Examples with VCFHeader htsjdk.variant.vcf.VCFHeader used on opensource projects

Example 61 with VCFHeader

use of htsjdk.variant.vcf.VCFHeader in project jvarkit by lindenb.

the class VcfCompareCallersOneSample method doWork.

@Override
public int doWork(List<String> args) {
    File inputFile = null;
    List<EqualRangeVcfIterator> listChallengers = new ArrayList<>();
    VariantContextWriter vcw = null;
    VcfIterator in = null;
    try {
        in = super.openVcfIterator(oneFileOrNull(args));
        VCFHeader header = in.getHeader();
        if (header.getNGenotypeSamples() != 1) {
            LOG.error("vcf.must.have.only.one.sample");
            return -1;
        }
        VCFHeader h2 = new VCFHeader(header);
        h2.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "CmdLine", String.valueOf(getProgramCommandLine())));
        h2.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "Version", String.valueOf(getVersion())));
        h2.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "HtsJdkVersion", HtsjdkVersion.getVersion()));
        h2.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "HtsJdkHome", HtsjdkVersion.getHome()));
        SAMSequenceDictionary dict = header.getSequenceDictionary();
        if (dict == null) {
            LOG.error("no.dict.in.vcf");
            return -1;
        }
        Comparator<VariantContext> ctxComparator = VCFUtils.createTidPosComparator(dict);
        /* load files to be challenged */
        for (File cf : this.challengerVcf) {
            // do not challenge vs itself
            if (inputFile != null && inputFile.equals(cf)) {
                LOG.error("Ignoring challenger (self): " + cf);
                continue;
            }
            VcfIterator cin = VCFUtils.createVcfIteratorFromFile(cf);
            VCFHeader ch = cin.getHeader();
            if (ch.getNGenotypeSamples() != 1) {
                LOG.warning("vcf.must.have.only.one.sample");
                cin.close();
                continue;
            }
            if (!header.getSampleNamesInOrder().get(0).equals(ch.getSampleNamesInOrder().get(0))) {
                LOG.warning("Ignoring " + cf + " because not the same sample.");
                cin.close();
                continue;
            }
            SAMSequenceDictionary hdict = ch.getSequenceDictionary();
            if (hdict == null || !SequenceUtil.areSequenceDictionariesEqual(dict, hdict)) {
                LOG.error("not.the.same.sequence.dictionaries");
                return -1;
            }
            listChallengers.add(new EqualRangeVcfIterator(cin, ctxComparator));
        }
        vcw = super.openVariantContextWriter(outputFile);
        vcw.writeHeader(h2);
        SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(dict);
        VariantContext prev_ctx = null;
        while (in.hasNext() && !vcw.checkError()) {
            VariantContext ctx = progress.watch(in.next());
            // check input order
            if (prev_ctx != null && ctxComparator.compare(prev_ctx, ctx) > 0) {
                LOG.error("bad sort order : got\n\t" + prev_ctx + "\nbefore\n\t" + ctx + "\n");
                return -1;
            }
            prev_ctx = ctx;
            int countInOtherFiles = 0;
            for (EqualRangeVcfIterator citer : listChallengers) {
                boolean foundInThatFile = false;
                List<VariantContext> ctxChallenging = citer.next(ctx);
                for (VariantContext ctx2 : ctxChallenging) {
                    if (!ctx2.getReference().equals(ctx.getReference()))
                        continue;
                    boolean ok = true;
                    if (!this.ignoreAlternate) {
                        Set<Allele> myAlt = new HashSet<Allele>(ctx.getAlternateAlleles());
                        myAlt.removeAll(ctx2.getAlternateAlleles());
                        if (!myAlt.isEmpty())
                            ok = false;
                    }
                    if (ok) {
                        foundInThatFile = true;
                        break;
                    }
                }
                countInOtherFiles += (foundInThatFile ? 1 : 0);
            }
            if (countInOtherFiles >= minCountInclusive && countInOtherFiles <= maxCountInclusive) {
                vcw.add(ctx);
            }
        }
        progress.finish();
        return 0;
    } catch (Exception err) {
        LOG.error(err);
        return -1;
    } finally {
        CloserUtil.close(vcw);
        CloserUtil.close(listChallengers);
        CloserUtil.close(in);
    }
}

Example 62 with VCFHeader

use of htsjdk.variant.vcf.VCFHeader in project jvarkit by lindenb.

the class VcfIn method scanUsingTabix.

