Merge branch 'feature-unittest_core' into 'develop'

Flexiprep unit test + fix for small bugs

See subject

See merge request !94

log4j.properties

+#
+# Biopet is built on top of GATK Queue for building bioinformatic
+# pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+# SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+# should also be able to execute Biopet tools and pipelines.
+#
+# Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+#
+# Contact us at: sasc@lumc.nl
+#
+# A dual licensing mode is applied. The source code within this project that are
+# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# license; For commercial users or users who do not want to follow the AGPL
+# license, please contact us to obtain a separate license.
+#
+
+# Set root logger level to DEBUG and its only appender to A1.
+log4j.rootLogger=ERROR, A1
+
+# A1 is set to be a ConsoleAppender.
+log4j.appender.A1=org.apache.log4j.ConsoleAppender
+
+# A1 uses PatternLayout.
+log4j.appender.A1.layout=org.apache.log4j.PatternLayout
+log4j.appender.A1.layout.ConversionPattern=%-5p [%d] [%C{1}] - %m%n
\ No newline at end of file

protected/basty/src/main/scala/nl/lumc/sasc/biopet/pipelines/basty/Basty.scala

@@ -60,28 +60,34 @@ class Basty(val root: Configurable) extends QScript with MultiSampleQScript {
   }
 
   def addMultiSampleJobs(): Unit = {
-    val refVariants = addGenerateFasta(null, outputDir + "reference/", outputName = "reference")
-    val refVariantSnps = addGenerateFasta(null, outputDir + "reference/", outputName = "reference", snpsOnly = true)
+    val refVariants = addGenerateFasta(null, new File(outputDir, "reference"), outputName = "reference")
+    val refVariantSnps = addGenerateFasta(null, new File(outputDir, "reference"), outputName = "reference", snpsOnly = true)
 
-    val catVariants = Cat(this, refVariants.variants :: samples.map(_._2.output.variants).toList, outputDir + "fastas/variant.fasta")
+    val catVariants = Cat(this, refVariants.variants :: samples.map(_._2.output.variants).toList,
+      new File(outputDir, "fastas" + File.separator + "variant.fasta"))
     add(catVariants)
-    val catVariantsSnps = Cat(this, refVariantSnps.variants :: samples.map(_._2.outputSnps.variants).toList, outputDir + "fastas/variant.snps_only.fasta")
+    val catVariantsSnps = Cat(this, refVariantSnps.variants :: samples.map(_._2.outputSnps.variants).toList,
+      new File(outputDir, "fastas" + File.separator + "variant.snps_only.fasta"))
     add(catVariantsSnps)
 
-    val catConsensus = Cat(this, refVariants.consensus :: samples.map(_._2.output.consensus).toList, outputDir + "fastas/consensus.fasta")
+    val catConsensus = Cat(this, refVariants.consensus :: samples.map(_._2.output.consensus).toList,
+      new File(outputDir, "fastas" + File.separator + "consensus.fasta"))
     add(catConsensus)
-    val catConsensusSnps = Cat(this, refVariantSnps.consensus :: samples.map(_._2.outputSnps.consensus).toList, outputDir + "fastas/consensus.snps_only.fasta")
+    val catConsensusSnps = Cat(this, refVariantSnps.consensus :: samples.map(_._2.outputSnps.consensus).toList,
+      new File(outputDir, "fastas" + File.separator + "consensus.snps_only.fasta"))
     add(catConsensusSnps)
 
-    val catConsensusVariants = Cat(this, refVariants.consensusVariants :: samples.map(_._2.output.consensusVariants).toList, outputDir + "fastas/consensus.variant.fasta")
+    val catConsensusVariants = Cat(this, refVariants.consensusVariants :: samples.map(_._2.output.consensusVariants).toList,
+      new File(outputDir, "fastas" + File.separator + "consensus.variant.fasta"))
     add(catConsensusVariants)
-    val catConsensusVariantsSnps = Cat(this, refVariantSnps.consensusVariants :: samples.map(_._2.outputSnps.consensusVariants).toList, outputDir + "fastas/consensus.variant.snps_only.fasta")
+    val catConsensusVariantsSnps = Cat(this, refVariantSnps.consensusVariants :: samples.map(_._2.outputSnps.consensusVariants).toList,
+      new File(outputDir, "fastas" + File.separator + "consensus.variant.snps_only.fasta"))
     add(catConsensusVariantsSnps)
 
