Merge pull request #1 from TRON-Bioinformatics/parametrise-memory-and…

…-cpus

Parametrise memory and cpus

.github/workflows/automated_tests.yml

@@ -0,0 +1,22 @@
+name: Automated tests
+
+on: [push]
+
+jobs:
+  test:
+    runs-on: ubuntu-20.04
+
+    steps:
+    - uses: actions/checkout@v2
+    - uses: actions/setup-java@v2
+      with:
+        distribution: 'zulu' # See 'Supported distributions' for available options
+        java-version: '11'
+    - uses: conda-incubator/setup-miniconda@v2
+    - name: Install dependencies
+      run: |
+        apt-get update && apt-get --assume-yes install wget make procps software-properties-common
+        wget -qO- https://get.nextflow.io | bash && cp nextflow /usr/local/bin/nextflow
+    - name: Run tests
+      run: |
+        make

Makefile

@@ -18,9 +18,12 @@ test:
 	nextflow main.nf -profile test,conda --disable_common_germline_filter --output output/test2 --input_files test_data/test_input.txt
 	echo "sample_name_with_replicates\t"`pwd`"/test_data/TESTX_S1_L001.bam,"`pwd`"/test_data/TESTX_S1_L001.bam\t"`pwd`"/test_data/TESTX_S1_L002.bam,"`pwd`"/test_data/TESTX_S1_L002.bam" > test_data/test_input_with_replicates.txt
 	nextflow main.nf -profile test,conda --input_files test_data/test_input_with_replicates.txt --output output/test3
+	nextflow main.nf -profile test,conda --output output/test4 --input_files test_data/test_input.txt --intervals false
 
 
 check:
 	test -s output/test1/sample_name/sample_name.mutect2.vcf || { echo "Missing test 1 output file!"; exit 1; }
 	test -s output/test2/sample_name/sample_name.mutect2.vcf || { echo "Missing test 2 output file!"; exit 1; }
 	test -s output/test3/sample_name_with_replicates/sample_name_with_replicates.mutect2.vcf || { echo "Missing test 3 output file!"; exit 1; }
+	test -s output/test4/sample_name/sample_name.mutect2.vcf || { echo "Missing test 4 output file!"; exit 1; }
+

README.md

@@ -1,6 +1,7 @@
 # TronFlow Mutect2
 
 ![GitHub tag (latest SemVer)](https://img.shields.io/github/v/release/tron-bioinformatics/tronflow-mutect2?sort=semver)
+[![Run tests](https://github.com/TRON-Bioinformatics/tronflow-mutect2/actions/workflows/automated_tests.yml/badge.svg?branch=master)](https://github.com/TRON-Bioinformatics/tronflow-mutect2/actions/workflows/automated_tests.yml)
 [![DOI](https://zenodo.org/badge/355860788.svg)](https://zenodo.org/badge/latestdoi/355860788)
 [![License](https://img.shields.io/badge/license-MIT-green)](https://opensource.org/licenses/MIT)
 [![Powered by Nextflow](https://img.shields.io/badge/powered%20by-Nextflow-orange.svg?style=flat&colorA=E1523D&colorB=007D8A)](https://www.nextflow.io/)
@@ -27,13 +28,21 @@ Input:
     name1	tumor_bam1	normal_bam1
     name2	tumor_bam2	normal_bam2
     * reference: path to the FASTA genome reference (indexes expected *.fai, *.dict)
-    * intervals: path to a BED file containing the regions to analyse
-    * gnomad: path to the gnomad VCF
+    * gnomad: path to the gnomad VCF or other germline resource
     
 Optional input:
+    * intervals: path to a BED file containing the regions to analyse
     * output: the folder where to publish output
-    * memory: the ammount of memory used by each job (default: 16g)
-    * cpus: the number of CPUs used by each job (default: 2)
+    * memory_mutect2: the ammount of memory used by mutect2 (default: 16g)
+    * cpus_mutect2: the number of CPUs used by mutect2 (default: 2)
+    * memory_read_orientation: the ammount of memory used by learn read orientation (default: 16g)
+    * cpus_read_orientation: the number of CPUs used by learn read orientation (default: 2)
+    * memory_pileup: the ammount of memory used by pileup (default: 32g)
+    * cpus_pileup: the number of CPUs used by pileup (default: 2)
+    * memory_contamination: the ammount of memory used by contamination (default: 16g)
+    * cpus_contamination: the number of CPUs used by contamination (default: 2)
+    * memory_filter: the ammount of memory used by filter (default: 16g)
+    * cpus_filter: the number of CPUs used by filter (default: 2)
     * disable_common_germline_filter: disable the use of GnomAD to filter out common variants in the population
     from the somatic calls. The GnomAD resource is still required though as this common SNPs are used elsewhere to
     calculate the contamination (default: false)

