarity (parameter threshold) features score >= threshold grouped feature group. group features based similarity EICs two samples highest total signal respective feature groups. default, Pearson correlation coefficient used similarity score similarity/distance metric function used instead (parameter FUN EicSimilarityParam - see respective help page ?EicSimilarityParam details options). define threshold correlation coefficient 0.7. computationally intense approach since involves also loading raw MS data extract ion chromatograms feature. results grouping shown . many cases, pre-defined feature groups (retention time similarity abundance correlation) subdivided. evaluate feature groups, starting FG.008.001 split two different feature groups based EIC correlation. first extract EICs features initial feature group. n = 1 specify extract EIC sample highest intensity. Next plot EICs using different color subgroups. peakType = \"none\" disable highlighting detected chromatographic peaks. Feature EICs per sample features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. Feature EICs per sample normalized absolute intensity 1 features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. One features (highlighted red plots ) within original feature group separated two low similarity EICs. fact, feature’s EIC shifted samples along retention time dimension can thus represent signal compound. evaluate next sub-grouping another feature group. Next plot EICs using different color subgroups. Feature EICs per sample features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. Feature EICs per sample normalized absolute intensity 1 features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. Based EIC correlation, initial feature group FG.045.001 grouped two separate sub-groups. grouping based EIC correlation pre-defined feature groups previous sections grouped 351 features 272 feature groups.","code":"xmse <- groupFeatures(xmse, EicSimilarityParam(threshold = 0.7, n = 2)) table(featureGroups(xmse)) ##  ## FG.001.001.001 FG.001.001.002 FG.001.002.001 FG.002.001.001 FG.002.002.001  ##              2              1              1              3              1  ## FG.002.002.002 FG.003.001.001 FG.003.002.001 FG.003.003.001 FG.004.001.001  ##              1              1              1              1              2  ## FG.004.002.001 FG.004.003.001 FG.005.001.001 FG.005.002.001 FG.006.001.001  ##              1              1              1              1              2  ## FG.006.002.001 FG.006.003.001 FG.007.001.001 FG.007.002.001 FG.008.001.001  ##              1              1              3              1              1  ## FG.008.002.001 FG.008.003.001 FG.008.004.001 FG.009.001.001 FG.009.002.001  ##              1              1              1              3              1  ## FG.010.001.001 FG.010.002.001 FG.011.001.001 FG.011.002.001 FG.012.001.001  ##              2              1              2              1              2  ## FG.012.002.001 FG.012.003.001 FG.012.004.001 FG.013.001.001 FG.013.001.002  ##              1              1              1              2              1  ## FG.013.002.001 FG.013.003.001 FG.013.004.001 FG.014.001.001 FG.015.001.001  ##              1              1              1              2              3  ## FG.015.002.001 FG.016.001.001 FG.017.001.001 FG.017.002.001 FG.018.001.001  ##              1              3              1              1              1  ## FG.018.002.001 FG.019.001.001 FG.019.002.001 FG.020.001.001 FG.020.002.001  ##              1              2              1              1              1  ## FG.021.001.001 FG.021.002.001 FG.021.003.001 FG.022.001.001 FG.022.002.001  ##              1              1              1              1              1  ## FG.022.003.001 FG.022.004.001 FG.023.001.001 FG.023.002.001 FG.023.003.001  ##              1              1              1              1              1  ## FG.024.001.001 FG.024.002.001 FG.024.003.001 FG.024.004.001 FG.024.005.001  ##              2              2              1              1              1  ## FG.024.006.001 FG.025.001.001 FG.026.001.001 FG.026.002.001 FG.027.001.001  ##              1              6              2              1              2  ## FG.027.002.001 FG.028.001.001 FG.029.001.001 FG.029.002.001 FG.030.001.001  ##              1              3              2              1              1  ## FG.030.002.001 FG.031.001.001 FG.031.002.001 FG.032.001.001 FG.032.002.001  ##              1              2              1              1              1  ## FG.032.003.001 FG.033.001.001 FG.033.002.001 FG.033.003.001 FG.034.001.001  ##              1              2              1              1              1  ## FG.034.002.001 FG.035.001.001 FG.035.002.001 FG.036.001.001 FG.036.001.002  ##              1              2              1              1              1  ## FG.036.002.001 FG.036.003.001 FG.037.001.001 FG.037.002.001 FG.037.003.001  ##              2              1              1              1              1  ## FG.038.001.001 FG.038.002.001 FG.038.003.001 FG.039.001.001 FG.039.002.001  ##              1              1              1              1              1  ## FG.039.003.001 FG.039.004.001 FG.040.001.001 FG.040.002.001 FG.040.003.001  ##              1              1              4              1              1  ## FG.041.001.001 FG.041.002.001 FG.042.001.001 FG.042.002.001 FG.042.003.001  ##              1              1              2              1              1  ## FG.043.001.001 FG.044.001.001 FG.044.001.002 FG.044.002.001 FG.044.002.002  ##              2              2              1              1              1  ## FG.045.001.001 FG.045.001.002 FG.045.002.001 FG.046.001.001 FG.046.002.001  ##              2              2              2              2              1  ## FG.047.001.001 FG.047.002.001 FG.047.003.001 FG.047.004.001 FG.048.001.001  ##              1              1              1              1              2  ## FG.048.002.001 FG.049.001.001 FG.049.002.001 FG.050.001.001 FG.050.002.001  ##              1              1              1              1              1  ## FG.051.001.001 FG.051.002.001 FG.052.001.001 FG.052.002.001 FG.052.003.001  ##              1              1              1              1              1  ## FG.052.004.001 FG.053.001.001 FG.053.002.001 FG.053.003.001 FG.054.001.001  ##              1              1              1              1              1  ## FG.054.002.001 FG.054.003.001 FG.055.001.001 FG.055.002.001 FG.056.001.001  ##              1              1              2              1              1  ## FG.056.001.002 FG.056.002.001 FG.057.001.001 FG.057.001.002 FG.057.002.001  ##              1              1              1              1              1  ## FG.058.001.001 FG.058.001.002 FG.058.002.001 FG.059.001.001 FG.059.002.001  ##              3              1              1              2              1  ## FG.059.003.001 FG.060.001.001 FG.060.002.001 FG.061.001.001 FG.061.002.001  ##              1              1              1              2              2  ## FG.062.001.001 FG.062.002.001 FG.063.001.001 FG.063.002.001 FG.063.003.001  ##              1              1              1              1              1  ## FG.064.001.001 FG.065.001.001 FG.065.002.001 FG.065.003.001 FG.066.001.001  ##              2              1              1              1              2  ## FG.066.002.001 FG.067.001.001 FG.067.001.002 FG.067.002.001 FG.067.003.001  ##              1              1              1              1              1  ## FG.068.001.001 FG.068.002.001 FG.069.001.001 FG.069.001.002 FG.069.002.001  ##              2              1              1              1              1  ## FG.070.001.001 FG.070.002.001 FG.071.001.001 FG.071.002.001 FG.072.001.001  ##              2              1              3              1              2  ## FG.073.001.001 FG.073.002.001 FG.073.003.001 FG.074.001.001 FG.074.002.001  ##              2              2              1              1              1  ## FG.075.001.001 FG.076.001.001 FG.076.002.001 FG.077.001.001 FG.078.001.001  ##              2              1              1              2              1  ## FG.078.002.001 FG.079.001.001 FG.079.002.001 FG.079.003.001 FG.080.001.001  ##              1              1              1              1              2  ## FG.080.002.001 FG.081.001.001 FG.081.002.001 FG.082.001.001 FG.082.002.001  ##              1              2              1              2              1  ## FG.083.001.001 FG.083.002.001 FG.083.003.001 FG.084.001.001 FG.084.002.001  ##              1              1              1              1              1  ## FG.085.001.001 FG.085.002.001 FG.085.003.001 FG.086.001.001 FG.086.002.001  ##              1              1              1              2              1  ## FG.086.003.001 FG.087.001.001 FG.087.001.002 FG.088.001.001 FG.088.002.001  ##              1              2              1              2              1  ## FG.088.003.001 FG.089.001.001 FG.089.001.002 FG.089.002.001 FG.090.001.001  ##              1              1              1              1              2  ## FG.090.002.001 FG.090.002.002 FG.091.001.001 FG.091.002.001 FG.092.001.001  ##              1              1              1              1              1  ## FG.092.002.001 FG.093.001.001 FG.093.002.001 FG.094.001.001 FG.094.002.001  ##              1              1              1              2              1  ## FG.094.003.001 FG.095.001.001 FG.096.001.001 FG.096.002.001 FG.097.001.001  ##              1              2              2              1              2  ## FG.097.001.002 FG.098.001.001 FG.098.002.001 FG.099.001.001 FG.100.001.001  ##              1              1              1              2              1  ## FG.100.002.001 FG.101.001.001 FG.101.001.002 FG.101.002.001 FG.102.001.001  ##              1              1              1              1              1  ## FG.102.002.001 FG.103.001.001 FG.103.001.002 FG.104.001.001 FG.105.001.001  ##              1              1              1              2              2  ## FG.105.002.001 FG.106.001.001 FG.106.001.002 FG.106.002.001 FG.107.001.001  ##              1              1              1              1              1  ## FG.107.002.001 FG.108.001.001 FG.109.001.001 FG.109.002.001 FG.110.001.001  ##              1              2              1              1              2  ## FG.111.001.001 FG.112.001.001 FG.113.001.001 FG.114.001.001 FG.115.001.001  ##              1              1              1              1              1  ## FG.116.001.001 FG.117.001.001  ##              1              1 fidx <- grep(\"FG.013.001.\", featureGroups(xmse)) eics <- featureChromatograms(     xmse, features = rownames(featureDefinitions(xmse))[fidx],     filled = TRUE, n = 1) cols <- c(\"#ff000080\", \"#0000ff80\") names(cols) <- unique(featureGroups(xmse)[fidx])  plotChromatogramsOverlay(eics, col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\") plotChromatogramsOverlay(normalize(eics),                          col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\") fidx <- grep(\"FG.045.001.\", featureGroups(xmse)) eics <- featureChromatograms(     xmse, features = rownames(featureDefinitions(xmse))[fidx],     filled = TRUE, n = 1) cols <- c(\"#ff000080\", \"#0000ff80\") names(cols) <- unique(featureGroups(xmse)[fidx])  plotChromatogramsOverlay(eics, col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\") plotChromatogramsOverlay(normalize(eics),                          col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\")"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-subsets-of-features","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of subsets of features","title":"Compounding (grouping) of LC-MS features","text":"previous sections always considering features data set, sometimes desirable just group pre-defined set features, example features found particular interest certain experiment (e.g. significant features). can easily achieved assigning features interest initial feature group, using NA group ID features. illustrate reset feature groups setting NA assign features interest (example just 30 randomly selected features) initial feature group \"FG\". call groupFeatures now simply sub-group set 30 features. feature NA \"feature_group\" column ignored.","code":"featureDefinitions(xmse)$feature_group <- NA_character_  set.seed(123) fts_idx <- sample(1:nrow(featureDefinitions(xmse)), 30) featureDefinitions(xmse)$feature_group[fts_idx] <- \"FG\" xmse <- groupFeatures(xmse, SimilarRtimeParam(diffRt = 20)) xmse <- groupFeatures(xmse, AbundanceSimilarityParam(threshold = 0.7)) table(featureGroups(xmse)) ##  ## FG.001.001 FG.001.002 FG.002.001 FG.002.002 FG.003.001 FG.004.001 FG.004.002  ##          1          1          1          1          2          1          1  ## FG.005.001 FG.005.002 FG.005.003 FG.006.001 FG.007.001 FG.007.002 FG.008.001  ##          1          1          1          2          1          1          1  ## FG.009.001 FG.010.001 FG.011.001 FG.012.001 FG.013.001 FG.014.001 FG.015.001  ##          1          1          1          1          1          1          1  ## FG.016.001 FG.017.001 FG.018.001 FG.019.001 FG.020.001 FG.021.001 FG.022.001  ##          1          1          1          1          1          1          1"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"Compounding (grouping) of LC-MS features","text":"","code":"sessionInfo() ## R Under development (unstable) (2023-11-22 r85609) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] pheatmap_1.0.12     faahKO_1.43.0       MsFeatures_1.11.0   ##  [4] xcms_4.1.4          MSnbase_2.29.1      ProtGenerics_1.35.0 ##  [7] S4Vectors_0.41.2    mzR_2.37.0          Rcpp_1.0.11         ## [10] Biobase_2.63.0      BiocGenerics_0.49.1 BiocParallel_1.37.0 ## [13] BiocStyle_2.31.0    ##  ## loaded via a namespace (and not attached): ##   [1] bitops_1.0-7                rlang_1.1.2                 ##   [3] magrittr_2.0.3              clue_0.3-65                 ##   [5] MassSpecWavelet_1.69.0      matrixStats_1.1.0           ##   [7] compiler_4.4.0              systemfonts_1.0.5           ##   [9] vctrs_0.6.5                 stringr_1.5.1               ##  [11] MetaboCoreUtils_1.11.0      pkgconfig_2.0.3             ##  [13] crayon_1.5.2                fastmap_1.1.1               ##  [15] XVector_0.43.0              utf8_1.2.4                  ##  [17] rmarkdown_2.25              preprocessCore_1.65.0       ##  [19] ragg_1.2.6                  purrr_1.0.2                 ##  [21] MultiAssayExperiment_1.29.0 xfun_0.41                   ##  [23] zlibbioc_1.49.0             cachem_1.0.8                ##  [25] GenomeInfoDb_1.39.1         jsonlite_1.8.8              ##  [27] progress_1.2.3              highr_0.10                  ##  [29] DelayedArray_0.29.0         prettyunits_1.2.0           ##  [31] parallel_4.4.0              cluster_2.1.6               ##  [33] R6_2.5.1                    bslib_0.6.1                 ##  [35] stringi_1.8.2               RColorBrewer_1.1-3          ##  [37] limma_3.59.1                GenomicRanges_1.55.1        ##  [39] jquerylib_0.1.4             bookdown_0.37               ##  [41] SummarizedExperiment_1.33.1 iterators_1.0.14            ##  [43] knitr_1.45                  IRanges_2.37.0              ##  [45] igraph_1.5.1                splines_4.4.0               ##  [47] Matrix_1.6-4                tidyselect_1.2.0            ##  [49] abind_1.4-5                 yaml_2.3.7                  ##  [51] doParallel_1.0.17           codetools_0.2-19            ##  [53] affy_1.81.0                 lattice_0.22-5              ##  [55] tibble_3.2.1                plyr_1.8.9                  ##  [57] evaluate_0.23               survival_3.5-7              ##  [59] desc_1.4.2                  Spectra_1.13.2              ##  [61] pillar_1.9.0                affyio_1.73.0               ##  [63] BiocManager_1.30.22         MatrixGenerics_1.15.0       ##  [65] foreach_1.5.2               MALDIquant_1.22.1           ##  [67] ncdf4_1.22                  generics_0.1.3              ##  [69] rprojroot_2.0.4             RCurl_1.98-1.13             ##  [71] hms_1.1.3                   ggplot2_3.4.4               ##  [73] munsell_0.5.0               scales_1.3.0                ##  [75] MsExperiment_1.5.1          glue_1.6.2                  ##  [77] lazyeval_0.2.2              tools_4.4.0                 ##  [79] mzID_1.41.0                 robustbase_0.99-1           ##  [81] QFeatures_1.13.1            vsn_3.71.0                  ##  [83] RANN_2.6.1                  fs_1.6.3                    ##  [85] XML_3.99-0.16               grid_4.4.0                  ##  [87] impute_1.77.0               MsCoreUtils_1.15.1          ##  [89] colorspace_2.1-0            GenomeInfoDbData_1.2.11     ##  [91] cli_3.6.1                   textshaping_0.3.7           ##  [93] fansi_1.0.5                 S4Arrays_1.3.1              ##  [95] dplyr_1.1.4                 AnnotationFilter_1.27.0     ##  [97] pcaMethods_1.95.0           gtable_0.3.4                ##  [99] DEoptimR_1.1-3              sass_0.4.8                  ## [101] digest_0.6.33               SparseArray_1.3.1           ## [103] farver_2.1.1                multtest_2.59.0             ## [105] memoise_2.0.1               htmltools_0.5.7             ## [107] pkgdown_2.0.7.9000          lifecycle_1.0.4             ## [109] statmod_1.5.0               MASS_7.3-60.1"},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"Grouping FTICR-MS data with xcms","text":"document describes use xcms analysis direct injection mass spec data, including peak detection, calibration correspondence (grouping peaks across samples).","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"peak-detection","dir":"Articles","previous_headings":"","what":"Peak detection","title":"Grouping FTICR-MS data with xcms","text":"Prior analysis step, peaks identified mass spec data. contrast typical metabolomics workflow, peaks identified chromatographic (time) dimension, direct injection mass spec data sets peaks identified m/z dimension. xcms uses functionality MassSpecWavelet package identify peaks. load required packages. information parallel processing setup please see BiocParallel vignette. documentation use example data set msdata package. Assuming msdata installed, locate path package load data set. create also data.frame describing experimental setup based file names. data files direct injection mass spectrometry experiments, .e. single spectrum available sample retention times. Peaks identified within spectrum using mass spec wavelet method.","code":"library(xcms) library(MassSpecWavelet)  register(SerialParam()) mzML_path <- system.file(\"fticr-mzML\", package = \"msdata\") mzML_files <- list.files(mzML_path, recursive = TRUE, full.names = TRUE)  ## We're subsetting to 2 samples per condition mzML_files <- mzML_files[c(1, 2, 6, 7)]  ## Create a data.frame assigning samples to sample groups, i.e. ham4 and ham5. grp <- rep(\"ham4\", length(mzML_files)) grp[grep(basename(mzML_files), pattern = \"^HAM005\")] <- \"ham5\" pd <- data.frame(filename = basename(mzML_files), sample_group = grp)  ## Load the data. ham_raw <- readMSData(files = mzML_files,                       pdata = new(\"NAnnotatedDataFrame\", pd),                       mode = \"onDisk\") ## Only a single spectrum with an *artificial* retention time is available ## for each sample rtime(ham_raw) ## F1.S1 F2.S1 F3.S1 F4.S1  ##    -1    -1    -1    -1 ## Define the parameters for the peak detection msw <- MSWParam(scales = c(1, 4, 9), nearbyPeak = TRUE, winSize.noise = 500,                 SNR.method = \"data.mean\", snthresh = 10)  ham_prep <- findChromPeaks(ham_raw, param = msw)  head(chromPeaks(ham_prep)) ##            mz    mzmin    mzmax rt rtmin rtmax    into     maxo       sn intf ## CP01 403.2367 403.2279 403.2447 -1    -1    -1 4735258 372259.4 22.97534   NA ## CP02 409.1845 409.1747 409.1936 -1    -1    -1 4158404 310572.1 20.61382   NA ## CP03 413.2677 413.2585 413.2769 -1    -1    -1 6099006 435462.6 27.21723   NA ## CP04 423.2363 423.2266 423.2459 -1    -1    -1 2708391 174252.7 14.74527   NA ## CP05 427.2681 427.2574 427.2779 -1    -1    -1 6302089 461385.6 32.50050   NA ## CP06 437.2375 437.2254 437.2488 -1    -1    -1 7523070 517917.6 34.37645   NA ##           maxf sample ## CP01  814693.1      1 ## CP02  732119.9      1 ## CP03 1018994.8      1 ## CP04  435858.5      1 ## CP05 1125644.3      1 ## CP06 1282906.5      1"},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"calibration","dir":"Articles","previous_headings":"","what":"Calibration","title":"Grouping FTICR-MS data with xcms","text":"calibrate method can used correct m/z values identified peaks. currently implemented method requires identified peaks list m/z values known calibrants. identified peaks m/z values adjusted based differences calibrants’ m/z values m/z values closest peaks (within user defined permitted maximal distance). Note method presently calibrate identified peaks, original m/z values spectra. demonstrate calibrate method one data files artificially defined calibration m/z values. first subset data set first data file, extract m/z values 3 peaks modify values slightly. Next calibrate data set using previously defined artificial calibrants. using \"edgeshift\" method calibration adjusts peaks within range m/z values calibrants using linear interpolation shifts chromatographic peaks outside range constant factor (difference lowest respectively largest calibrant m/z closest peak’s m/z). Note real use case, m/z values obviously represent known m/z calibrants defined actual data. evaluate calibration plot difference adjusted raw m/z values (y-axis) raw m/z values.","code":"## Subset to the first file. first_file <- filterFile(ham_prep, file = 1)  ## Extract 3 m/z values calib_mz <- chromPeaks(first_file)[c(1, 4, 7), \"mz\"] calib_mz <- calib_mz + 0.00001 * runif(1, 0, 0.4) * calib_mz + 0.0001 ## Set-up the parameter class for the calibration prm <- CalibrantMassParam(mz = calib_mz, method = \"edgeshift\",                           mzabs = 0.0001, mzppm = 5) first_file_calibrated <- calibrate(first_file, param = prm) diffs <- chromPeaks(first_file_calibrated)[, \"mz\"] -     chromPeaks(first_file)[, \"mz\"]  plot(x = chromPeaks(first_file)[, \"mz\"], xlab = expression(m/z[raw]),      y = diffs, ylab = expression(m/z[calibrated] - m/z[raw]))"},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"correspondence","dir":"Articles","previous_headings":"","what":"Correspondence","title":"Grouping FTICR-MS data with xcms","text":"Correspondence aims group peaks across samples define features (ions m/z values). Peaks single spectrum, direct injection MS experiments can grouped MZclust method. perform correspondence analysis groupChromPeaks method using default settings. Getting overview performed processings: peak group information, .e. feature definitions can accessed featureDefinitions method. Plotting raw data direct injection samples involves little processing LC/GC-MS data can simply use chromatogram method extract data. extract m/z-intensity pairs peaks associated first feature. thus first identify peaks feature define m/z values range. Using range can subsequently use filterMz function sub-set full data set signal associated feature’s peaks. object can call mz intensity functions extract data.  access actual intensity values feature sample featureValue method can used. setting value = \"\" tells function return integrated signal peak (one representative peak) per sample. NA reported features samples peak identified feature’s m/z value. instances might still signal feature’s position raw data files, peak detection failed identify peak. cases signal can recovered using fillChromPeaks method integrates raw signal feature’s location. signal location NA reported.","code":"## Using default settings but define sample group assignment mzc_prm <- MzClustParam(sampleGroups = ham_prep$sample_group) ham_prep <- groupChromPeaks(ham_prep, param = mzc_prm) ham_prep ## MSn experiment data (\"XCMSnExp\") ## Object size in memory: 0.04 Mb ## - - - Spectra data - - - ##  MS level(s): 1  ##  Number of spectra: 4  ##  MSn retention times: -1:59 - -1:59 minutes ## - - - Processing information - - - ## Data loaded [Thu Dec  7 08:45:30 2023]  ##  MSnbase version: 2.29.1  ## - - - Meta data  - - - ## phenoData ##   rowNames: 1 2 3 4 ##   varLabels: filename sample_group ##   varMetadata: labelDescription ## Loaded from: ##   [1] HAM004_641fE_14-11-07--Exp1.extracted.mzML...  [4] HAM005_641fE_14-11-07--Exp2.extracted.mzML ##   Use 'fileNames(.)' to see all files. ## protocolData: none ## featureData ##   featureNames: F1.S1 F2.S1 F3.S1 F4.S1 ##   fvarLabels: fileIdx spIdx ... spectrum (35 total) ##   fvarMetadata: labelDescription ## experimentData: use 'experimentData(object)' ## - - - xcms preprocessing - - - ## Chromatographic peak detection: ##  method: MSW  ##  38 peaks identified in 4 samples. ##  On average 9.5 chromatographic peaks per sample. ## Correspondence: ##  method: mzClust  ##  20 features identified. ##  Median mz range of features: 9.1553e-05 ##  Median rt range of features: 0 featureDefinitions(ham_prep) ## DataFrame with 20 rows and 10 columns ##          mzmed     mzmin     mzmax     rtmed     rtmin     rtmax    npeaks ##      <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> ## FT01   402.285   402.285   402.286        -1        -1        -1         2 ## FT02   403.237   403.237   403.237        -1        -1        -1         4 ## FT03   405.109   405.109   405.109        -1        -1        -1         2 ## FT04   409.184   409.184   409.185        -1        -1        -1         2 ## FT05   410.144   410.144   410.145        -1        -1        -1         2 ## ...        ...       ...       ...       ...       ...       ...       ... ## FT16   437.238   437.238   437.238        -1        -1        -1         2 ## FT17   438.240   438.240   438.240        -1        -1        -1         2 ## FT18   439.151   439.151   439.151        -1        -1        -1         2 ## FT19   441.130   441.130   441.131        -1        -1        -1         2 ## FT20   445.293   445.292   445.293        -1        -1        -1         2 ##           ham4      ham5     peakidx ##      <numeric> <numeric>      <list> ## FT01         0         2       16,28 ## FT02         2         2 17,29,1,... ## FT03         0         2       18,30 ## FT04         2         0        10,2 ## FT05         0         2       19,31 ## ...        ...       ...         ... ## FT16         2         0        6,13 ## FT17         2         0        7,14 ## FT18         0         2       26,37 ## FT19         0         2       38,27 ## FT20         2         0        15,8 ## Get the peaks belonging to the first feature pks <- chromPeaks(ham_prep)[featureDefinitions(ham_prep)$peakidx[[1]], ]  ## Define the m/z range mzr <- c(min(pks[, \"mzmin\"]) - 0.001, max(pks[, \"mzmax\"]) + 0.001)  ## Subset the object to the m/z range ham_prep_sub <- filterMz(ham_prep, mz = mzr)  ## Extract the mz and intensity values mzs <- mz(ham_prep_sub, bySample = TRUE) ints <- intensity(ham_prep_sub, bySample = TRUE)  ## Plot the data plot(3, 3, pch = NA, xlim = range(mzs), ylim = range(ints), main = \"FT01\",      xlab = \"m/z\", ylab = \"intensity\") ## Define colors cols <- rep(\"#ff000080\", length(mzs)) cols[ham_prep_sub$sample_group == \"ham5\"] <- \"#0000ff80\" tmp <- mapply(mzs, ints, cols, FUN = function(x, y, col) {     points(x, y, col = col, type = \"l\") }) feat_vals <- featureValues(ham_prep, value = \"into\") head(feat_vals) ##      HAM004_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                         NA ## FT02                                    4735258 ## FT03                                         NA ## FT04                                    4158404 ## FT05                                         NA ## FT06                                    6099006 ##      HAM004_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                         NA ## FT02                                    6202418 ## FT03                                         NA ## FT04                                    5004546 ## FT05                                         NA ## FT06                                    4950642 ##      HAM005_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                    4095293 ## FT02                                    4811391 ## FT03                                    2982453 ## FT04                                         NA ## FT05                                    2872023 ## FT06                                         NA ##      HAM005_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                    4804763 ## FT02                                    2581183 ## FT03                                    2268984 ## FT04                                         NA ## FT05                                    2133219 ## FT06                                         NA ham_prep <- fillChromPeaks(ham_prep, param = FillChromPeaksParam())  head(featureValues(ham_prep, value = \"into\")) ##      HAM004_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                   768754.0 ## FT02                                  4735257.5 ## FT03                                   652566.6 ## FT04                                  4158404.5 ## FT05                                   652201.1 ## FT06                                  6099006.3 ##      HAM004_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                  1230140.4 ## FT02                                  6202417.6 ## FT03                                   374109.9 ## FT04                                  5004546.3 ## FT05                                   403448.4 ## FT06                                  4950641.7 ##      HAM005_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                    4095293 ## FT02                                    4811391 ## FT03                                    2982453 ## FT04                                    1221031 ## FT05                                    2872023 ## FT06                                    1573988 ##      HAM005_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                  4804762.5 ## FT02                                  2581183.1 ## FT03                                  2268984.5 ## FT04                                  1241294.4 ## FT05                                  2133219.4 ## FT06                                   977694.5"},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"further-analysis","dir":"Articles","previous_headings":"","what":"Further analysis","title":"Grouping FTICR-MS data with xcms","text":"analysis, .e. detection features/metabolites significantly different abundances, PCA analyses can performed feature matrix using functionality R packages, limma.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"Grouping FTICR-MS data with xcms","text":"","code":"sessionInfo() ## R Under development (unstable) (2023-11-22 r85609) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] MassSpecWavelet_1.69.0 xcms_4.1.4             MSnbase_2.29.1         ##  [4] ProtGenerics_1.35.0    S4Vectors_0.41.2       mzR_2.37.0             ##  [7] Rcpp_1.0.11            Biobase_2.63.0         BiocGenerics_0.49.1    ## [10] BiocParallel_1.37.0    BiocStyle_2.31.0       ##  ## loaded via a namespace (and not attached): ##   [1] bitops_1.0-7                rlang_1.1.2                 ##   [3] magrittr_2.0.3              clue_0.3-65                 ##   [5] matrixStats_1.1.0           compiler_4.4.0              ##   [7] systemfonts_1.0.5           vctrs_0.6.5                 ##   [9] stringr_1.5.1               MetaboCoreUtils_1.11.0      ##  [11] pkgconfig_2.0.3             crayon_1.5.2                ##  [13] fastmap_1.1.1               XVector_0.43.0              ##  [15] utf8_1.2.4                  rmarkdown_2.25              ##  [17] preprocessCore_1.65.0       ragg_1.2.6                  ##  [19] purrr_1.0.2                 MultiAssayExperiment_1.29.0 ##  [21] xfun_0.41                   zlibbioc_1.49.0             ##  [23] cachem_1.0.8                GenomeInfoDb_1.39.1         ##  [25] jsonlite_1.8.8              progress_1.2.3              ##  [27] highr_0.10                  DelayedArray_0.29.0         ##  [29] prettyunits_1.2.0           parallel_4.4.0              ##  [31] cluster_2.1.6               R6_2.5.1                    ##  [33] bslib_0.6.1                 stringi_1.8.2               ##  [35] RColorBrewer_1.1-3          limma_3.59.1                ##  [37] GenomicRanges_1.55.1        jquerylib_0.1.4             ##  [39] bookdown_0.37               SummarizedExperiment_1.33.1 ##  [41] iterators_1.0.14            knitr_1.45                  ##  [43] IRanges_2.37.0              igraph_1.5.1                ##  [45] splines_4.4.0               Matrix_1.6-4                ##  [47] tidyselect_1.2.0            abind_1.4-5                 ##  [49] yaml_2.3.7                  doParallel_1.0.17           ##  [51] codetools_0.2-19            affy_1.81.0                 ##  [53] lattice_0.22-5              tibble_3.2.1                ##  [55] plyr_1.8.9                  signal_1.8-0                ##  [57] evaluate_0.23               survival_3.5-7              ##  [59] desc_1.4.2                  Spectra_1.13.2              ##  [61] pillar_1.9.0                affyio_1.73.0               ##  [63] BiocManager_1.30.22         MatrixGenerics_1.15.0       ##  [65] foreach_1.5.2               MALDIquant_1.22.1           ##  [67] ncdf4_1.22                  generics_0.1.3              ##  [69] rprojroot_2.0.4             RCurl_1.98-1.13             ##  [71] hms_1.1.3                   ggplot2_3.4.4               ##  [73] munsell_0.5.0               scales_1.3.0                ##  [75] MsExperiment_1.5.1          glue_1.6.2                  ##  [77] lazyeval_0.2.2              MsFeatures_1.11.0           ##  [79] tools_4.4.0                 mzID_1.41.0                 ##  [81] robustbase_0.99-1           QFeatures_1.13.1            ##  [83] vsn_3.71.0                  RANN_2.6.1                  ##  [85] fs_1.6.3                    XML_3.99-0.16               ##  [87] grid_4.4.0                  impute_1.77.0               ##  [89] MsCoreUtils_1.15.1          colorspace_2.1-0            ##  [91] GenomeInfoDbData_1.2.11     cli_3.6.1                   ##  [93] textshaping_0.3.7           fansi_1.0.5                 ##  [95] S4Arrays_1.3.1              dplyr_1.1.4                 ##  [97] AnnotationFilter_1.27.0     pcaMethods_1.95.0           ##  [99] gtable_0.3.4                DEoptimR_1.1-3              ## [101] sass_0.4.8                  digest_0.6.33               ## [103] SparseArray_1.3.1           multtest_2.59.0             ## [105] memoise_2.0.1               htmltools_0.5.7             ## [107] pkgdown_2.0.7.9000          lifecycle_1.0.4             ## [109] statmod_1.5.0               MASS_7.3-60.1"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"LC-MS/MS data analysis with xcms","text":"Metabolite identification important step non-targeted metabolomics requires different steps. One involves use tandem mass spectrometry generate fragmentation spectra detected metabolites (LC-MS/MS), compared fragmentation spectra known metabolites. Different approaches exist generation fragmentation spectra, whereas used data dependent acquisition (DDA) also known top-n method. method top N intense ions (m/z values) MS1 scan selected fragmentation next N scans cycle starts . method allows generate clean MS2 fragmentation spectra fly acquisition without need experiments, suffers poor coverage detected metabolites (since limited number ions fragmented) less accurate quantification compounds (since fewer MS1 scans generated). Data independent approaches (DIA) like Bruker bbCID, Agilent AllIons Waters MSe don’t use preselection, rather fragment detected molecules . using alternating schemes scan low high collision energy collect MS1 MS2 data. Using approach, problem coverage, relation precursor fragment masses lost leading chimeric spectra. Sequential Window Acquisition Theoretical Mass Spectra (SWATH [1]) combines approaches middle-way approach. precursor selection acquisition independent acquired data, rather isolating precusors , defined windows (.e. ranges m/z values) used scanned. reduces overlap fragment spectra still keeping high coverage. document showcases analysis two small LC-MS/MS data sets using xcms. data files used reversed-phase LC-MS/MS runs Agilent Pesticide mix obtained Sciex 6600 Triple ToF operated SWATH acquisition mode. comparison DDA file sample included.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"analysis-of-dda-data","dir":"Articles","previous_headings":"","what":"Analysis of DDA data","title":"LC-MS/MS data analysis with xcms","text":"load example DDA data set create total ion chromatogram MS1 data. According TIC signal measured ~ 200 600 seconds (see plot ). thus filter DDA data retention time range.  variable dda_data contains now MS1 MS2 spectra specified mzML file. number spectra MS level listed . Note subset experiment first data file (using [1]) access directly spectra within sample spectra function (returns Spectra object Spectra package). Note use pipe operator |> better readability. MS2 spectra can get m/z precursor ion precursorMz function. first subset data set single sample filter spectra MS level 2 extracting precursor m/z values. precursorIntensity function also possible extract intensity precursor ion. manufacturers (like Sciex) don’t define/export precursor intensity thus either NA 0 reported. can however use estimatePrecursorIntensity function Spectra package determine precursor intensity MS 2 spectrum based intensity respective ion previous MS1 scan (note method = \"interpolation\" precursor intensity defined based interpolation intensity previous subsequent MS1 scan). estimate precursor intensities, full data (MS1 spectra NA value reported). next set precursor intensity spectrum metadata dda_data. can extracted later precursorIntensity function. Next perform chromatographic peak detection MS level 1 data findChromPeaks method. define settings centWave-based peak detection perform analysis. total 114 peaks identified present data set. advantage LC-MS/MS data (MS1) ions fragmented corresponding MS2 spectra measured. Thus, ions (identified MS1 chromatographic peaks) MS2 spectra available. can facilitate annotation respective MS1 chromatographic peaks (MS1 features correspondence analysis). Spectra identified chromatographic peaks can extracted chromPeakSpectra method. MS2 spectra precursor m/z retention time within rt m/z range chromatographic peak returned. default chromPeakSpectra returns spectra associated MS1 chromatographic peak, parameter method allows choose return one spectrum per peak (look ?chromPeakSpectra help page details). Also, possible extract MS1 spectra peak specifying msLevel = 1L call (e.g. evaluate full MS1 signal peak’s apex position). returned Spectra contains also reference respective chromatographic peak additional spectra variable \"peak_id\" contains identifier chromatographic peak (.e. row name chromPeaks matrix). Note also return.type = \"List\" list parallel chromPeaks matrix returned, .e. element list contain spectra chromatographic peak index. data representation might eventually simplify processing. next use MS2 information aid annotation chromatographic peak. example use chromatographic peak ion m/z 304.1131 extract code block . search potential ions similar m/z reference database (e.g. Metlin) returned large list potential hits, small ppm. two hits, Flumazenil (Metlin ID 2724) Fenamiphos (Metlin ID 72445) experimental MS2 spectra available. Thus, match MS2 spectrum identified chromatographic peak annotate ion. extract MS2 spectra associated candidate chromatographic peak using ID peak present data set. 5 MS2 spectra representing fragmentation ion(s) measured candidate chromatographic peak. next reduce single MS2 spectrum using combineSpectra method employing combinePeaks function determine peaks keep resulting spectrum (look ?combinePeaks help page details). Parameter f allows specify spectra input object combined one. Note combination multiple fragment spectra single spectrum might generally best approach suggested types data. Mass peaks input spectra difference m/z smaller 20 ppm (parameter ppm) combined one peak median m/z intensity reported . Due parameter minProp = 0.8, resulting MS2 spectrum contains peaks present 80% input spectra. plot consensus spectrum shown . Consensus MS2 spectrum created measured MS2 spectra ions chromatographic peak CP53. now match consensus spectrum database MS2 spectra. example simply load MS2 spectra two compounds matching m/z exported Metlin. compounds MS2 spectra created collision energies 0V, 10V, 20V 40V available. import respective data plot candidate spectrum MS2 spectra Flumanezil Fenamiphos (collision energy 20V). import files MGF format load MsBackendMgf Bioconductor package adds MGF file support Spectra package. Prior plotting scale experimental spectra replace peak intensities values relative maximum peak intensity (set value 100). Mirror plots candidate MS2 spectrum Flumanezil (left) Fenamiphos (right). upper panel represents candidate MS2 spectrum, lower target MS2 spectrum. Matching peaks indicated dot. candidate spectrum matches Fenamiphos, thus, example chromatographic peak represents signal measured compound. addition plotting spectra, can also calculate similarities compareSpectra method (uses default normalized dot-product calculate similarity). Clearly, candidate spectrum match Flumanezil, high similarity Fenamiphos. performed MS2-based annotation single chromatographic peak, easily extended full list MS2 spectra (returned chromPeakSpectra) chromatographic peaks experiment. See also alternative tutorials matching experimental fragment spectra reference. present example used single data file thus need perform sample alignment correspondence analysis. tasks however performed similarly plain LC-MS data, retention times recorded MS2 spectra however also adjusted alignment based MS1 data. correspondence analysis (peak grouping) MS2 spectra features can extracted featureSpectra function returns MS2 spectra associated chromatographic peak feature. Note also workflow can included Feature-Based Molecular Networking FBMN match MS2 spectra GNPS. See details examples.","code":"library(xcms) library(MsExperiment)  dda_file <- system.file(\"TripleTOF-SWATH\", \"PestMix1_DDA.mzML\",                         package = \"msdata\") dda_data <- readMsExperiment(dda_file) chr <- chromatogram(dda_data, aggregationFun = \"sum\", msLevel = 1L) plot(chr) abline(v = c(230, 610)) ## filter the data dda_data <- filterRt(dda_data, rt = c(230, 610)) ## Filter spectra dda_data[1] |> spectra() |> msLevel() |> table() ##  ##    1    2  ## 1389 2214 dda_data[1] |> spectra() |> filterMsLevel(2) |> precursorMz() |> head() ## [1] 182.18777 182.18893  55.00579 182.19032 237.12296 237.11987 dda_data[1] |> spectra() |> filterMsLevel(2) |> precursorIntensity() |> head() ## [1] 0 0 0 0 0 0 prec_int <- estimatePrecursorIntensity(spectra(dda_data)) spectra(dda_data)$precursorIntensity <- prec_int  dda_data[1] |> spectra() |> filterMsLevel(2) |> precursorIntensity() |> head() ## [1]        NA  9.198155  2.773988 27.590797  3.443145  7.621923 cwp <- CentWaveParam(snthresh = 5, noise = 100, ppm = 10,                      peakwidth = c(3, 30)) dda_data <- findChromPeaks(dda_data, param = cwp, msLevel = 1L) library(Spectra) dda_spectra <- chromPeakSpectra(dda_data, msLevel = 2L) dda_spectra ## MSn data (Spectra) with 142 spectra in a MsBackendMzR backend: ##       msLevel     rtime scanIndex ##     <integer> <numeric> <integer> ## 1           2   237.869      1812 ## 2           2   241.299      1846 ## 3           2   325.763      2439 ## 4           2   326.583      2446 ## 5           2   327.113      2450 ## ...       ...       ...       ... ## 138         2   574.725      5110 ## 139         2   575.255      5115 ## 140         2   596.584      5272 ## 141         2   592.424      5236 ## 142         2   596.054      5266 ##  ... 34 more variables/columns. ##  ## file(s): ## PestMix1_DDA.mzML ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023] ##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023] ##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023] dda_spectra$peak_id ##   [1] \"CP004\" \"CP004\" \"CP005\" \"CP005\" \"CP006\" \"CP006\" \"CP008\" \"CP008\" \"CP011\" ##  [10] \"CP011\" \"CP012\" \"CP012\" \"CP013\" \"CP013\" \"CP013\" \"CP013\" \"CP014\" \"CP014\" ##  [19] \"CP014\" \"CP014\" \"CP018\" \"CP022\" \"CP022\" \"CP022\" \"CP025\" \"CP025\" \"CP025\" ##  [28] \"CP025\" \"CP026\" \"CP026\" \"CP026\" \"CP026\" \"CP033\" \"CP033\" \"CP034\" \"CP034\" ##  [37] \"CP034\" \"CP034\" \"CP034\" \"CP035\" \"CP035\" \"CP035\" \"CP041\" \"CP041\" \"CP041\" ##  [46] \"CP042\" \"CP042\" \"CP042\" \"CP043\" \"CP047\" \"CP047\" \"CP049\" \"CP049\" \"CP049\" ##  [55] \"CP049\" \"CP050\" \"CP050\" \"CP050\" \"CP051\" \"CP051\" \"CP051\" \"CP054\" \"CP055\" ##  [64] \"CP055\" \"CP055\" \"CP056\" \"CP056\" \"CP056\" \"CP056\" \"CP056\" \"CP060\" \"CP060\" ##  [73] \"CP060\" \"CP060\" \"CP064\" \"CP064\" \"CP065\" \"CP065\" \"CP066\" \"CP066\" \"CP069\" ##  [82] \"CP069\" \"CP069\" \"CP070\" \"CP070\" \"CP070\" \"CP072\" \"CP072\" \"CP072\" \"CP073\" ##  [91] \"CP074\" \"CP074\" \"CP074\" \"CP074\" \"CP075\" \"CP075\" \"CP075\" \"CP077\" \"CP077\" ## [100] \"CP077\" \"CP079\" \"CP079\" \"CP079\" \"CP079\" \"CP080\" \"CP080\" \"CP080\" \"CP081\" ## [109] \"CP086\" \"CP086\" \"CP086\" \"CP086\" \"CP086\" \"CP088\" \"CP088\" \"CP088\" \"CP089\" ## [118] \"CP089\" \"CP091\" \"CP091\" \"CP093\" \"CP093\" \"CP094\" \"CP094\" \"CP094\" \"CP095\" ## [127] \"CP095\" \"CP095\" \"CP096\" \"CP096\" \"CP096\" \"CP098\" \"CP098\" \"CP098\" \"CP098\" ## [136] \"CP098\" \"CP099\" \"CP099\" \"CP099\" \"CP100\" \"CP101\" \"CP101\" ex_mz <- 304.1131 chromPeaks(dda_data, mz = ex_mz, ppm = 20) ##             mz    mzmin    mzmax      rt   rtmin   rtmax    into     intb ## CP056 304.1133 304.1126 304.1143 425.024 417.985 441.773 13040.8 13007.79 ##           maxo  sn sample ## CP056 3978.987 232      1 ex_id <- rownames(chromPeaks(dda_data, mz = ex_mz, ppm = 20)) ex_spectra <- dda_spectra[dda_spectra$peak_id == ex_id] ex_spectra ## MSn data (Spectra) with 5 spectra in a MsBackendMzR backend: ##     msLevel     rtime scanIndex ##   <integer> <numeric> <integer> ## 1         2   418.926      3505 ## 2         2   419.306      3510 ## 3         2   423.036      3582 ## 4         2   423.966      3603 ## 5         2   424.296      3609 ##  ... 34 more variables/columns. ##  ## file(s): ## PestMix1_DDA.mzML ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023] ##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023] ##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023] ex_spectrum <- combineSpectra(ex_spectra, FUN = combinePeaks, ppm = 20,                               peaks = \"intersect\", minProp = 0.8,                               intensityFun = median, mzFun = median,                               f = ex_spectra$peak_id) ## Warning: 'combinePeaks' for lists of peak matrices is deprecated; please use ## 'combinePeaksData' instead. ex_spectrum ## MSn data (Spectra) with 1 spectra in a MsBackendMemory backend: ##     msLevel     rtime scanIndex ##   <integer> <numeric> <integer> ## 1         2   418.926      3505 ##  ... 34 more variables/columns. ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023] ##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023] ##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023] ##  ...1 more processings. Use 'processingLog' to list all. plotSpectra(ex_spectrum) scale_fun <- function(z, ...) {     z[, \"intensity\"] <- z[, \"intensity\"] /         max(z[, \"intensity\"], na.rm = TRUE) * 100     z } ex_spectrum <- addProcessing(ex_spectrum, FUN = scale_fun) library(MsBackendMgf) flumanezil <- Spectra(     system.file(\"mgf\", \"metlin-2724.mgf\", package = \"xcms\"),     source = MsBackendMgf()) ## Start data import from 1 files ... done fenamiphos <- Spectra(     system.file(\"mgf\", \"metlin-72445.mgf\", package = \"xcms\"),     source = MsBackendMgf()) ## Start data import from 1 files ... done par(mfrow = c(1, 2)) plotSpectraMirror(ex_spectrum, flumanezil[3], main = \"against Flumanezil\",                   ppm = 40) plotSpectraMirror(ex_spectrum, fenamiphos[3], main = \"against Fenamiphos\",                   ppm = 40) compareSpectra(ex_spectrum, flumanezil, ppm = 40) ## [1] 4.520957e-02 3.283806e-02 2.049379e-03 3.374354e-05 compareSpectra(ex_spectrum, fenamiphos, ppm = 40) ## [1] 0.1326234432 0.4879399946 0.7198406271 0.3997922658 0.0004876129 ## [6] 0.0028408885 0.0071030051 0.0053809736"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"dia-swath-data-analysis","dir":"Articles","previous_headings":"","what":"DIA (SWATH) data analysis","title":"LC-MS/MS data analysis with xcms","text":"section analyze small SWATH data set consisting single mzML file data sample analyzed previous section recorded SWATH mode. read data readMsExperiment function. resulting object contain recorded MS1 MS2 spectra specified file. Similar previous data file, filter file signal 230 610 seconds. determine number MS level 1 2 spectra present data set. described introduction, SWATH mode ions within pre-defined isolation windows fragmented MS2 spectra measured. definition isolation windows (SWATH pockets) imported mzML files available additional spectra variables. inspect respective information first spectra. upper lower isolation window m/z available spectra variables \"isolationWindowLowerMz\" \"isolationWindowUpperMz\" respectively target m/z isolation window \"isolationWindowTargetMz\". can use spectraData function extract information spectra within swath_data object. also access variables directly dedicated isolationWindowLowerMz isolationWindowUpperMz functions. present data set use value isolation window target m/z define individual SWATH pockets. list number spectra recorded pocket/isolation window. thus 422 423 MS2 spectra measured isolation window. inspect data can also extract chromatograms measured MS1 well MS2 data. MS2 data set parameter msLevel = 2L addition also specify isolation window want extract data. extract TIC MS1 data one isolation windows (isolation window target m/z 270.85) plot . TIC MS1 (upper panel) MS2 data isolation window target m/z 270.85 (lower panel). Note also DIA data manufacturers (e.g. Waters MSe) supported long spectra variable isolationWindowTargetMz available. variable provided raw data files can also manually assigned spectrum data set. example precursor m/z individual spectra used \"isolationWindowTargetMz\" (note: line shown example evaluated).","code":"swath_file <- system.file(\"TripleTOF-SWATH\",                           \"PestMix1_SWATH.mzML\",                           package = \"msdata\")  swath_data <- readMsExperiment(swath_file) swath_data <- filterRt(swath_data, rt = c(230, 610)) spectra(swath_data) |> msLevel() |> table() ##  ##    1    2  ##  422 3378 spectra(swath_data) |> spectraData(c(\"isolationWindowTargetMz\", \"isolationWindowLowerMz\",               \"isolationWindowUpperMz\", \"msLevel\", \"rtime\")) |> head() ## DataFrame with 6 rows and 5 columns ##   isolationWindowTargetMz isolationWindowLowerMz isolationWindowUpperMz ##                 <numeric>              <numeric>              <numeric> ## 1                  299.10                  283.5                  314.7 ## 2                  329.80                  313.7                  345.9 ## 3                  367.35                  344.9                  389.8 ## 4                  601.85                  388.8                  814.9 ## 5                      NA                     NA                     NA ## 6                  163.75                  139.5                  188.0 ##     msLevel     rtime ##   <integer> <numeric> ## 1         2   230.073 ## 2         2   230.170 ## 3         2   230.267 ## 4         2   230.364 ## 5         1   230.459 ## 6         2   230.585 head(isolationWindowLowerMz(spectra(swath_data))) ## [1] 283.5 313.7 344.9 388.8    NA 139.5 head(isolationWindowUpperMz(spectra(swath_data))) ## [1] 314.7 345.9 389.8 814.9    NA 188.0 table(isolationWindowTargetMz(spectra(swath_data))) ##  ## 163.75 208.95 244.05 270.85  299.1  329.8 367.35 601.85  ##    422    422    422    422    423    423    422    422 tic_ms1 <- chromatogram(swath_data, msLevel = 1L, aggregationFun = \"sum\") tic_ms2 <- chromatogram(swath_data, msLevel = 2L, aggregationFun = \"sum\",                         isolationWindowTargetMz = 270.85) par(mfrow = c(2, 1)) plot(tic_ms1, main = \"MS1\") plot(tic_ms2, main = \"MS2, isolation window m/z 270.85\") spectra(swath_data)$isolationWindowTargetMz <- precursorMz(spectra(swath_data))"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"chromatographic-peak-detection-in-ms1-and-ms2-data","dir":"Articles","previous_headings":"DIA (SWATH) data analysis","what":"Chromatographic peak detection in MS1 and MS2 data","title":"LC-MS/MS data analysis with xcms","text":"Similar conventional LC-MS analysis, perform first chromatographic peak detection (MS level 1 data) findChromPeaks method. define settings centWave-based peak detection perform analysis. Next perform chromatographic peak detection MS level 2 data separately individual isolation window. use findChromPeaksIsolationWindow function employing peak detection algorithm reducing however required signal--noise ratio. isolationWindow parameter allows specify MS2 spectra belong isolation window hence defines set MS2 spectra chromatographic peak detection performed. default \"isolationWindowTargetMz\" variable object’s spectra used. findChromPeaksIsolationWindow function added peaks identified individual isolation windows chromPeaks matrix containing already MS1 chromatographic peaks. newly added peaks can identified \"isolationWindow\" column object’s chromPeakData. count number chromatographic peaks identified within isolation window (number chromatographic peaks identified MS1 62). thus successfully identified chromatographic peaks different MS levels isolation windows. next step identify measured signals represents data original compound reconstruct fragment spectra MS1 signal (chromatographic peak).","code":"cwp <- CentWaveParam(snthresh = 5, noise = 100, ppm = 10,                      peakwidth = c(3, 30)) swath_data <- findChromPeaks(swath_data, param = cwp) swath_data ## Object of class XcmsExperiment  ##  Spectra: MS1 (422) MS2 (3378)  ##  Experiment data: 1 sample(s) ##  Sample data links: ##   - spectra: 1 sample(s) to 3800 element(s). ##  xcms results: ##   - chromatographic peaks: 62 in MS level(s): 1 cwp <- CentWaveParam(snthresh = 3, noise = 10, ppm = 10,                      peakwidth = c(3, 30)) swath_data <- findChromPeaksIsolationWindow(swath_data, param = cwp) swath_data ## Object of class XcmsExperiment  ##  Spectra: MS1 (422) MS2 (3378)  ##  Experiment data: 1 sample(s) ##  Sample data links: ##   - spectra: 1 sample(s) to 3800 element(s). ##  xcms results: ##   - chromatographic peaks: 370 in MS level(s): 1, 2 chromPeakData(swath_data) ## DataFrame with 370 rows and 6 columns ##        ms_level is_filled isolationWindow isolationWindowTargetMZ ##       <integer> <logical>       <numeric>               <numeric> ## CP01          1     FALSE              NA                      NA ## CP02          1     FALSE              NA                      NA ## CP03          1     FALSE              NA                      NA ## CP04          1     FALSE              NA                      NA ## CP05          1     FALSE              NA                      NA ## ...         ...       ...             ...                     ... ## CP366         2     FALSE          601.85                  601.85 ## CP367         2     FALSE          601.85                  601.85 ## CP368         2     FALSE          601.85                  601.85 ## CP369         2     FALSE          601.85                  601.85 ## CP370         2     FALSE          601.85                  601.85 ##       isolationWindowLowerMz isolationWindowUpperMz ##                    <numeric>              <numeric> ## CP01                      NA                     NA ## CP02                      NA                     NA ## CP03                      NA                     NA ## CP04                      NA                     NA ## CP05                      NA                     NA ## ...                      ...                    ... ## CP366                  388.8                  814.9 ## CP367                  388.8                  814.9 ## CP368                  388.8                  814.9 ## CP369                  388.8                  814.9 ## CP370                  388.8                  814.9 table(chromPeakData(swath_data)$isolationWindow) ##  ## 163.75 208.95 244.05 270.85  299.1  329.8 367.35 601.85  ##      2     38     32     14    105     23     62     32"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"reconstruction-of-ms2-spectra","dir":"Articles","previous_headings":"DIA (SWATH) data analysis","what":"Reconstruction of MS2 spectra","title":"LC-MS/MS data analysis with xcms","text":"Identifying signal fragment ions precursor measured MS1 chromatographic peak non-trivial task. MS2 spectrum fragment ion MS1 chromatographic peak reconstructed available MS2 signal (.e. chromatographic peaks identified MS level 2). SWATH data, fragment ion signal present isolation window contains m/z precursor ion chromatographic peak shape MS2 chromatographic peaks fragment ions specific precursor similar retention time peak shape precursor’s MS1 chromatographic peak. detection MS1 MS2 chromatographic peaks performed, can reconstruct MS2 spectra using reconstructChromPeakSpectra function. function defines MS2 spectrum MS1 chromatographic peak based following approach: Identify MS2 chromatographic peaks isolation window containing m/z ion (MS1 chromatographic peak) approximately retention time MS1 chromatographic peak (accepted difference retention time can defined diffRt parameter). Extract MS1 chromatographic peak MS2 chromatographic peaks identified previous step correlate peak shapes candidate MS2 chromatographic peaks shape MS1 peak. MS2 chromatographic peaks correlation coefficient larger minCor retained. Reconstruct MS2 spectrum using m/z selected MS2 chromatographic peaks intensity; MS2 chromatographic peak selected MS1 peak thus represent one mass peak reconstructed spectrum. illustrate process perform individual steps example fenamiphos (exact mass 303.105800777 m/z [M+H]+ adduct 304.113077). first step extract chromatographic peak ion. Next identify MS2 chromatographic peaks identified isolation window containing m/z fenamiphos. information isolation window chromatographic peak identified available chromPeakData. also require retention time MS2 chromatographic peaks similar retention time MS1 peak extract corresponding peak information. thus select chromatographic peaks retention time range contains retention time apex position MS1 chromatographic peak. total 24 MS2 chromatographic peaks match conditions. Next extract ion chromatogram MS1 peak selected candidate MS2 signals. ensure chromatograms extracted spectra correct isolation window need specify respective isolation window passing isolation window target m/z chromatogram function (addition setting msLevel = 2). can done either getting isolationWindowTargetMz spectra data subset using filterIsolationWindow (done ) selecting isolationWindowTargetMz closest m/z compound interest. target m/z isolation window containing m/z interest thus 299.1 can use chromatogram call extract data correct (MS2) spectra. can now plot extracted ion chromatogram MS1 extracted MS2 data. Extracted ion chromatograms Fenamiphos MS1 (blue) potentially related signal MS2 (grey). Next can calculate correlations peak shapes MS2 chromatogram MS1 peak. illustrate process example one MS2 chromatographic peaks. Note , MS1 MS2 spectra recorded consecutively, retention times individual data points differ MS2 MS1 chromatographic data data points thus matched (aligned) performing correlation analysis. done automatically correlate function. See help align method information alignment options. identifying MS2 chromatographic peaks shapes enough high similarity MS1 chromatographic peaks, MS2 spectrum reconstructed based m/z intensities MS2 chromatographic peaks (.e., using \"mz\" \"maxo\" \"\" values). Instead performing assignment MS2 signal MS1 chromatographic peaks manually , can use reconstructChromPeakSpectra function performs exact steps MS1 chromatographic peaks DIA data set. use function reconstruct MS2 spectra example data requiring peak shape correlation higher 0.9 candidate MS2 chromatographic peak target MS1 chromatographic peak. result got Spectra object length equal number MS1 peaks data. length spectrum represents number peaks contains. Thus, length 0 indicates matching peak (MS2 signal) found respective MS1 chromatographic peak. reconstructed spectra additional annotations available IDs MS2 chromatographic peaks spectrum reconstructed (\"ms2_peak_id\") well correlation coefficient chromatographic peak shape precursor’s shape (\"ms2_peak_cor\"). Metadata column \"peak_id\" contains ID MS1 chromatographic peak: next extract MS2 spectrum example peak likely representing [M+H]+ ions Fenamiphos using chromatographic peak ID: can now compare reconstructed spectrum example consensus spectrum DDA experiment previous section (variable ex_spectrum) well MS2 spectrum Fenamiphos Metlin (collision energy 10V). better visualization normalize also peak intensities reconstructed SWATH spectrum function used experimental DDA spectrum. Mirror plot comparing reconstructed MS2 spectrum Fenamiphos (upper panel) measured spectrum DDA data Fenamiphhos spectrum Metlin. wanted get EICs MS2 chromatographic peaks used generate MS2 spectrum can use IDs peaks provided $ms2_peak_id result spectrum. peak IDs available can extract retention time window m/z ranges chromPeaks matrix use chromatogram function extract EIC. Note however SWATH data MS2 signal different isolation windows. Thus first filter swath_data object isolation window containing precursor m/z filterIsolationWindow subset data MS2 spectra related ion interest. addition, use msLevel = 2L chromatogram call chromatogram extracts default data MS1 spectra need specify target m/z isolation window containing fragment data compound interest. row ms2_eics contains now EIC one MS2 chromatographic peaks. can also plot overlay plot. Overlay EICs chromatographic peaks used reconstruct MS2 spectrum fenamiphos. second example analyze signal [M+H]+ ion m/z 376.0381 (match Prochloraz). first identify MS1 chromatographic peak m/z retrieve reconstructed MS2 spectrum peak. MS2 spectrum (tentative) MS1 signal prochloraz reconstructed SWATH MS2 data thus 9 peaks. addition identify corresponding MS1 peak DDA data set, extract measured MS2 chromatographic peaks build consensus spectrum . retention times chromatographic peaks DDA SWATH data match almost perfectly. Next get MS2 spectra peak. total 5 spectra measured, relatively high number peaks. Next combine consensus spectrum. last load also Prochloraz MS2 spectra (different collision energies) Metlin. validate reconstructed spectrum plot corresponding DDA spectrum MS2 spectrum Prochloraz (collision energy 10V) Metlin. Mirror plot comparing reconstructed MS2 spectrum Prochloraz (upper panel) measured spectrum DDA data Prochloraz spectrum Metlin. spectra fit relatively well. Interestingly, peak representing precursor (right-peak) seems slightly shifted m/z value reconstructed spectrum. Also, closer inspecting spectrum two groups peaks small differences m/z can observed (see plot ). SWATH-derived MS2 spectrum prochloraz. represent fragments isotopes original compound. DIA MS2 data, since ions given retention time fragmented, can contain fragments isotopes. thus use isotopologues function MetaboCoreUtils package check presence potential isotope peaks reconstructed MS2 spectrum prochloraz. Indeed, peaks 3, 4 5 well 6 7 assigned group potential isotope peaks. proof peaks indeed fragment isotopes prochloraz highly likely (given difference m/z relative intensity differences). thus define function keeps monoisotopic peak isotope group spectrum. SWATH MS2 spectrum prochloraz (left) deisotoping (right). Removing isotope peaks SWATH MS2 spectrum increases also spectra similarity score (since reference spectra generally contain fragments ion interest, isotopes). Similar DDA data, reconstructed MS2 spectra SWATH data used annotation MS1 chromatographic peaks.","code":"fenamiphos_mz <- 304.113077 fenamiphos_ms1_peak <- chromPeaks(swath_data, mz = fenamiphos_mz, ppm = 2) fenamiphos_ms1_peak ##            mz    mzmin    mzmax      rt   rtmin   rtmax     into     intb ## CP34 304.1124 304.1121 304.1126 423.945 419.445 428.444 10697.34 10688.34 ##          maxo  sn sample ## CP34 2401.849 618      1 keep <- chromPeakData(swath_data)$isolationWindowLowerMz < fenamiphos_mz &         chromPeakData(swath_data)$isolationWindowUpperMz > fenamiphos_mz keep <- keep &     chromPeaks(swath_data)[, \"rtmin\"] < fenamiphos_ms1_peak[, \"rt\"] &     chromPeaks(swath_data)[, \"rtmax\"] > fenamiphos_ms1_peak[, \"rt\"]  fenamiphos_ms2_peak <- chromPeaks(swath_data)[which(keep), ] rtr <- fenamiphos_ms1_peak[, c(\"rtmin\", \"rtmax\")] mzr <- fenamiphos_ms1_peak[, c(\"mzmin\", \"mzmax\")] fenamiphos_ms1_chr <- chromatogram(swath_data, rt = rtr, mz = mzr)  rtr <- fenamiphos_ms2_peak[, c(\"rtmin\", \"rtmax\")] mzr <- fenamiphos_ms2_peak[, c(\"mzmin\", \"mzmax\")] ## Get the isolationWindowTargetMz for spectra containing the m/z of the ## compound of interest swath_data |> filterIsolationWindow(mz = fenamiphos_mz) |> spectra() |> isolationWindowTargetMz() |> table() ##  ## 299.1  ##   423 fenamiphos_ms2_chr <- chromatogram(     swath_data, rt = rtr, mz = mzr, msLevel = 2L,     isolationWindowTargetMz = rep(299.1, nrow(rtr))) plot(rtime(fenamiphos_ms1_chr[1, 1]),      intensity(fenamiphos_ms1_chr[1, 1]),      xlab = \"retention time [s]\", ylab = \"intensity\", pch = 16,      ylim = c(0, 5000), col = \"blue\", type = \"b\", lwd = 2) #' Add data from all MS2 peaks tmp <- lapply(fenamiphos_ms2_chr@.Data,               function(z) points(rtime(z), intensity(z),                                  col = \"#00000080\",                                  type = \"b\", pch = 16)) compareChromatograms(fenamiphos_ms2_chr[1, 1],                fenamiphos_ms1_chr[1, 1],                ALIGNFUNARGS = list(method = \"approx\")) ## [1] 0.9997871 swath_spectra <- reconstructChromPeakSpectra(swath_data, minCor = 0.9) swath_spectra ## MSn data (Spectra) with 62 spectra in a MsBackendMemory backend: ##        msLevel     rtime scanIndex ##      <integer> <numeric> <integer> ## CP01         2   239.458        NA ## CP02         2   240.358        NA ## CP03         2   329.577        NA ## CP04         2   329.771        NA ## CP05         2   346.164        NA ## ...        ...       ...       ... ## CP58         2   551.735        NA ## CP59         2   551.735        NA ## CP60         2   575.134        NA ## CP61         2   575.134        NA ## CP62         2   574.942        NA ##  ... 20 more variables/columns. ## Processing: ##  Merge 1 Spectra into one [Thu Dec  7 08:46:06 2023] lengths(swath_spectra) ##  [1]  0  0  1  1  1  0  0  0  0  0  0  0  3  0  3  4  0  3  0  1  0  9 14  1  0 ## [26]  0 15  4  1  1  2  4  6 15 12 11  2  4 13  0  0  0  0  1  2  0  1  0  0  0 ## [51]  3  0  2  1  7  7  0  0  0  0  0  2 spectraData(swath_spectra, c(\"peak_id\", \"ms2_peak_id\", \"ms2_peak_cor\")) ## DataFrame with 62 rows and 3 columns ##          peak_id ms2_peak_id      ms2_peak_cor ##      <character>      <list>            <list> ## CP01        CP01                               ## CP02        CP02                               ## CP03        CP03       CP063          0.950582 ## CP04        CP04       CP105           0.95157 ## CP05        CP05       CP153          0.924545 ## ...          ...         ...               ... ## CP58        CP58                               ## CP59        CP59                               ## CP60        CP60                               ## CP61        CP61                               ## CP62        CP62 CP334,CP329 0.918915,0.911944 fenamiphos_swath_spectrum <- swath_spectra[     swath_spectra$peak_id == rownames(fenamiphos_ms1_peak)] fenamiphos_swath_spectrum <- addProcessing(fenamiphos_swath_spectrum,                                            scale_fun) par(mfrow = c(1, 2)) plotSpectraMirror(fenamiphos_swath_spectrum, ex_spectrum,      ppm = 50, main = \"against DDA\") plotSpectraMirror(fenamiphos_swath_spectrum, fenamiphos[2],      ppm = 50, main = \"against Metlin\") pk_ids <- fenamiphos_swath_spectrum$ms2_peak_id[[1]] pk_ids ##  [1] \"CP199\" \"CP201\" \"CP211\" \"CP208\" \"CP200\" \"CP202\" \"CP217\" \"CP215\" \"CP205\" ## [10] \"CP212\" \"CP221\" \"CP223\" \"CP213\" \"CP207\" \"CP220\" rt_range <- chromPeaks(swath_data)[pk_ids, c(\"rtmin\", \"rtmax\")] mz_range <- chromPeaks(swath_data)[pk_ids, c(\"mzmin\", \"mzmax\")]  pmz <- precursorMz(fenamiphos_swath_spectrum)[1] ## Determine the isolation window target m/z tmz <- swath_data |> filterIsolationWindow(mz = pmz) |> spectra() |> isolationWindowTargetMz() |> unique()  ms2_eics <- chromatogram(     swath_data, rt = rt_range, mz = mz_range, msLevel = 2L,     isolationWindowTargetMz = rep(tmz, nrow(rt_range))) plotChromatogramsOverlay(ms2_eics) prochloraz_mz <- 376.0381  prochloraz_ms1_peak <- chromPeaks(swath_data, msLevel = 1L,                                   mz = prochloraz_mz, ppm = 5) prochloraz_ms1_peak ##            mz   mzmin    mzmax      rt   rtmin   rtmax     into     intb ## CP22 376.0373 376.037 376.0374 405.046 401.446 409.546 3664.051 3655.951 ##          maxo  sn sample ## CP22 897.3923 278      1 prochloraz_swath_spectrum <- swath_spectra[     swath_spectra$peak_id == rownames(prochloraz_ms1_peak)] lengths(prochloraz_swath_spectrum) ## [1] 9 prochloraz_dda_peak <- chromPeaks(dda_data, msLevel = 1L,                                   mz = prochloraz_mz, ppm = 5) prochloraz_dda_peak ##             mz    mzmin    mzmax      rt   rtmin   rtmax     into    intb ## CP034 376.0385 376.0378 376.0391 405.295 401.166 410.145 5082.157 5072.77 ##           maxo  sn sample ## CP034 1350.633 310      1 prochloraz_dda_spectra <- dda_spectra[     dda_spectra$peak_id == rownames(prochloraz_dda_peak)] prochloraz_dda_spectra ## MSn data (Spectra) with 5 spectra in a MsBackendMzR backend: ##     msLevel     rtime scanIndex ##   <integer> <numeric> <integer> ## 1         2   401.438      3253 ## 2         2   402.198      3259 ## 3         2   404.677      3306 ## 4         2   405.127      3316 ## 5         2   405.877      3325 ##  ... 34 more variables/columns. ##  ## file(s): ## PestMix1_DDA.mzML ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Thu Dec  7 08:45:51 2023] ##  Filter: select MS level(s) 2 [Thu Dec  7 08:45:58 2023] ##  Merge 1 Spectra into one [Thu Dec  7 08:45:58 2023] prochloraz_dda_spectrum <- combineSpectra(     prochloraz_dda_spectra, FUN = combinePeaks, ppm = 20,     peaks = \"intersect\", minProp = 0.8, intensityFun = median, mzFun = median,     f = prochloraz_dda_spectra$peak_id) ## Backend of the input object is read-only, will change that to an 'MsBackendMemory' ## Warning: 'combinePeaks' for lists of peak matrices is deprecated; please use ## 'combinePeaksData' instead. prochloraz <- Spectra(     system.file(\"mgf\", \"metlin-68898.mgf\", package = \"xcms\"),     source = MsBackendMgf()) ## Start data import from 1 files ... done prochloraz_swath_spectrum <- addProcessing(prochloraz_swath_spectrum, scale_fun) prochloraz_dda_spectrum <- addProcessing(prochloraz_dda_spectrum, scale_fun)  par(mfrow = c(1, 2)) plotSpectraMirror(prochloraz_swath_spectrum, prochloraz_dda_spectrum,                   ppm = 40, main = \"against DDA\") plotSpectraMirror(prochloraz_swath_spectrum, prochloraz[2],                   ppm = 40, main = \"against Metlin\") plotSpectra(prochloraz_swath_spectrum) library(MetaboCoreUtils) isotopologues(peaksData(prochloraz_swath_spectrum)[[1]]) ## [[1]] ## [1] 3 4 5 ##  ## [[2]] ## [1] 6 7 ## Function to keep only the first (monoisotopic) peak for potential ## isotopologue peak groups. rem_iso <- function(x, ...) {     idx <- isotopologues(x)     idx <- unlist(lapply(idx, function(z) z[-1]), use.names = FALSE)     if (length(idx))         x[-idx, , drop = FALSE]     else x } prochloraz_swath_spectrum2 <- addProcessing(prochloraz_swath_spectrum,                                             rem_iso) par(mfrow = c(1, 2)) plotSpectra(prochloraz_swath_spectrum) plotSpectra(prochloraz_swath_spectrum2) compareSpectra(prochloraz_swath_spectrum, prochloraz_dda_spectrum) ## [1] 0.4623719 compareSpectra(prochloraz_swath_spectrum2, prochloraz_dda_spectrum) ## [1] 0.5932303"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"outlook","dir":"Articles","previous_headings":"","what":"Outlook","title":"LC-MS/MS data analysis with xcms","text":"Currently, spectra data representation, handling processing re-implemented part RforMassSpectrometry initiative aiming increasing performance methods simplifying use. Thus, parts workflow described changed (improved) future. Along developments, improved matching strategies larger data sets implemented well functionality compare Spectra directly reference MS2 spectra public annotation resources (e.g. Massbank HMDB). See example information. Regarding SWATH data analysis, future development involve improved selection correct MS2 chromatographic peaks considering also correlation intensity values across several samples.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"LC-MS/MS data analysis with xcms","text":"","code":"sessionInfo() ## R Under development (unstable) (2023-11-22 r85609) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] MetaboCoreUtils_1.11.0 MsBackendMgf_1.11.0    MsExperiment_1.5.1     ##  [4] pander_0.6.5           Spectra_1.13.2         xcms_4.1.4             ##  [7] MSnbase_2.29.1         ProtGenerics_1.35.0    S4Vectors_0.41.2       ## [10] mzR_2.37.0             Rcpp_1.0.11            Biobase_2.63.0         ## [13] BiocGenerics_0.49.1    BiocParallel_1.37.0    BiocStyle_2.31.0       ##  ## loaded via a namespace (and not attached): ##   [1] bitops_1.0-7                rlang_1.1.2                 ##   [3] magrittr_2.0.3              clue_0.3-65                 ##   [5] MassSpecWavelet_1.69.0      matrixStats_1.1.0           ##   [7] compiler_4.4.0              systemfonts_1.0.5           ##   [9] vctrs_0.6.5                 stringr_1.5.1               ##  [11] pkgconfig_2.0.3             crayon_1.5.2                ##  [13] fastmap_1.1.1               XVector_0.43.0              ##  [15] utf8_1.2.4                  rmarkdown_2.25              ##  [17] preprocessCore_1.65.0       ragg_1.2.6                  ##  [19] purrr_1.0.2                 MultiAssayExperiment_1.29.0 ##  [21] xfun_0.41                   zlibbioc_1.49.0             ##  [23] cachem_1.0.8                GenomeInfoDb_1.39.1         ##  [25] jsonlite_1.8.8              progress_1.2.3              ##  [27] highr_0.10                  DelayedArray_0.29.0         ##  [29] prettyunits_1.2.0           parallel_4.4.0              ##  [31] cluster_2.1.6               R6_2.5.1                    ##  [33] bslib_0.6.1                 stringi_1.8.2               ##  [35] RColorBrewer_1.1-3          limma_3.59.1                ##  [37] GenomicRanges_1.55.1        jquerylib_0.1.4             ##  [39] bookdown_0.37               SummarizedExperiment_1.33.1 ##  [41] iterators_1.0.14            knitr_1.45                  ##  [43] IRanges_2.37.0              igraph_1.5.1                ##  [45] splines_4.4.0               Matrix_1.6-4                ##  [47] tidyselect_1.2.0            abind_1.4-5                 ##  [49] yaml_2.3.7                  doParallel_1.0.17           ##  [51] codetools_0.2-19            affy_1.81.0                 ##  [53] lattice_0.22-5              tibble_3.2.1                ##  [55] plyr_1.8.9                  evaluate_0.23               ##  [57] survival_3.5-7              desc_1.4.2                  ##  [59] pillar_1.9.0                affyio_1.73.0               ##  [61] BiocManager_1.30.22         MatrixGenerics_1.15.0       ##  [63] foreach_1.5.2               MALDIquant_1.22.1           ##  [65] ncdf4_1.22                  generics_0.1.3              ##  [67] rprojroot_2.0.4             RCurl_1.98-1.13             ##  [69] hms_1.1.3                   ggplot2_3.4.4               ##  [71] munsell_0.5.0               scales_1.3.0                ##  [73] glue_1.6.2                  lazyeval_0.2.2              ##  [75] MsFeatures_1.11.0           tools_4.4.0                 ##  [77] mzID_1.41.0                 robustbase_0.99-1           ##  [79] QFeatures_1.13.1            vsn_3.71.0                  ##  [81] RANN_2.6.1                  fs_1.6.3                    ##  [83] XML_3.99-0.16               grid_4.4.0                  ##  [85] impute_1.77.0               MsCoreUtils_1.15.1          ##  [87] colorspace_2.1-0            GenomeInfoDbData_1.2.11     ##  [89] cli_3.6.1                   textshaping_0.3.7           ##  [91] fansi_1.0.5                 S4Arrays_1.3.1              ##  [93] dplyr_1.1.4                 AnnotationFilter_1.27.0     ##  [95] pcaMethods_1.95.0           gtable_0.3.4                ##  [97] DEoptimR_1.1-3              sass_0.4.8                  ##  [99] digest_0.6.33               SparseArray_1.3.1           ## [101] multtest_2.59.0             memoise_2.0.1               ## [103] htmltools_0.5.7             pkgdown_2.0.7.9000          ## [105] lifecycle_1.0.4             statmod_1.5.0               ## [107] MASS_7.3-60.1"},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"LC-MS data pre-processing and analysis with xcms","text":"xcms package provides functionality perform pre-processing LC-MS, GC-MS LC-MS/MS data raw signals mzML, mzXML CDF files processed feature abundances. pre-processing includes chromatographic peak detection, sample alignment correspondence analysis. first version package already published 2006 [1] since updated modernized several rounds better integrate R-based packages analysis untargeted metabolomics data. includes version 3 xcms used MSnbase package MS data representation [2]. recent update (xcms version 4) enables addition pre-processing MS data represented modern MsExperiment Spectra packages provides even better integration RforMassSpectrometry R package ecosystem simplifying e.g. also compound annotation [3]. document describes data import, exploration pre-processing simple test LC-MS data set xcms package version >= 4. functions can applied older MSnbase-based workflows (xcms version 3). Additional documents tutorials covering also topics untargeted metabolomics analysis listed end document.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"data-import","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Data import","title":"LC-MS data pre-processing and analysis with xcms","text":"xcms supports analysis LC-MS(/MS) data can imported Spectra package. data typically provided (AIA/ANDI) NetCDF, mzXML mzML format can, dedicated extensions Spectra package, also imported sources, e.g. also directly raw data files manufacturer’s formats. demonstration purpose analyze document small subset data [4] metabolic consequences knock-fatty acid amide hydrolase (FAAH) gene mice investigated. raw data files (NetCDF format) provided faahKO data package. data set consists samples spinal cords 6 knock-6 wild-type mice. file contains data centroid mode acquired positive ion polarity 200-600 m/z 2500-4500 seconds. speed-processing vignette restrict analysis 8 files. load required packages, locate raw CDF files within faahKO package build phenodata data.frame describing experimental setup. Generally, data frames contain relevant experimental variables sample descriptions (including also names raw data files) imported R using either read.table function (file csv tabulator delimited text file format) also using functions readxl R package Excel file format. next load data using readMsExperiment function MsExperiment package. MS spectra data experiment now available Spectra object within faahko. Note MsExperiment container addition spectra data also contain types data also references files. See vignette MsExperiment details. Also, loading data mzML, mzXML CDF files, default general spectra data loaded memory actual peaks data, .e. m/z intensity values retrieved --fly raw files needed (similar MSnbase -disk mode described [2]). guarantees low memory footprint hence allowing analyze also large experiments without need high performance computing environments. Note also different alternative backends (hence data representations) used Spectra object within faahko eventually even lower memory footprint, higher performance. See package vignette Spectra package SpectraTutorials tutorial details Spectra backends change .","code":"library(xcms) library(faahKO) library(RColorBrewer) library(pander) library(pheatmap) library(MsExperiment)  ## Get the full path to the CDF files cdfs <- dir(system.file(\"cdf\", package = \"faahKO\"), full.names = TRUE,             recursive = TRUE)[c(1, 2, 5, 6, 7, 8, 11, 12)] ## Create a phenodata data.frame pd <- data.frame(sample_name = sub(basename(cdfs), pattern = \".CDF\",                                    replacement = \"\", fixed = TRUE),                  sample_group = c(rep(\"KO\", 4), rep(\"WT\", 4)),                  stringsAsFactors = FALSE) faahko <- readMsExperiment(spectraFiles = cdfs, sampleData = pd) faahko ## Object of class MsExperiment  ##  Spectra: MS1 (10224)  ##  Experiment data: 8 sample(s) ##  Sample data links: ##   - spectra: 8 sample(s) to 10224 element(s)."},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"initial-data-inspection","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Initial data inspection","title":"LC-MS data pre-processing and analysis with xcms","text":"MsExperiment object simple flexible container MS experiments. raw MS data stored Spectra object can accessed spectra function. spectra organized sequentially (.e., file) fromFile function can used get spectrum information data files belongs. simply count number spectra per file. Information samples can retrieved sampleData function. row DataFrame represents one sample (input file). Using [ possible subset MsExperiment object sample. subset faahko 3rd sample (file) access spectra sample data. first evaluation data plot base peak chromatogram (BPC) file experiment. use chromatogram method set aggregationFun \"max\" return spectrum maximal intensity hence create BPC raw data. create total ion chromatogram set aggregationFun \"sum\". Base peak chromatogram. chromatogram method returned MChromatograms object organizes individual Chromatogram objects (fact contain chromatographic data) two-dimensional array: columns represent samples rows (optionally) m/z /retention time ranges. extract chromatogram first sample access retention time intensity values. BPC seems around 4200 seconds signal measured anymore. Thus, filter full data set retention time range 2550 4250 seconds using filterRt function. Note present subset spectra within MsExperiment. Subsequently re-create also BPC. next create boxplots representing distribution total ion currents per data file. plots can useful spot potentially problematic MS runs. extract information, use tic function Spectra object within faahko split values file using fromFile. Distribution total ion currents per file. addition, can also cluster samples based similarity base peak chromatograms. Samples thus grouped based similarity LC runs. need however bin data along retention time axis, since retention times generally differ samples. use bin function BPC bin intensities 2 second wide retention time bins. clustering performed using complete linkage hierarchical clustering pairwise correlations binned base peak chromatograms. Grouping samples based similarity base peak chromatogram. samples cluster pairwise manner, KO WT samples sample index similar BPC.","code":"spectra(faahko) ## MSn data (Spectra) with 10224 spectra in a MsBackendMzR backend: ##         msLevel     rtime scanIndex ##       <integer> <numeric> <integer> ## 1             1   2501.38         1 ## 2             1   2502.94         2 ## 3             1   2504.51         3 ## 4             1   2506.07         4 ## 5             1   2507.64         5 ## ...         ...       ...       ... ## 10220         1   4493.56      1274 ## 10221         1   4495.13      1275 ## 10222         1   4496.69      1276 ## 10223         1   4498.26      1277 ## 10224         1   4499.82      1278 ##  ... 33 more variables/columns. ##  ## file(s): ## ko15.CDF ## ko16.CDF ## ko21.CDF ##  ... 5 more files table(fromFile(faahko)) ##  ##    1    2    3    4    5    6    7    8  ## 1278 1278 1278 1278 1278 1278 1278 1278 sampleData(faahko) ## DataFrame with 8 rows and 3 columns ##   sample_name sample_group spectraOrigin ##   <character>  <character>   <character> ## 1        ko15           KO /__w/_temp... ## 2        ko16           KO /__w/_temp... ## 3        ko21           KO /__w/_temp... ## 4        ko22           KO /__w/_temp... ## 5        wt15           WT /__w/_temp... ## 6        wt16           WT /__w/_temp... ## 7        wt21           WT /__w/_temp... ## 8        wt22           WT /__w/_temp... faahko_3 <- faahko[3] spectra(faahko_3) ## MSn data (Spectra) with 1278 spectra in a MsBackendMzR backend: ##        msLevel     rtime scanIndex ##      <integer> <numeric> <integer> ## 1            1   2501.38         1 ## 2            1   2502.94         2 ## 3            1   2504.51         3 ## 4            1   2506.07         4 ## 5            1   2507.64         5 ## ...        ...       ...       ... ## 1274         1   4493.56      1274 ## 1275         1   4495.13      1275 ## 1276         1   4496.69      1276 ## 1277         1   4498.26      1277 ## 1278         1   4499.82      1278 ##  ... 33 more variables/columns. ##  ## file(s): ## ko21.CDF sampleData(faahko_3) ## DataFrame with 1 row and 3 columns ##   sample_name sample_group spectraOrigin ##   <character>  <character>   <character> ## 1        ko21           KO /__w/_temp... ## Get the base peak chromatograms. This reads data from the files. bpis <- chromatogram(faahko, aggregationFun = \"max\") ## Define colors for the two groups group_colors <- paste0(brewer.pal(3, \"Set1\")[1:2], \"60\") names(group_colors) <- c(\"KO\", \"WT\")  ## Plot all chromatograms. plot(bpis, col = group_colors[sampleData(faahko)$sample_group]) bpi_1 <- bpis[1, 1] rtime(bpi_1) |> head() ## [1] 2501.378 2502.943 2504.508 2506.073 2507.638 2509.203 intensity(bpi_1) |> head() ## [1] 43888 43960 43392 42632 42200 42288 faahko <- filterRt(faahko, rt = c(2550, 4250)) ## Filter spectra ## creating the BPC on the subsetted data bpis <- chromatogram(faahko, aggregationFun = \"max\") ## Get the total ion current by file tc <- spectra(faahko) |>     tic() |>     split(f = fromFile(faahko)) boxplot(tc, col = group_colors[sampleData(faahko)$sample_group],         ylab = \"intensity\", main = \"Total ion current\") ## Bin the BPC bpis_bin <- bin(bpis, binSize = 2)  ## Calculate correlation on the log2 transformed base peak intensities cormat <- cor(log2(do.call(cbind, lapply(bpis_bin, intensity)))) colnames(cormat) <- rownames(cormat) <- bpis_bin$sample_name  ## Define which phenodata columns should be highlighted in the plot ann <- data.frame(group = bpis_bin$sample_group) rownames(ann) <- bpis_bin$sample_name  ## Perform the cluster analysis pheatmap(cormat, annotation = ann,          annotation_color = list(group = group_colors))"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"chromatographic-peak-detection","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Chromatographic peak detection","title":"LC-MS data pre-processing and analysis with xcms","text":"Chromatographic peak detection aims identifying signal sample created ions originating compound species. Chromatographic peak detection can performed xcms findChromPeaks function parameter object defines configures algorithm used (see ?findChromPeaks list supported algorithms). running peak detection however strongly suggested first visually inspect extracted ion chromatogram e.g. internal standards compounds known present samples order evaluate adapt settings peak detection algorithm since default settings appropriate LC-MS setups. extract EIC one compound using chromatogram function specifying addition m/z retention time range expect signal compound. Extracted ion chromatogram one peak. Note Chromatogram objects extracted chromatogram method contain NA value certain scan (.e. spectrum specific retention time) signal measured respective m/z range. reflected lines drawn continuous lines plot . peak thus width 50 seconds. can use information define peakwidth parameter centWave peak detection method [5] use chromatographic peak detection data. peakwidth parameter allows define expected lower upper width (retention time dimension) chromatographic peaks. data set thus peakwidth = c(20, 80). second important parameter centWave ppm defines expected maximum deviation m/z values centroids included one chromatographic peak (note precision m/z values provided MS instrument manufacturer). ppm parameter extract full MS data (intensity, retention time m/z values) corresponding peak. end first filter raw object retention time, m/z finally plot object type = \"XIC\" produce plot . use pipe (|>) operator better illustrate corresponding workflow. Visualization raw MS data one peak. plot: upper panel: chromatogram plotting intensity values retention time, lower panel m/z retention time plot. individual data points colored according intensity. present data actually variation m/z values. Usually m/z values individual centroids (lower panel) plots scatter around real m/z value compound (intensity dependent manner). first step centWave algorithm defines regions interest (ROI) subsequently screened presence chromatographic peaks. ROIs defined based difference m/z values centroids consecutive scans (spectra). detail, centroids consecutive scans included ROI difference m/z mean m/z ROI smaller user defined ppm parameter. reasonable choice ppm thus maximal m/z difference data points neighboring scans/spectra part chromatographic peak internal standard known compound. suggested inspect ranges m/z values several compounds (either internal standards compounds known present sample) define ppm parameter centWave according . See also tutorial additional information examples choosing testing peak detection settings. Chromatographic peak detection can also performed extracted ion chromatograms, helps testing tuning peak detection settings. Note however peak detection EICs involve first step centWave described thus consider ppm parameter. Also, since less data available algorithms, background signal estimation performed differently different settings snthresh need used (generally lower snthresh used EICs since estimated background signal tends higher data subsets full data). perform peak detection findChromPeaks function EIC generated . submitted parameter object defines algorithm used allows define settings algorithm. use CentWaveParam parameter object use configure centWave algorithm default settings, except snthresh. can access identified chromatographic peaks chromPeaks function. Parallel chromPeaks matrix also chromPeakData data frame allows add arbitrary annotations chromatographic peak, e.g. MS level peak detected: plot EIC along identified chromatographic peaks using plot function result object . Additional parameters peakCol peakBg allow define foreground background (fill) color identified chromatographic peak provided result object (.e., need define one color row chromPeaks(xchr) - column \"column\" (\"sample\" present) peak matrix specifies sample peak identified). Signal example peak. Red blue colors represent KO wild type samples, respectively. Peak area identified chromatographic peaks highlighted sample group color. happy settings can use peak detection full data set (.e. MsExperiment object data whole experiment). Note set argument prefilter c(6, 5000) noise 5000 reduce run time vignette. setting consider ROIs least 6 centroids intensity larger 5000 centWave chromatographic peak detection. results findChromPeaks MsExperiment object returned XcmsExperiment object. object extends MsExperiment directly (hence providing access raw data) contains xcms pre-processing results. Note also additional rounds chromatographic peak detections performed results added existing peak detection results additional calls findChromPeaks result object using parameter add = TRUE. chromPeaks function can also used access results chromatographic peak detection. show first 6 identified chromatographic peaks. Columns chromPeaks matrix might differ depending used peak detection algorithm. Columns algorithms provide : \"mz\", \"mzmin\", \"mzmax\", \"rt\", \"rtmin\" \"rtmax\" define m/z retention time range chromatographic peak (.e. mass peaks within area used integrate peak signal) well m/z retention time peak apex. mandatory columns \"\" \"maxo\" absolute integrated peak signal maximum peak intensity. Finally, \"sample\" provides index sample peak identified. Additional annotations individual peak can extracted chromPeakData function. data frame also used add/store arbitrary annotations detected peak (don’t necessarily need numeric). Peak detection always work perfectly types peak shapes present data set leading peak detection artifacts, (partially completely) overlapping peaks artificially split peaks (common issues especially centWave). xcms provides refineChromPeaks function can called peak detection results order refine (clean) peak detection results either removing identified peaks passing certain criteria merging artificially split partially completely overlapping chromatographic peaks. Different algorithms available can configured respective parameter objects: CleanPeaksParam FilterIntensityParam allow remove peaks retention time range intensity specified threshold, respectively. MergeNeighboringPeaksParam possible merge chromatographic peaks hence remove many mentioned (centWave) peak detection artifacts. See also ?refineChromPeaks information help different methods. post-process peak detection results merging peaks overlap 4 second window per file signal lower 75% smaller peak’s maximal intensity. See ?MergeNeighboringPeaksParam help page detailed description settings approach. example merged peak given . Result peak refinement step. Left: data processing, right: refinement. splitted peak merged one. centWave thus detected originally 3 chromatographic peaks m/z slice (left panel plot ) peak refinement MergeNeighboringPeaksParam specified settings merged first two peaks single one (right panel figure ). close peaks, lower intensity , however merged (see ). Result peak refinement step. Left: data processing, right: refinement. peaks merged. also possible perform peak refinement extracted ion chromatograms. used test fine-tune settings parameter. illustrate perform peak refinement extracted ion chromatogram chr_1 reducing minProp parameter force joining two peaks.  proceeding next replace faahko object results peak refinement step. use data chromPeaks matrix calculate per-file summaries peak detection results, number peaks per file well distribution retention time widths. Summary statistics identified chromatographic peaks. Shown number identified peaks per sample widths/duration chromatographic peaks. default chromPeaks return identified chromatographic peaks result object also possible extract chromatographic peaks specified m/z /rt range: can also plot location identified chromatographic peaks m/z - retention time space one file using plotChromPeaks function. plot information third sample. Identified chromatographic peaks m/z retention time space one sample. general overview peak detection results can addition visualize number identified chromatographic peaks per file along retention time axis. Parameter binSize allows define width bins rt dimension peaks counted. number chromatographic peaks within bin shown color-coded resulting plot. Frequency identified chromatographic peaks along retention time axis. frequency color coded higher frequency represented yellow-white. line shows peak frequency one file. Note extracting ion chromatorams xcms result object addition chromatographic data also extract identified chromatographic peaks within region. can thus also used validate verify used peak detection settings identified e.g. peaks known compounds internal standards properly. extract ion chromatogram m/z - rt region access detected peaks region using chromPeaks function. can also plot extracted ion chromatogram also visualize identified chromatographic peaks region. Signal example peak. Red blue colors represent KO wild type samples, respectively. signal area identified chromatographic peaks filled color. Chromatographic peaks can visualized also ways: setting peakType = \"rectangle\" indicated rectangle instead default highlighting option (peakType = \"polygon\") used . third alternative also possible just indicate apex position identified chromatographic peak single point (peakType = \"point\"). plot data using peakType = \"rectangle\". Signal example peak. Red blue colors represent KO wild type samples, respectively. rectangles indicate identified chromatographic peaks per sample. Finally plot also distribution peak intensity per sample. allows investigate whether systematic differences peak signals samples present. Peak intensity distribution per sample. signal identified chromatographic peaks comparable across samples, exception samples 3, 4 , degree, also 7 show slightly higher average intensities, also lower number detected peaks (indicated smaller width boxes). Note addition described identification chromatographic peaks, also possible manually define add chromatographic peaks manualChromPeaks function (see ?manualChromPeaks help page information).","code":"## Define the rt and m/z range of the peak area rtr <- c(2700, 2900) mzr <- c(334.9, 335.1) ## extract the chromatogram chr_raw <- chromatogram(faahko, mz = mzr, rt = rtr) plot(chr_raw, col = group_colors[chr_raw$sample_group]) faahko |> filterRt(rt = rtr) |> filterMz(mz = mzr) |> plot(type = \"XIC\") xchr <- findChromPeaks(chr_raw, param = CentWaveParam(snthresh = 2)) chromPeaks(xchr) ##             rt    rtmin    rtmax        into        intb  maxo   sn row column ##  [1,] 2781.505 2761.160 2809.674  412134.255  355516.374 16856   13   1      1 ##  [2,] 2786.199 2764.290 2812.803 1496244.206 1391821.332 58736   20   1      2 ##  [3,] 2734.556 2714.211 2765.855   21579.367   18449.428   899    4   1      3 ##  [4,] 2797.154 2775.245 2815.933  159058.782  150289.314  6295   12   1      3 ##  [5,] 2784.635 2761.160 2808.109   54947.545   37923.532  2715    2   1      4 ##  [6,] 2859.752 2845.668 2878.532   13895.212   13874.868   905  904   1      4 ##  [7,] 2897.311 2891.051 2898.876    5457.155    5450.895   995  994   1      4 ##  [8,] 2819.064 2808.109 2834.713   19456.914   19438.134  1347 1576   1      4 ##  [9,] 2789.329 2762.725 2808.109  174473.311   91114.975  8325    3   1      5 ## [10,] 2786.199 2764.290 2812.803  932645.211  569236.246 35856    2   1      6 ## [11,] 2792.461 2768.987 2823.760  876585.527  511324.134 27200    2   1      7 ## [12,] 2789.329 2773.680 2806.544   89582.569   73871.386  5427    6   1      8 chromPeakData(xchr) ## DataFrame with 12 rows and 4 columns ##      ms_level is_filled       row    column ##     <integer> <logical> <integer> <integer> ## 1           1     FALSE         1         1 ## 2           1     FALSE         1         2 ## 3           1     FALSE         1         3 ## 4           1     FALSE         1         3 ## 5           1     FALSE         1         4 ## ...       ...       ...       ...       ... ## 8           1     FALSE         1         4 ## 9           1     FALSE         1         5 ## 10          1     FALSE         1         6 ## 11          1     FALSE         1         7 ## 12          1     FALSE         1         8 ## Define a color for each sample sample_colors <- group_colors[xchr$sample_group] ## Define the background color for each chromatographic peak bg <- sample_colors[chromPeaks(xchr)[, \"column\"]] ## Parameter `col` defines the color of each sample/line, `peakBg` of each ## chromatographic peak. plot(xchr, col = sample_colors, peakBg = bg) cwp <- CentWaveParam(peakwidth = c(20, 80), noise = 5000,                      prefilter = c(6, 5000)) faahko <- findChromPeaks(faahko, param = cwp) chromPeaks(faahko) |> head() ##           mz mzmin mzmax       rt    rtmin    rtmax     into     intb  maxo sn ## CP0001 594.0 594.0 594.0 2601.535 2581.191 2637.529 161042.2 146073.3  7850 11 ## CP0002 577.0 577.0 577.0 2604.665 2581.191 2626.574 136105.2 128067.9  6215 11 ## CP0003 307.0 307.0 307.0 2618.750 2592.145 2645.354 284782.4 264907.0 16872 20 ## CP0004 302.0 302.0 302.0 2617.185 2595.275 2640.659 687146.6 669778.1 30552 43 ## CP0005 370.1 370.1 370.1 2673.523 2643.789 2700.127 449284.6 417225.3 25672 17 ## CP0006 427.0 427.0 427.0 2675.088 2643.789 2684.478 283334.7 263943.2 11025 13 ##        sample ## CP0001      1 ## CP0002      1 ## CP0003      1 ## CP0004      1 ## CP0005      1 ## CP0006      1 chromPeakData(faahko) ## DataFrame with 2908 rows and 2 columns ##         ms_level is_filled ##        <integer> <logical> ## CP0001         1     FALSE ## CP0002         1     FALSE ## CP0003         1     FALSE ## CP0004         1     FALSE ## CP0005         1     FALSE ## ...          ...       ... ## CP2904         1     FALSE ## CP2905         1     FALSE ## CP2906         1     FALSE ## CP2907         1     FALSE ## CP2908         1     FALSE mpp <- MergeNeighboringPeaksParam(expandRt = 4) faahko_pp <- refineChromPeaks(faahko, mpp) mzr_1 <- 305.1 + c(-0.01, 0.01) chr_1 <- chromatogram(faahko[1], mz = mzr_1) chr_2 <- chromatogram(faahko_pp[1], mz = mzr_1) par(mfrow = c(1, 2)) plot(chr_1) plot(chr_2) mzr_1 <- 496.2 + c(-0.01, 0.01) chr_1 <- chromatogram(faahko[1], mz = mzr_1) chr_2 <- chromatogram(faahko_pp[1], mz = mzr_1) par(mfrow = c(1, 2)) plot(chr_1) plot(chr_2) res <- refineChromPeaks(chr_1, MergeNeighboringPeaksParam(minProp = 0.05)) chromPeaks(res) ##         mz  mzmin  mzmax       rt    rtmin    rtmax     into intb    maxo  sn ## CPM1 496.2 496.19 496.21 3384.012 3294.809 3412.181 45940118   NA 1128960 177 ##      sample row column ## CPM1      1   1      1 plot(res) faahko <- faahko_pp summary_fun <- function(z)     c(peak_count = nrow(z), rt = quantile(z[, \"rtmax\"] - z[, \"rtmin\"]))  T <- chromPeaks(faahko) |>     split.data.frame(f = chromPeaks(faahko)[, \"sample\"]) |>     lapply(FUN = summary_fun) |>     do.call(what = rbind) rownames(T) <- basename(fileNames(faahko)) pandoc.table(     T,     caption = paste0(\"Summary statistics on identified chromatographic\",                      \" peaks. Shown are number of identified peaks per\",                      \" sample and widths/duration of chromatographic \",                      \"peaks.\")) chromPeaks(faahko, mz = c(334.9, 335.1), rt = c(2700, 2900)) ##         mz mzmin mzmax       rt    rtmin    rtmax      into      intb  maxo sn ## CP0038 335   335   335 2781.505 2761.160 2809.674  412134.3  383167.4 16856 23 ## CP0494 335   335   335 2786.199 2764.290 2812.803 1496244.2 1461187.3 58736 72 ## CP1025 335   335   335 2797.154 2775.245 2815.933  159058.8  149229.6  6295 13 ## CP1964 335   335   335 2786.199 2764.290 2812.803  932645.2  915333.8 35856 66 ## CP2349 335   335   335 2792.461 2768.987 2823.760  876585.5  848569.1 27200 36 ##        sample ## CP0038      1 ## CP0494      2 ## CP1025      3 ## CP1964      6 ## CP2349      7 plotChromPeaks(faahko, file = 3) plotChromPeakImage(faahko, binSize = 10) chr_ex <- chromatogram(faahko, mz = mzr, rt = rtr) chromPeaks(chr_ex) ##         mz mzmin mzmax       rt    rtmin    rtmax      into      intb  maxo sn ## CP0038 335   335   335 2781.505 2761.160 2809.674  412134.3  383167.4 16856 23 ## CP0494 335   335   335 2786.199 2764.290 2812.803 1496244.2 1461187.3 58736 72 ## CP1025 335   335   335 2797.154 2775.245 2815.933  159058.8  149229.6  6295 13 ## CP1964 335   335   335 2786.199 2764.290 2812.803  932645.2  915333.8 35856 66 ## CP2349 335   335   335 2792.461 2768.987 2823.760  876585.5  848569.1 27200 36 ##        sample row column ## CP0038      1   1      1 ## CP0494      2   1      2 ## CP1025      3   1      3 ## CP1964      6   1      6 ## CP2349      7   1      7 sample_colors <- group_colors[chr_ex$sample_group] plot(chr_ex, col = group_colors[chr_raw$sample_group], lwd = 2,      peakBg = sample_colors[chromPeaks(chr_ex)[, \"sample\"]]) plot(chr_ex, col = sample_colors, peakType = \"rectangle\",      peakCol = sample_colors[chromPeaks(chr_ex)[, \"sample\"]],      peakBg = NA) ## Extract a list of per-sample peak intensities (in log2 scale) ints <- split(log2(chromPeaks(faahko)[, \"into\"]),               f = chromPeaks(faahko)[, \"sample\"]) boxplot(ints, varwidth = TRUE, col = sample_colors,         ylab = expression(log[2]~intensity), main = \"Peak intensities\") grid(nx = NA, ny = NULL)"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"alignment","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Alignment","title":"LC-MS data pre-processing and analysis with xcms","text":"time analytes elute chromatography can vary samples (even compounds). differences already visible BPC even extracted ion chromatogram plot previous section. alignment step, also referred retention time correction, aims adjust differences shifting signals along retention time axis aligning different samples within experiment. plethora alignment algorithms exist (see [6]), also implemented xcms. Alignment LC-MS data can performed xcms using adjustRtime method algorithm-specific parameter class (see ?adjustRtime overview available methods xcms). example use obiwarp method [7] align samples. use binSize = 0.6 creates warping functions m/z bins 0.6. Also advisable modify adapt settings experiment. Note adjustRtime, besides calculating adjusted retention times spectrum, adjusts also retention times identified chromatographic peaks xcms result object. Adjusted retention times individual spectra can extracted result object using either adjustedRtime function using rtime parameter adjusted = TRUE (default): evaluate impact alignment plot BPC adjusted data. addition plot also differences adjusted raw retention times per sample using plotAdjustedRtime function. disable automatic extraction identified chromatographic peaks chromatogram function (make much sense BPC) use chromPeaks = \"none\" . Obiwarp aligned data. Base peak chromatogram (top) alignment (middle) difference adjusted raw retention times along retention time axis (bottom). large differences adjusted raw retention times indicate poorly performing samples alignment. last evaluate also impact alignment test peak. Example extracted ion chromatogram (top) alignment (bottom). Note: xcms result objects (XcmsExperiment well XCMSnExp) contain raw adjusted retention times spectra subset operation many cases drop adjusted retention times. Thus might sometimes useful immediately replace raw retention times data using applyAdjustedRtime function.","code":"faahko <- adjustRtime(faahko, param = ObiwarpParam(binSize = 0.6)) ## Extract adjusted retention times adjustedRtime(faahko) |> head() ## [1] 2551.457 2553.089 2554.720 2556.352 2557.983 2559.615 ## Or simply use the rtime method rtime(faahko) |> head() ## [1] 2551.457 2553.089 2554.720 2556.352 2557.983 2559.615 ## Get raw (unadjusted) retention times rtime(faahko, adjusted = FALSE) |> head() ## [1] 2551.457 2553.022 2554.586 2556.151 2557.716 2559.281 ## Get the base peak chromatograms. bpis_adj <- chromatogram(faahko, aggregationFun = \"max\", chromPeaks = \"none\") par(mfrow = c(3, 1), mar = c(4.5, 4.2, 1, 0.5)) plot(bpis, col = sample_colors) grid() plot(bpis_adj, col = sample_colors) grid() ## Plot also the difference of adjusted to raw retention time. plotAdjustedRtime(faahko, col = sample_colors) grid() par(mfrow = c(2, 1)) ## Plot the raw data plot(chr_raw, col = sample_colors) grid() ## Extract the chromatogram from the adjusted object chr_adj <- chromatogram(faahko, rt = rtr, mz = mzr) plot(chr_adj, col = sample_colors, peakType = \"none\") grid()"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"subset-based-alignment","dir":"Articles","previous_headings":"Pre-processing of LC-MS data > Alignment","what":"Subset-based alignment","title":"LC-MS data pre-processing and analysis with xcms","text":"experiments might better perform alignment based subset available samples, e.g. pooled QC samples injected regular intervals experiment contains blanks. alignment methods xcms support subset-based alignment retention time shifts estimated specified subset samples followed alignment whole data set based aligned subset. subset samples alignment can specified parameter subset PeakGroupsParam ObiwarpParam object. Parameter subsetAdjust allows specify method left-samples adjusted. currently two options available: subsetAdjust = \"previous\": adjust retention times non-subset sample based alignment results previous subset sample (e.g. QC sample). samples e.g. order A1, B1, B2, A2, B3, B4 representing QC samples B study samples, using subset = c(1, 4) subsetAdjust = \"previous\" result samples aligned non-subset samples B1 B2 adjusted based alignment result subset samples A1 B3 B4 A2. subsetAdjust = \"average\": adjust retention times non-subset samples based interpolation alignment results previous subsequent subset sample. example , B1 adjusted based average adjusted retention times subset (QC) samples A1 A2. Since subset sample non-subset samples B3 B4 adjusted based alignment results A2 alone. Note weighted average used calculate adjusted retention time averages, uses inverse difference index non-subset sample subset samples weights. Thus, setup like A1, B1, B2, A2 adjusted retention times A1 get larger weight A2 adjustment non-subset sample B1 causing ’s adjusted retention times closer A1 A2. See examples. Important: cases require meaningful/correct ordering samples within object (.e., samples ordered injection index). examples aim illustrate effect alignment options. assume samples 1, 4 7 faahKO data set QC pool samples. thus want perform alignment based samples subsequently adjust retention times left-samples (2, 3, 5, 6 8) based interpolation results neighboring subset (QC) samples. initial peak grouping perform subset-based alignment peak groups method passing indices QC samples subset parameter PeakGroupsParam function set subsetAdjust = \"average\" adjust study samples based interpolation alignment results neighboring subset/QC samples. Note subset-alignment parameters minFraction relative subset, full experiment! first remove previous alignment results dropAdjustedRtime function allow fresh alignment using subset-based option outlined . addition removing adjusted retention times spectra, function also restore original retention times identified chromatographic peaks. alignment peak groups method initial peak grouping (correspondence) analysis required, algorithm estimates retention times shifts samples using retention times hook peaks (.e. chromatographic peaks present /samples). use default settings peak density method-based correspondence, strongly advised adapt parameters data set (details next section). definition sample groups (.e. assignment individual samples sample groups experiment) mandatory PeakDensityParam. sample groups experiment, sampleGroups set single value file (e.g. rep(1, length(fileNames(faahko))). plot results alignment highlighting subset samples green. nicely shows interpolation subsetAdjust = \"average\" works: retention times sample 2 adjusted based subset sample 1 4, giving however weight closer subset sample 1 results adjusted retention times 2 similar sample 1. Sample 3 hand gets adjusted giving weight second subset sample (4). Subset-alignment results option average. Difference adjusted raw retention times along retention time axis. Samples alignment models estimated shown green, study samples grey. Option subsetAdjust = \"previous\" adjust retention times non-subset sample based single subset sample (previous), results cases adjusted retention times non-subset sample highly similar subset sample used adjustment.","code":"faahko <- dropAdjustedRtime(faahko)  ## Define the experimental layout sampleData(faahko)$sample_type <- \"study\" sampleData(faahko)$sample_type[c(1, 4, 7)] <- \"QC\" ## Initial peak grouping. Use sample_type as grouping variable pdp_subs <- PeakDensityParam(sampleGroups = sampleData(faahko)$sample_type,                              minFraction = 0.9) faahko <- groupChromPeaks(faahko, param = pdp_subs)  ## Define subset-alignment options and perform the alignment pgp_subs <- PeakGroupsParam(     minFraction = 0.85,     subset = which(sampleData(faahko)$sample_type == \"QC\"),     subsetAdjust = \"average\", span = 0.4) faahko <- adjustRtime(faahko, param = pgp_subs) clrs <- rep(\"#00000040\", 8) clrs[sampleData(faahko)$sample_type == \"QC\"] <- c(\"#00ce0080\") par(mfrow = c(2, 1), mar = c(4, 4.5, 1, 0.5)) plot(chromatogram(faahko, aggregationFun = \"max\", chromPeaks = \"none\"),      col = clrs) grid() plotAdjustedRtime(faahko, col = clrs, peakGroupsPch = 1,                   peakGroupsCol = \"#00ce0040\") grid()"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"correspondence","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Correspondence","title":"LC-MS data pre-processing and analysis with xcms","text":"Correspondence usually final step LC-MS data pre-processing data, presumably representing signal originating ions, matched across samples. result, chromatographic peaks different samples similar m/z retention times get grouped LC-MS features. function perform correspondence xcms called groupChromPeaks supports different algorithms can selected configured specific parameter object (see ?groupChromPeaks overview). example use peak density method [1] , within small slices along m/z dimension, combines chromatographic peaks depending density peaks along retention time axis. illustrate , simulate peak grouping m/z slice containing multiple chromatoghaphic peaks within sample using plotChromPeakDensity function PeakDensityParam object parameter minFraction = 0.4 (features defined least 40% samples chromatographic peak present) - parameter sampleGroups used define sample group sample belongs. Example peak density correspondence. Upper panel: chromatogram mz slice multiple chromatographic peaks. lower panel: identified chromatographic peaks retention time (x-axis) index within samples experiments (y-axis) different values bw parameter. black line represents peak density estimate. Grouping peaks (based provided settings) indicated grey rectangles. upper panel plot shows extracted ion chromatogram sample detected peaks highlighted. retention times individual chromatographic peaks sample (y-axis index sample data set) shown lower panel solid black line representing density estimate distribution detected peaks along retention time. Parameter bw defines smoothness estimation. grey rectangles indicate chromatographic peaks grouped feature (grey rectangle thus representing one feature). type visualization thus ideal test, validate optimize correspondence settings manually defined m/z slices applying full data set. tested m/z slice settings seemed OK thus applying full data set . Especially parameter bw data set dependent (specifically LC-dependent) adapted data set. See Metabolomics pre-processing xcms tutorial examples details. Results correspondence analysis can accessed featureDefinitions featureValues function. former returns data frame general information defined features, row one feature columns providing information median m/z retention time well indices chromatographic peaks assigned feature column \"peakidx\". show information first 6 features. featureValues function returns matrix rows features columns samples. content matrix can defined using value argument can column name chromPeaks matrix. default value = \"\" matrix integrated signal peaks corresponding feature sample returned. generally used intensity matrix downstream analysis. extract intensities first 6 features. can see several missing values feature matrix. Missing values reported one sample chromatographic peak detected m/z - rt region feature. however necessarily mean signal specific ion sample. chromatographic peak detection algorithm also just failed identify peak region, e.g. signal noisy low. Thus advisable perform, correspondence, also gap-filling (see next section). performance peak detection, alignment correspondence always evaluated inspecting extracted ion chromatograms e.g. known compounds, internal standards identified features general. featureChromatograms function allows extract chromatograms feature present featureDefinitions. returned MChromatograms object contains ion chromatogram feature (row containing data one feature) sample (column representing containing data one sample). Parameter features allows define specific features EIC returned. can specified index ID (.e. row name featureDefinitions data frame. features defined, EICs returned features data set, can take also considerable amount time. extract chromatograms first 4 features. plot extracted ion chromatograms. use group color identified peak fill area. Extracted ion chromatograms features 1 4. access EICs second feature can simply subset feature_chroms object.","code":"## Define the mz slice. mzr <- c(305.05, 305.15)  ## Extract and plot the chromatograms chr_mzr <- chromatogram(faahko, mz = mzr) ## Define the parameters for the peak density method pdp <- PeakDensityParam(sampleGroups = sampleData(faahko)$sample_group,                         minFraction = 0.4, bw = 30) plotChromPeakDensity(chr_mzr, col = sample_colors, param = pdp,                      peakBg = sample_colors[chromPeaks(chr_mzr)[, \"sample\"]],                      peakCol = sample_colors[chromPeaks(chr_mzr)[, \"sample\"]],                      peakPch = 16) ## Perform the correspondence pdp <- PeakDensityParam(sampleGroups = sampleData(faahko)$sample_group,                         minFraction = 0.4, bw = 30) faahko <- groupChromPeaks(faahko, param = pdp) featureDefinitions(faahko) |> head() ##       mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks KO WT      peakidx ## FT001 200.1 200.1 200.1 2902.634 2882.603 2922.664      2  2  0    458, 1161 ## FT002 205.0 205.0 205.0 2789.901 2782.955 2796.531      8  4  4 44, 443,.... ## FT003 206.0 206.0 206.0 2789.405 2781.389 2794.219      7  3  4 29, 430,.... ## FT004 207.1 207.1 207.1 2718.560 2714.047 2727.347      7  4  3 16, 420,.... ## FT005 233.0 233.0 233.1 3023.579 3015.145 3043.959      7  3  4 69, 959,.... ## FT006 241.1 241.1 241.2 3683.299 3661.586 3695.886      8  3  4 276, 284.... ##       ms_level ## FT001        1 ## FT002        1 ## FT003        1 ## FT004        1 ## FT005        1 ## FT006        1 featureValues(faahko, value = \"into\") |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001        NA  506848.9        NA  169955.6        NA        NA        NA ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7        NA  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9        NA ## FT005  286221.4        NA  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5        NA  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001        NA ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 feature_chroms <- featureChromatograms(faahko, features = 1:4)  feature_chroms ## XChromatograms with 4 rows and 8 columns ##             ko15.CDF        ko16.CDF        ko21.CDF        ko22.CDF ##      <XChromatogram> <XChromatogram> <XChromatogram> <XChromatogram> ## [1,]        peaks: 0        peaks: 1        peaks: 0        peaks: 1 ## [2,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ## [3,]        peaks: 1        peaks: 1        peaks: 0        peaks: 1 ## [4,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ##             wt15.CDF        wt16.CDF        wt21.CDF        wt22.CDF ##      <XChromatogram> <XChromatogram> <XChromatogram> <XChromatogram> ## [1,]        peaks: 0        peaks: 0        peaks: 0        peaks: 0 ## [2,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ## [3,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ## [4,]        peaks: 1        peaks: 1        peaks: 0        peaks: 1 ## phenoData with 4 variables ## featureData with 4 variables ## - - - xcms preprocessing - - - ## Chromatographic peak detection: ##  method: centWave  ## Correspondence: ##  method: chromatographic peak density  ##  4 feature(s) identified. plot(feature_chroms, col = sample_colors,      peakBg = sample_colors[chromPeaks(feature_chroms)[, \"sample\"]]) eic_2 <- feature_chroms[2, ] chromPeaks(eic_2) ##         mz mzmin mzmax       rt    rtmin    rtmax    into    intb  maxo sn ## CP0048 205   205   205 2791.873 2771.300 2815.623 1924712 1850331 84280 64 ## CP0495 205   205   205 2795.796 2773.702 2821.043 1757151 1711473 68384 69 ## CP1033 205   205   205 2796.531 2774.506 2821.697 1383417 1334570 47384 54 ## CP1255 205   205   205 2789.405 2769.019 2812.924 1180288 1126958 48336 32 ## CP1535 205   205   205 2782.955 2762.596 2806.444 2129885 2054677 93312 44 ## CP1967 205   205   205 2787.464 2767.135 2812.486 1634342 1566379 67984 53 ## CP2350 205   205   205 2790.396 2763.847 2821.630 1623589 1531573 49208 28 ## CP2601 205   205   205 2787.273 2766.970 2812.261 1354005 1299188 55712 35 ##        sample row column ## CP0048      1   1      1 ## CP0495      2   1      2 ## CP1033      3   1      3 ## CP1255      4   1      4 ## CP1535      5   1      5 ## CP1967      6   1      6 ## CP2350      7   1      7 ## CP2601      8   1      8"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"gap-filling","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Gap filling","title":"LC-MS data pre-processing and analysis with xcms","text":"Missing values LC-MS data many cases result chromatographic peak detection algorithm failing identify peaks (noisy low intensity signal). aim gap filling step reduce number missing values integrating signals original data files samples chromatographic peak found m/z - rt region signal ion expected. Gap filling can performed xcms fillChromPeaks function parameter object selecting configuring gap filling algorithm. method choice ChromPeakAreaParam integrates signal (samples chromatographic peak found feature) m/z - rt region defined based m/z retention time ranges detected chromatographic peaks specific feature. lower m/z limit area defined lower quartile (25% quantile) \"mzmin\" values peaks feature, upper m/z value upper quartile (75% quantile) \"mzmax\" values, lower rt value lower quartile (25% quantile) \"rtmin\" upper rt value upper quartile (75% quantile) \"rtmax\" values. perform gap filling test data extract feature values first 6 features gap filling. NA reported signal measured specific sample.","code":"faahko <- fillChromPeaks(faahko, param = ChromPeakAreaParam())  featureValues(faahko, value = \"into\") |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001  135162.4  506848.9  111657.3  169955.6  209929.4  141607.9  226853.7 ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7  164380.7  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  226234.4 ## FT005  286221.4  285857.6  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5 1129426.4  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001  138341.2 ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"final-result","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Final result","title":"LC-MS data pre-processing and analysis with xcms","text":"can continue using xcms result set analysis (e.g. also feature grouping MsFeatures package; see LC-MS feature grouping vignette details) also extract results SummarizedExperiment object. standard data container Bioconductor defined SummarizedExperiment package integration Bioconductor packages might thus easier using type object. use quantify function extract xcms pre-processing results SummarizedExperiment object. Internally, featureValues function used generate feature value matrix. can pass parameters function quantify call. use value = \"\" method = \"sum\" report integrated peak signal intensity sum values samples one chromatographic peak assigned feature (option important run refineChromPeaks like described merge overlapping peaks sample). information featureDefinitions now stored rowData object. rowData provides annotations information row SummarizedExperiment (case features). Annotations columns (case samples) stored colData. data frame row contains annotations one sample (hence one column feature values matrix). Finally, feature matrix stored assay within object. Note SummarizedExperiment can multiple assays numeric matrices number rows columns matching number features samples, respectively. list names available assays. can access actual data using assay function, optionally also providing name assay want access. show first 6 lines matrix. Since SummarizedExperiment supports multiple assays, addition add also feature value matrix without filled-values (.e. feature intensities added gap filling step). now two assays result object. can extract feature values without gap-filling: Finally, history full processing xcms available metadata SummarizedExperiment. information can also extracted xcms result object using processHistory function. extract information first processing step. processing steps contain also individual parameter objects used analysis, hence allowing exactly reproduce analysis. last perform also principal component analysis evaluate grouping samples experiment. Note perform data normalization hence grouping might () also influenced technical biases. PCA faahKO data set, un-normalized intensities. can see expected separation KO WT samples PC2. PC1 samples separate based ID, samples ID <= 18 samples ID > 18. separation might caused technical bias (e.g. measurements performed different days/weeks) due biological properties mice analyzed (sex, age, litter mates etc).","code":"library(SummarizedExperiment) ## Loading required package: MatrixGenerics ## Loading required package: matrixStats ##  ## Attaching package: 'matrixStats' ## The following objects are masked from 'package:Biobase': ##  ##     anyMissing, rowMedians ##  ## Attaching package: 'MatrixGenerics' ## The following objects are masked from 'package:matrixStats': ##  ##     colAlls, colAnyNAs, colAnys, colAvgsPerRowSet, colCollapse, ##     colCounts, colCummaxs, colCummins, colCumprods, colCumsums, ##     colDiffs, colIQRDiffs, colIQRs, colLogSumExps, colMadDiffs, ##     colMads, colMaxs, colMeans2, colMedians, colMins, colOrderStats, ##     colProds, colQuantiles, colRanges, colRanks, colSdDiffs, colSds, ##     colSums2, colTabulates, colVarDiffs, colVars, colWeightedMads, ##     colWeightedMeans, colWeightedMedians, colWeightedSds, ##     colWeightedVars, rowAlls, rowAnyNAs, rowAnys, rowAvgsPerColSet, ##     rowCollapse, rowCounts, rowCummaxs, rowCummins, rowCumprods, ##     rowCumsums, rowDiffs, rowIQRDiffs, rowIQRs, rowLogSumExps, ##     rowMadDiffs, rowMads, rowMaxs, rowMeans2, rowMedians, rowMins, ##     rowOrderStats, rowProds, rowQuantiles, rowRanges, rowRanks, ##     rowSdDiffs, rowSds, rowSums2, rowTabulates, rowVarDiffs, rowVars, ##     rowWeightedMads, rowWeightedMeans, rowWeightedMedians, ##     rowWeightedSds, rowWeightedVars ## The following object is masked from 'package:Biobase': ##  ##     rowMedians ## Loading required package: GenomicRanges ## Loading required package: IRanges ##  ## Attaching package: 'IRanges' ## The following object is masked from 'package:xcms': ##  ##     distance ## Loading required package: GenomeInfoDb res <- quantify(faahko, value = \"into\", method = \"sum\") res ## class: SummarizedExperiment  ## dim: 351 8  ## metadata(6): '' '' ... '' '' ## assays(1): raw ## rownames(351): FT001 FT002 ... FT350 FT351 ## rowData names(10): mzmed mzmin ... WT ms_level ## colnames(8): ko15.CDF ko16.CDF ... wt21.CDF wt22.CDF ## colData names(4): sample_name sample_group spectraOrigin sample_type rowData(res) ## DataFrame with 351 rows and 10 columns ##           mzmed     mzmin     mzmax     rtmed     rtmin     rtmax    npeaks ##       <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> ## FT001     200.1     200.1     200.1   2902.63   2882.60   2922.66         2 ## FT002     205.0     205.0     205.0   2789.90   2782.95   2796.53         8 ## FT003     206.0     206.0     206.0   2789.40   2781.39   2794.22         7 ## FT004     207.1     207.1     207.1   2718.56   2714.05   2727.35         7 ## FT005     233.0     233.0     233.1   3023.58   3015.14   3043.96         7 ## ...         ...       ...       ...       ...       ...       ...       ... ## FT347    595.25     595.2     595.3   3010.60   2992.76   3014.37         6 ## FT348    596.20     596.2     596.2   2997.78   2992.76   3002.79         2 ## FT349    596.30     596.3     596.3   3819.83   3811.52   3836.39         4 ## FT350    597.40     597.4     597.4   3820.88   3817.76   3826.13         3 ## FT351    599.30     599.3     599.3   4071.41   4044.94   4125.32         3 ##              KO        WT  ms_level ##       <numeric> <numeric> <integer> ## FT001         2         0         1 ## FT002         4         4         1 ## FT003         3         4         1 ## FT004         4         3         1 ## FT005         3         4         1 ## ...         ...       ...       ... ## FT347         2         3         1 ## FT348         0         2         1 ## FT349         2         2         1 ## FT350         1         2         1 ## FT351         1         2         1 colData(res) ## DataFrame with 8 rows and 4 columns ##          sample_name sample_group spectraOrigin sample_type ##          <character>  <character>   <character> <character> ## ko15.CDF        ko15           KO /__w/_temp...          QC ## ko16.CDF        ko16           KO /__w/_temp...       study ## ko21.CDF        ko21           KO /__w/_temp...       study ## ko22.CDF        ko22           KO /__w/_temp...          QC ## wt15.CDF        wt15           WT /__w/_temp...       study ## wt16.CDF        wt16           WT /__w/_temp...       study ## wt21.CDF        wt21           WT /__w/_temp...          QC ## wt22.CDF        wt22           WT /__w/_temp...       study assayNames(res) ## [1] \"raw\" assay(res) |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001  135162.4  506848.9  111657.3  169955.6  209929.4  141607.9  226853.7 ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7  164380.7  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  226234.4 ## FT005  286221.4  285857.6  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1923307.8 1129426.4  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001  138341.2 ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 assays(res)$raw_nofill <- featureValues(faahko, filled = FALSE, method = \"sum\") assayNames(res) ## [1] \"raw\"        \"raw_nofill\" assay(res, \"raw_nofill\") |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001        NA  506848.9        NA  169955.6        NA        NA        NA ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7        NA  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9        NA ## FT005  286221.4        NA  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1923307.8        NA  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001        NA ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 metadata(res) ## [[1]] ## Object of class \"XProcessHistory\" ##  type: Peak detection  ##  date: Thu Dec  7 08:46:45 2023  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: CentWaveParam  ##  MS level(s) 1  ##  ## [[2]] ## Object of class \"XProcessHistory\" ##  type: Peak refinement  ##  date: Thu Dec  7 08:46:48 2023  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: MergeNeighboringPeaksParam  ##  MS level(s) 1  ##  ## [[3]] ## Object of class \"XProcessHistory\" ##  type: Peak grouping  ##  date: Thu Dec  7 08:46:59 2023  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: PeakDensityParam  ##  MS level(s) 1  ##  ## [[4]] ## Object of class \"XProcessHistory\" ##  type: Retention time correction  ##  date: Thu Dec  7 08:47:00 2023  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: PeakGroupsParam  ##  MS level(s) 1  ##  ## [[5]] ## Object of class \"XProcessHistory\" ##  type: Peak grouping  ##  date: Thu Dec  7 08:47:04 2023  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: PeakDensityParam  ##  MS level(s) 1  ##  ## [[6]] ## Object of class \"XProcessHistory\" ##  type: Missing peak filling  ##  date: Thu Dec  7 08:47:08 2023  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: ChromPeakAreaParam  ##  MS level(s) 1 processHistory(faahko)[[1]] ## Object of class \"XProcessHistory\" ##  type: Peak detection  ##  date: Thu Dec  7 08:46:45 2023  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: CentWaveParam  ##  MS level(s) 1 processHistory(faahko)[[1]] |> processParam() ## Object of class:  CentWaveParam  ##  Parameters: ##  - ppm: [1] 25 ##  - peakwidth: [1] 20 80 ##  - snthresh: [1] 10 ##  - prefilter: [1]    6 5000 ##  - mzCenterFun: [1] \"wMean\" ##  - integrate: [1] 1 ##  - mzdiff: [1] -0.001 ##  - fitgauss: [1] FALSE ##  - noise: [1] 5000 ##  - verboseColumns: [1] FALSE ##  - roiList: list() ##  - firstBaselineCheck: [1] TRUE ##  - roiScales: numeric(0) ##  - extendLengthMSW: [1] FALSE ##  - verboseBetaColumns: [1] FALSE ## Extract the features and log2 transform them ft_ints <- log2(assay(res, \"raw\"))  ## Perform the PCA omitting all features with an NA in any of the ## samples. Also, the intensities are mean centered. pc <- prcomp(t(na.omit(ft_ints)), center = TRUE)  ## Plot the PCA pcSummary <- summary(pc) plot(pc$x[, 1], pc$x[,2], pch = 21, main = \"\",      xlab = paste0(\"PC1: \", format(pcSummary$importance[2, 1] * 100,                                    digits = 3), \" % variance\"),      ylab = paste0(\"PC2: \", format(pcSummary$importance[2, 2] * 100,                                    digits = 3), \" % variance\"),      col = \"darkgrey\", bg = sample_colors, cex = 2) grid() text(pc$x[, 1], pc$x[,2], labels = res$sample_name, col = \"darkgrey\",      pos = 3, cex = 2)"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"further-data-processing-and-analysis","dir":"Articles","previous_headings":"","what":"Further data processing and analysis","title":"LC-MS data pre-processing and analysis with xcms","text":"Normalizing features’ signal intensities required, present (yet) supported xcms (methods might added near future). advised use SummarizedExperiment returned quantify method data processing, type object stores feature definitions, sample annotations well feature abundances object. identification e.g. features significant different intensities/abundances suggested use functionality provided R packages, Bioconductor’s excellent limma package.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"subsetting-and-filtering","dir":"Articles","previous_headings":"Additional details and notes","what":"Subsetting and filtering","title":"LC-MS data pre-processing and analysis with xcms","text":"xcms result objects can subset/filtered sample using [ method one filter* functions (although XcmsExperiment supports present selected filter functions). cases filtering can remove pre-processing results, filter functions support parameters keepFeatures keepAdjustedRtime can set TRUE avoid removal.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"parallel-processing","dir":"Articles","previous_headings":"Additional details and notes","what":"Parallel processing","title":"LC-MS data pre-processing and analysis with xcms","text":"functions xcms support parallel processing, enabled default performed, operations, per-file basis. Parallel processing handled configured BiocParallel Bioconductor package can globally defined R session. Note , data objects designed low memory footprint loading peak data memory needed, parallel processing increase demand: general, full data many files parallel processes loaded memory. needs also considered, number parallel processes defined. Unix-based systems (Linux, macOS) support multicore-based parallel processing. configure globally register parameter class. Note also bpstart used initialize parallel processes. Windows supports socket-based parallel processing:","code":"register(bpstart(MulticoreParam(2))) register(bpstart(SnowParam(2)))"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"main-differences-to-the-msnbase-based-xcms-version-3","dir":"Articles","previous_headings":"Additional details and notes","what":"Main differences to the MSnbase-based xcms version 3","title":"LC-MS data pre-processing and analysis with xcms","text":"Spectra used main container (raw) MS data. Thus changing backends supported stage analysis (e.g. load data memory load data remote databases).","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"additional-documentation-resources","dir":"Articles","previous_headings":"","what":"Additional documentation resources","title":"LC-MS data pre-processing and analysis with xcms","text":"documentations listed still based xcms version 3 subsequently updated. Metabolomics pre-processing xcms: detailed description pre-processing LC-MS data xcms. MetaboAnnotationTutorials: examples annotation metabolomics data [3]. [2]: describes concept -disk data modes used also xcms. SpectraTutorials: tutorials describing Spectra package functionality.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"LC-MS data pre-processing and analysis with xcms","text":"R packages used document listed .","code":"sessionInfo() ## R Under development (unstable) (2023-11-22 r85609) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] SummarizedExperiment_1.33.1 GenomicRanges_1.55.1        ##  [3] GenomeInfoDb_1.39.1         IRanges_2.37.0              ##  [5] MatrixGenerics_1.15.0       matrixStats_1.1.0           ##  [7] MsExperiment_1.5.1          pheatmap_1.0.12             ##  [9] RColorBrewer_1.1-3          pander_0.6.5                ## [11] faahKO_1.43.0               xcms_4.1.4                  ## [13] MSnbase_2.29.1              ProtGenerics_1.35.0         ## [15] S4Vectors_0.41.2            mzR_2.37.0                  ## [17] Rcpp_1.0.11                 Biobase_2.63.0              ## [19] BiocGenerics_0.49.1         BiocParallel_1.37.0         ## [21] BiocStyle_2.31.0            ##  ## loaded via a namespace (and not attached): ##   [1] bitops_1.0-7                rlang_1.1.2                 ##   [3] magrittr_2.0.3              clue_0.3-65                 ##   [5] MassSpecWavelet_1.69.0      compiler_4.4.0              ##   [7] systemfonts_1.0.5           vctrs_0.6.5                 ##   [9] stringr_1.5.1               MetaboCoreUtils_1.11.0      ##  [11] pkgconfig_2.0.3             crayon_1.5.2                ##  [13] fastmap_1.1.1               XVector_0.43.0              ##  [15] utf8_1.2.4                  rmarkdown_2.25              ##  [17] preprocessCore_1.65.0       ragg_1.2.6                  ##  [19] purrr_1.0.2                 MultiAssayExperiment_1.29.0 ##  [21] xfun_0.41                   zlibbioc_1.49.0             ##  [23] cachem_1.0.8                jsonlite_1.8.8              ##  [25] progress_1.2.3              highr_0.10                  ##  [27] DelayedArray_0.29.0         prettyunits_1.2.0           ##  [29] parallel_4.4.0              cluster_2.1.6               ##  [31] R6_2.5.1                    bslib_0.6.1                 ##  [33] stringi_1.8.2               limma_3.59.1                ##  [35] jquerylib_0.1.4             bookdown_0.37               ##  [37] iterators_1.0.14            knitr_1.45                  ##  [39] igraph_1.5.1                splines_4.4.0               ##  [41] Matrix_1.6-4                tidyselect_1.2.0            ##  [43] abind_1.4-5                 yaml_2.3.7                  ##  [45] doParallel_1.0.17           codetools_0.2-19            ##  [47] affy_1.81.0                 lattice_0.22-5              ##  [49] tibble_3.2.1                plyr_1.8.9                  ##  [51] evaluate_0.23               survival_3.5-7              ##  [53] desc_1.4.2                  Spectra_1.13.2              ##  [55] pillar_1.9.0                affyio_1.73.0               ##  [57] BiocManager_1.30.22         foreach_1.5.2               ##  [59] MALDIquant_1.22.1           ncdf4_1.22                  ##  [61] generics_0.1.3              rprojroot_2.0.4             ##  [63] RCurl_1.98-1.13             hms_1.1.3                   ##  [65] ggplot2_3.4.4               munsell_0.5.0               ##  [67] scales_1.3.0                glue_1.6.2                  ##  [69] lazyeval_0.2.2              MsFeatures_1.11.0           ##  [71] tools_4.4.0                 mzID_1.41.0                 ##  [73] robustbase_0.99-1           QFeatures_1.13.1            ##  [75] vsn_3.71.0                  RANN_2.6.1                  ##  [77] fs_1.6.3                    XML_3.99-0.16               ##  [79] grid_4.4.0                  impute_1.77.0               ##  [81] MsCoreUtils_1.15.1          colorspace_2.1-0            ##  [83] GenomeInfoDbData_1.2.11     cli_3.6.1                   ##  [85] textshaping_0.3.7           fansi_1.0.5                 ##  [87] S4Arrays_1.3.1              dplyr_1.1.4                 ##  [89] AnnotationFilter_1.27.0     pcaMethods_1.95.0           ##  [91] gtable_0.3.4                DEoptimR_1.1-3              ##  [93] sass_0.4.8                  digest_0.6.33               ##  [95] SparseArray_1.3.1           farver_2.1.1                ##  [97] multtest_2.59.0             memoise_2.0.1               ##  [99] htmltools_0.5.7             pkgdown_2.0.7.9000          ## [101] lifecycle_1.0.4             statmod_1.5.0               ## [103] MASS_7.3-60.1"},{"path":[]},{"path":"https://sneumann.github.io/xcms/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Colin . Smith. Author. Ralf Tautenhahn. Author. Steffen Neumann. Author, maintainer. Paul Benton. Author. Christopher Conley. Author. Johannes Rainer. Author. Michael Witting. Contributor. William Kumler. Author. Philippine Louail. Author.","code":""},{"path":"https://sneumann.github.io/xcms/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Smith, C.. Want, E.J. O'Maille, G. Abagyan,R. Siuzdak, G.: XCMS: Processing mass spectrometry data metabolite profiling using nonlinear peak alignment, matching identification, Analytical Chemistry, 78:779-787 (2006) Ralf Tautenhahn, Christoph Boettcher, Steffen Neumann: Highly sensitive feature detection high resolution LC/MS BMC Bioinformatics, 9:504 (2008) H. Paul Benton, Elizabeth J. Want Timothy M. D. Ebbels Correction mass calibration gaps liquid chromatography-mass spectrometry metabolomics data Bioinformatics, 26:2488 (2010)","code":"@Article{,   author = {{Smith} and {C.A.} and {Want} and {E.J.} and {O'Maille} and {G.} and {Abagyan,R.} and {Siuzdak} and {G.}},   title = {XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching and identification},   journal = {Analytical Chemistry},   year = {2006},   volume = {78},   pages = {779--787}, } @Article{,   author = {Ralf Tautenhahn and Christoph Boettcher and Steffen Neumann},   title = {Highly sensitive feature detection for high resolution LC/MS},   journal = {BMC Bioinformatics},   year = {2008},   volume = {9},   pages = {504}, } @Article{,   author = {H. Paul Benton and Elizabeth J. Want and Timothy M. D. Ebbels},   title = {Correction of mass calibration gaps in liquid chromatography-mass spectrometry metabolomics data},   journal = {BIOINFORMATICS},   year = {2010},   volume = {26},   pages = {2488}, }"},{"path":"https://sneumann.github.io/xcms/index.html","id":"the-xcms-package-pre-processing-gclc-msms-data","dir":"","previous_headings":"","what":"LC-MS and GC-MS Data Analysis","title":"LC-MS and GC-MS Data Analysis","text":"Please see package documentation information examples news latest changes.","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"version-4","dir":"","previous_headings":"","what":"Version 4","title":"LC-MS and GC-MS Data Analysis","text":"Version 4 adds native support Spectra package xcms allows perform pre-processing MsExperiment objects (MsExperiment. new supported data containers (Spectra, MsExperiment XcmsExperiment) allow flexible analyses seamless future extensions additional types data (ion mobility data). Ultimately, changes also allow easier integration xcms R packages MsFeatures MetaboAnnotation. suggested users switch newer data result objects, functionality version 3 remain fully supported.","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"version-3","dir":"","previous_headings":"","what":"Version 3","title":"LC-MS and GC-MS Data Analysis","text":"Version >= 3 xcms package updated partially re-written versions original xcms package. version number 3 selected avoid confusions xcms2 (http://pubs.acs.org/doi/abs/10.1021/ac800795f) software. providing original software’s functionality, xcms version >= 3 aims : Better integration Bioconductor framework: Make use extend classes defined MSnbase package. Implement class versioning (Biobase’s Versioned class). Use BiocParallel parallel processing. Implementation validation methods classes ensure data integrity. Easier faster access raw spectra data. Cleanup source code: Remove obsolete redundant functionality (getEIC, rawEIC etc). Unify interfaces, .e. implement layer base functions accessing analysis methods (implemented C, C++ R). Using consistent naming scheme methods follows established naming conventions (e.g. correspondence instead grouping). Update, improve extend documentation. Establishing layer base R-functions interface analysis methods. take M/Z, retention time (scan index) intensity values input along optional arguments downstream functions (implemented C, C++ R). input arguments basic R objects (numeric vectors) thus enabling easy integration analysis methods R packages. user interface’s analysis methods take (raw) data object parameter class, used dispatching corresponding analysis algorithm. Add unit tests. Discussions suggestions welcome: https://github.com/sneumann/xcms/issues","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"contribution","dir":"","previous_headings":"","what":"Contribution","title":"LC-MS and GC-MS Data Analysis","text":"Contributions xcms package welcome, whether form ideas, documentation, code, packages, … contribution guideline please see guideline RforMassSpectrometry initiative. seamless integration, contributors expected adhere RforMassSpectrometry coding syle.","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"code-of-conduct","dir":"","previous_headings":"","what":"Code of Conduct","title":"LC-MS and GC-MS Data Analysis","text":"contributors maintainers package, pledge respect people contribute reporting issues, posting feature requests, updating documentation, submitting pull requests patches, activities. See RforMassSpectrometry Code Conduct information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"AutoLockMass  - function decides lock mass scans \txcmsRaw object. done using scan time differences.","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"object = \"xcmsRaw\" signature(object = \"xcmsRaw\")","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"object xcmsRaw-class object","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"AutoLockMass numeric vector scan locations corresponding lock Mass scans","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"Paul Benton, hpaul.benton08@imperial.ac.uk","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"","code":"if (FALSE) library(xcms)     library(faahKO)     ## These files do not have this problem     ## to correct for but just for an example     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xr<-xcmsRaw(cdffiles[1]) #> Create profile matrix with method 'bin' and step 1 ...  #> OK     xr #> An \"xcmsRaw\" object with 1278 mass spectra #>  #> Time range: 2501.4-4499.8 seconds (41.7-75 minutes) #> Mass range: 200-600 m/z #> Intensity range: 70-1373180  #>  #> MSn data on  0  mass(es) #> \twith  0  MSn spectra #> Profile method: bin  #> Profile step: 1 m/z (401 grid points from 200 to 600 m/z) #>  #> Memory usage: 10.8 MB     ##Lets assume that the lockmass starts at 1 and is every 100 scans     lockMass<-xcms:::makeacqNum(xr, freq=100, start=1)     ## these are equalvent     lockmass2<-AutoLockMass(xr) #> Warning:  #> Lock mass frequency wasn't detected     all((lockmass == lockmass2) == TRUE) #> Error in eval(expr, envir, enclos): object 'lockmass' not found      ob<-stitch(xr, lockMass) #> Warning: no non-missing arguments to min; returning Inf #> Warning: no non-missing arguments to max; returning -Inf"},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":null,"dir":"Reference","previous_headings":"","what":"Generic parameter class — GenericParam-class","title":"Generic parameter class — GenericParam-class","text":"GenericParam class allows store generic parameter     information name function /called     (slot fun) arguments (slot args). object     used track process history data processings     XCMSnExp object. contrast e.g.     CentWaveParam object passed actual     processing method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generic parameter class — GenericParam-class","text":"","code":"GenericParam(fun = character(), args = list())"},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generic parameter class — GenericParam-class","text":"fun character representing name function. args list (ideally named) arguments function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generic parameter class — GenericParam-class","text":"GenericParam function returns GenericParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Generic parameter class — GenericParam-class","text":"fun character specifying function name. args list (ideally named) arguments function.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generic parameter class — GenericParam-class","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generic parameter class — GenericParam-class","text":"","code":"prm <- GenericParam(fun = \"mean\")  prm <- GenericParam(fun = \"mean\", args = list(na.rm = TRUE))"},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":null,"dir":"Reference","previous_headings":"","what":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"`PlainTextParam` class method enable users save `MsExperiment` `XcmsExperiment` objects collections plain text files specified folder. Note , xcms results within `XcmsExperiment` can exported, full raw MS data (object's `Spectra` object) currently exported plain text format. `Spectra` using [MsBackendMzR()] backend, names raw data files however exported (enables *restore* full `Spectra` respectively `MsExperiment` objects). `MsExperiment` object, exported files include: - [sampleData()] stored text file named *sample_data.txt*. - [fileNames()] *Spectra* object stored tabular format text file named *spectra_files.txt*.file names exported `Spectra` object uses [MsBackendMzR()] backend. backends information raw spectra data currently exported `PlainTextParam`. - Processing queue `Spectra` object, ensuring spectra data modifications retained. stored `json` file named *spectra_processing_queue.json*. `XcmsExperiment` object, exported files `MsExperiment` object, addition following: - [processHistory()] information object, stored `json` file named *process_history.json*. - chromatographic peak information obtained [chromPeaks()] [chromPeaksData()], stored tabular format text files *chrom_peaks.txt* *chrom_peak_data.txt* respectively. - retention time information obtained [adjustedRtime()] stored text file named *rtime_adjusted.txt*. - [featureDefinitions()] stored text file named *feature_definitions.txt*. Additionally, second file named *feature_peak_index.txt* generated connect features' definitions names. `param` class method part possible dispatch generic function `storeResults`. folder created calling `storeResults`. folder already exists, previous exports folder might get overwritten.","code":""},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"","code":"PlainTextParam(path = tempdir())  # S4 method for MsExperiment,PlainTextParam storeResults(object, param)  # S4 method for XcmsExperiment,PlainTextParam storeResults(object, param)"},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"path `PlainTextParam` `character(1)`, defining files going stored. default `tempdir()`. object MsExperiment XcmsExperiment data object needs saved. param parameter object selecting configuring format saving. can one following classes: RDataParam, PlainTextParam, MzTabMParam.","code":""},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"`PlainTextParam`: `PlainTextParam` class. `storeResults` return anything saves object collections different plain text files folder.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"Philippine Louail, Johannes Rainer.","code":""},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"","code":"## Load test data set of class `MsExperiment` library(MsExperiment) fls <- dir(system.file(\"sciex\", package = \"msdata\"), full.names = TRUE) pd <- data.frame(file = basename(fls),                  sample = c(\"POOL_1\", \"POOL_2\"),                  injection_index = c(1, 19),                 group = \"POOL\") rownames(pd) <- c(\"1\", \"2\") mse <- readMsExperiment(fls, sampleData = pd)  ## Define param pth = file.path(tempdir(), \"test\") param <- PlainTextParam(path = pth)  ## Save as a collection of plain text files storeResults(object = mse, param = param)  ## Load a test data set with detected peaks, of class `XcmsExperiment` faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Define param pth = file.path(tempdir(), \"test\") param <- PlainTextParam(path = pth)  ## Save as a collection of plain text files storeResults(object = faahko_sub, param = param) #> Warning: '/tmp/RtmpZ5Cxxr/test' already exists"},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Tracking data processing — ProcessHistory-class","title":"Tracking data processing — ProcessHistory-class","text":"Objects type ProcessHistory allow keep track     data processing step metabolomics experiment.     created data processing methods,     findChromPeaks added corresponding results     objects. Thus, usually, users need create . XProcessHistory extends ProcessHistory     adding slot param allows store actual parameter     class processing step. processParam, processParam<-: get set     parameter class XProcessHistory object. msLevel: returns MS level certain analysis     performed, NA defined. processType method returns character specifying     processing step type. processDate extracts start date processing     step. processInfo extracts optional additional information     processing step. fileIndex extracts indices files     processing step applied.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tracking data processing — ProcessHistory-class","text":"","code":"# S4 method for ProcessHistory show(object)  # S4 method for XProcessHistory show(object)  # S4 method for XProcessHistory processParam(object)  # S4 method for XProcessHistory msLevel(object)  # S4 method for ProcessHistory processType(object)  # S4 method for ProcessHistory processDate(object)  # S4 method for ProcessHistory processInfo(object)  # S4 method for ProcessHistory fileIndex(object)"},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tracking data processing — ProcessHistory-class","text":"object ProcessHistory XProcessHistory object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tracking data processing — ProcessHistory-class","text":"processParam: parameter object extending  Param class. processType method returns character string     processing step type. processDate method returns character string     time stamp processing step start. processInfo method returns character string     optional additional informations. fileIndex method returns integer vector index     files/samples processing step applied.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Tracking data processing — ProcessHistory-class","text":"type character(1): string defining type processing step. string match predefined values. Use processHistoryTypes list . date character(1): date time stamp processing step started. info character(1): optional additional information. fileIndex integer length 1 > 1 specify samples object processing performed. error (): used store eventual calculation errors. param (Param): object type Param (e.g. CentWaveParam) specifying settings processing step. msLevel: integer definining MS level(s) analysis performed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Tracking data processing — ProcessHistory-class","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":null,"dir":"Reference","previous_headings":"","what":"Store `XcmsExperiment` object as .RData file — RDataParam","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"`RDataParam` class method allow users save `XcmsExperiment` object .RData file chosen filename. object gets exported using [`save()`] function.  `param` class method part possible dispatch generic function `storeResults`.","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"","code":"RDataParam(fileName = tempfile())  # S4 method for XcmsExperiment,RDataParam storeResults(object, param)"},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"fileName `RDataParam` `character(1)`, defining file name. default `tempfile()`. object MsExperiment XcmsExperiment data object needs saved. param parameter object selecting configuring format saving. can one following classes: RDataParam, PlainTextParam, MzTabMParam.","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"`RDataParam`: `RDataParam` class. `storeResults` return anything saves object RData file.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"Philippine Louail","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Define param  param <- RDataParam(fileName = \"example_xcms_object\")  ## Save as RData storeResults(object = faahko_sub, param = param)"},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":null,"dir":"Reference","previous_headings":"","what":"Gaussian Model — SSgauss","title":"Gaussian Model — SSgauss","text":"selfStart model evalueates Gaussian model   gradient. initial attribute evalueate   inital estimates parameters mu, sigma,   h.","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Gaussian Model — SSgauss","text":"","code":"SSgauss(x, mu, sigma, h)"},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Gaussian Model — SSgauss","text":"x numeric vector values evaluate model mu mean  distribution function sigma standard deviation distribution fuction h height distribution function","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Gaussian Model — SSgauss","text":"Initial values mu h chosen   maximal value x. initial value sigma   determined area x divided h*sqrt(2*pi).","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Gaussian Model — SSgauss","text":"numeric vector length x. value   expression h*exp(-(x-mu)^2/(2*sigma^2),   modified gaussian function maximum height treated   separate parameter dependent sigma. arguments  mu, sigma, h names objects,   gradient matrix respect names attached   attribute named gradient.","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Gaussian Model — SSgauss","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Data container storing xcms preprocessing results — XCMSnExp-class","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"XCMSnExp object container results G/LC-MS data preprocessing comprises chromatographic peak detection, alignment correspondence. results can accessed chromPeaks, adjustedRtime featureDefinitions functions; see (Usage, Arguments, Value Slots sections) details). Along results, object contains processing history allows track processing step along used settings. can extracted processHistory method. XCMSnExp objects, directly extending OnDiskMSnExp object MSnbase package, inherit functionality allows thus easy access full raw data stage analysis. support interaction packages requiring old objects, XCMSnExp objects can coerced xcmsSet objects using method (see examples ). preprocessing results passed along resulting xcmsSet object. General functions XCMSnExp objects (see specific function handle chromatographic peak data, alignment correspondence results): processHistoryTypes returns available types     process histories. can passed argument type     processHistory method extract specific process step(s). hasFilledChromPeaks: whether filled-peaks present . profMat: creates profile matrix,     n x m matrix, n (rows) representing equally spaced m/z values (bins)     m (columns) retention time corresponding scans. cell     contains maximum intensity measured specific scan m/z     values. See profMat details description     various binning methods. hasAdjustedRtime: whether object provides adjusted retention times. hasFeatures: whether object contains correspondence results (.e. features). hasChromPeaks: whether object contains peak detection results. hasFilledChromPeaks: whether object contains filled-chromatographic peaks. adjustedRtime,adjustedRtime<-: extract/set adjusted retention times. adjustedRtime<- called manually, called internally adjustRtime methods. XCMSnExp objects, adjustedRtime<- also apply retention time adjustments eventually present chromatographic peaks. bySample parameter allows specify whether adjusted retention time grouped sample (file). featureDefinitions, featureDefinitions<-: extract set correspondence results, .e. mz-rt features (peak groups). Similar chromPeaks possible extract features specified m/z /rt ranges. function supports also parameter type allows specify features returned rt mz specified. details see help chromPeaks. See also featureSummary function calculate simple feature summaries. chromPeaks, chromPeaks<-: extract set matrix containing information identified chromatographic peaks. Rownames matrix represent unique IDs respective peaks within experiment. Parameter bySample allows specify whether peaks returned ungrouped (default bySample = FALSE) grouped sample (bySample = TRUE). chromPeaks<- method XCMSnExp objects removes also correspondence (peak grouping) retention time correction (alignment) results. optional arguments rt, mz, ppm type allow extract chromatographic peaks overlapping defined retention time /m/z ranges. Argument type allows define overlapping determined: type == \"\" (default), peaks even partially overlapping region returned (.e. either \"mzmin\" \"mzmax\" chromPeaks featureDefinitions matrix within provided m/z range), type == \"within\" full peak within region (.e. \"mzmin\" \"mzmax\" within m/z range) type == \"apex_within\" peak's apex position (highest signal peak) within region (.e. peak's features m/z within m/z range). See description return value details returned matrix. Users usually use chromPeaks<- method directly detected chromatographic peaks added object findChromPeaks method. Also, chromPeaks<- replace existing chromPeakData. chromPeakData chromPeakData<- allow get set arbitrary chromatographic peak annotations. returned ar returned DataFrame. Note number rows rownames DataFrame match chromPeaks. rtime: extracts retention time scan. bySample parameter allows return values grouped sample/file adjusted whether adjusted raw retention times returned. default method returns adjusted retention times, available (.e. retention times adjusted using adjustRtime method). mz: extracts mz values scan files within XCMSnExp object. values extracted original data files eventual processing steps applied fly. Using bySample parameter possible switch default grouping mz values spectrum/scan grouping sample/file. intensity: extracts intensity values scan files within XCMSnExp object. values extracted original data files eventual processing steps applied fly. Using bySample parameter possible switch default grouping intensity values spectrum/scan grouping sample/file. spectra: extracts Spectrum objects containing data object. values extracted original data files eventual processing steps applied fly. setting bySample = TRUE, spectra returned grouped sample/file. XCMSnExp object contains adjusted retention times, returned default Spectrum objects (can overwritten setting adjusted = FALSE). processHistory: returns list ProcessHistory objects (objects inheriting base class) representing individual processing steps performed, eventually along settings (Param parameter class). Optional arguments fileIndex, type msLevel allow restrict process steps certain type performed certain file MS level. dropChromPeaks: drops identified chromatographic peaks returns object without information. Note XCMSnExp objects method drops default also results correspondence (peak grouping) analysis. Adjusted retention times removed alignment performed peak detection. can overruled keepAdjustedRtime = TRUE. dropFeatureDefinitions: drops results correspondence (peak grouping) analysis, .e. definition mz-rt features returns object without information. Note XCMSnExp objects method also default drop retention time adjustment results, performed last peak grouping (.e. base results peak grouping going removed). related process history steps removed well eventually filled peaks (fillChromPeaks). parameter keepAdjustedRtime can used avoid removal adjusted retention times. dropAdjustedRtime: drops retention time adjustment information returns object without adjusted retention time. XCMSnExp objects, also reverts retention times reported chromatographic peaks peak matrix original, raw, ones (chromatographic peak detection). Note XCMSnExp objects method drops also peak grouping results performed retention time adjustment. related process history steps removed . findChromPeaks performs chromatographic peak detection provided XCMSnExp objects. details see method XCMSnExp. Note default (parameter add = FALSE) previous peak detection results removed. Use add = TRUE perform second round peak detection add newly identified peaks previous peak detection results. Correspondence results (features) always removed prior peak detection. Previous alignment (retention time adjustment) results kept, .e. chromatographic peak detection performed using adjusted retention times data first aligned using e.g. obiwarp (adjustRtime). dropFilledChromPeaks: drops filled-chromatographic peaks (filled fillChromPeaks method) related process history steps. spectrapply applies provided function Spectrum object returns results. function specified function simply returns list Spectrum objects. XCMSnExp objects can combined c function. combines identified chromatographic peaks objects' pheno data discards alignment results feature definitions. plot plots spectrum data (see plot MSnExp objects MSnbase package details. type = \"XIC\", identified chromatographic peaks indicated rectangles border color peakCol.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"","code":"processHistoryTypes()  # S4 method for XCMSnExp hasFilledChromPeaks(object)  # S4 method for OnDiskMSnExp profMat(   object,   method = \"bin\",   step = 0.1,   baselevel = NULL,   basespace = NULL,   mzrange. = NULL,   fileIndex,   ... )  # S4 method for XCMSnExp show(object)  # S4 method for XCMSnExp hasAdjustedRtime(object)  # S4 method for XCMSnExp hasFeatures(object, msLevel = integer())  # S4 method for XCMSnExp hasChromPeaks(object, msLevel = integer())  # S4 method for XCMSnExp hasFilledChromPeaks(object)  # S4 method for XCMSnExp adjustedRtime(object, bySample = FALSE)  # S4 method for XCMSnExp adjustedRtime(object) <- value  # S4 method for XCMSnExp featureDefinitions(   object,   mz = numeric(),   rt = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel = integer() )  # S4 method for XCMSnExp featureDefinitions(object) <- value  # S4 method for XCMSnExp chromPeaks(   object,   bySample = FALSE,   rt = numeric(),   mz = numeric(),   ppm = 0,   msLevel = integer(),   type = c(\"any\", \"within\", \"apex_within\"),   isFilledColumn = FALSE )  # S4 method for XCMSnExp chromPeaks(object) <- value  # S4 method for XCMSnExp rtime(object, bySample = FALSE, adjusted = hasAdjustedRtime(object))  # S4 method for XCMSnExp mz(object, bySample = FALSE, BPPARAM = bpparam())  # S4 method for XCMSnExp intensity(object, bySample = FALSE, BPPARAM = bpparam())  # S4 method for XCMSnExp spectra(   object,   bySample = FALSE,   adjusted = hasAdjustedRtime(object),   BPPARAM = bpparam() )  # S4 method for XCMSnExp processHistory(object, fileIndex, type, msLevel)  # S4 method for XCMSnExp dropChromPeaks(object, keepAdjustedRtime = FALSE)  # S4 method for XCMSnExp dropFeatureDefinitions(object, keepAdjustedRtime = FALSE, dropLastN = -1)  # S4 method for XCMSnExp dropAdjustedRtime(object)  # S4 method for XCMSnExp profMat(   object,   method = \"bin\",   step = 0.1,   baselevel = NULL,   basespace = NULL,   mzrange. = NULL,   fileIndex,   ... )  # S4 method for XCMSnExp,Param findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   add = FALSE )  # S4 method for XCMSnExp dropFilledChromPeaks(object)  # S4 method for XCMSnExp spectrapply(object, FUN = NULL, BPPARAM = bpparam(), ...)  # S3 method for XCMSnExp c(...)  # S4 method for XCMSnExp chromPeakData(object)  # S4 method for XCMSnExp chromPeakData(object) <- value  # S4 method for XCMSnExp,missing plot(x, y, type = c(\"spectra\", \"XIC\"), peakCol = \"#ff000060\", ...)"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"object adjustedRtime, featureDefinitions, chromPeaks, hasAdjustedRtime, hasFeatures hasChromPeaks either MsFeatureData XCMSnExp object, methods XCMSnExp object. method character(1) defining profile matrix generation method. Allowed \"bin\", \"binlin\", \"binlinbase\" \"intlin\". See details section information. step numeric(1) representing m/z bin size. baselevel numeric(1) representing base value empty elements (.e. m/z bins without measured intensity) set. considered method = \"binlinbase\". See baseValue parameter imputeLinInterpol() details. basespace numeric(1) representing m/z length signal drop base level. Linear interpolation used consecutive data points falling within 2 * basespace . considered method = \"binlinbase\". specified, defaults 0.075. Internally parameter translated distance parameter imputeLinInterpol() function distance = floor(basespace / step). See distance parameter imputeLinInterpol() details. mzrange. Optional numeric(2) manually specifying mz value range used binnind. provided, whole m/z value range used. fileIndex processHistory: optional integer specifying index files/samples ProcessHistory objects retrieved. ... Additional parameters. msLevel integer specifying MS level(s) identified chromatographic peaks returned. bySample logical(1) specifying whether results grouped sample. value adjustedRtime<-: list (length equal     number samples) numeric vectors representing adjusted     retention times per scan. featureDefinitions<-: DataFrame peak     grouping information. See return value featureDefinitions     method expected format. chromPeaks<-: matrix information     detected peaks. See return value chromPeaks method     expected format. mz optional numeric(2) defining mz range chromatographic peaks returned. rt optional numeric(2) defining retention time range chromatographic peaks returned. ppm optional numeric(1) specifying ppm mz range extended. value ppm = 10, peaks within mz[1] - ppm / 1e6 mz[2] + ppm / 1e6 returned. type processHistory: restrict returned ProcessHistory objects analysis steps certain type. Use processHistoryTypes list supported values. chromPeaks: character specifying peaks return rt mz defined. type = \"\" chromatographic peaks partially overlapping range defined mz /rt returned, type = \"within\" returns peaks completely within region type = \"apex_within\" peaks peak's apex within region. isFilledColumn logical(1) whether column \"is_filled\" included returned \"matrix\" providing information peak filled . Alternatively, information provided chromPeakData data frame. adjusted logical(1) whether adjusted raw (.e. original retention times reported files) returned. BPPARAM Parameter class parallel processing. See bpparam. keepAdjustedRtime dropFeatureDefinitions,XCMSnExp: logical(1) defining whether eventually present retention time adjustment dropped. default dropping feature definitions drops retention time adjustment results . dropLastN dropFeatureDefinitions,XCMSnExp: numeric(1) defining number peak grouping related process history steps remove. default dropLastN = -1, dropping chromatographic peaks removes process history steps related peak grouping. Setting e.g. dropLastN = 1 remove recent peak grouping related process history step. param CentWaveParam, MatchedFilterParam, MassifquantParam, MSWParam CentWavePredIsoParam object settings chromatographic peak detection algorithm. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". add findChromPeaks: newly identified chromatographic peaks added peak matrix already identified chromatographic peaks. default (add = FALSE) previous peak detection results removed. FUN spectrapply: function applied spectrum object. x plot: XCMSnExp object. y plot: used. peakCol plot: color used indicate identified chromatographic peaks (combination type = \"XIC\" chromatographic peaks present).","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"profMat: list profile matrix  matrix (matrices fileIndex specified  length(fileIndex) > 1). See profile-matrix     general help information profile matrix. adjustedRtime: bySample = FALSE numeric vector adjusted retention spectrum files/samples within object. bySample = TRUE  list (length equal number samples) adjusted retention times grouped sample. Returns NULL adjusted retention times present. featureDefinitions: DataFrame peak grouping information, row corresponding one mz-rt feature (grouped peaks within across samples) columns \"mzmed\" (median mz value), \"mzmin\" (minimal mz value), \"mzmax\" (maximum mz value), \"rtmed\" (median retention time), \"rtmin\" (minimal retention time), \"rtmax\" (maximal retention time) \"peakidx\". Column \"peakidx\" contains list indices chromatographic peaks (rows) matrix returned chromPeaks method belong feature group. method returns NULL feature definitions present. featureDefinitions supports also parameters mz, rt, ppm type return features within certain ranges (see description chromPeaks details). chromPeaks: bySample = FALSE matrix (row chromatographic peak, rownames representing unique IDs peaks) least following columns: \"mz\" (intensity-weighted mean mz values peak across scans/retention times), \"mzmin\" (minimal mz value), \"mzmax\" (maximal mz value), \"rt\" (retention time peak apex), \"rtmin\" (minimal retention time), \"rtmax\" (maximal retention time), \"\" (integrated, original, intensity peak), \"maxo\" (maximum intentity peak), \"sample\" (sample index peak identified) Depending employed peak detection algorithm verboseColumns parameter , additional columns might returned. parameter isFilledColumn set TRUE column named \"is_filled\" also returned. bySample = TRUE chromatographic peaks returned list matrices, containing chromatographic peaks specific sample. samples peaks detected matrix 0 rows returned. rtime: bySample = FALSE numeric vector retention times scan, bySample = TRUE list numeric vectors retention times per sample. mz: bySample = FALSE list mz values (numeric vectors) scan. bySample = TRUE list mz values per sample. intensity: bySample = FALSE list intensity values (numeric vectors) scan. bySample = TRUE list intensity values per sample. spectra: bySample = FALSE list Spectrum objects. bySample = TRUE result grouped sample, .e. list lists, element outer  list list spectra specific file. processHistory: list ProcessHistory objects providing details individual data processing steps performed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":".processHistory list XProcessHistory objects tracking individual analysis steps performed. msFeatureData MsFeatureData class extending environment containing results chromatographic peak detection (element \"chromPeaks\"), peak grouping (element \"featureDefinitions\") retention time correction (element \"adjustedRtime\") steps. object manipulated directly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"\"chromPeaks\" element msFeatureData slot     equivalent @peaks slot xcmsSet object,     \"featureDefinitions\" contains information @groups     @groupidx slots xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"chromatographic-peak-data","dir":"Reference","previous_headings":"","what":"Chromatographic peak data","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"Chromatographic peak data added XCMSnExp object findChromPeaks function. Functions access chromatographic peak data : hasChromPeaks whether chromatographic peak data available, see help function. chromPeaks access chromatographic peaks (see help). dropChromPeaks remove chromatographic peaks (see help). dropFilledChromPeaks remove filled-peaks (see help). fillChromPeaks fill-missing peaks (see respective help page). plotChromPeaks plot identified peaks file (see respective help page). plotChromPeakImage plot distribution peaks along retention time axis (see respective help page). highlightChromPeaks add chromatographic peaks existing plot Chromatogram (see respective help page).","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"adjusted-retention-times","dir":"Reference","previous_headings":"","what":"Adjusted retention times","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"Adjusted retention times stored XCMSnExp object besides original, raw, retention times, allowing switch raw adjusted times. also possible replace raw retention times adjusted ones applyAdjustedRtime. adjusted retention times added XCMSnExp adjustRtime function. functions related access  adjusted retention times : hasAdjustedRtime whether adjusted retention times available (see help). dropAdjustedRtime remove adjusted retention times (see help). applyAdjustedRtime replace raw retention times adjusted ones (see respective help page). plotAdjustedRtime plot differences adjusted raw retention times (see respective help page).","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"correspondence-results-features","dir":"Reference","previous_headings":"","what":"Correspondence results, features","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"correspondence analysis (groupChromPeaks) adds feature definitions XCMSnExp object. functions related listed : hasFeatures whether correspondence results available (see help). featureDefinitions access definitions features (see help). dropFeatureDefinitions remove correspondence results (see help). featureValues access values features (see respective help page). featureSummary perform simple summary defined features (see respective help page). overlappingFeatures identify features overlapping close m/z - rt space (see respective help page). quantify extract feature intensities put , along feature definitions phenodata information, SummarizedExperiment. See help page details.