     val seed: Int = config("seed", default = 12345)
-    def addTreeJobs(variants: File, concensusVariants: File, outputDir: String, outputName: String) {
-      val dirSufixRaxml = if (outputDir.endsWith(File.separator)) "raxml" else File.separator + "raxml"
-      val dirSufixGubbins = if (outputDir.endsWith(File.separator)) "gubbins" else File.separator + "gubbins"
+    def addTreeJobs(variants: File, concensusVariants: File, outputDir: File, outputName: String) {
+      val dirSufixRaxml = new File(outputDir, "raxml")
+      val dirSufixGubbins = new File(outputDir, "gubbins")
 
       val raxmlMl = new Raxml(this)
       raxmlMl.input = variants
@@ -101,7 +107,7 @@ class Basty(val root: Configurable) extends QScript with MultiSampleQScript {
         raxmlBoot.m = config("raxml_ml_model", default = "GTRGAMMAX")
         raxmlBoot.p = seed
         raxmlBoot.b = math.abs(r.nextInt)
-        raxmlBoot.w = outputDir + dirSufixRaxml
+        raxmlBoot.w = dirSufixRaxml
         raxmlBoot.N = 1
         raxmlBoot.n = outputName + "_boot_" + t
         add(raxmlBoot)
@@ -125,16 +131,18 @@ class Basty(val root: Configurable) extends QScript with MultiSampleQScript {
       val gubbins = new RunGubbins(this)
       gubbins.fastafile = concensusVariants
       gubbins.startingTree = Some(raxmlBi.getBipartitionsFile)
-      gubbins.outputDirectory = outputDir + dirSufixGubbins
+      gubbins.outputDirectory = dirSufixGubbins
       add(gubbins)
     }
 
-    addTreeJobs(catVariantsSnps.output, catConsensusVariantsSnps.output, outputDir + "trees" + File.separator + "snps_only", "snps_only")
-    addTreeJobs(catVariants.output, catConsensusVariants.output, outputDir + "trees" + File.separator + "snps_indels", "snps_indels")
+    addTreeJobs(catVariantsSnps.output, catConsensusVariantsSnps.output,
+      new File(outputDir, "trees" + File.separator + "snps_only"), "snps_only")
+    addTreeJobs(catVariants.output, catConsensusVariants.output,
+      new File(outputDir, "trees" + File.separator + "snps_indels"), "snps_indels")
 
   }
 
-  def addGenerateFasta(sampleName: String, outputDir: String, outputName: String = null,
+  def addGenerateFasta(sampleName: String, outputDir: File, outputName: String = null,
                        snpsOnly: Boolean = false): FastaOutput = {
     val bastyGenerateFasta = new BastyGenerateFasta(this)
     bastyGenerateFasta.outputName = if (outputName != null) outputName else sampleName

protected/basty/src/test/resources/log4j.properties

+#
+# Biopet is built on top of GATK Queue for building bioinformatic
+# pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+# SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+# should also be able to execute Biopet tools and pipelines.
+#
+# Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+#
+# Contact us at: sasc@lumc.nl
+#
+# A dual licensing mode is applied. The source code within this project that are
+# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# license; For commercial users or users who do not want to follow the AGPL
+# license, please contact us to obtain a separate license.
+#
+
+# Set root logger level to DEBUG and its only appender to A1.
+log4j.rootLogger=ERROR, A1
+
+# A1 is set to be a ConsoleAppender.
+log4j.appender.A1=org.apache.log4j.ConsoleAppender
+
+# A1 uses PatternLayout.
+log4j.appender.A1.layout=org.apache.log4j.PatternLayout
+log4j.appender.A1.layout.ConversionPattern=%-5p [%d] [%C{1}] - %m%n
\ No newline at end of file

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

@@ -11,8 +11,8 @@ import nl.lumc.sasc.biopet.core.config.Configurable
 class ApplyRecalibration(val root: Configurable) extends org.broadinstitute.gatk.queue.extensions.gatk.ApplyRecalibration with GatkGeneral {
   scatterCount = config("scattercount", default = 0)
 
-  override def afterGraph {
-    super.afterGraph
+  override def beforeGraph {
+    super.beforeGraph
 
     nt = Option(getThreads(3))
     memoryLimit = Option(nt.getOrElse(1) * 2)

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

@@ -22,7 +22,7 @@ object BaseRecalibrator {
     val br = new BaseRecalibrator(root)
     br.input_file :+= input
     br.out = output
-    br.afterGraph
+    br.beforeGraph
     return br
   }
 }
\ No newline at end of file

