Adjusted Vignettes and removed unnecessary dependencies

DESCRIPTION

@@ -1,14 +1,14 @@
 Package: JACUSA2helper
 Type: Package
 Title: Post-processing for JACUSA2 output
-Version: 1.9.9.9100
+Version: 1.9.9.9200
 Depends: R (>= 3.5)
-Date: 2022-01-26
+Date: 2022-02-11
 Authors@R: c(person("Michael", "Piechotta", email = "[email protected]", role = c("aut", "cre")),
          person("Qi", "Wang", role = c("ctb")),
          person("Christoph", "Dieterich", role = c("ctb")),
          person("Dieterich Lab", role = c("fnd")))
-Description: This package enables post-processing of JACUSA2 output: read, filter, plot, write.
+Description: This package enables post-processing of JACUSA2 output: read, filter, and plot.
 URL: https://github.com/dieterich-lab/JACUSA2helper, https://dieterich-lab.github.io/JACUSA2helper/
 BugReports: https://github.com/dieterich-lab/JACUSA2helper/issues
 License: GPL-3 + file LICENSE
@@ -18,27 +18,25 @@ LazyData: true
 LazyDataCompression: bzip2
 biocViews:
 Imports: 
+  curl,
   data.table, 
   tibble,
   tidyr,
   dplyr, 
-  rlang, 
   magrittr,
-  stringr,
   Rcpp,
   GenomicRanges,
   IRanges,
-  plyranges,
-  methods
+  plyranges
 Suggests: 
   ggplot2, 
   scales,
   Rsamtools,
   testthat (>= 2.1.0),
   knitr,
+  VennDiagram,
   reshape2,
   rmarkdown,
-  markdown,
   NMF,
   BSgenome,
   BSgenome.Hsapiens.NCBI.GRCh38

NAMESPACE

@@ -13,8 +13,6 @@ export(clean_tag)
 export(coverage)
 export(expand_tag)
 export(extend)
-export(extract_info)
-export(extract_score)
 export(fast_unpack_info)
 export(filter_artefact)
 export(gather_repl)
@@ -38,5 +36,4 @@ export(variant_bc)
 export(write_bedGraph)
 importFrom(Rcpp,sourceCpp)
 importFrom(magrittr,"%>%")
-importFrom(methods,as)
 useDynLib(JACUSA2helper, .registration = TRUE)