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## The results from the peak detection are now stored in the XCMSnExp ## object faahko_sub #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 1.22 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 3834  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 1 2 3 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   [1] ko15.CDF...  [3] ko18.CDF #>   Use 'fileNames(.)' to see all files. #> protocolData: none #> featureData #>   featureNames: F1.S0001 F1.S0002 ... F3.S1278 (3834 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave  #>  248 peaks identified in 3 samples. #>  On average 82.7 chromatographic peaks per sample.  ## The detected peaks can be accessed with the chromPeaks method. head(chromPeaks(faahko_sub)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1  ## The settings of the chromatographic peak detection can be accessed with ## the processHistory method processHistory(faahko_sub) #> [[1]] #> Object of class \"XProcessHistory\" #>  type: Peak detection  #>  date: Tue Aug  1 15:54:19 2023  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: CentWaveParam  #>  MS level(s) 1  #>   ## Also the parameter class for the peak detection can be accessed processParam(processHistory(faahko_sub)[[1]]) #> Object of class:  CentWaveParam  #>  Parameters: #>  - ppm: [1] 25 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 40 #>  - prefilter: [1]     3 10000 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 10000 #>  - verboseColumns: [1] FALSE #>  - roiList: list() #>  - firstBaselineCheck: [1] TRUE #>  - roiScales: numeric(0) #>  - extendLengthMSW: [1] FALSE #>  - verboseBetaColumns: [1] FALSE  ## The XCMSnExp inherits all methods from the pSet and OnDiskMSnExp classes ## defined in Bioconductor's MSnbase package. To access the (raw) retention ## time for each spectrum we can use the rtime method. Setting bySample = TRUE ## would cause the retention times to be grouped by sample head(rtime(faahko_sub)) #> F1.S0001 F1.S0002 F1.S0003 F1.S0004 F1.S0005 F1.S0006  #> 2501.378 2502.943 2504.508 2506.073 2507.638 2509.203   ## Similarly it is possible to extract the mz values or the intensity values ## using the mz and intensity method, respectively, also with the option to ## return the results grouped by sample instead of the default, which is ## grouped by spectrum. Finally, to extract all of the data we can use the ## spectra method which returns Spectrum objects containing all raw data. ## Note that all these methods read the information from the original input ## files and subsequently apply eventual data processing steps to them. mzs <- mz(faahko_sub, bySample = TRUE) length(mzs) #> [1] 3 lengths(mzs) #>      1      2      3  #> 445769 440794 442516   ## The full data could also be read using the spectra data, which returns ## a list of Spectrum object containing the mz, intensity and rt values. ## spctr <- spectra(faahko_sub) ## To get all spectra of the first file we can split them by file ## head(split(spctr, fromFile(faahko_sub))[[1]])  ############ ## Filtering ## ## XCMSnExp objects can be filtered by file, retention time, mz values or ## MS level. For some of these filter preprocessing results (mostly ## retention time correction and peak grouping results) will be dropped. ## Below we filter the XCMSnExp object by file to extract the results for ## only the second file. xod_2 <- filterFile(faahko_sub, file = 2) xod_2 #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #> Filter: select file(s) 2. [Thu Dec  7 08:42:31 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 2 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko16.CDF  #> protocolData: none #> featureData #>   featureNames: F2.S0001 F2.S0002 ... F2.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave  #>  100 peaks identified in 1 samples. #>  On average 100 chromatographic peaks per sample.  ## Now the objects contains only the idenfified peaks for the second file head(chromPeaks(xod_2)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into        intb #> CP088 384.1 384.1 384.1 2603.100 2592.145 2614.054   274906.3   274885.96 #> CP089 533.3 533.3 533.3 2686.042 2681.347 2689.172    70456.3    70450.04 #> CP090 344.0 344.0 344.0 2686.042 2648.483 2714.211  5700652.0  5574744.87 #> CP091 366.0 366.0 366.0 2686.042 2639.094 2725.166  3448376.1  3338483.07 #> CP092 343.0 343.0 343.0 2686.042 2645.353 2717.341 26229546.1 25615117.17 #> CP093 365.0 365.0 365.0 2686.042 2639.094 2726.731 15565868.0 14884380.84 #>         maxo    sn sample #> CP088  15187 15186      1 #> CP089  11819 11818      1 #> CP090 151744    87      1 #> CP091  77640    42      1 #> CP092 692352    68      1 #> CP093 345536    48      1  ########## ## Coercing to an xcmsSet object ## ## We can also coerce the XCMSnExp object into an xcmsSet object: xs <- as(faahko_sub, \"xcmsSet\") #> Note: you might want to set/adjust the 'sampclass' of the returned xcmSet object before proceeding with the analysis. head(peaks(xs)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"methods listed page allow filter subset XCMSnExp objects. inherited OnDiskMSnExp object defined MSnbase package adapted XCMSnExp enable correct subsetting preprocessing results. [: subset XCMSnExp object spectra. aware removes preprocessing results, except adjusted retention times keepAdjustedRtime = TRUE passed method. [[: extracts single Spectrum object (defined MSnbase). reported retention time adjusted retention time alignment performed. filterChromPeaks: subset chromPeaks matrix object. Parameter method allows specify chromatographic peaks filtered. Currently, method = \"keep\" supported allows specify chromatographic peaks keep parameter keep (.e. provide logical, integer character defining chromatographic peaks keep). Feature definitions (present) updated correspondingly. filterFeatureDefinitions: allows subset feature definitions XCMSnExp object. Parameter features allow define features keep. can logical, integer (index features keep) character (feature IDs) vector. filterFile: allows reduce XCMSnExp data selected files. Identified chromatographic peaks files retained correspondence results (feature definitions) removed default. force keeping feature definitions use keepFeatures = TRUE. Adjusted retention times (present) retained default present. Use keepAdjustedRtime = FALSE drop . filterMsLevel: reduces XCMSnExp object spectra specified MS level(s). Chromatographic peaks identified features also subsetted respective MS level. See also filterMsLevel documentation MSnbase details examples. filterMz: filters data set based provided m/z value range. chromatographic peaks features (grouped peaks) apex falling within provided mz value range retained (.e. chromPeaks(object)[, \"mz\"] >= mz[1] <= mz[2]). Adjusted retention times, present, kept. filterRt: filters data set based provided retention time range. chromatographic peaks features (grouped peaks) within specified retention time window retained (.e. retention time corresponding peak's apex within specified rt range). retention time correction performed, method default filter object adjusted retention times. argument adjusted allows specify manually whether filtering performed raw adjusted retention times. Filtering retention time drop preprocessing results remove change alignment results (.e. adjusted retention times). method returns empty object spectrum feature within specified retention time range. split: splits XCMSnExp object list XCMSnExp objects based provided parameter f. Note default pre-processing results removed splitting, except adjusted retention times, optional argument keepAdjustedRtime = TRUE provided.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"","code":"# S4 method for XCMSnExp,ANY,ANY,ANY [(x, i, j, ..., drop = TRUE)  # S4 method for XCMSnExp,ANY,ANY [[(x, i, j, drop = FALSE)  # S4 method for XCMSnExp filterMsLevel(object, msLevel., keepAdjustedRtime = hasAdjustedRtime(object))  # S4 method for XCMSnExp filterFile(   object,   file,   keepAdjustedRtime = hasAdjustedRtime(object),   keepFeatures = FALSE )  # S4 method for XCMSnExp filterMz(object, mz, msLevel., ...)  # S4 method for XCMSnExp filterRt(object, rt, msLevel., adjusted = hasAdjustedRtime(object))  # S4 method for XCMSnExp,ANY split(x, f, drop = FALSE, ...)  # S4 method for XCMSnExp filterChromPeaks(   object,   keep = rep(TRUE, nrow(chromPeaks(object))),   method = \"keep\",   ... )  # S4 method for XCMSnExp filterFeatureDefinitions(object, features = integer())"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"x [ [[: XCMSnExp object. [: numeric logical vector specifying spectra data set reduced. [[: single integer character. j [ [[: supported. ... Optional additional arguments. drop [ [[: supported. object XCMSnExp object. msLevel. filterMz, filterRt: numeric defining MS level(s) operations applied object subsetted. keepAdjustedRtime filterFile, filterMsLevel, [, split: logical(1) defining whether adjusted retention times kept, even e.g. features removed (retention time correction performed features). file filterFile: integer defining file index within object subset object file character specifying file names sub set. indices expected increasingly ordered, ordered internally. keepFeatures filterFile: logical(1) whether correspondence results (feature definitions) kept dropped. Defaults keepFeatures = FALSE hence feature definitions removed returned object default. mz filterMz: numeric(2) defining lower upper mz value filtering. rt filterRt: numeric(2) defining retention time window (lower upper bound) filtering. adjusted filterRt: logical indicating whether object filtered original (adjusted = FALSE) adjusted retention times (adjusted = TRUE). spectra: whether retention times individual Spectrum objects adjusted raw retention times. f split vector length equal length x defining x splitted. converted internally factor. keep filterChromPeaks: logical, integer character defining chromatographic peaks retained. method filterChromPeaks: character(1) allowing specify method chromatographic peaks filtered. Currently method = \"keep\" supported (.e. specify parameter keep chromatographic peaks retained). features filterFeatureDefinitions: either integer specifying indices features (rows) keep, logical length matching number rows featureDefinitions character feature (row) names.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"methods return XCMSnExp object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"subsetting methods try ensure returned data consistent. Correspondence results example removed default data set sub-setted file, since correspondence results dependent files correspondence performed. can changed setting keepFeatures = TRUE. adjusted retention times, subsetting methods support argument keepAdjustedRtime (even [ method) forces adjusted retention times retained even default drop .","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"filterFile method removes also process history steps related files object sub-setted updates fileIndex attribute accordingly. Also, method allow arbitrary ordering files re-ordering files within object. Note also filtering methods, also subsetting operations [ drop selected preprocessing results. consolidate alignment results, .e. ensure adjusted retention times always preserved, use applyAdjustedRtime() function object contains alignment results. replaces raw retention times adjusted ones.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"","code":"## Loading a test data set with identified chromatographic peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Subset the dataset to the first and third file. xod_sub <- filterFile(faahko_sub, file = c(1, 3))  ## The number of chromatographic peaks per file for the full object table(chromPeaks(faahko_sub)[, \"sample\"]) #>  #>   1   2   3  #>  87 100  61   ## The number of chromatographic peaks per file for the subset table(chromPeaks(xod_sub)[, \"sample\"]) #>  #>  1  2  #> 87 61   basename(fileNames(faahko_sub)) #> [1] \"ko15.CDF\" \"ko16.CDF\" \"ko18.CDF\" basename(fileNames(xod_sub)) #> [1] \"ko15.CDF\" \"ko18.CDF\"  ## Filter on mz values; chromatographic peaks and features within the ## mz range are retained (as well as adjusted retention times). xod_sub <- filterMz(faahko_sub, mz = c(300, 400)) head(chromPeaks(xod_sub)) #>        mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP002 302   302   302 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344   344   344 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP005 366   366   366 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343   343   343 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #> CP007 365   365   365 2679.783 2634.399 2717.342 14975760.8 14525699.1 357632 #> CP009 313   313   313 2783.070 2768.985 2806.544  1744615.5  1698756.0  74080 #>        sn sample #> CP002  46      1 #> CP003  68      1 #> CP005  49      1 #> CP006  87      1 #> CP007 111      1 #> CP009  44      1 nrow(chromPeaks(xod_sub)) #> [1] 89 nrow(chromPeaks(faahko_sub)) #> [1] 248  ## Filter on rt values. All chromatographic peaks and features within the ## retention time range are retained. Filtering is performed by default on ## adjusted retention times, if present. xod_sub <- filterRt(faahko_sub, rt = c(2700, 2900))  range(rtime(xod_sub)) #> [1] 2700.127 2898.877 head(chromPeaks(xod_sub)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP008 579.1 579.1 579.1 2786.200 2765.855 2806.544  1669436.8  1642951.6  87656 #> CP009 313.0 313.0 313.0 2783.070 2768.985 2806.544  1744615.5  1698756.0  74080 #> CP010 301.0 301.0 301.0 2786.200 2765.855 2809.674  3051847.8  2958809.7 118512 #> CP011 279.0 279.0 279.0 2787.765 2764.290 2814.369 17140627.0 16526792.4 805248 #> CP012 279.1 279.1 279.1 2825.323 2820.629 2828.453   106769.4   106763.2  17760 #> CP018 453.2 453.2 453.2 2720.472 2678.218 2786.200  2987076.3  2986966.8  34208 #>          sn sample #> CP008    57      1 #> CP009    44      1 #> CP010    62      1 #> CP011    45      1 #> CP012 17759      1 #> CP018 33983      1 range(chromPeaks(xod_sub)[, \"rt\"]) #> [1] 2701.693 2895.746  nrow(chromPeaks(faahko_sub)) #> [1] 248 nrow(chromPeaks(xod_sub)) #> [1] 21  ## Extract a single Spectrum faahko_sub[[4]] #> Object of class \"Spectrum1\" #>  Retention time: 41:46  #>  MSn level: 1  #>  Total ion count: 427  #>  Polarity: -1   ## Subsetting using [ removes all preprocessing results - using ## keepAdjustedRtime = TRUE would keep adjusted retention times, if present. xod_sub <- faahko_sub[fromFile(faahko_sub) == 1] #> Warning: Removed preprocessing results xod_sub #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 1 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko15.CDF  #> protocolData: none #> featureData #>   featureNames: F1.S0001 F1.S0002 ... F1.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - -  ## Using split does also remove preprocessing results, but it supports the ## optional parameter keepAdjustedRtime. ## Split the object into a list of XCMSnExp objects, one per file xod_list <- split(faahko_sub, f = fromFile(faahko_sub)) xod_list #> $`1` #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 1 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko15.CDF  #> protocolData: none #> featureData #>   featureNames: F1.S0001 F1.S0002 ... F1.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #>  #> $`2` #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 2 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko16.CDF  #> protocolData: none #> featureData #>   featureNames: F2.S0001 F2.S0002 ... F2.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #>  #> $`3` #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 3 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko18.CDF  #> protocolData: none #> featureData #>   featureNames: F3.S0001 F3.S0002 ... F3.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #>"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"methods listed page XCMSnExp methods inherited parent, OnDiskMSnExp class MSnbase package, alter raw data related data subsetting. Thus calling methods causes xcms pre-processing results removed XCMSnExp object ensure data integrity. bin: allows bin spectra. See bin documentation MSnbase package details examples. clean: removes unused 0 intensity data points. See clean documentation MSnbase package details examples. filterAcquisitionNum: filters XCMSnExp object keeping spectra provided acquisition numbers. See filterAcquisitionNum details examples. normalize method performs basic normalization spectra intensities. See normalize documentation MSnbase package details examples. pickPeaks method performs peak picking. See pickPeaks documentation details examples. removePeaks method removes mass peaks (intensities) lower threshold. Note peaks refer mass peaks, different chromatographic peaks detected analyzed metabolomics experiment! See removePeaks documentation details examples. smooth method smooths spectra. See smooth documentation MSnbase details examples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp bin(x, binSize = 1L, msLevel.)  # S4 method for XCMSnExp clean(object, all = FALSE, verbose = FALSE, msLevel.)  # S4 method for XCMSnExp filterAcquisitionNum(object, n, file)  # S4 method for XCMSnExp normalize(object, method = c(\"max\", \"sum\"), ...)  # S4 method for XCMSnExp pickPeaks(   object,   halfWindowSize = 3L,   method = c(\"MAD\", \"SuperSmoother\"),   SNR = 0L,   ... )  # S4 method for XCMSnExp removePeaks(object, t = \"min\", verbose = FALSE, msLevel.)  # S4 method for XCMSnExp smooth(   x,   method = c(\"SavitzkyGolay\", \"MovingAverage\"),   halfWindowSize = 2L,   verbose = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"x XCMSnExp OnDiskMSnExp object. binSize numeric(1) defining size bin (Dalton). msLevel. bin, clean, filterMsLevel, removePeaks: numeric(1) defining MS level(s) operations applied object subsetted. object XCMSnExp OnDiskMSnExp object. clean: logical(1), TRUE zeros removed. verbose logical(1) whether progress information displayed. n filterAcquisitionNum: integer defining acquisition numbers spectra data set sub-setted. file filterAcquisitionNum: integer defining file index within object subset object file. method normalize: character(1) specifying normalization method. See normalize MSnbase package details. pickPeaks: character(1) defining method. See pickPeaks options. smooth: character(1) defining method. See smooth MSnbase package options details. ... Optional additional arguments. halfWindowSize pickPeaks smooth: integer(1) defining window size peak picking. See pickPeaks smooth MSnbase package details options. SNR pickPeaks: numeric(1) defining signal noise ratio considered. See pickPeaks documentation details. t removePeaks: either numeric(1) \"min\" defining threshold (method) used. See removePeaks details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"methods: XCMSnExp object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":null,"dir":"Reference","previous_headings":"","what":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"featureValues,XCMSnExp : extract matrix feature values rows representing features columns samples. Parameter value allows define column chromPeaks matrix returned. Multiple chromatographic peaks sample can assigned feature. Parameter method allows specify method used cases chose peaks value returned. Parameter `msLevel` allows choose specific MS level feature values returned (given features defined MS level). quantify,XCMSnExp: return preprocessing results SummarizedExperiment object containing feature abundances assay matrix, feature definitions (returned featureDefinitions) rowData phenotype information colData. ideal container processing data. Internally, featureValues method used extract feature abundances, parameters method can passed quantify ....","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp quantify(object, ...)  # S4 method for XCMSnExp featureValues(   object,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   filled = TRUE,   missing = NA,   msLevel = integer() )"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"object XCMSnExp object providing feature definitions. ... quantify: additional parameters passed featureValues method. method character specifying method resolve multi-peak mappings within sample, .e. define representative peak feature samples one peak assigned feature. \"medret\": select peak closest median retention time feature. \"maxint\": select peak yielding largest signal. \"sum\": sum values (value \"\" \"maxo\". value character specifying name column chromPeaks(object) returned. Defaults \"\" case integrated peak area returned. get index peak chromPeaks(object) matrix use \"index\". intensity character specifying name column chromPeaks(objects) matrix containing intensity value peak used conflict resolution method = \"maxint\". filled logical(1) specifying whether values filled-peaks returned . filled = FALSE, NA returned matrix respective peak. See fillChromPeaks details peak filling. missing missing values reported. Allowed values NA (default), numeric missing = \"rowmin_half\". latter replaces NA half row's minimal (non-missing) value. msLevel `featureValues`: `integer` defining MS level(s) feature values returned. default, values features defined MS levels returned.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"featureValues: matrix feature values, columns representing samples, rows features. order features matches order found featureDefinitions(object)  DataFrame. rownames matrix featureDefinitions DataFrame. NA reported features without corresponding chromatographic peak respective sample(s). quantify: SummarizedExperiment representing preprocessing results.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"method equivalent groupval xcmsSet objects. Note missing = 0 used get behaviour groupval, .e. report missing values 0 call fillPeaks.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Containers for chromatographic and peak detection data — XChromatograms","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"XChromatogram object allows store chromatographic data (e.g. extracted ion chromatogram) along identified chromatographic peaks within data. object inherits functions Chromatogram() object MSnbase package. Multiple XChromatogram objects can stored XChromatograms object. class extends MChromatograms() MSnbase package allows thus arrange chromatograms matrix-like structure, columns representing samples rows m/z-retention time ranges. functions described (grouped topic-related sections) Arguments section.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"","code":"XChromatograms(data, phenoData, featureData, chromPeaks, chromPeakData, ...)  XChromatogram(   rtime = numeric(),   intensity = numeric(),   mz = c(NA_real_, NA_real_),   filterMz = c(NA_real_, NA_real_),   precursorMz = c(NA_real_, NA_real_),   productMz = c(NA_real_, NA_real_),   fromFile = integer(),   aggregationFun = character(),   msLevel = 1L,   chromPeaks,   chromPeakData )  # S4 method for XChromatogram show(object)  # S4 method for XChromatogram chromPeaks(   object,   rt = numeric(),   mz = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel )  # S4 method for XChromatogram chromPeaks(object) <- value  # S4 method for XChromatogram,ANY plot(   x,   col = \"#00000060\",   lty = 1,   type = \"l\",   xlab = \"retention time\",   ylab = \"intensity\",   main = NULL,   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakCol = \"#00000060\",   peakBg = \"#00000020\",   peakPch = 1,   ... )  # S4 method for XChromatogram filterMz(object, mz, ...)  # S4 method for XChromatogram filterRt(object, rt, ...)  # S4 method for XChromatogram hasChromPeaks(object)  # S4 method for XChromatogram dropFilledChromPeaks(object)  # S4 method for XChromatogram chromPeakData(object)  # S4 method for XChromatogram chromPeakData(object) <- value  # S4 method for XChromatogram,MergeNeighboringPeaksParam refineChromPeaks(object, param = MergeNeighboringPeaksParam())  # S4 method for XChromatogram filterChromPeaks(object, method = c(\"keepTop\"), ...)  # S4 method for XChromatogram transformIntensity(object, FUN = identity)  # S4 method for XChromatograms show(object)  # S4 method for XChromatograms hasChromPeaks(object)  # S4 method for XChromatograms hasFilledChromPeaks(object)  # S4 method for XChromatograms chromPeaks(   object,   rt = numeric(),   mz = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel )  # S4 method for XChromatograms chromPeakData(object)  # S4 method for XChromatograms filterMz(object, mz, ...)  # S4 method for XChromatograms filterRt(object, rt, ...)  # S4 method for XChromatograms,ANY plot(   x,   col = \"#00000060\",   lty = 1,   type = \"l\",   xlab = \"retention time\",   ylab = \"intensity\",   main = NULL,   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakCol = \"#00000060\",   peakBg = \"#00000020\",   peakPch = 1,   ... )  # S4 method for XChromatograms processHistory(object, fileIndex, type)  # S4 method for XChromatograms hasFeatures(object, ...)  # S4 method for XChromatograms dropFeatureDefinitions(object, ...)  # S4 method for XChromatograms,PeakDensityParam groupChromPeaks(object, param)  # S4 method for XChromatograms featureDefinitions(   object,   mz = numeric(),   rt = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\") )  # S4 method for XChromatograms,ANY,ANY,ANY [(x, i, j, drop = TRUE)  # S4 method for XChromatograms featureValues(   object,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   missing = NA,   ... )  # S4 method for XChromatograms plotChromPeakDensity(   object,   param,   col = \"#00000060\",   xlab = \"retention time\",   main = NULL,   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakCol = \"#00000060\",   peakBg = \"#00000020\",   peakPch = 1,   simulate = TRUE,   ... )  # S4 method for XChromatograms dropFilledChromPeaks(object)  # S4 method for XChromatograms,MergeNeighboringPeaksParam refineChromPeaks(object, param = MergeNeighboringPeaksParam())  # S4 method for XChromatograms filterChromPeaks(object, method = c(\"keepTop\"), ...)  # S4 method for XChromatograms transformIntensity(object, FUN = identity)"},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"data XChromatograms: list Chromatogram XChromatogram objects. phenoData XChromatograms: either data.frame, AnnotatedDataFrame NAnnotatedDataFrame describing phenotypical information samples. featureData XChromatograms: either data.frame AnnotatedDataFrame additional information row chromatograms. chromPeaks XChromatogram: matrix required columns \"rt\", \"rtmin\", \"rtmax\", \"\", \"maxo\" \"sn\". XChromatograms: list, length data, chromatographic peaks chromatogram. element matrix, ordering match order chromatograms data. chromPeakData XChromatogram: DataFrame optional additional annotations chromatographic peak. number rows match number chromatographic peaks. ... filterChromPeaks: additional parameters defining filter chromatographic peaks. See function description details. rtime XChromatogram: numeric retention times (length equal length intensity). intensity XChromatogram: numeric intensity values (length equal length rtime).   mz XChromatogram: numeric(2) representing m/z value range (min, max) chromatogram created. supposed contain real range m/z values contrast filterMz . chromPeaks featureDefinitions: numeric(2) defining m/z range chromatographic peaks features returned. filterMz: numeric(2) defining m/z range chromatographic peaks retained.#' filterMz XChromatogram: numeric(2) representing m/z value range (min, max) used filter original object m/z dimension. applicable use filterMz = c(0, 0). precursorMz XChromatogram: numeric(2) SRM/MRM transitions. Represents mz precursor ion. See details information. productMz XChromatogram: numeric(2) SRM/MRM transitions. Represents mz product. See details information. fromFile XChromatogram: integer(1) index file within OnDiskMSnExp MSnExp object chromatogram extracted. aggregationFun XChromatogram: character(1) specifying function used aggregate intensity values retention time across m/z range. msLevel XChromatogram: integer MS level chromatogram extracted. chromPeaks chromPeakData: extract chromatographic peaks certain MS level. object XChromatogram XChromatograms object. rt chromPeaks featureDefinitions: numeric(2) defining retention time range chromatographic peaks features returned. filterRt: numeric(2) defining retention time range reduce object . ppm chromPeaks featureDefinitions: numeric(1) defining ppm expand provided m/z range. type chromPeaks featureDefinitions: character(1) defining peaks features return rt mz provided: \"\" (default) return peaks even partially overlapping rt, \"within\" return peaks completely within rt \"apex_within\" return peaks apex within rt.   value chromPeaks<-: numeric matrix required columns \"rt\", \"rtmin\", \"rtmax\", \"\" \"maxo\".   x plot: XChromatogram XChromatograms object. col plot: color used draw chromatogram. lty plot plotChromPeakDensity: line type. xlab plot plotChromPeakDensity: x axis label. ylab plot: y axis label. main plot plotChromPeakDensity: optional title plot. peakType plot plotChromPeakDensity: character(1) defining () identified chromatographic peak within chromatogram plotted. Options \"polygon\" (default): draw peak borders peakCol color fill peak area peakBg color, \"point\": indicate peak's apex point, \"rectangle\": draw rectangle around identified peak \"none\": draw peaks. peakCol plot plotChromPeakDensity: foreground color peaks. peakType = \"polygon\" peakType = \"rectangle\" color border. Use NA use foreground color. either single color vector colors length chromPeaks(x) rows. peakBg plot plotChromPeakDensity: background color peaks. peakType = \"polygon\" peakType = \"rectangle\" peak rectangle filled color. Use NA skip. either single color vector colors length chromPeaks(x) rows. peakPch plot plotChromPeakDensity: point character used peakType = \"point\". See plot() graphics package details. param groupChromPeaks plotChromPeakDensity: PeakDensityParam() object settings peak density correspondence analysis algorithm. method featureValues: character(1) specifying method resolve multi-peak mappings within sample sample, .e. select representative peak feature one peak assigned one sample. Options \"medret\" (default): select peak closest median retention time feature, \"maxint\": select peak largest signal \"sum\": sum values peaks (value \"\" \"maxo\"). filterChromPeaks: character(1) defining method used filter chromatographic peaks. See help filterChromPeaks details. FUN transformIntensity: function transform intensity values object. fileIndex processHistory: optional integer specifying index files/samples ProcessHistory objects returned. [: integer row indices subset XChromatograms object. j [: integer column indices subset XChromatograms object. drop [: logical(1) whether dimensionality dropped (possible). Defaults drop = TRUE, thus, length j 1 XChromatogram returned. Note drop ignored length j larger 1, thus XChromatograms returned. missing featureValues: missing values reported. Allowed values NA (default), numeric(1) replace NAs value missing = \"rowmin_half\" replace NAs half row's minimal (non-missing) value. simulate plotChromPeakDensity: logical(1) whether correspondence analysis simulated based available data provided PeakDensityParam() param argument. See section Correspondence analysis details.","code":"For `featureValues`: `character(1)` specifying the name of the column in `chromPeaks(object)` containing the intensity value of the peak that should be used for the `method = \"maxint\"` conflict resolution if. For `plot`: what type of plot should be used for the chromatogram (such as `\"l\"` for lines, `\"p\"` for points etc), see help of [plot()] in the `graphics` package for more details. For `processHistory`: restrict returned processing steps to specific types. Use [processHistoryTypes()] to list all supported values. For `featureValues`: `character(1)` specifying the name of the column in `chromPeaks(object)` that should be returned or `\"index\"` (default) to return the index of the peak associated with the feature in each sample. To return the integrated peak area instead of the index use `value = \"into\"`."},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"See help individual functions.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Highlighting peak area(s) XChromatogram XChromatograms object (plot peakType = \"polygon\") draws polygon representing displayed chromatogram peak's minimal retention time maximal retention time. XChromatograms extracted XCMSnExp() object chromatogram() function might represent actual identified peak area m/z range used extract chromatogram larger peak's m/z.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"creation-of-objects","dir":"Reference","previous_headings":"","what":"Creation of objects","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Objects can created contructor function XChromatogram XChromatograms, respectively. Also, can coerced Chromatogram MChromatograms() objects using (object, \"XChromatogram\") (object, \"XChromatograms\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"filtering-and-subsetting","dir":"Reference","previous_headings":"","what":"Filtering and subsetting","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Besides classical subsetting [ specific filter operations MChromatograms() XChromatograms objects available. See filterColumnsIntensityAbove() details. [ allows subset XChromatograms object row () column (j), j type integer. featureDefinitions also subsetted accordingly peakidx column updated. filterMz filters chromatographic peaks within XChromatogram XChromatograms, column \"mz\" present chromPeaks matrix. case XChromatogram extracted XCMSnExp() object chromatogram() function. chromatographic peaks m/z within m/z range defined mz retained. Also feature definitions (present) subset accordingly. function returns filtered XChromatogram XChromatograms object. filterRt filters chromatogram(s) provided retention time range. eventually present chromatographic peaks apex within retention time range specified rt retained. Also feature definitions, present, filtered accordingly. function returns filtered XChromatogram XChromatograms object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"accessing-data","dir":"Reference","previous_headings":"","what":"Accessing data","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"See also help Chromatogram MSnbase package general information data access. methods listed specific XChromatogram XChromatograms objects. chromPeaks, chromPeaks<-: extract set matrix chromatographic peak definitions. Parameter rt allows specify retention time range peaks returned along parameter type defines overlapping defined (parameter description details). XChromatogram objects function returns matrix columns \"rt\" (retention time peak apex), \"rtmin\" (lower peak boundary), \"rtmax\" (upper peak boundary), \"\" (ingegrated peak signal/area peak), \"maxo\" (maximum instensity peak \"sn\" (signal noise ratio). Note , depending peak detection algorithm, matrix may contain additional columns. XChromatograms objects matrix contains also columns \"row\" \"column\" specifying chromatogram object peak identified. Chromatographic peaks ordered row. chromPeakData, chromPeakData<-: extract set DataFrame() optional chromatographic peak annotations. hasChromPeaks: infer whether XChromatogram (XChromatograms) chromatographic peaks. XChromatogram: returns logical(1), XChromatograms: returns matrix, dimensions object either TRUE FALSE chromatographic peaks available chromatogram respective position. hasFilledChromPeaks: whether XChromatogram (XChromatogram XChromatograms) filled-chromatographic peaks. XChromatogram: returns logical(1), XChromatograms: returns matrix, dimensions object either TRUE FALSE chromatographic peaks available chromatogram respective position. dropFilledChromPeaks: removes filled-chromatographic peaks. See dropFilledChromPeaks() help XCMSnExp() objects information. hasFeatures: XChromatograms objects : correspondence analysis performed m/z-rt feature definitions present. Returns logical(1). dropFeatureDefinitions: XChrmomatograms objects : delete correspondence analysis results (related process history). featureDefinitions: XChromatograms objects . Extract results correspondence analysis (performed groupChromPeaks). Returns DataFrame properties defined m/z-rt features: m/z retention time range. Columns peakidx row contain index chromatographic peaks chromPeaks matrix associated feature row XChromatograms object feature defined. Similar chromPeaks method possible filter returned feature matrix mz, rt ppm parameters. featureValues: XChromatograms objects . Extract abundance estimates individuals features. Note default (parameter value = \"index\" matrix indices peaks chromPeaks matrix associated feature returned. extract integrated peak area use value = \"\". function returns matrix one row per feature (featureDefinitions) column sample (.e. column object). features without peak associated certain sample NA returned. can changed missing argument function. filterChromPeaks: filters chromatographic peaks object depending parameter method method-specific parameters passed additional arguments .... Available methods : method = \"keepTop\": keep top n (default n = 1L) peaks chromatogram ordered column order (defaults order = \"maxo\"). Parameter decreasing (default decreasing = TRUE) can used order peaks descending (decreasing = TRUE) ascending ( decreasing = FALSE) order keep top n peaks largest smallest values, respectively. processHistory: returns list ProcessHistory objects representing individual performed processing steps. Optional parameters type fileIndex allow specify processing steps return.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"manipulating-data","dir":"Reference","previous_headings":"","what":"Manipulating data","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"transformIntensity: transforms intensity values chromatograms provided function FUN. See transformIntensity() MSnbase package details. XChromatogram XChromatograms addition intensity values also columns \"\" \"maxo\" object's chromPeaks matrix transformed function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"plotting-and-visualizing","dir":"Reference","previous_headings":"","what":"Plotting and visualizing","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"plot draws chromatogram highlights addition chromatographic peaks present XChromatogram XChromatograms (unless peakType = \"none\" specified). draw peaks different colors vector color definitions length equal nrow(chromPeaks(x)) submitted  peakCol /peakBg defining one color peak (order peaks chromPeaks(x)). base peak chromatograms total ion chromatograms might better set peakType = \"none\" avoid generating busy plots. plotChromPeakDensity: visualize peak density-based correspondence analysis results. See section Correspondence analysis details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"chromatographic-peak-detection","dir":"Reference","previous_headings":"","what":"Chromatographic peak detection","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"See findChromPeaks-Chromatogram-CentWaveParam information. chromatographic peak detection also possible refine identified chromatographic peaks refineChromPeaks method (e.g. reduce peak detection artifacts). Currently, peak refinement using merge neighboring peaks method available (see MergeNeighboringPeaksParam() detailed description approach.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"correspondence-analysis","dir":"Reference","previous_headings":"","what":"Correspondence analysis","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Identified chromatographic peaks XChromatograms object can grouped features groupChromPeaks function. Currently, correspondence analysis can performed peak density method (see groupChromPeaks details) specifying algorithm settings PeakDensityParam() object. correspondence analysis performed separately row XChromatograms object grouping chromatographic peaks across samples (columns). analysis results stored returned XChromatograms object can accessed featureDefinitions method returns DataFrame one row feature. Column \"row\" specifies row XChromatograms object feature identified. plotChromPeakDensity method can used visualize peak density correspondence results, simulate peak density correspondence analysis chromatographic data. resulting plot consists two panels, upper panel showing chromatographic data well identified chromatographic peaks, lower panel distribution peaks (peak density) along retention time axis. plot shows peak point peak's retention time x-axis, sample found y-axis. distribution peaks along retention time axis visualized density estimate. Grouped chromatographic peaks indicated grey shaded rectangles. Parameter simulate allows define whether correspondence analysis simulated ( simulate=TRUE, based available data provided PeakDensityParam() parameter class) (simulate=FALSE). latter assumed correspondence analysis performed peak density method object. See examples . Abundance estimates feature can extracted featureValues function using parameter value = \"\" extract integrated peak area feature. result matrix, columns samples rows features.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"","code":"## ---- Creation of XChromatograms ---- ## ## Create a XChromatograms from Chromatogram objects dta <- list(Chromatogram(rtime = 1:7, c(3, 4, 6, 12, 8, 3, 2)),     Chromatogram(1:10, c(4, 6, 3, 4, 7, 13, 43, 34, 23, 9)))  ## Create an XChromatograms without peak data xchrs <- XChromatograms(dta)  ## Create an XChromatograms with peaks data pks <- list(matrix(c(4, 2, 5, 30, 12, NA), nrow = 1,     dimnames = list(NULL, c(\"rt\", \"rtmin\", \"rtmax\", \"into\", \"maxo\", \"sn\"))),     NULL) xchrs <- XChromatograms(dta, chromPeaks = pks)  ## Create an XChromatograms from XChromatogram objects dta <- lapply(dta, as, \"XChromatogram\") chromPeaks(dta[[1]]) <- pks[[1]]  xchrs <- XChromatograms(dta, nrow = 1)  hasChromPeaks(xchrs) #>      1     2 #> 1 TRUE FALSE  ## Loading a test data set with identified chromatographic peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Subset the dataset to the first and third file. xod_sub <- filterFile(faahko_sub, file = c(1, 3))  od <- as(xod_sub, \"MsExperiment\")  ## Extract chromatograms for a m/z - retention time slice chrs <- chromatogram(od, mz = 344, rt = c(2500, 3500)) chrs #> MChromatograms with 1 row and 2 columns #>                   1              2 #>      <Chromatogram> <Chromatogram> #> [1,]    length: 639    length: 639 #> phenoData with 2 variables #> featureData with 4 variables  ## --------------------------------------------------- ## ##       Chromatographic peak detection                ## ## --------------------------------------------------- ## ## Perform peak detection using CentWave xchrs <- findChromPeaks(chrs, param = CentWaveParam()) xchrs #> XChromatograms with 1 row and 2 columns #>                    1               2 #>      <XChromatogram> <XChromatogram> #> [1,]        peaks: 2        peaks: 1 #> phenoData with 2 variables #> featureData with 4 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Do we have chromatographic peaks? hasChromPeaks(xchrs) #>      1    2 #> 1 TRUE TRUE  ## Process history processHistory(xchrs) #> [[1]] #> Object of class \"XProcessHistory\" #>  type: Peak detection  #>  date: Thu Dec  7 08:42:34 2023  #>  info:   #>  fileIndex: 1,2  #>  Parameter class: CentWaveParam  #>   ## The chromatographic peaks, columns \"row\" and \"column\" provide information ## in which sample the peak was identified. chromPeaks(xchrs) #>            rt    rtmin    rtmax      into      intb   maxo sn row column #> [1,] 2589.015 2571.801 2612.490  592297.3  554533.9  27600 14   1      1 #> [2,] 2679.783 2646.919 2709.517 5210015.9 5071346.6 152320 30   1      1 #> [3,] 2682.914 2643.790 2731.427 5255689.5 4983388.9 125632 28   1      2  ## Spectifically extract chromatographic peaks for one sample/chromatogram chromPeaks(xchrs[1, 2]) #>            rt   rtmin    rtmax    into    intb   maxo sn #> [1,] 2682.914 2643.79 2731.427 5255689 4983389 125632 28  ## Plot the results plot(xchrs)   ## Plot the results using a different color for each sample sample_colors <- c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\") cols <- sample_colors[chromPeaks(xchrs)[, \"column\"]] plot(xchrs, col = sample_colors, peakBg = cols)   ## Indicate the peaks with a rectangle plot(xchrs, col = sample_colors, peakCol = cols, peakType = \"rectangle\",     peakBg = NA)   ## --------------------------------------------------- ## ##       Correspondence analysis                       ## ## --------------------------------------------------- ## ## Group chromatographic peaks across samples prm <- PeakDensityParam(sampleGroup = rep(1, 2)) res <- groupChromPeaks(xchrs, param = prm)  hasFeatures(res) #> [1] TRUE featureDefinitions(res) #> DataFrame with 2 rows and 10 columns #>         mzmed     mzmin     mzmax     rtmed     rtmin     rtmax    npeaks #>     <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> #> FT1        NA        NA        NA   2681.35   2679.78   2682.91         2 #> FT2        NA        NA        NA   2589.01   2589.01   2589.01         1 #>            X1 peakidx       row #>     <numeric>  <list> <integer> #> FT1         2     2,3         1 #> FT2         1       1         1  ## Plot the correspondence results. Use simulate = FALSE to show the ## actual results. Grouped chromatographic peaks are indicated with ## grey shaded rectangles. plotChromPeakDensity(res, simulate = FALSE)   ## Simulate a correspondence analysis based on different settings. Larger ## bw will increase the smoothing of the density estimate hence grouping ## chromatographic peaks that are more apart on the retention time axis. prm <- PeakDensityParam(sampleGroup = rep(1, 3), bw = 60) plotChromPeakDensity(res, param = prm)   ## Delete the identified feature definitions res <- dropFeatureDefinitions(res) hasFeatures(res) #> [1] FALSE  ## Create a XChromatogram object pks <- matrix(nrow = 1, ncol = 6) colnames(pks) <- c(\"rt\", \"rtmin\", \"rtmax\", \"into\", \"maxo\", \"sn\") pks[, \"rtmin\"] <- 2 pks[, \"rtmax\"] <- 9 pks[, \"rt\"] <- 4 pks[, \"maxo\"] <- 19 pks[, \"into\"] <- 93  xchr <- XChromatogram(rtime = 1:10,     intensity = c(4, 8, 14, 19, 18, 12, 9, 8, 5, 2),     chromPeaks = pks) xchr #> Object of class: XChromatogram #> length of object: 10 #> from file:  #> mz range: [NA, NA] #> rt range: [1, 10] #> MS level: 1 #> Identified chromatographic peaks (1): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn  #>  4\t2\t9\t93\t19\tNA  ## Add arbitrary peak annotations df <- DataFrame(peak_id = c(\"a\")) xchr <- XChromatogram(rtime = 1:10,     intensity = c(4, 8, 14, 19, 18, 12, 9, 8, 5, 2),     chromPeaks = pks, chromPeakData = df) xchr #> Object of class: XChromatogram #> length of object: 10 #> from file:  #> mz range: [NA, NA] #> rt range: [1, 10] #> MS level: 1 #> Identified chromatographic peaks (1): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn  #>  4\t2\t9\t93\t19\tNA chromPeakData(xchr) #> DataFrame with 1 row and 3 columns #>       peak_id  ms_level is_filled #>   <character> <integer> <logical> #> 1           a         1     FALSE  ## Extract the chromatographic peaks chromPeaks(xchr) #>      rt rtmin rtmax into maxo sn #> [1,]  4     2     9   93   19 NA  ## Plotting of a single XChromatogram object ## o Don't highlight chromatographic peaks plot(xchr, peakType = \"none\")  ## o Indicate peaks with a polygon plot(xchr)   ## Add a second peak to the data. pks <- rbind(chromPeaks(xchr), c(7, 7, 10, NA, 15, NA)) chromPeaks(xchr) <- pks  ## Plot the peaks in different colors plot(xchr, peakCol = c(\"#ff000080\", \"#0000ff80\"),     peakBg = c(\"#ff000020\", \"#0000ff20\"))  ## Indicate the peaks as rectangles plot(xchr, peakCol = c(\"#ff000060\", \"#0000ff60\"), peakBg = NA,     peakType = \"rectangle\")   ## Filter the XChromatogram by retention time xchr_sub <- filterRt(xchr, rt = c(4, 6)) xchr_sub #> Object of class: XChromatogram #> length of object: 3 #> from file:  #> mz range: [NA, NA] #> rt range: [4, 6] #> MS level: 1 #> Identified chromatographic peaks (1): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn  #>  4\t2\t9\t93\t19\tNA plot(xchr_sub)"},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":null,"dir":"Reference","previous_headings":"","what":"Next Generation xcms Result Object — filterFeatureDefinitions","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"XcmsExperiment data container xcms preprocessing results (.e. results chromatographic peak detection, alignment correspondence analysis). provides functionality XCMSnExp object, uses advanced modern MS infrastructure provided MsExperiment Spectra Bioconductor packages. comes higher flexibility store data. Documentation various functions XcmsExperiment objects grouped topic provided sections . default xcms workflow perform chromatographic peak detection using findChromPeaks() optionally refine identified chromatographic peaks using refineChromPeaks() perform alignment (retention time adjustment) using adjustRtime(). Depending method used requires run correspondence analysis first perform correspondence analysis using groupChromPeaks() function group chromatographic peaks across samples define LC-MS features. optionally perform gap-filling rescue signal samples chromatographic peak identified hence missing value reported. can performed using fillChromPeaks() function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"","code":"filterFeatureDefinitions(object, ...)  # S4 method for MsExperiment filterRt(object, rt = numeric(), ...)  # S4 method for MsExperiment filterMzRange(object, mz = numeric(), msLevel. = uniqueMsLevels(object))  # S4 method for MsExperiment filterMz(object, mz = numeric(), msLevel. = uniqueMsLevels(object))  # S4 method for MsExperiment filterMsLevel(object, msLevel. = uniqueMsLevels(object))  # S4 method for MsExperiment uniqueMsLevels(object)  # S4 method for MsExperiment filterFile(object, file = integer(), ...)  # S4 method for MsExperiment rtime(object)  # S4 method for MsExperiment fromFile(object)  # S4 method for MsExperiment fileNames(object)  # S4 method for MsExperiment polarity(object)  # S4 method for MsExperiment filterIsolationWindow(object, mz = numeric())  # S4 method for MsExperiment chromatogram(   object,   rt = matrix(nrow = 0, ncol = 2),   mz = matrix(nrow = 0, ncol = 2),   aggregationFun = \"sum\",   msLevel = 1L,   isolationWindowTargetMz = NULL,   chunkSize = 2L,   return.type = \"MChromatograms\",   BPPARAM = bpparam() )  featureArea(   object,   mzmin = min,   mzmax = max,   rtmin = min,   rtmax = max,   msLevel = integer(),   features = character() )  # S4 method for MsExperiment,missing plot(x, y, msLevel = 1L, peakCol = \"#ff000060\", ...)  # S4 method for XcmsExperiment,ANY,ANY,ANY [(x, i, j, ..., drop = TRUE)  # S4 method for XcmsExperiment filterIsolationWindow(object, mz = numeric())  # S4 method for XcmsExperiment filterRt(object, rt, msLevel.)  # S4 method for XcmsExperiment filterMzRange(object, mz = numeric(), msLevel. = uniqueMsLevels(object))  # S4 method for XcmsExperiment filterMsLevel(object, msLevel. = uniqueMsLevels(object))  # S4 method for XcmsExperiment hasChromPeaks(object, msLevel = integer())  # S4 method for XcmsExperiment dropChromPeaks(object, keepAdjustedRtime = FALSE)  # S4 method for XcmsExperiment chromPeaks(object) <- value  # S4 method for XcmsExperiment chromPeaks(   object,   rt = numeric(),   mz = numeric(),   ppm = 0,   msLevel = integer(),   type = c(\"any\", \"within\", \"apex_within\"),   isFilledColumn = FALSE )  # S4 method for XcmsExperiment chromPeakData(object) <- value  # S4 method for XcmsExperiment chromPeakData(   object,   msLevel = integer(),   return.type = c(\"DataFrame\", \"data.frame\") )  # S4 method for XcmsExperiment filterChromPeaks(   object,   keep = rep(TRUE, nrow(chromPeaks(object))),   method = \"keep\",   ... )  # S4 method for XcmsExperiment dropAdjustedRtime(object)  # S4 method for MsExperiment hasAdjustedRtime(object)  # S4 method for XcmsExperiment rtime(object, adjusted = hasAdjustedRtime(object))  # S4 method for XcmsExperiment adjustedRtime(object)  # S4 method for XcmsExperiment hasFeatures(object, msLevel = integer())  # S4 method for XcmsExperiment featureDefinitions(object) <- value  # S4 method for XcmsExperiment featureDefinitions(   object,   mz = numeric(),   rt = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel = integer() )  # S4 method for XcmsExperiment dropFeatureDefinitions(object, keepAdjustedRtime = FALSE)  # S4 method for XcmsExperiment filterFeatureDefinitions(object, features = integer())  # S4 method for XcmsExperiment hasFilledChromPeaks(object)  # S4 method for XcmsExperiment dropFilledChromPeaks(object)  # S4 method for XcmsExperiment quantify(object, ...)  # S4 method for XcmsExperiment featureValues(   object,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   filled = TRUE,   missing = NA_real_,   msLevel = integer() )  # S4 method for XcmsExperiment chromatogram(   object,   rt = matrix(nrow = 0, ncol = 2),   mz = matrix(nrow = 0, ncol = 2),   aggregationFun = \"sum\",   msLevel = 1L,   chunkSize = 2L,   isolationWindowTargetMz = NULL,   return.type = c(\"XChromatograms\", \"MChromatograms\"),   include = character(),   chromPeaks = c(\"apex_within\", \"any\", \"none\"),   BPPARAM = bpparam() )  # S4 method for XcmsExperiment processHistory(object, type)  # S4 method for XcmsExperiment filterFile(   object,   file,   keepAdjustedRtime = hasAdjustedRtime(object),   keepFeatures = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"object XcmsExperiment object. ... Additional optional parameters. quantify: parameter featureValues call used extract feature value matrix. rt chromPeaks featureDefinitions: numeric(2) defining retention time range chromatographic peaks features returned. full range used default. chromatogram: two column numerical matrix row representing lower upper retention time window(s) chromatograms. provided full retention time range used. mz chromPeaks featureDefinitions: numeric(2) optionally defining m/z range chromatographic peaks feature definitions returned. full m/z range used default. chromatogram: two-column numerical matrix row representing m/z range aggregated chromatogram. provided full m/z range data used (hence total ion chromatogram returned aggregationFun = \"sum\" used). filterIsolationWindow: numeric(1) defining m/z contained within spectra's isolation window. msLevel. filterRt: ignored. filterRt always filter retention times MS levels regardless parameter. chromatogram: integer MS level chromatogram(s) extracted. either length 1 length equal numer rows parameters mz rt defining m/z rt regions chromatograms created. Defaults msLevel = 1L. filterMsLevel: integer defining MS level(s) data subset. file filterFile: integer indices samples (files) data subsetted. aggregationFun chromatogram: character(1) defining function used aggregate intensities retention time (.e. spectrum) along specified m/z range (parameter mz). Defaults aggregationFun = \"sum\" hence intensities summed . Alternatively, use aggregationFun = \"max\" use maximal intensity per m/z range create base peak chromatogram (BPC). msLevel integer defining MS level (multiple MS level function supports ). isolationWindowTargetMz chromatogram: numeric (length equal number rows rt mz) isolation window target m/z MS2 spectra chromatgrom generated. MS1 data (msLevel = 1L, default), parameter ignored. See examples chromatogram information. chunkSize chromatogram: integer(1) defining number files data loaded time memory. Defaults chunkSize = 2L. return.type chromPeakData: character(1) defining class returned object. Can either \"DataFrame\" (default) \"data.frame\". chromatogram: character(1) defining type returned object. Currently return.type = \"MChromatograms\" supported. BPPARAM chromatogram: parallel processing setup. Defaults BPPARAM = bpparam(). See bpparam() information. mzmin featureArea: function calculate \"mzmin\" feature based \"mzmin\" values individual chromatographic peaks assigned feature. Defaults mzmin = min. mzmax featureArea: function calculate \"mzmax\" feature based \"mzmax\" values individual chromatographic peaks assigned feature. Defaults mzmax = max. rtmin featureArea: function calculate \"rtmin\" feature based \"rtmin\" values individual chromatographic peaks assigned feature. Defaults rtmin = min. rtmax featureArea: function calculate \"rtmax\" feature based \"rtmax\" values individual chromatographic peaks assigned feature. Defaults rtmax = max. features filterFeatureDefinitions featureArea: logical, integer character defining features keep extract feature , respectively. See function description information. x XcmsExperiment object. y plot: defined supported. peakCol plot: defines border color rectangles indicating identified chromatographic peaks. single color supported. Defaults `peakCol = \"#ff000060\". [: integer logical defining samples/files subset. j [: supported. drop [: ignored. keepAdjustedRtime logical(1): whether adjusted retention times (present) retained. value featureValues: character(1) defining value reported feature sample. Can column chromPeaks matrix \"index\" simply index assigned peak returned. Defaults value = \"\" thus integrated peak area reported. ppm chromPeaks featureDefinitions: optional numeric(1) specifying ppm m/z range (defined mz extended. value ppm = 10, peaks within mz[1] - ppm / 1e6 mz[2] + ppm / 1e6 returned. type chromPeaks featureDefinitions either mz rt defined : character(1): defining peaks (features) returned. type = \"\": returns chromatographic peaks features also partially overlapping provided ranges. type = \"within\": returns peaks features completely within region defined mz /rt. type = \"apex_within\": returns peaks features m/z retention time peak's apex within region defined mz /rt. processHistory: restrict returned processing steps specific types. Use processHistoryTypes() list supported values. isFilledColumn chromPeaks: logical(1) whether column \"is_filled\" included returned matrix information whether peak detected filled-. Note information also provided chromPeakData data frame. keep filterChromPeaks: logical, integer character specifying chromatographic peaks keep. logical length keep needs match number rows chromPeaks. Alternatively, keep allows specify index (row) peaks keep ID (.e. row name chromPeaks). method featureValues: character(1) specifying method resolve multi-peak mappings within sample (correspondence analysis can assign one chromatographic peak within sample feature, e.g. close retention time). Options: method = \"medret\": report value chromatographic peak closest feature's median retention time. method = \"maxint\": report value chromatographic peak largest signal (parameter intensity allows select column chromPeaks used signal). method = \"sum\": sum value chromatographic peaks sample assigned feature. default method = \"medret\". filterChromPeaks: currently method = \"keep\" supported. adjusted rtime,XcmsExperiment: whether adjusted raw retention times returned. default return adjusted retention times, available. intensity featureValues: character(1) specifying name column chromPeaks(objects) matrix containing intensity value peak used conflict resolution method = \"maxint\". filled featureValues: logical(1) specifying whether values filled-peaks reported. filled = TRUE (default) filled peak values returned, otherwise NA reported respective features samples peak detected. missing featureValues: default value missing values. Allows define value reported missing peak intensity. Defaults missing = NA_real_. include chromatogram: deprecated; use parameter chromPeaks instead. chromPeaks chromatogram: character(1) defining chromatographic peaks returned. Can either chromPeaks = \"apex_within\" (default) return chromatographic peaks m/z RT apex within m/z retention time window, chromPeaks = \"\" chromatographic peaks overlapping m/z - retention time window chromPeaks = \"none\" include chromatographic peaks. See also parameter type additional information. keepFeatures subsetting functions ([, filterFile): logical(1): wheter eventually present feature definitions retained returned (filtered) object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"subsetting-and-filtering","dir":"Reference","previous_headings":"","what":"Subsetting and filtering","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"[: subset XcmsExperiment sample (parameter ). Subsetting default drop correspondence results (subsetting samples obviously affect feature definition) alignment results (adjusted retention times) identified chromatographic peaks (selected samples) retained. preprocessing results kept dropped can also configured optional parameters keepChromPeaks (default TRUE), keepAdjustedRtime (default FALSE) keepFeatures (default FALSE). filterChromPeaks: filter chromatographic peaks XcmsExperiment keeping specified parameter keep. Returns XcmsExperiment filtered data. Chromatographic peaks retain can specified either providing index chromPeaks matrix, ID (rowname chromPeaks) logical vector length number rows chromPeaks. Assignment chromatographic peaks updated eventually present feature definitions filtering. filterFeatureDefinitions: filter feature definitions XcmsExperiment keeping defined parameter features, can logical length equal number features, integer index features featureDefinitions(object) keep character feature IDs (.e. row names featureDefinitions(object)). filterFile: filter XcmsExperiment (MsExperiment) file (sample). index samples data subsetted can specified parameter file. sole purpose function provide backward compatibility MSnbase package. Wherever possible, [ function used instead sample-based subsetting. Parameters keepChromPeaks, keepAdjustedRtime keepChromPeaks can passed using .... Note also contrast [, filterFile support subsetting arbitrary order. filterIsolationWindow: filter spectra within MsExperiment XcmsExperiment object keeping isolation window containing specified m/z (.e., keeping spectra \"isolationWindowLowerMz\" smaller user-provided mz \"isolationWindowUpperMz\" larger mz). XcmsExperiment also chromatographic peaks (subsequently also features) removed range \"isolationWindowLowerMz\" \"isolationWindowUpperMz\" (columns chromPeakData) contain user provided mz. filterMsLevel: filter data XcmsExperiment MsExperiment keep data MS level(s) specified parameter msLevel.. filterMz, filterMzRange: filter spectra within XcmsExperiment MsExperiment specified m/z range (parameter mz). XcmsExperiment also identified chromatographic peaks features filtered keeping within specified m/z range (.e. m/z peak apex within m/z range). Parameter msLevels. allows restrict filtering specified MS levels. default data MS levels filtered. filterRt: filter XcmsExperiment keeping data within specified retention time range (parameter rt). function keep preprocessing results present within retention time range: identified chromatographic peaks retention time apex position within retention time range rt retained along, present, associated features. Parameter msLevel. currently ignored, .e. filtering always performed MS levels object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"functionality-related-to-chromatographic-peaks","dir":"Reference","previous_headings":"","what":"Functionality related to chromatographic peaks","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"chromatogram: extract chromatographic data data set. Parameters mz rt allow define specific m/z - retention time regions extract data (e.g. extracted ion chromatograms EICs). parameters expected numerical two-column matrices first column defining lower second upper margin. row can define separate m/z - retention time region. Currently function returns MChromatograms() object object MsExperiment , object XcmsExperiment, either MChromatograms XChromatograms() depending parameter return.type (can either \"MChromatograms\" \"XChromatograms\"). latter also chromatographic peaks detected within provided m/z retention times returned. Parameter chromPeaks allows specify chromatographic peaks reported. See documentation chromPeaks parameter information. XcmsExperiment contains correspondence results, also associated feature definitions included returned XChromatograms. default function returns chromatograms MS1 data, setting parameter msLevel = 2L possible e.g. extract also MS2 chromatograms. msLevel 1 addition important also specify isolationWindowTargetMz MS2 data extracted (e.g. SWATH data MS2 spectra created different m/z isolation windows isolationWindowTargetMz parameter allows define MS2 chromatogram extracted. Note future efficient data structures chromatographic data available well. chromPeaks: returns numeric matrix identified chromatographic peaks. row represents chromatographic peak identified one sample (file). number columns depends peak detection algorithm (see findChromPeaks()) methods return following columns: \"mz\" (intensity-weighted mean m/z values mass peaks included chromatographic peak), \"mzmin\" ( smallest m/z value mass peak chromatographic peak), \"mzmax\" (largest m/z value mass peak chromatographic peak), \"rt\" (retention time peak apex), \"rtmin\" (retention time first scan/mass peak chromatographic peak), \"rtmax\" (retention time last scan/mass peak chromatographic peak), \"\" (integrated intensity chromatographic peak), \"maxo\" (maximal intensity mass peak chromatographic peak), \"sample\" (index sample object peak identified). Parameters rt, mz, ppm, msLevel type allow extract subsets identified chromatographic peaks object. See parameter description details. chromPeakData: returns DataFrame potential additional annotations identified chromatographic peaks. row DataFrame corresponds row (index row name) chromPeaks matrix. default annotations \"ms_level\" (MS level peak identified) \"is_filled\" (whether chromatographic peak detected (findChromPeaks) filled-(fillChromPeaks). chromPeakSpectra: extract MS spectra identified chromatographic peaks. can either (full scan) MS1 spectra retention times retention time range chromatographic peak, MS2 spectra (present) retention time within retention time range (MS1) chromatographic peak precursor m/z within m/z range chromatographic peak single, selected spectra depending total signal highest signal. Parameter msLevel allows define MS level spectra extracted, parameter method allows define selected spectra returned. See chromPeakSpectra() details. dropChromPeaks: removes () chromatographic peak detection results object. also remove correspondence results (.e. features) eventually present adjusted retention times object alignment performed peak detection. Alignment results (adjusted retention times) can retained parameter keepAdjustedRtime set TRUE. dropFilledChromPeaks: removes chromatographic peaks added gap filling fillChromPeaks. fillChromPeaks: perform gap filling integrate signal missing values samples chromatographic peak found. depends correspondence results, hence groupChromPeaks needs called first. details options see fillChromPeaks(). findChromPeaks: perform chromatographic peak detection. See findChromPeaks() details. hasChromPeaks: whether object contains peak detection results. Parameter msLevel allows check whether peak detection results available specified MS level(s). hasFilledChromPeaks: whether gap-filling results (.e., filled-chromatographic peaks) present. manualChromPeaks: manually add chromatographic peaks defining m/z retention time ranges. See manualChromPeaks() details examples. plotChromPeakImage: show density identified chromatographic peaks per file along retention time. See plotChromPeakImage() details. plotChromPeaks: indicate identified chromatographic peaks one sample RT-m/z space. See plotChromPeaks() details. refineChromPeaks: refines identified chromatographic peaks object. See refineChromPeaks() details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"functionality-related-to-alignment","dir":"Reference","previous_headings":"","what":"Functionality related to alignment","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"adjustedRtime: extract adjusted retention times. just alias rtime(object, adjusted = TRUE). adjustRtime: performs retention time adjustment (alignment) data. See adjustRtime() details. applyAdjustedRtime: replaces original (raw) retention times adjusted ones. See applyAdjustedRtime() information. dropAdjustedRtime: drops alignment results (adjusted retention time) result object. also reverts retention times identified chromatographic peaks present result object. Note results correspondence analysis (.e. feature definitions) dropped (correspondence analysis performed alignment). can overruled keepAdjustedRtime = TRUE. hasAdjustedRtime: whether alignment performed object (.e., object contains alignment results). plotAdjustedRtime: plot alignment results; see plotAdjustedRtime() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"functionality-related-to-correspondence-analysis","dir":"Reference","previous_headings":"","what":"Functionality related to correspondence analysis","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"dropFeatureDefinitions: removes correspondence analysis results object well filled-chromatographic peaks. default (parameter keepAdjustedRtime = FALSE) also alignment results removed alignment performed correspondence analysis. can overruled keepAdjustedRtime = TRUE. featureArea: returns matrix columns \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" m/z retention time range feature (row) object. default represent minimal m/z retention times well maximal m/z retention times chromatographi peaks assigned feature. Note one sample one chromatographic peak assigned feature one higher intensity considered. Parameter features allows extract values selected features . Parameters mzmin, mzmax, rtmin rtmax allow define function calculate reported \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" values. featureChromatograms: extract ion chromatograms (EICs) feature object. See featureChromatograms() details. featureDefinitions: returns data.frame feature definitions empty data.frame correspondence analysis results present. Parameters msLevel, mz, ppm rt allow define subsets feature definitions returned parameter type defining parameters used subset returned data.frame. See parameter descriptions details. featureSpectra: returns Spectra() List Spectra (MS1 MS2) spectra associated feature. See featureSpectra() details available parameters. featuresSummary: calculate simple summary features. See featureSummary() details. groupChromPeaks: performs correspondence analysis (.e., grouping chromatographic peaks LC-MS features). See groupChromPeaks() details. hasFeatures: whether correspondence analysis results presentin object. optional parameter msLevel allows define  MS level(s) determined feature definitions available. overlappingFeatures: identify features overlapping close m/z - rt dimension. See overlappingFeatures() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"extracting-data-and-results-from-an-xcmsexperiment","dir":"Reference","previous_headings":"","what":"Extracting data and results from an XcmsExperiment","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"Preprocessing results can extracted using following functions: chromPeaks: extract identified chromatographic peaks. See section chromatographic peak detection details. featureDefinitions: extract definition features (chromatographic peaks grouped across samples). See section correspondence analysis details. featureValues: extract matrix values features sample (file). Rows features, columns samples. value returned can defined parameter value, can column chromPeaks matrix. default (value = \"\") integrated chromatographic peak intensities returned. parameter msLevel possible extract values features certain MS levels. correspondence analysis, one chromatographic peak per sample can assigned feature (e.g. close retention time). Parameter method allows define strategy deal cases: method = \"medret\": report value chromatographic peak apex position closest feautre's median retention time. method = \"maxint\": report value chromatographic peak largest signal (parameter intensity allows define column chromPeaks selected; defaults intensity = \"\"). method = \"sum\"`: sum values chromatographic peaks assigned feature sample. quantify: extract correspondence analysis results SummarizedExperiment(). feature values used assay returned SummarizedExperiment, rowData contains featureDefinitions (without column \"peakidx\") colData sampleData object. Additional parameters featureValues function (used extract feature value matrix) can passed via ....","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"visualization","dir":"Reference","previous_headings":"","what":"Visualization","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"plot: plot file position individual peaks m/z - retention time space (color-coded intensity) base peak chromatogram. function ideally called data subset (.e. using filterRt filterMz restrict region interest). Parameter msLevel allows define MS level plot created. x XcmsExperiment available identified chromatographic peaks, also region defining peaks indicated rectangle. Parameter peakCol allows define color border rectangles. plotAdjustedRtime: plot alignment results; see plotAdjustedRtime() information. plotChromPeakImage: show density identified chromatographic peaks per file along retention time. See plotChromPeakImage() details. plotChromPeaks: indicate identified chromatographic peaks one sample RT-m/z space. See plotChromPeaks() details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"general-functionality-and-functions-for-backward-compatibility","dir":"Reference","previous_headings":"","what":"General functionality and functions for backward compatibility","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"uniqueMsLevels: returns unique MS levels spectra object. functions listed ensure compatibility older XCMSnExp() xcms result object. fileNames: returns original data file names spectra data. Ideally, dataOrigin dataStorage spectra variables object's spectra used instead. fromFile: returns file (sample) index spectrum within object. Generally, subsetting sample using [ preferred way get spectra specific sample. polarity: returns polarity information spectrum object. processHistory: returns list ProcessHistory process history objects contain also parameter object used different processings. Optional parameter type allows query specific processing steps. rtime: extract retention times spectra MsExperiment XcmsExperiment object. thus shortcut rtime(spectra(object)) preferred way extract retention times MsExperiment. rtime method XcmsExperiment additional parameter adjusted allows define whether adjusted retention times (present - adjusted = TRUE) raw retention times (adjusted = FALSE) returned. default adjusted retention times returned available.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"differences-compared-to-the-xcmsnexp-object","dir":"Reference","previous_headings":"","what":"Differences compared to the XCMSnExp() object","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"Subsetting [ supports arbitrary ordering.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"","code":"## Creating a MsExperiment object representing the data from an LC-MS ## experiment. library(MsExperiment)  ## Defining the raw data files fls <- c(system.file('cdf/KO/ko15.CDF', package = \"faahKO\"),          system.file('cdf/KO/ko16.CDF', package = \"faahKO\"),          system.file('cdf/KO/ko18.CDF', package = \"faahKO\"))  ## Defining a data frame with the sample characterization df <- data.frame(mzML_file = basename(fls),                 sample = c(\"ko15\", \"ko16\", \"ko18\")) ## Importing the data. This will initialize a `Spectra` object representing ## the raw data and assign these to the individual samples. mse <- readMsExperiment(spectraFiles = fls, sampleData = df)  ## Extract a total ion chromatogram and base peak chromatogram ## from the data bpc <- chromatogram(mse, aggregationFun = \"max\") tic <- chromatogram(mse)  ## Plot them par(mfrow = c(2, 1)) plot(bpc, main = \"BPC\") plot(tic, main = \"TIC\")   ## Extracting MS2 chromatographic data ## ## To show how MS2 chromatograms can be extracted we first load a DIA ## (SWATH) data set. mse_dia <- readMsExperiment(system.file(\"TripleTOF-SWATH\",     \"PestMix1_SWATH.mzML\", package = \"msdata\"))  ## Extracting MS2 chromatogram requires also to specify the isolation ## window from which to extract the data. Without that chromatograms ## will be empty: chr_ms2 <- chromatogram(mse_dia, msLevel = 2L) intensity(chr_ms2[[1L]]) #> numeric(0)  ## First we list available isolation windows table(isolationWindowTargetMz(spectra(mse_dia))) #>  #> 163.75 208.95 244.05 270.85  299.1  329.8 367.35 601.85  #>   1000   1000   1000   1000   1000   1000   1000   1000   ## We can then extract the TIC of MS2 data for a specific isolation window chr_ms2 <- chromatogram(mse_dia, msLevel = 2L,     isolationWindowTargetMz = 244.05) plot(chr_ms2)  #### ## Chromatographic peak detection  ## Perform peak detection on the data using the centWave algorith. Note ## that the parameters are chosen to reduce the run time of the example. p <- CentWaveParam(noise = 10000, snthresh = 40, prefilter = c(3, 10000)) xmse <- findChromPeaks(mse, param = p) xmse #> Object of class XcmsExperiment  #>  Spectra: MS1 (3834)  #>  Experiment data: 3 sample(s) #>  Sample data links: #>   - spectra: 3 sample(s) to 3834 element(s). #>  xcms results: #>   - chromatographic peaks: 248 in MS level(s): 1   ## Have a quick look at the identified chromatographic peaks head(chromPeaks(xmse)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1  ## Extract chromatographic peaks identified between 3000 and 3300 seconds chromPeaks(xmse, rt = c(3000, 3300), type = \"within\") #>          mz mzmin mzmax       rt    rtmin    rtmax        into        intb #> CP021 453.2 453.2 453.2 3063.196 3035.027 3114.840  3001594.78  3001514.97 #> CP030 361.1 361.1 361.1 3147.704 3141.444 3153.964   367361.84   362278.28 #> CP031 340.3 340.3 340.3 3230.646 3225.952 3233.776    68671.25    68664.99 #> CP032 526.2 526.2 526.2 3243.166 3238.471 3246.296   374698.56   374692.30 #> CP033 313.1 313.1 313.1 3276.030 3254.121 3290.115  1199152.63  1162839.05 #> CP034 454.1 454.1 454.1 3276.030 3261.945 3294.809 12283448.95 12024951.77 #> CP108 307.1 307.1 307.1 3143.009 3121.099 3164.918  2191519.65  2063202.03 #> CP109 278.1 278.1 278.1 3196.217 3180.568 3213.432   657840.25   651048.36 #> CP110 526.1 526.1 526.1 3179.003 3150.833 3210.302 21334966.98 21147392.01 #> CP111 380.1 380.1 380.1 3152.398 3132.054 3174.308  2201517.91  2201474.09 #> CP112 380.1 380.1 380.1 3210.302 3210.302 3216.561   187031.44   187023.61 #> CP113 380.1 380.1 380.1 3202.477 3174.308 3216.561  1156780.73  1156736.91 #> CP114 380.1 380.1 380.1 3222.821 3216.561 3240.036   485818.25   485793.21 #> CP115 286.2 286.2 286.2 3258.815 3246.296 3280.725  1264118.64  1247233.70 #> CP116 308.1 308.1 308.1 3261.945 3241.601 3285.419  2066854.37  2025917.82 #> CP204 380.1 380.1 380.1 3150.835 3128.925 3171.179  2036803.99  1937655.24 #> CP205 286.2 286.2 286.2 3250.992 3233.777 3257.252   732016.23   722185.86 #> CP206 568.2 568.2 568.2 3207.173 3185.264 3232.212  3951832.25  3866503.74 #>         maxo    sn sample #> CP021  53096 53095      1 #> CP030  49240    57      1 #> CP031  11391 11390      1 #> CP032  60800 60799      1 #> CP033  55392    50      1 #> CP034 554112    58      1 #> CP108 101400    43      2 #> CP109  30208    54      2 #> CP110 622144   251      2 #> CP111  95176 95175      2 #> CP112  25144 28791      2 #> CP113  28920 28919      2 #> CP114  23176 23535      2 #> CP115  67600    41      2 #> CP116  96272    78      2 #> CP204  82624    49      3 #> CP205  74824    68      3 #> CP206 164352   113      3  ## Extract ion chromatograms (EIC) for the first two chromatographic ## peaks. chrs <- chromatogram(xmse,     mz = chromPeaks(xmse)[1:2, c(\"mzmin\", \"mzmax\")],     rt = chromPeaks(xmse)[1:2, c(\"rtmin\", \"rtmax\")])  ## An EIC for each sample and each of the two regions was extracted. ## Identified chromatographic peaks in the defined regions are extracted ## as well. chrs #> XChromatograms with 2 rows and 3 columns #>                    1               2               3 #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 1        peaks: 0        peaks: 0 #> [2,]        peaks: 1        peaks: 0        peaks: 0 #> phenoData with 3 variables #> featureData with 4 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Plot the EICs for the second defined region plot(chrs[2, ])   ## Subsetting the data to the results (and data) for the second sample a <- xmse[2] nrow(chromPeaks(xmse)) #> [1] 248 nrow(chromPeaks(a)) #> [1] 100  ## Filtering the result by retention time: keeping all spectra and ## chromatographic peaks within 3000 and 3500 seconds. xmse_sub <- filterRt(xmse, rt = c(3000, 3500)) #> Filter spectra xmse_sub #> Object of class XcmsExperiment  #>  Spectra: MS1 (960)  #>  Experiment data: 3 sample(s) #>  Sample data links: #>   - spectra: 3 sample(s) to 960 element(s). #>  xcms results: #>   - chromatographic peaks: 79 in MS level(s): 1  nrow(chromPeaks(xmse_sub)) #> [1] 79  ## Perform an initial feature grouping to allow alignment using the ## peak groups method: pdp <- PeakDensityParam(sampleGroups = rep(1, 3)) xmse <- groupChromPeaks(xmse, param = pdp)  ## Perform alignment using the peak groups method. pgp <- PeakGroupsParam(span = 0.4) xmse <- adjustRtime(xmse, param = pgp) #> Performing retention time correction using 19 peak groups.  ## Visualizing the alignment results plotAdjustedRtime(xmse)  ## Performing the final correspondence analysis xmse <- groupChromPeaks(xmse, param = pdp)  ## Show the definition of the first 6 features featureDefinitions(xmse) |> head() #>      mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks 1      peakidx #> FT01 279.0 279.0 279.0 2789.588 2787.430 2791.746      2 2      11, 199 #> FT02 286.2 286.2 286.2 3253.923 3245.811 3262.034      2 2     115, 205 #> FT03 300.2 300.2 300.2 3385.835 3384.068 3390.895      4 3 35, 125,.... #> FT04 301.0 301.0 301.0 2789.066 2787.430 2790.180      3 3  10, 97, 198 #> FT05 305.1 305.1 305.1 2927.922 2922.158 2933.686      2 2      14, 202 #> FT06 305.1 305.1 305.1 3000.543 2991.470 3009.616      2 2      15, 203 #>      ms_level #> FT01        1 #> FT02        1 #> FT03        1 #> FT04        1 #> FT05        1 #> FT06        1  ## Extract the feature values; show the results for the first 6 rows. featureValues(xmse) |> head() #>        ko15.CDF ko16.CDF   ko18.CDF #> FT01 17140627.0       NA 16919266.9 #> FT02         NA  1264119   732016.2 #> FT03  4700903.2  5313736  5169558.2 #> FT04  3051847.8  1964444  2774885.3 #> FT05  1070389.9       NA  1983342.5 #> FT06   847473.1       NA  1003750.6  ## The full results can also be extracted as a `SummarizedExperiment` ## that would eventually simplify subsequent analyses with other packages. ## Any additional parameters passed to the function are passed to the ## `featureValues` function that is called to generate the feature value ## matrix. se <- quantify(xmse, method = \"sum\")  ## EICs for all features can be extracted with the `featureChromatograms` ## function. Note that, depending on the data set, extracting this for ## all features might take some time. Below we extract EICs for the ## first 10 features by providing the feature IDs. chrs <- featureChromatograms(xmse,     features = rownames(featureDefinitions(xmse))[1:10]) chrs #> XChromatograms with 10 rows and 3 columns #>              ko15.CDF        ko16.CDF        ko18.CDF #>       <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]         peaks: 1        peaks: 0        peaks: 1 #> [2,]         peaks: 0        peaks: 1        peaks: 1 #> ...              ...             ...             ...  #> [9,]         peaks: 1        peaks: 2        peaks: 0 #> [10,]        peaks: 1        peaks: 1        peaks: 1 #> phenoData with 3 variables #> featureData with 4 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave  #> Correspondence: #>  method: chromatographic peak density  #>  10 feature(s) identified.  plot(chrs[3, ])"},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine which peaks are absent / present in a sample class — absent-methods","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"Determine peaks absent / present sample class","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"object = \"xcmsSet\" absent(object, ...)     present(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"object xcmsSet-class object class Name sample class sampclass minfrac minimum fraction samples necessary class absent/present","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"Determine peaks absent / present sample class   functions treat peaks present   fillPeaks correctly, .e. count present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"logical vector length nrow(groups(object)).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Alignment: Retention time correction methods. — adjustRtime","title":"Alignment: Retention time correction methods. — adjustRtime","text":"adjustRtime method(s) perform retention time correction (alignment) chromatograms different samples. Alignment performed defaul MS level 1 data. Retention times spectra MS levels, present, subsequently adjusted based adjusted retention times MS1 spectra. Note calling adjustRtime xcms result object remove eventually present previous alignment results well correspondence analysis results. run second round alignment, raw retention times need replaced adjusted ones using applyAdjustedRtime() function. alignment method can specified (configured) using dedicated param argument. Supported param objects : ObiwarpParam: performs retention time adjustment based full m/z - rt data using obiwarp method (Prince (2006)). based original code supports addition alignment multiple samples aligning center sample. alignment performed directly profile-matrix can hence performed independently peak detection peak grouping. PeakGroupsParam: performs retention time correctoin based alignment features defined /samples (corresponding house keeping compounds marker compounds) (Smith 2006). First retention time deviation features described fitting either polynomial (smooth = \"loess\") linear (smooth = \"linear\") function data points. subsequently used adjust retention time spectrum sample (even spectra MS levels different MS 1). Since function based features (.e. chromatographic peaks grouped across samples) initial correspondence analysis performed using groupChromPeaks() function. Alternatively, also possible manually define numeric matrix retention times markers samples used alignment. matrix can passed alignment function using peakGroupsMatrix parameter PeakGroupsParam parameter object. default adjustRtimePeakGroups function used define matrix. function identifies peak groups (features) alignment object based parameters defined param. See also do_adjustRtime_peakGroups() core API function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Alignment: Retention time correction methods. — adjustRtime","text":"","code":"adjustRtime(object, param, ...)  # S4 method for MsExperiment,ObiwarpParam adjustRtime(object, param, chunkSize = 2L, BPPARAM = bpparam())  # S4 method for MsExperiment,PeakGroupsParam adjustRtime(object, param, msLevel = 1L, ...)  PeakGroupsParam(   minFraction = 0.9,   extraPeaks = 1,   smooth = \"loess\",   span = 0.2,   family = \"gaussian\",   peakGroupsMatrix = matrix(nrow = 0, ncol = 0),   subset = integer(),   subsetAdjust = c(\"average\", \"previous\") )  ObiwarpParam(   binSize = 1,   centerSample = integer(),   response = 1L,   distFun = \"cor_opt\",   gapInit = numeric(),   gapExtend = numeric(),   factorDiag = 2,   factorGap = 1,   localAlignment = FALSE,   initPenalty = 0,   subset = integer(),   subsetAdjust = c(\"average\", \"previous\") )  adjustRtimePeakGroups(object, param = PeakGroupsParam(), msLevel = 1L)  # S4 method for OnDiskMSnExp,ObiwarpParam adjustRtime(object, param, msLevel = 1L)  # S4 method for PeakGroupsParam minFraction(object)  # S4 method for PeakGroupsParam minFraction(object) <- value  # S4 method for PeakGroupsParam extraPeaks(object)  # S4 method for PeakGroupsParam extraPeaks(object) <- value  # S4 method for PeakGroupsParam smooth(x)  # S4 method for PeakGroupsParam smooth(object) <- value  # S4 method for PeakGroupsParam span(object)  # S4 method for PeakGroupsParam span(object) <- value  # S4 method for PeakGroupsParam family(object)  # S4 method for PeakGroupsParam family(object) <- value  # S4 method for PeakGroupsParam peakGroupsMatrix(object)  # S4 method for PeakGroupsParam peakGroupsMatrix(object) <- value  # S4 method for PeakGroupsParam subset(x)  # S4 method for PeakGroupsParam subset(object) <- value  # S4 method for PeakGroupsParam subsetAdjust(object)  # S4 method for PeakGroupsParam subsetAdjust(object) <- value  # S4 method for ObiwarpParam binSize(object)  # S4 method for ObiwarpParam binSize(object) <- value  # S4 method for ObiwarpParam centerSample(object)  # S4 method for ObiwarpParam centerSample(object) <- value  # S4 method for ObiwarpParam response(object)  # S4 method for ObiwarpParam response(object) <- value  # S4 method for ObiwarpParam distFun(object)  # S4 method for ObiwarpParam distFun(object) <- value  # S4 method for ObiwarpParam gapInit(object)  # S4 method for ObiwarpParam gapInit(object) <- value  # S4 method for ObiwarpParam gapExtend(object)  # S4 method for ObiwarpParam gapExtend(object) <- value  # S4 method for ObiwarpParam factorDiag(object)  # S4 method for ObiwarpParam factorDiag(object) <- value  # S4 method for ObiwarpParam factorGap(object)  # S4 method for ObiwarpParam factorGap(object) <- value  # S4 method for ObiwarpParam localAlignment(object)  # S4 method for ObiwarpParam localAlignment(object) <- value  # S4 method for ObiwarpParam initPenalty(object)  # S4 method for ObiwarpParam initPenalty(object) <- value  # S4 method for ObiwarpParam subset(x)  # S4 method for ObiwarpParam subset(object) <- value  # S4 method for ObiwarpParam subsetAdjust(object)  # S4 method for ObiwarpParam subsetAdjust(object) <- value  # S4 method for XCMSnExp,PeakGroupsParam adjustRtime(object, param, msLevel = 1L)  # S4 method for XCMSnExp,ObiwarpParam adjustRtime(object, param, msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Alignment: Retention time correction methods. — adjustRtime","text":"object adjustRtime: OnDiskMSnExp(), XCMSnExp(), MsExperiment() XcmsExperiment() object. param parameter object defining alignment method (setting). ... ignored. chunkSize adjustRtime object either MsExperiment XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. Alignment performed parallel (per sample) subset loaded data. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM parallel processing setup. Defaults BPPARAM = bpparam(). See bpparam() details. msLevel adjustRtime: integer(1) defining MS level alignment performed. minFraction PeakGroupsParam: numeric(1) 0 1 defining minimum required proportion samples peaks peak group identified. Peak groups passing criteria aligned across samples retention times individual spectra adjusted based alignment. minFraction = 1 peak group contain peaks samples experiment. Note subset provided, specified fraction relative defined subset samples total number samples within experiment (.e., peak present specified proportion subset samples). extraPeaks PeakGroupsParam: numeric(1) defining maximal number additional peaks samples assigned peak group (feature) retention time correction. data set 6 samples, extraPeaks = 1 uses peak groups total peak count <= 6 + 1. total peak count total number peaks assigned peak group considers also multiple peaks within sample assigned group. smooth PeakGroupsParam: character(1) defining function used interpolate corrected retention times peak groups. Can either \"loess\" \"linear\". span PeakGroupsParam: numeric(1) defining degree smoothing (smooth = \"loess\"). parameter passed internal call loess(). family PeakGroupsParam: character(1) defining method loess smoothing. Allowed values \"gaussian\" \"symmetric\". See loess() information. peakGroupsMatrix PeakGroupsParam: optional matrix (raw) retention times (marker) peak groups alignment performed. column represents sample, row feature/peak group. adjustRtimePeakGroups method used default determine matrix provided object. subset ObiwarpParam PeakGroupsParam: integer indices samples within experiment alignment models estimated. Samples part subset adjusted based closest subset sample. See Subset-based alignment section details. subsetAdjust ObiwarpParam PeakGroupsParam: character(1) specifying method non-subset samples adjusted. Supported options \"previous\" \"average\" (default). See Subset-based alignment section details. binSize numeric(1) defining bin size (mz dimension) used profile matrix generation. See step parameter profile-matrix documentation details. centerSample integer(1) defining index center sample experiment. defaults floor(median(1:length(fileNames(object)))). Note subset used, index passed centerSample within subset samples. response ObiwarpParam: numeric(1) defining responsiveness warping response = 0 giving linear warping start end points response = 100 warping using bijective anchors. distFun ObiwarpParam: character(1) defining distance function used. Allowed values \"cor\" (Pearson's correlation), \"cor_opt\" (calculate 10% diagonal band distance matrix; better runtime), \"cov\" (covariance), \"prd\" (product) \"euc\" (Euclidian distance). default value distFun = \"cor_opt\". gapInit ObiwarpParam: numeric(1) defining penalty gap opening. default value depends value distFun: distFun = \"cor\" distFun = \"cor_opt\" 0.3, distFun = \"cov\" distFun = \"prd\" 0.0 distFun = \"euc\" 0.9. gapExtend ObiwarpParam: numeric(1) defining penalty gap enlargement. default value gapExtend depends value distFun: distFun = \"cor\" distFun = \"cor_opt\" 2.4, distFun = \"cov\" 11.7, distFun = \"euc\" 1.8 distFun = \"prd\" 7.8. factorDiag ObiwarpParam: numeric(1) defining local weight applied diagonal moves alignment. factorGap ObiwarpParam: numeric(1) defining local weight gap moves alignment. localAlignment ObiwarpParam: logical(1) whether local alignment performed instead default global alignment. initPenalty ObiwarpParam: numeric(1) defining penalty initiating alignment (local alignment ). value value slot. x ObiwarpParam PeakGroupsParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Alignment: Retention time correction methods. — adjustRtime","text":"adjustRtime OnDiskMSnExp XCMSnExp object return XCMSnExp object alignment results. adjustRtime MsExperiment XcmsExperiment return XcmsExperiment adjusted retention times stored new spectra variable  rtime_adjusted object's spectra. ObiwarpParam PeakGroupsParam return respective parameter object. adjustRtimeGroups returns matrix retention times marker features sample (row one feature, row one sample).","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"subset-based-alignment","dir":"Reference","previous_headings":"","what":"Subset-based alignment","title":"Alignment: Retention time correction methods. — adjustRtime","text":"alignment methods allow perform retention time correction user-selected subset samples (e.g. QC samples) samples part subset adjusted based adjusted retention times closest subset sample (close terms index within object hence possibly injection index). thus suggested load MS data files order samples injected measurement run(s). non-subset samples adjusted depends also parameter subsetAdjust: subsetAdjust = \"previous\", non-subset sample adjusted based closest previous subset sample results cases adjusted retention times non-subset sample identical subset sample adjustment bases. second, default, option subsetAdjust = \"average\" case non subset sample adjusted based average retention time adjustment previous following subset sample. average, weighted mean used weights inverse distance non-subset sample subset samples used alignment. See also section Alignment experiments including blanks xcms vignette details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Alignment: Retention time correction methods. — adjustRtime","text":"Prince, J. T., Marcotte, E. M. (2006) \"Chromatographic Alignment ESI-LC-MS Proteomic Data Sets Ordered Bijective Interpolated Warping\" Anal. Chem., 78 (17), 6140-6152. Smith, C.., Want, E.J., O'Maille, G., Abagyan, R. Siuzdak, G. (2006). \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Alignment: Retention time correction methods. — adjustRtime","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Replace raw with adjusted retention times — applyAdjustedRtime","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Replaces raw retention times adjusted retention time returns object unchanged none present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"","code":"applyAdjustedRtime(object)"},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"object XCMSnExp XcmsExperiment object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"XCMSnExp XcmsExperiment object raw (original) retention times replaced adjusted retention time.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Adjusted retention times stored parallel adjusted retention times XCMSnExp XcmsExperiment objects. applyAdjustedRtime replaces raw (original) retention times adjusted retention times.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Replacing raw retention times adjusted retention times disables possibility restore raw retention times using dropAdjustedRtime() method. function remove retention time processing step settings alignment processHistory() object ensure processing history preserved.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  xod <- adjustRtime(faahko_sub, param = ObiwarpParam()) #> Sample number 2 used as center sample. #> Aligning ko15.CDF against ko16.CDF ...  #> OK #> Aligning ko18.CDF against ko16.CDF ...  #> OK #> Applying retention time adjustment to the identified chromatographic peaks ...  #> OK  hasAdjustedRtime(xod) #> [1] TRUE  ## Replace raw retention times with adjusted retention times. xod <- applyAdjustedRtime(xod)  ## No adjusted retention times present hasAdjustedRtime(xod) #> [1] FALSE  ## Raw retention times have been replaced with adjusted retention times plot(split(rtime(faahko_sub), fromFile(faahko_sub))[[1]] -     split(rtime(xod), fromFile(xod))[[1]], type = \"l\")   ## And the process history still contains the settings for the alignment processHistory(xod) #> [[1]] #> Object of class \"XProcessHistory\" #>  type: Peak detection  #>  date: Tue Aug  1 15:54:19 2023  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: CentWaveParam  #>  MS level(s) 1  #>  #> [[2]] #> Object of class \"XProcessHistory\" #>  type: Retention time correction  #>  date: Thu Dec  7 08:42:45 2023  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: ObiwarpParam  #>  MS level(s) 1  #>"},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":null,"dir":"Reference","previous_headings":"","what":"Aggregate values in y for bins defined on x — binYonX","title":"Aggregate values in y for bins defined on x — binYonX","text":"functions takes two -sized numeric vectors x     y, bins/cuts x bins (either pre-defined number     equal-sized bins bins pre-defined size) aggregates values     y corresponding x values falling within bin.     default (.e. method = \"max\") maximal y value     corresponding x values identified. x expected     incrementally sorted , , internally sorted (    case also y ordered according order x).","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Aggregate values in y for bins defined on x — binYonX","text":"","code":"binYonX(   x,   y,   breaks,   nBins,   binSize,   binFromX,   binToX,   fromIdx = 1L,   toIdx = length(x),   method = \"max\",   baseValue,   sortedX = !is.unsorted(x),   shiftByHalfBinSize = FALSE,   returnIndex = FALSE,   returnX = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Aggregate values in y for bins defined on x — binYonX","text":"x Numeric vector used binning. y Numeric vector (length x) maximum values bin defined. provided, x used. breaks Numeric vector defining breaks bins, .e. lower upper values bin. See examples . nBins integer(1) defining number desired bins. binSize numeric(1) defining desired bin size. binFromX Optional numeric(1) allowing manually specify range x-values used binning. affect calculation breaks bins (.e. nBins binSize provided). provided minimal value sub-set fromIdx-toIdx input vector x used. binToX binFromX, defining maximum x-value used binning. fromIdx Integer vector defining start position one multiple sub-sets input vector x used binning. toIdx toIdx, defining maximum index (indices) x used binning. method character string specifying method used aggregate values y. Allowed \"max\", \"min\", \"sum\" \"mean\" identify maximal minimal value sum values within bin calculate mean value. baseValue base value empty bins (.e. bins either values x fall, NA values y assigned). default (.e. specified), NA assigned bins. sortedX Whether x sorted. shiftByHalfBinSize Logical specifying whether bins shifted half bin size left. Thus, first bin center fromX lower upper boundary fromX - binSize/2 fromX + binSize/2. argument ignored breaks provided. returnIndex Logical indicating whether index max (method = \"max\") min (method = \"min\") value within bin input vector x also reported. methods \"max\" \"min\" argument ignored. returnX logical allowing avoid returning $x, .e. mid-points bins. returnX = FALSE might useful cases breaks pre-defined considerably reduces memory demand.","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Aggregate values in y for bins defined on x — binYonX","text":"Returns list length 2, first element (named \"x\")     contains bin mid-points, second element (named \"y\")     aggregated values input vector y within bin.  returnIndex = TRUE list contains additional element  \"index\" index max min (depending whether  method = \"max\" method = \"min\") value within bin     input vector x.","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Aggregate values in y for bins defined on x — binYonX","text":"breaks defining boundary bin can either passed     directly function argument breaks,     calculated data based arguments nBins binSize     along fromIdx, toIdx optionally binFromX     binToX.     Arguments fromIdx toIdx allow specify subset(s)     input vector x bins calculated.     default full x vector considered. Also, specified     otherwise arguments binFromX binToX , range     bins within sub-sets x[fromIdx]     x[toIdx]. Arguments binFromX binToX allow     overwrite manually defining range breaks     calculated. See examples details. Calculation breaks: nBins breaks correspond     seq(min(x[fromIdx])), max(x[fromIdx], length.= (nBins + 1)).     binSize breaks correspond     seq(min(x[fromIdx]), max(x[toIdx]), = binSize)     exception last break value forced equal     max(x[toIdx]). ensures values specified     range covered breaks defining bins. last bin     however instances slightly larger binSize. See     breaks_on_binSize breaks_on_nBins     details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Aggregate values in y for bins defined on x — binYonX","text":"function ensures values within range used define     breaks considered binning (assigned bin).     means bins except last one values x     >= xlower < xupper (xlower     xupper lower upper boundary, respectively).     last bin condition x >= xlower & x <= xupper.     Note also shiftByHalfBinSize TRUE range     values used binning expanded binSize (.e.     lower boundary fromX - binSize/2, upper     toX + binSize/2). Setting argument TRUE resembles     binning /used profBin function     xcms < 1.51. NA handling: default function ignores NA values     y (thus inherently assumes na.rm = TRUE). NA     values allowed x.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Aggregate values in y for bins defined on x — binYonX","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Aggregate values in y for bins defined on x — binYonX","text":"","code":"######## ## Simple example illustrating the breaks and the binning. ## ## Define breaks for 5 bins: brks <- seq(2, 12, length.out = 6) ## The first bin is then [2,4), the second [4,6) and so on. brks #> [1]  2  4  6  8 10 12 ## Get the max value falling within each bin. binYonX(x = 1:16, y = 1:16, breaks = brks) #> $x #> [1]  3  5  7  9 11 #>  #> $y #> [1]  3  5  7  9 12 #>  ## Thus, the largest value in x = 1:16 falling into the bin [2,4) (i.e. being ## >= 2 and < 4) is 3, the largest one falling into [4,6) is 5 and so on. ## Note however the function ensures that the minimal and maximal x-value ## (in this example 1 and 12) fall within a bin, i.e. 12 is considered for ## the last bin.  ####### ## Performing the binning ons sub-set of x ## X <- 1:16 ## Bin X from element 4 to 10 into 5 bins. X[4:10] #> [1]  4  5  6  7  8  9 10 binYonX(X, X, nBins = 5L, fromIdx = 4, toIdx = 10) #> $x #> [1] 4.6 5.8 7.0 8.2 9.4 #>  #> $y #> [1]  5  6  7  8 10 #>  ## This defines breaks for 5 bins on the values from 4 to 10 and bins ## the values into these 5 bins. Alternatively, we could manually specify ## the range for the binning, i.e. the minimal and maximal value for the ## breaks: binYonX(X, X, nBins = 5L, fromIdx = 4, toIdx = 10, binFromX = 1, binToX = 16) #> $x #> [1]  2.5  5.5  8.5 11.5 14.5 #>  #> $y #> [1] NA  6  9 10 NA #>  ## In this case the breaks for 5 bins were defined from a value 1 to 16 and ## the values 4 to 10 were binned based on these breaks.  ####### ## Bin values within a sub-set of x, second example ## ## This example illustrates how the fromIdx and toIdx parameters can be used. ## x defines 3 times the sequence form 1 to 10, while y is the sequence from ## 1 to 30. In this very simple example x is supposed to represent M/Z values ## from 3 consecutive scans and y the intensities measured for each M/Z in ## each scan. We want to get the maximum intensities for M/Z value bins only ## for the second scan, and thus we use fromIdx = 11 and toIdx = 20. The breaks ## for the bins are defined with the nBins, binFromX and binToX. X <- rep(1:10, 3) Y <- 1:30 ## Bin the M/Z values in the second scan into 5 bins and get the maximum ## intensity for each bin. Note that we have to specify sortedX = TRUE as ## the x and y vectors would be sorted otherwise. binYonX(X, Y, nBins = 5L, sortedX = TRUE, fromIdx = 11, toIdx = 20) #> $x #> [1] 1.9 3.7 5.5 7.3 9.1 #>  #> $y #> [1] 12 14 16 18 20 #>   ####### ## Bin in overlapping sub-sets of X ## ## In this example we define overlapping sub-sets of X and perform the binning ## within these. X <- 1:30 ## Define the start and end indices of the sub-sets. fIdx <- c(2, 8, 21) tIdx <- c(10, 25, 30) binYonX(X, nBins = 5L, fromIdx = fIdx, toIdx = tIdx) #> [[1]] #> [[1]]$x #> [1] 2.8 4.4 6.0 7.6 9.2 #>  #> [[1]]$y #> [1]  3  5  6  8 10 #>  #>  #> [[2]] #> [[2]]$x #> [1]  9.7 13.1 16.5 19.9 23.3 #>  #> [[2]]$y #> [1] 11 14 18 21 25 #>  #>  #> [[3]] #> [[3]]$x #> [1] 21.9 23.7 25.5 27.3 29.1 #>  #> [[3]]$y #> [1] 22 24 26 28 30 #>  #>  ## The same, but pre-defining also the desired range of the bins. binYonX(X, nBins = 5L, fromIdx = fIdx, toIdx = tIdx, binFromX = 4, binToX = 28) #> [[1]] #> [[1]]$x #> [1]  6.4 11.2 16.0 20.8 25.6 #>  #> [[1]]$y #> [1]  8 10 NA NA NA #>  #>  #> [[2]] #> [[2]]$x #> [1]  6.4 11.2 16.0 20.8 25.6 #>  #> [[2]]$y #> [1]  8 13 18 23 25 #>  #>  #> [[3]] #> [[3]]$x #> [1]  6.4 11.2 16.0 20.8 25.6 #>  #> [[3]]$y #> [1] NA NA NA 23 28 #>  #>  ## The same bins are thus used for each sub-set."},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"Defines breaks binSize sized bins values ranging fromX toX.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"","code":"breaks_on_binSize(fromX, toX, binSize)"},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"fromX numeric(1) specifying lowest value bins. toX numeric(1) specifying largest value bins. binSize numeric(1) defining size bin.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"numeric vector defining lower upper bounds bins.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"function creates breaks bins size binSize. function ensures full data range included bins, .e. last value (upper boundary last bin) always equal toX. however means size last bin always equal desired bin size. See examples details comparisom R's seq function.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"","code":"## Define breaks with a size of 0.13 for a data range from 1 to 10: breaks_on_binSize(1, 10, 0.13) #>  [1]  1.00  1.13  1.26  1.39  1.52  1.65  1.78  1.91  2.04  2.17  2.30  2.43 #> [13]  2.56  2.69  2.82  2.95  3.08  3.21  3.34  3.47  3.60  3.73  3.86  3.99 #> [25]  4.12  4.25  4.38  4.51  4.64  4.77  4.90  5.03  5.16  5.29  5.42  5.55 #> [37]  5.68  5.81  5.94  6.07  6.20  6.33  6.46  6.59  6.72  6.85  6.98  7.11 #> [49]  7.24  7.37  7.50  7.63  7.76  7.89  8.02  8.15  8.28  8.41  8.54  8.67 #> [61]  8.80  8.93  9.06  9.19  9.32  9.45  9.58  9.71  9.84 10.00 ## The size of the last bin is however larger than 0.13: diff(breaks_on_binSize(1, 10, 0.13)) #>  [1] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [16] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [31] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [46] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [61] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.16 ## If we would use seq, the max value would not be included: seq(1, 10, by = 0.13) #>  [1] 1.00 1.13 1.26 1.39 1.52 1.65 1.78 1.91 2.04 2.17 2.30 2.43 2.56 2.69 2.82 #> [16] 2.95 3.08 3.21 3.34 3.47 3.60 3.73 3.86 3.99 4.12 4.25 4.38 4.51 4.64 4.77 #> [31] 4.90 5.03 5.16 5.29 5.42 5.55 5.68 5.81 5.94 6.07 6.20 6.33 6.46 6.59 6.72 #> [46] 6.85 6.98 7.11 7.24 7.37 7.50 7.63 7.76 7.89 8.02 8.15 8.28 8.41 8.54 8.67 #> [61] 8.80 8.93 9.06 9.19 9.32 9.45 9.58 9.71 9.84 9.97  ## In the next example we use binSize that leads to an additional last bin with ## a smaller binSize: breaks_on_binSize(1, 10, 0.51) #>  [1]  1.00  1.51  2.02  2.53  3.04  3.55  4.06  4.57  5.08  5.59  6.10  6.61 #> [13]  7.12  7.63  8.14  8.65  9.16  9.67 10.00 ## Again, the max value is included, but the size of the last bin is < 0.51. diff(breaks_on_binSize(1, 10, 0.51)) #>  [1] 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 #> [16] 0.51 0.51 0.33 ## Using just seq would result in the following bin definition: seq(1, 10, by = 0.51) #>  [1] 1.00 1.51 2.02 2.53 3.04 3.55 4.06 4.57 5.08 5.59 6.10 6.61 7.12 7.63 8.14 #> [16] 8.65 9.16 9.67 ## Thus it defines one bin (break) less."},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate breaks for binning — breaks_on_nBins","title":"Generate breaks for binning — breaks_on_nBins","text":"Calculate breaks -sized bins data values fromX toX.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate breaks for binning — breaks_on_nBins","text":"","code":"breaks_on_nBins(fromX, toX, nBins, shiftByHalfBinSize = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate breaks for binning — breaks_on_nBins","text":"fromX numeric(1) specifying lowest value bins. toX numeric(1) specifying largest value bins. nBins numeric(1) defining number bins. shiftByHalfBinSize Logical indicating whether bins shifted left half bin size. results centered bins, .e. first bin centered fromX last around toX.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate breaks for binning — breaks_on_nBins","text":"numeric vector length nBins + 1 defining lower upper bounds bins.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generate breaks for binning — breaks_on_nBins","text":"generates bins call seq(fromX, toX, length.= nBins) . first second element result vector thus defines lower upper boundary first bin, second third value second bin .","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate breaks for binning — breaks_on_nBins","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generate breaks for binning — breaks_on_nBins","text":"","code":"## Create breaks to bin values from 3 to 20 into 20 bins breaks_on_nBins(3, 20, nBins = 20) #>  [1]  3.00  3.85  4.70  5.55  6.40  7.25  8.10  8.95  9.80 10.65 11.50 12.35 #> [13] 13.20 14.05 14.90 15.75 16.60 17.45 18.30 19.15 20.00 ## The same call but using shiftByHalfBinSize breaks_on_nBins(3, 20, nBins = 20, shiftByHalfBinSize = TRUE) #>  [1]  2.552632  3.447368  4.342105  5.236842  6.131579  7.026316  7.921053 #>  [8]  8.815789  9.710526 10.605263 11.500000 12.394737 13.289474 14.184211 #> [15] 15.078947 15.973684 16.868421 17.763158 18.657895 19.552632 20.447368"},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"Combine xcmsSet objects — c-methods","title":"Combine xcmsSet objects — c-methods","text":"Combines samples peaks multiple xcmsSet objects   single object. Group retention time correction data   discarded. profinfo list set equal   first object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Combine xcmsSet objects — c-methods","text":"xs1 = \"xcmsRaw\" c(xs1, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Combine xcmsSet objects — c-methods","text":"xs1 xcmsSet object ... xcmsSet objects","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Combine xcmsSet objects — c-methods","text":"xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Combine xcmsSet objects — c-methods","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":null,"dir":"Reference","previous_headings":"","what":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"Calibrate peaks using mz values known masses/calibrants. mz values identified peaks adjusted based peaks close provided mz values. See details information. isCalibrated function returns TRUE chromatographic peaks XCMSnExp object x calibrated FALSE otherwise.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"","code":"CalibrantMassParam(   mz = list(),   mzabs = 1e-04,   mzppm = 5,   neighbors = 3,   method = \"linear\" )  isCalibrated(object)  # S4 method for XCMSnExp calibrate(object, param)"},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"mz numeric list numeric vectors reference mz values. numeric vector provided, used sample XCMSnExp object. list provided, length equal number samples experiment. mzabs numeric(1) absolute error/deviation matching peaks calibrants (Da). mzppm numeric(1) relative error matching peaks calibrants ppm (parts per million). neighbors integer(1) maximal number peaks within permitted distance calibrants considered. Among mz value peak largest intensity used calibration function estimation. method character(1) defining method used estimate calibration function. Can \"shift\", \"linear\" (default) \"edgeshift\". object XCMSnExp object. param CalibrantMassParam object calibration settings.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"CalibrantMassParam: CalibrantMassParam instance. calibrate: XCMSnExp object chromatographic peaks calibrated. aware actual raw mz values (yet) calibrated, identified chromatographic peaks. CalibrantMassParam function returns instance CalibrantMassParam class settings properties set. calibrate method returns XCMSnExp object chromatographic peaks calibrated. Note detected peaks calibrated, individual mz values spectrum.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"method first identify peaks close provided mz values , given difference calibrants smaller user provided cut (based arguments mzabs mzppm), mz values replaced provided mz values. mz values peaks either globally shifted (method = \"shift\" estimated linear model calibrants. Peaks considered close calibrant mz difference calibrant mz <= mzabs + mz * mzppm /1e6. Adjustment methods: adjustment function/factor estimated using difference calibrant peak mz values peaks close enough calibrants. availabel methods : shift: shifts m/z peak global factor corresponds average difference peak mz calibrant mz. linear: fits linear model throught differences calibrant peak mz values adjusts mz values peaks using . edgeshift: performs adjustment linear peaks within mz range calibrants shift outside . information, details examples refer xcms-direct-injection vignette.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"CalibrantMassParam classes exported getter setter methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"Joachim Bargsten, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":null,"dir":"Reference","previous_headings":"","what":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"Calibrate peaks xcmsSet via set known masses","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"object = \"xcmsSet\" calibrate(object, calibrants,method=\"linear\",     mzabs=0.0001, mzppm=5,     neighbours=3, plotres=FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"object xcmsSet object uncalibrated mz calibrants vector list vectors reference m/z-values method used calibrating-method, see mzppm relative error used matching peaks ppm (parts per million) mzabs absolute error used matching peaks Da neighbours number neighbours wich one highest intensity used (instead nearest) plotres can set TRUE wanted result-plot showing found m/z distances regression","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"object xcmsSet one ore samples calibrants sample different calibrants can used, list m/z-vectors given. length list must number samples, alternatively single vector masses can given used samples. method \"shift\" shifting m/z, \"linear\" linear regression adds linear term m/z. \"edgeshift\" linear regression within range mz-calibrants shift outside.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"Extract ion chromatogram (EIC) chromatographic peak XcmsExperiment() object. result returned XChromatograms() length equal number chromatographic peaks (one column).","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"","code":"chromPeakChromatograms(object, ...)  # S4 method for XcmsExperiment chromPeakChromatograms(   object,   expandRt = 0,   expandMz = 0,   aggregationFun = \"max\",   peaks = character(),   return.type = c(\"XChromatograms\", \"MChromatograms\"),   ...,   progressbar = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"object XcmsExperiment() identified chromatographic peaks. ... currently ignored. expandRt numeric(1) eventually expand retention time range signal integrated. chromatogram contain signal chromPeaks[, \"rtmin\"] - expandRt chromPeaks[, \"rtmax\"] + expandRt. default expandRt = 0. expandMz numeric(1) eventually expand m/z range signal integrated. chromatogram contain signal chromPeaks[, \"mzmin\"] - expandMz chromPeaks[, \"mzmax\"] + expandMz. default expandMz = 0. aggregationFun character(1) defining function signals within m/z range spectrum (.e. discrete retention time) aggregated. default (aggregationFun = \"max\") reports largest signal spectrum. peaks optional character providing IDs chromatographic peaks (.e. row names peaks chromPeaks(object)) chromatograms returned. return.type character(1) specifying type returned object. Can either return.type = \"XChromatograms\" (default) return.type = \"MChromatograms\" return either chromatographic object without identified chromatographic peaks, respectively. progressbar logical(1) whether progress extraction process displayed.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Get EICs for every detected chromatographic peak chrs <- chromPeakChromatograms(faahko_sub) chrs #> XChromatograms with 248 rows and 1 column #>                      1 #>        <XChromatogram> #> [1,]          peaks: 1 #> [2,]          peaks: 1 #> ...               ...  #> [247,]        peaks: 1 #> [248,]        peaks: 1 #> phenoData with 2 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Order of EICs matches the order in chromPeaks chromPeaks(faahko_sub) |> head() #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1  ## variable \"sample_index\" provides the index of the sample the EIC was ## extracted from fData(chrs)$sample_index #>   [1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 #>  [38] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 #>  [75] 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 #> [112] 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 #> [149] 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 #> [186] 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 #> [223] 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3  ## Get the EIC for selected peaks only. pks <- rownames(chromPeaks(faahko_sub))[c(6, 12)] pks #> [1] \"CP006\" \"CP012\"  ## Expand the data on retention time dimension by 15 seconds (on each side) res <- chromPeakChromatograms(faahko_sub, peaks = pks, expandRt = 5) plot(res[1, ])"},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"Extract (MS1 MS2) spectra XcmsExperiment XCMSnExp object identified chromatographic peaks. return spectra selected chromatographic peaks, peak ID (.e., row name chromPeaks matrix) can provided parameter peaks. msLevel = 1L (supported return.type = \"Spectra\" return.type = \"List\") MS1 spectra within retention time boundaries (file peak detected) returned. msLevel = 2L MS2 spectra returned chromatographic peak precursor m/z within retention time m/z range chromatographic peak. Parameter method allows define whether single spectrum returned: method = \"\": (default): return spectra chromatographic peak. method = \"closest_rt\": return spectrum retention time closest peak's retention time (apex). method = \"closest_mz\": return spectrum precursor m/z closest peaks's m/z (apex); supported msLevel > 1. method = \"largest_tic\": return spectrum largest total signal (sum peaks intensities). method = \"largest_bpi\": return spectrum largest peak intensity (maximal peak intensity). method = \"signal\": object XCMSnExp: return spectrum sum intensities similar peak's apex signal (\"maxo\"); supported msLevel = 2L. Parameter return.type allows specify type result object. return.type = \"Spectra\" (default) Spectra object matching spectra returned. spectra variable \"peak_id\" returned Spectra contains ID chromatographic peak (.e., rowname peak chromPeaks matrix) spectrum. return.type = \"Spectra\" List Spectra returned. length list equal number rows chromPeaks. element list contains thus Spectra spectra one chromatographic peak (Spectra length 0 spectrum found respective chromatographic peak). See also LC-MS/MS data analysis vignette details examples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"","code":"chromPeakSpectra(object, ...)  # S4 method for XcmsExperiment chromPeakSpectra(   object,   method = c(\"all\", \"closest_rt\", \"closest_mz\", \"largest_tic\", \"largest_bpi\"),   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   skipFilled = FALSE,   peaks = character(),   return.type = c(\"Spectra\", \"List\"),   BPPARAM = bpparam() )  # S4 method for XCMSnExp chromPeakSpectra(   object,   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   method = c(\"all\", \"closest_rt\", \"closest_mz\", \"signal\", \"largest_tic\", \"largest_bpi\"),   skipFilled = FALSE,   return.type = c(\"Spectra\", \"MSpectra\", \"List\", \"list\"),   peaks = character() )"},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"object XcmsExperiment XCMSnExp object identified chromatographic peaks spectra returned. ... ignored. method character(1) specifying spectra include result. Defaults method = \"\". See function description details. msLevel integer(1) defining MS level spectra returned. expandRt numeric(1) expand retention time range peak constant value side. expandMz numeric(1) expand m/z range peak constant value side. ppm numeric(1) expand m/z range peak (side) value dependent peak's m/z. skipFilled logical(1) whether spectra filled-peaks reported . peaks character, logical integer allowing specify subset chromatographic peaks chromPeaks spectra returned (providing either ID, logical vector length nrow(chromPeaks(x)) index chromPeaks(x)). parameter overrides skipFilled. return.type character(1) defining type result object returned. BPPARAM parallel processing setup. Defaults bpparam().","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"parameter return.type allow specify type returned object: return.type = \"Spectra\" (default): Spectra object (defined Spectra package). result contains spectra peaks. Metadata column \"peak_id\" provides ID respective peak (.e. rowname chromPeaks(). return.type = \"List\": List length equal number chromatographic peaks returned, element Spectra spectra one chromatographic peak. backward compatibility options \"MSpectra\" \"list\" also supported suggested. return.type = \"MSpectra\" (deprecated): MSpectra object elements Spectrum objects. result objects contains spectra peaks. Metadata column \"peak_id\" provides ID respective peak (.e. rowname chromPeaks()). return.type = \"list\": list lists either length 0 contain Spectrum2 object(s) within m/z-rt range. length list matches number peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"","code":"## Read a file with DDA LC-MS/MS data library(MsExperiment) fl <- system.file(\"TripleTOF-SWATH/PestMix1_DDA.mzML\", package = \"msdata\")  dda <- readMsExperiment(fl)  ## Perform MS1 peak detection dda <- findChromPeaks(dda, CentWaveParam(peakwidth = c(5, 15),     prefilter = c(5, 1000)))  ## Return all MS2 spectro for each chromatographic peaks as a Spectra object ms2_sps <- chromPeakSpectra(dda) ms2_sps #> MSn data (Spectra) with 93 spectra in a MsBackendMzR backend: #>       msLevel     rtime scanIndex #>     <integer> <numeric> <integer> #> 1           2   237.460      1810 #> 2           2   237.869      1812 #> 3           2   240.759      1842 #> 4           2   241.299      1846 #> 5           2   318.444      2391 #> ...       ...       ...       ... #> 89          2   552.147      4948 #> 90          2   570.506      5077 #> 91          2   574.435      5108 #> 92          2   574.725      5110 #> 93          2   575.255      5115 #>  ... 34 more variables/columns. #>  #> file(s): #> PestMix1_DDA.mzML #> Processing: #>  Filter: select MS level(s) 2 [Thu Dec  7 08:42:51 2023] #>  Merge 1 Spectra into one [Thu Dec  7 08:42:51 2023]   ## spectra variable *peak_id* contain the row names of the peaks in the ## chromPeak matrix and allow thus to map chromatographic peaks to the ## returned MS2 spectra ms2_sps$peak_id #>  [1] \"CP01\" \"CP01\" \"CP01\" \"CP01\" \"CP02\" \"CP03\" \"CP03\" \"CP03\" \"CP03\" \"CP04\" #> [11] \"CP04\" \"CP04\" \"CP04\" \"CP06\" \"CP06\" \"CP06\" \"CP06\" \"CP06\" \"CP06\" \"CP07\" #> [21] \"CP07\" \"CP07\" \"CP07\" \"CP07\" \"CP07\" \"CP07\" \"CP08\" \"CP08\" \"CP09\" \"CP09\" #> [31] \"CP10\" \"CP10\" \"CP11\" \"CP11\" \"CP11\" \"CP11\" \"CP11\" \"CP12\" \"CP12\" \"CP12\" #> [41] \"CP12\" \"CP12\" \"CP12\" \"CP13\" \"CP13\" \"CP13\" \"CP13\" \"CP13\" \"CP14\" \"CP14\" #> [51] \"CP14\" \"CP15\" \"CP15\" \"CP15\" \"CP15\" \"CP15\" \"CP15\" \"CP17\" \"CP17\" \"CP17\" #> [61] \"CP17\" \"CP17\" \"CP18\" \"CP18\" \"CP18\" \"CP18\" \"CP18\" \"CP19\" \"CP19\" \"CP19\" #> [71] \"CP19\" \"CP19\" \"CP19\" \"CP20\" \"CP20\" \"CP20\" \"CP21\" \"CP21\" \"CP21\" \"CP22\" #> [81] \"CP22\" \"CP22\" \"CP23\" \"CP23\" \"CP23\" \"CP23\" \"CP25\" \"CP25\" \"CP25\" \"CP27\" #> [91] \"CP27\" \"CP27\" \"CP27\" chromPeaks(dda) #>            mz    mzmin    mzmax      rt   rtmin   rtmax        into        intb #> CP01 219.0961 219.0933 219.0996 240.897 233.595 247.874 15236.49207 15222.54500 #> CP02 219.0947 219.0926 219.0963 319.236 311.979 324.874    76.97947    63.82131 #> CP03 227.9913 227.9902 227.9921 358.349 351.649 367.218  5511.96730  5496.32244 #> CP04 300.0312 300.0289 300.0321 366.568 360.029 375.738 12659.45398 12643.63044 #> CP05 302.0284 302.0273 302.0295 366.568 360.449 375.738  7953.70018  7938.19680 #> CP06 256.1102 256.1095 256.1109 388.167 379.097 397.326  7879.79234  7861.19876 #> CP07 224.0836 224.0831 224.0843 388.587 379.517 397.446  8904.69865  8886.41107 #> CP08 309.9980 309.9962 309.9987 405.715 399.527 415.525  5381.59741  5365.38152 #> CP09 378.0356 378.0341 378.0364 405.715 399.527 415.525  5117.90702  5101.66617 #> CP10 298.2751 298.2739 298.2781 345.711 339.830 354.979 39574.83769 39559.51278 #> CP11 308.0012 307.9983 308.0018 405.715 399.527 415.525  5630.65430  5614.25375 #> CP12 376.0385 376.0359 376.0391 405.715 399.527 415.525  5198.61440  5182.23188 #> CP13 292.1216 292.1204 292.1223 410.555 403.625 420.035 15783.11419 15766.32540 #> CP14 289.1217 289.1202 289.1229 425.434 415.525 433.478 10049.05516 10031.22736 #> CP15 304.1141 304.1119 304.1192 424.614 417.985 434.313 14704.14891 14687.51150 #> CP16 384.9700 384.9684 384.9706 440.533 433.893 449.312  8075.24022  8059.47631 #> CP17 382.9727 382.9715 382.9739 440.533 433.478 449.312  8307.76833  8291.81980 #> CP18 231.0144 231.0134 231.0151 450.992 444.342 460.691  5612.28545  5595.90874 #> CP19 306.1659 306.1624 306.1694 463.786 453.922 471.431 31273.95023 31256.40745 #> CP20 205.0970 205.0956 205.0976 459.421 453.092 466.461  5609.65542  5596.09130 #> CP21 230.9875 230.9862 230.9879 495.109 486.210 502.807  8036.59181  8019.64904 #> CP22 216.1414 216.1398 216.1418 508.678 502.680 516.408  4406.91863  4392.98419 #> CP23 330.2065 330.2008 330.2078 422.514 416.754 431.858   614.91821   599.73617 #> CP24 330.2059 330.2036 330.2081 474.408 460.261 479.107   248.31997   229.20853 #> CP25 330.2067 330.2054 330.2077 552.815 541.355 561.837  5890.30116  5869.52232 #> CP26 330.2057 330.2038 330.2076 505.738 495.519 515.868   315.87230   295.17246 #> CP27 373.0416 373.0406 373.0430 576.468 569.174 580.636  6184.87790  6173.40597 #>            maxo   sn sample #> CP01 4877.11621 4876      1 #> CP02   19.35541   18      1 #> CP03 1459.38428 1458      1 #> CP04 3156.89258 3156      1 #> CP05 1974.66846 1974      1 #> CP06 1836.12695 1835      1 #> CP07 2268.62915 2268      1 #> CP08 1345.32336 1344      1 #> CP09 1300.64075 1300      1 #> CP10 8060.13037 8059      1 #> CP11 1450.15808 1449      1 #> CP12 1350.63330 1350      1 #> CP13 3203.84229 3203      1 #> CP14 2972.63916 2972      1 #> CP15 5009.70654 5009      1 #> CP16 2224.37622 2223      1 #> CP17 2361.33594 2360      1 #> CP18 1671.33203 1670      1 #> CP19 6300.43652 6299      1 #> CP20 1717.72607 1717      1 #> CP21 2244.02979 2243      1 #> CP22 1576.85681 1576      1 #> CP23  135.81871  135      1 #> CP24   19.64630   19      1 #> CP25 2102.92871 2102      1 #> CP26   27.88095   27      1 #> CP27 2511.13501 2510      1  ## Alternatively, return the result as a List of Spectra objects. This list ## is parallel to chromPeaks hence the mapping between chromatographic peaks ## and MS2 spectra is easier. ms2_sps <- chromPeakSpectra(dda, return.type = \"List\") names(ms2_sps) #>  [1] \"CP01\" \"CP02\" \"CP03\" \"CP04\" \"CP05\" \"CP06\" \"CP07\" \"CP08\" \"CP09\" \"CP10\" #> [11] \"CP11\" \"CP12\" \"CP13\" \"CP14\" \"CP15\" \"CP16\" \"CP17\" \"CP18\" \"CP19\" \"CP20\" #> [21] \"CP21\" \"CP22\" \"CP23\" \"CP24\" \"CP25\" \"CP26\" \"CP27\" rownames(chromPeaks(dda)) #>  [1] \"CP01\" \"CP02\" \"CP03\" \"CP04\" \"CP05\" \"CP06\" \"CP07\" \"CP08\" \"CP09\" \"CP10\" #> [11] \"CP11\" \"CP12\" \"CP13\" \"CP14\" \"CP15\" \"CP16\" \"CP17\" \"CP18\" \"CP19\" \"CP20\" #> [21] \"CP21\" \"CP22\" \"CP23\" \"CP24\" \"CP25\" \"CP26\" \"CP27\" ms2_sps[[1L]] #> MSn data (Spectra) with 4 spectra in a MsBackendMzR backend: #>     msLevel     rtime scanIndex #>   <integer> <numeric> <integer> #> 1         2   237.460      1810 #> 2         2   237.869      1812 #> 3         2   240.759      1842 #> 4         2   241.299      1846 #>  ... 34 more variables/columns. #>  #> file(s): #> PestMix1_DDA.mzML #> Processing: #>  Filter: select MS level(s) 2 [Thu Dec  7 08:42:51 2023] #>  Merge 1 Spectra into one [Thu Dec  7 08:42:51 2023]   ## Parameter `msLevel` allows to define from which MS level spectra should ## be returned. By default `msLevel = 2L` but with `msLevel = 1L` all ## MS1 spectra with a retention time within the retention time range of ## a chromatographic peak can be returned. Alternatively, selected ## spectra can be returned by specifying the selection criteria/method ## with the `method` parameter. Below we extract for each chromatographic ## peak the MS1 spectra with a retention time closest to the ## chromatographic peak's apex position. Alternatively it would also be ## possible to select the spectrum with the highest total signal or ## highest (maximal) intensity. ms1_sps <- chromPeakSpectra(dda, msLevel = 1L, method = \"closest_rt\") ms1_sps #> MSn data (Spectra) with 27 spectra in a MsBackendMzR backend: #>       msLevel     rtime scanIndex #>     <integer> <numeric> <integer> #> 1           1   240.897      1843 #> 2           1   319.236      2396 #> 3           1   358.349      2813 #> 4           1   366.568      2924 #> 5           1   366.568      2924 #> ...       ...       ...       ... #> 23          1   422.514      3571 #> 24          1   474.408      4266 #> 25          1   552.815      4958 #> 26          1   505.738      4549 #> 27          1   576.468      5122 #>  ... 34 more variables/columns. #>  #> file(s): #> PestMix1_DDA.mzML #> Processing: #>  Filter: select MS level(s) 1 [Thu Dec  7 08:42:51 2023] #>  Merge 1 Spectra into one [Thu Dec  7 08:42:51 2023]   ## Parameter peaks would allow to extract spectra for specific peaks only. ## Peaks can be defined with parameter `peaks` which can be either an ## `integer` with the index of the peak in the `chromPeaks` matrix or a ## `character` with its rowname in `chromPeaks`. chromPeakSpectra(dda, msLevel = 1L, method = \"closest_rt\", peaks = c(3, 5))"},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Extracting chromatograms — chromatogram,XCMSnExp-method","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"chromatogram: extract chromatographic data (extracted ion chromatogram, base peak chromatogram total ion chromatogram) OnDiskMSnExp XCMSnExp objects. See also help page chromatogram function MSnbase package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp chromatogram(   object,   rt,   mz,   aggregationFun = \"sum\",   missing = NA_real_,   msLevel = 1L,   BPPARAM = bpparam(),   adjustedRtime = hasAdjustedRtime(object),   filled = FALSE,   include = c(\"apex_within\", \"any\", \"none\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"object Either OnDiskMSnExp XCMSnExp object chromatograms extracted. rt numeric(2) two-column matrix defining lower upper boundary retention time range(s). specified, full retention time range original data used. mz numeric(2) two-column matrix defining lower upper mz value MS data slice(s). specified, chromatograms calculated full mz range. aggregationFun character(1) specifying function used aggregate intensity values across mz value range retention time. Allowed values \"sum\" (default), \"max\", \"mean\" \"min\". missing numeric(1) allowing specify intensity value used given retention time signal measured within mz range corresponding scan. Defaults NA_real_ (see also Details Notes sections ). Use missing = 0 resemble behaviour getEIC old user interface. msLevel integer(1) specifying MS level chromatogram extracted. Defaults msLevel = 1L. BPPARAM Parallelisation backend used, depend architecture. Default BiocParallel::bparam(). adjustedRtime chromatogram,XCMSnExp: whether adjusted (adjustedRtime = TRUE) raw retention times (adjustedRtime = FALSE) used filtering returned resulting MChromatograms object. Adjusted retention times used default available. filled logical(1) whether filled-peaks also returned. Defaults filled = FALSE, .e. returns detected chromatographic peaks result object. include character(1) defining chromatographic peaks returned. Supported include = \"apex_within\" (default) returns chromatographic peaks apex within mz rt range, include = \"\" return chromatographic peaks m/z rt ranges overlap mz rt include = \"none\" include chromatographic peaks. ... optional parameters - currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"chromatogram returns XChromatograms object number columns corresponding number files object number rows number specified ranges (.e. number rows matrices provided arguments mz /rt). chromatographic peaks apex position within m/z retention time range also retained well feature definitions peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"Arguments rt mz allow specify MS data slice (.e. m/z range retention time window) chromatogram extracted. parameters can either numeric length 2 lower upper limit, matrix two columns lower upper limits extract multiple EICs . parameter aggregationSum allows specify function used aggregate intensities across m/z range retention time. Setting aggregationFun = \"sum\" e.g. allow calculate total ion chromatogram (TIC), aggregationFun = \"max\" base peak chromatogram (BPC). given retention time intensity measured spectrum NA intensity value returned default. can changed parameter missing, setting missing = 0 result 0 intensity returned cases.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"XCMSnExp objects, adjusted retention times available, chromatogram method default report use (subsetting based provided parameter rt). can changed setting adjustedRtime = FALSE.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"","code":"## Load a test data set with identified chromatographic peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the ion chromatogram for one chromatographic peak in the data. chrs <- chromatogram(faahko_sub, rt = c(2700, 2900), mz = 335)  chrs #> XChromatograms with 1 row and 3 columns #>                    1               2               3 #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 0        peaks: 1        peaks: 0 #> phenoData with 1 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Identified chromatographic peaks chromPeaks(chrs) #>        mz mzmin mzmax       rt   rtmin    rtmax    into    intb  maxo sn sample #> CP095 335   335   335 2786.199 2764.29 2812.803 1496244 1451652 58736 55      2 #>       row column #> CP095   1      2  ## Plot the chromatogram plot(chrs)  ## Extract chromatograms for multiple ranges. mzr <- matrix(c(335, 335, 344, 344), ncol = 2, byrow = TRUE) rtr <- matrix(c(2700, 2900, 2600, 2750), ncol = 2, byrow = TRUE) chrs <- chromatogram(faahko_sub, mz = mzr, rt = rtr)  chromPeaks(chrs) #>        mz mzmin mzmax       rt    rtmin    rtmax    into    intb   maxo sn #> CP095 335   335   335 2786.199 2764.290 2812.803 1496244 1451652  58736 55 #> CP003 344   344   344 2679.783 2646.919 2709.517 5210016 5135917 152320 68 #> CP090 344   344   344 2686.042 2648.483 2714.211 5700652 5574745 151744 87 #> CP193 344   344   344 2682.914 2643.790 2731.427 5255689 4946143 125632 41 #>       sample row column #> CP095      2   1      2 #> CP003      1   2      1 #> CP090      2   2      2 #> CP193      3   2      3  plot(chrs)   ## Get access to all chromatograms for the second mz/rt range chrs[1, ] #> XChromatograms with 1 row and 3 columns #>                    1               2               3 #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 0        peaks: 1        peaks: 0 #> phenoData with 1 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Plot just that one plot(chrs[1, , drop = FALSE])"},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":null,"dir":"Reference","previous_headings":"","what":"Find row and column maximum values — colMax","title":"Find row and column maximum values — colMax","text":"Find row column maximum values numeric arrays.","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find row and column maximum values — colMax","text":"","code":"colMax(x, na.rm = FALSE, dims = 1) rowMax(x, na.rm = FALSE, dims = 1) which.colMax(x, na.rm = FALSE, dims = 1) which.rowMax(x, na.rm = FALSE, dims = 1)"},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find row and column maximum values — colMax","text":"x array two dimensions, containing numeric values na.rm logical. missing values (including 'NaN') omitted     calculations? (currently implemented) dims dimensions regarded \"rows\" \"columns\" maximize.     rowMax, maximum dimensions dims+1,     ...; colMax dimensions 1:dims.","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find row and column maximum values — colMax","text":"functions designed act like colSums series   functions except currently handle real arrays    remove NA values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find row and column maximum values — colMax","text":"numeric array suitable size, vector result   one-dimensional. dimnames (names   vector result) taken original array. .* functions, integer array suitable   size, vector result one-dimensional. indecies   returned accessing x one-dimensionally (.e.  x[index]). .colMax(), actual row   indecies determined using (.colMax(x)-1) %%   nrow(x) + 1. .rowMax(), actual column indecies   may determined using ceiling(rowMax(x)/nrow(x)).","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find row and column maximum values — colMax","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Collect MS^n peaks into xcmsFragments — collect-methods","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"Collecting Peaks xcmsFragmentss several   MS-runs using xcmsSet   xcmsRaw.","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"object = \"xcmsFragments\" collect(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"object (empty) xcmsFragments-class object xs xcmsSet-class object contains     picked ms1-peaks several experiments compMethod (\"floor\", \"round\", \"none\"): compare-method     used find parent peak MSnpeak comparing     MZ-values  MS1peaks MSnParentPeaks. snthresh, mzgap, uniq parameters     getspec-peakpicker included xcmsRaw.","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"running collect(xFragments,xSet) peak table   xcmsFragments includes ms1Peaks experiments stored   xcmsSet-object. contains relevant msN-peaks   xcmsRaw-objects, created temporarily paths   xcmsSet.","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"matrix columns: peakID unique identifier every peak MSnParentPeakID PeakID parent peak msLevel>1 - peak, 0 peak msLevel 1. msLevel msLevel peak. rt retention time peak midpoint mz mz-Value peak intensity intensity peak sample number sample xcmsSet GroupPeakMSn Used grouped xcmsSet groups CollisionEnergy collision energy fragment","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"xcms >= 3.15.3 please use compareChromatograms() instead correlate Correlate intensities two chromatograms . two Chromatogram objects different retention times first aligned match data points first data points second chromatogram. See help alignRt MSnbase::Chromatogram() details. correlate called single MChromatograms() object pairwise correlation chromatogram performed matrix correlation coefficients returned. Note correlation two chromatograms depends also order, e.g. correlate(chr1, chr2) might identical correlate(chr2, chr1). lower upper triangular part correlation matrix might thus different.","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"","code":"# S4 method for Chromatogram,Chromatogram correlate(   x,   y,   use = \"pairwise.complete.obs\",   method = c(\"pearson\", \"kendall\", \"spearman\"),   align = c(\"closest\", \"approx\"),   ... )  # S4 method for MChromatograms,missing correlate(   x,   y = NULL,   use = \"pairwise.complete.obs\",   method = c(\"pearson\", \"kendall\", \"spearman\"),   align = c(\"closest\", \"approx\"),   ... )  # S4 method for MChromatograms,MChromatograms correlate(   x,   y = NULL,   use = \"pairwise.complete.obs\",   method = c(\"pearson\", \"kendall\", \"spearman\"),   align = c(\"closest\", \"approx\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"x Chromatogram() MChromatograms() object. y Chromatogram() MChromatograms() object. use character(1) passed cor function. See cor() details. method character(1) passed cor function. See cor() details. align character(1) defining alignment method used. See help alignRt MSnbase::Chromatogram() details. value parameter passed method parameter alignRt. ... optional parameters passed along alignRt method tolerance , set 0 requires retention times identical.","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"numeric(1) matrix (called MChromatograms objects) correlation coefficient. matrix returned, rows represent chromatograms x columns chromatograms y.","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"Michael Witting, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"","code":"chr1 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(5, 29, 50, NA, 100, 12, 3, 4, 1, 3)) chr2 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(80, 50, 20, 10, 9, 4, 3, 4, 1, 3)) chr3 <- Chromatogram(rtime = 3:9 + rnorm(7, sd = 0.3),     intensity = c(53, 80, 130, 15, 5, 3, 2))  chrs <- MChromatograms(list(chr1, chr2, chr3))  ## Using `compareChromatograms` instead of `correlate`. compareChromatograms(chr1, chr2) #> [1] -0.02314823 compareChromatograms(chr2, chr1) #> [1] -0.02314823  compareChromatograms(chrs, chrs) #>             [,1]        [,2]      [,3] #> [1,]  1.00000000 -0.02314823 0.9940933 #> [2,] -0.02314823  1.00000000 0.7977335 #> [3,]  0.99409327  0.79773352 1.0000000"},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":null,"dir":"Reference","previous_headings":"","what":"Find start and end points of a peak — descendZero","title":"Find start and end points of a peak — descendZero","text":"Decends sides data peak finds either points   greater equal zero intercept, intercept   given value, bottom first valley side.","code":""},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find start and end points of a peak — descendZero","text":"","code":"descendZero(y, istart = which.max(y)) descendValue(y, value, istart = which.max(y)) descendMin(y, istart = which.max(y))"},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find start and end points of a peak — descendZero","text":"y numeric vector values istart starting point descent value numeric value descend ","code":""},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find start and end points of a peak — descendZero","text":"integer vector length 2 starting ending indicies   peak start end points.","code":""},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find start and end points of a peak — descendZero","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Find start and end points of a peak — descendZero","text":"","code":"normdist <- dnorm(seq(-4, 4, .1)) - .1 xcms:::descendZero(normdist) #> ilower iupper  #>     25     57  normdist[xcms:::descendZero(normdist)] #> [1] 0.01092083 0.01092083 xcms:::descendValue(normdist, .15) #> ilower iupper  #>     32     50  normdist[xcms:::descendValue(normdist, .15)] #> [1] 0.1660852 0.1660852 xcms:::descendMin(normdist) #> ilower iupper  #>      1     81"},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Create report of analyte differences — diffreport-methods","title":"Create report of analyte differences — diffreport-methods","text":"Create report showing significant differences   two sets samples. Optionally create extracted ion chromatograms   significant differences.","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create report of analyte differences — diffreport-methods","text":"object = \"xcmsSet\" diffreport(object, class1 = levels(sampclass(object))[1],                    class2 = levels(sampclass(object))[2],                    filebase = character(), eicmax = 0, eicwidth = 200,                    sortpval = TRUE, classeic = c(class1,class2),                    value=c(\"\",\"maxo\",\"intb\"), metlin = FALSE, \t\t\t\t   h=480,w=640, mzdec=2, missing = \t\t\t\t   numeric(), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create report of analyte differences — diffreport-methods","text":"object xcmsSet object class1 character vector first set sample classes     compared class2 character vector second set sample classes     compared filebase base file name save report, .tsv file _eic     appended name tabular report EIC     directory, respectively. blank nothing saved eicmax number significantly different analytes create     EICs eicwidth width (seconds) EICs produced sortpval logical indicating whether reports sorted     p-value classeic character vector sample classes include EICs value intensity values used diffreport.      value=\"\", integrated peak intensities used.      value=\"maxo\", maximum peak intensities used.      value=\"intb\", baseline corrected integrated peak intensities used (available peak detection done findPeaks.centWave). metlin mass uncertainty use generating link Metlin metabolite     database. sign uncertainty indicates negative     positive mode data M+H M-H calculation. value FALSE     0 removes column h Numeric variable height eic boxplots printed . w Numeric variable width eic boxplots print made. mzdec Number decimal places title m/z values eic plot. missing numeric(1) defining optional value missing     values. missing = 0 e.g. replace NA values     feature matrix 0. Note also call     fillPeaks results feature matrix     NA values replaced 0. ... optional arguments passed mt.teststat     multtest package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create report of analyte differences — diffreport-methods","text":"method handles creation summary reports statistics   analytes significantly different   two sets samples. computes Welch's two-sample t-statistic   analyte ranks p-value. returns summary   report can optionally written tab-separated file. Additionally, heavy lifting involved creating   superimposed extracted ion chromatograms given number   analytes.  reading raw data files associated   samples interest one time. ,   prints name sample currently reading. Depending   number size samples, process can take   long time. base file name provided, report (see Value section)   saved tab separated file. EICs generated,   saved 640x480 PNG files newly created   subdirectory. However parameter can changed   commands arguments. numbered file names correspond   rows report. Chromatographic traces EICs colored labeled   sample class. Sample classes take color   current palette. color sample class assigned dependent   order xcmsSet object, order given   class arguments. Thus levels(sampclass(object))[1]   use color palette()[1] . way, sample   classes maintain color across number different   generated reports. multiple sample classes, xcms produce boxplots   different classes generate single anova p-value statistic.   Like eic's plot number corresponds row number   report.","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create report of analyte differences — diffreport-methods","text":"data frame following columns: fold mean fold change (always greater 1, see tstat     set sample classes higher) tstat Welch's two sample t-statistic, positive analytes     greater intensity class2, negative analytes     greater intensity class1 pvalue p-value t-statistic anova p-value anova statistic multiple classes mzmed median m/z peaks group mzmin minimum m/z peaks group mzmax maximum m/z peaks group rtmed median retention time peaks group rtmin minimum retention time peaks group rtmax maximum retention time peaks group npeaks number peaks assigned group Sample Classes number samples sample class represented group metlin URL metlin mass ... one column every sample class Sample Names integrated intensity value every sample ... one column every sample","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":null,"dir":"Reference","previous_headings":"","what":"Change the file path of an OnDiskMSnExp object — dirname","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"dirname allows get set path directory containing source files OnDiskMSnExp (XCMSnExp) object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"","code":"# S4 method for OnDiskMSnExp dirname(path)  # S4 method for OnDiskMSnExp dirname(path) <- value"},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"path OnDiskMSnExp. value character length 1 length equal number files defining new path files.","code":""},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":null,"dir":"Reference","previous_headings":"","what":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"function performs retention time correction assessing retention time deviation across samples using peak groups (features) containg chromatographic peaks present /samples. retention time deviation features sample described fitting either polynomial (smooth = \"loess\") linear (smooth = \"linear\") model data points. models subsequently used adjust retention time spectrum sample.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"","code":"do_adjustRtime_peakGroups(   peaks,   peakIndex,   rtime = list(),   minFraction = 0.9,   extraPeaks = 1,   smooth = c(\"loess\", \"linear\"),   span = 0.2,   family = c(\"gaussian\", \"symmetric\"),   peakGroupsMatrix = matrix(ncol = 0, nrow = 0),   subset = integer(),   subsetAdjust = c(\"average\", \"previous\") )"},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"peaks matrix data.frame identified chromatographic peaks samples. peakIndex list indices provides grouping information chromatographic peaks (across within samples). rtime list numeric vectors retention times per file/sample. minFraction PeakGroupsParam: numeric(1) 0 1 defining minimum required proportion samples peaks peak group identified. Peak groups passing criteria aligned across samples retention times individual spectra adjusted based alignment. minFraction = 1 peak group contain peaks samples experiment. Note subset provided, specified fraction relative defined subset samples total number samples within experiment (.e., peak present specified proportion subset samples). extraPeaks PeakGroupsParam: numeric(1) defining maximal number additional peaks samples assigned peak group (feature) retention time correction. data set 6 samples, extraPeaks = 1 uses peak groups total peak count <= 6 + 1. total peak count total number peaks assigned peak group considers also multiple peaks within sample assigned group. smooth PeakGroupsParam: character(1) defining function used interpolate corrected retention times peak groups. Can either \"loess\" \"linear\". span PeakGroupsParam: numeric(1) defining degree smoothing (smooth = \"loess\"). parameter passed internal call loess(). family PeakGroupsParam: character(1) defining method loess smoothing. Allowed values \"gaussian\" \"symmetric\". See loess() information. peakGroupsMatrix optional matrix (raw) retention times peak groups alignment performed. column represents sample, row feature/peak group. provided, matrix determined depending parameters minFraction extraPeaks. provided, minFraction extraPeaks ignored. subset ObiwarpParam PeakGroupsParam: integer indices samples within experiment alignment models estimated. Samples part subset adjusted based closest subset sample. See Subset-based alignment section details. subsetAdjust ObiwarpParam PeakGroupsParam: character(1) specifying method non-subset samples adjusted. Supported options \"previous\" \"average\" (default). See Subset-based alignment section details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"list numeric vectors adjusted     retention times grouped sample.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"alignment bases presence compounds can found /samples experiment. retention times individual spectra adjusted based alignment features corresponding house keeping compounds. paraneters minFraction extraPeaks can used fine tune features used alignment (.e. features likely correspond mentioned house keeping compounds). Parameter subset allows define subset samples within experiment aligned. samples part subset aligned based adjustment closest sample within subset. allows e.g. exclude blank samples alignment process retention times still adjusted based alignment results real samples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"method ensures returned adjusted retention times     increasingly ordered, just raw retention times.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for centWave peak detection — do_findChromPeaks_centWave","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"function performs peak density wavelet based     chromatographic peak detection high resolution LC/MS data centroid     mode [Tautenhahn 2008].","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"","code":"do_findChromPeaks_centWave(   mz,   int,   scantime,   valsPerSpect,   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = NULL,   sleep = 0,   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. sleep numeric(1) defining number seconds wait iterations. Defaults sleep = 0. > 0 plot generated visualizing identified chromatographic peak. Note: argument backward compatibility removed future. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685  https://doi.org/10.1186/s12859-023-05533-4 information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"matrix, row representing identified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peak across     scans. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intb Per-peak baseline corrected integrated peak intensity. maxo Maximum intensity peak. sn Signal noise ratio, defined (maxo - baseline)/sd,     sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit. Additional columns verboseColumns = TRUE:  mu Gaussian parameter mu. sigma Gaussian parameter sigma. h Gaussian parameter h. f Region number m/z ROI peak localized. dppm m/z deviation mass trace across scans ppm. scale Scale peak localized. scpos Peak position found wavelet analysis (scan number). scmin Left peak limit found wavelet analysis (scan number). scmax Right peak limit found wavelet analysis (scan numer). Additional columns verboseBetaColumns = TRUE:  beta_cor Correlation \"ideal\" bell curve raw data beta_snr Signal--noise residuals calculated beta_cor fit","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method identifies regions interest (ROIs) representing mass traces characterized regions less ppm m/z deviation consecutive scans LC/MS map. detail, starting single m/z, ROI extended m/z can found next scan (spectrum) difference mean m/z ROI smaller user defined ppm m/z. mean m/z ROI updated considering also newly included m/z value. ROIs , cleanup, analyzed using continuous wavelet transform (CWT) locate chromatographic peaks different scales. first analysis step skipped, regions interest passed roiList parameter.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"centWave designed work centroided mode, thus     expected data presented function. function exposes core chromatographic peak detection functionality     centWave method. function can called     directly, users generally call corresponding method     data object instead.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann \"Highly     sensitive feature detection high resolution LC/MS\"     BMC Bioinformatics 2008, 9:504","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"Ralf Tautenhahn, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and restrict to a certain retention time range data <- filterRt(filterFile(faahko_sub, 1), c(2500, 3000))  ## Get m/z and intensity values mzs <- mz(data) ints <- intensity(data)  ## Define the values per spectrum: valsPerSpect <- lengths(mzs)  ## Calling the function. We're using a large value for noise and prefilter ## to speed up the call in the example - in a real use case we would either ## set the value to a reasonable value or use the default value. res <- do_findChromPeaks_centWave(mz = unlist(mzs), int = unlist(ints),     scantime = rtime(data), valsPerSpect = valsPerSpect, noise = 10000,     prefilter = c(3, 10000)) #> Detecting mass traces at 25 ppm ...  #> OK #> Detecting chromatographic peaks in 186 regions of interest ... #>  OK: 47 found. head(res) #>         mz mzmin mzmax       rt    rtmin    rtmax      into      intb   maxo #> [1,] 453.2 453.2 453.2 2506.073 2501.378 2527.982 1007409.0 1007380.8  38152 #> [2,] 307.0 307.0 307.0 2618.750 2592.145 2645.354  284782.4  268039.8  16872 #> [3,] 302.0 302.0 302.0 2617.185 2595.275 2640.659  687146.6  671297.8  30552 #> [4,] 360.0 360.0 360.0 2682.913 2668.828 2698.562 5641322.3 5420634.7 317568 #> [5,] 361.1 361.1 361.1 2684.478 2665.698 2698.562 1158340.2 1116522.0  72272 #> [6,] 416.1 416.1 416.1 2682.913 2635.964 2709.517  487698.6  446552.1  12036 #>         sn #> [1,] 38151 #> [2,]    20 #> [3,]    46 #> [4,]    11 #> [5,]    11 #> [6,]    11"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"do_findChromPeaks_centWaveWithPredIsoROIs performs     two-step centWave based peak detection: chromatographic peaks     identified using centWave followed prediction location     identified peaks' isotopes mz-retention time space.     locations fed regions interest (ROIs) subsequent     centWave run. non overlapping peaks two peak detection     runs reported final list identified peaks. do_findChromPeaks_centWaveAddPredIsoROIs performs     centWave based peak detection based regions interest (ROIs)     representing predicted isotopes peaks submitted argument     peaks.. function returns matrix identified peaks     consisting input peaks peaks representing predicted isotopes     (found centWave algorithm).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"","code":"do_findChromPeaks_centWaveWithPredIsoROIs(   mz,   int,   scantime,   valsPerSpect,   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = NULL,   snthreshIsoROIs = 6.25,   maxCharge = 3,   maxIso = 5,   mzIntervalExtension = TRUE,   polarity = \"unknown\",   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE )  do_findChromPeaks_addPredIsoROIs(   mz,   int,   scantime,   valsPerSpect,   ppm = 25,   peakwidth = c(20, 50),   snthresh = 6.25,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   peaks. = NULL,   maxCharge = 3,   maxIso = 5,   mzIntervalExtension = TRUE,   polarity = \"unknown\" )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh do_findChromPeaks_addPredIsoROIs: numeric(1) defining signal noise threshold centWave algorithm. do_findChromPeaks_centWaveWithPredIsoROIs: numeric(1) defining signal noise threshold initial (first) centWave run. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. snthreshIsoROIs numeric(1) defining signal noise ratio cutoff used second centWave run identify peaks predicted isotope ROIs. maxCharge integer(1) defining maximal isotope charge. Isotopes defined charges 1:maxCharge. maxIso integer(1) defining number isotope peaks predicted peak identified first centWave run. mzIntervalExtension logical(1) whether mz range predicted isotope ROIs extended increase detection low intensity peaks. polarity character(1) specifying polarity data. Currently used, \"positive\", \"negative\" \"unknown\" provided. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685  https://doi.org/10.1186/s12859-023-05533-4 information. peaks. matrix xcmsPeaks object one returned call link{do_findChromPeaks_centWave} link{findPeaks.centWave} (verboseColumns = TRUE) peaks isotopes predicted used additional peak detectoin using centWave method. Required columns : \"mz\", \"mzmin\", \"mzmax\", \"scmin\", \"scmax\", \"scale\" \"\".","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"matrix, row representing identified chromatographic peak.     non-overlapping peaks identified centWave runs reported.     matrix columns :  mz Intensity weighted mean m/z values peaks across scans. mzmin Minimum m/z peaks. mzmax Maximum m/z peaks. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intb Per-peak baseline corrected integrated peak intensity. maxo Maximum intensity peak. sn Signal noise ratio, defined (maxo - baseline)/sd,     sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit. Additional columns verboseColumns = TRUE:  mu Gaussian parameter mu. sigma Gaussian parameter sigma. h Gaussian parameter h. f Region number m/z ROI peak localized. dppm m/z deviation mass trace across scans ppm. scale Scale peak localized. scpos Peak position found wavelet analysis (scan number). scmin Left peak limit found wavelet analysis (scan number). scmax Right peak limit found wavelet analysis (scan numer). Additional columns verboseBetaColumns = TRUE:  beta_cor Correlation \"ideal\" bell curve raw data beta_snr Signal--noise residuals calculated beta_cor fit","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"details centWave algorithm see     centWave.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"Hendrik Treutler, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"Massifquant Kalman filter (KF)-based chromatographic peak     detection XC-MS data centroid mode. identified peaks     can refined centWave method (see     do_findChromPeaks_centWave details centWave)     specifying withWave = TRUE.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"","code":"do_findChromPeaks_massifquant(   mz,   int,   scantime,   valsPerSpect,   ppm = 10,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   criticalValue = 1.125,   consecMissedLimit = 2,   unions = 1,   checkBack = 0,   withWave = FALSE )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. criticalValue numeric(1). Suggested values: (0.1-3.0). setting helps determine Kalman Filter prediciton margin error. real centroid belonging bonafide peak must fall within KF prediction margin error. Much like construction confidence interval, criticalVal loosely translates multiplier standard error prediction reported Kalman Filter. peak XC-MS sample small mass deviance ppm error, smaller critical value might better vice versa. consecMissedLimit integer(1) Suggested values: (1,2,3). peak proces detected Kalman Filter, Kalman Filter may find predicted centroid every scan. 1 consecutive failed predictions, setting informs Massifquant stop Kalman Filter following candidate peak. unions integer(1) set 1 apply t-test union segmentation; set 0 t-test applied chromatographically continous peaks sharing m/z range. Explanation: data points, sometimes Kalman Filter stops tracking peak prematurely. Another Kalman Filter instantiated begins following rest signal. tracking done backwards forwards, algorithmic defect leaves real peak divided two segments . option turned , program identifies segmented peaks combines (merges ) one two sample t-test. potential danger option truly distinct peaks may merged. checkBack integer(1) set 1 turned ; set 0 turned . convergence Kalman Filter peak's precise m/z mapping fast, sometimes incorporates erroneous centroids part peak (especially early ). scanBack option attempt remove occasional outlier lies beyond converged bounds Kalman Filter. option directly affect identification peak postprocessing measure; shown extremely useful thus far default set turned . withWave logical(1) TRUE, peaks identified first Massifquant subsequently filtered second step centWave algorithm, includes wavelet estimation.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"matrix, row representing identified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peaks across     scans. mzmin Minumum m/z peak. mzmax Maximum m/z peak. rtmin Minimum retention time peak. rtmax Maximum retention time peak. rt Retention time peak's midpoint. Integrated (original) intensity peak. maxo Maximum intensity peak. withWave set TRUE, result     returned do_findChromPeaks_centWave method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"algorithm's performance tested rigorously     high resolution LC/OrbiTrap, TOF-MS data centroid mode.     Simultaneous kalman filters identify peaks calculate     area curve. default parameters set operate     complex LC-MS Orbitrap sample. Users find useful     simple exploratory data analysis find set minimum     intensity, identify many scans average peak spans.     consecMissedLimit parameter yielded good performance     Orbitrap data set (2) TOF data found best     (1). may change algorithm yet     tested many samples. criticalValue parameter perhaps     dificult dial appropriately visual inspection peak     identification best suggested tool quick optimization.     ppm checkBack parameters shown less influence     parameters exist give users flexibility     better accuracy.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"Conley CJ, Smith R, Torgrip RJ, Taylor RM, Tautenhahn R Prince JT \"Massifquant: open-source Kalman filter-based XC-MS isotope trace feature detection\" Bioinformatics 2014, 30(18):2636-43.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"Christopher Conley","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and restrict to a certain retention time range data <- filterRt(filterFile(faahko_sub, 1), c(2500, 3000))  ## Get m/z and intensity values mzs <- mz(data) ints <- intensity(data)  ## Define the values per spectrum: valsPerSpect <- lengths(mzs)  ## Perform the peak detection using massifquant - setting prefilter to ## a high value to speed up the call for the example res <- do_findChromPeaks_massifquant(mz = unlist(mzs), int = unlist(ints),     scantime = rtime(data), valsPerSpect = valsPerSpect,     prefilter = c(3, 10000)) #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 10ppm ...  #> OK #>  69 Peaks. head(res) #>         mz mzmin mzmax    rtmin    rtmax       rt    into  maxo #> [1,] 426.1 426.1 426.1 2963.039 2999.033 2977.124  129958 29408 #> [2,] 590.3 590.3 590.3 2963.039 2999.033 2978.689  255317 19072 #> [3,] 454.1 454.1 454.1 2934.870 2975.559 2953.650  372685 20432 #> [4,] 309.1 309.1 309.1 2905.136 2967.734 2936.435  652140 38888 #> [5,] 475.2 475.2 475.2 2880.097 2999.033 2934.870 2415224 37160 #> [6,] 532.2 532.2 532.2 2853.493 2906.701 2878.532  232493 16480"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"function identifies peaks chromatographic     time domain described [Smith 2006]. intensity values     binned cutting LC/MS data slices (bins) mass unit     (binSize m/z) wide. Within bin maximal intensity     selected. peak detection performed bin     extending based steps parameter generate slices     comprising bins current_bin - steps +1     current_bin + steps - 1.     slices filtered matched filtration using     second-derative Gaussian model peak shape. filtration     peaks detected using signal--ration cut-. details     illustrations see [Smith 2006].","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"","code":"do_findChromPeaks_matchedFilter(   mz,   int,   scantime,   valsPerSpect,   binSize = 0.1,   impute = \"none\",   baseValue,   distance,   fwhm = 30,   sigma = fwhm/2.3548,   max = 5,   snthresh = 10,   steps = 2,   mzdiff = 0.8 - binSize * steps,   index = FALSE,   sleep = 0 )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. binSize numeric(1) specifying width bins/slices m/z dimension. impute Character string specifying method used missing value imputation. Allowed values \"none\" (linear interpolation), \"lin\" (linear interpolation), \"linbase\" (linear interpolation within certain bin-neighborhood) \"intlin\". See imputeLinInterpol details. baseValue base value empty elements set. considered method = \"linbase\" corresponds profBinLinBase's baselevel argument. distance method = \"linbase\": number non-empty neighboring element empty element considered linear interpolation. See details section information. fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) representing maximum number peaks expected/identified per slice. snthresh numeric(1) defining signal noise ratio cutoff. steps numeric(1) defining number bins merged filtration (.e. number neighboring bins joined slice filtration peak detection performed). mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. index logical(1) specifying whether indicies returned instead values m/z retention times. sleep numeric(1) defining number seconds wait iterations. Defaults sleep = 0. > 0 plot generated visualizing identified chromatographic peak. Note: argument backward compatibility removed future.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"matrix, row representing identified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peak across scans. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intf Integrated intensity filtered peak. maxo Maximum intensity peak. maxf Maximum intensity filtered peak. Rank peak merged EIC (<= max). sn Signal noise ratio peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"intensities binned provided m/z values within     spectrum (scan). Binning performed bins centered     around m/z values (.e. first bin includes m/z values     min(mz) - bin_size/2 min(mz) + bin_size/2). details binning missing value imputation see     binYonX imputeLinInterpol methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"function exposes core peak detection functionality     matchedFilter method. function can called     directly, users generally call corresponding method     data object instead (e.g. link{findPeaks.matchedFilter}     method).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and restrict to a certain retention time range data <- filterRt(filterFile(faahko_sub, 1), c(2500, 3000))  ## Get m/z and intensity values mzs <- mz(data) ints <- intensity(data)  ## Define the values per spectrum: valsPerSpect <- lengths(mzs)  res <- do_findChromPeaks_matchedFilter(mz = unlist(mzs), int = unlist(ints),     scantime = rtime(data), valsPerSpect = valsPerSpect) head(res) #>            mz mzmin mzmax       rt    rtmin    rtmax      into      intf  maxo #> [1,] 205.0000 205.0 205.0 2784.635 2770.550 2800.284 1778568.9 3610062.2 84280 #> [2,] 205.9819 205.9 206.0 2786.200 2772.115 2800.284  237993.6  448580.3 10681 #> [3,] 207.0821 207.0 207.1 2712.647 2698.562 2726.731  380873.0  730981.4 18800 #> [4,] 236.0956 236.0 236.1 2518.593 2504.508 2534.242  252282.0  458747.7 12957 #> [5,] 244.1000 244.1 244.1 2828.453 2814.369 2844.103  612169.9 1279308.9 31312 #> [6,] 266.0751 266.0 266.1 2828.453 2815.934 2844.103  113219.0  214886.8  5801 #>           maxf i       sn #> [1,] 195026.48 1 28.33394 #> [2,]  23860.11 1 16.53987 #> [3,]  40065.74 1 12.87314 #> [4,]  24536.55 1 14.99012 #> [5,]  69898.24 1 24.20989 #> [6,]  11773.56 1 10.83870"},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"function performs peak detection mass spectrometry     direct injection spectrum using wavelet based algorithm.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"","code":"do_findPeaks_MSW(   mz,   int,   snthresh = 3,   verboseColumns = FALSE,   scantime = numeric(),   valsPerSpect = integer(),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. snthresh numeric(1) defining signal noise ratio cutoff. verboseColumns logical(1) whether additional peak meta data columns returned. scantime ignored. valsPerSpect ignored. ... Additional parameters passed peakDetectionCWT function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"matrix, row representing identified peak, columns:  mz m/z value peak centroid position. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Always -1. rtmin Always -1. rtmax Always -1. Integrated (original) intensity peak. maxo Maximum intensity peak. intf Always NA. maxf Maximum MSW-filter response peak. sn Signal noise ratio.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"wrapper around peak picker Bioconductor's     MassSpecWavelet package calling     peakDetectionCWT     tuneInPeakInfo functions. See     xcmsDirect vignette information.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"Joachim Kutzera, Steffen Neumann, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"do_groupChromPeaks_density function performs chromatographic peak grouping based density (distribution) peaks, found different samples, along retention time axis slices overlapping mz ranges.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"","code":"do_groupChromPeaks_density(   peaks,   sampleGroups,   bw = 30,   minFraction = 0.5,   minSamples = 1,   binSize = 0.25,   maxFeatures = 50,   sleep = 0,   index = seq_len(nrow(peaks)) )"},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"peaks matrix data.frame mz values retention times identified chromatographic peaks samples experiment. Required columns \"mz\", \"rt\" \"sample\". latter contain numeric values representing index sample peak found. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). bw PeakDensityParam: numeric(1) defining bandwidth (standard deviation ot smoothing kernel) used. argument passed [density() method. minFraction PeakDensityParam: numeric(1) defining minimum fraction samples least one sample group peaks present considered peak group (feature). minSamples PeakDensityParam: numeric(1) minimum number samples least one sample group peaks detected considered peak group (feature). binSize PeakDensityParam: numeric(1) defining size overlapping slices m/z dimension. maxFeatures PeakDensityParam: numeric(1) maximum number peak groups identified single mz slice. sleep numeric(1) defining time sleep iterations plot result current iteration. index optional integer providing indices peaks original peak matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"data.frame, row representing (mz-rt) feature (.e. peak group) columns: \"mzmed\": median peaks' apex mz values. \"mzmin\": smallest mz value peaks' apex within feature. \"mzmax\":largest mz value peaks' apex within feature. \"rtmed\": median peaks' retention times. \"rtmin\": smallest retention time peaks group. \"rtmax\": largest retention time peaks group. \"npeaks\": total number peaks assigned feature. \"peakidx\": list indices peaks feature peaks input matrix. Note number can larger total number samples, since multiple peaks sample assigned feature.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"overlapping slices along mz dimension, function calculates density distribution identified peaks along retention time axis groups peaks different samples close . See (Smith 2006) details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"default settings might appropriate LC/GC-MS setups, especially bw binSize parameter adjusted accordingly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the matrix with the identified peaks from the xcmsSet: pks <- chromPeaks(faahko_sub)  ## Perform the peak grouping with default settings: res <- do_groupChromPeaks_density(pks, sampleGroups = rep(1, 3))  ## The feature definitions: head(res) #>   mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks 1      peakidx #> 1 279.0 279.0 279.0 2787.765 2787.765 2787.766      2 2      11, 199 #> 2 286.2 286.2 286.2 3254.904 3250.992 3258.815      2 2     115, 205 #> 3 300.2 300.2 300.2 3387.143 3379.317 3390.271      4 3 35, 125,.... #> 4 301.0 301.0 301.0 2787.766 2786.200 2792.459      3 3  10, 97, 198 #> 5 305.1 305.1 305.1 2994.338 2994.338 2994.339      2 2      15, 203 #> 6 305.1 305.1 305.1 2923.917 2923.916 2923.917      2 2      14, 202"},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"do_groupChromPeaks_nearest function groups peaks across samples creating master peak list assigning corresponding peaks samples peak group (.e. feature). method inspired correspondence algorithm mzMine (Katajamaa 2006).