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

@@ -35,13 +35,13 @@ class HaplotypeCaller(val root: Configurable) extends org.broadinstitute.gatk.qu
     stand_emit_conf = config("stand_emit_conf", default = 0)
   }
 
-  override def afterGraph {
-    super.afterGraph
+  override def beforeGraph {
+    super.beforeGraph
     if (bamOutput != null && nct.getOrElse(1) > 1) {
       threads = 1
       logger.warn("BamOutput is on, nct/threads is forced to set on 1, this option is only for debug")
     }
-    nct = Some(threads)
+    nct = Some(getThreads(1))
     memoryLimit = Option(memoryLimit.getOrElse(2.0) * nct.getOrElse(1))
   }
 

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

@@ -13,7 +13,7 @@ class IndelRealigner(val root: Configurable) extends org.broadinstitute.gatk.que
 }
 
 object IndelRealigner {
-  def apply(root: Configurable, input: File, targetIntervals: File, outputDir: String): IndelRealigner = {
+  def apply(root: Configurable, input: File, targetIntervals: File, outputDir: File): IndelRealigner = {
     val ir = new IndelRealigner(root)
     ir.input_file :+= input
     ir.targetIntervals = targetIntervals

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

@@ -18,7 +18,7 @@ class RealignerTargetCreator(val root: Configurable) extends org.broadinstitute.
 }
 
 object RealignerTargetCreator {
-  def apply(root: Configurable, input: File, outputDir: String): RealignerTargetCreator = {
+  def apply(root: Configurable, input: File, outputDir: File): RealignerTargetCreator = {
     val re = new RealignerTargetCreator(root)
     re.input_file :+= input
     re.out = new File(outputDir, input.getName.stripSuffix(".bam") + ".realign.intervals")

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

@@ -8,14 +8,14 @@ package nl.lumc.sasc.biopet.extensions.gatk
 import nl.lumc.sasc.biopet.core.config.Configurable
 
 class UnifiedGenotyper(val root: Configurable) extends org.broadinstitute.gatk.queue.extensions.gatk.UnifiedGenotyper with GatkGeneral {
-  override def afterGraph {
-    super.afterGraph
+  override def beforeGraph {
+    super.beforeGraph
 
     genotype_likelihoods_model = org.broadinstitute.gatk.tools.walkers.genotyper.GenotypeLikelihoodsCalculationModel.Model.BOTH
     if (config.contains("scattercount")) scatterCount = config("scattercount")
     if (config.contains("dbsnp")) this.dbsnp = config("dbsnp")
     this.sample_ploidy = config("ploidy")
-    nct = config("threads", default = 1)
+    nct = Some(getThreads(1))
     memoryLimit = Option(nct.getOrElse(1) * 2)
     if (config.contains("allSitePLs")) this.allSitePLs = config("allSitePLs")
     if (config.contains("output_mode")) {

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

@@ -9,8 +9,8 @@ import java.io.File
 import nl.lumc.sasc.biopet.core.config.Configurable
 
 class VariantEval(val root: Configurable) extends org.broadinstitute.gatk.queue.extensions.gatk.VariantEval with GatkGeneral {
-  override def afterGraph {
-    super.afterGraph
+  override def beforeGraph {
+    super.beforeGraph
   }
 }
 
@@ -21,7 +21,7 @@ object VariantEval {
     vareval.eval = Seq(sample)
     vareval.comp = Seq(compareWith)
     vareval.out = output
-    vareval.afterGraph
+    vareval.beforeGraph
     return vareval
   }
 
@@ -35,7 +35,7 @@ object VariantEval {
     vareval.ST = ST
     vareval.noEV = true
     vareval.EV = EV
-    vareval.afterGraph
+    vareval.beforeGraph
     return vareval
   }
 

protected/biopet-gatk-extensions/src/test/resources/log4j.properties

+#
+# Biopet is built on top of GATK Queue for building bioinformatic
+# pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+# SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+# should also be able to execute Biopet tools and pipelines.
+#
+# Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+#
+# Contact us at: sasc@lumc.nl
+#
+# A dual licensing mode is applied. The source code within this project that are
+# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# license; For commercial users or users who do not want to follow the AGPL
+# license, please contact us to obtain a separate license.
+#
+
+# Set root logger level to DEBUG and its only appender to A1.
+log4j.rootLogger=ERROR, A1
+
+# A1 is set to be a ConsoleAppender.
+log4j.appender.A1=org.apache.log4j.ConsoleAppender
+
+# A1 uses PatternLayout.
+log4j.appender.A1.layout=org.apache.log4j.PatternLayout
+log4j.appender.A1.layout.ConversionPattern=%-5p [%d] [%C{1}] - %m%n
\ No newline at end of file

protected/biopet-gatk-pipelines/src/main/scala/nl/lumc/sasc/biopet/pipelines/gatk/GatkBenchmarkGenotyping.scala

