Merge branch 'QC' into 'develop'

Changes for flexiprep QC pipeline

See merge request sasc/wdl-tasks!1

bioconda.wdl

@@ -20,5 +20,6 @@ task installPrefix {
     }
     output {
         File condaEnvPath=prefix
+        File condaJson=stdout()
     }
  }

centrifuge.wdl

@@ -6,7 +6,7 @@ task download {
     String libraryPath
     Array[String]? domain
     String? executable = "centrifuge-download"
-    File? condaEnvironment
+    String? preCommand
     String? seqTaxMapPath
     String? database = "refseq"
     String? assemblyLevel
@@ -22,11 +22,12 @@ task download {
     # The bash statement at the beginning is to make sure
     # the directory for the SeqTaxMapPath exists.
     command {
+        set -e -o pipefail
+        ${preCommand}
         ${'if [ ! -f ' + seqTaxMapPath +
         ' ]; then mkdir -p ' + seqTaxMapPath +
         '; rm -d ' + seqTaxMapPath +
         '; fi' }
-        ${"source activate " + condaEnvironment}
         ${executable} \
         -o ${libraryPath} \
         ${true='-d ' false='' defined(domain)}${sep=','  domain} \
@@ -50,9 +51,10 @@ task download {
 task downloadTaxonomy {
     String centrifugeTaxonomyDir
     String? executable = "centrifuge-download"
-    File? condaEnvironment
+    String? preCommand
     command {
-        ${"source activate " + condaEnvironment}
+        set -e -o pipefail
+        ${preCommand}
         ${executable} \
         -o ${centrifugeTaxonomyDir} \
         taxonomy
@@ -69,7 +71,7 @@ task build {
     File taxonomyTree
     File inputFasta
     String centrifugeIndexBase
-    File? condaEnvironment
+    String? preCommand
     String? centrifugeBuildExecutable = "centrifuge-build"
     #Boolean? c = false
     Boolean? largeIndex = false
@@ -88,9 +90,10 @@ task build {
     Int? kmerCount
 
     command {
+        set -e -o pipefail
+        ${preCommand}
         mkdir -p  ${centrifugeIndexBase}
         rm -d ${centrifugeIndexBase}
-        ${"source activate " + condaEnvironment}
         ${centrifugeBuildExecutable} \
         ${true='--large-index' false='' largeIndex} \
         ${true='--noauto' false='' noAuto} \