private int scanUsingTabix(final VariantContextWriter vcw, final String databaseFile, final VcfIterator in2) {
    VCFFileReader tabix = null;
    try {
        tabix = new VCFFileReader(new File(databaseFile), true);
        final VCFHeader header1 = new VCFHeader(in2.getHeader());
        this.addMetaData(header1);
        vcw.writeHeader(header1);
        final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(header1.getSequenceDictionary()).logger(LOG);
        while (in2.hasNext() && !vcw.checkError()) {
            final VariantContext userCtx = progress.watch(in2.next());
            final CloseableIterator<VariantContext> iter = tabix.query(userCtx.getContig(), Math.max(1, userCtx.getStart() - 1), userCtx.getEnd() + 1);
            boolean keep = false;
            while (iter.hasNext()) {
                final VariantContext dbctx = iter.next();
                if (!sameContext(userCtx, dbctx))
                    continue;
                if (!allUserAltFoundInDatabase(userCtx, dbctx))
                    continue;
                keep = true;
                break;
            }
            iter.close();
            addVariant(vcw, userCtx, keep);
            if (vcw.checkError())
                break;
        }
        progress.finish();
        return RETURN_OK;
    } catch (final Exception err) {
        LOG.error(err);
        return -1;
    } finally {
        CloserUtil.close(tabix);
        CloserUtil.close(in2);
    }
}

Example 63 with VCFHeader

use of htsjdk.variant.vcf.VCFHeader in project jvarkit by lindenb.

the class VcfTreePack method scan.

void scan(final VcfIterator iter) {
    final VCFHeader header = iter.getHeader();
    super.bindings.put("header", header);
    final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(header.getSequenceDictionary());
    while (iter.hasNext()) {
        final VariantContext ctx = progress.watch(iter.next());
        bindings.put("variant", ctx);
        super.nodeFactoryChain.watch(super.rootNode, ctx);
    }
    progress.finish();
}

Example 64 with VCFHeader

use of htsjdk.variant.vcf.VCFHeader in project jvarkit by lindenb.

the class VcfLiftOver method doVcfToVcf.