@@ -29,11 +29,8 @@ class GatkBenchmarkGenotyping(val root: Configurable) extends QScript with Biope
   var dbsnp: File = config("dbsnp")
 
   def init() {
-    if (config.contains("gvcffiles")) for (file <- config("gvcffiles").asList) {
+    if (config.contains("gvcffiles")) for (file <- config("gvcffiles").asList)
       gvcfFiles ::= file.toString
-    }
-    if (outputDir == null) throw new IllegalStateException("Missing Output directory on gatk module")
-    else if (!outputDir.endsWith("/")) outputDir += "/"
   }
 
   def biopetScript() {
@@ -53,7 +50,7 @@ class GatkBenchmarkGenotyping(val root: Configurable) extends QScript with Biope
     val gatkGenotyping = new GatkGenotyping(this)
     gatkGenotyping.inputGvcfs = sampleGvcf :: gvcfPool
     gatkGenotyping.samples :+= sampleName
-    gatkGenotyping.outputDir = outputDir + "samples_" + gvcfPool.size + "/"
+    gatkGenotyping.outputDir = new File(outputDir, "samples_" + gvcfPool.size)
     gatkGenotyping.init
     gatkGenotyping.biopetScript
     addAll(gatkGenotyping.functions)

protected/biopet-gatk-pipelines/src/main/scala/nl/lumc/sasc/biopet/pipelines/gatk/GatkGenotyping.scala

@@ -33,16 +33,15 @@ class GatkGenotyping(val root: Configurable) extends QScript with BiopetQScript
   var samples: List[String] = Nil
 
   def init() {
-    if (outputFile == null) outputFile = outputDir + outputName + ".vcf.gz"
-    if (outputDir == null) throw new IllegalStateException("Missing Output directory on gatk module")
-    else if (!outputDir.endsWith("/")) outputDir += "/"
+    require(outputName != null, "Outputname is null")
+    if (outputFile == null) outputFile = new File(outputDir, outputName + ".vcf.gz")
   }
 
   def biopetScript() {
     addGenotypeGVCFs(inputGvcfs, outputFile)
     if (!samples.isEmpty) {
-      if (samples.size > 1) addSelectVariants(outputFile, samples, outputDir + "samples/", "all")
-      for (sample <- samples) addSelectVariants(outputFile, List(sample), outputDir + "samples/", sample)
+      if (samples.size > 1) addSelectVariants(outputFile, samples, new File(outputDir, "samples/"), "all")
+      for (sample <- samples) addSelectVariants(outputFile, List(sample), new File(outputDir, "samples/"), sample)
     }
   }
 
@@ -52,8 +51,8 @@ class GatkGenotyping(val root: Configurable) extends QScript with BiopetQScript
     return genotypeGVCFs.out
   }
 
-  def addSelectVariants(inputFile: File, samples: List[String], outputDir: String, name: String) {
-    val selectVariants = SelectVariants(this, inputFile, outputDir + name + ".vcf.gz")
+  def addSelectVariants(inputFile: File, samples: List[String], outputDir: File, name: String) {
+    val selectVariants = SelectVariants(this, inputFile, new File(outputDir, name + ".vcf.gz"))
     selectVariants.excludeNonVariants = true
     for (sample <- samples) selectVariants.sample_name :+= sample
     add(selectVariants)

protected/biopet-gatk-pipelines/src/main/scala/nl/lumc/sasc/biopet/pipelines/gatk/GatkPipeline.scala

@@ -47,13 +47,13 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
       val mapping = new Mapping(qscript)
       mapping.sampleId = sampleId
       mapping.libId = libId
-      mapping.outputDir = libDir + "/variantcalling/"
+      mapping.outputDir = libDir
 
