Merge branch 'master' of ssh://gsa1/humgen/gsa-scr1/gsa-engineering/git/unstable

public/java/src/org/broadinstitute/sting/gatk/walkers/beagle/BeagleOutputToVCFWalker.java

@@ -79,24 +79,45 @@ public class BeagleOutputToVCFWalker  extends RodWalker<Integer, Integer> {
     @ArgumentCollection
     protected StandardVariantContextInputArgumentCollection variantCollection = new StandardVariantContextInputArgumentCollection();
 
+    /**
+     * If this argument is present, the original allele frequencies and counts from this vcf are added as annotations ACH,AFH and ANH. at each record present in this vcf
+     */
     @Input(fullName="comp", shortName = "comp", doc="Comparison VCF file", required=false)
     public RodBinding<VariantContext> comp;
 
+
+    /**
+     * This required argument is used to annotate each site in the vcf INFO field with R2 annotation. Will be NaN if Beagle determined there are no variant samples.
+     */
     @Input(fullName="beagleR2", shortName = "beagleR2", doc="Beagle-produced .r2 file containing R^2 values for all markers", required=true)
     public RodBinding<BeagleFeature> beagleR2;
 
+    /**
+     * These values will populate the GL field for each sample and contain the posterior probability of each genotype given the data after phasing and imputation.
+     */
     @Input(fullName="beagleProbs", shortName = "beagleProbs", doc="Beagle-produced .probs file containing posterior genotype probabilities", required=true)
     public RodBinding<BeagleFeature> beagleProbs;
 
+    /**
+     * By default, all genotypes will be marked in the VCF as "phased", using the "|" separator after Beagle.
+     */
     @Input(fullName="beaglePhased", shortName = "beaglePhased", doc="Beagle-produced .phased file containing phased genotypes", required=true)
     public RodBinding<BeagleFeature> beaglePhased;
 
     @Output(doc="VCF File to which variants should be written",required=true)
     protected VCFWriter vcfWriter = null;
 
+    /**
+     * If this argument is absent, and if Beagle determines that there is no sample in a site that has a variant genotype, the site will be marked as filtered (Default behavior).
+     * If the argument is present, the site won't be marked as filtered under this condition even if there are no variant genotypes.
+     */
     @Argument(fullName="dont_mark_monomorphic_sites_as_filtered", shortName="keep_monomorphic", doc="If provided, we won't filter sites that beagle tags as monomorphic.  Useful for imputing a sample's genotypes from a reference panel" ,required=false)
     public boolean DONT_FILTER_MONOMORPHIC_SITES = false;
 
+    /**
+     * Value between 0 and 1. If the probability of getting a genotype correctly (based on the posterior genotype probabilities and the actual genotype) is below this threshold,
+     * a genotype will be substitute by a no-call.
+     */
     @Argument(fullName="no" +
             "call_threshold", shortName="ncthr", doc="Threshold of confidence at which a genotype won't be called", required=false)
     private double noCallThreshold = 0.0;

public/java/src/org/broadinstitute/sting/gatk/walkers/beagle/ProduceBeagleInputWalker.java

@@ -112,6 +112,9 @@ public class ProduceBeagleInputWalker extends RodWalker<Integer, Integer> {
      @Argument(fullName = "bootstrap_vcf",shortName = "bvcf", doc = "Output a VCF with the records used for bootstrapping filtered out", required = false)
     VCFWriter bootstrapVCFOutput = null;
 
+    /**
+     * If sample gender is known, this flag should be set to true to ensure that Beagle treats male Chr X properly.
+     */
     @Argument(fullName = "checkIsMaleOnChrX", shortName = "checkIsMaleOnChrX", doc = "Set to true when Beagle-ing chrX and want to ensure male samples don't have heterozygous calls.", required = false)
     public boolean CHECK_IS_MALE_ON_CHR_X = false;
 

public/java/src/org/broadinstitute/sting/utils/variantcontext/VariantContext.java

@@ -983,7 +983,7 @@ public class VariantContext implements Feature { // to enable tribble intergrati
      * @return true if it's monomorphic
      */
     public boolean isMonomorphic() {
-        return ! isVariant() || getChromosomeCount(getReference()) == getChromosomeCount();
+        return ! isVariant() || (hasGenotypes() && getHomRefCount() + getNoCallCount() == getNSamples());
     }
 
     /**

public/scala/qscript/org/broadinstitute/sting/queue/qscripts/DataProcessingPipeline.scala

@@ -37,11 +37,6 @@ class DataProcessingPipeline extends QScript {
   * Optional Parameters
   ****************************************************************************/
 
-
-//  @Input(doc="path to Picard's SortSam.jar (if re-aligning a previously processed BAM file)", fullName="path_to_sort_jar", shortName="sort", required=false)
-//  var sortSamJar: File = _
-//
-
   @Input(doc="extra VCF files to use as reference indels for Indel Realignment", fullName="extra_indels", shortName="indels", required=false)
   var indels: List[File] = List()
 
@@ -132,24 +127,6 @@ class DataProcessingPipeline extends QScript {
       }
     }
     return sampleTable.toMap
-
-//    println("\n\n*** INPUT FILES ***\n")
-//    // Creating one file for each sample in the dataset
-//    val sampleBamFiles = scala.collection.mutable.Map.empty[String, File]
-//    for ((sample, flist) <- sampleTable) {
-//
-//      println(sample + ":")
-//      for (f <- flist)
-//        println (f)
-//      println()
-//
-//      val sampleFileName = new File(qscript.outputDir + qscript.projectName + "." + sample + ".list")
-//      sampleBamFiles(sample) = sampleFileName
-//      //add(writeList(flist, sampleFileName))
-//    }
-//    println("*** INPUT FILES ***\n\n")
-//
-//    return sampleBamFiles.toMap
   }
 
   // Rebuilds the Read Group string to give BWA