Commit changes before removing branch

biopet-framework/src/main/scala/nl/lumc/sasc/biopet/extensions/sambamba/SambambaMarkdup.scala

+package nl.lumc.sasc.biopet.extensions.sambamba
+
+import nl.lumc.sasc.biopet.core.config.Configurable
+import org.broadinstitute.gatk.utils.commandline.{ Input, Output }
+import java.io.File
+
+class SambambaMarkdup(val root: Configurable) extends Sambamba {
+  override val defaultThreads = 4
+
+  @Input(doc = "Bam File")
+  var input: File = _
+
+  @Output(doc = "Markdup output bam")
+  var output: File = _
+
+  
+  var remove_duplicates: Boolean = config("remove_duplicates", default = false)
+  
+  // @doc: compression_level 6 is average, 0 = no compression, 9 = best
+  val compression_level: Option[Int] = config("compression_level", default=6) 
+  val hash_table_size: Option[Int] = config("hash-table-size", default=262144)
+  val overflow_list_size: Option[Int] = config("overflow-list-size", default=200000)
+  val io_buffer_size: Option[Int] = config("io-buffer-size", default=128)
+  
+  
+  def cmdLine = required(executable) + 
+  required("markdup") + 
+  conditional(remove_duplicates, "--remove-duplicates") +
+  optional("-t", nCoresRequest) +
+  optional("-l", compression_level) +
+  optional("--hash-table-size=", hash_table_size, spaceSeparated=false ) +
+  optional("--overflow-list-size=", overflow_list_size, spaceSeparated=false ) +
+  optional("--io-buffer-size=", io_buffer_size, spaceSeparated=false ) +
+  required(input) + 
+  required(output)
+}
+
+object SambambaMarkdup {
+  def apply(root: Configurable, input: File, output: File): SambambaMarkdup = {
+    val flagstat = new SambambaMarkdup(root)
+    flagstat.input = input
+    flagstat.output = output
+    return flagstat
+  }
+
+  def apply(root: Configurable, input: File, outputDir: String): SambambaMarkdup = {
+    val dir = if (outputDir.endsWith("/")) outputDir else outputDir + "/"
+    val outputFile = new File(dir + swapExtension(input.getName))
+    return apply(root, input, outputFile)
+  }
+
+  def apply(root: Configurable, input: File): SambambaMarkdup = {
+    return apply(root, input, new File(swapExtension(input.getAbsolutePath)))
+  }
+
+  private def swapExtension(inputFile: String) = inputFile.stripSuffix(".bam") + ".bam.bai"
+}
\ No newline at end of file

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

@@ -11,33 +11,34 @@ import java.io.File
 
 class BreakdancerConfig(val root: Configurable) extends BiopetCommandLineFunction {
   executable = config("exe", default = "bam2cfg.pl", freeVar = false)
+
   @Input(doc = "Bam File")
   var input: File = _
 
   @Output(doc = "Output File")
   var output: File = _
 
-  var MIN_MQ: Option[Int] = config("min_mq", default = 20) // minimum of MQ to consider for taking read into histogram
-  var USE_MQ: Boolean = config("use_mq", default = false)
-  var MIN_INSERTSIZE: Option[Int] = config("min_insertsize", default = 450)
-  var SOLID_DATA: Boolean = config("solid", default = false)
-  var SD_CUTOFF: Option[Int] = config("sd_cutoff", default = 4) // Cutoff in unit of standard deviation [4]
+  var min_mq: Option[Int] = config("min_mq", default = 20) // minimum of MQ to consider for taking read into histogram
+  var use_mq: Boolean = config("use_mq", default = false)
+  var min_insertsize: Option[Int] = config("min_insertsize", default = 450)
+  var solid_data: Boolean = config("solid", default = false)
+  var sd_cutoff: Option[Int] = config("sd_cutoff", default = 4) // Cutoff in unit of standard deviation [4]
   