       /** Library variantcalling */
       val gatkVariantcalling = new GatkVariantcalling(qscript)
       gatkVariantcalling.doublePreProces = false
       gatkVariantcalling.sampleID = sampleId
-      gatkVariantcalling.outputDir = libDir
+      gatkVariantcalling.outputDir = new File(libDir, "variantcalling")
 
       protected def addJobs(): Unit = {
         val bamFile: Option[File] = if (config.contains("R1")) {
@@ -124,7 +124,7 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
     /** sample variantcalling */
     val gatkVariantcalling = new GatkVariantcalling(qscript)
     gatkVariantcalling.sampleID = sampleId
-    gatkVariantcalling.outputDir = sampleDir + "/variantcalling/"
+    gatkVariantcalling.outputDir = new File(sampleDir, "variantcalling/")
 
     protected def addJobs(): Unit = {
       addPerLibJobs()
@@ -143,8 +143,6 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
   }
 
   def init() {
-    if (outputDir == null) throw new IllegalStateException("Missing Output directory on gatk module")
-    else if (!outputDir.endsWith("/")) outputDir += "/"
   }
 
   val multisampleVariantcalling = new GatkVariantcalling(this) {
@@ -158,7 +156,7 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
 
   def addMultiSampleJobs(): Unit = {
     val gvcfFiles: List[File] = if (mergeGvcfs && externalGvcfs.size + samples.size > 1) {
-      val newFile = outputDir + "merged.gvcf.vcf.gz"
+      val newFile = new File(outputDir, "merged.gvcf.vcf.gz")
       add(CombineGVCFs(this, externalGvcfs ++ samples.map(_._2.gatkVariantcalling.scriptOutput.gvcfFile), newFile))
       List(newFile)
     } else externalGvcfs ++ samples.map(_._2.gatkVariantcalling.scriptOutput.gvcfFile)
@@ -167,7 +165,7 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
       if (jointGenotyping) {
         val gatkGenotyping = new GatkGenotyping(this)
         gatkGenotyping.inputGvcfs = gvcfFiles
-        gatkGenotyping.outputDir = outputDir + "genotyping/"
+        gatkGenotyping.outputDir = new File(outputDir, "genotyping")
         gatkGenotyping.init
         gatkGenotyping.biopetScript
         addAll(gatkGenotyping.functions)
@@ -185,7 +183,7 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
       }
 
       if (gatkVariantcalling.useMpileup) {
-        val cvRaw = CombineVariants(this, allRawVcfFiles.toList, outputDir + "variantcalling/multisample.raw.vcf.gz")
+        val cvRaw = CombineVariants(this, allRawVcfFiles.toList, new File(outputDir, "variantcalling/multisample.raw.vcf.gz"))
         add(cvRaw)
         gatkVariantcalling.rawVcfInput = cvRaw.out
       }
@@ -193,7 +191,7 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
       multisampleVariantcalling.preProcesBams = false
       multisampleVariantcalling.doublePreProces = false
       multisampleVariantcalling.inputBams = allBamfiles.toList
-      multisampleVariantcalling.outputDir = outputDir + "variantcalling"
+      multisampleVariantcalling.outputDir = new File(outputDir, "variantcalling")
       multisampleVariantcalling.outputName = "multisample"
       multisampleVariantcalling.init
       multisampleVariantcalling.biopetScript
@@ -203,7 +201,7 @@ class GatkPipeline(val root: Configurable) extends QScript with MultiSampleQScri
         val recalibration = new GatkVariantRecalibration(this)
         recalibration.inputVcf = multisampleVariantcalling.scriptOutput.finalVcfFile
         recalibration.bamFiles = allBamfiles
-        recalibration.outputDir = outputDir + "recalibration/"
+        recalibration.outputDir = new File(outputDir, "recalibration")
         recalibration.init
         recalibration.biopetScript
       }

protected/biopet-gatk-pipelines/src/main/scala/nl/lumc/sasc/biopet/pipelines/gatk/GatkVariantRecalibration.scala

@@ -26,9 +26,7 @@ class GatkVariantRecalibration(val root: Configurable) extends QScript with Biop
   var outputVcf: File = _
 
   def init() {
-    if (inputVcf == null) throw new IllegalStateException("Missing Output directory on gatk module")
-    if (outputDir == null) throw new IllegalStateException("Missing Output directory on gatk module")
-    else if (!outputDir.endsWith("/")) outputDir += "/"
+    require(inputVcf != null, "Missing Output directory on gatk module")
   }
 
   def biopetScript() {
@@ -37,7 +35,7 @@ class GatkVariantRecalibration(val root: Configurable) extends QScript with Biop
     vcfFile = addIndelVariantRecalibrator(vcfFile, outputDir)
   }
 
