Java Code Examples for htsjdk.variant.vcf.VCFHeader#getSampleNamesInOrder()

@Override
public void onTraversalStart() {
    super.onTraversalStart();

    if (snpmaskPriority && snpmask == null){
        throw new CommandLineException("Cannot specify --" + SNP_MASK_PRIORITY_LONG_NAME + " without " + " --" + SNP_MASK_LONG_NAME);
    }

    if ( iupacSample != null ) {
        final VCFHeader variantHeader = (VCFHeader) getHeaderForFeatures(variants);
        final ArrayList<String> samples = variantHeader.getSampleNamesInOrder();
        if( samples == null || !samples.contains(iupacSample)) {
            throw new UserException.BadInput("the IUPAC sample specified is not present in the provided VCF file");
        }
    }
}
 

@Override public void setHeader(VCFHeader header) {
	genoFieldDecoders = new BCF2GenotypeFieldDecoders(header);
	fieldDict = BCF2Utils.makeDictionary(header);

	builders = new GenotypeBuilder[header.getNGenotypeSamples()];
	final List<String> genotypeSamples = header.getGenotypeSamples();
	for (int i = 0; i < builders.length; ++i)
		builders[i] = new GenotypeBuilder(genotypeSamples.get(i));

	sampleNamesInOrder = header.getSampleNamesInOrder();
	sampleNameToOffset = header.getSampleNameToOffset();
}
 

private static void validateCNVcallDataSource(final SAMFileHeader headerForReads,
                                              final String sampleId,
                                              final FeatureDataSource<VariantContext> dataSource) {
    final VCFHeader cnvCallHeader = (VCFHeader) dataSource.getHeader();
    final ArrayList<String> sampleNamesInOrder = cnvCallHeader.getSampleNamesInOrder();
    Utils.validate(sampleNamesInOrder.size() == 1, "CNV call VCF should be single sample");
    Utils.validate(sampleNamesInOrder.contains(sampleId), ("CNV call VCF does not contain calls for sample " + sampleId));
    Utils.validate(cnvCallHeader.getSequenceDictionary() != null,
            "CNV calls file does not have a valid sequence dictionary");
    Utils.validate(cnvCallHeader.getSequenceDictionary().isSameDictionary(headerForReads.getSequenceDictionary()),
            "CNV calls file does not have the same sequence dictionary as the read evidence");
}
 

/**
 * This method should not be needed in germline use cases, but should produce an empty list in such cases.
 *
 * Pairing is done as follows:
 *
 * - Using the M2 convention:  "tumor_sample" in the header and "normal_sample" if applicable
 * - If there is only one sample, the pair will be the sample name, "".
 * - The samples have names such as "TUMOR" and "NORMAL".  See {@link SamplePairExtractor#TUMOR_SAMPLE_NAME_LIST} and
 *     {@link SamplePairExtractor#NORMAL_SAMPLE_NAME_LIST}
 *
 * If none of these are fulfilled, the returned list will be empty.
 *
 * @param vcfHeader Never {@code null}.  Use blank constructor in {@link VCFHeader} rather than null.
 * @return Tumor-normal sample names.  In tumor-only cases, the normal sample will have an empty string.
 *  Never {@code null}
 */
public static List<TumorNormalPair> extractPossibleTumorNormalPairs(final VCFHeader vcfHeader) {
    Utils.nonNull(vcfHeader);
    // First attempt the M2 nomenclature
    if (vcfHeader.getMetaDataLine(Mutect2Engine.TUMOR_SAMPLE_KEY_IN_VCF_HEADER) != null) {
        final VCFHeaderLine normalMetaDataLine = vcfHeader.getMetaDataLine(Mutect2Engine.NORMAL_SAMPLE_KEY_IN_VCF_HEADER);
        return Collections.singletonList(new TumorNormalPair(vcfHeader.getMetaDataLine(Mutect2Engine.TUMOR_SAMPLE_KEY_IN_VCF_HEADER).getValue(),
                (normalMetaDataLine == null ?
                        NO_NORMAL : normalMetaDataLine.getValue())));
    }

    // Then try sample names (e.g. "TUMOR", "NORMAL")
    final List<String> sampleNames = vcfHeader.getSampleNamesInOrder();
    if (sampleNames.size() == 1) {
        return Collections.singletonList(new TumorNormalPair(sampleNames.get(0), NO_NORMAL));
    }
    if (sampleNames.size() > 0) {
        final List<String> tumorSamples = sampleNames.stream()
                .filter(s -> TUMOR_SAMPLE_NAME_LIST.contains(s)).collect(Collectors.toList());
        final List<String> normalSamples = sampleNames.stream()
                .filter(s -> NORMAL_SAMPLE_NAME_LIST.contains(s)).collect(Collectors.toList());

        return tumorSamples.stream().flatMap(t ->
                normalSamples.stream().map(n -> new TumorNormalPair(t,n))).collect(Collectors.toList());
    }

    return Collections.emptyList();
}
 

private static VCFHeader warpHeader(VCFHeader in, Map<String, Long> namesAndLength)
    throws IllegalArgumentException {
  Set<VCFHeaderLine> newLines = new HashSet<>();
  boolean hasSource = false;
  VCFHeaderVersion version = DEFAULT_VCF_VERSION;

  for (VCFHeaderLine line : in.getMetaDataInInputOrder()) {
    if (line.getKey().equals("reference")) {
      newLines.add(new VCFHeaderLine(line.getKey(), ARGS.refTargetFASTA));
    } else if (line.getKey().equals("source")) {
      newLines.add(new VCFHeaderLine(line.getKey(), line.getValue() + "_and_"
          + GENOME_WARP_VERSION));
      hasSource = true;
    } else if (line.getKey().equals("fileformat")) {
      version = VCFHeaderVersion.toHeaderVersion(line.getValue());
      if (version == null) {
        throw new IllegalArgumentException("malformed version: " + line.getValue());
      }
    } else if (line.getKey().equals(VCFConstants.CONTIG_HEADER_KEY)) {
      continue;
    } else {
      newLines.add(line);
    }
  }

  if (!hasSource) {
    newLines.add(new VCFHeaderLine("source", GENOME_WARP_VERSION));
  }

  // Add contigs
  int i = 0;
  for (Map.Entry<String, Long> entry : namesAndLength.entrySet()) {
    String currName = entry.getKey();
    long chrSize = entry.getValue();

    newLines.add(new VCFContigHeaderLine(VCFHeaderLine.toStringEncoding(
        createContigEntry(currName, chrSize, ARGS.targetAssembly, ARGS.species)),
        version, VCFConstants.CONTIG_HEADER_KEY, i++));
  }

  return new VCFHeader(newLines, in.getSampleNamesInOrder());
}