Move scatter functions to biopet extensions

biopet-core/src/main/scala/nl/lumc/sasc/biopet/core/BiopetCommandLineFunction.scala

@@ -35,10 +35,6 @@ trait BiopetCommandLineFunction extends CommandLineResources { biopetFunction =>
   @Input(doc = "deps", required = false)
   var deps: List[File] = Nil
 
-  @Output
-  @Gather(enabled = false)
-  var outputFiles: List[File] = Nil
-
   var executable: String = _
 
   /** This is the default shell for drmaa jobs */

biopet-core/src/main/scala/nl/lumc/sasc/biopet/core/BiopetFifoPipe.scala

@@ -61,9 +61,6 @@ class BiopetFifoPipe(val root: Configurable,
 
     val fifoFiles = fifos
 
-    outputFiles :::= outputs.values.toList.flatten.filter(!fifoFiles.contains(_))
-    outputFiles = outputFiles.distinct
-
     deps :::= inputs.values.toList.flatten.filter(!fifoFiles.contains(_))
     deps = deps.distinct
 

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/Flash.scala

@@ -17,9 +17,9 @@ package nl.lumc.sasc.biopet.extensions
 
 import java.io.File
 
-import nl.lumc.sasc.biopet.core.{ Version, BiopetCommandLineFunction }
+import nl.lumc.sasc.biopet.core.{ BiopetCommandLineFunction, Version }
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.utils.commandline.Input
+import org.broadinstitute.gatk.utils.commandline.{ Input, Output }
 
 import scala.util.matching.Regex
 
@@ -64,16 +64,33 @@ class Flash(val root: Configurable) extends BiopetCommandLineFunction with Versi
 
   private def suffix = outputSuffix.getOrElse("fastq") + (if (compress) ".gz" else "")
 
-  def combinedFastq = new File(outputDirectory, s"$outputPrefix.extendedFrags.$suffix")
-  def notCombinedR1 = new File(outputDirectory, s"$outputPrefix.notCombined_1.$suffix")
-  def notCombinedR2 = new File(outputDirectory, s"$outputPrefix.notCombined_2.$suffix")
-  def outputHistogramTable = new File(outputDirectory, s"$outputPrefix.hist")
-  def outputHistogram = new File(outputDirectory, s"$outputPrefix.histogram")
+  @Output
+  private var _combinedFastq: File = _
+  def combinedFastq = _combinedFastq
+
+  @Output
+  private var _notCombinedR1: File = _
+  def notCombinedR1 = _notCombinedR1
+
+  @Output
+  private var _notCombinedR2: File = _
+  def notCombinedR2 = _notCombinedR2
+
+  @Output
+  private var _outputHistogramTable: File = _
+  def outputHistogramTable = _outputHistogramTable
+
+  @Output
+  private var _outputHistogram: File = _
+  def outputHistogram = _outputHistogram
 
   override def beforeGraph(): Unit = {
     super.beforeGraph()
-    outputFiles :::= combinedFastq :: notCombinedR1 ::
-      notCombinedR2 :: outputHistogramTable :: outputHistogram :: Nil
+    _combinedFastq = new File(outputDirectory, s"$outputPrefix.extendedFrags.$suffix")
+    _notCombinedR1 = new File(outputDirectory, s"$outputPrefix.notCombined_1.$suffix")
+    _notCombinedR2 = new File(outputDirectory, s"$outputPrefix.notCombined_2.$suffix")
+    _outputHistogramTable = new File(outputDirectory, s"$outputPrefix.hist")
+    _outputHistogram = new File(outputDirectory, s"$outputPrefix.histogram")
   }
 
   def cmdLine = executable +

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/RunGubbins.scala

@@ -49,6 +49,9 @@ class RunGubbins(val root: Configurable) extends BiopetCommandLineFunction {
   var verbose: Boolean = config("verbose", default = false)
   var noCleanup: Boolean = config("no_cleanup", default = false)
 