environment.yml

@@ -1,6 +1,6 @@
 # You can use this file to create a conda environment for this pipeline:
 #   conda env create -f environment.yml
-name: tronflow-mutect2-0.3.1
+name: tronflow-mutect2-1.2.0
 channels:
   - conda-forge
   - bioconda

main.nf

@@ -8,8 +8,16 @@ params.intervals = false
 params.gnomad = false
 params.output = 'output'
 params.pon = false
-params.memory = "16g"
-params.cpus = 2
+params.memory_mutect2 = "16g"
+params.cpus_mutect2 = 2
+params.memory_read_orientation = "16g"
+params.cpus_read_orientation = 2
+params.memory_pileup = "32g"
+params.cpus_pileup = 2
+params.memory_contamination = "16g"
+params.cpus_contamination = 2
+params.memory_filter = "16g"
+params.cpus_filter = 2
 params.disable_common_germline_filter = false
 
 def helpMessage() {
@@ -24,10 +32,6 @@ if (!params.reference) {
     log.error "--reference is required"
     exit 1
 }
-if (!params.intervals) {
-    log.error "--intervals is required"
-    exit 1
-}
 if (!params.gnomad) {
     log.error "--gnomad is required"
     exit 1
@@ -51,8 +55,8 @@ if (params.input_files) {
 }
 
 process mutect2 {
-    cpus params.cpus
-    memory params.memory
+    cpus params.cpus_mutect2
+    memory params.memory_mutect2
     tag "${name}"
     publishDir "${params.output}/${name}", mode: "copy"
 
@@ -68,10 +72,11 @@ process mutect2 {
     	germline_filter = params.disable_common_germline_filter ? "" : "--germline-resource ${params.gnomad}"
     	normal_inputs = normal_bam.split(",").collect({v -> "--input $v"}).join(" ")
     	tumor_inputs = tumor_bam.split(",").collect({v -> "--input $v"}).join(" ")
+    	intervals_option = params.intervals ? "--intervals ${params.intervals}" : ""
 	"""
-    gatk --java-options '-Xmx${params.memory}' Mutect2 \
+    gatk --java-options '-Xmx${params.memory_mutect2}' Mutect2 \
 	--reference ${params.reference} \
-	--intervals ${params.intervals} \
+	${intervals_option} \
 	${germline_filter} \
 	${normal_panel_option} \
 	${normal_inputs} --normal-sample normal \
@@ -82,8 +87,8 @@ process mutect2 {
 }
 
 process learnReadOrientationModel {
-  cpus params.cpus
-  memory params.memory
+  cpus params.cpus_read_orientation
+  memory params.memory_read_orientation
   tag "${name}"
   publishDir "${params.output}/${name}", mode: "copy"
 
@@ -94,15 +99,15 @@ process learnReadOrientationModel {
     set name, file("${name}.read-orientation-model.tar.gz") into read_orientation_model
 
   """
-  gatk --java-options '-Xmx${params.memory}' LearnReadOrientationModel \
+  gatk --java-options '-Xmx${params.memory_read_orientation}' LearnReadOrientationModel \
   --input ${f1r2_stats} \
   --output ${name}.read-orientation-model.tar.gz
   """
 }
 
 process pileUpSummaries {
-    cpus params.cpus
-    memory params.memory
+    cpus params.cpus_pileup
+    memory params.memory_pileup
     tag "${name}"
     publishDir "${params.output}/${name}", mode: "copy"
 
@@ -115,7 +120,7 @@ process pileUpSummaries {
     script:
     tumor_inputs = tumor_bam.split(",").collect({v -> "--input $v"}).join(" ")
 	"""
-    gatk --java-options '-Xmx${params.memory}' GetPileupSummaries  \
+    gatk --java-options '-Xmx${params.memory_pileup}' GetPileupSummaries  \
 	--intervals ${params.gnomad} \
 	--variant ${params.gnomad} \
 	${tumor_inputs} \
@@ -124,8 +129,8 @@ process pileUpSummaries {
 }
 
 process calculateContamination {
-    cpus params.cpus
-    memory params.memory
+    cpus params.cpus_contamination
+    memory params.memory_contamination
     tag "${name}"
     publishDir "${params.output}/${name}", mode: "copy"
 