   // we set this to a higher number to avoid biases in small numbers in sorted bams
-  var MIN_OBSERVATIONS: Option[Int] = config("min_observations", default = 1000000) //  Number of observation required to estimate mean and s.d. insert size [10_000]
-  var COEFVAR_CUTOFF: Option[Int] = config("coef_cutoff", default = 1) // Cutoff on coefficients of variation [1]
-  var HISTOGRAM_BINS: Option[Int] = config("histogram_bins", default = 50) // Number of bins in the histogram [50]
+  var min_observations: Option[Int] = config("min_observations", default = 10000) //  Number of observation required to estimate mean and s.d. insert size [10_000]
+  var coefvar_cutoff: Option[Int] = config("coef_cutoff", default = 1) // Cutoff on coefficients of variation [1]
+  var histogram_bins: Option[Int] = config("histogram_bins", default = 50) // Number of bins in the histogram [50]
   
   def cmdLine = required(executable) +
-      optional("-q", MIN_MQ) +
-      conditional(USE_MQ, "-m") +
-      optional("-s", MIN_INSERTSIZE) +
-      conditional(SOLID_DATA, "-s") +
-      optional("-c", SD_CUTOFF) +
-      optional("-n", MIN_OBSERVATIONS) +
-      optional("-v", COEFVAR_CUTOFF) +
-      optional("-b", HISTOGRAM_BINS)
-      required(input) + " > " + required(output)
+      optional("-q", min_mq) +
+      conditional(use_mq, "-m") +
+      optional("-s", min_insertsize) +
+      conditional(solid_data, "-s") +
+      optional("-c", sd_cutoff) +
+      optional("-n", min_observations) +
+      optional("-v", coefvar_cutoff) +
+      optional("-b", histogram_bins) +
+      required(input) + " 1> " + required(output)
 }
 
 object BreakdancerConfig {
@@ -74,18 +75,18 @@ class BreakdancerCaller(val root: Configurable) extends BiopetCommandLineFunctio
   executable = config("exe", default = "breakdancer-max", freeVar = false)
   
   override val defaultVmem = "4G"
-  override val defaultThreads = 8
+  override val defaultThreads = 1 // breakdancer can only work on 1 single thread
   
   override val versionRegex = """.*[Vv]ersion:? (.*)""".r
-//  override val versionExitcode = List(0, 1)
+  override val versionExitcode = List(1)
   override def versionCommand = executable
   
   
   @Input(doc = "The breakdancer configuration file")
   var input: File = _
   
-  @Argument(doc = "Work directory")
-  var workdir: String = _
+//  @Argument(doc = "Work directory")
+//  var workdir: String = _
   
   @Output(doc = "Breakdancer VCF output")
   var output: File = _
@@ -125,8 +126,6 @@ class BreakdancerCaller(val root: Configurable) extends BiopetCommandLineFunctio
   var y: Option[Int] = config("y", default = 30)  
   
   override def beforeCmd {
-    if (workdir == null) throw new Exception("Breakdancer :: Workdirectory is not defined")
-//    if (input.getName.endsWith(".sort.bam")) sorted = true
   }
 
   def cmdLine = required(executable) + 
@@ -176,6 +175,10 @@ class Breakdancer(val root: Configurable) extends QScript with BiopetQScript {
 //    new File(workdir + "/" + input.getName.substring(0, input.getName.lastIndexOf(".bam")) + ".breakdancer.vcf")
 //  }
   
+  @Output(doc = "Breakdancer config")
+  lazy val configfile: File = {
+    new File(workdir + "/" + input.getName.substring(0, input.getName.lastIndexOf(".bam")) + ".breakdancer.cfg")
+  }
   @Output(doc = "Breakdancer raw output")
   lazy val outputraw: File = {
     new File(workdir + "/" + input.getName.substring(0, input.getName.lastIndexOf(".bam")) + ".breakdancer.tsv")
@@ -188,7 +191,7 @@ class Breakdancer(val root: Configurable) extends QScript with BiopetQScript {
     // read config and set all parameters for the pipeline
     logger.info("Starting Breakdancer configuration")
     