+  @Output
+  var outputFiles: List[File] = Nil
+
   /** Set correct output files */
   override def beforeGraph(): Unit = {
     super.beforeGraph()

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/Tabix.scala

@@ -65,6 +65,9 @@ class Tabix(val root: Configurable) extends BiopetCommandLineFunction with Versi
   /** Formats that tabix can handle */
   private val validFormats: Set[String] = Set("gff", "bed", "sam", "vcf", "psltbl")
 
+  @Output
+  var outputFiles: List[File] = Nil
+
   override def beforeGraph(): Unit = {
     super.beforeGraph()
     p match {

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/VariantEffectPredictor.scala

@@ -150,6 +150,9 @@ class VariantEffectPredictor(val root: Configurable) extends BiopetCommandLineFu
 
   override def defaultCoreMemory = 4.0
 
+  @Output
+  private var _summary: File = null
+
   override def beforeGraph(): Unit = {
     super.beforeGraph()
     if (!cache && !database) {
@@ -157,7 +160,7 @@ class VariantEffectPredictor(val root: Configurable) extends BiopetCommandLineFu
     } else if (cache && dir.isEmpty) {
       Logging.addError("Must supply dir to cache for VariantEffectPredictor")
     }
-    if (statsText) outputFiles :+= new File(output.getAbsolutePath + "_summary.txt")
+    if (statsText) _summary = new File(output.getAbsolutePath + "_summary.txt")
   }
 
   /** Returns command to execute */

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/bowtie/Bowtie2Build.scala

@@ -17,9 +17,9 @@ package nl.lumc.sasc.biopet.extensions.bowtie
 
 import java.io.File
 
-import nl.lumc.sasc.biopet.core.{ Version, BiopetCommandLineFunction }
+import nl.lumc.sasc.biopet.core.{ BiopetCommandLineFunction, Version }
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.utils.commandline.{ Argument, Input }
+import org.broadinstitute.gatk.utils.commandline.{ Argument, Input, Output }
 
 /**
  * Created by pjvan_thof on 8/15/15.
@@ -37,6 +37,9 @@ class Bowtie2Build(val root: Configurable) extends BiopetCommandLineFunction wit
 
   override def defaultCoreMemory = 15.0
 
+  @Output
+  private var outputFiles: List[File] = Nil
+
   override def beforeGraph: Unit = {
     outputFiles ::= new File(reference.getParentFile, baseName + ".1.bt2")
     outputFiles ::= new File(reference.getParentFile, baseName + ".2.bt2")

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/bowtie/BowtieBuild.scala

@@ -17,9 +17,9 @@ package nl.lumc.sasc.biopet.extensions.bowtie
 
 import java.io.File
 
-import nl.lumc.sasc.biopet.core.{ Version, BiopetCommandLineFunction }
+import nl.lumc.sasc.biopet.core.{ BiopetCommandLineFunction, Version }
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.utils.commandline.{ Argument, Input }
+import org.broadinstitute.gatk.utils.commandline.{ Argument, Input, Output }
 
 /**
  * Created by pjvan_thof on 8/15/15.
@@ -37,6 +37,9 @@ class BowtieBuild(val root: Configurable) extends BiopetCommandLineFunction with
 
   override def defaultCoreMemory = 15.0
 
+  @Output
+  private var outputFiles: List[File] = Nil
+
   override def beforeGraph: Unit = {
     outputFiles ::= new File(reference.getParentFile, baseName + ".1.ebwt")
     outputFiles ::= new File(reference.getParentFile, baseName + ".2.ebwt")

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/CombineVariants.scala

@@ -17,8 +17,9 @@ package nl.lumc.sasc.biopet.extensions.gatk
 
 import java.io.File
 
+import nl.lumc.sasc.biopet.utils.VcfUtils
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.utils.commandline.{ Input, Output }
+import org.broadinstitute.gatk.utils.commandline.{ Gather, Input, Output }
 
