Merge branch 'develop' into 'master'

Release v1.0.0

See merge request tron/tronflow-mutect2!8

.gitignore

@@ -2,7 +2,7 @@
 /shelf/
 /workspace.xml
 work
-.nextflow.log
+.nextflow.log*
 report.html
 timeline.html
 trace.txt

Makefile

@@ -1,6 +1,9 @@
+
+all : clean test check
+
 clean:
 	rm -rf output
-	rm -rf work
+	#rm -rf work
 	rm -f report.html*
 	rm -f timeline.html*
 	rm -f trace.txt*
@@ -10,5 +13,12 @@ clean:
 
 
 test:
-	nextflow main.nf -profile test,conda
-	nextflow main.nf -profile test,conda --disable_common_germline_filter
+	nextflow main.nf -profile test,conda --output output/test1
+	nextflow main.nf -profile test,conda --disable_common_germline_filter --output output/test2
+	nextflow main.nf -profile test,conda --input_files test_data/test_input_with_replicates.txt --output output/test3
+
+
+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; }

README.md

@@ -10,7 +10,7 @@ A nextflow (Di Tommaso, 2017) pipeline implementing the Mutect2 (Benjamin, 2019)
 ```
 $ nextflow run tron-bioinformatics/tronflow-mutect2 -profile conda --help
 Usage:
-    nextflow main.nf --input_files input_files
+    nextflow run tron-bioinformatics/tronflow-mutect2 -profile conda --input_files input_files [--reference reference.fasta]
 
 This workflow is based on the implementation at /code/iCaM/scripts/mutect2_ID.sh
 

main.nf

@@ -15,35 +15,7 @@ params.cpus = 2
 params.disable_common_germline_filter = false
 
 def helpMessage() {
-    log.info"""
-Usage:
-    nextflow run tron-bioinformatics/tronflow-mutect2 -profile conda --input_files input_files [--reference reference.fasta]
-
-This workflow is based on the implementation at /code/iCaM/scripts/mutect2_ID.sh
-
-Input:
-    * input_files: the path to a tab-separated values file containing in each row the sample name, tumor bam and normal bam
-    The input file does not have header!
-    Example input file:
-    name1	tumor_bam1	normal_bam1
-    name2	tumor_bam2	normal_bam2
-
-Optional input:
-    * 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
-    * NOTE: if any of the above parameters is not provided, default hg19 resources will be used
-    * 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
-    from the somatic calls. The GnomAD resource is still required though as this common SNPs are used elsewhere to
-    calculate the contamination (default: false)
-
-Output:
-    * Output VCF
-    * Other intermediate files
-    """
+    log.info params.help_message
 }
 
 if (params.help) {
@@ -56,19 +28,13 @@ if (params.input_files) {
   Channel
     .fromPath(params.input_files)
     .splitCsv(header: ['name', 'tumor_bam', 'normal_bam'], sep: "\t")
-    .map{ row->
-	    tuple(row.name,
-	    file(row.tumor_bam), file(row.tumor_bam + ".bai"),
-	    file(row.normal_bam), file(row.normal_bam + ".bai")
-	    ) }
+    .map{ row-> tuple(row.name, row.tumor_bam, row.normal_bam) }
     .set { input_files }
 
   Channel
       .fromPath(params.input_files)
       .splitCsv(header: ['name', 'tumor_bam', 'normal_bam'], sep: "\t")
-      .map{ row->
-  	    tuple(row.name,
-  	    file(row.tumor_bam), file(row.tumor_bam + ".bai")) }
+      .map{ row-> tuple(row.name, row.tumor_bam) }
       .set { tumor_bams }
 } else {
   exit 1, "Input file not specified!"
@@ -78,38 +44,38 @@ process mutect2 {
     cpus params.cpus
     memory params.memory
     tag "${name}"
-    publishDir "${params.output}", mode: "copy"
+    publishDir "${params.output}/${name}", mode: "copy"
 
