Ejemplo n.º 1
0
 private void outputMissingInterval(final IntervalStratification stats) {
   final GenomeLoc interval = stats.getInterval();
   final boolean missing[] = new boolean[interval.size()];
   Arrays.fill(missing, true);
   for (AbstractStratification sample : stats.getElements()) {
     if (hasMissingStatuses(sample)) {
       int pos = 0;
       for (AbstractStratification locus : sample.getElements()) {
         if (locus.callableStatuses().isEmpty()) {
           missing[pos] = false;
         }
         pos++;
       }
     }
   }
   int start = -1;
   boolean insideMissing = false;
   for (int i = 0; i < missing.length; i++) {
     if (missing[i] && !insideMissing) {
       start = interval.getStart() + i;
       insideMissing = true;
     } else if (!missing[i] && insideMissing) {
       final int stop = interval.getStart() + i - 1;
       outputMissingInterval(interval.getContig(), start, stop);
       insideMissing = false;
     }
   }
   if (insideMissing) {
     outputMissingInterval(interval.getContig(), start, interval.getStop());
   }
 }
Ejemplo n.º 2
0
  @Override
  public Long map(
      final RefMetaDataTracker tracker,
      final ReferenceContext ref,
      final AlignmentContext context) {
    GenomeLoc refLocus = ref.getLocus();

    // process and remove any intervals in the map that are don't overlap the current locus anymore
    // and add all new intervals that may overlap this reference locus
    addNewOverlappingIntervals(refLocus);
    outputFinishedIntervals(refLocus, ref.getBase());

    // at this point, all intervals in intervalMap overlap with this locus, so update all of them
    for (IntervalStratification intervalStratification : intervalMap.values())
      intervalStratification.addLocus(context, ref);

    return 1L;
  }
Ejemplo n.º 3
0
  /**
   * Takes the interval, finds it in the stash, prints it to the VCF
   *
   * @param stats The statistics of the interval
   * @param refAllele the reference allele
   */
  private void outputStatsToVCF(final IntervalStratification stats, final Allele refAllele) {
    GenomeLoc interval = stats.getInterval();

    final List<Allele> alleles = new ArrayList<>();
    final Map<String, Object> attributes = new HashMap<>();
    final ArrayList<Genotype> genotypes = new ArrayList<>();

    for (String sample : samples) {
      final GenotypeBuilder gb = new GenotypeBuilder(sample);

      SampleStratification sampleStat = stats.getSampleStatistics(sample);
      gb.attribute(
          GATKVCFConstants.AVG_INTERVAL_DP_BY_SAMPLE_KEY,
          sampleStat.averageCoverage(interval.size()));
      gb.attribute(GATKVCFConstants.LOW_COVERAGE_LOCI, sampleStat.getNLowCoveredLoci());
      gb.attribute(GATKVCFConstants.ZERO_COVERAGE_LOCI, sampleStat.getNUncoveredLoci());
      gb.filters(statusToStrings(stats.getSampleStatistics(sample).callableStatuses(), false));

      genotypes.add(gb.make());
    }
    alleles.add(refAllele);
    alleles.add(SYMBOLIC_ALLELE);
    VariantContextBuilder vcb =
        new VariantContextBuilder(
            "DiagnoseTargets",
            interval.getContig(),
            interval.getStart(),
            interval.getStop(),
            alleles);

    vcb = vcb.log10PError(VariantContext.NO_LOG10_PERROR);
    vcb.filters(new LinkedHashSet<>(statusToStrings(stats.callableStatuses(), true)));

    attributes.put(VCFConstants.END_KEY, interval.getStop());
    attributes.put(GATKVCFConstants.AVG_INTERVAL_DP_KEY, stats.averageCoverage(interval.size()));
    attributes.put(GATKVCFConstants.INTERVAL_GC_CONTENT_KEY, stats.gcContent());

    vcb = vcb.attributes(attributes);
    vcb = vcb.genotypes(genotypes);

    vcfWriter.add(vcb.make());
  }