diff --git a/biopet/biopet.wdl b/biopet/biopet.wdl
index 5c9242b49e7acaf5b0e052a52fa1fdd84d4a2a1c..149d8bfdf33d75a2b1a86b7d2ab06294df7410ae 100644
--- a/biopet/biopet.wdl
+++ b/biopet/biopet.wdl
@@ -452,6 +452,45 @@ task VcfStats {
"/sample_compare/genotype.non_ref.abs.tsv"
File? sampleCompareGenotypeRefAbs = outputDir + "/sample_compare/genotype.ref.abs.tsv"
File? sampleCompareGenotypeRel = outputDir + "/sample_compare/genotype.rel.tsv"
+ # A glob is easier, but duplicates all the outputs
+ Array[File] allStats = select_all([
+ general,
+ genotype,
+ sampleDistributionAvailableAggregate,
+ sampleDistributionAvailable,
+ sampleDistributionCalledAggregate,
+ sampleDistributionCalled,
+ sampleDistributionFilteredAggregate,
+ sampleDistributionFiltered,
+ sampleDistributionHetAggregate,
+ sampleDistributionHetNoNRefAggregate,
+ sampleDistributionHetNonRef,
+ sampleDistributionHet,
+ sampleDistributionHomAggregate,
+ sampleDistributionHomRefAggregate,
+ sampleDistributionHomRef,
+ sampleDistributionHom,
+ sampleDistributionHomVarAggregate,
+ sampleDistributionHomVar,
+ sampleDistributionMixedAggregate,
+ sampleDistributionMixed,
+ sampleDistributionNoCallAggregate,
+ sampleDistributionNoCall,
+ sampleDistributionNonInformativeAggregate,
+ sampleDistributionNonInformative,
+ sampleDistributionToalAggregate,
+ sampleDistributionTotal,
+ sampleDistributionVariantAggregate,
+ sampleDistributionVariant,
+ sampleCompareAlleleAbs,
+ sampleCompareAlleleNonRefAbs,
+ sampleCompareAlleleRefAbs,
+ sampleCompareAlleleRel,
+ sampleCompareGenotypeAbs,
+ sampleCompareGenotypeNonRefAbs,
+ sampleCompareGenotypeRefAbs,
+ sampleCompareGenotypeRel
+ ])
}
runtime {
diff --git a/bwa.wdl b/bwa.wdl
index edc8206f8124fa0506a7142260d155befea47f05..e45c4dbc31de14114c9e3feb84938fa62e3ff3d7 100644
--- a/bwa.wdl
+++ b/bwa.wdl
@@ -1,7 +1,5 @@
version 1.0
-import "common.wdl" as common
-
task Mem {
input {
File read1
@@ -35,10 +33,8 @@ task Mem {
}
output {
- IndexedBamFile bamFile = object {
- file: outputPath,
- index: sub(outputPath, ".bam$", ".bai")
- }
+ File outputBam = outputPath
+ File outputBamIndex = sub(outputPath, "\.bam$", ".bai")
}
runtime{
diff --git a/common.wdl b/common.wdl
index e684dfbcbd0eaf19992abd544075a6c9d516a850..2d099aaa9d921e453b3b2b500acb9d953bcb40e2 100644
--- a/common.wdl
+++ b/common.wdl
@@ -25,8 +25,10 @@ task CheckFileMD5 {
input {
File file
String md5
- # Version not that important as long as it is stable.
- String dockerTag = "5.0.2"
+ # By default cromwell expects /bin/bash to be present in the container
+ # The 'bash' container does not fill this requirement. (It is in /usr/local/bin/bash)
+ # Use a stable version of debian:stretch-slim for this. (Smaller than ubuntu)
+ String dockerImage = "debian@sha256:f05c05a218b7a4a5fe979045b1c8e2a9ec3524e5611ebfdd0ef5b8040f9008fa"
}
command {
@@ -36,8 +38,7 @@ task CheckFileMD5 {
}
runtime {
- # Apparently there is a bash container for this sort of stuff.