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"","code":"do_groupChromPeaks_nearest(   peaks,   sampleGroups,   mzVsRtBalance = 10,   absMz = 0.2,   absRt = 15,   kNN = 10 )"},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"peaks matrix data.frame mz values retention times identified chromatographic peaks samples experiment. Required columns \"mz\", \"rt\" \"sample\". latter contain numeric values representing index sample peak found. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). mzVsRtBalance NearestPeaksParam: numeric(1) representing factor m/z values multiplied calculating (euclician) distance two peaks. absMz NearestPeaksParam MzClustParam: numeric(1) maximum tolerated distance m/z values. absRt NearestPeaksParam: numeric(1) maximum tolerated distance retention times. kNN NearestPeaksParam: integer(1) representing number nearest neighbors check.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"list elements \"featureDefinitions\" \"peakIndex\". \"featureDefinitions\" matrix, row representing (mz-rt) feature (.e. peak group) columns: \"mzmed\": median peaks' apex mz values. \"mzmin\": smallest mz value peaks' apex within feature. \"mzmax\":largest mz value peaks' apex within feature. \"rtmed\": median peaks' retention times. \"rtmin\": smallest retention time peaks feature. \"rtmax\": largest retention time peaks feature. \"npeaks\": total number peaks assigned feature. \"peakIndex\" list indices peaks feature peaks input matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"Katajamaa M, Miettinen J, Oresic M: MZmine: Toolbox processing visualization mass spectrometry based molecular profile data. Bioinformatics 2006, 22:634-636.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"do_groupPeaks_mzClust function performs high resolution correspondence single spectra samples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"","code":"do_groupPeaks_mzClust(   peaks,   sampleGroups,   ppm = 20,   absMz = 0,   minFraction = 0.5,   minSamples = 1 )"},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"peaks matrix data.frame mz values retention times identified chromatographic peaks samples experiment. Required columns \"mz\", \"rt\" \"sample\". latter contain numeric values representing index sample peak found. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). ppm MzClustParam: numeric(1) representing relative m/z error clustering/grouping (parts per million). absMz NearestPeaksParam MzClustParam: numeric(1) maximum tolerated distance m/z values. minFraction PeakDensityParam: numeric(1) defining minimum fraction samples least one sample group peaks present considered peak group (feature). minSamples PeakDensityParam: numeric(1) minimum number samples least one sample group peaks detected considered peak group (feature).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"list elements \"featureDefinitions\" \"peakIndex\". \"featureDefinitions\" matrix, row representing (mz-rt) feature (.e. peak group) columns: \"mzmed\": median peaks' apex mz values. \"mzmin\": smallest mz value peaks' apex within feature. \"mzmax\": largest mz value peaks' apex within feature. \"rtmed\": always -1. \"rtmin\": always -1. \"rtmax\": always -1. \"npeaks\": total number peaks assigned feature. Note number can larger total number samples, since multiple peaks sample assigned group. \"peakIndex\" list indices peaks peak group peaks input matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"Saira . Kazmi, Samiran Ghosh, Dong-Guk Shin, Dennis W. Hill David F. GrantAlignment high resolution mass spectra: development heuristic approach metabolomics. Metabolomics, Vol. 2, . 2, 75-83 (2006)","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":null,"dir":"Reference","previous_headings":"","what":"Allocate double, integer, or logical matricies — doubleMatrix","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"Allocate double, integer, logical matricies one step without   copying memory around.","code":""},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"","code":"doubleMatrix(nrow = 0, ncol = 0) integerMatrix(nrow = 0, ncol = 0) logicalMatrix(nrow = 0, ncol = 0)"},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"nrow number matrix rows ncol number matrix columns","code":""},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"Matrix double, integer, logical values. Memory zeroed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":null,"dir":"Reference","previous_headings":"","what":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"estimatePrecursorIntensity determines precursor intensity MS 2 spectrum based intensity respective signal neighboring MS 1 spectra (.e. based peak m/z matching precursor m/z MS 2 spectrum). Based parameter method either intensity peak previous MS 1 scan used (method = \"previous\") interpolation intensity previous subsequent MS1 scan used (method = \"interpolation\", considers also retention times two MS1 scans retention time MS2 spectrum).","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"","code":"estimatePrecursorIntensity(   x,   ppm = 10,   method = c(\"previous\", \"interpolation\"),   BPPARAM = bpparam() )"},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"x OnDiskMSnExp XCMSnExp object. ppm numeric(1) defining maximal acceptable difference (ppm) precursor m/z m/z corresponding peak MS 1 scan. method character(1) defining method precursor intensity determined (see description details). Defaults method = \"previous\". BPPARAM parallel processing setup. See bpparam() details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"numeric length equal number spectra x. NA returned MS 1 spectra matching peak MS 1 scan can found MS 2 spectrum","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":null,"dir":"Reference","previous_headings":"","what":"Empirically Transformed Gaussian function — etg","title":"Empirically Transformed Gaussian function — etg","text":"general function asymmetric chromatographic peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Empirically Transformed Gaussian function — etg","text":"","code":"etg(x, H, t1, tt, k1, kt, lambda1, lambdat, alpha, beta)"},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Empirically Transformed Gaussian function — etg","text":"x times evaluate function H peak height t1 time leading edge inflection point tt time trailing edge inflection point k1 leading edge parameter kt trailing edge parameter lambda1 leading edge parameter lambdat trailing edge parameter alpha leading edge parameter beta trailing edge parameter","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Empirically Transformed Gaussian function — etg","text":"function evaluated times x.","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Empirically Transformed Gaussian function — etg","text":"Jianwei Li. Development Evaluation Flexible Empirical Peak   Functions Processing Chromatographic Peaks. Anal. Chem., 69   (21), 4452-4462, 1997.   http://dx.doi.org/10.1021/ac970481d","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Empirically Transformed Gaussian function — etg","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":null,"dir":"Reference","previous_headings":"","what":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"Export feature table analysis MetaboAnalyst software (MetaboAnalystR R package).","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"","code":"exportMetaboAnalyst(   x,   file = NULL,   label,   value = \"into\",   digits = NULL,   groupnames = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"x XCMSnExp object identified chromatographic peaks grouped across samples. file character(1) defining file name. specified, matrix content returned. label either character(1) specifying phenodata column x defining sample grouping vector length samples x defining group assignment samples. value character(1) specifying value returned feature. See featureValues() details. digits integer(1) defining number significant digits used numeric. default NULL uses getOption(\"digits\"). See format() information. groupnames logical(1) whether row names resulting matrix feature IDs (groupnames = FALSE; default) IDs composed m/z retention time features (format M<m/z>T<rt> (groupnames = TRUE). See help groupnames function details. ... additional parameters passed featureValues() function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"file specified, function returns matrix format supported MetaboAnalyst.","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":null,"dir":"Reference","previous_headings":"","what":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"UPDATE: extractMsData plotMsData functions deprecated (x, \"data.frame\") plot(x, type = \"XIC\") (x OnDiskMSnExp XCMSnExp object) used instead. See examples . aware filtering raw object might however drop adjusted retention times. cases advisable use applyAdjustedRtime() function prior filtering. Extract data.frame retention time, mz intensity values file/sample provided rt-mz range (full data range rt mz defined).","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"","code":"# S4 method for OnDiskMSnExp extractMsData(object, rt, mz, msLevel = 1L)  # S4 method for XCMSnExp extractMsData(   object,   rt,   mz,   msLevel = 1L,   adjustedRtime = hasAdjustedRtime(object) )"},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"object XCMSnExp OnDiskMSnExp object. rt numeric(2) retention time range data extracted. mz numeric(2) mz range. msLevel integer defining MS level(s) data sub-setted prior extraction; defaults msLevel = 1L. adjustedRtime (extractMsData,XCMSnExp): logical(1) specifying adjusted raw retention times reported. Defaults adjusted retention times, present object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"list length equal number samples/files object. element data.frame columns \"rt\", \"mz\" \"\" retention time, mz intensity tuples file. data available mz-rt range file data.frame 0 rows returned file.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the full MS data for a certain retention time range ## as a data.frame tmp <- filterRt(faahko_sub, rt = c(2800, 2900)) ms_all <- as(tmp, \"data.frame\") head(ms_all) #>   file       rt    mz     i #> 1    1 2800.284 200.1   968 #> 2    1 2800.284 201.0  1273 #> 3    1 2800.284 201.9  1729 #> 4    1 2800.284 203.0   859 #> 5    1 2800.284 205.0 24136 #> 6    1 2800.284 206.0  3913 nrow(ms_all) #> [1] 79960"},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding of LC-MS features — feature-grouping","title":"Compounding of LC-MS features — feature-grouping","text":"Feature compounding aims identifying grouping LC-MS features representing different ions adducts (including isotopes) originating compound. MsFeatures package provides general framework functionality group features based different properties. groupFeatures methods XcmsExperiment() XCMSnExp objects implemented xcms extend enable compounding LC-MS data considering also e.g. feature peak shaped. Note functions simply define feature groups actually aggregate combine features. See MsFeatures::groupFeatures() overview general feature grouping concept well details individual settings parameters. available options groupFeatures xcms preprocessing results (.e. XcmsExperiment XCMSnExp objects correspondence analysis groupChromPeaks()) : Grouping similar retention times: groupFeatures-similar-rtime(). Grouping similar feature values across samples: AbundanceSimilarityParam(). Grouping similar peak shape extracted ion chromatograms: EicSimilarityParam(). ideal workflow grouping features sequentially perform methods (listed order). Compounded feature groups can accessed featureGroups function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding of LC-MS features — feature-grouping","text":"","code":"# S4 method for XcmsResult featureGroups(object)  # S4 method for XcmsResult featureGroups(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding of LC-MS features — feature-grouping","text":"object XcmsExperiment() XCMSnExp() object LC-MS pre-processing results. value featureGroups<-: replacement feature groups object. length 1 length equal number features object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding of LC-MS features — feature-grouping","text":"Johannes Rainer, Mar Garcia-Aloy, Vinicius Veri Hernandes","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract ion chromatograms for each feature — featureChromatograms","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"Extract ion chromatograms features XcmsExperiment XCMSnExp object. function returns feature extracted ion chromatograms (along associated chromatographic peaks) sample. chromatogram extracted m/z - rt region including chromatographic peaks features (.e. based ranges \"mzmin\", \"mzmax\", \"rtmin\", \"rtmax\" chromatographic peaks feature). default chromatographic peaks associated feature included. object XCMSnExp object parameter include allows also return chromatographic peaks apex position within selected region (include = \"apex_within\") chromatographic peak overlapping m/z retention time range (include = \"\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"","code":"featureChromatograms(object, ...)  # S4 method for XcmsExperiment featureChromatograms(   object,   expandRt = 0,   expandMz = 0,   aggregationFun = \"max\",   features = character(),   return.type = \"XChromatograms\",   chunkSize = 2L,   ...,   progressbar = TRUE,   BPPARAM = bpparam() )  # S4 method for XCMSnExp featureChromatograms(   object,   expandRt = 0,   aggregationFun = \"max\",   features,   include = c(\"feature_only\", \"apex_within\", \"any\", \"all\"),   filled = FALSE,   n = length(fileNames(object)),   value = c(\"maxo\", \"into\"),   expandMz = 0,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"object XcmsExperiment XCMSnExp object grouped chromatographic peaks. ... optional arguments passed along chromatogram() function. expandRt numeric(1) expand retention time range chromatographic peak constant value side. expandMz numeric(1) expand m/z range chromatographic peak constant value side. aware extending m/z range extracted EIC might longer represent actual identified chromatographic peak intensities potential additional mass peaks within spectra aggregated final reported intensity value per spectrum (retention time). aggregationFun character(1) specifying name used aggregate intensity values across m/z value range retention time. default \"max\" returns base peak chromatogram. features integer, character logical defining subset features chromatograms returned. Can index features featureDefinitions, feature IDs (row names featureDefinitions) logical vector. return.type character(1) defining result returned. present return.type = \"XChromatograms\" supported results thus returned XChromatograms() object. chunkSize object XcmsExperiment: integer(1) defining number files data loaded time memory. Defaults chunkSize = 2L. progressbar logical(1) defining whether progress bar shown. BPPARAM object XcmsExperiment: parallel processing setup. Defaults BPPARAM = bpparam(). See bpparam() information. include object XCMSnExp: character(1) defining chromatographic peaks (related feature definitions) included returned XChromatograms(). Defaults \"feature_only\"; See description options details. filled object XCMSnExp: logical(1) whether filled-peaks included result object. default filled = FALSE, .e. detected peaks reported. n object XCMSnExp: integer(1) optionally specify number top n samples EIC extracted. value object XCMSnExp: character(1) specifying column used sort samples. Can either \"maxo\" (default) \"\" use maximal peak intensity integrated peak area, respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"XChromatograms() object. future, depending parameter return.type, data might returned different object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"Parameters include, filled, n value supported object XCMSnExp. extracting EICs top n samples can happen one features specified features dropped detected peak top n samples. chance happen smaller x contains also filled-peaks (fillChromPeaks).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Subset the object to a smaller retention time range xdata <- filterRt(faahko_sub, c(2500, 3500)) #> Filter spectra  xdata <- groupChromPeaks(xdata,     param = PeakDensityParam(minFraction = 0.8, sampleGroups = rep(1, 3)))  ## Get the feature definitions featureDefinitions(xdata) #>      mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks 1      peakidx #> FT01 300.2 300.2 300.2 3387.143 3379.317 3390.271      4 3 35, 91, .... #> FT02 301.0 301.0 301.0 2787.766 2786.200 2792.459      3 3  10, 63, 117 #> FT03 326.2 326.2 326.2 3416.877 3415.311 3424.700      3 3  37, 94, 132 #> FT04 343.0 343.0 343.0 2684.479 2678.218 2686.042      3 3   6, 58, 114 #> FT05 344.0 344.0 344.0 2682.914 2679.783 2686.042      3 3   3, 56, 112 #> FT06 365.0 365.0 365.0 2684.479 2679.783 2686.042      3 3   7, 59, 115 #> FT07 496.2 496.2 496.2 3335.498 3316.719 3340.194      3 3 47, 103, 138 #> FT08 496.2 496.2 496.2 3402.791 3384.012 3410.617      3 3 48, 104, 139 #> FT09 522.2 522.2 522.2 3387.142 3344.888 3434.091      6 3 52, 53, .... #> FT10 523.2 523.2 523.2 3387.142 3344.888 3434.091      6 3 38, 39, .... #> FT11 532.2 532.2 532.2 3482.605 3476.344 3488.863      3 3 49, 102, 137 #>      ms_level #> FT01        1 #> FT02        1 #> FT03        1 #> FT04        1 #> FT05        1 #> FT06        1 #> FT07        1 #> FT08        1 #> FT09        1 #> FT10        1 #> FT11        1  ## Extract ion chromatograms for the first 3 features. Parameter ## `features` can be either the feature IDs or feature indices. chrs <- featureChromatograms(xdata,     features = rownames(featureDefinitions)[1:3])  ## Plot the XIC for the first feature using different colors for each file plot(chrs[1, ], col = c(\"red\", \"green\", \"blue\"))"},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract spectra associated with features — featureSpectra","title":"Extract spectra associated with features — featureSpectra","text":"function returns spectra associated identified features input object. default, spectra returned features (MS levels), parameter features allows specify/select features result returned. Parameter msLevel allows define whether MS level 1 2 spectra returned. msLevel = 1L MS1 spectra within retention time range chromatographic peak (respective data file) associated feature returned. Note samples peak identified (even filled-) spectra returned. msLevel = 2L MS2 spectra retention time within retention time range precursor m/z within m/z range chromatographic peak feature returned. See also chromPeakSpectra() (used internally extract spectra chromatographic peak feature) additional information, specifically also parameter method. default (method = \"\") spectra associated chromatographic peaks feature returned. option method, single spectrum per chromatographic peak returned (hence multiple spectra per feature). ID chromatographic peak (.e. row name chromPeaks) feature (.e., row name featureDefinitions) available returned Spectra() spectra variables \"peak_id\" \"feature_id\", respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract spectra associated with features — featureSpectra","text":"","code":"featureSpectra(object, ...)  # S4 method for XcmsExperiment featureSpectra(   object,   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   skipFilled = FALSE,   return.type = c(\"Spectra\", \"List\"),   features = character(),   ... )  # S4 method for XCMSnExp featureSpectra(   object,   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   skipFilled = FALSE,   return.type = c(\"MSpectra\", \"Spectra\", \"list\", \"List\"),   features = character(),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract spectra associated with features — featureSpectra","text":"object XcmsExperiment XCMSnExp object feature defitions. ... additional arguments passed along chromPeakSpectra(), method. msLevel integer(1) defining MS level spectra returned. expandRt numeric(1) expand retention time range peak constant value side. expandMz numeric(1) expand m/z range peak constant value side. ppm numeric(1) expand m/z range peak (side) value dependent peak's m/z. skipFilled logical(1) whether spectra filled-peaks reported . return.type character(1) defining type result object returned. features character, logical integer allowing specify subset features featureDefinitions spectra returned (providing either ID, logical vector length nrow(featureDefinitions(x)) index featureDefinitions(x)). parameter overrides skipFilled supported return.type either \"Spectra\" \"List\".","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract spectra associated with features — featureSpectra","text":"function returns either Spectra() (return.type = \"Spectra\") List Spectra (return.type = \"List\"). latter, order length matches parameter features (features defined order features featureDefinitions(object)). Spectra variables \"peak_id\" \"feature_id\" define chromatographic peak feature individual spectrum associated .","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract spectra associated with features — featureSpectra","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":null,"dir":"Reference","previous_headings":"","what":"Simple feature summaries — featureSummary","title":"Simple feature summaries — featureSummary","text":"Simple function calculate feature summaries. include counts percentages samples chromatographic peak present feature counts percentages samples one chromatographic peak annotated feature. Also relative standard deviations (RSD) calculated integrated peak areas per feature across samples. `perSampleCounts = TRUE` also individual chromatographic peak counts per sample returned.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simple feature summaries — featureSummary","text":"","code":"featureSummary(   x,   group,   perSampleCounts = FALSE,   method = \"maxint\",   skipFilled = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simple feature summaries — featureSummary","text":"x [XcmsExperiment()] [XCMSnExp()] object correspondence results. group `numeric`, `logical`, `character` `factor` length `x` samples aggregate counts groups defined `group`. perSampleCounts `logical(1)` whether feature wise individual peak counts per sample returned . method `character` passed [featureValues()] function. See respective help page information. skipFilled `logical(1)` whether filled-peaks excluded (default) included summary calculation.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simple feature summaries — featureSummary","text":"`matrix` one row per feature columns: - `\"count\"`: total number samples peak found. - `\"perc\"`: percentage samples peak found. - `\"multi_count\"`: total number samples one peak   assigned feature. - `\"multi_perc\"`: percentage samples peak found,   also multiple peaks annotated feature. Example:   feature, least one peak detected 50 samples. 5 2 peaks   assigned feature. `\"multi_perc\"` case 10 - `\"rsd\"`: relative standard deviation (coefficient variation)   integrated peak area feature's peaks. - 4 columns repeated unique element (level) `group`   `group` provided. `perSampleCounts = TRUE` also one column sample returned peak counts per sample.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Simple feature summaries — featureSummary","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Gap Filling — fillChromPeaks","title":"Gap Filling — fillChromPeaks","text":"Gap filling integrate signal m/z-rt area feature (.e., chromatographic peak group) samples chromatographic peak feature identified add chromPeaks() matrix. filled-peaks indicated TRUE column \"is_filled\" result object's chromPeakData() data frame. method gap filling along settings can defined param argument. Two different approaches available: param = FillChromPeaksParam(): default original xcms code. Signal integrated m/z retention time range defined featureDefinitions() data frame, .e. \"rtmin\", \"rtmax\", \"mzmin\" \"mzmax\". method suggested underestimates actual peak area also available object XcmsExperiment object. See details information settings method. param = ChromPeakAreaParam(): area signal feature integrated defined based feature's chromatographic peak areas. m/z range default defined lower quartile chromatographic peaks' \"mzmin\" value upper quartile chromatographic peaks' \"mzmax\" values. retention time range area defined analogously. Alternatively, setting mzmin = median, mzmax = median, rtmin = median rtmax = median ChromPeakAreaParam, median \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" values detected chromatographic peaks feature used instead. contrast  FillChromPeaksParam approach method uses actual identified chromatographic peaks feature define area signal integrated. expandMz,expandMz<-: getter setter     expandMz slot object. expandRt,expandRt<-: getter setter     expandRt slot object. ppm,ppm<-: getter setter     ppm slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Gap Filling — fillChromPeaks","text":"","code":"fillChromPeaks(object, param, ...)  # S4 method for XcmsExperiment,ChromPeakAreaParam fillChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  FillChromPeaksParam(   expandMz = 0,   expandRt = 0,   ppm = 0,   fixedMz = 0,   fixedRt = 0 )  fixedRt(object)  fixedMz(object)  ChromPeakAreaParam(   mzmin = function(z) quantile(z, probs = 0.25),   mzmax = function(z) quantile(z, probs = 0.75),   rtmin = function(z) quantile(z, probs = 0.25),   rtmax = function(z) quantile(z, probs = 0.75) )  # S4 method for FillChromPeaksParam expandMz(object)  # S4 method for FillChromPeaksParam expandMz(object) <- value  # S4 method for FillChromPeaksParam expandRt(object)  # S4 method for FillChromPeaksParam expandRt(object) <- value  # S4 method for FillChromPeaksParam ppm(object)  # S4 method for FillChromPeaksParam ppm(object) <- value  # S4 method for XCMSnExp,FillChromPeaksParam fillChromPeaks(object, param, msLevel = 1L, BPPARAM = bpparam())  # S4 method for XCMSnExp,ChromPeakAreaParam fillChromPeaks(object, param, msLevel = 1L, BPPARAM = bpparam())  # S4 method for XCMSnExp,missing fillChromPeaks(object, param, BPPARAM = bpparam(), msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Gap Filling — fillChromPeaks","text":"object XcmsExperiment XCMSnExp object identified grouped chromatographic peaks. param ChromPeakAreaParam FillChromPeaksParam object defining approach used (see details section). ... currently ignored. msLevel integer(1) defining MS level peak filling performed (defaults msLevel = 1L). peak filling one MS level time supported, fill peaks MS level 1 2 run first using msLevel = 1 (returned result object) msLevel = 2. chunkSize fillChromPeaks object XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM Parallel processing settings. expandMz FillChromPeaksParam: numeric(1) defining value mz width peaks expanded. peak expanded mz direction expandMz * original m/z width. value 0 means expansion, value 1 grows peak 1 * m/z width peak resulting peaks twice original size m/z direction (expansion half m/z width sides). expandRt FillChromPeaksParam: numeric(1), expandMz retention time width. ppm FillChromPeaksParam: numeric(1) optionally specifying ppm m/z width peak region expanded. peaks m/z width smaller mean(c(mzmin, mzmax)) * ppm / 1e6, mzmin replaced mean(c(mzmin, mzmax)) - (mean(c(mzmin, mzmax)) * ppm / 2 / 1e6) mzmax mean(c(mzmin, mzmax)) + (mean(c(mzmin, mzmax)) * ppm / 2 / 1e6). applied eventually expanding m/z width using expandMz parameter. fixedMz FillChromPeaksParam: numeric(1) defining constant factor m/z width feature expanded. m/z width expanded sides fixedMz (.e. fixedMz subtracted lower m/z added upper m/z). expansion applied expandMz ppm. fixedRt FillChromPeaksParam: numeric(1) defining constant factor retention time width factor expanded. rt width expanded sides fixedRt (.e. fixedRt subtracted lower rt added upper rt). expansion applied expandRt. mzmin function applied values \"mzmin\" column chromatographic peaks feature define lower m/z value area signal feature integrated. Defaults mzmin = function(z) quantile(z, probs = 0.25) hence using 25% quantile values. mzmax function applied values \"mzmax\" column chromatographic peaks feature define upper m/z value area signal feature integrated. Defaults  mzmax = function(z) quantile(z, probs = 0.75) hence using 75% quantile values. rtmin function applied values \"rtmin\" column chromatographic peaks feature define lower rt value area signal feature integrated. Defaults rtmin = function(z) quantile(z, probs = 0.25) hence using 25% quantile values. rtmax function applied values \"rtmax\" column chromatographic peaks feature define upper rt value area signal feature integrated. Defaults rtmax = function(z) quantile(z, probs = 0.75) hence using 75% quantile values. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Gap Filling — fillChromPeaks","text":"XcmsExperiment  XCMSnExp object previously missing chromatographic peaks features filled chromPeaks() matrix. FillChromPeaksParam function returns  FillChromPeaksParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Gap Filling — fillChromPeaks","text":"correspondence (.e. grouping chromatographic peaks across samples) always features (peak groups) include peaks every sample. fillChromPeaks method defines intensity values features missing samples integrating signal m/z-rt region feature. Two different approaches define region available: ChromPeakAreaParam region defined based detected chromatographic peaks feature, FillChromPeaksParam region defined based m/z retention times feature (represent m/z retentention times apex position associated chromatographic peaks). latter approach various parameters available increase area signal integrated, either constant value (fixedMz fixedRt) feature-relative amount (expandMz expandRt). Adjusted retention times used available. Based peak finding algorithm used identify (chromatographic) peaks, different internal functions used guarantee integrated peak signal matches much possible peak signal integration used peak detection. peaks identified matchedFilter() method, signal integration performed profile matrix generated settings used also peak finding (using bin size example). direct injection data peaks identified MSW algorithm signal integrated along mz dimension. methods complete (raw) signal within area used.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Gap Filling — fillChromPeaks","text":"expandMz,expandRt,ppm,fixedMz,fixedRt See corresponding parameter . rtmin,rtmax,mzmin,mzmax See corresponding parameter .","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Gap Filling — fillChromPeaks","text":"reported \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" filled peaks represents actual MS area signal integrated. peak filled signal present file/sample respective mz-rt area. samples still show NA matrix returned featureValues() method.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Gap Filling — fillChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Gap Filling — fillChromPeaks","text":"","code":"## Load a test data set with identified chromatographic peaks res <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Perform the correspondence. We assign all samples to the same group. res <- groupChromPeaks(res,     param = PeakDensityParam(sampleGroups = rep(1, length(fileNames(res)))))  ## For how many features do we lack an integrated peak signal? sum(is.na(featureValues(res))) #> [1] 26  ## Filling missing peak data using the peak area from identified ## chromatographic peaks. res <- fillChromPeaks(res, param = ChromPeakAreaParam())  ## How many missing values do we have after peak filling? sum(is.na(featureValues(res))) #> [1] 3  ## Get the peaks that have been filled in: fp <- chromPeaks(res)[chromPeakData(res)$is_filled, ] head(fp) #>          mz mzmin mzmax       rt    rtmin    rtmax      into intb   maxo sn #> CP249 286.2 286.2 286.2 3252.556 3236.907 3274.857 1288521.3   NA  78048 NA #> CP250 380.1 380.1 380.1 3172.743 3129.707 3173.526  677133.8   NA  27040 NA #> CP251 447.2 447.2 447.2 3858.193 3828.068 3914.140 2156784.3   NA  61256 NA #> CP252 497.2 497.2 497.2 3382.447 3379.317 3438.786 7672169.4   NA 310912 NA #> CP253 510.2 510.2 510.2 3750.211 3741.213 3799.898 1927179.7   NA  61368 NA #> CP254 531.2 531.2 531.2 3340.193 3328.848 3383.620 7589092.3   NA 254464 NA #>       sample #> CP249      1 #> CP250      1 #> CP251      1 #> CP252      1 #> CP253      1 #> CP254      1  ## Get the process history step along with the parameters used to perform ## The peak filling: ph <- processHistory(res, type = \"Missing peak filling\")[[1]] ph #> Object of class \"XProcessHistory\" #>  type: Missing peak filling  #>  date: Thu Dec  7 08:43:03 2023  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: ChromPeakAreaParam  #>  MS level(s) 1   ## The parameter class: ph@param #> Object of class:  ChromPeakAreaParam  #>  Parameters: #>  - rtmin: function (z)  #> quantile(z, probs = 0.25) #> <environment: 0x55feb8ba0e40> #>  - rtmax: function (z)  #> quantile(z, probs = 0.75) #> <environment: 0x55feb8ba0e40> #>  - mzmin: function (z)  #> quantile(z, probs = 0.25) #> <environment: 0x55feb8ba0e40> #>  - mzmax: function (z)  #> quantile(z, probs = 0.75) #> <environment: 0x55feb8ba0e40>  ## It is also possible to remove filled-in peaks: res <- dropFilledChromPeaks(res)  sum(is.na(featureValues(res))) #> [1] 26"},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Integrate areas of missing peaks — fillPeaks-methods","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"sample, identify peak groups sample   represented. peak groups, integrate signal   region peak group create new peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"object = \"xcmsSet\" fillPeaks(object, method=\"\")","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"object xcmsSet object method filling method","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"peak grouping, always peak groups   include peaks every sample. method produces intensity   values missing samples integrating raw data peak   group region. According type raw-data 2 different   methods available. filling gcms/lcms data method \"chrom\" integrates   raw-data chromatographic domain, whereas \"MSW\" used   peaklists without retention-time information like   direct-infusion spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"xcmsSet objects filled peak groups.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"sample, identify peak groups sample   represented. peak groups, integrate signal   region peak group create new peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"object = \"xcmsSet\" fillPeaks.MSW(object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"object xcmsSet object","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"peak grouping, always peak groups   include peaks every sample. method produces intensity   values missing samples integrating raw data peak   group region. given group, start ending m/z values   integration defined median start   end points detected peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"contrast fillPeaks.chrom method maximum   intensity reported column \"maxo\" maximum   intensity measured expected peak area (defined columns   \"mzmin\" \"mzmax\"), largest intensity mz   value(s) closest \"mzmed\" feature.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"xcmsSet objects filled peak groups.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Integrate areas of missing peaks — fillPeaks.chrom-methods","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"sample, identify peak groups sample   represented. peak groups, integrate signal   region peak group create new peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"object = \"xcmsSet\" fillPeaks.chrom(object, nSlaves=0,expand.mz=1,expand.rt=1,     BPPARAM = bpparam())","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"object xcmsSet object nSlaves (DEPRECATED): number slaves/cores used     parallel peak filling.     MPI used installed, otherwise snow package employed     multicore support. none two packages available uses     parallel package parallel processing multiple CPUs     current machine. Users advised use BPPARAM     parameter instead. expand.mz Expansion factor m/z range used integration. expand.rt Expansion factor rentention time range used     integration. BPPARAM allows define specific parallel processing setup   current task (see bpparam   BiocParallel package help information). default uses   globally defined parallel setup.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"peak grouping, always peak groups   include peaks every sample. method produces intensity   values missing samples integrating raw data peak   group region. given group, start ending retention   time points integration defined median start   end points detected peaks. start end m/z   values similarly determined. Intensities   can still zero, rather unusual intensity peak.   case e.g. raw data threshholded,   integration area contains actual raw intensities, one sample   miscalibrated, thet raw data points (just) outside   integration area. Importantly, retention time correction data available,   alignment information used precisely integrate   propper region raw data. corrected retention time   beyond end raw data, value --number (NaN).","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"xcmsSet objects filled peak groups (maxo).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":null,"dir":"Reference","previous_headings":"","what":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"functions allow filter (subset) MChromatograms() XChromatograms() objects, .e. sets chromatographic data, without changing data (intensity retention times) within individual chromatograms (Chromatogram() objects). filterColumnsIntensityAbove: subsets MChromatograms objects keeping columns (samples) value larger provided threshold rows (.e. = \"\" column kept chromatograms column value larger threshold = \"\" chromatograms column fulfill criteria). Parameter value allows define value comparison performed, value = \"bpi\" maximum intensity chromatogram compared threshold, value = \"tic\" total sum intensities chromatogram compared threshold. XChromatogramsobject,value = \"maxo\"andvalue = \"\"supported compares largest intensity identified chromatographic peaks chromatogram withthreshold`, integrated peak area, respectively. filterColumnsKeepTop: subsets MChromatograms object keeping top n columns sorted value specified sortBy. detail, column value defined sortBy extracted chromatogram aggregated using aggregationFun. Thus, default, chromatogram maximum intensity determined (sortBy = \"bpi\") values summed chromatograms column (aggregationFun = sum). columns sorted values top n columns retained returned MChromatograms. Similar filterColumnsIntensityAbove function, function allows use XChromatograms objects sort columns column sortBy = \"maxo\" sortBy = \"\" chromPeaks matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"","code":"# S4 method for MChromatograms filterColumnsIntensityAbove(   object,   threshold = 0,   value = c(\"bpi\", \"tic\"),   which = c(\"any\", \"all\") )  # S4 method for MChromatograms filterColumnsKeepTop(   object,   n = 1L,   sortBy = c(\"bpi\", \"tic\"),   aggregationFun = sum )  # S4 method for XChromatograms filterColumnsIntensityAbove(   object,   threshold = 0,   value = c(\"bpi\", \"tic\", \"maxo\", \"into\"),   which = c(\"any\", \"all\") )  # S4 method for XChromatograms filterColumnsKeepTop(   object,   n = 1L,   sortBy = c(\"bpi\", \"tic\", \"maxo\", \"into\"),   aggregationFun = sum )"},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"object MChromatograms() XChromatograms() object. threshold filterColumnsIntensityAbove: numeric(1) threshold value compare . value character(1) defining value used comparison sorting. Can value = \"bpi\" (default) use maximum intensity per chromatogram value = \"tic\" use sum intensities per chromatogram. XChromatograms() objects also value = \"maxo\" value = \"\" supported use maximum intensity integrated area identified chromatographic peaks chromatogram. filterColumnsIntensityAbove: character(1) defining whether (= \"\", default) (= \"\") chromatograms column fulfill criteria column kept. n filterColumnsKeepTop: integer(1) specifying number columns returned. n rounded closest (larger) integer value. sortBy filterColumnsKeepTop: value columns ordered determine top n columns. Can either sortBy = \"bpi\" (default), case maximum intensity column's chromatograms used, sortBy = \"tic\" use total intensity sum chromatograms. XChromatograms() objects also value = \"maxo\" value = \"\" supported use maximum intensity integrated area identified chromatographic peaks chromatogram. aggregationFun filterColumnsKeepTop: function used aggregate (combine) values chromatograms column. Defaults aggregationFun = sum case sum values used rank columns. Alternatively mean, median similar function can used.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"filtered MChromatograms (XChromatograms) object number rows (EICs) eventually lower number columns (samples).","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"","code":"chr1 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(5, 29, 50, NA, 100, 12, 3, 4, 1, 3)) chr2 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(80, 50, 20, 10, 9, 4, 3, 4, 1, 3)) chr3 <- Chromatogram(rtime = 3:9 + rnorm(7, sd = 0.3),     intensity = c(53, 80, 130, 15, 5, 3, 2))  chrs <- MChromatograms(list(chr1, chr2, chr1, chr3, chr2, chr3),     ncol = 3, byrow = FALSE) chrs #> MChromatograms with 2 rows and 3 columns #>                   1              2              3 #>      <Chromatogram> <Chromatogram> <Chromatogram> #> [1,]     length: 10     length: 10     length: 10 #> [2,]     length: 10      length: 7      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  #### filterColumnsIntensityAbove ## ## Keep all columns with for which the maximum intensity of any of its ## chromatograms is larger 90 filterColumnsIntensityAbove(chrs, threshold = 90) #> MChromatograms with 2 rows and 3 columns #>                   1              2              3 #>      <Chromatogram> <Chromatogram> <Chromatogram> #> [1,]     length: 10     length: 10     length: 10 #> [2,]     length: 10      length: 7      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  ## Require that ALL chromatograms in a column have a value larger 90 filterColumnsIntensityAbove(chrs, threshold = 90, which = \"all\") #> MChromatograms with 2 rows and 1 column #>                   2 #>      <Chromatogram> #> [1,]     length: 10 #> [2,]      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  ## If none of the columns fulfills the criteria no columns are returned filterColumnsIntensityAbove(chrs, threshold = 900) #> MChromatograms with 2 rows and 0 columns #> phenoData with 0 variables #> featureData with 0 variables  ## Filtering XChromatograms allow in addition to filter on the columns ## \"maxo\" or \"into\" of the identified chromatographic peaks within each ## chromatogram.  #### filterColumnsKeepTop ## ## Keep the 2 columns with the highest sum of maximal intensities in their ## chromatograms filterColumnsKeepTop(chrs, n = 1) #> MChromatograms with 2 rows and 1 column #>                   2 #>      <Chromatogram> #> [1,]     length: 10 #> [2,]      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  ## Keep the 50 percent of columns with the highest total sum of signal. Note ## that n will be rounded to the next larger integer value filterColumnsKeepTop(chrs, n = 0.5 * ncol(chrs), sortBy = \"tic\") #> MChromatograms with 2 rows and 2 columns #>                   2              3 #>      <Chromatogram> <Chromatogram> #> [1,]     length: 10     length: 10 #> [2,]      length: 7      length: 7 #> phenoData with 0 variables #> featureData with 0 variables"},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":null,"dir":"Reference","previous_headings":"","what":"Apply an convolution filter using an FFT — filtfft","title":"Apply an convolution filter using an FFT — filtfft","text":"Expands vector length filter convolutes   using two successive FFTs.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Apply an convolution filter using an FFT — filtfft","text":"","code":"filtfft(y, filt)"},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Apply an convolution filter using an FFT — filtfft","text":"y numeric vector data filtered filt filter length nextn(length(y))","code":""},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Apply an convolution filter using an FFT — filtfft","text":"numeric vector length y.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Apply an convolution filter using an FFT — filtfft","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":null,"dir":"Reference","previous_headings":"","what":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"findChromPeaks Chromatogram MChromatograms object CentWaveParam parameter object performs centWave-based peak detection purely chromatographic data. See centWave details method CentWaveParam details parameter class. Note settings CentWaveParam used. See peaksWithCentWave() arguments used peak detection purely chromatographic data. chromatographic peak detection, identified peaks can also refined refineChromPeaks() method, can help reduce peak detection artifacts.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"","code":"# S4 method for Chromatogram,CentWaveParam findChromPeaks(object, param, ...)  # S4 method for MChromatograms,CentWaveParam findChromPeaks(object, param, BPPARAM = bpparam(), ...)  # S4 method for MChromatograms,MatchedFilterParam findChromPeaks(object, param, BPPARAM = BPPARAM, ...)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"object Chromatogram MChromatograms object. param CentWaveParam object specifying settings peak detection. See peaksWithCentWave() description arguments used peak detection. ... currently ignored. BPPARAM parameter class specifying parallel processing performed (XChromatograms objects). defaults bpparam(). See bpparam() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"called Chromatogram object, method returns XChromatogram object identified peaks. See peaksWithCentWave() details peak matrix content.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"","code":"## Loading a test data set with identified chromatographic peaks faahko_sub <- loadXcmsData(\"faahko_sub2\") faahko_sub <- filterRt(faahko_sub, c(2500, 3700)) #> Filter spectra  ## od <- as(filterFile(faahko_sub, 1L), \"MsExperiment\")  ## Extract chromatographic data for a small m/z range chr <- chromatogram(od, mz = c(272.1, 272.3))[1, 1]  ## Identify peaks with default settings xchr <- findChromPeaks(chr, CentWaveParam()) xchr #> Object of class: XChromatogram #> Intensity values aggregated using: sum  #> length of object: 766 #> from file: 1 #> mz range: [272.1, 272.3] #> rt range: [2501.378, 3698.567] #> MS level: 1 #> Identified chromatographic peaks (3): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn (1 more column(s)) #>  2734.556\t2709.517\t2770.55\t286653.221615386\t8907\t15 #>  3117.97\t3097.625\t3139.879\t372049.599925925\t16496\t33 #>  3487.299\t3457.565\t3509.208\t93049.7314242425\t2651\t15  ## Plot data and identified peaks. plot(xchr)   ## Perform peak detection on an MChromatograms object od3 <- readMSData(c(system.file(\"cdf/KO/ko15.CDF\", package = \"faahKO\"),     system.file(\"cdf/KO/ko16.CDF\", package = \"faahKO\"),     system.file(\"cdf/KO/ko18.CDF\", package = \"faahKO\")),     mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Disable parallel processing for this example register(SerialParam())  ## Extract chromatograms for a m/z - retention time slice chrs <- chromatogram(od3, mz = 344, rt = c(2500, 3500))  ## Perform peak detection using CentWave xchrs <- findChromPeaks(chrs, param = CentWaveParam()) xchrs #> XChromatograms with 1 row and 3 columns #>             ko15.CDF        ko16.CDF        ko18.CDF #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 2        peaks: 2        peaks: 1 #> phenoData with 1 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Extract the identified chromatographic peaks chromPeaks(xchrs) #>            rt    rtmin    rtmax      into      intb   maxo sn row column #> [1,] 2589.015 2571.801 2612.490  592297.3  554533.9  27600 14   1      1 #> [2,] 2679.783 2646.919 2709.517 5210015.9 5071346.6 152320 30   1      1 #> [3,] 2593.710 2559.281 2637.529  722415.3  640838.6  30640 14   1      2 #> [4,] 2686.042 2648.483 2714.211 5700652.0 5587464.2 151744 45   1      2 #> [5,] 2682.914 2643.790 2731.427 5255689.5 4983388.9 125632 28   1      3  ## plot the result plot(xchrs)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"findChromPeaks Chromatogram MChromatograms object MatchedFilterParam parameter object performs matchedFilter-based peak detection purely chromatographic data. See matchedFilter details method MatchedFilterParam details parameter class. Note settings MatchedFilterParam used. See peaksWithMatchedFilter() arguments used peak detection purely chromatographic data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"","code":"# S4 method for Chromatogram,MatchedFilterParam findChromPeaks(object, param, ...)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"object Chromatogram MChromatograms object. param MatchedFilterParam object specifying settings peak detection. See peaksWithMatchedFilter() description arguments used peak detection. ... currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"called Chromatogram object, method returns matrix identified peaks. See peaksWithMatchedFilter() details matrix content.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"","code":"## Loading a test data set with identified chromatographic peaks faahko_sub <- loadXcmsData(\"faahko_sub2\") faahko_sub <- filterRt(faahko_sub, c(2500, 3700)) #> Filter spectra  ## od <- as(filterFile(faahko_sub, 1L), \"MsExperiment\")  ## Extract chromatographic data for a small m/z range chr <- chromatogram(od, mz = c(272.1, 272.3))[1, 1]  ## Identify peaks with default settings xchr <- findChromPeaks(chr, MatchedFilterParam())  ## Plot the identified peaks plot(xchr)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":null,"dir":"Reference","previous_headings":"","what":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"centWave algorithm perform peak density wavelet based     chromatographic peak detection high resolution LC/MS data centroid     mode [Tautenhahn 2008]. CentWaveParam class allows specify settings     chromatographic peak detection using centWave method. Instances     created CentWaveParam constructor. findChromPeaks,OnDiskMSnExp,CentWaveParam method     performs chromatographic peak detection using centWave     algorithm samples OnDiskMSnExp     object. OnDiskMSnExp objects encapsule     experiment specific data load spectra data (mz intensity     values) fly original files applying also eventual     data manipulations. ppm,ppm<-: getter setter ppm     slot object. peakwidth,peakwidth<-: getter setter     peakwidth slot object. snthresh,snthresh<-: getter setter     snthresh slot object. prefilter,prefilter<-: getter setter     prefilter slot object. mzCenterFun,mzCenterFun<-: getter setter     mzCenterFun slot object. integrate,integrate<-: getter setter     integrate slot object. mzdiff,mzdiff<-: getter setter     mzdiff slot object. fitgauss,fitgauss<-: getter setter     fitgauss slot object. noise,noise<-: getter setter     noise slot object. verboseColumns,verboseColumns<-: getter     setter verboseColumns slot object. roiList,roiList<-: getter setter     roiList slot object. fistBaselineCheck,firstBaselineCheck<-: getter     setter firstBaselineCheck slot object. roiScales,roiScales<-: getter setter     roiScales slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"","code":"CentWaveParam(   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1L,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = numeric(),   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE )  # S4 method for OnDiskMSnExp,CentWaveParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for CentWaveParam ppm(object)  # S4 method for CentWaveParam ppm(object) <- value  # S4 method for CentWaveParam peakwidth(object)  # S4 method for CentWaveParam peakwidth(object) <- value  # S4 method for CentWaveParam snthresh(object)  # S4 method for CentWaveParam snthresh(object) <- value  # S4 method for CentWaveParam prefilter(object)  # S4 method for CentWaveParam prefilter(object) <- value  # S4 method for CentWaveParam mzCenterFun(object)  # S4 method for CentWaveParam mzCenterFun(object) <- value  # S4 method for CentWaveParam integrate(f)  # S4 method for CentWaveParam integrate(object) <- value  # S4 method for CentWaveParam mzdiff(object)  # S4 method for CentWaveParam mzdiff(object) <- value  # S4 method for CentWaveParam fitgauss(object)  # S4 method for CentWaveParam fitgauss(object) <- value  # S4 method for CentWaveParam noise(object)  # S4 method for CentWaveParam noise(object) <- value  # S4 method for CentWaveParam verboseColumns(object)  # S4 method for CentWaveParam verboseColumns(object) <- value  # S4 method for CentWaveParam roiList(object)  # S4 method for CentWaveParam roiList(object) <- value  # S4 method for CentWaveParam firstBaselineCheck(object)  # S4 method for CentWaveParam firstBaselineCheck(object) <- value  # S4 method for CentWaveParam roiScales(object)  # S4 method for CentWaveParam roiScales(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685  https://doi.org/10.1186/s12859-023-05533-4 information. object findChromPeaks:     OnDiskMSnExp  object containing MS-     experiment-relevant data. methods: parameter object. param CentWaveParam object containing settings centWave algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot. f integrate: CentWaveParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"CentWaveParam function returns CentWaveParam class instance settings specified chromatographic     peak detection centWave method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"centWave algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method identifies regions interest (ROIs) representing mass traces characterized regions less ppm m/z deviation consecutive scans LC/MS map. detail, starting single m/z, ROI extended m/z can found next scan (spectrum) difference mean m/z ROI smaller user defined ppm m/z. mean m/z ROI updated considering also newly included m/z value. ROIs , cleanup, analyzed using continuous wavelet transform (CWT) locate chromatographic peaks different scales. first analysis step skipped, regions interest passed via param parameter. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"ppm,peakwidth,snthresh,prefilter,mzCenterFun,integrate,mzdiff,fitgauss,noise,verboseColumns,roiList,firstBaselineCheck,roiScales,extendLengthMSW,verboseBetaColumns See corresponding parameter . Slots values exclusively accessed via corresponding getter setter methods listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports peak detection     OnDiskMSnExp objects (defined MSnbase     package). settings centWave algorithm can passed     CentWaveParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann \"Highly sensitive feature detection high resolution LC/MS\" BMC Bioinformatics 2008, 9:504","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"Ralf Tautenhahn, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"","code":"## Create a CentWaveParam object. Note that the noise is set to 10000 to ## speed up the execution of the example - in a real use case the default ## value should be used, or it should be set to a reasonable value. cwp <- CentWaveParam(ppm = 20, noise = 10000, prefilter = c(3, 10000)) ## Change snthresh parameter snthresh(cwp) <- 25 cwp #> Object of class:  CentWaveParam  #>  Parameters: #>  - ppm: [1] 20 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 25 #>  - prefilter: [1]     3 10000 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 10000 #>  - verboseColumns: [1] FALSE #>  - roiList: list() #>  - firstBaselineCheck: [1] TRUE #>  - roiScales: numeric(0) #>  - extendLengthMSW: [1] FALSE #>  - verboseBetaColumns: [1] FALSE  ## Perform the peak detection using centWave on some of the files from the ## faahKO package. Files are read using the readMSData from the MSnbase ## package library(faahKO) library(xcms) fls <- dir(system.file(\"cdf/KO\", package = \"faahKO\"), recursive = TRUE,            full.names = TRUE) raw_data <- readMSData(fls[1], mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Perform the peak detection using the settings defined above. res <- findChromPeaks(raw_data, param = cwp) #> Detecting mass traces at 20 ppm ...  #> OK #> Detecting chromatographic peaks in 937 regions of interest ... #>  OK: 120 found. head(chromPeaks(res)) #>          mz mzmin mzmax       rt    rtmin    rtmax      into      intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982 1007409.0 1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659  687146.6  671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517 5210015.9 5135916.9 152320 #> CP004 381.0 381.0 381.0 2678.218 2637.529 2720.472 2180565.2 2023571.9  52504 #> CP005 430.1 430.1 430.1 2681.348 2639.094 2712.647 2395840.3 2299899.6  65752 #> CP006 366.0 366.0 366.0 2679.783 2642.224 2718.907 3365174.0 3279468.3  79928 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    37      1 #> CP005    42      1 #> CP006    49      1"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":null,"dir":"Reference","previous_headings":"","what":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"method performs two-step centWave-based chromatographic     peak detection: first centWave run peaks identified     location potential isotopes mz-retention time     predicted. second centWave run performed     regions interest (ROIs). final list chromatographic     peaks comprises non-overlapping peaks centWave runs. CentWavePredIsoParam class allows specify     settings two-step centWave-based peak detection considering also     predicted isotopes peaks identified first centWave run.     Instances created CentWavePredIsoParam     constructor. See also documentation     CentWaveParam methods arguments class     inherits. findChromPeaks,OnDiskMSnExp,CentWavePredIsoParam     method performs two-step centWave-based chromatographic peak detection     samples OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. snthreshIsoROIs,snthreshIsoROIs<-: getter     setter snthreshIsoROIs slot object. maxCharge,maxCharge<-: getter     setter maxCharge slot object. maxIso,maxIso<-: getter     setter maxIso slot object. mzIntervalExtension,mzIntervalExtension<-: getter     setter mzIntervalExtension slot object. polarity,polarity<-: getter     setter polarity slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"","code":"CentWavePredIsoParam(   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1L,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = numeric(),   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE,   snthreshIsoROIs = 6.25,   maxCharge = 3,   maxIso = 5,   mzIntervalExtension = TRUE,   polarity = \"unknown\" )  # S4 method for OnDiskMSnExp,CentWavePredIsoParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for CentWavePredIsoParam snthreshIsoROIs(object)  # S4 method for CentWavePredIsoParam snthreshIsoROIs(object) <- value  # S4 method for CentWavePredIsoParam maxCharge(object)  # S4 method for CentWavePredIsoParam maxCharge(object) <- value  # S4 method for CentWavePredIsoParam maxIso(object)  # S4 method for CentWavePredIsoParam maxIso(object) <- value  # S4 method for CentWavePredIsoParam mzIntervalExtension(object)  # S4 method for CentWavePredIsoParam mzIntervalExtension(object) <- value  # S4 method for CentWavePredIsoParam polarity(object)  # S4 method for CentWavePredIsoParam polarity(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685  https://doi.org/10.1186/s12859-023-05533-4 information. snthreshIsoROIs numeric(1) defining signal noise ratio cutoff used second centWave run identify peaks predicted isotope ROIs. maxCharge integer(1) defining maximal isotope charge. Isotopes defined charges 1:maxCharge. maxIso integer(1) defining number isotope peaks predicted peak identified first centWave run. mzIntervalExtension logical(1) whether mz range predicted isotope ROIs extended increase detection low intensity peaks. polarity character(1) specifying polarity data. Currently used, \"positive\", \"negative\" \"unknown\" provided. object findChromPeaks:     OnDiskMSnExp  object containing MS-     experiment-relevant data. methods: parameter object. param CentWavePredIsoParam object settings chromatographic peak detection algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"CentWavePredIsoParam function returns  CentWavePredIsoParam class instance settings     specified two-step centWave-based peak detection considering also     isotopes. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"See centWave details centWave method. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"ppm,peakwidth,snthresh,prefilter,mzCenterFun,integrate,mzdiff,fitgauss,noise,verboseColumns,roiList,firstBaselineCheck,roiScales,extendLengthMSW,verboseBetaColumns,snthreshIsoROIs,maxCharge,maxIso,mzIntervalExtension,polarity See corresponding parameter .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports chromatographic peak     detection     OnDiskMSnExp objects (defined     MSnbase package). settings algorithm can     passed CentWavePredIsoParam object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"Hendrik Treutler, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"","code":"## Create a param object p <- CentWavePredIsoParam(maxCharge = 4) ## Change snthresh parameter snthresh(p) <- 25 p #> Object of class:  CentWavePredIsoParam  #>  Parameters: #>  - snthreshIsoROIs: [1] 6.25 #>  - maxCharge: [1] 4 #>  - maxIso: [1] 5 #>  - mzIntervalExtension: [1] TRUE #>  - polarity: [1] \"unknown\" #>  - ppm: [1] 25 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 25 #>  - prefilter: [1]   3 100 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 0 #>  - verboseColumns: [1] FALSE #>  - roiList: list() #>  - firstBaselineCheck: [1] TRUE #>  - roiScales: numeric(0) #>  - extendLengthMSW: [1] FALSE #>  - verboseBetaColumns: [1] FALSE"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":null,"dir":"Reference","previous_headings":"","what":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"Massifquant Kalman filter (KF)-based chromatographic peak     detection XC-MS data centroid mode. identified peaks     can refined centWave method (see     findChromPeaks-centWave details centWave)     specifying withWave = TRUE. MassifquantParam class allows specify     settings chromatographic peak detection using massifquant     method eventually combination centWave algorithm. Instances     created MassifquantParam constructor. findChromPeaks,OnDiskMSnExp,MassifquantParam     method performs chromatographic peak detection using     massifquant algorithm samples     OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. ppm,ppm<-: getter setter ppm     slot object. peakwidth,peakwidth<-: getter setter     peakwidth slot object. snthresh,snthresh<-: getter setter     snthresh slot object. prefilter,prefilter<-: getter setter     prefilter slot object. mzCenterFun,mzCenterFun<-: getter setter     mzCenterFun slot object. integrate,integrate<-: getter setter     integrate slot object. mzdiff,mzdiff<-: getter setter     mzdiff slot object. fitgauss,fitgauss<-: getter setter     fitgauss slot object. noise,noise<-: getter setter     noise slot object. verboseColumns,verboseColumns<-: getter     setter verboseColumns slot object. criticalValue,criticalValue<-: getter     setter criticalValue slot object. consecMissedLimit,consecMissedLimit<-: getter     setter consecMissedLimit slot object. unions,unions<-: getter     setter unions slot object. checkBack,checkBack<-: getter     setter checkBack slot object. withWave,withWave<-: getter     setter withWave slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"","code":"MassifquantParam(   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1L,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   criticalValue = 1.125,   consecMissedLimit = 2,   unions = 1,   checkBack = 0,   withWave = FALSE )  # S4 method for OnDiskMSnExp,MassifquantParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for MassifquantParam ppm(object)  # S4 method for MassifquantParam ppm(object) <- value  # S4 method for MassifquantParam peakwidth(object)  # S4 method for MassifquantParam peakwidth(object) <- value  # S4 method for MassifquantParam snthresh(object)  # S4 method for MassifquantParam snthresh(object) <- value  # S4 method for MassifquantParam prefilter(object)  # S4 method for MassifquantParam prefilter(object) <- value  # S4 method for MassifquantParam mzCenterFun(object)  # S4 method for MassifquantParam mzCenterFun(object) <- value  # S4 method for MassifquantParam integrate(f)  # S4 method for MassifquantParam integrate(object) <- value  # S4 method for MassifquantParam mzdiff(object)  # S4 method for MassifquantParam mzdiff(object) <- value  # S4 method for MassifquantParam fitgauss(object)  # S4 method for MassifquantParam fitgauss(object) <- value  # S4 method for MassifquantParam noise(object)  # S4 method for MassifquantParam noise(object) <- value  # S4 method for MassifquantParam verboseColumns(object)  # S4 method for MassifquantParam verboseColumns(object) <- value  # S4 method for MassifquantParam criticalValue(object)  # S4 method for MassifquantParam criticalValue(object) <- value  # S4 method for MassifquantParam consecMissedLimit(object)  # S4 method for MassifquantParam consecMissedLimit(object) <- value  # S4 method for MassifquantParam unions(object)  # S4 method for MassifquantParam unions(object) <- value  # S4 method for MassifquantParam checkBack(object)  # S4 method for MassifquantParam checkBack(object) <- value  # S4 method for MassifquantParam withWave(object)  # S4 method for MassifquantParam withWave(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2). first element used massifquant, specifices minimum peak length time scans. withWave = TRUE second argument represents maximum peak length subject greater mininum peak length (see also documentation do_findChromPeaks_centWave). snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2). first argument used (withWave = TRUE); see findChromPeaks-centWave details. second argument specifies minimum threshold maximum intensity chromatographic peak must met. mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. criticalValue numeric(1). Suggested values: (0.1-3.0). setting helps determine Kalman Filter prediciton margin error. real centroid belonging bonafide peak must fall within KF prediction margin error. Much like construction confidence interval, criticalVal loosely translates multiplier standard error prediction reported Kalman Filter. peak XC-MS sample small mass deviance ppm error, smaller critical value might better vice versa. consecMissedLimit integer(1) Suggested values: (1,2,3). peak proces detected Kalman Filter, Kalman Filter may find predicted centroid every scan. 1 consecutive failed predictions, setting informs Massifquant stop Kalman Filter following candidate peak. unions integer(1) set 1 apply t-test union segmentation; set 0 t-test applied chromatographically continous peaks sharing m/z range. Explanation: data points, sometimes Kalman Filter stops tracking peak prematurely. Another Kalman Filter instantiated begins following rest signal. tracking done backwards forwards, algorithmic defect leaves real peak divided two segments . option turned , program identifies segmented peaks combines (merges ) one two sample t-test. potential danger option truly distinct peaks may merged. checkBack integer(1) set 1 turned ; set 0 turned . convergence Kalman Filter peak's precise m/z mapping fast, sometimes incorporates erroneous centroids part peak (especially early ). scanBack option attempt remove occasional outlier lies beyond converged bounds Kalman Filter. option directly affect identification peak postprocessing measure; shown extremely useful thus far default set turned . withWave logical(1) TRUE, peaks identified first Massifquant subsequently filtered second step centWave algorithm, includes wavelet estimation. object findChromPeaks:     OnDiskMSnExp object containing MS-     experiment-relevant data. methods: parameter object. param MassifquantParam object containing settings massifquant algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot. f integrate: MassifquantParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"MassifquantParam function returns  MassifquantParam class instance settings     specified chromatographic peak detection massifquant method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"algorithm's performance tested rigorously     high resolution LC/OrbiTrap, TOF-MS data centroid mode.     Simultaneous kalman filters identify chromatographic peaks calculate     area curve. default parameters set operate     complex LC-MS Orbitrap sample. Users find useful     simple exploratory data analysis find set minimum     intensity, identify many scans average peak spans.     consecMissedLimit parameter yielded good performance     Orbitrap data set (2) TOF data found best     (1). may change algorithm yet     tested many samples. criticalValue parameter perhaps     dificult dial appropriately visual inspection peak     identification best suggested tool quick optimization.     ppm checkBack parameters shown less influence     parameters exist give users flexibility     better accuracy. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"ppm,peakwidth,snthresh,prefilter,mzCenterFun,integrate,mzdiff,fitgauss,noise,verboseColumns,criticalValue,consecMissedLimit,unions,checkBack,withWave See corresponding parameter . Slots values exclusively accessed via corresponding getter setter methods listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports chromatographic peak     detection     OnDiskMSnExp objects (defined     MSnbase package). settings massifquant     centWave algorithm can passed MassifquantParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"Conley CJ, Smith R, Torgrip RJ, Taylor RM, Tautenhahn R Prince JT \"Massifquant: open-source Kalman filter-based XC-MS isotope trace feature detection\" Bioinformatics 2014, 30(18):2636-43.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"Christopher Conley, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"","code":"## Create a MassifquantParam object. mqp <- MassifquantParam() ## Change snthresh prefilter parameters snthresh(mqp) <- 30 prefilter(mqp) <- c(6, 10000) mqp #> Object of class:  MassifquantParam  #>  Parameters: #>  - ppm: [1] 25 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 30 #>  - prefilter: [1]     6 10000 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 0 #>  - verboseColumns: [1] FALSE #>  - criticalValue: [1] 1.125 #>  - consecMissedLimit: [1] 2 #>  - unions: [1] 1 #>  - checkBack: [1] 0 #>  - withWave: [1] FALSE  ## Perform the peak detection using massifquant on the files from the ## faahKO package. Files are read using the readMSData from the MSnbase ## package library(faahKO) library(MSnbase) fls <- dir(system.file(\"cdf/KO\", package = \"faahKO\"), recursive = TRUE,            full.names = TRUE) raw_data <- readMSData(fls[1], mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method. ## Perform the peak detection using the settings defined above. res <- findChromPeaks(raw_data, param = mqp) #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 25ppm ...  #> OK #>  334 Peaks. head(chromPeaks(res)) #>          mz mzmin mzmax    rtmin    rtmax       rt    into   maxo sample #> CP001 578.4 578.4 578.4 4103.890 4141.449 4119.540 1748295 187328      1 #> CP002 577.3 577.3 577.3 4100.760 4130.495 4114.845 3854210 457024      1 #> CP003 367.2 367.2 367.2 4099.196 4164.924 4122.670  140753  11330      1 #> CP004 459.1 459.1 459.1 4099.196 4128.930 4113.280  144739  15874      1 #> CP005 401.2 401.2 401.2 4083.546 4221.262 4147.709  472407  11783      1 #> CP006 412.3 412.3 412.3 4072.591 4111.715 4089.806 2192350 183936      1"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"matchedFilter algorithm identifies peaks     chromatographic time domain described [Smith 2006]. intensity     values binned cutting LC/MS data slices (bins) mass     unit (binSize m/z) wide. Within bin maximal intensity     selected. chromatographic peak detection performed     bin extending based steps parameter generate     slices comprising bins current_bin - steps +1     current_bin + steps - 1. slices filtered     matched filtration using second-derative Gaussian model     peak shape. filtration peaks detected using signal--ratio     cut-. details illustrations see [Smith 2006]. MatchedFilterParam class allows specify     settings chromatographic peak detection using matchedFilter     method. Instances created MatchedFilterParam     constructor. findChromPeaks,OnDiskMSnExp,MatchedFilterParam     method performs peak detection using matchedFilter algorithm     samples OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. binSize,binSize<-: getter setter     binSize slot object. impute,impute<-: getter setter     impute slot object. baseValue,baseValue<-: getter setter     baseValue slot object. distance,distance<-: getter setter     distance slot object. fwhm,fwhm<-: getter setter     fwhm slot object. sigma,sigma<-: getter setter     sigma slot object. max,max<-: getter setter      max slot object. snthresh,snthresh<-: getter setter     snthresh slot object. steps,steps<-: getter setter     steps slot object. mzdiff,mzdiff<-: getter setter      mzdiff slot object. index,index<-: getter setter     index slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"","code":"MatchedFilterParam(   binSize = 0.1,   impute = \"none\",   baseValue = numeric(),   distance = numeric(),   fwhm = 30,   sigma = fwhm/2.3548,   max = 5,   snthresh = 10,   steps = 2,   mzdiff = 0.8 - binSize * steps,   index = FALSE )  # S4 method for OnDiskMSnExp,MatchedFilterParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for MatchedFilterParam binSize(object)  # S4 method for MatchedFilterParam binSize(object) <- value  # S4 method for MatchedFilterParam impute(object)  # S4 method for MatchedFilterParam impute(object) <- value  # S4 method for MatchedFilterParam baseValue(object)  # S4 method for MatchedFilterParam baseValue(object) <- value  # S4 method for MatchedFilterParam distance(object)  # S4 method for MatchedFilterParam distance(object) <- value  # S4 method for MatchedFilterParam fwhm(object)  # S4 method for MatchedFilterParam fwhm(object) <- value  # S4 method for MatchedFilterParam sigma(object)  # S4 method for MatchedFilterParam sigma(object) <- value  # S4 method for MatchedFilterParam max(x)  # S4 method for MatchedFilterParam max(object) <- value  # S4 method for MatchedFilterParam snthresh(object)  # S4 method for MatchedFilterParam snthresh(object) <- value  # S4 method for MatchedFilterParam steps(object)  # S4 method for MatchedFilterParam steps(object) <- value  # S4 method for MatchedFilterParam mzdiff(object)  # S4 method for MatchedFilterParam mzdiff(object) <- value  # S4 method for MatchedFilterParam index(object)  # S4 method for MatchedFilterParam index(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"binSize numeric(1) specifying width bins/slices m/z dimension. impute Character string specifying method used missing value imputation. Allowed values \"none\" (linear interpolation), \"lin\" (linear interpolation), \"linbase\" (linear interpolation within certain bin-neighborhood) \"intlin\". See imputeLinInterpol details. baseValue base value empty elements set. considered method = \"linbase\" corresponds profBinLinBase's baselevel argument. distance method = \"linbase\": number non-empty neighboring element empty element considered linear interpolation. See details section information. fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) representing maximum number peaks expected/identified per slice. snthresh numeric(1) defining signal noise cutoff used chromatographic peak detection step. steps numeric(1) defining number bins merged filtration (.e. number neighboring bins joined slice filtration peak detection performed). mzdiff numeric(1) defining minimum difference m/z peaks overlapping retention times index logical(1) specifying whether indicies returned instead values m/z retention times. object findChromPeaks:     OnDiskMSnExp object containing MS-     experiment-relevant data. methods: parameter object. param MatchedFilterParam object containing settings matchedFilter algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot. x max: MatchedFilterParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"MatchedFilterParam function returns  MatchedFilterParam class instance settings     specified chromatographic detection matchedFilter method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"intensities binned provided m/z values within     spectrum (scan). Binning performed bins centered     around m/z values (.e. first bin includes m/z values     min(mz) - bin_size/2 min(mz) + bin_size/2). details binning missing value imputation see     binYonX imputeLinInterpol methods. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"binSize,impute,baseValue,distance,fwhm,sigma,max,snthresh,steps,mzdiff,index See corresponding parameter . Slots values exclusively accessed via corresponding getter setter methods listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports chromatographic peak     detection     OnDiskMSnExp objects (defined     MSnbase package). settings matchedFilter     algorithm can passed MatchedFilterParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"","code":"## Create a MatchedFilterParam object. Note that we use a unnecessarily large ## binSize parameter to reduce the run-time of the example. mfp <- MatchedFilterParam(binSize = 5) ## Change snthresh parameter snthresh(mfp) <- 15 mfp #> Object of class:  MatchedFilterParam  #>  Parameters: #>  - binSize: [1] 5 #>  - impute: [1] \"none\" #>  - baseValue: numeric(0) #>  - distance: numeric(0) #>  - fwhm: [1] 30 #>  - sigma: [1] 12.73994 #>  - max: [1] 5 #>  - snthresh: [1] 15 #>  - steps: [1] 2 #>  - mzdiff: [1] -9.2 #>  - index: [1] FALSE  ## Perform the peak detection using matchecFilter on the files from the ## faahKO package. Files are read using the readMSData from the MSnbase ## package library(faahKO) library(MSnbase) fls <- dir(system.file(\"cdf/KO\", package = \"faahKO\"), recursive = TRUE,            full.names = TRUE) raw_data <- readMSData(fls[1], mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method. ## Perform the chromatographic peak detection using the settings defined ## above. Note that we are also disabling parallel processing in this ## example by registering a \"SerialParam\" res <- findChromPeaks(raw_data, param = mfp) head(chromPeaks(res)) #>             mz mzmin mzmax       rt    rtmin    rtmax      into    intf  maxo #> CP001 205.0000 205.0 205.0 2784.635 2770.550 2800.284 1778568.9 3580020 84280 #> CP002 205.0000 205.0 205.0 2784.635 2770.550 2800.284 1778568.9 3577971 84280 #> CP003 241.1460 241.1 241.2 3662.574 3646.924 3682.918 1465988.7 2234510 49728 #> CP004 241.1460 241.1 241.2 3662.574 3646.924 3682.918 1465988.7 2234510 49728 #> CP005 244.1000 244.1 244.1 2828.453 2814.369 2842.538  598990.3 1145078 31312 #> CP006 249.1591 249.1 249.2 3659.444 3643.794 3678.223 1435000.7 2367467 49040 #>            maxf i       sn sample #> CP001 194233.12 1 63.28090      1 #> CP002 194213.46 1 66.00099      1 #> CP003  96022.23 1 25.42409      1 #> CP004  96022.23 1 25.42643      1 #> CP005  64181.64 2 16.99513      1 #> CP006 104291.09 1 36.83500      1"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Chromatographic Peak Detection — findChromPeaks","title":"Chromatographic Peak Detection — findChromPeaks","text":"findChromPeaks method performs chromatographic peak detection LC/GC-MS data. peak detection algorithm can selected, configured, using param argument. Supported param objects : CentWaveParam(): chromatographic peak detection using centWave algorithm. CentWavePredIsoParam(): centWave predicted isotopes. Peak detection uses two-step centWave-based approach considering also feature isotopes. MatchedFilterParam(): peak detection using matched filter algorithm. MassifquantParam(): peak detection using Kalman filter-based massifquant method. MSWParam(): single-spectrum non-chromatography MS data peak detection. specific examples see help pages individual parameter classes listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Chromatographic Peak Detection — findChromPeaks","text":"","code":"findChromPeaks(object, param, ...)  # S4 method for MsExperiment,Param findChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   ...,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment,Param findChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   add = FALSE,   ...,   BPPARAM = bpparam() )"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Chromatographic Peak Detection — findChromPeaks","text":"object data object perform peak detection. Can OnDiskMSnExp(), XCMSnExp(), MChromatograms() MsExperiment() object. param parameter object selecting configuring algorithm. ... Optional parameters. msLevel integer(1) defining MS level chromatographic peak detection performed. chunkSize integer(1) object MsExperiment XcmsExperiment(): defines number files (samples) full peaks data (m/z intensity values) loaded memory time. Peak detection performed parallel (per sample) subset loaded data. setting thus allows balance memory demand speed (due parallel processing) peak detection. parallel processing can performed subset data loaded currently memory (iteration), value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSizeto total number samples experiment load full MS data memory thus settings cause --memory error. settingchunkSize = -1the peak detection performed separately, parallel, sample. however work allSpectra` backends (see eventually Spectra() details). BPPARAM Parallel processing setup. Uses default system-wide default setup. See bpparam() details. add logical(1) (object contains already chromatographic peaks, .e. either XCMSnExp XcmsExperiment) whether chromatographic peak detection results added existing results. default (add = FALSE) additional findChromPeaks call result object remove previous results.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Chromatographic Peak Detection — findChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":null,"dir":"Reference","previous_headings":"","what":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"findChromPeaksIsolationWindow function allows perform chromatographic peak detection MS level > 1 spectra certain isolation windows (e.g. SWATH pockets). function performs peak detection, separately spectra belonging isolation window adds chromPeaks() matrix result object. Information isolation window detected added chromPeakData() data frame. Note peak detection method remove previously identified chromatographic peaks (e.g. MS1 level using findChromPeaks() function adds newly identified peaks existing chromPeaks() matrix. Isolation windows can defined isolationWindow parameter, default uses definition isolationWindowTargetMz(), .e. chromatographic peak detection performed spectra isolation window target m/z (seprarately file). parameter param allows define configure peak detection algorithm (see findChromPeaks() information).","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"","code":"findChromPeaksIsolationWindow(object, ...)  # S4 method for MsExperiment findChromPeaksIsolationWindow(   object,   param,   msLevel = 2L,   isolationWindow = isolationWindowTargetMz(spectra(object)),   chunkSize = 2L,   ...,   BPPARAM = bpparam() )  # S4 method for OnDiskMSnExp findChromPeaksIsolationWindow(   object,   param,   msLevel = 2L,   isolationWindow = isolationWindowTargetMz(object),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"object MsExperiment, XcmsExperiment, OnDiskMSnExp XCMSnExp object DIA data. ... currently used. param Peak detection parameter object, CentWaveParam object defining configuring chromographic peak detection algorithm. See also findChromPeaks() details. msLevel integer(1) specifying MS level peak detection performed. default msLevel = 2L. isolationWindow factor similar defining isolation windows peak detection performed length equal number spectra object. chunkSize object MsExperiment XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. See findChromPeaks() information. BPPARAM object MsExperiment XcmsExperiment: parallel processing setup. See bpparam() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"XcmsExperiment XCMSnExp object chromatographic peaks identified spectra isolation window file added chromPeaks matrix. Isolation window definition identified peak stored additional columns chromPeakData().","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"Johannes Rainer, Michael Witting","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":null,"dir":"Reference","previous_headings":"","what":"Find values in sorted vectors — findEqualGreater","title":"Find values in sorted vectors — findEqualGreater","text":"Find values sorted vectors.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find values in sorted vectors — findEqualGreater","text":"","code":"findEqualGreater(x, value) findEqualLess(x, value) findEqualGreaterM(x, values) findRange(x, values, NAOK = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find values in sorted vectors — findEqualGreater","text":"x numeric vector sorted increasing order value value find x values numeric values find x NAOK check NA values x","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find values in sorted vectors — findEqualGreater","text":"findEqualGreater finds index first value   x equal greater value. findEqualLess   except finds equal less. findEqualGreaterM   creates index vector finding specified values.   findRange locates start stop indicides range   two x values. things save time point   findeEqualGreaterM (length values approaches   lenght x) findRange (NAOK set   TRUE). run log(N) N time, respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find values in sorted vectors — findEqualGreater","text":"integer vector position(s) values(s).","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find values in sorted vectors — findEqualGreater","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":null,"dir":"Reference","previous_headings":"","what":"Find fragment ions in xcmsFragment objects — findMZ","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"method find fragment mass ppm window xcmsFragment object","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"","code":"findMZ(object, find, ppmE=25, print=TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"object xcmsFragment object type find fragment ion found ppmE ppm error window searching print print nice little report","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"method simply searches given fragment ion xcmsFragment object type given certain ppm error window","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"data frame following columns: PrecursorMz precursor m/z fragment MSnParentPeakID index ID location precursor peak xcmsFragment object msLevel level found fragment ion rt Retention time found ion mz actual m/z found fragment ion intensity intensity fragment ion sample sample fragment ion came GroupPeakMSn ID peaks grouped xcmsSet grouping CollisionEnergy collision energy precursor scan","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"H. Paul Benton, D.M. Wong, S..Strauger, G. Siuzdak \"XC\\(MS^2\\)\"   Analytical Chemistry 2008","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"H. Paul Benton, hpaul.beonton08@imperial.ac.uk","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"","code":"if (FALSE) {     library(msdata)     mzMLpath <- system.file(\"iontrap\", package = \"msdata\")     mzMLfiles<-list.files(mzMLpath, pattern = \"extracted.mzML\",                             recursive = TRUE, full.names = TRUE)     xs <- xcmsSet(mzMLfiles, method = \"MS1\")     ##takes only one file from the file set     xfrag <- xcmsFragments(xs)     found<-findMZ(xfrag, 657.3433, 50) }"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":null,"dir":"Reference","previous_headings":"","what":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"Perform peak detection mass spectrometry     direct injection spectrum using wavelet based algorithm. MSWParam class allows specify     settings peak detection using MSW method. Instances     created MSWParam constructor. findChromPeaks,OnDiskMSnExp,MSWParam     method performs peak detection single-spectrum non-chromatography MS     data using functionality MassSpecWavelet package     samples OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. snthresh,snthresh<-: getter setter     snthresh slot object. verboseColumns,verboseColumns<-: getter setter     verboseColumns slot object. scales,scales<-: getter setter     scales slot object. nearbyPeak,nearbyPeak<-: getter setter     nearbyPeak slot object. peakScaleRange,peakScaleRange<-: getter setter     peakScaleRange slot object. ampTh,ampTh<-: getter setter     ampTh slot object. minNoiseLevel,minNoiseLevel<-: getter setter     minNoiseLevel slot object. ridgeLength,ridgeLength<-: getter setter     ridgeLength slot object. peakThr,peakThr<-: getter setter     peakThr slot object. tuneIn,tuneIn<-: getter setter     tuneIn slot object. addParams,addParams<-: getter setter     addParams slot object. slot stores optional additional     parameters passed     identifyMajorPeaks     peakDetectionCWT functions     MassSpecWavelet package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"","code":"MSWParam(   snthresh = 3,   verboseColumns = FALSE,   scales = c(1, seq(2, 30, 2), seq(32, 64, 4)),   nearbyPeak = TRUE,   peakScaleRange = 5,   ampTh = 0.01,   minNoiseLevel = ampTh/snthresh,   ridgeLength = 24,   peakThr = NULL,   tuneIn = FALSE,   ... )  # S4 method for OnDiskMSnExp,MSWParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for MSWParam snthresh(object)  # S4 method for MSWParam snthresh(object) <- value  # S4 method for MSWParam verboseColumns(object)  # S4 method for MSWParam verboseColumns(object) <- value  # S4 method for MSWParam scales(object)  # S4 method for MSWParam scales(object) <- value  # S4 method for MSWParam nearbyPeak(object)  # S4 method for MSWParam nearbyPeak(object) <- value  # S4 method for MSWParam peakScaleRange(object)  # S4 method for MSWParam peakScaleRange(object) <- value  # S4 method for MSWParam ampTh(object)  # S4 method for MSWParam ampTh(object) <- value  # S4 method for MSWParam minNoiseLevel(object)  # S4 method for MSWParam minNoiseLevel(object) <- value  # S4 method for MSWParam ridgeLength(object)  # S4 method for MSWParam ridgeLength(object) <- value  # S4 method for MSWParam peakThr(object)  # S4 method for MSWParam peakThr(object) <- value  # S4 method for MSWParam tuneIn(object)  # S4 method for MSWParam tuneIn(object) <- value  # S4 method for MSWParam addParams(object)  # S4 method for MSWParam addParams(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"snthresh numeric(1) defining signal noise ratio cutoff. verboseColumns logical(1) whether additional peak meta data columns returned. scales Numeric defining scales continuous wavelet transform (CWT). nearbyPeak logical(1) whether include nearby peaks major peaks. peakScaleRange numeric(1) defining scale range peak (larger 5 default). ampTh numeric(1) defining minimum required relative amplitude peak (ratio maximum CWT coefficients). minNoiseLevel numeric(1) defining minimum noise level used computing SNR. ridgeLength numeric(1) defining minimum highest scale peak 2-D CWT coefficient matrix. peakThr numeric(1) minimum absolute intensity (baseline) peaks picked. provided, smoothing Savitzky-Golay filter used (MassSpecWavelet) package estimate local intensity. tuneIn logical(1) whther tune parameter estimation detected peaks. ... Additional parameters passed peakDetectionCWT identifyMajorPeaks functions MassSpecWavelet package. object findChromPeaks:     OnDiskMSnExp object containing MS-     experiment-relevant data. methods: parameter object. param MSWParam object containing settings algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"MSWParam function returns MSWParam class instance settings specified peak detection     MSW method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"wrapper peak picker Bioconductor's     MassSpecWavelet package calling     peakDetectionCWT     tuneInPeakInfo functions. See     xcmsDirect vignette information. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"snthresh,verboseColumns,scales,nearbyPeak,peakScaleRange,ampTh,minNoiseLevel,ridgeLength,peakThr,tuneIn,addParams See corresponding parameter .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports peak detection     OnDiskMSnExp     objects (defined MSnbase package). settings     algorithm can passed MSWParam object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"Joachim Kutzera, Steffen Neumann, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"","code":"## Create a MSWParam object mp <- MSWParam() ## Change snthresh parameter snthresh(mp) <- 15 mp #> Object of class:  MSWParam  #>  Parameters: #>  - snthresh: [1] 15 #>  - verboseColumns: [1] FALSE #>  - scales:  [1]  1  2  4  6  8 10 12 14 16 18 20 22 24 26 28 30 32 36 40 44 48 52 56 60 64 #>  - nearbyPeak: [1] TRUE #>  - peakScaleRange: [1] 5 #>  - amp.Th: [1] 0.01 #>  - minNoiseLevel: [1] 0.003333333 #>  - ridgeLength: [1] 24 #>  - peakThr: numeric(0) #>  - tuneIn: [1] FALSE  ## Loading a small subset of direct injection, single spectrum files library(msdata) fticrf <- list.files(system.file(\"fticr-mzML\", package = \"msdata\"),                     recursive = TRUE, full.names = TRUE) fticr <- readMSData(fticrf[1], msLevel. = 1, mode = \"onDisk\")  ## Perform the MSW peak detection on these: p <- MSWParam(scales = c(1, 7), peakThr = 80000, ampTh = 0.005,              SNR.method = \"data.mean\", winSize.noise = 500) fticr <- findChromPeaks(fticr, param = p)  head(chromPeaks(fticr)) #>            mz    mzmin    mzmax rt rtmin rtmax    into     maxo        sn intf #> CP01 403.2367 403.2279 403.2447 -1    -1    -1 4735258 372259.4 22.975342   NA #> CP02 409.1845 409.1755 409.1936 -1    -1    -1 4158404 310572.1 20.597978   NA #> CP03 413.2677 413.2585 413.2769 -1    -1    -1 6099006 435462.6 27.217229   NA #> CP04 414.2719 414.2652 414.2786 -1    -1    -1 1916658 157613.1  8.777433   NA #> CP05 423.2363 423.2266 423.2459 -1    -1    -1 2708391 174252.7 14.745275   NA #> CP06 427.2681 427.2583 427.2779 -1    -1    -1 6302089 461385.6 32.468892   NA #>           maxf sample #> CP01  814693.1      1 #> CP02  731557.2      1 #> CP03 1018994.8      1 #> CP04  280826.9      1 #> CP05  435858.5      1 #> CP06 1124549.4      1"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"number peak pickers exist XCMS. findPeaks   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"object = \"xcmsRaw\" findPeaks(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"object xcmsRaw-class object method Method use peak detection. See details. ... Optional arguments passed along","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"Different algorithms can used specifying   method argument. example use matched filter   approach described Smith et al (2006) one use:   findPeaks(object, method=\"matchedFilter\"). also   default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$findPeaks.methods.   nickname method called \"centWave\",   help page specific method can   accessed ?findPeaks.centWave.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z minimum step mzmax m/z maximum step rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated area original (raw) peak maxo maximum intensity original (raw) peak additional columns depending choosen method.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"Collecting Tandem MS MS$^n$ Mass Spectrometry   precursor peaks annotated XML raw file","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"object = \"xcmsRaw\" findPeaks.MS1(object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"mass spectrometers can acquire MS1 MS2 (MS$^n$ scans)   quasi simultaneously, e.g. data dependent tandem MS DDIT mode. Since xcmsFragments attaches MS$^n$ peaks MS1 peaks xcmsSet,   important findPeaks xcmsSet miss MS1 precursor peak. sure MS1 precursor peaks xcmsSet,   findPeaks.MS1 actual peak picking, simply uses   annotation stored mzXML, mzData mzML raw files. relies following XML tags: mzData:      <spectrum id=\"463\">         <spectrumInstrument msLevel=\"2\"> \t  <cvParam cvLabel=\"psi\" accession=\"PSI:1000039\" name=\"TimeInSeconds\" value=\"92.7743\"/>         <\/spectrumInstrument>         <precursor msLevel=\"1\" spectrumRef=\"461\">             <cvParam cvLabel=\"psi\" accession=\"PSI:1000040\" name=\"MassToChargeRatio\" value=\"462.091\"/>             <cvParam cvLabel=\"psi\" accession=\"PSI:1000042\" name=\"Intensity\" value=\"366.674\"/>         <\/precursor>   <\/spectrum> mzXML:         <scan num=\"17\" msLevel=\"2\" retentionTime=\"PT1.5224S\">      <precursorMz precursorIntensity=\"125245\">220.1828003<\/precursorMz>      <\/scan> Several mzXML mzData converters known create incomplete   files, either without intensities (set 0) without   precursor retention time (reasonably close rt   chosen. NYI).","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"object xcmsRaw object","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"matrix columns: mz, mzmin, mzmax annotated MS1 precursor selection mass rt, rtmin, rtmax annotated MS1 precursor retention time , maxo, sn annotated MS1 precursor intensity","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"Steffen Neumann, sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"method performs peak detection mass spectrometry     direct injection spectrum using wavelet based algorithm.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"","code":"# S4 method for xcmsRaw findPeaks.MSW(object, snthresh = 3, verbose.columns = FALSE, ...)"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"object xcmsRaw object peak detection performed. snthresh numeric(1) defining signal noise ratio cutoff. verbose.columns Logical whether additional peak meta data columns returned. ... Additional parameters passed peakDetectionCWT identifyMajorPeaks functions MassSpecWavelet package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"matrix, row representing intentified peak, columns:  mz m/z value peak centroid position. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Always -1. rtmin Always -1. rtmax Always -1. Integrated (original) intensity peak. maxo Maximum intensity peak. intf Always NA. maxf Maximum MSW-filter response peak. sn Signal noise ratio.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"wrapper around peak picker Bioconductor's     MassSpecWavelet package calling     peakDetectionCWT     tuneInPeakInfo functions.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"Joachim Kutzera, Steffen Neumann, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"Peak density wavelet based feature detection aiming isotope peaks high resolution LC/MS data centroid mode","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"object = \"xcmsRaw\" findPeaks.centWave(object, ppm=25, peakwidth=c(20,50),     prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001, fitgauss=FALSE,     scanrange= numeric(), noise=0, sleep=0, verbose.columns=FALSE, xcmsPeaks, snthresh=6.25, maxcharge=3, maxiso=5, mzIntervalExtension=TRUE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method isotope ROIs (regions interest) LC/MS map predicted.   second phase mass traces analysed.   Continuous wavelet transform (CWT) used locate chromatographic peaks different scales.   resulting peak list given peak list (xcmsPeaks) merged redundant peaks removed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"object xcmsSet object ppm maxmial tolerated m/z deviation consecutive scans, ppm (parts per million) peakwidth Chromatographic peak width, given range (min,max) seconds prefilter prefilter=c(k,). Prefilter step first phase. Mass traces retained contain least k peaks intensity >= . mzCenterFun Function calculate m/z center feature: wMean intensity weighted mean feature m/z values, mean mean feature m/z values, apex use m/z value peak apex,         wMeanApex3 intensity weighted mean m/z value peak apex m/z value left right , meanApex3 mean m/z value peak apex m/z value left right . integrate Integration method. =1 peak limits found descent mexican hat filtered data, =2 descent done real data. Method 2 accurate prone noise, method 1 robust noise less exact. mzdiff minimum difference m/z peaks overlapping retention     times, can negative allow overlap fitgauss logical, TRUE Gaussian fitted peak scanrange scan range process noise optional argument useful data centroided without intensity threshold,     centroids intensity < noise omitted ROI detection sleep number seconds pause plotting peak finding cycles verbose.columns logical, TRUE additional peak meta data columns returned xcmsPeaks peak list picked using centWave algorithm parameter verbose.columns set TRUE (columns scmin scmax needed) snthresh signal noise ratio cutoff, definition see . maxcharge max. number isotope charge. maxiso max. number isotope peaks predict detected feature. mzIntervalExtension logical, TRUE predicted isotope ROIs (regions interest) extended m/z dimension increase detection low intensity hence noisy peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z peak minimum mzmax m/z peak maximum rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated peak intensity intb baseline corrected integrated peak intensity maxo maximum peak intensity sn Signal/Noise ratio, defined (maxo - baseline)/sd, maxo maximum peak intensity, baseline estimated baseline value sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit verbose.columns TRUE additionally : mu Gaussian parameter mu sigma Gaussian parameter sigma h Gaussian parameter h f Region number m/z ROI peak localised dppm m/z deviation mass trace across scans ppm scale Scale peak localised scpos Peak position found wavelet analysis scmin Left peak limit found wavelet analysis (scan number) scmax Right peak limit found wavelet analysis (scan number)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"Ralf Tautenhahn","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann   \"Highly sensitive feature detection high resolution LC/MS\"   BMC Bioinformatics 2008, 9:504\\   Hendrik Treutler Steffen Neumann.   \"Prediction, detection, validation isotope clusters mass spectrometry data\"   Submitted Metabolites 2016, Special Issue \"Bioinformatics Data Analysis\"","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"Peak density wavelet based feature detection high resolution LC/MS data centroid mode","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"object = \"xcmsRaw\" findPeaks.centWave(object, ppm=25, peakwidth=c(20,50), snthresh=10,     prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001, fitgauss=FALSE,     scanrange= numeric(), noise=0, sleep=0, verbose.columns=FALSE, ROI.list=list()),     firstBaselineCheck=TRUE, roiScales=NULL","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method mass traces (characterised regions less ppm m/z deviation consecutive scans) LC/MS map located.   second phase mass traces analysed.   Continuous wavelet transform (CWT) used locate chromatographic peaks different scales.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"object xcmsSet object ppm maxmial tolerated m/z deviation consecutive scans, ppm (parts per million) peakwidth Chromatographic peak width, given range (min,max) seconds snthresh signal noise ratio cutoff, definition see . prefilter prefilter=c(k,). Prefilter step first phase. Mass traces retained contain least k peaks intensity >= . mzCenterFun Function calculate m/z center feature: wMean intensity weighted mean feature m/z values, mean mean feature m/z values, apex use m/z value peak apex,         wMeanApex3 intensity weighted mean m/z value peak apex m/z value left right , meanApex3 mean m/z value peak apex m/z value left right . integrate Integration method. =1 peak limits found descent mexican hat filtered data, =2 descent done real data. Method 2 accurate prone noise, method 1 robust noise less exact. mzdiff minimum difference m/z peaks overlapping retention     times, can negative allow overlap fitgauss logical, TRUE Gaussian fitted peak scanrange scan range process noise optional argument useful data centroided without intensity threshold,     centroids intensity < noise omitted ROI detection sleep number seconds pause plotting peak finding cycles verbose.columns logical, TRUE additional peak meta data columns returned ROI.list optional list ROIs represents detected mass traces (ROIs).  list empty (default) centWave detects mass trace ROIs,     otherwise step skipped supplied ROIs used peak detection phase.  ROI object list following slots:     scmin start scan index,     scmax end scan index,     mzmin minimum m/z,     mzmax maximum m/z,     length number scans,     intensity summed intensity. firstBaselineCheck logical, TRUE continuous data within ROI checked 1st baseline roiScales numeric, optional vector scales ROI ROI.list used centWave-wavelets","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z peak minimum mzmax m/z peak maximum rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated peak intensity intb baseline corrected integrated peak intensity maxo maximum peak intensity sn Signal/Noise ratio, defined (maxo - baseline)/sd, maxo maximum peak intensity, baseline estimated baseline value sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit verbose.columns TRUE additionally : mu Gaussian parameter mu sigma Gaussian parameter sigma h Gaussian parameter h f Region number m/z ROI peak localised dppm m/z deviation mass trace across scans ppm scale Scale peak localised scpos Peak position found wavelet analysis scmin Left peak limit found wavelet analysis (scan number) scmax Right peak limit found wavelet analysis (scan number)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"Ralf Tautenhahn","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann   \"Highly sensitive feature detection high resolution LC/MS\"   BMC Bioinformatics 2008, 9:504","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"Peak density wavelet based feature detection high resolution LC/MS data centroid mode additional peak picking isotope features basis isotope peak predictions","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"object = \"xcmsRaw\" findPeaks.centWaveWithPredictedIsotopeROIs(object, ppm=25, peakwidth=c(20,50), snthresh=10,     prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001, fitgauss=FALSE,     scanrange= numeric(), noise=0, sleep=0, verbose.columns=FALSE, ROI.list=list(),     firstBaselineCheck=TRUE, roiScales=NULL, snthreshIsoROIs=6.25, maxcharge=3, maxiso=5, mzIntervalExtension=TRUE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode.   centWave algorithm applied two peak picking steps follows. first peak picking step ROIs (regions interest, characterised regions less ppm m/z deviation consecutive scans) LC/MS map located analysed using continuous wavelet transform (CWT) localization chromatographic peaks different scales.   second peak picking step isotope ROIs LC/MS map predicted analysed using continuous wavelet transform (CWT) localization chromatographic peaks different scales.   peak lists resulting peak picking steps merged redundant peaks removed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"object xcmsSet object ppm maxmial tolerated m/z deviation consecutive scans, ppm (parts per million) peakwidth Chromatographic peak width, given range (min,max) seconds snthresh signal noise ratio cutoff, definition see . prefilter prefilter=c(k,). Prefilter step first phase. Mass traces retained contain least k peaks intensity >= . mzCenterFun Function calculate m/z center feature: wMean intensity weighted mean feature m/z values, mean mean feature m/z values, apex use m/z value peak apex,         wMeanApex3 intensity weighted mean m/z value peak apex m/z value left right , meanApex3 mean m/z value peak apex m/z value left right . integrate Integration method. =1 peak limits found descent mexican hat filtered data, =2 descent done real data. Method 2 accurate prone noise, method 1 robust noise less exact. mzdiff minimum difference m/z peaks overlapping retention     times, can negative allow overlap fitgauss logical, TRUE Gaussian fitted peak scanrange scan range process noise optional argument useful data centroided without intensity threshold,     centroids intensity < noise omitted ROI detection sleep number seconds pause plotting peak finding cycles verbose.columns logical, TRUE additional peak meta data columns returned ROI.list optional list ROIs represents detected mass traces (ROIs).  list empty (default) centWave detects mass trace ROIs,     otherwise step skipped supplied ROIs used peak detection phase.  ROI object list following slots:     scmin start scan index,     scmax end scan index,     mzmin minimum m/z,     mzmax maximum m/z,     length number scans,     intensity summed intensity. firstBaselineCheck logical, TRUE continuous data within ROI checked 1st baseline roiScales numeric, optional vector scales ROI ROI.list used centWave-wavelets snthreshIsoROIs signal noise ratio cutoff predicted isotope ROIs, definition see . maxcharge max. number isotope charge. maxiso max. number isotope peaks predict detected feature. mzIntervalExtension logical, TRUE predicted isotope ROIs (regions interest) extended m/z dimension increase detection low intensity hence noisy peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z peak minimum mzmax m/z peak maximum rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated peak intensity intb baseline corrected integrated peak intensity maxo maximum peak intensity sn Signal/Noise ratio, defined (maxo - baseline)/sd, maxo maximum peak intensity, baseline estimated baseline value sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit verbose.columns TRUE additionally : mu Gaussian parameter mu sigma Gaussian parameter sigma h Gaussian parameter h f Region number m/z ROI peak localised dppm m/z deviation mass trace across scans ppm scale Scale peak localised scpos Peak position found wavelet analysis scmin Left peak limit found wavelet analysis (scan number) scmax Right peak limit found wavelet analysis (scan number)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"Ralf Tautenhahn","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann   \"Highly sensitive feature detection high resolution LC/MS\"   BMC Bioinformatics 2008, 9:504\\   Hendrik Treutler Steffen Neumann.   \"Prediction, detection, validation isotope clusters mass spectrometry data\"   Submitted Metabolites 2016, Special Issue \"Bioinformatics Data Analysis\"","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"Massifquant Kalman filter (KF) based feature detection XC-MS data centroid mode (currently experimental stage). Optionally allows calling method \"centWave\" features discovered Massifquant refine feature detection; , supply additional parameters specific centWave (even experimental). method may conveniently called xcmsSet(...) method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"object = \"xcmsRaw\" findPeaks.massifquant(object, ppm=10, peakwidth=c(20,50), snthresh=10,      prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001,      fitgauss=FALSE, scanrange= numeric(), noise=0,      sleep=0, verbose.columns=FALSE, criticalValue = 1.125, consecMissedLimit = 2,      unions = 1, checkBack = 0, withWave = 0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"algorithm's performance tested rigorously high resolution LC/{OrbiTrap, TOF}-MS data centroid mode. Simultaneous kalman filters identify features calculate area curve. default parameters set operate complex LC-MS Orbitrap sample. Users find useful simple exploratory data analysis find set minimum intensity, identify many scans average feature spans. \"consecMissedLimit\" parameter yielded good performance Orbitrap data set (2) TOF data found best (1). may change algorithm yet tested many samples. \"criticalValue\" parameter perhaps dificult dial appropriately visual inspection peak identification best suggested tool quick optimization.  \"ppm\" \"checkBack\" parameters shown less influence parameters exist give users flexibility better accuracy.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"following arguments specific Massifquant. additional arguments supplied must correspond specified method findPeaks.centWave. object xcmsRaw object. criticalValue Numeric: Suggested values: (0.1-3.0). setting helps determine Kalman Filter prediciton margin error. real centroid belonging bonafide feature must fall within KF prediction margin error. Much like construction confidence interval, criticalVal loosely translates  multiplier standard error prediction reported Kalman Filter. features XC-MS sample small mass deviance ppm error, smaller critical value might better vice versa. consecMissedLimit Integer: Suggested values:(1,2,3). feature proces detected Kalman Filter, Kalman Filter may find predicted centroid every scan. 1 consecutive failed predictions, setting informs Massifquant stop Kalman Filter following candidate feature. prefilter Numeric Vector: (Positive Integer, Positive Numeric): first argument used (withWave = 1); see centWave details. second argument specifies minimum threshold maximum intensity feature must met. peakwidth Integer Vector: (Positive Integer, Positive Integer): first argument used Massifquant, specifices minimum feature length time scans. centWave used, second argument maximum feature length  subject greater mininum feature length. ppm minimum estimated parts per million mass resolution feature must possess. unions Integer: set 1 apply t-test union segmentation; set 0 t-test applied chromatographically continous features sharing m/z range. Explanation: data points, sometimes Kalman Filter stops tracking feature prematurely. Another Kalman Filter instantiated begins following rest signal. tracking done backwards forwards, algorithmic defect leaves real feature divided two segments . option turned , program identifies segmented features combines (merges ) one two sample t-test. potential danger option truly distinct features may merged. withWave Integer: set 1 turned ; set 0 turned . Allows user find features first Massifquant filter features second phase centWave, includes wavelet estimation. checkBack Integer: set 1 turned ; set 0 turned . convergence Kalman Filter feature's precise m/z mapping  fast, sometimes incorporates erroneous centroids part feature (especially early ). \"scanBack\" option attempt remove occasional outlier lies beyond converged bounds Kalman Filter. option directly affect identification feature postprocessing measure; shown extremely useful thus far default set turned .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"method findPeaks.massifquant(...) used, matrix returned rows corresponding features, properties features listed following column names. Otherwise, centWave feature used also (withWave = 1), Massifquant called xcmsSet(...) method, corresponding return values used. mz weighted m/z mean (weighted intensity) feature mzmin m/z  lower boundary feature mzmax m/z  upper boundary feature rtmin starting scan time feature rtmax starting scan time feature raw quantitation (area curve) feature. area feature area normalized scan rate.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"Christopher Conley","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"Submitted review. Christopher Conley, Ralf J .O Torgrip. Ryan Taylor, John T. Prince.  \"Massifquant: open-source Kalman filter based XC-MS feature detection\". August 2013.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"","code":"library(faahKO) library(xcms) #load all the wild type and Knock out samples cdfpath <- system.file(\"cdf\", package = \"faahKO\") ## Subset to only the first 2 files. cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)[1:2]  ## Run the massifquant analysis. Setting the noise level to 10000 to speed up ## execution of the examples - in a real use case it should be set to a reasoable ## value. xset <- xcmsSet(cdffiles, method = \"massifquant\",                 consecMissedLimit = 1,                 snthresh = 10,                 criticalValue = 1.73,                 ppm = 10,                 peakwidth= c(30, 60),                 prefilter= c(1,3000),                 noise = 10000,                 withWave = 0) #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 10ppm ...  #> OK #>  242 Peaks. #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 10ppm ...  #> OK #>  350 Peaks."},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"Find peaks chromatographic time domain     profile matrix. details see     do_findChromPeaks_matchedFilter.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"","code":"# S4 method for xcmsRaw findPeaks.matchedFilter(   object,   fwhm = 30,   sigma = fwhm/2.3548,   max = 5,   snthresh = 10,   step = 0.1,   steps = 2,   mzdiff = 0.8 - step * steps,   index = FALSE,   sleep = 0,   scanrange = numeric() )"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"object xcmsRaw object peak detection performed. fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) representing maximum number peaks expected/identified per slice. snthresh numeric(1) defining signal noise cutoff used chromatographic peak detection step. step numeric(1) specifying width bins/slices m/z dimension. steps numeric(1) defining number bins merged filtration (.e. number neighboring bins joined slice filtration peak detection performed). mzdiff numeric(1) defining minimum difference m/z peaks overlapping retention times index logical(1) specifying whether indicies returned instead values m/z retention times. sleep (DEPRECATED). use parameter highly discouraged, cause problems parallel processing mode. scanrange Numeric vector defining range scans original object sub-setted peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"matrix, row representing intentified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peak across     scans. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intf Integrated intensity filtered peak. maxo Maximum intensity peak. maxf Maximum intensity filtered peak. Rank peak merged EIC (<= max). sn Signal noise ratio peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan     Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite     Profiling Using Nonlinear Peak Alignment, Matching, Identification\"     Anal. Chem. 2006, 78:779-787.     @family Old peak detection methods","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"Colin . Smith","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":null,"dir":"Reference","previous_headings":"","what":"Find neutral losses in xcmsFragment objects — findneutral","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"method find neutral loss ppm window xcmsFragment object","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"","code":"findneutral(object, find, ppmE=25, print=TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"object xcmsFragment object type find neutral loss found ppmE ppm error window searching print print nice little report","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"method searches given neutral loss xcmsFragment object type given certain ppm error window. neutral losses generated neighbouring ions. resulting data frame shows whole scan neutral loss found.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"data frame following columns: PrecursorMz precursor m/z neutral losses MSnParentPeakID index ID location precursor peak xcmsFragment object msLevel level found fragment ion rt Retention time found ion mz actual m/z found fragment ion intensity intensity fragment ion sample sample fragment ion came GroupPeakMSn ID peaks grouped xcmsSet grouping CollisionEnergy collision energy precursor scan","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"H. Paul Benton, D.M. Wong, S..Strauger, G. Siuzdak \"XC\\(MS^2\\)\"   Analytical Chemistry 2008","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"H. Paul Benton, hpbenton@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"","code":"if (FALSE) {     library(msdata)     mzMLpath <- system.file(\"iontrap\", package = \"msdata\")     mzMLfiles<-list.files(mzMLpath, pattern = \"extracted.mzML\",                             recursive = TRUE, full.names = TRUE)     xs <- xcmsSet(mzMLfiles, method = \"MS1\")     ##takes only one file from the file set     xfrag <- xcmsFragments(xs)     found<-findneutral(xfrag, 58.1455, 50) }"},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"Generate multiple extracted ion chromatograms m/z values   interest. xcmsSet objects, reread original raw data   apply precomputed retention time correction, applicable. Note method always return profile, raw   data (profile data binned data along M/Z). See details   information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"object = \"xcmsRaw\" getEIC(object, mzrange, rtrange = NULL, step = 0.1) object = \"xcmsSet\" getEIC(object, mzrange, rtrange = 200, groupidx,                sampleidx = sampnames(object), rt = c(\"corrected\", \"raw\"))","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"object xcmsRaw xcmsSet object mzrange Either two column matrix minimum maximum m/z     matrix dimensions containing columns mzmin     mzmax. specified, method xcmsRaw returns     base peak chromatogram (BPC, .e. intense signal     RT across m/z). xcmsSet objects group data used     mzrange provided. rtrange two column matrix size mzrange minimum     maximum retention times return EIC data     points. specified, method returns chromatogram     full RT range. xcmsSet objects, may also single number     specifying time window around peak return EIC data     points step step (bin) size use profile generation. Note     value step = 0 supported. groupidx either character vector names integer vector     indicies peak groups get EICs sampleidx either character vector names integer vector     indicies samples get EICs rt \"corrected\" using corrected retention times,     \"raw\" using raw retention times","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"contrast rawEIC method, extracts   actual raw values, method extracts object's profile   matrix (provided step argument match   profStep object profile matrix calculated   fly values returned).","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"xcmsSet xcmsRaw objects, xcmsEIC object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get peak intensities for specified regions — getPeaks-methods","title":"Get peak intensities for specified regions — getPeaks-methods","text":"Integrate extracted ion chromatograms pre-defined defined   regions. Return output similar findPeaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get peak intensities for specified regions — getPeaks-methods","text":"object = \"xcmsRaw\" getPeaks(object, peakrange, step = 0.1)","code":""},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get peak intensities for specified regions — getPeaks-methods","text":"object xcmsSet object peakrange matrix data frame 4 columns: mzmin, mzmax,     rtmin, rtmax (must order named) step step size use profile generation","code":""},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get peak intensities for specified regions — getPeaks-methods","text":"matrix columns: rank peak identified merged EIC (<= max), always NA mz weighted (intensity) mean peak m/z across scans mzmin m/z minimum step mzmax m/z maximum step ret retention time peak midpoint retmin leading edge peak retention time retmax trailing edge peak retention time integrated area original (raw) peak intf integrated area filtered peak, always NA maxo maximum intensity original (raw) peak maxf maximum intensity filtered peak, always NA","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get m/z and intensity values for a single mass scan — getScan-methods","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"Return data single mass scan using numeric index   scan reference.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"object = \"xcmsRaw\" getScan(object, scan, mzrange = numeric())   getMsnScan(object, scan, mzrange = numeric())","code":""},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"object xcmsRaw object scan integer index scan. negative, index numbered     end mzrange limit data points returned range,     range(mzrange)","code":""},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"matrix two columns: mz m/z values intensity intensity values","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"Return full-resolution averaged data multiple mass scans.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"object = \"xcmsRaw\" getSpec(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"object xcmsRaw object ... arguments passed profRange used sepecify     spectral segments interest averaging","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"Based mass points spectra selected, master   unique list masses generated. Every spectra interpolated   masses averaged.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"matrix two columns: mz m/z values intensity intensity values","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"Reads raw data applies evential retention time corrections   waters Lock mass correction   returns xcmsRaw object (list xcmsRaw   objects) one files xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"object = \"xcmsSet\" getXcmsRaw(object, sampleidx=1,     profmethod=profinfo(object)$method, profstep=profinfo(object)$step,     rt=c(\"corrected\", \"raw\"), ...   )","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"object xcmsSet object sampleidx index sample raw data     returned. Can single number numeric vector     indices. Alternatively, file name can specified. profmethod profile method. profstep profile step. rt Whether corrected raw retention times returned. ... Additional arguments submitted xcmsRaw function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"single xcmsRaw object list xcmsRaw objects.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"Johannes Rainer, johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Group peaks from different samples together — group-methods","title":"Group peaks from different samples together — group-methods","text":"number grouping (alignment) methods exist XCMS. group   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group peaks from different samples together — group-methods","text":"object = \"xcmsSet\" group(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group peaks from different samples together — group-methods","text":"object xcmsSet-class object method Method use grouping. See details. ... Optional arguments passed along","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Group peaks from different samples together — group-methods","text":"Different algorithms can used specifying   method argument. example use density-based   approach described Smith et al (2006) one use:   group(object, method=\"density\"). also default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$group.methods.   nickname method called \"mzClust\",   help page specific method can   accessed ?group.mzClust.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group peaks from different samples together — group-methods","text":"xcmsSet object peak group assignments statistics.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":null,"dir":"Reference","previous_headings":"","what":"Group peaks from different samples together — group.density","title":"Group peaks from different samples together — group.density","text":"Group peaks together across samples using overlapping m/z bins  calculation smoothed peak distributions chromatographic time.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group peaks from different samples together — group.density","text":"object = \"xcmsSet\" group(object, bw = 30, minfrac = 0.5, minsamp = 1,     mzwid = 0.25, max = 50, sleep = 0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group peaks from different samples together — group.density","text":"object xcmsSet object minfrac minimum fraction samples necessary least one sample     groups valid group minsamp minimum number samples necessary least one sample     groups valid group bw bandwidth (standard deviation half width half maximum)     gaussian smoothing kernel apply peak density     chromatogram mzwid width overlapping m/z slices use creating peak density     chromatograms grouping peaks across samples max maximum number groups identify single m/z slice sleep seconds pause plotting successive steps peak     grouping algorithm. peaks plotted points showing relative     intensity. identified groups flanked dotted vertical     lines.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group peaks from different samples together — group.density","text":"xcmsSet object peak group assignments statistics.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":null,"dir":"Reference","previous_headings":"","what":"Group Peaks via High Resolution Alignment — group.mzClust","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"Runs high resolution alignment single spectra samples stored \tgiven xcmsSet.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"object = \"xcmsSet\" group(object, method=\"mzClust\", mzppm = 20, mzabs = 0, minsamp = 1, minfrac=0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"object xcmsSet peaks mzppm relative error used clustering/grouping ppm (parts per million) mzabs absolute error used clustering/grouping minsamp set minimum number samples one bin minfrac set minimum fraction class one bin","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"Returns xcmsSet slots groups groupindex set.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"Saira . Kazmi, Samiran Ghosh, Dong-Guk Shin, \tDennis W. Hill David F. GrantAlignment high resolution mass spectra: development heuristic \tapproach metabolomics. \tMetabolomics, Vol. 2, . 2, 75-83 (2006)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"","code":"if (FALSE) { library(msdata) mzMLpath <- system.file(\"fticr-mzML\", package = \"msdata\") mzMLfiles <- list.files(mzMLpath, recursive = TRUE, full.names = TRUE)  xs <- xcmsSet(method=\"MSW\", files=mzMLfiles, scales=c(1,7),               SNR.method='data.mean' , winSize.noise=500,               peakThr=80000,  amp.Th=0.005)  xsg <- group(xs, method=\"mzClust\") }"},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":null,"dir":"Reference","previous_headings":"","what":"Group peaks from different samples together — group.nearest","title":"Group peaks from different samples together — group.nearest","text":"Group peaks together across samples creating master peak list  assigning corresponding peaks samples. inspired  alignment algorithm mzMine. details check  http://mzmine.sourceforge.net/ Katajamaa M, Miettinen J, Oresic M: MZmine: Toolbox processing  visualization mass spectrometry based molecular profile  data. Bioinformatics (Oxford, England) 2006, 22:634?636. Currently, equivalent minfrac minsamp.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group peaks from different samples together — group.nearest","text":"object = \"xcmsSet\" group(object, mzVsRTbalance=10, mzCheck=0.2, rtCheck=15, kNN=10)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group peaks from different samples together — group.nearest","text":"object xcmsSet object mzVsRTbalance Multiplicator mz value calculating     (euclidean) distance two peaks. mzCheck Maximum tolerated distance mz. rtCheck Maximum tolerated distance RT. kNN Number nearest Neighbours check","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group peaks from different samples together — group.nearest","text":"xcmsSet object peak group assignments statistics.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Group peaks from different samples together — group.nearest","text":"","code":"if (FALSE) library(xcms)     library(faahKO)     ## These files do not have this problem to correct for     ## but just for an example     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)      xset<-xcmsSet(cdffiles)      gxset<-group(xset, method=\"nearest\") #> Processing sample number 8 ...  #> OK #> Processing sample number 1 ...  #> OK #> Processing sample number 7 ...  #> OK #> Processing sample number 3 ...  #> OK #> Processing sample number 12 ...  #> OK #> Processing sample number 9 ...  #> OK #> Processing sample number 6 ...  #> OK #> Processing sample number 5 ...  #> OK #> Processing sample number 4 ...  #> OK #> Processing sample number 11 ...  #> OK #> Processing sample number 10 ...  #> OK     nrow(gxset@groups) == 1096 ## the number of features before minFrac #> [1] FALSE      post.minFrac<-function(object, minFrac=0.5){         ix.minFrac<-sapply(1:length(unique(sampclass(object))),                            function(x, object, mf){                                meta<-groups(object)                                minFrac.idx<-numeric(length=nrow(meta))                                idx<-which(                                    meta[,levels(sampclass(object))[x]] >=                                    mf*length(which(levels(sampclass(object))[x]                                                    == sampclass(object)) ))                                minFrac.idx[idx]<-1                                return(minFrac.idx)                            }, object, minFrac)         ix.minFrac<-as.logical(apply(ix.minFrac, 1, sum))         ix<-which(ix.minFrac == TRUE)         return(ix)     }      ## using the above function we can get a post processing minFrac     idx<-post.minFrac(gxset)      gxset.post<-gxset ## copy the xcmsSet object     gxset.post@groupidx<-gxset@groupidx[idx]     gxset.post@groups<-gxset@groups[idx,]      nrow(gxset.post@groups) == 465 ## number of features after minFrac #> [1] FALSE"},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"groupChromPeaks method performs correspondence analysis .e., groups chromatographic peaks across samples define LC-MS features. correspondence algorithm can selected, configured, using param argument. See documentation XcmsExperiment() XCMSnExp() information access extract correspondence results. correspondence analysis can performed chromatographic peaks MS level (present chromatographic peak detection performed MS level) defining features combining peaks. MS level can selected parameter msLevel. default, calling groupChromPeaks remove previous correspondence results. can disabled add = TRUE, add newly defined features already present feature definitions. Supported param objects : PeakDensityParam: correspondence using peak density method (Smith 2006) groups chromatographic peaks along retention time axis within slices (partially overlapping) m/z ranges. peaks (different samples) apex position close retention time axis grouped LC-MS feature. See addition do_groupChromPeaks_density() core API function. NearestPeaksParam: performs peak grouping based proximity chromatographic peaks different samples m/z - rt space similar correspondence method mzMine (Katajamaa 2006). method creates first master peak list consisting chromatographic peaks sample detected peaks iteratively calculates distances peaks sample next number peaks grouping peaks together distance smaller provided thresholds. See addition do_groupChromPeaks_nearest() core API function. MzClustParam: performs high resolution peak grouping single spectrum metabolomics data (Kazmi 2006). method used data retention time considered correspondence analysis. See addition do_groupPeaks_mzClust() core API function. specific examples description method settings see help pages individual parameter classes listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"","code":"groupChromPeaks(object, param, ...)  # S4 method for XcmsExperiment,Param groupChromPeaks(object, param, msLevel = 1L, add = FALSE)  PeakDensityParam(   sampleGroups = numeric(),   bw = 30,   minFraction = 0.5,   minSamples = 1,   binSize = 0.25,   maxFeatures = 50 )  MzClustParam(   sampleGroups = numeric(),   ppm = 20,   absMz = 0,   minFraction = 0.5,   minSamples = 1 )  NearestPeaksParam(   sampleGroups = numeric(),   mzVsRtBalance = 10,   absMz = 0.2,   absRt = 15,   kNN = 10 )  # S4 method for PeakDensityParam sampleGroups(object)  # S4 method for PeakDensityParam sampleGroups(object) <- value  # S4 method for PeakDensityParam bw(object)  # S4 method for PeakDensityParam bw(object) <- value  # S4 method for PeakDensityParam minFraction(object)  # S4 method for PeakDensityParam minFraction(object) <- value  # S4 method for PeakDensityParam minSamples(object)  # S4 method for PeakDensityParam minSamples(object) <- value  # S4 method for PeakDensityParam binSize(object)  # S4 method for PeakDensityParam binSize(object) <- value  # S4 method for PeakDensityParam maxFeatures(object)  # S4 method for PeakDensityParam maxFeatures(object) <- value  # S4 method for MzClustParam sampleGroups(object)  # S4 method for MzClustParam sampleGroups(object) <- value  # S4 method for MzClustParam ppm(object)  # S4 method for MzClustParam ppm(object) <- value  # S4 method for MzClustParam absMz(object)  # S4 method for MzClustParam absMz(object) <- value  # S4 method for MzClustParam minFraction(object)  # S4 method for MzClustParam minFraction(object) <- value  # S4 method for MzClustParam minSamples(object)  # S4 method for MzClustParam minSamples(object) <- value  # S4 method for NearestPeaksParam sampleGroups(object)  # S4 method for NearestPeaksParam sampleGroups(object) <- value  # S4 method for NearestPeaksParam mzVsRtBalance(object)  # S4 method for NearestPeaksParam mzVsRtBalance(object) <- value  # S4 method for NearestPeaksParam absMz(object)  # S4 method for NearestPeaksParam absMz(object) <- value  # S4 method for NearestPeaksParam absRt(object)  # S4 method for NearestPeaksParam absRt(object) <- value  # S4 method for NearestPeaksParam kNN(object)  # S4 method for NearestPeaksParam kNN(object) <- value  # S4 method for XCMSnExp,PeakDensityParam groupChromPeaks(object, param, msLevel = 1L, add = FALSE)  # S4 method for XCMSnExp,MzClustParam groupChromPeaks(object, param, msLevel = 1L)  # S4 method for XCMSnExp,NearestPeaksParam groupChromPeaks(object, param, msLevel = 1L, add = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"object data object correspondence analysis performed. Can XCMSnExp(), XcmsExperiment() object. param parameter object selecting configuring algorithm. ... Optional parameters. msLevel integer(1) defining MS level chromatographic peak detection performed. add logical(1) (object contains already chromatographic peaks, .e. either XCMSnExp XcmsExperiment) whether chromatographic peak detection results added existing results. default (add = FALSE) additional findChromPeaks call result object remove previous results. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). bw PeakDensityParam: numeric(1) defining bandwidth (standard deviation ot smoothing kernel) used. argument passed [density() method. minFraction PeakDensityParam: numeric(1) defining minimum fraction samples least one sample group peaks present considered peak group (feature). minSamples PeakDensityParam: numeric(1) minimum number samples least one sample group peaks detected considered peak group (feature). binSize PeakDensityParam: numeric(1) defining size overlapping slices m/z dimension. maxFeatures PeakDensityParam: numeric(1) maximum number peak groups identified single mz slice. ppm MzClustParam: numeric(1) representing relative m/z error clustering/grouping (parts per million). absMz NearestPeaksParam MzClustParam: numeric(1) maximum tolerated distance m/z values. mzVsRtBalance NearestPeaksParam: numeric(1) representing factor m/z values multiplied calculating (euclician) distance two peaks. absRt NearestPeaksParam: numeric(1) maximum tolerated distance retention times. kNN NearestPeaksParam: integer(1) representing number nearest neighbors check. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"groupChromPeaks: either XcmsExperiment() XCMSnExp() object correspondence result.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"Smith, C.., Want E.J., O'Maille G., Abagyan R., Siuzdak G. (2006) \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 78:779-787. Katajamaa, M., Miettinen, J., Oresic, M. (2006) \"MZmine: Toolbox processing visualization mass spectrometry based molecular profile data\". Bioinformatics, 22:634-636. Kazmi S. ., Ghosh, S., Shin, D., Hill, D.W., Grant, D.F. (2006) \"Alignment high resolution mass spectra: development heuristic approach metabolomics. Metabolomics Vol. 2, . 2, 75-83.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"Features originating compound expected similar intensities across samples. method thus groups features based similarity abundances (.e. feature values) across samples data set. See also AbundanceSimilarityParam() additional information details. help page lists parameters specific xcms result objects (.e. XcmsExperiment() XCMSnExp() objects). Documentation parameters similarity calculation available AbundanceSimilarityParam() help page MsFeatures package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"","code":"# S4 method for XcmsResult,AbundanceSimilarityParam groupFeatures(   object,   param,   msLevel = 1L,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   filled = TRUE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"object XcmsExperiment() XCMSnExp() object containing LC-MS pre-processing results. param AbudanceSimilarityParam object settings method. See AbundanceSimilarityParam() details grouping method parameters. msLevel integer(1) defining MS level features grouped. method character(1) passed featureValues call. See featureValues() details. Defaults method = \"medret\". value character(1) passed featureValues call. See featureValues() details. Defaults value = \"\". intensity character(1) passed featureValues call. See featureValues() details. Defaults intensity = \"\". filled logical(1) whether filled-values included correlation analysis. Defaults filled = TRUE. ... additional parameters passed groupFeatures method matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"input object feature group definitions added (updated ) column \"feature_group\" featureDefinitions data frame.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"","code":"library(MsFeatures) ## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Group chromatographic peaks across samples xodg <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = rep(1, 3)))  ## Group features based on correlation of feature values (integrated ## peak area) across samples. Note that there are many missing values ## in the feature value which influence grouping of features in the present ## data set. xodg_grp <- groupFeatures(xodg,     param = AbundanceSimilarityParam(threshold = 0.8)) table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008  #>      8      8      8      6     12      3      1      1   ## Group based on the maximal peak intensity per feature xodg_grp <- groupFeatures(xodg,     param = AbundanceSimilarityParam(threshold = 0.8, value = \"maxo\")) table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008  #>      8      8      8      6     12      3      1      1"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"Features originating compound expected share elution pattern (.e. chromatographic peak shape) . Thus, methods allows group features based similarity extracted ion chromatograms (EICs). similarity calculation performed separately sample similarity score aggregated across samples final generation similarity matrix grouping (considering parameter threshold) performed. compareChromatograms() function used similarity calculation default calculates Pearson's correlation coefficient. settings compareChromatograms can specified parameters ALIGNFUN, ALIGNFUNARGS, FUN FUNARGS. ALIGNFUN defaults alignRt() function used align chromatograms comparison. ALIGNFUNARGS allows specify additional arguments ALIGNFUN function. defaults ALIGNFUNARGS = list(tolerance = 0, method = \"closest\") ensures data points spectrum (scan, .e. retention time) compared EICs sample. Parameter FUN defines function calculate similarity score defaults FUN = cor FUNARGS allows pass additional arguments function (defaults FUNARGS = list(use = \"pairwise.complete.obs\"). See also compareChromatograms() information. grouping features based EIC similarity matrix performed function specified parameter groupFun defaults groupFun = groupSimilarityMatrix groups rows (features) similarity matrix similarity score larger threshold cluster. creates clusters features features similarity score >= threshold feature cluster. See groupSimilarityMatrix() details. Additional parameters function can passed ... argument. feature grouping called initial feature grouping retention time (see SimilarRtimeParam()). feature groups defined columns \"feature_group\" featureDefinitions(object) (features matching msLevel) used refined method. Features value NA featureDefinitions(object)$feature_group skipped/considered feature grouping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"","code":"EicSimilarityParam(   threshold = 0.9,   n = 1,   onlyPeak = TRUE,   value = c(\"maxo\", \"into\"),   groupFun = groupSimilarityMatrix,   ALIGNFUN = alignRt,   ALIGNFUNARGS = list(tolerance = 0, method = \"closest\"),   FUN = cor,   FUNARGS = list(use = \"pairwise.complete.obs\"),   ... )  # S4 method for XcmsResult,EicSimilarityParam groupFeatures(object, param, msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"threshold numeric(1) minimal required similarity score group featues. passed groupFun function. n numeric(1) defining total number samples per feature group similarity calculation performed. value rounded next larger integer value. onlyPeak logical(1) whether correlation performed signals within identified chromatographic peaks (onlyPeak = TRUE, default) signal extracted ion chromatogram. value character(1) defining whether samples grouped based sum maximal peak intensity (value = \"maxo\", default) integrated peak area (value = \"\") feature. groupFun function defining function used group rows based pairwise similarity matrix. Defaults groupSimilarityMatrix(). ALIGNFUN function defining function used align chromatograms prior similarity calculation. Defaults ALIGNFUN = alignRt. See alignRt() compareChromatograms() information. ALIGNFUNARGS named list arguments ALIGNFUN. Defaults ALIGNFUNARGS = list(tolerance = 0, method = \"closest\"). FUN function defining function used calculate similarity (aligned) chromatograms. Defaults FUN = cor. See cor() compareChromatograms() information. FUNARGS named list arguments FUN. Defaults FUN = list(use = \"pairwise.complete.obs\"). ... EicSimilarityParam: additional arguments passed groupFun featureChromatograms (expandRt expand retention time range feature). object XcmsExperiment() XCMSnExp() object LC-MS pre-processing results. param EicSimilarityParam object settings method. msLevel integer(1) defining MS level features grouped.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"input object feature groups added (.e. column \"feature_group\" featureDefinitions data frame.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"possible performed full data set without prior feature grouping, suggested following reasons: ) selection top n samples highest signal feature group biased abundant compounds performed full data set (.e. samples highest overall intensities used correlation features) II) computationally much expensive pairwise correlation features performed. also suggested perform correlation subset samples per feature highest intensities peaks (feature) although also possible run correlation samples setting n equal total number samples data set. EIC correlation however performed ideally samples original compound highly abundant avoid correlation missing values noisy peak shapes much possible. default also signal outside identified chromatographic peaks excluded correlation.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"","code":"library(MsFeatures) ## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Group chromatographic peaks across samples xodg <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = rep(1, 3)))  ## Performing a feature grouping based on EIC similarities on a single ## sample xodg_grp <- groupFeatures(xodg, param = EicSimilarityParam(n = 1))  table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  #>      2      2      2      4      3      2      3      2      1      1      1  #> FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.034 FG.035  #>      1      1   ## Usually it is better to perform this correlation on pre-grouped features ## e.g. based on similar retention time. xodg_grp <- groupFeatures(xodg, param = SimilarRtimeParam(diffRt = 4)) xodg_grp <- groupFeatures(xodg_grp, param = EicSimilarityParam(n = 1))  table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001.001 FG.002.001 FG.003.001 FG.003.002 FG.004.001 FG.004.002 FG.005.001  #>          2          2          2          1          3          1          3  #> FG.006.001 FG.006.002 FG.007.001 FG.008.001 FG.009.001 FG.009.002 FG.010.001  #>          1          1          2          2          1          1          2  #> FG.011.001 FG.012.001 FG.013.001 FG.014.001 FG.015.001 FG.016.001 FG.017.001  #>          1          1          1          1          1          1          1  #> FG.018.001 FG.019.001 FG.020.001 FG.021.001 FG.022.001 FG.023.001 FG.024.001  #>          1          1          1          1          1          1          1  #> FG.025.001 FG.026.001 FG.027.001 FG.028.001 FG.029.001 FG.030.001 FG.031.001  #>          1          1          1          1          1          1          1  #> FG.032.001 FG.033.001  #>          1          1"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"Group features based similar retention time. method supposed used initial crude grouping features based median retention time chromatographic peaks. features difference retention time <= parameter diffRt parameter object grouped together. column \"feature_group\" found featureDefinitions() sub-grouped method. See MsFeatures::SimilarRtimeParam() MsFeatures details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"","code":"# S4 method for XcmsResult,SimilarRtimeParam groupFeatures(object, param, msLevel = 1L, ...)"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"object XcmsExperiment() XCMSnExp() object containing also correspondence results. param SimilarRtimeParam object settings method. See MsFeatures::SimilarRtimeParam() details options. msLevel integer(1) defining MS level features grouped. ... passed groupFeatures function numeric values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"input object feature groups added (.e. column \"feature_group\" featureDefinitions data frame.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"","code":"library(MsFeatures) ## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Group chromatographic peaks across samples xodg <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = rep(1, 3)))  ## Group features based on similar retention time (i.e. difference <= 2 seconds) xodg_grp <- groupFeatures(xodg, param = SimilarRtimeParam(diffRt = 2))  ## Feature grouping get added to the featureDefinitions in column \"feature_group\" head(featureDefinitions(xodg_grp)$feature_group) #> [1] \"FG.004\" \"FG.009\" \"FG.003\" \"FG.004\" \"FG.010\" \"FG.011\"  table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  #>      2      2      3      3      3      2      2      2      1      1      1  #> FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.034 FG.035 FG.036  #>      1      1      1  length(unique(featureDefinitions(xodg_grp)$feature_group)) #> [1] 36  ## Using an alternative groupiing method that creates larger groups xodg_grp <- groupFeatures(xodg,     param = SimilarRtimeParam(diffRt = 2, groupFun = MsCoreUtils::group))  length(unique(featureDefinitions(xodg_grp)$feature_group)) #> [1] 35"},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":null,"dir":"Reference","previous_headings":"","what":"Group overlapping ranges — groupOverlaps","title":"Group overlapping ranges — groupOverlaps","text":"groupOverlaps identifies overlapping ranges input data groups returning indices xmin xmax.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Group overlapping ranges — groupOverlaps","text":"","code":"groupOverlaps(xmin, xmax)"},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group overlapping ranges — groupOverlaps","text":"xmin numeric (length xmax) lower boundary range. xmax numeric (length xmin) upper boundary range.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group overlapping ranges — groupOverlaps","text":"list indices grouped elements.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Group overlapping ranges — groupOverlaps","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Group overlapping ranges — groupOverlaps","text":"","code":"x <- c(2, 12, 34.2, 12.4) y <- c(3, 16, 35, 36)  groupOverlaps(x, y) #> [[1]] #> [1] 1 #>  #> [[2]] #> [1] 2 3 4 #>"},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"groupnames generates names identified features correspondence analysis based mass retention time. generates feature names equivalent group names old user interface (aka xcms1).","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp groupnames(object, mzdec = 0, rtdec = 0, template = NULL)"},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"object XCMSnExp object containing correspondence results. mzdec integer(1) number decimal places use m/z ( defaults 0). rtdec integer(1) number decimal places use retention time (defaults 0). template character existing group names whose format emulated.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"character unique names feature object. format M(m/z)T(time seconds).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate unque names for peak groups — groupnames-methods","title":"Generate unque names for peak groups — groupnames-methods","text":"Allow linking peak group data classes using unique   group names remain long re-grouping occurs.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Generate unque names for peak groups — groupnames-methods","text":"object = \"xcmsSet\" (object, mzdec = 0, rtdec = 0, template = NULL) object = \"xcmsEIC\" (object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate unque names for peak groups — groupnames-methods","text":"object xcmsSet xcmsEIC object mzdec number decimal places use m/z rtdec number decimal places use retention time template character vector existing group names whose format     emulated","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate unque names for peak groups — groupnames-methods","text":"character vector unique names peak group   object. format M[m/z]T[time seconds].","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract a matrix of peak values for each group — groupval-methods","title":"Extract a matrix of peak values for each group — groupval-methods","text":"Generate matrix peak values rows every group   columns every sample. value included matrix can   columns xcmsSet peaks slot   matrix. Collisions one peak single sample   group get resolved one several user-selectable   methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Extract a matrix of peak values for each group — groupval-methods","text":"object = \"xcmsSet\" groupval(object, method = c(\"medret\", \"maxint\"),                  value = \"index\", intensity = \"\")","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract a matrix of peak values for each group — groupval-methods","text":"object xcmsSet object method conflict resolution method, \"medret\" use peak     closest median retention time \"maxint\" use     peak highest intensity value name peak column enter returned matrix, \"index\"     index corresponding row peaks slot     matrix intensity method == \"maxint\", name peak column use     intensity","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract a matrix of peak values for each group — groupval-methods","text":"matrix rows every group columns every   sample. Missing peaks NA values.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"highlightChromPeaks function adds chromatographic peak definitions existing plot, one created plot method Chromatogram MChromatograms object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"","code":"highlightChromPeaks(   x,   rt,   mz,   peakIds = character(),   border = rep(\"00000040\", length(fileNames(x))),   lwd = 1,   col = NA,   type = c(\"rect\", \"point\", \"polygon\"),   whichPeaks = c(\"any\", \"within\", \"apex_within\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"x highlightChromPeaks: XCMSnExp object detected peaks. rt highlightChromPeaks: numeric(2) retention time range peaks extracted plotted. mz numeric(2) mz range peaks extracted plotted. peakIds character defining IDs (.e. rownames peak chromPeaks table) chromatographic peaks highlighted plot. border colors used color border rectangles/peaks. equal number samples x. lwd numeric(1) defining width line/border. col highlightChromPeaks: color used fill rectangle (type = \"rect\") peak (type = \"polygon\"). type plotting type. See plot base grapics details. highlightChromPeaks: character(1) defining peak highlighted: type = \"rect\" draws rectangle representing peak definition, type = \"point\" indicates chromatographic peak single point position peak's \"rt\" \"maxo\" type = \"polygon\" highlight peak shape. type = \"polygon\" color border area can defined parameters \"border\" \"col\", respectively. whichPeaks character(1) specifying peaks called located within region defined mz rt. Can one \"\", \"within\", \"apex_within\" peaks even partially overlapping region, peaks completely within region, peaks apex within region. parameter passed type argument chromPeaks function. See related documentation information examples. ... additional parameters matplot plot function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the ion chromatogram for one chromatographic peak in the data. chrs <- chromatogram(faahko_sub, rt = c(2700, 2900), mz = 335)  plot(chrs)  ## Extract chromatographic peaks for the mz/rt range (if any). chromPeaks(faahko_sub, rt = c(2700, 2900), mz = 335) #>        mz mzmin mzmax       rt   rtmin    rtmax    into    intb  maxo sn sample #> CP095 335   335   335 2786.199 2764.29 2812.803 1496244 1451652 58736 55      2  ## Highlight the chromatographic peaks in the area ## Show the peak definition with a rectangle highlightChromPeaks(faahko_sub, rt = c(2700, 2900), mz = 335)  ## Color the actual peak highlightChromPeaks(faahko_sub, rt = c(2700, 2900), mz = 335,     col = c(\"#ff000020\", \"#00ff0020\"), type = \"polygon\")"},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot log intensity image of a xcmsRaw object — image-methods","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"Create log intensity false-color image xcmsRaw object plotted   m/z retention time axes","code":""},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"x = \"xcmsRaw\" image(x, col = rainbow(256), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"x xcmsRaw object col vector colors use image ... arguments profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":null,"dir":"Reference","previous_headings":"","what":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"function provides missing value imputation based linear     interpolation resembles functionality     profBinLin profBinLinBase functions deprecated     version 1.51 .","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"","code":"imputeLinInterpol(   x,   baseValue,   method = \"lin\",   distance = 1L,   noInterpolAtEnds = FALSE )"},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"x numeric vector eventual missing (NA) values. baseValue base value empty elements set. considered method = \"linbase\" corresponds profBinLinBase's baselevel argument. method One \"none\", \"lin\" \"linbase\". distance method = \"linbase\": number non-empty neighboring element empty element considered linear interpolation. See details section information. noInterpolAtEnds method = \"lin\": Logical indicating whether linear interpolation also performed ends data vector (.e. missing values present beginning end vector).","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"numeric vector empty values imputed based selected  method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"Values NAs input vector x can imputed using methods     \"lin\" \"linbase\": impute = \"lin\" uses simple linear imputation derive value     empty element input vector x neighboring     non-empty elements. method equivalent linear     interpolation profBinLin method. Whether interpolation     performed missing values present beginning end     x can set argument noInterpolAtEnds. default     interpolation also performed ends interpolating 0     beginning towards 0 end.     noInterpolAtEnds = TRUE interpolation performed     ends replacing missing values beginning /end     x 0. impute = \"linbase\" uses linear interpolation impute values     empty elements within user-definable proximity non-empty elements     setting element's value baseValue otherwise.     default baseValue half smallest value     x (NAs removed). Whether linear interpolation based     imputation performed missing value depends     distance argument. Interpolation performed one     next distance closest neighbors current empty element     value NA. interpolation takes place     distance = 0, distance = 1 means value     empty element interpolated directly adjacent non-empty     elements , next neighbors current empty element     also NA, vale set baseValue.     corresponds linear interpolation performed     profBinLinBase method. details see examples .","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"","code":"####### ## Impute missing values by linearly interpolating from neighboring ## non-empty elements x <- c(3, NA, 1, 2, NA, NA, 4, NA, NA, NA, 3, NA, NA, NA, NA, 2) imputeLinInterpol(x, method = \"lin\") #>  [1] 3.000000 2.000000 1.000000 2.000000 2.666667 3.333333 4.000000 3.750000 #>  [9] 3.500000 3.250000 3.000000 2.800000 2.600000 2.400000 2.200000 2.000000 ## visualize the interpolation: plot(x = 1:length(x), y = x) points(x = 1:length(x), y = imputeLinInterpol(x, method = \"lin\"), type = \"l\", col = \"grey\")   ## If the first or last elements are NA, interpolation is performed from 0 ## to the first non-empty element. x <- c(NA, 2, 1, 4, NA) imputeLinInterpol(x, method = \"lin\") #> [1] 1 2 1 4 2 ## visualize the interpolation: plot(x = 1:length(x), y = x) points(x = 1:length(x), y = imputeLinInterpol(x, method = \"lin\"), type = \"l\", col = \"grey\")   ## If noInterpolAtEnds is TRUE no interpolation is performed at both ends imputeLinInterpol(x, method = \"lin\", noInterpolAtEnds = TRUE) #> [1] 0 2 1 4 0  ###### ## method = \"linbase\" ## \"linbase\" performs imputation by interpolation for empty elements based on ## 'distance' adjacent non-empty elements, setting all remaining empty elements ## to the baseValue x <- c(3, NA, 1, 2, NA, NA, 4, NA, NA, NA, 3, NA, NA, NA, NA, 2) ## Setting distance = 0 skips imputation by linear interpolation imputeLinInterpol(x, method = \"linbase\", distance = 0) #>  [1] 3.0 0.5 1.0 2.0 0.5 0.5 4.0 0.5 0.5 0.5 3.0 0.5 0.5 0.5 0.5 2.0  ## With distance = 1 for all empty elements next to a non-empty element the value ## is imputed by linear interpolation. xInt <- imputeLinInterpol(x, method = \"linbase\", distance = 1L) xInt #>  [1] 3.000000 2.000000 1.000000 2.000000 2.666667 3.333333 4.000000 2.250000 #>  [9] 0.500000 1.750000 3.000000 1.750000 0.500000 0.500000 1.250000 2.000000  plot(x = 1:length(x), y = x, ylim = c(0, max(x, na.rm = TRUE))) points(x = 1:length(x), y = xInt, type = \"l\", col = \"grey\")   ## Setting distance = 2L would cause that for all empty elements for which the ## distance to the next non-empty element is <= 2 the value is imputed by ## linear interpolation: xInt <- imputeLinInterpol(x, method = \"linbase\", distance = 2L) xInt #>  [1] 3.000000 2.000000 1.000000 2.000000 2.666667 3.333333 4.000000 3.750000 #>  [9] 3.500000 3.250000 3.000000 2.800000 2.600000 2.400000 2.200000 2.000000  plot(x = 1:length(x), y = x, ylim = c(0, max(x, na.rm = TRUE))) points(x = 1:length(x), y = xInt, type = \"l\", col = \"grey\")"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":null,"dir":"Reference","previous_headings":"","what":"Replace missing values with a proportion of the row minimum — imputeRowMin","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"imputeRowMin imputes missing values x replacing NAs row proportion minimal value row (.e. min_fraction * min(x[, ])).","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"","code":"imputeRowMin(x, min_fraction = 1/2)"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"x matrix abundances, rows features/metabolites columns samples. min_fraction numeric(1) fraction row minimum used replace NA values row.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"","code":"library(faahKO) data(\"faahko\")  xset <- group(faahko) mat <- groupval(xset, value = \"into\")  mat_imp <- imputeRowMin(mat)  head(mat) #>                 ko15      ko16       ko18      ko19       ko21      ko22 #> 200.1/2926  147887.5  451600.7   65290.38        NA   91635.45  162012.4 #> 205/2791   1778568.9 1567038.1 1482796.38 1039129.8 1223132.35 1072037.7 #> 206/2790    237993.6  269714.0  201393.42  150107.3  176989.65  156797.0 #> 207.1/2719  380873.0  460629.7  351750.14  219288.0  286848.56  235022.6 #> 219.1/2525  235544.9  173623.4         NA        NA  185792.43  174458.8 #> 231/2517          NA        NA  222609.07  286232.1  435094.49        NA #>                 wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.1   82619.48         NA   69198.22  153273.5   98144.28 #> 205/2791   1950287.5 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.8  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.6  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.5  161184.05   72029.38         NA  238194.4  173829.95 #> 231/2517          NA         NA         NA  240261.21  201316.2  179437.72 head(mat_imp) #>                  ko15       ko16       ko18       ko19       ko21       ko22 #> 200.1/2926  147887.53  451600.71   65290.38   32645.19   91635.45  162012.44 #> 205/2791   1778568.94 1567038.14 1482796.38 1039129.82 1223132.35 1072037.70 #> 206/2790    237993.62  269713.98  201393.42  150107.31  176989.65  156797.04 #> 207.1/2719  380873.05  460629.74  351750.14  219287.97  286848.56  235022.63 #> 219.1/2525  235544.92  173623.38   36014.69   36014.69  185792.43  174458.77 #> 231/2517     89718.86   89718.86  222609.07  286232.15  435094.49   89718.86 #>                  wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.08   82619.48   32645.19   69198.22  153273.5   98144.28 #> 205/2791   1950287.49 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.85  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.58  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.47  161184.05   72029.38   36014.69  238194.4  173829.95 #> 231/2517     89718.86   89718.86   89718.86  240261.21  201316.2  179437.72  ## Replace with 1/8 of the row mimimum head(imputeRowMin(mat, min_fraction = 1/8)) #>                  ko15       ko16        ko18        ko19       ko21       ko22 #> 200.1/2926  147887.53  451600.71   65290.385    8161.298   91635.45  162012.44 #> 205/2791   1778568.94 1567038.14 1482796.385 1039129.818 1223132.35 1072037.70 #> 206/2790    237993.62  269713.98  201393.415  150107.310  176989.65  156797.04 #> 207.1/2719  380873.05  460629.74  351750.138  219287.968  286848.56  235022.63 #> 219.1/2525  235544.92  173623.38    9003.672    9003.672  185792.43  174458.77 #> 231/2517     22429.72   22429.72  222609.068  286232.146  435094.49   22429.72 #>                  wt15       wt16        wt18        wt19      wt21       wt22 #> 200.1/2926  175177.08   82619.48    8161.298   69198.222  153273.5   98144.28 #> 205/2791   1950287.49 1466780.60 1572679.161 1275312.758 1356014.3 1231442.16 #> 206/2790    276541.85  222366.15  211717.713  186850.878  188285.9  172348.76 #> 207.1/2719  417169.58  324892.46  277990.701  220972.352  252874.0  236728.16 #> 219.1/2525  244584.47  161184.05   72029.378    9003.672  238194.4  173829.95 #> 231/2517     22429.72   22429.72   22429.715  240261.212  201316.2  179437.72"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":null,"dir":"Reference","previous_headings":"","what":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"Replace missing values random numbers. using method = \"mean_sd\", random numbers generated normal distribution based (fraction ) row min standard deviation estimated linear relationship row standard deviation mean full data set. Parameter sd_fraction allows reduce estimated standard deviation. using method method = \"from_to\", random numbers 2 specific values generated.","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"","code":"imputeRowMinRand(   x,   method = c(\"mean_sd\", \"from_to\"),   min_fraction = 1/2,   min_fraction_from = 1/1000,   sd_fraction = 1,   abs = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"x matrix abundances, rows features/metabolites columns samples. method method character(1) defining imputation method. See description details. Defaults method = \"mean_sd\". min_fraction numeric(1) fraction row minimum used replace NA values row case mean_sd method specified.  using from_to method, value one used calculate maximum value replace NA values row. min_fraction_from numeric(1) fraction row minimum used calculate minimum value replace NA values row. parameter used case from_to method specified. sd_fraction numeric(1) factor reduce estimated standard deviation. parameter used case mean_sd method specified. abs logical(1) force imputed values strictly positive.","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"method mean_sd, imputed values taken normal distribution mean user defined fraction row minimum standard deviation estimated mean based linear relationship row standard deviations row means full matrix x. largely avoid imputed values negative larger real values, standard deviation random number generation estimated ignoring intercept linear model estimating relationship standard deviation mean. abs = TRUE NA values replaced absolute value random values. method from_to, imputed values taken 2 user defined fractions row minimum.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"Johannes Rainer, Mar Garcia-Aloy","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"","code":"library(faahKO) data(\"faahko\")  xset <- group(faahko) mat <- groupval(xset, value = \"into\")  ## Estimate the relationship between row sd and mean. The standard deviation ## of the random distribution is estimated on this relationship. mns <- rowMeans(mat, na.rm = TRUE) sds <- apply(mat, MARGIN = 1, sd, na.rm = TRUE) plot(mns, sds) abline(lm(sds ~ mns))   mat_imp_meansd <- imputeRowMinRand(mat, method = \"mean_sd\") mat_imp_fromto <- imputeRowMinRand(mat, method = \"from_to\")  head(mat) #>                 ko15      ko16       ko18      ko19       ko21      ko22 #> 200.1/2926  147887.5  451600.7   65290.38        NA   91635.45  162012.4 #> 205/2791   1778568.9 1567038.1 1482796.38 1039129.8 1223132.35 1072037.7 #> 206/2790    237993.6  269714.0  201393.42  150107.3  176989.65  156797.0 #> 207.1/2719  380873.0  460629.7  351750.14  219288.0  286848.56  235022.6 #> 219.1/2525  235544.9  173623.4         NA        NA  185792.43  174458.8 #> 231/2517          NA        NA  222609.07  286232.1  435094.49        NA #>                 wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.1   82619.48         NA   69198.22  153273.5   98144.28 #> 205/2791   1950287.5 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.8  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.6  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.5  161184.05   72029.38         NA  238194.4  173829.95 #> 231/2517          NA         NA         NA  240261.21  201316.2  179437.72 head(mat_imp_meansd) #>                 ko15      ko16       ko18       ko19       ko21       ko22 #> 200.1/2926  147887.5  451600.7   65290.38   40787.93   91635.45  162012.44 #> 205/2791   1778568.9 1567038.1 1482796.38 1039129.82 1223132.35 1072037.70 #> 206/2790    237993.6  269714.0  201393.42  150107.31  176989.65  156797.04 #> 207.1/2719  380873.0  460629.7  351750.14  219287.97  286848.56  235022.63 #> 219.1/2525  235544.9  173623.4   35882.08   48149.57  185792.43  174458.77 #> 231/2517    140725.6   84084.1  222609.07  286232.15  435094.49   81878.46 #>                 wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.1   82619.48   34637.57   69198.22  153273.5   98144.28 #> 205/2791   1950287.5 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.8  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.6  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.5  161184.05   72029.38   28146.13  238194.4  173829.95 #> 231/2517    122405.0   86251.20   48322.54  240261.21  201316.2  179437.72 head(mat_imp_fromto) #>                  ko15       ko16       ko18        ko19       ko21       ko22 #> 200.1/2926  147887.53  451600.71   65290.38    2778.231   91635.45  162012.44 #> 205/2791   1778568.94 1567038.14 1482796.38 1039129.818 1223132.35 1072037.70 #> 206/2790    237993.62  269713.98  201393.42  150107.310  176989.65  156797.04 #> 207.1/2719  380873.05  460629.74  351750.14  219287.968  286848.56  235022.63 #> 219.1/2525  235544.92  173623.38   32793.36    2387.531  185792.43  174458.77 #> 231/2517     38276.79   37741.04  222609.07  286232.146  435094.49   46964.75 #>                  wt15       wt16       wt18        wt19      wt21       wt22 #> 200.1/2926  175177.08   82619.48   31192.57   69198.222  153273.5   98144.28 #> 205/2791   1950287.49 1466780.60 1572679.16 1275312.758 1356014.3 1231442.16 #> 206/2790    276541.85  222366.15  211717.71  186850.878  188285.9  172348.76 #> 207.1/2719  417169.58  324892.46  277990.70  220972.352  252874.0  236728.16 #> 219.1/2525  244584.47  161184.05   72029.38    4217.932  238194.4  173829.95 #> 231/2517     44101.44   78784.93   14870.97  240261.212  201316.2  179437.72"},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"isolationWindowTargetMz extracts isolation window target m/z definition spectrum object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"","code":"# S4 method for OnDiskMSnExp isolationWindowTargetMz(object)"},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"object OnDiskMSnExp object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"numeric length equal number spectra object isolation window target m/z NA specified/available.","code":""},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot log intensity image of a xcmsRaw object — levelplot-methods","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"Create image raw (profile) data m/z retention   time, intensity color coded.","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"x = \"xcmsRaw\" levelplot(x, log=TRUE, col.regions=colorRampPalette(brewer.pal(9,     \"YlOrRd\"))(256), ...) x = \"xcmsSet\" levelplot(x, log=TRUE, col.regions=colorRampPalette(brewer.pal(9,     \"YlOrRd\"))(256), rt=\"raw\", ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"x xcmsRaw object. log Whether intensity log transformed. col.regions color ramp used encoding     intensity. rt wheter original (rt=\"raw\") corrected     (rt=\"corrected\") retention times used. ... Arguments profRange.","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"Johannes Rainer, johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Read binary data from a source — loadRaw-methods","title":"Read binary data from a source — loadRaw-methods","text":"function extracts raw data used   xcmsRaw object. processing data   done xcmsRaw constructor.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Read binary data from a source — loadRaw-methods","text":"signature(object = \"xcmsSource\") Uses loadRaw,xcmsSource-method extract raw data.       Subclasses xcmsSource can provide different       ways fetching data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Read binary data from a source — loadRaw-methods","text":"object Specification data source (file name     database query)","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Read binary data from a source — loadRaw-methods","text":"implementing methods decide gather data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Read binary data from a source — loadRaw-methods","text":"list containing elements describing data source. rt,  scanindex, tic, acquisitionNum components   one entry per scan. parallel sense  rt[1], scanindex[1], acquisitionNum[1]   refer scan. list containst following components: rt Numeric vector acquisition time (seconds) scan tic Numeric vector Total Ion Count scan scanindex Integer vector starting positions scan mz     intensity components. exclusive offset,     scanindex[] offset mz intensity     beginning scan . means     mz (respectively intensity) values scan     scanindex[] + 1 scanindex[+ 1] mz Concatenated vector m/z values scans intensity Concatenated vector intensity values scans","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Read binary data from a source — loadRaw-methods","text":"Daniel Hackney, dan@haxney.org","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":null,"dir":"Reference","previous_headings":"","what":"LC-MS preprocessing result test data sets — loadXcmsData","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"Data sets `xcms` preprocessing results provided within `xcms` package can loaded `loadXcmsData` function. available Test data sets : - `xdata`: [XCMSnExp()] object results `xcms`-based   pre-processing LC-MS untargeted metabolomics data set. raw data   files provided `faahKO` R package. - `xmse`: [XcmsExperiment()] object results `xcms`-based   pre-processing LC-MS untargeted metabolomics data set (original   data set pre-processing settings `xdata` data set).   