Merge branch 'develop' into fix-BIOPET-656

biopet-tools-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/tools/VcfStatsForSv.scala

+/**
+ * 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 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.
+ */
+package nl.lumc.sasc.biopet.extensions.tools
+
+import java.io.File
+
+import nl.lumc.sasc.biopet.core.summary.Summarizable
+import nl.lumc.sasc.biopet.core.ToolCommandFunction
+import nl.lumc.sasc.biopet.utils.config.Configurable
+import nl.lumc.sasc.biopet.utils.ConfigUtils
+import org.broadinstitute.gatk.utils.commandline.{ Argument, Input, Output }
+
+class VcfStatsForSv(val parent: Configurable) extends ToolCommandFunction with Summarizable {
+  def toolObject = nl.lumc.sasc.biopet.tools.vcfstats.VcfStatsForSv
+
+  mainFunction = false
+
+  @Input(required = true)
+  var inputFile: File = _
+
+  @Argument(required = true)
+  var histogramBinBoundaries: Array[Int] = _
+
+  @Output(required = true)
+  var outputFile: File = _
+
+  override def defaultCoreMemory = 1.0
+
+  override def cmdLine = super.cmdLine +
+    required("-i", inputFile) +
+    required("-o", outputFile) +
+    repeat("--histBinBoundaries", histogramBinBoundaries)
+
+  def summaryStats: Map[String, Any] = ConfigUtils.fileToConfigMap(outputFile)
+
+  def summaryFiles: Map[String, File] = Map.empty
+
+}

biopet-tools/src/main/scala/nl/lumc/sasc/biopet/tools/vcfstats/VcfStatsForSv.scala

+package nl.lumc.sasc.biopet.tools.vcfstats
+
+import java.io.File
+
+import htsjdk.variant.vcf.VCFFileReader
+import nl.lumc.sasc.biopet.utils.{ ConfigUtils, ToolCommand }
+
+import scala.collection.JavaConversions._
+
+object VcfStatsForSv extends ToolCommand {
+  /** Commandline arguments */
+  case class Args(inputFile: File = null, outputFile: File = null, histBinBoundaries: Array[Int] = Array()) extends AbstractArgs
+
+  /** Parsing commandline arguments */
+  class OptParser extends AbstractOptParser {
+    opt[File]('i', "inputFile") required () maxOccurs 1 valueName "<file>" action { (x, c) =>
+      c.copy(inputFile = x)
+    } validate {
+      x => if (x.exists) success else failure("Input VCF required")
+    } text "Input VCF file (required)"
+
+    opt[File]('o', "outputFile") required () maxOccurs 1 valueName "<file>" action { (x, c) =>
+      c.copy(outputFile = x)
+    } text "Output file (required)"
+
+    opt[Int]("histBinBoundaries") required () unbounded () action { (x, c) =>
+      c.copy(histBinBoundaries = c.histBinBoundaries :+ x)
+    } text "When counting the records, sv-s are divided to different size classes, this parameter should give the boundaries between these classes."
+  }
+
+  protected var cmdArgs: Args = _
+
+  def main(args: Array[String]): Unit = {
+    val argsParser = new OptParser
+    cmdArgs = argsParser.parse(args, Args()) getOrElse (throw new IllegalArgumentException)
+
+    logger.info(s"Parsing file: ${cmdArgs.inputFile}")
+
+    val stats: Map[String, Any] = getVariantCounts(cmdArgs.inputFile, cmdArgs.histBinBoundaries)
+
+    ConfigUtils.mapToYamlFile(stats, cmdArgs.outputFile)
+
+  }
+
+  /** Parses a vcf-file and counts sv-s by type and size. Sv-s are divided to different size classes, the parameter histogramBinBoundaries gives the boundaries between these classes. */
+  def getVariantCounts(vcfFile: File, histogramBinBoundaries: Array[Int]): Map[String, Any] = {
+    val delCounts, insCounts, dupCounts, invCounts = Array.fill(histogramBinBoundaries.size + 1) { 0 }
+    var traCount = 0
+
+    val reader = new VCFFileReader(vcfFile, false)
+    for (record <- reader) {
+      record.getAttributeAsString("SVTYPE", "") match {
+        case "TRA" | "CTX" | "ITX" => traCount += 1
+        case svType => {
+          val size = record.getEnd - record.getStart
+          var i = 0
+          while (i < histogramBinBoundaries.size && size > histogramBinBoundaries(i)) i += 1
+          svType match {
+            case "DEL" => delCounts(i) += 1
+            case "INS" => insCounts(i) += 1
+            case "DUP" => dupCounts(i) += 1
+            case "INV" => invCounts(i) += 1
+            case _     => logger.warn(s"Vcf file contains a record of unknown type: file-$vcfFile, type-$svType")
+          }
+        }
+      }
+    }
+    reader.close()
+
+    Map("DEL" -> delCounts, "INS" -> insCounts, "DUP" -> dupCounts, "INV" -> invCounts, "TRA" -> traCount)
+  }
+
+}

