Merge pull request #556 from MaxUlysse/StrelkaBP

Follow Strelka BP and use Manta small indels candidates as input

Author: Szilveszter Juhos

annotate.nf

@@ -77,16 +77,18 @@ vcfNotToAnnotate = Channel.create()
 
 if (annotateVCF == []) {
   Channel.empty().mix(
-    Channel.fromPath("${params.outDir}/VariantCalling/HaplotypeCaller/*.vcf.gz")
+    Channel.fromPath("${directoryMap.haplotypecaller}/*.vcf.gz")
       .flatten().map{vcf -> ['haplotypecaller',vcf]},
-    Channel.fromPath("${params.outDir}/VariantCalling/Manta/*SV.vcf.gz")
+    Channel.fromPath("${directoryMap.manta}/*SV.vcf.gz")
       .flatten().map{vcf -> ['manta',vcf]},
-    Channel.fromPath("${params.outDir}/VariantCalling/MuTect1/*.vcf.gz")
+    Channel.fromPath("${directoryMap.mutect1}/*.vcf.gz")
       .flatten().map{vcf -> ['mutect1',vcf]},
-    Channel.fromPath("${params.outDir}/VariantCalling/MuTect2/*.vcf.gz")
+    Channel.fromPath("${directoryMap.mutect2}/*.vcf.gz")
       .flatten().map{vcf -> ['mutect2',vcf]},
-    Channel.fromPath("${params.outDir}/VariantCalling/Strelka/*{somatic,variants}*.vcf.gz")
-      .flatten().map{vcf -> ['strelka',vcf]}
+    Channel.fromPath("${directoryMap.strelka}/*{somatic,variants}*.vcf.gz")
+      .flatten().map{vcf -> ['strelka',vcf]},
+    Channel.fromPath("${directoryMap.strelkabp}/*{somatic,variants}*.vcf.gz")
+      .flatten().map{vcf -> ['strelkabp',vcf]}
   ).choice(vcfToAnnotate, vcfNotToAnnotate) {
     annotateTools == [] || (annotateTools != [] && it[0] in annotateTools) ? 0 : 1
   }
@@ -213,6 +215,12 @@ def checkUppmaxProject() {
 
 def defineDirectoryMap() {
   return [
+    'haplotypecaller'  : "${params.outDir}/VariantCalling/HaplotypeCaller",
+    'manta'            : "${params.outDir}/VariantCalling/Manta",
+    'mutect1'          : "${params.outDir}/VariantCalling/MuTect1",
+    'mutect2'          : "${params.outDir}/VariantCalling/MuTect2",
+    'strelka'          : "${params.outDir}/VariantCalling/Strelka",
+    'strelkabp'        : "${params.outDir}/VariantCalling/StrelkaBP",
     'bcftoolsStats'    : "${params.outDir}/Reports/BCFToolsStats",
     'snpeff'           : "${params.outDir}/Annotation/SnpEff",
     'vep'              : "${params.outDir}/Annotation/VEP"

configuration/base.config

@@ -42,6 +42,7 @@ params {
   sampleDir = '' // samples directory (for Germline only)
   singularity = false // Don't use singularity to build buildContainers.nf
   step = 'mapping' // Default step is mapping
+  strelkaBP = false // Don't use Manta's candidate indels as input to Strelka
   tag = 'latest' // Default tag is latest, to be overwritten by --tag <version>
   test = false // Not testing by default
   tools = '' // List of tools to use

configuration/containers.config

@@ -40,5 +40,6 @@ process {
   $RunSingleStrelka.container         = "${params.repository}/sarek:${params.tag}"
   $RunSnpeff.container = {params.genome == 'GRCh38' ? "${params.repository}/snpeffgrch38:${params.tag}" : "${params.repository}/snpeffgrch37:${params.tag}"}
   $RunStrelka.container               = "${params.repository}/sarek:${params.tag}"
+  $RunStrelkaBP.container             = "${params.repository}/sarek:${params.tag}"
   $RunVEP.container = {params.genome == 'GRCh38' ? "${params.repository}/vepgrch38:${params.tag}" : "${params.repository}/vepgrch37:${params.tag}"}
 }

configuration/singularity-path.config

@@ -46,5 +46,6 @@ process {
   $RunSingleStrelka.container         = "${params.containerPath}/sarek-${params.tag}.img"
   $RunSnpeff.container                = {params.genome == 'GRCh38' ? "${params.containerPath}/snpeffgrch38-${params.tag}.img" : "${params.containerPath}/snpeffgrch37-${params.tag}.img"}
   $RunStrelka.container               = "${params.containerPath}/sarek-${params.tag}.img"
+  $RunStrelkaBP.container             = "${params.containerPath}/sarek-${params.tag}.img"
   $RunVEP.container                   = {params.genome == 'GRCh38' ? "${params.containerPath}/vepgrch38-${params.tag}.img" : "${params.containerPath}/vepgrch37-${params.tag}.img"}
 }

