Allow rasusa random seed to be configured (#20)

BCCDC-PHL · Mar 27, 2024 · a903c06 · a903c06
1 parent 0a424e8
commit a903c06
Show file tree

Hide file tree

Showing 3 changed files with 28 additions and 15 deletions.
diff --git a/README.md b/README.md
@@ -144,6 +144,13 @@ nextflow run BCCDC-PHL/downsample-reads \
   --outdir </path/to/output_dir>
 ```
 
+### Random Seed
+
+rasusa allows users to specify a [random seed](https://github.com/mbhall88/rasusa?tab=readme-ov-file#random-seed) to be used
+for its random subsampling algorithm. By default, rasusa generates a random seed at runtime using inputs from the operating system.
+This pipeline sets the default random seed to `0`, which ensures that the same set of reads will be sampled given the same inputs.
+A different random seed can be set using the `--random_seed` flag.
+
 ## Output
 
 A pair of fastq.gz files will be produced for each target coverage, for each sample.
@@ -220,6 +227,8 @@ In the output directory for each sample, a provenance file will be written with
           value: 10
         - parameter: --genome-size
           value: 4.4m
+        - parameter: --seed
+          value: 0
 - filename: NC000962-downsample-10x_R1.fastq.gz
   file_type: fastq-output
   sha256: 2fe74753d889d1b6f02832a09b10a1cab51b1fb2e16a2af20577277aded07a83

diff --git a/modules/downsample_reads.nf b/modules/downsample_reads.nf
@@ -81,23 +81,26 @@ process downsample {
 
     script:
     """
-    printf -- "- process_name: downsample\\n"         >> ${sample_id}_${coverage}x_downsample_provenance.yml
-    printf -- "  tools:\\n"                           >> ${sample_id}_${coverage}x_downsample_provenance.yml
-    printf -- "    - tool_name: rasusa\\n"            >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "- process_name: downsample\\n"             >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "  tools:\\n"                               >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "    - tool_name: rasusa\\n"                >> ${sample_id}_${coverage}x_downsample_provenance.yml
     printf -- "      tool_version: \$(rasusa --version 2>&1 | cut -d ' ' -f 2)\\n" >> ${sample_id}_${coverage}x_downsample_provenance.yml
-    printf -- "      parameters:\\n"                  >> ${sample_id}_${coverage}x_downsample_provenance.yml
-    printf -- "        - parameter: --coverage\\n"    >> ${sample_id}_${coverage}x_downsample_provenance.yml
-    printf -- "          value: ${coverage}\\n"       >> ${sample_id}_${coverage}x_downsample_provenance.yml
-    printf -- "        - parameter: --genome-size\\n" >> ${sample_id}_${coverage}x_downsample_provenance.yml
-    printf -- "          value: ${genome_size}\\n"    >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "      parameters:\\n"                      >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "        - parameter: --coverage\\n"        >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "          value: ${coverage}\\n"           >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "        - parameter: --genome-size\\n"     >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "          value: ${genome_size}\\n"        >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "        - parameter: --seed\\n"            >> ${sample_id}_${coverage}x_downsample_provenance.yml
+    printf -- "          value: ${params.random_seed}\\n" >> ${sample_id}_${coverage}x_downsample_provenance.yml
     
     rasusa \
-      -i ${reads[0]} \
-      -i ${reads[1]} \
-      --coverage ${coverage} \
-      --genome-size ${genome_size} \
-      -o ${sample_id}-downsample-${coverage}x_R1.fastq.gz \
-      -o ${sample_id}-downsample-${coverage}x_R2.fastq.gz
+        --seed ${params.random_seed} \
+        -i ${reads[0]} \
+        -i ${reads[1]} \
+        --coverage ${coverage} \
+        --genome-size ${genome_size} \
+        -o ${sample_id}-downsample-${coverage}x_R1.fastq.gz \
+        -o ${sample_id}-downsample-${coverage}x_R2.fastq.gz
     """
 }
 
diff --git a/nextflow.config b/nextflow.config
@@ -1,7 +1,7 @@
 manifest {
     author = 'Dan Fornika'
     name = 'BCCDC-PHL/downsample-reads'
-    version = '0.1.1'
+    version = '0.2.0'
     description = 'Downsample Reads'
     mainScript = 'main.nf'
     nextflowVersion = '>=20.01.0'
@@ -17,6 +17,7 @@ params {
     coverages = 'NO_FILE'
     coverage = 30
     genome_size = '5m'
+    random_seed = 0
     enable_quality_trimming = false
     disable_quality_filtering = false
     collect_outputs = false