-  def addSnpVariantRecalibrator(inputVcf: File, dir: String): File = {
+  def addSnpVariantRecalibrator(inputVcf: File, dir: File): File = {
     val snpRecal = VariantRecalibrator(this, inputVcf, swapExt(dir, inputVcf, ".vcf", ".indel.recal"),
       swapExt(dir, inputVcf, ".vcf", ".indel.tranches"), indel = false)
     if (!snpRecal.resource.isEmpty) {
@@ -54,7 +52,7 @@ class GatkVariantRecalibration(val root: Configurable) extends QScript with Biop
     }
   }
 
-  def addIndelVariantRecalibrator(inputVcf: File, dir: String): File = {
+  def addIndelVariantRecalibrator(inputVcf: File, dir: File): File = {
     val indelRecal = VariantRecalibrator(this, inputVcf, swapExt(dir, inputVcf, ".vcf", ".indel.recal"),
       swapExt(dir, inputVcf, ".vcf", ".indel.tranches"), indel = true)
     if (!indelRecal.resource.isEmpty) {
@@ -71,7 +69,7 @@ class GatkVariantRecalibration(val root: Configurable) extends QScript with Biop
     }
   }
 
-  def addVariantAnnotator(inputvcf: File, bamfiles: List[File], dir: String): File = {
+  def addVariantAnnotator(inputvcf: File, bamfiles: List[File], dir: File): File = {
     val variantAnnotator = VariantAnnotator(this, inputvcf, bamfiles, swapExt(dir, inputvcf, ".vcf", ".anotated.vcf"))
     add(variantAnnotator)
     return variantAnnotator.out

protected/biopet-gatk-pipelines/src/main/scala/nl/lumc/sasc/biopet/pipelines/gatk/GatkVariantcalling.scala

@@ -52,8 +52,6 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
   def init() {
     if (outputName == null && sampleID != null) outputName = sampleID
     else if (outputName == null) outputName = config("output_name", default = "noname")
-    if (outputDir == null) throw new IllegalStateException("Missing Output directory on gatk module")
-    else if (!outputDir.endsWith("/")) outputDir += "/"
 
     val baseRecalibrator = new BaseRecalibrator(this)
     if (preProcesBams && useBaseRecalibration && baseRecalibrator.knownSites.isEmpty) {
@@ -66,7 +64,7 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
     if (files.isEmpty) throw new IllegalStateException("Files can't be empty")
     else if (!doublePreProces) files
     else if (files.size == 1) {
-      val bamFile: File = outputDir + files.head.getName
+      val bamFile = new File(outputDir, files.head.getName)
       if (bamFile != files.head) {
         val oldIndex: File = files.head.getAbsolutePath.stripSuffix(".bam") + ".bai"
         val newIndex: File = bamFile.getAbsolutePath.stripSuffix(".bam") + ".bai"
@@ -113,20 +111,20 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
         val hcGvcf = new HaplotypeCaller(this)
         hcGvcf.useGvcf
         hcGvcf.input_file = scriptOutput.bamFiles
-        hcGvcf.out = outputDir + outputName + ".hc.discovery.gvcf.vcf.gz"
+        hcGvcf.out = new File(outputDir, outputName + ".hc.discovery.gvcf.vcf.gz")
         add(hcGvcf)
         scriptOutput.gvcfFile = hcGvcf.out
       }
 