lib/SarekUtils.groovy

@@ -55,6 +55,8 @@ class MyUtils {
       'singleCPUMem',
       'singularity',
       'step',
+      'strelka-BP',
+      'strelkaBP',
       'tag',
       'test',
       'tools',

scripts/test.sh

@@ -94,12 +94,13 @@ fi
 
 if [[ ALL,TOOLS =~ $TEST ]]
 then
-  run_wrapper --somatic --sample $SAMPLE --variantCalling  --tools FreeBayes,HaplotypeCaller,MuTect1,MuTect2,Strelka
+  run_wrapper --somatic --sample $SAMPLE --variantCalling  --tools FreeBayes,HaplotypeCaller,MuTect1,MuTect2
 fi
 
 if [[ ALL,MANTA =~ $TEST ]]
 then
   run_wrapper --somatic --sample data/tsv/tiny-manta.tsv --variantCalling --tools Manta --noReports
+  run_wrapper --somatic --sample data/tsv/tiny-manta.tsv --variantCalling --tools Manta,Strelka --noReports --strelkaBP
   clean_repo
 fi
 

somaticVC.nf

@@ -546,6 +546,7 @@ process RunManta {
 
   output:
     set val("manta"), idPatient, idSampleNormal, idSampleTumor, file("*.vcf.gz"), file("*.vcf.gz.tbi") into mantaOutput
+    set idPatient, idSampleNormal, file(bamNormal), file(baiNormal), idSampleTumor, file(bamTumor), file(baiTumor), file("*.candidateSmallIndels.vcf.gz"), file("*.candidateSmallIndels.vcf.gz.tbi") into mantaToStrelka
 
   when: 'manta' in tools && !params.onlyQC
 
@@ -633,6 +634,54 @@ if (params.verbose) singleMantaOutput = singleMantaOutput.view {
   Index : ${it[4].fileName}"
 }
 
+
+process RunStrelkaBP {
+  tag {idSampleTumor + "_vs_" + idSampleNormal}
+
+  publishDir directoryMap.strelkabp, mode: 'link'
+
+  input:
+    set idPatient, idSampleNormal, file(bamNormal), file(baiNormal), idSampleTumor, file(bamTumor), file(baiTumor), file(mantaCSI), file(mantaCSIi) from mantaToStrelka
+    set file(genomeFile), file(genomeIndex), file(genomeDict) from Channel.value([
+      referenceMap.genomeFile,
+      referenceMap.genomeIndex,
+      referenceMap.genomeDict
+    ])
+
+  output:
+    set val("strelkaBP"), idPatient, idSampleNormal, idSampleTumor, file("*.vcf.gz"), file("*.vcf.gz.tbi") into strelkaBPOutput
+
+  when: 'strelka' in tools && 'manta' in tools && params.strelkaBP && !params.onlyQC
+
+  script:
+  """
+  \$STRELKA_INSTALL_PATH/bin/configureStrelkaSomaticWorkflow.py \
+  --tumor ${bamTumor} \
+  --normal ${bamNormal} \
+  --referenceFasta ${genomeFile} \
+  --indelCandidates ${mantaCSI} \
+  --runDir Strelka
+
+  python Strelka/runWorkflow.py -m local -j ${task.cpus}
+
+  mv Strelka/results/variants/somatic.indels.vcf.gz \
+    Strelka_${idSampleTumor}_vs_${idSampleNormal}_somatic_indels.vcf.gz
+  mv Strelka/results/variants/somatic.indels.vcf.gz.tbi \
+    Strelka_${idSampleTumor}_vs_${idSampleNormal}_somatic_indels.vcf.gz.tbi
+  mv Strelka/results/variants/somatic.snvs.vcf.gz \
+    Strelka_${idSampleTumor}_vs_${idSampleNormal}_somatic_snvs.vcf.gz
+  mv Strelka/results/variants/somatic.snvs.vcf.gz.tbi \
+    Strelka_${idSampleTumor}_vs_${idSampleNormal}_somatic_snvs.vcf.gz.tbi
+  """
+}
+
+if (params.verbose) strelkaBPOutput = strelkaBPOutput.view {
+  "Variant Calling output:\n\
+  Tool  : ${it[0]}\tID    : ${it[1]}\tSample: [${it[3]}, ${it[2]}]\n\
+  Files : ${it[4].fileName}\n\
+  Index : ${it[5].fileName}"
+}
+
 // Run commands and code from Malin Larsson
 // Based on Jesper Eisfeldt's code
 process RunAlleleCount {
@@ -844,7 +893,8 @@ def defineDirectoryMap() {
     'manta'            : "${params.outDir}/VariantCalling/Manta",
     'mutect1'          : "${params.outDir}/VariantCalling/MuTect1",
     'mutect2'          : "${params.outDir}/VariantCalling/MuTect2",
-    'strelka'          : "${params.outDir}/VariantCalling/Strelka"
+    'strelka'          : "${params.outDir}/VariantCalling/Strelka",
+    'strelkabp'        : "${params.outDir}/VariantCalling/StrelkaBP"
   ]
 }