@@ -136,16 +141,16 @@ process calculateContamination {
       set name, file("${name}.segments.table"), file("${name}.calculatecontamination.table") into contaminationTables
 
     """
-    gatk --java-options '-Xmx${params.memory}' CalculateContamination \
+    gatk --java-options '-Xmx${params.memory_contamination}' CalculateContamination \
     --input ${table} \
     -tumor-segmentation ${name}.segments.table \
     --output ${name}.calculatecontamination.table
     """
 }
 
 process filterCalls {
-    cpus params.cpus
-    memory params.memory
+    cpus params.cpus_filter
+    memory params.memory_filter
     tag "${name}"
     publishDir "${params.output}/${name}", mode: "copy"
 
@@ -158,7 +163,7 @@ process filterCalls {
       file "${name}.mutect2.vcf"
 
     """
-    gatk --java-options '-Xmx${params.memory}' FilterMutectCalls \
+    gatk --java-options '-Xmx${params.memory_filter}' FilterMutectCalls \
     -V ${unfiltered_vcf} \
     --reference ${params.reference} \
     --tumor-segmentation ${segments_table} \

nextflow.config

@@ -12,8 +12,16 @@ profiles {
     params.reference = "$baseDir/test_data/ucsc.hg19.minimal.fasta"
     params.intervals = "$baseDir/test_data/intervals.minimal.bed"
     params.gnomad = "$baseDir/test_data/gnomad.minimal.vcf.gz"
-    params.cpus = 1
-    params.memory = "2g"
+    params.memory_mutect2 = "2g"
+    params.cpus_mutect2 = 1
+    params.memory_read_orientation = "2g"
+    params.cpus_read_orientation = 1
+    params.memory_pileup = "2g"
+    params.cpus_pileup = 1
+    params.memory_contamination = "2g"
+    params.cpus_contamination = 1
+    params.memory_filter = "2g"
+    params.cpus_filter = 1
     timeline.enabled = false
     report.enabled = false
     trace.enabled = false
@@ -29,7 +37,7 @@ env {
 // Capture exit codes from upstream processes when piping
 process.shell = ['/bin/bash', '-euo', 'pipefail']
 
-VERSION = '1.1.0'
+VERSION = '1.2.0'
 DOI = 'https://zenodo.org/badge/latestdoi/355860788'
 
 manifest {
@@ -58,14 +66,22 @@ Input:
     name1	tumor_bam1	normal_bam1
     name2	tumor_bam2	normal_bam2
     * reference: path to the FASTA genome reference (indexes expected *.fai, *.dict)
-    * intervals: path to a BED file containing the regions to analyse
-    * gnomad: path to the gnomad VCF
+    * gnomad: path to the gnomad VCF or other germline resource
 
 Optional input:
+    * intervals: path to a BED file containing the regions to analyse
     * output: the folder where to publish output
-    * memory: the ammount of memory used by each job (default: 16g)
-    * cpus: the number of CPUs used by each job (default: 2)
-        * disable_common_germline_filter: disable the use of GnomAD to filter out common variants in the population
+    * memory_mutect2: the ammount of memory used by mutect2 (default: 16g)
+    * cpus_mutect2: the number of CPUs used by mutect2 (default: 2)
+    * memory_read_orientation: the ammount of memory used by learn read orientation (default: 16g)
+    * cpus_read_orientation: the number of CPUs used by learn read orientation (default: 2)
+    * memory_pileup: the ammount of memory used by pileup (default: 32g)
+    * cpus_pileup: the number of CPUs used by pileup (default: 2)
+    * memory_contamination: the ammount of memory used by contamination (default: 16g)
+    * cpus_contamination: the number of CPUs used by contamination (default: 2)
+    * memory_filter: the ammount of memory used by filter (default: 16g)
+    * cpus_filter: the number of CPUs used by filter (default: 2)
+    * disable_common_germline_filter: disable the use of GnomAD to filter out common variants in the population
     from the somatic calls. The GnomAD resource is still required though as this common SNPs are used elsewhere to
     calculate the contamination (default: false)