 /**
  * Extension for CombineVariants from GATK
@@ -47,9 +48,13 @@ class CombineVariants(val root: Configurable) extends Gatk {
     inputMap += file -> name
   }
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
   override def beforeGraph(): Unit = {
     super.beforeGraph()
-    if (outputFile.getName.endsWith(".vcf.gz")) outputFiles :+= new File(outputFile.getAbsolutePath + ".tbi")
+    outputIndex = VcfUtils.getVcfIndexFile(outputFile)
     genotypeMergeOptions match {
       case Some("UNIQUIFY") | Some("PRIORITIZE") | Some("UNSORTED") | Some("REQUIRE_UNIQUE") | None =>
       case _ => throw new IllegalArgumentException("Wrong option for genotypeMergeOptions")

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/SelectVariants.scala

@@ -17,8 +17,9 @@ package nl.lumc.sasc.biopet.extensions.gatk
 
 import java.io.File
 
+import nl.lumc.sasc.biopet.utils.VcfUtils
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.utils.commandline.{ Input, Output }
+import org.broadinstitute.gatk.utils.commandline.{ Gather, Input, Output }
 
 /**
  * Extension for CombineVariants from GATK
@@ -43,9 +44,13 @@ class SelectVariants(val root: Configurable) extends Gatk {
     inputMap += file -> name
   }
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
   override def beforeGraph(): Unit = {
     super.beforeGraph()
-    if (outputFile.getName.endsWith(".vcf.gz")) outputFiles :+= new File(outputFile.getAbsolutePath + ".tbi")
+    outputIndex = VcfUtils.getVcfIndexFile(outputFile)
     deps :::= inputFiles.filter(_.getName.endsWith("vcf.gz")).map(x => new File(x.getAbsolutePath + ".tbi"))
     deps = deps.distinct
   }

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/ApplyRecalibration.scala

@@ -82,14 +82,17 @@ class ApplyRecalibration(val root: Configurable) extends CommandLineGATK with Sc
   @Argument(fullName = "filter_bases_not_stored", shortName = "filterNoBases", doc = "Filter out reads with no stored bases (i.e. '*' where the sequence should be), instead of failing with an error", required = false, exclusiveOf = "", validation = "")
   var filter_bases_not_stored: Boolean = config("filter_bases_not_stored", default = false)
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
   override def beforeGraph() {
     super.beforeGraph()
     deps ++= input.filter(orig => orig != null && (!orig.getName.endsWith(".list"))).map(orig => VcfUtils.getVcfIndexFile(orig))
     if (recal_file != null)
       deps :+= VcfUtils.getVcfIndexFile(recal_file)
     if (out != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(out))
-      if (!org.broadinstitute.gatk.utils.commandline.ArgumentTypeDescriptor.isCompressed(out.getPath))
-        outputFiles :+= VcfUtils.getVcfIndexFile(out)
+      outputIndex = VcfUtils.getVcfIndexFile(out)
     num_threads = Option(getThreads)
   }
 

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/BamGatherFunction.scala

+package nl.lumc.sasc.biopet.extensions.gatk.broad
+
+import org.broadinstitute.gatk.queue.function.scattergather.GatherFunction
+import org.broadinstitute.gatk.queue.function.RetryMemoryLimit
+import java.io.File
+
+import nl.lumc.sasc.biopet.core.BiopetCommandLineFunction
+import nl.lumc.sasc.biopet.extensions.picard.MergeSamFiles
+
+/**
+ * Merges BAM files using htsjdk.samtools.MergeSamFiles.
+ */
+class BamGatherFunction extends MergeSamFiles(null) with GatherFunction {
+
+  override val root = originalFunction match {
+    case b: BiopetCommandLineFunction => b
+    case _                            => null
+  }
+
+  this.assumeSorted = true
+
+  override def freezeFieldValues() {
+    this.input = this.gatherParts.toList
+    this.output = this.originalOutput
+    this.sortOrder = "coordinate"
+    //Left to its own devices (ie, MergeSamFiles.freezeFieldValues), outputIndex
+    //will be in the gather directory.  Ensure that it actually matches this.output
+
+    val originalGATK = originalFunction.asInstanceOf[CommandLineGATK]
+
+    // Whatever the original function can handle, merging *should* do less.
+    this.createIndex = !originalGATK.disable_bam_indexing
+
+    super.freezeFieldValues()
+  }
+}

