Yamsvp pipeline changes, including the Clever as pipeline instead as...

Yamsvp pipeline changes, including the Clever as pipeline instead as CommandLine function. Allowing post processing of Clever results and such..

biopet-framework/nbactions-yamsvp.xml

@@ -10,7 +10,7 @@
                 <goal>org.codehaus.mojo:exec-maven-plugin:1.2.1:exec</goal>
             </goals>
             <properties>
-                <exec.args>-classpath %classpath nl.lumc.sasc.biopet.pipelines.yamsvp.Yamsvp -outDir ../test -config ../test/samples.json -run</exec.args>
+                <exec.args>-classpath %classpath nl.lumc.sasc.biopet.pipelines.yamsvp.Yamsvp -outDir /usr/local/waiyi/ -config ../test/samples.json -run -l debug</exec.args>
                 <exec.executable>java</exec.executable>
                 <exec.workingdir>../test</exec.workingdir>
             </properties>
@@ -25,7 +25,7 @@
                 <goal>org.codehaus.mojo:exec-maven-plugin:1.2.1:exec</goal>
             </goals>
             <properties>
-                <exec.args>-Xdebug -Xrunjdwp:transport=dt_socket,server=n,address=${jpda.address} -classpath %classpath nl.lumc.sasc.biopet.pipelines.yamsvp.Yamsvp -outDir ../test -config ../test/samples.json -run</exec.args>
+                <exec.args>-Xdebug -Xrunjdwp:transport=dt_socket,server=n,address=${jpda.address} -classpath %classpath nl.lumc.sasc.biopet.pipelines.yamsvp.Yamsvp -outDir /usr/local/waiyi/ -config ../test/samples.json -run -l debug</exec.args>
                 <exec.executable>java</exec.executable>
                 <jpda.listen>true</jpda.listen>
                 <exec.workingdir>../test</exec.workingdir>
@@ -41,7 +41,7 @@
                 <goal>org.codehaus.mojo:exec-maven-plugin:1.2.1:exec</goal>
             </goals>
             <properties>
-                <exec.args>-classpath %classpath nl.lumc.sasc.biopet.pipelines.yamsvp.Yamsvp -outDir ../test -config ../test/samples.json -run</exec.args>
+                <exec.args>-classpath %classpath nl.lumc.sasc.biopet.pipelines.yamsvp.Yamsvp -outDir /usr/local/waiyi/ -config ../test/samples.json -run -l debug</exec.args>
                 <exec.executable>java</exec.executable>
                 <exec.workingdir>../test</exec.workingdir>
             </properties>

biopet-framework/src/main/scala/nl/lumc/sasc/biopet/extensions/svcallers/Clever.scala

-/*
- * To change this license header, choose License Headers in Project Properties.
- * To change this template file, choose Tools | Templates
- * and open the template in the editor.
- */
-
 package nl.lumc.sasc.biopet.extensions.svcallers
 
+
 import nl.lumc.sasc.biopet.core.BiopetCommandLineFunction
+import org.broadinstitute.gatk.queue.QScript
+import nl.lumc.sasc.biopet.core.BiopetQScript
 import nl.lumc.sasc.biopet.core.config.Configurable
-import org.broadinstitute.gatk.utils.commandline.{ Input, Output, Argument , MissingArgumentException}
+import nl.lumc.sasc.biopet.core.PipelineCommand
+import org.broadinstitute.gatk.utils.commandline.{ Input, Output, Argument }
 import java.io.File
 
 class Clever(val root: Configurable) extends BiopetCommandLineFunction  {
@@ -16,17 +14,17 @@ class Clever(val root: Configurable) extends BiopetCommandLineFunction  {
   
   private lazy val versionexecutable: File = config("version_exe", default = (new File(executable).getParent + "/ctk-version"))
   
+  
   override val defaultVmem = "4G"
   override val defaultThreads = 8
   
   override def versionCommand = versionexecutable.getAbsolutePath
   override val versionRegex = """(.*)""".r
   
-  
   @Input(doc = "Input file (bam)")
   var input: File = _
-
-  @Input(doc = "Reference Fasta file")
+  
+  @Input(doc = "Reference")
   var reference: File = _
   
