Merge pull request #123 from nf-cmgg/panel-coverage-efficiency

update efficiency of panel coverage

CHANGELOG.md

@@ -5,7 +5,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## dev
 
-# v2.0.4
+- Panel coverage will now run all genelists' coverage analysis in one process per sample
+
+## v2.0.4
 
 - Add flag to `snap` to allow for more expansion of compressed data
 - Parse readgroup samplenames for `snp_` prefixed samples

modules/local/panelcoverage/main.nf

@@ -1,5 +1,5 @@
 process PANELCOVERAGE {
-    tag "$meta.id - $genelist"
+    tag "$meta.id"
     label 'process_single'
 
     conda "${moduleDir}/environment.yml"
@@ -8,7 +8,8 @@ process PANELCOVERAGE {
         'biocontainers/bedtools:2.31.1--hf5e1c6e_1' }"
 
     input:
-    tuple val(meta), path(perbase), path(perbase_index), path(genelist)
+    tuple val(meta), path(perbase), path(perbase_index)
+    path(genelists)
 
     output:
     tuple val(meta), path("*.mosdepth.region.dist.txt"), emit: regiondist
@@ -20,7 +21,10 @@ process PANELCOVERAGE {
     script:
     def prefix = task.ext.prefix ?: "${meta.id}"
     """
-    cmgg_genelists regiondist --samplename ${prefix} --perbase ${perbase} --genelist ${genelist}
+    for GENELIST in $genelists
+    do
+        cmgg_genelists regiondist --samplename ${prefix} --perbase ${perbase} --genelist \$GENELIST
+    done
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
@@ -32,7 +36,11 @@ process PANELCOVERAGE {
     stub:
     def prefix = task.ext.prefix ?: "${meta.id}"
     """
-    touch ${prefix}_genelist.mosdepth.region.dist.txt
+    for GENELIST in $genelists
+    do
+        name=\$(basename \$GENELIST .bed)
+        touch ${prefix}_\${name}.mosdepth.region.dist.txt
+    done
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

subworkflows/local/coverage/main.nf

@@ -33,8 +33,8 @@ workflow COVERAGE {
 
         PANELCOVERAGE(
             MOSDEPTH.out.per_base_bed
-            .join(MOSDEPTH.out.per_base_csi)
-            .combine(ch_genelists)
+            .join(MOSDEPTH.out.per_base_csi),
+            ch_genelists
         )
         ch_versions = ch_versions.mix(PANELCOVERAGE.out.versions.first())
         ch_coverageqc_files = ch_coverageqc_files.mix(PANELCOVERAGE.out.regiondist)

tests/modules/local/panelcoverage/main.nf.test

@@ -17,6 +17,8 @@ nextflow_process {
                     [id: "test", single_end: false ],
                     file("https://github.com/nf-cmgg/test-datasets/raw/preprocessing/data/genomics/homo_sapiens/illumina/bed/sample1.per-base.bed.gz", checkIfExists:true),
                     file("https://github.com/nf-cmgg/test-datasets/raw/preprocessing/data/genomics/homo_sapiens/illumina/bed/sample1.per-base.bed.gz.csi", checkIfExists:true),
+                ]
+                input[1] = [
                     file("https://github.com/nf-cmgg/test-datasets/raw/preprocessing/data/genomics/homo_sapiens/illumina/regions/genelists/genelist_chr21_per_exon.bed",checkIfExists:true)
                 ]
                 """

workflows/preprocessing.nf

@@ -60,7 +60,7 @@ workflow PREPROCESSING {
 
     roi = roi ? file(roi, checkIfExists:true) : null
 
-    genelists = genelists ? file(genelists + "/*.bed", checkIfExists:true) : []
+    genelists = genelists ? Channel.value(file(genelists + "/*.bed", checkIfExists:true)) : Channel.empty()
 
 /*
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~