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/CatVariants.scala

+package nl.lumc.sasc.biopet.extensions.gatk.broad
+
+import java.io.File
+
+import nl.lumc.sasc.biopet.core.BiopetJavaCommandLineFunction
+import nl.lumc.sasc.biopet.utils.config.Configurable
+import org.broadinstitute.gatk.utils.commandline.Argument
+import org.broadinstitute.gatk.utils.commandline.Gather
+import org.broadinstitute.gatk.utils.commandline.Input
+import org.broadinstitute.gatk.utils.commandline.Output
+
+class CatVariants(val root: Configurable) extends BiopetJavaCommandLineFunction {
+  analysisName = "CatVariants"
+  javaMainClass = "org.broadinstitute.gatk.tools.CatVariants"
+
+  /** genome reference file <name>.fasta */
+  @Input(fullName = "reference", shortName = "R", doc = "genome reference file <name>.fasta", required = true, exclusiveOf = "", validation = "")
+  var reference: File = _
+
+  /** Input VCF file/s */
+  @Input(fullName = "variant", shortName = "V", doc = "Input VCF file/s", required = true, exclusiveOf = "", validation = "")
+  var variant: Seq[File] = Nil
+
+  /** output file */
+  @Output(fullName = "outputFile", shortName = "out", doc = "output file", required = true, exclusiveOf = "", validation = "")
+  @Gather(classOf[org.broadinstitute.gatk.queue.function.scattergather.SimpleTextGatherFunction])
+  var outputFile: File = _
+
+  /** assumeSorted should be true if the input files are already sorted (based on the position of the variants) */
+  @Argument(fullName = "assumeSorted", shortName = "assumeSorted", doc = "assumeSorted should be true if the input files are already sorted (based on the position of the variants)", required = false, exclusiveOf = "", validation = "")
+  var assumeSorted: Boolean = _
+
+  /** which type of IndexCreator to use for VCF/BCF indices */
+  @Argument(fullName = "variant_index_type", shortName = "", doc = "which type of IndexCreator to use for VCF/BCF indices", required = false, exclusiveOf = "", validation = "")
+  var variant_index_type: Option[String] = None
+
+  /** the parameter (bin width or features per bin) to pass to the VCF/BCF IndexCreator */
+  @Argument(fullName = "variant_index_parameter", shortName = "", doc = "the parameter (bin width or features per bin) to pass to the VCF/BCF IndexCreator", required = false, exclusiveOf = "", validation = "")
+  var variant_index_parameter: Option[Int] = None
+
+  /** Set the minimum level of logging */
+  @Argument(fullName = "logging_level", shortName = "l", doc = "Set the minimum level of logging", required = false, exclusiveOf = "", validation = "")
+  var logging_level: String = _
+
+  /** Set the logging location */
+  @Output(fullName = "log_to_file", shortName = "log", doc = "Set the logging location", required = false, exclusiveOf = "", validation = "")
+  @Gather(classOf[org.broadinstitute.gatk.queue.function.scattergather.SimpleTextGatherFunction])
+  var log_to_file: File = _
+
+  override def cmdLine = super.cmdLine +
+    required("-R", reference, spaceSeparated = true, escape = true, format = "%s") +
+    repeat("-V", variant, spaceSeparated = true, escape = true, format = "%s") +
+    required("-out", outputFile, spaceSeparated = true, escape = true, format = "%s") +
+    conditional(assumeSorted, "-assumeSorted", escape = true, format = "%s") +
+    optional("--variant_index_type", variant_index_type, spaceSeparated = true, escape = true, format = "%s") +
+    optional("--variant_index_parameter", variant_index_parameter, spaceSeparated = true, escape = true, format = "%s") +
+    optional("-l", logging_level, spaceSeparated = true, escape = true, format = "%s") +
+    optional("-log", log_to_file, spaceSeparated = true, escape = true, format = "%s")
+}