       if (useHaplotypecaller) {
         if (sampleID != null) {
-          val genotypeGVCFs = GenotypeGVCFs(this, List(scriptOutput.gvcfFile), outputDir + outputName + ".hc.discovery.vcf.gz")
+          val genotypeGVCFs = GenotypeGVCFs(this, List(scriptOutput.gvcfFile), new File(outputDir, outputName + ".hc.discovery.vcf.gz"))
           add(genotypeGVCFs)
           scriptOutput.hcVcfFile = genotypeGVCFs.out
         } else {
           val hcGvcf = new HaplotypeCaller(this)
           hcGvcf.input_file = scriptOutput.bamFiles
-          hcGvcf.out = outputDir + outputName + ".hc.discovery.vcf.gz"
+          hcGvcf.out = new File(outputDir, outputName + ".hc.discovery.vcf.gz")
           add(hcGvcf)
           scriptOutput.hcVcfFile = hcGvcf.out
         }
@@ -136,7 +134,7 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
       if (useUnifiedGenotyper) {
         val ugVcf = new UnifiedGenotyper(this)
         ugVcf.input_file = scriptOutput.bamFiles
-        ugVcf.out = outputDir + outputName + ".ug.discovery.vcf.gz"
+        ugVcf.out = new File(outputDir, outputName + ".ug.discovery.vcf.gz")
         add(ugVcf)
         scriptOutput.ugVcfFile = ugVcf.out
         mergBuffer += ("2.UG-Discovery" -> scriptOutput.ugVcfFile)
@@ -148,7 +146,7 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
           val m2v = new MpileupToVcf(this)
           m2v.inputBam = scriptOutput.bamFiles.head
           m2v.sample = sampleID
-          m2v.output = outputDir + outputName + ".raw.vcf"
+          m2v.output = new File(outputDir, outputName + ".raw.vcf")
           add(m2v)
           scriptOutput.rawVcfFile = m2v.output
 
@@ -160,7 +158,7 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
             ), super.defaults)
           }
           vcfFilter.inputVcf = m2v.output
-          vcfFilter.outputVcf = this.swapExt(outputDir, m2v.output, ".vcf", ".filter.vcf.gz")
+          vcfFilter.outputVcf = swapExt(outputDir, m2v.output, ".vcf", ".filter.vcf.gz")
           add(vcfFilter)
           scriptOutput.rawFilterVcfFile = vcfFilter.outputVcf
         } else if (rawVcfInput != null) scriptOutput.rawFilterVcfFile = rawVcfInput
@@ -176,7 +174,7 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
         if (useHaplotypecaller) {
           val hcAlleles = new HaplotypeCaller(this)
           hcAlleles.input_file = scriptOutput.bamFiles
-          hcAlleles.out = outputDir + outputName + ".hc.allele.vcf.gz"
+          hcAlleles.out = new File(outputDir, outputName + ".hc.allele.vcf.gz")
           hcAlleles.alleles = mergeAlleles.output
           hcAlleles.genotyping_mode = org.broadinstitute.gatk.tools.walkers.genotyper.GenotypingOutputMode.GENOTYPE_GIVEN_ALLELES
           add(hcAlleles)
@@ -187,7 +185,7 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
         if (useUnifiedGenotyper) {
           val ugAlleles = new UnifiedGenotyper(this)
           ugAlleles.input_file = scriptOutput.bamFiles
-          ugAlleles.out = outputDir + outputName + ".ug.allele.vcf.gz"
+          ugAlleles.out = new File(outputDir, outputName + ".ug.allele.vcf.gz")
           ugAlleles.alleles = mergeAlleles.output
           ugAlleles.genotyping_mode = org.broadinstitute.gatk.tools.walkers.genotyper.GenotypingOutputMode.GENOTYPE_GIVEN_ALLELES
           add(ugAlleles)
@@ -206,32 +204,32 @@ class GatkVariantcalling(val root: Configurable) extends QScript with BiopetQScr
         sv.out
       }
 
-      val cvFinal = CombineVariants(this, mergeList.toList, outputDir + outputName + ".final.vcf.gz")
+      val cvFinal = CombineVariants(this, mergeList.toList, new File(outputDir, outputName + ".final.vcf.gz"))
       cvFinal.genotypemergeoption = org.broadinstitute.gatk.utils.variant.GATKVariantContextUtils.GenotypeMergeType.UNSORTED
       add(cvFinal)
 
       val vcfStats = new VcfStats(this)
       vcfStats.input = cvFinal.out
-      vcfStats.setOutputDir(outputDir + File.separator + "vcfstats")
+      vcfStats.setOutputDir(new File(outputDir, "vcfstats"))
       add(vcfStats)
 
       scriptOutput.finalVcfFile = cvFinal.out
     }
   }
 
-  def addIndelRealign(inputBam: File, dir: String, isIntermediate: Boolean = true): File = {
+  def addIndelRealign(inputBam: File, dir: File, isIntermediate: Boolean = true): File = {
     val realignerTargetCreator = RealignerTargetCreator(this, inputBam, dir)
     realignerTargetCreator.isIntermediate = true
     add(realignerTargetCreator)
 
