finished responding to comments

broadinstitute · Mar 4, 2022 · addef19 · addef19
1 parent cd103b2
commit addef19
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 28 deletions.
diff --git a/...org/broadinstitute/hellbender/tools/walkers/haplotypecaller/AssemblyBasedCallerUtils.java b/...org/broadinstitute/hellbender/tools/walkers/haplotypecaller/AssemblyBasedCallerUtils.java
@@ -499,21 +499,9 @@ static Set<Haplotype> filterPileupHaplotypes(final List<Haplotype> onlyNewHaplot
                                                  final Map<Kmer, Integer> kmerReadCounts,
                                                  final int numPileupHaplotypes,
                                                  final int kmerSize ) {
-//        // make sure we're supposed to look for high entropy
-//        if ( lookForMismatchEntropy &&
-//                pileup.getNumberOfElements() >= minReadsAtLocus &&
-//                (double)mismatchQualities / (double)totalQualities >= mismatchThreshold )
-//            hasPointEvent = true;
-
-        // AH & BG filter
-        // get reads from assemblyResultSet.regionForGenotyping.reads[x].samRecord mReadBases and mBaseQualities and kmerize them.
-        // Map<kmer, Integer> # times Kmer in all the reads
-        // for each kmer in reads - find hapotypes that contain it and
-        // filter haplotypes that don't have 10% coverage of read kmers
-        // check if finalizeRegion methods is already applied and it removes the softclipped bases
-
-        // get haplotypes from assemblyResultSet and kmerize. for each haplotype create a set of kmers.
-        // for each haplotype, look up the kmers in the read-map and sum thee counts fo the haplotype score
+
+        // Get haplotypes from assemblyResultSet and kmerize. for each haplotype create a set of kmers.
+        // for each haplotype, look up the kmers in the read-map and sum the counts fo the haplotype score
         // create a Map<Haplytope, Score>
         LinkedHashMap<Haplotype, Integer> haplotypeScores = new LinkedHashMap<>();
         for (Haplotype haplotype : onlyNewHaplotypes) {
@@ -527,6 +515,7 @@ static Set<Haplotype> filterPileupHaplotypes(final List<Haplotype> onlyNewHaplot
             haplotypeScores.put(haplotype, hapKmerCount);
         }
 
+        // Select the top haplotypes in order of how many of their kmers have supports in the underlying reads
         Map<Haplotype,Integer> topScoringHaplotypes =
                 haplotypeScores.entrySet().stream()
                         .sorted(Collections.reverseOrder(Map.Entry.comparingByValue()))

diff --git a/...va/org/broadinstitute/hellbender/tools/walkers/haplotypecaller/HaplotypeCallerEngine.java b/...va/org/broadinstitute/hellbender/tools/walkers/haplotypecaller/HaplotypeCallerEngine.java
@@ -768,8 +768,6 @@ private boolean containsCalls(final CalledHaplotypes calledHaplotypes) {
      */
     private List<VariantContext> referenceModelForNoVariation(final AssemblyRegion region, final boolean needsToBeFinalized, final List<VariantContext> VCpriors) {
         if ( emitReferenceConfidence() ) {
-            //TODO - why the activeRegion cannot manage its own one-time finalization and filtering?
-            //TODO - perhaps we can remove the last parameter of this method and the three lines bellow?
             if ( needsToBeFinalized ) {
                 AssemblyBasedCallerUtils.finalizeRegion(region, hcArgs.assemblerArgs.errorCorrectReads, hcArgs.dontUseSoftClippedBases, minTailQuality, readsHeader, samplesList, ! hcArgs.doNotCorrectOverlappingBaseQualities, hcArgs.softClipLowQualityEnds, hcArgs.pileupDetectionArgs.usePileupDetection);
             }

diff --git a/...institute/hellbender/tools/walkers/haplotypecaller/PileupDetectionArgumentCollection.java b/...institute/hellbender/tools/walkers/haplotypecaller/PileupDetectionArgumentCollection.java
@@ -44,49 +44,49 @@ public final class PileupDetectionArgumentCollection {
      * Enables detection of indels from the pileups in. (EXPERIMENTAL FEATURE)
      */
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_ENABLE_INDELS, doc = "Pileup Detection: If enabled, pileup detection code will attempt to detect indels missing from assembly. (Requires `--pileup-detection` argument)", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_ENABLE_INDELS, doc = "Pileup Detection: If enabled, pileup detection code will attempt to detect indels missing from assembly. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public boolean detectIndels = false;
 