   @Argument(doc = "Work directory")
@@ -42,10 +40,6 @@ class Clever(val root: Configurable) extends BiopetCommandLineFunction  {
     new File(workdir + "predictions.raw.txt")
   }
   
-//  var vcf: File = output
-  if ( reference == null ) reference = config("reference")
-  
-  
 //  var T: Option[Int] = config("T", default = defaultThreads)
   var f: Boolean = config("f", default = true) // delete work directory before running
 //  var w: String = config("w", default = workdir + "/work")
@@ -59,6 +53,8 @@ class Clever(val root: Configurable) extends BiopetCommandLineFunction  {
   }
 
   def cmdLine = required(executable) +
+      "--sorted " +
+      "--use_xa " +
       optional("-T", nCoresRequest) +
       conditional(f ,"-f") +
       conditional(a ,"-a") +
@@ -77,4 +73,44 @@ object Clever {
     clever.workdir = runDir
     return clever
   }
+}
+
+
+
+class CleverPipeline(val root: Configurable) extends QScript with BiopetQScript {
+  def this() = this(null)
+  
+  @Input(doc = "Input file (bam)")
+  var input: File = _
+
+  @Input(doc = "Reference")
+  var reference: File = _
+  
+  @Argument(doc = "Work directory")
+  var workdir: String = _
+  
+  override def init() {
+  }
+
+  def biopetScript() {
+    // write the pipeline here
+    logger.info("Starting Clever Pipeline")
+    
+    /// start clever and then copy the vcf into the root directory "<sample>.clever/"
+    val clever = Clever(this, input, reference, workdir )
+    outputFiles += ("clever_vcf" -> clever.outputvcf )
+    add( clever )
+  }
+}
+
+object CleverPipeline extends PipelineCommand {
+  override val pipeline = "/nl/lumc/sasc/biopet/extensions/svcallers/Clever/Clever.class"
+  
+  def apply(root: Configurable, input: File, runDir: String): CleverPipeline = {
+    val cleverpipeline = new CleverPipeline( root )
+    cleverpipeline.input = input
+    cleverpipeline.workdir = runDir
+    return cleverpipeline
+  }
+  
 }
\ No newline at end of file

biopet-framework/src/main/scala/nl/lumc/sasc/biopet/pipelines/yamsvp/Yamsvp.scala

@@ -10,6 +10,7 @@ import nl.lumc.sasc.biopet.core.PipelineCommand
 
 import nl.lumc.sasc.biopet.extensions.Cat
 import nl.lumc.sasc.biopet.extensions.picard.MergeSamFiles
+import nl.lumc.sasc.biopet.extensions.svcallers.Breakdancer
 import nl.lumc.sasc.biopet.extensions.svcallers.Clever
 
 import nl.lumc.sasc.biopet.pipelines.mapping.Mapping
@@ -17,15 +18,13 @@ import nl.lumc.sasc.biopet.pipelines.mapping.Mapping
 import org.broadinstitute.gatk.queue.QScript
 import org.broadinstitute.gatk.queue.function._
 
-
-
 class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
   def this() = this(null)
 
   var reference: File = _
   var finalBamFiles: List[File] = Nil
-  
-  def init() {
+
+  override def init() {
     for (file <- configfiles) globalConfig.loadConfigFile(file)
     reference = config("reference", required = true)
     if (outputDir == null)
@@ -44,52 +43,66 @@ class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
     // 
 