pre-processing data set described detail *xcms*   vignette `xcms` package. - `faahko_sub`: [XCMSnExp()] object identified   chromatographic peaks 3 samples data files `faahKO`   R package. - `faahko_sub2`: [XcmsExperiment()] object identified   chromatographic peaks 3 samples data files `faahKO`   R package. Data sets can also loaded using `data`, however require update objects point location raw data files. `loadXcmsData` loads data ensures paths updated accordingly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"","code":"loadXcmsData(x = c(\"xmse\", \"xdata\", \"faahko_sub\", \"faahko_sub2\"))"},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"x `loadXcmsData`: `character(1)` name data file (object) load.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"","code":"library(xcms) xdata <- loadXcmsData()"},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Manual peak integration and feature definition — manualChromPeaks","title":"Manual peak integration and feature definition — manualChromPeaks","text":"manualChromPeaks function allows manually define chromatographic peaks, integrate intensities within specified peak area add object's chromPeaks matrix. peak added sample signal found respective data file. chromatographic peaks added eventually previously identified peaks, suggested run refineChromPeaks() MergeNeighboringPeaksParam() approach merge potentially overlapping peaks. manualFeatures function allows manually group identified chromatographic peaks features providing index object's chromPeaks matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Manual peak integration and feature definition — manualChromPeaks","text":"","code":"manualChromPeaks(object, ...)  manualFeatures(object, ...)  # S4 method for MsExperiment manualChromPeaks(   object,   chromPeaks = matrix(numeric()),   samples = seq_along(object),   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment manualChromPeaks(   object,   chromPeaks = matrix(numeric()),   samples = seq_along(object),   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment manualFeatures(object, peakIdx = list(), msLevel = 1L)  # S4 method for OnDiskMSnExp manualChromPeaks(   object,   chromPeaks = matrix(),   samples = seq_along(fileNames(object)),   msLevel = 1L,   BPPARAM = bpparam() )  # S4 method for XCMSnExp manualChromPeaks(   object,   chromPeaks = matrix(),   samples = seq_along(fileNames(object)),   msLevel = 1L,   BPPARAM = bpparam() )  # S4 method for XCMSnExp manualFeatures(object, peakIdx = list(), msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Manual peak integration and feature definition — manualChromPeaks","text":"object XcmsExperiment, XCMSnExp OnDiskMSnExp object. ... ignored. chromPeaks manualChromPeaks: matrix defining boundaries chromatographic peaks one row per chromatographic peak columns \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" defining m/z retention time region peak. samples manualChromPeaks: optional integer defining individual samples peak integration performed. Defaults samples. msLevel integer(1) defining MS level peak integration performed. single MS level time supported. Defaults msLevel = 1L. chunkSize integer(1) defining number files (samples) loaded memory processed time. Peak integration performed parallel (per sample) subset data. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM parallel processing settings (see bpparam() details). peakIdx manualFeatures: list integer vectors indices chromatographic peaks object's chromPeaks matrix grouped features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Manual peak integration and feature definition — manualChromPeaks","text":"XcmsExperiment XCMSnExp manually added chromatographic peaks features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Manual peak integration and feature definition — manualChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Manual peak integration and feature definition — manualChromPeaks","text":"","code":"## Read a test dataset. fls <- c(system.file(\"microtofq/MM14.mzML\", package = \"msdata\"),          system.file(\"microtofq/MM8.mzML\", package = \"msdata\"))  ## Define a data frame with some sample annotations ann <- data.frame(     injection_index = 1:2,     sample_id = c(\"MM14\", \"MM8\"))  ## Import the data library(MsExperiment) mse <- readMsExperiment(fls)  ## Define some arbitrary peak areas pks <- cbind(     mzmin = c(512, 234.3), mzmax = c(513, 235),     rtmin = c(10, 33), rtmax = c(19, 50) ) pks #>      mzmin mzmax rtmin rtmax #> [1,] 512.0   513    10    19 #> [2,] 234.3   235    33    50  res <- manualChromPeaks(mse, pks) chromPeaks(res) #>           mz mzmin mzmax       rt rtmin rtmax      into sample maxo #> CP1 512.6294 512.0   513 18.66000    10    19  848.4231      2   85 #> CP2 234.8539 234.3   235 41.54298    33    50 2690.4200      2  245  ## Peaks were only found in the second file."},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Apply a median filter to a matrix — medianFilter","title":"Apply a median filter to a matrix — medianFilter","text":"element matix, replace median   values around .","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Apply a median filter to a matrix — medianFilter","text":"","code":"medianFilter(x, mrad, nrad)"},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Apply a median filter to a matrix — medianFilter","text":"x numeric matrix median filter mrad number rows either side value use median     calculation nrad number rows either side value use median     calculation","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Apply a median filter to a matrix — medianFilter","text":"matrix whose values median filtered","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Apply a median filter to a matrix — medianFilter","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Apply a median filter to a matrix — medianFilter","text":"","code":"mat <- matrix(1:25, nrow=5) mat #>      [,1] [,2] [,3] [,4] [,5] #> [1,]    1    6   11   16   21 #> [2,]    2    7   12   17   22 #> [3,]    3    8   13   18   23 #> [4,]    4    9   14   19   24 #> [5,]    5   10   15   20   25 medianFilter(mat, 1, 1) #>      [,1] [,2] [,3] [,4] [,5] #> [1,]  4.0  6.5 11.5 16.5 19.0 #> [2,]  4.5  7.0 12.0 17.0 19.5 #> [3,]  5.5  8.0 13.0 18.0 20.5 #> [4,]  6.5  9.0 14.0 19.0 21.5 #> [5,]  7.0  9.5 14.5 19.5 22.0"},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"MS2 MSn data stored separate slots,   can directly used e.g. findPeaks().   msn2xcmsRaw() copy MSn spectra   \"normal\" xcmsRaw slots.","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"","code":"msn2xcmsRaw(xmsn)"},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"xmsn object class xcmsRaw contains     spectra read includeMSn=TRUE","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"default gap value determined 90th percentile   pair-wise differences adjacent mass values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"xcmsRaw object","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"Steffen Neumann sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"","code":"msnfile <- system.file(\"microtofq/MSMSpos20_6.mzML\", package = \"msdata\")  xrmsn <- xcmsRaw(msnfile, includeMSn=TRUE) #> Create profile matrix with method 'bin' and step 1 ...  #> OK  xr <- msn2xcmsRaw(xrmsn)  p <- findPeaks(xr, method=\"centWave\") #> Detecting mass traces at 25 ppm ...  #> OK #> Detecting chromatographic peaks in 10 regions of interest ... #>  OK: 10 found."},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":null,"dir":"Reference","previous_headings":"","what":"Fill in NA values at the extremes of a vector — na.flatfill","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"Extend first last real values vector fill   NA values present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"","code":"na.flatfill(x)"},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"x numeric vector NA values","code":""},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"Modified vector.","code":""},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify overlapping features — overlappingFeatures","title":"Identify overlapping features — overlappingFeatures","text":"overlappingFeatures identifies features overlapping close m/z - rt space.","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify overlapping features — overlappingFeatures","text":"","code":"overlappingFeatures(x, expandMz = 0, expandRt = 0, ppm = 0)"},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify overlapping features — overlappingFeatures","text":"x XcmsExperiment() XCMSnExp() object features. expandMz numeric(1) value expand feature (side) m/z dimension identifying overlapping features. resulting \"mzmin\" feature thus mzmin - expandMz \"mzmax\" mzmax + expandMz. expandRt numeric(1) value expand feature (side) retention time dimension identifying overlapping features. resulting \"rtmin\" feature thus rtmin - expandRt \"rtmax\" rtmax + expandRt. ppm numeric(1) grow m/z width feature relative value: mzmin - mzmin * ppm / 2e6, mzmax + mzmax * ppm / 2e6. feature thus expanded m/z dimension ppm/2 side identifying overlapping features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify overlapping features — overlappingFeatures","text":"list indices features (featureDefinitions()) overlapping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify overlapping features — overlappingFeatures","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Identify overlapping features — overlappingFeatures","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Correspondence analysis xdata <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = c(1, 1, 1)))  ## Identify overlapping features overlappingFeatures(xdata) #> list()  ## Identify features that are separated on retention time by less than ## 2 minutes overlappingFeatures(xdata, expandRt = 60) #> [[1]] #> [1] 5 6 #>  #> [[2]] #> [1] 31 32 #>"},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":null,"dir":"Reference","previous_headings":"","what":"Correlation coefficient panel for pairs function — panel.cor","title":"Correlation coefficient panel for pairs function — panel.cor","text":"Correlation coefficient panel pairs function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Correlation coefficient panel for pairs function — panel.cor","text":"","code":"panel.cor(x, y, digits = 2, prefix = \"\", cex.cor)"},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correlation coefficient panel for pairs function — panel.cor","text":"x first data series y second data series digits number digits plot prefix text prefix coefficients cex.cor character expansion factor","code":""},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correlation coefficient panel for pairs function — panel.cor","text":"Colin . Smith, csmith@scripps.edu, based   pairs example code","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot a grid of a large number of peaks — peakPlots-methods","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"Plot extracted ion chromatograms many peaks simultaneously, indicating peak integration start end points vertical grey lines.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"signature(object = \"xcmsSet\") plotPeaks(object, peaks, figs, width = 200)","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"object xcmsRaw object peaks matrix peak information produced findPeaks figs two-element vector describing number rows number     columns peaks plot, missing approximately     square grid fit number peaks supplied width width chromatogram retention time plot peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"function intended help graphically analyze results   peak picking. can help estimate number false positives   improper integration start end points. output   compact tries waste little space possible. plot   labeled rounded m/z retention time separated   space.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Create report of aligned peak intensities — peakTable-methods","title":"Create report of aligned peak intensities — peakTable-methods","text":"Create report showing aligned peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create report of aligned peak intensities — peakTable-methods","text":"object = \"xcmsSet\" peakTable(object, filebase = character(), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create report of aligned peak intensities — peakTable-methods","text":"object xcmsSet object filebase base file name save report, .tsv file _eic     appended name tabular report EIC     directory, respectively. blank nothing saved ... arguments passed groupval,     provides actual intensities.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create report of aligned peak intensities — peakTable-methods","text":"method handles creation summary reports similar   diffreport. returns summary report can   optionally written tab-separated file. base file name provided, report (see Value section)   saved tab separated file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create report of aligned peak intensities — peakTable-methods","text":"data frame following columns: mz median m/z peaks group mzmin minimum m/z peaks group mzmax maximum m/z peaks group rt median retention time peaks group rtmin minimum retention time peaks group rtmax maximum retention time peaks group npeaks number peaks assigned group Sample Classes number samples sample class represented group ... one column every sample class Sample Names integrated intensity value every sample ... one column every sample","code":""},{"path":[]},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"peaksWithCentWave identifies (chromatographic) peaks purely chromatographic data, .e. based intensity retention time values without m/z values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"","code":"peaksWithCentWave(   int,   rt,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   integrate = 1,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   firstBaselineCheck = TRUE,   extendLengthMSW = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"int numeric intensity values. rt numeric retention time intensities. Length equal length(int). peakwidth numeric(2) lower upper bound expected peak width. snthresh numeric(1) defining signal noise ratio cutoff. Peaks signal noise ratio < snthresh omitted. prefilter numeric(2) (c(k, )): regions interest least k centroids signal >= returned first step. integrate numeric(1), integration method. integrate = 1 peak limits found descending mexican hat filtered data, integrate = 2 descend done real data. latter method accurate prone noise, former robust, less exact. fitgauss logical(1) whether Gaussian fitted peak. noise numeric(1) defining minimum required intensity centroids considered first analysis step (definition regions interest). verboseColumns logical(1): whether additional peak meta data columns returned. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. extendLengthMSW logical(1). TRUE \"open\" method EIC extension used, rather default \"reflect\" method. ... currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"matrix, row representing identified chromatographic peak, columns: \"rt\": retention time peak's midpoint (time maximum signal). \"rtmin\": minimum retention time peak. \"rtmax\": maximum retention time peak. \"\": integrated (original) intensity peak. \"intb\": per-peak baseline corrected integrated peak intensity. \"maxo\": maximum (original) intensity peak. \"sn\": signal noise ratio peak defined (maxo - baseline)/sd sd standard deviation local chromatographic noise. Additional columns verboseColumns = TRUE: \"mu\": gaussian parameter mu. \"sigma\": gaussian parameter sigma. \"h\": gaussian parameter h. \"f\": region number m/z ROI peak localized. \"dppm\": m/z deviation mass trace across scans ppm (always NA). \"scale\": scale peak localized. \"scpos\": peak position found wavelet analysis (index int). \"scmin\": left peak limit found wavelet analysis (index int). \"scmax\": right peak limit found wavelet analysis (index int).","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"method uses algorithm peak detection centWave, employs however different approach identify initial regions peak detection performed (.e. regions interest ROI). method first identifies local maxima chromatographic data defines corresponding positions +/- peakwidth[2] ROIs. Noise estimation bases also ROIs can thus different centWave resulting different signal noise ratios.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"","code":"## Reading a file library(xcms) od <- readMSData(system.file(\"cdf/KO/ko15.CDF\", package = \"faahKO\"),     mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Extract chromatographic data for a small m/z range mzr <- c(272.1, 272.2) chr <- chromatogram(od, mz = mzr, rt = c(3000, 3300))[1, 1]  int <- intensity(chr) rt <- rtime(chr)  ## Plot the region plot(chr, type = \"h\")  ## Identify peaks in the chromatographic data pks <- peaksWithCentWave(intensity(chr), rtime(chr)) pks #>            rt    rtmin    rtmax     into     intb  maxo sn #> [1,] 3119.535 3097.625 3127.359 247211.6 227217.9 16496 24  ## Highlight the peaks rect(xleft = pks[, \"rtmin\"], xright = pks[, \"rtmax\"],     ybottom = rep(0, nrow(pks)), ytop = pks[, \"maxo\"], col = \"#ff000040\",     border = \"#00000040\")"},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"function performs peak detection using matchedFilter algorithm chromatographic data (.e. intensities retention time).","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"","code":"peaksWithMatchedFilter(   int,   rt,   fwhm = 30,   sigma = fwhm/2.3548,   max = 20,   snthresh = 10,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"int numeric intensity values. rt numeric retention time intensities. Length equal length(int). fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) maximal number peaks expected/ bbe detected data snthresh numeric(1) defining signal noise cut-used peak detection step. ... currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"matrix, row representing identified chromatographic peak, columns: \"rt\": retention time peak's midpoint (time maximum signal). \"rtmin\": minimum retention time peak. \"rtmax\": maximum retention time peak. \"\": integrated (original) intensity peak. \"intf\": integrated intensity filtered peak. \"maxo\": maximum (original) intensity peak. \"maxf\"\" maximum intensity filtered peak. \"sn\": signal noise ratio peak.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and drop identified chromatographic peaks data <- dropChromPeaks(filterFile(faahko_sub, 1))  ## Extract chromatographic data for a small m/z range chr <- chromatogram(data, mz = c(272.1, 272.3), rt = c(3000, 3200))[1, 1]  pks <- peaksWithMatchedFilter(intensity(chr), rtime(chr)) pks #>            rt   rtmin    rtmax     into     intf  maxo     maxf       sn #> [1,] 3119.535 3105.45 3133.619 332583.6 668444.6 16496 36129.39 12.33813  ## Plotting the data plot(rtime(chr), intensity(chr), type = \"h\") rect(xleft = pks[, \"rtmin\"], xright = pks[, \"rtmax\"], ybottom = c(0, 0),     ytop = pks[, \"maxo\"], border = \"red\")"},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":null,"dir":"Reference","previous_headings":"","what":"Derive experimental design from file paths — phenoDataFromPaths","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"phenoDataFromPaths function builds data.frame representing experimental design folder structure files experiment located.","code":""},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"","code":"phenoDataFromPaths(paths)"},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"paths character representing file names (including full path) experiment's files.","code":""},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"function used old xcmsSet function guess experimental design (.e. group assignment files) folders files experiment can found.","code":""},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"","code":"## List the files available in the faahKO package base_dir <- system.file(\"cdf\", package = \"faahKO\") cdf_files <- list.files(base_dir, recursive = TRUE, full.names = TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"Batch plot list extracted ion chromatograms current   graphics device.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"x = \"xcmsEIC\" plot.xcmsEIC(x, y, groupidx = groupnames(x), sampleidx = sampnames(x), rtrange = x@rtrange,              col = rep(1, length(sampleidx)), legtext = NULL, peakint = TRUE, sleep = 0, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"x xcmsEIC object y optional xcmsSet object peak integration data groupidx either character vector names integer vector     indicies peak groups plot EICs sampleidx either character vector names integer vector     indicies samples plot EICs rtrange two column matrix minimum maximum retention times     return EIC data points number rows number groups     xcmsEIC object, sampleidx used subset     . otherwise, repeated length sampleidx may also single number specifying time window around     peak plot EIC data col color use plotting extracted ion chromatograms. missing     y specified, colors taken     unclass(sampclass(y)) default palette length number groups xcmsEIC     object, sampleidx used subset . otherwise,     repeated length sampleidx legtext text use legend. NULL y specified,     legend text taken sample class information found     xcmsSet peakint logical, plot integrated peak area darkened lines (requires     y also specified) sleep seconds pause plotting EICs ... graphical parameters","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Visualization of Alignment Results — plotAdjustedRtime","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"`plotAdjustedRtime` function plots difference adjusted *raw* retention times y-axis raw retention times x-axis. line represents results one sample (file). alignment performed using *peak groups* method (see [adjustRtime()] infromation) also peak groups used alignment visualized.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"","code":"plotAdjustedRtime(   object,   col = \"#00000080\",   lty = 1,   lwd = 1,   type = \"l\",   adjustedRtime = TRUE,   xlab = ifelse(adjustedRtime, yes = expression(rt[adj]), no = expression(rt[raw])),   ylab = expression(rt[adj] - rt[raw]),   peakGroupsCol = \"#00000060\",   peakGroupsPch = 16,   peakGroupsLty = 3,   ylim,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"object [XcmsExperiment()] [XCMSnExp()] object alignment results. col color(s) individual lines. length 1 equal number samples. lty line type lines individual samples. lwd line width lines individual samples. type plot *type* (see [par()] options; defaults `type = \"l\"`). adjustedRtime `logical(1)` whether adjusted raw retention times shown x-axis. xlab label x-axis. ylab label y-axis. peakGroupsCol color used peak groups (alignment performed using *peak groups* method. peakGroupsPch point character (`pch`) used peak groups (alignment performed using *peak groups* method. peakGroupsLty line type (`lty`) used connect points peak groups (alignment performed using *peak groups* method. ylim optional `numeric(2)` upper lower limits y-axis.b ... Additional arguments passed `plot` function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Performing the peak grouping using the \"peak density\" method. p <- PeakDensityParam(sampleGroups = c(1, 1, 1)) res <- groupChromPeaks(faahko_sub, param = p)  ## Perform the retention time adjustment using peak groups found in both ## files. fgp <- PeakGroupsParam(minFraction = 1) res <- adjustRtime(res, param = fgp) #> Performing retention time correction using 19 peak groups. #> Warning: Span too small for 'loess' and the available number of peak groups, resetting to 0.21 #> Warning: Adjusted retention times had to be re-adjusted for some files to ensure them being in the same order than the raw retention times. A call to 'dropAdjustedRtime' might thus fail to restore retention times of chromatographic peaks to their original values. Eventually consider to increase the value of the 'span' parameter.  ## Visualize the impact of the alignment. plotAdjustedRtime(res, adjusted = FALSE) grid()"},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"Uses pre-generated profile mode matrix plot averaged   base peak extracted ion chromatograms specified mass range.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"object = \"xcmsRaw\" plotChrom(object, base = FALSE, ident = FALSE,                   fitgauss = FALSE, vline = numeric(0), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"object xcmsRaw object base logical, plot base-peak chromatogram ident logical, use mouse identify label peaks fitgauss logical, fit gaussian largest peak vline numeric vector locations vertical lines ... arguments passed profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"ident == TRUE, integer vector indecies   points identified. fitgauss == TRUE,  nls model fitted gaussian. Otherwise two-column   matrix plotted points.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"Plot density chromatographic peaks along retention time axis indicate peaks () grouped feature based using peak density correspondence method. Settings peak density method can passed PeakDensityParam object parameter param. object contains correspondence results correspondence performed peak groups method, results correspondence can visualized setting simulate = FALSE.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp plotChromPeakDensity(   object,   mz,   rt,   param,   simulate = TRUE,   col = \"#00000080\",   xlab = \"retention time\",   ylab = \"sample\",   xlim = range(rt),   main = NULL,   type = c(\"any\", \"within\", \"apex_within\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"object XCMSnExp object identified chromatographic peaks. mz numeric(2) defining mz range peak density plotted. rt numeric(2) defining optional rt range peak density plotted. Defaults absolute retention time range object. param PeakDensityParam parameters peak density correspondence algorithm can extracted. provided object contains feature definitions correspondence/ peak grouping performed peak density method, corresponding parameter class stored object used. simulate logical(1) defining whether correspondence simulated within specified m/z / rt region (simulate = FALSE) whether results already performed correspondence shown. col Color used individual samples. Length 1 equal number samples object. xlab character(1) label x-axis. ylab character(1) label y-axis. xlim numeric(2) representing limits x-axis. Defaults range rt parameter. main character(1) defining title plot. default (main = NULL) mz-range used. type character(1) specifying peaks called located within region defined mz rt. Can one \"\", \"within\", \"apex_within\" peaks even partially overlapping region, peaks completely within region, peaks apex within region. parameter passed chromPeaks function. See related documentation information examples. ... Additional parameters passed plot function. Data point specific parameters bg pch length 1 equal number samples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"function called side effect, .e. create plot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"plotChromPeakDensity function allows evaluate different settings peak density mz slice interest (e.g. containing chromatographic peaks corresponding known metabolite). plot shows individual peaks detected within specified mz slice retention time (x-axis) sample detected (y-axis). density function plotted black line. Parameters density function taken param object. Grey rectangles indicate chromatographic peaks grouped feature peak density correspondence method. Parameters algorithm also taken param. See groupChromPeaks() information algorithm supported settings.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Plot the chromatographic peak density for a specific mz range to evaluate ## different peak density correspondence settings. mzr <- c(305.05, 305.15)  plotChromPeakDensity(faahko_sub, mz = mzr, pch = 16,     param = PeakDensityParam(sampleGroups = rep(1, length(fileNames(faahko_sub))))) #> Warning: Use of 'plotChromPeakDensity' on 'XCMSnExp' isdiscouraged. Please extract chromatographic data first and call 'plotChromPeakDensity' directly on the 'XChromatograms' object. See ?XChromatograms, section 'Correspondence analysis' for more details."},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"General visualizations of peak detection results — plotChromPeaks","title":"General visualizations of peak detection results — plotChromPeaks","text":"`plotChromPeaks` plots identified chromatographic peaks one file plane spanned retention time (x-axis) m/z (y-axis) dimension. chromatographic peak plotted rectangle representing width RT m/z dimension. `plotChromPeakImage` plots number detected peaks sample along retention time axis *image* plot, .e. number peaks detected bin along retention time represented color respective cell.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"General visualizations of peak detection results — plotChromPeaks","text":"","code":"plotChromPeaks(   x,   file = 1,   xlim = NULL,   ylim = NULL,   add = FALSE,   border = \"#00000060\",   col = NA,   xlab = \"retention time\",   ylab = \"mz\",   main = NULL,   msLevel = 1L,   ... )  plotChromPeakImage(   x,   binSize = 30,   xlim = NULL,   log = FALSE,   xlab = \"retention time\",   yaxt = par(\"yaxt\"),   main = \"Chromatographic peak counts\",   msLevel = 1L,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"General visualizations of peak detection results — plotChromPeaks","text":"x [XcmsExperiment()] [XCMSnExp()] object. file `plotChromPeaks`: `integer(1)` specifying index file within `x` plot created. Defaults `file = 1`. xlim `numeric(2)` specifying x-axis limits (retention time dimension). Defaults `xlim = NULL` case full retention time range file used. ylim `plotChromPeaks`: `numeric(2)` specifying y-axis limits (m/z dimension). Defaults `ylim = NULL` case full m/z range file used. add `plotChromPeaks`: `logical(1)` whether plot added existing plot new plot created. border `plotChromPeaks`: color rectangles' border. col `plotChromPeaks`: color used fill rectangles. xlab `character(1)` defining x-axis label. ylab `plotChromPeaks`: `character(1)` defining y-axis label. main `character(1)` defining plot title. default (.e. `main = NULL`) name file used title. msLevel `integer(1)` defining MS level peaks visualized. ... Additional arguments passed `plot` (`plotChromPeaks`) `image` (`plotChromPeakImage`) functions. Ignored `add = TRUE`. binSize `plotChromPeakImage`: `numeric(1)` defining size bins along x-axis (retention time). Defaults `binSize = 30`, peaks within 30 seconds thus counted plotted. log `plotChromPeakImage`: `logical(1)` whether peak counts log2 transformed plotting. yaxt `plotChromPeakImage`: `character(1)` defining whether y-axis labels added. disable y-axis use `yaxt = \"n\"`. value `yaxt` axis drawn. See [par()] help page details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"General visualizations of peak detection results — plotChromPeaks","text":"width line type rectangles indicating detected chromatographic peaks `plotChromPeaks` function can specified using `par` function, .e. `par(lwd = 3)` `par(lty = 2)`, respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"General visualizations of peak detection results — plotChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"General visualizations of peak detection results — plotChromPeaks","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## plotChromPeakImage: plot an image for the identified peaks per file plotChromPeakImage(faahko_sub)   ## Show all detected chromatographic peaks from the first file plotChromPeaks(faahko_sub)   ## Plot all detected peaks from the second file and restrict the plot to a ## mz-rt slice plotChromPeaks(faahko_sub, file = 2, xlim = c(3500, 3600), ylim = c(400, 600))"},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"plotOverlay draws chromatographic peak data multiple (different) extracted ion chromatograms (EICs) plot. allows directly compare peak shape EICs sample. contrast plot function MChromatograms() object, draws data EIC across multiple samples plot, function draws different EICs sample plot. plotChromatogramsOverlay called XChromatograms object present chromatographic peaks also highlighted/drawn depending parameters peakType, peakCol, peakBg peakPch (see also help plot function XChromatogram() object details).","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"","code":"# S4 method for MChromatograms plotChromatogramsOverlay(   object,   col = \"#00000060\",   type = \"l\",   main = NULL,   xlab = \"rtime\",   ylab = \"intensity\",   xlim = numeric(),   ylim = numeric(),   stacked = 0,   transform = identity,   ... )  # S4 method for XChromatograms plotChromatogramsOverlay(   object,   col = \"#00000060\",   type = \"l\",   main = NULL,   xlab = \"rtime\",   ylab = \"intensity\",   xlim = numeric(),   ylim = numeric(),   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakBg = NULL,   peakCol = NULL,   peakPch = 1,   stacked = 0,   transform = identity,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"object MChromatograms() XChromatograms() object. col definition color chromatograms drawn. Can length 1 equal nrow(object) plot overlayed chromatogram different color. type character(1) defing type plot. default (type = \"l\") chromatogram drawn line. main optional title plot. defined, range m/z values used. xlab character(1) defining x-axis label. ylab character(1) defining y-axis label. xlim optional numeric(2) defining x-axis limits. ylim optional numeric(2) defining y-axis limits. stacked numeric(1) defining part (proportion) y-axis use stack EICs depending m/z values. stacked = 0 (default) stacking performed. stacked = 1 half y-axis used stacking half intensity y-axis (.e. ratio stacking intensity y-axis 1:1). Note stacking different 0 y-axis label drawn. transform function transform intensity values plotting. Defaults transform = identity plots data . transform = log10 intensity values log10 transformed plotting. ... optional arguments passed plotting functions (see help base R plot function. peakType object XChromatograms object: chromatographic peaks drawn: peakType = \"polygon\" (default): label full chromatographic peak area, peakType = \"rectangle\": indicate chromatographic peak rectangle peakType  = \"point\": label chromatographic peaks' apex position point. peakBg object XChromatograms object: definition background color(s) chromatographic peak. either length 1 equal number peaks object. specified, peak drawn color defined col. peakCol object XChromatograms object: definition color(s) chromatographic peak. either length 1 equal number peaks object. specified, peak drawn color defined col. peakPch object XChromatograms object: point character used label apex position chromatographic peak peakType = \"point\".","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"silently returns list (length equal ncol(object) numeric (length equal nrow(object)) y position EIC.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"","code":"## Load preprocessed data and extract EICs for some features. library(xcms) xdata <- loadXcmsData() data(xdata) ## Update the path to the files for the local system dirname(xdata) <- c(rep(system.file(\"cdf\", \"KO\", package = \"faahKO\"), 4),                     rep(system.file(\"cdf\", \"WT\", package = \"faahKO\"), 4)) #> Error: unable to find an inherited method for function ‘path’ for signature ‘object = \"XcmsExperiment\"’ ## Subset to the first 3 files. xdata <- filterFile(xdata, 1:3, keepFeatures = TRUE)  ## Define features for which to extract EICs fts <- c(\"FT097\", \"FT163\", \"FT165\") chrs <- featureChromatograms(xdata, features = fts)  plotChromatogramsOverlay(chrs)   ## plot the overlay of EICs in the first sample plotChromatogramsOverlay(chrs[, 1])  ## Define a different color for each feature (row in chrs). By default, also ## all chromatographic peaks of a feature is labeled in the same color. plotChromatogramsOverlay(chrs[, 1],     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"))  ## Alternatively, we can define a color for each individual chromatographic ## peak and provide this with the `peakBg` and `peakCol` parameters. chromPeaks(chrs[, 1]) #>              mz mzmin mzmax       rt    rtmin    rtmax     into     intb  maxo #> CP2973 354.1357 354.1 354.2 2789.781 2755.258 2826.270 299005.1       NA  9700 #> CP0312 413.1000 413.1 413.1 3637.671 3614.978 3665.850 879851.5 852195.3 30920 #> CP0089 414.2000 414.2 414.2 3066.457 3038.628 3089.311 234308.5 218923.4  9669 #>        sn sample row column #> CP2973 NA      1   1      1 #> CP0312 34      1   2      1 #> CP0089 15      1   3      1  ## Use a color for each of the two identified peaks in that sample plotChromatogramsOverlay(chrs[, 1],     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"),     peakBg = c(\"#ffff0020\", \"#00ffff20\"))   ## Plotting the data in all samples. plotChromatogramsOverlay(chrs,     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"))   ## Creating a \"stacked\" EIC plot: the EICs are placed along the y-axis ## relative to their m/z value. With `stacked = 1` the y-axis is split in ## half, the lower half being used for the stacking of the EICs, the upper ## half being used for the *original* intensity axis. res <- plotChromatogramsOverlay(chrs[, 1], stacked = 1,     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\")) ## add horizontal lines for the m/z values of each EIC abline(h = res[[1]], col = \"grey\", lty = 2)  ## Note that this type of visualization is different than the conventional ## plot function for chromatographic data, which will draw the EICs for ## multiple samples into the same plot plot(chrs)   ## Converting the object to a MChromatograms without detected peaks chrs <- as(chrs, \"MChromatograms\")  plotChromatogramsOverlay(chrs,     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"))"},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"Plot extracted ion chromatogram m/z values interest. raw data used contrast  plotChrom uses data profile matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"object = \"xcmsRaw\" plotEIC(object, mzrange = numeric(), rtrange = numeric(),     scanrange = numeric(), mzdec=2, type=\"l\", add=FALSE, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"object xcmsRaw object mzrange m/z range EIC. Uses full m/z range default. rtrange retention time range EIC. Uses full retention time range default. scanrange scan range EIC mzdec Number decimal places title m/z values eic plot. type Speficies data plotted (default line). add EIC added existing plot. ... Additional parameters passed plotting function     (e.g. col etc).","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"two-column matrix plotted points.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"Ralf Tautenhahn","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"plotFeatureGroups visualizes defined feature groups m/z retention time space. Features indicated points features feature group connected line. See featureGroups() details options feature grouping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"","code":"plotFeatureGroups(   x,   xlim = numeric(),   ylim = numeric(),   xlab = \"retention time\",   ylab = \"m/z\",   pch = 4,   col = \"#00000060\",   type = \"o\",   main = \"Feature groups\",   featureGroups = character(),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"x XcmsExperiment XCMSnExp() object grouped features (.e. calling groupFeatures(). xlim numeric(2) lower upper limit x-axis. ylim numeric(2) lower upper limit y-axis. xlab character(1) label x-axis. ylab character(1) label y-axis. pch plotting character. Defaults pch = 4 .e. plotting features crosses. See par() information. col color used draw features. present single color supported. type plotting type (see par()). Defaults type = \"o\" draws feature point connecting features feature group line. main character(1) title plot. featureGroups optional character feature group IDs draw specified feature group(s). provided, feature groups drawn. ... additional parameters passed lines function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":null,"dir":"Reference","previous_headings":"","what":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"UPDATE: please use plot(x, type = \"XIC\") MSnbase package instead. See examples . plotMsData creates plot combines (base peak ) extracted ion chromatogram top (rt intensity) plot rt m/z values bottom.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"","code":"plotMsData(   x,   main = \"\",   cex = 1,   mfrow = c(2, 1),   grid.color = \"lightgrey\",   colramp = colorRampPalette(rev(brewer.pal(9, \"YlGnBu\"))) )"},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"x data.frame returned extractMsData() function. single data.frame supported. main character(1) specifying title. cex numeric(1) defining size points. Passed directly plot function. mfrow numeric(2) defining plot layout. passed directly par(mfrow = mfrow). See par information. Setting mfrow = NULL avoids calling par(mfrow = mfrow) hence allowing pre-define plot layout. grid.color color definition grid line (NA skip creating ). colramp color ramp palette used color data points based intensity. See argument col.regions lattice::level.colors documentation.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"","code":"## Read two files from the faahKO package library(faahKO) cdfs <- dir(system.file(\"cdf\", package = \"faahKO\"), full.names = TRUE,     recursive = TRUE)[1:2] raw_data <- readMSData(cdfs, mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Subset the object to a rt and mz range and plot the data. raw_data |>     filterRt(rt = c(2700, 2900)) |>     filterMz(mz = c(334.9, 335.1)) |>     plot(type = \"XIC\") #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 128. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 131. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 132. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 140. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 141. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 142. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 143. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 149. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 150. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 151. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 152. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 153. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 154. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 155. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 156. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 157. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 159. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 160. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 161. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 162. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 163. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 164. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 170. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 171. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 172. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 199. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 200. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 201. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 206. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 207. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 208. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 209. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 217. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 218. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 219. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 220. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 221. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 222. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 223. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 235. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 236. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 237. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 238. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 239. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 240. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 241. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 242. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 243. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 244. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 245. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 246. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 247. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 248. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 249. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 250. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 129. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 130. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 131. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 132. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 133. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 134. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 135. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 136. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 138. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 139. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 140. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 141. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 144. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 145. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 146. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 147. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 148. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 149. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 155. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 156. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 157. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 158. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 159. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 160. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 165. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 166. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 167. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 207. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 208. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 219. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 220. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 221. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 222. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 223. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 224. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 227. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 228. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 229. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 233. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 234. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 235. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 247. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 248. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 249. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 250. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 251. Returning empty spectrum."},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot a grid of a large number of peaks — plotPeaks-methods","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"Plot extracted ion chromatograms many peaks simultaneously,   indicating peak integration start end points vertical   grey lines.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"object = \"xcmsRaw\" plotPeaks(object, peaks, figs, width = 200)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"object xcmsRaw object peaks matrix peak information produced findPeaks figs two-element vector describing number rows number     columns peaks plot, missing approximately     square grid fit number peaks supplied width width chromatogram retention time plot peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"function intended help graphically analyze results   peak picking. can help estimate number false positives   improper integration start end points. output   compact tries waste little space possible. plot   labeled rounded m/z retention time separated   space.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"Use \"democracy\" determine average m/z RT deviations grouped xcmsSet, dependency sample absolute m/z","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"","code":"plotQC(object, sampNames, sampColors, sampOrder, what)"},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"object grouped xcmsSet sampNames Override sample names (e.g. simplified names) sampColors Provide set colors (default: monochrome ?) sampOrder Override order samples, e.g. bring order measurement detect time drift vector QC plots generate. \"mzdevhist\": histogram mz deviations. gaussian shaped. multimodal, peaks seem systematically higher m/z deviation \"rtdevhist\":  histogram RT deviations. gaussian shaped. multimodal, peaks seem systematically higher RT deviation \"mzdevmass\": Shows whether m/z deviations absolute m/z dependent, indicate miscalibration \"mzdevtime\": Shows whether m/z deviations RT dependent, indicate instrument drift \"mzdevsample\": median mz deviation sample, indicates outliers \"rtdevsample\": median RT deviation sample, indicates outliers","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"List four matrices, dimension features * samples:  \"mz\":    median mz deviation sample  \"mzdev\": median mz deviation sample  \"rt\":    median RT deviation sample  \"rtdev\": median RT deviation sample","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"plotQC() warpper create set diagnostic plots. m/z deviations, median m/z withon one group assumed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"Michael Wenk, Michael Wenk <michael.wenk@student.uni-halle.de>","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"","code":"library(faahKO) xsg <- group(faahko)  plotQC(xsg, what=\"mzdevhist\")  plotQC(xsg, what=\"rtdevhist\")  plotQC(xsg, what=\"mzdevmass\")  plotQC(xsg, what=\"mzdevtime\")  plotQC(xsg, what=\"mzdevsample\")  plotQC(xsg, what=\"rtdevsample\")"},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Scatterplot of raw data points — plotRaw-methods","title":"Scatterplot of raw data points — plotRaw-methods","text":"Produce scatterplot showing raw data point location retention   time m/z. plot useful centroided data   continuum data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Scatterplot of raw data points — plotRaw-methods","text":"object = \"xcmsRaw\" plotRaw(object, mzrange = numeric(), rtrange = numeric(),                 scanrange = numeric(), log=FALSE, title='Raw Data')","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scatterplot of raw data points — plotRaw-methods","text":"object xcmsRaw object mzrange numeric vector length >= 2 whose range used select     masses plot rtrange numeric vector length >= 2 whose range used select     retention times plot scanrange numeric vector length >= 2 whose range used select     scans plot log logical, log transform intensity title main title plot","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scatterplot of raw data points — plotRaw-methods","text":"matrix points plotted.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotScan-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot a single mass scan — plotScan-methods","title":"Plot a single mass scan — plotScan-methods","text":"Plot single mass scan using impulse representation.   useful centroided data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotScan-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot a single mass scan — plotScan-methods","text":"object = \"xcmsRaw\" plotScan(object, scan, mzrange = numeric(), ident = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotScan-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot a single mass scan — plotScan-methods","text":"object xcmsRaw object scan integer number scan plot mzrange numeric vector length >= 2 whose range used select     masses plot ident logical, use mouse interactively identify label individual     masses","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot mass spectra from the profile matrix — plotSpec-methods","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"Uses pre-generated profile mode matrix plot mass spectra   specified retention time range.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"object = \"xcmsRaw\" plotSpec(object, ident = FALSE, vline = numeric(0), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"object xcmsRaw object ident logical, use mouse identify label peaks vline numeric vector locations vertical lines ... arguments passed profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"ident == TRUE, integer vector indecies   points identified. Otherwise two-column matrix   plotted points.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"method uses rgl package create interactive three   dimensonal representations profile matrix. uses terrain   color scheme.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"object = \"xcmsRaw\" plotSurf(object, log = FALSE, aspect = c(1, 1, .5), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"object xcmsRaw object log logical, log transform intensity aspect numeric vector aspect ratio m/z, retention time     intensity components plot ... arguments passed profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"rgl package still development imposes limitations   output format. bug axis label code means   axis labels go 0 aspect ratio constant   axis. Additionally axes labeled . important plot small portion profile matrix.   Large portions can quickly overwhelm CPU memory.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot total ion count — plotTIC-methods","title":"Plot total ion count — plotTIC-methods","text":"Plot chromatogram total ion count. Optionally allow identification   target peaks viewing/identification individual spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot total ion count — plotTIC-methods","text":"object = \"xcmsRaw\" plotTIC(object, ident = FALSE, msident = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot total ion count — plotTIC-methods","text":"object xcmsRaw object ident logical, use mouse identify label chromatographic peaks msident logical, use mouse identify label spectral peaks","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot total ion count — plotTIC-methods","text":"ident == TRUE, integer vector indecies   points identified. Otherwise two-column matrix   plotted points.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotrt-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot retention time deviation profiles — plotrt-methods","title":"Plot retention time deviation profiles — plotrt-methods","text":"Use corrected retention times sample calculate retention   time deviation profiles plot graph.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotrt-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot retention time deviation profiles — plotrt-methods","text":"object = \"xcmsSet\" plotrt(object, col = NULL, ty = NULL, leg = TRUE,                densplit = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotrt-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot retention time deviation profiles — plotrt-methods","text":"object xcmsSet object col vector colors plotting sample ty vector line point types plotting sample leg logical plot legend sample labels densplit logical, also plot peak overall peak density","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":null,"dir":"Reference","previous_headings":"","what":"Generation of profile data — profGenerate","title":"Generation of profile data — profGenerate","text":"Generates profile (binned) data given range indexed pair   vectors.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generation of profile data — profGenerate","text":"","code":"profBin(x, y, num, xstart = min(x), xend = max(x), param = list()) profBinM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,          param = list()) profBinLin(x, y, num, xstart = min(x), xend = max(x), param = list()) profBinLinM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,             param = list()) profBinLinBase(x, y, num, xstart = min(x), xend = max(x), param = list()) profBinLinBaseM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,                 param = list()) profIntLin(x, y, num, xstart = min(x), xend = max(x), param = list()) profIntLinM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,             param = list()) profMaxIdx(x, y, num, xstart = min(x), xend = max(x), param = list()) profMaxIdxM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,             param = list())"},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generation of profile data — profGenerate","text":"x numeric vector value positions y numeric vector values bin zidx starting position new segment num number equally spaced x bins xstart starting x value xend ending x value NAOK allow NA values (faster) param parameters profile generation","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generation of profile data — profGenerate","text":"functions take vector unequally spaced y values   transform either vector matrix, depending   whether index . point vector   matrix represents data point centered corresponding   x value, plus minus half x step size   (xend-xstart/(num-1)). Bin functions set matrix vector value   maximal point gets binned . BinLin functions except linearly   interpolate values nothing binned. BinLinBase functions except populate   empty parts spectra base value. take two parameters:   1) baselevel, intensity level fill empty parts   spectra. defaluts half minimum intensity. 2)   basespace, m/z length signal drop   base level. Linear interpolation used consecuitive   data points falling within 2*basespace eachother. defaluts   0.075. IntLin functions set matrix vector value   integral linearly interpolated data plus minus   half step size. MaxIdx functions work similarly Bin   functions execpt return integer index x,y   pair placed particular cell.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Generation of profile data — profGenerate","text":"issues profBinLin method, see   https://github.com/sneumann/xcms/issues/46   https://github.com/sneumann/xcms/issues/49. Thus suggested   use functions binYonX combination   imputeLinInterpol instead.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generation of profile data — profGenerate","text":"prof*, numeric vector length num. prof*M, matrix dimensions num  length(zidx). MaxIdx, data type integer, others   double.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generation of profile data — profGenerate","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generation of profile data — profGenerate","text":"","code":"if (FALSE) {     library(faahKO)     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xraw <- xcmsRaw(cdffiles[1])      image(xraw) ## not how with intLin the intensity's blur     profMethod(xraw) <- \"bin\"     image(xraw) ## now with 'bin' there is no blurring good for centroid data     ##try binlinbase for profile data }"},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"The profile matrix — profMat,MsExperiment-method","title":"The profile matrix — profMat,MsExperiment-method","text":"profile matrix n x m matrix, n (rows) representing equally spaced m/z values (bins) m (columns) retention time corresponding scans. cell contains maximum intensity measured specific scan m/z values falling within m/z bin.","code":"The `profMat` method creates a new profile matrix or returns the profile matrix within the object's `@env` slot, if available. Settings for the profile matrix generation, such as `step` (the bin size), `method` or additional settings are extracted from the respective slots of the `xcmsRaw` object. Alternatively it is possible to specify all of the settings as additional parameters.  For [MsExperiment()] or [XcmsExperiment()] objects, the method returns a `list` of profile matrices, one for each sample in `object`. Using parameter `fileIndex` it is also possible to create a profile matrix only for selected samples (files)."},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"The profile matrix — profMat,MsExperiment-method","text":"","code":"# S4 method for MsExperiment profMat(   object,   method = \"bin\",   step = 0.1,   baselevel = NULL,   basespace = NULL,   mzrange. = NULL,   fileIndex = seq_along(object),   chunkSize = 1L,   msLevel = 1L,   BPPARAM = bpparam(),   ... )  # S4 method for xcmsRaw profMat(object, method, step, baselevel, basespace, mzrange.)"},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"The profile matrix — profMat,MsExperiment-method","text":"object xcmsRaw, OnDiskMSnExp, XCMSnExp, MsExperiment XcmsExperiment  object. method character(1) defining profile matrix generation method. Allowed \"bin\", \"binlin\", \"binlinbase\" \"intlin\". See details section information. step numeric(1) representing m/z bin size. baselevel numeric(1) representing base value empty elements (.e. m/z bins without measured intensity) set. considered method = \"binlinbase\". See baseValue parameter imputeLinInterpol() details. basespace numeric(1) representing m/z length signal drop base level. Linear interpolation used consecutive data points falling within 2 * basespace . considered method = \"binlinbase\". specified, defaults 0.075. Internally parameter translated distance parameter imputeLinInterpol() function distance = floor(basespace / step). See distance parameter imputeLinInterpol() details. mzrange. Optional numeric(2) manually specifying mz value range used binnind. provided, whole m/z value range used. fileIndex MsExperiment XcmsExperiment: integer defining idex (indices) sample(s) profile matrix created. chunkSize MsExperiment XcmsExperiment: integer(1) defining number files data loaded processed one iteration. default one file time processed chunkSize = 1L requires less memory. parallel processing, chunkSize >= number parallel processes used. msLevel MsExperiment XcmsExperiment: integer(1) defining MS level profile matrix generated. BPPARAM MsExperiment XcmsExperiment: parallel processing setup. See bpparam() details. Defaults BPPARAM = bpparam(). ... ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"The profile matrix — profMat,MsExperiment-method","text":"profMat returns profile matrix (rows representing scans, columns equally spaced m/z values). object MsExperiment XcmsExperiment, method returns list profile matrices, one file (sample).","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"The profile matrix — profMat,MsExperiment-method","text":"Profile matrix generation methods: \"bin\": default profile matrix generation method simple binning, .e. aggregating intensity values falling within m/z bin. \"binlin\": Binning followed linear interpolation impute missing values. value m/z bins without measured intensity inferred linear interpolation neighboring bins measured intensity. \"binlinbase\": Binning followed linear interpolation impute values empty elements (m/z bins) within user-definable proximity non-empty elements stetting element's value baselevel otherwise. See impute = \"linbase\" parameter imputeLinInterpol() details. \"intlin\": Set elements' values integral linearly interpolated data plus minus half step size.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"The profile matrix — profMat,MsExperiment-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"The profile matrix — profMat,MsExperiment-method","text":"","code":"file <- system.file('cdf/KO/ko15.CDF', package = \"faahKO\") ## Load the data without generating the profile matrix (profstep = 0) xraw <- xcmsRaw(file, profstep = 0) ## Extract the profile matrix profmat <- profMat(xraw, step = 0.3) #> Create profile matrix with method 'bin' and step 0.3 ...  #> OK dim(profmat) #> [1] 1335 1278 ## If not otherwise specified, the settings from the xraw object are used: profinfo(xraw) #> $method #> [1] \"bin\" #>  #> $step #> [1] 0 #>  ## To extract a profile matrix with linear interpolation use profmat <- profMat(xraw, step = 0.3, method = \"binlin\") #> Create profile matrix with method 'binlin' and step 0.3 ...  #> OK ## Alternatively, the profMethod of the xraw objects could be changed profMethod(xraw) <- \"binlin\" profmat_2 <- profMat(xraw, step = 0.3) #> Create profile matrix with method 'binlin' and step 0.3 ...  #> OK all.equal(profmat, profmat_2) #> [1] TRUE"},{"path":"https://sneumann.github.io/xcms/reference/profMedFilt-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Median filtering of the profile matrix — profMedFilt-methods","title":"Median filtering of the profile matrix — profMedFilt-methods","text":"Apply median filter given size profile matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMedFilt-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Median filtering of the profile matrix — profMedFilt-methods","text":"object = \"xcmsRaw\" profMedFilt(object, massrad = 0, scanrad = 0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMedFilt-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Median filtering of the profile matrix — profMedFilt-methods","text":"object xcmsRaw object massrad number m/z grid points either side use median     calculation scanrad number scan grid points either side use median     calculation","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profMethod-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get and set method for generating profile data — profMethod-methods","title":"Get and set method for generating profile data — profMethod-methods","text":"methods get set method generating profile   (matrix) data raw mass spectral data. can currently   bin, binlin, binlinbase, intlin.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMethod-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get and set method for generating profile data — profMethod-methods","text":"object = \"xcmsRaw\" profMethod(object)","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Specify a subset of profile mode data — profRange-methods","title":"Specify a subset of profile mode data — profRange-methods","text":"Specify subset profile mode matrix given mass, time,   scan range. Allow flexible user entry functions.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Specify a subset of profile mode data — profRange-methods","text":"object = \"xcmsRaw\" profRange(object, mzrange = numeric(),                   rtrange = numeric(), scanrange = numeric(),                   ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Specify a subset of profile mode data — profRange-methods","text":"object xcmsRaw object mzrange single numeric mass vector masses rtrange single numeric time (seconds) vector times scanrange single integer scan index vector indecies ... arguments functions","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Specify a subset of profile mode data — profRange-methods","text":"function handles selection mass/time subsets profile   matrix functions. allows user specify   subsets variety flexible ways minimal typing. R partial argument matching, mzrange,   scanrange, rtrange can specified short   form using m=, s=, t=, respectively.   scanrange rtrange specified,   rtrange specification takes precedence. specifying ranges, may either enter single number   numeric vector. single number entered, closest   single scan mass value selected. vector entered,   range set range() values entered.   allows specification ranges using shortened, slightly   non-standard syntax. example, one specify 400 500   seconds using following: t=c(400,500),   t=c(500,400), t=400:500. Use sequence   operator (:) can save several keystrokes specifying   ranges. However, sequence operator works well   specifying integer ranges, fractional ranges always work   well.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Specify a subset of profile mode data — profRange-methods","text":"list folloing items: mzrange numeric vector start end mass masslab textual label mass range massidx integer vector mass indecies scanrange integer vector stat ane end scans scanlab textual label scan range scanidx integer vector scan range rtrange numeric vector start end times timelab textual label time range","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profStep-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get and set m/z step for generating profile data — profStep-methods","title":"Get and set m/z step for generating profile data — profStep-methods","text":"methods get set m/z step generating profile   (matrix) data raw mass spectral data. Smaller steps yield   precision cost greater memory usage.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profStep-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get and set m/z step for generating profile data — profStep-methods","text":"object = \"xcmsRaw\" profStep(object)","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profStep-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get and set m/z step for generating profile data — profStep-methods","text":"","code":"if (FALSE) {     library(faahKO)     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xset <- xcmsRaw(cdffiles[1])      xset     plotSurf(xset, mass=c(200,500))      profStep(xset)<-0.1 ## decrease the bin size to get better resolution     plotSurf(xset, mass=c(200, 500))     ##works nicer on high resolution data.   }"},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate p-values for a vector of t-statistics — pval","title":"Generate p-values for a vector of t-statistics — pval","text":"Generate p-values vector Welch's two-sample t-statistics   based t distribution.","code":""},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate p-values for a vector of t-statistics — pval","text":"","code":"pval(X, classlabel, teststat)"},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate p-values for a vector of t-statistics — pval","text":"X original data matrix classlabel integer vector classlabel teststat numeric vector Welch's two-sample t-statistics","code":""},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate p-values for a vector of t-statistics — pval","text":"numeric vector p-values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate p-values for a vector of t-statistics — pval","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"Generate extracted ion chromatogram m/z values interest.     raw data used contrast  getEIC uses     data profile matrix (.e. values binned along M/Z     dimension).","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"object = \"xcmsRaw\" rawEIC(object, mzrange = numeric(), rtrange = numeric(), scanrange = numeric())","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"object xcmsRaw object mzrange m/z range EIC rtrange retention time range EIC scanrange scan range EIC","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"list : scan scan number intensity added intensity values","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"Ralf Tautenhahn","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get a raw data matrix — rawMat-methods","title":"Get a raw data matrix — rawMat-methods","text":"Returns matrix columns time, m/z, intensity   represents raw data chromatography mass spectrometry experiment.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get a raw data matrix — rawMat-methods","text":"object = \"xcmsRaw\" rawMat(object, mzrange = numeric(), rtrange = numeric(),       scanrange = numeric(), log=FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get a raw data matrix — rawMat-methods","text":"object container raw data mzrange Subset m/z range rtrange Subset retention time range scanrange Subset scan index range log Whether log transform intensities","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get a raw data matrix — rawMat-methods","text":"numeric matrix three columns: time, mz intensity.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get a raw data matrix — rawMat-methods","text":"Michael Lawrence","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":null,"dir":"Reference","previous_headings":"","what":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"Reconstructs MS2 spectra MS1 chromatographic peak (possible) data independent acquisition (DIA) data (SWATH). See LC-MS/MS analysis vignette details examples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"","code":"reconstructChromPeakSpectra(object, ...)  # S4 method for XcmsExperiment reconstructChromPeakSpectra(   object,   expandRt = 0,   diffRt = 2,   minCor = 0.8,   intensity = \"maxo\",   peakId = rownames(chromPeaks(object, msLevel = 1L)),   BPPARAM = bpparam() )  # S4 method for XCMSnExp reconstructChromPeakSpectra(   object,   expandRt = 0,   diffRt = 2,   minCor = 0.8,   intensity = \"maxo\",   peakId = rownames(chromPeaks(object, msLevel = 1L)),   BPPARAM = bpparam(),   return.type = c(\"Spectra\", \"MSpectra\") )"},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"object XCMSnExp identified chromatographic peaks. ... ignored. expandRt numeric(1) allowing expand retention time range extracted ion chromatograms constant value (peak shape correlation). Defaults expandRt = 0 hence correlates signal included identified chromatographic peaks. diffRt numeric(1) defining maximal allowed difference retention time chromatographic peak (apex) retention times MS2 chromatographic peaks (apex) consider representing candidate fragments original ion. minCor numeric(1) defining minimal required correlation coefficient MS2 chromatographic peaks considered MS2 spectrum reconstruction. intensity character(1) defining column chromPeaks matrix used intensities reconstructed spectra's peaks. value MS1 chromatographic peaks used precursorIntensity resulting spectra. peakId optional character vector peak IDs (.e. rownames chromPeaks) MS1 peaks MS2 spectra reconstructed. default reconstructed MS1 chromatographic peaks. BPPARAM parallel processing setup. See bpparam() information. return.type character(1) defining type returned object. return.type = \"Spectra\" supported, return.type = \"MSpectra\" deprecated.","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"Spectra() object (defined Spectra package) reconstructed MS2 spectra MS1 peaks object. Contains empty spectra (.e. without m/z intensity values) MS1 peaks reconstruction possible (either MS2 signal recorded correlation MS2 chromatographic peaks MS1 chromatographic peak threshold minCor. Spectra variables \"ms2_peak_id\" \"ms2_peak_cor\" (type CharacterList() NumericList() length equal number peaks per reconstructed MS2 spectrum) providing IDs correlation MS2 chromatographic peaks MS2 spectrum reconstructed. retention time median retention times MS2 chromatographic peaks used spectrum reconstruction reported. MS1 chromatographic peak intensity reported reconstructed spectrum's precursorIntensity value (see parameter intensity ).","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"detail, function performs MS1 chromatographic peak: Identify MS2 chromatographic peaks isolation window containing m/z ion (.e. MS1 chromatographic peak) approximately retention time MS1 peak (accepted rt shift can specified diffRt parameter). Correlate peak shapes candidate MS2 chromatographic peaks peak shape MS1 peak retaining MS2 chromatographic peaks correlation > minCor. Reconstruct MS2 spectrum using m/z selected MS2 chromatographic peaks intensity (either \"maxo\" \"\"). MS2 chromatographic peak selected MS1 peak thus represent one mass peak reconstructed spectrum. resulting Spectra() object provides also peak IDs MS2 chromatographic peaks spectrum well correlation value spectra variables ms2_peak_id ms2_peak_cor.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"Johannes Rainer, Michael Witting","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"Given matrix rectangular areas, function determines subset   rectangles overlap. Rectangles preserved   first come, first served basis, user control order   rectangles processed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"","code":"rectUnique(m, order = seq(length = nrow(m)), xdiff = 0, ydiff = 0)"},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"m four column matrix defining rectangular areas order order matrix columns scanned xdiff maximum space overlapping rectangles x dimension ydiff maximum space overlapping rectangles y dimension","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"m matrix must contain four colums defining position   rectangle sides folloing order: left, right, bottom, top.   function currently implemented C using algorithm   quadratic running time.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"logical vector indicating rows kept.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"","code":"m <- rbind(c(0,4,0,3), c(1,3,2,6), c(3,6,4,6)) plot(0, 0, type = \"n\", xlim=range(m[,1:2]), ylim=range(m[,3:4])) rect(m[,1], m[,3], m[,2], m[,4])  xcms:::rectUnique(m) #> [1]  TRUE FALSE  TRUE # Changing order of processing xcms:::rectUnique(m, c(2,1,3)) #> [1] FALSE  TRUE FALSE # Requiring border spacing xcms:::rectUnique(m, ydiff = 1) #> [1]  TRUE FALSE FALSE # Allowing adjacent boxes xcms:::rectUnique(m, c(2,1,3), xdiff = -0.00001) #> [1] FALSE  TRUE  TRUE # Allowing interpenetration xcms:::rectUnique(m, xdiff = -1.00001, ydiff = -1.00001) #> [1] TRUE TRUE TRUE"},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Refine Identified Chromatographic Peaks — refineChromPeaks","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"refineChromPeaks method performs post-processing chromatographic peak detection step eventually clean improve results. function can applied XcmsExperiment() XCMSnExp() object peak detection findChromPeaks(). type peak refinement cleaning can defined, along settings, using one following parameter objects: CleanPeaksParam: remove chromatographic peaks retention time range larger provided maximal acceptable width (maxPeakwidth). FilterIntensityParam: remove chromatographic peaks intensities specified threshold. default (nValues = 1) values chromPeaks matrix evaluated: peaks value column defined parameter value >= threshold (defined parameter threshold) retained. nValues larger 1, individual peak intensities raw MS files evaluated: chromatographic peaks least nValues mass peaks >= threshold retained. MergeNeighboringPeaksParam: peak detection sometimes fails identify chromatographic peak correctly, especially broad peaks peak shape irregular (mostly HILIC data). cases several smaller peaks reported. Also, peak detection centWave can result partially completely overlapping peaks. method aims reduce peak detection artifacts merging chromatographic peaks overlapping close RT m/z dimension (considering also measured signal ). See section Details MergeNeighboringPeaksParam details comprehensive description approach. refineChromPeaks methods always remove feature definitions, call method can change remove identified chromatographic peaks, may part features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"","code":"refineChromPeaks(object, param, ...)  # S4 method for XcmsExperiment,CleanPeaksParam refineChromPeaks(object, param = CleanPeaksParam(), msLevel = 1L)  # S4 method for XcmsExperiment,MergeNeighboringPeaksParam refineChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment,FilterIntensityParam refineChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  CleanPeaksParam(maxPeakwidth = 10)  MergeNeighboringPeaksParam(   expandRt = 2,   expandMz = 0,   ppm = 10,   minProp = 0.75 )  FilterIntensityParam(threshold = 0, nValues = 1L, value = \"maxo\")  # S4 method for XCMSnExp,CleanPeaksParam refineChromPeaks(object, param = CleanPeaksParam(), msLevel = 1L)  # S4 method for XCMSnExp,MergeNeighboringPeaksParam refineChromPeaks(   object,   param = MergeNeighboringPeaksParam(),   msLevel = 1L,   BPPARAM = bpparam() )  # S4 method for XCMSnExp,FilterIntensityParam refineChromPeaks(   object,   param = FilterIntensityParam(),   msLevel = 1L,   BPPARAM = bpparam() )"},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"object XCMSnExp XcmsExperiment object identified chromatographic peaks. param Object defining refinement method settings. ... ignored. msLevel integer defining MS level(s) chromatographic peaks cleaned. chunkSize refineChromPeaks object either XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. Peak refinement performed parallel (per sample) subset data. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM parameter object set parallel processing. Uses default parallel processing setup returned bpparam(). See bpparam() details examples. maxPeakwidth CleanPeaksParam: numeric(1) defining maximal allowed peak width (retention time). expandRt MergeNeighboringPeaksParam: numeric(1) defining many seconds retention time window expanded sides check overlapping peaks. expandMz MergeNeighboringPeaksParam: numeric(1) constant value m/z range chromatographic peak expanded (sides!) check overlapping peaks. ppm MergeNeighboringPeaksParam: numeric(1) defining m/z relative value (parts per million) m/z range chromatographic peak expanded check overlapping peaks. minProp MergeNeighboringPeaksParam: numeric(1) 0 1 representing proporion intensity required peaks joined. See description details. default (minProp = 0.75) peaks joined signal half way larger 75% smallest two peak's \"maxo\" (maximal intensity peak apex). threshold FilterIntensityParam: numeric(1) defining threshold peaks removed. nValues FilterIntensityParam: integer(1) defining number data points (chromatographic peak) >= threshold. Defaults nValues = 1. value FilterIntensityParam: character(1) defining name column chromPeaks contains values used filtering.","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"XCMSnExp XcmsExperiment object refined chomatographic peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"details-for-mergeneighboringpeaksparam","dir":"Reference","previous_headings":"","what":"Details for MergeNeighboringPeaksParam","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"peak refinement using MergeNeighboringPeaksParam, chromatographic peaks first expanded m/z retention time dimension (based parameters expandMz, ppm expandRt) subsequently grouped sets merge candidates (expansion) overlapping m/z rt (within sample). Note peak gets expanded expandRt expandMz, thus peaks differing less 2 * expandMz (2 * expandRt) evaluated merging. Peak merging performed along retention time axis, .e., peaks first ordered \"rtmin\" merge candidates defined iteratively starting first peak. Candidate peaks merged average intensity 3 data points middle position (.e., half distance \"rtmax\" first \"rtmin\" second peak) larger certain proportion (minProp) smaller (\"maxo\") intensity peaks. cases calculated mid point located apexes two peaks (e.g., peaks largely overlapping) average signal intensity half way apexes used instead. Candidate peaks merged 3 data points NA intensities. Merged peaks get \"mz\", \"rt\", \"sn\" \"maxo\" values peak largest signal (\"maxo\") well row metadata peak (chromPeakData). \"rtmin\" \"rtmax\" merged peaks updated \"\" recalculated based signal \"rtmin\" \"rtmax\" newly defined \"mzmin\" \"mzmax\" (range \"mzmin\" \"mzmax\" merged peaks expanding expandMz ppm). reported \"mzmin\" \"mzmax\" merged peak represents m/z range non-NA intensities used calculation peak signal (\"\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"Johannes Rainer, Mar Garcia-Aloy","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  #### ## CleanPeaksParam:  ## Distribution of chromatographic peak widths quantile(chromPeaks(faahko_sub)[, \"rtmax\"] - chromPeaks(faahko_sub)[, \"rtmin\"]) #>      0%     25%     50%     75%    100%  #>   6.259  29.734  43.819  57.903 173.710   ## Remove all chromatographic peaks with a width larger 60 seconds data <- refineChromPeaks(faahko_sub, param = CleanPeaksParam(60)) #> Removed 54 of 248 chromatographic peaks.  quantile(chromPeaks(data)[, \"rtmax\"] - chromPeaks(data)[, \"rtmin\"]) #>     0%    25%    50%    75%   100%  #>  6.259 23.475 39.906 46.949 59.469   #### ## FilterIntensityParam:  ## Remove all peaks with a maximal intensity below 50000 res <- refineChromPeaks(faahko_sub,     param = FilterIntensityParam(threshold = 50000)) #> Reduced from 248 to 155 chromatographic peaks.  nrow(chromPeaks(faahko_sub)) #> [1] 248 nrow(chromPeaks(res)) #> [1] 155  #### ## MergeNeighboringPeaksParam:  ## Subset to a single file xd <- filterFile(faahko_sub, file = 1)  ## Example of a split peak that will be merged mzr <- 305.1 + c(-0.01, 0.01) chr <- chromatogram(xd, mz = mzr, rt = c(2700, 3700)) plot(chr)   ## Combine the peaks res <- refineChromPeaks(xd, param = MergeNeighboringPeaksParam(expandRt = 4)) #> Reduced from 87 to 70 chromatographic peaks. chr_res <- chromatogram(res, mz = mzr, rt = c(2700, 3700)) plot(chr_res)   ## Example of a peak that was not merged, because the signal between them ## is lower than the cut-off minProp mzr <- 496.2 + c(-0.01, 0.01) chr <- chromatogram(xd, mz = mzr, rt = c(3200, 3500)) plot(chr) chr_res <- chromatogram(res, mz = mzr, rt = c(3200, 3500)) plot(chr_res)"},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"removeIntensities allows remove intensities chromatographic data matching certain conditions (depending parameter ). intensities actually removed replaced NA_real_. actually remove intensities (associated retention times) use clean() afterwards. Parameter allows specify intensities replaced NA_real_. default (= \"below_threshod\" intensities threshold removed. x XChromatogram XChromatograms object (hence provides also chromatographic peak definitions within object) = \"outside_chromPeak\" can selected removes intensity outside boundaries identified chromatographic peak(s) chromatographic data. Note filterIntensity() might better approach subset/filter chromatographic data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"","code":"# S4 method for Chromatogram removeIntensity(object, which = \"below_threshold\", threshold = 0)  # S4 method for MChromatograms removeIntensity(object, which = \"below_threshold\", threshold = 0)  # S4 method for XChromatogram removeIntensity(   object,   which = c(\"below_threshold\", \"outside_chromPeak\"),   threshold = 0 )"},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"object object representing chromatographic data. Can Chromatogram(), MChromatograms(), XChromatogram() XChromatograms() object. character(1) defining condition remove intensities. See description details options. threshold numeric(1) defining threshold intensities removed (= \"below_threshold\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"input object matching intensities replaced NA.","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"","code":"chr <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(5, 29, 50, NA, 100, 12, 3, 4, 1, 3))  ## Remove all intensities below 20 res <- removeIntensity(chr, threshold = 20) intensity(res) #>  [1]  NA  29  50  NA 100  NA  NA  NA  NA  NA"},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Correct retention time from different samples — retcor-methods","title":"Correct retention time from different samples — retcor-methods","text":"correct differences retention times different   samples, number methods exist XCMS. retcor   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Correct retention time from different samples — retcor-methods","text":"object = \"xcmsSet\" retcor(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correct retention time from different samples — retcor-methods","text":"object xcmsSet-class object method Method use retention time correction. See details. ... Optional arguments passed along","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Correct retention time from different samples — retcor-methods","text":"Different algorithms can used specifying   method argument. example use approach described   Smith et al (2006) one use: retcor(object,   method=\"loess\"). also default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$retcor.methods.   nickname method called \"loess\",   help page specific method can   accessed ?retcor.loess.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correct retention time from different samples — retcor-methods","text":"xcmsSet object corrected retntion times.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Align retention times across samples with Obiwarp — retcor.obiwarp","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"Calculate retention time deviations sample.   based code http://obi-warp.sourceforge.net/.   However, function able align multiple samples,   center-star strategy. original publication see Chromatographic Alignment ESI-LC-MS Proteomics Data Sets       Ordered Bijective Interpolated Warping John T. Prince , Edward       M. Marcotte Analytical Chemistry 2006 78 (17), 6140-6152","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"object = \"xcmsSet\" retcor(object, method=\"obiwarp\", plottype = c(\"none\", \"deviation\"),              profStep=1, center=NULL, col = NULL, ty = NULL,              response=1, distFunc=\"cor_opt\", gapInit=NULL, gapExtend=NULL,             factorDiag=2, factorGap=1, localAlignment=0, initPenalty=0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"object xcmsSet object plottype deviation plot retention time deviation profStep step size (m/z) use profile generation     raw data files center index sample others aligned   . center==NULL, sample peaks chosen default. col vector colors plotting sample ty vector line point types plotting sample response Responsiveness warping.  0 give linear warp based     start end points.  100 use bijective anchors distFunc DistFunc function: cor (Pearson's R) cor_opt (default,     calculate 10% diagonal band distance matrix, better     runtime), cov (covariance), prd (product), euc (Euclidean distance) gapInit Penalty Gap opening, see gapExtend Penalty Gap enlargement, see factorDiag Local weighting applied diagonal moves alignment. factorGap Local weighting applied gap moves alignment. localAlignment Local rather global alignment initPenalty Penalty initiating alignment (local   alignment ) Default: 0 Default gap penalties: (gapInit, gapExtend) [distFunc type]: 'cor' = '0.3,2.4' 'cov' = '0,11.7' 'prd' = '0,7.8' 'euc' = '0.9,1.8'","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"xcmsSet object","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Align retention times across samples — retcor.peakgroups-methods","title":"Align retention times across samples — retcor.peakgroups-methods","text":"two methods use ``well behaved'' peak groups calculate   retention time deviations every time point sample. Use   smoothed deviations align retention times.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Align retention times across samples — retcor.peakgroups-methods","text":"object = \"xcmsSet\" retcor(object, missing = 1, extra = 1,                smooth = c(\"loess\", \"linear\"), span = .2,                family = c(\"gaussian\", \"symmetric\"),                plottype = c(\"none\", \"deviation\", \"mdevden\"),                col = NULL, ty = NULL)","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Align retention times across samples — retcor.peakgroups-methods","text":"object xcmsSet object missing number missing samples allow retention time correction     groups extra number extra peaks allow retention time correction     correction groups smooth either \"loess\" non-linear alignment \"linear\"     linear alignment span degree smoothing local polynomial regression fitting family gaussian fitting least-squares outlier     removal, symmetric re-descending M estimator     used Tukey's biweight function, allowing outlier removal plottype deviation plot retention time deviation points     regression fit, mdevden also plot peak overall     peak density retention time correction peak density col vector colors plotting sample ty vector line point types plotting sample","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Align retention times across samples — retcor.peakgroups-methods","text":"xcmsSet object","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":null,"dir":"Reference","previous_headings":"","what":"Set retention time window to a specified width — retexp","title":"Set retention time window to a specified width — retexp","text":"Expands (contracts) retention time window row   matrix defined retmin retmax columns.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Set retention time window to a specified width — retexp","text":"","code":"retexp(peakrange, width = 200)"},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Set retention time window to a specified width — retexp","text":"peakrange maxtrix columns retmin retmax width new width window","code":""},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Set retention time window to a specified width — retexp","text":"altered matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Set retention time window to a specified width — retexp","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate relative log abundances — rla","title":"Calculate relative log abundances — rla","text":"rla calculates relative log abundances (RLA, see reference) numeric vector.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate relative log abundances — rla","text":"","code":"rla(x, group, log.transform = TRUE)  rowRla(x, group, log.transform = TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate relative log abundances — rla","text":"x numeric (rla) matrix (rowRla) abundances (natural scale) RLA calculated. group factor, numeric character length x groups values x. omitted values considered group. log.transform logical(1) whether x log2 transformed. Set log.transform = FALSE x already log scale.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate relative log abundances — rla","text":"numeric length x (rla) matrix dimensions x (rowRla).","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate relative log abundances — rla","text":"RLA defines (log) abundance analyte relative median across abundances group.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Calculate relative log abundances — rla","text":"De Livera , Dias DA, De Souza D, Rupasinghe T, Pyke J, Tull D, Roessner U, McConville M, Speed TP. Normalizing integrating metabolomics data. Anal Chem 2012 Dec 18;84(24):10768-76.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Calculate relative log abundances — rla","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate relative log abundances — rla","text":"","code":"x <- c(3, 4, 5, 1, 2, 3, 7, 8, 9)  grp <- c(1, 1, 1, 2, 2, 2, 3, 3, 3)  rla(x, grp) #>          1          1          1          2          2          2          3  #> -0.4150375  0.0000000  0.3219281 -1.0000000  0.0000000  0.5849625 -0.1926451  #>          3          3  #>  0.0000000  0.1699250"},{"path":"https://sneumann.github.io/xcms/reference/sampnames-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get sample names — sampnames-methods","title":"Get sample names — sampnames-methods","text":"Return sample names object","code":""},{"path":"https://sneumann.github.io/xcms/reference/sampnames-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get sample names — sampnames-methods","text":"object = \"xcmsEIC\" sampnames(object) object = \"xcmsSet\" sampnames(object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/sampnames-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get sample names — sampnames-methods","text":"character vector sample names.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract processing errors — showError,xcmsSet-method","title":"Extract processing errors — showError,xcmsSet-method","text":"peak detection performed findPeaks     setting argument stopOnError = FALSE eventual errors     process cause stop processing recorded inside     resulting xcmsSet object. errors can     accessed showError method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract processing errors — showError,xcmsSet-method","text":"","code":"# S4 method for xcmsSet showError(object, message. = TRUE, ...)"},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract processing errors — showError,xcmsSet-method","text":"object xcmsSet object. message. Logical indicating whether error message, error returned. ... Additional arguments.","code":""},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract processing errors — showError,xcmsSet-method","text":"list error messages (message. = TRUE) errors     empty list errors present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract processing errors — showError,xcmsSet-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"several methods calculating distance two sets peaks xcms. specDist   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"object = \"xcmsSet\" specDist(object, peakIDs1, peakIDs2,...)  object = \"xsAnnotate\" specDist(object, PSpec1, PSpec2,...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"object xcmsSet xcmsRaw. method Method use distance calculation. See details. ... mzabs, mzppm parameters distance function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"Different algorithms can used specifying   method argument. example use \"meanMZmatch\"   approach xcmsSet one use:   specDist(object, peakIDs1, peakIDs2, method=\"meanMZmatch\"). also   default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$specDist.methods.   nickname method called \"meanMZmatch\",   help page specific method can   accessed ?specDist.meanMZmatch.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"a Distance function based on matching peaks — specDist.cosine","title":"a Distance function based on matching peaks — specDist.cosine","text":"method calculates distance two sets peaks   using cosine-distance.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"a Distance function based on matching peaks — specDist.cosine","text":"","code":"specDist.cosine(peakTable1, peakTable2, mzabs=0.001, mzppm=10, mzExp=0.6,                 intExp=3, nPdiff=2, nPmin=8, symmetric=FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"a Distance function based on matching peaks — specDist.cosine","text":"peakTable1 = \"matrix\", peakTable2 = \"matrix\" specDist.cosine(peakTable1, peakTable2, mzabs = 0.001, mzppm = 10,       mzExp = 0.6, intExp = 3, nPdiff = 2, nPmin = 8,       symmetric = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"a Distance function based on matching peaks — specDist.cosine","text":"result cosine-distance product weighted factors mz intensity matching peaks two peaktables. factors calculated wFact = mz^mzExp * int^intExp. distance calculated (example matching peaks found) return-value NA.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"a Distance function based on matching peaks — specDist.cosine","text":"peakTable1 Matrix containing least m/z-values, \trow must called \"mz\" peakTable2 matrix mz-values mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match mzExp exponent used mz intExp exponent used intensity nPdiff maximum nrow-difference two peaktables nPmin minimum absolute sum peaks praktables","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"a Distance function based on matching peaks — specDist.cosine","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"a Distance function based on matching peaks — specDist.meanMZmatch","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"method calculates distance two sets peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"","code":"specDist.meanMZmatch(peakTable1, peakTable2, matchdist=1, matchrate=1,                      mzabs=0.001, mzppm=10, symmetric=TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"peakTable1 = \"matrix\", peakTable2 = \"matrix\" specDist.meanMZmatch(peakTable1, peakTable2,     matchdist=1, matchrate=1,     mzabs=0.001, mzppm=10, symmetric=TRUE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"result calculation weighted sum two values. Value one mean absolute difference matching peaks, value two relation matching peaks non matching peaks. distance calculated (example matching peaks found) return-value NA.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"peakTable1 Matrix containing least m/z-values, \trow must called \"mz\" peakTable2 matrix mz-values mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match matchdist weight value one (see details) matchrate weight value two","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"a Distance function based on matching peaks — specDist.peakCount-methods","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"method calculates distance two sets peaks just returning number matching peaks (m/z-values).","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"","code":"specDist.peakCount(peakTable1, peakTable2, mzabs=0.001, mzppm=10, symmetric=FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"peakTable1 = \"matrix\", peakTable2 = \"matrix\" specDist.peakCount(peakTable1, peakTable2, mzppm=10,symmetric=FALSE )","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"peakTable1 Matrix containing least m/z-values, \trow must called \"mz\" peakTable2 matrix mz-values mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate noise for a sparse continuum mass spectrum — specNoise","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"Given sparse continuum mass spectrum, determine regions   signal present, substituting half minimum intensity   regions. Calculate noise level weighted mean   regions signal regions without signal.   one raw peak, return zero.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"","code":"specNoise(spec, gap = quantile(diff(spec[, \"mz\"]), 0.9))"},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"spec matrix named columns mz intensity gap threshold data points considerd separated     blank region bridged interpolating line","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"default gap value determined 90th percentile   pair-wise differences adjacent mass values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"numeric noise level","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify peaks in a sparse continuum mode spectrum — specPeaks","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"Given spectrum, identify list significant peaks determined   several criteria.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"","code":"specPeaks(spec, sn = 20, mzgap = 0.2)"},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"spec matrix named columns mz intensity sn minimum signal noise ratio mzgap minimal distance adjacent peaks, smaller peaks     excluded","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"Peaks must meet two criteria considered peaks: 1)   s/n ratio must exceed certain threshold. 2) must   within given distance greater intensity peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"matrix columns: mz m/z maximum peak intensity intensity maximum intensity peak fwhm full width half max peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Divide an xcmsRaw object — split.xcmsRaw","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"Divides scans xcmsRaw object   list multiple objects. MS$^n$ data discarded.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"xr = \"xcmsRaw\" split(x, f, drop = TRUE, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"x xcmsRaw object f factor factor(f) defines scans go     new xcmsRaw objects drop logical indicating levels occur dropped     ('f' 'factor' list). ... potential arguments passed methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"list xcmsRaw objects.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"Steffen Neumann, sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"Divide an xcmsSet object — split.xcmsSet","title":"Divide an xcmsSet object — split.xcmsSet","text":"Divides samples peaks xcmsSet object   list multiple objects. Group data discarded.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Divide an xcmsSet object — split.xcmsSet","text":"xs = \"xcmsSet\" split(x, f, drop = TRUE, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Divide an xcmsSet object — split.xcmsSet","text":"xs xcmsSet object f factor factor(f) defines grouping drop logical indicating levels occur dropped     ('f' 'factor' list). ... potential arguments passed methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Divide an xcmsSet object — split.xcmsSet","text":"list xcmsSet objects.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Divide an xcmsSet object — split.xcmsSet","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Correct gaps in data — stitch-methods","title":"Correct gaps in data — stitch-methods","text":"Fixes gaps data due calibration scans lock mass. Automatically detects file type   calls relevant method. mzXML file keeps data length time   overwrites lock mass scans. netCDF version adds scans back data thereby   increasing length data correcting unseen gap.","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Correct gaps in data — stitch-methods","text":"object = \"xcmsRaw\" stitch(object, lockMass=numeric())  object = \"xcmsRaw\" makeacqNum(object, freq=numeric(), start=1)","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correct gaps in data — stitch-methods","text":"object xcmsRaw-class object lockMass dataframe locations gaps freq intervals lock mass scans start starting lock mass scan location, default 1","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Correct gaps in data — stitch-methods","text":"makeacqNum takes locates gap using starting lock mass scan intervals. data frame used \tstitch correct gap caused lock mass. Correction works using scans either side gap fill .","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correct gaps in data — stitch-methods","text":"stitch corrected xcmsRaw-class object  makeacqNum numeric vector scan locations corresponding lock Mass scans","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correct gaps in data — stitch-methods","text":"Paul Benton, hpaul.benton08@imperial.ac.uk","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Correct gaps in data — stitch-methods","text":"","code":"if (FALSE) library(xcms)     library(faahKO)     ## These files do not have this problem to correct for but just     ## for an example     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xr<-xcmsRaw(cdffiles[1]) #> Create profile matrix with method 'bin' and step 1 ...  #> OK     xr #> An \"xcmsRaw\" object with 1278 mass spectra #>  #> Time range: 2501.4-4499.8 seconds (41.7-75 minutes) #> Mass range: 200-600 m/z #> Intensity range: 70-1373180  #>  #> MSn data on  0  mass(es) #> \twith  0  MSn spectra #> Profile method: bin  #> Profile step: 1 m/z (401 grid points from 200 to 600 m/z) #>  #> Memory usage: 10.8 MB     ##Lets assume that the lockmass starts at 1 and is every 100 scans     lockMass<-xcms:::makeacqNum(xr, freq=100, start=1)     ## these are equcal     lockmass<-AutoLockMass(xr) #> Warning:  #> Lock mass frequency wasn't detected     ob<-stitch(xr, lockMass)     ob #> An \"xcmsRaw\" object with 1304 mass spectra #>  #> Time range: 1.6-2040.7 seconds (0-34 minutes) #> Mass range: 200-600 m/z #> Intensity range: 70-1373180  #>  #> MSn data on  0  mass(es) #> \twith  0  MSn spectra #> Profile method: bin  #> Profile step: no profile data #>  #> Memory usage: 6.96 MB      ## plot the old data before correction     foo<-rawEIC(xr, m=c(200,210), scan=c(80,140))     plot(foo$scan, foo$intensity, type=\"h\")       ## plot the new corrected data to see what changed     foo<-rawEIC(ob, m=c(200,210), scan=c(80,140))     plot(foo$scan, foo$intensity, type=\"h\")"},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":null,"dir":"Reference","previous_headings":"","what":"Save xcms result objects in a specified format — storeResults","title":"Save xcms result objects in a specified format — storeResults","text":"storeResults function saves object resulting processing xcms package (mainly XcmsExperiment). Multiple formats storing exporting available can defined param argument. Supported param objects : RDataParam: Save .RData format file. name file can specified fileName argument. PlainTextParam: Store MsExperiment XcmsExperiment objects folder plain text files, folder path defined path argument. MzTabMParam: Save MzTab format (defined). specific examples, see help pages individual parameter classes listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Save xcms result objects in a specified format — storeResults","text":"","code":"storeResults(object, param, ...)"},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save xcms result objects in a specified format — storeResults","text":"object MsExperiment XcmsExperiment data object needs saved. param parameter object selecting configuring format saving. can one following classes: RDataParam, PlainTextParam, MzTabMParam. ... Optional parameters.","code":""},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Save xcms result objects in a specified format — storeResults","text":"Philippine Louail","code":""},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Save xcms result objects in a specified format — storeResults","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Set up parameter to save as .RData file param <- RDataParam(fileName = \"example_xcms_results\")  ## save as .RData storeResults(object = faahko_sub, param = param)  ## Set up parameter to save as a collection of plain text file param <- PlainTextParam(path = \"test/path/\")  ## Save as a collection of plain text files storeResults(object = faahko_sub, param = param)"},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"Subset xcmsRaw object scans.     returned xcmsRaw object contains values     scans specified argument . Note scanrange     slot returned xcmsRaw     c(1, length(object@scantime)) hence range().","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"","code":"# S4 method for xcmsRaw,logicalOrNumeric,missing,missing [(x, i, j, drop)"},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"x xcmsRaw object sub-setted. Integer logical vector specifying scans/spectra x sub-setted. j supported. drop supported.","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"sub-setted xcmsRaw object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"subsetting scan index increasing order logical     vector supported. ordered, argument sorted     automatically. Indices larger total number scans     discarded.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"","code":"## Load a test file file <- system.file('cdf/KO/ko15.CDF', package = \"faahKO\") xraw <- xcmsRaw(file, profstep = 0) ## The number of scans/spectra: length(xraw@scantime) #> [1] 1278  ## Subset the object to scans with a scan time from 3500 to 4000. xsub <- xraw[xraw@scantime >= 3500 & xraw@scantime <= 4000] range(xsub@scantime) #> [1] 3501.383 3999.039 ## The number of scans: length(xsub@scantime) #> [1] 319 ## The number of values of the subset: length(xsub@env$mz) #> [1] 130569"},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Update an xcmsSet object — updateObject,xcmsSet-method","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"method updates old xcmsSet     object latest definition.","code":""},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"","code":"# S4 method for xcmsSet updateObject(object, ..., verbose = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"object xcmsSet object update. ... Optional additional arguments. Currently ignored. verbose Currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"updated xcmsSet containing data input object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":null,"dir":"Reference","previous_headings":"","what":"Enable usage of old xcms code — useOriginalCode","title":"Enable usage of old xcms code — useOriginalCode","text":"function allows enable usage old, partially deprecated code xcms setting corresponding global option. See details functions affected.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Enable usage of old xcms code — useOriginalCode","text":"","code":"useOriginalCode(x)"},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Enable usage of old xcms code — useOriginalCode","text":"x logical(1) specify whether original old code used corresponding functions. provided function simply returns value global option.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Enable usage of old xcms code — useOriginalCode","text":"logical(1) indicating whether old code used.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Enable usage of old xcms code — useOriginalCode","text":"functions/methods affected option : do_findChromPeaks_matchedFilter: use original code iteratively creates subset binned (profile) matrix. helpful computers limited memory matchedFilter settings small bin size. getPeaks","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Enable usage of old xcms code — useOriginalCode","text":"parallel processing using SOCKS method (e.g. SnowParam() Windows computers) option might passed individual R processes performing calculations. cases suggested specify option manually system-wide adding line options(XCMSuseOriginalCode = TRUE) file called .Rprofile folder new R processes started (usually user's home directory; ensure option correctly read add new line file ). See also Startup base R documentation specify system-wide options R. Usage old code strongly dicouraged. function thought used mainly transition phase xcms xcms version 3.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Enable usage of old xcms code — useOriginalCode","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":null,"dir":"Reference","previous_headings":"","what":"Verify an mzQuantML file — verify.mzQuantM","title":"Verify an mzQuantML file — verify.mzQuantM","text":"Export XML data formats: verify written data","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Verify an mzQuantML file — verify.mzQuantM","text":"","code":"verify.mzQuantML(filename, xsdfilename)"},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Verify an mzQuantML file — verify.mzQuantM","text":"filename filename (may include full path)     output file. Pipes URLs allowed. xsdfilename Filename XSD verify (may include full path)","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Verify an mzQuantML file — verify.mzQuantM","text":"verify.mzQuantML() function verify PSI standard format mzQuantML   document XSD schemda, see http://www.psidev.info/mzquantml","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Verify an mzQuantML file — verify.mzQuantM","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Save an xcmsRaw object to file — write.cdf-methods","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"Write raw data (simple) CDF file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"object = \"xcmsRaw\" write.cdf(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"object xcmsRaw object filename filename (may include full path)     CDF file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"Currently application known read resulting file   XCMS. Others, especially build AndiMS library,   refuse load output.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":null,"dir":"Reference","previous_headings":"","what":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"Export XML data formats:    Write processed data xcmsSet mzQuantML.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"object = \"xcmsSet\" write.mzQuantML(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"object xcmsRaw xcmsSet object filename filename (may include full path)     output file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"write.mzQuantML() function write (grouped) xcmsSet   PSI standard format mzQuantML, see http://www.psidev.info/mzquantml","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Save an xcmsRaw object to a file — write.mzdata-methods","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"Write raw data (simple) mzData file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"object = \"xcmsRaw\" write.mzdata(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"object xcmsRaw object filename filename (may include full path)     mzData file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"function export given xcmsRaw object mzData file.   mzData file contain <spectrumList> containing   <spectrum> mass intensity values 32 bit precision.   formats currently supported. header information   (e.g. additional <software> information <cvParams>) lost.   Currently, also MSn information stored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"writeMSData exports mass spectrometry data mzML mzXML format. adjusted retention times present, used retention time exported spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"","code":"# S4 method for XCMSnExp,character writeMSData(   object,   file,   outformat = c(\"mzml\", \"mzxml\"),   copy = FALSE,   software_processing = NULL,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"object XCMSnExp object mass spectrometry data. file character file name(s). length parameter match number files/samples object. outformat character(1) defining format output files ( either \"mzml\" \"mzxml\"). copy logical(1) metadata (data processing, software used, original file names etc) copied original files. software_processing optionally provide specific data processing steps. See documentation software_processing parameter mzR::writeMSData(). ... Additional parameters pass writeMSData() function MSnbase package, outformat specify output format (\"mzml\" \"mzxml\") copy specify whether general information original MS data files (data processing, software etc) copied new files.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":null,"dir":"Reference","previous_headings":"","what":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"Write grouped xcmsSet mzTab file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"object = \"xcmsSet\" writeMzTab(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"object xcmsSet object filename filename (may include full path)     mzTab file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"mzTab file format MS-based metabolomics (proteomics)    lightweight    supplement existing standard XML-based file formats (mzML,    mzIdentML, mzQuantML), providing comprehensive summary, similar    concept supplemental material scientific    publication. mzTab files xcms contain small    molecule sections together experimental metadata    basic quantitative information. format intended store    simple summary final results.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"","code":"library(faahKO)     xs <- group(faahko)      mzt <- data.frame(character(0))     mzt <- xcms:::mzTabHeader(mzt,                        version=\"1.1.0\", mode=\"Complete\", type=\"Quantification\",                        description=\"faahKO\",                        xset=xs)     mzt <- xcms:::mzTabAddSME(mzt, xs)          xcms:::writeMzTab(mzt, \"faahKO.mzTab\")"},{"path":"https://sneumann.github.io/xcms/reference/xcms-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated functions in package ‘xcms’ — xcms-deprecated","title":"Deprecated functions in package ‘xcms’ — xcms-deprecated","text":"functions provided compatibility older versions   ‘xcms’ , defunct next release.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcms-deprecated.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Deprecated functions in package ‘xcms’ — xcms-deprecated","text":"following functions/methods deprecated.  profBin, profBinM, profBinLin,       profBinLinM, profBinLinBase, profBinLinBaseM       deprecated binYonX combination     imputeLinInterpol used instead. extractMsData: replaced (x, \"data.frame\"). plotMsData: replaced plot(x, type = \"XIC\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"class used store plot parallel extracted ion   chromatograms multiple sample files. integrates   xcmsSet class display peak area integrated peak   identification fill-.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"Objects can created getEIC method   xcmsSet class. Objects can also created calls   form new(\"xcmsEIC\", ...).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"eic: list containing named entries every sample. entry,       list two column EIC matricies retention time intensity mzrange: two column matrix containing starting ending m/z EIC rtrange: two column matrix containing starting ending time       EIC rt: either \"raw\" \"corrected\" specify retention       times contained object groupnames: group names xcmsSet object used generate EICs","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"groupnames signature(object = \"xcmsEIC\"): get groupnames slot mzrange signature(object = \"xcmsEIC\"): get mzrange slot plot signature(x = \"xcmsEIC\"): plot extracted ion       chromatograms rtrange signature(object = \"xcmsEIC\"): get rtrange slot sampnames signature(object = \"xcmsEIC\"): get sample names","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"notes yet.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Base class for loading raw data from a file — xcmsFileSource-class","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"Data sources read data file inherit   class. xcms package provides classes read   netCDF, mzData, mzXML, mzML files   using xcmsFileSource. class considered virtual work passed   loadRaw-methods. reason explicitly   virtual appear way class   virtual data part (lets functions treat objects   character strings). class validates file exists path given.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"xcmsFileSource objects instantiated directly.   Instead, create subclasses instantiate .","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":".Data: Object class \"character\". File path       file read raw data object's data part","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"extends","dir":"Reference","previous_headings":"","what":"Extends","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"Class \"character\", data part.   Class \"xcmsSource\", directly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"xcmsSource signature(object = \"character\"): Create       xcmsFileSource object referencing given file name.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"Daniel Hackney dan@haxney.org","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"class similar xcmsSet stores peaks   number individual files. However,  xcmsFragments keeps   Tandem MS e.g. Ion Trap Orbitrap MS$^n$ peaks, including   parent ion relationships.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"Objects can created xcmsFragments   constructor filled peaks using collect method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"peaks: matrix colmns peakID (MS1 parent corresponding xcmsSet),       MSnParentPeakID (parent peak within xcmsFragments), msLevel       (e.g. 2 Tandem MS), rt (retention time case LC data), mz       (fragment mass--charge), intensity (peak intensity extracted       original xcmsSet), sample (index rawData-file). MS2spec: list matrixes. matrix list single collected spectra collect. column ID's mz, intensity, full width half maximum(fwhm). fwhm column relevant spectra came profile data. specinfo: matrix reference data spectra MS2spec. column id's preMZ, AccMZ, rtmin, rtmax, ref, CollisionEnergy. preMZ precursor mass MS1 scan. mass given XML file. instruments mass given nominal mass, therefore AccMZ given weighted average mass MS1 scan collected spectra. retention time given rtmin rtmax. ref column pointer MS2spec matrix spectra. collisionEnergy column collision Energy spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"collect signature(object = \"xcmsFragments\"): gets xcmsSet-object, collects ms1-peaks msn-peaks corresponding xcmsRaw-files. plotTree signature(object = \"xcmsFragments\"): prints (text based) pseudo-tree peaktable display dependencies peaks among . show signature(object = \"xcmsFragments\"): print human-readable       description object console.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"S. Neumann, J. Kutzera","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":null,"dir":"Reference","previous_headings":"","what":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"EXPERIMANTAL FEATURE xcmsFragments object similar xcmsSet, holds peaks   picked (collected) one several xcmsRaw objects. still discussions going exact API MS$^n$   data, likely change future. code yet   pipeline-ified.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"","code":"xcmsFragments(xs, ...)"},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"xs xcmsSet-class object contains     picked ms1-peaks one several experiments ... arguments collect method","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"running collect(xFragments,xSet) peaktable   xcmsFragments includes ms1Peaks experinemts stored   xcmsSet-object. contains relevant MSn-peaks   xcmsRaw-objects, created temporarily paths   xcmsSet.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"xcmsFragments object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"Joachim Kutzera, Steffen Neumann, sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":null,"dir":"Reference","previous_headings":"","what":"A matrix of peaks — xcmsPeaks-class","title":"A matrix of peaks — xcmsPeaks-class","text":"matrix peak information. actual columns depend   generated (.e. findPeaks method).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"A matrix of peaks — xcmsPeaks-class","text":"Objects can created calls form new(\"xcmsPeaks\", ...).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"A matrix of peaks — xcmsPeaks-class","text":".Data: matrix holding peak information","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"extends","dir":"Reference","previous_headings":"","what":"Extends","title":"A matrix of peaks — xcmsPeaks-class","text":"Class \"matrix\", data part. Class \"array\", class \"matrix\", distance 2. Class \"structure\", class \"matrix\", distance 3. Class \"vector\", class \"matrix\", distance 4, explicit coerce.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"A matrix of peaks — xcmsPeaks-class","text":"None yet. utilities working peak data nice.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"A matrix of peaks — xcmsPeaks-class","text":"Michael Lawrence","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"class handles processing visualization raw data   single LC/MS GS/MS run. includes methods producing   standard suite plots including individual spectra, multi-scan   average spectra, TIC, EIC. also produce feature   list significant peaks using matched filtration.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"Objects can created xcmsRaw constructor   reads data NetCDF file new object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"acquisitionNum: Numeric representing acquisition       number individual scans/spectra. Length       acquisitionNum equal number spectra/scans       object hence equal scantime slot. Note however       information available mzML files. env: environment three variables: mz - concatenated       m/z values scans, intensity - corresponding       signal intensity m/z value, profile -       matrix represention intensity values columns       representing scans rows representing equally spaced m/z       values. profile matrix extracted       profMat method. filepath: Path raw data file gradient: matrix first row, time, containing time point       interpolation successive columns representing solvent       fractions point msnAcquisitionNum: scan unique acquisition       number reported via \"spectrum id\" (mzData) \"<scan num=...>\"       \"<scanOrigin num=...>\" (mzXML) msnCollisionEnergy: \"CollisionEnergy\" (mzData)       \"collisionEnergy\" (mzXML) msnLevel: scan \"msLevel\" (mzData mzXML) msnPrecursorCharge: \"ChargeState\" (mzData)       \"precursorCharge\" (mzXML) msnPrecursorIntensity: \"Intensity\" (mzData)       \"precursorIntensity\" (mzXML) msnPrecursorMz: \"MassToChargeRatio\" (mzData)       \"precursorMz\" (mzXML) msnPrecursorScan: \"spectrumRef\" (mzData mzXML) msnRt: Retention time scan msnScanindex: msnScanindex mzrange: numeric vector length 2 minimum maximum m/z values       represented profile matrix polarity: polarity profmethod: characer value name method used generating       profile matrix. profparam: list store additional profile matrix       generation settings. Use profinfo method       extract profile matrix creation relevant information. scanindex: integer vector starting positions scan       mz intensity variables (note index       values based 0 initial position instead 1). scantime: numeric vector acquisition time (seconds)       scan. tic: numeric vector total ion count (intensity) scan mslevel: Numeric representing MS level present MS1       slot. slot accessed getter method       mslevel. scanrange: Numeric length 2 specifying scan range (NULL       full range). slot accessed getter       method scanrange. Note scanrange       always 1 number scans within xcmsRaw       object, necessarily match scan index       original mzML file (e.g. original data sub-setted).       acquisitionNum information can used track       original position scan mzML file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"findPeaks signature(object = \"xcmsRaw\"): feature detection using       matched filtration chromatographic time domain getEIC signature(object = \"xcmsRaw\"): get extracted ion       chromatograms specified m/z ranges. return total   ion chromatogram (TIC) m/z range corresponds full m/z   range (.e. sum signals per retention time across m/z). getPeaks signature(object = \"xcmsRaw\"): get data peaks       specified m/z time ranges getScan signature(object = \"xcmsRaw\"): get m/z intensity       values single mass scan getSpec signature(object = \"xcmsRaw\"): get average m/z       intensity values multiple mass scans image signature(x = \"xcmsRaw\"): get data peaks       specified m/z time ranges levelplot Create image raw (profile) data m/z retention   time, intensity color coded. mslevel Getter method mslevel slot. plotChrom signature(object = \"xcmsRaw\"): plot chromatogram       profile data plotRaw signature(object = \"xcmsRaw\"): plot locations raw       intensity data points plotScan signature(object = \"xcmsRaw\"): plot mass spectrum       individual scan raw data plotSpec signature(object = \"xcmsRaw\"): plot mass spectrum       profile data plotSurf signature(object = \"xcmsRaw\"): experimental method       plotting 3D surface profile data rgl. plotTIC signature(object = \"xcmsRaw\"): plot total ion count       chromatogram profinfo signature(object = \"xcmsRaw\"): returns list containing   profile generation method step (profile m/z step size)   eventual additional parameters profile function. profMedFilt signature(object = \"xcmsRaw\"): median filter profile       data time m/z dimensions profMethod<- signature(object = \"xcmsRaw\"): change method       generating profile matrix profMethod signature(object = \"xcmsRaw\"): get method       generating profile matrix profMz signature(object = \"xcmsRaw\"): get vector m/z values       row profile matrix profRange signature(object = \"xcmsRaw\"): interpret flexible ways       specifying subsets profile matrix profStep<- signature(object = \"xcmsRaw\"): change m/z step       used generating profile matrix profStep signature(object = \"xcmsRaw\"): get m/z step used       generating profile matrix revMz signature(object = \"xcmsRaw\"): reverse order       data points scan scanrange Getter method scanrange slot. See slot description       information. sortMz signature(object = \"xcmsRaw\"): sort data points       increasing m/z scan stitch signature(object = \"xcmsRaw\"): Raw data correction       lock mass calibration gaps. findmzROI signature(object = \"xcmsRaw\"): \tinternal function identify regions interest raw \tdata part first step centWave-based peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"Colin . Smith, csmith@scripps.edu,   Johannes Rainer johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"function handles task reading NetCDF/mzXML file containing   LC/MS GC/MS data new xcmsRaw object. also   transforms data profile (maxrix) mode efficient   plotting data exploration.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"","code":"xcmsRaw(filename, profstep = 1, profmethod = \"bin\", profparam = list(), includeMSn=FALSE, mslevel=NULL, scanrange=NULL)  deepCopy(object)"},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"filename path name NetCDF mzXML file read profstep step size (m/z) use profile generation profmethod method use profile generation. See     profile-matrix details supported values. profparam extra parameters use profile generation includeMSn XML file formats: also read MS$^n$     (Tandem-MS Ion-/Orbi- Trap spectra) mslevel move data mslevel normal MS1 slots, e.g.     peak picking visualisation scanrange scan range read object xcmsRaw object","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"See profile-matrix details profile matrix   generation methods settings. scanrange import can restricted, otherwise MS1 data   read. profstep set 0, profile matrix generated.   Unless includeMSn = TRUE first level MS data read, MS/MS,   etc. deepCopy(xraw) create copy xcmsRaw object   copy mz intensity data xraw@env.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"xcmsRaw object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"NetCDF file format:   https://www.unidata.ucar.edu/software/netcdf/   http://www.astm.org/Standards/E2077.htm   http://www.astm.org/Standards/E2078.htm mzXML file format:   http://sashimi.sourceforge.net/software_glossolalia.html PSI-MS working group developed mzData mzML file formats:   http://www.psidev.info/index.php?q=node/80 Parser used XML file formats:   http://tools.proteomecenter.org/wiki/index.php?title=Software:RAMP","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"","code":"if (FALSE) {     library(xcms)     library(faahKO)     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xr<-xcmsRaw(cdffiles[1])     xr     ##This gives some information about the file     names(attributes(xr))     ## Lets have a look at the structure of the object      str(xr)     ##same but with a preview of each slot in the object     ##SO... lets have a look at how this works     head(xr@scanindex)     ##[1]    0  429  860 1291 1718 2140     xr@env$mz[425:430]     ##[1] 596.3 597.0 597.3 598.1 599.3 200.1     ##We can see that the 429 index is the last mz of scan 1 therefore...      mz.scan1<-xr@env$mz[(1+xr@scanindex[1]):xr@scanindex[2]]     intensity.scan1<-xr@env$intensity[(1+xr@scanindex[1]):xr@scanindex[2]]     plot(mz.scan1, intensity.scan1, type=\"h\",          main=paste(\"Scan 1 of file\", basename(cdffiles[1]), sep=\"\"))     ##the easier way :p     scan1<-getScan(xr, 1)     head(scan1)     plotScan(xr, 1)   }"},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"class transforms set peaks multiple LC/MS GC/MS   samples matrix preprocessed data. groups peaks   nonlinear retention time correction without internal   standards. fills missing peak values raw data. Lastly,   generates extracted ion chromatograms ions interest.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"Objects can created xcmsSet constructor   gathers peaks set NetCDF files. Objects can also   created calls form new(\"xcmsSet\", ...).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"peaks matrix containing peak data. filled vector peak indices peaks added       fillPeaks method. groups Matrix containing statistics peak groups. groupidx List containing indices peaks group. phenoData data.frame containing experimental design factors. rt list containing two lists, raw corrected,       containing retention times every scan every sample. filepaths Character vector absolute path name NetCDF file. profinfo list containing values method - profile generation       method, step - profile m/z step size eventual       additional parameters profile function. dataCorrection logical vector filled waters Lock mass correction       parameter used. polarity string (\"positive\" \"negative\" NULL) describing whether       positive negative scans used reading raw data. progressInfo Progress informations xcms functions (GUI). progressCallback Function called, progressInfo changes (GUI). mslevel Numeric representing MS level peak picking       performed (default MS1). slot accessed       getter method mslevel. scanrange Numeric length 2 specifying scan range (NULL       full range). slot accessed getter       method scanrange. scan range provided slot       represents scans whole raw data subsetted. .processHistory Internal slot used keep track performed processing       steps. slot directly accessed user.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"c signature(\"xcmsSet\"): combine objects together filepaths<- signature(object = \"xcmsSet\"): set filepaths slot filepaths signature(object = \"xcmsSet\"): get filepaths slot diffreport signature(object = \"xcmsSet\"): create report       differentially regulated ions including EICs fillPeaks signature(object = \"xcmsSet\"): fill peak data       groups missing peaks getEIC signature(object = \"xcmsSet\"): get list EICs       sample set getXcmsRaw signature(object = \"xcmsSet\", sampleidx = 1, \tprofmethod = profMethod(object), profstep = profStep(object), \tprofparam=profinfo(object), mslevel = NULL, scanrange = NULL, \trt=c(\"corrected\", \"raw\"), BPPARAM = bpparam()): read raw       data one files xcmsSet return       . default parameters apply settings used       original xcmsSet call generate xcmsSet       object applied also raw data. Parameter       sampleidx allows specify raw file(s)       loaded. Argument BPPARAM allows setup parallel       processing. groupidx<- signature(object = \"xcmsSet\"): set groupidx slot groupidx signature(object = \"xcmsSet\"): get groupidx slot groupnames signature(object = \"xcmsSet\"): get textual names       peak groups groups<- signature(object = \"xcmsSet\"): set groups slot groups signature(object = \"xcmsSet\"): get groups slot groupval signature(object = \"xcmsSet\"): get matrix values       peak data row peak group group signature(object = \"xcmsSet\"): find groups peaks       across samples share similar m/z retention times mslevel Getter method mslevel slot. peaks<- signature(object = \"xcmsSet\"): set peaks slot peaks signature(object = \"xcmsSet\"): get peaks slot plotrt signature(object = \"xcmsSet\"): plot retention time       deviation profiles profinfo<- signature(object = \"xcmsSet\"): set profinfo slot profinfo signature(object = \"xcmsSet\"): get profinfo slot profMethod signature(object = \"xcmsSet\"): extract method used       generate profile matrix. profStep signature(object = \"xcmsSet\"): extract profile step       used generation profile matrix. retcor signature(object = \"xcmsSet\"): use initial grouping       peaks nonlinear loess retention time correction sampclass<- signature(object = \"xcmsSet\"): Replaces column       “class” phenoData slot. See details information. sampclass signature(object = \"xcmsSet\"): Returns content       column “class” phenoData slot ,       present, interaction experimental design factors       (.e. phenoData data.frame). See details       information. phenoData<- signature(object = \"xcmsSet\"): set phenoData slot phenoData signature(object = \"xcmsSet\"): get phenoData slot progressCallback<- signature(object = \"xcmsSet\"): set progressCallback slot progressCallback signature(object = \"xcmsSet\"): get progressCallback slot scanrange Getter method scanrange slot. See scanrange slot       description details. sampnames<- signature(object = \"xcmsSet\"): set rownames       phenoData slot sampnames signature(object = \"xcmsSet\"): get rownames       phenoData slot split signature(\"xcmsSet\"): divide xcmsSet list       xcmsSet objects depending provided factor. Note       peak data preserved, .e. eventual peak grouping information       lost. object$name, object$name<-value Access set name column phenoData object[, ] Conducts subsetting xcmsSet instance. subsetting       columns, .e. samples, supported. Subsetting performed       slots, also groups groupidx. Parameter       can integer vector, logical vector character       vector sample names (matching sampnames).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"phenoData slot (phenoData parameter   xcmsSet function) intended contain data.frame describing   experimental factors, .e. samples along   properties. data.frame contains column named   “class”, returned sampclass method   thus used methods determine sample   grouping/class assignment (e.g. define colors various plots   group method). sampclass<- method adds replaces “class”   column phenoData slot. data.frame   submitted method, interaction columns   stored “class” column. Also, similar classes Bioconductor, $ method   can used directly access columns phenoData   slot (e.g. use xset$name xcmsSet object called   “xset” extract values column named “name” phenoData slot).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"Colin . Smith, csmith@scripps.edu, Johannes Rainer johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"function handles construction xcmsSet objects.   finds peaks batch mode pre-sorts files subdirectories   different classes suitable grouping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"","code":"xcmsSet(files = NULL, snames = NULL, sclass = NULL, phenoData = NULL,         profmethod = \"bin\", profparam = list(),         polarity = NULL, lockMassFreq=FALSE,   mslevel=NULL, nSlaves=0, progressCallback=NULL,         scanrange = NULL, BPPARAM = bpparam(),         stopOnError = TRUE, ...)"},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"files path names NetCDF/mzXML files read snames sample names. default file name without extension     used. sclass sample classes. phenoData data.frame AnnotatedDataFrame     defining sample names classes sample related     properties. provided, argument sclass     subdirectories samples stored     used specify sample grouping. profmethod Method use profile generation. Supported     values \"bin\", \"binlin\", \"binlinbase\"     \"intlin\" (methods profBin,     profBinLin, profBinLinBase     profIntLin, respectively). See help     profBin complete list available methods     supported parameters. profparam parameters use profile generation. polarity filter raw data positive/negative scans lockMassFreq Performs correction Waters LockMass function mslevel perform peak picking data given mslevel nSlaves DEPRECATED, use BPPARAM argument       instead. progressCallback function called, progressInfo changes (useful GUIs) scanrange scan range read BPPARAM BiocParallel parameter object control     parallel processing performed. objects can     created SerialParam,     MulticoreParam     SnowParam functions. stopOnError Logical specifying whether feature detection     call stop first encountered error (default),     whether feature detection performed files regardless     eventual failures individual files case errors     reported warnings. ... arguments findPeaks method     xcmsRaw class","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"default values files, snames, sclass,   phenoData arguments cause function recursively search   readable files. filename without extention used   sample name. subdirectory path used sample class.   files contain positive negative spectra, polarity   can selected explicitly. default (NULL) read scans. phenoData provided, stored phenoData   slot returned xcmsSet class. data.frame   contains column named “class”, content returned   sampclass method thus used   group/class assignment individual files (e.g. peak grouping   etc.). details see help xcmsSet-class. step size (m/z) use profile generation can submitted   either using profparam argument   (e.g. profparam=list(step=0.1)) submitting   step=0.1. specifying value 0 profile matrix   generation can skipped. feature/peak detection algorithm can specified   method argument defaults \"matchFilter\"   method (findPeaks.matchedFilter). Possible values   returned getOption(\"BioC\")$xcms$findPeaks.methods. lock mass correction allows lock mass scan added back   last working scan. correction gives better reproducibility   sample sets.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"arguments profmethod profparam influence   feature/peak detection. step size parameter step   profile generation findPeaks.matchedFilter   peak detection algorithm can passed using ....","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Virtual class for raw data sources — xcmsSource-class","title":"Virtual class for raw data sources — xcmsSource-class","text":"virtual class provides implementation-independent way load   mass spectrometer data various sources use   xcmsRaw object. Subclasses can defined   enable data loaded user-specified sources. virtual   class xcmsFileSource included box   contains file name character string. implementing child classes xcmsSource, corresponding   loadRaw-methods method must provided accepts   xcmsSource child class returns list format   described loadRaw-methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Virtual class for raw data sources — xcmsSource-class","text":"virtual Class: objects may created .","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Virtual class for raw data sources — xcmsSource-class","text":"Daniel Hackney, dan@haxney.org","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Create an xcmsSource object in a flexible way — xcmsSource-methods","title":"Create an xcmsSource object in a flexible way — xcmsSource-methods","text":"Users can define alternate means reading data   xcmsRaw objects creating new implementations   method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create an xcmsSource object in a flexible way — xcmsSource-methods","text":"signature(object = \"xcmsSource\") Pass object unmodified.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Create an xcmsSource object in a flexible way — xcmsSource-methods","text":"Daniel Hackney, dan@haxney.org","code":""},{"path":[]}]
+[{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"Compounding (grouping) of LC-MS features","text":"typical LC-MS-based metabolomics experiment compounds eluting chromatography first ionized measured mass spectrometry (MS). ionization different (multiple) ions can generated compound measured MS. general, resulting data pre-processed identify chromatographic peaks data group across samples correspondence analysis. result distinct LC-MS features, characterized specific m/z retention time range. Different ions generated ionization detected different features. Compounding aims now grouping features presumably representing signal originating compound reduce data set complexity (aid subsequent annotation steps). General MS feature grouping functionality defined MsFeatures package additional functionality implemented xcms package enable compounding LC-MS data. document provides simple compounding workflow using xcms. Note present functionality (yet) aggregate combine actual features per values, define feature groups (one per compound).","code":""},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"compounding-of-lc-ms-data","dir":"Articles","previous_headings":"","what":"Compounding of LC-MS data","title":"Compounding (grouping) of LC-MS features","text":"demonstrate compounding (feature grouping) functionality simple toy data set used also xcms package provided faahKO package. data set consists samples 4 mice knock-fatty acid amide hydrolase (FAAH) 4 wild type mice. Pre-processing data set described detail main vignette xcms package. load required packages result pre-processing updating also location respective raw data files current machine. performing feature grouping inspect result object. featureDefinitions can extract results correspondence analysis. row data frame represents definition one feature, average range m/z retention time. Column \"peakidx\" provides index chromatographic peak assigned feature chromPeaks matrix result object. featureValues function allows extract feature values, .e. matrix feature abundances, one row per feature columns representing samples present data set. extract feature values without filled-peak data. Without gap-filled data abundances detected chromatographic peaks reported. gap-filled data, samples chromatographic peak feature detected, signal m/z - retention time range defined based detected chromatographic peaks integrated. total 351 features defined present data set, many likely represent signal different ions (adducts isotopes) compound. aim grouping functions now define features likely come original compound. feature grouping functions base following assumptions/properties LC-MS data: Features (ions) compound similar retention time. abundance features (ions) compound similar pattern across samples, .e. compound highly concentrated one sample low another, ions follow pattern. peak shape extracted ion chromatograms (EIC) features compound similar follow elution pattern original compound LC. main method perform feature grouping called groupFeatures takes xcms result object (.e., XcmsExperiment XCMSnExp) input well parameter object chose grouping algorithm specify settings. xcms provides supports following grouping approaches: SimilarRtimeParam: perform initial grouping based similar retention time. AbundanceSimilarityParam: perform feature grouping based correlation feature abundances (values) across samples. EicSimilarityParam: perform feature grouping based correlation EICs. Calling groupFeatures xcms result object perform feature grouping assigning feature data set feature group. feature groups stored additional column called \"feature_group\" featureDefinition data frame result object can accessed featureGroups function. subsequent groupFeature call sub-group (refine) identified feature groups . thus possible use single grouping approach, combine multiple generate desired feature grouping incremental fashion. individual feature grouping algorithms can called order, advisable use EicSimilarityParam last refinement step, computationally expensive, especially applied result object without pre-defined feature groups feature groups large.","code":"library(xcms) library(faahKO) library(MsFeatures)  xmse <- loadXcmsData(\"xmse\") featureDefinitions(xmse) |> head() ##       mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks KO WT      peakidx ## FT001 200.1 200.1 200.1 2902.177 2882.133 2922.222      2  2  0 458, 116.... ## FT002 205.0 205.0 205.0 2789.457 2782.376 2795.823      8  4  4 44, 443,.... ## FT003 206.0 206.0 206.0 2788.770 2780.807 2793.606      7  3  4 29, 430,.... ## FT004 207.1 207.1 207.1 2718.476 2713.314 2726.487      7  4  3 16, 420,.... ## FT005 233.0 233.0 233.1 3023.353 3014.971 3043.942      7  3  4 69, 959,.... ## FT006 241.1 241.1 241.2 3680.200 3661.067 3695.533      8  3  4 276, 284.... ##       ms_level ## FT001        1 ## FT002        1 ## FT003        1 ## FT004        1 ## FT005        1 ## FT006        1 head(featureValues(xmse, filled = FALSE)) ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001        NA  506848.9        NA  169955.6        NA        NA        NA ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7        NA  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9        NA ## FT005  286221.4        NA  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5        NA  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001        NA ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 head(featureValues(xmse, filled = TRUE)) ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001  135422.4  506848.9  111872.0  169955.6  210333.1  141880.3  233271.3 ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7  167516.7  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  219142.3 ## FT005  286221.4  286594.4  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5 1145753.0  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001  142254.8 ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-features-by-similar-retention-time","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of features by similar retention time","title":"Compounding (grouping) of LC-MS features","text":"intuitive simple way group features based retention time. perform initial grouping evaluate retention times m/z features present data set. Plot retention times m/z features data set. Several features retention time (different m/z) can spotted, especially beginning LC. thus group features within retention time window 10 seconds feature groups. results feature grouping can accessed featureGroups function. determine size feature groups (.e. many features grouped together). addition visualize feature groups plotFeatureGroups function shows features m/z - retention time space grouped features connected line. Feature groups defined rt window 10 seconds Let’s assume don’t agree feature grouping, also knowing quite large shifts retention times runs. thus first remove defined feature groups assigning value NULL re-perform feature grouping using larger rt window. Feature groups defined rt window 20 seconds Grouping similar retention time grouped total 351 features 117 feature groups.","code":"plot(featureDefinitions(xmse)$rtmed, featureDefinitions(xmse)$mzmed,      xlab = \"retention time\", ylab = \"m/z\", main = \"features\",      col = \"#00000080\", pch = 21, bg = \"#00000040\") grid() xmse <- groupFeatures(xmse, param = SimilarRtimeParam(10)) table(featureGroups(xmse)) ##  ## FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  ##      4      5      3      4      2      4      4      4      4      3      3  ## FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  ##      5      6      2      4      3      2      2      3      2      3      4  ## FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  ##      3      8      6      3      3      3      3      2      3      3      4  ## FG.034 FG.035 FG.036 FG.037 FG.038 FG.039 FG.040 FG.041 FG.042 FG.043 FG.044  ##      2      3      5      3      3      4      5      2      4      2      4  ## FG.045 FG.046 FG.047 FG.048 FG.049 FG.050 FG.051 FG.052 FG.053 FG.054 FG.055  ##      6      3      3      3      2      2      2      4      3      3      3  ## FG.056 FG.057 FG.058 FG.059 FG.060 FG.061 FG.062 FG.063 FG.064 FG.065 FG.066  ##      2      3      5      4      2      4      2      3      2      3      3  ## FG.067 FG.068 FG.069 FG.070 FG.071 FG.072 FG.073 FG.074 FG.075 FG.076 FG.077  ##      3      3      3      3      3      2      4      2      2      2      2  ## FG.078 FG.079 FG.080 FG.081 FG.082 FG.083 FG.084 FG.085 FG.086 FG.087 FG.088  ##      2      3      3      3      3      3      2      3      4      2      3  ## FG.089 FG.090 FG.091 FG.092 FG.093 FG.094 FG.095 FG.096 FG.097 FG.098 FG.099  ##      2      3      2      2      2      3      2      3      3      2      2  ## FG.100 FG.101 FG.102 FG.103 FG.104 FG.105 FG.106 FG.107 FG.108 FG.109 FG.110  ##      2      2      2      2      2      2      2      2      2      2      2  ## FG.111 FG.112 FG.113 FG.114 FG.115 FG.116 FG.117 FG.118 FG.119 FG.120 FG.121  ##      1      1      1      1      1      1      1      1      1      1      1  ## FG.122 FG.123 FG.124 FG.125 FG.126 FG.127 FG.128 FG.129 FG.130 FG.131 FG.132  ##      1      1      1      1      1      1      1      1      1      1      1 plotFeatureGroups(xmse, pch = 21, lwd = 2, col = \"#00000040\",                   bg = \"#00000020\") grid() ## Remove previous feature grouping results to repeat the rtime-based ## feature grouping with different setting featureDefinitions(xmse)$feature_group <- NULL  ## Repeat the grouping xmse <- groupFeatures(xmse, SimilarRtimeParam(20)) table(featureGroups(xmse)) ##  ## FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  ##      4      5      3      4      2      4      4      4      4      3      3  ## FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  ##      5      6      2      4      3      2      2      3      2      3      4  ## FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  ##      3      8      6      3      3      3      3      2      3      3      4  ## FG.034 FG.035 FG.036 FG.037 FG.038 FG.039 FG.040 FG.041 FG.042 FG.043 FG.044  ##      2      3      5      3      3      4      6      2      4      2      5  ## FG.045 FG.046 FG.047 FG.048 FG.049 FG.050 FG.051 FG.052 FG.053 FG.054 FG.055  ##      6      3      4      3      2      2      2      4      3      3      3  ## FG.056 FG.057 FG.058 FG.059 FG.060 FG.061 FG.062 FG.063 FG.064 FG.065 FG.066  ##      3      3      5      4      2      4      2      3      2      3      3  ## FG.067 FG.068 FG.069 FG.070 FG.071 FG.072 FG.073 FG.074 FG.075 FG.076 FG.077  ##      4      3      3      3      4      2      5      2      2      2      2  ## FG.078 FG.079 FG.080 FG.081 FG.082 FG.083 FG.084 FG.085 FG.086 FG.087 FG.088  ##      2      3      3      3      3      3      2      3      4      3      4  ## FG.089 FG.090 FG.091 FG.092 FG.093 FG.094 FG.095 FG.096 FG.097 FG.098 FG.099  ##      3      4      2      2      2      4      2      3      3      2      2  ## FG.100 FG.101 FG.102 FG.103 FG.104 FG.105 FG.106 FG.107 FG.108 FG.109 FG.110  ##      2      3      2      2      2      3      3      2      2      2      2  ## FG.111 FG.112 FG.113 FG.114 FG.115 FG.116 FG.117  ##      1      1      1      1      1      1      1 plotFeatureGroups(xmse, pch = 21, lwd = 2, col = \"#00000040\", bg = \"#00000020\") grid()"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-features-by-abundance-correlation-across-samples","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of features by abundance correlation across samples","title":"Compounding (grouping) of LC-MS features","text":"Assuming OK crude initial feature grouping previous section, can next refine feature groups considering also feature abundances across samples. can use groupFeatures method AbundanceSimilarityParam object. approach performs pairwise correlation (non-missing) feature values (abundances; across samples) features predefined feature group (defined previous section). Features correlation >= threshold grouped together. Feature grouping based approach works best features higher variability concentration across samples. Parameter subset allows restrict analysis subset samples allows thus e.g. exclude QC sample pools correlation artificially increase correlation. parameters passed directly internal featureValues call extracts feature values correlation performed. performing grouping also evaluate correlation features based (log2 transformed) abundances across samples heatmap. Correlation features based feature abundances. large correlations can observed several groups features, many within feature group defined previous section (.e. eluting time). might thus represent correlated metabolites, ions adducts metabolite. use groupFeatures AbundanceSimilarityParam group features correlation coefficient higher 0.7 including detected filled-signal. Whether filled-detected signal used correlation analysis evaluated data set data set. specifying transform = log2 tell function log2 transform abundance prior correlation analysis. See help page groupFeatures AbundanceSimilarityParam xcms package details options. Many larger retention time-based feature groups splitted two sub-groups based correlation feature abundances (e.g. feature group FG.004 split feature groups FG.004.001, FG.004.002 FG.004.003). evaluate refinement feature group \"FG.040\" plotting pairwise correlation. better visualize feature grouping addition define custom panel plot pairs function plots data points blue features correlation coefficient selected threshold (0.7) red otherwise. Pairwise correlation plot features initially grouped feature group FG.040. Indeed, correlation seems present features retention time feature group, e.g. clearly FT273 FT274 also FT143 FT273. Note however abundance correlation suffers relatively samples (8 total), relatively small variance abundances across samples. feature grouping abundance correlation, 351 features grouped 253 feature groups.","code":"library(pheatmap) fvals <- log2(featureValues(xmse, filled = TRUE))  cormat <- cor(t(fvals), use = \"pairwise.complete.obs\") ann <- data.frame(fgroup = featureGroups(xmse)) rownames(ann) <- rownames(cormat)  res <- pheatmap(cormat, annotation_row = ann, cluster_rows = TRUE,                 cluster_cols = TRUE) xmse <- groupFeatures(     xmse, AbundanceSimilarityParam(threshold = 0.7, transform = log2),     filled = TRUE) table(featureGroups(xmse)) ##  ## FG.001.001 FG.001.002 FG.002.001 FG.002.002 FG.003.001 FG.003.002 FG.003.003  ##          3          1          3          2          1          1          1  ## FG.004.001 FG.004.002 FG.004.003 FG.005.001 FG.005.002 FG.006.001 FG.006.002  ##          2          1          1          1          1          2          1  ## FG.006.003 FG.007.001 FG.007.002 FG.008.001 FG.008.002 FG.008.003 FG.008.004  ##          1          3          1          1          1          1          1  ## FG.009.001 FG.009.002 FG.010.001 FG.010.002 FG.011.001 FG.011.002 FG.012.001  ##          3          1          2          1          2          1          2  ## FG.012.002 FG.012.003 FG.012.004 FG.013.001 FG.013.002 FG.013.003 FG.013.004  ##          1          1          1          3          1          1          1  ## FG.014.001 FG.015.001 FG.015.002 FG.016.001 FG.017.001 FG.017.002 FG.018.001  ##          2          3          1          3          1          1          1  ## FG.018.002 FG.019.001 FG.019.002 FG.020.001 FG.020.002 FG.021.001 FG.021.002  ##          1          2          1          1          1          1          1  ## FG.021.003 FG.022.001 FG.022.002 FG.022.003 FG.022.004 FG.023.001 FG.023.002  ##          1          1          1          1          1          1          1  ## FG.023.003 FG.024.001 FG.024.002 FG.024.003 FG.024.004 FG.024.005 FG.024.006  ##          1          2          2          1          1          1          1  ## FG.025.001 FG.026.001 FG.026.002 FG.027.001 FG.027.002 FG.028.001 FG.029.001  ##          6          2          1          2          1          3          2  ## FG.029.002 FG.030.001 FG.030.002 FG.031.001 FG.031.002 FG.032.001 FG.032.002  ##          1          1          1          2          1          1          1  ## FG.032.003 FG.033.001 FG.033.002 FG.033.003 FG.034.001 FG.034.002 FG.035.001  ##          1          2          1          1          1          1          2  ## FG.035.002 FG.036.001 FG.036.002 FG.036.003 FG.037.001 FG.037.002 FG.037.003  ##          1          2          2          1          1          1          1  ## FG.038.001 FG.038.002 FG.038.003 FG.039.001 FG.039.002 FG.039.003 FG.039.004  ##          1          1          1          1          1          1          1  ## FG.040.001 FG.040.002 FG.040.003 FG.041.001 FG.041.002 FG.042.001 FG.042.002  ##          4          1          1          1          1          2          1  ## FG.042.003 FG.043.001 FG.044.001 FG.044.002 FG.045.001 FG.045.002 FG.046.001  ##          1          2          3          2          4          2          2  ## FG.046.002 FG.047.001 FG.047.002 FG.047.003 FG.047.004 FG.048.001 FG.048.002  ##          1          1          1          1          1          2          1  ## FG.049.001 FG.049.002 FG.050.001 FG.050.002 FG.051.001 FG.051.002 FG.052.001  ##          1          1          1          1          1          1          1  ## FG.052.002 FG.052.003 FG.052.004 FG.053.001 FG.053.002 FG.053.003 FG.054.001  ##          1          1          1          1          1          1          1  ## FG.054.002 FG.054.003 FG.055.001 FG.055.002 FG.056.001 FG.056.002 FG.057.001  ##          1          1          2          1          2          1          2  ## FG.057.002 FG.058.001 FG.058.002 FG.059.001 FG.059.002 FG.059.003 FG.060.001  ##          1          4          1          2          1          1          1  ## FG.060.002 FG.061.001 FG.061.002 FG.062.001 FG.062.002 FG.063.001 FG.063.002  ##          1          2          2          1          1          1          1  ## FG.063.003 FG.064.001 FG.065.001 FG.065.002 FG.065.003 FG.066.001 FG.066.002  ##          1          2          1          1          1          2          1  ## FG.067.001 FG.067.002 FG.067.003 FG.068.001 FG.068.002 FG.069.001 FG.069.002  ##          2          1          1          2          1          2          1  ## FG.070.001 FG.070.002 FG.071.001 FG.071.002 FG.072.001 FG.073.001 FG.073.002  ##          2          1          3          1          2          2          2  ## FG.073.003 FG.074.001 FG.074.002 FG.075.001 FG.076.001 FG.076.002 FG.077.001  ##          1          1          1          2          1          1          2  ## FG.078.001 FG.078.002 FG.079.001 FG.079.002 FG.079.003 FG.080.001 FG.080.002  ##          1          1          1          1          1          2          1  ## FG.081.001 FG.081.002 FG.082.001 FG.082.002 FG.083.001 FG.083.002 FG.083.003  ##          2          1          2          1          1          1          1  ## FG.084.001 FG.084.002 FG.085.001 FG.085.002 FG.085.003 FG.086.001 FG.086.002  ##          1          1          1          1          1          2          1  ## FG.086.003 FG.087.001 FG.088.001 FG.088.002 FG.088.003 FG.089.001 FG.089.002  ##          1          3          2          1          1          2          1  ## FG.090.001 FG.090.002 FG.091.001 FG.091.002 FG.092.001 FG.092.002 FG.093.001  ##          2          2          1          1          1          1          1  ## FG.093.002 FG.094.001 FG.094.002 FG.094.003 FG.095.001 FG.096.001 FG.096.002  ##          1          2          1          1          2          2          1  ## FG.097.001 FG.098.001 FG.098.002 FG.099.001 FG.100.001 FG.100.002 FG.101.001  ##          3          1          1          2          1          1          2  ## FG.101.002 FG.102.001 FG.102.002 FG.103.001 FG.104.001 FG.105.001 FG.105.002  ##          1          1          1          2          2          2          1  ## FG.106.001 FG.106.002 FG.107.001 FG.107.002 FG.108.001 FG.109.001 FG.109.002  ##          2          1          1          1          2          1          1  ## FG.110.001 FG.111.001 FG.112.001 FG.113.001 FG.114.001 FG.115.001 FG.116.001  ##          2          1          1          1          1          1          1  ## FG.117.001  ##          1 cor_plot <- function(x, y) {     C <- cor(x, y, use = \"pairwise.complete.obs\")     col <- ifelse(C >= 0.7, yes = \"#0000ff80\", no = \"#ff000080\")     points(x, y, pch = 16, col = col)     grid() } fts <- grep(\"FG.040\", featureGroups(xmse)) pairs(t(fvals[fts, ]), gap = 0.1, main = \"FG.040\", panel = cor_plot)"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-features-by-similarity-of-their-eics","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of features by similarity of their EICs","title":"Compounding (grouping) of LC-MS features","text":"chromatographic peak shape ion compound highly similar elution pattern compound. Thus, features compound assumed similar peak shapes EICs within sample. grouping features based similarity EICs can performed groupFeatures EicSimilarityParam object. advisable perform peak shape correlation subset samples (peak shape correlation computationally intense chromatographic peaks low intensity features notoriously noisy). EicSimilarityParam approach allows select number top n samples (ordered total intensity feature abundances per feature group) correlation performed parameter n. n =3, 3 samples highest signal features respective feature group first identified pairwise similarity calculation performed EICs samples. resulting similarity score 3 samples aggregated single score taking 75% quantile across 3 samples. value subsequently compared cut-similarity (parameter threshold) features score >= threshold grouped feature group. group features based similarity EICs two samples highest total signal respective feature groups. default, Pearson correlation coefficient used similarity score similarity/distance metric function used instead (parameter FUN EicSimilarityParam - see respective help page ?EicSimilarityParam details options). define threshold correlation coefficient 0.7. computationally intense approach since involves also loading raw MS data extract ion chromatograms feature. results grouping shown . many cases, pre-defined feature groups (retention time similarity abundance correlation) subdivided. evaluate feature groups, starting FG.008.001 split two different feature groups based EIC correlation. first extract EICs features initial feature group. n = 1 specify extract EIC sample highest intensity. Next plot EICs using different color subgroups. peakType = \"none\" disable highlighting detected chromatographic peaks. Feature EICs per sample features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. Feature EICs per sample normalized absolute intensity 1 features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. One features (highlighted red plots ) within original feature group separated two low similarity EICs. fact, feature’s EIC shifted samples along retention time dimension can thus represent signal compound. evaluate next sub-grouping another feature group. Next plot EICs using different color subgroups. Feature EICs per sample features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. Feature EICs per sample normalized absolute intensity 1 features feature group defined rentention time feature abudances across samples. Features high correlation EICs shown color. Based EIC correlation, initial feature group FG.045.001 grouped two separate sub-groups. grouping based EIC correlation pre-defined feature groups previous sections grouped 351 features 272 feature groups.","code":"xmse <- groupFeatures(xmse, EicSimilarityParam(threshold = 0.7, n = 2)) table(featureGroups(xmse)) ##  ## FG.001.001.001 FG.001.001.002 FG.001.002.001 FG.002.001.001 FG.002.002.001  ##              2              1              1              3              1  ## FG.002.002.002 FG.003.001.001 FG.003.002.001 FG.003.003.001 FG.004.001.001  ##              1              1              1              1              2  ## FG.004.002.001 FG.004.003.001 FG.005.001.001 FG.005.002.001 FG.006.001.001  ##              1              1              1              1              2  ## FG.006.002.001 FG.006.003.001 FG.007.001.001 FG.007.002.001 FG.008.001.001  ##              1              1              3              1              1  ## FG.008.002.001 FG.008.003.001 FG.008.004.001 FG.009.001.001 FG.009.002.001  ##              1              1              1              3              1  ## FG.010.001.001 FG.010.002.001 FG.011.001.001 FG.011.002.001 FG.012.001.001  ##              2              1              2              1              2  ## FG.012.002.001 FG.012.003.001 FG.012.004.001 FG.013.001.001 FG.013.001.002  ##              1              1              1              2              1  ## FG.013.002.001 FG.013.003.001 FG.013.004.001 FG.014.001.001 FG.015.001.001  ##              1              1              1              2              3  ## FG.015.002.001 FG.016.001.001 FG.017.001.001 FG.017.002.001 FG.018.001.001  ##              1              3              1              1              1  ## FG.018.002.001 FG.019.001.001 FG.019.002.001 FG.020.001.001 FG.020.002.001  ##              1              2              1              1              1  ## FG.021.001.001 FG.021.002.001 FG.021.003.001 FG.022.001.001 FG.022.002.001  ##              1              1              1              1              1  ## FG.022.003.001 FG.022.004.001 FG.023.001.001 FG.023.002.001 FG.023.003.001  ##              1              1              1              1              1  ## FG.024.001.001 FG.024.002.001 FG.024.003.001 FG.024.004.001 FG.024.005.001  ##              2              2              1              1              1  ## FG.024.006.001 FG.025.001.001 FG.026.001.001 FG.026.002.001 FG.027.001.001  ##              1              6              2              1              2  ## FG.027.002.001 FG.028.001.001 FG.029.001.001 FG.029.002.001 FG.030.001.001  ##              1              3              2              1              1  ## FG.030.002.001 FG.031.001.001 FG.031.002.001 FG.032.001.001 FG.032.002.001  ##              1              2              1              1              1  ## FG.032.003.001 FG.033.001.001 FG.033.002.001 FG.033.003.001 FG.034.001.001  ##              1              2              1              1              1  ## FG.034.002.001 FG.035.001.001 FG.035.002.001 FG.036.001.001 FG.036.001.002  ##              1              2              1              1              1  ## FG.036.002.001 FG.036.003.001 FG.037.001.001 FG.037.002.001 FG.037.003.001  ##              2              1              1              1              1  ## FG.038.001.001 FG.038.002.001 FG.038.003.001 FG.039.001.001 FG.039.002.001  ##              1              1              1              1              1  ## FG.039.003.001 FG.039.004.001 FG.040.001.001 FG.040.002.001 FG.040.003.001  ##              1              1              4              1              1  ## FG.041.001.001 FG.041.002.001 FG.042.001.001 FG.042.002.001 FG.042.003.001  ##              1              1              2              1              1  ## FG.043.001.001 FG.044.001.001 FG.044.001.002 FG.044.002.001 FG.044.002.002  ##              2              2              1              1              1  ## FG.045.001.001 FG.045.001.002 FG.045.002.001 FG.046.001.001 FG.046.002.001  ##              2              2              2              2              1  ## FG.047.001.001 FG.047.002.001 FG.047.003.001 FG.047.004.001 FG.048.001.001  ##              1              1              1              1              2  ## FG.048.002.001 FG.049.001.001 FG.049.002.001 FG.050.001.001 FG.050.002.001  ##              1              1              1              1              1  ## FG.051.001.001 FG.051.002.001 FG.052.001.001 FG.052.002.001 FG.052.003.001  ##              1              1              1              1              1  ## FG.052.004.001 FG.053.001.001 FG.053.002.001 FG.053.003.001 FG.054.001.001  ##              1              1              1              1              1  ## FG.054.002.001 FG.054.003.001 FG.055.001.001 FG.055.002.001 FG.056.001.001  ##              1              1              2              1              1  ## FG.056.001.002 FG.056.002.001 FG.057.001.001 FG.057.001.002 FG.057.002.001  ##              1              1              1              1              1  ## FG.058.001.001 FG.058.001.002 FG.058.002.001 FG.059.001.001 FG.059.002.001  ##              3              1              1              2              1  ## FG.059.003.001 FG.060.001.001 FG.060.002.001 FG.061.001.001 FG.061.002.001  ##              1              1              1              2              2  ## FG.062.001.001 FG.062.002.001 FG.063.001.001 FG.063.002.001 FG.063.003.001  ##              1              1              1              1              1  ## FG.064.001.001 FG.065.001.001 FG.065.002.001 FG.065.003.001 FG.066.001.001  ##              2              1              1              1              2  ## FG.066.002.001 FG.067.001.001 FG.067.001.002 FG.067.002.001 FG.067.003.001  ##              1              1              1              1              1  ## FG.068.001.001 FG.068.002.001 FG.069.001.001 FG.069.001.002 FG.069.002.001  ##              2              1              1              1              1  ## FG.070.001.001 FG.070.002.001 FG.071.001.001 FG.071.002.001 FG.072.001.001  ##              2              1              3              1              2  ## FG.073.001.001 FG.073.002.001 FG.073.003.001 FG.074.001.001 FG.074.002.001  ##              2              2              1              1              1  ## FG.075.001.001 FG.076.001.001 FG.076.002.001 FG.077.001.001 FG.078.001.001  ##              2              1              1              2              1  ## FG.078.002.001 FG.079.001.001 FG.079.002.001 FG.079.003.001 FG.080.001.001  ##              1              1              1              1              2  ## FG.080.002.001 FG.081.001.001 FG.081.002.001 FG.082.001.001 FG.082.002.001  ##              1              2              1              2              1  ## FG.083.001.001 FG.083.002.001 FG.083.003.001 FG.084.001.001 FG.084.002.001  ##              1              1              1              1              1  ## FG.085.001.001 FG.085.002.001 FG.085.003.001 FG.086.001.001 FG.086.002.001  ##              1              1              1              2              1  ## FG.086.003.001 FG.087.001.001 FG.087.001.002 FG.088.001.001 FG.088.002.001  ##              1              2              1              2              1  ## FG.088.003.001 FG.089.001.001 FG.089.001.002 FG.089.002.001 FG.090.001.001  ##              1              1              1              1              2  ## FG.090.002.001 FG.090.002.002 FG.091.001.001 FG.091.002.001 FG.092.001.001  ##              1              1              1              1              1  ## FG.092.002.001 FG.093.001.001 FG.093.002.001 FG.094.001.001 FG.094.002.001  ##              1              1              1              2              1  ## FG.094.003.001 FG.095.001.001 FG.096.001.001 FG.096.002.001 FG.097.001.001  ##              1              2              2              1              2  ## FG.097.001.002 FG.098.001.001 FG.098.002.001 FG.099.001.001 FG.100.001.001  ##              1              1              1              2              1  ## FG.100.002.001 FG.101.001.001 FG.101.001.002 FG.101.002.001 FG.102.001.001  ##              1              1              1              1              1  ## FG.102.002.001 FG.103.001.001 FG.103.001.002 FG.104.001.001 FG.105.001.001  ##              1              1              1              2              2  ## FG.105.002.001 FG.106.001.001 FG.106.001.002 FG.106.002.001 FG.107.001.001  ##              1              1              1              1              1  ## FG.107.002.001 FG.108.001.001 FG.109.001.001 FG.109.002.001 FG.110.001.001  ##              1              2              1              1              2  ## FG.111.001.001 FG.112.001.001 FG.113.001.001 FG.114.001.001 FG.115.001.001  ##              1              1              1              1              1  ## FG.116.001.001 FG.117.001.001  ##              1              1 fidx <- grep(\"FG.013.001.