- docker: "bash:" + dockerTag
+ docker: dockerImage
}
}
diff --git a/gatk.wdl b/gatk.wdl
index a02a2069436e065f6f9282fc2deefe7816a694a9..e76a93e8c979c6ebee0e8857d669da2e13827458 100644
--- a/gatk.wdl
+++ b/gatk.wdl
@@ -1,15 +1,16 @@
version 1.0
-import "common.wdl"
-
# Apply Base Quality Score Recalibration (BQSR) model
task ApplyBQSR {
input {
- IndexedBamFile inputBam
+ File inputBam
+ File inputBamIndex
String outputBamPath
File recalibrationReport
Array[File]+ sequenceGroupInterval
- Reference reference
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
Int memory = 4
Float memoryMultiplier = 3.0
@@ -23,8 +24,8 @@ task ApplyBQSR {
ApplyBQSR \
--create-output-bam-md5 \
--add-output-sam-program-record \
- -R ~{reference.fasta} \
- -I ~{inputBam.file} \
+ -R ~{referenceFasta} \
+ -I ~{inputBam} \
--use-original-qualities \
-O ~{outputBamPath} \
-bqsr ~{recalibrationReport} \
@@ -35,11 +36,9 @@ task ApplyBQSR {
}
output {
- IndexedBamFile recalibratedBam = {
- "file": outputBamPath,
- "index": sub(outputBamPath, "\.bam$", ".bai"),
- "md5": outputBamPath + ".md5"
- }
+ File recalibratedBam = outputBamPath
+ File recalibratedBamIndex = sub(outputBamPath, "\.bam$", ".bai")
+ File recalibratedBamMd5 = outputBamPath + ".md5"
}
runtime {
@@ -51,13 +50,17 @@ task ApplyBQSR {
# Generate Base Quality Score Recalibration (BQSR) model
task BaseRecalibrator {
input {
- IndexedBamFile inputBam
+ File inputBam
+ File inputBamIndex
String recalibrationReportPath
Array[File]+ sequenceGroupInterval
Array[File]? knownIndelsSitesVCFs
Array[File]? knownIndelsSitesVCFIndexes
- IndexedVcfFile? dbsnpVCF
- Reference reference
+ File? dbsnpVCF
+ File? dbsnpVCFIndex
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
Int memory = 4
Float memoryMultiplier = 3.0
@@ -66,7 +69,7 @@ task BaseRecalibrator {
Array[File]+ knownIndelsSitesVCFsArg = flatten([
select_first([knownIndelsSitesVCFs, []]),
- [select_first([dbsnpVCF]).file]
+ [select_first([dbsnpVCF])]
])
command {
@@ -74,8 +77,8 @@ task BaseRecalibrator {
mkdir -p $(dirname ~{recalibrationReportPath})
gatk --java-options -Xmx~{memory}G \
BaseRecalibrator \
- -R ~{reference.fasta} \
- -I ~{inputBam.file} \
+ -R ~{referenceFasta} \
+ -I ~{inputBam} \
--use-original-qualities \
-O ~{recalibrationReportPath} \
--known-sites ~{sep=" --known-sites " knownIndelsSitesVCFsArg} \
@@ -98,7 +101,9 @@ task CombineGVCFs {
Array[File]+ gvcfFilesIndex
Array[File]+ intervals
String outputPath
- Reference reference
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
Int memory = 4
Float memoryMultiplier = 3.0
@@ -110,17 +115,15 @@ task CombineGVCFs {
mkdir -p $(dirname ~{outputPath})
gatk --java-options -Xmx~{memory}G \
CombineGVCFs \
- -R ~{reference.fasta} \
+ -R ~{referenceFasta} \
-O ~{outputPath} \
-V ~{sep=' -V ' gvcfFiles} \
-L ~{sep=' -L ' intervals}
}
output {
- IndexedVcfFile outputVCF = {
- "file": outputPath,
- "index": outputPath + ".tbi"
- }
+ File outputVcf = outputPath
+ File outputVcfIndex = outputPath + ".tbi"
}
runtime {
@@ -165,26 +168,24 @@ task GenotypeGVCFs {
Array[File]+ gvcfFilesIndex
Array[File]+ intervals
String outputPath
- Reference reference
- IndexedVcfFile? dbsnpVCF
-
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
+ File? dbsnpVCF
+ File? dbsnpVCFIndex
Int memory = 6
Float memoryMultiplier = 2.0
String dockerTag = "4.1.0.0--0"
}
- File dbsnpFile = if (defined(dbsnpVCF))
- then select_first([dbsnpVCF]).file
- else ""
-
command {
set -e
mkdir -p $(dirname ~{outputPath})
gatk --java-options -Xmx~{memory}G \
GenotypeGVCFs \
- -R ~{reference.fasta} \
+ -R ~{referenceFasta} \
-O ~{outputPath} \
- ~{true="-D" false="" defined(dbsnpVCF)} ~{dbsnpFile} \
+ ~{true="-D" false="" defined(dbsnpVCF)} ~{dbsnpVCF} \
-G StandardAnnotation \
--only-output-calls-starting-in-intervals \
-new-qual \
@@ -193,10 +194,9 @@ task GenotypeGVCFs {