     @Advanced
     @Hidden
-    @Argument(fullName= "num-artificial-haplotypes-to-add-per-allele", doc = "Pileup Detection: This argument limits the maximum number of novel haplotypes to be added to the assembly haplotypes per pileup allele added", optional = true)
+    @Argument(fullName= "num-artificial-haplotypes-to-add-per-allele", doc = "Pileup Detection: This argument limits the maximum number of novel haplotypes to be added to the assembly haplotypes per pileup allele added. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"'` argument)", optional = true)
     public int numHaplotypesToIterate = 5;
     @Advanced
     @Hidden
-    @Argument(fullName= "artifical-haplotype-filtering-kmer-size", doc = "Pileup Detection: Controls what size to kmerize reads to in order to select best supported artificial haplotypes", optional = true)
+    @Argument(fullName= "artifical-haplotype-filtering-kmer-size", doc = "Pileup Detection: Controls what size to kmerize reads to in order to select best supported artificial haplotypes. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public int filteringKmerSize = 10;
 
     /**
      * Percentage of reads required to support the alt for a variant to be considered
      */
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_SNP_THRESHOLD, doc = "Pileup Detection: Percentage of alt supporting reads in order to consider alt SNP", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_SNP_THRESHOLD, doc = "Pileup Detection: Percentage of alt supporting reads in order to consider alt SNP. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public double snpThreshold = 0.1;
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_INDEL_THRESHOLD, doc = "Pileup Detection: Percentage of alt supporting reads in order to consider alt indel", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_INDEL_THRESHOLD, doc = "Pileup Detection: Percentage of alt supporting reads in order to consider alt indel. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public double indelThreshold = 0.5;
 
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_ABSOLUTE_ALT_DEPTH, doc = "Pileup Detection: Absolute number of alt reads necessary to be included in pileup events", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_ABSOLUTE_ALT_DEPTH, doc = "Pileup Detection: Absolute number of alt reads necessary to be included in pileup events. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public double pileupAbsoluteDepth = 0;
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_INDEL_SNP_BLOCKING_RANGE, doc = "Pileup Detection: Filters out pileup snps within this many bases of an assembled indel", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_INDEL_SNP_BLOCKING_RANGE, doc = "Pileup Detection: Filters out pileup snps within this many bases of an assembled indel. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public int snpAdajacentToAssemblyIndel = 5;
 
     /**
      * Arguments related to the "bad read filtering" where alleles that are supported primarily by reads that fail at least one of a number of heuristics will be filtered out
      */
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_BAD_READ_RATIO_LONG_NAME, doc = "Pileup Detection: Threshold of Alt reads rejected by bad reads heuristics to allow the variant", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_BAD_READ_RATIO_LONG_NAME, doc = "Pileup Detection: Threshold of Alt reads rejected by bad reads heuristics to allow the variant. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public double badReadThreshold = 0.0;
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_PROPER_PAIR_READ_BADNESS_LONG_NAME, doc = "Pileup Detection: Reject alt reads not in proper-pairs", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_PROPER_PAIR_READ_BADNESS_LONG_NAME, doc = "Pileup Detection: Reject alt reads not in proper-pairs. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public boolean badReadProperPair = true;
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_EDIT_DISTANCE_BADNESS_LONG_NAME, doc = "Pileup Detection: Reject alt reads with greater than this fraction edit distance from the reference", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_EDIT_DISTANCE_BADNESS_LONG_NAME, doc = "Pileup Detection: Reject alt reads with greater than this fraction of mismatching bases from the reference (proxied using the NM tag). (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public double badReadEditDistance = 0.08;
     @Hidden
-    @Argument(fullName= PILEUP_DETECTION_CHIMERIC_READ_BADNESS_LONG_NAME, doc = "Pileup Detection: Reject reads that are chimeric or supplementary", optional = true)
+    @Argument(fullName= PILEUP_DETECTION_CHIMERIC_READ_BADNESS_LONG_NAME, doc = "Pileup Detection: Reject reads that are chimeric or supplementary. (Requires '--"+PILEUP_DETECTION_LONG_NAME+"' argument)", optional = true)
     public boolean badReadSecondaryOrSupplementary = true;
     @Hidden
     @Argument(fullName= PILEUP_DETECTION_TLEN_MEAN_LONG_NAME, doc = "Pileup Detection: Mean template length (T LEN) to consider for read badness. Requires '--"+PILEUP_DETECTION_TLEN_STD_LONG_NAME+"' to also be set.", optional = true)