cutadapt.wdl

+task cutadapt {
+    File read1
+    File? read2
+    String read1output
+    String? read2output
+    String? format
+    String? preCommand
+    Int? cores = 1
+    String? memory = "4G"
+    Array[String]? adapter
+    Array[String]? front
+    Array[String]? anywhere
+    Array[String]? adapterRead2
+    Array[String]? frontRead2
+    Array[String]? anywhereRead2
+    Boolean? interleaved
+    String? pairFilter
+    Float? errorRate
+    Boolean? noIndels
+    Int? times
+    Int? overlap
+    Boolean? matchReadWildcards
+    Boolean? noMatchAdapterWildcards
+    Boolean? noTrim
+    Boolean? maskAdapter
+    Int? cut
+    String? nextseqTrim
+    String? qualityCutoff
+    Int? qualityBase
+    Int? length
+    Boolean? trimN
+    String? lengthTag
+    String? stripSuffix
+    String? prefix
+    String? suffix
+    Int? minimumLength
+    Int? maximumLength
+    Int? maxN
+    Boolean? discardTrimmed
+    Boolean? discardUntrimmed
+    String? infoFilePath
+    String? restFilePath
+    String? wildcardFilePath
+    String? tooShortOutputPath
+    String? tooLongOutputPath
+    String? untrimmedOutputPath
+    String? tooShortPairedOutputPath
+    String? tooLongPairedOutputPath
+    String? untrimmedPairedOutputPath
+    Boolean? colorspace
+    Boolean? doubleEncode
+    Boolean? trimPrimer
+    Boolean? stripF3
+    Boolean? maq
+    Boolean? bwa
+    Boolean? zeroCap
+    Boolean? noZeroCap
+    String? reportPath
+
+    command {
+        set -e -o pipefail
+        mkdir -p $(dirname ${read1output})
+        ${preCommand}
+        cutadapt \
+        ${"--cores=" + cores} \
+        ${true="-a " false="" defined(adapter)} ${sep=" -a " adapter} \
+        ${true="-A " false="" defined(adapterRead2)} ${sep=" -A " adapterRead2} \
+        ${true="-g " false="" defined(front)} ${sep=" -g " front} \
+        ${true="-G " false="" defined(frontRead2)} ${sep=" -G " frontRead2} \
+        ${true="-b " false="" defined(anywhere)} ${sep=" -a " anywhere} \
+        ${true="-B " false="" defined(anywhereRead2)} ${sep=" -B " anywhereRead2} \
+        --output ${read1output} ${"--paired-output " + read2output} \
+        ${"--to-short-output " + tooShortOutputPath} ${"--to-short-paired-output " + tooShortPairedOutputPath} \
+        ${"--to-long-output " + tooLongOutputPath} ${"--to-long-paired-output " + tooLongPairedOutputPath} \
+        ${"--untrimmed-output " + untrimmedOutputPath} ${"--untrimmed-paired-output " + untrimmedPairedOutputPath} \
+        ${"--pair-filter " + pairFilter} \
+        ${"--error-rate " + errorRate} \
+        ${"--times " + times} \
+        ${"--overlap " + overlap} \
+        ${"--cut " + cut} \
+        ${"--nextseq-trim " + nextseqTrim} \
+        ${"--quality-cutoff " + qualityCutoff} \
+        ${"--quality-base " + qualityBase} \
+        ${"--length " + length} \
+        ${"--length-tag " + lengthTag} \
+        ${"--strip-suffix " + stripSuffix} \
+        ${"--prefix " + prefix} \
+        ${"--suffix " + suffix} \
+        ${"--minimum-length " + minimumLength} \
+        ${"--maximum-length " + maximumLength} \
+        ${"--max-n " + maxN} \
+        ${true="--discard-untrimmed" false="" discardUntrimmed} \
+        ${"--info-file " + infoFilePath } \
+        ${"--rest-file " + restFilePath } \
+        ${"--wildcard-file " + wildcardFilePath} \
+        ${true="--match-read-wildcards" false="" matchReadWildcards} ${true="--no-match-adapter-wildcards" false="" noMatchAdapterWildcards} \
+        ${true="--no-trim" false="" noTrim} ${true="--mask-adapter" false="" maskAdapter} \
+        ${true="--no-indels" false="" noIndels} ${true="--trim-n" false="" trimN}  \
+        ${true="--interleaved" false="" interleaved} ${true="--discard-trimmed" false="" discardTrimmed } \
+        ${true="--colorspace" false="" colorspace} ${true="--double-encode" false="" doubleEncode} \
+        ${true="--strip-f3" false="" stripF3} ${true="--maq" false="" maq} ${true="--bwa" false="" bwa} \
+        ${true="--zero-cap" false="" zeroCap} ${true="--no-zero-cap" false="" noZeroCap} \
+        ${read1} ${read2} ${"> " + reportPath}
+    }
+    output{
+        File report = if defined(reportPath) then select_first([reportPath]) else stdout()
+        File cutRead1 = read1output
+        File? cutRead2 = read2output
+        File? tooLongOutput=tooLongOutputPath
+        File? tooShortOutput=tooShortOutputPath
+        File? untrimmedOutput=untrimmedOutputPath
+        File? tooLongPairedOutput=tooLongPairedOutputPath
+        File? tooShortPairedOutput=tooShortPairedOutputPath
+        File? untrimmedPairedOutput=untrimmedPairedOutputPath
+        File? infoFile=infoFilePath
+        File? restFile=restFilePath
+        File? wildcardFile=wildcardFilePath
+    }
+    runtime {
+        cpu: select_first([cores])
+        memory: select_first([memory])
+    }
+}
\ No newline at end of file