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/CatVariantsGather.scala

+package nl.lumc.sasc.biopet.extensions.gatk.broad
+
+import nl.lumc.sasc.biopet.core.BiopetCommandLineFunction
+import org.broadinstitute.gatk.queue.extensions.gatk.TaggedFile
+import org.broadinstitute.gatk.queue.function.RetryMemoryLimit
+import org.broadinstitute.gatk.queue.function.scattergather.GatherFunction
+
+/**
+ *
+ * Currently this is the default gather for VCFs.
+ * One can set a specific gatherer to use by adding @Gather before any output argument.
+ * For example (used to be part of UG):
+ *           \@Gather(className = "org.broadinstitute.gatk.queue.extensions.gatk.CatVariantsGatherer")
+ *           \@Output(doc="File to which variants should be written",required=true)
+ *           protected VariantContextWriter writer = null;
+ */
+class CatVariantsGatherer extends CatVariants(null) with GatherFunction {
+  this.assumeSorted = true
+
+  analysisName = "Gather_CatVariants"
+
+  override val root = originalFunction match {
+    case b: BiopetCommandLineFunction => b
+    case _                            => null
+  }
+
+  override def freezeFieldValues() {
+    val originalGATK = this.originalFunction.asInstanceOf[CommandLineGATK]
+
+    this.reference = originalGATK.reference_sequence
+    this.variant = this.gatherParts.zipWithIndex map { case (input, index) => new TaggedFile(input, "input" + index) }
+    this.outputFile = this.originalOutput
+    this.assumeSorted = true
+    this.variant_index_type = originalGATK.variant_index_type
+    this.variant_index_parameter = originalGATK.variant_index_parameter
+
+    super.freezeFieldValues()
+  }
+}

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/CombineGVCFs.scala

@@ -8,7 +8,7 @@ package nl.lumc.sasc.biopet.extensions.gatk.broad
 import java.io.File
 
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.queue.extensions.gatk.{ CatVariantsGatherer, TaggedFile }
+import org.broadinstitute.gatk.queue.extensions.gatk.TaggedFile
 import nl.lumc.sasc.biopet.core.ScatterGatherableFunction
 import nl.lumc.sasc.biopet.utils.VcfUtils
 import org.broadinstitute.gatk.utils.commandline.{ Gather, Input, Output, _ }
@@ -59,13 +59,16 @@ class CombineGVCFs(val root: Configurable) extends CommandLineGATK with ScatterG
   @Argument(fullName = "filter_bases_not_stored", shortName = "filterNoBases", doc = "Filter out reads with no stored bases (i.e. '*' where the sequence should be), instead of failing with an error", required = false, exclusiveOf = "", validation = "")
   var filter_bases_not_stored: Boolean = config("filter_bases_not_stored", default = false)
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
   override def beforeGraph() {
     super.beforeGraph()
     dbsnp.foreach(deps :+= VcfUtils.getVcfIndexFile(_))
     deps ++= variant.filter(orig => orig != null && (!orig.getName.endsWith(".list"))).map(orig => VcfUtils.getVcfIndexFile(orig))
     if (out != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(out))
-      if (!org.broadinstitute.gatk.utils.commandline.ArgumentTypeDescriptor.isCompressed(out.getPath))
-        outputFiles :+= VcfUtils.getVcfIndexFile(out)
+      outputIndex = VcfUtils.getVcfIndexFile(out)
   }
 
   override def cmdLine = super.cmdLine +

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/CombineVariants.scala

@@ -8,7 +8,7 @@ package nl.lumc.sasc.biopet.extensions.gatk.broad
 import java.io.File
 