biopet-utils/src/main/resources/nl/lumc/sasc/biopet/utils/rscript/plotXY.R

@@ -13,6 +13,11 @@ parser$add_argument('--llabel', dest='llabel', type='character')
 parser$add_argument('--title', dest='title', type='character')
 parser$add_argument('--hideLegend', dest='hideLegend', type='character', default="false")
 parser$add_argument('--removeZero', dest='removeZero', type='character', default="false")
+parser$add_argument('--xLog10', dest='xLog10', type='character', default="false")
+parser$add_argument('--yLog10', dest='yLog10', type='character', default="false")
+
+parser$add_argument('--xLog10Breaks', dest='xLog10Breaks', nargs='+', type='integer')
+parser$add_argument('--xLog10Labels', dest='xLog10Labels', nargs='+', type='character')
 
 arguments <- parser$parse_args()
 
@@ -31,14 +36,29 @@ DF1 <- melt(DF, id.var="Rank")
 if (arguments$removeZero == "true") DF1 <- DF1[DF1$value > 0, ]
 if (arguments$removeZero == "true") print("Removed 0 values")
 
-ggplot(DF1, aes(x = Rank, y = value, group = variable, color = variable)) +
+plot = ggplot(DF1, aes(x = Rank, y = value, group = variable, color = variable)) +
   xlab(xlab) +
   ylab(arguments$ylabel) +
-  guides(fill=guide_legend(title=arguments$llabel)) +
+  guides(color=guide_legend(title=arguments$llabel)) +
   theme(axis.text.x = element_text(angle = 90, hjust = 1, size = 8)) +
   ggtitle(arguments$title) +
   theme_bw() +
   theme(legend.position = legendPosition) +
   geom_line()
 
+
+if (arguments$xLog10 == "true") {
+  if (!is.null(arguments$xLog10Labels)) {
+    scale_x <- scale_x_log10(breaks = arguments$xLog10Breaks, labels=arguments$xLog10Labels)
+  } else {
+    scale_x <- scale_x_log10()
+  }
+  plot <- plot + scale_x
+}
+if (arguments$yLog10 == "true") {
+  plot <- plot + scale_y_log10()
+}
+
+plot
+
 dev.off()

biopet-utils/src/main/scala/nl/lumc/sasc/biopet/utils/rscript/LinePlot.scala

@@ -39,6 +39,13 @@ class LinePlot(val parent: Configurable) extends Rscript {
   var hideLegend: Boolean = config("hide_legend", default = false)
   var removeZero: Boolean = config("removeZero", default = false)
 