}
output {
- IndexedVcfFile outputVCF = {
- "file": outputPath,
- "index": outputPath + ".tbi"
- }
+ File outputVCF = outputPath
+ File outputVCFIndex = outputPath + ".tbi"
+
}
runtime {
@@ -212,38 +212,34 @@ task HaplotypeCallerGvcf {
Array[File]+ inputBamsIndex
Array[File]+ intervalList
String gvcfPath
- Reference reference
+ File referenceFasta
+ File referenceFastaIndex
+ File referenceFastaDict
Float contamination = 0.0
- IndexedVcfFile? dbsnpVCF
-
+ File? dbsnpVCF
+ File? dbsnpVCFIndex
Int memory = 4
Float memoryMultiplier = 3
String dockerTag = "4.1.0.0--0"
}
- File dbsnpFile = if (defined(dbsnpVCF))
- then select_first([dbsnpVCF]).file
- else ""
-
command {
set -e
mkdir -p $(dirname ~{gvcfPath})
gatk --java-options -Xmx~{memory}G \
HaplotypeCaller \
- -R ~{reference.fasta} \
+ -R ~{referenceFasta} \
-O ~{gvcfPath} \
-I ~{sep=" -I " inputBams} \
-L ~{sep=' -L ' intervalList} \
- ~{true="-D" false="" defined(dbsnpVCF)} ~{dbsnpFile} \
+ ~{true="-D" false="" defined(dbsnpVCF)} ~{dbsnpVCF} \
-contamination ~{contamination} \
-ERC GVCF
}
output {
- IndexedVcfFile outputGVCF = {
- "file": gvcfPath,
- "index": gvcfPath + ".tbi"
- }
+ File outputGVCF = gvcfPath
+ File outputGVCFIndex = gvcfPath + ".tbi"
}
runtime {
@@ -256,7 +252,9 @@ task MuTect2 {
input {
Array[File]+ inputBams
Array[File]+ inputBamsIndex
- Reference reference
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
String outputVcf
String tumorSample
String? normalSample
@@ -272,7 +270,7 @@ task MuTect2 {
mkdir -p $(dirname ~{outputVcf})
gatk --java-options -Xmx~{memory}G \
Mutect2 \
- -R ~{reference.fasta} \
+ -R ~{referenceFasta} \
-I ~{sep=" -I " inputBams} \
-tumor ~{tumorSample} \
~{"-normal " + normalSample} \
@@ -281,10 +279,8 @@ task MuTect2 {
}
output {
- IndexedVcfFile vcfFile = {
- "file": outputVcf,
- "index": outputVcf + ".tbi"
- }
+ File vcfFile = outputVcf
+ File vcfFileIndex = outputVcf + ".tbi"
}
runtime {
@@ -295,8 +291,11 @@ task MuTect2 {
task SplitNCigarReads {
input {
- IndexedBamFile inputBam
- Reference reference
+ File inputBam
+ File inputBamIndex
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
String outputBam
Array[File]+ intervals
@@ -310,17 +309,15 @@ task SplitNCigarReads {
mkdir -p $(dirname ~{outputBam})
gatk --java-options -Xmx~{memory}G \
SplitNCigarReads \
- -I ~{inputBam.file} \
- -R ~{reference.fasta} \
+ -I ~{inputBam} \
+ -R ~{referenceFasta} \
-O ~{outputBam} \
-L ~{sep=' -L ' intervals}
}
output {
- IndexedBamFile bam = {
- "file": outputBam,
- "index": sub(outputBam, "\.bam$", ".bai")
- }
+ File bam = outputBam
+ File bamIndex = sub(outputBam, "\.bam$", ".bai")
}
runtime {
diff --git a/picard.wdl b/picard.wdl
index 78c7e8bd597fa31d9f8723f3a4cb0fc2d5711c92..2628ca8125e78087b4fc57e98118e15a9cde47a0 100644
--- a/picard.wdl
+++ b/picard.wdl
@@ -1,7 +1,5 @@
version 1.0
-import "common.wdl"
-
task BedToIntervalList {
input {
File bedFile
@@ -35,8 +33,11 @@ task BedToIntervalList {
task CollectMultipleMetrics {
input {
- IndexedBamFile bamFile
- Reference reference
+ File inputBam
+ File inputBamIndex
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
String basename
Boolean collectAlignmentSummaryMetrics = true
@@ -60,8 +61,8 @@ task CollectMultipleMetrics {
mkdir -p $(dirname "~{basename}")
picard -Xmx~{memory}G \
CollectMultipleMetrics \
- I=~{bamFile.file} \
- R=~{reference.fasta} \
+ I=~{inputBam} \
+ R=~{referenceFasta} \
O=~{basename} \
PROGRAM=null \
~{true="PROGRAM=CollectAlignmentSummaryMetrics" false="" collectAlignmentSummaryMetrics} \
@@ -94,6 +95,26 @@ task CollectMultipleMetrics {
File qualityDistribution = basename + ".quality_distribution_metrics"
File qualityDistributionPdf = basename + ".quality_distribution.pdf"
File qualityYield = basename + ".quality_yield_metrics"
+ # Using a glob is easier. But will lead to very ugly output directories.