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.queue.extensions.gatk.{ CatVariantsGatherer, TaggedFile }
+import org.broadinstitute.gatk.queue.extensions.gatk.TaggedFile
 import nl.lumc.sasc.biopet.core.ScatterGatherableFunction
 import nl.lumc.sasc.biopet.utils.VcfUtils
 import org.broadinstitute.gatk.utils.commandline.{ Argument, Gather, Output, _ }
@@ -91,12 +91,15 @@ class CombineVariants(val root: Configurable) extends CommandLineGATK with Scatt
   @Argument(fullName = "filter_bases_not_stored", shortName = "filterNoBases", doc = "Filter out reads with no stored bases (i.e. '*' where the sequence should be), instead of failing with an error", required = false, exclusiveOf = "", validation = "")
   var filter_bases_not_stored: Boolean = config("filter_bases_not_stored", default = false)
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
   override def beforeGraph() {
     super.beforeGraph()
     deps ++= variant.filter(orig => orig != null && (!orig.getName.endsWith(".list"))).map(orig => VcfUtils.getVcfIndexFile(orig))
     if (out != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(out))
-      if (!org.broadinstitute.gatk.utils.commandline.ArgumentTypeDescriptor.isCompressed(out.getPath))
-        outputFiles :+= VcfUtils.getVcfIndexFile(out)
+      outputIndex = VcfUtils.getVcfIndexFile(out)
   }
 
   override def cmdLine = super.cmdLine +

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/ContigScatterFunction.scala

@@ -5,8 +5,8 @@ import org.broadinstitute.gatk.utils.interval.IntervalUtils
 import org.broadinstitute.gatk.queue.function.InProcessFunction
 
 /**
-  * Splits intervals by contig instead of evenly.
-  */
+ * Splits intervals by contig instead of evenly.
+ */
 class ContigScatterFunction extends GATKScatterFunction with InProcessFunction {
 
   override def scatterCount = if (intervalFilesExist) super.scatterCount min this.maxIntervals else super.scatterCount

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/GenotypeGVCFs.scala

@@ -8,7 +8,7 @@ package nl.lumc.sasc.biopet.extensions.gatk.broad
 import java.io.File
 
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.queue.extensions.gatk.{ CatVariantsGatherer, TaggedFile }
+import org.broadinstitute.gatk.queue.extensions.gatk.TaggedFile
 import nl.lumc.sasc.biopet.core.ScatterGatherableFunction
 import nl.lumc.sasc.biopet.utils.VcfUtils
 import org.broadinstitute.gatk.utils.commandline.{ Argument, Gather, Output, _ }
@@ -107,11 +107,15 @@ class GenotypeGVCFs(val root: Configurable) extends CommandLineGATK with Scatter
   @Argument(fullName = "filter_bases_not_stored", shortName = "filterNoBases", doc = "Filter out reads with no stored bases (i.e. '*' where the sequence should be), instead of failing with an error", required = false, exclusiveOf = "", validation = "")
   var filter_bases_not_stored: Boolean = config("filter_bases_not_stored", default = false)
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
   override def freezeFieldValues() {
     super.freezeFieldValues()
     deps ++= variant.filter(orig => orig != null && (!orig.getName.endsWith(".list"))).map(orig => VcfUtils.getVcfIndexFile(orig))
     if (out != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(out))
-      outputFiles :+= VcfUtils.getVcfIndexFile(out)
+      outputIndex = VcfUtils.getVcfIndexFile(out)
     dbsnp.foreach(VcfUtils.getVcfIndexFile(_))
   }
 

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/HaplotypeCaller.scala

@@ -8,10 +8,10 @@ package nl.lumc.sasc.biopet.extensions.gatk.broad
 import java.io.File
 