+  // whether to use log scale for x and y axis
+  var xLog10: Boolean = false
+  var yLog10: Boolean = false
+
+  var xLog10AxisTicks: Seq[String] = Seq.empty
+  var xLog10AxisLabels: Seq[String] = Seq.empty
+
   override def cmd = super.cmd ++
     Seq("--input", input.getAbsolutePath) ++
     Seq("--output", output.getAbsolutePath) ++
@@ -49,7 +56,11 @@ class LinePlot(val parent: Configurable) extends Rscript {
     llabel.map(Seq("--llabel", _)).getOrElse(Seq()) ++
     title.map(Seq("--title", _)).getOrElse(Seq()) ++
     (if (hideLegend) Seq("--hideLegend", "true") else Seq()) ++
-    (if (removeZero) Seq("--removeZero", "true") else Seq())
+    (if (removeZero) Seq("--removeZero", "true") else Seq()) ++
+    (if (xLog10) Seq("--xLog10", "true") else Seq()) ++
+    (if (yLog10) Seq("--yLog10", "true") else Seq()) ++
+    (if (xLog10AxisTicks.nonEmpty) xLog10AxisTicks.+:("--xLog10Breaks") else Seq()) ++
+    (if (xLog10AxisLabels.nonEmpty) xLog10AxisLabels.+:("--xLog10Labels") else Seq())
 }
 
 object LinePlot {

shiva/src/main/resources/nl/lumc/sasc/biopet/pipelines/shiva/sampleVariantsSv.ssp

+#import(nl.lumc.sasc.biopet.utils.summary.db.SummaryDb)
+#import(nl.lumc.sasc.biopet.utils.summary.db.Schema.Sample)
+#import(java.io.File)
+<%@ var summary: SummaryDb %>
+<%@ var allSamples: Seq[Sample]%>
+<%@ var rootPath: String %>
+<%@ var outputDir: File %>
+<%@ var runId: Int %>
+<%@ var sampleId: Option[Int] = None %>
+<%@ var showPlot: Boolean = false %>
+<%@ var showTable: Boolean = true %>
+<%@ var showIntro: Boolean = true %>
+
+#{
+    val sampleNames: Seq[String] = sampleId match {
+       case Some(sampleId) => Seq(allSamples.filter(s => s.id == sampleId).head.name)
+       case _ => allSamples.collect({case s: Sample => s.name}).sorted
+    }
+
+    val counts: Map[String, Map[String, Array[Long]]] = ShivaSvCallingReport.parseSummaryForSvCounts(summary, runId, sampleNames)
+    val traCounts: Map[String, Long] = ShivaSvCallingReport.parseSummaryForTranslocations(summary, runId, sampleNames)
+
+    var svTypes = List(
+    SvTypeForReport("DEL", "Deletions", "svSummaryDeletions.tsv", "svSummaryDeletions.png"),
+    SvTypeForReport("DUP", "Duplications", "svSummaryDuplications.tsv", "svSummaryDuplications.png"),
+    SvTypeForReport("INS", "Insertions", "svSummaryInsertions.tsv", "svSummaryInsertions.png"),
+    SvTypeForReport("INV", "Inversions", "svSummaryInversions.tsv", "svSummaryInversions.png"))
+    svTypes = svTypes.filter(sv => counts.contains(sv.svType))
+    val tsvAllTypes = "svSummary.tsv"
+
+    ShivaSvCallingReport.writeTsvFiles(sampleNames, counts, svTypes, tsvAllTypes, outputDir)
+    ShivaSvCallingReport.createPlots(svTypes, outputDir)
+}#
+
+
+#if (showPlot)
+    <div class="panel-body">
+    #for (sv <- svTypes)
+        <img src="${sv.pngFileName}" />
+    #end
+    </div>
+    #if (traCounts.nonEmpty)
+        <div class="panel-body">
+        <h5>Number of translocation events detected:</h5>
+        <table class="table table-condensed" style="width:auto">
+            <thead><tr>
+            #for (sampleName <- sampleNames)
+                <th>${sampleName}</th>
+            #end
+            </tr></thead>
+            <tbody><tr>
+            #for (sampleName <- sampleNames)
+                #{
+                    val sampleCount: String = traCounts.get(sampleName) match {
+                        case Some(c) => c.toString()
+                        case None => "-"
+                    }
+                }#
+                <td>${sampleCount}</td>
+            #end
+            </tr></tbody>
+        </table>
+        </div>
+    #end
+    <div class="panel-footer">
+        <button type="button" class="btn btn-info" data-toggle="collapse" data-target="#svSummaryTable">
+            #if (showTable)
+                <i class="glyphicon glyphicon-eye-close"></i> Hide tables
+            #else
+                <i class="glyphicon glyphicon-eye-open"></i> Show tables
+            #end
+        </button>
+        <a href="${tsvAllTypes}"><button type="button" class="btn btn-info"><i class="glyphicon glyphicon-cloud-download"></i> TSV file</button></a>
+    </div>
+#end
+
+<div class="panel-body collapse #if (showTable)in#end" id="svSummaryTable">
+
+#for (sv <- svTypes)
+#{
+    val countsForSvType: Map[String, Array[Long]] = counts(sv.svType)
+    val missingCounts: Array[String] = Array.fill(ShivaSvCallingReport.histogramText.size) { "-" }
+}#
+<h3>${sv.displayText}</h3>
+<table class="table sortable-theme-bootstrap" data-sortable>
+    <thead><tr><th data-sorted="true" data-sorted-direction="ascending">Sample</th>
+        #for (text <- ShivaSvCallingReport.histogramText)
+            <th>${text}</th>
+         #end
+    </tr></thead>
+    <tbody>
+        #for (sampleName <- sampleNames)
+            <tr>
+            <td><a href="${rootPath}Samples/${sampleName}/index.html">${sampleName}</a></td>
+            #{
+                val sampleCounts: Array[String] = countsForSvType.get(sampleName) match {
+                    case Some(c: Array[_]) => c.collect({case x => x.toString()})
+                    case None => missingCounts
+                }
+            }#
+
+            #for (countForSize <- sampleCounts)
+                <td>${countForSize}</td>
+            #end
+            </tr>
+        #end
+    </tbody>
+</table>
+
+#end
+</div>