\", featureGroups(xmse)) eics <- featureChromatograms(     xmse, features = rownames(featureDefinitions(xmse))[fidx],     filled = TRUE, n = 1) cols <- c(\"#ff000080\", \"#0000ff80\") names(cols) <- unique(featureGroups(xmse)[fidx])  plotChromatogramsOverlay(eics, col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\") plotChromatogramsOverlay(normalize(eics),                          col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\") fidx <- grep(\"FG.045.001.\", featureGroups(xmse)) eics <- featureChromatograms(     xmse, features = rownames(featureDefinitions(xmse))[fidx],     filled = TRUE, n = 1) cols <- c(\"#ff000080\", \"#0000ff80\") names(cols) <- unique(featureGroups(xmse)[fidx])  plotChromatogramsOverlay(eics, col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\") plotChromatogramsOverlay(normalize(eics),                          col = cols[featureGroups(xmse)[fidx]],                          lwd = 2, peakType = \"none\")"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"grouping-of-subsets-of-features","dir":"Articles","previous_headings":"Compounding of LC-MS data","what":"Grouping of subsets of features","title":"Compounding (grouping) of LC-MS features","text":"previous sections always considering features data set, sometimes desirable just group pre-defined set features, example features found particular interest certain experiment (e.g. significant features). can easily achieved assigning features interest initial feature group, using NA group ID features. illustrate reset feature groups setting NA assign features interest (example just 30 randomly selected features) initial feature group \"FG\". call groupFeatures now simply sub-group set 30 features. feature NA \"feature_group\" column ignored.","code":"featureDefinitions(xmse)$feature_group <- NA_character_  set.seed(123) fts_idx <- sample(1:nrow(featureDefinitions(xmse)), 30) featureDefinitions(xmse)$feature_group[fts_idx] <- \"FG\" xmse <- groupFeatures(xmse, SimilarRtimeParam(diffRt = 20)) xmse <- groupFeatures(xmse, AbundanceSimilarityParam(threshold = 0.7)) table(featureGroups(xmse)) ##  ## FG.001.001 FG.001.002 FG.002.001 FG.002.002 FG.003.001 FG.004.001 FG.004.002  ##          1          1          1          1          2          1          1  ## FG.005.001 FG.005.002 FG.005.003 FG.006.001 FG.007.001 FG.007.002 FG.008.001  ##          1          1          1          2          1          1          1  ## FG.009.001 FG.010.001 FG.011.001 FG.012.001 FG.013.001 FG.014.001 FG.015.001  ##          1          1          1          1          1          1          1  ## FG.016.001 FG.017.001 FG.018.001 FG.019.001 FG.020.001 FG.021.001 FG.022.001  ##          1          1          1          1          1          1          1"},{"path":"https://sneumann.github.io/xcms/articles/LC-MS-feature-grouping.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"Compounding (grouping) of LC-MS features","text":"","code":"sessionInfo() ## R Under development (unstable) (2024-01-17 r85813) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] pheatmap_1.0.12     faahKO_1.43.0       MsFeatures_1.11.0   ##  [4] xcms_4.1.6          MSnbase_2.29.3      ProtGenerics_1.35.2 ##  [7] S4Vectors_0.41.3    mzR_2.37.0          Rcpp_1.0.12         ## [10] Biobase_2.63.0      BiocGenerics_0.49.1 BiocParallel_1.37.0 ## [13] BiocStyle_2.31.0    ##  ## loaded via a namespace (and not attached): ##   [1] DBI_1.2.1                   bitops_1.0-7                ##   [3] rlang_1.1.3                 magrittr_2.0.3              ##   [5] clue_0.3-65                 MassSpecWavelet_1.69.0      ##   [7] matrixStats_1.2.0           compiler_4.4.0              ##   [9] systemfonts_1.0.5           vctrs_0.6.5                 ##  [11] stringr_1.5.1               MetaboCoreUtils_1.11.2      ##  [13] pkgconfig_2.0.3             crayon_1.5.2                ##  [15] fastmap_1.1.1               XVector_0.43.1              ##  [17] utf8_1.2.4                  rmarkdown_2.25              ##  [19] preprocessCore_1.65.0       ragg_1.2.7                  ##  [21] purrr_1.0.2                 xfun_0.41                   ##  [23] MultiAssayExperiment_1.29.0 zlibbioc_1.49.0             ##  [25] cachem_1.0.8                GenomeInfoDb_1.39.5         ##  [27] jsonlite_1.8.8              progress_1.2.3              ##  [29] highr_0.10                  DelayedArray_0.29.0         ##  [31] prettyunits_1.2.0           parallel_4.4.0              ##  [33] cluster_2.1.6               R6_2.5.1                    ##  [35] RColorBrewer_1.1-3          bslib_0.6.1                 ##  [37] stringi_1.8.3               limma_3.59.1                ##  [39] GenomicRanges_1.55.2        jquerylib_0.1.4             ##  [41] bookdown_0.37               SummarizedExperiment_1.33.3 ##  [43] iterators_1.0.14            knitr_1.45                  ##  [45] IRanges_2.37.1              splines_4.4.0               ##  [47] Matrix_1.6-5                igraph_1.6.0                ##  [49] tidyselect_1.2.0            abind_1.4-5                 ##  [51] yaml_2.3.8                  doParallel_1.0.17           ##  [53] codetools_0.2-19            affy_1.81.0                 ##  [55] lattice_0.22-5              tibble_3.2.1                ##  [57] plyr_1.8.9                  evaluate_0.23               ##  [59] survival_3.5-7              desc_1.4.3                  ##  [61] Spectra_1.13.3              pillar_1.9.0                ##  [63] affyio_1.73.0               BiocManager_1.30.22         ##  [65] MatrixGenerics_1.15.0       foreach_1.5.2               ##  [67] MALDIquant_1.22.2           ncdf4_1.22                  ##  [69] generics_0.1.3              RCurl_1.98-1.14             ##  [71] hms_1.1.3                   ggplot2_3.4.4               ##  [73] munsell_0.5.0               scales_1.3.0                ##  [75] MsExperiment_1.5.4          glue_1.7.0                  ##  [77] lazyeval_0.2.2              tools_4.4.0                 ##  [79] robustbase_0.99-1           mzID_1.41.0                 ##  [81] QFeatures_1.13.2            vsn_3.71.0                  ##  [83] RANN_2.6.1                  fs_1.6.3                    ##  [85] XML_3.99-0.16.1             grid_4.4.0                  ##  [87] impute_1.77.0               MsCoreUtils_1.15.3          ##  [89] colorspace_2.1-0            GenomeInfoDbData_1.2.11     ##  [91] PSMatch_1.7.1               cli_3.6.2                   ##  [93] textshaping_0.3.7           fansi_1.0.6                 ##  [95] S4Arrays_1.3.2              dplyr_1.1.4                 ##  [97] AnnotationFilter_1.27.0     DEoptimR_1.1-3              ##  [99] pcaMethods_1.95.0           gtable_0.3.4                ## [101] sass_0.4.8                  digest_0.6.34               ## [103] SparseArray_1.3.3           farver_2.1.1                ## [105] multtest_2.59.0             memoise_2.0.1               ## [107] htmltools_0.5.7             pkgdown_2.0.7.9000          ## [109] lifecycle_1.0.4             statmod_1.5.0               ## [111] MASS_7.3-60.2"},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"Grouping FTICR-MS data with xcms","text":"document describes use xcms analysis direct injection mass spec data, including peak detection, calibration correspondence (grouping peaks across samples).","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"peak-detection","dir":"Articles","previous_headings":"","what":"Peak detection","title":"Grouping FTICR-MS data with xcms","text":"Prior analysis step, peaks identified mass spec data. contrast typical metabolomics workflow, peaks identified chromatographic (time) dimension, direct injection mass spec data sets peaks identified m/z dimension. xcms uses functionality MassSpecWavelet package identify peaks. load required packages. information parallel processing setup please see BiocParallel vignette. documentation use example data set msdata package. Assuming msdata installed, locate path package load data set. create also data.frame describing experimental setup based file names. data files direct injection mass spectrometry experiments, .e. single spectrum available sample retention times. Peaks identified within spectrum using mass spec wavelet method.","code":"library(xcms) library(MassSpecWavelet)  register(SerialParam()) mzML_path <- system.file(\"fticr-mzML\", package = \"msdata\") mzML_files <- list.files(mzML_path, recursive = TRUE, full.names = TRUE)  ## We're subsetting to 2 samples per condition mzML_files <- mzML_files[c(1, 2, 6, 7)]  ## Create a data.frame assigning samples to sample groups, i.e. ham4 and ham5. grp <- rep(\"ham4\", length(mzML_files)) grp[grep(basename(mzML_files), pattern = \"^HAM005\")] <- \"ham5\" pd <- data.frame(filename = basename(mzML_files), sample_group = grp)  ## Load the data. ham_raw <- readMSData(files = mzML_files,                       pdata = new(\"NAnnotatedDataFrame\", pd),                       mode = \"onDisk\") ## Only a single spectrum with an *artificial* retention time is available ## for each sample rtime(ham_raw) ## F1.S1 F2.S1 F3.S1 F4.S1  ##    -1    -1    -1    -1 ## Define the parameters for the peak detection msw <- MSWParam(scales = c(1, 4, 9), nearbyPeak = TRUE, winSize.noise = 500,                 SNR.method = \"data.mean\", snthresh = 10)  ham_prep <- findChromPeaks(ham_raw, param = msw)  head(chromPeaks(ham_prep)) ##            mz    mzmin    mzmax rt rtmin rtmax    into     maxo       sn intf ## CP01 403.2367 403.2279 403.2447 -1    -1    -1 4735258 372259.4 22.97534   NA ## CP02 409.1845 409.1747 409.1936 -1    -1    -1 4158404 310572.1 20.61382   NA ## CP03 413.2677 413.2585 413.2769 -1    -1    -1 6099006 435462.6 27.21723   NA ## CP04 423.2363 423.2266 423.2459 -1    -1    -1 2708391 174252.7 14.74527   NA ## CP05 427.2681 427.2574 427.2779 -1    -1    -1 6302089 461385.6 32.50050   NA ## CP06 437.2375 437.2254 437.2488 -1    -1    -1 7523070 517917.6 34.37645   NA ##           maxf sample ## CP01  814693.1      1 ## CP02  732119.9      1 ## CP03 1018994.8      1 ## CP04  435858.5      1 ## CP05 1125644.3      1 ## CP06 1282906.5      1"},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"calibration","dir":"Articles","previous_headings":"","what":"Calibration","title":"Grouping FTICR-MS data with xcms","text":"calibrate method can used correct m/z values identified peaks. currently implemented method requires identified peaks list m/z values known calibrants. identified peaks m/z values adjusted based differences calibrants’ m/z values m/z values closest peaks (within user defined permitted maximal distance). Note method presently calibrate identified peaks, original m/z values spectra. demonstrate calibrate method one data files artificially defined calibration m/z values. first subset data set first data file, extract m/z values 3 peaks modify values slightly. Next calibrate data set using previously defined artificial calibrants. using \"edgeshift\" method calibration adjusts peaks within range m/z values calibrants using linear interpolation shifts chromatographic peaks outside range constant factor (difference lowest respectively largest calibrant m/z closest peak’s m/z). Note real use case, m/z values obviously represent known m/z calibrants defined actual data. evaluate calibration plot difference adjusted raw m/z values (y-axis) raw m/z values.","code":"## Subset to the first file. first_file <- filterFile(ham_prep, file = 1)  ## Extract 3 m/z values calib_mz <- chromPeaks(first_file)[c(1, 4, 7), \"mz\"] calib_mz <- calib_mz + 0.00001 * runif(1, 0, 0.4) * calib_mz + 0.0001 ## Set-up the parameter class for the calibration prm <- CalibrantMassParam(mz = calib_mz, method = \"edgeshift\",                           mzabs = 0.0001, mzppm = 5) first_file_calibrated <- calibrate(first_file, param = prm) diffs <- chromPeaks(first_file_calibrated)[, \"mz\"] -     chromPeaks(first_file)[, \"mz\"]  plot(x = chromPeaks(first_file)[, \"mz\"], xlab = expression(m/z[raw]),      y = diffs, ylab = expression(m/z[calibrated] - m/z[raw]))"},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"correspondence","dir":"Articles","previous_headings":"","what":"Correspondence","title":"Grouping FTICR-MS data with xcms","text":"Correspondence aims group peaks across samples define features (ions m/z values). Peaks single spectrum, direct injection MS experiments can grouped MZclust method. perform correspondence analysis groupChromPeaks method using default settings. Getting overview performed processings: peak group information, .e. feature definitions can accessed featureDefinitions method. Plotting raw data direct injection samples involves little processing LC/GC-MS data can simply use chromatogram method extract data. extract m/z-intensity pairs peaks associated first feature. thus first identify peaks feature define m/z values range. Using range can subsequently use filterMz function sub-set full data set signal associated feature’s peaks. object can call mz intensity functions extract data.  access actual intensity values feature sample featureValue method can used. setting value = \"\" tells function return integrated signal peak (one representative peak) per sample. NA reported features samples peak identified feature’s m/z value. instances might still signal feature’s position raw data files, peak detection failed identify peak. cases signal can recovered using fillChromPeaks method integrates raw signal feature’s location. signal location NA reported.","code":"## Using default settings but define sample group assignment mzc_prm <- MzClustParam(sampleGroups = ham_prep$sample_group) ham_prep <- groupChromPeaks(ham_prep, param = mzc_prm) ham_prep ## MSn experiment data (\"XCMSnExp\") ## Object size in memory: 0.04 Mb ## - - - Spectra data - - - ##  MS level(s): 1  ##  Number of spectra: 4  ##  MSn retention times: -1:59 - -1:59 minutes ## - - - Processing information - - - ## Data loaded [Fri Jan 26 20:08:58 2024]  ##  MSnbase version: 2.29.3  ## - - - Meta data  - - - ## phenoData ##   rowNames: 1 2 3 4 ##   varLabels: filename sample_group ##   varMetadata: labelDescription ## Loaded from: ##   [1] HAM004_641fE_14-11-07--Exp1.extracted.mzML...  [4] HAM005_641fE_14-11-07--Exp2.extracted.mzML ##   Use 'fileNames(.)' to see all files. ## protocolData: none ## featureData ##   featureNames: F1.S1 F2.S1 F3.S1 F4.S1 ##   fvarLabels: fileIdx spIdx ... spectrum (35 total) ##   fvarMetadata: labelDescription ## experimentData: use 'experimentData(object)' ## - - - xcms preprocessing - - - ## Chromatographic peak detection: ##  method: MSW  ##  38 peaks identified in 4 samples. ##  On average 9.5 chromatographic peaks per sample. ## Correspondence: ##  method: mzClust  ##  20 features identified. ##  Median mz range of features: 9.1553e-05 ##  Median rt range of features: 0 featureDefinitions(ham_prep) ## DataFrame with 20 rows and 10 columns ##          mzmed     mzmin     mzmax     rtmed     rtmin     rtmax    npeaks ##      <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> ## FT01   402.285   402.285   402.286        -1        -1        -1         2 ## FT02   403.237   403.237   403.237        -1        -1        -1         4 ## FT03   405.109   405.109   405.109        -1        -1        -1         2 ## FT04   409.184   409.184   409.185        -1        -1        -1         2 ## FT05   410.144   410.144   410.145        -1        -1        -1         2 ## ...        ...       ...       ...       ...       ...       ...       ... ## FT16   437.238   437.238   437.238        -1        -1        -1         2 ## FT17   438.240   438.240   438.240        -1        -1        -1         2 ## FT18   439.151   439.151   439.151        -1        -1        -1         2 ## FT19   441.130   441.130   441.131        -1        -1        -1         2 ## FT20   445.293   445.292   445.293        -1        -1        -1         2 ##           ham4      ham5     peakidx ##      <numeric> <numeric>      <list> ## FT01         0         2       16,28 ## FT02         2         2 17,29,1,... ## FT03         0         2       18,30 ## FT04         2         0        10,2 ## FT05         0         2       19,31 ## ...        ...       ...         ... ## FT16         2         0        6,13 ## FT17         2         0        7,14 ## FT18         0         2       26,37 ## FT19         0         2       38,27 ## FT20         2         0        15,8 ## Get the peaks belonging to the first feature pks <- chromPeaks(ham_prep)[featureDefinitions(ham_prep)$peakidx[[1]], ]  ## Define the m/z range mzr <- c(min(pks[, \"mzmin\"]) - 0.001, max(pks[, \"mzmax\"]) + 0.001)  ## Subset the object to the m/z range ham_prep_sub <- filterMz(ham_prep, mz = mzr)  ## Extract the mz and intensity values mzs <- mz(ham_prep_sub, bySample = TRUE) ints <- intensity(ham_prep_sub, bySample = TRUE)  ## Plot the data plot(3, 3, pch = NA, xlim = range(mzs), ylim = range(ints), main = \"FT01\",      xlab = \"m/z\", ylab = \"intensity\") ## Define colors cols <- rep(\"#ff000080\", length(mzs)) cols[ham_prep_sub$sample_group == \"ham5\"] <- \"#0000ff80\" tmp <- mapply(mzs, ints, cols, FUN = function(x, y, col) {     points(x, y, col = col, type = \"l\") }) feat_vals <- featureValues(ham_prep, value = \"into\") head(feat_vals) ##      HAM004_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                         NA ## FT02                                    4735258 ## FT03                                         NA ## FT04                                    4158404 ## FT05                                         NA ## FT06                                    6099006 ##      HAM004_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                         NA ## FT02                                    6202418 ## FT03                                         NA ## FT04                                    5004546 ## FT05                                         NA ## FT06                                    4950642 ##      HAM005_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                    4095293 ## FT02                                    4811391 ## FT03                                    2982453 ## FT04                                         NA ## FT05                                    2872023 ## FT06                                         NA ##      HAM005_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                    4804763 ## FT02                                    2581183 ## FT03                                    2268984 ## FT04                                         NA ## FT05                                    2133219 ## FT06                                         NA ham_prep <- fillChromPeaks(ham_prep, param = FillChromPeaksParam())  head(featureValues(ham_prep, value = \"into\")) ##      HAM004_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                   768754.0 ## FT02                                  4735257.5 ## FT03                                   652566.6 ## FT04                                  4158404.5 ## FT05                                   652201.1 ## FT06                                  6099006.3 ##      HAM004_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                  1230140.4 ## FT02                                  6202417.6 ## FT03                                   374109.9 ## FT04                                  5004546.3 ## FT05                                   403448.4 ## FT06                                  4950641.7 ##      HAM005_641fE_14-11-07--Exp1.extracted.mzML ## FT01                                    4095293 ## FT02                                    4811391 ## FT03                                    2982453 ## FT04                                    1221031 ## FT05                                    2872023 ## FT06                                    1573988 ##      HAM005_641fE_14-11-07--Exp2.extracted.mzML ## FT01                                  4804762.5 ## FT02                                  2581183.1 ## FT03                                  2268984.5 ## FT04                                  1241294.4 ## FT05                                  2133219.4 ## FT06                                   977694.5"},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"further-analysis","dir":"Articles","previous_headings":"","what":"Further analysis","title":"Grouping FTICR-MS data with xcms","text":"analysis, .e. detection features/metabolites significantly different abundances, PCA analyses can performed feature matrix using functionality R packages, limma.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-direct-injection.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"Grouping FTICR-MS data with xcms","text":"","code":"sessionInfo() ## R Under development (unstable) (2024-01-17 r85813) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] MassSpecWavelet_1.69.0 xcms_4.1.6             MSnbase_2.29.3         ##  [4] ProtGenerics_1.35.2    S4Vectors_0.41.3       mzR_2.37.0             ##  [7] Rcpp_1.0.12            Biobase_2.63.0         BiocGenerics_0.49.1    ## [10] BiocParallel_1.37.0    BiocStyle_2.31.0       ##  ## loaded via a namespace (and not attached): ##   [1] DBI_1.2.1                   bitops_1.0-7                ##   [3] rlang_1.1.3                 magrittr_2.0.3              ##   [5] clue_0.3-65                 matrixStats_1.2.0           ##   [7] compiler_4.4.0              systemfonts_1.0.5           ##   [9] vctrs_0.6.5                 stringr_1.5.1               ##  [11] MetaboCoreUtils_1.11.2      pkgconfig_2.0.3             ##  [13] crayon_1.5.2                fastmap_1.1.1               ##  [15] XVector_0.43.1              utf8_1.2.4                  ##  [17] rmarkdown_2.25              preprocessCore_1.65.0       ##  [19] ragg_1.2.7                  purrr_1.0.2                 ##  [21] xfun_0.41                   MultiAssayExperiment_1.29.0 ##  [23] zlibbioc_1.49.0             cachem_1.0.8                ##  [25] GenomeInfoDb_1.39.5         jsonlite_1.8.8              ##  [27] progress_1.2.3              highr_0.10                  ##  [29] DelayedArray_0.29.0         prettyunits_1.2.0           ##  [31] parallel_4.4.0              cluster_2.1.6               ##  [33] R6_2.5.1                    RColorBrewer_1.1-3          ##  [35] bslib_0.6.1                 stringi_1.8.3               ##  [37] limma_3.59.1                GenomicRanges_1.55.2        ##  [39] jquerylib_0.1.4             bookdown_0.37               ##  [41] SummarizedExperiment_1.33.3 iterators_1.0.14            ##  [43] knitr_1.45                  IRanges_2.37.1              ##  [45] splines_4.4.0               Matrix_1.6-5                ##  [47] igraph_1.6.0                tidyselect_1.2.0            ##  [49] abind_1.4-5                 yaml_2.3.8                  ##  [51] doParallel_1.0.17           codetools_0.2-19            ##  [53] affy_1.81.0                 lattice_0.22-5              ##  [55] tibble_3.2.1                plyr_1.8.9                  ##  [57] signal_1.8-0                evaluate_0.23               ##  [59] survival_3.5-7              desc_1.4.3                  ##  [61] Spectra_1.13.3              pillar_1.9.0                ##  [63] affyio_1.73.0               BiocManager_1.30.22         ##  [65] MatrixGenerics_1.15.0       foreach_1.5.2               ##  [67] MALDIquant_1.22.2           ncdf4_1.22                  ##  [69] generics_0.1.3              RCurl_1.98-1.14             ##  [71] hms_1.1.3                   ggplot2_3.4.4               ##  [73] munsell_0.5.0               scales_1.3.0                ##  [75] MsExperiment_1.5.4          glue_1.7.0                  ##  [77] MsFeatures_1.11.0           lazyeval_0.2.2              ##  [79] tools_4.4.0                 robustbase_0.99-1           ##  [81] mzID_1.41.0                 QFeatures_1.13.2            ##  [83] vsn_3.71.0                  RANN_2.6.1                  ##  [85] fs_1.6.3                    XML_3.99-0.16.1             ##  [87] grid_4.4.0                  impute_1.77.0               ##  [89] MsCoreUtils_1.15.3          colorspace_2.1-0            ##  [91] GenomeInfoDbData_1.2.11     PSMatch_1.7.1               ##  [93] cli_3.6.2                   textshaping_0.3.7           ##  [95] fansi_1.0.6                 S4Arrays_1.3.2              ##  [97] dplyr_1.1.4                 AnnotationFilter_1.27.0     ##  [99] DEoptimR_1.1-3              pcaMethods_1.95.0           ## [101] gtable_0.3.4                sass_0.4.8                  ## [103] digest_0.6.34               SparseArray_1.3.3           ## [105] multtest_2.59.0             memoise_2.0.1               ## [107] htmltools_0.5.7             pkgdown_2.0.7.9000          ## [109] lifecycle_1.0.4             statmod_1.5.0               ## [111] MASS_7.3-60.2"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"LC-MS/MS data analysis with xcms","text":"Metabolite identification important step non-targeted metabolomics requires different steps. One involves use tandem mass spectrometry generate fragmentation spectra detected metabolites (LC-MS/MS), compared fragmentation spectra known metabolites. Different approaches exist generation fragmentation spectra, whereas used data dependent acquisition (DDA) also known top-n method. method top N intense ions (m/z values) MS1 scan selected fragmentation next N scans cycle starts . method allows generate clean MS2 fragmentation spectra fly acquisition without need experiments, suffers poor coverage detected metabolites (since limited number ions fragmented) less accurate quantification compounds (since fewer MS1 scans generated). Data independent approaches (DIA) like Bruker bbCID, Agilent AllIons Waters MSe don’t use preselection, rather fragment detected molecules . using alternating schemes scan low high collision energy collect MS1 MS2 data. Using approach, problem coverage, relation precursor fragment masses lost leading chimeric spectra. Sequential Window Acquisition Theoretical Mass Spectra (SWATH [1]) combines approaches middle-way approach. precursor selection acquisition independent acquired data, rather isolating precusors , defined windows (.e. ranges m/z values) used scanned. reduces overlap fragment spectra still keeping high coverage. document showcases analysis two small LC-MS/MS data sets using xcms. data files used reversed-phase LC-MS/MS runs Agilent Pesticide mix obtained Sciex 6600 Triple ToF operated SWATH acquisition mode. comparison DDA file sample included.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"analysis-of-dda-data","dir":"Articles","previous_headings":"","what":"Analysis of DDA data","title":"LC-MS/MS data analysis with xcms","text":"load example DDA data set create total ion chromatogram MS1 data. According TIC signal measured ~ 200 600 seconds (see plot ). thus filter DDA data retention time range.  variable dda_data contains now MS1 MS2 spectra specified mzML file. number spectra MS level listed . Note subset experiment first data file (using [1]) access directly spectra within sample spectra function (returns Spectra object Spectra package). Note use pipe operator |> better readability. MS2 spectra can get m/z precursor ion precursorMz function. first subset data set single sample filter spectra MS level 2 extracting precursor m/z values. precursorIntensity function also possible extract intensity precursor ion. manufacturers (like Sciex) don’t define/export precursor intensity thus either NA 0 reported. can however use estimatePrecursorIntensity function Spectra package determine precursor intensity MS 2 spectrum based intensity respective ion previous MS1 scan (note method = \"interpolation\" precursor intensity defined based interpolation intensity previous subsequent MS1 scan). estimate precursor intensities, full data (MS1 spectra NA value reported). next set precursor intensity spectrum metadata dda_data. can extracted later precursorIntensity function. Next perform chromatographic peak detection MS level 1 data findChromPeaks method. define settings centWave-based peak detection perform analysis. total 114 peaks identified present data set. advantage LC-MS/MS data (MS1) ions fragmented corresponding MS2 spectra measured. Thus, ions (identified MS1 chromatographic peaks) MS2 spectra available. can facilitate annotation respective MS1 chromatographic peaks (MS1 features correspondence analysis). Spectra identified chromatographic peaks can extracted chromPeakSpectra method. MS2 spectra precursor m/z retention time within rt m/z range chromatographic peak returned. default chromPeakSpectra returns spectra associated MS1 chromatographic peak, parameter method allows choose return one spectrum per peak (look ?chromPeakSpectra help page details). Also, possible extract MS1 spectra peak specifying msLevel = 1L call (e.g. evaluate full MS1 signal peak’s apex position). returned Spectra contains also reference respective chromatographic peak additional spectra variable \"peak_id\" contains identifier chromatographic peak (.e. row name chromPeaks matrix). Note also return.type = \"List\" list parallel chromPeaks matrix returned, .e. element list contain spectra chromatographic peak index. data representation might eventually simplify processing. next use MS2 information aid annotation chromatographic peak. example use chromatographic peak ion m/z 304.1131 extract code block . search potential ions similar m/z reference database (e.g. Metlin) returned large list potential hits, small ppm. two hits, Flumazenil (Metlin ID 2724) Fenamiphos (Metlin ID 72445) experimental MS2 spectra available. Thus, match MS2 spectrum identified chromatographic peak annotate ion. extract MS2 spectra associated candidate chromatographic peak using ID peak present data set. 5 MS2 spectra representing fragmentation ion(s) measured candidate chromatographic peak. next reduce single MS2 spectrum using combineSpectra method employing combinePeaks function determine peaks keep resulting spectrum (look ?combinePeaks help page details). Parameter f allows specify spectra input object combined one. Note combination multiple fragment spectra single spectrum might generally best approach suggested types data. Mass peaks input spectra difference m/z smaller 20 ppm (parameter ppm) combined one peak median m/z intensity reported . Due parameter minProp = 0.8, resulting MS2 spectrum contains peaks present 80% input spectra. plot consensus spectrum shown . Consensus MS2 spectrum created measured MS2 spectra ions chromatographic peak CP53. now match consensus spectrum database MS2 spectra. example simply load MS2 spectra two compounds matching m/z exported Metlin. compounds MS2 spectra created collision energies 0V, 10V, 20V 40V available. import respective data plot candidate spectrum MS2 spectra Flumanezil Fenamiphos (collision energy 20V). import files MGF format load MsBackendMgf Bioconductor package adds MGF file support Spectra package. Prior plotting scale experimental spectra replace peak intensities values relative maximum peak intensity (set value 100). Mirror plots candidate MS2 spectrum Flumanezil (left) Fenamiphos (right). upper panel represents candidate MS2 spectrum, lower target MS2 spectrum. Matching peaks indicated dot. candidate spectrum matches Fenamiphos, thus, example chromatographic peak represents signal measured compound. addition plotting spectra, can also calculate similarities compareSpectra method (uses default normalized dot-product calculate similarity). Clearly, candidate spectrum match Flumanezil, high similarity Fenamiphos. performed MS2-based annotation single chromatographic peak, easily extended full list MS2 spectra (returned chromPeakSpectra) chromatographic peaks experiment. See also alternative tutorials matching experimental fragment spectra reference. present example used single data file thus need perform sample alignment correspondence analysis. tasks however performed similarly plain LC-MS data, retention times recorded MS2 spectra however also adjusted alignment based MS1 data. correspondence analysis (peak grouping) MS2 spectra features can extracted featureSpectra function returns MS2 spectra associated chromatographic peak feature. Note also workflow can included Feature-Based Molecular Networking FBMN match MS2 spectra GNPS. See details examples.","code":"library(xcms) library(MsExperiment)  dda_file <- system.file(\"TripleTOF-SWATH\", \"PestMix1_DDA.mzML\",                         package = \"msdata\") dda_data <- readMsExperiment(dda_file) chr <- chromatogram(dda_data, aggregationFun = \"sum\", msLevel = 1L) plot(chr) abline(v = c(230, 610)) ## filter the data dda_data <- filterRt(dda_data, rt = c(230, 610)) ## Filter spectra dda_data[1] |> spectra() |> msLevel() |> table() ##  ##    1    2  ## 1389 2214 dda_data[1] |> spectra() |> filterMsLevel(2) |> precursorMz() |> head() ## [1] 182.18777 182.18893  55.00579 182.19032 237.12296 237.11987 dda_data[1] |> spectra() |> filterMsLevel(2) |> precursorIntensity() |> head() ## [1] 0 0 0 0 0 0 prec_int <- estimatePrecursorIntensity(spectra(dda_data)) spectra(dda_data)$precursorIntensity <- prec_int  dda_data[1] |> spectra() |> filterMsLevel(2) |> precursorIntensity() |> head() ## [1]        NA  9.198155  2.773988 27.590797  3.443145  7.621923 cwp <- CentWaveParam(snthresh = 5, noise = 100, ppm = 10,                      peakwidth = c(3, 30)) dda_data <- findChromPeaks(dda_data, param = cwp, msLevel = 1L) library(Spectra) dda_spectra <- chromPeakSpectra(dda_data, msLevel = 2L) dda_spectra ## MSn data (Spectra) with 142 spectra in a MsBackendMzR backend: ##       msLevel     rtime scanIndex ##     <integer> <numeric> <integer> ## 1           2   237.869      1812 ## 2           2   241.299      1846 ## 3           2   325.763      2439 ## 4           2   326.583      2446 ## 5           2   327.113      2450 ## ...       ...       ...       ... ## 138         2   574.725      5110 ## 139         2   575.255      5115 ## 140         2   596.584      5272 ## 141         2   592.424      5236 ## 142         2   596.054      5266 ##  ... 34 more variables/columns. ##  ## file(s): ## PestMix1_DDA.mzML ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024] ##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024] ##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024] dda_spectra$peak_id ##   [1] \"CP004\" \"CP004\" \"CP005\" \"CP005\" \"CP006\" \"CP006\" \"CP008\" \"CP008\" \"CP011\" ##  [10] \"CP011\" \"CP012\" \"CP012\" \"CP013\" \"CP013\" \"CP013\" \"CP013\" \"CP014\" \"CP014\" ##  [19] \"CP014\" \"CP014\" \"CP018\" \"CP022\" \"CP022\" \"CP022\" \"CP025\" \"CP025\" \"CP025\" ##  [28] \"CP025\" \"CP026\" \"CP026\" \"CP026\" \"CP026\" \"CP033\" \"CP033\" \"CP034\" \"CP034\" ##  [37] \"CP034\" \"CP034\" \"CP034\" \"CP035\" \"CP035\" \"CP035\" \"CP041\" \"CP041\" \"CP041\" ##  [46] \"CP042\" \"CP042\" \"CP042\" \"CP043\" \"CP047\" \"CP047\" \"CP049\" \"CP049\" \"CP049\" ##  [55] \"CP049\" \"CP050\" \"CP050\" \"CP050\" \"CP051\" \"CP051\" \"CP051\" \"CP054\" \"CP055\" ##  [64] \"CP055\" \"CP055\" \"CP056\" \"CP056\" \"CP056\" \"CP056\" \"CP056\" \"CP060\" \"CP060\" ##  [73] \"CP060\" \"CP060\" \"CP064\" \"CP064\" \"CP065\" \"CP065\" \"CP066\" \"CP066\" \"CP069\" ##  [82] \"CP069\" \"CP069\" \"CP070\" \"CP070\" \"CP070\" \"CP072\" \"CP072\" \"CP072\" \"CP073\" ##  [91] \"CP074\" \"CP074\" \"CP074\" \"CP074\" \"CP075\" \"CP075\" \"CP075\" \"CP077\" \"CP077\" ## [100] \"CP077\" \"CP079\" \"CP079\" \"CP079\" \"CP079\" \"CP080\" \"CP080\" \"CP080\" \"CP081\" ## [109] \"CP086\" \"CP086\" \"CP086\" \"CP086\" \"CP086\" \"CP088\" \"CP088\" \"CP088\" \"CP089\" ## [118] \"CP089\" \"CP091\" \"CP091\" \"CP093\" \"CP093\" \"CP094\" \"CP094\" \"CP094\" \"CP095\" ## [127] \"CP095\" \"CP095\" \"CP096\" \"CP096\" \"CP096\" \"CP098\" \"CP098\" \"CP098\" \"CP098\" ## [136] \"CP098\" \"CP099\" \"CP099\" \"CP099\" \"CP100\" \"CP101\" \"CP101\" ex_mz <- 304.1131 chromPeaks(dda_data, mz = ex_mz, ppm = 20) ##             mz    mzmin    mzmax      rt   rtmin   rtmax    into     intb ## CP056 304.1133 304.1126 304.1143 425.024 417.985 441.773 13040.8 13007.79 ##           maxo  sn sample ## CP056 3978.987 232      1 ex_id <- rownames(chromPeaks(dda_data, mz = ex_mz, ppm = 20)) ex_spectra <- dda_spectra[dda_spectra$peak_id == ex_id] ex_spectra ## MSn data (Spectra) with 5 spectra in a MsBackendMzR backend: ##     msLevel     rtime scanIndex ##   <integer> <numeric> <integer> ## 1         2   418.926      3505 ## 2         2   419.306      3510 ## 3         2   423.036      3582 ## 4         2   423.966      3603 ## 5         2   424.296      3609 ##  ... 34 more variables/columns. ##  ## file(s): ## PestMix1_DDA.mzML ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024] ##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024] ##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024] ex_spectrum <- combineSpectra(ex_spectra, FUN = combinePeaks, ppm = 20,                               peaks = \"intersect\", minProp = 0.8,                               intensityFun = median, mzFun = median,                               f = ex_spectra$peak_id) ## Warning: 'combinePeaks' for lists of peak matrices is deprecated; please use ## 'combinePeaksData' instead. ex_spectrum ## MSn data (Spectra) with 1 spectra in a MsBackendMemory backend: ##     msLevel     rtime scanIndex ##   <integer> <numeric> <integer> ## 1         2   418.926      3505 ##  ... 34 more variables/columns. ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024] ##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024] ##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024] ##  ...1 more processings. Use 'processingLog' to list all. plotSpectra(ex_spectrum) scale_fun <- function(z, ...) {     z[, \"intensity\"] <- z[, \"intensity\"] /         max(z[, \"intensity\"], na.rm = TRUE) * 100     z } ex_spectrum <- addProcessing(ex_spectrum, FUN = scale_fun) library(MsBackendMgf) flumanezil <- Spectra(     system.file(\"mgf\", \"metlin-2724.mgf\", package = \"xcms\"),     source = MsBackendMgf()) ## Start data import from 1 files ... done fenamiphos <- Spectra(     system.file(\"mgf\", \"metlin-72445.mgf\", package = \"xcms\"),     source = MsBackendMgf()) ## Start data import from 1 files ... done par(mfrow = c(1, 2)) plotSpectraMirror(ex_spectrum, flumanezil[3], main = \"against Flumanezil\",                   ppm = 40) plotSpectraMirror(ex_spectrum, fenamiphos[3], main = \"against Fenamiphos\",                   ppm = 40) compareSpectra(ex_spectrum, flumanezil, ppm = 40) ## [1] 4.520957e-02 3.283806e-02 2.049379e-03 3.374354e-05 compareSpectra(ex_spectrum, fenamiphos, ppm = 40) ## [1] 0.1326234432 0.4879399946 0.7198406271 0.3997922658 0.0004876129 ## [6] 0.0028408885 0.0071030051 0.0053809736"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"dia-swath-data-analysis","dir":"Articles","previous_headings":"","what":"DIA (SWATH) data analysis","title":"LC-MS/MS data analysis with xcms","text":"section analyze small SWATH data set consisting single mzML file data sample analyzed previous section recorded SWATH mode. read data readMsExperiment function. resulting object contain recorded MS1 MS2 spectra specified file. Similar previous data file, filter file signal 230 610 seconds. determine number MS level 1 2 spectra present data set. described introduction, SWATH mode ions within pre-defined isolation windows fragmented MS2 spectra measured. definition isolation windows (SWATH pockets) imported mzML files available additional spectra variables. inspect respective information first spectra. upper lower isolation window m/z available spectra variables \"isolationWindowLowerMz\" \"isolationWindowUpperMz\" respectively target m/z isolation window \"isolationWindowTargetMz\". can use spectraData function extract information spectra within swath_data object. also access variables directly dedicated isolationWindowLowerMz isolationWindowUpperMz functions. present data set use value isolation window target m/z define individual SWATH pockets. list number spectra recorded pocket/isolation window. thus 422 423 MS2 spectra measured isolation window. inspect data can also extract chromatograms measured MS1 well MS2 data. MS2 data set parameter msLevel = 2L addition also specify isolation window want extract data. extract TIC MS1 data one isolation windows (isolation window target m/z 270.85) plot . TIC MS1 (upper panel) MS2 data isolation window target m/z 270.85 (lower panel). Note also DIA data manufacturers (e.g. Waters MSe) supported long spectra variable isolationWindowTargetMz available. variable provided raw data files can also manually assigned spectrum data set. example precursor m/z individual spectra used \"isolationWindowTargetMz\" (note: line shown example evaluated).","code":"swath_file <- system.file(\"TripleTOF-SWATH\",                           \"PestMix1_SWATH.mzML\",                           package = \"msdata\")  swath_data <- readMsExperiment(swath_file) swath_data <- filterRt(swath_data, rt = c(230, 610)) spectra(swath_data) |> msLevel() |> table() ##  ##    1    2  ##  422 3378 spectra(swath_data) |> spectraData(c(\"isolationWindowTargetMz\", \"isolationWindowLowerMz\",               \"isolationWindowUpperMz\", \"msLevel\", \"rtime\")) |> head() ## DataFrame with 6 rows and 5 columns ##   isolationWindowTargetMz isolationWindowLowerMz isolationWindowUpperMz ##                 <numeric>              <numeric>              <numeric> ## 1                  299.10                  283.5                  314.7 ## 2                  329.80                  313.7                  345.9 ## 3                  367.35                  344.9                  389.8 ## 4                  601.85                  388.8                  814.9 ## 5                      NA                     NA                     NA ## 6                  163.75                  139.5                  188.0 ##     msLevel     rtime ##   <integer> <numeric> ## 1         2   230.073 ## 2         2   230.170 ## 3         2   230.267 ## 4         2   230.364 ## 5         1   230.459 ## 6         2   230.585 head(isolationWindowLowerMz(spectra(swath_data))) ## [1] 283.5 313.7 344.9 388.8    NA 139.5 head(isolationWindowUpperMz(spectra(swath_data))) ## [1] 314.7 345.9 389.8 814.9    NA 188.0 table(isolationWindowTargetMz(spectra(swath_data))) ##  ## 163.75 208.95 244.05 270.85  299.1  329.8 367.35 601.85  ##    422    422    422    422    423    423    422    422 tic_ms1 <- chromatogram(swath_data, msLevel = 1L, aggregationFun = \"sum\") tic_ms2 <- chromatogram(swath_data, msLevel = 2L, aggregationFun = \"sum\",                         isolationWindowTargetMz = 270.85) par(mfrow = c(2, 1)) plot(tic_ms1, main = \"MS1\") plot(tic_ms2, main = \"MS2, isolation window m/z 270.85\") spectra(swath_data)$isolationWindowTargetMz <- precursorMz(spectra(swath_data))"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"chromatographic-peak-detection-in-ms1-and-ms2-data","dir":"Articles","previous_headings":"DIA (SWATH) data analysis","what":"Chromatographic peak detection in MS1 and MS2 data","title":"LC-MS/MS data analysis with xcms","text":"Similar conventional LC-MS analysis, perform first chromatographic peak detection (MS level 1 data) findChromPeaks method. define settings centWave-based peak detection perform analysis. Next perform chromatographic peak detection MS level 2 data separately individual isolation window. use findChromPeaksIsolationWindow function employing peak detection algorithm reducing however required signal--noise ratio. isolationWindow parameter allows specify MS2 spectra belong isolation window hence defines set MS2 spectra chromatographic peak detection performed. default \"isolationWindowTargetMz\" variable object’s spectra used. findChromPeaksIsolationWindow function added peaks identified individual isolation windows chromPeaks matrix containing already MS1 chromatographic peaks. newly added peaks can identified \"isolationWindow\" column object’s chromPeakData. count number chromatographic peaks identified within isolation window (number chromatographic peaks identified MS1 62). thus successfully identified chromatographic peaks different MS levels isolation windows. next step identify measured signals represents data original compound reconstruct fragment spectra MS1 signal (chromatographic peak).","code":"cwp <- CentWaveParam(snthresh = 5, noise = 100, ppm = 10,                      peakwidth = c(3, 30)) swath_data <- findChromPeaks(swath_data, param = cwp) swath_data ## Object of class XcmsExperiment  ##  Spectra: MS1 (422) MS2 (3378)  ##  Experiment data: 1 sample(s) ##  Sample data links: ##   - spectra: 1 sample(s) to 3800 element(s). ##  xcms results: ##   - chromatographic peaks: 62 in MS level(s): 1 cwp <- CentWaveParam(snthresh = 3, noise = 10, ppm = 10,                      peakwidth = c(3, 30)) swath_data <- findChromPeaksIsolationWindow(swath_data, param = cwp) swath_data ## Object of class XcmsExperiment  ##  Spectra: MS1 (422) MS2 (3378)  ##  Experiment data: 1 sample(s) ##  Sample data links: ##   - spectra: 1 sample(s) to 3800 element(s). ##  xcms results: ##   - chromatographic peaks: 370 in MS level(s): 1, 2 chromPeakData(swath_data) ## DataFrame with 370 rows and 6 columns ##        ms_level is_filled isolationWindow isolationWindowTargetMZ ##       <integer> <logical>       <numeric>               <numeric> ## CP01          1     FALSE              NA                      NA ## CP02          1     FALSE              NA                      NA ## CP03          1     FALSE              NA                      NA ## CP04          1     FALSE              NA                      NA ## CP05          1     FALSE              NA                      NA ## ...         ...       ...             ...                     ... ## CP366         2     FALSE          601.85                  601.85 ## CP367         2     FALSE          601.85                  601.85 ## CP368         2     FALSE          601.85                  601.85 ## CP369         2     FALSE          601.85                  601.85 ## CP370         2     FALSE          601.85                  601.85 ##       isolationWindowLowerMz isolationWindowUpperMz ##                    <numeric>              <numeric> ## CP01                      NA                     NA ## CP02                      NA                     NA ## CP03                      NA                     NA ## CP04                      NA                     NA ## CP05                      NA                     NA ## ...                      ...                    ... ## CP366                  388.8                  814.9 ## CP367                  388.8                  814.9 ## CP368                  388.8                  814.9 ## CP369                  388.8                  814.9 ## CP370                  388.8                  814.9 table(chromPeakData(swath_data)$isolationWindow) ##  ## 163.75 208.95 244.05 270.85  299.1  329.8 367.35 601.85  ##      2     38     32     14    105     23     62     32"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"reconstruction-of-ms2-spectra","dir":"Articles","previous_headings":"DIA (SWATH) data analysis","what":"Reconstruction of MS2 spectra","title":"LC-MS/MS data analysis with xcms","text":"Identifying signal fragment ions precursor measured MS1 chromatographic peak non-trivial task. MS2 spectrum fragment ion MS1 chromatographic peak reconstructed available MS2 signal (.e. chromatographic peaks identified MS level 2). SWATH data, fragment ion signal present isolation window contains m/z precursor ion chromatographic peak shape MS2 chromatographic peaks fragment ions specific precursor similar retention time peak shape precursor’s MS1 chromatographic peak. detection MS1 MS2 chromatographic peaks performed, can reconstruct MS2 spectra using reconstructChromPeakSpectra function. function defines MS2 spectrum MS1 chromatographic peak based following approach: Identify MS2 chromatographic peaks isolation window containing m/z ion (MS1 chromatographic peak) approximately retention time MS1 chromatographic peak (accepted difference retention time can defined diffRt parameter). Extract MS1 chromatographic peak MS2 chromatographic peaks identified previous step correlate peak shapes candidate MS2 chromatographic peaks shape MS1 peak. MS2 chromatographic peaks correlation coefficient larger minCor retained. Reconstruct MS2 spectrum using m/z selected MS2 chromatographic peaks intensity; MS2 chromatographic peak selected MS1 peak thus represent one mass peak reconstructed spectrum. illustrate process perform individual steps example fenamiphos (exact mass 303.105800777 m/z [M+H]+ adduct 304.113077). first step extract chromatographic peak ion. Next identify MS2 chromatographic peaks identified isolation window containing m/z fenamiphos. information isolation window chromatographic peak identified available chromPeakData. also require retention time MS2 chromatographic peaks similar retention time MS1 peak extract corresponding peak information. thus select chromatographic peaks retention time range contains retention time apex position MS1 chromatographic peak. total 24 MS2 chromatographic peaks match conditions. Next extract ion chromatogram MS1 peak selected candidate MS2 signals. ensure chromatograms extracted spectra correct isolation window need specify respective isolation window passing isolation window target m/z chromatogram function (addition setting msLevel = 2). can done either getting isolationWindowTargetMz spectra data subset using filterIsolationWindow (done ) selecting isolationWindowTargetMz closest m/z compound interest. target m/z isolation window containing m/z interest thus 299.1 can use chromatogram call extract data correct (MS2) spectra. can now plot extracted ion chromatogram MS1 extracted MS2 data. Extracted ion chromatograms Fenamiphos MS1 (blue) potentially related signal MS2 (grey). Next can calculate correlations peak shapes MS2 chromatogram MS1 peak. illustrate process example one MS2 chromatographic peaks. Note , MS1 MS2 spectra recorded consecutively, retention times individual data points differ MS2 MS1 chromatographic data data points thus matched (aligned) performing correlation analysis. done automatically correlate function. See help align method information alignment options. identifying MS2 chromatographic peaks shapes enough high similarity MS1 chromatographic peaks, MS2 spectrum reconstructed based m/z intensities MS2 chromatographic peaks (.e., using \"mz\" \"maxo\" \"\" values). Instead performing assignment MS2 signal MS1 chromatographic peaks manually , can use reconstructChromPeakSpectra function performs exact steps MS1 chromatographic peaks DIA data set. use function reconstruct MS2 spectra example data requiring peak shape correlation higher 0.9 candidate MS2 chromatographic peak target MS1 chromatographic peak. result got Spectra object length equal number MS1 peaks data. length spectrum represents number peaks contains. Thus, length 0 indicates matching peak (MS2 signal) found respective MS1 chromatographic peak. reconstructed spectra additional annotations available IDs MS2 chromatographic peaks spectrum reconstructed (\"ms2_peak_id\") well correlation coefficient chromatographic peak shape precursor’s shape (\"ms2_peak_cor\"). Metadata column \"peak_id\" contains ID MS1 chromatographic peak: next extract MS2 spectrum example peak likely representing [M+H]+ ions Fenamiphos using chromatographic peak ID: can now compare reconstructed spectrum example consensus spectrum DDA experiment previous section (variable ex_spectrum) well MS2 spectrum Fenamiphos Metlin (collision energy 10V). better visualization normalize also peak intensities reconstructed SWATH spectrum function used experimental DDA spectrum. Mirror plot comparing reconstructed MS2 spectrum Fenamiphos (upper panel) measured spectrum DDA data Fenamiphhos spectrum Metlin. wanted get EICs MS2 chromatographic peaks used generate MS2 spectrum can use IDs peaks provided $ms2_peak_id result spectrum. peak IDs available can extract retention time window m/z ranges chromPeaks matrix use chromatogram function extract EIC. Note however SWATH data MS2 signal different isolation windows. Thus first filter swath_data object isolation window containing precursor m/z filterIsolationWindow subset data MS2 spectra related ion interest. addition, use msLevel = 2L chromatogram call chromatogram extracts default data MS1 spectra need specify target m/z isolation window containing fragment data compound interest. row ms2_eics contains now EIC one MS2 chromatographic peaks. can also plot overlay plot. Overlay EICs chromatographic peaks used reconstruct MS2 spectrum fenamiphos. second example analyze signal [M+H]+ ion m/z 376.0381 (match Prochloraz). first identify MS1 chromatographic peak m/z retrieve reconstructed MS2 spectrum peak. MS2 spectrum (tentative) MS1 signal prochloraz reconstructed SWATH MS2 data thus 9 peaks. addition identify corresponding MS1 peak DDA data set, extract measured MS2 chromatographic peaks build consensus spectrum . retention times chromatographic peaks DDA SWATH data match almost perfectly. Next get MS2 spectra peak. total 5 spectra measured, relatively high number peaks. Next combine consensus spectrum. last load also Prochloraz MS2 spectra (different collision energies) Metlin. validate reconstructed spectrum plot corresponding DDA spectrum MS2 spectrum Prochloraz (collision energy 10V) Metlin. Mirror plot comparing reconstructed MS2 spectrum Prochloraz (upper panel) measured spectrum DDA data Prochloraz spectrum Metlin. spectra fit relatively well. Interestingly, peak representing precursor (right-peak) seems slightly shifted m/z value reconstructed spectrum. Also, closer inspecting spectrum two groups peaks small differences m/z can observed (see plot ). SWATH-derived MS2 spectrum prochloraz. represent fragments isotopes original compound. DIA MS2 data, since ions given retention time fragmented, can contain fragments isotopes. thus use isotopologues function MetaboCoreUtils package check presence potential isotope peaks reconstructed MS2 spectrum prochloraz. Indeed, peaks 3, 4 5 well 6 7 assigned group potential isotope peaks. proof peaks indeed fragment isotopes prochloraz highly likely (given difference m/z relative intensity differences). thus define function keeps monoisotopic peak isotope group spectrum. SWATH MS2 spectrum prochloraz (left) deisotoping (right). Removing isotope peaks SWATH MS2 spectrum increases also spectra similarity score (since reference spectra generally contain fragments ion interest, isotopes). Similar DDA data, reconstructed MS2 spectra SWATH data used annotation MS1 chromatographic peaks.","code":"fenamiphos_mz <- 304.113077 fenamiphos_ms1_peak <- chromPeaks(swath_data, mz = fenamiphos_mz, ppm = 2) fenamiphos_ms1_peak ##            mz    mzmin    mzmax      rt   rtmin   rtmax     into     intb ## CP34 304.1124 304.1121 304.1126 423.945 419.445 428.444 10697.34 10688.34 ##          maxo  sn sample ## CP34 2401.849 618      1 keep <- chromPeakData(swath_data)$isolationWindowLowerMz < fenamiphos_mz &         chromPeakData(swath_data)$isolationWindowUpperMz > fenamiphos_mz keep <- keep &     chromPeaks(swath_data)[, \"rtmin\"] < fenamiphos_ms1_peak[, \"rt\"] &     chromPeaks(swath_data)[, \"rtmax\"] > fenamiphos_ms1_peak[, \"rt\"]  fenamiphos_ms2_peak <- chromPeaks(swath_data)[which(keep), ] rtr <- fenamiphos_ms1_peak[, c(\"rtmin\", \"rtmax\")] mzr <- fenamiphos_ms1_peak[, c(\"mzmin\", \"mzmax\")] fenamiphos_ms1_chr <- chromatogram(swath_data, rt = rtr, mz = mzr) ## Processing chromatographic peaks rtr <- fenamiphos_ms2_peak[, c(\"rtmin\", \"rtmax\")] mzr <- fenamiphos_ms2_peak[, c(\"mzmin\", \"mzmax\")] ## Get the isolationWindowTargetMz for spectra containing the m/z of the ## compound of interest swath_data |> filterIsolationWindow(mz = fenamiphos_mz) |> spectra() |> isolationWindowTargetMz() |> table() ##  ## 299.1  ##   423 fenamiphos_ms2_chr <- chromatogram(     swath_data, rt = rtr, mz = mzr, msLevel = 2L,     isolationWindowTargetMz = rep(299.1, nrow(rtr))) ## Processing chromatographic peaks plot(rtime(fenamiphos_ms1_chr[1, 1]),      intensity(fenamiphos_ms1_chr[1, 1]),      xlab = \"retention time [s]\", ylab = \"intensity\", pch = 16,      ylim = c(0, 5000), col = \"blue\", type = \"b\", lwd = 2) #' Add data from all MS2 peaks tmp <- lapply(fenamiphos_ms2_chr@.Data,               function(z) points(rtime(z), intensity(z),                                  col = \"#00000080\",                                  type = \"b\", pch = 16)) compareChromatograms(fenamiphos_ms2_chr[1, 1],                fenamiphos_ms1_chr[1, 1],                ALIGNFUNARGS = list(method = \"approx\")) ## [1] 0.9997871 swath_spectra <- reconstructChromPeakSpectra(swath_data, minCor = 0.9) swath_spectra ## MSn data (Spectra) with 62 spectra in a MsBackendMemory backend: ##        msLevel     rtime scanIndex ##      <integer> <numeric> <integer> ## CP01         2   239.458        NA ## CP02         2   240.358        NA ## CP03         2   329.577        NA ## CP04         2   329.771        NA ## CP05         2   346.164        NA ## ...        ...       ...       ... ## CP58         2   551.735        NA ## CP59         2   551.735        NA ## CP60         2   575.134        NA ## CP61         2   575.134        NA ## CP62         2   574.942        NA ##  ... 20 more variables/columns. ## Processing: ##  Merge 1 Spectra into one [Fri Jan 26 20:09:40 2024] lengths(swath_spectra) ##  [1]  0  0  1  1  1  0  0  0  0  0  0  0  3  0  3  4  0  3  0  1  0  9 14  1  0 ## [26]  0 15  4  1  1  2  4  6 15 12 11  2  4 13  0  0  0  0  1  2  0  1  0  0  0 ## [51]  3  0  2  1  7  7  0  0  0  0  0  2 spectraData(swath_spectra, c(\"peak_id\", \"ms2_peak_id\", \"ms2_peak_cor\")) ## DataFrame with 62 rows and 3 columns ##          peak_id ms2_peak_id      ms2_peak_cor ##      <character>      <list>            <list> ## CP01        CP01                               ## CP02        CP02                               ## CP03        CP03       CP063          0.950582 ## CP04        CP04       CP105           0.95157 ## CP05        CP05       CP153          0.924545 ## ...          ...         ...               ... ## CP58        CP58                               ## CP59        CP59                               ## CP60        CP60                               ## CP61        CP61                               ## CP62        CP62 CP334,CP329 0.918915,0.911944 fenamiphos_swath_spectrum <- swath_spectra[     swath_spectra$peak_id == rownames(fenamiphos_ms1_peak)] fenamiphos_swath_spectrum <- addProcessing(fenamiphos_swath_spectrum,                                            scale_fun) par(mfrow = c(1, 2)) plotSpectraMirror(fenamiphos_swath_spectrum, ex_spectrum,      ppm = 50, main = \"against DDA\") plotSpectraMirror(fenamiphos_swath_spectrum, fenamiphos[2],      ppm = 50, main = \"against Metlin\") pk_ids <- fenamiphos_swath_spectrum$ms2_peak_id[[1]] pk_ids ##  [1] \"CP199\" \"CP201\" \"CP211\" \"CP208\" \"CP200\" \"CP202\" \"CP217\" \"CP215\" \"CP205\" ## [10] \"CP212\" \"CP221\" \"CP223\" \"CP213\" \"CP207\" \"CP220\" rt_range <- chromPeaks(swath_data)[pk_ids, c(\"rtmin\", \"rtmax\")] mz_range <- chromPeaks(swath_data)[pk_ids, c(\"mzmin\", \"mzmax\")]  pmz <- precursorMz(fenamiphos_swath_spectrum)[1] ## Determine the isolation window target m/z tmz <- swath_data |> filterIsolationWindow(mz = pmz) |> spectra() |> isolationWindowTargetMz() |> unique()  ms2_eics <- chromatogram(     swath_data, rt = rt_range, mz = mz_range, msLevel = 2L,     isolationWindowTargetMz = rep(tmz, nrow(rt_range))) ## Processing chromatographic peaks plotChromatogramsOverlay(ms2_eics) prochloraz_mz <- 376.0381  prochloraz_ms1_peak <- chromPeaks(swath_data, msLevel = 1L,                                   mz = prochloraz_mz, ppm = 5) prochloraz_ms1_peak ##            mz   mzmin    mzmax      rt   rtmin   rtmax     into     intb ## CP22 376.0373 376.037 376.0374 405.046 401.446 409.546 3664.051 3655.951 ##          maxo  sn sample ## CP22 897.3923 278      1 prochloraz_swath_spectrum <- swath_spectra[     swath_spectra$peak_id == rownames(prochloraz_ms1_peak)] lengths(prochloraz_swath_spectrum) ## [1] 9 prochloraz_dda_peak <- chromPeaks(dda_data, msLevel = 1L,                                   mz = prochloraz_mz, ppm = 5) prochloraz_dda_peak ##             mz    mzmin    mzmax      rt   rtmin   rtmax     into    intb ## CP034 376.0385 376.0378 376.0391 405.295 401.166 410.145 5082.157 5072.77 ##           maxo  sn sample ## CP034 1350.633 310      1 prochloraz_dda_spectra <- dda_spectra[     dda_spectra$peak_id == rownames(prochloraz_dda_peak)] prochloraz_dda_spectra ## MSn data (Spectra) with 5 spectra in a MsBackendMzR backend: ##     msLevel     rtime scanIndex ##   <integer> <numeric> <integer> ## 1         2   401.438      3253 ## 2         2   402.198      3259 ## 3         2   404.677      3306 ## 4         2   405.127      3316 ## 5         2   405.877      3325 ##  ... 34 more variables/columns. ##  ## file(s): ## PestMix1_DDA.mzML ## Processing: ##  Filter: select retention time [230..610] on MS level(s) 1 2 [Fri Jan 26 20:09:22 2024] ##  Filter: select MS level(s) 2 [Fri Jan 26 20:09:31 2024] ##  Merge 1 Spectra into one [Fri Jan 26 20:09:31 2024] prochloraz_dda_spectrum <- combineSpectra(     prochloraz_dda_spectra, FUN = combinePeaks, ppm = 20,     peaks = \"intersect\", minProp = 0.8, intensityFun = median, mzFun = median,     f = prochloraz_dda_spectra$peak_id) ## Backend of the input object is read-only, will change that to an 'MsBackendMemory' ## Warning: 'combinePeaks' for lists of peak matrices is deprecated; please use ## 'combinePeaksData' instead. prochloraz <- Spectra(     system.file(\"mgf\", \"metlin-68898.mgf\", package = \"xcms\"),     source = MsBackendMgf()) ## Start data import from 1 files ... done prochloraz_swath_spectrum <- addProcessing(prochloraz_swath_spectrum, scale_fun) prochloraz_dda_spectrum <- addProcessing(prochloraz_dda_spectrum, scale_fun)  par(mfrow = c(1, 2)) plotSpectraMirror(prochloraz_swath_spectrum, prochloraz_dda_spectrum,                   ppm = 40, main = \"against DDA\") plotSpectraMirror(prochloraz_swath_spectrum, prochloraz[2],                   ppm = 40, main = \"against Metlin\") plotSpectra(prochloraz_swath_spectrum) library(MetaboCoreUtils) isotopologues(peaksData(prochloraz_swath_spectrum)[[1]]) ## [[1]] ## [1] 3 4 5 ##  ## [[2]] ## [1] 6 7 ## Function to keep only the first (monoisotopic) peak for potential ## isotopologue peak groups. rem_iso <- function(x, ...) {     idx <- isotopologues(x)     idx <- unlist(lapply(idx, function(z) z[-1]), use.names = FALSE)     if (length(idx))         x[-idx, , drop = FALSE]     else x } prochloraz_swath_spectrum2 <- addProcessing(prochloraz_swath_spectrum,                                             rem_iso) par(mfrow = c(1, 2)) plotSpectra(prochloraz_swath_spectrum) plotSpectra(prochloraz_swath_spectrum2) compareSpectra(prochloraz_swath_spectrum, prochloraz_dda_spectrum) ## [1] 0.4623719 compareSpectra(prochloraz_swath_spectrum2, prochloraz_dda_spectrum) ## [1] 0.5932303"},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"outlook","dir":"Articles","previous_headings":"","what":"Outlook","title":"LC-MS/MS data analysis with xcms","text":"Currently, spectra data representation, handling processing re-implemented part RforMassSpectrometry initiative aiming increasing performance methods simplifying use. Thus, parts workflow described changed (improved) future. Along developments, improved matching strategies larger data sets implemented well functionality compare Spectra directly reference MS2 spectra public annotation resources (e.g. Massbank HMDB). See example information. Regarding SWATH data analysis, future development involve improved selection correct MS2 chromatographic peaks considering also correlation intensity values across several samples.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms-lcms-ms.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"LC-MS/MS data analysis with xcms","text":"","code":"sessionInfo() ## R Under development (unstable) (2024-01-17 r85813) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] MetaboCoreUtils_1.11.2 MsBackendMgf_1.11.1    MsExperiment_1.5.4     ##  [4] pander_0.6.5           Spectra_1.13.3         xcms_4.1.6             ##  [7] MSnbase_2.29.3         ProtGenerics_1.35.2    S4Vectors_0.41.3       ## [10] mzR_2.37.0             Rcpp_1.0.12            Biobase_2.63.0         ## [13] BiocGenerics_0.49.1    BiocParallel_1.37.0    BiocStyle_2.31.0       ##  ## loaded via a namespace (and not attached): ##   [1] DBI_1.2.1                   bitops_1.0-7                ##   [3] rlang_1.1.3                 magrittr_2.0.3              ##   [5] clue_0.3-65                 MassSpecWavelet_1.69.0      ##   [7] matrixStats_1.2.0           compiler_4.4.0              ##   [9] systemfonts_1.0.5           vctrs_0.6.5                 ##  [11] stringr_1.5.1               pkgconfig_2.0.3             ##  [13] crayon_1.5.2                fastmap_1.1.1               ##  [15] XVector_0.43.1              utf8_1.2.4                  ##  [17] rmarkdown_2.25              preprocessCore_1.65.0       ##  [19] ragg_1.2.7                  purrr_1.0.2                 ##  [21] xfun_0.41                   MultiAssayExperiment_1.29.0 ##  [23] zlibbioc_1.49.0             cachem_1.0.8                ##  [25] GenomeInfoDb_1.39.5         jsonlite_1.8.8              ##  [27] progress_1.2.3              highr_0.10                  ##  [29] DelayedArray_0.29.0         prettyunits_1.2.0           ##  [31] parallel_4.4.0              cluster_2.1.6               ##  [33] R6_2.5.1                    RColorBrewer_1.1-3          ##  [35] bslib_0.6.1                 stringi_1.8.3               ##  [37] limma_3.59.1                GenomicRanges_1.55.2        ##  [39] jquerylib_0.1.4             bookdown_0.37               ##  [41] SummarizedExperiment_1.33.3 iterators_1.0.14            ##  [43] knitr_1.45                  IRanges_2.37.1              ##  [45] splines_4.4.0               Matrix_1.6-5                ##  [47] igraph_1.6.0                tidyselect_1.2.0            ##  [49] abind_1.4-5                 yaml_2.3.8                  ##  [51] doParallel_1.0.17           codetools_0.2-19            ##  [53] affy_1.81.0                 lattice_0.22-5              ##  [55] tibble_3.2.1                plyr_1.8.9                  ##  [57] evaluate_0.23               survival_3.5-7              ##  [59] desc_1.4.3                  pillar_1.9.0                ##  [61] affyio_1.73.0               BiocManager_1.30.22         ##  [63] MatrixGenerics_1.15.0       foreach_1.5.2               ##  [65] MALDIquant_1.22.2           ncdf4_1.22                  ##  [67] generics_0.1.3              RCurl_1.98-1.14             ##  [69] hms_1.1.3                   ggplot2_3.4.4               ##  [71] munsell_0.5.0               scales_1.3.0                ##  [73] glue_1.7.0                  MsFeatures_1.11.0           ##  [75] lazyeval_0.2.2              tools_4.4.0                 ##  [77] robustbase_0.99-1           mzID_1.41.0                 ##  [79] QFeatures_1.13.2            vsn_3.71.0                  ##  [81] RANN_2.6.1                  fs_1.6.3                    ##  [83] XML_3.99-0.16.1             grid_4.4.0                  ##  [85] impute_1.77.0               MsCoreUtils_1.15.3          ##  [87] colorspace_2.1-0            GenomeInfoDbData_1.2.11     ##  [89] PSMatch_1.7.1               cli_3.6.2                   ##  [91] textshaping_0.3.7           fansi_1.0.6                 ##  [93] S4Arrays_1.3.2              dplyr_1.1.4                 ##  [95] AnnotationFilter_1.27.0     DEoptimR_1.1-3              ##  [97] pcaMethods_1.95.0           gtable_0.3.4                ##  [99] sass_0.4.8                  digest_0.6.34               ## [101] SparseArray_1.3.3           multtest_2.59.0             ## [103] memoise_2.0.1               htmltools_0.5.7             ## [105] pkgdown_2.0.7.9000          lifecycle_1.0.4             ## [107] statmod_1.5.0               MASS_7.3-60.2"},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"LC-MS data pre-processing and analysis with xcms","text":"xcms package provides functionality perform pre-processing LC-MS, GC-MS LC-MS/MS data raw signals mzML, mzXML CDF files processed feature abundances. pre-processing includes chromatographic peak detection, sample alignment correspondence analysis. first version package already published 2006 [1] since updated modernized several rounds better integrate R-based packages analysis untargeted metabolomics data. includes version 3 xcms used MSnbase package MS data representation [2]. recent update (xcms version 4) enables addition pre-processing MS data represented modern MsExperiment Spectra packages provides even better integration RforMassSpectrometry R package ecosystem simplifying e.g. also compound annotation [3]. document describes data import, exploration pre-processing simple test LC-MS data set xcms package version >= 4. functions can applied older MSnbase-based workflows (xcms version 3). Additional documents tutorials covering also topics untargeted metabolomics analysis listed end document. also xcms tutorial available examples details.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"data-import","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Data import","title":"LC-MS data pre-processing and analysis with xcms","text":"xcms supports analysis LC-MS(/MS) data can imported Spectra package. data typically provided (AIA/ANDI) NetCDF, mzXML mzML format can, dedicated extensions Spectra package, also imported sources, e.g. also directly raw data files manufacturer’s formats. demonstration purpose analyze document small subset data [4] metabolic consequences knock-fatty acid amide hydrolase (FAAH) gene mice investigated. raw data files (NetCDF format) provided faahKO data package. data set consists samples spinal cords 6 knock-6 wild-type mice. file contains data centroid mode acquired positive ion polarity 200-600 m/z 2500-4500 seconds. speed-processing vignette restrict analysis 8 files. load required packages, locate raw CDF files within faahKO package build phenodata data.frame describing experimental setup. Generally, data frames contain relevant experimental variables sample descriptions (including also names raw data files) imported R using either read.table function (file csv tabulator delimited text file format) also using functions readxl R package Excel file format. next load data using readMsExperiment function MsExperiment package. MS spectra data experiment now available Spectra object within faahko. Note MsExperiment container addition spectra data also contain types data also references files. See vignette MsExperiment details. Also, loading data mzML, mzXML CDF files, default general spectra data loaded memory actual peaks data, .e. m/z intensity values retrieved --fly raw files needed (similar MSnbase -disk mode described [2]). guarantees low memory footprint hence allowing analyze also large experiments without need high performance computing environments. Note also different alternative backends (hence data representations) used Spectra object within faahko eventually even lower memory footprint, higher performance. See package vignette Spectra package SpectraTutorials tutorial details Spectra backends change .","code":"library(xcms) library(faahKO) library(RColorBrewer) library(pander) library(pheatmap) library(MsExperiment)  ## Get the full path to the CDF files cdfs <- dir(system.file(\"cdf\", package = \"faahKO\"), full.names = TRUE,             recursive = TRUE)[c(1, 2, 5, 6, 7, 8, 11, 12)] ## Create a phenodata data.frame pd <- data.frame(sample_name = sub(basename(cdfs), pattern = \".CDF\",                                    replacement = \"\", fixed = TRUE),                  sample_group = c(rep(\"KO\", 4), rep(\"WT\", 4)),                  stringsAsFactors = FALSE) faahko <- readMsExperiment(spectraFiles = cdfs, sampleData = pd) faahko ## Object of class MsExperiment  ##  Spectra: MS1 (10224)  ##  Experiment data: 8 sample(s) ##  Sample data links: ##   - spectra: 8 sample(s) to 10224 element(s)."},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"initial-data-inspection","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Initial data inspection","title":"LC-MS data pre-processing and analysis with xcms","text":"MsExperiment object simple flexible container MS experiments. raw MS data stored Spectra object can accessed spectra function. spectra organized sequentially (.e., file) fromFile function can used get spectrum information data files belongs. simply count number spectra per file. Information samples can retrieved sampleData function. row DataFrame represents one sample (input file). Using [ possible subset MsExperiment object sample. subset faahko 3rd sample (file) access spectra sample data. first evaluation data plot base peak chromatogram (BPC) file experiment. use chromatogram method set aggregationFun \"max\" return spectrum maximal intensity hence create BPC raw data. create total ion chromatogram set aggregationFun \"sum\". Base peak chromatogram. chromatogram method returned MChromatograms object organizes individual Chromatogram objects (fact contain chromatographic data) two-dimensional array: columns represent samples rows (optionally) m/z /retention time ranges. extract chromatogram first sample access retention time intensity values. BPC seems around 4200 seconds signal measured anymore. Thus, filter full data set retention time range 2550 4250 seconds using filterRt function. Note present subset spectra within MsExperiment. Subsequently re-create also BPC. next create boxplots representing distribution total ion currents per data file. plots can useful spot potentially problematic MS runs. extract information, use tic function Spectra object within faahko split values file using fromFile. Distribution total ion currents per file. addition, can also cluster samples based similarity base peak chromatograms. Samples thus grouped based similarity LC runs. need however bin data along retention time axis, since retention times generally differ samples. use bin function BPC bin intensities 2 second wide retention time bins. clustering performed using complete linkage hierarchical clustering pairwise correlations binned base peak chromatograms. Grouping samples based similarity base peak chromatogram. samples cluster pairwise manner, KO WT samples sample index similar BPC.","code":"spectra(faahko) ## MSn data (Spectra) with 10224 spectra in a MsBackendMzR backend: ##         msLevel     rtime scanIndex ##       <integer> <numeric> <integer> ## 1             1   2501.38         1 ## 2             1   2502.94         2 ## 3             1   2504.51         3 ## 4             1   2506.07         4 ## 5             1   2507.64         5 ## ...         ...       ...       ... ## 10220         1   4493.56      1274 ## 10221         1   4495.13      1275 ## 10222         1   4496.69      1276 ## 10223         1   4498.26      1277 ## 10224         1   4499.82      1278 ##  ... 33 more variables/columns. ##  ## file(s): ## ko15.CDF ## ko16.CDF ## ko21.CDF ##  ... 5 more files table(fromFile(faahko)) ##  ##    1    2    3    4    5    6    7    8  ## 1278 1278 1278 1278 1278 1278 1278 1278 sampleData(faahko) ## DataFrame with 8 rows and 3 columns ##   sample_name sample_group spectraOrigin ##   <character>  <character>   <character> ## 1        ko15           KO /__w/_temp... ## 2        ko16           KO /__w/_temp... ## 3        ko21           KO /__w/_temp... ## 4        ko22           KO /__w/_temp... ## 5        wt15           WT /__w/_temp... ## 6        wt16           WT /__w/_temp... ## 7        wt21           WT /__w/_temp... ## 8        wt22           WT /__w/_temp... faahko_3 <- faahko[3] spectra(faahko_3) ## MSn data (Spectra) with 1278 spectra in a MsBackendMzR backend: ##        msLevel     rtime scanIndex ##      <integer> <numeric> <integer> ## 1            1   2501.38         1 ## 2            1   2502.94         2 ## 3            1   2504.51         3 ## 4            1   2506.07         4 ## 5            1   2507.64         5 ## ...        ...       ...       ... ## 1274         1   4493.56      1274 ## 1275         1   4495.13      1275 ## 1276         1   4496.69      1276 ## 1277         1   4498.26      1277 ## 1278         1   4499.82      1278 ##  ... 33 more variables/columns. ##  ## file(s): ## ko21.CDF sampleData(faahko_3) ## DataFrame with 1 row and 3 columns ##   sample_name sample_group spectraOrigin ##   <character>  <character>   <character> ## 1        ko21           KO /__w/_temp... ## Get the base peak chromatograms. This reads data from the files. bpis <- chromatogram(faahko, aggregationFun = \"max\") ## Define colors for the two groups group_colors <- paste0(brewer.pal(3, \"Set1\")[1:2], \"60\") names(group_colors) <- c(\"KO\", \"WT\")  ## Plot all chromatograms. plot(bpis, col = group_colors[sampleData(faahko)$sample_group]) bpi_1 <- bpis[1, 1] rtime(bpi_1) |> head() ## [1] 2501.378 2502.943 2504.508 2506.073 2507.638 2509.203 intensity(bpi_1) |> head() ## [1] 43888 43960 43392 42632 42200 42288 faahko <- filterRt(faahko, rt = c(2550, 4250)) ## Filter spectra ## creating the BPC on the subsetted data bpis <- chromatogram(faahko, aggregationFun = \"max\") ## Get the total ion current by file tc <- spectra(faahko) |>     tic() |>     split(f = fromFile(faahko)) boxplot(tc, col = group_colors[sampleData(faahko)$sample_group],         ylab = \"intensity\", main = \"Total ion current\") ## Bin the BPC bpis_bin <- bin(bpis, binSize = 2)  ## Calculate correlation on the log2 transformed base peak intensities cormat <- cor(log2(do.call(cbind, lapply(bpis_bin, intensity)))) colnames(cormat) <- rownames(cormat) <- bpis_bin$sample_name  ## Define which phenodata columns should be highlighted in the plot ann <- data.frame(group = bpis_bin$sample_group) rownames(ann) <- bpis_bin$sample_name  ## Perform the cluster analysis pheatmap(cormat, annotation = ann,          annotation_color = list(group = group_colors))"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"chromatographic-peak-detection","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Chromatographic peak detection","title":"LC-MS data pre-processing and analysis with xcms","text":"Chromatographic peak detection aims identifying signal sample created ions originating compound species. Chromatographic peak detection can performed xcms findChromPeaks function parameter object defines configures algorithm used (see ?findChromPeaks list supported algorithms). running peak detection however strongly suggested first visually inspect extracted ion chromatogram e.g. internal standards compounds known present samples order evaluate adapt settings peak detection algorithm since default settings appropriate LC-MS setups. extract EIC one compound using chromatogram function specifying addition m/z retention time range expect signal compound. Extracted ion chromatogram one peak. Note Chromatogram objects extracted chromatogram method contain NA value certain scan (.e. spectrum specific retention time) signal measured respective m/z range. reflected lines drawn continuous lines plot . peak thus width 50 seconds. can use information define peakwidth parameter centWave peak detection method [5] use chromatographic peak detection data. peakwidth parameter allows define expected lower upper width (retention time dimension) chromatographic peaks. data set thus peakwidth = c(20, 80). second important parameter centWave ppm defines expected maximum deviation m/z values centroids included one chromatographic peak (note precision m/z values provided MS instrument manufacturer). ppm parameter extract full MS data (intensity, retention time m/z values) corresponding peak. end first filter raw object retention time, m/z finally plot object type = \"XIC\" produce plot . use pipe (|>) operator better illustrate corresponding workflow. Visualization raw MS data one peak. plot: upper panel: chromatogram plotting intensity values retention time, lower panel m/z retention time plot. individual data points colored according intensity. present data actually variation m/z values. Usually m/z values individual centroids (lower panel) plots scatter around real m/z value compound (intensity dependent manner). first step centWave algorithm defines regions interest (ROI) subsequently screened presence chromatographic peaks. ROIs defined based difference m/z values centroids consecutive scans (spectra). detail, centroids consecutive scans included ROI difference m/z mean m/z ROI smaller user defined ppm parameter. reasonable choice ppm thus maximal m/z difference data points neighboring scans/spectra part chromatographic peak internal standard known compound. suggested inspect ranges m/z values several compounds (either internal standards compounds known present sample) define ppm parameter centWave according . See also tutorial additional information examples choosing testing peak detection settings. Chromatographic peak detection can also performed extracted ion chromatograms, helps testing tuning peak detection settings. Note however peak detection EICs involve first step centWave described thus consider ppm parameter. Also, since less data available algorithms, background signal estimation performed differently different settings snthresh need used (generally lower snthresh used EICs since estimated background signal tends higher data subsets full data). perform peak detection findChromPeaks function EIC generated . submitted parameter object defines algorithm used allows define settings algorithm. use CentWaveParam parameter object use configure centWave algorithm default settings, except snthresh. can access identified chromatographic peaks chromPeaks function. Parallel chromPeaks matrix also chromPeakData data frame allows add arbitrary annotations chromatographic peak, e.g. MS level peak detected: plot EIC along identified chromatographic peaks using plot function result object . Additional parameters peakCol peakBg allow define foreground background (fill) color identified chromatographic peak provided result object (.e., need define one color row chromPeaks(xchr) - column \"column\" (\"sample\" present) peak matrix specifies sample peak identified). Signal example peak. Red blue colors represent KO wild type samples, respectively. Peak area identified chromatographic peaks highlighted sample group color. happy settings can use peak detection full data set (.e. MsExperiment object data whole experiment). Note set argument prefilter c(6, 5000) noise 5000 reduce run time vignette. setting consider ROIs least 6 centroids intensity larger 5000 centWave chromatographic peak detection. results findChromPeaks MsExperiment object returned XcmsExperiment object. object extends MsExperiment directly (hence providing access raw data) contains xcms pre-processing results. Note also additional rounds chromatographic peak detections performed results added existing peak detection results additional calls findChromPeaks result object using parameter add = TRUE. chromPeaks function can also used access results chromatographic peak detection. show first 6 identified chromatographic peaks. Columns chromPeaks matrix might differ depending used peak detection algorithm. Columns algorithms provide : \"mz\", \"mzmin\", \"mzmax\", \"rt\", \"rtmin\" \"rtmax\" define m/z retention time range chromatographic peak (.e. mass peaks within area used integrate peak signal) well m/z retention time peak apex. mandatory columns \"\" \"maxo\" absolute integrated peak signal maximum peak intensity. Finally, \"sample\" provides index sample peak identified. Additional annotations individual peak can extracted chromPeakData function. data frame also used add/store arbitrary annotations detected peak (don’t necessarily need numeric). Peak detection always work perfectly types peak shapes present data set leading peak detection artifacts, (partially completely) overlapping peaks artificially split peaks (common issues especially centWave). xcms provides refineChromPeaks function can called peak detection results order refine (clean) peak detection results either removing identified peaks passing certain criteria merging artificially split partially completely overlapping chromatographic peaks. Different algorithms available can configured respective parameter objects: CleanPeaksParam FilterIntensityParam allow remove peaks retention time range intensity specified threshold, respectively. MergeNeighboringPeaksParam possible merge chromatographic peaks hence remove many mentioned (centWave) peak detection artifacts. See also ?refineChromPeaks information help different methods. post-process peak detection results merging peaks overlap 4 second window per file signal lower 75% smaller peak’s maximal intensity. See ?MergeNeighboringPeaksParam help page detailed description settings approach. example merged peak given . Result peak refinement step. Left: data processing, right: refinement. splitted peak merged one. centWave thus detected originally 3 chromatographic peaks m/z slice (left panel plot ) peak refinement MergeNeighboringPeaksParam specified settings merged first two peaks single one (right panel figure ). close peaks, lower intensity , however merged (see ). Result peak refinement step. Left: data processing, right: refinement. peaks merged. also possible perform peak refinement extracted ion chromatograms. used test fine-tune settings parameter. illustrate perform peak refinement extracted ion chromatogram chr_1 reducing minProp parameter force joining two peaks.  proceeding next replace faahko object results peak refinement step. use data chromPeaks matrix calculate per-file summaries peak detection results, number peaks per file well distribution retention time widths. Summary statistics identified chromatographic peaks. Shown number identified peaks per sample widths/duration chromatographic peaks. default chromPeaks return identified chromatographic peaks result object also possible extract chromatographic peaks specified m/z /rt range: can also plot location identified chromatographic peaks m/z - retention time space one file using plotChromPeaks function. plot information third sample. Identified chromatographic peaks m/z retention time space one sample. general overview peak detection results can addition visualize number identified chromatographic peaks per file along retention time axis. Parameter binSize allows define width bins rt dimension peaks counted. number chromatographic peaks within bin shown color-coded resulting plot. Frequency identified chromatographic peaks along retention time axis. frequency color coded higher frequency represented yellow-white. line shows peak frequency one file. Note extracting ion chromatorams xcms result object addition chromatographic data also extract identified chromatographic peaks within region. can thus also used validate verify used peak detection settings identified e.g. peaks known compounds internal standards properly. extract ion chromatogram m/z - rt region access detected peaks region using chromPeaks function. can also plot extracted ion chromatogram also visualize identified chromatographic peaks region. Signal example peak. Red blue colors represent KO wild type samples, respectively. signal area identified chromatographic peaks filled color. Chromatographic peaks can visualized also ways: setting peakType = \"rectangle\" indicated rectangle instead default highlighting option (peakType = \"polygon\") used . third alternative also possible just indicate apex position identified chromatographic peak single point (peakType = \"point\"). plot data using peakType = \"rectangle\". Signal example peak. Red blue colors represent KO wild type samples, respectively. rectangles indicate identified chromatographic peaks per sample. Finally plot also distribution peak intensity per sample. allows investigate whether systematic differences peak signals samples present. Peak intensity distribution per sample. signal identified chromatographic peaks comparable across samples, exception samples 3, 4 , degree, also 7 show slightly higher average intensities, also lower number detected peaks (indicated smaller width boxes). Note addition described identification chromatographic peaks, also possible manually define add chromatographic peaks manualChromPeaks function (see ?manualChromPeaks help page information).","code":"## Define the rt and m/z range of the peak area rtr <- c(2700, 2900) mzr <- c(334.9, 335.1) ## extract the chromatogram chr_raw <- chromatogram(faahko, mz = mzr, rt = rtr) plot(chr_raw, col = group_colors[chr_raw$sample_group]) faahko |> filterRt(rt = rtr) |> filterMz(mz = mzr) |> plot(type = \"XIC\") xchr <- findChromPeaks(chr_raw, param = CentWaveParam(snthresh = 2)) chromPeaks(xchr) ##             rt    rtmin    rtmax        into        intb  maxo   sn row column ##  [1,] 2781.505 2761.160 2809.674  412134.255  355516.374 16856   13   1      1 ##  [2,] 2786.199 2764.290 2812.803 1496244.206 1391821.332 58736   20   1      2 ##  [3,] 2734.556 2714.211 2765.855   21579.367   18449.428   899    4   1      3 ##  [4,] 2797.154 2775.245 2815.933  159058.782  150289.314  6295   12   1      3 ##  [5,] 2784.635 2761.160 2808.109   54947.545   37923.532  2715    2   1      4 ##  [6,] 2859.752 2845.668 2878.532   13895.212   13874.868   905  904   1      4 ##  [7,] 2897.311 2891.051 2898.876    5457.155    5450.895   995  994   1      4 ##  [8,] 2819.064 2808.109 2834.713   19456.914   19438.134  1347 1576   1      4 ##  [9,] 2789.329 2762.725 2808.109  174473.311   91114.975  8325    3   1      5 ## [10,] 2786.199 2764.290 2812.803  932645.211  569236.246 35856    2   1      6 ## [11,] 2792.461 2768.987 2823.760  876585.527  511324.134 27200    2   1      7 ## [12,] 2789.329 2773.680 2806.544   89582.569   73871.386  5427    6   1      8 chromPeakData(xchr) ## DataFrame with 12 rows and 4 columns ##      ms_level is_filled       row    column ##     <integer> <logical> <integer> <integer> ## 1           1     FALSE         1         1 ## 2           1     FALSE         1         2 ## 3           1     FALSE         1         3 ## 4           1     FALSE         1         3 ## 5           1     FALSE         1         4 ## ...       ...       ...       ...       ... ## 8           1     FALSE         1         4 ## 9           1     FALSE         1         5 ## 10          1     FALSE         1         6 ## 11          1     FALSE         1         7 ## 12          1     FALSE         1         8 ## Define a color for each sample sample_colors <- group_colors[xchr$sample_group] ## Define the background color for each chromatographic peak bg <- sample_colors[chromPeaks(xchr)[, \"column\"]] ## Parameter `col` defines the color of each sample/line, `peakBg` of each ## chromatographic peak. plot(xchr, col = sample_colors, peakBg = bg) cwp <- CentWaveParam(peakwidth = c(20, 80), noise = 5000,                      prefilter = c(6, 5000)) faahko <- findChromPeaks(faahko, param = cwp) chromPeaks(faahko) |> head() ##           mz mzmin mzmax       rt    rtmin    rtmax     into     intb  maxo sn ## CP0001 594.0 594.0 594.0 2601.535 2581.191 2637.529 161042.2 146073.3  7850 11 ## CP0002 577.0 577.0 577.0 2604.665 2581.191 2626.574 136105.2 128067.9  6215 11 ## CP0003 307.0 307.0 307.0 2618.750 2592.145 2645.354 284782.4 264907.0 16872 20 ## CP0004 302.0 302.0 302.0 2617.185 2595.275 2640.659 687146.6 669778.1 30552 43 ## CP0005 370.1 370.1 370.1 2673.523 2643.789 2700.127 449284.6 417225.3 25672 17 ## CP0006 427.0 427.0 427.0 2675.088 2643.789 2684.478 283334.7 263943.2 11025 13 ##        sample ## CP0001      1 ## CP0002      1 ## CP0003      1 ## CP0004      1 ## CP0005      1 ## CP0006      1 chromPeakData(faahko) ## DataFrame with 2908 rows and 2 columns ##         ms_level is_filled ##        <integer> <logical> ## CP0001         1     FALSE ## CP0002         1     FALSE ## CP0003         1     FALSE ## CP0004         1     FALSE ## CP0005         1     FALSE ## ...          ...       ... ## CP2904         1     FALSE ## CP2905         1     FALSE ## CP2906         1     FALSE ## CP2907         1     FALSE ## CP2908         1     FALSE mpp <- MergeNeighboringPeaksParam(expandRt = 4) faahko_pp <- refineChromPeaks(faahko, mpp) mzr_1 <- 305.1 + c(-0.01, 0.01) chr_1 <- chromatogram(faahko[1], mz = mzr_1) ## Processing chromatographic peaks chr_2 <- chromatogram(faahko_pp[1], mz = mzr_1) ## Processing chromatographic peaks par(mfrow = c(1, 2)) plot(chr_1) plot(chr_2) mzr_1 <- 496.2 + c(-0.01, 0.01) chr_1 <- chromatogram(faahko[1], mz = mzr_1) ## Processing chromatographic peaks chr_2 <- chromatogram(faahko_pp[1], mz = mzr_1) ## Processing chromatographic peaks par(mfrow = c(1, 2)) plot(chr_1) plot(chr_2) res <- refineChromPeaks(chr_1, MergeNeighboringPeaksParam(minProp = 0.05)) chromPeaks(res) ##         mz  mzmin  mzmax       rt    rtmin    rtmax     into intb    maxo  sn ## CPM1 496.2 496.19 496.21 3384.012 3294.809 3412.181 45940118   NA 1128960 177 ##      sample row column ## CPM1      1   1      1 plot(res) faahko <- faahko_pp summary_fun <- function(z)     c(peak_count = nrow(z), rt = quantile(z[, \"rtmax\"] - z[, \"rtmin\"]))  T <- chromPeaks(faahko) |>     split.data.frame(f = chromPeaks(faahko)[, \"sample\"]) |>     lapply(FUN = summary_fun) |>     do.call(what = rbind) rownames(T) <- basename(fileNames(faahko)) pandoc.table(     T,     caption = paste0(\"Summary statistics on identified chromatographic\",                      \" peaks. Shown are number of identified peaks per\",                      \" sample and widths/duration of chromatographic \",                      \"peaks.\")) chromPeaks(faahko, mz = c(334.9, 335.1), rt = c(2700, 2900)) ##         mz mzmin mzmax       rt    rtmin    rtmax      into      intb  maxo sn ## CP0038 335   335   335 2781.505 2761.160 2809.674  412134.3  383167.4 16856 23 ## CP0494 335   335   335 2786.199 2764.290 2812.803 1496244.2 1461187.3 58736 72 ## CP1025 335   335   335 2797.154 2775.245 2815.933  159058.8  149229.6  6295 13 ## CP1964 335   335   335 2786.199 2764.290 2812.803  932645.2  915333.8 35856 66 ## CP2349 335   335   335 2792.461 2768.987 2823.760  876585.5  848569.1 27200 36 ##        sample ## CP0038      1 ## CP0494      2 ## CP1025      3 ## CP1964      6 ## CP2349      7 plotChromPeaks(faahko, file = 3) plotChromPeakImage(faahko, binSize = 10) chr_ex <- chromatogram(faahko, mz = mzr, rt = rtr) chromPeaks(chr_ex) ##         mz mzmin mzmax       rt    rtmin    rtmax      into      intb  maxo sn ## CP0038 335   335   335 2781.505 2761.160 2809.674  412134.3  383167.4 16856 23 ## CP0494 335   335   335 2786.199 2764.290 2812.803 1496244.2 1461187.3 58736 72 ## CP1025 335   335   335 2797.154 2775.245 2815.933  159058.8  149229.6  6295 13 ## CP1964 335   335   335 2786.199 2764.290 2812.803  932645.2  915333.8 35856 66 ## CP2349 335   335   335 2792.461 2768.987 2823.760  876585.5  848569.1 27200 36 ##        sample row column ## CP0038      1   1      1 ## CP0494      2   1      2 ## CP1025      3   1      3 ## CP1964      6   1      6 ## CP2349      7   1      7 sample_colors <- group_colors[chr_ex$sample_group] plot(chr_ex, col = group_colors[chr_raw$sample_group], lwd = 2,      peakBg = sample_colors[chromPeaks(chr_ex)[, \"sample\"]]) plot(chr_ex, col = sample_colors, peakType = \"rectangle\",      peakCol = sample_colors[chromPeaks(chr_ex)[, \"sample\"]],      peakBg = NA) ## Extract a list of per-sample peak intensities (in log2 scale) ints <- split(log2(chromPeaks(faahko)[, \"into\"]),               f = chromPeaks(faahko)[, \"sample\"]) boxplot(ints, varwidth = TRUE, col = sample_colors,         ylab = expression(log[2]~intensity), main = \"Peak intensities\") grid(nx = NA, ny = NULL)"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"alignment","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Alignment","title":"LC-MS data pre-processing and analysis with xcms","text":"time analytes elute chromatography can vary samples (even compounds). differences already visible BPC even extracted ion chromatogram plot previous section. alignment step, also referred retention time correction, aims adjust differences shifting signals along retention time axis aligning different samples within experiment. plethora alignment algorithms exist (see [6]), also implemented xcms. Alignment LC-MS data can performed xcms using adjustRtime method algorithm-specific parameter class (see ?adjustRtime overview available methods xcms). example use obiwarp method [7] align samples. use binSize = 0.6 creates warping functions m/z bins 0.6. Also advisable modify adapt settings experiment. Note adjustRtime, besides calculating adjusted retention times spectrum, adjusts also retention times identified chromatographic peaks xcms result object. Adjusted retention times individual spectra can extracted result object using either adjustedRtime function using rtime parameter adjusted = TRUE (default): evaluate impact alignment plot BPC adjusted data. addition plot also differences adjusted raw retention times per sample using plotAdjustedRtime function. disable automatic extraction identified chromatographic peaks chromatogram function (make much sense BPC) use chromPeaks = \"none\" . Obiwarp aligned data. Base peak chromatogram (top) alignment (middle) difference adjusted raw retention times along retention time axis (bottom). large differences adjusted raw retention times indicate poorly performing samples alignment. last evaluate also impact alignment test peak. Example extracted ion chromatogram (top) alignment (bottom). Note: xcms result objects (XcmsExperiment well XCMSnExp) contain raw adjusted retention times spectra subset operation many cases drop adjusted retention times. Thus might sometimes useful immediately replace raw retention times data using applyAdjustedRtime function.","code":"faahko <- adjustRtime(faahko, param = ObiwarpParam(binSize = 0.6)) ## Extract adjusted retention times adjustedRtime(faahko) |> head() ## [1] 2551.457 2553.089 2554.720 2556.352 2557.983 2559.615 ## Or simply use the rtime method rtime(faahko) |> head() ## [1] 2551.457 2553.089 2554.720 2556.352 2557.983 2559.615 ## Get raw (unadjusted) retention times rtime(faahko, adjusted = FALSE) |> head() ## [1] 2551.457 2553.022 2554.586 2556.151 2557.716 2559.281 ## Get the base peak chromatograms. bpis_adj <- chromatogram(faahko, aggregationFun = \"max\", chromPeaks = \"none\") par(mfrow = c(3, 1), mar = c(4.5, 4.2, 1, 0.5)) plot(bpis, col = sample_colors) grid() plot(bpis_adj, col = sample_colors) grid() ## Plot also the difference of adjusted to raw retention time. plotAdjustedRtime(faahko, col = sample_colors) grid() par(mfrow = c(2, 1)) ## Plot the raw data plot(chr_raw, col = sample_colors) grid() ## Extract the chromatogram from the adjusted object chr_adj <- chromatogram(faahko, rt = rtr, mz = mzr) plot(chr_adj, col = sample_colors, peakType = \"none\") grid()"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"subset-based-alignment","dir":"Articles","previous_headings":"Pre-processing of LC-MS data > Alignment","what":"Subset-based alignment","title":"LC-MS data pre-processing and analysis with xcms","text":"experiments might better perform alignment based subset available samples, e.g. pooled QC samples injected regular intervals experiment contains blanks. alignment methods xcms support subset-based alignment retention time shifts estimated specified subset samples followed alignment whole data set based aligned subset. subset samples alignment can specified parameter subset PeakGroupsParam ObiwarpParam object. Parameter subsetAdjust allows specify method left-samples adjusted. currently two options available: subsetAdjust = \"previous\": adjust retention times non-subset sample based alignment results previous subset sample (e.g. QC sample). samples e.g. order A1, B1, B2, A2, B3, B4 representing QC samples B study samples, using subset = c(1, 4) subsetAdjust = \"previous\" result samples aligned non-subset samples B1 B2 adjusted based alignment result subset samples A1 B3 B4 A2. subsetAdjust = \"average\": adjust retention times non-subset samples based interpolation alignment results previous subsequent subset sample. example , B1 adjusted based average adjusted retention times subset (QC) samples A1 A2. Since subset sample non-subset samples B3 B4 adjusted based alignment results A2 alone. Note weighted average used calculate adjusted retention time averages, uses inverse difference index non-subset sample subset samples weights. Thus, setup like A1, B1, B2, A2 adjusted retention times A1 get larger weight A2 adjustment non-subset sample B1 causing ’s adjusted retention times closer A1 A2. See examples. Important: cases require meaningful/correct ordering samples within object (.e., samples ordered injection index). examples aim illustrate effect alignment options. assume samples 1, 4 7 faahKO data set QC pool samples. thus want perform alignment based samples subsequently adjust retention times left-samples (2, 3, 5, 6 8) based interpolation results neighboring subset (QC) samples. initial peak grouping perform subset-based alignment peak groups method passing indices QC samples subset parameter PeakGroupsParam function set subsetAdjust = \"average\" adjust study samples based interpolation alignment results neighboring subset/QC samples. Note subset-alignment parameters minFraction relative subset, full experiment! first remove previous alignment results dropAdjustedRtime function allow fresh alignment using subset-based option outlined . addition removing adjusted retention times spectra, function also restore original retention times identified chromatographic peaks. alignment peak groups method initial peak grouping (correspondence) analysis required, algorithm estimates retention times shifts samples using retention times hook peaks (.e. chromatographic peaks present /samples). use default settings peak density method-based correspondence, strongly advised adapt parameters data set (details next section). definition sample groups (.e. assignment individual samples sample groups experiment) mandatory PeakDensityParam. sample groups experiment, sampleGroups set single value file (e.g. rep(1, length(fileNames(faahko))). plot results alignment highlighting subset samples green. nicely shows interpolation subsetAdjust = \"average\" works: retention times sample 2 adjusted based subset sample 1 4, giving however weight closer subset sample 1 results adjusted retention times 2 similar sample 1. Sample 3 hand gets adjusted giving weight second subset sample (4). Subset-alignment results option average. Difference adjusted raw retention times along retention time axis. Samples alignment models estimated shown green, study samples grey. Option subsetAdjust = \"previous\" adjust retention times non-subset sample based single subset sample (previous), results cases adjusted retention times non-subset sample highly similar subset sample used adjustment.","code":"faahko <- dropAdjustedRtime(faahko)  ## Define the experimental layout sampleData(faahko)$sample_type <- \"study\" sampleData(faahko)$sample_type[c(1, 4, 7)] <- \"QC\" ## Initial peak grouping. Use sample_type as grouping variable pdp_subs <- PeakDensityParam(sampleGroups = sampleData(faahko)$sample_type,                              minFraction = 0.9) faahko <- groupChromPeaks(faahko, param = pdp_subs)  ## Define subset-alignment options and perform the alignment pgp_subs <- PeakGroupsParam(     minFraction = 0.85,     subset = which(sampleData(faahko)$sample_type == \"QC\"),     subsetAdjust = \"average\", span = 0.4) faahko <- adjustRtime(faahko, param = pgp_subs) clrs <- rep(\"#00000040\", 8) clrs[sampleData(faahko)$sample_type == \"QC\"] <- c(\"#00ce0080\") par(mfrow = c(2, 1), mar = c(4, 4.5, 1, 0.5)) plot(chromatogram(faahko, aggregationFun = \"max\", chromPeaks = \"none\"),      col = clrs) grid() plotAdjustedRtime(faahko, col = clrs, peakGroupsPch = 1,                   peakGroupsCol = \"#00ce0040\") grid()"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"correspondence","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Correspondence","title":"LC-MS data pre-processing and analysis with xcms","text":"Correspondence usually final step LC-MS data pre-processing data, presumably representing signal originating ions, matched across samples. result, chromatographic peaks different samples similar m/z retention times get grouped LC-MS features. function perform correspondence xcms called groupChromPeaks supports different algorithms can selected configured specific parameter object (see ?groupChromPeaks overview). example use peak density method [1] , within small slices along m/z dimension, combines chromatographic peaks depending density peaks along retention time axis. illustrate , simulate peak grouping m/z slice containing multiple chromatoghaphic peaks within sample using plotChromPeakDensity function PeakDensityParam object parameter minFraction = 0.4 (features defined least 40% samples chromatographic peak present) - parameter sampleGroups used define sample group sample belongs. Example peak density correspondence. Upper panel: chromatogram mz slice multiple chromatographic peaks. lower panel: identified chromatographic peaks retention time (x-axis) index within samples experiments (y-axis) different values bw parameter. black line represents peak density estimate. Grouping peaks (based provided settings) indicated grey rectangles. upper panel plot shows extracted ion chromatogram sample detected peaks highlighted. retention times individual chromatographic peaks sample (y-axis index sample data set) shown lower panel solid black line representing density estimate distribution detected peaks along retention time. Parameter bw defines smoothness estimation. grey rectangles indicate chromatographic peaks grouped feature (grey rectangle thus representing one feature). type visualization thus ideal test, validate optimize correspondence settings manually defined m/z slices applying full data set. tested m/z slice settings seemed OK thus applying full data set . Especially parameter bw data set dependent (specifically LC-dependent) adapted data set. Another important parameter binSize defines size m/z slices (bins) within peaks grouped. parameter thus defines required similarity m/z values chromatographic peaks assumed represent signal (type ion ) compound hence evaluated grouping. default, constant m/z bin size used, changing parameter ppm value larger 0, m/z-relative bin sizes used instead (.e., bin size increase m/z value hence better representing measurement error/precision MS instruments). bin sizes (subsequently m/z width defined features) reach maximal value binSize plus ppm parts-per-million largest m/z value chromatographic peak data set. See also xcms tutorial examples details. alternative perform correspondence using m/z relative bin sizes. results mostly similar, except higher m/z range (larger m/z bins used). plot m/z range features median m/z. present data set (acquired triple quad instrument) clear difference can seen two approaches hence proceed analysis fixed bin size setting. stronger relationship expected example data measured TOF instruments. Relationship feature’s m/z m/z width (max - min m/z) feature. Red points represent results fixed m/z bin size, blue m/z-relative bin size. Results correspondence analysis can accessed featureDefinitions featureValues function. former returns data frame general information defined features, row one feature columns providing information median m/z retention time well indices chromatographic peaks assigned feature column \"peakidx\". show information first 6 features. featureValues function returns matrix rows features columns samples. content matrix can defined using value argument can column name chromPeaks matrix. default value = \"\" matrix integrated signal peaks corresponding feature sample returned. generally used intensity matrix downstream analysis. extract intensities first 6 features. can see several missing values feature matrix. Missing values reported one sample chromatographic peak detected m/z - rt region feature. however necessarily mean signal specific ion sample. chromatographic peak detection algorithm also just failed identify peak region, e.g. signal noisy low. Thus advisable perform, correspondence, also gap-filling (see next section). performance peak detection, alignment correspondence always evaluated inspecting extracted ion chromatograms e.g. known compounds, internal standards identified features general. featureChromatograms function allows extract chromatograms feature present featureDefinitions. returned MChromatograms object contains ion chromatogram feature (row containing data one feature) sample (column representing containing data one sample). Parameter features allows define specific features EIC returned. can specified index ID (.e. row name featureDefinitions data frame. features defined, EICs returned features data set, can take also considerable amount time. extract chromatograms first 4 features. plot extracted ion chromatograms. use group color identified peak fill area. Extracted ion chromatograms features 1 4. access EICs second feature can simply subset feature_chroms object.","code":"## Define the mz slice. mzr <- c(305.05, 305.15)  ## Extract and plot the chromatograms chr_mzr <- chromatogram(faahko, mz = mzr) ## Define the parameters for the peak density method pdp <- PeakDensityParam(sampleGroups = sampleData(faahko)$sample_group,                         minFraction = 0.4, bw = 30) plotChromPeakDensity(chr_mzr, col = sample_colors, param = pdp,                      peakBg = sample_colors[chromPeaks(chr_mzr)[, \"sample\"]],                      peakCol = sample_colors[chromPeaks(chr_mzr)[, \"sample\"]],                      peakPch = 16) ## Perform the correspondence using fixed m/z bin sizes. pdp <- PeakDensityParam(sampleGroups = sampleData(faahko)$sample_group,                         minFraction = 0.4, bw = 30) faahko <- groupChromPeaks(faahko, param = pdp) ## Drop feature definitions and re-perform the correspondence ## using m/z-relative bin sizes. faahko_ppm <- groupChromPeaks(     dropFeatureDefinitions(faahko),     PeakDensityParam(sampleGroups = sampleData(faahko)$sample_group,                      minFraction = 0.4, bw = 30, ppm = 10)) ## Calculate m/z width of features mzw <- featureDefinitions(faahko)$mzmax - featureDefinitions(faahko)$mzmin mzw_ppm <- featureDefinitions(faahko_ppm)$mzmax -                                         featureDefinitions(faahko_ppm)$mzmin plot(featureDefinitions(faahko_ppm)$mzmed, mzw_ppm,      xlab = \"m/z\", ylab = \"m/z width\", pch = 21,      col = \"#0000ff20\", bg = \"#0000ff10\") points(featureDefinitions(faahko)$mzmed, mzw, pch = 21,      col = \"#ff000020\", bg = \"#ff000010\") featureDefinitions(faahko) |> head() ##       mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks KO WT      peakidx ## FT001 200.1 200.1 200.1 2902.634 2882.603 2922.664      2  2  0    458, 1161 ## FT002 205.0 205.0 205.0 2789.901 2782.955 2796.531      8  4  4 44, 443,.... ## FT003 206.0 206.0 206.0 2789.405 2781.389 2794.219      7  3  4 29, 430,.... ## FT004 207.1 207.1 207.1 2718.560 2714.047 2727.347      7  4  3 16, 420,.... ## FT005 233.0 233.0 233.1 3023.579 3015.145 3043.959      7  3  4 69, 959,.... ## FT006 241.1 241.1 241.2 3683.299 3661.586 3695.886      8  3  4 276, 284.... ##       ms_level ## FT001        1 ## FT002        1 ## FT003        1 ## FT004        1 ## FT005        1 ## FT006        1 featureValues(faahko, value = \"into\") |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001        NA  506848.9        NA  169955.6        NA        NA        NA ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7        NA  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9        NA ## FT005  286221.4        NA  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5        NA  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001        NA ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 feature_chroms <- featureChromatograms(faahko, features = 1:4)  feature_chroms ## XChromatograms with 4 rows and 8 columns ##             ko15.CDF        ko16.CDF        ko21.CDF        ko22.CDF ##      <XChromatogram> <XChromatogram> <XChromatogram> <XChromatogram> ## [1,]        peaks: 0        peaks: 1        peaks: 0        peaks: 1 ## [2,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ## [3,]        peaks: 1        peaks: 1        peaks: 0        peaks: 1 ## [4,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ##             wt15.CDF        wt16.CDF        wt21.CDF        wt22.CDF ##      <XChromatogram> <XChromatogram> <XChromatogram> <XChromatogram> ## [1,]        peaks: 0        peaks: 0        peaks: 0        peaks: 0 ## [2,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ## [3,]        peaks: 1        peaks: 1        peaks: 1        peaks: 1 ## [4,]        peaks: 1        peaks: 1        peaks: 0        peaks: 1 ## phenoData with 4 variables ## featureData with 4 variables ## - - - xcms preprocessing - - - ## Chromatographic peak detection: ##  method: centWave  ## Correspondence: ##  method: chromatographic peak density  ##  4 feature(s) identified. plot(feature_chroms, col = sample_colors,      peakBg = sample_colors[chromPeaks(feature_chroms)[, \"sample\"]]) eic_2 <- feature_chroms[2, ] chromPeaks(eic_2) ##         mz mzmin mzmax       rt    rtmin    rtmax    into    intb  maxo sn ## CP0048 205   205   205 2791.873 2771.300 2815.623 1924712 1850331 84280 64 ## CP0495 205   205   205 2795.796 2773.702 2821.043 1757151 1711473 68384 69 ## CP1033 205   205   205 2796.531 2774.506 2821.697 1383417 1334570 47384 54 ## CP1255 205   205   205 2789.405 2769.019 2812.924 1180288 1126958 48336 32 ## CP1535 205   205   205 2782.955 2762.596 2806.444 2129885 2054677 93312 44 ## CP1967 205   205   205 2787.464 2767.135 2812.486 1634342 1566379 67984 53 ## CP2350 205   205   205 2790.396 2763.847 2821.630 1623589 1531573 49208 28 ## CP2601 205   205   205 2787.273 2766.970 2812.261 1354005 1299188 55712 35 ##        sample row column ## CP0048      1   1      1 ## CP0495      2   1      2 ## CP1033      3   1      3 ## CP1255      4   1      4 ## CP1535      5   1      5 ## CP1967      6   1      6 ## CP2350      7   1      7 ## CP2601      8   1      8"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"gap-filling","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Gap filling","title":"LC-MS data pre-processing and analysis with xcms","text":"Missing values LC-MS data many cases result chromatographic peak detection algorithm failing identify peaks (noisy low intensity signal). aim gap filling step reduce number missing values integrating signals original data files samples chromatographic peak found m/z - rt region signal ion expected. Gap filling can performed xcms fillChromPeaks function parameter object selecting configuring gap filling algorithm. method choice ChromPeakAreaParam integrates signal (samples chromatographic peak found feature) m/z - rt region defined based m/z retention time ranges detected chromatographic peaks specific feature. lower m/z limit area defined lower quartile (25% quantile) \"mzmin\" values peaks feature, upper m/z value upper quartile (75% quantile) \"mzmax\" values, lower rt value lower quartile (25% quantile) \"rtmin\" upper rt value upper quartile (75% quantile) \"rtmax\" values. perform gap filling test data extract feature values first 6 features gap filling. NA reported signal measured specific sample.","code":"faahko <- fillChromPeaks(faahko, param = ChromPeakAreaParam())  featureValues(faahko, value = \"into\") |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001  135162.4  506848.9  111657.3  169955.6  209929.4  141607.9  226853.7 ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7  164380.7  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  226234.4 ## FT005  286221.4  285857.6  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1160580.5 1102832.6  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001  138341.2 ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"final-result","dir":"Articles","previous_headings":"Pre-processing of LC-MS data","what":"Final result","title":"LC-MS data pre-processing and analysis with xcms","text":"can continue using xcms result set analysis (e.g. also feature grouping MsFeatures package; see LC-MS feature grouping vignette details) also extract results SummarizedExperiment object. standard data container Bioconductor defined SummarizedExperiment package integration Bioconductor packages might thus easier using type object. use quantify function extract xcms pre-processing results SummarizedExperiment object. Internally, featureValues function used generate feature value matrix. can pass parameters function quantify call. use value = \"\" method = \"sum\" report integrated peak signal intensity sum values samples one chromatographic peak assigned feature (option important run refineChromPeaks like described merge overlapping peaks sample). information featureDefinitions now stored rowData object. rowData provides annotations information row SummarizedExperiment (case features). Annotations columns (case samples) stored colData. data frame row contains annotations one sample (hence one column feature values matrix). Finally, feature matrix stored assay within object. Note SummarizedExperiment can multiple assays numeric matrices number rows columns matching number features samples, respectively. list names available assays. can access actual data using assay function, optionally also providing name assay want access. show first 6 lines matrix. Since SummarizedExperiment supports multiple assays, addition add also feature value matrix without filled-values (.e. feature intensities added gap filling step). now two assays result object. can extract feature values without gap-filling: Finally, history full processing xcms available metadata SummarizedExperiment. information can also extracted xcms result object using processHistory function. extract information first processing step. processing steps contain also individual parameter objects used analysis, hence allowing exactly reproduce analysis. last perform also principal component analysis evaluate grouping samples experiment. Note perform data normalization hence grouping might () also influenced technical biases. PCA faahKO data set, un-normalized intensities. can see expected separation KO WT samples PC2. PC1 samples separate based ID, samples ID <= 18 samples ID > 18. separation might caused technical bias (e.g. measurements performed different days/weeks) due biological properties mice analyzed (sex, age, litter mates etc).","code":"library(SummarizedExperiment) ## Loading required package: MatrixGenerics ## Loading required package: matrixStats ##  ## Attaching package: 'matrixStats' ## The following objects are masked from 'package:Biobase': ##  ##     anyMissing, rowMedians ##  ## Attaching package: 'MatrixGenerics' ## The following objects are masked from 'package:matrixStats': ##  ##     colAlls, colAnyNAs, colAnys, colAvgsPerRowSet, colCollapse, ##     colCounts, colCummaxs, colCummins, colCumprods, colCumsums, ##     colDiffs, colIQRDiffs, colIQRs, colLogSumExps, colMadDiffs, ##     colMads, colMaxs, colMeans2, colMedians, colMins, colOrderStats, ##     colProds, colQuantiles, colRanges, colRanks, colSdDiffs, colSds, ##     colSums2, colTabulates, colVarDiffs, colVars, colWeightedMads, ##     colWeightedMeans, colWeightedMedians, colWeightedSds, ##     colWeightedVars, rowAlls, rowAnyNAs, rowAnys, rowAvgsPerColSet, ##     rowCollapse, rowCounts, rowCummaxs, rowCummins, rowCumprods, ##     rowCumsums, rowDiffs, rowIQRDiffs, rowIQRs, rowLogSumExps, ##     rowMadDiffs, rowMads, rowMaxs, rowMeans2, rowMedians, rowMins, ##     rowOrderStats, rowProds, rowQuantiles, rowRanges, rowRanks, ##     rowSdDiffs, rowSds, rowSums2, rowTabulates, rowVarDiffs, rowVars, ##     rowWeightedMads, rowWeightedMeans, rowWeightedMedians, ##     rowWeightedSds, rowWeightedVars ## The following object is masked from 'package:Biobase': ##  ##     rowMedians ## Loading required package: GenomicRanges ## Loading required package: IRanges ##  ## Attaching package: 'IRanges' ## The following object is masked from 'package:xcms': ##  ##     distance ## Loading required package: GenomeInfoDb res <- quantify(faahko, value = \"into\", method = \"sum\") res ## class: SummarizedExperiment  ## dim: 351 8  ## metadata(6): '' '' ... '' '' ## assays(1): raw ## rownames(351): FT001 FT002 ... FT350 FT351 ## rowData names(10): mzmed mzmin ... WT ms_level ## colnames(8): ko15.CDF ko16.CDF ... wt21.CDF wt22.CDF ## colData names(4): sample_name sample_group spectraOrigin sample_type rowData(res) ## DataFrame with 351 rows and 10 columns ##           mzmed     mzmin     mzmax     rtmed     rtmin     rtmax    npeaks ##       <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> ## FT001     200.1     200.1     200.1   2902.63   2882.60   2922.66         2 ## FT002     205.0     205.0     205.0   2789.90   2782.95   2796.53         8 ## FT003     206.0     206.0     206.0   2789.40   2781.39   2794.22         7 ## FT004     207.1     207.1     207.1   2718.56   2714.05   2727.35         7 ## FT005     233.0     233.0     233.1   3023.58   3015.14   3043.96         7 ## ...         ...       ...       ...       ...       ...       ...       ... ## FT347    595.25     595.2     595.3   3010.60   2992.76   3014.37         6 ## FT348    596.20     596.2     596.2   2997.78   2992.76   3002.79         2 ## FT349    596.30     596.3     596.3   3819.83   3811.52   3836.39         4 ## FT350    597.40     597.4     597.4   3820.88   3817.76   3826.13         3 ## FT351    599.30     599.3     599.3   4071.41   4044.94   4125.32         3 ##              KO        WT  ms_level ##       <numeric> <numeric> <integer> ## FT001         2         0         1 ## FT002         4         4         1 ## FT003         3         4         1 ## FT004         4         3         1 ## FT005         3         4         1 ## ...         ...       ...       ... ## FT347         2         3         1 ## FT348         0         2         1 ## FT349         2         2         1 ## FT350         1         2         1 ## FT351         1         2         1 colData(res) ## DataFrame with 8 rows and 4 columns ##          sample_name sample_group spectraOrigin sample_type ##          <character>  <character>   <character> <character> ## ko15.CDF        ko15           KO /__w/_temp...          QC ## ko16.CDF        ko16           KO /__w/_temp...       study ## ko21.CDF        ko21           KO /__w/_temp...       study ## ko22.CDF        ko22           KO /__w/_temp...          QC ## wt15.CDF        wt15           WT /__w/_temp...       study ## wt16.CDF        wt16           WT /__w/_temp...       study ## wt21.CDF        wt21           WT /__w/_temp...          QC ## wt22.CDF        wt22           WT /__w/_temp...       study assayNames(res) ## [1] \"raw\" assay(res) |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001  135162.4  506848.9  111657.3  169955.6  209929.4  141607.9  226853.7 ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7  164380.7  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9  226234.4 ## FT005  286221.4  285857.6  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1923307.8 1102832.6  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001  138341.2 ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 assays(res)$raw_nofill <- featureValues(faahko, filled = FALSE, method = \"sum\") assayNames(res) ## [1] \"raw\"        \"raw_nofill\" assay(res, \"raw_nofill\") |> head() ##        ko15.CDF  ko16.CDF  ko21.CDF  ko22.CDF  wt15.CDF  wt16.CDF  wt21.CDF ## FT001        NA  506848.9        NA  169955.6        NA        NA        NA ## FT002 1924712.0 1757151.0 1383416.7 1180288.2 2129885.1 1634342.0 1623589.2 ## FT003  213659.3  289500.7        NA  178285.7  253825.6  241844.4  240606.0 ## FT004  349011.5  451863.7  343897.8  208002.8  364609.8  360908.9        NA ## FT005  286221.4        NA  164009.0  149097.6  255697.7  311296.8  366441.5 ## FT006 1923307.8        NA  380970.3  588986.4 1286883.0 1739516.6  639755.3 ##        wt22.CDF ## FT001        NA ## FT002 1354004.9 ## FT003  185399.5 ## FT004  221937.5 ## FT005  271128.0 ## FT006  508546.4 metadata(res) ## [[1]] ## Object of class \"XProcessHistory\" ##  type: Peak detection  ##  date: Fri Jan 26 20:10:29 2024  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: CentWaveParam  ##  MS level(s) 1  ##  ## [[2]] ## Object of class \"XProcessHistory\" ##  type: Peak refinement  ##  date: Fri Jan 26 20:10:32 2024  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: MergeNeighboringPeaksParam  ##  MS level(s) 1  ##  ## [[3]] ## Object of class \"XProcessHistory\" ##  type: Peak grouping  ##  date: Fri Jan 26 20:10:46 2024  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: PeakDensityParam  ##  MS level(s) 1  ##  ## [[4]] ## Object of class \"XProcessHistory\" ##  type: Retention time correction  ##  date: Fri Jan 26 20:10:46 2024  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: PeakGroupsParam  ##  MS level(s) 1  ##  ## [[5]] ## Object of class \"XProcessHistory\" ##  type: Peak grouping  ##  date: Fri Jan 26 20:10:51 2024  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: PeakDensityParam  ##  MS level(s) 1  ##  ## [[6]] ## Object of class \"XProcessHistory\" ##  type: Missing peak filling  ##  date: Fri Jan 26 20:10:58 2024  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: ChromPeakAreaParam  ##  MS level(s) 1 processHistory(faahko)[[1]] ## Object of class \"XProcessHistory\" ##  type: Peak detection  ##  date: Fri Jan 26 20:10:29 2024  ##  info:   ##  fileIndex: 1,2,3,4,5,6,7,8  ##  Parameter class: CentWaveParam  ##  MS level(s) 1 processHistory(faahko)[[1]] |> processParam() ## Object of class:  CentWaveParam  ##  Parameters: ##  - ppm: [1] 25 ##  - peakwidth: [1] 20 80 ##  - snthresh: [1] 10 ##  - prefilter: [1]    6 5000 ##  - mzCenterFun: [1] \"wMean\" ##  - integrate: [1] 1 ##  - mzdiff: [1] -0.001 ##  - fitgauss: [1] FALSE ##  - noise: [1] 5000 ##  - verboseColumns: [1] FALSE ##  - roiList: list() ##  - firstBaselineCheck: [1] TRUE ##  - roiScales: numeric(0) ##  - extendLengthMSW: [1] FALSE ##  - verboseBetaColumns: [1] FALSE ## Extract the features and log2 transform them ft_ints <- log2(assay(res, \"raw\"))  ## Perform the PCA omitting all features with an NA in any of the ## samples. Also, the intensities are mean centered. pc <- prcomp(t(na.omit(ft_ints)), center = TRUE)  ## Plot the PCA pcSummary <- summary(pc) plot(pc$x[, 1], pc$x[,2], pch = 21, main = \"\",      xlab = paste0(\"PC1: \", format(pcSummary$importance[2, 1] * 100,                                    digits = 3), \" % variance\"),      ylab = paste0(\"PC2: \", format(pcSummary$importance[2, 2] * 100,                                    digits = 3), \" % variance\"),      col = \"darkgrey\", bg = sample_colors, cex = 2) grid() text(pc$x[, 1], pc$x[,2], labels = res$sample_name, col = \"darkgrey\",      pos = 3, cex = 2)"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"further-data-processing-and-analysis","dir":"Articles","previous_headings":"","what":"Further data processing and analysis","title":"LC-MS data pre-processing and analysis with xcms","text":"Normalizing features’ signal intensities required, present (yet) supported xcms (methods might added near future). advised use SummarizedExperiment returned quantify method data processing, type object stores feature definitions, sample annotations well feature abundances object. identification e.g. features significant different intensities/abundances suggested use functionality provided R packages, Bioconductor’s excellent limma package.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"subsetting-and-filtering","dir":"Articles","previous_headings":"Additional details and notes","what":"Subsetting and filtering","title":"LC-MS data pre-processing and analysis with xcms","text":"xcms result objects can subset/filtered sample using [ method one filter* functions (although XcmsExperiment supports present selected filter functions). cases filtering can remove pre-processing results, filter functions support parameters keepFeatures keepAdjustedRtime can set TRUE avoid removal.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"parallel-processing","dir":"Articles","previous_headings":"Additional details and notes","what":"Parallel processing","title":"LC-MS data pre-processing and analysis with xcms","text":"functions xcms support parallel processing, enabled default performed, operations, per-file basis. Parallel processing handled configured BiocParallel Bioconductor package can globally defined R session. Note , data objects designed low memory footprint loading peak data memory needed, parallel processing increase demand: general, full data many files parallel processes loaded memory. needs also considered, number parallel processes defined. Unix-based systems (Linux, macOS) support multicore-based parallel processing. configure globally register parameter class. Note also bpstart used initialize parallel processes. Windows supports socket-based parallel processing:","code":"register(bpstart(MulticoreParam(2))) register(bpstart(SnowParam(2)))"},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"main-differences-to-the-msnbase-based-xcms-version-3","dir":"Articles","previous_headings":"Additional details and notes","what":"Main differences to the MSnbase-based xcms version 3","title":"LC-MS data pre-processing and analysis with xcms","text":"Spectra used main container (raw) MS data. Thus changing backends supported stage analysis (e.g. load data memory load data remote databases).","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"additional-documentation-resources","dir":"Articles","previous_headings":"","what":"Additional documentation resources","title":"LC-MS data pre-processing and analysis with xcms","text":"documentations listed still based xcms version 3 subsequently updated. Metabolomics pre-processing xcms: detailed description pre-processing LC-MS data xcms. MetaboAnnotationTutorials: examples annotation metabolomics data [3]. [2]: describes concept -disk data modes used also xcms. SpectraTutorials: tutorials describing Spectra package functionality.","code":""},{"path":"https://sneumann.github.io/xcms/articles/xcms.html","id":"session-information","dir":"Articles","previous_headings":"","what":"Session information","title":"LC-MS data pre-processing and analysis with xcms","text":"R packages used document listed .","code":"sessionInfo() ## R Under development (unstable) (2024-01-17 r85813) ## Platform: x86_64-pc-linux-gnu ## Running under: Ubuntu 22.04.3 LTS ##  ## Matrix products: default ## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  ## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 ##  ## locale: ##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               ##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     ##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    ##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  ##  [9] LC_ADDRESS=C               LC_TELEPHONE=C             ## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        ##  ## time zone: UTC ## tzcode source: system (glibc) ##  ## attached base packages: ## [1] stats4    stats     graphics  grDevices utils     datasets  methods   ## [8] base      ##  ## other attached packages: ##  [1] SummarizedExperiment_1.33.3 GenomicRanges_1.55.2        ##  [3] GenomeInfoDb_1.39.5         IRanges_2.37.1              ##  [5] MatrixGenerics_1.15.0       matrixStats_1.2.0           ##  [7] MsExperiment_1.5.4          pheatmap_1.0.12             ##  [9] RColorBrewer_1.1-3          pander_0.6.5                ## [11] faahKO_1.43.0               xcms_4.1.6                  ## [13] MSnbase_2.29.3              ProtGenerics_1.35.2         ## [15] S4Vectors_0.41.3            mzR_2.37.0                  ## [17] Rcpp_1.0.12                 Biobase_2.63.0              ## [19] BiocGenerics_0.49.1         BiocParallel_1.37.0         ## [21] BiocStyle_2.31.0            ##  ## loaded via a namespace (and not attached): ##   [1] DBI_1.2.1                   bitops_1.0-7                ##   [3] rlang_1.1.3                 magrittr_2.0.3              ##   [5] clue_0.3-65                 MassSpecWavelet_1.69.0      ##   [7] compiler_4.4.0              systemfonts_1.0.5           ##   [9] vctrs_0.6.5                 stringr_1.5.1               ##  [11] MetaboCoreUtils_1.11.2      pkgconfig_2.0.3             ##  [13] crayon_1.5.2                fastmap_1.1.1               ##  [15] XVector_0.43.1              utf8_1.2.4                  ##  [17] rmarkdown_2.25              preprocessCore_1.65.0       ##  [19] ragg_1.2.7                  purrr_1.0.2                 ##  [21] xfun_0.41                   MultiAssayExperiment_1.29.0 ##  [23] zlibbioc_1.49.0             cachem_1.0.8                ##  [25] jsonlite_1.8.8              progress_1.2.3              ##  [27] highr_0.10                  DelayedArray_0.29.0         ##  [29] prettyunits_1.2.0           parallel_4.4.0              ##  [31] cluster_2.1.6               R6_2.5.1                    ##  [33] bslib_0.6.1                 stringi_1.8.3               ##  [35] limma_3.59.1                jquerylib_0.1.4             ##  [37] bookdown_0.37               iterators_1.0.14            ##  [39] knitr_1.45                  splines_4.4.0               ##  [41] Matrix_1.6-5                igraph_1.6.0                ##  [43] tidyselect_1.2.0            abind_1.4-5                 ##  [45] yaml_2.3.8                  doParallel_1.0.17           ##  [47] codetools_0.2-19            affy_1.81.0                 ##  [49] lattice_0.22-5              tibble_3.2.1                ##  [51] plyr_1.8.9                  evaluate_0.23               ##  [53] survival_3.5-7              desc_1.4.3                  ##  [55] Spectra_1.13.3              pillar_1.9.0                ##  [57] affyio_1.73.0               BiocManager_1.30.22         ##  [59] foreach_1.5.2               MALDIquant_1.22.2           ##  [61] ncdf4_1.22                  generics_0.1.3              ##  [63] RCurl_1.98-1.14             hms_1.1.3                   ##  [65] ggplot2_3.4.4               munsell_0.5.0               ##  [67] scales_1.3.0                glue_1.7.0                  ##  [69] MsFeatures_1.11.0           lazyeval_0.2.2              ##  [71] tools_4.4.0                 robustbase_0.99-1           ##  [73] mzID_1.41.0                 QFeatures_1.13.2            ##  [75] vsn_3.71.0                  RANN_2.6.1                  ##  [77] fs_1.6.3                    XML_3.99-0.16.1             ##  [79] grid_4.4.0                  impute_1.77.0               ##  [81] MsCoreUtils_1.15.3          colorspace_2.1-0            ##  [83] GenomeInfoDbData_1.2.11     PSMatch_1.7.1               ##  [85] cli_3.6.2                   textshaping_0.3.7           ##  [87] fansi_1.0.6                 S4Arrays_1.3.2              ##  [89] dplyr_1.1.4                 AnnotationFilter_1.27.0     ##  [91] DEoptimR_1.1-3              pcaMethods_1.95.0           ##  [93] gtable_0.3.4                sass_0.4.8                  ##  [95] digest_0.6.34               SparseArray_1.3.3           ##  [97] farver_2.1.1                multtest_2.59.0             ##  [99] memoise_2.0.1               htmltools_0.5.7             ## [101] pkgdown_2.0.7.9000          lifecycle_1.0.4             ## [103] statmod_1.5.0               MASS_7.3-60.2"},{"path":[]},{"path":"https://sneumann.github.io/xcms/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Colin . Smith. Author. Ralf Tautenhahn. Author. Steffen Neumann. Author, maintainer. Paul Benton. Author. Christopher Conley. Author. Johannes Rainer. Author. Michael Witting. Contributor. William Kumler. Author. Philippine Louail. Author.","code":""},{"path":"https://sneumann.github.io/xcms/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Smith, C.. Want, E.J. O'Maille, G. Abagyan,R. Siuzdak, G.: XCMS: Processing mass spectrometry data metabolite profiling using nonlinear peak alignment, matching identification, Analytical Chemistry, 78:779-787 (2006) Ralf Tautenhahn, Christoph Boettcher, Steffen Neumann: Highly sensitive feature detection high resolution LC/MS BMC Bioinformatics, 9:504 (2008) H. Paul Benton, Elizabeth J. Want Timothy M. D. Ebbels Correction mass calibration gaps liquid chromatography-mass spectrometry metabolomics data Bioinformatics, 26:2488 (2010)","code":"@Article{,   author = {{Smith} and {C.A.} and {Want} and {E.J.} and {O'Maille} and {G.} and {Abagyan,R.} and {Siuzdak} and {G.}},   title = {XCMS: Processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching and identification},   journal = {Analytical Chemistry},   year = {2006},   volume = {78},   pages = {779--787}, } @Article{,   author = {Ralf Tautenhahn and Christoph Boettcher and Steffen Neumann},   title = {Highly sensitive feature detection for high resolution LC/MS},   journal = {BMC Bioinformatics},   year = {2008},   volume = {9},   pages = {504}, } @Article{,   author = {H. Paul Benton and Elizabeth J. Want and Timothy M. D. Ebbels},   title = {Correction of mass calibration gaps in liquid chromatography-mass spectrometry metabolomics data},   journal = {BIOINFORMATICS},   year = {2010},   volume = {26},   pages = {2488}, }"},{"path":"https://sneumann.github.io/xcms/index.html","id":"the-xcms-package-pre-processing-gclc-msms-data","dir":"","previous_headings":"","what":"LC-MS and GC-MS Data Analysis","title":"LC-MS and GC-MS Data Analysis","text":"Please see package documentation information examples news latest changes.","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"version-4","dir":"","previous_headings":"","what":"Version 4","title":"LC-MS and GC-MS Data Analysis","text":"Version 4 adds native support Spectra package xcms allows perform pre-processing MsExperiment objects (MsExperiment. new supported data containers (Spectra, MsExperiment XcmsExperiment) allow flexible analyses seamless future extensions additional types data (ion mobility data). Ultimately, changes also allow easier integration xcms R packages MsFeatures MetaboAnnotation. suggested users switch newer data result objects, functionality version 3 remain fully supported.","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"version-3","dir":"","previous_headings":"","what":"Version 3","title":"LC-MS and GC-MS Data Analysis","text":"Version >= 3 xcms package updated partially re-written versions original xcms package. version number 3 selected avoid confusions xcms2 (http://pubs.acs.org/doi/abs/10.1021/ac800795f) software. providing original software’s functionality, xcms version >= 3 aims : Better integration Bioconductor framework: Make use extend classes defined MSnbase package. Implement class versioning (Biobase’s Versioned class). Use BiocParallel parallel processing. Implementation validation methods classes ensure data integrity. Easier faster access raw spectra data. Cleanup source code: Remove obsolete redundant functionality (getEIC, rawEIC etc). Unify interfaces, .e. implement layer base functions accessing analysis methods (implemented C, C++ R). Using consistent naming scheme methods follows established naming conventions (e.g. correspondence instead grouping). Update, improve extend documentation. Establishing layer base R-functions interface analysis methods. take M/Z, retention time (scan index) intensity values input along optional arguments downstream functions (implemented C, C++ R). input arguments basic R objects (numeric vectors) thus enabling easy integration analysis methods R packages. user interface’s analysis methods take (raw) data object parameter class, used dispatching corresponding analysis algorithm. Add unit tests. Discussions suggestions welcome: https://github.com/sneumann/xcms/issues","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"contribution","dir":"","previous_headings":"","what":"Contribution","title":"LC-MS and GC-MS Data Analysis","text":"Contributions xcms package welcome, whether form ideas, documentation, code, packages, … contribution guideline please see guideline RforMassSpectrometry initiative. seamless integration, contributors expected adhere RforMassSpectrometry coding syle.","code":""},{"path":"https://sneumann.github.io/xcms/index.html","id":"code-of-conduct","dir":"","previous_headings":"","what":"Code of Conduct","title":"LC-MS and GC-MS Data Analysis","text":"contributors maintainers package, pledge respect people contribute reporting issues, posting feature requests, updating documentation, submitting pull requests patches, activities. See RforMassSpectrometry Code Conduct information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"AutoLockMass  - function decides lock mass scans \txcmsRaw object. done using scan time differences.","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"object = \"xcmsRaw\" signature(object = \"xcmsRaw\")","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"object xcmsRaw-class object","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"AutoLockMass numeric vector scan locations corresponding lock Mass scans","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"Paul Benton, hpaul.benton08@imperial.ac.uk","code":""},{"path":"https://sneumann.github.io/xcms/reference/AutoLockMass-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Automatic parameter for Lock mass fixing AutoLockMass ~~ — AutoLockMass-methods","text":"","code":"if (FALSE) library(xcms)     library(faahKO)     ## These files do not have this problem     ## to correct for but just for an example     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xr<-xcmsRaw(cdffiles[1]) #> Create profile matrix with method 'bin' and step 1 ...  #> OK     xr #> An \"xcmsRaw\" object with 1278 mass spectra #>  #> Time range: 2501.4-4499.8 seconds (41.7-75 minutes) #> Mass range: 200-600 m/z #> Intensity range: 70-1373180  #>  #> MSn data on  0  mass(es) #> \twith  0  MSn spectra #> Profile method: bin  #> Profile step: 1 m/z (401 grid points from 200 to 600 m/z) #>  #> Memory usage: 10.8 MB     ##Lets assume that the lockmass starts at 1 and is every 100 scans     lockMass<-xcms:::makeacqNum(xr, freq=100, start=1)     ## these are equalvent     lockmass2<-AutoLockMass(xr) #> Warning:  #> Lock mass frequency wasn't detected     all((lockmass == lockmass2) == TRUE) #> Error in eval(expr, envir, enclos): object 'lockmass' not found      ob<-stitch(xr, lockMass) #> Warning: no non-missing arguments to min; returning Inf #> Warning: no non-missing arguments to max; returning -Inf"},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":null,"dir":"Reference","previous_headings":"","what":"Generic parameter class — GenericParam-class","title":"Generic parameter class — GenericParam-class","text":"GenericParam class allows store generic parameter     information name function /called     (slot fun) arguments (slot args). object     used track process history data processings     XCMSnExp object. contrast e.g.     CentWaveParam object passed actual     processing method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generic parameter class — GenericParam-class","text":"","code":"GenericParam(fun = character(), args = list())"},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generic parameter class — GenericParam-class","text":"fun character representing name function. args list (ideally named) arguments function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generic parameter class — GenericParam-class","text":"GenericParam function returns GenericParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Generic parameter class — GenericParam-class","text":"fun character specifying function name. args list (ideally named) arguments function.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generic parameter class — GenericParam-class","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/GenericParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generic parameter class — GenericParam-class","text":"","code":"prm <- GenericParam(fun = \"mean\")  prm <- GenericParam(fun = \"mean\", args = list(na.rm = TRUE))"},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":null,"dir":"Reference","previous_headings":"","what":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"`PlainTextParam` class method enable users save `MsExperiment` `XcmsExperiment` objects collections plain text files specified folder. Note , xcms results within `XcmsExperiment` can exported, full raw MS data (object's `Spectra` object) currently exported plain text format. `Spectra` using [MsBackendMzR()] backend, names raw data files however exported (enables *restore* full `Spectra` respectively `MsExperiment` objects). `MsExperiment` object, exported files include: - [sampleData()] stored text file named *sample_data.txt*. - [fileNames()] *Spectra* object stored tabular format text file named *spectra_files.txt*.file names exported `Spectra` object uses [MsBackendMzR()] backend. backends information raw spectra data currently exported `PlainTextParam`. - Processing queue `Spectra` object, ensuring spectra data modifications retained. stored `json` file named *spectra_processing_queue.json*. `XcmsExperiment` object, exported files `MsExperiment` object, addition following: - [processHistory()] information object, stored `json` file named *process_history.json*. - chromatographic peak information obtained [chromPeaks()] [chromPeaksData()], stored tabular format text files *chrom_peaks.txt* *chrom_peak_data.txt* respectively. - retention time information obtained [adjustedRtime()] stored text file named *rtime_adjusted.txt*. - [featureDefinitions()] stored text file named *feature_definitions.txt*. Additionally, second file named *feature_peak_index.txt* generated connect features' definitions names. `param` class method part possible dispatch generic function `storeResults`. folder created calling `storeResults`. folder already exists, previous exports folder might get overwritten.","code":""},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"","code":"PlainTextParam(path = tempdir())  # S4 method for MsExperiment,PlainTextParam storeResults(object, param)  # S4 method for XcmsExperiment,PlainTextParam storeResults(object, param)"},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"path `PlainTextParam` `character(1)`, defining files going stored. default `tempdir()`. object MsExperiment XcmsExperiment data object needs saved. param parameter object selecting configuring format saving. can one following classes: RDataParam, PlainTextParam, MzTabMParam.","code":""},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"`PlainTextParam`: `PlainTextParam` class. `storeResults` return anything saves object collections different plain text files folder.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"Philippine Louail, Johannes Rainer.","code":""},{"path":"https://sneumann.github.io/xcms/reference/PlainTextParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Store contents of `MsExperiment` and `XcmsExperiment` objects as plain text files — PlainTextParam","text":"","code":"## Load test data set of class `MsExperiment` library(MsExperiment) fls <- dir(system.file(\"sciex\", package = \"msdata\"), full.names = TRUE) pd <- data.frame(file = basename(fls),                  sample = c(\"POOL_1\", \"POOL_2\"),                  injection_index = c(1, 19),                 group = \"POOL\") rownames(pd) <- c(\"1\", \"2\") mse <- readMsExperiment(fls, sampleData = pd)  ## Define param pth = file.path(tempdir(), \"test\") param <- PlainTextParam(path = pth)  ## Save as a collection of plain text files storeResults(object = mse, param = param)  ## Load a test data set with detected peaks, of class `XcmsExperiment` faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Define param pth = file.path(tempdir(), \"test\") param <- PlainTextParam(path = pth)  ## Save as a collection of plain text files storeResults(object = faahko_sub, param = param) #> Warning: '/tmp/RtmpPglK9Y/test' already exists"},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Tracking data processing — ProcessHistory-class","title":"Tracking data processing — ProcessHistory-class","text":"Objects type ProcessHistory allow keep track     data processing step metabolomics experiment.     created data processing methods,     findChromPeaks added corresponding results     objects. Thus, usually, users need create . XProcessHistory extends ProcessHistory     adding slot param allows store actual parameter     class processing step. processParam, processParam<-: get set     parameter class XProcessHistory object. msLevel: returns MS level certain analysis     performed, NA defined. processType method returns character specifying     processing step type. processDate extracts start date processing     step. processInfo extracts optional additional information     processing step. fileIndex extracts indices files     processing step applied.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tracking data processing — ProcessHistory-class","text":"","code":"# S4 method for ProcessHistory show(object)  # S4 method for XProcessHistory show(object)  # S4 method for XProcessHistory processParam(object)  # S4 method for XProcessHistory msLevel(object)  # S4 method for ProcessHistory processType(object)  # S4 method for ProcessHistory processDate(object)  # S4 method for ProcessHistory processInfo(object)  # S4 method for ProcessHistory fileIndex(object)"},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tracking data processing — ProcessHistory-class","text":"object ProcessHistory XProcessHistory object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tracking data processing — ProcessHistory-class","text":"processParam: parameter object extending  Param class. processType method returns character string     processing step type. processDate method returns character string     time stamp processing step start. processInfo method returns character string     optional additional informations. fileIndex method returns integer vector index     files/samples processing step applied.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Tracking data processing — ProcessHistory-class","text":"type character(1): string defining type processing step. string match predefined values. Use processHistoryTypes list . date character(1): date time stamp processing step started. info character(1): optional additional information. fileIndex integer length 1 > 1 specify samples object processing performed. error (): used store eventual calculation errors. param (Param): object type Param (e.g. CentWaveParam) specifying settings processing step. msLevel: integer definining MS level(s) analysis performed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/ProcessHistory-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Tracking data processing — ProcessHistory-class","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":null,"dir":"Reference","previous_headings":"","what":"Store `XcmsExperiment` object as .RData file — RDataParam","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"`RDataParam` class method allow users save `XcmsExperiment` object .RData file chosen filename. object gets exported using [`save()`] function.  `param` class method part possible dispatch generic function `storeResults`.","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"","code":"RDataParam(fileName = tempfile())  # S4 method for XcmsExperiment,RDataParam storeResults(object, param)"},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"fileName `RDataParam` `character(1)`, defining file name. default `tempfile()`. object MsExperiment XcmsExperiment data object needs saved. param parameter object selecting configuring format saving. can one following classes: RDataParam, PlainTextParam, MzTabMParam.","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"`RDataParam`: `RDataParam` class. `storeResults` return anything saves object RData file.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"Philippine Louail","code":""},{"path":"https://sneumann.github.io/xcms/reference/RDataParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Store `XcmsExperiment` object as .RData file — RDataParam","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Define param  param <- RDataParam(fileName = \"example_xcms_object\")  ## Save as RData storeResults(object = faahko_sub, param = param)"},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":null,"dir":"Reference","previous_headings":"","what":"Gaussian Model — SSgauss","title":"Gaussian Model — SSgauss","text":"selfStart model evalueates Gaussian model   gradient. initial attribute evalueate   inital estimates parameters mu, sigma,   h.","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Gaussian Model — SSgauss","text":"","code":"SSgauss(x, mu, sigma, h)"},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Gaussian Model — SSgauss","text":"x numeric vector values evaluate model mu mean  distribution function sigma standard deviation distribution fuction h height distribution function","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Gaussian Model — SSgauss","text":"Initial values mu h chosen   maximal value x. initial value sigma   determined area x divided h*sqrt(2*pi).","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Gaussian Model — SSgauss","text":"numeric vector length x. value   expression h*exp(-(x-mu)^2/(2*sigma^2),   modified gaussian function maximum height treated   separate parameter dependent sigma. arguments  mu, sigma, h names objects,   gradient matrix respect names attached   attribute named gradient.","code":""},{"path":"https://sneumann.github.io/xcms/reference/SSgauss.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Gaussian Model — SSgauss","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Data container storing xcms preprocessing results — XCMSnExp-class","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"XCMSnExp object container results G/LC-MS data preprocessing comprises chromatographic peak detection, alignment correspondence. results can accessed chromPeaks, adjustedRtime featureDefinitions functions; see (Usage, Arguments, Value Slots sections) details). Along results, object contains processing history allows track processing step along used settings. can extracted processHistory method. XCMSnExp objects, directly extending OnDiskMSnExp object MSnbase package, inherit functionality allows thus easy access full raw data stage analysis. support interaction packages requiring old objects, XCMSnExp objects can coerced xcmsSet objects using method (see examples ). preprocessing results passed along resulting xcmsSet object. General functions XCMSnExp objects (see specific function handle chromatographic peak data, alignment correspondence results): processHistoryTypes returns available types     process histories. can passed argument type     processHistory method extract specific process step(s). hasFilledChromPeaks: whether filled-peaks present . profMat: creates profile matrix,     n x m matrix, n (rows) representing equally spaced m/z values (bins)     m (columns) retention time corresponding scans. cell     contains maximum intensity measured specific scan m/z     values. See profMat details description     various binning methods. hasAdjustedRtime: whether object provides adjusted retention times. hasFeatures: whether object contains correspondence results (.e. features). hasChromPeaks: whether object contains peak detection results. hasFilledChromPeaks: whether object contains filled-chromatographic peaks. adjustedRtime,adjustedRtime<-: extract/set adjusted retention times. adjustedRtime<- called manually, called internally adjustRtime methods. XCMSnExp objects, adjustedRtime<- also apply retention time adjustments eventually present chromatographic peaks. bySample parameter allows specify whether adjusted retention time grouped sample (file). featureDefinitions, featureDefinitions<-: extract set correspondence results, .e. mz-rt features (peak groups). Similar chromPeaks possible extract features specified m/z /rt ranges. function supports also parameter type allows specify features returned rt mz specified. details see help chromPeaks. See also featureSummary function calculate simple feature summaries. chromPeaks, chromPeaks<-: extract set matrix containing information identified chromatographic peaks. Rownames matrix represent unique IDs respective peaks within experiment. Parameter bySample allows specify whether peaks returned ungrouped (default bySample = FALSE) grouped sample (bySample = TRUE). chromPeaks<- method XCMSnExp objects removes also correspondence (peak grouping) retention time correction (alignment) results. optional arguments rt, mz, ppm type allow extract chromatographic peaks overlapping defined retention time /m/z ranges. Argument type allows define overlapping determined: type == \"\" (default), peaks even partially overlapping region returned (.e. either \"mzmin\" \"mzmax\" chromPeaks featureDefinitions matrix within provided m/z range), type == \"within\" full peak within region (.e. \"mzmin\" \"mzmax\" within m/z range) type == \"apex_within\" peak's apex position (highest signal peak) within region (.e. peak's features m/z within m/z range). See description return value details returned matrix. Users usually use chromPeaks<- method directly detected chromatographic peaks added object findChromPeaks method. Also, chromPeaks<- replace existing chromPeakData. chromPeakData chromPeakData<- allow get set arbitrary chromatographic peak annotations. returned ar returned DataFrame. Note number rows rownames DataFrame match chromPeaks. rtime: extracts retention time scan. bySample parameter allows return values grouped sample/file adjusted whether adjusted raw retention times returned. default method returns adjusted retention times, available (.e. retention times adjusted using adjustRtime method). mz: extracts mz values scan files within XCMSnExp object. values extracted original data files eventual processing steps applied fly. Using bySample parameter possible switch default grouping mz values spectrum/scan grouping sample/file. intensity: extracts intensity values scan files within XCMSnExp object. values extracted original data files eventual processing steps applied fly. Using bySample parameter possible switch default grouping intensity values spectrum/scan grouping sample/file. spectra: extracts Spectrum objects containing data object. values extracted original data files eventual processing steps applied fly. setting bySample = TRUE, spectra returned grouped sample/file. XCMSnExp object contains adjusted retention times, returned default Spectrum objects (can overwritten setting adjusted = FALSE). processHistory: returns list ProcessHistory objects (objects inheriting base class) representing individual processing steps performed, eventually along settings (Param parameter class). Optional arguments fileIndex, type msLevel allow restrict process steps certain type performed certain file MS level. dropChromPeaks: drops identified chromatographic peaks returns object without information. Note XCMSnExp objects method drops default also results correspondence (peak grouping) analysis. Adjusted retention times removed alignment performed peak detection. can overruled keepAdjustedRtime = TRUE. dropFeatureDefinitions: drops results correspondence (peak grouping) analysis, .e. definition mz-rt features returns object without information. Note XCMSnExp objects method also default drop retention time adjustment results, performed last peak grouping (.e. base results peak grouping going removed). related process history steps removed well eventually filled peaks (fillChromPeaks). parameter keepAdjustedRtime can used avoid removal adjusted retention times. dropAdjustedRtime: drops retention time adjustment information returns object without adjusted retention time. XCMSnExp objects, also reverts retention times reported chromatographic peaks peak matrix original, raw, ones (chromatographic peak detection). Note XCMSnExp objects method drops also peak grouping results performed retention time adjustment. related process history steps removed . findChromPeaks performs chromatographic peak detection provided XCMSnExp objects. details see method XCMSnExp. Note default (parameter add = FALSE) previous peak detection results removed. Use add = TRUE perform second round peak detection add newly identified peaks previous peak detection results. Correspondence results (features) always removed prior peak detection. Previous alignment (retention time adjustment) results kept, .e. chromatographic peak detection performed using adjusted retention times data first aligned using e.g. obiwarp (adjustRtime). dropFilledChromPeaks: drops filled-chromatographic peaks (filled fillChromPeaks method) related process history steps. spectrapply applies provided function Spectrum object returns results. function specified function simply returns list Spectrum objects. XCMSnExp objects can combined c function. combines identified chromatographic peaks objects' pheno data discards alignment results feature definitions. plot plots spectrum data (see plot MSnExp objects MSnbase package details. type = \"XIC\", identified chromatographic peaks indicated rectangles border color peakCol.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"","code":"processHistoryTypes()  # S4 method for XCMSnExp hasFilledChromPeaks(object)  # S4 method for OnDiskMSnExp profMat(   object,   method = \"bin\",   step = 0.1,   baselevel = NULL,   basespace = NULL,   mzrange. = NULL,   fileIndex,   ... )  # S4 method for XCMSnExp show(object)  # S4 method for XCMSnExp hasAdjustedRtime(object)  # S4 method for XCMSnExp hasFeatures(object, msLevel = integer())  # S4 method for XCMSnExp hasChromPeaks(object, msLevel = integer())  # S4 method for XCMSnExp hasFilledChromPeaks(object)  # S4 method for XCMSnExp adjustedRtime(object, bySample = FALSE)  # S4 method for XCMSnExp adjustedRtime(object) <- value  # S4 method for XCMSnExp featureDefinitions(   object,   mz = numeric(),   rt = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel = integer() )  # S4 method for XCMSnExp featureDefinitions(object) <- value  # S4 method for XCMSnExp chromPeaks(   object,   bySample = FALSE,   rt = numeric(),   mz = numeric(),   ppm = 0,   msLevel = integer(),   type = c(\"any\", \"within\", \"apex_within\"),   isFilledColumn = FALSE )  # S4 method for XCMSnExp chromPeaks(object) <- value  # S4 method for XCMSnExp rtime(object, bySample = FALSE, adjusted = hasAdjustedRtime(object))  # S4 method for XCMSnExp mz(object, bySample = FALSE, BPPARAM = bpparam())  # S4 method for XCMSnExp intensity(object, bySample = FALSE, BPPARAM = bpparam())  # S4 method for XCMSnExp spectra(   object,   bySample = FALSE,   adjusted = hasAdjustedRtime(object),   BPPARAM = bpparam() )  # S4 method for XCMSnExp processHistory(object, fileIndex, type, msLevel)  # S4 method for XCMSnExp dropChromPeaks(object, keepAdjustedRtime = FALSE)  # S4 method for XCMSnExp dropFeatureDefinitions(object, keepAdjustedRtime = FALSE, dropLastN = -1)  # S4 method for XCMSnExp dropAdjustedRtime(object)  # S4 method for XCMSnExp profMat(   object,   method = \"bin\",   step = 0.1,   baselevel = NULL,   basespace = NULL,   mzrange. = NULL,   fileIndex,   ... )  # S4 method for XCMSnExp,Param findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   add = FALSE )  # S4 method for XCMSnExp dropFilledChromPeaks(object)  # S4 method for XCMSnExp spectrapply(object, FUN = NULL, BPPARAM = bpparam(), ...)  # S3 method for XCMSnExp c(...)  # S4 method for XCMSnExp chromPeakData(object, ...)  # S4 method for XCMSnExp chromPeakData(object) <- value  # S4 method for XCMSnExp,missing plot(x, y, type = c(\"spectra\", \"XIC\"), peakCol = \"#ff000060\", ...)"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"object adjustedRtime, featureDefinitions, chromPeaks, hasAdjustedRtime, hasFeatures hasChromPeaks either MsFeatureData XCMSnExp object, methods XCMSnExp object. method character(1) defining profile matrix generation method. Allowed \"bin\", \"binlin\", \"binlinbase\" \"intlin\". See details section information. step numeric(1) representing m/z bin size. baselevel numeric(1) representing base value empty elements (.e. m/z bins without measured intensity) set. considered method = \"binlinbase\". See baseValue parameter imputeLinInterpol() details. basespace numeric(1) representing m/z length signal drop base level. Linear interpolation used consecutive data points falling within 2 * basespace . considered method = \"binlinbase\". specified, defaults 0.075. Internally parameter translated distance parameter imputeLinInterpol() function distance = floor(basespace / step). See distance parameter imputeLinInterpol() details. mzrange. Optional numeric(2) manually specifying mz value range used binnind. provided, whole m/z value range used. fileIndex processHistory: optional integer specifying index files/samples ProcessHistory objects retrieved. ... Additional parameters. msLevel integer specifying MS level(s) identified chromatographic peaks returned. bySample logical(1) specifying whether results grouped sample. value adjustedRtime<-: list (length equal     number samples) numeric vectors representing adjusted     retention times per scan. featureDefinitions<-: DataFrame peak     grouping information. See return value featureDefinitions     method expected format. chromPeaks<-: matrix information     detected peaks. See return value chromPeaks method     expected format. mz optional numeric(2) defining mz range chromatographic peaks returned. rt optional numeric(2) defining retention time range chromatographic peaks returned. ppm optional numeric(1) specifying ppm mz range extended. value ppm = 10, peaks within mz[1] - ppm / 1e6 mz[2] + ppm / 1e6 returned. type processHistory: restrict returned ProcessHistory objects analysis steps certain type. Use processHistoryTypes list supported values. chromPeaks: character specifying peaks return rt mz defined. type = \"\" chromatographic peaks partially overlapping range defined mz /rt returned, type = \"within\" returns peaks completely within region type = \"apex_within\" peaks peak's apex within region. isFilledColumn logical(1) whether column \"is_filled\" included returned \"matrix\" providing information peak filled . Alternatively, information provided chromPeakData data frame. adjusted logical(1) whether adjusted raw (.e. original retention times reported files) returned. BPPARAM Parameter class parallel processing. See bpparam. keepAdjustedRtime dropFeatureDefinitions,XCMSnExp: logical(1) defining whether eventually present retention time adjustment dropped. default dropping feature definitions drops retention time adjustment results . dropLastN dropFeatureDefinitions,XCMSnExp: numeric(1) defining number peak grouping related process history steps remove. default dropLastN = -1, dropping chromatographic peaks removes process history steps related peak grouping. Setting e.g. dropLastN = 1 remove recent peak grouping related process history step. param CentWaveParam, MatchedFilterParam, MassifquantParam, MSWParam CentWavePredIsoParam object settings chromatographic peak detection algorithm. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". add findChromPeaks: newly identified chromatographic peaks added peak matrix already identified chromatographic peaks. default (add = FALSE) previous peak detection results removed. FUN spectrapply: function applied spectrum object. x plot: XCMSnExp object. y plot: used. peakCol plot: color used indicate identified chromatographic peaks (combination type = \"XIC\" chromatographic peaks present).","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"profMat: list profile matrix  matrix (matrices fileIndex specified  length(fileIndex) > 1). See profile-matrix     general help information profile matrix. adjustedRtime: bySample = FALSE numeric vector adjusted retention spectrum files/samples within object. bySample = TRUE  list (length equal number samples) adjusted retention times grouped sample. Returns NULL adjusted retention times present. featureDefinitions: DataFrame peak grouping information, row corresponding one mz-rt feature (grouped peaks within across samples) columns \"mzmed\" (median mz value), \"mzmin\" (minimal mz value), \"mzmax\" (maximum mz value), \"rtmed\" (median retention time), \"rtmin\" (minimal retention time), \"rtmax\" (maximal retention time) \"peakidx\". Column \"peakidx\" contains list indices chromatographic peaks (rows) matrix returned chromPeaks method belong feature group. method returns NULL feature definitions present. featureDefinitions supports also parameters mz, rt, ppm type return features within certain ranges (see description chromPeaks details). chromPeaks: bySample = FALSE matrix (row chromatographic peak, rownames representing unique IDs peaks) least following columns: \"mz\" (intensity-weighted mean mz values peak across scans/retention times), \"mzmin\" (minimal mz value), \"mzmax\" (maximal mz value), \"rt\" (retention time peak apex), \"rtmin\" (minimal retention time), \"rtmax\" (maximal retention time), \"\" (integrated, original, intensity peak), \"maxo\" (maximum intentity peak), \"sample\" (sample index peak identified) Depending employed peak detection algorithm verboseColumns parameter , additional columns might returned. parameter isFilledColumn set TRUE column named \"is_filled\" also returned. bySample = TRUE chromatographic peaks returned list matrices, containing chromatographic peaks specific sample. samples peaks detected matrix 0 rows returned. rtime: bySample = FALSE numeric vector retention times scan, bySample = TRUE list numeric vectors retention times per sample. mz: bySample = FALSE list mz values (numeric vectors) scan. bySample = TRUE list mz values per sample. intensity: bySample = FALSE list intensity values (numeric vectors) scan. bySample = TRUE list intensity values per sample. spectra: bySample = FALSE list Spectrum objects. bySample = TRUE result grouped sample, .e. list lists, element outer  list list spectra specific file. processHistory: list ProcessHistory objects providing details individual data processing steps performed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":".processHistory list XProcessHistory objects tracking individual analysis steps performed. msFeatureData MsFeatureData class extending environment containing results chromatographic peak detection (element \"chromPeaks\"), peak grouping (element \"featureDefinitions\") retention time correction (element \"adjustedRtime\") steps. object manipulated directly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"\"chromPeaks\" element msFeatureData slot     equivalent @peaks slot xcmsSet object,     \"featureDefinitions\" contains information @groups     @groupidx slots xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"chromatographic-peak-data","dir":"Reference","previous_headings":"","what":"Chromatographic peak data","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"Chromatographic peak data added XCMSnExp object findChromPeaks function. Functions access chromatographic peak data : hasChromPeaks whether chromatographic peak data available, see help function. chromPeaks access chromatographic peaks (see help). dropChromPeaks remove chromatographic peaks (see help). dropFilledChromPeaks remove filled-peaks (see help). fillChromPeaks fill-missing peaks (see respective help page). plotChromPeaks plot identified peaks file (see respective help page). plotChromPeakImage plot distribution peaks along retention time axis (see respective help page). highlightChromPeaks add chromatographic peaks existing plot Chromatogram (see respective help page).","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"adjusted-retention-times","dir":"Reference","previous_headings":"","what":"Adjusted retention times","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"Adjusted retention times stored XCMSnExp object besides original, raw, retention times, allowing switch raw adjusted times. also possible replace raw retention times adjusted ones applyAdjustedRtime. adjusted retention times added XCMSnExp adjustRtime function. functions related access  adjusted retention times : hasAdjustedRtime whether adjusted retention times available (see help). dropAdjustedRtime remove adjusted retention times (see help). applyAdjustedRtime replace raw retention times adjusted ones (see respective help page). plotAdjustedRtime plot differences adjusted raw retention times (see respective help page).","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"correspondence-results-features","dir":"Reference","previous_headings":"","what":"Correspondence results, features","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"correspondence analysis (groupChromPeaks) adds feature definitions XCMSnExp object. functions related listed : hasFeatures whether correspondence results available (see help). featureDefinitions access definitions features (see help). dropFeatureDefinitions remove correspondence results (see help). featureValues access values features (see respective help page). featureSummary perform simple summary defined features (see respective help page). overlappingFeatures identify features overlapping close m/z - rt space (see respective help page). quantify extract feature intensities put , along feature definitions phenodata information, SummarizedExperiment. See help page details.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-class.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data container storing xcms preprocessing results — XCMSnExp-class","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## The results from the peak detection are now stored in the XCMSnExp ## object faahko_sub #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 1.22 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 3834  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 1 2 3 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   [1] ko15.CDF...  [3] ko18.CDF #>   Use 'fileNames(.)' to see all files. #> protocolData: none #> featureData #>   featureNames: F1.S0001 F1.S0002 ... F3.S1278 (3834 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave  #>  248 peaks identified in 3 samples. #>  On average 82.7 chromatographic peaks per sample.  ## The detected peaks can be accessed with the chromPeaks method. head(chromPeaks(faahko_sub)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1  ## The settings of the chromatographic peak detection can be accessed with ## the processHistory method processHistory(faahko_sub) #> [[1]] #> Object of class \"XProcessHistory\" #>  type: Peak detection  #>  date: Tue Aug  1 15:54:19 2023  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: CentWaveParam  #>  MS level(s) 1  #>   ## Also the parameter class for the peak detection can be accessed processParam(processHistory(faahko_sub)[[1]]) #> Object of class:  CentWaveParam  #>  Parameters: #>  - ppm: [1] 25 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 40 #>  - prefilter: [1]     3 10000 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 10000 #>  - verboseColumns: [1] FALSE #>  - roiList: list() #>  - firstBaselineCheck: [1] TRUE #>  - roiScales: numeric(0) #>  - extendLengthMSW: [1] FALSE #>  - verboseBetaColumns: [1] FALSE  ## The XCMSnExp inherits all methods from the pSet and OnDiskMSnExp classes ## defined in Bioconductor's MSnbase package. To access the (raw) retention ## time for each spectrum we can use the rtime method. Setting bySample = TRUE ## would cause the retention times to be grouped by sample head(rtime(faahko_sub)) #> F1.S0001 F1.S0002 F1.S0003 F1.S0004 F1.S0005 F1.S0006  #> 2501.378 2502.943 2504.508 2506.073 2507.638 2509.203   ## Similarly it is possible to extract the mz values or the intensity values ## using the mz and intensity method, respectively, also with the option to ## return the results grouped by sample instead of the default, which is ## grouped by spectrum. Finally, to extract all of the data we can use the ## spectra method which returns Spectrum objects containing all raw data. ## Note that all these methods read the information from the original input ## files and subsequently apply eventual data processing steps to them. mzs <- mz(faahko_sub, bySample = TRUE) length(mzs) #> [1] 3 lengths(mzs) #>      1      2      3  #> 445769 440794 442516   ## The full data could also be read using the spectra data, which returns ## a list of Spectrum object containing the mz, intensity and rt values. ## spctr <- spectra(faahko_sub) ## To get all spectra of the first file we can split them by file ## head(split(spctr, fromFile(faahko_sub))[[1]])  ############ ## Filtering ## ## XCMSnExp objects can be filtered by file, retention time, mz values or ## MS level. For some of these filter preprocessing results (mostly ## retention time correction and peak grouping results) will be dropped. ## Below we filter the XCMSnExp object by file to extract the results for ## only the second file. xod_2 <- filterFile(faahko_sub, file = 2) xod_2 #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #> Filter: select file(s) 2. [Fri Jan 26 20:05:21 2024]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 2 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko16.CDF  #> protocolData: none #> featureData #>   featureNames: F2.S0001 F2.S0002 ... F2.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave  #>  100 peaks identified in 1 samples. #>  On average 100 chromatographic peaks per sample.  ## Now the objects contains only the idenfified peaks for the second file head(chromPeaks(xod_2)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into        intb #> CP088 384.1 384.1 384.1 2603.100 2592.145 2614.054   274906.3   274885.96 #> CP089 533.3 533.3 533.3 2686.042 2681.347 2689.172    70456.3    70450.04 #> CP090 344.0 344.0 344.0 2686.042 2648.483 2714.211  5700652.0  5574744.87 #> CP091 366.0 366.0 366.0 2686.042 2639.094 2725.166  3448376.1  3338483.07 #> CP092 343.0 343.0 343.0 2686.042 2645.353 2717.341 26229546.1 25615117.17 #> CP093 365.0 365.0 365.0 2686.042 2639.094 2726.731 15565868.0 14884380.84 #>         maxo    sn sample #> CP088  15187 15186      1 #> CP089  11819 11818      1 #> CP090 151744    87      1 #> CP091  77640    42      1 #> CP092 692352    68      1 #> CP093 345536    48      1  ########## ## Coercing to an xcmsSet object ## ## We can also coerce the XCMSnExp object into an xcmsSet object: xs <- as(faahko_sub, \"xcmsSet\") #> Note: you might want to set/adjust the 'sampclass' of the returned xcmSet object before proceeding with the analysis. head(peaks(xs)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"methods listed page allow filter subset XCMSnExp objects. inherited OnDiskMSnExp object defined MSnbase package adapted XCMSnExp enable correct subsetting preprocessing results. [: subset XCMSnExp object spectra. aware removes preprocessing results, except adjusted retention times keepAdjustedRtime = TRUE passed method. [[: extracts single Spectrum object (defined MSnbase). reported retention time adjusted retention time alignment performed. filterChromPeaks: subset chromPeaks matrix object. Parameter method allows specify chromatographic peaks filtered. Currently, method = \"keep\" supported allows specify chromatographic peaks keep parameter keep (.e. provide logical, integer character defining chromatographic peaks keep). Feature definitions (present) updated correspondingly. filterFeatureDefinitions: allows subset feature definitions XCMSnExp object. Parameter features allow define features keep. can logical, integer (index features keep) character (feature IDs) vector. filterFile: allows reduce XCMSnExp data selected files. Identified chromatographic peaks files retained correspondence results (feature definitions) removed default. force keeping feature definitions use keepFeatures = TRUE. Adjusted retention times (present) retained default present. Use keepAdjustedRtime = FALSE drop . filterMsLevel: reduces XCMSnExp object spectra specified MS level(s). Chromatographic peaks identified features also subsetted respective MS level. See also filterMsLevel documentation MSnbase details examples. filterMz: filters data set based provided m/z value range. chromatographic peaks features (grouped peaks) apex falling within provided mz value range retained (.e. chromPeaks(object)[, \"mz\"] >= mz[1] <= mz[2]). Adjusted retention times, present, kept. filterRt: filters data set based provided retention time range. chromatographic peaks features (grouped peaks) within specified retention time window retained (.e. retention time corresponding peak's apex within specified rt range). retention time correction performed, method default filter object adjusted retention times. argument adjusted allows specify manually whether filtering performed raw adjusted retention times. Filtering retention time drop preprocessing results remove change alignment results (.e. adjusted retention times). method returns empty object spectrum feature within specified retention time range. split: splits XCMSnExp object list XCMSnExp objects based provided parameter f. Note default pre-processing results removed splitting, except adjusted retention times, optional argument keepAdjustedRtime = TRUE provided.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"","code":"# S4 method for XCMSnExp,ANY,ANY,ANY [(x, i, j, ..., drop = TRUE)  # S4 method for XCMSnExp,ANY,ANY [[(x, i, j, drop = FALSE)  # S4 method for XCMSnExp filterMsLevel(object, msLevel., keepAdjustedRtime = hasAdjustedRtime(object))  # S4 method for XCMSnExp filterFile(   object,   file,   keepAdjustedRtime = hasAdjustedRtime(object),   keepFeatures = FALSE )  # S4 method for XCMSnExp filterMz(object, mz, msLevel., ...)  # S4 method for XCMSnExp filterRt(object, rt, msLevel., adjusted = hasAdjustedRtime(object))  # S4 method for XCMSnExp,ANY split(x, f, drop = FALSE, ...)  # S4 method for XCMSnExp filterChromPeaks(   object,   keep = rep(TRUE, nrow(chromPeaks(object))),   method = \"keep\",   ... )  # S4 method for XCMSnExp filterFeatureDefinitions(object, features = integer())"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"x [ [[: XCMSnExp object. [: numeric logical vector specifying spectra data set reduced. [[: single integer character. j [ [[: supported. ... Optional additional arguments. drop [ [[: supported. object XCMSnExp object. msLevel. filterMz, filterRt: numeric defining MS level(s) operations applied object subsetted. keepAdjustedRtime filterFile, filterMsLevel, [, split: logical(1) defining whether adjusted retention times kept, even e.g. features removed (retention time correction performed features). file filterFile: integer defining file index within object subset object file character specifying file names sub set. indices expected increasingly ordered, ordered internally. keepFeatures filterFile: logical(1) whether correspondence results (feature definitions) kept dropped. Defaults keepFeatures = FALSE hence feature definitions removed returned object default. mz filterMz: numeric(2) defining lower upper mz value filtering. rt filterRt: numeric(2) defining retention time window (lower upper bound) filtering. adjusted filterRt: logical indicating whether object filtered original (adjusted = FALSE) adjusted retention times (adjusted = TRUE). spectra: whether retention times individual Spectrum objects adjusted raw retention times. f split vector length equal length x defining x splitted. converted internally factor. keep filterChromPeaks: logical, integer character defining chromatographic peaks retained. method filterChromPeaks: character(1) allowing specify method chromatographic peaks filtered. Currently method = \"keep\" supported (.e. specify parameter keep chromatographic peaks retained). features filterFeatureDefinitions: either integer specifying indices features (rows) keep, logical length matching number rows featureDefinitions character feature (row) names.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"methods return XCMSnExp object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"subsetting methods try ensure returned data consistent. Correspondence results example removed default data set sub-setted file, since correspondence results dependent files correspondence performed. can changed setting keepFeatures = TRUE. adjusted retention times, subsetting methods support argument keepAdjustedRtime (even [ method) forces adjusted retention times retained even default drop .","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"filterFile method removes also process history steps related files object sub-setted updates fileIndex attribute accordingly. Also, method allow arbitrary ordering files re-ordering files within object. Note also filtering methods, also subsetting operations [ drop selected preprocessing results. consolidate alignment results, .e. ensure adjusted retention times always preserved, use applyAdjustedRtime() function object contains alignment results. replaces raw retention times adjusted ones.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-filter-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"XCMSnExp filtering and subsetting — [,XCMSnExp,ANY,ANY,ANY-method","text":"","code":"## Loading a test data set with identified chromatographic peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Subset the dataset to the first and third file. xod_sub <- filterFile(faahko_sub, file = c(1, 3))  ## The number of chromatographic peaks per file for the full object table(chromPeaks(faahko_sub)[, \"sample\"]) #>  #>   1   2   3  #>  87 100  61   ## The number of chromatographic peaks per file for the subset table(chromPeaks(xod_sub)[, \"sample\"]) #>  #>  1  2  #> 87 61   basename(fileNames(faahko_sub)) #> [1] \"ko15.CDF\" \"ko16.CDF\" \"ko18.CDF\" basename(fileNames(xod_sub)) #> [1] \"ko15.CDF\" \"ko18.CDF\"  ## Filter on mz values; chromatographic peaks and features within the ## mz range are retained (as well as adjusted retention times). xod_sub <- filterMz(faahko_sub, mz = c(300, 400)) head(chromPeaks(xod_sub)) #>        mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP002 302   302   302 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344   344   344 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP005 366   366   366 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343   343   343 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #> CP007 365   365   365 2679.783 2634.399 2717.342 14975760.8 14525699.1 357632 #> CP009 313   313   313 2783.070 2768.985 2806.544  1744615.5  1698756.0  74080 #>        sn sample #> CP002  46      1 #> CP003  68      1 #> CP005  49      1 #> CP006  87      1 #> CP007 111      1 #> CP009  44      1 nrow(chromPeaks(xod_sub)) #> [1] 89 nrow(chromPeaks(faahko_sub)) #> [1] 248  ## Filter on rt values. All chromatographic peaks and features within the ## retention time range are retained. Filtering is performed by default on ## adjusted retention times, if present. xod_sub <- filterRt(faahko_sub, rt = c(2700, 2900))  range(rtime(xod_sub)) #> [1] 2700.127 2898.877 head(chromPeaks(xod_sub)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP008 579.1 579.1 579.1 2786.200 2765.855 2806.544  1669436.8  1642951.6  87656 #> CP009 313.0 313.0 313.0 2783.070 2768.985 2806.544  1744615.5  1698756.0  74080 #> CP010 301.0 301.0 301.0 2786.200 2765.855 2809.674  3051847.8  2958809.7 118512 #> CP011 279.0 279.0 279.0 2787.765 2764.290 2814.369 17140627.0 16526792.4 805248 #> CP012 279.1 279.1 279.1 2825.323 2820.629 2828.453   106769.4   106763.2  17760 #> CP018 453.2 453.2 453.2 2720.472 2678.218 2786.200  2987076.3  2986966.8  34208 #>          sn sample #> CP008    57      1 #> CP009    44      1 #> CP010    62      1 #> CP011    45      1 #> CP012 17759      1 #> CP018 33983      1 range(chromPeaks(xod_sub)[, \"rt\"]) #> [1] 2701.693 2895.746  nrow(chromPeaks(faahko_sub)) #> [1] 248 nrow(chromPeaks(xod_sub)) #> [1] 21  ## Extract a single Spectrum faahko_sub[[4]] #> Object of class \"Spectrum1\" #>  Retention time: 41:46  #>  MSn level: 1  #>  Total ion count: 427  #>  Polarity: -1   ## Subsetting using [ removes all preprocessing results - using ## keepAdjustedRtime = TRUE would keep adjusted retention times, if present. xod_sub <- faahko_sub[fromFile(faahko_sub) == 1] #> Warning: Removed preprocessing results xod_sub #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 1 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko15.CDF  #> protocolData: none #> featureData #>   featureNames: F1.S0001 F1.S0002 ... F1.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - -  ## Using split does also remove preprocessing results, but it supports the ## optional parameter keepAdjustedRtime. ## Split the object into a list of XCMSnExp objects, one per file xod_list <- split(faahko_sub, f = fromFile(faahko_sub)) xod_list #> $`1` #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 1 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko15.CDF  #> protocolData: none #> featureData #>   featureNames: F1.S0001 F1.S0002 ... F1.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #>  #> $`2` #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 2 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko16.CDF  #> protocolData: none #> featureData #>   featureNames: F2.S0001 F2.S0002 ... F2.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #>  #> $`3` #> MSn experiment data (\"XCMSnExp\") #> Object size in memory: 0.48 Mb #> - - - Spectra data - - - #>  MS level(s): 1  #>  Number of spectra: 1278  #>  MSn retention times: 41:41 - 74:60 minutes #> - - - Processing information - - - #> Data loaded [Tue Aug  1 15:54:19 2023]  #> Filter: select MS level(s) 1. [Tue Aug  1 15:54:19 2023]  #>  MSnbase version: 2.26.0  #> - - - Meta data  - - - #> phenoData #>   rowNames: 3 #>   varLabels: sampleNames #>   varMetadata: labelDescription #> Loaded from: #>   ko18.CDF  #> protocolData: none #> featureData #>   featureNames: F3.S0001 F3.S0002 ... F3.S1278 (1278 total) #>   fvarLabels: fileIdx spIdx ... spectrum (35 total) #>   fvarMetadata: labelDescription #> experimentData: use 'experimentData(object)' #> - - - xcms preprocessing - - - #>"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"methods listed page XCMSnExp methods inherited parent, OnDiskMSnExp class MSnbase package, alter raw data related data subsetting. Thus calling methods causes xcms pre-processing results removed XCMSnExp object ensure data integrity. bin: allows bin spectra. See bin documentation MSnbase package details examples. clean: removes unused 0 intensity data points. See clean documentation MSnbase package details examples. filterAcquisitionNum: filters XCMSnExp object keeping spectra provided acquisition numbers. See filterAcquisitionNum details examples. normalize method performs basic normalization spectra intensities. See normalize documentation MSnbase package details examples. pickPeaks method performs peak picking. See pickPeaks documentation details examples. removePeaks method removes mass peaks (intensities) lower threshold. Note peaks refer mass peaks, different chromatographic peaks detected analyzed metabolomics experiment! See removePeaks documentation details examples. smooth method smooths spectra. See smooth documentation MSnbase details examples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp bin(x, binSize = 1L, msLevel.)  # S4 method for XCMSnExp clean(object, all = FALSE, verbose = FALSE, msLevel.)  # S4 method for XCMSnExp filterAcquisitionNum(object, n, file)  # S4 method for XCMSnExp normalize(object, method = c(\"max\", \"sum\"), ...)  # S4 method for XCMSnExp pickPeaks(   object,   halfWindowSize = 3L,   method = c(\"MAD\", \"SuperSmoother\"),   SNR = 0L,   ... )  # S4 method for XCMSnExp removePeaks(object, t = \"min\", verbose = FALSE, msLevel.)  # S4 method for XCMSnExp smooth(   x,   method = c(\"SavitzkyGolay\", \"MovingAverage\"),   halfWindowSize = 2L,   verbose = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"x XCMSnExp OnDiskMSnExp object. binSize numeric(1) defining size bin (Dalton). msLevel. bin, clean, filterMsLevel, removePeaks: numeric(1) defining MS level(s) operations applied object subsetted. object XCMSnExp OnDiskMSnExp object. clean: logical(1), TRUE zeros removed. verbose logical(1) whether progress information displayed. n filterAcquisitionNum: integer defining acquisition numbers spectra data set sub-setted. file filterAcquisitionNum: integer defining file index within object subset object file. method normalize: character(1) specifying normalization method. See normalize MSnbase package details. pickPeaks: character(1) defining method. See pickPeaks options. smooth: character(1) defining method. See smooth MSnbase package options details. ... Optional additional arguments. halfWindowSize pickPeaks smooth: integer(1) defining window size peak picking. See pickPeaks smooth MSnbase package details options. SNR pickPeaks: numeric(1) defining signal noise ratio considered. See pickPeaks documentation details. t removePeaks: either numeric(1) \"min\" defining threshold (method) used. See removePeaks details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"methods: XCMSnExp object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-inherited-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"XCMSnExp data manipulation methods inherited from MSnbase — bin,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":null,"dir":"Reference","previous_headings":"","what":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"featureValues,XCMSnExp : extract matrix feature values rows representing features columns samples. Parameter value allows define column chromPeaks matrix returned. Multiple chromatographic peaks sample can assigned feature. Parameter method allows specify method used cases chose peaks value returned. Parameter `msLevel` allows choose specific MS level feature values returned (given features defined MS level). quantify,XCMSnExp: return preprocessing results SummarizedExperiment object containing feature abundances assay matrix, feature definitions (returned featureDefinitions) rowData phenotype information colData. ideal container processing data. Internally, featureValues method used extract feature abundances, parameters method can passed quantify ....","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp quantify(object, ...)  # S4 method for XCMSnExp featureValues(   object,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   filled = TRUE,   missing = NA,   msLevel = integer() )"},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"object XCMSnExp object providing feature definitions. ... quantify: additional parameters passed featureValues method. method character specifying method resolve multi-peak mappings within sample, .e. define representative peak feature samples one peak assigned feature. \"medret\": select peak closest median retention time feature. \"maxint\": select peak yielding largest signal. \"sum\": sum values (value \"\" \"maxo\". value character specifying name column chromPeaks(object) returned. Defaults \"\" case integrated peak area returned. get index peak chromPeaks(object) matrix use \"index\". intensity character specifying name column chromPeaks(objects) matrix containing intensity value peak used conflict resolution method = \"maxint\". filled logical(1) specifying whether values filled-peaks returned . filled = FALSE, NA returned matrix respective peak. See fillChromPeaks details peak filling. missing missing values reported. Allowed values NA (default), numeric missing = \"rowmin_half\". latter replaces NA half row's minimal (non-missing) value. msLevel `featureValues`: `integer` defining MS level(s) feature values returned. default, values features defined MS levels returned.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"featureValues: matrix feature values, columns representing samples, rows features. order features matches order found featureDefinitions(object)  DataFrame. rownames matrix featureDefinitions DataFrame. NA reported features without corresponding chromatographic peak respective sample(s). quantify: SummarizedExperiment representing preprocessing results.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"method equivalent groupval xcmsSet objects. Note missing = 0 used get behaviour groupval, .e. report missing values 0 call fillPeaks.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XCMSnExp-peak-grouping-results.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Accessing mz-rt feature data values — quantify,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Containers for chromatographic and peak detection data — XChromatograms","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"XChromatogram object allows store chromatographic data (e.g. extracted ion chromatogram) along identified chromatographic peaks within data. object inherits functions Chromatogram() object MSnbase package. Multiple XChromatogram objects can stored XChromatograms object. class extends MChromatograms() MSnbase package allows thus arrange chromatograms matrix-like structure, columns representing samples rows m/z-retention time ranges. functions described (grouped topic-related sections) Arguments section.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"","code":"XChromatograms(data, phenoData, featureData, chromPeaks, chromPeakData, ...)  XChromatogram(   rtime = numeric(),   intensity = numeric(),   mz = c(NA_real_, NA_real_),   filterMz = c(NA_real_, NA_real_),   precursorMz = c(NA_real_, NA_real_),   productMz = c(NA_real_, NA_real_),   fromFile = integer(),   aggregationFun = character(),   msLevel = 1L,   chromPeaks,   chromPeakData )  # S4 method for XChromatogram show(object)  # S4 method for XChromatogram chromPeaks(   object,   rt = numeric(),   mz = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel )  # S4 method for XChromatogram chromPeaks(object) <- value  # S4 method for XChromatogram,ANY plot(   x,   col = \"#00000060\",   lty = 1,   type = \"l\",   xlab = \"retention time\",   ylab = \"intensity\",   main = NULL,   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakCol = \"#00000060\",   peakBg = \"#00000020\",   peakPch = 1,   ... )  # S4 method for XChromatogram filterMz(object, mz, ...)  # S4 method for XChromatogram filterRt(object, rt, ...)  # S4 method for XChromatogram hasChromPeaks(object)  # S4 method for XChromatogram dropFilledChromPeaks(object)  # S4 method for XChromatogram chromPeakData(object)  # S4 method for XChromatogram chromPeakData(object) <- value  # S4 method for XChromatogram,MergeNeighboringPeaksParam refineChromPeaks(object, param = MergeNeighboringPeaksParam())  # S4 method for XChromatogram filterChromPeaks(object, method = c(\"keepTop\"), ...)  # S4 method for XChromatogram transformIntensity(object, FUN = identity)  # S4 method for XChromatograms show(object)  # S4 method for XChromatograms hasChromPeaks(object)  # S4 method for XChromatograms hasFilledChromPeaks(object)  # S4 method for XChromatograms chromPeaks(   object,   rt = numeric(),   mz = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel )  # S4 method for XChromatograms chromPeakData(object)  # S4 method for XChromatograms filterMz(object, mz, ...)  # S4 method for XChromatograms filterRt(object, rt, ...)  # S4 method for XChromatograms,ANY plot(   x,   col = \"#00000060\",   lty = 1,   type = \"l\",   xlab = \"retention time\",   ylab = \"intensity\",   main = NULL,   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakCol = \"#00000060\",   peakBg = \"#00000020\",   peakPch = 1,   ... )  # S4 method for XChromatograms processHistory(object, fileIndex, type)  # S4 method for XChromatograms hasFeatures(object, ...)  # S4 method for XChromatograms dropFeatureDefinitions(object, ...)  # S4 method for XChromatograms,PeakDensityParam groupChromPeaks(object, param)  # S4 method for XChromatograms featureDefinitions(   object,   mz = numeric(),   rt = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\") )  # S4 method for XChromatograms,ANY,ANY,ANY [(x, i, j, drop = TRUE)  # S4 method for XChromatograms featureValues(   object,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   missing = NA,   ... )  # S4 method for XChromatograms plotChromPeakDensity(   object,   param,   col = \"#00000060\",   xlab = \"retention time\",   main = NULL,   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakCol = \"#00000060\",   peakBg = \"#00000020\",   peakPch = 1,   simulate = TRUE,   ... )  # S4 method for XChromatograms dropFilledChromPeaks(object)  # S4 method for XChromatograms,MergeNeighboringPeaksParam refineChromPeaks(object, param = MergeNeighboringPeaksParam())  # S4 method for XChromatograms filterChromPeaks(object, method = c(\"keepTop\"), ...)  # S4 method for XChromatograms transformIntensity(object, FUN = identity)"},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"data XChromatograms: list Chromatogram XChromatogram objects. phenoData XChromatograms: either data.frame, AnnotatedDataFrame NAnnotatedDataFrame describing phenotypical information samples. featureData XChromatograms: either data.frame AnnotatedDataFrame additional information row chromatograms. chromPeaks XChromatogram: matrix required columns \"rt\", \"rtmin\", \"rtmax\", \"\", \"maxo\" \"sn\". XChromatograms: list, length data, chromatographic peaks chromatogram. element matrix, ordering match order chromatograms data. chromPeakData XChromatogram: DataFrame optional additional annotations chromatographic peak. number rows match number chromatographic peaks. ... filterChromPeaks: additional parameters defining filter chromatographic peaks. See function description details. rtime XChromatogram: numeric retention times (length equal length intensity). intensity XChromatogram: numeric intensity values (length equal length rtime).   mz XChromatogram: numeric(2) representing m/z value range (min, max) chromatogram created. supposed contain real range m/z values contrast filterMz . chromPeaks featureDefinitions: numeric(2) defining m/z range chromatographic peaks features returned. filterMz: numeric(2) defining m/z range chromatographic peaks retained.#' filterMz XChromatogram: numeric(2) representing m/z value range (min, max) used filter original object m/z dimension. applicable use filterMz = c(0, 0). precursorMz XChromatogram: numeric(2) SRM/MRM transitions. Represents mz precursor ion. See details information. productMz XChromatogram: numeric(2) SRM/MRM transitions. Represents mz product. See details information. fromFile XChromatogram: integer(1) index file within OnDiskMSnExp MSnExp object chromatogram extracted. aggregationFun XChromatogram: character(1) specifying function used aggregate intensity values retention time across m/z range. msLevel XChromatogram: integer MS level chromatogram extracted. chromPeaks chromPeakData: extract chromatographic peaks certain MS level. object XChromatogram XChromatograms object. rt chromPeaks featureDefinitions: numeric(2) defining retention time range chromatographic peaks features returned. filterRt: numeric(2) defining retention time range reduce object . ppm chromPeaks featureDefinitions: numeric(1) defining ppm expand provided m/z range. type chromPeaks featureDefinitions: character(1) defining peaks features return rt mz provided: \"\" (default) return peaks even partially overlapping rt, \"within\" return peaks completely within rt \"apex_within\" return peaks apex within rt.   value chromPeaks<-: numeric matrix required columns \"rt\", \"rtmin\", \"rtmax\", \"\" \"maxo\".   x plot: XChromatogram XChromatograms object. col plot: color used draw chromatogram. lty plot plotChromPeakDensity: line type. xlab plot plotChromPeakDensity: x axis label. ylab plot: y axis label. main plot plotChromPeakDensity: optional title plot. peakType plot plotChromPeakDensity: character(1) defining () identified chromatographic peak within chromatogram plotted. Options \"polygon\" (default): draw peak borders peakCol color fill peak area peakBg color, \"point\": indicate peak's apex point, \"rectangle\": draw rectangle around identified peak \"none\": draw peaks. peakCol plot plotChromPeakDensity: foreground color peaks. peakType = \"polygon\" peakType = \"rectangle\" color border. Use NA use foreground color. either single color vector colors length chromPeaks(x) rows. peakBg plot plotChromPeakDensity: background color peaks. peakType = \"polygon\" peakType = \"rectangle\" peak rectangle filled color. Use NA skip. either single color vector colors length chromPeaks(x) rows. peakPch plot plotChromPeakDensity: point character used peakType = \"point\". See plot() graphics package details. param groupChromPeaks plotChromPeakDensity: PeakDensityParam() object settings peak density correspondence analysis algorithm. method featureValues: character(1) specifying method resolve multi-peak mappings within sample sample, .e. select representative peak feature one peak assigned one sample. Options \"medret\" (default): select peak closest median retention time feature, \"maxint\": select peak largest signal \"sum\": sum values peaks (value \"\" \"maxo\"). filterChromPeaks: character(1) defining method used filter chromatographic peaks. See help filterChromPeaks details. FUN transformIntensity: function transform intensity values object. fileIndex processHistory: optional integer specifying index files/samples ProcessHistory objects returned. [: integer row indices subset XChromatograms object. j [: integer column indices subset XChromatograms object. drop [: logical(1) whether dimensionality dropped (possible). Defaults drop = TRUE, thus, length j 1 XChromatogram returned. Note drop ignored length j larger 1, thus XChromatograms returned. missing featureValues: missing values reported. Allowed values NA (default), numeric(1) replace NAs value missing = \"rowmin_half\" replace NAs half row's minimal (non-missing) value. simulate plotChromPeakDensity: logical(1) whether correspondence analysis simulated based available data provided PeakDensityParam() param argument. See section Correspondence analysis details.","code":"For `featureValues`: `character(1)` specifying the name of the column in `chromPeaks(object)` containing the intensity value of the peak that should be used for the `method = \"maxint\"` conflict resolution if. For `plot`: what type of plot should be used for the chromatogram (such as `\"l\"` for lines, `\"p\"` for points etc), see help of [plot()] in the `graphics` package for more details. For `processHistory`: restrict returned processing steps to specific types. Use [processHistoryTypes()] to list all supported values. For `featureValues`: `character(1)` specifying the name of the column in `chromPeaks(object)` that should be returned or `\"index\"` (default) to return the index of the peak associated with the feature in each sample. To return the integrated peak area instead of the index use `value = \"into\"`."},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"See help individual functions.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Highlighting peak area(s) XChromatogram XChromatograms object (plot peakType = \"polygon\") draws polygon representing displayed chromatogram peak's minimal retention time maximal retention time. XChromatograms extracted XCMSnExp() object chromatogram() function might represent actual identified peak area m/z range used extract chromatogram larger peak's m/z.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"creation-of-objects","dir":"Reference","previous_headings":"","what":"Creation of objects","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Objects can created contructor function XChromatogram XChromatograms, respectively. Also, can coerced Chromatogram MChromatograms() objects using (object, \"XChromatogram\") (object, \"XChromatograms\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"filtering-and-subsetting","dir":"Reference","previous_headings":"","what":"Filtering and subsetting","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Besides classical subsetting [ specific filter operations MChromatograms() XChromatograms objects available. See filterColumnsIntensityAbove() details. [ allows subset XChromatograms object row () column (j), j type integer. featureDefinitions also subsetted accordingly peakidx column updated. filterMz filters chromatographic peaks within XChromatogram XChromatograms, column \"mz\" present chromPeaks matrix. case XChromatogram extracted XCMSnExp() object chromatogram() function. chromatographic peaks m/z within m/z range defined mz retained. Also feature definitions (present) subset accordingly. function returns filtered XChromatogram XChromatograms object. filterRt filters chromatogram(s) provided retention time range. eventually present chromatographic peaks apex within retention time range specified rt retained. Also feature definitions, present, filtered accordingly. function returns filtered XChromatogram XChromatograms object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"accessing-data","dir":"Reference","previous_headings":"","what":"Accessing data","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"See also help Chromatogram MSnbase package general information data access. methods listed specific XChromatogram XChromatograms objects. chromPeaks, chromPeaks<-: extract set matrix chromatographic peak definitions. Parameter rt allows specify retention time range peaks returned along parameter type defines overlapping defined (parameter description details). XChromatogram objects function returns matrix columns \"rt\" (retention time peak apex), \"rtmin\" (lower peak boundary), \"rtmax\" (upper peak boundary), \"\" (ingegrated peak signal/area peak), \"maxo\" (maximum instensity peak \"sn\" (signal noise ratio). Note , depending peak detection algorithm, matrix may contain additional columns. XChromatograms objects matrix contains also columns \"row\" \"column\" specifying chromatogram object peak identified. Chromatographic peaks ordered row. chromPeakData, chromPeakData<-: extract set DataFrame() optional chromatographic peak annotations. hasChromPeaks: infer whether XChromatogram (XChromatograms) chromatographic peaks. XChromatogram: returns logical(1), XChromatograms: returns matrix, dimensions object either TRUE FALSE chromatographic peaks available chromatogram respective position. hasFilledChromPeaks: whether XChromatogram (XChromatogram XChromatograms) filled-chromatographic peaks. XChromatogram: returns logical(1), XChromatograms: returns matrix, dimensions object either TRUE FALSE chromatographic peaks available chromatogram respective position. dropFilledChromPeaks: removes filled-chromatographic peaks. See dropFilledChromPeaks() help XCMSnExp() objects information. hasFeatures: XChromatograms objects : correspondence analysis performed m/z-rt feature definitions present. Returns logical(1). dropFeatureDefinitions: XChrmomatograms objects : delete correspondence analysis results (related process history). featureDefinitions: XChromatograms objects . Extract results correspondence analysis (performed groupChromPeaks). Returns DataFrame properties defined m/z-rt features: m/z retention time range. Columns peakidx row contain index chromatographic peaks chromPeaks matrix associated feature row XChromatograms object feature defined. Similar chromPeaks method possible filter returned feature matrix mz, rt ppm parameters. featureValues: XChromatograms objects . Extract abundance estimates individuals features. Note default (parameter value = \"index\" matrix indices peaks chromPeaks matrix associated feature returned. extract integrated peak area use value = \"\". function returns matrix one row per feature (featureDefinitions) column sample (.e. column object). features without peak associated certain sample NA returned. can changed missing argument function. filterChromPeaks: filters chromatographic peaks object depending parameter method method-specific parameters passed additional arguments .... Available methods : method = \"keepTop\": keep top n (default n = 1L) peaks chromatogram ordered column order (defaults order = \"maxo\"). Parameter decreasing (default decreasing = TRUE) can used order peaks descending (decreasing = TRUE) ascending ( decreasing = FALSE) order keep top n peaks largest smallest values, respectively. processHistory: returns list ProcessHistory objects representing individual performed processing steps. Optional parameters type fileIndex allow specify processing steps return.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"manipulating-data","dir":"Reference","previous_headings":"","what":"Manipulating data","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"transformIntensity: transforms intensity values chromatograms provided function FUN. See transformIntensity() MSnbase package details. XChromatogram XChromatograms addition intensity values also columns \"\" \"maxo\" object's chromPeaks matrix transformed function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"plotting-and-visualizing","dir":"Reference","previous_headings":"","what":"Plotting and visualizing","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"plot draws chromatogram highlights addition chromatographic peaks present XChromatogram XChromatograms (unless peakType = \"none\" specified). draw peaks different colors vector color definitions length equal nrow(chromPeaks(x)) submitted  peakCol /peakBg defining one color peak (order peaks chromPeaks(x)). base peak chromatograms total ion chromatograms might better set peakType = \"none\" avoid generating busy plots. plotChromPeakDensity: visualize peak density-based correspondence analysis results. See section Correspondence analysis details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"chromatographic-peak-detection","dir":"Reference","previous_headings":"","what":"Chromatographic peak detection","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"See findChromPeaks-Chromatogram-CentWaveParam information. chromatographic peak detection also possible refine identified chromatographic peaks refineChromPeaks method (e.g. reduce peak detection artifacts). Currently, peak refinement using merge neighboring peaks method available (see MergeNeighboringPeaksParam() detailed description approach.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"correspondence-analysis","dir":"Reference","previous_headings":"","what":"Correspondence analysis","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Identified chromatographic peaks XChromatograms object can grouped features groupChromPeaks function. Currently, correspondence analysis can performed peak density method (see groupChromPeaks details) specifying algorithm settings PeakDensityParam() object. correspondence analysis performed separately row XChromatograms object grouping chromatographic peaks across samples (columns). analysis results stored returned XChromatograms object can accessed featureDefinitions method returns DataFrame one row feature. Column \"row\" specifies row XChromatograms object feature identified. plotChromPeakDensity method can used visualize peak density correspondence results, simulate peak density correspondence analysis chromatographic data. resulting plot consists two panels, upper panel showing chromatographic data well identified chromatographic peaks, lower panel distribution peaks (peak density) along retention time axis. plot shows peak point peak's retention time x-axis, sample found y-axis. distribution peaks along retention time axis visualized density estimate. Grouped chromatographic peaks indicated grey shaded rectangles. Parameter simulate allows define whether correspondence analysis simulated ( simulate=TRUE, based available data provided PeakDensityParam() parameter class) (simulate=FALSE). latter assumed correspondence analysis performed peak density method object. See examples . Abundance estimates feature can extracted featureValues function using parameter value = \"\" extract integrated peak area feature. result matrix, columns samples rows features.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XChromatogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Containers for chromatographic and peak detection data — XChromatograms","text":"","code":"## ---- Creation of XChromatograms ---- ## ## Create a XChromatograms from Chromatogram objects dta <- list(Chromatogram(rtime = 1:7, c(3, 4, 6, 12, 8, 3, 2)),     Chromatogram(1:10, c(4, 6, 3, 4, 7, 13, 43, 34, 23, 9)))  ## Create an XChromatograms without peak data xchrs <- XChromatograms(dta)  ## Create an XChromatograms with peaks data pks <- list(matrix(c(4, 2, 5, 30, 12, NA), nrow = 1,     dimnames = list(NULL, c(\"rt\", \"rtmin\", \"rtmax\", \"into\", \"maxo\", \"sn\"))),     NULL) xchrs <- XChromatograms(dta, chromPeaks = pks)  ## Create an XChromatograms from XChromatogram objects dta <- lapply(dta, as, \"XChromatogram\") chromPeaks(dta[[1]]) <- pks[[1]]  xchrs <- XChromatograms(dta, nrow = 1)  hasChromPeaks(xchrs) #>      1     2 #> 1 TRUE FALSE  ## Loading a test data set with identified chromatographic peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Subset the dataset to the first and third file. xod_sub <- filterFile(faahko_sub, file = c(1, 3))  od <- as(xod_sub, \"MsExperiment\")  ## Extract chromatograms for a m/z - retention time slice chrs <- chromatogram(od, mz = 344, rt = c(2500, 3500)) chrs #> MChromatograms with 1 row and 2 columns #>                   1              2 #>      <Chromatogram> <Chromatogram> #> [1,]    length: 639    length: 639 #> phenoData with 2 variables #> featureData with 4 variables  ## --------------------------------------------------- ## ##       Chromatographic peak detection                ## ## --------------------------------------------------- ## ## Perform peak detection using CentWave xchrs <- findChromPeaks(chrs, param = CentWaveParam()) xchrs #> XChromatograms with 1 row and 2 columns #>                    1               2 #>      <XChromatogram> <XChromatogram> #> [1,]        peaks: 2        peaks: 1 #> phenoData with 2 variables #> featureData with 4 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Do we have chromatographic peaks? hasChromPeaks(xchrs) #>      1    2 #> 1 TRUE TRUE  ## Process history processHistory(xchrs) #> [[1]] #> Object of class \"XProcessHistory\" #>  type: Peak detection  #>  date: Fri Jan 26 20:05:24 2024  #>  info:   #>  fileIndex: 1,2  #>  Parameter class: CentWaveParam  #>   ## The chromatographic peaks, columns \"row\" and \"column\" provide information ## in which sample the peak was identified. chromPeaks(xchrs) #>            rt    rtmin    rtmax      into      intb   maxo sn row column #> [1,] 2589.015 2571.801 2612.490  592297.3  554533.9  27600 14   1      1 #> [2,] 2679.783 2646.919 2709.517 5210015.9 5071346.6 152320 30   1      1 #> [3,] 2682.914 2643.790 2731.427 5255689.5 4983388.9 125632 28   1      2  ## Spectifically extract chromatographic peaks for one sample/chromatogram chromPeaks(xchrs[1, 2]) #>            rt   rtmin    rtmax    into    intb   maxo sn #> [1,] 2682.914 2643.79 2731.427 5255689 4983389 125632 28  ## Plot the results plot(xchrs)   ## Plot the results using a different color for each sample sample_colors <- c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\") cols <- sample_colors[chromPeaks(xchrs)[, \"column\"]] plot(xchrs, col = sample_colors, peakBg = cols)   ## Indicate the peaks with a rectangle plot(xchrs, col = sample_colors, peakCol = cols, peakType = \"rectangle\",     peakBg = NA)   ## --------------------------------------------------- ## ##       Correspondence analysis                       ## ## --------------------------------------------------- ## ## Group chromatographic peaks across samples prm <- PeakDensityParam(sampleGroup = rep(1, 2)) res <- groupChromPeaks(xchrs, param = prm)  hasFeatures(res) #> [1] TRUE featureDefinitions(res) #> DataFrame with 2 rows and 10 columns #>         mzmed     mzmin     mzmax     rtmed     rtmin     rtmax    npeaks #>     <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> <numeric> #> FT1        NA        NA        NA   2681.35   2679.78   2682.91         2 #> FT2        NA        NA        NA   2589.01   2589.01   2589.01         1 #>            X1 peakidx       row #>     <numeric>  <list> <integer> #> FT1         2     2,3         1 #> FT2         1       1         1  ## Plot the correspondence results. Use simulate = FALSE to show the ## actual results. Grouped chromatographic peaks are indicated with ## grey shaded rectangles. plotChromPeakDensity(res, simulate = FALSE)   ## Simulate a correspondence analysis based on different settings. Larger ## bw will increase the smoothing of the density estimate hence grouping ## chromatographic peaks that are more apart on the retention time axis. prm <- PeakDensityParam(sampleGroup = rep(1, 3), bw = 60) plotChromPeakDensity(res, param = prm)   ## Delete the identified feature definitions res <- dropFeatureDefinitions(res) hasFeatures(res) #> [1] FALSE  ## Create a XChromatogram object pks <- matrix(nrow = 1, ncol = 6) colnames(pks) <- c(\"rt\", \"rtmin\", \"rtmax\", \"into\", \"maxo\", \"sn\") pks[, \"rtmin\"] <- 2 pks[, \"rtmax\"] <- 9 pks[, \"rt\"] <- 4 pks[, \"maxo\"] <- 19 pks[, \"into\"] <- 93  xchr <- XChromatogram(rtime = 1:10,     intensity = c(4, 8, 14, 19, 18, 12, 9, 8, 5, 2),     chromPeaks = pks) xchr #> Object of class: XChromatogram #> length of object: 10 #> from file:  #> mz range: [NA, NA] #> rt range: [1, 10] #> MS level: 1 #> Identified chromatographic peaks (1): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn  #>  4\t2\t9\t93\t19\tNA  ## Add arbitrary peak annotations df <- DataFrame(peak_id = c(\"a\")) xchr <- XChromatogram(rtime = 1:10,     intensity = c(4, 8, 14, 19, 18, 12, 9, 8, 5, 2),     chromPeaks = pks, chromPeakData = df) xchr #> Object of class: XChromatogram #> length of object: 10 #> from file:  #> mz range: [NA, NA] #> rt range: [1, 10] #> MS level: 1 #> Identified chromatographic peaks (1): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn  #>  4\t2\t9\t93\t19\tNA chromPeakData(xchr) #> DataFrame with 1 row and 3 columns #>       peak_id  ms_level is_filled #>   <character> <integer> <logical> #> 1           a         1     FALSE  ## Extract the chromatographic peaks chromPeaks(xchr) #>      rt rtmin rtmax into maxo sn #> [1,]  4     2     9   93   19 NA  ## Plotting of a single XChromatogram object ## o Don't highlight chromatographic peaks plot(xchr, peakType = \"none\")  ## o Indicate peaks with a polygon plot(xchr)   ## Add a second peak to the data. pks <- rbind(chromPeaks(xchr), c(7, 7, 10, NA, 15, NA)) chromPeaks(xchr) <- pks  ## Plot the peaks in different colors plot(xchr, peakCol = c(\"#ff000080\", \"#0000ff80\"),     peakBg = c(\"#ff000020\", \"#0000ff20\"))  ## Indicate the peaks as rectangles plot(xchr, peakCol = c(\"#ff000060\", \"#0000ff60\"), peakBg = NA,     peakType = \"rectangle\")   ## Filter the XChromatogram by retention time xchr_sub <- filterRt(xchr, rt = c(4, 6)) xchr_sub #> Object of class: XChromatogram #> length of object: 3 #> from file:  #> mz range: [NA, NA] #> rt range: [4, 6] #> MS level: 1 #> Identified chromatographic peaks (1): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn  #>  4\t2\t9\t93\t19\tNA plot(xchr_sub)"},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":null,"dir":"Reference","previous_headings":"","what":"Next Generation xcms Result Object — filterFeatureDefinitions","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"XcmsExperiment data container xcms preprocessing results (.e. results chromatographic peak detection, alignment correspondence analysis). provides functionality XCMSnExp object, uses advanced modern MS infrastructure provided MsExperiment Spectra Bioconductor packages. comes higher flexibility store data. Documentation various functions XcmsExperiment objects grouped topic provided sections . default xcms workflow perform chromatographic peak detection using findChromPeaks() optionally refine identified chromatographic peaks using refineChromPeaks() perform alignment (retention time adjustment) using adjustRtime(). Depending method used requires run correspondence analysis first perform correspondence analysis using groupChromPeaks() function group chromatographic peaks across samples define LC-MS features. optionally perform gap-filling rescue signal samples chromatographic peak identified hence missing value reported. can performed using fillChromPeaks() function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"","code":"filterFeatureDefinitions(object, ...)  # S4 method for MsExperiment filterRt(object, rt = numeric(), ...)  # S4 method for MsExperiment filterMzRange(object, mz = numeric(), msLevel. = uniqueMsLevels(object))  # S4 method for MsExperiment filterMz(object, mz = numeric(), msLevel. = uniqueMsLevels(object))  # S4 method for MsExperiment filterMsLevel(object, msLevel. = uniqueMsLevels(object))  # S4 method for MsExperiment uniqueMsLevels(object)  # S4 method for MsExperiment filterFile(object, file = integer(), ...)  # S4 method for MsExperiment rtime(object)  # S4 method for MsExperiment fromFile(object)  # S4 method for MsExperiment fileNames(object)  # S4 method for MsExperiment polarity(object)  # S4 method for MsExperiment filterIsolationWindow(object, mz = numeric())  # S4 method for MsExperiment chromatogram(   object,   rt = matrix(nrow = 0, ncol = 2),   mz = matrix(nrow = 0, ncol = 2),   aggregationFun = \"sum\",   msLevel = 1L,   isolationWindowTargetMz = NULL,   chunkSize = 2L,   return.type = \"MChromatograms\",   BPPARAM = bpparam() )  featureArea(   object,   mzmin = min,   mzmax = max,   rtmin = min,   rtmax = max,   features = character() )  # S4 method for MsExperiment,missing plot(x, y, msLevel = 1L, peakCol = \"#ff000060\", ...)  # S4 method for XcmsExperiment,ANY,ANY,ANY [(x, i, j, ..., drop = TRUE)  # S4 method for XcmsExperiment filterIsolationWindow(object, mz = numeric())  # S4 method for XcmsExperiment filterRt(object, rt, msLevel.)  # S4 method for XcmsExperiment filterMzRange(object, mz = numeric(), msLevel. = uniqueMsLevels(object))  # S4 method for XcmsExperiment filterMsLevel(object, msLevel. = uniqueMsLevels(object))  # S4 method for XcmsExperiment hasChromPeaks(object, msLevel = integer())  # S4 method for XcmsExperiment dropChromPeaks(object, keepAdjustedRtime = FALSE)  # S4 method for XcmsExperiment chromPeaks(object) <- value  # S4 method for XcmsExperiment chromPeaks(   object,   rt = numeric(),   mz = numeric(),   ppm = 0,   msLevel = integer(),   type = c(\"any\", \"within\", \"apex_within\"),   isFilledColumn = FALSE )  # S4 method for XcmsExperiment chromPeakData(object) <- value  # S4 method for XcmsExperiment chromPeakData(   object,   msLevel = integer(),   return.type = c(\"DataFrame\", \"data.frame\") )  # S4 method for XcmsExperiment filterChromPeaks(   object,   keep = rep(TRUE, nrow(.chromPeaks(object))),   method = \"keep\",   ... )  # S4 method for XcmsExperiment dropAdjustedRtime(object)  # S4 method for MsExperiment hasAdjustedRtime(object)  # S4 method for XcmsExperiment rtime(object, adjusted = hasAdjustedRtime(object))  # S4 method for XcmsExperiment adjustedRtime(object)  # S4 method for XcmsExperiment hasFeatures(object, msLevel = integer())  # S4 method for XcmsExperiment featureDefinitions(object) <- value  # S4 method for XcmsExperiment featureDefinitions(   object,   mz = numeric(),   rt = numeric(),   ppm = 0,   type = c(\"any\", \"within\", \"apex_within\"),   msLevel = integer() )  # S4 method for XcmsExperiment dropFeatureDefinitions(object, keepAdjustedRtime = FALSE)  # S4 method for XcmsExperiment filterFeatureDefinitions(object, features = integer())  # S4 method for XcmsExperiment hasFilledChromPeaks(object)  # S4 method for XcmsExperiment dropFilledChromPeaks(object)  # S4 method for XcmsExperiment quantify(object, ...)  # S4 method for XcmsExperiment featureValues(   object,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   filled = TRUE,   missing = NA_real_,   msLevel = integer() )  # S4 method for XcmsExperiment chromatogram(   object,   rt = matrix(nrow = 0, ncol = 2),   mz = matrix(nrow = 0, ncol = 2),   aggregationFun = \"sum\",   msLevel = 1L,   chunkSize = 2L,   isolationWindowTargetMz = NULL,   return.type = c(\"XChromatograms\", \"MChromatograms\"),   include = character(),   chromPeaks = c(\"apex_within\", \"any\", \"none\"),   BPPARAM = bpparam() )  # S4 method for XcmsExperiment processHistory(object, type)  # S4 method for XcmsExperiment filterFile(   object,   file,   keepAdjustedRtime = hasAdjustedRtime(object),   keepFeatures = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"object XcmsExperiment object. ... Additional optional parameters. quantify: parameter featureValues call used extract feature value matrix. rt chromPeaks featureDefinitions: numeric(2) defining retention time range chromatographic peaks features returned. full range used default. chromatogram: two column numerical matrix row representing lower upper retention time window(s) chromatograms. provided full retention time range used. mz chromPeaks featureDefinitions: numeric(2) optionally defining m/z range chromatographic peaks feature definitions returned. full m/z range used default. chromatogram: two-column numerical matrix row representing m/z range aggregated chromatogram. provided full m/z range data used (hence total ion chromatogram returned aggregationFun = \"sum\" used). filterIsolationWindow: numeric(1) defining m/z contained within spectra's isolation window. msLevel. filterRt: ignored. filterRt always filter retention times MS levels regardless parameter. chromatogram: integer MS level chromatogram(s) extracted. either length 1 length equal numer rows parameters mz rt defining m/z rt regions chromatograms created. Defaults msLevel = 1L. filterMsLevel: integer defining MS level(s) data subset. file filterFile: integer indices samples (files) data subsetted. aggregationFun chromatogram: character(1) defining function used aggregate intensities retention time (.e. spectrum) along specified m/z range (parameter mz). Defaults aggregationFun = \"sum\" hence intensities summed . Alternatively, use aggregationFun = \"max\" use maximal intensity per m/z range create base peak chromatogram (BPC). msLevel integer defining MS level (multiple MS level function supports ). isolationWindowTargetMz chromatogram: numeric (length equal number rows rt mz) isolation window target m/z MS2 spectra chromatgrom generated. MS1 data (msLevel = 1L, default), parameter ignored. See examples chromatogram information. chunkSize chromatogram: integer(1) defining number files data loaded time memory. Defaults chunkSize = 2L. return.type chromPeakData: character(1) defining class returned object. Can either \"DataFrame\" (default) \"data.frame\". chromatogram: character(1) defining type returned object. Currently return.type = \"MChromatograms\" supported. BPPARAM chromatogram: parallel processing setup. Defaults BPPARAM = bpparam(). See bpparam() information. mzmin featureArea: function calculate \"mzmin\" feature based \"mzmin\" values individual chromatographic peaks assigned feature. Defaults mzmin = min. mzmax featureArea: function calculate \"mzmax\" feature based \"mzmax\" values individual chromatographic peaks assigned feature. Defaults mzmax = max. rtmin featureArea: function calculate \"rtmin\" feature based \"rtmin\" values individual chromatographic peaks assigned feature. Defaults rtmin = min. rtmax featureArea: function calculate \"rtmax\" feature based \"rtmax\" values individual chromatographic peaks assigned feature. Defaults rtmax = max. features filterFeatureDefinitions featureArea: logical, integer character defining features keep extract feature area, respectively. See function description information. x XcmsExperiment object. y plot: defined supported. peakCol plot: defines border color rectangles indicating identified chromatographic peaks. single color supported. Defaults `peakCol = \"#ff000060\". [: integer logical defining samples/files subset. j [: supported. drop [: ignored. keepAdjustedRtime logical(1): whether adjusted retention times (present) retained. value featureValues: character(1) defining value reported feature sample. Can column chromPeaks matrix \"index\" simply index assigned peak returned. Defaults value = \"\" thus integrated peak area reported. ppm chromPeaks featureDefinitions: optional numeric(1) specifying ppm m/z range (defined mz extended. value ppm = 10, peaks within mz[1] - ppm / 1e6 mz[2] + ppm / 1e6 returned. type chromPeaks featureDefinitions either mz rt defined : character(1): defining peaks (features) returned. type = \"\": returns chromatographic peaks features also partially overlapping provided ranges. type = \"within\": returns peaks features completely within region defined mz /rt. type = \"apex_within\": returns peaks features m/z retention time peak's apex within region defined mz /rt. processHistory: restrict returned processing steps specific types. Use processHistoryTypes() list supported values. isFilledColumn chromPeaks: logical(1) whether column \"is_filled\" included returned matrix information whether peak detected filled-. Note information also provided chromPeakData data frame. keep filterChromPeaks: logical, integer character specifying chromatographic peaks keep. logical length keep needs match number rows chromPeaks. Alternatively, keep allows specify index (row) peaks keep ID (.e. row name chromPeaks). method featureValues: character(1) specifying method resolve multi-peak mappings within sample (correspondence analysis can assign one chromatographic peak within sample feature, e.g. close retention time). Options: method = \"medret\": report value chromatographic peak closest feature's median retention time. method = \"maxint\": report value chromatographic peak largest signal (parameter intensity allows select column chromPeaks used signal). method = \"sum\": sum value chromatographic peaks sample assigned feature. default method = \"medret\". filterChromPeaks: currently method = \"keep\" supported. adjusted rtime,XcmsExperiment: whether adjusted raw retention times returned. default return adjusted retention times, available. intensity featureValues: character(1) specifying name column chromPeaks(objects) matrix containing intensity value peak used conflict resolution method = \"maxint\". filled featureValues: logical(1) specifying whether values filled-peaks reported. filled = TRUE (default) filled peak values returned, otherwise NA reported respective features samples peak detected. missing featureValues: default value missing values. Allows define value reported missing peak intensity. Defaults missing = NA_real_. include chromatogram: deprecated; use parameter chromPeaks instead. chromPeaks chromatogram: character(1) defining chromatographic peaks returned. Can either chromPeaks = \"apex_within\" (default) return chromatographic peaks m/z RT apex within m/z retention time window, chromPeaks = \"\" chromatographic peaks overlapping m/z - retention time window chromPeaks = \"none\" include chromatographic peaks. See also parameter type additional information. keepFeatures subsetting functions ([, filterFile): logical(1): wheter eventually present feature definitions retained returned (filtered) object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"subsetting-and-filtering","dir":"Reference","previous_headings":"","what":"Subsetting and filtering","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"[: subset XcmsExperiment sample (parameter ). Subsetting default drop correspondence results (subsetting samples obviously affect feature definition) alignment results (adjusted retention times) identified chromatographic peaks (selected samples) retained. preprocessing results kept dropped can also configured optional parameters keepChromPeaks (default TRUE), keepAdjustedRtime (default FALSE) keepFeatures (default FALSE). filterChromPeaks: filter chromatographic peaks XcmsExperiment keeping specified parameter keep. Returns XcmsExperiment filtered data. Chromatographic peaks retain can specified either providing index chromPeaks matrix, ID (rowname chromPeaks) logical vector length number rows chromPeaks. Assignment chromatographic peaks updated eventually present feature definitions filtering. filterFeatureDefinitions: filter feature definitions XcmsExperiment keeping defined parameter features, can logical length equal number features, integer index features featureDefinitions(object) keep character feature IDs (.e. row names featureDefinitions(object)). filterFile: filter XcmsExperiment (MsExperiment) file (sample). index samples data subsetted can specified parameter file. sole purpose function provide backward compatibility MSnbase package. Wherever possible, [ function used instead sample-based subsetting. Parameters keepChromPeaks, keepAdjustedRtime keepChromPeaks can passed using .... Note also contrast [, filterFile support subsetting arbitrary order. filterIsolationWindow: filter spectra within MsExperiment XcmsExperiment object keeping isolation window containing specified m/z (.e., keeping spectra \"isolationWindowLowerMz\" smaller user-provided mz \"isolationWindowUpperMz\" larger mz). XcmsExperiment also chromatographic peaks (subsequently also features) removed range \"isolationWindowLowerMz\" \"isolationWindowUpperMz\" (columns chromPeakData) contain user provided mz. filterMsLevel: filter data XcmsExperiment MsExperiment keep data MS level(s) specified parameter msLevel.. filterMz, filterMzRange: filter spectra within XcmsExperiment MsExperiment specified m/z range (parameter mz). XcmsExperiment also identified chromatographic peaks features filtered keeping within specified m/z range (.e. m/z peak apex within m/z range). Parameter msLevels. allows restrict filtering specified MS levels. default data MS levels filtered. filterRt: filter XcmsExperiment keeping data within specified retention time range (parameter rt). function keep preprocessing results present within retention time range: identified chromatographic peaks retention time apex position within retention time range rt retained along, present, associated features. Parameter msLevel. currently ignored, .e. filtering always performed MS levels object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"functionality-related-to-chromatographic-peaks","dir":"Reference","previous_headings":"","what":"Functionality related to chromatographic peaks","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"chromatogram: extract chromatographic data data set. Parameters mz rt allow define specific m/z - retention time regions extract data (e.g. extracted ion chromatograms EICs). parameters expected numerical two-column matrices first column defining lower second upper margin. row can define separate m/z - retention time region. Currently function returns MChromatograms() object object MsExperiment , object XcmsExperiment, either MChromatograms XChromatograms() depending parameter return.type (can either \"MChromatograms\" \"XChromatograms\"). latter also chromatographic peaks detected within provided m/z retention times returned. Parameter chromPeaks allows specify chromatographic peaks reported. See documentation chromPeaks parameter information. XcmsExperiment contains correspondence results, also associated feature definitions included returned XChromatograms. default function returns chromatograms MS1 data, setting parameter msLevel = 2L possible e.g. extract also MS2 chromatograms. msLevel 1 addition important also specify isolationWindowTargetMz MS2 data extracted (e.g. SWATH data MS2 spectra created different m/z isolation windows isolationWindowTargetMz parameter allows define MS2 chromatogram extracted. Note future efficient data structures chromatographic data available well. chromPeaks: returns numeric matrix identified chromatographic peaks. row represents chromatographic peak identified one sample (file). number columns depends peak detection algorithm (see findChromPeaks()) methods return following columns: \"mz\" (intensity-weighted mean m/z values mass peaks included chromatographic peak), \"mzmin\" ( smallest m/z value mass peak chromatographic peak), \"mzmax\" (largest m/z value mass peak chromatographic peak), \"rt\" (retention time peak apex), \"rtmin\" (retention time first scan/mass peak chromatographic peak), \"rtmax\" (retention time last scan/mass peak chromatographic peak), \"\" (integrated intensity chromatographic peak), \"maxo\" (maximal intensity mass peak chromatographic peak), \"sample\" (index sample object peak identified). Parameters rt, mz, ppm, msLevel type allow extract subsets identified chromatographic peaks object. See parameter description details. chromPeakData: returns DataFrame potential additional annotations identified chromatographic peaks. row DataFrame corresponds row (index row name) chromPeaks matrix. default annotations \"ms_level\" (MS level peak identified) \"is_filled\" (whether chromatographic peak detected (findChromPeaks) filled-(fillChromPeaks). chromPeakSpectra: extract MS spectra identified chromatographic peaks. can either (full scan) MS1 spectra retention times retention time range chromatographic peak, MS2 spectra (present) retention time within retention time range (MS1) chromatographic peak precursor m/z within m/z range chromatographic peak single, selected spectra depending total signal highest signal. Parameter msLevel allows define MS level spectra extracted, parameter method allows define selected spectra returned. See chromPeakSpectra() details. dropChromPeaks: removes () chromatographic peak detection results object. also remove correspondence results (.e. features) eventually present adjusted retention times object alignment performed peak detection. Alignment results (adjusted retention times) can retained parameter keepAdjustedRtime set TRUE. dropFilledChromPeaks: removes chromatographic peaks added gap filling fillChromPeaks. fillChromPeaks: perform gap filling integrate signal missing values samples chromatographic peak found. depends correspondence results, hence groupChromPeaks needs called first. details options see fillChromPeaks(). findChromPeaks: perform chromatographic peak detection. See findChromPeaks() details. hasChromPeaks: whether object contains peak detection results. Parameter msLevel allows check whether peak detection results available specified MS level(s). hasFilledChromPeaks: whether gap-filling results (.e., filled-chromatographic peaks) present. manualChromPeaks: manually add chromatographic peaks defining m/z retention time ranges. See manualChromPeaks() details examples. plotChromPeakImage: show density identified chromatographic peaks per file along retention time. See plotChromPeakImage() details. plotChromPeaks: indicate identified chromatographic peaks one sample RT-m/z space. See plotChromPeaks() details. refineChromPeaks: refines identified chromatographic peaks object. See refineChromPeaks() details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"functionality-related-to-alignment","dir":"Reference","previous_headings":"","what":"Functionality related to alignment","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"adjustedRtime: extract adjusted retention times. just alias rtime(object, adjusted = TRUE). adjustRtime: performs retention time adjustment (alignment) data. See adjustRtime() details. applyAdjustedRtime: replaces original (raw) retention times adjusted ones. See applyAdjustedRtime() information. dropAdjustedRtime: drops alignment results (adjusted retention time) result object. also reverts retention times identified chromatographic peaks present result object. Note results correspondence analysis (.e. feature definitions) dropped (correspondence analysis performed alignment). can overruled keepAdjustedRtime = TRUE. hasAdjustedRtime: whether alignment performed object (.e., object contains alignment results). plotAdjustedRtime: plot alignment results; see plotAdjustedRtime() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"functionality-related-to-correspondence-analysis","dir":"Reference","previous_headings":"","what":"Functionality related to correspondence analysis","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"dropFeatureDefinitions: removes correspondence analysis results object well filled-chromatographic peaks. default (parameter keepAdjustedRtime = FALSE) also alignment results removed alignment performed correspondence analysis. can overruled keepAdjustedRtime = TRUE. featureArea: returns matrix columns \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" m/z retention time range feature (row) object. default represent minimal m/z retention times well maximal m/z retention times chromatographic peaks assigned feature. Parameter features allows extract values selected features . Parameters mzmin, mzmax, rtmin rtmax allow define function calculate reported \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" values. featureChromatograms: extract ion chromatograms (EICs) feature object. See featureChromatograms() details. featureDefinitions: returns data.frame feature definitions empty data.frame correspondence analysis results present. Parameters msLevel, mz, ppm rt allow define subsets feature definitions returned parameter type defining parameters used subset returned data.frame. See parameter descriptions details. featureSpectra: returns Spectra() List Spectra (MS1 MS2) spectra associated feature. See featureSpectra() details available parameters. featuresSummary: calculate simple summary features. See featureSummary() details. groupChromPeaks: performs correspondence analysis (.e., grouping chromatographic peaks LC-MS features). See groupChromPeaks() details. hasFeatures: whether correspondence analysis results presentin object. optional parameter msLevel allows define  MS level(s) determined feature definitions available. overlappingFeatures: identify features overlapping close m/z - rt dimension. See overlappingFeatures() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"extracting-data-and-results-from-an-xcmsexperiment","dir":"Reference","previous_headings":"","what":"Extracting data and results from an XcmsExperiment","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"Preprocessing results can extracted using following functions: chromPeaks: extract identified chromatographic peaks. See section chromatographic peak detection details. featureDefinitions: extract definition features (chromatographic peaks grouped across samples). See section correspondence analysis details. featureValues: extract matrix values features sample (file). Rows features, columns samples. value returned can defined parameter value, can column chromPeaks matrix. default (value = \"\") integrated chromatographic peak intensities returned. parameter msLevel possible extract values features certain MS levels. correspondence analysis, one chromatographic peak per sample can assigned feature (e.g. close retention time). Parameter method allows define strategy deal cases: method = \"medret\": report value chromatographic peak apex position closest feautre's median retention time. method = \"maxint\": report value chromatographic peak largest signal (parameter intensity allows define column chromPeaks selected; defaults intensity = \"\"). method = \"sum\"`: sum values chromatographic peaks assigned feature sample. quantify: extract correspondence analysis results SummarizedExperiment(). feature values used assay returned SummarizedExperiment, rowData contains featureDefinitions (without column \"peakidx\") colData sampleData object. Additional parameters featureValues function (used extract feature value matrix) can passed via ....","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"visualization","dir":"Reference","previous_headings":"","what":"Visualization","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"plot: plot file position individual peaks m/z - retention time space (color-coded intensity) base peak chromatogram. function ideally called data subset (.e. using filterRt filterMz restrict region interest). Parameter msLevel allows define MS level plot created. x XcmsExperiment available identified chromatographic peaks, also region defining peaks indicated rectangle. Parameter peakCol allows define color border rectangles. plotAdjustedRtime: plot alignment results; see plotAdjustedRtime() information. plotChromPeakImage: show density identified chromatographic peaks per file along retention time. See plotChromPeakImage() details. plotChromPeaks: indicate identified chromatographic peaks one sample RT-m/z space. See plotChromPeaks() details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"general-functionality-and-functions-for-backward-compatibility","dir":"Reference","previous_headings":"","what":"General functionality and functions for backward compatibility","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"uniqueMsLevels: returns unique MS levels spectra object. functions listed ensure compatibility older XCMSnExp() xcms result object. fileNames: returns original data file names spectra data. Ideally, dataOrigin dataStorage spectra variables object's spectra used instead. fromFile: returns file (sample) index spectrum within object. Generally, subsetting sample using [ preferred way get spectra specific sample. polarity: returns polarity information spectrum object. processHistory: returns list ProcessHistory process history objects contain also parameter object used different processings. Optional parameter type allows query specific processing steps. rtime: extract retention times spectra MsExperiment XcmsExperiment object. thus shortcut rtime(spectra(object)) preferred way extract retention times MsExperiment. rtime method XcmsExperiment additional parameter adjusted allows define whether adjusted retention times (present - adjusted = TRUE) raw retention times (adjusted = FALSE) returned. default adjusted retention times returned available.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"differences-compared-to-the-xcmsnexp-object","dir":"Reference","previous_headings":"","what":"Differences compared to the XCMSnExp() object","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"Subsetting [ supports arbitrary ordering.","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/XcmsExperiment.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Next Generation xcms Result Object — filterFeatureDefinitions","text":"","code":"## Creating a MsExperiment object representing the data from an LC-MS ## experiment. library(MsExperiment)  ## Defining the raw data files fls <- c(system.file('cdf/KO/ko15.CDF', package = \"faahKO\"),          system.file('cdf/KO/ko16.CDF', package = \"faahKO\"),          system.file('cdf/KO/ko18.CDF', package = \"faahKO\"))  ## Defining a data frame with the sample characterization df <- data.frame(mzML_file = basename(fls),                 sample = c(\"ko15\", \"ko16\", \"ko18\")) ## Importing the data. This will initialize a `Spectra` object representing ## the raw data and assign these to the individual samples. mse <- readMsExperiment(spectraFiles = fls, sampleData = df)  ## Extract a total ion chromatogram and base peak chromatogram ## from the data bpc <- chromatogram(mse, aggregationFun = \"max\") tic <- chromatogram(mse)  ## Plot them par(mfrow = c(2, 1)) plot(bpc, main = \"BPC\") plot(tic, main = \"TIC\")   ## Extracting MS2 chromatographic data ## ## To show how MS2 chromatograms can be extracted we first load a DIA ## (SWATH) data set. mse_dia <- readMsExperiment(system.file(\"TripleTOF-SWATH\",     \"PestMix1_SWATH.mzML\", package = \"msdata\"))  ## Extracting MS2 chromatogram requires also to specify the isolation ## window from which to extract the data. Without that chromatograms ## will be empty: chr_ms2 <- chromatogram(mse_dia, msLevel = 2L) intensity(chr_ms2[[1L]]) #> numeric(0)  ## First we list available isolation windows table(isolationWindowTargetMz(spectra(mse_dia))) #>  #> 163.75 208.95 244.05 270.85  299.1  329.8 367.35 601.85  #>   1000   1000   1000   1000   1000   1000   1000   1000   ## We can then extract the TIC of MS2 data for a specific isolation window chr_ms2 <- chromatogram(mse_dia, msLevel = 2L,     isolationWindowTargetMz = 244.05) plot(chr_ms2)  #### ## Chromatographic peak detection  ## Perform peak detection on the data using the centWave algorith. Note ## that the parameters are chosen to reduce the run time of the example. p <- CentWaveParam(noise = 10000, snthresh = 40, prefilter = c(3, 10000)) xmse <- findChromPeaks(mse, param = p) xmse #> Object of class XcmsExperiment  #>  Spectra: MS1 (3834)  #>  Experiment data: 3 sample(s) #>  Sample data links: #>   - spectra: 3 sample(s) to 3834 element(s). #>  xcms results: #>   - chromatographic peaks: 248 in MS level(s): 1   ## Have a quick look at the identified chromatographic peaks head(chromPeaks(xmse)) #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1  ## Extract chromatographic peaks identified between 3000 and 3300 seconds chromPeaks(xmse, rt = c(3000, 3300), type = \"within\") #>          mz mzmin mzmax       rt    rtmin    rtmax        into        intb #> CP021 453.2 453.2 453.2 3063.196 3035.027 3114.840  3001594.78  3001514.97 #> CP030 361.1 361.1 361.1 3147.704 3141.444 3153.964   367361.84   362278.28 #> CP031 340.3 340.3 340.3 3230.646 3225.952 3233.776    68671.25    68664.99 #> CP032 526.2 526.2 526.2 3243.166 3238.471 3246.296   374698.56   374692.30 #> CP033 313.1 313.1 313.1 3276.030 3254.121 3290.115  1199152.63  1162839.05 #> CP034 454.1 454.1 454.1 3276.030 3261.945 3294.809 12283448.95 12024951.77 #> CP108 307.1 307.1 307.1 3143.009 3121.099 3164.918  2191519.65  2063202.03 #> CP109 278.1 278.1 278.1 3196.217 3180.568 3213.432   657840.25   651048.36 #> CP110 526.1 526.1 526.1 3179.003 3150.833 3210.302 21334966.98 21147392.01 #> CP111 380.1 380.1 380.1 3152.398 3132.054 3174.308  2201517.91  2201474.09 #> CP112 380.1 380.1 380.1 3210.302 3210.302 3216.561   187031.44   187023.61 #> CP113 380.1 380.1 380.1 3202.477 3174.308 3216.561  1156780.73  1156736.91 #> CP114 380.1 380.1 380.1 3222.821 3216.561 3240.036   485818.25   485793.21 #> CP115 286.2 286.2 286.2 3258.815 3246.296 3280.725  1264118.64  1247233.70 #> CP116 308.1 308.1 308.1 3261.945 3241.601 3285.419  2066854.37  2025917.82 #> CP204 380.1 380.1 380.1 3150.835 3128.925 3171.179  2036803.99  1937655.24 #> CP205 286.2 286.2 286.2 3250.992 3233.777 3257.252   732016.23   722185.86 #> CP206 568.2 568.2 568.2 3207.173 3185.264 3232.212  3951832.25  3866503.74 #>         maxo    sn sample #> CP021  53096 53095      1 #> CP030  49240    57      1 #> CP031  11391 11390      1 #> CP032  60800 60799      1 #> CP033  55392    50      1 #> CP034 554112    58      1 #> CP108 101400    43      2 #> CP109  30208    54      2 #> CP110 622144   251      2 #> CP111  95176 95175      2 #> CP112  25144 28791      2 #> CP113  28920 28919      2 #> CP114  23176 23535      2 #> CP115  67600    41      2 #> CP116  96272    78      2 #> CP204  82624    49      3 #> CP205  74824    68      3 #> CP206 164352   113      3  ## Extract ion chromatograms (EIC) for the first two chromatographic ## peaks. chrs <- chromatogram(xmse,     mz = chromPeaks(xmse)[1:2, c(\"mzmin\", \"mzmax\")],     rt = chromPeaks(xmse)[1:2, c(\"rtmin\", \"rtmax\")]) #> Processing chromatographic peaks  ## An EIC for each sample and each of the two regions was extracted. ## Identified chromatographic peaks in the defined regions are extracted ## as well. chrs #> XChromatograms with 2 rows and 3 columns #>                    1               2               3 #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 1        peaks: 0        peaks: 0 #> [2,]        peaks: 1        peaks: 0        peaks: 0 #> phenoData with 3 variables #> featureData with 4 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Plot the EICs for the second defined region plot(chrs[2, ])   ## Subsetting the data to the results (and data) for the second sample a <- xmse[2] nrow(chromPeaks(xmse)) #> [1] 248 nrow(chromPeaks(a)) #> [1] 100  ## Filtering the result by retention time: keeping all spectra and ## chromatographic peaks within 3000 and 3500 seconds. xmse_sub <- filterRt(xmse, rt = c(3000, 3500)) #> Filter spectra xmse_sub #> Object of class XcmsExperiment  #>  Spectra: MS1 (960)  #>  Experiment data: 3 sample(s) #>  Sample data links: #>   - spectra: 3 sample(s) to 960 element(s). #>  xcms results: #>   - chromatographic peaks: 79 in MS level(s): 1  nrow(chromPeaks(xmse_sub)) #> [1] 79  ## Perform an initial feature grouping to allow alignment using the ## peak groups method: pdp <- PeakDensityParam(sampleGroups = rep(1, 3)) xmse <- groupChromPeaks(xmse, param = pdp)  ## Perform alignment using the peak groups method. pgp <- PeakGroupsParam(span = 0.4) xmse <- adjustRtime(xmse, param = pgp) #> Performing retention time correction using 19 peak groups.  ## Visualizing the alignment results plotAdjustedRtime(xmse)  ## Performing the final correspondence analysis xmse <- groupChromPeaks(xmse, param = pdp)  ## Show the definition of the first 6 features featureDefinitions(xmse) |> head() #>      mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks 1      peakidx #> FT01 279.0 279.0 279.0 2789.588 2787.430 2791.746      2 2      11, 199 #> FT02 286.2 286.2 286.2 3253.923 3245.811 3262.034      2 2     115, 205 #> FT03 300.2 300.2 300.2 3385.835 3384.068 3390.895      4 3 35, 125,.... #> FT04 301.0 301.0 301.0 2789.066 2787.430 2790.180      3 3  10, 97, 198 #> FT05 305.1 305.1 305.1 2927.922 2922.158 2933.686      2 2      14, 202 #> FT06 305.1 305.1 305.1 3000.543 2991.470 3009.616      2 2      15, 203 #>      ms_level #> FT01        1 #> FT02        1 #> FT03        1 #> FT04        1 #> FT05        1 #> FT06        1  ## Extract the feature values; show the results for the first 6 rows. featureValues(xmse) |> head() #>        ko15.CDF ko16.CDF   ko18.CDF #> FT01 17140627.0       NA 16919266.9 #> FT02         NA  1264119   732016.2 #> FT03  4700903.2  5313736  5169558.2 #> FT04  3051847.8  1964444  2774885.3 #> FT05  1070389.9       NA  1983342.5 #> FT06   847473.1       NA  1003750.6  ## The full results can also be extracted as a `SummarizedExperiment` ## that would eventually simplify subsequent analyses with other packages. ## Any additional parameters passed to the function are passed to the ## `featureValues` function that is called to generate the feature value ## matrix. se <- quantify(xmse, method = \"sum\")  ## EICs for all features can be extracted with the `featureChromatograms` ## function. Note that, depending on the data set, extracting this for ## all features might take some time. Below we extract EICs for the ## first 10 features by providing the feature IDs. chrs <- featureChromatograms(xmse,     features = rownames(featureDefinitions(xmse))[1:10]) chrs #> XChromatograms with 10 rows and 3 columns #>              ko15.CDF        ko16.CDF        ko18.CDF #>       <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]         peaks: 1        peaks: 0        peaks: 1 #> [2,]         peaks: 0        peaks: 1        peaks: 1 #> ...              ...             ...             ...  #> [9,]         peaks: 1        peaks: 2        peaks: 0 #> [10,]        peaks: 1        peaks: 1        peaks: 1 #> phenoData with 3 variables #> featureData with 4 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave  #> Correspondence: #>  method: chromatographic peak density  #>  10 feature(s) identified.  plot(chrs[3, ])"},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine which peaks are absent / present in a sample class — absent-methods","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"Determine peaks absent / present sample class","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"object = \"xcmsSet\" absent(object, ...)     present(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"object xcmsSet-class object class Name sample class sampclass minfrac minimum fraction samples necessary class absent/present","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"Determine peaks absent / present sample class   functions treat peaks present   fillPeaks correctly, .e. count present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/absent-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine which peaks are absent / present in a sample class — absent-methods","text":"logical vector length nrow(groups(object)).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Alignment: Retention time correction methods. — adjustRtime","title":"Alignment: Retention time correction methods. — adjustRtime","text":"adjustRtime method(s) perform retention time correction (alignment) chromatograms different samples. Alignment performed defaul MS level 1 data. Retention times spectra MS levels, present, subsequently adjusted based adjusted retention times MS1 spectra. Note calling adjustRtime xcms result object remove eventually present previous alignment results well correspondence analysis results. run second round alignment, raw retention times need replaced adjusted ones using applyAdjustedRtime() function. alignment method can specified (configured) using dedicated param argument. Supported param objects : ObiwarpParam: performs retention time adjustment based full m/z - rt data using obiwarp method (Prince (2006)). based original code supports addition alignment multiple samples aligning center sample. alignment performed directly profile-matrix can hence performed independently peak detection peak grouping. PeakGroupsParam: performs retention time correctoin based alignment features defined /samples (corresponding house keeping compounds marker compounds) (Smith 2006). First retention time deviation features described fitting either polynomial (smooth = \"loess\") linear (smooth = \"linear\") function data points. subsequently used adjust retention time spectrum sample (even spectra MS levels different MS 1). Since function based features (.e. chromatographic peaks grouped across samples) initial correspondence analysis performed using groupChromPeaks() function. Alternatively, also possible manually define numeric matrix retention times markers samples used alignment. matrix can passed alignment function using peakGroupsMatrix parameter PeakGroupsParam parameter object. default adjustRtimePeakGroups function used define matrix. function identifies peak groups (features) alignment object based parameters defined param. See also do_adjustRtime_peakGroups() core API function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Alignment: Retention time correction methods. — adjustRtime","text":"","code":"adjustRtime(object, param, ...)  # S4 method for MsExperiment,ObiwarpParam adjustRtime(object, param, chunkSize = 2L, BPPARAM = bpparam())  # S4 method for MsExperiment,PeakGroupsParam adjustRtime(object, param, msLevel = 1L, ...)  PeakGroupsParam(   minFraction = 0.9,   extraPeaks = 1,   smooth = \"loess\",   span = 0.2,   family = \"gaussian\",   peakGroupsMatrix = matrix(nrow = 0, ncol = 0),   subset = integer(),   subsetAdjust = c(\"average\", \"previous\") )  ObiwarpParam(   binSize = 1,   centerSample = integer(),   response = 1L,   distFun = \"cor_opt\",   gapInit = numeric(),   gapExtend = numeric(),   factorDiag = 2,   factorGap = 1,   localAlignment = FALSE,   initPenalty = 0,   subset = integer(),   subsetAdjust = c(\"average\", \"previous\") )  adjustRtimePeakGroups(object, param = PeakGroupsParam(), msLevel = 1L)  # S4 method for OnDiskMSnExp,ObiwarpParam adjustRtime(object, param, msLevel = 1L)  # S4 method for PeakGroupsParam minFraction(object)  # S4 method for PeakGroupsParam minFraction(object) <- value  # S4 method for PeakGroupsParam extraPeaks(object)  # S4 method for PeakGroupsParam extraPeaks(object) <- value  # S4 method for PeakGroupsParam smooth(x)  # S4 method for PeakGroupsParam smooth(object) <- value  # S4 method for PeakGroupsParam span(object)  # S4 method for PeakGroupsParam span(object) <- value  # S4 method for PeakGroupsParam family(object)  # S4 method for PeakGroupsParam family(object) <- value  # S4 method for PeakGroupsParam peakGroupsMatrix(object)  # S4 method for PeakGroupsParam peakGroupsMatrix(object) <- value  # S4 method for PeakGroupsParam subset(x)  # S4 method for PeakGroupsParam subset(object) <- value  # S4 method for PeakGroupsParam subsetAdjust(object)  # S4 method for PeakGroupsParam subsetAdjust(object) <- value  # S4 method for ObiwarpParam binSize(object)  # S4 method for ObiwarpParam binSize(object) <- value  # S4 method for ObiwarpParam centerSample(object)  # S4 method for ObiwarpParam centerSample(object) <- value  # S4 method for ObiwarpParam response(object)  # S4 method for ObiwarpParam response(object) <- value  # S4 method for ObiwarpParam distFun(object)  # S4 method for ObiwarpParam distFun(object) <- value  # S4 method for ObiwarpParam gapInit(object)  # S4 method for ObiwarpParam gapInit(object) <- value  # S4 method for ObiwarpParam gapExtend(object)  # S4 method for ObiwarpParam gapExtend(object) <- value  # S4 method for ObiwarpParam factorDiag(object)  # S4 method for ObiwarpParam factorDiag(object) <- value  # S4 method for ObiwarpParam factorGap(object)  # S4 method for ObiwarpParam factorGap(object) <- value  # S4 method for ObiwarpParam localAlignment(object)  # S4 method for ObiwarpParam localAlignment(object) <- value  # S4 method for ObiwarpParam initPenalty(object)  # S4 method for ObiwarpParam initPenalty(object) <- value  # S4 method for ObiwarpParam subset(x)  # S4 method for ObiwarpParam subset(object) <- value  # S4 method for ObiwarpParam subsetAdjust(object)  # S4 method for ObiwarpParam subsetAdjust(object) <- value  # S4 method for XCMSnExp,PeakGroupsParam adjustRtime(object, param, msLevel = 1L)  # S4 method for XCMSnExp,ObiwarpParam adjustRtime(object, param, msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Alignment: Retention time correction methods. — adjustRtime","text":"object adjustRtime: OnDiskMSnExp(), XCMSnExp(), MsExperiment() XcmsExperiment() object. param parameter object defining alignment method (setting). ... ignored. chunkSize adjustRtime object either MsExperiment XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. Alignment performed parallel (per sample) subset loaded data. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM parallel processing setup. Defaults BPPARAM = bpparam(). See bpparam() details. msLevel adjustRtime: integer(1) defining MS level alignment performed. minFraction PeakGroupsParam: numeric(1) 0 1 defining minimum required proportion samples peaks peak group identified. Peak groups passing criteria aligned across samples retention times individual spectra adjusted based alignment. minFraction = 1 peak group contain peaks samples experiment. Note subset provided, specified fraction relative defined subset samples total number samples within experiment (.e., peak present specified proportion subset samples). extraPeaks PeakGroupsParam: numeric(1) defining maximal number additional peaks samples assigned peak group (feature) retention time correction. data set 6 samples, extraPeaks = 1 uses peak groups total peak count <= 6 + 1. total peak count total number peaks assigned peak group considers also multiple peaks within sample assigned group. smooth PeakGroupsParam: character(1) defining function used interpolate corrected retention times peak groups. Can either \"loess\" \"linear\". span PeakGroupsParam: numeric(1) defining degree smoothing (smooth = \"loess\"). parameter passed internal call loess(). family PeakGroupsParam: character(1) defining method loess smoothing. Allowed values \"gaussian\" \"symmetric\". See loess() information. peakGroupsMatrix PeakGroupsParam: optional matrix (raw) retention times (marker) peak groups alignment performed. column represents sample, row feature/peak group. adjustRtimePeakGroups method used default determine matrix provided object. subset ObiwarpParam PeakGroupsParam: integer indices samples within experiment alignment models estimated. Samples part subset adjusted based closest subset sample. See Subset-based alignment section details. subsetAdjust ObiwarpParam PeakGroupsParam: character(1) specifying method non-subset samples adjusted. Supported options \"previous\" \"average\" (default). See Subset-based alignment section details. binSize numeric(1) defining bin size (mz dimension) used profile matrix generation. See step parameter profile-matrix documentation details. centerSample integer(1) defining index center sample experiment. defaults floor(median(1:length(fileNames(object)))). Note subset used, index passed centerSample within subset samples. response ObiwarpParam: numeric(1) defining responsiveness warping response = 0 giving linear warping start end points response = 100 warping using bijective anchors. distFun ObiwarpParam: character(1) defining distance function used. Allowed values \"cor\" (Pearson's correlation), \"cor_opt\" (calculate 10% diagonal band distance matrix; better runtime), \"cov\" (covariance), \"prd\" (product) \"euc\" (Euclidian distance). default value distFun = \"cor_opt\". gapInit ObiwarpParam: numeric(1) defining penalty gap opening. default value depends value distFun: distFun = \"cor\" distFun = \"cor_opt\" 0.3, distFun = \"cov\" distFun = \"prd\" 0.0 distFun = \"euc\" 0.9. gapExtend ObiwarpParam: numeric(1) defining penalty gap enlargement. default value gapExtend depends value distFun: distFun = \"cor\" distFun = \"cor_opt\" 2.4, distFun = \"cov\" 11.7, distFun = \"euc\" 1.8 distFun = \"prd\" 7.8. factorDiag ObiwarpParam: numeric(1) defining local weight applied diagonal moves alignment. factorGap ObiwarpParam: numeric(1) defining local weight gap moves alignment. localAlignment ObiwarpParam: logical(1) whether local alignment performed instead default global alignment. initPenalty ObiwarpParam: numeric(1) defining penalty initiating alignment (local alignment ). value value slot. x ObiwarpParam PeakGroupsParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Alignment: Retention time correction methods. — adjustRtime","text":"adjustRtime OnDiskMSnExp XCMSnExp object return XCMSnExp object alignment results. adjustRtime MsExperiment XcmsExperiment return XcmsExperiment adjusted retention times stored new spectra variable  rtime_adjusted object's spectra. ObiwarpParam PeakGroupsParam return respective parameter object. adjustRtimeGroups returns matrix retention times marker features sample (row one feature, row one sample).","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"subset-based-alignment","dir":"Reference","previous_headings":"","what":"Subset-based alignment","title":"Alignment: Retention time correction methods. — adjustRtime","text":"alignment methods allow perform retention time correction user-selected subset samples (e.g. QC samples) samples part subset adjusted based adjusted retention times closest subset sample (close terms index within object hence possibly injection index). thus suggested load MS data files order samples injected measurement run(s). non-subset samples adjusted depends also parameter subsetAdjust: subsetAdjust = \"previous\", non-subset sample adjusted based closest previous subset sample results cases adjusted retention times non-subset sample identical subset sample adjustment bases. second, default, option subsetAdjust = \"average\" case non subset sample adjusted based average retention time adjustment previous following subset sample. average, weighted mean used weights inverse distance non-subset sample subset samples used alignment. See also section Alignment experiments including blanks xcms vignette details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Alignment: Retention time correction methods. — adjustRtime","text":"Prince, J. T., Marcotte, E. M. (2006) \"Chromatographic Alignment ESI-LC-MS Proteomic Data Sets Ordered Bijective Interpolated Warping\" Anal. Chem., 78 (17), 6140-6152. Smith, C.., Want, E.J., O'Maille, G., Abagyan, R. Siuzdak, G. (2006). \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/adjustRtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Alignment: Retention time correction methods. — adjustRtime","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Replace raw with adjusted retention times — applyAdjustedRtime","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Replaces raw retention times adjusted retention time returns object unchanged none present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"","code":"applyAdjustedRtime(object)"},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"object XCMSnExp XcmsExperiment object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"XCMSnExp XcmsExperiment object raw (original) retention times replaced adjusted retention time.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Adjusted retention times stored parallel adjusted retention times XCMSnExp XcmsExperiment objects. applyAdjustedRtime replaces raw (original) retention times adjusted retention times.","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Replacing raw retention times adjusted retention times disables possibility restore raw retention times using dropAdjustedRtime() method. function remove retention time processing step settings alignment processHistory() object ensure processing history preserved.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/applyAdjustedRtime.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Replace raw with adjusted retention times — applyAdjustedRtime","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  xod <- adjustRtime(faahko_sub, param = ObiwarpParam()) #> Sample number 2 used as center sample. #> Aligning ko15.CDF against ko16.CDF ...  #> OK #> Aligning ko18.CDF against ko16.CDF ...  #> OK #> Applying retention time adjustment to the identified chromatographic peaks ...  #> OK  hasAdjustedRtime(xod) #> [1] TRUE  ## Replace raw retention times with adjusted retention times. xod <- applyAdjustedRtime(xod)  ## No adjusted retention times present hasAdjustedRtime(xod) #> [1] FALSE  ## Raw retention times have been replaced with adjusted retention times plot(split(rtime(faahko_sub), fromFile(faahko_sub))[[1]] -     split(rtime(xod), fromFile(xod))[[1]], type = \"l\")   ## And the process history still contains the settings for the alignment processHistory(xod) #> [[1]] #> Object of class \"XProcessHistory\" #>  type: Peak detection  #>  date: Tue Aug  1 15:54:19 2023  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: CentWaveParam  #>  MS level(s) 1  #>  #> [[2]] #> Object of class \"XProcessHistory\" #>  type: Retention time correction  #>  date: Fri Jan 26 20:05:36 2024  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: ObiwarpParam  #>  MS level(s) 1  #>"},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":null,"dir":"Reference","previous_headings":"","what":"Aggregate values in y for bins defined on x — binYonX","title":"Aggregate values in y for bins defined on x — binYonX","text":"functions takes two -sized numeric vectors x     y, bins/cuts x bins (either pre-defined number     equal-sized bins bins pre-defined size) aggregates values     y corresponding x values falling within bin.     default (.e. method = \"max\") maximal y value     corresponding x values identified. x expected     incrementally sorted , , internally sorted (    case also y ordered according order x).","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Aggregate values in y for bins defined on x — binYonX","text":"","code":"binYonX(   x,   y,   breaks,   nBins,   binSize,   binFromX,   binToX,   fromIdx = 1L,   toIdx = length(x),   method = \"max\",   baseValue,   sortedX = !is.unsorted(x),   shiftByHalfBinSize = FALSE,   returnIndex = FALSE,   returnX = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Aggregate values in y for bins defined on x — binYonX","text":"x Numeric vector used binning. y Numeric vector (length x) maximum values bin defined. provided, x used. breaks Numeric vector defining breaks bins, .e. lower upper values bin. See examples . nBins integer(1) defining number desired bins. binSize numeric(1) defining desired bin size. binFromX Optional numeric(1) allowing manually specify range x-values used binning. affect calculation breaks bins (.e. nBins binSize provided). provided minimal value sub-set fromIdx-toIdx input vector x used. binToX binFromX, defining maximum x-value used binning. fromIdx Integer vector defining start position one multiple sub-sets input vector x used binning. toIdx toIdx, defining maximum index (indices) x used binning. method character string specifying method used aggregate values y. Allowed \"max\", \"min\", \"sum\" \"mean\" identify maximal minimal value sum values within bin calculate mean value. baseValue base value empty bins (.e. bins either values x fall, NA values y assigned). default (.e. specified), NA assigned bins. sortedX Whether x sorted. shiftByHalfBinSize Logical specifying whether bins shifted half bin size left. Thus, first bin center fromX lower upper boundary fromX - binSize/2 fromX + binSize/2. argument ignored breaks provided. returnIndex Logical indicating whether index max (method = \"max\") min (method = \"min\") value within bin input vector x also reported. methods \"max\" \"min\" argument ignored. returnX logical allowing avoid returning $x, .e. mid-points bins. returnX = FALSE might useful cases breaks pre-defined considerably reduces memory demand.","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Aggregate values in y for bins defined on x — binYonX","text":"Returns list length 2, first element (named \"x\")     contains bin mid-points, second element (named \"y\")     aggregated values input vector y within bin.  returnIndex = TRUE list contains additional element  \"index\" index max min (depending whether  method = \"max\" method = \"min\") value within bin     input vector x.","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Aggregate values in y for bins defined on x — binYonX","text":"breaks defining boundary bin can either passed     directly function argument breaks,     calculated data based arguments nBins binSize     along fromIdx, toIdx optionally binFromX     binToX.     Arguments fromIdx toIdx allow specify subset(s)     input vector x bins calculated.     default full x vector considered. Also, specified     otherwise arguments binFromX binToX , range     bins within sub-sets x[fromIdx]     x[toIdx]. Arguments binFromX binToX allow     overwrite manually defining range breaks     calculated. See examples details. Calculation breaks: nBins breaks correspond     seq(min(x[fromIdx])), max(x[fromIdx], length.= (nBins + 1)).     binSize breaks correspond     seq(min(x[fromIdx]), max(x[toIdx]), = binSize)     exception last break value forced equal     max(x[toIdx]). ensures values specified     range covered breaks defining bins. last bin     however instances slightly larger binSize. See     breaks_on_binSize breaks_on_nBins     details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Aggregate values in y for bins defined on x — binYonX","text":"function ensures values within range used define     breaks considered binning (assigned bin).     means bins except last one values x     >= xlower < xupper (xlower     xupper lower upper boundary, respectively).     last bin condition x >= xlower & x <= xupper.     Note also shiftByHalfBinSize TRUE range     values used binning expanded binSize (.e.     lower boundary fromX - binSize/2, upper     toX + binSize/2). Setting argument TRUE resembles     binning /used profBin function     xcms < 1.51. NA handling: default function ignores NA values     y (thus inherently assumes na.rm = TRUE). NA     values allowed x.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Aggregate values in y for bins defined on x — binYonX","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/binYonX.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Aggregate values in y for bins defined on x — binYonX","text":"","code":"######## ## Simple example illustrating the breaks and the binning. ## ## Define breaks for 5 bins: brks <- seq(2, 12, length.out = 6) ## The first bin is then [2,4), the second [4,6) and so on. brks #> [1]  2  4  6  8 10 12 ## Get the max value falling within each bin. binYonX(x = 1:16, y = 1:16, breaks = brks) #> $x #> [1]  3  5  7  9 11 #>  #> $y #> [1]  3  5  7  9 12 #>  ## Thus, the largest value in x = 1:16 falling into the bin [2,4) (i.e. being ## >= 2 and < 4) is 3, the largest one falling into [4,6) is 5 and so on. ## Note however the function ensures that the minimal and maximal x-value ## (in this example 1 and 12) fall within a bin, i.e. 12 is considered for ## the last bin.  ####### ## Performing the binning ons sub-set of x ## X <- 1:16 ## Bin X from element 4 to 10 into 5 bins. X[4:10] #> [1]  4  5  6  7  8  9 10 binYonX(X, X, nBins = 5L, fromIdx = 4, toIdx = 10) #> $x #> [1] 4.6 5.8 7.0 8.2 9.4 #>  #> $y #> [1]  5  6  7  8 10 #>  ## This defines breaks for 5 bins on the values from 4 to 10 and bins ## the values into these 5 bins. Alternatively, we could manually specify ## the range for the binning, i.e. the minimal and maximal value for the ## breaks: binYonX(X, X, nBins = 5L, fromIdx = 4, toIdx = 10, binFromX = 1, binToX = 16) #> $x #> [1]  2.5  5.5  8.5 11.5 14.5 #>  #> $y #> [1] NA  6  9 10 NA #>  ## In this case the breaks for 5 bins were defined from a value 1 to 16 and ## the values 4 to 10 were binned based on these breaks.  ####### ## Bin values within a sub-set of x, second example ## ## This example illustrates how the fromIdx and toIdx parameters can be used. ## x defines 3 times the sequence form 1 to 10, while y is the sequence from ## 1 to 30. In this very simple example x is supposed to represent M/Z values ## from 3 consecutive scans and y the intensities measured for each M/Z in ## each scan. We want to get the maximum intensities for M/Z value bins only ## for the second scan, and thus we use fromIdx = 11 and toIdx = 20. The breaks ## for the bins are defined with the nBins, binFromX and binToX. X <- rep(1:10, 3) Y <- 1:30 ## Bin the M/Z values in the second scan into 5 bins and get the maximum ## intensity for each bin. Note that we have to specify sortedX = TRUE as ## the x and y vectors would be sorted otherwise. binYonX(X, Y, nBins = 5L, sortedX = TRUE, fromIdx = 11, toIdx = 20) #> $x #> [1] 1.9 3.7 5.5 7.3 9.1 #>  #> $y #> [1] 12 14 16 18 20 #>   ####### ## Bin in overlapping sub-sets of X ## ## In this example we define overlapping sub-sets of X and perform the binning ## within these. X <- 1:30 ## Define the start and end indices of the sub-sets. fIdx <- c(2, 8, 21) tIdx <- c(10, 25, 30) binYonX(X, nBins = 5L, fromIdx = fIdx, toIdx = tIdx) #> [[1]] #> [[1]]$x #> [1] 2.8 4.4 6.0 7.6 9.2 #>  #> [[1]]$y #> [1]  3  5  6  8 10 #>  #>  #> [[2]] #> [[2]]$x #> [1]  9.7 13.1 16.5 19.9 23.3 #>  #> [[2]]$y #> [1] 11 14 18 21 25 #>  #>  #> [[3]] #> [[3]]$x #> [1] 21.9 23.7 25.5 27.3 29.1 #>  #> [[3]]$y #> [1] 22 24 26 28 30 #>  #>  ## The same, but pre-defining also the desired range of the bins. binYonX(X, nBins = 5L, fromIdx = fIdx, toIdx = tIdx, binFromX = 4, binToX = 28) #> [[1]] #> [[1]]$x #> [1]  6.4 11.2 16.0 20.8 25.6 #>  #> [[1]]$y #> [1]  8 10 NA NA NA #>  #>  #> [[2]] #> [[2]]$x #> [1]  6.4 11.2 16.0 20.8 25.6 #>  #> [[2]]$y #> [1]  8 13 18 23 25 #>  #>  #> [[3]] #> [[3]]$x #> [1]  6.4 11.2 16.0 20.8 25.6 #>  #> [[3]]$y #> [1] NA NA NA 23 28 #>  #>  ## The same bins are thus used for each sub-set."},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"Defines breaks binSize sized bins values ranging fromX toX.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"","code":"breaks_on_binSize(fromX, toX, binSize)"},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"fromX numeric(1) specifying lowest value bins. toX numeric(1) specifying largest value bins. binSize numeric(1) defining size bin.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"numeric vector defining lower upper bounds bins.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"function creates breaks bins size binSize. function ensures full data range included bins, .e. last value (upper boundary last bin) always equal toX. however means size last bin always equal desired bin size. See examples details comparisom R's seq function.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_binSize.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generate breaks for binning using a defined bin size. — breaks_on_binSize","text":"","code":"## Define breaks with a size of 0.13 for a data range from 1 to 10: breaks_on_binSize(1, 10, 0.13) #>  [1]  1.00  1.13  1.26  1.39  1.52  1.65  1.78  1.91  2.04  2.17  2.30  2.43 #> [13]  2.56  2.69  2.82  2.95  3.08  3.21  3.34  3.47  3.60  3.73  3.86  3.99 #> [25]  4.12  4.25  4.38  4.51  4.64  4.77  4.90  5.03  5.16  5.29  5.42  5.55 #> [37]  5.68  5.81  5.94  6.07  6.20  6.33  6.46  6.59  6.72  6.85  6.98  7.11 #> [49]  7.24  7.37  7.50  7.63  7.76  7.89  8.02  8.15  8.28  8.41  8.54  8.67 #> [61]  8.80  8.93  9.06  9.19  9.32  9.45  9.58  9.71  9.84 10.00 ## The size of the last bin is however larger than 0.13: diff(breaks_on_binSize(1, 10, 0.13)) #>  [1] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [16] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [31] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [46] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 #> [61] 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.13 0.16 ## If we would use seq, the max value would not be included: seq(1, 10, by = 0.13) #>  [1] 1.00 1.13 1.26 1.39 1.52 1.65 1.78 1.91 2.04 2.17 2.30 2.43 2.56 2.69 2.82 #> [16] 2.95 3.08 3.21 3.34 3.47 3.60 3.73 3.86 3.99 4.12 4.25 4.38 4.51 4.64 4.77 #> [31] 4.90 5.03 5.16 5.29 5.42 5.55 5.68 5.81 5.94 6.07 6.20 6.33 6.46 6.59 6.72 #> [46] 6.85 6.98 7.11 7.24 7.37 7.50 7.63 7.76 7.89 8.02 8.15 8.28 8.41 8.54 8.67 #> [61] 8.80 8.93 9.06 9.19 9.32 9.45 9.58 9.71 9.84 9.97  ## In the next example we use binSize that leads to an additional last bin with ## a smaller binSize: breaks_on_binSize(1, 10, 0.51) #>  [1]  1.00  1.51  2.02  2.53  3.04  3.55  4.06  4.57  5.08  5.59  6.10  6.61 #> [13]  7.12  7.63  8.14  8.65  9.16  9.67 10.00 ## Again, the max value is included, but the size of the last bin is < 0.51. diff(breaks_on_binSize(1, 10, 0.51)) #>  [1] 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 0.51 #> [16] 0.51 0.51 0.33 ## Using just seq would result in the following bin definition: seq(1, 10, by = 0.51) #>  [1] 1.00 1.51 2.02 2.53 3.04 3.55 4.06 4.57 5.08 5.59 6.10 6.61 7.12 7.63 8.14 #> [16] 8.65 9.16 9.67 ## Thus it defines one bin (break) less."},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate breaks for binning — breaks_on_nBins","title":"Generate breaks for binning — breaks_on_nBins","text":"Calculate breaks -sized bins data values fromX toX.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate breaks for binning — breaks_on_nBins","text":"","code":"breaks_on_nBins(fromX, toX, nBins, shiftByHalfBinSize = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate breaks for binning — breaks_on_nBins","text":"fromX numeric(1) specifying lowest value bins. toX numeric(1) specifying largest value bins. nBins numeric(1) defining number bins. shiftByHalfBinSize Logical indicating whether bins shifted left half bin size. results centered bins, .e. first bin centered fromX last around toX.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate breaks for binning — breaks_on_nBins","text":"numeric vector length nBins + 1 defining lower upper bounds bins.","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generate breaks for binning — breaks_on_nBins","text":"generates bins call seq(fromX, toX, length.= nBins) . first second element result vector thus defines lower upper boundary first bin, second third value second bin .","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate breaks for binning — breaks_on_nBins","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/breaks_on_nBins.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generate breaks for binning — breaks_on_nBins","text":"","code":"## Create breaks to bin values from 3 to 20 into 20 bins breaks_on_nBins(3, 20, nBins = 20) #>  [1]  3.00  3.85  4.70  5.55  6.40  7.25  8.10  8.95  9.80 10.65 11.50 12.35 #> [13] 13.20 14.05 14.90 15.75 16.60 17.45 18.30 19.15 20.00 ## The same call but using shiftByHalfBinSize breaks_on_nBins(3, 20, nBins = 20, shiftByHalfBinSize = TRUE) #>  [1]  2.552632  3.447368  4.342105  5.236842  6.131579  7.026316  7.921053 #>  [8]  8.815789  9.710526 10.605263 11.500000 12.394737 13.289474 14.184211 #> [15] 15.078947 15.973684 16.868421 17.763158 18.657895 19.552632 20.447368"},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"Combine xcmsSet objects — c-methods","title":"Combine xcmsSet objects — c-methods","text":"Combines samples peaks multiple xcmsSet objects   single object. Group retention time correction data   discarded. profinfo list set equal   first object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Combine xcmsSet objects — c-methods","text":"xs1 = \"xcmsRaw\" c(xs1, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Combine xcmsSet objects — c-methods","text":"xs1 xcmsSet object ... xcmsSet objects","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Combine xcmsSet objects — c-methods","text":"xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/c.xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Combine xcmsSet objects — c-methods","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":null,"dir":"Reference","previous_headings":"","what":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"Calibrate peaks using mz values known masses/calibrants. mz values identified peaks adjusted based peaks close provided mz values. See details information. isCalibrated function returns TRUE chromatographic peaks XCMSnExp object x calibrated FALSE otherwise.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"","code":"CalibrantMassParam(   mz = list(),   mzabs = 1e-04,   mzppm = 5,   neighbors = 3,   method = \"linear\" )  isCalibrated(object)  # S4 method for XCMSnExp calibrate(object, param)"},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"mz numeric list numeric vectors reference mz values. numeric vector provided, used sample XCMSnExp object. list provided, length equal number samples experiment. mzabs numeric(1) absolute error/deviation matching peaks calibrants (Da). mzppm numeric(1) relative error matching peaks calibrants ppm (parts per million). neighbors integer(1) maximal number peaks within permitted distance calibrants considered. Among mz value peak largest intensity used calibration function estimation. method character(1) defining method used estimate calibration function. Can \"shift\", \"linear\" (default) \"edgeshift\". object XCMSnExp object. param CalibrantMassParam object calibration settings.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"CalibrantMassParam: CalibrantMassParam instance. calibrate: XCMSnExp object chromatographic peaks calibrated. aware actual raw mz values (yet) calibrated, identified chromatographic peaks. CalibrantMassParam function returns instance CalibrantMassParam class settings properties set. calibrate method returns XCMSnExp object chromatographic peaks calibrated. Note detected peaks calibrated, individual mz values spectrum.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"method first identify peaks close provided mz values , given difference calibrants smaller user provided cut (based arguments mzabs mzppm), mz values replaced provided mz values. mz values peaks either globally shifted (method = \"shift\" estimated linear model calibrants. Peaks considered close calibrant mz difference calibrant mz <= mzabs + mz * mzppm /1e6. Adjustment methods: adjustment function/factor estimated using difference calibrant peak mz values peaks close enough calibrants. availabel methods : shift: shifts m/z peak global factor corresponds average difference peak mz calibrant mz. linear: fits linear model throught differences calibrant peak mz values adjusts mz values peaks using . edgeshift: performs adjustment linear peaks within mz range calibrants shift outside . information, details examples refer xcms-direct-injection vignette.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"CalibrantMassParam classes exported getter setter methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate-calibrant-mass.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Calibrant mass based calibration of chromatgraphic peaks — CalibrantMassParam-class","text":"Joachim Bargsten, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":null,"dir":"Reference","previous_headings":"","what":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"Calibrate peaks xcmsSet via set known masses","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"object = \"xcmsSet\" calibrate(object, calibrants,method=\"linear\",     mzabs=0.0001, mzppm=5,     neighbours=3, plotres=FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"object xcmsSet object uncalibrated mz calibrants vector list vectors reference m/z-values method used calibrating-method, see mzppm relative error used matching peaks ppm (parts per million) mzabs absolute error used matching peaks Da neighbours number neighbours wich one highest intensity used (instead nearest) plotres can set TRUE wanted result-plot showing found m/z distances regression","code":""},{"path":"https://sneumann.github.io/xcms/reference/calibrate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calibrate peaks for correcting unprecise m/z values — calibrate-methods","text":"object xcmsSet one ore samples calibrants sample different calibrants can used, list m/z-vectors given. length list must number samples, alternatively single vector masses can given used samples. method \"shift\" shifting m/z, \"linear\" linear regression adds linear term m/z. \"edgeshift\" linear regression within range mz-calibrants shift outside.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"Extract ion chromatogram (EIC) chromatographic peak XcmsExperiment() object. result returned XChromatograms() length equal number chromatographic peaks (one column).","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"","code":"chromPeakChromatograms(object, ...)  # S4 method for XcmsExperiment chromPeakChromatograms(   object,   expandRt = 0,   expandMz = 0,   aggregationFun = \"max\",   peaks = character(),   return.type = c(\"XChromatograms\", \"MChromatograms\"),   ...,   progressbar = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"object XcmsExperiment() identified chromatographic peaks. ... currently ignored. expandRt numeric(1) eventually expand retention time range signal integrated. chromatogram contain signal chromPeaks[, \"rtmin\"] - expandRt chromPeaks[, \"rtmax\"] + expandRt. default expandRt = 0. expandMz numeric(1) eventually expand m/z range signal integrated. chromatogram contain signal chromPeaks[, \"mzmin\"] - expandMz chromPeaks[, \"mzmax\"] + expandMz. default expandMz = 0. aggregationFun character(1) defining function signals within m/z range spectrum (.e. discrete retention time) aggregated. default (aggregationFun = \"max\") reports largest signal spectrum. peaks optional character providing IDs chromatographic peaks (.e. row names peaks chromPeaks(object)) chromatograms returned. return.type character(1) specifying type returned object. Can either return.type = \"XChromatograms\" (default) return.type = \"MChromatograms\" return either chromatographic object without identified chromatographic peaks, respectively. progressbar logical(1) whether progress extraction process displayed.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakChromatograms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract an ion chromatogram for each chromatographic peak — chromPeakChromatograms","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Get EICs for every detected chromatographic peak chrs <- chromPeakChromatograms(faahko_sub) chrs #> XChromatograms with 248 rows and 1 column #>                      1 #>        <XChromatogram> #> [1,]          peaks: 1 #> [2,]          peaks: 1 #> ...               ...  #> [247,]        peaks: 1 #> [248,]        peaks: 1 #> phenoData with 2 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Order of EICs matches the order in chromPeaks chromPeaks(faahko_sub) |> head() #>          mz mzmin mzmax       rt    rtmin    rtmax       into       intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982  1007409.0  1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659   687146.6   671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517  5210015.9  5135916.9 152320 #> CP004 430.1 430.1 430.1 2681.348 2639.094 2712.647  2395840.3  2299899.6  65752 #> CP005 366.0 366.0 366.0 2679.783 2642.224 2718.907  3365174.0  3279468.3  79928 #> CP006 343.0 343.0 343.0 2678.218 2637.529 2712.647 24147443.2 23703761.7 672064 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    42      1 #> CP005    49      1 #> CP006    87      1  ## variable \"sample_index\" provides the index of the sample the EIC was ## extracted from fData(chrs)$sample_index #>   [1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 #>  [38] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 #>  [75] 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 #> [112] 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 #> [149] 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 #> [186] 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 #> [223] 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3  ## Get the EIC for selected peaks only. pks <- rownames(chromPeaks(faahko_sub))[c(6, 12)] pks #> [1] \"CP006\" \"CP012\"  ## Expand the data on retention time dimension by 15 seconds (on each side) res <- chromPeakChromatograms(faahko_sub, peaks = pks, expandRt = 5) plot(res[1, ])"},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"Extract (MS1 MS2) spectra XcmsExperiment XCMSnExp object identified chromatographic peaks. return spectra selected chromatographic peaks, peak ID (.e., row name chromPeaks matrix) can provided parameter peaks. msLevel = 1L (supported return.type = \"Spectra\" return.type = \"List\") MS1 spectra within retention time boundaries (file peak detected) returned. msLevel = 2L MS2 spectra returned chromatographic peak precursor m/z within retention time m/z range chromatographic peak. Parameter method allows define whether single spectrum returned: method = \"\": (default): return spectra chromatographic peak. method = \"closest_rt\": return spectrum retention time closest peak's retention time (apex). method = \"closest_mz\": return spectrum precursor m/z closest peaks's m/z (apex); supported msLevel > 1. method = \"largest_tic\": return spectrum largest total signal (sum peaks intensities). method = \"largest_bpi\": return spectrum largest peak intensity (maximal peak intensity). method = \"signal\": object XCMSnExp: return spectrum sum intensities similar peak's apex signal (\"maxo\"); supported msLevel = 2L. Parameter return.type allows specify type result object. return.type = \"Spectra\" (default) Spectra object matching spectra returned. spectra variable \"peak_id\" returned Spectra contains ID chromatographic peak (.e., rowname peak chromPeaks matrix) spectrum. return.type = \"Spectra\" List Spectra returned. length list equal number rows chromPeaks. element list contains thus Spectra spectra one chromatographic peak (Spectra length 0 spectrum found respective chromatographic peak). See also LC-MS/MS data analysis vignette details examples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"","code":"chromPeakSpectra(object, ...)  # S4 method for XcmsExperiment chromPeakSpectra(   object,   method = c(\"all\", \"closest_rt\", \"closest_mz\", \"largest_tic\", \"largest_bpi\"),   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   skipFilled = FALSE,   peaks = character(),   return.type = c(\"Spectra\", \"List\"),   BPPARAM = bpparam() )  # S4 method for XCMSnExp chromPeakSpectra(   object,   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   method = c(\"all\", \"closest_rt\", \"closest_mz\", \"signal\", \"largest_tic\", \"largest_bpi\"),   skipFilled = FALSE,   return.type = c(\"Spectra\", \"MSpectra\", \"List\", \"list\"),   peaks = character() )"},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"object XcmsExperiment XCMSnExp object identified chromatographic peaks spectra returned. ... ignored. method character(1) specifying spectra include result. Defaults method = \"\". See function description details. msLevel integer(1) defining MS level spectra returned. expandRt numeric(1) expand retention time range peak constant value side. expandMz numeric(1) expand m/z range peak constant value side. ppm numeric(1) expand m/z range peak (side) value dependent peak's m/z. skipFilled logical(1) whether spectra filled-peaks reported . peaks character, logical integer allowing specify subset chromatographic peaks chromPeaks spectra returned (providing either ID, logical vector length nrow(chromPeaks(x)) index chromPeaks(x)). parameter overrides skipFilled. return.type character(1) defining type result object returned. BPPARAM parallel processing setup. Defaults bpparam().","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"parameter return.type allow specify type returned object: return.type = \"Spectra\" (default): Spectra object (defined Spectra package). result contains spectra peaks. Metadata column \"peak_id\" provides ID respective peak (.e. rowname chromPeaks(). return.type = \"List\": List length equal number chromatographic peaks returned, element Spectra spectra one chromatographic peak. backward compatibility options \"MSpectra\" \"list\" also supported suggested. return.type = \"MSpectra\" (deprecated): MSpectra object elements Spectrum objects. result objects contains spectra peaks. Metadata column \"peak_id\" provides ID respective peak (.e. rowname chromPeaks()). return.type = \"list\": list lists either length 0 contain Spectrum2 object(s) within m/z-rt range. length list matches number peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromPeakSpectra.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract spectra associated with chromatographic peaks — chromPeakSpectra","text":"","code":"## Read a file with DDA LC-MS/MS data library(MsExperiment) fl <- system.file(\"TripleTOF-SWATH/PestMix1_DDA.mzML\", package = \"msdata\")  dda <- readMsExperiment(fl)  ## Perform MS1 peak detection dda <- findChromPeaks(dda, CentWaveParam(peakwidth = c(5, 15),     prefilter = c(5, 1000)))  ## Return all MS2 spectro for each chromatographic peaks as a Spectra object ms2_sps <- chromPeakSpectra(dda) ms2_sps #> MSn data (Spectra) with 93 spectra in a MsBackendMzR backend: #>       msLevel     rtime scanIndex #>     <integer> <numeric> <integer> #> 1           2   237.460      1810 #> 2           2   237.869      1812 #> 3           2   240.759      1842 #> 4           2   241.299      1846 #> 5           2   318.444      2391 #> ...       ...       ...       ... #> 89          2   552.147      4948 #> 90          2   570.506      5077 #> 91          2   574.435      5108 #> 92          2   574.725      5110 #> 93          2   575.255      5115 #>  ... 34 more variables/columns. #>  #> file(s): #> PestMix1_DDA.mzML #> Processing: #>  Filter: select MS level(s) 2 [Fri Jan 26 20:05:44 2024] #>  Merge 1 Spectra into one [Fri Jan 26 20:05:44 2024]   ## spectra variable *peak_id* contain the row names of the peaks in the ## chromPeak matrix and allow thus to map chromatographic peaks to the ## returned MS2 spectra ms2_sps$peak_id #>  [1] \"CP01\" \"CP01\" \"CP01\" \"CP01\" \"CP02\" \"CP03\" \"CP03\" \"CP03\" \"CP03\" \"CP04\" #> [11] \"CP04\" \"CP04\" \"CP04\" \"CP06\" \"CP06\" \"CP06\" \"CP06\" \"CP06\" \"CP06\" \"CP07\" #> [21] \"CP07\" \"CP07\" \"CP07\" \"CP07\" \"CP07\" \"CP07\" \"CP08\" \"CP08\" \"CP09\" \"CP09\" #> [31] \"CP10\" \"CP10\" \"CP11\" \"CP11\" \"CP11\" \"CP11\" \"CP11\" \"CP12\" \"CP12\" \"CP12\" #> [41] \"CP12\" \"CP12\" \"CP12\" \"CP13\" \"CP13\" \"CP13\" \"CP13\" \"CP13\" \"CP14\" \"CP14\" #> [51] \"CP14\" \"CP15\" \"CP15\" \"CP15\" \"CP15\" \"CP15\" \"CP15\" \"CP17\" \"CP17\" \"CP17\" #> [61] \"CP17\" \"CP17\" \"CP18\" \"CP18\" \"CP18\" \"CP18\" \"CP18\" \"CP19\" \"CP19\" \"CP19\" #> [71] \"CP19\" \"CP19\" \"CP19\" \"CP20\" \"CP20\" \"CP20\" \"CP21\" \"CP21\" \"CP21\" \"CP22\" #> [81] \"CP22\" \"CP22\" \"CP23\" \"CP23\" \"CP23\" \"CP23\" \"CP25\" \"CP25\" \"CP25\" \"CP27\" #> [91] \"CP27\" \"CP27\" \"CP27\" chromPeaks(dda) #>            mz    mzmin    mzmax      rt   rtmin   rtmax        into        intb #> CP01 219.0961 219.0933 219.0996 240.897 233.595 247.874 15236.49207 15222.54500 #> CP02 219.0947 219.0926 219.0963 319.236 311.979 324.874    76.97947    63.82131 #> CP03 227.9913 227.9902 227.9921 358.349 351.649 367.218  5511.96730  5496.32244 #> CP04 300.0312 300.0289 300.0321 366.568 360.029 375.738 12659.45398 12643.63044 #> CP05 302.0284 302.0273 302.0295 366.568 360.449 375.738  7953.70018  7938.19680 #> CP06 256.1102 256.1095 256.1109 388.167 379.097 397.326  7879.79234  7861.19876 #> CP07 224.0836 224.0831 224.0843 388.587 379.517 397.446  8904.69865  8886.41107 #> CP08 309.9980 309.9962 309.9987 405.715 399.527 415.525  5381.59741  5365.38152 #> CP09 378.0356 378.0341 378.0364 405.715 399.527 415.525  5117.90702  5101.66617 #> CP10 298.2751 298.2739 298.2781 345.711 339.830 354.979 39574.83769 39559.51278 #> CP11 308.0012 307.9983 308.0018 405.715 399.527 415.525  5630.65430  5614.25375 #> CP12 376.0385 376.0359 376.0391 405.715 399.527 415.525  5198.61440  5182.23188 #> CP13 292.1216 292.1204 292.1223 410.555 403.625 420.035 15783.11419 15766.32540 #> CP14 289.1217 289.1202 289.1229 425.434 415.525 433.478 10049.05516 10031.22736 #> CP15 304.1141 304.1119 304.1192 424.614 417.985 434.313 14704.14891 14687.51150 #> CP16 384.9700 384.9684 384.9706 440.533 433.893 449.312  8075.24022  8059.47631 #> CP17 382.9727 382.9715 382.9739 440.533 433.478 449.312  8307.76833  8291.81980 #> CP18 231.0144 231.0134 231.0151 450.992 444.342 460.691  5612.28545  5595.90874 #> CP19 306.1659 306.1624 306.1694 463.786 453.922 471.431 31273.95023 31256.40745 #> CP20 205.0970 205.0956 205.0976 459.421 453.092 466.461  5609.65542  5596.09130 #> CP21 230.9875 230.9862 230.9879 495.109 486.210 502.807  8036.59181  8019.64904 #> CP22 216.1414 216.1398 216.1418 508.678 502.680 516.408  4406.91863  4392.98419 #> CP23 330.2065 330.2008 330.2078 422.514 416.754 431.858   614.91821   599.73617 #> CP24 330.2059 330.2036 330.2081 474.408 460.261 479.107   248.31997   229.20853 #> CP25 330.2067 330.2054 330.2077 552.815 541.355 561.837  5890.30116  5869.52232 #> CP26 330.2057 330.2038 330.2076 505.738 495.519 515.868   315.87230   295.17246 #> CP27 373.0416 373.0406 373.0430 576.468 569.174 580.636  6184.87790  6173.40597 #>            maxo   sn sample #> CP01 4877.11621 4876      1 #> CP02   19.35541   18      1 #> CP03 1459.38428 1458      1 #> CP04 3156.89258 3156      1 #> CP05 1974.66846 1974      1 #> CP06 1836.12695 1835      1 #> CP07 2268.62915 2268      1 #> CP08 1345.32336 1344      1 #> CP09 1300.64075 1300      1 #> CP10 8060.13037 8059      1 #> CP11 1450.15808 1449      1 #> CP12 1350.63330 1350      1 #> CP13 3203.84229 3203      1 #> CP14 2972.63916 2972      1 #> CP15 5009.70654 5009      1 #> CP16 2224.37622 2223      1 #> CP17 2361.33594 2360      1 #> CP18 1671.33203 1670      1 #> CP19 6300.43652 6299      1 #> CP20 1717.72607 1717      1 #> CP21 2244.02979 2243      1 #> CP22 1576.85681 1576      1 #> CP23  135.81871  135      1 #> CP24   19.64630   19      1 #> CP25 2102.92871 2102      1 #> CP26   27.88095   27      1 #> CP27 2511.13501 2510      1  ## Alternatively, return the result as a List of Spectra objects. This list ## is parallel to chromPeaks hence the mapping between chromatographic peaks ## and MS2 spectra is easier. ms2_sps <- chromPeakSpectra(dda, return.type = \"List\") names(ms2_sps) #>  [1] \"CP01\" \"CP02\" \"CP03\" \"CP04\" \"CP05\" \"CP06\" \"CP07\" \"CP08\" \"CP09\" \"CP10\" #> [11] \"CP11\" \"CP12\" \"CP13\" \"CP14\" \"CP15\" \"CP16\" \"CP17\" \"CP18\" \"CP19\" \"CP20\" #> [21] \"CP21\" \"CP22\" \"CP23\" \"CP24\" \"CP25\" \"CP26\" \"CP27\" rownames(chromPeaks(dda)) #>  [1] \"CP01\" \"CP02\" \"CP03\" \"CP04\" \"CP05\" \"CP06\" \"CP07\" \"CP08\" \"CP09\" \"CP10\" #> [11] \"CP11\" \"CP12\" \"CP13\" \"CP14\" \"CP15\" \"CP16\" \"CP17\" \"CP18\" \"CP19\" \"CP20\" #> [21] \"CP21\" \"CP22\" \"CP23\" \"CP24\" \"CP25\" \"CP26\" \"CP27\" ms2_sps[[1L]] #> MSn data (Spectra) with 4 spectra in a MsBackendMzR backend: #>     msLevel     rtime scanIndex #>   <integer> <numeric> <integer> #> 1         2   237.460      1810 #> 2         2   237.869      1812 #> 3         2   240.759      1842 #> 4         2   241.299      1846 #>  ... 34 more variables/columns. #>  #> file(s): #> PestMix1_DDA.mzML #> Processing: #>  Filter: select MS level(s) 2 [Fri Jan 26 20:05:44 2024] #>  Merge 1 Spectra into one [Fri Jan 26 20:05:44 2024]   ## Parameter `msLevel` allows to define from which MS level spectra should ## be returned. By default `msLevel = 2L` but with `msLevel = 1L` all ## MS1 spectra with a retention time within the retention time range of ## a chromatographic peak can be returned. Alternatively, selected ## spectra can be returned by specifying the selection criteria/method ## with the `method` parameter. Below we extract for each chromatographic ## peak the MS1 spectra with a retention time closest to the ## chromatographic peak's apex position. Alternatively it would also be ## possible to select the spectrum with the highest total signal or ## highest (maximal) intensity. ms1_sps <- chromPeakSpectra(dda, msLevel = 1L, method = \"closest_rt\") ms1_sps #> MSn data (Spectra) with 27 spectra in a MsBackendMzR backend: #>       msLevel     rtime scanIndex #>     <integer> <numeric> <integer> #> 1           1   240.897      1843 #> 2           1   319.236      2396 #> 3           1   358.349      2813 #> 4           1   366.568      2924 #> 5           1   366.568      2924 #> ...       ...       ...       ... #> 23          1   422.514      3571 #> 24          1   474.408      4266 #> 25          1   552.815      4958 #> 26          1   505.738      4549 #> 27          1   576.468      5122 #>  ... 34 more variables/columns. #>  #> file(s): #> PestMix1_DDA.mzML #> Processing: #>  Filter: select MS level(s) 1 [Fri Jan 26 20:05:44 2024] #>  Merge 1 Spectra into one [Fri Jan 26 20:05:44 2024]   ## Parameter peaks would allow to extract spectra for specific peaks only. ## Peaks can be defined with parameter `peaks` which can be either an ## `integer` with the index of the peak in the `chromPeaks` matrix or a ## `character` with its rowname in `chromPeaks`. chromPeakSpectra(dda, msLevel = 1L, method = \"closest_rt\", peaks = c(3, 5))"},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Extracting chromatograms — chromatogram,XCMSnExp-method","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"chromatogram: extract chromatographic data (extracted ion chromatogram, base peak chromatogram total ion chromatogram) OnDiskMSnExp XCMSnExp objects. See also help page chromatogram function MSnbase package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp chromatogram(   object,   rt,   mz,   aggregationFun = \"sum\",   missing = NA_real_,   msLevel = 1L,   BPPARAM = bpparam(),   adjustedRtime = hasAdjustedRtime(object),   filled = FALSE,   include = c(\"apex_within\", \"any\", \"none\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"object Either OnDiskMSnExp XCMSnExp object chromatograms extracted. rt numeric(2) two-column matrix defining lower upper boundary retention time range(s). specified, full retention time range original data used. mz numeric(2) two-column matrix defining lower upper mz value MS data slice(s). specified, chromatograms calculated full mz range. aggregationFun character(1) specifying function used aggregate intensity values across mz value range retention time. Allowed values \"sum\" (default), \"max\", \"mean\" \"min\". missing numeric(1) allowing specify intensity value used given retention time signal measured within mz range corresponding scan. Defaults NA_real_ (see also Details Notes sections ). Use missing = 0 resemble behaviour getEIC old user interface. msLevel integer(1) specifying MS level chromatogram extracted. Defaults msLevel = 1L. BPPARAM Parallelisation backend used, depend architecture. Default BiocParallel::bparam(). adjustedRtime chromatogram,XCMSnExp: whether adjusted (adjustedRtime = TRUE) raw retention times (adjustedRtime = FALSE) used filtering returned resulting MChromatograms object. Adjusted retention times used default available. filled logical(1) whether filled-peaks also returned. Defaults filled = FALSE, .e. returns detected chromatographic peaks result object. include character(1) defining chromatographic peaks returned. Supported include = \"apex_within\" (default) returns chromatographic peaks apex within mz rt range, include = \"\" return chromatographic peaks m/z rt ranges overlap mz rt include = \"none\" include chromatographic peaks. ... optional parameters - currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"chromatogram returns XChromatograms object number columns corresponding number files object number rows number specified ranges (.e. number rows matrices provided arguments mz /rt). chromatographic peaks apex position within m/z retention time range also retained well feature definitions peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"Arguments rt mz allow specify MS data slice (.e. m/z range retention time window) chromatogram extracted. parameters can either numeric length 2 lower upper limit, matrix two columns lower upper limits extract multiple EICs . parameter aggregationSum allows specify function used aggregate intensities across m/z range retention time. Setting aggregationFun = \"sum\" e.g. allow calculate total ion chromatogram (TIC), aggregationFun = \"max\" base peak chromatogram (BPC). given retention time intensity measured spectrum NA intensity value returned default. can changed parameter missing, setting missing = 0 result 0 intensity returned cases.","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"XCMSnExp objects, adjusted retention times available, chromatogram method default report use (subsetting based provided parameter rt). can changed setting adjustedRtime = FALSE.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/chromatogram-method.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extracting chromatograms — chromatogram,XCMSnExp-method","text":"","code":"## Load a test data set with identified chromatographic peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the ion chromatogram for one chromatographic peak in the data. chrs <- chromatogram(faahko_sub, rt = c(2700, 2900), mz = 335)  chrs #> XChromatograms with 1 row and 3 columns #>                    1               2               3 #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 0        peaks: 1        peaks: 0 #> phenoData with 1 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Identified chromatographic peaks chromPeaks(chrs) #>        mz mzmin mzmax       rt   rtmin    rtmax    into    intb  maxo sn sample #> CP095 335   335   335 2786.199 2764.29 2812.803 1496244 1451652 58736 55      2 #>       row column #> CP095   1      2  ## Plot the chromatogram plot(chrs)  ## Extract chromatograms for multiple ranges. mzr <- matrix(c(335, 335, 344, 344), ncol = 2, byrow = TRUE) rtr <- matrix(c(2700, 2900, 2600, 2750), ncol = 2, byrow = TRUE) chrs <- chromatogram(faahko_sub, mz = mzr, rt = rtr)  chromPeaks(chrs) #>        mz mzmin mzmax       rt    rtmin    rtmax    into    intb   maxo sn #> CP095 335   335   335 2786.199 2764.290 2812.803 1496244 1451652  58736 55 #> CP003 344   344   344 2679.783 2646.919 2709.517 5210016 5135917 152320 68 #> CP090 344   344   344 2686.042 2648.483 2714.211 5700652 5574745 151744 87 #> CP193 344   344   344 2682.914 2643.790 2731.427 5255689 4946143 125632 41 #>       sample row column #> CP095      2   1      2 #> CP003      1   2      1 #> CP090      2   2      2 #> CP193      3   2      3  plot(chrs)   ## Get access to all chromatograms for the second mz/rt range chrs[1, ] #> XChromatograms with 1 row and 3 columns #>                    1               2               3 #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 0        peaks: 1        peaks: 0 #> phenoData with 1 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Plot just that one plot(chrs[1, , drop = FALSE])"},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":null,"dir":"Reference","previous_headings":"","what":"Find row and column maximum values — colMax","title":"Find row and column maximum values — colMax","text":"Find row column maximum values numeric arrays.","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find row and column maximum values — colMax","text":"","code":"colMax(x, na.rm = FALSE, dims = 1) rowMax(x, na.rm = FALSE, dims = 1) which.colMax(x, na.rm = FALSE, dims = 1) which.rowMax(x, na.rm = FALSE, dims = 1)"},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find row and column maximum values — colMax","text":"x array two dimensions, containing numeric values na.rm logical. missing values (including 'NaN') omitted     calculations? (currently implemented) dims dimensions regarded \"rows\" \"columns\" maximize.     rowMax, maximum dimensions dims+1,     ...; colMax dimensions 1:dims.","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find row and column maximum values — colMax","text":"functions designed act like colSums series   functions except currently handle real arrays    remove NA values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find row and column maximum values — colMax","text":"numeric array suitable size, vector result   one-dimensional. dimnames (names   vector result) taken original array. .* functions, integer array suitable   size, vector result one-dimensional. indecies   returned accessing x one-dimensionally (.e.  x[index]). .colMax(), actual row   indecies determined using (.colMax(x)-1) %%   nrow(x) + 1. .rowMax(), actual column indecies   may determined using ceiling(rowMax(x)/nrow(x)).","code":""},{"path":"https://sneumann.github.io/xcms/reference/colMax.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find row and column maximum values — colMax","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Collect MS^n peaks into xcmsFragments — collect-methods","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"Collecting Peaks xcmsFragmentss several   MS-runs using xcmsSet   xcmsRaw.","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"object = \"xcmsFragments\" collect(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"object (empty) xcmsFragments-class object xs xcmsSet-class object contains     picked ms1-peaks several experiments compMethod (\"floor\", \"round\", \"none\"): compare-method     used find parent peak MSnpeak comparing     MZ-values  MS1peaks MSnParentPeaks. snthresh, mzgap, uniq parameters     getspec-peakpicker included xcmsRaw.","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"running collect(xFragments,xSet) peak table   xcmsFragments includes ms1Peaks experiments stored   xcmsSet-object. contains relevant msN-peaks   xcmsRaw-objects, created temporarily paths   xcmsSet.","code":""},{"path":"https://sneumann.github.io/xcms/reference/collect-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Collect MS^n peaks into xcmsFragments — collect-methods","text":"matrix columns: peakID unique identifier every peak MSnParentPeakID PeakID parent peak msLevel>1 - peak, 0 peak msLevel 1. msLevel msLevel peak. rt retention time peak midpoint mz mz-Value peak intensity intensity peak sample number sample xcmsSet GroupPeakMSn Used grouped xcmsSet groups CollisionEnergy collision energy fragment","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"xcms >= 3.15.3 please use compareChromatograms() instead correlate Correlate intensities two chromatograms . two Chromatogram objects different retention times first aligned match data points first data points second chromatogram. See help alignRt MSnbase::Chromatogram() details. correlate called single MChromatograms() object pairwise correlation chromatogram performed matrix correlation coefficients returned. Note correlation two chromatograms depends also order, e.g. correlate(chr1, chr2) might identical correlate(chr2, chr1). lower upper triangular part correlation matrix might thus different.","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"","code":"# S4 method for Chromatogram,Chromatogram correlate(   x,   y,   use = \"pairwise.complete.obs\",   method = c(\"pearson\", \"kendall\", \"spearman\"),   align = c(\"closest\", \"approx\"),   ... )  # S4 method for MChromatograms,missing correlate(   x,   y = NULL,   use = \"pairwise.complete.obs\",   method = c(\"pearson\", \"kendall\", \"spearman\"),   align = c(\"closest\", \"approx\"),   ... )  # S4 method for MChromatograms,MChromatograms correlate(   x,   y = NULL,   use = \"pairwise.complete.obs\",   method = c(\"pearson\", \"kendall\", \"spearman\"),   align = c(\"closest\", \"approx\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"x Chromatogram() MChromatograms() object. y Chromatogram() MChromatograms() object. use character(1) passed cor function. See cor() details. method character(1) passed cor function. See cor() details. align character(1) defining alignment method used. See help alignRt MSnbase::Chromatogram() details. value parameter passed method parameter alignRt. ... optional parameters passed along alignRt method tolerance , set 0 requires retention times identical.","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"numeric(1) matrix (called MChromatograms objects) correlation coefficient. matrix returned, rows represent chromatograms x columns chromatograms y.","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"Michael Witting, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/correlate-Chromatogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Correlate chromatograms — correlate,Chromatogram,Chromatogram-method","text":"","code":"chr1 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(5, 29, 50, NA, 100, 12, 3, 4, 1, 3)) chr2 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(80, 50, 20, 10, 9, 4, 3, 4, 1, 3)) chr3 <- Chromatogram(rtime = 3:9 + rnorm(7, sd = 0.3),     intensity = c(53, 80, 130, 15, 5, 3, 2))  chrs <- MChromatograms(list(chr1, chr2, chr3))  ## Using `compareChromatograms` instead of `correlate`. compareChromatograms(chr1, chr2) #> [1] -0.09306005 compareChromatograms(chr2, chr1) #> [1] -0.09306005  compareChromatograms(chrs, chrs) #>             [,1]        [,2]      [,3] #> [1,]  1.00000000 -0.09306005 0.9887213 #> [2,] -0.09306005  1.00000000 0.5018044 #> [3,]  0.99355819  0.50180442 1.0000000"},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":null,"dir":"Reference","previous_headings":"","what":"Find start and end points of a peak — descendZero","title":"Find start and end points of a peak — descendZero","text":"Decends sides data peak finds either points   greater equal zero intercept, intercept   given value, bottom first valley side.","code":""},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find start and end points of a peak — descendZero","text":"","code":"descendZero(y, istart = which.max(y)) descendValue(y, value, istart = which.max(y)) descendMin(y, istart = which.max(y))"},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find start and end points of a peak — descendZero","text":"y numeric vector values istart starting point descent value numeric value descend ","code":""},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find start and end points of a peak — descendZero","text":"integer vector length 2 starting ending indicies   peak start end points.","code":""},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find start and end points of a peak — descendZero","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/descendZero.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Find start and end points of a peak — descendZero","text":"","code":"normdist <- dnorm(seq(-4, 4, .1)) - .1 xcms:::descendZero(normdist) #> ilower iupper  #>     25     57  normdist[xcms:::descendZero(normdist)] #> [1] 0.01092083 0.01092083 xcms:::descendValue(normdist, .15) #> ilower iupper  #>     32     50  normdist[xcms:::descendValue(normdist, .15)] #> [1] 0.1660852 0.1660852 xcms:::descendMin(normdist) #> ilower iupper  #>      1     81"},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Create report of analyte differences — diffreport-methods","title":"Create report of analyte differences — diffreport-methods","text":"Create report showing significant differences   two sets samples. Optionally create extracted ion chromatograms   significant differences.","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create report of analyte differences — diffreport-methods","text":"object = \"xcmsSet\" diffreport(object, class1 = levels(sampclass(object))[1],                    class2 = levels(sampclass(object))[2],                    filebase = character(), eicmax = 0, eicwidth = 200,                    sortpval = TRUE, classeic = c(class1,class2),                    value=c(\"\",\"maxo\",\"intb\"), metlin = FALSE, \t\t\t\t   h=480,w=640, mzdec=2, missing = \t\t\t\t   numeric(), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create report of analyte differences — diffreport-methods","text":"object xcmsSet object class1 character vector first set sample classes     compared class2 character vector second set sample classes     compared filebase base file name save report, .tsv file _eic     appended name tabular report EIC     directory, respectively. blank nothing saved eicmax number significantly different analytes create     EICs eicwidth width (seconds) EICs produced sortpval logical indicating whether reports sorted     p-value classeic character vector sample classes include EICs value intensity values used diffreport.      value=\"\", integrated peak intensities used.      value=\"maxo\", maximum peak intensities used.      value=\"intb\", baseline corrected integrated peak intensities used (available peak detection done findPeaks.centWave). metlin mass uncertainty use generating link Metlin metabolite     database. sign uncertainty indicates negative     positive mode data M+H M-H calculation. value FALSE     0 removes column h Numeric variable height eic boxplots printed . w Numeric variable width eic boxplots print made. mzdec Number decimal places title m/z values eic plot. missing numeric(1) defining optional value missing     values. missing = 0 e.g. replace NA values     feature matrix 0. Note also call     fillPeaks results feature matrix     NA values replaced 0. ... optional arguments passed mt.teststat     multtest package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create report of analyte differences — diffreport-methods","text":"method handles creation summary reports statistics   analytes significantly different   two sets samples. computes Welch's two-sample t-statistic   analyte ranks p-value. returns summary   report can optionally written tab-separated file. Additionally, heavy lifting involved creating   superimposed extracted ion chromatograms given number   analytes.  reading raw data files associated   samples interest one time. ,   prints name sample currently reading. Depending   number size samples, process can take   long time. base file name provided, report (see Value section)   saved tab separated file. EICs generated,   saved 640x480 PNG files newly created   subdirectory. However parameter can changed   commands arguments. numbered file names correspond   rows report. Chromatographic traces EICs colored labeled   sample class. Sample classes take color   current palette. color sample class assigned dependent   order xcmsSet object, order given   class arguments. Thus levels(sampclass(object))[1]   use color palette()[1] . way, sample   classes maintain color across number different   generated reports. multiple sample classes, xcms produce boxplots   different classes generate single anova p-value statistic.   Like eic's plot number corresponds row number   report.","code":""},{"path":"https://sneumann.github.io/xcms/reference/diffreport-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create report of analyte differences — diffreport-methods","text":"data frame following columns: fold mean fold change (always greater 1, see tstat     set sample classes higher) tstat Welch's two sample t-statistic, positive analytes     greater intensity class2, negative analytes     greater intensity class1 pvalue p-value t-statistic anova p-value anova statistic multiple classes mzmed median m/z peaks group mzmin minimum m/z peaks group mzmax maximum m/z peaks group rtmed median retention time peaks group rtmin minimum retention time peaks group rtmax maximum retention time peaks group npeaks number peaks assigned group Sample Classes number samples sample class represented group metlin URL metlin mass ... one column every sample class Sample Names integrated intensity value every sample ... one column every sample","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":null,"dir":"Reference","previous_headings":"","what":"Change the file path of an OnDiskMSnExp object — dirname","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"dirname allows get set path directory containing source files OnDiskMSnExp (XCMSnExp) object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"","code":"# S4 method for OnDiskMSnExp dirname(path)  # S4 method for OnDiskMSnExp dirname(path) <- value"},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"path OnDiskMSnExp. value character length 1 length equal number files defining new path files.","code":""},{"path":"https://sneumann.github.io/xcms/reference/dirname.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Change the file path of an OnDiskMSnExp object — dirname","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":null,"dir":"Reference","previous_headings":"","what":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"function performs retention time correction assessing retention time deviation across samples using peak groups (features) containg chromatographic peaks present /samples. retention time deviation features sample described fitting either polynomial (smooth = \"loess\") linear (smooth = \"linear\") model data points. models subsequently used adjust retention time spectrum sample.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"","code":"do_adjustRtime_peakGroups(   peaks,   peakIndex,   rtime = list(),   minFraction = 0.9,   extraPeaks = 1,   smooth = c(\"loess\", \"linear\"),   span = 0.2,   family = c(\"gaussian\", \"symmetric\"),   peakGroupsMatrix = matrix(ncol = 0, nrow = 0),   subset = integer(),   subsetAdjust = c(\"average\", \"previous\") )"},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"peaks matrix data.frame identified chromatographic peaks samples. peakIndex list indices provides grouping information chromatographic peaks (across within samples). rtime list numeric vectors retention times per file/sample. minFraction PeakGroupsParam: numeric(1) 0 1 defining minimum required proportion samples peaks peak group identified. Peak groups passing criteria aligned across samples retention times individual spectra adjusted based alignment. minFraction = 1 peak group contain peaks samples experiment. Note subset provided, specified fraction relative defined subset samples total number samples within experiment (.e., peak present specified proportion subset samples). extraPeaks PeakGroupsParam: numeric(1) defining maximal number additional peaks samples assigned peak group (feature) retention time correction. data set 6 samples, extraPeaks = 1 uses peak groups total peak count <= 6 + 1. total peak count total number peaks assigned peak group considers also multiple peaks within sample assigned group. smooth PeakGroupsParam: character(1) defining function used interpolate corrected retention times peak groups. Can either \"loess\" \"linear\". span PeakGroupsParam: numeric(1) defining degree smoothing (smooth = \"loess\"). parameter passed internal call loess(). family PeakGroupsParam: character(1) defining method loess smoothing. Allowed values \"gaussian\" \"symmetric\". See loess() information. peakGroupsMatrix optional matrix (raw) retention times peak groups alignment performed. column represents sample, row feature/peak group. provided, matrix determined depending parameters minFraction extraPeaks. provided, minFraction extraPeaks ignored. subset ObiwarpParam PeakGroupsParam: integer indices samples within experiment alignment models estimated. Samples part subset adjusted based closest subset sample. See Subset-based alignment section details. subsetAdjust ObiwarpParam PeakGroupsParam: character(1) specifying method non-subset samples adjusted. Supported options \"previous\" \"average\" (default). See Subset-based alignment section details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"list numeric vectors adjusted     retention times grouped sample.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"alignment bases presence compounds can found /samples experiment. retention times individual spectra adjusted based alignment features corresponding house keeping compounds. paraneters minFraction extraPeaks can used fine tune features used alignment (.e. features likely correspond mentioned house keeping compounds). Parameter subset allows define subset samples within experiment aligned. samples part subset aligned based adjustment closest sample within subset. allows e.g. exclude blank samples alignment process retention times still adjusted based alignment results real samples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"method ensures returned adjusted retention times     increasingly ordered, just raw retention times.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_adjustRtime_peakGroups.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Align spectrum retention times across samples using peak groups found in most samples — do_adjustRtime_peakGroups","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for centWave peak detection — do_findChromPeaks_centWave","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"function performs peak density wavelet based     chromatographic peak detection high resolution LC/MS data centroid     mode [Tautenhahn 2008].","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"","code":"do_findChromPeaks_centWave(   mz,   int,   scantime,   valsPerSpect,   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = NULL,   sleep = 0,   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. sleep numeric(1) defining number seconds wait iterations. Defaults sleep = 0. > 0 plot generated visualizing identified chromatographic peak. Note: argument backward compatibility removed future. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685 https://doi.org/10.1186/s12859-023-05533-4 information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"matrix, row representing identified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peak across     scans. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intb Per-peak baseline corrected integrated peak intensity. maxo Maximum intensity peak. sn Signal noise ratio, defined (maxo - baseline)/sd,     sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit. Additional columns verboseColumns = TRUE:  mu Gaussian parameter mu. sigma Gaussian parameter sigma. h Gaussian parameter h. f Region number m/z ROI peak localized. dppm m/z deviation mass trace across scans ppm. scale Scale peak localized. scpos Peak position found wavelet analysis (scan number). scmin Left peak limit found wavelet analysis (scan number). scmax Right peak limit found wavelet analysis (scan numer). Additional columns verboseBetaColumns = TRUE:  beta_cor Correlation \"ideal\" bell curve raw data beta_snr Signal--noise residuals calculated beta_cor fit","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method identifies regions interest (ROIs) representing mass traces characterized regions less ppm m/z deviation consecutive scans LC/MS map. detail, starting single m/z, ROI extended m/z can found next scan (spectrum) difference mean m/z ROI smaller user defined ppm m/z. mean m/z ROI updated considering also newly included m/z value. ROIs , cleanup, analyzed using continuous wavelet transform (CWT) locate chromatographic peaks different scales. first analysis step skipped, regions interest passed roiList parameter.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"centWave designed work centroided mode, thus     expected data presented function. function exposes core chromatographic peak detection functionality     centWave method. function can called     directly, users generally call corresponding method     data object instead.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann \"Highly     sensitive feature detection high resolution LC/MS\"     BMC Bioinformatics 2008, 9:504","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"Ralf Tautenhahn, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWave.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for centWave peak detection — do_findChromPeaks_centWave","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and restrict to a certain retention time range data <- filterRt(filterFile(faahko_sub, 1), c(2500, 3000))  ## Get m/z and intensity values mzs <- mz(data) ints <- intensity(data)  ## Define the values per spectrum: valsPerSpect <- lengths(mzs)  ## Calling the function. We're using a large value for noise and prefilter ## to speed up the call in the example - in a real use case we would either ## set the value to a reasonable value or use the default value. res <- do_findChromPeaks_centWave(mz = unlist(mzs), int = unlist(ints),     scantime = rtime(data), valsPerSpect = valsPerSpect, noise = 10000,     prefilter = c(3, 10000)) #> Detecting mass traces at 25 ppm ...  #> OK #> Detecting chromatographic peaks in 186 regions of interest ... #>  OK: 47 found. head(res) #>         mz mzmin mzmax       rt    rtmin    rtmax      into      intb   maxo #> [1,] 453.2 453.2 453.2 2506.073 2501.378 2527.982 1007409.0 1007380.8  38152 #> [2,] 307.0 307.0 307.0 2618.750 2592.145 2645.354  284782.4  268039.8  16872 #> [3,] 302.0 302.0 302.0 2617.185 2595.275 2640.659  687146.6  671297.8  30552 #> [4,] 360.0 360.0 360.0 2682.913 2668.828 2698.562 5641322.3 5420634.7 317568 #> [5,] 361.1 361.1 361.1 2684.478 2665.698 2698.562 1158340.2 1116522.0  72272 #> [6,] 416.1 416.1 416.1 2682.913 2635.964 2709.517  487698.6  446552.1  12036 #>         sn #> [1,] 38151 #> [2,]    20 #> [3,]    46 #> [4,]    11 #> [5,]    11 #> [6,]    11"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"do_findChromPeaks_centWaveWithPredIsoROIs performs     two-step centWave based peak detection: chromatographic peaks     identified using centWave followed prediction location     identified peaks' isotopes mz-retention time space.     locations fed regions interest (ROIs) subsequent     centWave run. non overlapping peaks two peak detection     runs reported final list identified peaks. do_findChromPeaks_centWaveAddPredIsoROIs performs     centWave based peak detection based regions interest (ROIs)     representing predicted isotopes peaks submitted argument     peaks.. function returns matrix identified peaks     consisting input peaks peaks representing predicted isotopes     (found centWave algorithm).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"","code":"do_findChromPeaks_centWaveWithPredIsoROIs(   mz,   int,   scantime,   valsPerSpect,   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = NULL,   snthreshIsoROIs = 6.25,   maxCharge = 3,   maxIso = 5,   mzIntervalExtension = TRUE,   polarity = \"unknown\",   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE )  do_findChromPeaks_addPredIsoROIs(   mz,   int,   scantime,   valsPerSpect,   ppm = 25,   peakwidth = c(20, 50),   snthresh = 6.25,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   peaks. = NULL,   maxCharge = 3,   maxIso = 5,   mzIntervalExtension = TRUE,   polarity = \"unknown\" )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh do_findChromPeaks_addPredIsoROIs: numeric(1) defining signal noise threshold centWave algorithm. do_findChromPeaks_centWaveWithPredIsoROIs: numeric(1) defining signal noise threshold initial (first) centWave run. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. snthreshIsoROIs numeric(1) defining signal noise ratio cutoff used second centWave run identify peaks predicted isotope ROIs. maxCharge integer(1) defining maximal isotope charge. Isotopes defined charges 1:maxCharge. maxIso integer(1) defining number isotope peaks predicted peak identified first centWave run. mzIntervalExtension logical(1) whether mz range predicted isotope ROIs extended increase detection low intensity peaks. polarity character(1) specifying polarity data. Currently used, \"positive\", \"negative\" \"unknown\" provided. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685 https://doi.org/10.1186/s12859-023-05533-4 information. peaks. matrix xcmsPeaks object one returned call link{do_findChromPeaks_centWave} link{findPeaks.centWave} (verboseColumns = TRUE) peaks isotopes predicted used additional peak detectoin using centWave method. Required columns : \"mz\", \"mzmin\", \"mzmax\", \"scmin\", \"scmax\", \"scale\" \"\".","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"matrix, row representing identified chromatographic peak.     non-overlapping peaks identified centWave runs reported.     matrix columns :  mz Intensity weighted mean m/z values peaks across scans. mzmin Minimum m/z peaks. mzmax Maximum m/z peaks. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intb Per-peak baseline corrected integrated peak intensity. maxo Maximum intensity peak. sn Signal noise ratio, defined (maxo - baseline)/sd,     sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit. Additional columns verboseColumns = TRUE:  mu Gaussian parameter mu. sigma Gaussian parameter sigma. h Gaussian parameter h. f Region number m/z ROI peak localized. dppm m/z deviation mass trace across scans ppm. scale Scale peak localized. scpos Peak position found wavelet analysis (scan number). scmin Left peak limit found wavelet analysis (scan number). scmax Right peak limit found wavelet analysis (scan numer). Additional columns verboseBetaColumns = TRUE:  beta_cor Correlation \"ideal\" bell curve raw data beta_snr Signal--noise residuals calculated beta_cor fit","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"details centWave algorithm see     centWave.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_centWaveWithPredIsoROIs.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for two-step centWave peak detection with isotopes — do_findChromPeaks_centWaveWithPredIsoROIs","text":"Hendrik Treutler, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"Massifquant Kalman filter (KF)-based chromatographic peak     detection XC-MS data centroid mode. identified peaks     can refined centWave method (see     do_findChromPeaks_centWave details centWave)     specifying withWave = TRUE.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"","code":"do_findChromPeaks_massifquant(   mz,   int,   scantime,   valsPerSpect,   ppm = 10,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   criticalValue = 1.125,   consecMissedLimit = 2,   unions = 1,   checkBack = 0,   withWave = FALSE )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. criticalValue numeric(1). Suggested values: (0.1-3.0). setting helps determine Kalman Filter prediciton margin error. real centroid belonging bonafide peak must fall within KF prediction margin error. Much like construction confidence interval, criticalVal loosely translates multiplier standard error prediction reported Kalman Filter. peak XC-MS sample small mass deviance ppm error, smaller critical value might better vice versa. consecMissedLimit integer(1) Suggested values: (1,2,3). peak proces detected Kalman Filter, Kalman Filter may find predicted centroid every scan. 1 consecutive failed predictions, setting informs Massifquant stop Kalman Filter following candidate peak. unions integer(1) set 1 apply t-test union segmentation; set 0 t-test applied chromatographically continous peaks sharing m/z range. Explanation: data points, sometimes Kalman Filter stops tracking peak prematurely. Another Kalman Filter instantiated begins following rest signal. tracking done backwards forwards, algorithmic defect leaves real peak divided two segments . option turned , program identifies segmented peaks combines (merges ) one two sample t-test. potential danger option truly distinct peaks may merged. checkBack integer(1) set 1 turned ; set 0 turned . convergence Kalman Filter peak's precise m/z mapping fast, sometimes incorporates erroneous centroids part peak (especially early ). scanBack option attempt remove occasional outlier lies beyond converged bounds Kalman Filter. option directly affect identification peak postprocessing measure; shown extremely useful thus far default set turned . withWave logical(1) TRUE, peaks identified first Massifquant subsequently filtered second step centWave algorithm, includes wavelet estimation.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"matrix, row representing identified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peaks across     scans. mzmin Minumum m/z peak. mzmax Maximum m/z peak. rtmin Minimum retention time peak. rtmax Maximum retention time peak. rt Retention time peak's midpoint. Integrated (original) intensity peak. maxo Maximum intensity peak. withWave set TRUE, result     returned do_findChromPeaks_centWave method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"algorithm's performance tested rigorously     high resolution LC/OrbiTrap, TOF-MS data centroid mode.     Simultaneous kalman filters identify peaks calculate     area curve. default parameters set operate     complex LC-MS Orbitrap sample. Users find useful     simple exploratory data analysis find set minimum     intensity, identify many scans average peak spans.     consecMissedLimit parameter yielded good performance     Orbitrap data set (2) TOF data found best     (1). may change algorithm yet     tested many samples. criticalValue parameter perhaps     dificult dial appropriately visual inspection peak     identification best suggested tool quick optimization.     ppm checkBack parameters shown less influence     parameters exist give users flexibility     better accuracy.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"Conley CJ, Smith R, Torgrip RJ, Taylor RM, Tautenhahn R Prince JT \"Massifquant: open-source Kalman filter-based XC-MS isotope trace feature detection\" Bioinformatics 2014, 30(18):2636-43.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"Christopher Conley","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_massifquant.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for massifquant peak detection — do_findChromPeaks_massifquant","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and restrict to a certain retention time range data <- filterRt(filterFile(faahko_sub, 1), c(2500, 3000))  ## Get m/z and intensity values mzs <- mz(data) ints <- intensity(data)  ## Define the values per spectrum: valsPerSpect <- lengths(mzs)  ## Perform the peak detection using massifquant - setting prefilter to ## a high value to speed up the call for the example res <- do_findChromPeaks_massifquant(mz = unlist(mzs), int = unlist(ints),     scantime = rtime(data), valsPerSpect = valsPerSpect,     prefilter = c(3, 10000)) #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 10ppm ...  #> OK #>  69 Peaks. head(res) #>         mz mzmin mzmax    rtmin    rtmax       rt    into  maxo #> [1,] 426.1 426.1 426.1 2963.039 2999.033 2977.124  129958 29408 #> [2,] 590.3 590.3 590.3 2963.039 2999.033 2978.689  255317 19072 #> [3,] 454.1 454.1 454.1 2934.870 2975.559 2953.650  372685 20432 #> [4,] 309.1 309.1 309.1 2905.136 2967.734 2936.435  652140 38888 #> [5,] 475.2 475.2 475.2 2880.097 2999.033 2934.870 2415224 37160 #> [6,] 532.2 532.2 532.2 2853.493 2906.701 2878.532  232493 16480"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"function identifies peaks chromatographic     time domain described [Smith 2006]. intensity values     binned cutting LC/MS data slices (bins) mass unit     (binSize m/z) wide. Within bin maximal intensity     selected. peak detection performed bin     extending based steps parameter generate slices     comprising bins current_bin - steps +1     current_bin + steps - 1.     slices filtered matched filtration using     second-derative Gaussian model peak shape. filtration     peaks detected using signal--ration cut-. details     illustrations see [Smith 2006].","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"","code":"do_findChromPeaks_matchedFilter(   mz,   int,   scantime,   valsPerSpect,   binSize = 0.1,   impute = \"none\",   baseValue,   distance,   fwhm = 30,   sigma = fwhm/2.3548,   max = 5,   snthresh = 10,   steps = 2,   mzdiff = 0.8 - binSize * steps,   index = FALSE,   sleep = 0 )"},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. scantime Numeric vector length equal number spectra/scans data representing retention time scan. valsPerSpect Numeric vector number values spectrum. binSize numeric(1) specifying width bins/slices m/z dimension. impute Character string specifying method used missing value imputation. Allowed values \"none\" (linear interpolation), \"lin\" (linear interpolation), \"linbase\" (linear interpolation within certain bin-neighborhood) \"intlin\". See imputeLinInterpol details. baseValue base value empty elements set. considered method = \"linbase\" corresponds profBinLinBase's baselevel argument. distance method = \"linbase\": number non-empty neighboring element empty element considered linear interpolation. See details section information. fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) representing maximum number peaks expected/identified per slice. snthresh numeric(1) defining signal noise ratio cutoff. steps numeric(1) defining number bins merged filtration (.e. number neighboring bins joined slice filtration peak detection performed). mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. index logical(1) specifying whether indicies returned instead values m/z retention times. sleep numeric(1) defining number seconds wait iterations. Defaults sleep = 0. > 0 plot generated visualizing identified chromatographic peak. Note: argument backward compatibility removed future.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"matrix, row representing identified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peak across scans. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intf Integrated intensity filtered peak. maxo Maximum intensity peak. maxf Maximum intensity filtered peak. Rank peak merged EIC (<= max). sn Signal noise ratio peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"intensities binned provided m/z values within     spectrum (scan). Binning performed bins centered     around m/z values (.e. first bin includes m/z values     min(mz) - bin_size/2 min(mz) + bin_size/2). details binning missing value imputation see     binYonX imputeLinInterpol methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"function exposes core peak detection functionality     matchedFilter method. function can called     directly, users generally call corresponding method     data object instead (e.g. link{findPeaks.matchedFilter}     method).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findChromPeaks_matchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for matchedFilter peak detection — do_findChromPeaks_matchedFilter","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and restrict to a certain retention time range data <- filterRt(filterFile(faahko_sub, 1), c(2500, 3000))  ## Get m/z and intensity values mzs <- mz(data) ints <- intensity(data)  ## Define the values per spectrum: valsPerSpect <- lengths(mzs)  res <- do_findChromPeaks_matchedFilter(mz = unlist(mzs), int = unlist(ints),     scantime = rtime(data), valsPerSpect = valsPerSpect) head(res) #>            mz mzmin mzmax       rt    rtmin    rtmax      into      intf  maxo #> [1,] 205.0000 205.0 205.0 2784.635 2770.550 2800.284 1778568.9 3610062.2 84280 #> [2,] 205.9819 205.9 206.0 2786.200 2772.115 2800.284  237993.6  448580.3 10681 #> [3,] 207.0821 207.0 207.1 2712.647 2698.562 2726.731  380873.0  730981.4 18800 #> [4,] 236.0956 236.0 236.1 2518.593 2504.508 2534.242  252282.0  458747.7 12957 #> [5,] 244.1000 244.1 244.1 2828.453 2814.369 2844.103  612169.9 1279308.9 31312 #> [6,] 266.0751 266.0 266.1 2828.453 2815.934 2844.103  113219.0  214886.8  5801 #>           maxf i       sn #> [1,] 195026.48 1 28.33394 #> [2,]  23860.11 1 16.53987 #> [3,]  40065.74 1 12.87314 #> [4,]  24536.55 1 14.99012 #> [5,]  69898.24 1 24.20989 #> [6,]  11773.56 1 10.83870"},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"function performs peak detection mass spectrometry     direct injection spectrum using wavelet based algorithm.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"","code":"do_findPeaks_MSW(   mz,   int,   snthresh = 3,   verboseColumns = FALSE,   scantime = numeric(),   valsPerSpect = integer(),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"mz Numeric vector individual m/z values scans/ spectra one file/sample. int Numeric vector individual intensity values scans/spectra one file/sample. snthresh numeric(1) defining signal noise ratio cutoff. verboseColumns logical(1) whether additional peak meta data columns returned. scantime ignored. valsPerSpect ignored. ... Additional parameters passed peakDetectionCWT function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"matrix, row representing identified peak, columns:  mz m/z value peak centroid position. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Always -1. rtmin Always -1. rtmax Always -1. Integrated (original) intensity peak. maxo Maximum intensity peak. intf Always NA. maxf Maximum MSW-filter response peak. sn Signal noise ratio.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"wrapper around peak picker Bioconductor's     MassSpecWavelet package calling     peakDetectionCWT     tuneInPeakInfo functions. See     xcmsDirect vignette information.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_findPeaks_MSW.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for single-spectrum non-chromatography MS data peak detection — do_findPeaks_MSW","text":"Joachim Kutzera, Steffen Neumann, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"do_groupChromPeaks_density function performs chromatographic peak grouping based density (distribution) peaks, found different samples, along retention time axis slices overlapping m/z ranges. default (parameter ppm = 0) m/z ranges (constant) size (depending parameter binSize). values ppm larger 0 m/z bins (ranges slices) increasing sizes depending m/z value. better models m/z-dependent measurement error/precision seen MS instruments.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"","code":"do_groupChromPeaks_density(   peaks,   sampleGroups,   bw = 30,   minFraction = 0.5,   minSamples = 1,   binSize = 0.25,   maxFeatures = 50,   sleep = 0,   index = seq_len(nrow(peaks)),   ppm = 0 )"},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"peaks matrix data.frame mz values retention times identified chromatographic peaks samples experiment. Required columns \"mz\", \"rt\" \"sample\". latter contain numeric values representing index sample peak found. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). bw PeakDensityParam: numeric(1) defining bandwidth (standard deviation ot smoothing kernel) used. argument passed [density() method. minFraction PeakDensityParam: numeric(1) defining minimum fraction samples least one sample group peaks present considered peak group (feature). minSamples PeakDensityParam: numeric(1) minimum number samples least one sample group peaks detected considered peak group (feature). binSize PeakDensityParam: numeric(1) defining size overlapping slices m/z dimension. maxFeatures PeakDensityParam: numeric(1) maximum number peak groups identified single mz slice. sleep numeric(1) defining time sleep iterations plot result current iteration. index optional integer providing indices peaks original peak matrix. ppm MzClustParam: numeric(1) representing relative m/z error clustering/grouping (parts per million). PeakDensityParam: numeric(1) define m/z-dependent, increasing m/z bin sizes. ppm = 0 (default) m/z bins defined sequence values smallest larges m/z value constant bin size binSize. ppm > 0 size bin increased addition ppm (upper) m/z boundary bin. maximal bin size (used largest m/z values) binSize plus ppm parts-per-million largest m/z value peaks data set.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"data.frame, row representing (mz-rt) feature (.e. peak group) columns: \"mzmed\": median peaks' apex mz values. \"mzmin\": smallest mz value peaks' apex within feature. \"mzmax\":largest mz value peaks' apex within feature. \"rtmed\": median peaks' retention times. \"rtmin\": smallest retention time peaks group. \"rtmax\": largest retention time peaks group. \"npeaks\": total number peaks assigned feature. \"peakidx\": list indices peaks feature peaks input matrix. Note number can larger total number samples, since multiple peaks sample assigned feature.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"overlapping slices along mz dimension, function calculates density distribution identified peaks along retention time axis groups peaks different samples close . See (Smith 2006) details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"default settings might appropriate LC/GC-MS setups, especially bw binSize parameter adjusted accordingly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_density.