Add region extraction

public/biopet-extensions/src/main/scala/nl/lumc/sasc/biopet/extensions/gatk/SelectVariants.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 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.
+ */
+package nl.lumc.sasc.biopet.extensions.gatk
+
+import java.io.File
+
+import nl.lumc.sasc.biopet.utils.config.Configurable
+import org.broadinstitute.gatk.utils.commandline.{Input, Output}
+
+/**
+ * Extension for CombineVariants from GATK
+ *
+ * Created by pjvan_thof on 2/26/15.
+ */
+class SelectVariants(val root: Configurable) extends Gatk {
+  val analysisType = "SelectVariants"
+
+  @Input(doc = "", required = true)
+  var inputFiles: List[File] = Nil
+
+  @Output(doc = "", required = true)
+  var outputFile: File = null
+
+  var excludeNonVariants: Boolean = false
+
+  var inputMap: Map[File, String] = Map()
+
+  def addInput(file: File, name: String): Unit = {
+    inputFiles :+= file
+    inputMap += file -> name
+  }
+
+  override def beforeGraph(): Unit = {
+    super.beforeGraph()
+    if (outputFile.getName.endsWith(".vcf.gz")) outputFiles :+= new File(outputFile.getAbsolutePath + ".tbi")
+    deps :::= inputFiles.filter(_.getName.endsWith("vcf.gz")).map(x => new File(x.getAbsolutePath + ".tbi"))
+    deps = deps.distinct
+  }
+
+  override def cmdLine = super.cmdLine +
+    (for (file <- inputFiles) yield {
+      inputMap.get(file) match {
+        case Some(name) => required("-V:" + name, file)
+        case _          => required("-V", file)
+      }
+    }).mkString +
+    required("-o", outputFile) +
+    conditional(excludeNonVariants, "--excludeNonVariants")
+}

public/gwas-test/src/main/scala/nl/lumc/sasc/biopet/pipelines/gwastest/GwasTest.scala

 package nl.lumc.sasc.biopet.pipelines.gwastest
 
+import java.io.File
+
 import nl.lumc.sasc.biopet.core.{PipelineCommand, Reference, BiopetQScript}
-import nl.lumc.sasc.biopet.extensions.gatk.CombineVariants
+import nl.lumc.sasc.biopet.extensions.gatk.{SelectVariants, CombineVariants}
 import nl.lumc.sasc.biopet.extensions.tools.GensToVcf
 import nl.lumc.sasc.biopet.utils.config.Configurable
 import nl.lumc.sasc.biopet.utils.intervals.BedRecordList
@@ -49,6 +51,7 @@ class GwasTest(val root: Configurable) extends QScript with BiopetQScript with R
         gensToVcf.contig = gen.contig
         gensToVcf.samplesFile = phenotypeFile
         gensToVcf.outputVcf = new File(outputDirGens, gen.genotypes.getName + ".vcf.gz")
+        gensToVcf.isIntermediate = true
         add(gensToVcf)
         cv.inputFiles :+= gensToVcf.outputVcf
       }
@@ -59,7 +62,18 @@ class GwasTest(val root: Configurable) extends QScript with BiopetQScript with R
     val snpTests = BedRecordList.fromReference(referenceFasta())
       .scatter(config("bin_size", default = 10^6))
       .allRecords.map { region =>
-      //TODO: bcftools view
+      val bedFile = new File(outputDir, "snptest"  + File.separator + region.chr + File.separator +
+        s"${region.chr}-${region.start + 1}-${region.end}.bed")
+      bedFile.getParentFile.mkdirs()
+      BedRecordList.fromList(List(region)).writeToFile(bedFile)
+
+      val sv = new SelectVariants(this)
+      sv.inputFiles :+= vcfFile
+      sv.outputFile = new File(outputDir, "snptest"  + File.separator + region.chr + File.separator +
+        s"${region.chr}-${region.start + 1}-${region.end}.vcf.gz")
+      sv.intervals :+= bedFile
+      sv.isIntermediate = true
+      add(sv)
 
       //TODO: snptest
       Map()