shiva/src/main/scala/nl/lumc/sasc/biopet/pipelines/shiva/ShivaReport.scala

@@ -48,13 +48,22 @@ trait ShivaReportTrait extends MultisampleMappingReportTrait {
       case _          => false
     }
 
+  def svCallingExecuted = summary.getSettingKeys(runId, "shiva", NoModule, keyValues = Map("sv_calling" -> List("sv_calling"))).get("sv_calling")
+    .flatten match {
+      case Some(true) => true
+      case _          => false
+    }
+
   override def frontSection = ReportSection("/nl/lumc/sasc/biopet/pipelines/shiva/shivaFront.ssp")
 
   override def pipelineName = "shiva"
 
-  override def additionalSections = super.additionalSections ++ (if (variantcallingExecuted) List("Variantcalling" -> ReportSection("/nl/lumc/sasc/biopet/pipelines/shiva/sampleVariants.ssp",
-    Map("showPlot" -> true, "showTable" -> false)))
-  else Nil)
+  override def additionalSections = {
+    val params = Map("showPlot" -> true, "showTable" -> false)
+    super.additionalSections ++
+      (if (variantcallingExecuted) List("SNV Calling" -> ReportSection("/nl/lumc/sasc/biopet/pipelines/shiva/sampleVariants.ssp", params)) else Nil) ++
+      (if (svCallingExecuted) List("SV Calling" -> ReportSection("/nl/lumc/sasc/biopet/pipelines/shiva/sampleVariantsSv.ssp", params)) else Nil)
+  }
 
   /** Root page for the shiva report */
   override def indexPage: Future[ReportPage] = Future {
@@ -107,9 +116,10 @@ trait ShivaReportTrait extends MultisampleMappingReportTrait {
 
   /** Single sample page */
   override def samplePage(sampleId: Int, args: Map[String, Any]): Future[ReportPage] = Future {
-    val variantcallingSection = if (variantcallingExecuted) List("Variantcalling" -> ReportSection("/nl/lumc/sasc/biopet/pipelines/shiva/sampleVariants.ssp")) else Nil
+    val variantcallingSection = if (variantcallingExecuted) List("SNV Calling" -> ReportSection("/nl/lumc/sasc/biopet/pipelines/shiva/sampleVariants.ssp")) else Nil
+    val svSection = if (svCallingExecuted) List("SV Calling" -> ReportSection("/nl/lumc/sasc/biopet/pipelines/shiva/sampleVariantsSv.ssp")) else Nil
     val oldPage: ReportPage = super.samplePage(sampleId, args)
-    oldPage.copy(sections = variantcallingSection ++ oldPage.sections)
+    oldPage.copy(sections = variantcallingSection ++ svSection ++ oldPage.sections)
   }
 