R/io.R

@@ -2,17 +2,32 @@
 #'
 #' \code{read_result()} reads data that was generated by JACUSA2 and creates a JACUSA2 result object.
 #'
-#' @param file String that represents the file name of the JACUSA2 output.
+#' @param file String that represents the file name of the JACUSA2 output or an URL.
 #' @param cond_desc Vector of strings that represent names/descriptions for conditions.
 #' @param unpack Boolean indicates if info column should be processed.
-#' @param ... parameters forwarded to `data.table::fread`
+#' @param tmpdir String @see `data.table::fread`.
+#' @param showProgress Boolean @see `data.table::fread`.
+#' @param ... parameters forwarded to `data.table::fread`.
 #'
 #' @importFrom magrittr %>%
 #'
 #' @export
-read_result <- function(file, cond_desc = c(), unpack = FALSE, ...) {
+read_result <- function(file, cond_desc = c(), unpack = FALSE, tmpdir = tempdir(), showProgress = FALSE, ...) {
+  # if url download first
+  if (grepl("ftp://|http://", file)) {
+    # partly adopted from data.table::fread
+    tmpFile <- tempfile(tmpdir = tmpdir)
+    if (! requireNamespace("curl", quietly = TRUE)) 
+      stop("Input URL requires https:// connection for which fread() requires 'curl' package which cannot be found. Please install 'curl' using 'install.packages('curl')'.")
+
+    tmpFile = tempfile(fileext = paste0(".", tools::file_ext(file)), tmpdir = tmpdir)
+    curl::curl_download(file, tmpFile, mode = "wb", quiet = ! showProgress)
+    file = tmpFile
+    on.exit(unlink(file), add = TRUE)
+  }
+
   # pre process file
-  # parse comments ^(#|##) to determine result/method type and number of conditions
+  # parse comments ^(#|##) to determine result/method type and number of conditions  
   fun <- ifelse(grepl('\\.gz$', file), base::gzfile, base::file)
   con <- fun(file, "r")
   cond_count <- 0
@@ -196,15 +211,21 @@ base_call <-function(bases) {
 .matches <- function(df, prefixes, cond_count) {
   # regex parts:
   prefix_regex = paste0("^(", paste0(prefixes, collapse = "|"), ")")
-  cond_regex = paste0("([:digit:]{", nchar(cond_count), "})")
-  repl_regex = "([:digit:]+)"
+  cond_regex = paste0("([0-9]{", nchar(cond_count), "})")
+  repl_regex = "([0-9]+)"
+
   # get relevant cols and disect to m(atch):
   # match, base_type, condition, replicate
-  cols <- stringr::str_subset(names(df), paste0(prefix_regex, cond_regex, repl_regex))
-  m <- stringr::str_match(
-    cols,
-    paste0(prefix_regex, cond_regex, repl_regex)
+  cols <- grep(paste0(prefix_regex, cond_regex, repl_regex), names(df), value = TRUE)
+  m <- do.call(
+    rbind, 
+    regmatches(names(df), regexec(paste0(prefix_regex, cond_regex, repl_regex), names(df)))
   )
+
+  col_names <- c("col", "prefix", "cond", "repl")
+  if (ncol(m) != length(col_names)) {
+    m <- matrix(character(), 0, length(col_names))
+  }
   colnames(m) <- c("col", "prefix", "cond", "repl")
   matches <- as.data.frame(m, stringsAsFactors = FALSE)
 
@@ -381,7 +402,6 @@ unpack_info <- function(info, cond_count, keys=get_info_keys(info)) {
 #' @param meta_conds Vector of string vectors that explain files.
 #' @param cond_descs Vector of strings that represent names/descriptions for conditions.
 #' @param unpack Boolean indicates if info column should be processed.
-#' @importFrom methods as
 #' @export
 read_results <- function(files, meta_conds, cond_descs = replicate(length(files), c()), unpack = FALSE) {
   stopifnot(length(files) == length(meta_conds))
@@ -401,7 +421,8 @@ read_results <- function(files, meta_conds, cond_descs = replicate(length(files)
   types <- lapply(l, "[[", 2) %>% unlist(use.names = FALSE)
 
   # combine read files
-  results <- unlist(as(lapply(l, "[[", 1), "GRangesList"))
+  #results <- unlist(as(lapply(l, "[[", 1), "GRangesList"))
+  results <- unlist(GenomicRanges::GRangesList(lapply(l, "[[", 1)))
   results$meta_cond <- as.factor(results$meta_cond)
 
   attr(results, .ATTR_TYPE) <- types

README.md

@@ -24,5 +24,4 @@ Check vignettes:
 * `vignette("Introduction to JACUSA2helper")`
 * `vignette("Analysis of DART-seq with JACUSA2helper")`
 * `vignette("Analysis of m6A mapping by MazF with JACUSA2helper")`
-* `vignette("Analysis of Nanopore HEK293 with JACUSA2helper")`
-* `vignette("Introduction to meta conditions with JACUSA2helper")`
+# * `vignette("Introduction to meta conditions with JACUSA2helper")`

vignettes/JACUSA2helper-dart-seq.Rmd

@@ -73,7 +73,7 @@ We observe a lot more C-to-U transitions in the active YTH binding domain as com
 
 ```{r}
 tidyr::tibble(
-  base_sub = c(ref2YTH,ref2YTHmut), 
+  base_sub = c(ref2YTH, ref2YTHmut), 
   Source = c(
     rep("YTH",length(ref2YTH)),
     rep("YTHmut",length(ref2YTHmut)))