   }
-  
+
   def runSingleSampleJobs(sampleConfig: Map[String, Any]): Map[String, List[File]] = {
     var outputFiles: Map[String, List[File]] = Map()
     var libraryBamfiles: List[File] = List()
     var libraryFastqFiles: List[File] = List()
     val sampleID: String = sampleConfig("ID").toString
-    val sampleDir: String = outputDir + sampleID + "/" 
+    val sampleDir: String = outputDir + sampleID + "/"
+
+    val svcallingDir: String = sampleDir + "svcalls/"
+
     for ((library, libraryFiles) <- runLibraryJobs(sampleConfig)) {
       libraryBamfiles +:= libraryFiles("FinalBam")
     }
-    
+
     val bamFile: File = if (libraryBamfiles.size == 1) libraryBamfiles.head
-                  else if (libraryBamfiles.size > 1) {
-                    val mergeSamFiles = MergeSamFiles(this, libraryBamfiles, sampleDir)
-                    add(mergeSamFiles)
-                    mergeSamFiles.output
-                  } else null
+    else if (libraryBamfiles.size > 1) {
+      val mergeSamFiles = MergeSamFiles(this, libraryBamfiles, sampleDir)
+      add(mergeSamFiles)
+      mergeSamFiles.output
+    } else null
     val fastqFile: File = if (libraryFastqFiles.size == 1) libraryFastqFiles.head
-                  else if (libraryFastqFiles.size > 1) {
-                    val cat = Cat.apply(this, libraryFastqFiles, sampleDir + sampleID + ".fastq")
-                    add(cat)
-                    cat.output
-                  } else null
+    else if (libraryFastqFiles.size > 1) {
+      val cat = Cat.apply(this, libraryFastqFiles, sampleDir + sampleID + ".fastq")
+      add(cat)
+      cat.output
+    } else null
 
     /// bamfile will be used as input for the SV callers. First run Clever
-//    val cleverVCF : File = sampleDir + "/" + sampleID + ".clever.vcf"
-//    
-//    val clever = Clever(this, bamFile, this.reference, sampleDir )
-//    outputFiles += ("CleverVCF" -> List(clever.outputvcf) )
-//    add(clever)
+    //    val cleverVCF : File = sampleDir + "/" + sampleID + ".clever.vcf"
+
+    val cleverDir = svcallingDir + sampleID + ".clever/"
+    val clever = Clever(this, bamFile, this.reference, cleverDir)
+    outputFiles += ("clever_vcf" -> List(clever.outputvcf))
+    add(clever)
+
+    //    val breakdancerDir = sampleDir + sampleID + ".breakdancer/"
+    //    val breakdancer = Breakdancer(this, bamFile, this.reference, breakdancerDir )
+    //    outputFiles += ("breakdancer_vcf" -> List(breakdancer.output) )
+    //    addAll(breakdancer.functions)
+
     return outputFiles
   }
-  
+
   // Called for each run from a sample
   def runSingleLibraryJobs(runConfig: Map[String, Any], sampleConfig: Map[String, Any]): Map[String, File] = {
     var outputFiles: Map[String, File] = Map()
     val runID: String = runConfig("ID").toString
     val sampleID: String = sampleConfig("ID").toString
-    val runDir: String = outputDir + sampleID + "/run_" + runID + "/"
+    val alignmentDir: String = outputDir + sampleID + "/alignment/"
+
+    val runDir: String = alignmentDir + "run_" + runID + "/"
     if (runConfig.contains("R1")) {
       val mapping = Mapping.loadFromLibraryConfig(this, runConfig, sampleConfig, runDir)
       mapping.skipFlexiprep = false
-      mapping.skipMarkduplicates = true
-      mapping.defaultAligner = "bwa"
-      mapping.chunking = true
-      mapping.numberChunks = 4
+      mapping.skipMarkduplicates = true // we do the dedup marking using Sambamba
+      mapping.defaultAligner = "stampy"
+      mapping.aligner = "stampy" // enfore stampy to be the aligner
+
+      //      mapping.chunking = true // align in chunks
+      //      mapping.numberChunks = 4 // could be a fixed value if taken from config
       mapping.RGLB = runConfig("ID").toString
       mapping.RGSM = sampleConfig("ID").toString
       if (runConfig.contains("PL")) mapping.RGPL = runConfig("PL").toString
@@ -102,7 +115,7 @@ class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
 
       outputFiles += ("FinalBam" -> mapping.outputFiles("finalBamFile"))
     } else this.logger.error("Sample: " + sampleID + ": No R1 found for run: " + runConfig)
-    logger.debug( outputFiles )
+    logger.debug(outputFiles)
     return outputFiles
   }
 }