   /** Name of the report */
@@ -163,4 +173,5 @@ trait ShivaReportTrait extends MultisampleMappingReportTrait {
     plot.width = Some(200 + (samples.count(s => sampleId.getOrElse(s) == s) * 10))
     plot.runLocal()
   }
+
 }

shiva/src/main/scala/nl/lumc/sasc/biopet/pipelines/shiva/ShivaSvCalling.scala

@@ -14,9 +14,10 @@
  */
 package nl.lumc.sasc.biopet.pipelines.shiva
 
-import nl.lumc.sasc.biopet.core.summary.SummaryQScript
-import nl.lumc.sasc.biopet.core.{ PipelineCommand, Reference, SampleLibraryTag }
+import nl.lumc.sasc.biopet.core.summary.{ Summarizable, SummaryQScript }
+import nl.lumc.sasc.biopet.core.{ PipelineCommand, Reference }
 import nl.lumc.sasc.biopet.extensions.Pysvtools
+import nl.lumc.sasc.biopet.extensions.tools.VcfStatsForSv
 import nl.lumc.sasc.biopet.pipelines.shiva.svcallers._
 import nl.lumc.sasc.biopet.utils.config.Configurable
 import nl.lumc.sasc.biopet.utils.{ BamUtils, Logging }
@@ -27,7 +28,7 @@ import org.broadinstitute.gatk.queue.QScript
  *
  * Created by pjvan_thof on 2/26/15.
  */
-class ShivaSvCalling(val parent: Configurable) extends QScript with SummaryQScript with SampleLibraryTag with Reference {
+class ShivaSvCalling(val parent: Configurable) extends QScript with SummaryQScript with Reference {
   qscript =>
 
   def this() = this(null)
@@ -96,6 +97,21 @@ class ShivaSvCalling(val parent: Configurable) extends QScript with SummaryQScri
     // group by "tags"
     // sample tagging is however not available within this pipeline
 
+    for ((sample, mergedResultFile) <- outputMergedVCFbySample) {
+      val vcfStats = new VcfStatsForSv(qscript)
+      vcfStats.inputFile = mergedResultFile
+      vcfStats.outputFile = new File(outputDir, s".$sample.merged.stats")
+      vcfStats.histogramBinBoundaries = ShivaSvCallingReport.histogramBinBoundaries
+
+      add(vcfStats)
+      addSummarizable(vcfStats, "vcfstats-sv", Some(sample))
+
+      addSummarizable(new Summarizable {
+        def summaryFiles = Map("output_vcf" -> mergedResultFile)
+        def summaryStats = Map.empty
+      }, "merge_variants", Some(sample))
+    }
+
     addSummaryJobs()
   }
 
@@ -103,10 +119,11 @@ class ShivaSvCalling(val parent: Configurable) extends QScript with SummaryQScri
   protected def callersList: List[SvCaller] = List(new Breakdancer(this), new Clever(this), new Delly(this), new Pindel(this))
 
   /** Settings for the summary */
-  def summarySettings = Map("sv_callers" -> configCallers.toList)
+  def summarySettings = Map("sv_callers" -> configCallers.toList, "hist_bin_boundaries" -> ShivaSvCallingReport.histogramBinBoundaries)
 
   /** Files for the summary */
-  def summaryFiles: Map[String, File] = Map("final_mergedvcf" -> (if (inputBams.size > 1) outputMergedVCF else outputMergedVCFbySample.values.head))
+  def summaryFiles: Map[String, File] = if (inputBams.size > 1) Map("final_mergedvcf" -> outputMergedVCF) else Map.empty
+
 }
 
 object ShivaSvCalling extends PipelineCommand
\ No newline at end of file

shiva/src/main/scala/nl/lumc/sasc/biopet/pipelines/shiva/ShivaSvCallingReport.scala