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Core API function for peak density based chromatographic peak grouping — do_groupChromPeaks_density","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the matrix with the identified peaks from the xcmsSet: pks <- chromPeaks(faahko_sub)  ## Perform the peak grouping with default settings: res <- do_groupChromPeaks_density(pks, sampleGroups = rep(1, 3))  ## The feature definitions: head(res) #>   mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks 1      peakidx #> 1 279.0 279.0 279.0 2787.765 2787.765 2787.766      2 2      11, 199 #> 2 286.2 286.2 286.2 3254.904 3250.992 3258.815      2 2     115, 205 #> 3 300.2 300.2 300.2 3387.143 3379.317 3390.271      4 3 35, 125,.... #> 4 301.0 301.0 301.0 2787.766 2786.200 2792.459      3 3  10, 97, 198 #> 5 305.1 305.1 305.1 2994.338 2994.338 2994.339      2 2      15, 203 #> 6 305.1 305.1 305.1 2923.917 2923.916 2923.917      2 2      14, 202"},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"do_groupChromPeaks_nearest function groups peaks across samples creating master peak list assigning corresponding peaks samples peak group (.e. feature). method inspired correspondence algorithm mzMine (Katajamaa 2006).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"","code":"do_groupChromPeaks_nearest(   peaks,   sampleGroups,   mzVsRtBalance = 10,   absMz = 0.2,   absRt = 15,   kNN = 10 )"},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"peaks matrix data.frame mz values retention times identified chromatographic peaks samples experiment. Required columns \"mz\", \"rt\" \"sample\". latter contain numeric values representing index sample peak found. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). mzVsRtBalance NearestPeaksParam: numeric(1) representing factor m/z values multiplied calculating (euclician) distance two peaks. absMz NearestPeaksParam MzClustParam: numeric(1) maximum tolerated distance m/z values. absRt NearestPeaksParam: numeric(1) maximum tolerated distance retention times. kNN NearestPeaksParam: integer(1) representing number nearest neighbors check.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"list elements \"featureDefinitions\" \"peakIndex\". \"featureDefinitions\" matrix, row representing (mz-rt) feature (.e. peak group) columns: \"mzmed\": median peaks' apex mz values. \"mzmin\": smallest mz value peaks' apex within feature. \"mzmax\":largest mz value peaks' apex within feature. \"rtmed\": median peaks' retention times. \"rtmin\": smallest retention time peaks feature. \"rtmax\": largest retention time peaks feature. \"npeaks\": total number peaks assigned feature. \"peakIndex\" list indices peaks feature peaks input matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupChromPeaks_nearest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for chromatic peak grouping using a nearest neighbor approach — do_groupChromPeaks_nearest","text":"Katajamaa M, Miettinen J, Oresic M: MZmine: Toolbox processing visualization mass spectrometry based molecular profile data. Bioinformatics 2006, 22:634-636.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":null,"dir":"Reference","previous_headings":"","what":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"do_groupPeaks_mzClust function performs high resolution correspondence single spectra samples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"","code":"do_groupPeaks_mzClust(   peaks,   sampleGroups,   ppm = 20,   absMz = 0,   minFraction = 0.5,   minSamples = 1 )"},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"peaks matrix data.frame mz values retention times identified chromatographic peaks samples experiment. Required columns \"mz\", \"rt\" \"sample\". latter contain numeric values representing index sample peak found. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). ppm MzClustParam: numeric(1) representing relative m/z error clustering/grouping (parts per million). PeakDensityParam: numeric(1) define m/z-dependent, increasing m/z bin sizes. ppm = 0 (default) m/z bins defined sequence values smallest larges m/z value constant bin size binSize. ppm > 0 size bin increased addition ppm (upper) m/z boundary bin. maximal bin size (used largest m/z values) binSize plus ppm parts-per-million largest m/z value peaks data set. absMz NearestPeaksParam MzClustParam: numeric(1) maximum tolerated distance m/z values. minFraction PeakDensityParam: numeric(1) defining minimum fraction samples least one sample group peaks present considered peak group (feature). minSamples PeakDensityParam: numeric(1) minimum number samples least one sample group peaks detected considered peak group (feature).","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"list elements \"featureDefinitions\" \"peakIndex\". \"featureDefinitions\" matrix, row representing (mz-rt) feature (.e. peak group) columns: \"mzmed\": median peaks' apex mz values. \"mzmin\": smallest mz value peaks' apex within feature. \"mzmax\": largest mz value peaks' apex within feature. \"rtmed\": always -1. \"rtmin\": always -1. \"rtmax\": always -1. \"npeaks\": total number peaks assigned feature. Note number can larger total number samples, since multiple peaks sample assigned group. \"peakIndex\" list indices peaks peak group peaks input matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/do_groupPeaks_mzClust.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Core API function for peak grouping using mzClust — do_groupPeaks_mzClust","text":"Saira . Kazmi, Samiran Ghosh, Dong-Guk Shin, Dennis W. Hill David F. GrantAlignment high resolution mass spectra: development heuristic approach metabolomics. Metabolomics, Vol. 2, . 2, 75-83 (2006)","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":null,"dir":"Reference","previous_headings":"","what":"Allocate double, integer, or logical matricies — doubleMatrix","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"Allocate double, integer, logical matricies one step without   copying memory around.","code":""},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"","code":"doubleMatrix(nrow = 0, ncol = 0) integerMatrix(nrow = 0, ncol = 0) logicalMatrix(nrow = 0, ncol = 0)"},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"nrow number matrix rows ncol number matrix columns","code":""},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"Matrix double, integer, logical values. Memory zeroed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/doubleMatrix.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Allocate double, integer, or logical matricies — doubleMatrix","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":null,"dir":"Reference","previous_headings":"","what":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"estimatePrecursorIntensity determines precursor intensity MS 2 spectrum based intensity respective signal neighboring MS 1 spectra (.e. based peak m/z matching precursor m/z MS 2 spectrum). Based parameter method either intensity peak previous MS 1 scan used (method = \"previous\") interpolation intensity previous subsequent MS1 scan used (method = \"interpolation\", considers also retention times two MS1 scans retention time MS2 spectrum).","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"","code":"estimatePrecursorIntensity(   x,   ppm = 10,   method = c(\"previous\", \"interpolation\"),   BPPARAM = bpparam() )"},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"x OnDiskMSnExp XCMSnExp object. ppm numeric(1) defining maximal acceptable difference (ppm) precursor m/z m/z corresponding peak MS 1 scan. method character(1) defining method precursor intensity determined (see description details). Defaults method = \"previous\". BPPARAM parallel processing setup. See bpparam() details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"numeric length equal number spectra x. NA returned MS 1 spectra matching peak MS 1 scan can found MS 2 spectrum","code":""},{"path":"https://sneumann.github.io/xcms/reference/estimatePrecursorIntensity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Estimate precursor intensity for MS level 2 spectra — estimatePrecursorIntensity","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":null,"dir":"Reference","previous_headings":"","what":"Empirically Transformed Gaussian function — etg","title":"Empirically Transformed Gaussian function — etg","text":"general function asymmetric chromatographic peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Empirically Transformed Gaussian function — etg","text":"","code":"etg(x, H, t1, tt, k1, kt, lambda1, lambdat, alpha, beta)"},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Empirically Transformed Gaussian function — etg","text":"x times evaluate function H peak height t1 time leading edge inflection point tt time trailing edge inflection point k1 leading edge parameter kt trailing edge parameter lambda1 leading edge parameter lambdat trailing edge parameter alpha leading edge parameter beta trailing edge parameter","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Empirically Transformed Gaussian function — etg","text":"function evaluated times x.","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Empirically Transformed Gaussian function — etg","text":"Jianwei Li. Development Evaluation Flexible Empirical Peak   Functions Processing Chromatographic Peaks. Anal. Chem., 69   (21), 4452-4462, 1997.   http://dx.doi.org/10.1021/ac970481d","code":""},{"path":"https://sneumann.github.io/xcms/reference/etg.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Empirically Transformed Gaussian function — etg","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":null,"dir":"Reference","previous_headings":"","what":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"Export feature table analysis MetaboAnalyst software (MetaboAnalystR R package).","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"","code":"exportMetaboAnalyst(   x,   file = NULL,   label,   value = \"into\",   digits = NULL,   groupnames = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"x XCMSnExp object identified chromatographic peaks grouped across samples. file character(1) defining file name. specified, matrix content returned. label either character(1) specifying phenodata column x defining sample grouping vector length samples x defining group assignment samples. value character(1) specifying value returned feature. See featureValues() details. digits integer(1) defining number significant digits used numeric. default NULL uses getOption(\"digits\"). See format() information. groupnames logical(1) whether row names resulting matrix feature IDs (groupnames = FALSE; default) IDs composed m/z retention time features (format M<m/z>T<rt> (groupnames = TRUE). See help groupnames function details. ... additional parameters passed featureValues() function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"file specified, function returns matrix format supported MetaboAnalyst.","code":""},{"path":"https://sneumann.github.io/xcms/reference/exportMetaboAnalyst.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Export data for use in MetaboAnalyst — exportMetaboAnalyst","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":null,"dir":"Reference","previous_headings":"","what":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"UPDATE: extractMsData plotMsData functions deprecated (x, \"data.frame\") plot(x, type = \"XIC\") (x OnDiskMSnExp XCMSnExp object) used instead. See examples . aware filtering raw object might however drop adjusted retention times. cases advisable use applyAdjustedRtime() function prior filtering. Extract data.frame retention time, mz intensity values file/sample provided rt-mz range (full data range rt mz defined).","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"","code":"# S4 method for OnDiskMSnExp extractMsData(object, rt, mz, msLevel = 1L)  # S4 method for XCMSnExp extractMsData(   object,   rt,   mz,   msLevel = 1L,   adjustedRtime = hasAdjustedRtime(object) )"},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"object XCMSnExp OnDiskMSnExp object. rt numeric(2) retention time range data extracted. mz numeric(2) mz range. msLevel integer defining MS level(s) data sub-setted prior extraction; defaults msLevel = 1L. adjustedRtime (extractMsData,XCMSnExp): logical(1) specifying adjusted raw retention times reported. Defaults adjusted retention times, present object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"list length equal number samples/files object. element data.frame columns \"rt\", \"mz\" \"\" retention time, mz intensity tuples file. data available mz-rt range file data.frame 0 rows returned file.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/extractMsData-method.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"DEPRECATED: Extract a data.frame containing MS data — extractMsData,OnDiskMSnExp-method","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the full MS data for a certain retention time range ## as a data.frame tmp <- filterRt(faahko_sub, rt = c(2800, 2900)) ms_all <- as(tmp, \"data.frame\") head(ms_all) #>   file       rt    mz     i #> 1    1 2800.284 200.1   968 #> 2    1 2800.284 201.0  1273 #> 3    1 2800.284 201.9  1729 #> 4    1 2800.284 203.0   859 #> 5    1 2800.284 205.0 24136 #> 6    1 2800.284 206.0  3913 nrow(ms_all) #> [1] 79960"},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding of LC-MS features — feature-grouping","title":"Compounding of LC-MS features — feature-grouping","text":"Feature compounding aims identifying grouping LC-MS features representing different ions adducts (including isotopes) originating compound. MsFeatures package provides general framework functionality group features based different properties. groupFeatures methods XcmsExperiment() XCMSnExp objects implemented xcms extend enable compounding LC-MS data considering also e.g. feature peak shaped. Note functions simply define feature groups actually aggregate combine features. See MsFeatures::groupFeatures() overview general feature grouping concept well details individual settings parameters. available options groupFeatures xcms preprocessing results (.e. XcmsExperiment XCMSnExp objects correspondence analysis groupChromPeaks()) : Grouping similar retention times: groupFeatures-similar-rtime(). Grouping similar feature values across samples: AbundanceSimilarityParam(). Grouping similar peak shape extracted ion chromatograms: EicSimilarityParam(). ideal workflow grouping features sequentially perform methods (listed order). Compounded feature groups can accessed featureGroups function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding of LC-MS features — feature-grouping","text":"","code":"# S4 method for XcmsResult featureGroups(object)  # S4 method for XcmsResult featureGroups(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding of LC-MS features — feature-grouping","text":"object XcmsExperiment() XCMSnExp() object LC-MS pre-processing results. value featureGroups<-: replacement feature groups object. length 1 length equal number features object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/feature-grouping.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding of LC-MS features — feature-grouping","text":"Johannes Rainer, Mar Garcia-Aloy, Vinicius Veri Hernandes","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract ion chromatograms for each feature — featureChromatograms","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"Extract ion chromatograms features XcmsExperiment XCMSnExp object. function returns feature extracted ion chromatograms (along associated chromatographic peaks) sample. chromatogram extracted m/z - rt region includes chromatographic peaks feature. default, region defined using range chromatographic peaks' m/z retention times (mzmin = min, mzmax = max, rtmin = min rtmax = max). features, depending data, m/z rt range can thus relatively large. boundaries m/z - rt region can also restricted changing parameters mzmin, mzmax, rtmin rtmax different functions, median. default chromatographic peaks associated feature included returned XChromatograms object. object XCMSnExp object parameter include allows also return chromatographic peaks apex position within selected region (include = \"apex_within\") chromatographic peak overlapping m/z retention time range (include = \"\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"","code":"featureChromatograms(object, ...)  # S4 method for XcmsExperiment featureChromatograms(   object,   expandRt = 0,   expandMz = 0,   aggregationFun = \"max\",   features = character(),   return.type = \"XChromatograms\",   chunkSize = 2L,   mzmin = min,   mzmax = max,   rtmin = min,   rtmax = max,   ...,   progressbar = TRUE,   BPPARAM = bpparam() )  # S4 method for XCMSnExp featureChromatograms(   object,   expandRt = 0,   aggregationFun = \"max\",   features,   include = c(\"feature_only\", \"apex_within\", \"any\", \"all\"),   filled = FALSE,   n = length(fileNames(object)),   value = c(\"maxo\", \"into\"),   expandMz = 0,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"object XcmsExperiment XCMSnExp object grouped chromatographic peaks. ... optional arguments passed along chromatogram() function. expandRt numeric(1) expand retention time range chromatographic peak constant value side. expandMz numeric(1) expand m/z range chromatographic peak constant value side. aware extending m/z range extracted EIC might longer represent actual identified chromatographic peak intensities potential additional mass peaks within spectra aggregated final reported intensity value per spectrum (retention time). aggregationFun character(1) specifying name used aggregate intensity values across m/z value range retention time. default \"max\" returns base peak chromatogram. features integer, character logical defining subset features chromatograms returned. Can index features featureDefinitions, feature IDs (row names featureDefinitions) logical vector. return.type character(1) defining result returned. present return.type = \"XChromatograms\" supported results thus returned XChromatograms() object. chunkSize object XcmsExperiment: integer(1) defining number files data loaded time memory. Defaults chunkSize = 2L. mzmin function defining lower boundary m/z region EIC integrated defined. Defaults mzmin = min thus smallest \"mzmin\" value chromatographic peaks feature used. mzmax function defining upper boundary m/z region EIC integrated defined. Defaults mzmax = max thus largest \"mzmax\" value chromatographic peaks feature used. rtmin function defining lower boundary rt region EIC integrated defined. Defaults rtmin = min thus smallest \"rtmin\" value chromatographic peaks feature used. rtmax function defining upper boundary rt region EIC integrated defined. Defaults rtmax = max thus largest \"rtmax\" value chromatographic peaks feature used. progressbar logical(1) defining whether progress bar shown. BPPARAM object XcmsExperiment: parallel processing setup. Defaults BPPARAM = bpparam(). See bpparam() information. include object XCMSnExp: character(1) defining chromatographic peaks (related feature definitions) included returned XChromatograms(). Defaults \"feature_only\"; See description options details. filled object XCMSnExp: logical(1) whether filled-peaks included result object. default filled = FALSE, .e. detected peaks reported. n object XCMSnExp: integer(1) optionally specify number top n samples EIC extracted. value object XCMSnExp: character(1) specifying column used sort samples. Can either \"maxo\" (default) \"\" use maximal peak intensity integrated peak area, respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"XChromatograms() object. future, depending parameter return.type, data might returned different object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"EIC data feature extracted every sample using m/z - rt area. EIC sample thus exactly represent signal actually identified chromatographic peak sample. chromPeakChromatograms() function allow extract actual EIC chromatographic peak specific sample. See also examples . Parameters include, filled, n value supported object XCMSnExp. extracting EICs top n samples can happen one features specified features dropped detected peak top n samples. chance happen smaller x contains also filled-peaks (fillChromPeaks).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureChromatograms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract ion chromatograms for each feature — featureChromatograms","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Perform correspondence analysis xdata <- groupChromPeaks(faahko_sub,     param = PeakDensityParam(minFraction = 0.8, sampleGroups = rep(1, 3)))  ## Get the feature definitions featureDefinitions(xdata) #>      mzmed mzmin mzmax    rtmed    rtmin    rtmax npeaks 1      peakidx #> FT01 300.2 300.2 300.2 3387.143 3379.317 3390.271      4 3 35, 125,.... #> FT02 301.0 301.0 301.0 2787.766 2786.200 2792.459      3 3  10, 97, 198 #> FT03 326.2 326.2 326.2 3416.877 3415.311 3424.700      3 3 37, 128, 213 #> FT04 328.2 328.2 328.2 3626.581 3618.755 3631.274      3 3 64, 159, 231 #> FT05 329.2 329.2 329.2 3626.581 3618.755 3631.274      3 3 63, 156, 230 #> FT06 343.0 343.0 343.0 2684.479 2678.218 2686.042      3 3   6, 92, 195 #> FT07 344.0 344.0 344.0 2682.914 2679.783 2686.042      3 3   3, 90, 193 #> FT08 354.2 354.2 354.2 3614.061 3610.930 3621.884      3 3 56, 153, 228 #> FT09 356.2 356.2 356.2 3830.025 3819.069 3833.153      3 3 78, 180, 241 #> FT10 365.0 365.0 365.0 2684.479 2679.783 2686.042      3 3   7, 93, 196 #> FT11 438.3 438.3 438.3 4052.248 4047.552 4060.071      3 3 81, 186, 248 #> FT12 496.2 496.2 496.2 3335.498 3316.719 3340.194      3 3 47, 143, 220 #> FT13 496.2 496.2 496.2 3402.791 3384.012 3410.617      3 3 48, 144, 221 #> FT14 508.2 508.2 508.2 3524.859 3515.468 3529.552      3 3 67, 163, 232 #> FT15 509.2 509.2 509.2 3523.294 3512.338 3531.117      3 3 54, 160, 229 #> FT16 522.2 522.2 522.2 3387.142 3344.888 3434.091      6 3 52, 53, .... #> FT17 523.2 523.2 523.2 3387.142 3344.888 3434.091      6 3 38, 39, .... #> FT18 524.2 524.2 524.2 3676.658 3598.410 3704.828      4 3 74, 75, .... #> FT19 525.2 525.2 525.2 3690.742 3662.574 3706.393      3 3 72, 173, 238 #> FT20 532.2 532.2 532.2 3482.605 3476.344 3488.863      3 3 49, 137, 219 #> FT21 536.2 536.2 536.2 3712.653 3703.262 3718.912      3 3 73, 174, 237 #>      ms_level #> FT01        1 #> FT02        1 #> FT03        1 #> FT04        1 #> FT05        1 #> FT06        1 #> FT07        1 #> FT08        1 #> FT09        1 #> FT10        1 #> FT11        1 #> FT12        1 #> FT13        1 #> FT14        1 #> FT15        1 #> FT16        1 #> FT17        1 #> FT18        1 #> FT19        1 #> FT20        1 #> FT21        1  ## Extract ion chromatograms for the first 3 features. Parameter ## `features` can be either the feature IDs or feature indices. chrs <- featureChromatograms(xdata,     features = rownames(featureDefinitions)[1:3])  ## Plot the EIC for the first feature using different colors for each file. plot(chrs[1, ], col = c(\"red\", \"green\", \"blue\"))   ## The EICs for all 3 samples use the same m/z and retention time range, ## which was defined using the `featureArea` function: featureArea(xdata, features = rownames(featureDefinitions(xdata))[1:3],     mzmin = min, mzmax = max, rtmin = min, rtmax = max) #>      mzmin mzmax    rtmin    rtmax #> FT01 300.2 300.2 3362.104 3413.746 #> FT02 301.0 301.0 2765.855 2820.630 #> FT03 326.2 326.2 3391.837 3449.740  ## To extract the actual (exact) EICs for each chromatographic peak of ## a feature in each sample, the `chromPeakChromatograms` function would ## need to be used instead. Below we extract the EICs for all ## chromatographic peaks of the first feature. We need to first get the ## IDs of all chromatographic peaks assigned to the first feature: peak_ids <- rownames(chromPeaks(xdata))[featureDefinitions(xdata)$peakidx[[1L]]]  ## We can now pass these to the `chromPeakChromatograms` function with ## parameter `peaks`: eic_1 <- chromPeakChromatograms(xdata, peaks = peak_ids)  ## To plot these into a single plot we need to use the ## `plotChromatogramsOverlay` function: plotChromatogramsOverlay(eic_1)"},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract spectra associated with features — featureSpectra","title":"Extract spectra associated with features — featureSpectra","text":"function returns spectra associated identified features input object. default, spectra returned features (MS levels), parameter features allows specify/select features result returned. Parameter msLevel allows define whether MS level 1 2 spectra returned. msLevel = 1L MS1 spectra within retention time range chromatographic peak (respective data file) associated feature returned. Note samples peak identified (even filled-) spectra returned. msLevel = 2L MS2 spectra retention time within retention time range precursor m/z within m/z range chromatographic peak feature returned. See also chromPeakSpectra() (used internally extract spectra chromatographic peak feature) additional information, specifically also parameter method. default (method = \"\") spectra associated chromatographic peaks feature returned. option method, single spectrum per chromatographic peak returned (hence multiple spectra per feature). ID chromatographic peak (.e. row name chromPeaks) feature (.e., row name featureDefinitions) available returned Spectra() spectra variables \"peak_id\" \"feature_id\", respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract spectra associated with features — featureSpectra","text":"","code":"featureSpectra(object, ...)  # S4 method for XcmsExperiment featureSpectra(   object,   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   skipFilled = FALSE,   return.type = c(\"Spectra\", \"List\"),   features = character(),   ... )  # S4 method for XCMSnExp featureSpectra(   object,   msLevel = 2L,   expandRt = 0,   expandMz = 0,   ppm = 0,   skipFilled = FALSE,   return.type = c(\"MSpectra\", \"Spectra\", \"list\", \"List\"),   features = character(),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract spectra associated with features — featureSpectra","text":"object XcmsExperiment XCMSnExp object feature defitions. ... additional arguments passed along chromPeakSpectra(), method. msLevel integer(1) defining MS level spectra returned. expandRt numeric(1) expand retention time range peak constant value side. expandMz numeric(1) expand m/z range peak constant value side. ppm numeric(1) expand m/z range peak (side) value dependent peak's m/z. skipFilled logical(1) whether spectra filled-peaks reported . return.type character(1) defining type result object returned. features character, logical integer allowing specify subset features featureDefinitions spectra returned (providing either ID, logical vector length nrow(featureDefinitions(x)) index featureDefinitions(x)). parameter overrides skipFilled supported return.type either \"Spectra\" \"List\".","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract spectra associated with features — featureSpectra","text":"function returns either Spectra() (return.type = \"Spectra\") List Spectra (return.type = \"List\"). latter, order length matches parameter features (features defined order features featureDefinitions(object)). Spectra variables \"peak_id\" \"feature_id\" define chromatographic peak feature individual spectrum associated .","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSpectra.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract spectra associated with features — featureSpectra","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":null,"dir":"Reference","previous_headings":"","what":"Simple feature summaries — featureSummary","title":"Simple feature summaries — featureSummary","text":"Simple function calculate feature summaries. include counts percentages samples chromatographic peak present feature counts percentages samples one chromatographic peak annotated feature. Also relative standard deviations (RSD) calculated integrated peak areas per feature across samples. `perSampleCounts = TRUE` also individual chromatographic peak counts per sample returned.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simple feature summaries — featureSummary","text":"","code":"featureSummary(   x,   group,   perSampleCounts = FALSE,   method = \"maxint\",   skipFilled = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simple feature summaries — featureSummary","text":"x [XcmsExperiment()] [XCMSnExp()] object correspondence results. group `numeric`, `logical`, `character` `factor` length `x` samples aggregate counts groups defined `group`. perSampleCounts `logical(1)` whether feature wise individual peak counts per sample returned . method `character` passed [featureValues()] function. See respective help page information. skipFilled `logical(1)` whether filled-peaks excluded (default) included summary calculation.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simple feature summaries — featureSummary","text":"`matrix` one row per feature columns: - `\"count\"`: total number samples peak found. - `\"perc\"`: percentage samples peak found. - `\"multi_count\"`: total number samples one peak   assigned feature. - `\"multi_perc\"`: percentage samples peak found,   also multiple peaks annotated feature. Example:   feature, least one peak detected 50 samples. 5 2 peaks   assigned feature. `\"multi_perc\"` case 10 - `\"rsd\"`: relative standard deviation (coefficient variation)   integrated peak area feature's peaks. - 4 columns repeated unique element (level) `group`   `group` provided. `perSampleCounts = TRUE` also one column sample returned peak counts per sample.","code":""},{"path":"https://sneumann.github.io/xcms/reference/featureSummary.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Simple feature summaries — featureSummary","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Gap Filling — fillChromPeaks","title":"Gap Filling — fillChromPeaks","text":"Gap filling integrate signal m/z-rt area feature (.e., chromatographic peak group) samples chromatographic peak feature identified add chromPeaks() matrix. filled-peaks indicated TRUE column \"is_filled\" result object's chromPeakData() data frame. method gap filling along settings can defined param argument. Two different approaches available: param = FillChromPeaksParam(): default original xcms code. Signal integrated m/z retention time range defined featureDefinitions() data frame, .e. \"rtmin\", \"rtmax\", \"mzmin\" \"mzmax\". method suggested underestimates actual peak area also available object XcmsExperiment object. See details information settings method. param = ChromPeakAreaParam(): area signal feature integrated defined based feature's chromatographic peak areas. m/z range default defined lower quartile chromatographic peaks' \"mzmin\" value upper quartile chromatographic peaks' \"mzmax\" values. retention time range area defined analogously. Alternatively, setting mzmin = median, mzmax = median, rtmin = median rtmax = median ChromPeakAreaParam, median \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" values detected chromatographic peaks feature used instead. contrast  FillChromPeaksParam approach method uses () identified chromatographic peaks feature define area signal integrated. expandMz,expandMz<-: getter setter     expandMz slot object. expandRt,expandRt<-: getter setter     expandRt slot object. ppm,ppm<-: getter setter     ppm slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Gap Filling — fillChromPeaks","text":"","code":"fillChromPeaks(object, param, ...)  # S4 method for XcmsExperiment,ChromPeakAreaParam fillChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  FillChromPeaksParam(   expandMz = 0,   expandRt = 0,   ppm = 0,   fixedMz = 0,   fixedRt = 0 )  fixedRt(object)  fixedMz(object)  ChromPeakAreaParam(   mzmin = function(z) quantile(z, probs = 0.25, names = FALSE),   mzmax = function(z) quantile(z, probs = 0.75, names = FALSE),   rtmin = function(z) quantile(z, probs = 0.25, names = FALSE),   rtmax = function(z) quantile(z, probs = 0.75, names = FALSE) )  # S4 method for FillChromPeaksParam expandMz(object)  # S4 method for FillChromPeaksParam expandMz(object) <- value  # S4 method for FillChromPeaksParam expandRt(object)  # S4 method for FillChromPeaksParam expandRt(object) <- value  # S4 method for FillChromPeaksParam ppm(object)  # S4 method for FillChromPeaksParam ppm(object) <- value  # S4 method for XCMSnExp,FillChromPeaksParam fillChromPeaks(object, param, msLevel = 1L, BPPARAM = bpparam())  # S4 method for XCMSnExp,ChromPeakAreaParam fillChromPeaks(object, param, msLevel = 1L, BPPARAM = bpparam())  # S4 method for XCMSnExp,missing fillChromPeaks(object, param, BPPARAM = bpparam(), msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Gap Filling — fillChromPeaks","text":"object XcmsExperiment XCMSnExp object identified grouped chromatographic peaks. param ChromPeakAreaParam FillChromPeaksParam object defining approach used (see details section). ... currently ignored. msLevel integer(1) defining MS level peak filling performed (defaults msLevel = 1L). peak filling one MS level time supported, fill peaks MS level 1 2 run first using msLevel = 1 (returned result object) msLevel = 2. chunkSize fillChromPeaks object XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM Parallel processing settings. expandMz FillChromPeaksParam: numeric(1) defining value mz width peaks expanded. peak expanded mz direction expandMz * original m/z width. value 0 means expansion, value 1 grows peak 1 * m/z width peak resulting peaks twice original size m/z direction (expansion half m/z width sides). expandRt FillChromPeaksParam: numeric(1), expandMz retention time width. ppm FillChromPeaksParam: numeric(1) optionally specifying ppm m/z width peak region expanded. peaks m/z width smaller mean(c(mzmin, mzmax)) * ppm / 1e6, mzmin replaced mean(c(mzmin, mzmax)) - (mean(c(mzmin, mzmax)) * ppm / 2 / 1e6) mzmax mean(c(mzmin, mzmax)) + (mean(c(mzmin, mzmax)) * ppm / 2 / 1e6). applied eventually expanding m/z width using expandMz parameter. fixedMz FillChromPeaksParam: numeric(1) defining constant factor m/z width feature expanded. m/z width expanded sides fixedMz (.e. fixedMz subtracted lower m/z added upper m/z). expansion applied expandMz ppm. fixedRt FillChromPeaksParam: numeric(1) defining constant factor retention time width factor expanded. rt width expanded sides fixedRt (.e. fixedRt subtracted lower rt added upper rt). expansion applied expandRt. mzmin function applied values \"mzmin\" column chromatographic peaks feature define lower m/z value area signal feature integrated. Defaults mzmin = function(z) quantile(z, probs = 0.25) hence using 25% quantile values. mzmax function applied values \"mzmax\" column chromatographic peaks feature define upper m/z value area signal feature integrated. Defaults  mzmax = function(z) quantile(z, probs = 0.75) hence using 75% quantile values. rtmin function applied values \"rtmin\" column chromatographic peaks feature define lower rt value area signal feature integrated. Defaults rtmin = function(z) quantile(z, probs = 0.25) hence using 25% quantile values. rtmax function applied values \"rtmax\" column chromatographic peaks feature define upper rt value area signal feature integrated. Defaults rtmax = function(z) quantile(z, probs = 0.75) hence using 75% quantile values. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Gap Filling — fillChromPeaks","text":"XcmsExperiment  XCMSnExp object previously missing chromatographic peaks features filled chromPeaks() matrix. FillChromPeaksParam function returns  FillChromPeaksParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Gap Filling — fillChromPeaks","text":"correspondence (.e. grouping chromatographic peaks across samples) always features (peak groups) include peaks every sample. fillChromPeaks method defines intensity values features missing samples integrating signal m/z-rt region feature. Two different approaches define region available: ChromPeakAreaParam region defined based detected chromatographic peaks feature, FillChromPeaksParam region defined based m/z retention times feature (represent m/z retentention times apex position associated chromatographic peaks). latter approach various parameters available increase area signal integrated, either constant value (fixedMz fixedRt) feature-relative amount (expandMz expandRt). Adjusted retention times used available. Based peak finding algorithm used identify (chromatographic) peaks, different internal functions used guarantee integrated peak signal matches much possible peak signal integration used peak detection. peaks identified matchedFilter() method, signal integration performed profile matrix generated settings used also peak finding (using bin size example). direct injection data peaks identified MSW algorithm signal integrated along mz dimension. methods complete (raw) signal within area used.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Gap Filling — fillChromPeaks","text":"expandMz,expandRt,ppm,fixedMz,fixedRt See corresponding parameter . rtmin,rtmax,mzmin,mzmax See corresponding parameter .","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Gap Filling — fillChromPeaks","text":"reported \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" filled peaks represents actual MS area signal integrated. peak filled signal present file/sample respective mz-rt area. samples still show NA matrix returned featureValues() method.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Gap Filling — fillChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Gap Filling — fillChromPeaks","text":"","code":"## Load a test data set with identified chromatographic peaks res <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Perform the correspondence. We assign all samples to the same group. res <- groupChromPeaks(res,     param = PeakDensityParam(sampleGroups = rep(1, length(fileNames(res)))))  ## For how many features do we lack an integrated peak signal? sum(is.na(featureValues(res))) #> [1] 26  ## Filling missing peak data using the peak area from identified ## chromatographic peaks. res <- fillChromPeaks(res, param = ChromPeakAreaParam())  ## How many missing values do we have after peak filling? sum(is.na(featureValues(res))) #> [1] 2  ## Get the peaks that have been filled in: fp <- chromPeaks(res)[chromPeakData(res)$is_filled, ] head(fp) #>          mz mzmin mzmax       rt    rtmin    rtmax     into intb   maxo sn #> CP249 286.2 286.2 286.2 3252.556 3236.907 3274.857  1288521   NA  78048 NA #> CP250 380.1 380.1 380.1 3193.087 3132.054 3216.561  2094879   NA  35928 NA #> CP251 447.2 447.2 447.2 3858.193 3828.068 3914.140  2156784   NA  61256 NA #> CP252 497.2 497.2 497.2 3382.447 3347.236 3438.785 10344783   NA 310912 NA #> CP253 510.2 510.2 510.2 3750.211 3741.213 3799.898  1927180   NA  61368 NA #> CP254 531.2 531.2 531.2 3340.193 3328.848 3383.620  7589092   NA 254464 NA #>       sample #> CP249      1 #> CP250      1 #> CP251      1 #> CP252      1 #> CP253      1 #> CP254      1  ## Get the process history step along with the parameters used to perform ## The peak filling: ph <- processHistory(res, type = \"Missing peak filling\")[[1]] ph #> Object of class \"XProcessHistory\" #>  type: Missing peak filling  #>  date: Fri Jan 26 20:05:57 2024  #>  info:   #>  fileIndex: 1,2,3  #>  Parameter class: ChromPeakAreaParam  #>  MS level(s) 1   ## The parameter class: ph@param #> Object of class:  ChromPeakAreaParam  #>  Parameters: #>  - rtmin: function (z)  #> quantile(z, probs = 0.25, names = FALSE) #> <environment: 0x55c9fa2bec18> #>  - rtmax: function (z)  #> quantile(z, probs = 0.75, names = FALSE) #> <environment: 0x55c9fa2bec18> #>  - mzmin: function (z)  #> quantile(z, probs = 0.25, names = FALSE) #> <environment: 0x55c9fa2bec18> #>  - mzmax: function (z)  #> quantile(z, probs = 0.75, names = FALSE) #> <environment: 0x55c9fa2bec18>  ## It is also possible to remove filled-in peaks: res <- dropFilledChromPeaks(res)  sum(is.na(featureValues(res))) #> [1] 26"},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Integrate areas of missing peaks — fillPeaks-methods","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"sample, identify peak groups sample   represented. peak groups, integrate signal   region peak group create new peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"object = \"xcmsSet\" fillPeaks(object, method=\"\")","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"object xcmsSet object method filling method","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"peak grouping, always peak groups   include peaks every sample. method produces intensity   values missing samples integrating raw data peak   group region. According type raw-data 2 different   methods available. filling gcms/lcms data method \"chrom\" integrates   raw-data chromatographic domain, whereas \"MSW\" used   peaklists without retention-time information like   direct-infusion spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Integrate areas of missing peaks — fillPeaks-methods","text":"xcmsSet objects filled peak groups.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"sample, identify peak groups sample   represented. peak groups, integrate signal   region peak group create new peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"object = \"xcmsSet\" fillPeaks.MSW(object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"object xcmsSet object","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"peak grouping, always peak groups   include peaks every sample. method produces intensity   values missing samples integrating raw data peak   group region. given group, start ending m/z values   integration defined median start   end points detected peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"contrast fillPeaks.chrom method maximum   intensity reported column \"maxo\" maximum   intensity measured expected peak area (defined columns   \"mzmin\" \"mzmax\"), largest intensity mz   value(s) closest \"mzmed\" feature.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.MSW-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Integrate areas of missing peaks in FTICR-MS data — fillPeaks.MSW-methods","text":"xcmsSet objects filled peak groups.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Integrate areas of missing peaks — fillPeaks.chrom-methods","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"sample, identify peak groups sample   represented. peak groups, integrate signal   region peak group create new peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"object = \"xcmsSet\" fillPeaks.chrom(object, nSlaves=0,expand.mz=1,expand.rt=1,     BPPARAM = bpparam())","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"object xcmsSet object nSlaves (DEPRECATED): number slaves/cores used     parallel peak filling.     MPI used installed, otherwise snow package employed     multicore support. none two packages available uses     parallel package parallel processing multiple CPUs     current machine. Users advised use BPPARAM     parameter instead. expand.mz Expansion factor m/z range used integration. expand.rt Expansion factor rentention time range used     integration. BPPARAM allows define specific parallel processing setup   current task (see bpparam   BiocParallel package help information). default uses   globally defined parallel setup.","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"peak grouping, always peak groups   include peaks every sample. method produces intensity   values missing samples integrating raw data peak   group region. given group, start ending retention   time points integration defined median start   end points detected peaks. start end m/z   values similarly determined. Intensities   can still zero, rather unusual intensity peak.   case e.g. raw data threshholded,   integration area contains actual raw intensities, one sample   miscalibrated, thet raw data points (just) outside   integration area. Importantly, retention time correction data available,   alignment information used precisely integrate   propper region raw data. corrected retention time   beyond end raw data, value --number (NaN).","code":""},{"path":"https://sneumann.github.io/xcms/reference/fillPeaks.chrom-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Integrate areas of missing peaks — fillPeaks.chrom-methods","text":"xcmsSet objects filled peak groups (maxo).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":null,"dir":"Reference","previous_headings":"","what":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"functions allow filter (subset) MChromatograms() XChromatograms() objects, .e. sets chromatographic data, without changing data (intensity retention times) within individual chromatograms (Chromatogram() objects). filterColumnsIntensityAbove: subsets MChromatograms objects keeping columns (samples) value larger provided threshold rows (.e. = \"\" column kept chromatograms column value larger threshold = \"\" chromatograms column fulfill criteria). Parameter value allows define value comparison performed, value = \"bpi\" maximum intensity chromatogram compared threshold, value = \"tic\" total sum intensities chromatogram compared threshold. XChromatogramsobject,value = \"maxo\"andvalue = \"\"supported compares largest intensity identified chromatographic peaks chromatogram withthreshold`, integrated peak area, respectively. filterColumnsKeepTop: subsets MChromatograms object keeping top n columns sorted value specified sortBy. detail, column value defined sortBy extracted chromatogram aggregated using aggregationFun. Thus, default, chromatogram maximum intensity determined (sortBy = \"bpi\") values summed chromatograms column (aggregationFun = sum). columns sorted values top n columns retained returned MChromatograms. Similar filterColumnsIntensityAbove function, function allows use XChromatograms objects sort columns column sortBy = \"maxo\" sortBy = \"\" chromPeaks matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"","code":"# S4 method for MChromatograms filterColumnsIntensityAbove(   object,   threshold = 0,   value = c(\"bpi\", \"tic\"),   which = c(\"any\", \"all\") )  # S4 method for MChromatograms filterColumnsKeepTop(   object,   n = 1L,   sortBy = c(\"bpi\", \"tic\"),   aggregationFun = sum )  # S4 method for XChromatograms filterColumnsIntensityAbove(   object,   threshold = 0,   value = c(\"bpi\", \"tic\", \"maxo\", \"into\"),   which = c(\"any\", \"all\") )  # S4 method for XChromatograms filterColumnsKeepTop(   object,   n = 1L,   sortBy = c(\"bpi\", \"tic\", \"maxo\", \"into\"),   aggregationFun = sum )"},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"object MChromatograms() XChromatograms() object. threshold filterColumnsIntensityAbove: numeric(1) threshold value compare . value character(1) defining value used comparison sorting. Can value = \"bpi\" (default) use maximum intensity per chromatogram value = \"tic\" use sum intensities per chromatogram. XChromatograms() objects also value = \"maxo\" value = \"\" supported use maximum intensity integrated area identified chromatographic peaks chromatogram. filterColumnsIntensityAbove: character(1) defining whether (= \"\", default) (= \"\") chromatograms column fulfill criteria column kept. n filterColumnsKeepTop: integer(1) specifying number columns returned. n rounded closest (larger) integer value. sortBy filterColumnsKeepTop: value columns ordered determine top n columns. Can either sortBy = \"bpi\" (default), case maximum intensity column's chromatograms used, sortBy = \"tic\" use total intensity sum chromatograms. XChromatograms() objects also value = \"maxo\" value = \"\" supported use maximum intensity integrated area identified chromatographic peaks chromatogram. aggregationFun filterColumnsKeepTop: function used aggregate (combine) values chromatograms column. Defaults aggregationFun = sum case sum values used rank columns. Alternatively mean, median similar function can used.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"filtered MChromatograms (XChromatograms) object number rows (EICs) eventually lower number columns (samples).","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/filter-MChromatograms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Filtering sets of chromatographic data — filterColumnsIntensityAbove,MChromatograms-method","text":"","code":"chr1 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(5, 29, 50, NA, 100, 12, 3, 4, 1, 3)) chr2 <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(80, 50, 20, 10, 9, 4, 3, 4, 1, 3)) chr3 <- Chromatogram(rtime = 3:9 + rnorm(7, sd = 0.3),     intensity = c(53, 80, 130, 15, 5, 3, 2))  chrs <- MChromatograms(list(chr1, chr2, chr1, chr3, chr2, chr3),     ncol = 3, byrow = FALSE) chrs #> MChromatograms with 2 rows and 3 columns #>                   1              2              3 #>      <Chromatogram> <Chromatogram> <Chromatogram> #> [1,]     length: 10     length: 10     length: 10 #> [2,]     length: 10      length: 7      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  #### filterColumnsIntensityAbove ## ## Keep all columns with for which the maximum intensity of any of its ## chromatograms is larger 90 filterColumnsIntensityAbove(chrs, threshold = 90) #> MChromatograms with 2 rows and 3 columns #>                   1              2              3 #>      <Chromatogram> <Chromatogram> <Chromatogram> #> [1,]     length: 10     length: 10     length: 10 #> [2,]     length: 10      length: 7      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  ## Require that ALL chromatograms in a column have a value larger 90 filterColumnsIntensityAbove(chrs, threshold = 90, which = \"all\") #> MChromatograms with 2 rows and 1 column #>                   2 #>      <Chromatogram> #> [1,]     length: 10 #> [2,]      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  ## If none of the columns fulfills the criteria no columns are returned filterColumnsIntensityAbove(chrs, threshold = 900) #> MChromatograms with 2 rows and 0 columns #> phenoData with 0 variables #> featureData with 0 variables  ## Filtering XChromatograms allow in addition to filter on the columns ## \"maxo\" or \"into\" of the identified chromatographic peaks within each ## chromatogram.  #### filterColumnsKeepTop ## ## Keep the 2 columns with the highest sum of maximal intensities in their ## chromatograms filterColumnsKeepTop(chrs, n = 1) #> MChromatograms with 2 rows and 1 column #>                   2 #>      <Chromatogram> #> [1,]     length: 10 #> [2,]      length: 7 #> phenoData with 0 variables #> featureData with 0 variables  ## Keep the 50 percent of columns with the highest total sum of signal. Note ## that n will be rounded to the next larger integer value filterColumnsKeepTop(chrs, n = 0.5 * ncol(chrs), sortBy = \"tic\") #> MChromatograms with 2 rows and 2 columns #>                   2              3 #>      <Chromatogram> <Chromatogram> #> [1,]     length: 10     length: 10 #> [2,]      length: 7      length: 7 #> phenoData with 0 variables #> featureData with 0 variables"},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":null,"dir":"Reference","previous_headings":"","what":"Apply an convolution filter using an FFT — filtfft","title":"Apply an convolution filter using an FFT — filtfft","text":"Expands vector length filter convolutes   using two successive FFTs.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Apply an convolution filter using an FFT — filtfft","text":"","code":"filtfft(y, filt)"},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Apply an convolution filter using an FFT — filtfft","text":"y numeric vector data filtered filt filter length nextn(length(y))","code":""},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Apply an convolution filter using an FFT — filtfft","text":"numeric vector length y.","code":""},{"path":"https://sneumann.github.io/xcms/reference/filtfft.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Apply an convolution filter using an FFT — filtfft","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":null,"dir":"Reference","previous_headings":"","what":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"findChromPeaks Chromatogram MChromatograms object CentWaveParam parameter object performs centWave-based peak detection purely chromatographic data. See centWave details method CentWaveParam details parameter class. Note settings CentWaveParam used. See peaksWithCentWave() arguments used peak detection purely chromatographic data. chromatographic peak detection, identified peaks can also refined refineChromPeaks() method, can help reduce peak detection artifacts.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"","code":"# S4 method for Chromatogram,CentWaveParam findChromPeaks(object, param, ...)  # S4 method for MChromatograms,CentWaveParam findChromPeaks(object, param, BPPARAM = bpparam(), ...)  # S4 method for MChromatograms,MatchedFilterParam findChromPeaks(object, param, BPPARAM = BPPARAM, ...)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"object Chromatogram MChromatograms object. param CentWaveParam object specifying settings peak detection. See peaksWithCentWave() description arguments used peak detection. ... currently ignored. BPPARAM parameter class specifying parallel processing performed (XChromatograms objects). defaults bpparam(). See bpparam() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"called Chromatogram object, method returns XChromatogram object identified peaks. See peaksWithCentWave() details peak matrix content.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-CentWaveParam.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"centWave-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,CentWaveParam-method","text":"","code":"## Loading a test data set with identified chromatographic peaks faahko_sub <- loadXcmsData(\"faahko_sub2\") faahko_sub <- filterRt(faahko_sub, c(2500, 3700)) #> Filter spectra  ## od <- as(filterFile(faahko_sub, 1L), \"MsExperiment\")  ## Extract chromatographic data for a small m/z range chr <- chromatogram(od, mz = c(272.1, 272.3))[1, 1]  ## Identify peaks with default settings xchr <- findChromPeaks(chr, CentWaveParam()) xchr #> Object of class: XChromatogram #> Intensity values aggregated using: sum  #> length of object: 766 #> from file: 1 #> mz range: [272.1, 272.3] #> rt range: [2501.378, 3698.567] #> MS level: 1 #> Identified chromatographic peaks (3): #>  rt\trtmin\trtmax\tinto\tmaxo\tsn (1 more column(s)) #>  2734.556\t2709.517\t2770.55\t286653.221615386\t8907\t15 #>  3117.97\t3097.625\t3139.879\t372049.599925925\t16496\t33 #>  3487.299\t3457.565\t3509.208\t93049.7314242425\t2651\t15  ## Plot data and identified peaks. plot(xchr)   ## Perform peak detection on an MChromatograms object od3 <- readMSData(c(system.file(\"cdf/KO/ko15.CDF\", package = \"faahKO\"),     system.file(\"cdf/KO/ko16.CDF\", package = \"faahKO\"),     system.file(\"cdf/KO/ko18.CDF\", package = \"faahKO\")),     mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Disable parallel processing for this example register(SerialParam())  ## Extract chromatograms for a m/z - retention time slice chrs <- chromatogram(od3, mz = 344, rt = c(2500, 3500))  ## Perform peak detection using CentWave xchrs <- findChromPeaks(chrs, param = CentWaveParam()) xchrs #> XChromatograms with 1 row and 3 columns #>             ko15.CDF        ko16.CDF        ko18.CDF #>      <XChromatogram> <XChromatogram> <XChromatogram> #> [1,]        peaks: 2        peaks: 2        peaks: 1 #> phenoData with 1 variables #> featureData with 5 variables #> - - - xcms preprocessing - - - #> Chromatographic peak detection: #>  method: centWave   ## Extract the identified chromatographic peaks chromPeaks(xchrs) #>            rt    rtmin    rtmax      into      intb   maxo sn row column #> [1,] 2589.015 2571.801 2612.490  592297.3  554533.9  27600 14   1      1 #> [2,] 2679.783 2646.919 2709.517 5210015.9 5071346.6 152320 30   1      1 #> [3,] 2593.710 2559.281 2637.529  722415.3  640838.6  30640 14   1      2 #> [4,] 2686.042 2648.483 2714.211 5700652.0 5587464.2 151744 45   1      2 #> [5,] 2682.914 2643.790 2731.427 5255689.5 4983388.9 125632 28   1      3  ## plot the result plot(xchrs)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"findChromPeaks Chromatogram MChromatograms object MatchedFilterParam parameter object performs matchedFilter-based peak detection purely chromatographic data. See matchedFilter details method MatchedFilterParam details parameter class. Note settings MatchedFilterParam used. See peaksWithMatchedFilter() arguments used peak detection purely chromatographic data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"","code":"# S4 method for Chromatogram,MatchedFilterParam findChromPeaks(object, param, ...)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"object Chromatogram MChromatograms object. param MatchedFilterParam object specifying settings peak detection. See peaksWithMatchedFilter() description arguments used peak detection. ... currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"called Chromatogram object, method returns matrix identified peaks. See peaksWithMatchedFilter() details matrix content.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-Chromatogram-MatchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"matchedFilter-based peak detection in purely chromatographic data — findChromPeaks,Chromatogram,MatchedFilterParam-method","text":"","code":"## Loading a test data set with identified chromatographic peaks faahko_sub <- loadXcmsData(\"faahko_sub2\") faahko_sub <- filterRt(faahko_sub, c(2500, 3700)) #> Filter spectra  ## od <- as(filterFile(faahko_sub, 1L), \"MsExperiment\")  ## Extract chromatographic data for a small m/z range chr <- chromatogram(od, mz = c(272.1, 272.3))[1, 1]  ## Identify peaks with default settings xchr <- findChromPeaks(chr, MatchedFilterParam())  ## Plot the identified peaks plot(xchr)"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":null,"dir":"Reference","previous_headings":"","what":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"centWave algorithm perform peak density wavelet based     chromatographic peak detection high resolution LC/MS data centroid     mode [Tautenhahn 2008]. CentWaveParam class allows specify settings     chromatographic peak detection using centWave method. Instances     created CentWaveParam constructor. findChromPeaks,OnDiskMSnExp,CentWaveParam method     performs chromatographic peak detection using centWave     algorithm samples OnDiskMSnExp     object. OnDiskMSnExp objects encapsule     experiment specific data load spectra data (mz intensity     values) fly original files applying also eventual     data manipulations. ppm,ppm<-: getter setter ppm     slot object. peakwidth,peakwidth<-: getter setter     peakwidth slot object. snthresh,snthresh<-: getter setter     snthresh slot object. prefilter,prefilter<-: getter setter     prefilter slot object. mzCenterFun,mzCenterFun<-: getter setter     mzCenterFun slot object. integrate,integrate<-: getter setter     integrate slot object. mzdiff,mzdiff<-: getter setter     mzdiff slot object. fitgauss,fitgauss<-: getter setter     fitgauss slot object. noise,noise<-: getter setter     noise slot object. verboseColumns,verboseColumns<-: getter     setter verboseColumns slot object. roiList,roiList<-: getter setter     roiList slot object. fistBaselineCheck,firstBaselineCheck<-: getter     setter firstBaselineCheck slot object. roiScales,roiScales<-: getter setter     roiScales slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"","code":"CentWaveParam(   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1L,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = numeric(),   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE )  # S4 method for OnDiskMSnExp,CentWaveParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for CentWaveParam ppm(object)  # S4 method for CentWaveParam ppm(object) <- value  # S4 method for CentWaveParam peakwidth(object)  # S4 method for CentWaveParam peakwidth(object) <- value  # S4 method for CentWaveParam snthresh(object)  # S4 method for CentWaveParam snthresh(object) <- value  # S4 method for CentWaveParam prefilter(object)  # S4 method for CentWaveParam prefilter(object) <- value  # S4 method for CentWaveParam mzCenterFun(object)  # S4 method for CentWaveParam mzCenterFun(object) <- value  # S4 method for CentWaveParam integrate(f)  # S4 method for CentWaveParam integrate(object) <- value  # S4 method for CentWaveParam mzdiff(object)  # S4 method for CentWaveParam mzdiff(object) <- value  # S4 method for CentWaveParam fitgauss(object)  # S4 method for CentWaveParam fitgauss(object) <- value  # S4 method for CentWaveParam noise(object)  # S4 method for CentWaveParam noise(object) <- value  # S4 method for CentWaveParam verboseColumns(object)  # S4 method for CentWaveParam verboseColumns(object) <- value  # S4 method for CentWaveParam roiList(object)  # S4 method for CentWaveParam roiList(object) <- value  # S4 method for CentWaveParam firstBaselineCheck(object)  # S4 method for CentWaveParam firstBaselineCheck(object) <- value  # S4 method for CentWaveParam roiScales(object)  # S4 method for CentWaveParam roiScales(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685 https://doi.org/10.1186/s12859-023-05533-4 information. object findChromPeaks:     OnDiskMSnExp  object containing MS-     experiment-relevant data. methods: parameter object. param CentWaveParam object containing settings centWave algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot. f integrate: CentWaveParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"CentWaveParam function returns CentWaveParam class instance settings specified chromatographic     peak detection centWave method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"centWave algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method identifies regions interest (ROIs) representing mass traces characterized regions less ppm m/z deviation consecutive scans LC/MS map. detail, starting single m/z, ROI extended m/z can found next scan (spectrum) difference mean m/z ROI smaller user defined ppm m/z. mean m/z ROI updated considering also newly included m/z value. ROIs , cleanup, analyzed using continuous wavelet transform (CWT) locate chromatographic peaks different scales. first analysis step skipped, regions interest passed via param parameter. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"ppm,peakwidth,snthresh,prefilter,mzCenterFun,integrate,mzdiff,fitgauss,noise,verboseColumns,roiList,firstBaselineCheck,roiScales,extendLengthMSW,verboseBetaColumns See corresponding parameter . Slots values exclusively accessed via corresponding getter setter methods listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports peak detection     OnDiskMSnExp objects (defined MSnbase     package). settings centWave algorithm can passed     CentWaveParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann \"Highly sensitive feature detection high resolution LC/MS\" BMC Bioinformatics 2008, 9:504","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"Ralf Tautenhahn, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWave.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Chromatographic peak detection using the centWave method — findChromPeaks-centWave","text":"","code":"## Create a CentWaveParam object. Note that the noise is set to 10000 to ## speed up the execution of the example - in a real use case the default ## value should be used, or it should be set to a reasonable value. cwp <- CentWaveParam(ppm = 20, noise = 10000, prefilter = c(3, 10000)) ## Change snthresh parameter snthresh(cwp) <- 25 cwp #> Object of class:  CentWaveParam  #>  Parameters: #>  - ppm: [1] 20 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 25 #>  - prefilter: [1]     3 10000 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 10000 #>  - verboseColumns: [1] FALSE #>  - roiList: list() #>  - firstBaselineCheck: [1] TRUE #>  - roiScales: numeric(0) #>  - extendLengthMSW: [1] FALSE #>  - verboseBetaColumns: [1] FALSE  ## Perform the peak detection using centWave on some of the files from the ## faahKO package. Files are read using the readMSData from the MSnbase ## package library(faahKO) library(xcms) fls <- dir(system.file(\"cdf/KO\", package = \"faahKO\"), recursive = TRUE,            full.names = TRUE) raw_data <- readMSData(fls[1], mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Perform the peak detection using the settings defined above. res <- findChromPeaks(raw_data, param = cwp) #> Detecting mass traces at 20 ppm ...  #> OK #> Detecting chromatographic peaks in 937 regions of interest ... #>  OK: 120 found. head(chromPeaks(res)) #>          mz mzmin mzmax       rt    rtmin    rtmax      into      intb   maxo #> CP001 453.2 453.2 453.2 2506.073 2501.378 2527.982 1007409.0 1007380.8  38152 #> CP002 302.0 302.0 302.0 2617.185 2595.275 2640.659  687146.6  671297.8  30552 #> CP003 344.0 344.0 344.0 2679.783 2646.919 2709.517 5210015.9 5135916.9 152320 #> CP004 381.0 381.0 381.0 2678.218 2637.529 2720.472 2180565.2 2023571.9  52504 #> CP005 430.1 430.1 430.1 2681.348 2639.094 2712.647 2395840.3 2299899.6  65752 #> CP006 366.0 366.0 366.0 2679.783 2642.224 2718.907 3365174.0 3279468.3  79928 #>          sn sample #> CP001 38151      1 #> CP002    46      1 #> CP003    68      1 #> CP004    37      1 #> CP005    42      1 #> CP006    49      1"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":null,"dir":"Reference","previous_headings":"","what":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"method performs two-step centWave-based chromatographic     peak detection: first centWave run peaks identified     location potential isotopes mz-retention time     predicted. second centWave run performed     regions interest (ROIs). final list chromatographic     peaks comprises non-overlapping peaks centWave runs. CentWavePredIsoParam class allows specify     settings two-step centWave-based peak detection considering also     predicted isotopes peaks identified first centWave run.     Instances created CentWavePredIsoParam     constructor. See also documentation     CentWaveParam methods arguments class     inherits. findChromPeaks,OnDiskMSnExp,CentWavePredIsoParam     method performs two-step centWave-based chromatographic peak detection     samples OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. snthreshIsoROIs,snthreshIsoROIs<-: getter     setter snthreshIsoROIs slot object. maxCharge,maxCharge<-: getter     setter maxCharge slot object. maxIso,maxIso<-: getter     setter maxIso slot object. mzIntervalExtension,mzIntervalExtension<-: getter     setter mzIntervalExtension slot object. polarity,polarity<-: getter     setter polarity slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"","code":"CentWavePredIsoParam(   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1L,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   roiList = list(),   firstBaselineCheck = TRUE,   roiScales = numeric(),   extendLengthMSW = FALSE,   verboseBetaColumns = FALSE,   snthreshIsoROIs = 6.25,   maxCharge = 3,   maxIso = 5,   mzIntervalExtension = TRUE,   polarity = \"unknown\" )  # S4 method for OnDiskMSnExp,CentWavePredIsoParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for CentWavePredIsoParam snthreshIsoROIs(object)  # S4 method for CentWavePredIsoParam snthreshIsoROIs(object) <- value  # S4 method for CentWavePredIsoParam maxCharge(object)  # S4 method for CentWavePredIsoParam maxCharge(object) <- value  # S4 method for CentWavePredIsoParam maxIso(object)  # S4 method for CentWavePredIsoParam maxIso(object) <- value  # S4 method for CentWavePredIsoParam mzIntervalExtension(object)  # S4 method for CentWavePredIsoParam mzIntervalExtension(object) <- value  # S4 method for CentWavePredIsoParam polarity(object)  # S4 method for CentWavePredIsoParam polarity(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2) expected approximate peak width chromatographic space. Given range (min, max) seconds. snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2): c(k, ) specifying prefilter step first analysis step (ROI detection). Mass traces retained contain least k peaks intensity >= . mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. roiList optional list regions--interest (ROI) representing detected mass traces. ROIs submitted first analysis step omitted chromatographic peak detection performed submitted ROIs. ROI expected following elements specified: scmin (start scan index), scmax (end scan index), mzmin (minimum m/z), mzmax (maximum m/z), length (number scans), intensity (summed intensity). ROI represented list elements single row data.frame. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. roiScales Optional numeric vector length equal roiList defining scale region interest roiList used centWave-wavelets. extendLengthMSW Option force centWave use scales running centWave rather truncating EIC length. Uses \"open\" method extend EIC integer base-2 length prior passed convolve rather default \"reflect\" method. See https://github.com/sneumann/xcms/issues/445 information. verboseBetaColumns Option calculate two additional metrics peak quality via comparison idealized bell curve. Adds beta_cor beta_snr chromPeaks output, corresponding Pearson correlation coefficient bell curve several degrees skew well estimate signal--noise using residuals best-fitting bell curve. See https://github.com/sneumann/xcms/pull/685 https://doi.org/10.1186/s12859-023-05533-4 information. snthreshIsoROIs numeric(1) defining signal noise ratio cutoff used second centWave run identify peaks predicted isotope ROIs. maxCharge integer(1) defining maximal isotope charge. Isotopes defined charges 1:maxCharge. maxIso integer(1) defining number isotope peaks predicted peak identified first centWave run. mzIntervalExtension logical(1) whether mz range predicted isotope ROIs extended increase detection low intensity peaks. polarity character(1) specifying polarity data. Currently used, \"positive\", \"negative\" \"unknown\" provided. object findChromPeaks:     OnDiskMSnExp  object containing MS-     experiment-relevant data. methods: parameter object. param CentWavePredIsoParam object settings chromatographic peak detection algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"CentWavePredIsoParam function returns  CentWavePredIsoParam class instance settings     specified two-step centWave-based peak detection considering also     isotopes. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"See centWave details centWave method. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"ppm,peakwidth,snthresh,prefilter,mzCenterFun,integrate,mzdiff,fitgauss,noise,verboseColumns,roiList,firstBaselineCheck,roiScales,extendLengthMSW,verboseBetaColumns,snthreshIsoROIs,maxCharge,maxIso,mzIntervalExtension,polarity See corresponding parameter .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports chromatographic peak     detection     OnDiskMSnExp objects (defined     MSnbase package). settings algorithm can     passed CentWavePredIsoParam object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"Hendrik Treutler, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-centWaveWithPredIsoROIs.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Two-step centWave peak detection considering also isotopes — findChromPeaks-centWaveWithPredIsoROIs","text":"","code":"## Create a param object p <- CentWavePredIsoParam(maxCharge = 4) ## Change snthresh parameter snthresh(p) <- 25 p #> Object of class:  CentWavePredIsoParam  #>  Parameters: #>  - snthreshIsoROIs: [1] 6.25 #>  - maxCharge: [1] 4 #>  - maxIso: [1] 5 #>  - mzIntervalExtension: [1] TRUE #>  - polarity: [1] \"unknown\" #>  - ppm: [1] 25 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 25 #>  - prefilter: [1]   3 100 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 0 #>  - verboseColumns: [1] FALSE #>  - roiList: list() #>  - firstBaselineCheck: [1] TRUE #>  - roiScales: numeric(0) #>  - extendLengthMSW: [1] FALSE #>  - verboseBetaColumns: [1] FALSE"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":null,"dir":"Reference","previous_headings":"","what":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"Massifquant Kalman filter (KF)-based chromatographic peak     detection XC-MS data centroid mode. identified peaks     can refined centWave method (see     findChromPeaks-centWave details centWave)     specifying withWave = TRUE. MassifquantParam class allows specify     settings chromatographic peak detection using massifquant     method eventually combination centWave algorithm. Instances     created MassifquantParam constructor. findChromPeaks,OnDiskMSnExp,MassifquantParam     method performs chromatographic peak detection using     massifquant algorithm samples     OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. ppm,ppm<-: getter setter ppm     slot object. peakwidth,peakwidth<-: getter setter     peakwidth slot object. snthresh,snthresh<-: getter setter     snthresh slot object. prefilter,prefilter<-: getter setter     prefilter slot object. mzCenterFun,mzCenterFun<-: getter setter     mzCenterFun slot object. integrate,integrate<-: getter setter     integrate slot object. mzdiff,mzdiff<-: getter setter     mzdiff slot object. fitgauss,fitgauss<-: getter setter     fitgauss slot object. noise,noise<-: getter setter     noise slot object. verboseColumns,verboseColumns<-: getter     setter verboseColumns slot object. criticalValue,criticalValue<-: getter     setter criticalValue slot object. consecMissedLimit,consecMissedLimit<-: getter     setter consecMissedLimit slot object. unions,unions<-: getter     setter unions slot object. checkBack,checkBack<-: getter     setter checkBack slot object. withWave,withWave<-: getter     setter withWave slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"","code":"MassifquantParam(   ppm = 25,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   mzCenterFun = \"wMean\",   integrate = 1L,   mzdiff = -0.001,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   criticalValue = 1.125,   consecMissedLimit = 2,   unions = 1,   checkBack = 0,   withWave = FALSE )  # S4 method for OnDiskMSnExp,MassifquantParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for MassifquantParam ppm(object)  # S4 method for MassifquantParam ppm(object) <- value  # S4 method for MassifquantParam peakwidth(object)  # S4 method for MassifquantParam peakwidth(object) <- value  # S4 method for MassifquantParam snthresh(object)  # S4 method for MassifquantParam snthresh(object) <- value  # S4 method for MassifquantParam prefilter(object)  # S4 method for MassifquantParam prefilter(object) <- value  # S4 method for MassifquantParam mzCenterFun(object)  # S4 method for MassifquantParam mzCenterFun(object) <- value  # S4 method for MassifquantParam integrate(f)  # S4 method for MassifquantParam integrate(object) <- value  # S4 method for MassifquantParam mzdiff(object)  # S4 method for MassifquantParam mzdiff(object) <- value  # S4 method for MassifquantParam fitgauss(object)  # S4 method for MassifquantParam fitgauss(object) <- value  # S4 method for MassifquantParam noise(object)  # S4 method for MassifquantParam noise(object) <- value  # S4 method for MassifquantParam verboseColumns(object)  # S4 method for MassifquantParam verboseColumns(object) <- value  # S4 method for MassifquantParam criticalValue(object)  # S4 method for MassifquantParam criticalValue(object) <- value  # S4 method for MassifquantParam consecMissedLimit(object)  # S4 method for MassifquantParam consecMissedLimit(object) <- value  # S4 method for MassifquantParam unions(object)  # S4 method for MassifquantParam unions(object) <- value  # S4 method for MassifquantParam checkBack(object)  # S4 method for MassifquantParam checkBack(object) <- value  # S4 method for MassifquantParam withWave(object)  # S4 method for MassifquantParam withWave(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"ppm numeric(1) defining maximal tolerated m/z deviation consecutive scans parts per million (ppm) initial ROI definition. peakwidth numeric(2). first element used massifquant, specifices minimum peak length time scans. withWave = TRUE second argument represents maximum peak length subject greater mininum peak length (see also documentation do_findChromPeaks_centWave). snthresh numeric(1) defining signal noise ratio cutoff. prefilter numeric(2). first argument used (withWave = TRUE); see findChromPeaks-centWave details. second argument specifies minimum threshold maximum intensity chromatographic peak must met. mzCenterFun Name function calculate m/z center chromatographic peak. Allowed : \"wMean\": intensity weighted mean peak's m/z values, \"mean\": mean peak's m/z values, \"apex\": use m/z value peak apex, \"wMeanApex3\": intensity weighted mean m/z value peak apex m/z values left right \"meanApex3\": mean m/z value peak apex m/z values left right . integrate Integration method. integrate = 1 peak limits found descent mexican hat filtered data, integrate = 2 descent done real data. latter method accurate prone noise, former robust, less exact. mzdiff numeric(1) representing minimum difference m/z dimension required peaks overlapping retention times; can negative allow overlap. peak post-processing, peaks defined overlapping reduced one peak largest signal. fitgauss logical(1) whether Gaussian fitted peak. affects mostly retention time position peak. noise numeric(1) allowing set minimum intensity required centroids considered first analysis step (centroids intensity < noise omitted ROI detection). verboseColumns logical(1) whether additional peak meta data columns returned. criticalValue numeric(1). Suggested values: (0.1-3.0). setting helps determine Kalman Filter prediciton margin error. real centroid belonging bonafide peak must fall within KF prediction margin error. Much like construction confidence interval, criticalVal loosely translates multiplier standard error prediction reported Kalman Filter. peak XC-MS sample small mass deviance ppm error, smaller critical value might better vice versa. consecMissedLimit integer(1) Suggested values: (1,2,3). peak proces detected Kalman Filter, Kalman Filter may find predicted centroid every scan. 1 consecutive failed predictions, setting informs Massifquant stop Kalman Filter following candidate peak. unions integer(1) set 1 apply t-test union segmentation; set 0 t-test applied chromatographically continous peaks sharing m/z range. Explanation: data points, sometimes Kalman Filter stops tracking peak prematurely. Another Kalman Filter instantiated begins following rest signal. tracking done backwards forwards, algorithmic defect leaves real peak divided two segments . option turned , program identifies segmented peaks combines (merges ) one two sample t-test. potential danger option truly distinct peaks may merged. checkBack integer(1) set 1 turned ; set 0 turned . convergence Kalman Filter peak's precise m/z mapping fast, sometimes incorporates erroneous centroids part peak (especially early ). scanBack option attempt remove occasional outlier lies beyond converged bounds Kalman Filter. option directly affect identification peak postprocessing measure; shown extremely useful thus far default set turned . withWave logical(1) TRUE, peaks identified first Massifquant subsequently filtered second step centWave algorithm, includes wavelet estimation. object findChromPeaks:     OnDiskMSnExp object containing MS-     experiment-relevant data. methods: parameter object. param MassifquantParam object containing settings massifquant algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot. f integrate: MassifquantParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"MassifquantParam function returns  MassifquantParam class instance settings     specified chromatographic peak detection massifquant method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"algorithm's performance tested rigorously     high resolution LC/OrbiTrap, TOF-MS data centroid mode.     Simultaneous kalman filters identify chromatographic peaks calculate     area curve. default parameters set operate     complex LC-MS Orbitrap sample. Users find useful     simple exploratory data analysis find set minimum     intensity, identify many scans average peak spans.     consecMissedLimit parameter yielded good performance     Orbitrap data set (2) TOF data found best     (1). may change algorithm yet     tested many samples. criticalValue parameter perhaps     dificult dial appropriately visual inspection peak     identification best suggested tool quick optimization.     ppm checkBack parameters shown less influence     parameters exist give users flexibility     better accuracy. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"ppm,peakwidth,snthresh,prefilter,mzCenterFun,integrate,mzdiff,fitgauss,noise,verboseColumns,criticalValue,consecMissedLimit,unions,checkBack,withWave See corresponding parameter . Slots values exclusively accessed via corresponding getter setter methods listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports chromatographic peak     detection     OnDiskMSnExp objects (defined     MSnbase package). settings massifquant     centWave algorithm can passed MassifquantParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"Conley CJ, Smith R, Torgrip RJ, Taylor RM, Tautenhahn R Prince JT \"Massifquant: open-source Kalman filter-based XC-MS isotope trace feature detection\" Bioinformatics 2014, 30(18):2636-43.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"Christopher Conley, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-massifquant.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Chromatographic peak detection using the massifquant method — findChromPeaks-massifquant","text":"","code":"## Create a MassifquantParam object. mqp <- MassifquantParam() ## Change snthresh prefilter parameters snthresh(mqp) <- 30 prefilter(mqp) <- c(6, 10000) mqp #> Object of class:  MassifquantParam  #>  Parameters: #>  - ppm: [1] 25 #>  - peakwidth: [1] 20 50 #>  - snthresh: [1] 30 #>  - prefilter: [1]     6 10000 #>  - mzCenterFun: [1] \"wMean\" #>  - integrate: [1] 1 #>  - mzdiff: [1] -0.001 #>  - fitgauss: [1] FALSE #>  - noise: [1] 0 #>  - verboseColumns: [1] FALSE #>  - criticalValue: [1] 1.125 #>  - consecMissedLimit: [1] 2 #>  - unions: [1] 1 #>  - checkBack: [1] 0 #>  - withWave: [1] FALSE  ## Perform the peak detection using massifquant on the files from the ## faahKO package. Files are read using the readMSData from the MSnbase ## package library(faahKO) library(MSnbase) fls <- dir(system.file(\"cdf/KO\", package = \"faahKO\"), recursive = TRUE,            full.names = TRUE) raw_data <- readMSData(fls[1], mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method. ## Perform the peak detection using the settings defined above. res <- findChromPeaks(raw_data, param = mqp) #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 25ppm ...  #> OK #>  334 Peaks. head(chromPeaks(res)) #>          mz mzmin mzmax    rtmin    rtmax       rt    into   maxo sample #> CP001 578.4 578.4 578.4 4103.890 4141.449 4119.540 1748295 187328      1 #> CP002 577.3 577.3 577.3 4100.760 4130.495 4114.845 3854210 457024      1 #> CP003 367.2 367.2 367.2 4099.196 4164.924 4122.670  140753  11330      1 #> CP004 459.1 459.1 459.1 4099.196 4128.930 4113.280  144739  15874      1 #> CP005 401.2 401.2 401.2 4083.546 4221.262 4147.709  472407  11783      1 #> CP006 412.3 412.3 412.3 4072.591 4111.715 4089.806 2192350 183936      1"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"matchedFilter algorithm identifies peaks     chromatographic time domain described [Smith 2006]. intensity     values binned cutting LC/MS data slices (bins) mass     unit (binSize m/z) wide. Within bin maximal intensity     selected. chromatographic peak detection performed     bin extending based steps parameter generate     slices comprising bins current_bin - steps +1     current_bin + steps - 1. slices filtered     matched filtration using second-derative Gaussian model     peak shape. filtration peaks detected using signal--ratio     cut-. details illustrations see [Smith 2006]. MatchedFilterParam class allows specify     settings chromatographic peak detection using matchedFilter     method. Instances created MatchedFilterParam     constructor. findChromPeaks,OnDiskMSnExp,MatchedFilterParam     method performs peak detection using matchedFilter algorithm     samples OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. binSize,binSize<-: getter setter     binSize slot object. impute,impute<-: getter setter     impute slot object. baseValue,baseValue<-: getter setter     baseValue slot object. distance,distance<-: getter setter     distance slot object. fwhm,fwhm<-: getter setter     fwhm slot object. sigma,sigma<-: getter setter     sigma slot object. max,max<-: getter setter      max slot object. snthresh,snthresh<-: getter setter     snthresh slot object. steps,steps<-: getter setter     steps slot object. mzdiff,mzdiff<-: getter setter      mzdiff slot object. index,index<-: getter setter     index slot object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"","code":"MatchedFilterParam(   binSize = 0.1,   impute = \"none\",   baseValue = numeric(),   distance = numeric(),   fwhm = 30,   sigma = fwhm/2.3548,   max = 5,   snthresh = 10,   steps = 2,   mzdiff = 0.8 - binSize * steps,   index = FALSE )  # S4 method for OnDiskMSnExp,MatchedFilterParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for MatchedFilterParam binSize(object)  # S4 method for MatchedFilterParam binSize(object) <- value  # S4 method for MatchedFilterParam impute(object)  # S4 method for MatchedFilterParam impute(object) <- value  # S4 method for MatchedFilterParam baseValue(object)  # S4 method for MatchedFilterParam baseValue(object) <- value  # S4 method for MatchedFilterParam distance(object)  # S4 method for MatchedFilterParam distance(object) <- value  # S4 method for MatchedFilterParam fwhm(object)  # S4 method for MatchedFilterParam fwhm(object) <- value  # S4 method for MatchedFilterParam sigma(object)  # S4 method for MatchedFilterParam sigma(object) <- value  # S4 method for MatchedFilterParam max(x)  # S4 method for MatchedFilterParam max(object) <- value  # S4 method for MatchedFilterParam snthresh(object)  # S4 method for MatchedFilterParam snthresh(object) <- value  # S4 method for MatchedFilterParam steps(object)  # S4 method for MatchedFilterParam steps(object) <- value  # S4 method for MatchedFilterParam mzdiff(object)  # S4 method for MatchedFilterParam mzdiff(object) <- value  # S4 method for MatchedFilterParam index(object)  # S4 method for MatchedFilterParam index(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"binSize numeric(1) specifying width bins/slices m/z dimension. impute Character string specifying method used missing value imputation. Allowed values \"none\" (linear interpolation), \"lin\" (linear interpolation), \"linbase\" (linear interpolation within certain bin-neighborhood) \"intlin\". See imputeLinInterpol details. baseValue base value empty elements set. considered method = \"linbase\" corresponds profBinLinBase's baselevel argument. distance method = \"linbase\": number non-empty neighboring element empty element considered linear interpolation. See details section information. fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) representing maximum number peaks expected/identified per slice. snthresh numeric(1) defining signal noise cutoff used chromatographic peak detection step. steps numeric(1) defining number bins merged filtration (.e. number neighboring bins joined slice filtration peak detection performed). mzdiff numeric(1) defining minimum difference m/z peaks overlapping retention times index logical(1) specifying whether indicies returned instead values m/z retention times. object findChromPeaks:     OnDiskMSnExp object containing MS-     experiment-relevant data. methods: parameter object. param MatchedFilterParam object containing settings matchedFilter algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. ... ignored. value value slot. x max: MatchedFilterParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"MatchedFilterParam function returns  MatchedFilterParam class instance settings     specified chromatographic detection matchedFilter method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"intensities binned provided m/z values within     spectrum (scan). Binning performed bins centered     around m/z values (.e. first bin includes m/z values     min(mz) - bin_size/2 min(mz) + bin_size/2). details binning missing value imputation see     binYonX imputeLinInterpol methods. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"binSize,impute,baseValue,distance,fwhm,sigma,max,snthresh,steps,mzdiff,index See corresponding parameter . Slots values exclusively accessed via corresponding getter setter methods listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports chromatographic peak     detection     OnDiskMSnExp objects (defined     MSnbase package). settings matchedFilter     algorithm can passed MatchedFilterParam object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 2006, 78:779-787.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks-matchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Peak detection in the chromatographic time domain — findChromPeaks-matchedFilter","text":"","code":"## Create a MatchedFilterParam object. Note that we use a unnecessarily large ## binSize parameter to reduce the run-time of the example. mfp <- MatchedFilterParam(binSize = 5) ## Change snthresh parameter snthresh(mfp) <- 15 mfp #> Object of class:  MatchedFilterParam  #>  Parameters: #>  - binSize: [1] 5 #>  - impute: [1] \"none\" #>  - baseValue: numeric(0) #>  - distance: numeric(0) #>  - fwhm: [1] 30 #>  - sigma: [1] 12.73994 #>  - max: [1] 5 #>  - snthresh: [1] 15 #>  - steps: [1] 2 #>  - mzdiff: [1] -9.2 #>  - index: [1] FALSE  ## Perform the peak detection using matchecFilter on the files from the ## faahKO package. Files are read using the readMSData from the MSnbase ## package library(faahKO) library(MSnbase) fls <- dir(system.file(\"cdf/KO\", package = \"faahKO\"), recursive = TRUE,            full.names = TRUE) raw_data <- readMSData(fls[1], mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method. ## Perform the chromatographic peak detection using the settings defined ## above. Note that we are also disabling parallel processing in this ## example by registering a \"SerialParam\" res <- findChromPeaks(raw_data, param = mfp) head(chromPeaks(res)) #>             mz mzmin mzmax       rt    rtmin    rtmax      into    intf  maxo #> CP001 205.0000 205.0 205.0 2784.635 2770.550 2800.284 1778568.9 3580020 84280 #> CP002 205.0000 205.0 205.0 2784.635 2770.550 2800.284 1778568.9 3577971 84280 #> CP003 241.1460 241.1 241.2 3662.574 3646.924 3682.918 1465988.7 2234510 49728 #> CP004 241.1460 241.1 241.2 3662.574 3646.924 3682.918 1465988.7 2234510 49728 #> CP005 244.1000 244.1 244.1 2828.453 2814.369 2842.538  598990.3 1145078 31312 #> CP006 249.1591 249.1 249.2 3659.444 3643.794 3678.223 1435000.7 2367467 49040 #>            maxf i       sn sample #> CP001 194233.12 1 63.28090      1 #> CP002 194213.46 1 66.00099      1 #> CP003  96022.23 1 25.42409      1 #> CP004  96022.23 1 25.42643      1 #> CP005  64181.64 2 16.99513      1 #> CP006 104291.09 1 36.83500      1"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Chromatographic Peak Detection — findChromPeaks","title":"Chromatographic Peak Detection — findChromPeaks","text":"findChromPeaks method performs chromatographic peak detection LC/GC-MS data. peak detection algorithm can selected, configured, using param argument. Supported param objects : CentWaveParam(): chromatographic peak detection using centWave algorithm. CentWavePredIsoParam(): centWave predicted isotopes. Peak detection uses two-step centWave-based approach considering also feature isotopes. MatchedFilterParam(): peak detection using matched filter algorithm. MassifquantParam(): peak detection using Kalman filter-based massifquant method. MSWParam(): single-spectrum non-chromatography MS data peak detection. specific examples see help pages individual parameter classes listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Chromatographic Peak Detection — findChromPeaks","text":"","code":"findChromPeaks(object, param, ...)  # S4 method for MsExperiment,Param findChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   ...,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment,Param findChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   add = FALSE,   ...,   BPPARAM = bpparam() )"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Chromatographic Peak Detection — findChromPeaks","text":"object data object perform peak detection. Can OnDiskMSnExp(), XCMSnExp(), MChromatograms() MsExperiment() object. param parameter object selecting configuring algorithm. ... Optional parameters. msLevel integer(1) defining MS level chromatographic peak detection performed. chunkSize integer(1) object MsExperiment XcmsExperiment(): defines number files (samples) full peaks data (m/z intensity values) loaded memory time. Peak detection performed parallel (per sample) subset loaded data. setting thus allows balance memory demand speed (due parallel processing) peak detection. parallel processing can performed subset data loaded currently memory (iteration), value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSizeto total number samples experiment load full MS data memory thus settings cause --memory error. settingchunkSize = -1the peak detection performed separately, parallel, sample. however work allSpectra` backends (see eventually Spectra() details). BPPARAM Parallel processing setup. Uses default system-wide default setup. See bpparam() details. add logical(1) (object contains already chromatographic peaks, .e. either XCMSnExp XcmsExperiment) whether chromatographic peak detection results added existing results. default (add = FALSE) additional findChromPeaks call result object remove previous results.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Chromatographic Peak Detection — findChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":null,"dir":"Reference","previous_headings":"","what":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"findChromPeaksIsolationWindow function allows perform chromatographic peak detection MS level > 1 spectra certain isolation windows (e.g. SWATH pockets). function performs peak detection, separately spectra belonging isolation window adds chromPeaks() matrix result object. Information isolation window detected added chromPeakData() data frame. Note peak detection method remove previously identified chromatographic peaks (e.g. MS1 level using findChromPeaks() function adds newly identified peaks existing chromPeaks() matrix. Isolation windows can defined isolationWindow parameter, default uses definition isolationWindowTargetMz(), .e. chromatographic peak detection performed spectra isolation window target m/z (seprarately file). parameter param allows define configure peak detection algorithm (see findChromPeaks() information).","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"","code":"findChromPeaksIsolationWindow(object, ...)  # S4 method for MsExperiment findChromPeaksIsolationWindow(   object,   param,   msLevel = 2L,   isolationWindow = isolationWindowTargetMz(spectra(object)),   chunkSize = 2L,   ...,   BPPARAM = bpparam() )  # S4 method for OnDiskMSnExp findChromPeaksIsolationWindow(   object,   param,   msLevel = 2L,   isolationWindow = isolationWindowTargetMz(object),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"object MsExperiment, XcmsExperiment, OnDiskMSnExp XCMSnExp object DIA data. ... currently used. param Peak detection parameter object, CentWaveParam object defining configuring chromographic peak detection algorithm. See also findChromPeaks() details. msLevel integer(1) specifying MS level peak detection performed. default msLevel = 2L. isolationWindow factor similar defining isolation windows peak detection performed length equal number spectra object. chunkSize object MsExperiment XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. See findChromPeaks() information. BPPARAM object MsExperiment XcmsExperiment: parallel processing setup. See bpparam() information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"XcmsExperiment XCMSnExp object chromatographic peaks identified spectra isolation window file added chromPeaks matrix. Isolation window definition identified peak stored additional columns chromPeakData().","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findChromPeaksIsolationWindow.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Data independent acquisition (DIA): peak detection in isolation windows — findChromPeaksIsolationWindow","text":"Johannes Rainer, Michael Witting","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":null,"dir":"Reference","previous_headings":"","what":"Find values in sorted vectors — findEqualGreater","title":"Find values in sorted vectors — findEqualGreater","text":"Find values sorted vectors.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find values in sorted vectors — findEqualGreater","text":"","code":"findEqualGreater(x, value) findEqualLess(x, value) findEqualGreaterM(x, values) findRange(x, values, NAOK = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find values in sorted vectors — findEqualGreater","text":"x numeric vector sorted increasing order value value find x values numeric values find x NAOK check NA values x","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find values in sorted vectors — findEqualGreater","text":"findEqualGreater finds index first value   x equal greater value. findEqualLess   except finds equal less. findEqualGreaterM   creates index vector finding specified values.   findRange locates start stop indicides range   two x values. things save time point   findeEqualGreaterM (length values approaches   lenght x) findRange (NAOK set   TRUE). run log(N) N time, respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find values in sorted vectors — findEqualGreater","text":"integer vector position(s) values(s).","code":""},{"path":"https://sneumann.github.io/xcms/reference/findEqualGreater.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find values in sorted vectors — findEqualGreater","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":null,"dir":"Reference","previous_headings":"","what":"Find fragment ions in xcmsFragment objects — findMZ","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"method find fragment mass ppm window xcmsFragment object","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"","code":"findMZ(object, find, ppmE=25, print=TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"object xcmsFragment object type find fragment ion found ppmE ppm error window searching print print nice little report","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"method simply searches given fragment ion xcmsFragment object type given certain ppm error window","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"data frame following columns: PrecursorMz precursor m/z fragment MSnParentPeakID index ID location precursor peak xcmsFragment object msLevel level found fragment ion rt Retention time found ion mz actual m/z found fragment ion intensity intensity fragment ion sample sample fragment ion came GroupPeakMSn ID peaks grouped xcmsSet grouping CollisionEnergy collision energy precursor scan","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"H. Paul Benton, D.M. Wong, S..Strauger, G. Siuzdak \"XC\\(MS^2\\)\"   Analytical Chemistry 2008","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"H. Paul Benton, hpaul.beonton08@imperial.ac.uk","code":""},{"path":"https://sneumann.github.io/xcms/reference/findMZ.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Find fragment ions in xcmsFragment objects — findMZ","text":"","code":"if (FALSE) {     library(msdata)     mzMLpath <- system.file(\"iontrap\", package = \"msdata\")     mzMLfiles<-list.files(mzMLpath, pattern = \"extracted.mzML\",                             recursive = TRUE, full.names = TRUE)     xs <- xcmsSet(mzMLfiles, method = \"MS1\")     ##takes only one file from the file set     xfrag <- xcmsFragments(xs)     found<-findMZ(xfrag, 657.3433, 50) }"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":null,"dir":"Reference","previous_headings":"","what":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"Perform peak detection mass spectrometry     direct injection spectrum using wavelet based algorithm. MSWParam class allows specify     settings peak detection using MSW method. Instances     created MSWParam constructor. findChromPeaks,OnDiskMSnExp,MSWParam     method performs peak detection single-spectrum non-chromatography MS     data using functionality MassSpecWavelet package     samples OnDiskMSnExp object.     OnDiskMSnExp objects encapsule experiment     specific data load spectra data (mz intensity values)     fly original files applying also eventual data     manipulations. snthresh,snthresh<-: getter setter     snthresh slot object. verboseColumns,verboseColumns<-: getter setter     verboseColumns slot object. scales,scales<-: getter setter     scales slot object. nearbyPeak,nearbyPeak<-: getter setter     nearbyPeak slot object. peakScaleRange,peakScaleRange<-: getter setter     peakScaleRange slot object. ampTh,ampTh<-: getter setter     ampTh slot object. minNoiseLevel,minNoiseLevel<-: getter setter     minNoiseLevel slot object. ridgeLength,ridgeLength<-: getter setter     ridgeLength slot object. peakThr,peakThr<-: getter setter     peakThr slot object. tuneIn,tuneIn<-: getter setter     tuneIn slot object. addParams,addParams<-: getter setter     addParams slot object. slot stores optional additional     parameters passed     identifyMajorPeaks     peakDetectionCWT functions     MassSpecWavelet package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"","code":"MSWParam(   snthresh = 3,   verboseColumns = FALSE,   scales = c(1, seq(2, 30, 2), seq(32, 64, 4)),   nearbyPeak = TRUE,   peakScaleRange = 5,   ampTh = 0.01,   minNoiseLevel = ampTh/snthresh,   ridgeLength = 24,   peakThr = NULL,   tuneIn = FALSE,   ... )  # S4 method for OnDiskMSnExp,MSWParam findChromPeaks(   object,   param,   BPPARAM = bpparam(),   return.type = \"XCMSnExp\",   msLevel = 1L,   ... )  # S4 method for MSWParam snthresh(object)  # S4 method for MSWParam snthresh(object) <- value  # S4 method for MSWParam verboseColumns(object)  # S4 method for MSWParam verboseColumns(object) <- value  # S4 method for MSWParam scales(object)  # S4 method for MSWParam scales(object) <- value  # S4 method for MSWParam nearbyPeak(object)  # S4 method for MSWParam nearbyPeak(object) <- value  # S4 method for MSWParam peakScaleRange(object)  # S4 method for MSWParam peakScaleRange(object) <- value  # S4 method for MSWParam ampTh(object)  # S4 method for MSWParam ampTh(object) <- value  # S4 method for MSWParam minNoiseLevel(object)  # S4 method for MSWParam minNoiseLevel(object) <- value  # S4 method for MSWParam ridgeLength(object)  # S4 method for MSWParam ridgeLength(object) <- value  # S4 method for MSWParam peakThr(object)  # S4 method for MSWParam peakThr(object) <- value  # S4 method for MSWParam tuneIn(object)  # S4 method for MSWParam tuneIn(object) <- value  # S4 method for MSWParam addParams(object)  # S4 method for MSWParam addParams(object) <- value"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"snthresh numeric(1) defining signal noise ratio cutoff. verboseColumns logical(1) whether additional peak meta data columns returned. scales Numeric defining scales continuous wavelet transform (CWT). nearbyPeak logical(1) whether include nearby peaks major peaks. peakScaleRange numeric(1) defining scale range peak (larger 5 default). ampTh numeric(1) defining minimum required relative amplitude peak (ratio maximum CWT coefficients). minNoiseLevel numeric(1) defining minimum noise level used computing SNR. ridgeLength numeric(1) defining minimum highest scale peak 2-D CWT coefficient matrix. peakThr numeric(1) minimum absolute intensity (baseline) peaks picked. provided, smoothing Savitzky-Golay filter used (MassSpecWavelet) package estimate local intensity. tuneIn logical(1) whther tune parameter estimation detected peaks. ... Additional parameters passed peakDetectionCWT identifyMajorPeaks functions MassSpecWavelet package. object findChromPeaks:     OnDiskMSnExp object containing MS-     experiment-relevant data. methods: parameter object. param MSWParam object containing settings algorithm. BPPARAM parameter class specifying parallel processing performed. defaults bpparam. See documentation BiocParallel details. parallel processing enabled, peak detection performed parallel several input samples. return.type Character specifying type object method return. Can either \"XCMSnExp\" (default), \"list\" \"xcmsSet\". msLevel integer(1) defining MS level peak detection performed. Defaults msLevel = 1. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"MSWParam function returns MSWParam class instance settings specified peak detection     MSW method. findChromPeaks: return.type = \"XCMSnExp\"  XCMSnExp object results peak detection.     return.type = \"list\" list length equal number     samples matrices specifying identified peaks.     return.type = \"xcmsSet\" xcmsSet object     results detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"wrapper peak picker Bioconductor's     MassSpecWavelet package calling     peakDetectionCWT     tuneInPeakInfo functions. See     xcmsDirect vignette information. Parallel processing (one process per sample) supported can     configured either BPPARAM parameter globally     defining parallel processing mode using     register method BiocParallel     package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"snthresh,verboseColumns,scales,nearbyPeak,peakScaleRange,ampTh,minNoiseLevel,ridgeLength,peakThr,tuneIn,addParams See corresponding parameter .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"methods classes part updated modernized     xcms user interface eventually replace     findPeaks methods. supports peak detection     OnDiskMSnExp     objects (defined MSnbase package). settings     algorithm can passed MSWParam object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"Joachim Kutzera, Steffen Neumann, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-MSW.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Single-spectrum non-chromatography MS data peak detection — findPeaks-MSW","text":"","code":"## Create a MSWParam object mp <- MSWParam() ## Change snthresh parameter snthresh(mp) <- 15 mp #> Object of class:  MSWParam  #>  Parameters: #>  - snthresh: [1] 15 #>  - verboseColumns: [1] FALSE #>  - scales:  [1]  1  2  4  6  8 10 12 14 16 18 20 22 24 26 28 30 32 36 40 44 48 52 56 60 64 #>  - nearbyPeak: [1] TRUE #>  - peakScaleRange: [1] 5 #>  - amp.Th: [1] 0.01 #>  - minNoiseLevel: [1] 0.003333333 #>  - ridgeLength: [1] 24 #>  - peakThr: numeric(0) #>  - tuneIn: [1] FALSE  ## Loading a small subset of direct injection, single spectrum files library(msdata) fticrf <- list.files(system.file(\"fticr-mzML\", package = \"msdata\"),                     recursive = TRUE, full.names = TRUE) fticr <- readMSData(fticrf[1], msLevel. = 1, mode = \"onDisk\")  ## Perform the MSW peak detection on these: p <- MSWParam(scales = c(1, 7), peakThr = 80000, ampTh = 0.005,              SNR.method = \"data.mean\", winSize.noise = 500) fticr <- findChromPeaks(fticr, param = p)  head(chromPeaks(fticr)) #>            mz    mzmin    mzmax rt rtmin rtmax    into     maxo        sn intf #> CP01 403.2367 403.2279 403.2447 -1    -1    -1 4735258 372259.4 22.975342   NA #> CP02 409.1845 409.1755 409.1936 -1    -1    -1 4158404 310572.1 20.597978   NA #> CP03 413.2677 413.2585 413.2769 -1    -1    -1 6099006 435462.6 27.217229   NA #> CP04 414.2719 414.2652 414.2786 -1    -1    -1 1916658 157613.1  8.777433   NA #> CP05 423.2363 423.2266 423.2459 -1    -1    -1 2708391 174252.7 14.745275   NA #> CP06 427.2681 427.2583 427.2779 -1    -1    -1 6302089 461385.6 32.468892   NA #>           maxf sample #> CP01  814693.1      1 #> CP02  731557.2      1 #> CP03 1018994.8      1 #> CP04  280826.9      1 #> CP05  435858.5      1 #> CP06 1124549.4      1"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"number peak pickers exist XCMS. findPeaks   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"object = \"xcmsRaw\" findPeaks(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"object xcmsRaw-class object method Method use peak detection. See details. ... Optional arguments passed along","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"Different algorithms can used specifying   method argument. example use matched filter   approach described Smith et al (2006) one use:   findPeaks(object, method=\"matchedFilter\"). also   default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$findPeaks.methods.   nickname method called \"centWave\",   help page specific method can   accessed ?findPeaks.centWave.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection for GC/MS and LC/MS Data - methods — findPeaks-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z minimum step mzmax m/z maximum step rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated area original (raw) peak maxo maximum intensity original (raw) peak additional columns depending choosen method.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"Collecting Tandem MS MS$^n$ Mass Spectrometry   precursor peaks annotated XML raw file","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"object = \"xcmsRaw\" findPeaks.MS1(object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"mass spectrometers can acquire MS1 MS2 (MS$^n$ scans)   quasi simultaneously, e.g. data dependent tandem MS DDIT mode. Since xcmsFragments attaches MS$^n$ peaks MS1 peaks xcmsSet,   important findPeaks xcmsSet miss MS1 precursor peak. sure MS1 precursor peaks xcmsSet,   findPeaks.MS1 actual peak picking, simply uses   annotation stored mzXML, mzData mzML raw files. relies following XML tags: mzData:      <spectrum id=\"463\">         <spectrumInstrument msLevel=\"2\"> \t  <cvParam cvLabel=\"psi\" accession=\"PSI:1000039\" name=\"TimeInSeconds\" value=\"92.7743\"/>         <\/spectrumInstrument>         <precursor msLevel=\"1\" spectrumRef=\"461\">             <cvParam cvLabel=\"psi\" accession=\"PSI:1000040\" name=\"MassToChargeRatio\" value=\"462.091\"/>             <cvParam cvLabel=\"psi\" accession=\"PSI:1000042\" name=\"Intensity\" value=\"366.674\"/>         <\/precursor>   <\/spectrum> mzXML:         <scan num=\"17\" msLevel=\"2\" retentionTime=\"PT1.5224S\">      <precursorMz precursorIntensity=\"125245\">220.1828003<\/precursorMz>      <\/scan> Several mzXML mzData converters known create incomplete   files, either without intensities (set 0) without   precursor retention time (reasonably close rt   chosen. NYI).","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"object xcmsRaw object","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"matrix columns: mz, mzmin, mzmax annotated MS1 precursor selection mass rt, rtmin, rtmax annotated MS1 precursor retention time , maxo, sn annotated MS1 precursor intensity","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MS1-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Collecting MS1 precursor peaks — findPeaks.MS1-methods","text":"Steffen Neumann, sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"method performs peak detection mass spectrometry     direct injection spectrum using wavelet based algorithm.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"","code":"# S4 method for xcmsRaw findPeaks.MSW(object, snthresh = 3, verbose.columns = FALSE, ...)"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"object xcmsRaw object peak detection performed. snthresh numeric(1) defining signal noise ratio cutoff. verbose.columns Logical whether additional peak meta data columns returned. ... Additional parameters passed peakDetectionCWT identifyMajorPeaks functions MassSpecWavelet package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"matrix, row representing intentified peak, columns:  mz m/z value peak centroid position. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Always -1. rtmin Always -1. rtmax Always -1. Integrated (original) intensity peak. maxo Maximum intensity peak. intf Always NA. maxf Maximum MSW-filter response peak. sn Signal noise ratio.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"wrapper around peak picker Bioconductor's     MassSpecWavelet package calling     peakDetectionCWT     tuneInPeakInfo functions.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.MSW-xcmsRaw-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Peak detection for single-spectrum non-chromatography MS data — findPeaks.MSW,xcmsRaw-method","text":"Joachim Kutzera, Steffen Neumann, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"Peak density wavelet based feature detection aiming isotope peaks high resolution LC/MS data centroid mode","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"object = \"xcmsRaw\" findPeaks.centWave(object, ppm=25, peakwidth=c(20,50),     prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001, fitgauss=FALSE,     scanrange= numeric(), noise=0, sleep=0, verbose.columns=FALSE, xcmsPeaks, snthresh=6.25, maxcharge=3, maxiso=5, mzIntervalExtension=TRUE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method isotope ROIs (regions interest) LC/MS map predicted.   second phase mass traces analysed.   Continuous wavelet transform (CWT) used locate chromatographic peaks different scales.   resulting peak list given peak list (xcmsPeaks) merged redundant peaks removed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"object xcmsSet object ppm maxmial tolerated m/z deviation consecutive scans, ppm (parts per million) peakwidth Chromatographic peak width, given range (min,max) seconds prefilter prefilter=c(k,). Prefilter step first phase. Mass traces retained contain least k peaks intensity >= . mzCenterFun Function calculate m/z center feature: wMean intensity weighted mean feature m/z values, mean mean feature m/z values, apex use m/z value peak apex,         wMeanApex3 intensity weighted mean m/z value peak apex m/z value left right , meanApex3 mean m/z value peak apex m/z value left right . integrate Integration method. =1 peak limits found descent mexican hat filtered data, =2 descent done real data. Method 2 accurate prone noise, method 1 robust noise less exact. mzdiff minimum difference m/z peaks overlapping retention     times, can negative allow overlap fitgauss logical, TRUE Gaussian fitted peak scanrange scan range process noise optional argument useful data centroided without intensity threshold,     centroids intensity < noise omitted ROI detection sleep number seconds pause plotting peak finding cycles verbose.columns logical, TRUE additional peak meta data columns returned xcmsPeaks peak list picked using centWave algorithm parameter verbose.columns set TRUE (columns scmin scmax needed) snthresh signal noise ratio cutoff, definition see . maxcharge max. number isotope charge. maxiso max. number isotope peaks predict detected feature. mzIntervalExtension logical, TRUE predicted isotope ROIs (regions interest) extended m/z dimension increase detection low intensity hence noisy peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z peak minimum mzmax m/z peak maximum rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated peak intensity intb baseline corrected integrated peak intensity maxo maximum peak intensity sn Signal/Noise ratio, defined (maxo - baseline)/sd, maxo maximum peak intensity, baseline estimated baseline value sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit verbose.columns TRUE additionally : mu Gaussian parameter mu sigma Gaussian parameter sigma h Gaussian parameter h f Region number m/z ROI peak localised dppm m/z deviation mass trace across scans ppm scale Scale peak localised scpos Peak position found wavelet analysis scmin Left peak limit found wavelet analysis (scan number) scmax Right peak limit found wavelet analysis (scan number)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"Ralf Tautenhahn","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.addPredictedIsotopeFeatures-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection based on predicted isotope features for high resolution LC/MS data — findPeaks.addPredictedIsotopeFeatures-methods","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann   \"Highly sensitive feature detection high resolution LC/MS\"   BMC Bioinformatics 2008, 9:504\\   Hendrik Treutler Steffen Neumann.   \"Prediction, detection, validation isotope clusters mass spectrometry data\"   Submitted Metabolites 2016, Special Issue \"Bioinformatics Data Analysis\"","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"Peak density wavelet based feature detection high resolution LC/MS data centroid mode","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"object = \"xcmsRaw\" findPeaks.centWave(object, ppm=25, peakwidth=c(20,50), snthresh=10,     prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001, fitgauss=FALSE,     scanrange= numeric(), noise=0, sleep=0, verbose.columns=FALSE, ROI.list=list()),     firstBaselineCheck=TRUE, roiScales=NULL","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode. first phase method mass traces (characterised regions less ppm m/z deviation consecutive scans) LC/MS map located.   second phase mass traces analysed.   Continuous wavelet transform (CWT) used locate chromatographic peaks different scales.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"object xcmsSet object ppm maxmial tolerated m/z deviation consecutive scans, ppm (parts per million) peakwidth Chromatographic peak width, given range (min,max) seconds snthresh signal noise ratio cutoff, definition see . prefilter prefilter=c(k,). Prefilter step first phase. Mass traces retained contain least k peaks intensity >= . mzCenterFun Function calculate m/z center feature: wMean intensity weighted mean feature m/z values, mean mean feature m/z values, apex use m/z value peak apex,         wMeanApex3 intensity weighted mean m/z value peak apex m/z value left right , meanApex3 mean m/z value peak apex m/z value left right . integrate Integration method. =1 peak limits found descent mexican hat filtered data, =2 descent done real data. Method 2 accurate prone noise, method 1 robust noise less exact. mzdiff minimum difference m/z peaks overlapping retention     times, can negative allow overlap fitgauss logical, TRUE Gaussian fitted peak scanrange scan range process noise optional argument useful data centroided without intensity threshold,     centroids intensity < noise omitted ROI detection sleep number seconds pause plotting peak finding cycles verbose.columns logical, TRUE additional peak meta data columns returned ROI.list optional list ROIs represents detected mass traces (ROIs).  list empty (default) centWave detects mass trace ROIs,     otherwise step skipped supplied ROIs used peak detection phase.  ROI object list following slots:     scmin start scan index,     scmax end scan index,     mzmin minimum m/z,     mzmax maximum m/z,     length number scans,     intensity summed intensity. firstBaselineCheck logical, TRUE continuous data within ROI checked 1st baseline roiScales numeric, optional vector scales ROI ROI.list used centWave-wavelets","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z peak minimum mzmax m/z peak maximum rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated peak intensity intb baseline corrected integrated peak intensity maxo maximum peak intensity sn Signal/Noise ratio, defined (maxo - baseline)/sd, maxo maximum peak intensity, baseline estimated baseline value sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit verbose.columns TRUE additionally : mu Gaussian parameter mu sigma Gaussian parameter sigma h Gaussian parameter h f Region number m/z ROI peak localised dppm m/z deviation mass trace across scans ppm scale Scale peak localised scpos Peak position found wavelet analysis scmin Left peak limit found wavelet analysis (scan number) scmax Right peak limit found wavelet analysis (scan number)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"Ralf Tautenhahn","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWave-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection for high resolution LC/MS data — findPeaks.centWave-methods","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann   \"Highly sensitive feature detection high resolution LC/MS\"   BMC Bioinformatics 2008, 9:504","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"Peak density wavelet based feature detection high resolution LC/MS data centroid mode additional peak picking isotope features basis isotope peak predictions","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"object = \"xcmsRaw\" findPeaks.centWaveWithPredictedIsotopeROIs(object, ppm=25, peakwidth=c(20,50), snthresh=10,     prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001, fitgauss=FALSE,     scanrange= numeric(), noise=0, sleep=0, verbose.columns=FALSE, ROI.list=list(),     firstBaselineCheck=TRUE, roiScales=NULL, snthreshIsoROIs=6.25, maxcharge=3, maxiso=5, mzIntervalExtension=TRUE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"algorithm suitable high resolution LC/{TOF,OrbiTrap,FTICR}-MS data centroid mode.   centWave algorithm applied two peak picking steps follows. first peak picking step ROIs (regions interest, characterised regions less ppm m/z deviation consecutive scans) LC/MS map located analysed using continuous wavelet transform (CWT) localization chromatographic peaks different scales.   second peak picking step isotope ROIs LC/MS map predicted analysed using continuous wavelet transform (CWT) localization chromatographic peaks different scales.   peak lists resulting peak picking steps merged redundant peaks removed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"object xcmsSet object ppm maxmial tolerated m/z deviation consecutive scans, ppm (parts per million) peakwidth Chromatographic peak width, given range (min,max) seconds snthresh signal noise ratio cutoff, definition see . prefilter prefilter=c(k,). Prefilter step first phase. Mass traces retained contain least k peaks intensity >= . mzCenterFun Function calculate m/z center feature: wMean intensity weighted mean feature m/z values, mean mean feature m/z values, apex use m/z value peak apex,         wMeanApex3 intensity weighted mean m/z value peak apex m/z value left right , meanApex3 mean m/z value peak apex m/z value left right . integrate Integration method. =1 peak limits found descent mexican hat filtered data, =2 descent done real data. Method 2 accurate prone noise, method 1 robust noise less exact. mzdiff minimum difference m/z peaks overlapping retention     times, can negative allow overlap fitgauss logical, TRUE Gaussian fitted peak scanrange scan range process noise optional argument useful data centroided without intensity threshold,     centroids intensity < noise omitted ROI detection sleep number seconds pause plotting peak finding cycles verbose.columns logical, TRUE additional peak meta data columns returned ROI.list optional list ROIs represents detected mass traces (ROIs).  list empty (default) centWave detects mass trace ROIs,     otherwise step skipped supplied ROIs used peak detection phase.  ROI object list following slots:     scmin start scan index,     scmax end scan index,     mzmin minimum m/z,     mzmax maximum m/z,     length number scans,     intensity summed intensity. firstBaselineCheck logical, TRUE continuous data within ROI checked 1st baseline roiScales numeric, optional vector scales ROI ROI.list used centWave-wavelets snthreshIsoROIs signal noise ratio cutoff predicted isotope ROIs, definition see . maxcharge max. number isotope charge. maxiso max. number isotope peaks predict detected feature. mzIntervalExtension logical, TRUE predicted isotope ROIs (regions interest) extended m/z dimension increase detection low intensity hence noisy peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"matrix columns: mz weighted (intensity) mean peak m/z across scans mzmin m/z peak minimum mzmax m/z peak maximum rt retention time peak midpoint rtmin leading edge peak retention time rtmax trailing edge peak retention time integrated peak intensity intb baseline corrected integrated peak intensity maxo maximum peak intensity sn Signal/Noise ratio, defined (maxo - baseline)/sd, maxo maximum peak intensity, baseline estimated baseline value sd standard deviation local chromatographic noise. egauss RMSE Gaussian fit verbose.columns TRUE additionally : mu Gaussian parameter mu sigma Gaussian parameter sigma h Gaussian parameter h f Region number m/z ROI peak localised dppm m/z deviation mass trace across scans ppm scale Scale peak localised scpos Peak position found wavelet analysis scmin Left peak limit found wavelet analysis (scan number) scmax Right peak limit found wavelet analysis (scan number)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"Ralf Tautenhahn","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.centWaveWithPredictedIsotopeROIs-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection with centWave and additional isotope features — findPeaks.centWaveWithPredictedIsotopeROIs-methods","text":"Ralf Tautenhahn, Christoph B\\\"ottcher, Steffen Neumann   \"Highly sensitive feature detection high resolution LC/MS\"   BMC Bioinformatics 2008, 9:504\\   Hendrik Treutler Steffen Neumann.   \"Prediction, detection, validation isotope clusters mass spectrometry data\"   Submitted Metabolites 2016, Special Issue \"Bioinformatics Data Analysis\"","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"Massifquant Kalman filter (KF) based feature detection XC-MS data centroid mode (currently experimental stage). Optionally allows calling method \"centWave\" features discovered Massifquant refine feature detection; , supply additional parameters specific centWave (even experimental). method may conveniently called xcmsSet(...) method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"object = \"xcmsRaw\" findPeaks.massifquant(object, ppm=10, peakwidth=c(20,50), snthresh=10,      prefilter=c(3,100), mzCenterFun=\"wMean\", integrate=1, mzdiff=-0.001,      fitgauss=FALSE, scanrange= numeric(), noise=0,      sleep=0, verbose.columns=FALSE, criticalValue = 1.125, consecMissedLimit = 2,      unions = 1, checkBack = 0, withWave = 0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"algorithm's performance tested rigorously high resolution LC/{OrbiTrap, TOF}-MS data centroid mode. Simultaneous kalman filters identify features calculate area curve. default parameters set operate complex LC-MS Orbitrap sample. Users find useful simple exploratory data analysis find set minimum intensity, identify many scans average feature spans. \"consecMissedLimit\" parameter yielded good performance Orbitrap data set (2) TOF data found best (1). may change algorithm yet tested many samples. \"criticalValue\" parameter perhaps dificult dial appropriately visual inspection peak identification best suggested tool quick optimization.  \"ppm\" \"checkBack\" parameters shown less influence parameters exist give users flexibility better accuracy.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"following arguments specific Massifquant. additional arguments supplied must correspond specified method findPeaks.centWave. object xcmsRaw object. criticalValue Numeric: Suggested values: (0.1-3.0). setting helps determine Kalman Filter prediciton margin error. real centroid belonging bonafide feature must fall within KF prediction margin error. Much like construction confidence interval, criticalVal loosely translates  multiplier standard error prediction reported Kalman Filter. features XC-MS sample small mass deviance ppm error, smaller critical value might better vice versa. consecMissedLimit Integer: Suggested values:(1,2,3). feature proces detected Kalman Filter, Kalman Filter may find predicted centroid every scan. 1 consecutive failed predictions, setting informs Massifquant stop Kalman Filter following candidate feature. prefilter Numeric Vector: (Positive Integer, Positive Numeric): first argument used (withWave = 1); see centWave details. second argument specifies minimum threshold maximum intensity feature must met. peakwidth Integer Vector: (Positive Integer, Positive Integer): first argument used Massifquant, specifices minimum feature length time scans. centWave used, second argument maximum feature length  subject greater mininum feature length. ppm minimum estimated parts per million mass resolution feature must possess. unions Integer: set 1 apply t-test union segmentation; set 0 t-test applied chromatographically continous features sharing m/z range. Explanation: data points, sometimes Kalman Filter stops tracking feature prematurely. Another Kalman Filter instantiated begins following rest signal. tracking done backwards forwards, algorithmic defect leaves real feature divided two segments . option turned , program identifies segmented features combines (merges ) one two sample t-test. potential danger option truly distinct features may merged. withWave Integer: set 1 turned ; set 0 turned . Allows user find features first Massifquant filter features second phase centWave, includes wavelet estimation. checkBack Integer: set 1 turned ; set 0 turned . convergence Kalman Filter feature's precise m/z mapping  fast, sometimes incorporates erroneous centroids part feature (especially early ). \"scanBack\" option attempt remove occasional outlier lies beyond converged bounds Kalman Filter. option directly affect identification feature postprocessing measure; shown extremely useful thus far default set turned .","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"method findPeaks.massifquant(...) used, matrix returned rows corresponding features, properties features listed following column names. Otherwise, centWave feature used also (withWave = 1), Massifquant called xcmsSet(...) method, corresponding return values used. mz weighted m/z mean (weighted intensity) feature mzmin m/z  lower boundary feature mzmax m/z  upper boundary feature rtmin starting scan time feature rtmax starting scan time feature raw quantitation (area curve) feature. area feature area normalized scan rate.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"Christopher Conley","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"Submitted review. Christopher Conley, Ralf J .O Torgrip. Ryan Taylor, John T. Prince.  \"Massifquant: open-source Kalman filter based XC-MS feature detection\". August 2013.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.massifquant-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Feature detection for XC-MS data. — findPeaks.massifquant-methods","text":"","code":"library(faahKO) library(xcms) #load all the wild type and Knock out samples cdfpath <- system.file(\"cdf\", package = \"faahKO\") ## Subset to only the first 2 files. cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)[1:2]  ## Run the massifquant analysis. Setting the noise level to 10000 to speed up ## execution of the examples - in a real use case it should be set to a reasoable ## value. xset <- xcmsSet(cdffiles, method = \"massifquant\",                 consecMissedLimit = 1,                 snthresh = 10,                 criticalValue = 1.73,                 ppm = 10,                 peakwidth= c(30, 60),                 prefilter= c(1,3000),                 noise = 10000,                 withWave = 0) #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 10ppm ...  #> OK #>  242 Peaks. #>  #>  Massifquant, Copyright (C) 2013 Brigham Young University. #>  Massifquant comes with ABSOLUTELY NO WARRANTY. See LICENSE for details. #>  #>  Detecting  mass traces at 10ppm ...  #> OK #>  350 Peaks."},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"Find peaks chromatographic time domain     profile matrix. details see     do_findChromPeaks_matchedFilter.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"","code":"# S4 method for xcmsRaw findPeaks.matchedFilter(   object,   fwhm = 30,   sigma = fwhm/2.3548,   max = 5,   snthresh = 10,   step = 0.1,   steps = 2,   mzdiff = 0.8 - step * steps,   index = FALSE,   sleep = 0,   scanrange = numeric() )"},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"object xcmsRaw object peak detection performed. fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) representing maximum number peaks expected/identified per slice. snthresh numeric(1) defining signal noise cutoff used chromatographic peak detection step. step numeric(1) specifying width bins/slices m/z dimension. steps numeric(1) defining number bins merged filtration (.e. number neighboring bins joined slice filtration peak detection performed). mzdiff numeric(1) defining minimum difference m/z peaks overlapping retention times index logical(1) specifying whether indicies returned instead values m/z retention times. sleep (DEPRECATED). use parameter highly discouraged, cause problems parallel processing mode. scanrange Numeric vector defining range scans original object sub-setted peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"matrix, row representing intentified chromatographic peak,     columns:  mz Intensity weighted mean m/z values peak across     scans. mzmin Minimum m/z peak. mzmax Maximum m/z peak. rt Retention time peak's midpoint. rtmin Minimum retention time peak. rtmax Maximum retention time peak. Integrated (original) intensity peak. intf Integrated intensity filtered peak. maxo Maximum intensity peak. maxf Maximum intensity filtered peak. Rank peak merged EIC (<= max). sn Signal noise ratio peak.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"Colin . Smith, Elizabeth J. Want, Grace O'Maille, Ruben Abagyan     Gary Siuzdak. \"XCMS: Processing Mass Spectrometry Data Metabolite     Profiling Using Nonlinear Peak Alignment, Matching, Identification\"     Anal. Chem. 2006, 78:779-787.     @family Old peak detection methods","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findPeaks.matchedFilter-xcmsRaw-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Peak detection in the chromatographic time domain — findPeaks.matchedFilter,xcmsRaw-method","text":"Colin . Smith","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":null,"dir":"Reference","previous_headings":"","what":"Find neutral losses in xcmsFragment objects — findneutral","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"method find neutral loss ppm window xcmsFragment object","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"","code":"findneutral(object, find, ppmE=25, print=TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"object xcmsFragment object type find neutral loss found ppmE ppm error window searching print print nice little report","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"method searches given neutral loss xcmsFragment object type given certain ppm error window. neutral losses generated neighbouring ions. resulting data frame shows whole scan neutral loss found.","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"data frame following columns: PrecursorMz precursor m/z neutral losses MSnParentPeakID index ID location precursor peak xcmsFragment object msLevel level found fragment ion rt Retention time found ion mz actual m/z found fragment ion intensity intensity fragment ion sample sample fragment ion came GroupPeakMSn ID peaks grouped xcmsSet grouping CollisionEnergy collision energy precursor scan","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"H. Paul Benton, D.M. Wong, S..Strauger, G. Siuzdak \"XC\\(MS^2\\)\"   Analytical Chemistry 2008","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"H. Paul Benton, hpbenton@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/findneutral.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Find neutral losses in xcmsFragment objects — findneutral","text":"","code":"if (FALSE) {     library(msdata)     mzMLpath <- system.file(\"iontrap\", package = \"msdata\")     mzMLfiles<-list.files(mzMLpath, pattern = \"extracted.mzML\",                             recursive = TRUE, full.names = TRUE)     xs <- xcmsSet(mzMLfiles, method = \"MS1\")     ##takes only one file from the file set     xfrag <- xcmsFragments(xs)     found<-findneutral(xfrag, 58.1455, 50) }"},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"Generate multiple extracted ion chromatograms m/z values   interest. xcmsSet objects, reread original raw data   apply precomputed retention time correction, applicable. Note method always return profile, raw   data (profile data binned data along M/Z). See details   information.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"object = \"xcmsRaw\" getEIC(object, mzrange, rtrange = NULL, step = 0.1) object = \"xcmsSet\" getEIC(object, mzrange, rtrange = 200, groupidx,                sampleidx = sampnames(object), rt = c(\"corrected\", \"raw\"))","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"object xcmsRaw xcmsSet object mzrange Either two column matrix minimum maximum m/z     matrix dimensions containing columns mzmin     mzmax. specified, method xcmsRaw returns     base peak chromatogram (BPC, .e. intense signal     RT across m/z). xcmsSet objects group data used     mzrange provided. rtrange two column matrix size mzrange minimum     maximum retention times return EIC data     points. specified, method returns chromatogram     full RT range. xcmsSet objects, may also single number     specifying time window around peak return EIC data     points step step (bin) size use profile generation. Note     value step = 0 supported. groupidx either character vector names integer vector     indicies peak groups get EICs sampleidx either character vector names integer vector     indicies samples get EICs rt \"corrected\" using corrected retention times,     \"raw\" using raw retention times","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"contrast rawEIC method, extracts   actual raw values, method extracts object's profile   matrix (provided step argument match   profStep object profile matrix calculated   fly values returned).","code":""},{"path":"https://sneumann.github.io/xcms/reference/getEIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get extracted ion chromatograms for specified m/z ranges — getEIC-methods","text":"xcmsSet xcmsRaw objects, xcmsEIC object.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get peak intensities for specified regions — getPeaks-methods","title":"Get peak intensities for specified regions — getPeaks-methods","text":"Integrate extracted ion chromatograms pre-defined defined   regions. Return output similar findPeaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get peak intensities for specified regions — getPeaks-methods","text":"object = \"xcmsRaw\" getPeaks(object, peakrange, step = 0.1)","code":""},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get peak intensities for specified regions — getPeaks-methods","text":"object xcmsSet object peakrange matrix data frame 4 columns: mzmin, mzmax,     rtmin, rtmax (must order named) step step size use profile generation","code":""},{"path":"https://sneumann.github.io/xcms/reference/getPeaks-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get peak intensities for specified regions — getPeaks-methods","text":"matrix columns: rank peak identified merged EIC (<= max), always NA mz weighted (intensity) mean peak m/z across scans mzmin m/z minimum step mzmax m/z maximum step ret retention time peak midpoint retmin leading edge peak retention time retmax trailing edge peak retention time integrated area original (raw) peak intf integrated area filtered peak, always NA maxo maximum intensity original (raw) peak maxf maximum intensity filtered peak, always NA","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get m/z and intensity values for a single mass scan — getScan-methods","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"Return data single mass scan using numeric index   scan reference.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"object = \"xcmsRaw\" getScan(object, scan, mzrange = numeric())   getMsnScan(object, scan, mzrange = numeric())","code":""},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"object xcmsRaw object scan integer index scan. negative, index numbered     end mzrange limit data points returned range,     range(mzrange)","code":""},{"path":"https://sneumann.github.io/xcms/reference/getScan-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get m/z and intensity values for a single mass scan — getScan-methods","text":"matrix two columns: mz m/z values intensity intensity values","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"Return full-resolution averaged data multiple mass scans.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"object = \"xcmsRaw\" getSpec(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"object xcmsRaw object ... arguments passed profRange used sepecify     spectral segments interest averaging","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"Based mass points spectra selected, master   unique list masses generated. Every spectra interpolated   masses averaged.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getSpec-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get average m/z and intensity values for multiple mass scans — getSpec-methods","text":"matrix two columns: mz m/z values intensity intensity values","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"Reads raw data applies evential retention time corrections   waters Lock mass correction   returns xcmsRaw object (list xcmsRaw   objects) one files xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"object = \"xcmsSet\" getXcmsRaw(object, sampleidx=1,     profmethod=profinfo(object)$method, profstep=profinfo(object)$step,     rt=c(\"corrected\", \"raw\"), ...   )","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"object xcmsSet object sampleidx index sample raw data     returned. Can single number numeric vector     indices. Alternatively, file name can specified. profmethod profile method. profstep profile step. rt Whether corrected raw retention times returned. ... Additional arguments submitted xcmsRaw function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"single xcmsRaw object list xcmsRaw objects.","code":""},{"path":"https://sneumann.github.io/xcms/reference/getXcmsRaw-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Load the raw data for one or more files in the xcmsSet — getXcmsRaw-methods","text":"Johannes Rainer, johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Group peaks from different samples together — group-methods","title":"Group peaks from different samples together — group-methods","text":"number grouping (alignment) methods exist XCMS. group   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group peaks from different samples together — group-methods","text":"object = \"xcmsSet\" group(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group peaks from different samples together — group-methods","text":"object xcmsSet-class object method Method use grouping. See details. ... Optional arguments passed along","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Group peaks from different samples together — group-methods","text":"Different algorithms can used specifying   method argument. example use density-based   approach described Smith et al (2006) one use:   group(object, method=\"density\"). also default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$group.methods.   nickname method called \"mzClust\",   help page specific method can   accessed ?group.mzClust.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group peaks from different samples together — group-methods","text":"xcmsSet object peak group assignments statistics.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":null,"dir":"Reference","previous_headings":"","what":"Group peaks from different samples together — group.density","title":"Group peaks from different samples together — group.density","text":"Group peaks together across samples using overlapping m/z bins  calculation smoothed peak distributions chromatographic time.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group peaks from different samples together — group.density","text":"object = \"xcmsSet\" group(object, bw = 30, minfrac = 0.5, minsamp = 1,     mzwid = 0.25, max = 50, sleep = 0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group peaks from different samples together — group.density","text":"object xcmsSet object minfrac minimum fraction samples necessary least one sample     groups valid group minsamp minimum number samples necessary least one sample     groups valid group bw bandwidth (standard deviation half width half maximum)     gaussian smoothing kernel apply peak density     chromatogram mzwid width overlapping m/z slices use creating peak density     chromatograms grouping peaks across samples max maximum number groups identify single m/z slice sleep seconds pause plotting successive steps peak     grouping algorithm. peaks plotted points showing relative     intensity. identified groups flanked dotted vertical     lines.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.density.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group peaks from different samples together — group.density","text":"xcmsSet object peak group assignments statistics.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":null,"dir":"Reference","previous_headings":"","what":"Group Peaks via High Resolution Alignment — group.mzClust","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"Runs high resolution alignment single spectra samples stored \tgiven xcmsSet.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"object = \"xcmsSet\" group(object, method=\"mzClust\", mzppm = 20, mzabs = 0, minsamp = 1, minfrac=0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"object xcmsSet peaks mzppm relative error used clustering/grouping ppm (parts per million) mzabs absolute error used clustering/grouping minsamp set minimum number samples one bin minfrac set minimum fraction class one bin","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"Returns xcmsSet slots groups groupindex set.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"Saira . Kazmi, Samiran Ghosh, Dong-Guk Shin, \tDennis W. Hill David F. GrantAlignment high resolution mass spectra: development heuristic \tapproach metabolomics. \tMetabolomics, Vol. 2, . 2, 75-83 (2006)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.mzClust.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Group Peaks via High Resolution Alignment — group.mzClust","text":"","code":"if (FALSE) { library(msdata) mzMLpath <- system.file(\"fticr-mzML\", package = \"msdata\") mzMLfiles <- list.files(mzMLpath, recursive = TRUE, full.names = TRUE)  xs <- xcmsSet(method=\"MSW\", files=mzMLfiles, scales=c(1,7),               SNR.method='data.mean' , winSize.noise=500,               peakThr=80000,  amp.Th=0.005)  xsg <- group(xs, method=\"mzClust\") }"},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":null,"dir":"Reference","previous_headings":"","what":"Group peaks from different samples together — group.nearest","title":"Group peaks from different samples together — group.nearest","text":"Group peaks together across samples creating master peak list  assigning corresponding peaks samples. inspired  alignment algorithm mzMine. details check  http://mzmine.sourceforge.net/ Katajamaa M, Miettinen J, Oresic M: MZmine: Toolbox processing  visualization mass spectrometry based molecular profile  data. Bioinformatics (Oxford, England) 2006, 22:634?636. Currently, equivalent minfrac minsamp.","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Group peaks from different samples together — group.nearest","text":"object = \"xcmsSet\" group(object, mzVsRTbalance=10, mzCheck=0.2, rtCheck=15, kNN=10)","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group peaks from different samples together — group.nearest","text":"object xcmsSet object mzVsRTbalance Multiplicator mz value calculating     (euclidean) distance two peaks. mzCheck Maximum tolerated distance mz. rtCheck Maximum tolerated distance RT. kNN Number nearest Neighbours check","code":""},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group peaks from different samples together — group.nearest","text":"xcmsSet object peak group assignments statistics.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/group.nearest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Group peaks from different samples together — group.nearest","text":"","code":"if (FALSE) library(xcms)     library(faahKO)     ## These files do not have this problem to correct for     ## but just for an example     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)      xset<-xcmsSet(cdffiles)      gxset<-group(xset, method=\"nearest\") #> Processing sample number 8 ...  #> OK #> Processing sample number 1 ...  #> OK #> Processing sample number 7 ...  #> OK #> Processing sample number 3 ...  #> OK #> Processing sample number 12 ...  #> OK #> Processing sample number 9 ...  #> OK #> Processing sample number 6 ...  #> OK #> Processing sample number 5 ...  #> OK #> Processing sample number 4 ...  #> OK #> Processing sample number 11 ...  #> OK #> Processing sample number 10 ...  #> OK     nrow(gxset@groups) == 1096 ## the number of features before minFrac #> [1] FALSE      post.minFrac<-function(object, minFrac=0.5){         ix.minFrac<-sapply(1:length(unique(sampclass(object))),                            function(x, object, mf){                                meta<-groups(object)                                minFrac.idx<-numeric(length=nrow(meta))                                idx<-which(                                    meta[,levels(sampclass(object))[x]] >=                                    mf*length(which(levels(sampclass(object))[x]                                                    == sampclass(object)) ))                                minFrac.idx[idx]<-1                                return(minFrac.idx)                            }, object, minFrac)         ix.minFrac<-as.logical(apply(ix.minFrac, 1, sum))         ix<-which(ix.minFrac == TRUE)         return(ix)     }      ## using the above function we can get a post processing minFrac     idx<-post.minFrac(gxset)      gxset.post<-gxset ## copy the xcmsSet object     gxset.post@groupidx<-gxset@groupidx[idx]     gxset.post@groups<-gxset@groups[idx,]      nrow(gxset.post@groups) == 465 ## number of features after minFrac #> [1] FALSE"},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"groupChromPeaks method performs correspondence analysis .e., groups chromatographic peaks across samples define LC-MS features. correspondence algorithm can selected, configured, using param argument. See documentation XcmsExperiment() XCMSnExp() information access extract correspondence results. correspondence analysis can performed chromatographic peaks MS level (present chromatographic peak detection performed MS level) defining features combining peaks. MS level can selected parameter msLevel. default, calling groupChromPeaks remove previous correspondence results. can disabled add = TRUE, add newly defined features already present feature definitions. Supported param objects : PeakDensityParam: correspondence using peak density method (Smith 2006) groups chromatographic peaks along retention time axis within slices (partially overlapping) m/z ranges. default, m/z ranges (bins) constant size. setting ppm value larger 0, m/z dependent bin sizes can used instead (better representing m/z dependent measurement error MS instruments). peaks (different samples) apex position close retention time axis grouped LC-MS feature. See addition do_groupChromPeaks_density() core API function. NearestPeaksParam: performs peak grouping based proximity chromatographic peaks different samples m/z - rt space similar correspondence method mzMine (Katajamaa 2006). method creates first master peak list consisting chromatographic peaks sample detected peaks iteratively calculates distances peaks sample next number peaks grouping peaks together distance smaller provided thresholds. See addition do_groupChromPeaks_nearest() core API function. MzClustParam: performs high resolution peak grouping single spectrum metabolomics data (Kazmi 2006). method used data retention time considered correspondence analysis. See addition do_groupPeaks_mzClust() core API function. specific examples description method settings see help pages individual parameter classes listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"","code":"groupChromPeaks(object, param, ...)  # S4 method for XcmsExperiment,Param groupChromPeaks(object, param, msLevel = 1L, add = FALSE)  PeakDensityParam(   sampleGroups = numeric(),   bw = 30,   minFraction = 0.5,   minSamples = 1,   binSize = 0.25,   ppm = 0,   maxFeatures = 50 )  MzClustParam(   sampleGroups = numeric(),   ppm = 20,   absMz = 0,   minFraction = 0.5,   minSamples = 1 )  NearestPeaksParam(   sampleGroups = numeric(),   mzVsRtBalance = 10,   absMz = 0.2,   absRt = 15,   kNN = 10 )  # S4 method for PeakDensityParam sampleGroups(object)  # S4 method for PeakDensityParam sampleGroups(object) <- value  # S4 method for PeakDensityParam bw(object)  # S4 method for PeakDensityParam bw(object) <- value  # S4 method for PeakDensityParam minFraction(object)  # S4 method for PeakDensityParam minFraction(object) <- value  # S4 method for PeakDensityParam minSamples(object)  # S4 method for PeakDensityParam minSamples(object) <- value  # S4 method for PeakDensityParam binSize(object)  # S4 method for PeakDensityParam binSize(object) <- value  # S4 method for PeakDensityParam maxFeatures(object)  # S4 method for PeakDensityParam maxFeatures(object) <- value  # S4 method for PeakDensityParam ppm(object)  # S4 method for MzClustParam sampleGroups(object)  # S4 method for MzClustParam sampleGroups(object) <- value  # S4 method for MzClustParam ppm(object)  # S4 method for MzClustParam ppm(object) <- value  # S4 method for MzClustParam absMz(object)  # S4 method for MzClustParam absMz(object) <- value  # S4 method for MzClustParam minFraction(object)  # S4 method for MzClustParam minFraction(object) <- value  # S4 method for MzClustParam minSamples(object)  # S4 method for MzClustParam minSamples(object) <- value  # S4 method for NearestPeaksParam sampleGroups(object)  # S4 method for NearestPeaksParam sampleGroups(object) <- value  # S4 method for NearestPeaksParam mzVsRtBalance(object)  # S4 method for NearestPeaksParam mzVsRtBalance(object) <- value  # S4 method for NearestPeaksParam absMz(object)  # S4 method for NearestPeaksParam absMz(object) <- value  # S4 method for NearestPeaksParam absRt(object)  # S4 method for NearestPeaksParam absRt(object) <- value  # S4 method for NearestPeaksParam kNN(object)  # S4 method for NearestPeaksParam kNN(object) <- value  # S4 method for XCMSnExp,PeakDensityParam groupChromPeaks(object, param, msLevel = 1L, add = FALSE)  # S4 method for XCMSnExp,MzClustParam groupChromPeaks(object, param, msLevel = 1L)  # S4 method for XCMSnExp,NearestPeaksParam groupChromPeaks(object, param, msLevel = 1L, add = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"object data object correspondence analysis performed. Can XCMSnExp(), XcmsExperiment() object. param parameter object selecting configuring algorithm. ... Optional parameters. msLevel integer(1) defining MS level chromatographic peak detection performed. add logical(1) (object contains already chromatographic peaks, .e. either XCMSnExp XcmsExperiment) whether chromatographic peak detection results added existing results. default (add = FALSE) additional findChromPeaks call result object remove previous results. sampleGroups PeakDensityParam: vector length samples defining sample group assignments (.e. samples belong sample group). parameter mandatory PeakDensityParam provided also sample grouping experiment (case samples assigned group). bw PeakDensityParam: numeric(1) defining bandwidth (standard deviation ot smoothing kernel) used. argument passed [density() method. minFraction PeakDensityParam: numeric(1) defining minimum fraction samples least one sample group peaks present considered peak group (feature). minSamples PeakDensityParam: numeric(1) minimum number samples least one sample group peaks detected considered peak group (feature). binSize PeakDensityParam: numeric(1) defining size overlapping slices m/z dimension. ppm MzClustParam: numeric(1) representing relative m/z error clustering/grouping (parts per million). PeakDensityParam: numeric(1) define m/z-dependent, increasing m/z bin sizes. ppm = 0 (default) m/z bins defined sequence values smallest larges m/z value constant bin size binSize. ppm > 0 size bin increased addition ppm (upper) m/z boundary bin. maximal bin size (used largest m/z values) binSize plus ppm parts-per-million largest m/z value peaks data set. maxFeatures PeakDensityParam: numeric(1) maximum number peak groups identified single mz slice. absMz NearestPeaksParam MzClustParam: numeric(1) maximum tolerated distance m/z values. mzVsRtBalance NearestPeaksParam: numeric(1) representing factor m/z values multiplied calculating (euclician) distance two peaks. absRt NearestPeaksParam: numeric(1) maximum tolerated distance retention times. kNN NearestPeaksParam: integer(1) representing number nearest neighbors check. value value slot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"groupChromPeaks: either XcmsExperiment() XCMSnExp() object correspondence result.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"Smith, C.., Want E.J., O'Maille G., Abagyan R., Siuzdak G. (2006) \"XCMS: Processing Mass Spectrometry Data Metabolite Profiling Using Nonlinear Peak Alignment, Matching, Identification\" Anal. Chem. 78:779-787. Katajamaa, M., Miettinen, J., Oresic, M. (2006) \"MZmine: Toolbox processing visualization mass spectrometry based molecular profile data\". Bioinformatics, 22:634-636. Kazmi S. ., Ghosh, S., Shin, D., Hill, D.W., Grant, D.F. (2006) \"Alignment high resolution mass spectra: development heuristic approach metabolomics. Metabolomics Vol. 2, . 2, 75-83.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correspondence: group chromatographic peaks across samples — groupChromPeaks","text":"Colin Smith, Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"Features originating compound expected similar intensities across samples. method thus groups features based similarity abundances (.e. feature values) across samples data set. See also AbundanceSimilarityParam() additional information details. help page lists parameters specific xcms result objects (.e. XcmsExperiment() XCMSnExp() objects). Documentation parameters similarity calculation available AbundanceSimilarityParam() help page MsFeatures package.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"","code":"# S4 method for XcmsResult,AbundanceSimilarityParam groupFeatures(   object,   param,   msLevel = 1L,   method = c(\"medret\", \"maxint\", \"sum\"),   value = \"into\",   intensity = \"into\",   filled = TRUE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"object XcmsExperiment() XCMSnExp() object containing LC-MS pre-processing results. param AbudanceSimilarityParam object settings method. See AbundanceSimilarityParam() details grouping method parameters. msLevel integer(1) defining MS level features grouped. method character(1) passed featureValues call. See featureValues() details. Defaults method = \"medret\". value character(1) passed featureValues call. See featureValues() details. Defaults value = \"\". intensity character(1) passed featureValues call. See featureValues() details. Defaults intensity = \"\". filled logical(1) whether filled-values included correlation analysis. Defaults filled = TRUE. ... additional parameters passed groupFeatures method matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"input object feature group definitions added (updated ) column \"feature_group\" featureDefinitions data frame.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-abundance-correlation.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compounding/feature grouping based on similarity of abundances across samples — groupFeatures-abundance-correlation","text":"","code":"library(MsFeatures) ## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Group chromatographic peaks across samples xodg <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = rep(1, 3)))  ## Group features based on correlation of feature values (integrated ## peak area) across samples. Note that there are many missing values ## in the feature value which influence grouping of features in the present ## data set. xodg_grp <- groupFeatures(xodg,     param = AbundanceSimilarityParam(threshold = 0.8)) table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008  #>      8      8      8      6     12      3      1      1   ## Group based on the maximal peak intensity per feature xodg_grp <- groupFeatures(xodg,     param = AbundanceSimilarityParam(threshold = 0.8, value = \"maxo\")) table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008  #>      8      8      8      6     12      3      1      1"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"Features originating compound expected share elution pattern (.e. chromatographic peak shape) . Thus, methods allows group features based similarity extracted ion chromatograms (EICs). similarity calculation performed separately sample similarity score aggregated across samples final generation similarity matrix grouping (considering parameter threshold) performed. compareChromatograms() function used similarity calculation default calculates Pearson's correlation coefficient. settings compareChromatograms can specified parameters ALIGNFUN, ALIGNFUNARGS, FUN FUNARGS. ALIGNFUN defaults alignRt() function used align chromatograms comparison. ALIGNFUNARGS allows specify additional arguments ALIGNFUN function. defaults ALIGNFUNARGS = list(tolerance = 0, method = \"closest\") ensures data points spectrum (scan, .e. retention time) compared EICs sample. Parameter FUN defines function calculate similarity score defaults FUN = cor FUNARGS allows pass additional arguments function (defaults FUNARGS = list(use = \"pairwise.complete.obs\"). See also compareChromatograms() information. grouping features based EIC similarity matrix performed function specified parameter groupFun defaults groupFun = groupSimilarityMatrix groups rows (features) similarity matrix similarity score larger threshold cluster. creates clusters features features similarity score >= threshold feature cluster. See groupSimilarityMatrix() details. Additional parameters function can passed ... argument. feature grouping called initial feature grouping retention time (see SimilarRtimeParam()). feature groups defined columns \"feature_group\" featureDefinitions(object) (features matching msLevel) used refined method. Features value NA featureDefinitions(object)$feature_group skipped/considered feature grouping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"","code":"EicSimilarityParam(   threshold = 0.9,   n = 1,   onlyPeak = TRUE,   value = c(\"maxo\", \"into\"),   groupFun = groupSimilarityMatrix,   ALIGNFUN = alignRt,   ALIGNFUNARGS = list(tolerance = 0, method = \"closest\"),   FUN = cor,   FUNARGS = list(use = \"pairwise.complete.obs\"),   ... )  # S4 method for XcmsResult,EicSimilarityParam groupFeatures(object, param, msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"threshold numeric(1) minimal required similarity score group featues. passed groupFun function. n numeric(1) defining total number samples per feature group similarity calculation performed. value rounded next larger integer value. onlyPeak logical(1) whether correlation performed signals within identified chromatographic peaks (onlyPeak = TRUE, default) signal extracted ion chromatogram. value character(1) defining whether samples grouped based sum maximal peak intensity (value = \"maxo\", default) integrated peak area (value = \"\") feature. groupFun function defining function used group rows based pairwise similarity matrix. Defaults groupSimilarityMatrix(). ALIGNFUN function defining function used align chromatograms prior similarity calculation. Defaults ALIGNFUN = alignRt. See alignRt() compareChromatograms() information. ALIGNFUNARGS named list arguments ALIGNFUN. Defaults ALIGNFUNARGS = list(tolerance = 0, method = \"closest\"). FUN function defining function used calculate similarity (aligned) chromatograms. Defaults FUN = cor. See cor() compareChromatograms() information. FUNARGS named list arguments FUN. Defaults FUN = list(use = \"pairwise.complete.obs\"). ... EicSimilarityParam: additional arguments passed groupFun featureChromatograms (expandRt expand retention time range feature). object XcmsExperiment() XCMSnExp() object LC-MS pre-processing results. param EicSimilarityParam object settings method. msLevel integer(1) defining MS level features grouped.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"input object feature groups added (.e. column \"feature_group\" featureDefinitions data frame.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"present featureChromatograms() function used extract EICs feature, however use one m/z rt range feature EICs thus exactly represent identified chromatographic peaks sample (.e. specific m/z retention time ranges). possible performed full data set without prior feature grouping, suggested following reasons: ) selection top n samples highest signal feature group biased abundant compounds performed full data set (.e. samples highest overall intensities used correlation features) II) computationally much expensive pairwise correlation features performed. also suggested perform correlation subset samples per feature highest intensities peaks (feature) although also possible run correlation samples setting n equal total number samples data set. EIC correlation however performed ideally samples original compound highly abundant avoid correlation missing values noisy peak shapes much possible. default also signal outside identified chromatographic peaks excluded correlation.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-eic-similarity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compounding/feature grouping based on similarity of extracted ion chromatograms — groupFeatures-eic-similarity","text":"","code":"library(MsFeatures) ## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Group chromatographic peaks across samples xodg <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = rep(1, 3)))  ## Performing a feature grouping based on EIC similarities on a single ## sample xodg_grp <- groupFeatures(xodg, param = EicSimilarityParam(n = 1))  table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  #>      2      2      2      3      4      2      3      2      1      1      1  #> FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.034 FG.035  #>      1      1   ## Usually it is better to perform this correlation on pre-grouped features ## e.g. based on similar retention time. xodg_grp <- groupFeatures(xodg, param = SimilarRtimeParam(diffRt = 4)) xodg_grp <- groupFeatures(xodg_grp, param = EicSimilarityParam(n = 1))  table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001.001 FG.002.001 FG.003.001 FG.003.002 FG.004.001 FG.004.002 FG.005.001  #>          2          2          2          1          3          1          3  #> FG.006.001 FG.006.002 FG.007.001 FG.008.001 FG.009.001 FG.009.002 FG.010.001  #>          1          1          2          2          1          1          2  #> FG.011.001 FG.012.001 FG.013.001 FG.014.001 FG.015.001 FG.016.001 FG.017.001  #>          1          1          1          1          1          1          1  #> FG.018.001 FG.019.001 FG.020.001 FG.021.001 FG.022.001 FG.023.001 FG.024.001  #>          1          1          1          1          1          1          1  #> FG.025.001 FG.026.001 FG.027.001 FG.028.001 FG.029.001 FG.030.001 FG.031.001  #>          1          1          1          1          1          1          1  #> FG.032.001 FG.033.001  #>          1          1"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"Group features based similar retention time. method supposed used initial crude grouping features based median retention time chromatographic peaks. features difference retention time <= parameter diffRt parameter object grouped together. column \"feature_group\" found featureDefinitions() sub-grouped method. See MsFeatures::SimilarRtimeParam() MsFeatures details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"","code":"# S4 method for XcmsResult,SimilarRtimeParam groupFeatures(object, param, msLevel = 1L, ...)"},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"object XcmsExperiment() XCMSnExp() object containing also correspondence results. param SimilarRtimeParam object settings method. See MsFeatures::SimilarRtimeParam() details options. msLevel integer(1) defining MS level features grouped. ... passed groupFeatures function numeric values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"input object feature groups added (.e. column \"feature_group\" featureDefinitions data frame.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupFeatures-similar-rtime.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compounding/feature grouping based on similar retention times — groupFeatures-similar-rtime","text":"","code":"library(MsFeatures) ## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Group chromatographic peaks across samples xodg <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = rep(1, 3)))  ## Group features based on similar retention time (i.e. difference <= 2 seconds) xodg_grp <- groupFeatures(xodg, param = SimilarRtimeParam(diffRt = 2))  ## Feature grouping get added to the featureDefinitions in column \"feature_group\" head(featureDefinitions(xodg_grp)$feature_group) #> [1] \"FG.004\" \"FG.009\" \"FG.003\" \"FG.004\" \"FG.010\" \"FG.011\"  table(featureDefinitions(xodg_grp)$feature_group) #>  #> FG.001 FG.002 FG.003 FG.004 FG.005 FG.006 FG.007 FG.008 FG.009 FG.010 FG.011  #>      2      2      3      3      3      2      2      2      1      1      1  #> FG.012 FG.013 FG.014 FG.015 FG.016 FG.017 FG.018 FG.019 FG.020 FG.021 FG.022  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.023 FG.024 FG.025 FG.026 FG.027 FG.028 FG.029 FG.030 FG.031 FG.032 FG.033  #>      1      1      1      1      1      1      1      1      1      1      1  #> FG.034 FG.035 FG.036  #>      1      1      1  length(unique(featureDefinitions(xodg_grp)$feature_group)) #> [1] 36  ## Using an alternative groupiing method that creates larger groups xodg_grp <- groupFeatures(xodg,     param = SimilarRtimeParam(diffRt = 2, groupFun = MsCoreUtils::group))  length(unique(featureDefinitions(xodg_grp)$feature_group)) #> [1] 35"},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":null,"dir":"Reference","previous_headings":"","what":"Group overlapping ranges — groupOverlaps","title":"Group overlapping ranges — groupOverlaps","text":"groupOverlaps identifies overlapping ranges input data groups returning indices xmin xmax.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Group overlapping ranges — groupOverlaps","text":"","code":"groupOverlaps(xmin, xmax)"},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Group overlapping ranges — groupOverlaps","text":"xmin numeric (length xmax) lower boundary range. xmax numeric (length xmin) upper boundary range.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Group overlapping ranges — groupOverlaps","text":"list indices grouped elements.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Group overlapping ranges — groupOverlaps","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupOverlaps.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Group overlapping ranges — groupOverlaps","text":"","code":"x <- c(2, 12, 34.2, 12.4) y <- c(3, 16, 35, 36)  groupOverlaps(x, y) #> [[1]] #> [1] 1 #>  #> [[2]] #> [1] 2 3 4 #>"},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"groupnames generates names identified features correspondence analysis based mass retention time. generates feature names equivalent group names old user interface (aka xcms1).","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp groupnames(object, mzdec = 0, rtdec = 0, template = NULL)"},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"object XCMSnExp object containing correspondence results. mzdec integer(1) number decimal places use m/z ( defaults 0). rtdec integer(1) number decimal places use retention time (defaults 0). template character existing group names whose format emulated.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-XCMSnExp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate unique group (feature) names based on mass and retention time — groupnames,XCMSnExp-method","text":"character unique names feature object. format M(m/z)T(time seconds).","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate unque names for peak groups — groupnames-methods","title":"Generate unque names for peak groups — groupnames-methods","text":"Allow linking peak group data classes using unique   group names remain long re-grouping occurs.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Generate unque names for peak groups — groupnames-methods","text":"object = \"xcmsSet\" (object, mzdec = 0, rtdec = 0, template = NULL) object = \"xcmsEIC\" (object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate unque names for peak groups — groupnames-methods","text":"object xcmsSet xcmsEIC object mzdec number decimal places use m/z rtdec number decimal places use retention time template character vector existing group names whose format     emulated","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupnames-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate unque names for peak groups — groupnames-methods","text":"character vector unique names peak group   object. format M[m/z]T[time seconds].","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract a matrix of peak values for each group — groupval-methods","title":"Extract a matrix of peak values for each group — groupval-methods","text":"Generate matrix peak values rows every group   columns every sample. value included matrix can   columns xcmsSet peaks slot   matrix. Collisions one peak single sample   group get resolved one several user-selectable   methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Extract a matrix of peak values for each group — groupval-methods","text":"object = \"xcmsSet\" groupval(object, method = c(\"medret\", \"maxint\"),                  value = \"index\", intensity = \"\")","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract a matrix of peak values for each group — groupval-methods","text":"object xcmsSet object method conflict resolution method, \"medret\" use peak     closest median retention time \"maxint\" use     peak highest intensity value name peak column enter returned matrix, \"index\"     index corresponding row peaks slot     matrix intensity method == \"maxint\", name peak column use     intensity","code":""},{"path":"https://sneumann.github.io/xcms/reference/groupval-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract a matrix of peak values for each group — groupval-methods","text":"matrix rows every group columns every   sample. Missing peaks NA values.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"highlightChromPeaks function adds chromatographic peak definitions existing plot, one created plot method Chromatogram MChromatograms object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"","code":"highlightChromPeaks(   x,   rt,   mz,   peakIds = character(),   border = rep(\"00000040\", length(fileNames(x))),   lwd = 1,   col = NA,   type = c(\"rect\", \"point\", \"polygon\"),   whichPeaks = c(\"any\", \"within\", \"apex_within\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"x highlightChromPeaks: XCMSnExp object detected peaks. rt highlightChromPeaks: numeric(2) retention time range peaks extracted plotted. mz numeric(2) mz range peaks extracted plotted. peakIds character defining IDs (.e. rownames peak chromPeaks table) chromatographic peaks highlighted plot. border colors used color border rectangles/peaks. equal number samples x. lwd numeric(1) defining width line/border. col highlightChromPeaks: color used fill rectangle (type = \"rect\") peak (type = \"polygon\"). type plotting type. See plot base grapics details. highlightChromPeaks: character(1) defining peak highlighted: type = \"rect\" draws rectangle representing peak definition, type = \"point\" indicates chromatographic peak single point position peak's \"rt\" \"maxo\" type = \"polygon\" highlight peak shape. type = \"polygon\" color border area can defined parameters \"border\" \"col\", respectively. whichPeaks character(1) specifying peaks called located within region defined mz rt. Can one \"\", \"within\", \"apex_within\" peaks even partially overlapping region, peaks completely within region, peaks apex within region. parameter passed type argument chromPeaks function. See related documentation information examples. ... additional parameters matplot plot function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/highlightChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add definition of chromatographic peaks to an extracted chromatogram plot — highlightChromPeaks","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Extract the ion chromatogram for one chromatographic peak in the data. chrs <- chromatogram(faahko_sub, rt = c(2700, 2900), mz = 335)  plot(chrs)  ## Extract chromatographic peaks for the mz/rt range (if any). chromPeaks(faahko_sub, rt = c(2700, 2900), mz = 335) #>        mz mzmin mzmax       rt   rtmin    rtmax    into    intb  maxo sn sample #> CP095 335   335   335 2786.199 2764.29 2812.803 1496244 1451652 58736 55      2  ## Highlight the chromatographic peaks in the area ## Show the peak definition with a rectangle highlightChromPeaks(faahko_sub, rt = c(2700, 2900), mz = 335)  ## Color the actual peak highlightChromPeaks(faahko_sub, rt = c(2700, 2900), mz = 335,     col = c(\"#ff000020\", \"#00ff0020\"), type = \"polygon\")"},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot log intensity image of a xcmsRaw object — image-methods","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"Create log intensity false-color image xcmsRaw object plotted   m/z retention time axes","code":""},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"x = \"xcmsRaw\" image(x, col = rainbow(256), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"x xcmsRaw object col vector colors use image ... arguments profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/image.xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot log intensity image of a xcmsRaw object — image-methods","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":null,"dir":"Reference","previous_headings":"","what":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"function provides missing value imputation based linear     interpolation resembles functionality     profBinLin profBinLinBase functions deprecated     version 1.51 .","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"","code":"imputeLinInterpol(   x,   baseValue,   method = \"lin\",   distance = 1L,   noInterpolAtEnds = FALSE )"},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"x numeric vector eventual missing (NA) values. baseValue base value empty elements set. considered method = \"linbase\" corresponds profBinLinBase's baselevel argument. method One \"none\", \"lin\" \"linbase\". distance method = \"linbase\": number non-empty neighboring element empty element considered linear interpolation. See details section information. noInterpolAtEnds method = \"lin\": Logical indicating whether linear interpolation also performed ends data vector (.e. missing values present beginning end vector).","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"numeric vector empty values imputed based selected  method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"Values NAs input vector x can imputed using methods     \"lin\" \"linbase\": impute = \"lin\" uses simple linear imputation derive value     empty element input vector x neighboring     non-empty elements. method equivalent linear     interpolation profBinLin method. Whether interpolation     performed missing values present beginning end     x can set argument noInterpolAtEnds. default     interpolation also performed ends interpolating 0     beginning towards 0 end.     noInterpolAtEnds = TRUE interpolation performed     ends replacing missing values beginning /end     x 0. impute = \"linbase\" uses linear interpolation impute values     empty elements within user-definable proximity non-empty elements     setting element's value baseValue otherwise.     