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.queue.extensions.gatk._
 import nl.lumc.sasc.biopet.core.ScatterGatherableFunction
 import nl.lumc.sasc.biopet.utils.VcfUtils
-import org.broadinstitute.gatk.utils.commandline.{ Argument, Gather, Input, _ }
+import org.broadinstitute.gatk.queue.extensions.gatk.TaggedFile
+import org.broadinstitute.gatk.utils.commandline.{ Argument, Gather, Input, Output }
 
 class HaplotypeCaller(val root: Configurable) extends CommandLineGATK with ScatterGatherableFunction {
   def analysis_type = "HaplotypeCaller"
@@ -396,15 +396,23 @@ class HaplotypeCaller(val root: Configurable) extends CommandLineGATK with Scatt
   @Argument(fullName = "filter_bases_not_stored", shortName = "filterNoBases", doc = "Filter out reads with no stored bases (i.e. '*' where the sequence should be), instead of failing with an error", required = false, exclusiveOf = "", validation = "")
   var filter_bases_not_stored: Boolean = config("filter_bases_not_stored", default = false)
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
+  @Output
+  @Gather(enabled = false)
+  private var outputBamIndex: File = _
+
   override def beforeGraph() {
     super.beforeGraph()
     if (out != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(out))
-      outputFiles :+= VcfUtils.getVcfIndexFile(out)
+      outputIndex = VcfUtils.getVcfIndexFile(out)
     dbsnp.foreach(deps :+= VcfUtils.getVcfIndexFile(_))
     deps ++= comp.filter(orig => orig != null && (!orig.getName.endsWith(".list"))).map(orig => new File(orig + ".idx"))
     if (bamOutput != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(bamOutput))
       if (!disable_bam_indexing)
-        outputFiles :+= new File(bamOutput.getPath.stripSuffix(".bam") + ".bai")
+        outputBamIndex = new File(bamOutput.getPath.stripSuffix(".bam") + ".bai")
     if (bamOutput != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(bamOutput))
       if (generate_md5)
         bamOutputMD5 = new File(bamOutput.getPath + ".md5")

biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/broad/IndelRealigner.scala

@@ -8,7 +8,7 @@ package nl.lumc.sasc.biopet.extensions.gatk.broad
 import java.io.File
 
 import nl.lumc.sasc.biopet.utils.config.Configurable
-import org.broadinstitute.gatk.queue.extensions.gatk.{ BamGatherFunction, TaggedFile }
+import org.broadinstitute.gatk.queue.extensions.gatk.TaggedFile
 import nl.lumc.sasc.biopet.core.ScatterGatherableFunction
 import nl.lumc.sasc.biopet.utils.VcfUtils
 import org.broadinstitute.gatk.utils.commandline.{ Argument, Gather, Output, _ }
@@ -131,12 +131,16 @@ class IndelRealigner(val root: Configurable) extends CommandLineGATK with Scatte
   @Argument(fullName = "filter_bases_not_stored", shortName = "filterNoBases", doc = "Filter out reads with no stored bases (i.e. '*' where the sequence should be), instead of failing with an error", required = false, exclusiveOf = "", validation = "")
   var filter_bases_not_stored: Boolean = config("filter_bases_not_stored", default = false)
 
+  @Output
+  @Gather(enabled = false)
+  private var outputIndex: File = _
+
   override def beforeGraph() {
     super.beforeGraph()
     deps ++= knownAlleles.filter(orig => orig != null && (!orig.getName.endsWith(".list"))).map(orig => VcfUtils.getVcfIndexFile(orig))
     if (out != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(out))
       if (!disable_bam_indexing)
-        outputFiles :+= new File(out.getPath.stripSuffix(".bam") + ".bai")
+        outputIndex = new File(out.getPath.stripSuffix(".bam") + ".bai")
     if (out != null && !org.broadinstitute.gatk.utils.io.IOUtils.isSpecialFile(out))
       if (generate_md5)
         outMD5 = new File(out.getPath + ".md5")