+package nl.lumc.sasc.biopet.pipelines.shiva
+
+import java.io.{ File, PrintWriter }
+
+import nl.lumc.sasc.biopet.core.report.ReportBuilder
+import nl.lumc.sasc.biopet.utils.Logging
+import nl.lumc.sasc.biopet.utils.rscript.LinePlot
+import nl.lumc.sasc.biopet.utils.summary.db.SummaryDb
+import nl.lumc.sasc.biopet.utils.summary.db.SummaryDb.{ ModuleName, PipelineName, SampleName }
+
+import scala.concurrent.Await
+import scala.concurrent.duration.Duration
+
+object ShivaSvCallingReport extends Logging {
+
+  implicit lazy val ec = ReportBuilder.ec
+
+  val histogramBinBoundaries: Array[Int] = Array(100, 1000, 10000, 100000, 1000000, 10000000)
+  val histogramPlotTicks: Array[Int] = Array(100, 1000, 10000, 100000, 1000000, 10000000, 100000000)
+  val histogramText: List[String] = List("<=100bp", "0.1-1kb", "1-10kb", "10-100kb", "0.1-1Mb", "1-10Mb", ">10Mb")
+
+  def parseSummaryForSvCounts(summary: SummaryDb, runId: Int, sampleNames: Seq[String]): Map[String, Map[String, Array[Long]]] = {
+    var delCounts, insCounts, dupCounts, invCounts: Map[String, Array[Long]] = Map()
+
+    for (sampleName <- sampleNames) {
+      val sampleCounts: Map[String, Any] = Await.result(summary.getStat(runId, PipelineName("shivasvcalling"), ModuleName("vcfstats-sv"), SampleName(sampleName)), Duration.Inf).get
+      for ((svType, counts) <- sampleCounts.collect({ case (k, v: List[_]) => (k, v.toArray[Any]) })) {
+        val elem: Tuple2[String, Array[Long]] = (sampleName, counts.collect({ case x: Long => x }))
+        svType match {
+          case "DEL" => delCounts += elem
+          case "INS" => insCounts += elem
+          case "DUP" => dupCounts += elem
+          case "INV" => invCounts += elem
+        }
+      }
+    }
+
+    var result: Map[String, Map[String, Array[Long]]] = Map()
+    if (delCounts.exists(elem => (elem._2.sum > 0))) result = Map("DEL" -> delCounts)
+    if (insCounts.exists(elem => (elem._2.sum > 0))) result += ("INS" -> insCounts)
+    if (dupCounts.exists(elem => (elem._2.sum > 0))) result += ("DUP" -> dupCounts)
+    if (invCounts.exists(elem => (elem._2.sum > 0))) result += ("INV" -> invCounts)
+    result
+  }
+
+  def parseSummaryForTranslocations(summary: SummaryDb, runId: Int, sampleNames: Seq[String]): Map[String, Long] = {
+    var traCounts: Map[String, Long] = Map()
+    for (sampleName <- sampleNames) {
+      val counts: Map[String, Any] = Await.result(summary.getStat(runId, PipelineName("shivasvcalling"), ModuleName("vcfstats-sv"), SampleName(sampleName)), Duration.Inf).get
+      counts.get("TRA") match {
+        case Some(c: Long) => traCounts += (sampleName -> c)
+        case Some(c)       => logger.error(s"Unable to parse translocation counts from summary db for sample $sampleName (type mismatch, type in the db: ${c.getClass})")
+        case _             => logger.error(s"Summary db doesn't have translocation counts for sample $sampleName")
+      }
+
+    }
+    if (traCounts.exists(elem => elem._2 > 0)) traCounts else Map.empty
+  }
+
+  def writeTsvFiles(sampleNames: Seq[String], counts: Map[String, Map[String, Array[Long]]], svTypes: List[SvTypeForReport], outFileAllTypes: String, outDir: File): Unit = {
+
+    val tsvWriter = new PrintWriter(new File(outDir, outFileAllTypes))
+    tsvWriter.print("sv_type\tsample")
+    histogramText.foreach(bin => tsvWriter.print("\t" + bin))