+ Array[File] allStats = select_all([
+ alignmentSummary,
+ baitBiasDetail,
+ baitBiasSummary,
+ baseDistributionByCycle,
+ baseDistributionByCyclePdf,
+ errorSummary,
+ gcBiasDetail,
+ gcBiasPdf,
+ gcBiasSummary,
+ insertSizeHistogramPdf,
+ insertSize,
+ preAdapterDetail,
+ qualityByCycle,
+ qualityByCyclePdf,
+ qualityDistribution,
+ qualityDistributionPdf,
+ qualityYield
+ ])
}
runtime {
@@ -106,7 +127,8 @@ task CollectMultipleMetrics {
task CollectRnaSeqMetrics {
input {
- IndexedBamFile bamFile
+ File inputBam
+ File inputBamIndex
File refRefflat
String basename
String strandSpecificity = "NONE"
@@ -121,7 +143,7 @@ task CollectRnaSeqMetrics {
mkdir -p $(dirname "~{basename}")
picard -Xmx~{memory}G \
CollectRnaSeqMetrics \
- I=~{bamFile.file} \
+ I=~{inputBam} \
O=~{basename}.RNA_Metrics \
CHART_OUTPUT=~{basename}.RNA_Metrics.pdf \
STRAND_SPECIFICITY=~{strandSpecificity} \
@@ -143,8 +165,11 @@ task CollectRnaSeqMetrics {
task CollectTargetedPcrMetrics {
input {
- IndexedBamFile bamFile
- Reference reference
+ File inputBam
+ File inputBamIndex
+ File referenceFasta
+ File referenceFastaDict
+ File referenceFastaFai
File ampliconIntervals
Array[File]+ targetIntervals
String basename
@@ -159,8 +184,8 @@ task CollectTargetedPcrMetrics {
mkdir -p $(dirname "~{basename}")
picard -Xmx~{memory}G \
CollectTargetedPcrMetrics \
- I=~{bamFile.file} \
- R=~{reference.fasta} \
+ I=~{inputBam} \
+ R=~{referenceFasta} \
AMPLICON_INTERVALS=~{ampliconIntervals} \
TARGET_INTERVALS=~{sep=" TARGET_INTERVALS=" targetIntervals} \
O=~{basename}.targetPcrMetrics \
@@ -194,6 +219,7 @@ task GatherBamFiles {
command {
set -e
+ mkdir -p $(dirname ~{outputBamPath})
picard -Xmx~{memory}G \
GatherBamFiles \
INPUT=~{sep=' INPUT=' inputBams} \
@@ -203,11 +229,9 @@ task GatherBamFiles {
}
output {
- IndexedBamFile outputBam = object {
- file: outputBamPath,
- index: sub(outputBamPath, ".bam$", ".bai"),
- md5: outputBamPath + ".md5"
- }
+ File outputBam = outputBamPath
+ File outputBamIndex = sub(outputBamPath, "\.bam$", ".bai")
+ File outputBamMd5 = outputBamPath + ".md5"
}
runtime {
@@ -220,7 +244,7 @@ task GatherVcfs {
input {
Array[File]+ inputVcfs
Array[File]+ inputVcfIndexes
- String outputVcfPath
+ String outputVcfPath = "out.vcf.gz"
Int memory = 4
Float memoryMultiplier = 3.0
@@ -229,6 +253,7 @@ task GatherVcfs {
command {
set -e
+ mkdir -p $(dirname ~{outputVcfPath})
picard -Xmx~{memory}G \
GatherVcfs \
INPUT=~{sep=' INPUT=' inputVcfs} \
@@ -287,11 +312,9 @@ task MarkDuplicates {
}
output {
- IndexedBamFile outputBam = object {
- file: outputBamPath,
- index: sub(outputBamPath, ".bam$", ".bai"),
- md5: outputBamPath + ".md5"
- }
+ File outputBam = outputBamPath