fastqc.wdl

+task fastqc {
+    File seqFile
+    String outdirPath
+    String? preCommand
+    Boolean? casava
+    Boolean? nano
+    Boolean? noFilter
+    Boolean? extract = true
+    Boolean? nogroup
+    Int? minLength
+    String? format
+    Int? threads = 1
+    File? contaminants
+    File? adapters
+    File? limits
+    Int? kmers
+    String? dir
+
+    command {
+    set -e -o pipefail
+    ${preCommand}
+    mkdir -p ${outdirPath}
+    fastqc \
+    ${"--outdir " + outdirPath} \
+    ${true="--casava" false="" casava} \
+    ${true="--nano" false="" nano} \
+    ${true="--nofilter" false="" noFilter} \
+    ${true="--extract" false="" extract} \
+    ${true="--nogroup" false="" nogroup} \
+    ${"--min_length " + minLength } \
+    ${"--format " + format} \
+    ${"--threads " + threads} \
+    ${"--contaminants " + contaminants} \
+    ${"--adapters " + adapters} \
+    ${"--limits " + limits} \
+    ${"--kmers " + kmers} \
+    ${"--dir " + dir} \
+    ${seqFile}
+
+    }
+
+    output {
+        # Chops of the .gz extension if present.
+        String name = sub(seqFile, "\\.gz$","")
+        # This regex chops of the extension and replaces it with _fastqc for the reportdir.
+        # Just as fastqc does it.
+        String reportDir = outdirPath + "/" + sub(basename(name), "\\.[^\\.]*$", "_fastqc")
+        File rawReport = reportDir + "/fastqc_data.txt"
+        File htmlReport = reportDir + "/fastqc_report.html"
+        File summary = reportDir + "/summary.txt"
+        File adapterContent = reportDir + "/Images/adapter_content.png"
+        File duplicationLevels = reportDir + "/Images/duplication_levels.png"
+        File perBaseNContent = reportDir + "/Images/per_base_n_content.png"
+        File perBaseQuality = reportDir + "/Images/per_base_quality.png"
+        File perBaseSequenceContent = reportDir + "/Images/per_base_sequence_content.png"
+        File perSequenceGCContent = reportDir + "/Images/per_sequence_gc_content.png"
+        File perSequenceQuality = reportDir + "/Images/per_sequence_quality.png"
+        File perTileQuality = reportDir + "/Images/per_tile_quality.png"
+        File sequenceLengthDistribution = reportDir + "/Images/sequence_length_distribution.png"
+    }
+
+    runtime {
+        cpu: select_first([threads])
+    }
+}
+
+task extractAdapters {
+    File extractAdaptersFastqcJar
+    File inputFile
+    String outputDir
+    String? adapterOutputFilePath = outputDir + "/adapter.list"
+    String? contamsOutputFilePath = outputDir + "/contaminations.list"
+    Boolean? skipContams
+    File? knownContamFile
+    File? knownAdapterFile
+    Float? adapterCutoff
+    Boolean? outputAsFasta
+    command {
+    set -e
+    mkdir -p ${outputDir}
+    java -jar ${extractAdaptersFastqcJar} \
+    --inputFile ${inputFile} \
+    ${"--adapterOutputFile " + adapterOutputFilePath } \
+    ${"--contamsOutputFile " + contamsOutputFilePath } \
+    ${"--knownContamFile " + knownContamFile} \
+    ${"--knownAdapterFile " + knownAdapterFile} \
+    ${"--adapterCutoff " + adapterCutoff} \
+    ${true="--skipContams" false="" skipContams} \
+    ${true="--outputAsFasta" false="" outputAsFasta}
+    }
+
+    output {
+        File adapterOutputFile = select_first([adapterOutputFilePath])
+        File contamsOutputFile = select_first([contamsOutputFilePath])
+        Array[String] adapterList = read_lines(select_first([adapterOutputFilePath]))
+        Array[String] contamsList = read_lines(select_first([contamsOutputFilePath]))
+    }
+}
+
+task getConfiguration {
+    String? preCommand
+    String? fastqcDirFile = "fastqcDir.txt"
+    command {
+        set -e -o pipefail
+        ${preCommand}
+        echo $(dirname $(readlink -f $(which fastqc))) > ${fastqcDirFile}
+    }
+    output {
+        String fastqcDir = read_string(fastqcDirFile)
+        File adapterList = fastqcDir + "/Configuration/adapter_list.txt"
+        File contaminantList = fastqcDir + "/Configuration/contaminant_list.txt"
+        File limits = fastqcDir + "/Configuration/limits.txt"
+    }
+}
\ No newline at end of file

ncbi.wdl

@@ -14,11 +14,11 @@ task genomeDownload {
     String? domain = "all"
 
     String? executable = "ncbi-genome-download"
-    File? condaEnvironment
+    String? preCommand
 
     command {
         set -e -o pipefail
-        ${"source activate " + condaEnvironment}
+        ${preCommand}
         ${executable} \
         ${"--section " + section} \
         ${"--format " + format} \