default baseValue half smallest value     x (NAs removed). Whether linear interpolation based     imputation performed missing value depends     distance argument. Interpolation performed one     next distance closest neighbors current empty element     value NA. interpolation takes place     distance = 0, distance = 1 means value     empty element interpolated directly adjacent non-empty     elements , next neighbors current empty element     also NA, vale set baseValue.     corresponds linear interpolation performed     profBinLinBase method. details see examples .","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeLinInterpol.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Impute values for empty elements in a vector using linear interpolation — imputeLinInterpol","text":"","code":"####### ## Impute missing values by linearly interpolating from neighboring ## non-empty elements x <- c(3, NA, 1, 2, NA, NA, 4, NA, NA, NA, 3, NA, NA, NA, NA, 2) imputeLinInterpol(x, method = \"lin\") #>  [1] 3.000000 2.000000 1.000000 2.000000 2.666667 3.333333 4.000000 3.750000 #>  [9] 3.500000 3.250000 3.000000 2.800000 2.600000 2.400000 2.200000 2.000000 ## visualize the interpolation: plot(x = 1:length(x), y = x) points(x = 1:length(x), y = imputeLinInterpol(x, method = \"lin\"), type = \"l\", col = \"grey\")   ## If the first or last elements are NA, interpolation is performed from 0 ## to the first non-empty element. x <- c(NA, 2, 1, 4, NA) imputeLinInterpol(x, method = \"lin\") #> [1] 1 2 1 4 2 ## visualize the interpolation: plot(x = 1:length(x), y = x) points(x = 1:length(x), y = imputeLinInterpol(x, method = \"lin\"), type = \"l\", col = \"grey\")   ## If noInterpolAtEnds is TRUE no interpolation is performed at both ends imputeLinInterpol(x, method = \"lin\", noInterpolAtEnds = TRUE) #> [1] 0 2 1 4 0  ###### ## method = \"linbase\" ## \"linbase\" performs imputation by interpolation for empty elements based on ## 'distance' adjacent non-empty elements, setting all remaining empty elements ## to the baseValue x <- c(3, NA, 1, 2, NA, NA, 4, NA, NA, NA, 3, NA, NA, NA, NA, 2) ## Setting distance = 0 skips imputation by linear interpolation imputeLinInterpol(x, method = \"linbase\", distance = 0) #>  [1] 3.0 0.5 1.0 2.0 0.5 0.5 4.0 0.5 0.5 0.5 3.0 0.5 0.5 0.5 0.5 2.0  ## With distance = 1 for all empty elements next to a non-empty element the value ## is imputed by linear interpolation. xInt <- imputeLinInterpol(x, method = \"linbase\", distance = 1L) xInt #>  [1] 3.000000 2.000000 1.000000 2.000000 2.666667 3.333333 4.000000 2.250000 #>  [9] 0.500000 1.750000 3.000000 1.750000 0.500000 0.500000 1.250000 2.000000  plot(x = 1:length(x), y = x, ylim = c(0, max(x, na.rm = TRUE))) points(x = 1:length(x), y = xInt, type = \"l\", col = \"grey\")   ## Setting distance = 2L would cause that for all empty elements for which the ## distance to the next non-empty element is <= 2 the value is imputed by ## linear interpolation: xInt <- imputeLinInterpol(x, method = \"linbase\", distance = 2L) xInt #>  [1] 3.000000 2.000000 1.000000 2.000000 2.666667 3.333333 4.000000 3.750000 #>  [9] 3.500000 3.250000 3.000000 2.800000 2.600000 2.400000 2.200000 2.000000  plot(x = 1:length(x), y = x, ylim = c(0, max(x, na.rm = TRUE))) points(x = 1:length(x), y = xInt, type = \"l\", col = \"grey\")"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":null,"dir":"Reference","previous_headings":"","what":"Replace missing values with a proportion of the row minimum — imputeRowMin","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"imputeRowMin imputes missing values x replacing NAs row proportion minimal value row (.e. min_fraction * min(x[, ])).","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"","code":"imputeRowMin(x, min_fraction = 1/2)"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"x matrix abundances, rows features/metabolites columns samples. min_fraction numeric(1) fraction row minimum used replace NA values row.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMin.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Replace missing values with a proportion of the row minimum — imputeRowMin","text":"","code":"library(faahKO) data(\"faahko\")  xset <- group(faahko) mat <- groupval(xset, value = \"into\")  mat_imp <- imputeRowMin(mat)  head(mat) #>                 ko15      ko16       ko18      ko19       ko21      ko22 #> 200.1/2926  147887.5  451600.7   65290.38        NA   91635.45  162012.4 #> 205/2791   1778568.9 1567038.1 1482796.38 1039129.8 1223132.35 1072037.7 #> 206/2790    237993.6  269714.0  201393.42  150107.3  176989.65  156797.0 #> 207.1/2719  380873.0  460629.7  351750.14  219288.0  286848.56  235022.6 #> 219.1/2525  235544.9  173623.4         NA        NA  185792.43  174458.8 #> 231/2517          NA        NA  222609.07  286232.1  435094.49        NA #>                 wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.1   82619.48         NA   69198.22  153273.5   98144.28 #> 205/2791   1950287.5 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.8  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.6  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.5  161184.05   72029.38         NA  238194.4  173829.95 #> 231/2517          NA         NA         NA  240261.21  201316.2  179437.72 head(mat_imp) #>                  ko15       ko16       ko18       ko19       ko21       ko22 #> 200.1/2926  147887.53  451600.71   65290.38   32645.19   91635.45  162012.44 #> 205/2791   1778568.94 1567038.14 1482796.38 1039129.82 1223132.35 1072037.70 #> 206/2790    237993.62  269713.98  201393.42  150107.31  176989.65  156797.04 #> 207.1/2719  380873.05  460629.74  351750.14  219287.97  286848.56  235022.63 #> 219.1/2525  235544.92  173623.38   36014.69   36014.69  185792.43  174458.77 #> 231/2517     89718.86   89718.86  222609.07  286232.15  435094.49   89718.86 #>                  wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.08   82619.48   32645.19   69198.22  153273.5   98144.28 #> 205/2791   1950287.49 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.85  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.58  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.47  161184.05   72029.38   36014.69  238194.4  173829.95 #> 231/2517     89718.86   89718.86   89718.86  240261.21  201316.2  179437.72  ## Replace with 1/8 of the row mimimum head(imputeRowMin(mat, min_fraction = 1/8)) #>                  ko15       ko16        ko18        ko19       ko21       ko22 #> 200.1/2926  147887.53  451600.71   65290.385    8161.298   91635.45  162012.44 #> 205/2791   1778568.94 1567038.14 1482796.385 1039129.818 1223132.35 1072037.70 #> 206/2790    237993.62  269713.98  201393.415  150107.310  176989.65  156797.04 #> 207.1/2719  380873.05  460629.74  351750.138  219287.968  286848.56  235022.63 #> 219.1/2525  235544.92  173623.38    9003.672    9003.672  185792.43  174458.77 #> 231/2517     22429.72   22429.72  222609.068  286232.146  435094.49   22429.72 #>                  wt15       wt16        wt18        wt19      wt21       wt22 #> 200.1/2926  175177.08   82619.48    8161.298   69198.222  153273.5   98144.28 #> 205/2791   1950287.49 1466780.60 1572679.161 1275312.758 1356014.3 1231442.16 #> 206/2790    276541.85  222366.15  211717.713  186850.878  188285.9  172348.76 #> 207.1/2719  417169.58  324892.46  277990.701  220972.352  252874.0  236728.16 #> 219.1/2525  244584.47  161184.05   72029.378    9003.672  238194.4  173829.95 #> 231/2517     22429.72   22429.72   22429.715  240261.212  201316.2  179437.72"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":null,"dir":"Reference","previous_headings":"","what":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"Replace missing values random numbers. using method = \"mean_sd\", random numbers generated normal distribution based (fraction ) row min standard deviation estimated linear relationship row standard deviation mean full data set. Parameter sd_fraction allows reduce estimated standard deviation. using method method = \"from_to\", random numbers 2 specific values generated.","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"","code":"imputeRowMinRand(   x,   method = c(\"mean_sd\", \"from_to\"),   min_fraction = 1/2,   min_fraction_from = 1/1000,   sd_fraction = 1,   abs = TRUE )"},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"x matrix abundances, rows features/metabolites columns samples. method method character(1) defining imputation method. See description details. Defaults method = \"mean_sd\". min_fraction numeric(1) fraction row minimum used replace NA values row case mean_sd method specified.  using from_to method, value one used calculate maximum value replace NA values row. min_fraction_from numeric(1) fraction row minimum used calculate minimum value replace NA values row. parameter used case from_to method specified. sd_fraction numeric(1) factor reduce estimated standard deviation. parameter used case mean_sd method specified. abs logical(1) force imputed values strictly positive.","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"method mean_sd, imputed values taken normal distribution mean user defined fraction row minimum standard deviation estimated mean based linear relationship row standard deviations row means full matrix x. largely avoid imputed values negative larger real values, standard deviation random number generation estimated ignoring intercept linear model estimating relationship standard deviation mean. abs = TRUE NA values replaced absolute value random values. method from_to, imputed values taken 2 user defined fractions row minimum.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"Johannes Rainer, Mar Garcia-Aloy","code":""},{"path":"https://sneumann.github.io/xcms/reference/imputeRowMinRand.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Impute missing values with random numbers based on the row minimum — imputeRowMinRand","text":"","code":"library(faahKO) data(\"faahko\")  xset <- group(faahko) mat <- groupval(xset, value = \"into\")  ## Estimate the relationship between row sd and mean. The standard deviation ## of the random distribution is estimated on this relationship. mns <- rowMeans(mat, na.rm = TRUE) sds <- apply(mat, MARGIN = 1, sd, na.rm = TRUE) plot(mns, sds) abline(lm(sds ~ mns))   mat_imp_meansd <- imputeRowMinRand(mat, method = \"mean_sd\") mat_imp_fromto <- imputeRowMinRand(mat, method = \"from_to\")  head(mat) #>                 ko15      ko16       ko18      ko19       ko21      ko22 #> 200.1/2926  147887.5  451600.7   65290.38        NA   91635.45  162012.4 #> 205/2791   1778568.9 1567038.1 1482796.38 1039129.8 1223132.35 1072037.7 #> 206/2790    237993.6  269714.0  201393.42  150107.3  176989.65  156797.0 #> 207.1/2719  380873.0  460629.7  351750.14  219288.0  286848.56  235022.6 #> 219.1/2525  235544.9  173623.4         NA        NA  185792.43  174458.8 #> 231/2517          NA        NA  222609.07  286232.1  435094.49        NA #>                 wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.1   82619.48         NA   69198.22  153273.5   98144.28 #> 205/2791   1950287.5 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.8  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.6  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.5  161184.05   72029.38         NA  238194.4  173829.95 #> 231/2517          NA         NA         NA  240261.21  201316.2  179437.72 head(mat_imp_meansd) #>                 ko15       ko16       ko18       ko19       ko21       ko22 #> 200.1/2926  147887.5  451600.71   65290.38   49246.37   91635.45  162012.44 #> 205/2791   1778568.9 1567038.14 1482796.38 1039129.82 1223132.35 1072037.70 #> 206/2790    237993.6  269713.98  201393.42  150107.31  176989.65  156797.04 #> 207.1/2719  380873.0  460629.74  351750.14  219287.97  286848.56  235022.63 #> 219.1/2525  235544.9  173623.38   19850.45   46023.31  185792.43  174458.77 #> 231/2517    149786.4   79591.63  222609.07  286232.15  435094.49   59448.34 #>                 wt15       wt16       wt18       wt19      wt21       wt22 #> 200.1/2926  175177.1   82619.48   16968.88   69198.22  153273.5   98144.28 #> 205/2791   1950287.5 1466780.60 1572679.16 1275312.76 1356014.3 1231442.16 #> 206/2790    276541.8  222366.15  211717.71  186850.88  188285.9  172348.76 #> 207.1/2719  417169.6  324892.46  277990.70  220972.35  252874.0  236728.16 #> 219.1/2525  244584.5  161184.05   72029.38   43139.31  238194.4  173829.95 #> 231/2517    120349.2  123319.72   95360.60  240261.21  201316.2  179437.72 head(mat_imp_fromto) #>                  ko15       ko16       ko18       ko19       ko21       ko22 #> 200.1/2926  147887.53  451600.71   65290.38   19617.55   91635.45  162012.44 #> 205/2791   1778568.94 1567038.14 1482796.38 1039129.82 1223132.35 1072037.70 #> 206/2790    237993.62  269713.98  201393.42  150107.31  176989.65  156797.04 #> 207.1/2719  380873.05  460629.74  351750.14  219287.97  286848.56  235022.63 #> 219.1/2525  235544.92  173623.38   14992.43   21335.01  185792.43  174458.77 #> 231/2517     42550.31   45065.53  222609.07  286232.15  435094.49   80160.63 #>                 wt15       wt16       wt18         wt19      wt21       wt22 #> 200.1/2926  175177.1   82619.48   12629.94   69198.2221  153273.5   98144.28 #> 205/2791   1950287.5 1466780.60 1572679.16 1275312.7582 1356014.3 1231442.16 #> 206/2790    276541.8  222366.15  211717.71  186850.8779  188285.9  172348.76 #> 207.1/2719  417169.6  324892.46  277990.70  220972.3520  252874.0  236728.16 #> 219.1/2525  244584.5  161184.05   72029.38     391.5481  238194.4  173829.95 #> 231/2517     67198.8   34599.03   52982.49  240261.2115  201316.2  179437.72"},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"isolationWindowTargetMz extracts isolation window target m/z definition spectrum object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"","code":"# S4 method for OnDiskMSnExp isolationWindowTargetMz(object)"},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"object OnDiskMSnExp object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"numeric length equal number spectra object isolation window target m/z NA specified/available.","code":""},{"path":"https://sneumann.github.io/xcms/reference/isolationWindowTargetMz-OnDiskMSnExp-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract isolation window target m/z definition — isolationWindowTargetMz,OnDiskMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot log intensity image of a xcmsRaw object — levelplot-methods","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"Create image raw (profile) data m/z retention   time, intensity color coded.","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"x = \"xcmsRaw\" levelplot(x, log=TRUE, col.regions=colorRampPalette(brewer.pal(9,     \"YlOrRd\"))(256), ...) x = \"xcmsSet\" levelplot(x, log=TRUE, col.regions=colorRampPalette(brewer.pal(9,     \"YlOrRd\"))(256), rt=\"raw\", ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"x xcmsRaw object. log Whether intensity log transformed. col.regions color ramp used encoding     intensity. rt wheter original (rt=\"raw\") corrected     (rt=\"corrected\") retention times used. ... Arguments profRange.","code":""},{"path":"https://sneumann.github.io/xcms/reference/levelplot.xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot log intensity image of a xcmsRaw object — levelplot-methods","text":"Johannes Rainer, johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Read binary data from a source — loadRaw-methods","title":"Read binary data from a source — loadRaw-methods","text":"function extracts raw data used   xcmsRaw object. processing data   done xcmsRaw constructor.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Read binary data from a source — loadRaw-methods","text":"signature(object = \"xcmsSource\") Uses loadRaw,xcmsSource-method extract raw data.       Subclasses xcmsSource can provide different       ways fetching data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Read binary data from a source — loadRaw-methods","text":"object Specification data source (file name     database query)","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Read binary data from a source — loadRaw-methods","text":"implementing methods decide gather data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Read binary data from a source — loadRaw-methods","text":"list containing elements describing data source. rt,  scanindex, tic, acquisitionNum components   one entry per scan. parallel sense  rt[1], scanindex[1], acquisitionNum[1]   refer scan. list containst following components: rt Numeric vector acquisition time (seconds) scan tic Numeric vector Total Ion Count scan scanindex Integer vector starting positions scan mz     intensity components. exclusive offset,     scanindex[] offset mz intensity     beginning scan . means     mz (respectively intensity) values scan     scanindex[] + 1 scanindex[+ 1] mz Concatenated vector m/z values scans intensity Concatenated vector intensity values scans","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadRaw-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Read binary data from a source — loadRaw-methods","text":"Daniel Hackney, dan@haxney.org","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":null,"dir":"Reference","previous_headings":"","what":"LC-MS preprocessing result test data sets — loadXcmsData","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"Data sets `xcms` preprocessing results provided within `xcms` package can loaded `loadXcmsData` function. available Test data sets : - `xdata`: [XCMSnExp()] object results `xcms`-based   pre-processing LC-MS untargeted metabolomics data set. raw data   files provided `faahKO` R package. - `xmse`: [XcmsExperiment()] object results `xcms`-based   pre-processing LC-MS untargeted metabolomics data set (original   data set pre-processing settings `xdata` data set).   pre-processing data set described detail *xcms*   vignette `xcms` package. - `faahko_sub`: [XCMSnExp()] object identified   chromatographic peaks 3 samples data files `faahKO`   R package. - `faahko_sub2`: [XcmsExperiment()] object identified   chromatographic peaks 3 samples data files `faahKO`   R package. Data sets can also loaded using `data`, however require update objects point location raw data files. `loadXcmsData` loads data ensures paths updated accordingly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"","code":"loadXcmsData(x = c(\"xmse\", \"xdata\", \"faahko_sub\", \"faahko_sub2\"))"},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"x `loadXcmsData`: `character(1)` name data file (object) load.","code":""},{"path":"https://sneumann.github.io/xcms/reference/loadXcmsData.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"LC-MS preprocessing result test data sets — loadXcmsData","text":"","code":"library(xcms) xdata <- loadXcmsData()"},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Manual peak integration and feature definition — manualChromPeaks","title":"Manual peak integration and feature definition — manualChromPeaks","text":"manualChromPeaks function allows manually define chromatographic peaks, integrate intensities within specified peak area add object's chromPeaks matrix. peak added sample signal found respective data file. chromatographic peaks added eventually previously identified peaks, suggested run refineChromPeaks() MergeNeighboringPeaksParam() approach merge potentially overlapping peaks. manualFeatures function allows manually group identified chromatographic peaks features providing index object's chromPeaks matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Manual peak integration and feature definition — manualChromPeaks","text":"","code":"manualChromPeaks(object, ...)  manualFeatures(object, ...)  # S4 method for MsExperiment manualChromPeaks(   object,   chromPeaks = matrix(numeric()),   samples = seq_along(object),   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment manualChromPeaks(   object,   chromPeaks = matrix(numeric()),   samples = seq_along(object),   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment manualFeatures(object, peakIdx = list(), msLevel = 1L)  # S4 method for OnDiskMSnExp manualChromPeaks(   object,   chromPeaks = matrix(),   samples = seq_along(fileNames(object)),   msLevel = 1L,   BPPARAM = bpparam() )  # S4 method for XCMSnExp manualChromPeaks(   object,   chromPeaks = matrix(),   samples = seq_along(fileNames(object)),   msLevel = 1L,   BPPARAM = bpparam() )  # S4 method for XCMSnExp manualFeatures(object, peakIdx = list(), msLevel = 1L)"},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Manual peak integration and feature definition — manualChromPeaks","text":"object XcmsExperiment, XCMSnExp OnDiskMSnExp object. ... ignored. chromPeaks manualChromPeaks: matrix defining boundaries chromatographic peaks one row per chromatographic peak columns \"mzmin\", \"mzmax\", \"rtmin\" \"rtmax\" defining m/z retention time region peak. samples manualChromPeaks: optional integer defining individual samples peak integration performed. Defaults samples. msLevel integer(1) defining MS level peak integration performed. single MS level time supported. Defaults msLevel = 1L. chunkSize integer(1) defining number files (samples) loaded memory processed time. Peak integration performed parallel (per sample) subset data. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM parallel processing settings (see bpparam() details). peakIdx manualFeatures: list integer vectors indices chromatographic peaks object's chromPeaks matrix grouped features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Manual peak integration and feature definition — manualChromPeaks","text":"XcmsExperiment XCMSnExp manually added chromatographic peaks features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Manual peak integration and feature definition — manualChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/manualChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Manual peak integration and feature definition — manualChromPeaks","text":"","code":"## Read a test dataset. fls <- c(system.file(\"microtofq/MM14.mzML\", package = \"msdata\"),          system.file(\"microtofq/MM8.mzML\", package = \"msdata\"))  ## Define a data frame with some sample annotations ann <- data.frame(     injection_index = 1:2,     sample_id = c(\"MM14\", \"MM8\"))  ## Import the data library(MsExperiment) mse <- readMsExperiment(fls)  ## Define some arbitrary peak areas pks <- cbind(     mzmin = c(512, 234.3), mzmax = c(513, 235),     rtmin = c(10, 33), rtmax = c(19, 50) ) pks #>      mzmin mzmax rtmin rtmax #> [1,] 512.0   513    10    19 #> [2,] 234.3   235    33    50  res <- manualChromPeaks(mse, pks) chromPeaks(res) #>           mz mzmin mzmax       rt rtmin rtmax      into sample maxo #> CP1 512.6294 512.0   513 18.66000    10    19  848.4231      2   85 #> CP2 234.8539 234.3   235 41.54298    33    50 2690.4200      2  245  ## Peaks were only found in the second file."},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Apply a median filter to a matrix — medianFilter","title":"Apply a median filter to a matrix — medianFilter","text":"element matix, replace median   values around .","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Apply a median filter to a matrix — medianFilter","text":"","code":"medianFilter(x, mrad, nrad)"},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Apply a median filter to a matrix — medianFilter","text":"x numeric matrix median filter mrad number rows either side value use median     calculation nrad number rows either side value use median     calculation","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Apply a median filter to a matrix — medianFilter","text":"matrix whose values median filtered","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Apply a median filter to a matrix — medianFilter","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/medianFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Apply a median filter to a matrix — medianFilter","text":"","code":"mat <- matrix(1:25, nrow=5) mat #>      [,1] [,2] [,3] [,4] [,5] #> [1,]    1    6   11   16   21 #> [2,]    2    7   12   17   22 #> [3,]    3    8   13   18   23 #> [4,]    4    9   14   19   24 #> [5,]    5   10   15   20   25 medianFilter(mat, 1, 1) #>      [,1] [,2] [,3] [,4] [,5] #> [1,]  4.0  6.5 11.5 16.5 19.0 #> [2,]  4.5  7.0 12.0 17.0 19.5 #> [3,]  5.5  8.0 13.0 18.0 20.5 #> [4,]  6.5  9.0 14.0 19.0 21.5 #> [5,]  7.0  9.5 14.5 19.5 22.0"},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"MS2 MSn data stored separate slots,   can directly used e.g. findPeaks().   msn2xcmsRaw() copy MSn spectra   \"normal\" xcmsRaw slots.","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"","code":"msn2xcmsRaw(xmsn)"},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"xmsn object class xcmsRaw contains     spectra read includeMSn=TRUE","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"default gap value determined 90th percentile   pair-wise differences adjacent mass values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"xcmsRaw object","code":""},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"Steffen Neumann sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/msn2xcmsRaw.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Copy MSn data in an xcmsRaw to the MS slots — msn2xcmsRaw","text":"","code":"msnfile <- system.file(\"microtofq/MSMSpos20_6.mzML\", package = \"msdata\")  xrmsn <- xcmsRaw(msnfile, includeMSn=TRUE) #> Create profile matrix with method 'bin' and step 1 ...  #> OK  xr <- msn2xcmsRaw(xrmsn)  p <- findPeaks(xr, method=\"centWave\") #> Detecting mass traces at 25 ppm ...  #> OK #> Detecting chromatographic peaks in 10 regions of interest ... #>  OK: 10 found."},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":null,"dir":"Reference","previous_headings":"","what":"Fill in NA values at the extremes of a vector — na.flatfill","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"Extend first last real values vector fill   NA values present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"","code":"na.flatfill(x)"},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"x numeric vector NA values","code":""},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"Modified vector.","code":""},{"path":"https://sneumann.github.io/xcms/reference/na.flatfill.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fill in NA values at the extremes of a vector — na.flatfill","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify overlapping features — overlappingFeatures","title":"Identify overlapping features — overlappingFeatures","text":"overlappingFeatures identifies features overlapping close m/z - rt space.","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify overlapping features — overlappingFeatures","text":"","code":"overlappingFeatures(x, expandMz = 0, expandRt = 0, ppm = 0)"},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify overlapping features — overlappingFeatures","text":"x XcmsExperiment() XCMSnExp() object features. expandMz numeric(1) value expand feature (side) m/z dimension identifying overlapping features. resulting \"mzmin\" feature thus mzmin - expandMz \"mzmax\" mzmax + expandMz. expandRt numeric(1) value expand feature (side) retention time dimension identifying overlapping features. resulting \"rtmin\" feature thus rtmin - expandRt \"rtmax\" rtmax + expandRt. ppm numeric(1) grow m/z width feature relative value: mzmin - mzmin * ppm / 2e6, mzmax + mzmax * ppm / 2e6. feature thus expanded m/z dimension ppm/2 side identifying overlapping features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify overlapping features — overlappingFeatures","text":"list indices features (featureDefinitions()) overlapping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify overlapping features — overlappingFeatures","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/overlappingFeatures.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Identify overlapping features — overlappingFeatures","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Disable parallel processing for this example register(SerialParam())  ## Correspondence analysis xdata <- groupChromPeaks(faahko_sub, param = PeakDensityParam(sampleGroups = c(1, 1, 1)))  ## Identify overlapping features overlappingFeatures(xdata) #> list()  ## Identify features that are separated on retention time by less than ## 2 minutes overlappingFeatures(xdata, expandRt = 60) #> [[1]] #> [1] 5 6 #>  #> [[2]] #> [1] 31 32 #>"},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":null,"dir":"Reference","previous_headings":"","what":"Correlation coefficient panel for pairs function — panel.cor","title":"Correlation coefficient panel for pairs function — panel.cor","text":"Correlation coefficient panel pairs function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Correlation coefficient panel for pairs function — panel.cor","text":"","code":"panel.cor(x, y, digits = 2, prefix = \"\", cex.cor)"},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correlation coefficient panel for pairs function — panel.cor","text":"x first data series y second data series digits number digits plot prefix text prefix coefficients cex.cor character expansion factor","code":""},{"path":"https://sneumann.github.io/xcms/reference/panel.cor.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correlation coefficient panel for pairs function — panel.cor","text":"Colin . Smith, csmith@scripps.edu, based   pairs example code","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot a grid of a large number of peaks — peakPlots-methods","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"Plot extracted ion chromatograms many peaks simultaneously, indicating peak integration start end points vertical grey lines.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"signature(object = \"xcmsSet\") plotPeaks(object, peaks, figs, width = 200)","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"object xcmsRaw object peaks matrix peak information produced findPeaks figs two-element vector describing number rows number     columns peaks plot, missing approximately     square grid fit number peaks supplied width width chromatogram retention time plot peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakPlots-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot a grid of a large number of peaks — peakPlots-methods","text":"function intended help graphically analyze results   peak picking. can help estimate number false positives   improper integration start end points. output   compact tries waste little space possible. plot   labeled rounded m/z retention time separated   space.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Create report of aligned peak intensities — peakTable-methods","title":"Create report of aligned peak intensities — peakTable-methods","text":"Create report showing aligned peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create report of aligned peak intensities — peakTable-methods","text":"object = \"xcmsSet\" peakTable(object, filebase = character(), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create report of aligned peak intensities — peakTable-methods","text":"object xcmsSet object filebase base file name save report, .tsv file _eic     appended name tabular report EIC     directory, respectively. blank nothing saved ... arguments passed groupval,     provides actual intensities.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create report of aligned peak intensities — peakTable-methods","text":"method handles creation summary reports similar   diffreport. returns summary report can   optionally written tab-separated file. base file name provided, report (see Value section)   saved tab separated file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peakTable-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create report of aligned peak intensities — peakTable-methods","text":"data frame following columns: mz median m/z peaks group mzmin minimum m/z peaks group mzmax maximum m/z peaks group rt median retention time peaks group rtmin minimum retention time peaks group rtmax maximum retention time peaks group npeaks number peaks assigned group Sample Classes number samples sample class represented group ... one column every sample class Sample Names integrated intensity value every sample ... one column every sample","code":""},{"path":[]},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"peaksWithCentWave identifies (chromatographic) peaks purely chromatographic data, .e. based intensity retention time values without m/z values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"","code":"peaksWithCentWave(   int,   rt,   peakwidth = c(20, 50),   snthresh = 10,   prefilter = c(3, 100),   integrate = 1,   fitgauss = FALSE,   noise = 0,   verboseColumns = FALSE,   firstBaselineCheck = TRUE,   extendLengthMSW = FALSE,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"int numeric intensity values. rt numeric retention time intensities. Length equal length(int). peakwidth numeric(2) lower upper bound expected peak width. snthresh numeric(1) defining signal noise ratio cutoff. Peaks signal noise ratio < snthresh omitted. prefilter numeric(2) (c(k, )): regions interest least k centroids signal >= returned first step. integrate numeric(1), integration method. integrate = 1 peak limits found descending mexican hat filtered data, integrate = 2 descend done real data. latter method accurate prone noise, former robust, less exact. fitgauss logical(1) whether Gaussian fitted peak. noise numeric(1) defining minimum required intensity centroids considered first analysis step (definition regions interest). verboseColumns logical(1): whether additional peak meta data columns returned. firstBaselineCheck logical(1). TRUE continuous data within regions interest checked first baseline. detail, first rough estimate noise calculated peak detection performed regions multiple sequential signals higher first estimated baseline/noise level. extendLengthMSW logical(1). TRUE \"open\" method EIC extension used, rather default \"reflect\" method. ... currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"matrix, row representing identified chromatographic peak, columns: \"rt\": retention time peak's midpoint (time maximum signal). \"rtmin\": minimum retention time peak. \"rtmax\": maximum retention time peak. \"\": integrated (original) intensity peak. \"intb\": per-peak baseline corrected integrated peak intensity. \"maxo\": maximum (original) intensity peak. \"sn\": signal noise ratio peak defined (maxo - baseline)/sd sd standard deviation local chromatographic noise. Additional columns verboseColumns = TRUE: \"mu\": gaussian parameter mu. \"sigma\": gaussian parameter sigma. \"h\": gaussian parameter h. \"f\": region number m/z ROI peak localized. \"dppm\": m/z deviation mass trace across scans ppm (always NA). \"scale\": scale peak localized. \"scpos\": peak position found wavelet analysis (index int). \"scmin\": left peak limit found wavelet analysis (index int). \"scmax\": right peak limit found wavelet analysis (index int).","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"method uses algorithm peak detection centWave, employs however different approach identify initial regions peak detection performed (.e. regions interest ROI). method first identifies local maxima chromatographic data defines corresponding positions +/- peakwidth[2] ROIs. Noise estimation bases also ROIs can thus different centWave resulting different signal noise ratios.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithCentWave.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Identify peaks in chromatographic data using centWave — peaksWithCentWave","text":"","code":"## Reading a file library(xcms) od <- readMSData(system.file(\"cdf/KO/ko15.CDF\", package = \"faahKO\"),     mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Extract chromatographic data for a small m/z range mzr <- c(272.1, 272.2) chr <- chromatogram(od, mz = mzr, rt = c(3000, 3300))[1, 1]  int <- intensity(chr) rt <- rtime(chr)  ## Plot the region plot(chr, type = \"h\")  ## Identify peaks in the chromatographic data pks <- peaksWithCentWave(intensity(chr), rtime(chr)) pks #>            rt    rtmin    rtmax     into     intb  maxo sn #> [1,] 3119.535 3097.625 3127.359 247211.6 227217.9 16496 24  ## Highlight the peaks rect(xleft = pks[, \"rtmin\"], xright = pks[, \"rtmax\"],     ybottom = rep(0, nrow(pks)), ytop = pks[, \"maxo\"], col = \"#ff000040\",     border = \"#00000040\")"},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"function performs peak detection using matchedFilter algorithm chromatographic data (.e. intensities retention time).","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"","code":"peaksWithMatchedFilter(   int,   rt,   fwhm = 30,   sigma = fwhm/2.3548,   max = 20,   snthresh = 10,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"int numeric intensity values. rt numeric retention time intensities. Length equal length(int). fwhm numeric(1) specifying full width half maximum matched filtration gaussian model peak. used calculate actual sigma, see . sigma numeric(1) specifying standard deviation (width) matched filtration model peak. max numeric(1) maximal number peaks expected/ bbe detected data snthresh numeric(1) defining signal noise cut-used peak detection step. ... currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"matrix, row representing identified chromatographic peak, columns: \"rt\": retention time peak's midpoint (time maximum signal). \"rtmin\": minimum retention time peak. \"rtmax\": maximum retention time peak. \"\": integrated (original) intensity peak. \"intf\": integrated intensity filtered peak. \"maxo\": maximum (original) intensity peak. \"maxf\"\" maximum intensity filtered peak. \"sn\": signal noise ratio peak.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/peaksWithMatchedFilter.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Identify peaks in chromatographic data using matchedFilter — peaksWithMatchedFilter","text":"","code":"## Load the test file faahko_sub <- loadXcmsData(\"faahko_sub\")  ## Subset to one file and drop identified chromatographic peaks data <- dropChromPeaks(filterFile(faahko_sub, 1))  ## Extract chromatographic data for a small m/z range chr <- chromatogram(data, mz = c(272.1, 272.3), rt = c(3000, 3200))[1, 1]  pks <- peaksWithMatchedFilter(intensity(chr), rtime(chr)) pks #>            rt   rtmin    rtmax     into     intf  maxo     maxf       sn #> [1,] 3119.535 3105.45 3133.619 332583.6 668444.6 16496 36129.39 12.33813  ## Plotting the data plot(rtime(chr), intensity(chr), type = \"h\") rect(xleft = pks[, \"rtmin\"], xright = pks[, \"rtmax\"], ybottom = c(0, 0),     ytop = pks[, \"maxo\"], border = \"red\")"},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":null,"dir":"Reference","previous_headings":"","what":"Derive experimental design from file paths — phenoDataFromPaths","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"phenoDataFromPaths function builds data.frame representing experimental design folder structure files experiment located.","code":""},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"","code":"phenoDataFromPaths(paths)"},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"paths character representing file names (including full path) experiment's files.","code":""},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"function used old xcmsSet function guess experimental design (.e. group assignment files) folders files experiment can found.","code":""},{"path":"https://sneumann.github.io/xcms/reference/phenoDataFromPaths.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Derive experimental design from file paths — phenoDataFromPaths","text":"","code":"## List the files available in the faahKO package base_dir <- system.file(\"cdf\", package = \"faahKO\") cdf_files <- list.files(base_dir, recursive = TRUE, full.names = TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"Batch plot list extracted ion chromatograms current   graphics device.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"x = \"xcmsEIC\" plot.xcmsEIC(x, y, groupidx = groupnames(x), sampleidx = sampnames(x), rtrange = x@rtrange,              col = rep(1, length(sampleidx)), legtext = NULL, peakint = TRUE, sleep = 0, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"x xcmsEIC object y optional xcmsSet object peak integration data groupidx either character vector names integer vector     indicies peak groups plot EICs sampleidx either character vector names integer vector     indicies samples plot EICs rtrange two column matrix minimum maximum retention times     return EIC data points number rows number groups     xcmsEIC object, sampleidx used subset     . otherwise, repeated length sampleidx may also single number specifying time window around     peak plot EIC data col color use plotting extracted ion chromatograms. missing     y specified, colors taken     unclass(sampclass(y)) default palette length number groups xcmsEIC     object, sampleidx used subset . otherwise,     repeated length sampleidx legtext text use legend. NULL y specified,     legend text taken sample class information found     xcmsSet peakint logical, plot integrated peak area darkened lines (requires     y also specified) sleep seconds pause plotting EICs ... graphical parameters","code":""},{"path":"https://sneumann.github.io/xcms/reference/plot.xcmsEIC.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot extracted ion chromatograms from multiple files — plot.xcmsEIC","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":null,"dir":"Reference","previous_headings":"","what":"Visualization of Alignment Results — plotAdjustedRtime","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"`plotAdjustedRtime` function plots difference adjusted *raw* retention times y-axis raw retention times x-axis. line represents results one sample (file). alignment performed using *peak groups* method (see [adjustRtime()] infromation) also peak groups used alignment visualized.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"","code":"plotAdjustedRtime(   object,   col = \"#00000080\",   lty = 1,   lwd = 1,   type = \"l\",   adjustedRtime = TRUE,   xlab = ifelse(adjustedRtime, yes = expression(rt[adj]), no = expression(rt[raw])),   ylab = expression(rt[adj] - rt[raw]),   peakGroupsCol = \"#00000060\",   peakGroupsPch = 16,   peakGroupsLty = 3,   ylim,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"object [XcmsExperiment()] [XCMSnExp()] object alignment results. col color(s) individual lines. length 1 equal number samples. lty line type lines individual samples. lwd line width lines individual samples. type plot *type* (see [par()] options; defaults `type = \"l\"`). adjustedRtime `logical(1)` whether adjusted raw retention times shown x-axis. xlab label x-axis. ylab label y-axis. peakGroupsCol color used peak groups (alignment performed using *peak groups* method. peakGroupsPch point character (`pch`) used peak groups (alignment performed using *peak groups* method. peakGroupsLty line type (`lty`) used connect points peak groups (alignment performed using *peak groups* method. ylim optional `numeric(2)` upper lower limits y-axis.b ... Additional arguments passed `plot` function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotAdjustedRtime.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Visualization of Alignment Results — plotAdjustedRtime","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  ## Performing the peak grouping using the \"peak density\" method. p <- PeakDensityParam(sampleGroups = c(1, 1, 1)) res <- groupChromPeaks(faahko_sub, param = p)  ## Perform the retention time adjustment using peak groups found in both ## files. fgp <- PeakGroupsParam(minFraction = 1) res <- adjustRtime(res, param = fgp) #> Performing retention time correction using 19 peak groups. #> Warning: Span too small for 'loess' and the available number of peak groups, resetting to 0.21 #> Warning: Adjusted retention times had to be re-adjusted for some files to ensure them being in the same order than the raw retention times. A call to 'dropAdjustedRtime' might thus fail to restore retention times of chromatographic peaks to their original values. Eventually consider to increase the value of the 'span' parameter.  ## Visualize the impact of the alignment. plotAdjustedRtime(res, adjusted = FALSE) grid()"},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"Uses pre-generated profile mode matrix plot averaged   base peak extracted ion chromatograms specified mass range.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"object = \"xcmsRaw\" plotChrom(object, base = FALSE, ident = FALSE,                   fitgauss = FALSE, vline = numeric(0), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"object xcmsRaw object base logical, plot base-peak chromatogram ident logical, use mouse identify label peaks fitgauss logical, fit gaussian largest peak vline numeric vector locations vertical lines ... arguments passed profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChrom-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot extracted ion chromatograms from the profile matrix — plotChrom-methods","text":"ident == TRUE, integer vector indecies   points identified. fitgauss == TRUE,  nls model fitted gaussian. Otherwise two-column   matrix plotted points.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"Plot density chromatographic peaks along retention time axis indicate peaks () grouped feature based using peak density correspondence method. Settings peak density method can passed PeakDensityParam object parameter param. object contains correspondence results correspondence performed peak groups method, results correspondence can visualized setting simulate = FALSE.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"","code":"# S4 method for XCMSnExp plotChromPeakDensity(   object,   mz,   rt,   param,   simulate = TRUE,   col = \"#00000080\",   xlab = \"retention time\",   ylab = \"sample\",   xlim = range(rt),   main = NULL,   type = c(\"any\", \"within\", \"apex_within\"),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"object XCMSnExp object identified chromatographic peaks. mz numeric(2) defining mz range peak density plotted. rt numeric(2) defining optional rt range peak density plotted. Defaults absolute retention time range object. param PeakDensityParam parameters peak density correspondence algorithm can extracted. provided object contains feature definitions correspondence/ peak grouping performed peak density method, corresponding parameter class stored object used. simulate logical(1) defining whether correspondence simulated within specified m/z / rt region (simulate = FALSE) whether results already performed correspondence shown. col Color used individual samples. Length 1 equal number samples object. xlab character(1) label x-axis. ylab character(1) label y-axis. xlim numeric(2) representing limits x-axis. Defaults range rt parameter. main character(1) defining title plot. default (main = NULL) mz-range used. type character(1) specifying peaks called located within region defined mz rt. Can one \"\", \"within\", \"apex_within\" peaks even partially overlapping region, peaks completely within region, peaks apex within region. parameter passed chromPeaks function. See related documentation information examples. ... Additional parameters passed plot function. Data point specific parameters bg pch length 1 equal number samples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"function called side effect, .e. create plot.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"plotChromPeakDensity function allows evaluate different settings peak density mz slice interest (e.g. containing chromatographic peaks corresponding known metabolite). plot shows individual peaks detected within specified mz slice retention time (x-axis) sample detected (y-axis). density function plotted black line. Parameters density function taken param object. Grey rectangles indicate chromatographic peaks grouped feature peak density correspondence method. Parameters algorithm also taken param. See groupChromPeaks() information algorithm supported settings.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeakDensity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot chromatographic peak density along the retention time axis — plotChromPeakDensity,XCMSnExp-method","text":"","code":"## Load a test data set with detected peaks data(faahko_sub) ## Update the path to the files for the local system dirname(faahko_sub) <- system.file(\"cdf/KO\", package = \"faahKO\")  ## Plot the chromatographic peak density for a specific mz range to evaluate ## different peak density correspondence settings. mzr <- c(305.05, 305.15)  plotChromPeakDensity(faahko_sub, mz = mzr, pch = 16,     param = PeakDensityParam(sampleGroups = rep(1, length(fileNames(faahko_sub))))) #> Warning: Use of 'plotChromPeakDensity' on 'XCMSnExp' isdiscouraged. Please extract chromatographic data first and call 'plotChromPeakDensity' directly on the 'XChromatograms' object. See ?XChromatograms, section 'Correspondence analysis' for more details."},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"General visualizations of peak detection results — plotChromPeaks","title":"General visualizations of peak detection results — plotChromPeaks","text":"`plotChromPeaks` plots identified chromatographic peaks one file plane spanned retention time (x-axis) m/z (y-axis) dimension. chromatographic peak plotted rectangle representing width RT m/z dimension. `plotChromPeakImage` plots number detected peaks sample along retention time axis *image* plot, .e. number peaks detected bin along retention time represented color respective cell.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"General visualizations of peak detection results — plotChromPeaks","text":"","code":"plotChromPeaks(   x,   file = 1,   xlim = NULL,   ylim = NULL,   add = FALSE,   border = \"#00000060\",   col = NA,   xlab = \"retention time\",   ylab = \"mz\",   main = NULL,   msLevel = 1L,   ... )  plotChromPeakImage(   x,   binSize = 30,   xlim = NULL,   log = FALSE,   xlab = \"retention time\",   yaxt = par(\"yaxt\"),   main = \"Chromatographic peak counts\",   msLevel = 1L,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"General visualizations of peak detection results — plotChromPeaks","text":"x [XcmsExperiment()] [XCMSnExp()] object. file `plotChromPeaks`: `integer(1)` specifying index file within `x` plot created. Defaults `file = 1`. xlim `numeric(2)` specifying x-axis limits (retention time dimension). Defaults `xlim = NULL` case full retention time range file used. ylim `plotChromPeaks`: `numeric(2)` specifying y-axis limits (m/z dimension). Defaults `ylim = NULL` case full m/z range file used. add `plotChromPeaks`: `logical(1)` whether plot added existing plot new plot created. border `plotChromPeaks`: color rectangles' border. col `plotChromPeaks`: color used fill rectangles. xlab `character(1)` defining x-axis label. ylab `plotChromPeaks`: `character(1)` defining y-axis label. main `character(1)` defining plot title. default (.e. `main = NULL`) name file used title. msLevel `integer(1)` defining MS level peaks visualized. ... Additional arguments passed `plot` (`plotChromPeaks`) `image` (`plotChromPeakImage`) functions. Ignored `add = TRUE`. binSize `plotChromPeakImage`: `numeric(1)` defining size bins along x-axis (retention time). Defaults `binSize = 30`, peaks within 30 seconds thus counted plotted. log `plotChromPeakImage`: `logical(1)` whether peak counts log2 transformed plotting. yaxt `plotChromPeakImage`: `character(1)` defining whether y-axis labels added. disable y-axis use `yaxt = \"n\"`. value `yaxt` axis drawn. See [par()] help page details.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"General visualizations of peak detection results — plotChromPeaks","text":"width line type rectangles indicating detected chromatographic peaks `plotChromPeaks` function can specified using `par` function, .e. `par(lwd = 3)` `par(lty = 2)`, respectively.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"General visualizations of peak detection results — plotChromPeaks","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"General visualizations of peak detection results — plotChromPeaks","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## plotChromPeakImage: plot an image for the identified peaks per file plotChromPeakImage(faahko_sub)   ## Show all detected chromatographic peaks from the first file plotChromPeaks(faahko_sub)   ## Plot all detected peaks from the second file and restrict the plot to a ## mz-rt slice plotChromPeaks(faahko_sub, file = 2, xlim = c(3500, 3600), ylim = c(400, 600))"},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"plotOverlay draws chromatographic peak data multiple (different) extracted ion chromatograms (EICs) plot. allows directly compare peak shape EICs sample. contrast plot function MChromatograms() object, draws data EIC across multiple samples plot, function draws different EICs sample plot. plotChromatogramsOverlay called XChromatograms object present chromatographic peaks also highlighted/drawn depending parameters peakType, peakCol, peakBg peakPch (see also help plot function XChromatogram() object details).","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"","code":"# S4 method for MChromatograms plotChromatogramsOverlay(   object,   col = \"#00000060\",   type = \"l\",   main = NULL,   xlab = \"rtime\",   ylab = \"intensity\",   xlim = numeric(),   ylim = numeric(),   stacked = 0,   transform = identity,   ... )  # S4 method for XChromatograms plotChromatogramsOverlay(   object,   col = \"#00000060\",   type = \"l\",   main = NULL,   xlab = \"rtime\",   ylab = \"intensity\",   xlim = numeric(),   ylim = numeric(),   peakType = c(\"polygon\", \"point\", \"rectangle\", \"none\"),   peakBg = NULL,   peakCol = NULL,   peakPch = 1,   stacked = 0,   transform = identity,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"object MChromatograms() XChromatograms() object. col definition color chromatograms drawn. Can length 1 equal nrow(object) plot overlayed chromatogram different color. type character(1) defing type plot. default (type = \"l\") chromatogram drawn line. main optional title plot. defined, range m/z values used. xlab character(1) defining x-axis label. ylab character(1) defining y-axis label. xlim optional numeric(2) defining x-axis limits. ylim optional numeric(2) defining y-axis limits. stacked numeric(1) defining part (proportion) y-axis use stack EICs depending m/z values. stacked = 0 (default) stacking performed. stacked = 1 half y-axis used stacking half intensity y-axis (.e. ratio stacking intensity y-axis 1:1). Note stacking different 0 y-axis label drawn. transform function transform intensity values plotting. Defaults transform = identity plots data . transform = log10 intensity values log10 transformed plotting. ... optional arguments passed plotting functions (see help base R plot function. peakType object XChromatograms object: chromatographic peaks drawn: peakType = \"polygon\" (default): label full chromatographic peak area, peakType = \"rectangle\": indicate chromatographic peak rectangle peakType  = \"point\": label chromatographic peaks' apex position point. peakBg object XChromatograms object: definition background color(s) chromatographic peak. either length 1 equal number peaks object. specified, peak drawn color defined col. peakCol object XChromatograms object: definition color(s) chromatographic peak. either length 1 equal number peaks object. specified, peak drawn color defined col. peakPch object XChromatograms object: point character used label apex position chromatographic peak peakType = \"point\".","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"silently returns list (length equal ncol(object) numeric (length equal nrow(object)) y position EIC.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotChromatogramsOverlay.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot multiple chromatograms into the same plot — plotChromatogramsOverlay","text":"","code":"## Load preprocessed data and extract EICs for some features. library(xcms) xdata <- loadXcmsData() data(xdata) ## Update the path to the files for the local system dirname(xdata) <- c(rep(system.file(\"cdf\", \"KO\", package = \"faahKO\"), 4),                     rep(system.file(\"cdf\", \"WT\", package = \"faahKO\"), 4)) #> Error: unable to find an inherited method for function ‘path’ for signature ‘object = \"XcmsExperiment\"’ ## Subset to the first 3 files. xdata <- filterFile(xdata, 1:3, keepFeatures = TRUE)  ## Define features for which to extract EICs fts <- c(\"FT097\", \"FT163\", \"FT165\") chrs <- featureChromatograms(xdata, features = fts)  plotChromatogramsOverlay(chrs)   ## plot the overlay of EICs in the first sample plotChromatogramsOverlay(chrs[, 1])  ## Define a different color for each feature (row in chrs). By default, also ## all chromatographic peaks of a feature is labeled in the same color. plotChromatogramsOverlay(chrs[, 1],     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"))  ## Alternatively, we can define a color for each individual chromatographic ## peak and provide this with the `peakBg` and `peakCol` parameters. chromPeaks(chrs[, 1]) #>              mz mzmin mzmax       rt    rtmin    rtmax     into     intb  maxo #> CP2973 354.1357 354.1 354.2 2789.781 2755.258 2826.270 299005.1       NA  9700 #> CP0312 413.1000 413.1 413.1 3637.671 3614.978 3665.850 879851.5 852195.3 30920 #> CP0089 414.2000 414.2 414.2 3066.457 3038.628 3089.311 234308.5 218923.4  9669 #>        sn sample row column #> CP2973 NA      1   1      1 #> CP0312 34      1   2      1 #> CP0089 15      1   3      1  ## Use a color for each of the two identified peaks in that sample plotChromatogramsOverlay(chrs[, 1],     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"),     peakBg = c(\"#ffff0020\", \"#00ffff20\"))   ## Plotting the data in all samples. plotChromatogramsOverlay(chrs,     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"))   ## Creating a \"stacked\" EIC plot: the EICs are placed along the y-axis ## relative to their m/z value. With `stacked = 1` the y-axis is split in ## half, the lower half being used for the stacking of the EICs, the upper ## half being used for the *original* intensity axis. res <- plotChromatogramsOverlay(chrs[, 1], stacked = 1,     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\")) ## add horizontal lines for the m/z values of each EIC abline(h = res[[1]], col = \"grey\", lty = 2)  ## Note that this type of visualization is different than the conventional ## plot function for chromatographic data, which will draw the EICs for ## multiple samples into the same plot plot(chrs)   ## Converting the object to a MChromatograms without detected peaks chrs <- as(chrs, \"MChromatograms\")  plotChromatogramsOverlay(chrs,     col = c(\"#ff000040\", \"#00ff0040\", \"#0000ff40\"))"},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"Plot extracted ion chromatogram m/z values interest. raw data used contrast  plotChrom uses data profile matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"object = \"xcmsRaw\" plotEIC(object, mzrange = numeric(), rtrange = numeric(),     scanrange = numeric(), mzdec=2, type=\"l\", add=FALSE, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"object xcmsRaw object mzrange m/z range EIC. Uses full m/z range default. rtrange retention time range EIC. Uses full retention time range default. scanrange scan range EIC mzdec Number decimal places title m/z values eic plot. type Speficies data plotted (default line). add EIC added existing plot. ... Additional parameters passed plotting function     (e.g. col etc).","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"two-column matrix plotted points.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotEIC-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot extracted ion chromatograms for specified m/z range — plotEIC-methods","text":"Ralf Tautenhahn","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"plotFeatureGroups visualizes defined feature groups m/z retention time space. Features indicated points features feature group connected line. See featureGroups() details options feature grouping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"","code":"plotFeatureGroups(   x,   xlim = numeric(),   ylim = numeric(),   xlab = \"retention time\",   ylab = \"m/z\",   pch = 4,   col = \"#00000060\",   type = \"o\",   main = \"Feature groups\",   featureGroups = character(),   ... )"},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"x XcmsExperiment XCMSnExp() object grouped features (.e. calling groupFeatures(). xlim numeric(2) lower upper limit x-axis. ylim numeric(2) lower upper limit y-axis. xlab character(1) label x-axis. ylab character(1) label y-axis. pch plotting character. Defaults pch = 4 .e. plotting features crosses. See par() information. col color used draw features. present single color supported. type plotting type (see par()). Defaults type = \"o\" draws feature point connecting features feature group line. main character(1) title plot. featureGroups optional character feature group IDs draw specified feature group(s). provided, feature groups drawn. ... additional parameters passed lines function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotFeatureGroups.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot feature groups in the m/z-retention time space — plotFeatureGroups","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":null,"dir":"Reference","previous_headings":"","what":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"UPDATE: please use plot(x, type = \"XIC\") MSnbase package instead. See examples . plotMsData creates plot combines (base peak ) extracted ion chromatogram top (rt intensity) plot rt m/z values bottom.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"","code":"plotMsData(   x,   main = \"\",   cex = 1,   mfrow = c(2, 1),   grid.color = \"lightgrey\",   colramp = colorRampPalette(rev(brewer.pal(9, \"YlGnBu\"))) )"},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"x data.frame returned extractMsData() function. single data.frame supported. main character(1) specifying title. cex numeric(1) defining size points. Passed directly plot function. mfrow numeric(2) defining plot layout. passed directly par(mfrow = mfrow). See par information. Setting mfrow = NULL avoids calling par(mfrow = mfrow) hence allowing pre-define plot layout. grid.color color definition grid line (NA skip creating ). colramp color ramp palette used color data points based intensity. See argument col.regions lattice::level.colors documentation.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotMsData.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"DEPRECATED: Create a plot that combines a XIC and a mz/rt 2D plot for one sample — plotMsData","text":"","code":"## Read two files from the faahKO package library(faahKO) cdfs <- dir(system.file(\"cdf\", package = \"faahKO\"), full.names = TRUE,     recursive = TRUE)[1:2] raw_data <- readMSData(cdfs, mode = \"onDisk\") #> Polarity can not be extracted from netCDF files, please set manually the polarity with the 'polarity' method.  ## Subset the object to a rt and mz range and plot the data. raw_data |>     filterRt(rt = c(2700, 2900)) |>     filterMz(mz = c(334.9, 335.1)) |>     plot(type = \"XIC\") #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 128. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 131. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 132. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 140. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 141. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 142. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 143. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 149. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 150. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 151. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 152. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 153. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 154. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 155. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 156. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 157. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 159. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 160. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 161. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 162. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 163. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 164. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 170. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 171. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 172. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 199. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 200. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 201. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 206. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 207. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 208. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 209. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 217. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 218. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 219. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 220. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 221. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 222. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 223. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 235. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 236. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 237. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 238. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 239. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 240. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 241. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 242. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 243. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 244. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 245. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 246. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 247. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 248. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 249. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 250. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 129. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 130. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 131. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 132. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 133. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 134. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 135. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 136. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 138. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 139. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 140. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 141. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 144. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 145. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 146. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 147. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 148. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 149. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 155. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 156. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 157. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 158. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 159. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 160. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 165. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 166. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 167. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 207. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 208. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 219. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 220. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 221. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 222. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 223. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 224. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 227. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 228. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 229. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 233. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 234. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 235. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 247. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 248. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 249. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 250. Returning empty spectrum. #> Warning: No data points between 334.9 and 335.1 for spectrum with acquisition #> number 251. Returning empty spectrum."},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot a grid of a large number of peaks — plotPeaks-methods","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"Plot extracted ion chromatograms many peaks simultaneously,   indicating peak integration start end points vertical   grey lines.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"object = \"xcmsRaw\" plotPeaks(object, peaks, figs, width = 200)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"object xcmsRaw object peaks matrix peak information produced findPeaks figs two-element vector describing number rows number     columns peaks plot, missing approximately     square grid fit number peaks supplied width width chromatogram retention time plot peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotPeaks-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot a grid of a large number of peaks — plotPeaks-methods","text":"function intended help graphically analyze results   peak picking. can help estimate number false positives   improper integration start end points. output   compact tries waste little space possible. plot   labeled rounded m/z retention time separated   space.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"Use \"democracy\" determine average m/z RT deviations grouped xcmsSet, dependency sample absolute m/z","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"","code":"plotQC(object, sampNames, sampColors, sampOrder, what)"},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"object grouped xcmsSet sampNames Override sample names (e.g. simplified names) sampColors Provide set colors (default: monochrome ?) sampOrder Override order samples, e.g. bring order measurement detect time drift vector QC plots generate. \"mzdevhist\": histogram mz deviations. gaussian shaped. multimodal, peaks seem systematically higher m/z deviation \"rtdevhist\":  histogram RT deviations. gaussian shaped. multimodal, peaks seem systematically higher RT deviation \"mzdevmass\": Shows whether m/z deviations absolute m/z dependent, indicate miscalibration \"mzdevtime\": Shows whether m/z deviations RT dependent, indicate instrument drift \"mzdevsample\": median mz deviation sample, indicates outliers \"rtdevsample\": median RT deviation sample, indicates outliers","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"List four matrices, dimension features * samples:  \"mz\":    median mz deviation sample  \"mzdev\": median mz deviation sample  \"rt\":    median RT deviation sample  \"rtdev\": median RT deviation sample","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"plotQC() warpper create set diagnostic plots. m/z deviations, median m/z withon one group assumed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"Michael Wenk, Michael Wenk <michael.wenk@student.uni-halle.de>","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotQC.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot m/z and RT deviations for QC purposes without external reference data — plotQC","text":"","code":"library(faahKO) xsg <- group(faahko)  plotQC(xsg, what=\"mzdevhist\")  plotQC(xsg, what=\"rtdevhist\")  plotQC(xsg, what=\"mzdevmass\")  plotQC(xsg, what=\"mzdevtime\")  plotQC(xsg, what=\"mzdevsample\")  plotQC(xsg, what=\"rtdevsample\")"},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Scatterplot of raw data points — plotRaw-methods","title":"Scatterplot of raw data points — plotRaw-methods","text":"Produce scatterplot showing raw data point location retention   time m/z. plot useful centroided data   continuum data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Scatterplot of raw data points — plotRaw-methods","text":"object = \"xcmsRaw\" plotRaw(object, mzrange = numeric(), rtrange = numeric(),                 scanrange = numeric(), log=FALSE, title='Raw Data')","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scatterplot of raw data points — plotRaw-methods","text":"object xcmsRaw object mzrange numeric vector length >= 2 whose range used select     masses plot rtrange numeric vector length >= 2 whose range used select     retention times plot scanrange numeric vector length >= 2 whose range used select     scans plot log logical, log transform intensity title main title plot","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotRaw-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scatterplot of raw data points — plotRaw-methods","text":"matrix points plotted.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotScan-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot a single mass scan — plotScan-methods","title":"Plot a single mass scan — plotScan-methods","text":"Plot single mass scan using impulse representation.   useful centroided data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotScan-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot a single mass scan — plotScan-methods","text":"object = \"xcmsRaw\" plotScan(object, scan, mzrange = numeric(), ident = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotScan-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot a single mass scan — plotScan-methods","text":"object xcmsRaw object scan integer number scan plot mzrange numeric vector length >= 2 whose range used select     masses plot ident logical, use mouse interactively identify label individual     masses","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot mass spectra from the profile matrix — plotSpec-methods","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"Uses pre-generated profile mode matrix plot mass spectra   specified retention time range.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"object = \"xcmsRaw\" plotSpec(object, ident = FALSE, vline = numeric(0), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"object xcmsRaw object ident logical, use mouse identify label peaks vline numeric vector locations vertical lines ... arguments passed profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSpec-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot mass spectra from the profile matrix — plotSpec-methods","text":"ident == TRUE, integer vector indecies   points identified. Otherwise two-column matrix   plotted points.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"method uses rgl package create interactive three   dimensonal representations profile matrix. uses terrain   color scheme.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"object = \"xcmsRaw\" plotSurf(object, log = FALSE, aspect = c(1, 1, .5), ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"object xcmsRaw object log logical, log transform intensity aspect numeric vector aspect ratio m/z, retention time     intensity components plot ... arguments passed profRange","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotSurf-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot profile matrix 3D surface using OpenGL — plotSurf-methods","text":"rgl package still development imposes limitations   output format. bug axis label code means   axis labels go 0 aspect ratio constant   axis. Additionally axes labeled . important plot small portion profile matrix.   Large portions can quickly overwhelm CPU memory.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot total ion count — plotTIC-methods","title":"Plot total ion count — plotTIC-methods","text":"Plot chromatogram total ion count. Optionally allow identification   target peaks viewing/identification individual spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot total ion count — plotTIC-methods","text":"object = \"xcmsRaw\" plotTIC(object, ident = FALSE, msident = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot total ion count — plotTIC-methods","text":"object xcmsRaw object ident logical, use mouse identify label chromatographic peaks msident logical, use mouse identify label spectral peaks","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotTIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot total ion count — plotTIC-methods","text":"ident == TRUE, integer vector indecies   points identified. Otherwise two-column matrix   plotted points.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/plotrt-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot retention time deviation profiles — plotrt-methods","title":"Plot retention time deviation profiles — plotrt-methods","text":"Use corrected retention times sample calculate retention   time deviation profiles plot graph.","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotrt-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Plot retention time deviation profiles — plotrt-methods","text":"object = \"xcmsSet\" plotrt(object, col = NULL, ty = NULL, leg = TRUE,                densplit = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/plotrt-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot retention time deviation profiles — plotrt-methods","text":"object xcmsSet object col vector colors plotting sample ty vector line point types plotting sample leg logical plot legend sample labels densplit logical, also plot peak overall peak density","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":null,"dir":"Reference","previous_headings":"","what":"Generation of profile data — profGenerate","title":"Generation of profile data — profGenerate","text":"Generates profile (binned) data given range indexed pair   vectors.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generation of profile data — profGenerate","text":"","code":"profBin(x, y, num, xstart = min(x), xend = max(x), param = list()) profBinM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,          param = list()) profBinLin(x, y, num, xstart = min(x), xend = max(x), param = list()) profBinLinM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,             param = list()) profBinLinBase(x, y, num, xstart = min(x), xend = max(x), param = list()) profBinLinBaseM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,                 param = list()) profIntLin(x, y, num, xstart = min(x), xend = max(x), param = list()) profIntLinM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,             param = list()) profMaxIdx(x, y, num, xstart = min(x), xend = max(x), param = list()) profMaxIdxM(x, y, zidx, num, xstart = min(x), xend = max(x), NAOK = FALSE,             param = list())"},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generation of profile data — profGenerate","text":"x numeric vector value positions y numeric vector values bin zidx starting position new segment num number equally spaced x bins xstart starting x value xend ending x value NAOK allow NA values (faster) param parameters profile generation","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generation of profile data — profGenerate","text":"functions take vector unequally spaced y values   transform either vector matrix, depending   whether index . point vector   matrix represents data point centered corresponding   x value, plus minus half x step size   (xend-xstart/(num-1)). Bin functions set matrix vector value   maximal point gets binned . BinLin functions except linearly   interpolate values nothing binned. BinLinBase functions except populate   empty parts spectra base value. take two parameters:   1) baselevel, intensity level fill empty parts   spectra. defaluts half minimum intensity. 2)   basespace, m/z length signal drop   base level. Linear interpolation used consecuitive   data points falling within 2*basespace eachother. defaluts   0.075. IntLin functions set matrix vector value   integral linearly interpolated data plus minus   half step size. MaxIdx functions work similarly Bin   functions execpt return integer index x,y   pair placed particular cell.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Generation of profile data — profGenerate","text":"issues profBinLin method, see   https://github.com/sneumann/xcms/issues/46   https://github.com/sneumann/xcms/issues/49. Thus suggested   use functions binYonX combination   imputeLinInterpol instead.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generation of profile data — profGenerate","text":"prof*, numeric vector length num. prof*M, matrix dimensions num  length(zidx). MaxIdx, data type integer, others   double.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generation of profile data — profGenerate","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/profGenerate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generation of profile data — profGenerate","text":"","code":"if (FALSE) {     library(faahKO)     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xraw <- xcmsRaw(cdffiles[1])      image(xraw) ## not how with intLin the intensity's blur     profMethod(xraw) <- \"bin\"     image(xraw) ## now with 'bin' there is no blurring good for centroid data     ##try binlinbase for profile data }"},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"The profile matrix — profMat,MsExperiment-method","title":"The profile matrix — profMat,MsExperiment-method","text":"profile matrix n x m matrix, n (rows) representing equally spaced m/z values (bins) m (columns) retention time corresponding scans. cell contains maximum intensity measured specific scan m/z values falling within m/z bin.","code":"The `profMat` method creates a new profile matrix or returns the profile matrix within the object's `@env` slot, if available. Settings for the profile matrix generation, such as `step` (the bin size), `method` or additional settings are extracted from the respective slots of the `xcmsRaw` object. Alternatively it is possible to specify all of the settings as additional parameters.  For [MsExperiment()] or [XcmsExperiment()] objects, the method returns a `list` of profile matrices, one for each sample in `object`. Using parameter `fileIndex` it is also possible to create a profile matrix only for selected samples (files)."},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"The profile matrix — profMat,MsExperiment-method","text":"","code":"# S4 method for MsExperiment profMat(   object,   method = \"bin\",   step = 0.1,   baselevel = NULL,   basespace = NULL,   mzrange. = NULL,   fileIndex = seq_along(object),   chunkSize = 1L,   msLevel = 1L,   BPPARAM = bpparam(),   ... )  # S4 method for xcmsRaw profMat(object, method, step, baselevel, basespace, mzrange.)"},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"The profile matrix — profMat,MsExperiment-method","text":"object xcmsRaw, OnDiskMSnExp, XCMSnExp, MsExperiment XcmsExperiment  object. method character(1) defining profile matrix generation method. Allowed \"bin\", \"binlin\", \"binlinbase\" \"intlin\". See details section information. step numeric(1) representing m/z bin size. baselevel numeric(1) representing base value empty elements (.e. m/z bins without measured intensity) set. considered method = \"binlinbase\". See baseValue parameter imputeLinInterpol() details. basespace numeric(1) representing m/z length signal drop base level. Linear interpolation used consecutive data points falling within 2 * basespace . considered method = \"binlinbase\". specified, defaults 0.075. Internally parameter translated distance parameter imputeLinInterpol() function distance = floor(basespace / step). See distance parameter imputeLinInterpol() details. mzrange. Optional numeric(2) manually specifying mz value range used binnind. provided, whole m/z value range used. fileIndex MsExperiment XcmsExperiment: integer defining idex (indices) sample(s) profile matrix created. chunkSize MsExperiment XcmsExperiment: integer(1) defining number files data loaded processed one iteration. default one file time processed chunkSize = 1L requires less memory. parallel processing, chunkSize >= number parallel processes used. msLevel MsExperiment XcmsExperiment: integer(1) defining MS level profile matrix generated. BPPARAM MsExperiment XcmsExperiment: parallel processing setup. See bpparam() details. Defaults BPPARAM = bpparam(). ... ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"The profile matrix — profMat,MsExperiment-method","text":"profMat returns profile matrix (rows representing scans, columns equally spaced m/z values). object MsExperiment XcmsExperiment, method returns list profile matrices, one file (sample).","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"The profile matrix — profMat,MsExperiment-method","text":"Profile matrix generation methods: \"bin\": default profile matrix generation method simple binning, .e. aggregating intensity values falling within m/z bin. \"binlin\": Binning followed linear interpolation impute missing values. value m/z bins without measured intensity inferred linear interpolation neighboring bins measured intensity. \"binlinbase\": Binning followed linear interpolation impute values empty elements (m/z bins) within user-definable proximity non-empty elements stetting element's value baselevel otherwise. See impute = \"linbase\" parameter imputeLinInterpol() details. \"intlin\": Set elements' values integral linearly interpolated data plus minus half step size.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"The profile matrix — profMat,MsExperiment-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMat-xcmsSet.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"The profile matrix — profMat,MsExperiment-method","text":"","code":"file <- system.file('cdf/KO/ko15.CDF', package = \"faahKO\") ## Load the data without generating the profile matrix (profstep = 0) xraw <- xcmsRaw(file, profstep = 0) ## Extract the profile matrix profmat <- profMat(xraw, step = 0.3) #> Create profile matrix with method 'bin' and step 0.3 ...  #> OK dim(profmat) #> [1] 1335 1278 ## If not otherwise specified, the settings from the xraw object are used: profinfo(xraw) #> $method #> [1] \"bin\" #>  #> $step #> [1] 0 #>  ## To extract a profile matrix with linear interpolation use profmat <- profMat(xraw, step = 0.3, method = \"binlin\") #> Create profile matrix with method 'binlin' and step 0.3 ...  #> OK ## Alternatively, the profMethod of the xraw objects could be changed profMethod(xraw) <- \"binlin\" profmat_2 <- profMat(xraw, step = 0.3) #> Create profile matrix with method 'binlin' and step 0.3 ...  #> OK all.equal(profmat, profmat_2) #> [1] TRUE"},{"path":"https://sneumann.github.io/xcms/reference/profMedFilt-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Median filtering of the profile matrix — profMedFilt-methods","title":"Median filtering of the profile matrix — profMedFilt-methods","text":"Apply median filter given size profile matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMedFilt-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Median filtering of the profile matrix — profMedFilt-methods","text":"object = \"xcmsRaw\" profMedFilt(object, massrad = 0, scanrad = 0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMedFilt-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Median filtering of the profile matrix — profMedFilt-methods","text":"object xcmsRaw object massrad number m/z grid points either side use median     calculation scanrad number scan grid points either side use median     calculation","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profMethod-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get and set method for generating profile data — profMethod-methods","title":"Get and set method for generating profile data — profMethod-methods","text":"methods get set method generating profile   (matrix) data raw mass spectral data. can currently   bin, binlin, binlinbase, intlin.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profMethod-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get and set method for generating profile data — profMethod-methods","text":"object = \"xcmsRaw\" profMethod(object)","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Specify a subset of profile mode data — profRange-methods","title":"Specify a subset of profile mode data — profRange-methods","text":"Specify subset profile mode matrix given mass, time,   scan range. Allow flexible user entry functions.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Specify a subset of profile mode data — profRange-methods","text":"object = \"xcmsRaw\" profRange(object, mzrange = numeric(),                   rtrange = numeric(), scanrange = numeric(),                   ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Specify a subset of profile mode data — profRange-methods","text":"object xcmsRaw object mzrange single numeric mass vector masses rtrange single numeric time (seconds) vector times scanrange single integer scan index vector indecies ... arguments functions","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Specify a subset of profile mode data — profRange-methods","text":"function handles selection mass/time subsets profile   matrix functions. allows user specify   subsets variety flexible ways minimal typing. R partial argument matching, mzrange,   scanrange, rtrange can specified short   form using m=, s=, t=, respectively.   scanrange rtrange specified,   rtrange specification takes precedence. specifying ranges, may either enter single number   numeric vector. single number entered, closest   single scan mass value selected. vector entered,   range set range() values entered.   allows specification ranges using shortened, slightly   non-standard syntax. example, one specify 400 500   seconds using following: t=c(400,500),   t=c(500,400), t=400:500. Use sequence   operator (:) can save several keystrokes specifying   ranges. However, sequence operator works well   specifying integer ranges, fractional ranges always work   well.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profRange-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Specify a subset of profile mode data — profRange-methods","text":"list folloing items: mzrange numeric vector start end mass masslab textual label mass range massidx integer vector mass indecies scanrange integer vector stat ane end scans scanlab textual label scan range scanidx integer vector scan range rtrange numeric vector start end times timelab textual label time range","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profStep-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get and set m/z step for generating profile data — profStep-methods","title":"Get and set m/z step for generating profile data — profStep-methods","text":"methods get set m/z step generating profile   (matrix) data raw mass spectral data. Smaller steps yield   precision cost greater memory usage.","code":""},{"path":"https://sneumann.github.io/xcms/reference/profStep-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get and set m/z step for generating profile data — profStep-methods","text":"object = \"xcmsRaw\" profStep(object)","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/profStep-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get and set m/z step for generating profile data — profStep-methods","text":"","code":"if (FALSE) {     library(faahKO)     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xset <- xcmsRaw(cdffiles[1])      xset     plotSurf(xset, mass=c(200,500))      profStep(xset)<-0.1 ## decrease the bin size to get better resolution     plotSurf(xset, mass=c(200, 500))     ##works nicer on high resolution data.   }"},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate p-values for a vector of t-statistics — pval","title":"Generate p-values for a vector of t-statistics — pval","text":"Generate p-values vector Welch's two-sample t-statistics   based t distribution.","code":""},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate p-values for a vector of t-statistics — pval","text":"","code":"pval(X, classlabel, teststat)"},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate p-values for a vector of t-statistics — pval","text":"X original data matrix classlabel integer vector classlabel teststat numeric vector Welch's two-sample t-statistics","code":""},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Generate p-values for a vector of t-statistics — pval","text":"numeric vector p-values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/pval.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate p-values for a vector of t-statistics — pval","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"Generate extracted ion chromatogram m/z values interest.     raw data used contrast  getEIC uses     data profile matrix (.e. values binned along M/Z     dimension).","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"object = \"xcmsRaw\" rawEIC(object, mzrange = numeric(), rtrange = numeric(), scanrange = numeric())","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"object xcmsRaw object mzrange m/z range EIC rtrange retention time range EIC scanrange scan range EIC","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"list : scan scan number intensity added intensity values","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawEIC-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get extracted ion chromatograms for specified m/z range — rawEIC-methods","text":"Ralf Tautenhahn","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get a raw data matrix — rawMat-methods","title":"Get a raw data matrix — rawMat-methods","text":"Returns matrix columns time, m/z, intensity   represents raw data chromatography mass spectrometry experiment.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get a raw data matrix — rawMat-methods","text":"object = \"xcmsRaw\" rawMat(object, mzrange = numeric(), rtrange = numeric(),       scanrange = numeric(), log=FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get a raw data matrix — rawMat-methods","text":"object container raw data mzrange Subset m/z range rtrange Subset retention time range scanrange Subset scan index range log Whether log transform intensities","code":""},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get a raw data matrix — rawMat-methods","text":"numeric matrix three columns: time, mz intensity.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/rawMat-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get a raw data matrix — rawMat-methods","text":"Michael Lawrence","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":null,"dir":"Reference","previous_headings":"","what":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"Reconstructs MS2 spectra MS1 chromatographic peak (possible) data independent acquisition (DIA) data (SWATH). See LC-MS/MS analysis vignette details examples.","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"","code":"reconstructChromPeakSpectra(object, ...)  # S4 method for XcmsExperiment reconstructChromPeakSpectra(   object,   expandRt = 0,   diffRt = 2,   minCor = 0.8,   intensity = \"maxo\",   peakId = rownames(chromPeaks(object, msLevel = 1L)),   BPPARAM = bpparam() )  # S4 method for XCMSnExp reconstructChromPeakSpectra(   object,   expandRt = 0,   diffRt = 2,   minCor = 0.8,   intensity = \"maxo\",   peakId = rownames(chromPeaks(object, msLevel = 1L)),   BPPARAM = bpparam(),   return.type = c(\"Spectra\", \"MSpectra\") )"},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"object XCMSnExp identified chromatographic peaks. ... ignored. expandRt numeric(1) allowing expand retention time range extracted ion chromatograms constant value (peak shape correlation). Defaults expandRt = 0 hence correlates signal included identified chromatographic peaks. diffRt numeric(1) defining maximal allowed difference retention time chromatographic peak (apex) retention times MS2 chromatographic peaks (apex) consider representing candidate fragments original ion. minCor numeric(1) defining minimal required correlation coefficient MS2 chromatographic peaks considered MS2 spectrum reconstruction. intensity character(1) defining column chromPeaks matrix used intensities reconstructed spectra's peaks. value MS1 chromatographic peaks used precursorIntensity resulting spectra. peakId optional character vector peak IDs (.e. rownames chromPeaks) MS1 peaks MS2 spectra reconstructed. default reconstructed MS1 chromatographic peaks. BPPARAM parallel processing setup. See bpparam() information. return.type character(1) defining type returned object. return.type = \"Spectra\" supported, return.type = \"MSpectra\" deprecated.","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"Spectra() object (defined Spectra package) reconstructed MS2 spectra MS1 peaks object. Contains empty spectra (.e. without m/z intensity values) MS1 peaks reconstruction possible (either MS2 signal recorded correlation MS2 chromatographic peaks MS1 chromatographic peak threshold minCor. Spectra variables \"ms2_peak_id\" \"ms2_peak_cor\" (type CharacterList() NumericList() length equal number peaks per reconstructed MS2 spectrum) providing IDs correlation MS2 chromatographic peaks MS2 spectrum reconstructed. retention time median retention times MS2 chromatographic peaks used spectrum reconstruction reported. MS1 chromatographic peak intensity reported reconstructed spectrum's precursorIntensity value (see parameter intensity ).","code":""},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"detail, function performs MS1 chromatographic peak: Identify MS2 chromatographic peaks isolation window containing m/z ion (.e. MS1 chromatographic peak) approximately retention time MS1 peak (accepted rt shift can specified diffRt parameter). Correlate peak shapes candidate MS2 chromatographic peaks peak shape MS1 peak retaining MS2 chromatographic peaks correlation > minCor. Reconstruct MS2 spectrum using m/z selected MS2 chromatographic peaks intensity (either \"maxo\" \"\"). MS2 chromatographic peak selected MS1 peak thus represent one mass peak reconstructed spectrum. resulting Spectra() object provides also peak IDs MS2 chromatographic peaks spectrum well correlation value spectra variables ms2_peak_id ms2_peak_cor.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/reconstructChromPeakSpectra.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Data independent acquisition (DIA): reconstruct MS2 spectra — reconstructChromPeakSpectra","text":"Johannes Rainer, Michael Witting","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"Given matrix rectangular areas, function determines subset   rectangles overlap. Rectangles preserved   first come, first served basis, user control order   rectangles processed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"","code":"rectUnique(m, order = seq(length = nrow(m)), xdiff = 0, ydiff = 0)"},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"m four column matrix defining rectangular areas order order matrix columns scanned xdiff maximum space overlapping rectangles x dimension ydiff maximum space overlapping rectangles y dimension","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"m matrix must contain four colums defining position   rectangle sides folloing order: left, right, bottom, top.   function currently implemented C using algorithm   quadratic running time.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"logical vector indicating rows kept.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/rectUnique.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine a subset of rectangles with unique, non-overlapping areas — rectUnique","text":"","code":"m <- rbind(c(0,4,0,3), c(1,3,2,6), c(3,6,4,6)) plot(0, 0, type = \"n\", xlim=range(m[,1:2]), ylim=range(m[,3:4])) rect(m[,1], m[,3], m[,2], m[,4])  xcms:::rectUnique(m) #> [1]  TRUE FALSE  TRUE # Changing order of processing xcms:::rectUnique(m, c(2,1,3)) #> [1] FALSE  TRUE FALSE # Requiring border spacing xcms:::rectUnique(m, ydiff = 1) #> [1]  TRUE FALSE FALSE # Allowing adjacent boxes xcms:::rectUnique(m, c(2,1,3), xdiff = -0.00001) #> [1] FALSE  TRUE  TRUE # Allowing interpenetration xcms:::rectUnique(m, xdiff = -1.00001, ydiff = -1.00001) #> [1] TRUE TRUE TRUE"},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Refine Identified Chromatographic Peaks — refineChromPeaks","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"refineChromPeaks method performs post-processing chromatographic peak detection step eventually clean improve results. function can applied XcmsExperiment() XCMSnExp() object peak detection findChromPeaks(). type peak refinement cleaning can defined, along settings, using one following parameter objects: CleanPeaksParam: remove chromatographic peaks retention time range larger provided maximal acceptable width (maxPeakwidth). FilterIntensityParam: remove chromatographic peaks intensities specified threshold. default (nValues = 1) values chromPeaks matrix evaluated: peaks value column defined parameter value >= threshold (defined parameter threshold) retained. nValues larger 1, individual peak intensities raw MS files evaluated: chromatographic peaks least nValues mass peaks >= threshold retained. MergeNeighboringPeaksParam: peak detection sometimes fails identify chromatographic peak correctly, especially broad peaks peak shape irregular (mostly HILIC data). cases several smaller peaks reported. Also, peak detection centWave can result partially completely overlapping peaks. method aims reduce peak detection artifacts merging chromatographic peaks overlapping close RT m/z dimension (considering also measured signal ). See section Details MergeNeighboringPeaksParam details comprehensive description approach. refineChromPeaks methods always remove feature definitions, call method can change remove identified chromatographic peaks, may part features.","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"","code":"refineChromPeaks(object, param, ...)  # S4 method for XcmsExperiment,CleanPeaksParam refineChromPeaks(object, param = CleanPeaksParam(), msLevel = 1L)  # S4 method for XcmsExperiment,MergeNeighboringPeaksParam refineChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  # S4 method for XcmsExperiment,FilterIntensityParam refineChromPeaks(   object,   param,   msLevel = 1L,   chunkSize = 2L,   BPPARAM = bpparam() )  CleanPeaksParam(maxPeakwidth = 10)  MergeNeighboringPeaksParam(   expandRt = 2,   expandMz = 0,   ppm = 10,   minProp = 0.75 )  FilterIntensityParam(threshold = 0, nValues = 1L, value = \"maxo\")  # S4 method for XCMSnExp,CleanPeaksParam refineChromPeaks(object, param = CleanPeaksParam(), msLevel = 1L)  # S4 method for XCMSnExp,MergeNeighboringPeaksParam refineChromPeaks(   object,   param = MergeNeighboringPeaksParam(),   msLevel = 1L,   BPPARAM = bpparam() )  # S4 method for XCMSnExp,FilterIntensityParam refineChromPeaks(   object,   param = FilterIntensityParam(),   msLevel = 1L,   BPPARAM = bpparam() )"},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"object XCMSnExp XcmsExperiment object identified chromatographic peaks. param Object defining refinement method settings. ... ignored. msLevel integer defining MS level(s) chromatographic peaks cleaned. chunkSize refineChromPeaks object either XcmsExperiment: integer(1) defining number files (samples) loaded memory processed time. Peak refinement performed parallel (per sample) subset data. setting thus allows balance memory demand speed (due parallel processing). parallel processing can performed subset data currently loaded memory iteration, value chunkSize match defined  parallel setting setup. Using parallel processing setup using 4 CPUs (separate processes) using chunkSize = 1will perform parallel processing, data one sample loaded memory time. hand, settingchunkSize` total number samples experiment load full MS data memory thus settings cause --memory error. BPPARAM parameter object set parallel processing. Uses default parallel processing setup returned bpparam(). See bpparam() details examples. maxPeakwidth CleanPeaksParam: numeric(1) defining maximal allowed peak width (retention time). expandRt MergeNeighboringPeaksParam: numeric(1) defining many seconds retention time window expanded sides check overlapping peaks. expandMz MergeNeighboringPeaksParam: numeric(1) constant value m/z range chromatographic peak expanded (sides!) check overlapping peaks. ppm MergeNeighboringPeaksParam: numeric(1) defining m/z relative value (parts per million) m/z range chromatographic peak expanded (side) check overlapping peaks. minProp MergeNeighboringPeaksParam: numeric(1) 0 1 representing proporion intensity required peaks joined. See description details. default (minProp = 0.75) peaks joined signal half way larger 75% smallest two peak's \"maxo\" (maximal intensity peak apex). threshold FilterIntensityParam: numeric(1) defining threshold peaks removed. nValues FilterIntensityParam: integer(1) defining number data points (chromatographic peak) >= threshold. Defaults nValues = 1. value FilterIntensityParam: character(1) defining name column chromPeaks contains values used filtering.","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"XCMSnExp XcmsExperiment object refined chomatographic peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"details-for-mergeneighboringpeaksparam","dir":"Reference","previous_headings":"","what":"Details for MergeNeighboringPeaksParam","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"peak refinement using MergeNeighboringPeaksParam, chromatographic peaks first expanded m/z retention time dimension (based parameters expandMz, ppm expandRt) subsequently grouped sets merge candidates (expansion) overlapping m/z rt (within sample). Note peak gets expanded expandRt expandMz, thus peaks differing less 2 * expandMz (2 * expandRt) evaluated merging. Peak merging performed along retention time axis, .e., peaks first ordered \"rtmin\" merge candidates defined iteratively starting first peak. Candidate peaks merged average intensity 3 data points middle position (.e., half distance \"rtmax\" first \"rtmin\" second peak) larger certain proportion (minProp) smaller (\"maxo\") intensity peaks. cases calculated mid point located apexes two peaks (e.g., peaks largely overlapping) average signal intensity half way apexes used instead. Candidate peaks merged 3 data points NA intensities. Merged peaks get \"mz\", \"rt\", \"sn\" \"maxo\" values peak largest signal (\"maxo\") well row metadata peak (chromPeakData). \"rtmin\" \"rtmax\" merged peaks updated \"\" recalculated based signal \"rtmin\" \"rtmax\" newly defined \"mzmin\" \"mzmax\" (range \"mzmin\" \"mzmax\" merged peaks expanding expandMz ppm). reported \"mzmin\" \"mzmax\" merged peak represents m/z range non-NA intensities used calculation peak signal (\"\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"Johannes Rainer, Mar Garcia-Aloy","code":""},{"path":"https://sneumann.github.io/xcms/reference/refineChromPeaks.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Refine Identified Chromatographic Peaks — refineChromPeaks","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Disable parallel processing for this example register(SerialParam())  #### ## CleanPeaksParam:  ## Distribution of chromatographic peak widths quantile(chromPeaks(faahko_sub)[, \"rtmax\"] - chromPeaks(faahko_sub)[, \"rtmin\"]) #>      0%     25%     50%     75%    100%  #>   6.259  29.734  43.819  57.903 173.710   ## Remove all chromatographic peaks with a width larger 60 seconds data <- refineChromPeaks(faahko_sub, param = CleanPeaksParam(60)) #> Removed 54 of 248 chromatographic peaks.  quantile(chromPeaks(data)[, \"rtmax\"] - chromPeaks(data)[, \"rtmin\"]) #>     0%    25%    50%    75%   100%  #>  6.259 23.475 39.906 46.949 59.469   #### ## FilterIntensityParam:  ## Remove all peaks with a maximal intensity below 50000 res <- refineChromPeaks(faahko_sub,     param = FilterIntensityParam(threshold = 50000)) #> Reduced from 248 to 155 chromatographic peaks.  nrow(chromPeaks(faahko_sub)) #> [1] 248 nrow(chromPeaks(res)) #> [1] 155  #### ## MergeNeighboringPeaksParam:  ## Subset to a single file xd <- filterFile(faahko_sub, file = 1)  ## Example of a split peak that will be merged mzr <- 305.1 + c(-0.01, 0.01) chr <- chromatogram(xd, mz = mzr, rt = c(2700, 3700)) #> Processing chromatographic peaks plot(chr)   ## Combine the peaks res <- refineChromPeaks(xd, param = MergeNeighboringPeaksParam(expandRt = 4)) #> Reduced from 87 to 70 chromatographic peaks. chr_res <- chromatogram(res, mz = mzr, rt = c(2700, 3700)) #> Processing chromatographic peaks plot(chr_res)   ## Example of a peak that was not merged, because the signal between them ## is lower than the cut-off minProp mzr <- 496.2 + c(-0.01, 0.01) chr <- chromatogram(xd, mz = mzr, rt = c(3200, 3500)) #> Processing chromatographic peaks plot(chr) chr_res <- chromatogram(res, mz = mzr, rt = c(3200, 3500)) #> Processing chromatographic peaks plot(chr_res)"},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"removeIntensities allows remove intensities chromatographic data matching certain conditions (depending parameter ). intensities actually removed replaced NA_real_. actually remove intensities (associated retention times) use clean() afterwards. Parameter allows specify intensities replaced NA_real_. default (= \"below_threshod\" intensities threshold removed. x XChromatogram XChromatograms object (hence provides also chromatographic peak definitions within object) = \"outside_chromPeak\" can selected removes intensity outside boundaries identified chromatographic peak(s) chromatographic data. Note filterIntensity() might better approach subset/filter chromatographic data.","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"","code":"# S4 method for Chromatogram removeIntensity(object, which = \"below_threshold\", threshold = 0)  # S4 method for MChromatograms removeIntensity(object, which = \"below_threshold\", threshold = 0)  # S4 method for XChromatogram removeIntensity(   object,   which = c(\"below_threshold\", \"outside_chromPeak\"),   threshold = 0 )"},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"object object representing chromatographic data. Can Chromatogram(), MChromatograms(), XChromatogram() XChromatograms() object. character(1) defining condition remove intensities. See description details options. threshold numeric(1) defining threshold intensities removed (= \"below_threshold\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"input object matching intensities replaced NA.","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/removeIntensity-Chromatogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Remove intensities from chromatographic data — removeIntensity,Chromatogram-method","text":"","code":"chr <- Chromatogram(rtime = 1:10 + rnorm(n = 10, sd = 0.3),     intensity = c(5, 29, 50, NA, 100, 12, 3, 4, 1, 3))  ## Remove all intensities below 20 res <- removeIntensity(chr, threshold = 20) intensity(res) #>  [1]  NA  29  50  NA 100  NA  NA  NA  NA  NA"},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Correct retention time from different samples — retcor-methods","title":"Correct retention time from different samples — retcor-methods","text":"correct differences retention times different   samples, number methods exist XCMS. retcor   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Correct retention time from different samples — retcor-methods","text":"object = \"xcmsSet\" retcor(object, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correct retention time from different samples — retcor-methods","text":"object xcmsSet-class object method Method use retention time correction. See details. ... Optional arguments passed along","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Correct retention time from different samples — retcor-methods","text":"Different algorithms can used specifying   method argument. example use approach described   Smith et al (2006) one use: retcor(object,   method=\"loess\"). also default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$retcor.methods.   nickname method called \"loess\",   help page specific method can   accessed ?retcor.loess.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correct retention time from different samples — retcor-methods","text":"xcmsSet object corrected retntion times.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Align retention times across samples with Obiwarp — retcor.obiwarp","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"Calculate retention time deviations sample.   based code http://obi-warp.sourceforge.net/.   However, function able align multiple samples,   center-star strategy. original publication see Chromatographic Alignment ESI-LC-MS Proteomics Data Sets       Ordered Bijective Interpolated Warping John T. Prince , Edward       M. Marcotte Analytical Chemistry 2006 78 (17), 6140-6152","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"object = \"xcmsSet\" retcor(object, method=\"obiwarp\", plottype = c(\"none\", \"deviation\"),              profStep=1, center=NULL, col = NULL, ty = NULL,              response=1, distFunc=\"cor_opt\", gapInit=NULL, gapExtend=NULL,             factorDiag=2, factorGap=1, localAlignment=0, initPenalty=0)","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"object xcmsSet object plottype deviation plot retention time deviation profStep step size (m/z) use profile generation     raw data files center index sample others aligned   . center==NULL, sample peaks chosen default. col vector colors plotting sample ty vector line point types plotting sample response Responsiveness warping.  0 give linear warp based     start end points.  100 use bijective anchors distFunc DistFunc function: cor (Pearson's R) cor_opt (default,     calculate 10% diagonal band distance matrix, better     runtime), cov (covariance), prd (product), euc (Euclidean distance) gapInit Penalty Gap opening, see gapExtend Penalty Gap enlargement, see factorDiag Local weighting applied diagonal moves alignment. factorGap Local weighting applied gap moves alignment. localAlignment Local rather global alignment initPenalty Penalty initiating alignment (local   alignment ) Default: 0 Default gap penalties: (gapInit, gapExtend) [distFunc type]: 'cor' = '0.3,2.4' 'cov' = '0,11.7' 'prd' = '0,7.8' 'euc' = '0.9,1.8'","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.obiwarp-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Align retention times across samples with Obiwarp — retcor.obiwarp","text":"xcmsSet object","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Align retention times across samples — retcor.peakgroups-methods","title":"Align retention times across samples — retcor.peakgroups-methods","text":"two methods use ``well behaved'' peak groups calculate   retention time deviations every time point sample. Use   smoothed deviations align retention times.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Align retention times across samples — retcor.peakgroups-methods","text":"object = \"xcmsSet\" retcor(object, missing = 1, extra = 1,                smooth = c(\"loess\", \"linear\"), span = .2,                family = c(\"gaussian\", \"symmetric\"),                plottype = c(\"none\", \"deviation\", \"mdevden\"),                col = NULL, ty = NULL)","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Align retention times across samples — retcor.peakgroups-methods","text":"object xcmsSet object missing number missing samples allow retention time correction     groups extra number extra peaks allow retention time correction     correction groups smooth either \"loess\" non-linear alignment \"linear\"     linear alignment span degree smoothing local polynomial regression fitting family gaussian fitting least-squares outlier     removal, symmetric re-descending M estimator     used Tukey's biweight function, allowing outlier removal plottype deviation plot retention time deviation points     regression fit, mdevden also plot peak overall     peak density retention time correction peak density col vector colors plotting sample ty vector line point types plotting sample","code":""},{"path":"https://sneumann.github.io/xcms/reference/retcor.peakgroups-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Align retention times across samples — retcor.peakgroups-methods","text":"xcmsSet object","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":null,"dir":"Reference","previous_headings":"","what":"Set retention time window to a specified width — retexp","title":"Set retention time window to a specified width — retexp","text":"Expands (contracts) retention time window row   matrix defined retmin retmax columns.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Set retention time window to a specified width — retexp","text":"","code":"retexp(peakrange, width = 200)"},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Set retention time window to a specified width — retexp","text":"peakrange maxtrix columns retmin retmax width new width window","code":""},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Set retention time window to a specified width — retexp","text":"altered matrix.","code":""},{"path":"https://sneumann.github.io/xcms/reference/retexp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Set retention time window to a specified width — retexp","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate relative log abundances — rla","title":"Calculate relative log abundances — rla","text":"rla calculates relative log abundances (RLA, see reference) numeric vector.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate relative log abundances — rla","text":"","code":"rla(x, group, log.transform = TRUE)  rowRla(x, group, log.transform = TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate relative log abundances — rla","text":"x numeric (rla) matrix (rowRla) abundances (natural scale) RLA calculated. group factor, numeric character length x groups values x. omitted values considered group. log.transform logical(1) whether x log2 transformed. Set log.transform = FALSE x already log scale.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate relative log abundances — rla","text":"numeric length x (rla) matrix dimensions x (rowRla).","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate relative log abundances — rla","text":"RLA defines (log) abundance analyte relative median across abundances group.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Calculate relative log abundances — rla","text":"De Livera , Dias DA, De Souza D, Rupasinghe T, Pyke J, Tull D, Roessner U, McConville M, Speed TP. Normalizing integrating metabolomics data. Anal Chem 2012 Dec 18;84(24):10768-76.","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Calculate relative log abundances — rla","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/rla.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate relative log abundances — rla","text":"","code":"x <- c(3, 4, 5, 1, 2, 3, 7, 8, 9)  grp <- c(1, 1, 1, 2, 2, 2, 3, 3, 3)  rla(x, grp) #>          1          1          1          2          2          2          3  #> -0.4150375  0.0000000  0.3219281 -1.0000000  0.0000000  0.5849625 -0.1926451  #>          3          3  #>  0.0000000  0.1699250"},{"path":"https://sneumann.github.io/xcms/reference/sampnames-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Get sample names — sampnames-methods","title":"Get sample names — sampnames-methods","text":"Return sample names object","code":""},{"path":"https://sneumann.github.io/xcms/reference/sampnames-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Get sample names — sampnames-methods","text":"object = \"xcmsEIC\" sampnames(object) object = \"xcmsSet\" sampnames(object)","code":""},{"path":"https://sneumann.github.io/xcms/reference/sampnames-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get sample names — sampnames-methods","text":"character vector sample names.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract processing errors — showError,xcmsSet-method","title":"Extract processing errors — showError,xcmsSet-method","text":"peak detection performed findPeaks     setting argument stopOnError = FALSE eventual errors     process cause stop processing recorded inside     resulting xcmsSet object. errors can     accessed showError method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract processing errors — showError,xcmsSet-method","text":"","code":"# S4 method for xcmsSet showError(object, message. = TRUE, ...)"},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract processing errors — showError,xcmsSet-method","text":"object xcmsSet object. message. Logical indicating whether error message, error returned. ... Additional arguments.","code":""},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract processing errors — showError,xcmsSet-method","text":"list error messages (message. = TRUE) errors     empty list errors present.","code":""},{"path":"https://sneumann.github.io/xcms/reference/showError-xcmsSet-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract processing errors — showError,xcmsSet-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"several methods calculating distance two sets peaks xcms. specDist   generic method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"object = \"xcmsSet\" specDist(object, peakIDs1, peakIDs2,...)  object = \"xsAnnotate\" specDist(object, PSpec1, PSpec2,...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"object xcmsSet xcmsRaw. method Method use distance calculation. See details. ... mzabs, mzppm parameters distance function.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"Different algorithms can used specifying   method argument. example use \"meanMZmatch\"   approach xcmsSet one use:   specDist(object, peakIDs1, peakIDs2, method=\"meanMZmatch\"). also   default. arguments given ...   passed function implementing   method. character vector nicknames   algorithms available returned   getOption(\"BioC\")$xcms$specDist.methods.   nickname method called \"meanMZmatch\",   help page specific method can   accessed ?specDist.meanMZmatch.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Distance methods for xcmsSet, xcmsRaw and xsAnnotate — specDist-methods","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"a Distance function based on matching peaks — specDist.cosine","title":"a Distance function based on matching peaks — specDist.cosine","text":"method calculates distance two sets peaks   using cosine-distance.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"a Distance function based on matching peaks — specDist.cosine","text":"","code":"specDist.cosine(peakTable1, peakTable2, mzabs=0.001, mzppm=10, mzExp=0.6,                 intExp=3, nPdiff=2, nPmin=8, symmetric=FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"a Distance function based on matching peaks — specDist.cosine","text":"peakTable1 = \"matrix\", peakTable2 = \"matrix\" specDist.cosine(peakTable1, peakTable2, mzabs = 0.001, mzppm = 10,       mzExp = 0.6, intExp = 3, nPdiff = 2, nPmin = 8,       symmetric = FALSE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"a Distance function based on matching peaks — specDist.cosine","text":"result cosine-distance product weighted factors mz intensity matching peaks two peaktables. factors calculated wFact = mz^mzExp * int^intExp. distance calculated (example matching peaks found) return-value NA.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"a Distance function based on matching peaks — specDist.cosine","text":"peakTable1 Matrix containing least m/z-values, \trow must called \"mz\" peakTable2 matrix mz-values mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match mzExp exponent used mz intExp exponent used intensity nPdiff maximum nrow-difference two peaktables nPmin minimum absolute sum peaks praktables","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.cosine-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"a Distance function based on matching peaks — specDist.cosine","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"a Distance function based on matching peaks — specDist.meanMZmatch","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"method calculates distance two sets peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"","code":"specDist.meanMZmatch(peakTable1, peakTable2, matchdist=1, matchrate=1,                      mzabs=0.001, mzppm=10, symmetric=TRUE)"},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"peakTable1 = \"matrix\", peakTable2 = \"matrix\" specDist.meanMZmatch(peakTable1, peakTable2,     matchdist=1, matchrate=1,     mzabs=0.001, mzppm=10, symmetric=TRUE)","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"result calculation weighted sum two values. Value one mean absolute difference matching peaks, value two relation matching peaks non matching peaks. distance calculated (example matching peaks found) return-value NA.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"peakTable1 Matrix containing least m/z-values, \trow must called \"mz\" peakTable2 matrix mz-values mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match matchdist weight value one (see details) matchrate weight value two","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.meanMZmatch-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"a Distance function based on matching peaks — specDist.meanMZmatch","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"a Distance function based on matching peaks — specDist.peakCount-methods","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"method calculates distance two sets peaks just returning number matching peaks (m/z-values).","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"","code":"specDist.peakCount(peakTable1, peakTable2, mzabs=0.001, mzppm=10, symmetric=FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"peakTable1 = \"matrix\", peakTable2 = \"matrix\" specDist.peakCount(peakTable1, peakTable2, mzppm=10,symmetric=FALSE )","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"peakTable1 Matrix containing least m/z-values, \trow must called \"mz\" peakTable2 matrix mz-values mzabs maximum absolute deviation two matching peaks mzppm relative deviations ppm two matching peaks symmetric use symmetric pairwise m/z-matches , match","code":""},{"path":"https://sneumann.github.io/xcms/reference/specDist.peakCount-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"a Distance function based on matching peaks — specDist.peakCount-methods","text":"Joachim Kutzera, jkutzer@ipb-halle.de","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate noise for a sparse continuum mass spectrum — specNoise","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"Given sparse continuum mass spectrum, determine regions   signal present, substituting half minimum intensity   regions. Calculate noise level weighted mean   regions signal regions without signal.   one raw peak, return zero.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"","code":"specNoise(spec, gap = quantile(diff(spec[, \"mz\"]), 0.9))"},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"spec matrix named columns mz intensity gap threshold data points considerd separated     blank region bridged interpolating line","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"default gap value determined 90th percentile   pair-wise differences adjacent mass values.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"numeric noise level","code":""},{"path":"https://sneumann.github.io/xcms/reference/specNoise.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Calculate noise for a sparse continuum mass spectrum — specNoise","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":null,"dir":"Reference","previous_headings":"","what":"Identify peaks in a sparse continuum mode spectrum — specPeaks","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"Given spectrum, identify list significant peaks determined   several criteria.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"","code":"specPeaks(spec, sn = 20, mzgap = 0.2)"},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"spec matrix named columns mz intensity sn minimum signal noise ratio mzgap minimal distance adjacent peaks, smaller peaks     excluded","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"Peaks must meet two criteria considered peaks: 1)   s/n ratio must exceed certain threshold. 2) must   within given distance greater intensity peaks.","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"matrix columns: mz m/z maximum peak intensity intensity maximum intensity peak fwhm full width half max peak","code":""},{"path":"https://sneumann.github.io/xcms/reference/specPeaks.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Identify peaks in a sparse continuum mode spectrum — specPeaks","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Divide an xcmsRaw object — split.xcmsRaw","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"Divides scans xcmsRaw object   list multiple objects. MS$^n$ data discarded.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"xr = \"xcmsRaw\" split(x, f, drop = TRUE, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"x xcmsRaw object f factor factor(f) defines scans go     new xcmsRaw objects drop logical indicating levels occur dropped     ('f' 'factor' list). ... potential arguments passed methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"list xcmsRaw objects.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Divide an xcmsRaw object — split.xcmsRaw","text":"Steffen Neumann, sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"Divide an xcmsSet object — split.xcmsSet","title":"Divide an xcmsSet object — split.xcmsSet","text":"Divides samples peaks xcmsSet object   list multiple objects. Group data discarded.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Divide an xcmsSet object — split.xcmsSet","text":"xs = \"xcmsSet\" split(x, f, drop = TRUE, ...)","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Divide an xcmsSet object — split.xcmsSet","text":"xs xcmsSet object f factor factor(f) defines grouping drop logical indicating levels occur dropped     ('f' 'factor' list). ... potential arguments passed methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Divide an xcmsSet object — split.xcmsSet","text":"list xcmsSet objects.","code":""},{"path":"https://sneumann.github.io/xcms/reference/split.xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Divide an xcmsSet object — split.xcmsSet","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Correct gaps in data — stitch-methods","title":"Correct gaps in data — stitch-methods","text":"Fixes gaps data due calibration scans lock mass. Automatically detects file type   calls relevant method. mzXML file keeps data length time   overwrites lock mass scans. netCDF version adds scans back data thereby   increasing length data correcting unseen gap.","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Correct gaps in data — stitch-methods","text":"object = \"xcmsRaw\" stitch(object, lockMass=numeric())  object = \"xcmsRaw\" makeacqNum(object, freq=numeric(), start=1)","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Correct gaps in data — stitch-methods","text":"object xcmsRaw-class object lockMass dataframe locations gaps freq intervals lock mass scans start starting lock mass scan location, default 1","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Correct gaps in data — stitch-methods","text":"makeacqNum takes locates gap using starting lock mass scan intervals. data frame used \tstitch correct gap caused lock mass. Correction works using scans either side gap fill .","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Correct gaps in data — stitch-methods","text":"stitch corrected xcmsRaw-class object  makeacqNum numeric vector scan locations corresponding lock Mass scans","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Correct gaps in data — stitch-methods","text":"Paul Benton, hpaul.benton08@imperial.ac.uk","code":""},{"path":"https://sneumann.github.io/xcms/reference/stitch-methods.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Correct gaps in data — stitch-methods","text":"","code":"if (FALSE) library(xcms)     library(faahKO)     ## These files do not have this problem to correct for but just     ## for an example     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xr<-xcmsRaw(cdffiles[1]) #> Create profile matrix with method 'bin' and step 1 ...  #> OK     xr #> An \"xcmsRaw\" object with 1278 mass spectra #>  #> Time range: 2501.4-4499.8 seconds (41.7-75 minutes) #> Mass range: 200-600 m/z #> Intensity range: 70-1373180  #>  #> MSn data on  0  mass(es) #> \twith  0  MSn spectra #> Profile method: bin  #> Profile step: 1 m/z (401 grid points from 200 to 600 m/z) #>  #> Memory usage: 10.8 MB     ##Lets assume that the lockmass starts at 1 and is every 100 scans     lockMass<-xcms:::makeacqNum(xr, freq=100, start=1)     ## these are equcal     lockmass<-AutoLockMass(xr) #> Warning:  #> Lock mass frequency wasn't detected     ob<-stitch(xr, lockMass)     ob #> An \"xcmsRaw\" object with 1304 mass spectra #>  #> Time range: 1.6-2040.7 seconds (0-34 minutes) #> Mass range: 200-600 m/z #> Intensity range: 70-1373180  #>  #> MSn data on  0  mass(es) #> \twith  0  MSn spectra #> Profile method: bin  #> Profile step: no profile data #>  #> Memory usage: 6.96 MB      ## plot the old data before correction     foo<-rawEIC(xr, m=c(200,210), scan=c(80,140))     plot(foo$scan, foo$intensity, type=\"h\")       ## plot the new corrected data to see what changed     foo<-rawEIC(ob, m=c(200,210), scan=c(80,140))     plot(foo$scan, foo$intensity, type=\"h\")"},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":null,"dir":"Reference","previous_headings":"","what":"Save xcms result objects in a specified format — storeResults","title":"Save xcms result objects in a specified format — storeResults","text":"storeResults function saves object resulting processing xcms package (mainly XcmsExperiment). Multiple formats storing exporting available can defined param argument. Supported param objects : RDataParam: Save .RData format file. name file can specified fileName argument. PlainTextParam: Store MsExperiment XcmsExperiment objects folder plain text files, folder path defined path argument. MzTabMParam: Save MzTab format (defined). specific examples, see help pages individual parameter classes listed .","code":""},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Save xcms result objects in a specified format — storeResults","text":"","code":"storeResults(object, param, ...)"},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save xcms result objects in a specified format — storeResults","text":"object MsExperiment XcmsExperiment data object needs saved. param parameter object selecting configuring format saving. can one following classes: RDataParam, PlainTextParam, MzTabMParam. ... Optional parameters.","code":""},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Save xcms result objects in a specified format — storeResults","text":"Philippine Louail","code":""},{"path":"https://sneumann.github.io/xcms/reference/storeResults.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Save xcms result objects in a specified format — storeResults","text":"","code":"## Load a test data set with detected peaks faahko_sub <- loadXcmsData(\"faahko_sub2\")  ## Set up parameter to save as .RData file param <- RDataParam(fileName = \"example_xcms_results\")  ## save as .RData storeResults(object = faahko_sub, param = param)  ## Set up parameter to save as a collection of plain text file param <- PlainTextParam(path = \"test/path/\")  ## Save as a collection of plain text files storeResults(object = faahko_sub, param = param)"},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"Subset xcmsRaw object scans.     returned xcmsRaw object contains values     scans specified argument . Note scanrange     slot returned xcmsRaw     c(1, length(object@scantime)) hence range().","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"","code":"# S4 method for xcmsRaw,logicalOrNumeric,missing,missing [(x, i, j, drop)"},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"x xcmsRaw object sub-setted. Integer logical vector specifying scans/spectra x sub-setted. j supported. drop supported.","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"sub-setted xcmsRaw object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"subsetting scan index increasing order logical     vector supported. ordered, argument sorted     automatically. Indices larger total number scans     discarded.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/sub-xcmsRaw-logicalOrNumeric-missing-missing-method.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Subset an xcmsRaw object by scans — [,xcmsRaw,logicalOrNumeric,missing,missing-method","text":"","code":"## Load a test file file <- system.file('cdf/KO/ko15.CDF', package = \"faahKO\") xraw <- xcmsRaw(file, profstep = 0) ## The number of scans/spectra: length(xraw@scantime) #> [1] 1278  ## Subset the object to scans with a scan time from 3500 to 4000. xsub <- xraw[xraw@scantime >= 3500 & xraw@scantime <= 4000] range(xsub@scantime) #> [1] 3501.383 3999.039 ## The number of scans: length(xsub@scantime) #> [1] 319 ## The number of values of the subset: length(xsub@env$mz) #> [1] 130569"},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Update an xcmsSet object — updateObject,xcmsSet-method","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"method updates old xcmsSet     object latest definition.","code":""},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"","code":"# S4 method for xcmsSet updateObject(object, ..., verbose = FALSE)"},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"object xcmsSet object update. ... Optional additional arguments. Currently ignored. verbose Currently ignored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"updated xcmsSet containing data input object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/updateObject-xcmsSet-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Update an xcmsSet object — updateObject,xcmsSet-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":null,"dir":"Reference","previous_headings":"","what":"Enable usage of old xcms code — useOriginalCode","title":"Enable usage of old xcms code — useOriginalCode","text":"function allows enable usage old, partially deprecated code xcms setting corresponding global option. See details functions affected.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Enable usage of old xcms code — useOriginalCode","text":"","code":"useOriginalCode(x)"},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Enable usage of old xcms code — useOriginalCode","text":"x logical(1) specify whether original old code used corresponding functions. provided function simply returns value global option.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Enable usage of old xcms code — useOriginalCode","text":"logical(1) indicating whether old code used.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Enable usage of old xcms code — useOriginalCode","text":"functions/methods affected option : do_findChromPeaks_matchedFilter: use original code iteratively creates subset binned (profile) matrix. helpful computers limited memory matchedFilter settings small bin size. getPeaks","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Enable usage of old xcms code — useOriginalCode","text":"parallel processing using SOCKS method (e.g. SnowParam() Windows computers) option might passed individual R processes performing calculations. cases suggested specify option manually system-wide adding line options(XCMSuseOriginalCode = TRUE) file called .Rprofile folder new R processes started (usually user's home directory; ensure option correctly read add new line file ). See also Startup base R documentation specify system-wide options R. Usage old code strongly dicouraged. function thought used mainly transition phase xcms xcms version 3.","code":""},{"path":"https://sneumann.github.io/xcms/reference/useOriginalCode.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Enable usage of old xcms code — useOriginalCode","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":null,"dir":"Reference","previous_headings":"","what":"Verify an mzQuantML file — verify.mzQuantM","title":"Verify an mzQuantML file — verify.mzQuantM","text":"Export XML data formats: verify written data","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Verify an mzQuantML file — verify.mzQuantM","text":"","code":"verify.mzQuantML(filename, xsdfilename)"},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Verify an mzQuantML file — verify.mzQuantM","text":"filename filename (may include full path)     output file. Pipes URLs allowed. xsdfilename Filename XSD verify (may include full path)","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Verify an mzQuantML file — verify.mzQuantM","text":"verify.mzQuantML() function verify PSI standard format mzQuantML   document XSD schemda, see http://www.psidev.info/mzquantml","code":""},{"path":"https://sneumann.github.io/xcms/reference/verify.mzQuantML.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Verify an mzQuantML file — verify.mzQuantM","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Save an xcmsRaw object to file — write.cdf-methods","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"Write raw data (simple) CDF file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"object = \"xcmsRaw\" write.cdf(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"object xcmsRaw object filename filename (may include full path)     CDF file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"Currently application known read resulting file   XCMS. Others, especially build AndiMS library,   refuse load output.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.cdf-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save an xcmsRaw object to file — write.cdf-methods","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":null,"dir":"Reference","previous_headings":"","what":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"Export XML data formats:    Write processed data xcmsSet mzQuantML.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"object = \"xcmsSet\" write.mzQuantML(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"object xcmsRaw xcmsSet object filename filename (may include full path)     output file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"write.mzQuantML() function write (grouped) xcmsSet   PSI standard format mzQuantML, see http://www.psidev.info/mzquantml","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzQuantML.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save an xcmsSet object to an PSI mzQuantML file — write.mzQuantML-methods","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Save an xcmsRaw object to a file — write.mzdata-methods","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"Write raw data (simple) mzData file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"object = \"xcmsRaw\" write.mzdata(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"object xcmsRaw object filename filename (may include full path)     mzData file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"function export given xcmsRaw object mzData file.   mzData file contain <spectrumList> containing   <spectrum> mass intensity values 32 bit precision.   formats currently supported. header information   (e.g. additional <software> information <cvParams>) lost.   Currently, also MSn information stored.","code":""},{"path":"https://sneumann.github.io/xcms/reference/write.mzdata-methods.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save an xcmsRaw object to a file — write.mzdata-methods","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":null,"dir":"Reference","previous_headings":"","what":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"writeMSData exports mass spectrometry data mzML mzXML format. adjusted retention times present, used retention time exported spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"","code":"# S4 method for XCMSnExp,character writeMSData(   object,   file,   outformat = c(\"mzml\", \"mzxml\"),   copy = FALSE,   software_processing = NULL,   ... )"},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"object XCMSnExp object mass spectrometry data. file character file name(s). length parameter match number files/samples object. outformat character(1) defining format output files ( either \"mzml\" \"mzxml\"). copy logical(1) metadata (data processing, software used, original file names etc) copied original files. software_processing optionally provide specific data processing steps. See documentation software_processing parameter mzR::writeMSData(). ... Additional parameters pass writeMSData() function MSnbase package, outformat specify output format (\"mzml\" \"mzxml\") copy specify whether general information original MS data files (data processing, software etc) copied new files.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/writeMSData-XCMSnExp-character-method.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Export MS data to mzML/mzXML files — writeMSData,XCMSnExp,character-method","text":"Johannes Rainer","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":null,"dir":"Reference","previous_headings":"","what":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"Write grouped xcmsSet mzTab file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"object = \"xcmsSet\" writeMzTab(object, filename)","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"object xcmsSet object filename filename (may include full path)     mzTab file. Pipes URLs allowed.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"mzTab file format MS-based metabolomics (proteomics)    lightweight    supplement existing standard XML-based file formats (mzML,    mzIdentML, mzQuantML), providing comprehensive summary, similar    concept supplemental material scientific    publication. mzTab files xcms contain small    molecule sections together experimental metadata    basic quantitative information. format intended store    simple summary final results.","code":""},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"None.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/writeMzTab.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Save a grouped xcmsSet object in mzTab-1.1 format file — writeMzTab","text":"","code":"library(faahKO)     xs <- group(faahko)      mzt <- data.frame(character(0))     mzt <- xcms:::mzTabHeader(mzt,                        version=\"1.1.0\", mode=\"Complete\", type=\"Quantification\",                        description=\"faahKO\",                        xset=xs)     mzt <- xcms:::mzTabAddSME(mzt, xs)          xcms:::writeMzTab(mzt, \"faahKO.mzTab\")"},{"path":"https://sneumann.github.io/xcms/reference/xcms-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated functions in package ‘xcms’ — xcms-deprecated","title":"Deprecated functions in package ‘xcms’ — xcms-deprecated","text":"functions provided compatibility older versions   ‘xcms’ , defunct next release.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcms-deprecated.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Deprecated functions in package ‘xcms’ — xcms-deprecated","text":"following functions/methods deprecated.  profBin, profBinM, profBinLin,       profBinLinM, profBinLinBase, profBinLinBaseM       deprecated binYonX combination     imputeLinInterpol used instead. extractMsData: replaced (x, \"data.frame\"). plotMsData: replaced plot(x, type = \"XIC\").","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"class used store plot parallel extracted ion   chromatograms multiple sample files. integrates   xcmsSet class display peak area integrated peak   identification fill-.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"Objects can created getEIC method   xcmsSet class. Objects can also created calls   form new(\"xcmsEIC\", ...).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"eic: list containing named entries every sample. entry,       list two column EIC matricies retention time intensity mzrange: two column matrix containing starting ending m/z EIC rtrange: two column matrix containing starting ending time       EIC rt: either \"raw\" \"corrected\" specify retention       times contained object groupnames: group names xcmsSet object used generate EICs","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"groupnames signature(object = \"xcmsEIC\"): get groupnames slot mzrange signature(object = \"xcmsEIC\"): get mzrange slot plot signature(x = \"xcmsEIC\"): plot extracted ion       chromatograms rtrange signature(object = \"xcmsEIC\"): get rtrange slot sampnames signature(object = \"xcmsEIC\"): get sample names","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsEIC-class.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Class xcmsEIC, a class for multi-sample extracted ion chromatograms — xcmsEIC-class","text":"notes yet.","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Base class for loading raw data from a file — xcmsFileSource-class","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"Data sources read data file inherit   class. xcms package provides classes read   netCDF, mzData, mzXML, mzML files   using xcmsFileSource. class considered virtual work passed   loadRaw-methods. reason explicitly   virtual appear way class   virtual data part (lets functions treat objects   character strings). class validates file exists path given.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"xcmsFileSource objects instantiated directly.   Instead, create subclasses instantiate .","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":".Data: Object class \"character\". File path       file read raw data object's data part","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"extends","dir":"Reference","previous_headings":"","what":"Extends","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"Class \"character\", data part.   Class \"xcmsSource\", directly.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"xcmsSource signature(object = \"character\"): Create       xcmsFileSource object referencing given file name.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFileSource-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Base class for loading raw data from a file — xcmsFileSource-class","text":"Daniel Hackney dan@haxney.org","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"class similar xcmsSet stores peaks   number individual files. However,  xcmsFragments keeps   Tandem MS e.g. Ion Trap Orbitrap MS$^n$ peaks, including   parent ion relationships.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"Objects can created xcmsFragments   constructor filled peaks using collect method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"peaks: matrix colmns peakID (MS1 parent corresponding xcmsSet),       MSnParentPeakID (parent peak within xcmsFragments), msLevel       (e.g. 2 Tandem MS), rt (retention time case LC data), mz       (fragment mass--charge), intensity (peak intensity extracted       original xcmsSet), sample (index rawData-file). MS2spec: list matrixes. matrix list single collected spectra collect. column ID's mz, intensity, full width half maximum(fwhm). fwhm column relevant spectra came profile data. specinfo: matrix reference data spectra MS2spec. column id's preMZ, AccMZ, rtmin, rtmax, ref, CollisionEnergy. preMZ precursor mass MS1 scan. mass given XML file. instruments mass given nominal mass, therefore AccMZ given weighted average mass MS1 scan collected spectra. retention time given rtmin rtmax. ref column pointer MS2spec matrix spectra. collisionEnergy column collision Energy spectra.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"collect signature(object = \"xcmsFragments\"): gets xcmsSet-object, collects ms1-peaks msn-peaks corresponding xcmsRaw-files. plotTree signature(object = \"xcmsFragments\"): prints (text based) pseudo-tree peaktable display dependencies peaks among . show signature(object = \"xcmsFragments\"): print human-readable       description object console.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsFragments, a class for handling Tandem MS and MS$^n$ data — xcmsFragments-class","text":"S. Neumann, J. Kutzera","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":null,"dir":"Reference","previous_headings":"","what":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"EXPERIMANTAL FEATURE xcmsFragments object similar xcmsSet, holds peaks   picked (collected) one several xcmsRaw objects. still discussions going exact API MS$^n$   data, likely change future. code yet   pipeline-ified.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"","code":"xcmsFragments(xs, ...)"},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"xs xcmsSet-class object contains     picked ms1-peaks one several experiments ... arguments collect method","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"running collect(xFragments,xSet) peaktable   xcmsFragments includes ms1Peaks experinemts stored   xcmsSet-object. contains relevant MSn-peaks   xcmsRaw-objects, created temporarily paths   xcmsSet.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"xcmsFragments object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsFragments.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Constructor for xcmsFragments objects which holds Tandem MS peaks — xcmsFragments","text":"Joachim Kutzera, Steffen Neumann, sneumann@ipb-halle.de","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":null,"dir":"Reference","previous_headings":"","what":"A matrix of peaks — xcmsPeaks-class","title":"A matrix of peaks — xcmsPeaks-class","text":"matrix peak information. actual columns depend   generated (.e. findPeaks method).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"A matrix of peaks — xcmsPeaks-class","text":"Objects can created calls form new(\"xcmsPeaks\", ...).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"A matrix of peaks — xcmsPeaks-class","text":".Data: matrix holding peak information","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"extends","dir":"Reference","previous_headings":"","what":"Extends","title":"A matrix of peaks — xcmsPeaks-class","text":"Class \"matrix\", data part. Class \"array\", class \"matrix\", distance 2. Class \"structure\", class \"matrix\", distance 3. Class \"vector\", class \"matrix\", distance 4, explicit coerce.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"A matrix of peaks — xcmsPeaks-class","text":"None yet. utilities working peak data nice.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsPeaks-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"A matrix of peaks — xcmsPeaks-class","text":"Michael Lawrence","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"class handles processing visualization raw data   single LC/MS GS/MS run. includes methods producing   standard suite plots including individual spectra, multi-scan   average spectra, TIC, EIC. also produce feature   list significant peaks using matched filtration.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"Objects can created xcmsRaw constructor   reads data NetCDF file new object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"acquisitionNum: Numeric representing acquisition       number individual scans/spectra. Length       acquisitionNum equal number spectra/scans       object hence equal scantime slot. Note however       information available mzML files. env: environment three variables: mz - concatenated       m/z values scans, intensity - corresponding       signal intensity m/z value, profile -       matrix represention intensity values columns       representing scans rows representing equally spaced m/z       values. profile matrix extracted       profMat method. filepath: Path raw data file gradient: matrix first row, time, containing time point       interpolation successive columns representing solvent       fractions point msnAcquisitionNum: scan unique acquisition       number reported via \"spectrum id\" (mzData) \"<scan num=...>\"       \"<scanOrigin num=...>\" (mzXML) msnCollisionEnergy: \"CollisionEnergy\" (mzData)       \"collisionEnergy\" (mzXML) msnLevel: scan \"msLevel\" (mzData mzXML) msnPrecursorCharge: \"ChargeState\" (mzData)       \"precursorCharge\" (mzXML) msnPrecursorIntensity: \"Intensity\" (mzData)       \"precursorIntensity\" (mzXML) msnPrecursorMz: \"MassToChargeRatio\" (mzData)       \"precursorMz\" (mzXML) msnPrecursorScan: \"spectrumRef\" (mzData mzXML) msnRt: Retention time scan msnScanindex: msnScanindex mzrange: numeric vector length 2 minimum maximum m/z values       represented profile matrix polarity: polarity profmethod: characer value name method used generating       profile matrix. profparam: list store additional profile matrix       generation settings. Use profinfo method       extract profile matrix creation relevant information. scanindex: integer vector starting positions scan       mz intensity variables (note index       values based 0 initial position instead 1). scantime: numeric vector acquisition time (seconds)       scan. tic: numeric vector total ion count (intensity) scan mslevel: Numeric representing MS level present MS1       slot. slot accessed getter method       mslevel. scanrange: Numeric length 2 specifying scan range (NULL       full range). slot accessed getter       method scanrange. Note scanrange       always 1 number scans within xcmsRaw       object, necessarily match scan index       original mzML file (e.g. original data sub-setted).       acquisitionNum information can used track       original position scan mzML file.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"findPeaks signature(object = \"xcmsRaw\"): feature detection using       matched filtration chromatographic time domain getEIC signature(object = \"xcmsRaw\"): get extracted ion       chromatograms specified m/z ranges. return total   ion chromatogram (TIC) m/z range corresponds full m/z   range (.e. sum signals per retention time across m/z). getPeaks signature(object = \"xcmsRaw\"): get data peaks       specified m/z time ranges getScan signature(object = \"xcmsRaw\"): get m/z intensity       values single mass scan getSpec signature(object = \"xcmsRaw\"): get average m/z       intensity values multiple mass scans image signature(x = \"xcmsRaw\"): get data peaks       specified m/z time ranges levelplot Create image raw (profile) data m/z retention   time, intensity color coded. mslevel Getter method mslevel slot. plotChrom signature(object = \"xcmsRaw\"): plot chromatogram       profile data plotRaw signature(object = \"xcmsRaw\"): plot locations raw       intensity data points plotScan signature(object = \"xcmsRaw\"): plot mass spectrum       individual scan raw data plotSpec signature(object = \"xcmsRaw\"): plot mass spectrum       profile data plotSurf signature(object = \"xcmsRaw\"): experimental method       plotting 3D surface profile data rgl. plotTIC signature(object = \"xcmsRaw\"): plot total ion count       chromatogram profinfo signature(object = \"xcmsRaw\"): returns list containing   profile generation method step (profile m/z step size)   eventual additional parameters profile function. profMedFilt signature(object = \"xcmsRaw\"): median filter profile       data time m/z dimensions profMethod<- signature(object = \"xcmsRaw\"): change method       generating profile matrix profMethod signature(object = \"xcmsRaw\"): get method       generating profile matrix profMz signature(object = \"xcmsRaw\"): get vector m/z values       row profile matrix profRange signature(object = \"xcmsRaw\"): interpret flexible ways       specifying subsets profile matrix profStep<- signature(object = \"xcmsRaw\"): change m/z step       used generating profile matrix profStep signature(object = \"xcmsRaw\"): get m/z step used       generating profile matrix revMz signature(object = \"xcmsRaw\"): reverse order       data points scan scanrange Getter method scanrange slot. See slot description       information. sortMz signature(object = \"xcmsRaw\"): sort data points       increasing m/z scan stitch signature(object = \"xcmsRaw\"): Raw data correction       lock mass calibration gaps. findmzROI signature(object = \"xcmsRaw\"): \tinternal function identify regions interest raw \tdata part first step centWave-based peak detection.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsRaw, a class for handling raw data — xcmsRaw-class","text":"Colin . Smith, csmith@scripps.edu,   Johannes Rainer johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":null,"dir":"Reference","previous_headings":"","what":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"function handles task reading NetCDF/mzXML file containing   LC/MS GC/MS data new xcmsRaw object. also   transforms data profile (maxrix) mode efficient   plotting data exploration.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"","code":"xcmsRaw(filename, profstep = 1, profmethod = \"bin\", profparam = list(), includeMSn=FALSE, mslevel=NULL, scanrange=NULL)  deepCopy(object)"},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"filename path name NetCDF mzXML file read profstep step size (m/z) use profile generation profmethod method use profile generation. See     profile-matrix details supported values. profparam extra parameters use profile generation includeMSn XML file formats: also read MS$^n$     (Tandem-MS Ion-/Orbi- Trap spectra) mslevel move data mslevel normal MS1 slots, e.g.     peak picking visualisation scanrange scan range read object xcmsRaw object","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"See profile-matrix details profile matrix   generation methods settings. scanrange import can restricted, otherwise MS1 data   read. profstep set 0, profile matrix generated.   Unless includeMSn = TRUE first level MS data read, MS/MS,   etc. deepCopy(xraw) create copy xcmsRaw object   copy mz intensity data xraw@env.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"xcmsRaw object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"NetCDF file format:   https://www.unidata.ucar.edu/software/netcdf/   http://www.astm.org/Standards/E2077.htm   http://www.astm.org/Standards/E2078.htm mzXML file format:   http://sashimi.sourceforge.net/software_glossolalia.html PSI-MS working group developed mzData mzML file formats:   http://www.psidev.info/index.php?q=node/80 Parser used XML file formats:   http://tools.proteomecenter.org/wiki/index.php?title=Software:RAMP","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsRaw.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Constructor for xcmsRaw objects which reads NetCDF/mzXML files — xcmsRaw","text":"","code":"if (FALSE) {     library(xcms)     library(faahKO)     cdfpath <- system.file(\"cdf\", package = \"faahKO\")     cdffiles <- list.files(cdfpath, recursive = TRUE, full.names = TRUE)     xr<-xcmsRaw(cdffiles[1])     xr     ##This gives some information about the file     names(attributes(xr))     ## Lets have a look at the structure of the object      str(xr)     ##same but with a preview of each slot in the object     ##SO... lets have a look at how this works     head(xr@scanindex)     ##[1]    0  429  860 1291 1718 2140     xr@env$mz[425:430]     ##[1] 596.3 597.0 597.3 598.1 599.3 200.1     ##We can see that the 429 index is the last mz of scan 1 therefore...      mz.scan1<-xr@env$mz[(1+xr@scanindex[1]):xr@scanindex[2]]     intensity.scan1<-xr@env$intensity[(1+xr@scanindex[1]):xr@scanindex[2]]     plot(mz.scan1, intensity.scan1, type=\"h\",          main=paste(\"Scan 1 of file\", basename(cdffiles[1]), sep=\"\"))     ##the easier way :p     scan1<-getScan(xr, 1)     head(scan1)     plotScan(xr, 1)   }"},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"class transforms set peaks multiple LC/MS GC/MS   samples matrix preprocessed data. groups peaks   nonlinear retention time correction without internal   standards. fills missing peak values raw data. Lastly,   generates extracted ion chromatograms ions interest.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"Objects can created xcmsSet constructor   gathers peaks set NetCDF files. Objects can also   created calls form new(\"xcmsSet\", ...).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"slots","dir":"Reference","previous_headings":"","what":"Slots","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"peaks matrix containing peak data. filled vector peak indices peaks added       fillPeaks method. groups Matrix containing statistics peak groups. groupidx List containing indices peaks group. phenoData data.frame containing experimental design factors. rt list containing two lists, raw corrected,       containing retention times every scan every sample. filepaths Character vector absolute path name NetCDF file. profinfo list containing values method - profile generation       method, step - profile m/z step size eventual       additional parameters profile function. dataCorrection logical vector filled waters Lock mass correction       parameter used. polarity string (\"positive\" \"negative\" NULL) describing whether       positive negative scans used reading raw data. progressInfo Progress informations xcms functions (GUI). progressCallback Function called, progressInfo changes (GUI). mslevel Numeric representing MS level peak picking       performed (default MS1). slot accessed       getter method mslevel. scanrange Numeric length 2 specifying scan range (NULL       full range). slot accessed getter       method scanrange. scan range provided slot       represents scans whole raw data subsetted. .processHistory Internal slot used keep track performed processing       steps. slot directly accessed user.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"c signature(\"xcmsSet\"): combine objects together filepaths<- signature(object = \"xcmsSet\"): set filepaths slot filepaths signature(object = \"xcmsSet\"): get filepaths slot diffreport signature(object = \"xcmsSet\"): create report       differentially regulated ions including EICs fillPeaks signature(object = \"xcmsSet\"): fill peak data       groups missing peaks getEIC signature(object = \"xcmsSet\"): get list EICs       sample set getXcmsRaw signature(object = \"xcmsSet\", sampleidx = 1, \tprofmethod = profMethod(object), profstep = profStep(object), \tprofparam=profinfo(object), mslevel = NULL, scanrange = NULL, \trt=c(\"corrected\", \"raw\"), BPPARAM = bpparam()): read raw       data one files xcmsSet return       . default parameters apply settings used       original xcmsSet call generate xcmsSet       object applied also raw data. Parameter       sampleidx allows specify raw file(s)       loaded. Argument BPPARAM allows setup parallel       processing. groupidx<- signature(object = \"xcmsSet\"): set groupidx slot groupidx signature(object = \"xcmsSet\"): get groupidx slot groupnames signature(object = \"xcmsSet\"): get textual names       peak groups groups<- signature(object = \"xcmsSet\"): set groups slot groups signature(object = \"xcmsSet\"): get groups slot groupval signature(object = \"xcmsSet\"): get matrix values       peak data row peak group group signature(object = \"xcmsSet\"): find groups peaks       across samples share similar m/z retention times mslevel Getter method mslevel slot. peaks<- signature(object = \"xcmsSet\"): set peaks slot peaks signature(object = \"xcmsSet\"): get peaks slot plotrt signature(object = \"xcmsSet\"): plot retention time       deviation profiles profinfo<- signature(object = \"xcmsSet\"): set profinfo slot profinfo signature(object = \"xcmsSet\"): get profinfo slot profMethod signature(object = \"xcmsSet\"): extract method used       generate profile matrix. profStep signature(object = \"xcmsSet\"): extract profile step       used generation profile matrix. retcor signature(object = \"xcmsSet\"): use initial grouping       peaks nonlinear loess retention time correction sampclass<- signature(object = \"xcmsSet\"): Replaces column       “class” phenoData slot. See details information. sampclass signature(object = \"xcmsSet\"): Returns content       column “class” phenoData slot ,       present, interaction experimental design factors       (.e. phenoData data.frame). See details       information. phenoData<- signature(object = \"xcmsSet\"): set phenoData slot phenoData signature(object = \"xcmsSet\"): get phenoData slot progressCallback<- signature(object = \"xcmsSet\"): set progressCallback slot progressCallback signature(object = \"xcmsSet\"): get progressCallback slot scanrange Getter method scanrange slot. See scanrange slot       description details. sampnames<- signature(object = \"xcmsSet\"): set rownames       phenoData slot sampnames signature(object = \"xcmsSet\"): get rownames       phenoData slot split signature(\"xcmsSet\"): divide xcmsSet list       xcmsSet objects depending provided factor. Note       peak data preserved, .e. eventual peak grouping information       lost. object$name, object$name<-value Access set name column phenoData object[, ] Conducts subsetting xcmsSet instance. subsetting       columns, .e. samples, supported. Subsetting performed       slots, also groups groupidx. Parameter       can integer vector, logical vector character       vector sample names (matching sampnames).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"phenoData slot (phenoData parameter   xcmsSet function) intended contain data.frame describing   experimental factors, .e. samples along   properties. data.frame contains column named   “class”, returned sampclass method   thus used methods determine sample   grouping/class assignment (e.g. define colors various plots   group method). sampclass<- method adds replaces “class”   column phenoData slot. data.frame   submitted method, interaction columns   stored “class” column. Also, similar classes Bioconductor, $ method   can used directly access columns phenoData   slot (e.g. use xset$name xcmsSet object called   “xset” extract values column named “name” phenoData slot).","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Class xcmsSet, a class for preprocessing peak data — xcmsSet-class","text":"Colin . Smith, csmith@scripps.edu, Johannes Rainer johannes.rainer@eurac.edu","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":null,"dir":"Reference","previous_headings":"","what":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"function handles construction xcmsSet objects.   finds peaks batch mode pre-sorts files subdirectories   different classes suitable grouping.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"","code":"xcmsSet(files = NULL, snames = NULL, sclass = NULL, phenoData = NULL,         profmethod = \"bin\", profparam = list(),         polarity = NULL, lockMassFreq=FALSE,   mslevel=NULL, nSlaves=0, progressCallback=NULL,         scanrange = NULL, BPPARAM = bpparam(),         stopOnError = TRUE, ...)"},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"files path names NetCDF/mzXML files read snames sample names. default file name without extension     used. sclass sample classes. phenoData data.frame AnnotatedDataFrame     defining sample names classes sample related     properties. provided, argument sclass     subdirectories samples stored     used specify sample grouping. profmethod Method use profile generation. Supported     values \"bin\", \"binlin\", \"binlinbase\"     \"intlin\" (methods profBin,     profBinLin, profBinLinBase     profIntLin, respectively). See help     profBin complete list available methods     supported parameters. profparam parameters use profile generation. polarity filter raw data positive/negative scans lockMassFreq Performs correction Waters LockMass function mslevel perform peak picking data given mslevel nSlaves DEPRECATED, use BPPARAM argument       instead. progressCallback function called, progressInfo changes (useful GUIs) scanrange scan range read BPPARAM BiocParallel parameter object control     parallel processing performed. objects can     created SerialParam,     MulticoreParam     SnowParam functions. stopOnError Logical specifying whether feature detection     call stop first encountered error (default),     whether feature detection performed files regardless     eventual failures individual files case errors     reported warnings. ... arguments findPeaks method     xcmsRaw class","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"default values files, snames, sclass,   phenoData arguments cause function recursively search   readable files. filename without extention used   sample name. subdirectory path used sample class.   files contain positive negative spectra, polarity   can selected explicitly. default (NULL) read scans. phenoData provided, stored phenoData   slot returned xcmsSet class. data.frame   contains column named “class”, content returned   sampclass method thus used   group/class assignment individual files (e.g. peak grouping   etc.). details see help xcmsSet-class. step size (m/z) use profile generation can submitted   either using profparam argument   (e.g. profparam=list(step=0.1)) submitting   step=0.1. specifying value 0 profile matrix   generation can skipped. feature/peak detection algorithm can specified   method argument defaults \"matchFilter\"   method (findPeaks.matchedFilter). Possible values   returned getOption(\"BioC\")$xcms$findPeaks.methods. lock mass correction allows lock mass scan added back   last working scan. correction gives better reproducibility   sample sets.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"xcmsSet object.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"Colin . Smith, csmith@scripps.edu","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSet.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Constructor for xcmsSet objects which finds peaks in NetCDF/mzXML files — xcmsSet","text":"arguments profmethod profparam influence   feature/peak detection. step size parameter step   profile generation findPeaks.matchedFilter   peak detection algorithm can passed using ....","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-class.html","id":null,"dir":"Reference","previous_headings":"","what":"Virtual class for raw data sources — xcmsSource-class","title":"Virtual class for raw data sources — xcmsSource-class","text":"virtual class provides implementation-independent way load   mass spectrometer data various sources use   xcmsRaw object. Subclasses can defined   enable data loaded user-specified sources. virtual   class xcmsFileSource included box   contains file name character string. implementing child classes xcmsSource, corresponding   loadRaw-methods method must provided accepts   xcmsSource child class returns list format   described loadRaw-methods.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-class.html","id":"objects-from-the-class","dir":"Reference","previous_headings":"","what":"Objects from the Class","title":"Virtual class for raw data sources — xcmsSource-class","text":"virtual Class: objects may created .","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-class.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Virtual class for raw data sources — xcmsSource-class","text":"Daniel Hackney, dan@haxney.org","code":""},{"path":[]},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-methods.html","id":null,"dir":"Reference","previous_headings":"","what":"Create an xcmsSource object in a flexible way — xcmsSource-methods","title":"Create an xcmsSource object in a flexible way — xcmsSource-methods","text":"Users can define alternate means reading data   xcmsRaw objects creating new implementations   method.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-methods.html","id":"methods","dir":"Reference","previous_headings":"","what":"Methods","title":"Create an xcmsSource object in a flexible way — xcmsSource-methods","text":"signature(object = \"xcmsSource\") Pass object unmodified.","code":""},{"path":"https://sneumann.github.io/xcms/reference/xcmsSource-methods.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Create an xcmsSource object in a flexible way — xcmsSource-methods","text":"Daniel Hackney, dan@haxney.org","code":""},{"path":[]}]