-    val bdcfg = BreakdancerConfig(this, input, workdir)
+    val bdcfg = BreakdancerConfig(this, input, this.configfile)
     outputFiles += ("breakdancer_cfg" -> bdcfg.output )
     add( bdcfg )
     
@@ -202,8 +205,6 @@ class Breakdancer(val root: Configurable) extends QScript with BiopetQScript {
     
     
   }
-  
-//  private def swapExtension(inputFile: String) = inputFile.substring(0, inputFile.lastIndexOf(".bam")) + ".breakdancer.tsv"
 }
 
 object Breakdancer extends PipelineCommand {

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

@@ -20,6 +20,7 @@ class Clever(val root: Configurable) extends BiopetCommandLineFunction  {
   
   override def versionCommand = versionexecutable.getAbsolutePath
   override val versionRegex = """(.*)""".r
+  override val versionExitcode = List(0, 1)
   
   @Input(doc = "Input file (bam)")
   var input: File = _
@@ -92,7 +93,6 @@ class CleverPipeline(val root: Configurable) extends QScript with BiopetQScript
   @Argument(doc = "Work directory")
   var workdir: String = _
   
-  
   @Argument(doc = "Current working directory")
   var cwd: String = _
   
@@ -117,6 +117,8 @@ object CleverPipeline extends PipelineCommand {
     val cleverpipeline = new CleverPipeline( root )
     cleverpipeline.input = input
     cleverpipeline.workdir = runDir
+    cleverpipeline.init
+    cleverpipeline.biopetScript
     return cleverpipeline
   }
   

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

@@ -76,9 +76,6 @@ class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
       // this is extending the libraryBamfiles list like '~=' in D or .append in Python or .push_back in C++
       libraryBamfiles ++= List(libraryOutput.mappedBamFile)
     }
-    
-    logger.debug( "We have the following bamfiles for merging" )
-    logger.debug( libraryBamfiles )
 
     val bamFile: File = 
       if (libraryBamfiles.size == 1) libraryBamfiles.head
@@ -93,8 +90,6 @@ class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
         
         mergeSamFiles.output
       } else null
-
-    logger.debug( bamFile )
     
     /// bamfile will be used as input for the SV callers. First run Clever
     //    val cleverVCF : File = sampleDir + "/" + sampleID + ".clever.vcf"
@@ -108,8 +103,8 @@ class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
     val breakdancer = Breakdancer(this, bamFile, this.reference, breakdancerDir)
     sampleOutput.vcf += ("breakdancer" -> List(breakdancer.outputraw))
     addAll(breakdancer.functions)
-//
-//    
+
+    // for pindel we should use per library config collected into one config file
 //    val pindelDir = svcallingDir + sampleID + ".pindel/"
 //    val pindel = Pindel(this, bamFile, this.reference, pindelDir)
 //    sampleOutput.vcf += ("pindel" -> List(pindel.outputvcf))
@@ -123,14 +118,13 @@ class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
 
   def runSingleLibraryJobs(runConfig: Map[String, Any], sampleConfig: Map[String, Any]): LibraryOutput = {
     val libraryOutput = new LibraryOutput
-    var outputFiles: Map[String, File] = Map()
+    
     val runID: String = runConfig("ID").toString
     val sampleID: String = sampleConfig("ID").toString
     val alignmentDir: String = outputDir + sampleID + "/alignment/"
-
     val runDir: String = alignmentDir + "run_" + runID + "/"
+    
     if (runConfig.contains("R1")) {
-      //      val mapping = Mapping.loadFromLibraryConfig(this, runConfig, sampleConfig, runDir)
       val mapping = new Mapping(this)
 
       mapping.defaultAligner = "stampy"
@@ -153,7 +147,6 @@ class Yamsvp(val root: Configurable) extends QScript with MultiSampleQScript {
 
       // start sambamba dedup
 
-      //      outputFiles += ("final_bam" -> mapping.outputFiles("finalBamFile"))
       libraryOutput.mappedBamFile = mapping.outputFiles("finalBamFile")
     } else this.logger.error("Sample: " + sampleID + ": No R1 found for run: " + runConfig)
     return libraryOutput