     input:
-    	set name, file(tumor_bam), file(tumor_bai), file(normal_bam), file(normal_bai) from input_files
+    	set name, tumor_bam, normal_bam from input_files
 
     output:
-	    set val("${name}"), file("${name}.unfiltered.vcf"), file("${name}.unfiltered.vcf.stats") into unfiltered_vcfs
-      set val("${name}"), file("${name}.f1r2.tar.gz") into f1r2_stats
+	    set val("${name}"), file("${name}.mutect2.unfiltered.vcf"), file("${name}.mutect2.unfiltered.vcf.stats") into unfiltered_vcfs
+        set val("${name}"), file("${name}.f1r2.tar.gz") into f1r2_stats
 
     script:
     	normal_panel_option = params.pon ? "--panel-of-normals ${params.pon}" : ""
     	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 $v"}).join(" ")
 	"""
     gatk --java-options '-Xmx${params.memory}' Mutect2 \
 	--reference ${params.reference} \
 	--intervals ${params.intervals} \
 	${germline_filter} \
 	${normal_panel_option} \
-	--input ${normal_bam} \
-  --normal-sample normal \
-  --input ${tumor_bam} \
-  --tumor-sample tumor \
-	--output ${name}.unfiltered.vcf \
-  --f1r2-tar-gz ${name}.f1r2.tar.gz
+	${normal_inputs} --normal-sample normal \
+    ${tumor_inputs} --tumor-sample tumor \
+	--output ${name}.mutect2.unfiltered.vcf \
+    --f1r2-tar-gz ${name}.f1r2.tar.gz
 	"""
 }
 
 process learnReadOrientationModel {
   cpus params.cpus
   memory params.memory
   tag "${name}"
-  publishDir "${params.output}", mode: "copy"
+  publishDir "${params.output}/${name}", mode: "copy"
 
   input:
     set name, file(f1r2_stats) from f1r2_stats
@@ -128,20 +94,21 @@ process pileUpSummaries {
     cpus params.cpus
     memory params.memory
     tag "${name}"
-    publishDir "${params.output}", mode: "copy"
+    publishDir "${params.output}/${name}", mode: "copy"
 
     input:
-    	set name, file(tumor_bam), file(tumor_bai) from tumor_bams
+    	set name, tumor_bam from tumor_bams
 
     output:
 	   set val("${name}"), file("${name}.pileupsummaries.table") into pileupsummaries
 
     script:
+    tumor_inputs = tumor_bam.split(",").collect({v -> "--input $v"}).join(" ")
 	"""
     gatk --java-options '-Xmx${params.memory}' GetPileupSummaries  \
 	--intervals ${params.gnomad} \
 	--variant ${params.gnomad} \
-	--input ${tumor_bam} \
+	${tumor_inputs} \
 	--output ${name}.pileupsummaries.table
 	"""
 }
@@ -150,7 +117,7 @@ process calculateContamination {
     cpus params.cpus
     memory params.memory
     tag "${name}"
-    publishDir "${params.output}", mode: "copy"
+    publishDir "${params.output}/${name}", mode: "copy"
 
     input:
       set name, file(table) from pileupsummaries
@@ -170,14 +137,15 @@ process filterCalls {
     cpus params.cpus
     memory params.memory
     tag "${name}"
-    publishDir "${params.output}", mode: "copy"
+    publishDir "${params.output}/${name}", mode: "copy"
 
     input:
       set name, file(segments_table), file(contamination_table), file(model),
       file(unfiltered_vcf), file(vcf_stats) from contaminationTables.join(read_orientation_model).join(unfiltered_vcfs)
 
     output:
-      set name, file("${name}.vcf") into final_vcfs
+      set name, val("${params.output}/${name}/${name}.mutect2.vcf") into final_vcfs
+      file "${name}.mutect2.vcf"
 
     """
     gatk --java-options '-Xmx${params.memory}' FilterMutectCalls \
@@ -186,7 +154,7 @@ process filterCalls {
     --tumor-segmentation ${segments_table} \
     --contamination-table ${contamination_table} \
