Пример #1
0
  /**
   * Function that process a set of statuses into strings
   *
   * @param statuses the set of statuses to be converted
   * @return a matching set of strings
   */
  private List<String> statusToStrings(
      Iterable<CallableStatus> statuses, final boolean isInfoField) {
    List<String> output = new LinkedList<>();

    for (CallableStatus status : statuses)
      if (isInfoField || status != CallableStatus.PASS) output.add(status.name());

    return output;
  }
Пример #2
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());
  }
Пример #3
0
 /**
  * Outputs all intervals that are behind the current reference locus
  *
  * @param refLocus the current reference locus
  * @param refBase the reference allele
  */
 private void outputFinishedIntervals(final GenomeLoc refLocus, final byte refBase) {
   // output any intervals that were finished
   final List<GenomeLoc> toRemove = new LinkedList<>();
   for (GenomeLoc key : intervalMap.keySet()) {
     if (key.isBefore(refLocus)) {
       final IntervalStratification intervalStats = intervalMap.get(key);
       outputStatsToVCF(intervalStats, Allele.create(refBase, true));
       if (hasMissingLoci(intervalStats)) {
         outputMissingInterval(intervalStats);
       }
       toRemove.add(key);
     }
   }
   for (GenomeLoc key : toRemove) {
     intervalMap.remove(key);
   }
 }