+ File outputBamIndex = sub(outputBamPath, "\.bam$", ".bai")
+ File outputBamMd5 = outputBamPath + ".md5"
File metricsFile = metricsPath
}
@@ -326,10 +349,8 @@ task MergeVCFs {
}
output {
- IndexedVcfFile outputVcf = object {
- file: outputVcfPath,
- index: outputVcfPath + ".tbi"
- }
+ File outputVcf = outputVcfPath
+ File outputVcfIndex = outputVcfPath + ".tbi"
}
runtime {
@@ -340,7 +361,8 @@ task MergeVCFs {
task SamToFastq {
input {
- IndexedBamFile inputBam
+ File inputBam
+ File inputBamIndex
String outputRead1
String? outputRead2
String? outputUnpaired
@@ -354,7 +376,7 @@ task SamToFastq {
set -e
picard -Xmx~{memory}G \
SamToFastq \
- I=~{inputBam.file} \
+ I=~{inputBam} \
~{"FASTQ=" + outputRead1} \
~{"SECOND_END_FASTQ=" + outputRead2} \
~{"UNPAIRED_FASTQ=" + outputUnpaired}
@@ -429,10 +451,8 @@ task SortVcf {
}
output {
- IndexedVcfFile outputVcf = object {
- file: outputVcfPath,
- index: outputVcfPath + ".tbi"
- }
+ File outputVcf = outputVcfPath
+ File outputVcfIndex = outputVcfPath + ".tbi"
}
runtime {
diff --git a/samtools.wdl b/samtools.wdl
index a14876d05ad814741c752e3af94a2e4d326a549b..a78f55353d5235463b8e3d9f9e255858c6861676 100644
--- a/samtools.wdl
+++ b/samtools.wdl
@@ -1,7 +1,5 @@
version 1.0
-import "common.wdl"
-
task BgzipAndIndex {
input {
File inputFile
@@ -42,10 +40,8 @@ task Index {
}
output {
- IndexedBamFile outputBam = object {
- file: bamFile,
- index: bamIndexPath
- }
+ File indexedBam = bamFile
+ File index = bamIndexPath
}
runtime {
@@ -56,18 +52,23 @@ task Index {
task Merge {
input {
Array[File]+ bamFiles
- String outputBamPath
+ String outputBamPath = "merged.bam"
Boolean force = true
String dockerTag = "1.8--h46bd0b3_5"
}
+ String indexPath = sub(outputBamPath, "\.bam$",".bai")
command {
+ set -e
+ mkdir -p $(dirname ~{outputBamPath})
samtools merge ~{true="-f" false="" force} ~{outputBamPath} ~{sep=' ' bamFiles}
+ samtools index ~{outputBamPath} ~{indexPath}
}
output {
File outputBam = outputBamPath
+ File outputBamIndex = indexPath
}
runtime {
@@ -177,17 +178,25 @@ task Fastq {
task Tabix {
input {
- String inputFile
+ File inputFile
+ String outputFilePath = "indexed.vcf.gz"
String type = "vcf"
String dockerTag = "0.2.6--ha92aebf_0"
}
-
+ # FIXME: It is better to do the indexing on VCF creation. Not in a separate task. With file localization this gets hairy fast.
command {
- tabix ~{inputFile} -p ~{type}
+ set -e
+ mkdir -p $(dirname ~{outputFilePath})
+ if [ ! -f ~{outputFilePath} ]
+ then
+ ln ~{inputFile} ~{outputFilePath}
+ fi
+ tabix ~{outputFilePath} -p ~{type}
}
output {
- File index = inputFile + ".tbi"
+ File indexedFile = outputFilePath
+ File index = outputFilePath + ".tbi"
}
runtime {