     --ob-priors ${model} \
-    --output ${name}.vcf
+    --output ${name}.mutect2.vcf
     """
 }
 

nextflow.config

@@ -5,20 +5,35 @@
  * Default config options for all environments.
  */
 
-// Container slug. Stable releases should specify release tag!
-// Developmental code should specify :dev
-process.container = 'tron-bioinformatics/tronflow-mutect2:0.3.0'
+test_bams = [
+    "sample_name",
+    "$baseDir/test_data/TESTX_S1_L001.bam",
+    "$baseDir/test_data/TESTX_S1_L002.bam"]
+test_input_file = new File("$baseDir/test_data/test_input.txt")
+test_input_file.write(test_bams.join("\t") + "\n")
+
+test_bams_with_replicates = [
+    "sample_name_with_replicates",
+    "$baseDir/test_data/TESTX_S1_L001.bam,$baseDir/test_data/TESTX_S1_L001.bam",
+    "$baseDir/test_data/TESTX_S1_L002.bam,$baseDir/test_data/TESTX_S1_L002.bam"]
+test_input_file_with_replicates = new File("$baseDir/test_data/test_input_with_replicates.txt")
+test_input_file_with_replicates.write(test_bams_with_replicates.join("\t") + "\n")
+
 
 profiles {
   conda { process.conda = "$baseDir/environment.yml" }
   debug { process.beforeScript = 'echo $HOSTNAME' }
   test {
-    params.input_files = "test_data/test_input.txt"
+    params.input_files = test_input_file
     params.reference = "$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"
+    timeline.enabled = false
+    report.enabled = false
+    trace.enabled = false
+    dag.enabled = false
   }
 }
 
@@ -30,29 +45,48 @@ env {
 // Capture exit codes from upstream processes when piping
 process.shell = ['/bin/bash', '-euo', 'pipefail']
 
-timeline {
-  enabled = true
-  //file = "${params.output}/execution_timeline.html"
-}
-report {
-  enabled = true
-  //file = "${params.output}/execution_report.html"
-}
-trace {
-  enabled = true
-  //file = "${params.output}/execution_trace.txt"
-}
-dag {
-  enabled = true
-  //file = "${params.output}/pipeline_dag.svg"
-}
+VERSION = '1.0.0'
+DOI = 'https://zenodo.org/badge/latestdoi/355860788'
 
 manifest {
   name = 'TRON-Bioinformatics/tronflow-mutect2'
-  author = 'Pablo Riesgo Ferreiro'
+  author = 'Pablo Riesgo-Ferreiro, Özlem Muslu, Luisa Bresadola'
   homePage = 'https://github.com/TRON-Bioinformatics/tronflow-mutect2'
   description = 'Mutect2 best practices workflow'
   mainScript = 'main.nf'
   nextflowVersion = '>=19.10.0'
-  version = '0.3.1'
+  version = VERSION
+  doi = DOI
 }
+
+params.help_message = """
+TronFlow Mutect2 v${VERSION} ${DOI}
+
+Usage:
+    nextflow run tron-bioinformatics/tronflow-mutect2 -profile conda --input_files input_files [--reference reference.fasta]
+
+This workflow is based on the implementation at /code/iCaM/scripts/mutect2_ID.sh
+
+Input:
+    * input_files: the path to a tab-separated values file containing in each row the sample name, tumor bam and normal bam
+    The input file does not have header!
+    Example input file:
+    name1	tumor_bam1	normal_bam1
+    name2	tumor_bam2	normal_bam2
+
+Optional input:
+    * 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
+    * NOTE: if any of the above parameters is not provided, default hg19 resources will be used
+    * 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
+    from the somatic calls. The GnomAD resource is still required though as this common SNPs are used elsewhere to
+    calculate the contamination (default: false)
+
+Output:
+    * Output VCF
+    * Other intermediate files
+    """