-    val indelRealigner = IndelRealigner.apply(this, inputBam, realignerTargetCreator.out, dir)
+    val indelRealigner = IndelRealigner(this, inputBam, realignerTargetCreator.out, dir)
     indelRealigner.isIntermediate = isIntermediate
     add(indelRealigner)
 
     return indelRealigner.o
   }
 
-  def addBaseRecalibrator(inputBam: File, dir: String, isIntermediate: Boolean = false): File = {
+  def addBaseRecalibrator(inputBam: File, dir: File, isIntermediate: Boolean = false): File = {
     val baseRecalibrator = BaseRecalibrator(this, inputBam, swapExt(dir, inputBam, ".bam", ".baserecal"))
 
     if (baseRecalibrator.knownSites.isEmpty) {

protected/biopet-gatk-pipelines/src/test/resources/log4j.properties

+#
+# Biopet is built on top of GATK Queue for building bioinformatic
+# pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+# SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+# should also be able to execute Biopet tools and pipelines.
+#
+# Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+#
+# Contact us at: sasc@lumc.nl
+#
+# A dual licensing mode is applied. The source code within this project that are
+# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# license; For commercial users or users who do not want to follow the AGPL
+# license, please contact us to obtain a separate license.
+#
+
+# Set root logger level to DEBUG and its only appender to A1.
+log4j.rootLogger=ERROR, A1
+
+# A1 is set to be a ConsoleAppender.
+log4j.appender.A1=org.apache.log4j.ConsoleAppender
+
+# A1 uses PatternLayout.
+log4j.appender.A1.layout=org.apache.log4j.PatternLayout
+log4j.appender.A1.layout.ConversionPattern=%-5p [%d] [%C{1}] - %m%n
\ No newline at end of file

protected/biopet-protected-package/src/test/resources/log4j.properties

+#
+# Biopet is built on top of GATK Queue for building bioinformatic
+# pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+# SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+# should also be able to execute Biopet tools and pipelines.
+#
+# Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+#
+# Contact us at: sasc@lumc.nl
+#
+# A dual licensing mode is applied. The source code within this project that are
+# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# license; For commercial users or users who do not want to follow the AGPL
+# license, please contact us to obtain a separate license.
+#
+
+# Set root logger level to DEBUG and its only appender to A1.
+log4j.rootLogger=ERROR, A1
+
+# A1 is set to be a ConsoleAppender.
+log4j.appender.A1=org.apache.log4j.ConsoleAppender
+
+# A1 uses PatternLayout.
+log4j.appender.A1.layout=org.apache.log4j.PatternLayout
+log4j.appender.A1.layout.ConversionPattern=%-5p [%d] [%C{1}] - %m%n
\ No newline at end of file

protected/log4j.properties

+#
+# Biopet is built on top of GATK Queue for building bioinformatic
+# pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+# SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+# should also be able to execute Biopet tools and pipelines.
+#
+# Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+#
+# Contact us at: sasc@lumc.nl
+#
+# A dual licensing mode is applied. The source code within this project that are
+# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# license; For commercial users or users who do not want to follow the AGPL
+# license, please contact us to obtain a separate license.
+#
+
+# Set root logger level to DEBUG and its only appender to A1.
+log4j.rootLogger=ERROR, A1
+
+# A1 is set to be a ConsoleAppender.
+log4j.appender.A1=org.apache.log4j.ConsoleAppender
+
+# A1 uses PatternLayout.
+log4j.appender.A1.layout=org.apache.log4j.PatternLayout
+log4j.appender.A1.layout.ConversionPattern=%-5p [%d] [%C{1}] - %m%n
\ No newline at end of file

protected/src/src/test/resources/log4j.properties

+#
+# Biopet is built on top of GATK Queue for building bioinformatic
+# pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+# SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+# should also be able to execute Biopet tools and pipelines.
+#
+# Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+#
+# Contact us at: sasc@lumc.nl
+#
+# A dual licensing mode is applied. The source code within this project that are
+# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# license; For commercial users or users who do not want to follow the AGPL
+# license, please contact us to obtain a separate license.
+#
+
+# Set root logger level to DEBUG and its only appender to A1.
+log4j.rootLogger=ERROR, A1
+
+# A1 is set to be a ConsoleAppender.
+log4j.appender.A1=org.apache.log4j.ConsoleAppender
+
+# A1 uses PatternLayout.
+log4j.appender.A1.layout=org.apache.log4j.PatternLayout
+log4j.appender.A1.layout.ConversionPattern=%-5p [%d] [%C{1}] - %m%n
\ No newline at end of file