+    tsvWriter.println()
+
+    val missingCounts: Array[String] = Array.fill(ShivaSvCallingReport.histogramText.size) { "-" }
+
+    for (sv <- svTypes) {
+      val countsForSvType: Map[String, Array[Long]] = counts.getOrElse(sv.svType, Map.empty)
+
+      if (countsForSvType.nonEmpty) {
+
+        writeTsvFileForSvType(sv, countsForSvType, sampleNames, outDir)
+
+        for (sampleName <- sampleNames) {
+          val sampleCounts: Array[String] = countsForSvType.get(sampleName) match {
+            case Some(c) => c.collect({ case x => x.toString() })
+            case None => {
+              logger.error(s"Internal error, missing sv counts, sample-$sampleName, sv type-${sv.svType}")
+              missingCounts
+            }
+          }
+
+          tsvWriter.print(sv.svType + "\t" + sampleName + "\t")
+          tsvWriter.println(sampleCounts.mkString("\t"))
+        }
+
+      } else {
+        logger.error(s"Internal error, skipping writing the tsv-file for sv type ${sv.svType}")
+      }
+
+    }
+    tsvWriter.close()
+  }
+
+  def writeTsvFileForSvType(svType: SvTypeForReport, counts: Map[String, Array[Long]], sampleNames: Seq[String], outDir: File): Unit = {
+    val tsvWriter = new PrintWriter(new File(outDir, svType.tsvFileName))
+
+    tsvWriter.print("histogramBin")
+    val samplesWithCounts: Seq[String] = sampleNames.filter(x => counts.contains(x))
+    samplesWithCounts.foreach(sampleName => tsvWriter.print("\t" + sampleName))
+    tsvWriter.println()
+
+    for (i <- histogramPlotTicks.indices) {
+      tsvWriter.print(histogramPlotTicks(i))
+      samplesWithCounts.foreach(sampleName => tsvWriter.print("\t" + counts.get(sampleName).get(i)))
+      tsvWriter.println()
+    }
+
+    tsvWriter.close()
+  }
+
+  def createPlots(svTypes: List[SvTypeForReport], outDir: File): Unit = {
+    for (sv <- svTypes) {
+      val tsvFile = new File(outDir, sv.tsvFileName)
+      val pngFile: File = new File(outDir, sv.pngFileName)
+      val plot = LinePlot(tsvFile, pngFile,
+        xlabel = Some(s"${sv.displayText.substring(0, sv.displayText.length - 1)} size"),
+        ylabel = Some("Number of loci"),
+        title = Some(sv.displayText),
+        width = 400,
+        removeZero = false)
+      plot.height = Some(300)
+      plot.llabel = Some("Sample")
+      plot.xLog10 = true
+      plot.yLog10 = true
+      plot.xLog10AxisTicks = histogramPlotTicks.collect({ case x => x.toString() })
+      plot.xLog10AxisLabels = histogramText
+      plot.runLocal()
+    }
+
+  }
+
+}
+
+case class SvTypeForReport(svType: String, displayText: String, tsvFileName: String, pngFileName: String)
\ No newline at end of file

shiva/src/test/scala/nl/lumc/sasc/biopet/pipelines/shiva/ShivaSvCallingTest.scala

@@ -90,7 +90,7 @@ class ShivaSvCallingTest extends TestNGSuite with Matchers {
       pipeline.init()
       pipeline.script()
 
-      val summaryCallers = pipeline.summarySettings("sv_callers")
+      val summaryCallers = pipeline.summarySettings.get("sv_callers").get.asInstanceOf[List[String]]
       if (delly) assert(summaryCallers.contains("delly"))
       else assert(!summaryCallers.contains("delly"))
       if (clever) assert(summaryCallers.contains("clever"))
@@ -182,7 +182,7 @@ class ShivaSvCallingTest extends TestNGSuite with Matchers {
     pipeline.init()