@Override
protected int doVcfToVcf(String inputName, VcfIterator in, VariantContextWriter out) {
    VariantContextWriter failed = null;
    GenomicSequence genomicSequence = null;
    try {
        final VCFHeader inputHeader = in.getHeader();
        final Set<VCFHeaderLine> headerLines = inputHeader.getMetaDataInInputOrder().stream().filter(V -> {
            if (!(V instanceof VCFInfoHeaderLine))
                return true;
            final VCFInfoHeaderLine vih = VCFInfoHeaderLine.class.cast(V);
            if (removeInfo.contains(vih.getID()))
                return false;
            return true;
        }).collect(Collectors.toSet());
        if (this.failedFile != null) {
            final VCFHeader header2 = new VCFHeader(headerLines, inputHeader.getSampleNamesInOrder());
            header2.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "CmdLine", String.valueOf(getProgramCommandLine())));
            header2.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "Version", String.valueOf(getVersion())));
            header2.addMetaDataLine(new VCFInfoHeaderLine(this.failedinfoTag, 1, VCFHeaderLineType.String, "Why the liftOver failed."));
            failed = super.openVariantContextWriter(failedFile);
            failed.writeHeader(header2);
        }
        final VCFHeader header3 = new VCFHeader(headerLines, inputHeader.getSampleNamesInOrder());
        header3.setSequenceDictionary(this.indexedFastaSequenceFile.getSequenceDictionary());
        header3.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "CmdLine", String.valueOf(getProgramCommandLine())));
        header3.addMetaDataLine(new VCFHeaderLine(getClass().getSimpleName() + "Version", String.valueOf(getVersion())));
        header3.addMetaDataLine(new VCFInfoHeaderLine(this.infoTag, 1, VCFHeaderLineType.String, "Chromosome|Position before liftOver."));
        out.writeHeader(header3);
        final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(in.getHeader());
        while (in.hasNext()) {
            VariantContext ctx = progress.watch(in.next());
            if (!this.removeInfo.isEmpty()) {
                VariantContextBuilder vcb = new VariantContextBuilder(ctx);
                for (final String tag : this.removeInfo) vcb.rmAttribute(tag);
                ctx = vcb.make();
            }
            if (ctx.isIndel() && this.ignoreIndels) {
                if (failed != null)
                    failed.add(new VariantContextBuilder(ctx).attribute(this.failedinfoTag, "Indel").make());
                continue;
            }
            if (adaptivematch) {
                double minAlleleLength = Math.min(0, ctx.getAlleles().stream().mapToInt(A -> A.length()).min().orElse(0));
                double maxAlleleLength = Math.max(1, ctx.getAlleles().stream().mapToInt(A -> A.length()).max().orElse(1));
                this.liftOver.setLiftOverMinMatch(minAlleleLength / maxAlleleLength);
            }
            final Interval lifted = liftOver.liftOver(new Interval(ctx.getContig(), ctx.getStart(), ctx.getEnd(), // negative strand
            false, String.join("|", ctx.getContig(), String.valueOf(ctx.getStart()), ctx.getReference().toString())));
            if (lifted == null) {
                if (failed != null)
                    failed.add(new VariantContextBuilder(ctx).attribute(this.failedinfoTag, "LiftOverFailed").make());
            } else if (this.indexedFastaSequenceFile.getSequenceDictionary().getSequence(lifted.getContig()) == null) {
                if (failed != null)
                    failed.add(new VariantContextBuilder(ctx).attribute(this.failedinfoTag, "ContigMissingDictionary|" + lifted.getContig()).make());
            } else {
                boolean alleleAreValidatedVsRef = true;
                // part of the code was copied from picard/liftovervcf
                final Map<Allele, Allele> reverseComplementAlleleMap = new HashMap<>();
                final List<Allele> alleles = new ArrayList<Allele>();
                for (final Allele oldAllele : ctx.getAlleles()) {
                    final Allele fixedAllele;
                    if (oldAllele.isSymbolic() || oldAllele.isNoCall() || oldAllele.equals(Allele.SPAN_DEL)) {
                        alleles.add(oldAllele);
                        continue;
                    } else if (lifted.isPositiveStrand()) {
                        fixedAllele = oldAllele;
                        alleles.add(oldAllele);
                    } else {
                        fixedAllele = Allele.create(SequenceUtil.reverseComplement(oldAllele.getBaseString()), oldAllele.isReference());
                        alleles.add(fixedAllele);
                        reverseComplementAlleleMap.put(oldAllele, fixedAllele);
                    }
                    if (this.checkAlleleSequence) {
                        if (genomicSequence == null || !genomicSequence.getChrom().equals(lifted.getContig())) {
                            genomicSequence = new GenomicSequence(this.indexedFastaSequenceFile, lifted.getContig());
                        }
                        final String alleleStr = fixedAllele.getBaseString();
                        int x = 0;
                        while (x < alleleStr.length() && lifted.getStart() - 1 + x < genomicSequence.length()) {
                            final char refChar = genomicSequence.charAt(lifted.getStart() - 1 + x);
                            if (Character.toLowerCase(refChar) != Character.toLowerCase(alleleStr.charAt(x))) {
                                alleleAreValidatedVsRef = false;
                                break;
                            }
                            ++x;
                        }
                        if (x != alleleStr.length()) {
                            alleleAreValidatedVsRef = false;
                            break;
                        }
                    }
                }
                if (!alleleAreValidatedVsRef) {
                    if (failed != null)
                        failed.add(new VariantContextBuilder(ctx).attribute(this.failedinfoTag, "AlleleMismatchRef").make());
                    continue;
                }
                if (lifted.getEnd() - lifted.getStart() != ctx.getEnd() - ctx.getStart()) {
                    if (failed != null)
                        failed.add(new VariantContextBuilder(ctx).attribute(this.failedinfoTag, "AlleleBadLength|" + lifted.length()).make());
                    continue;
                }
                final VariantContextBuilder vcb = new VariantContextBuilder(ctx.getSource(), lifted.getContig(), lifted.getStart(), lifted.getEnd(), alleles);
                vcb.id(ctx.getID());
                vcb.attributes(ctx.getAttributes());
                vcb.attribute(this.infoTag, ctx.getContig() + "|" + ctx.getStart() + "|" + ctx.getReference().getDisplayString());
                vcb.filters(ctx.getFilters());
                vcb.log10PError(ctx.getLog10PError());
                final GenotypesContext genotypeContext = ctx.getGenotypes();
                final GenotypesContext fixedGenotypes = GenotypesContext.create(genotypeContext.size());
                for (final Genotype genotype : genotypeContext) {
                    final List<Allele> fixedAlleles = new ArrayList<Allele>();
                    for (final Allele allele : genotype.getAlleles()) {
                        final Allele fixedAllele = reverseComplementAlleleMap.containsKey(allele) ? reverseComplementAlleleMap.get(allele) : allele;
                        fixedAlleles.add(fixedAllele);
                    }
                    fixedGenotypes.add(new GenotypeBuilder(genotype).alleles(fixedAlleles).make());
                }
                vcb.genotypes(fixedGenotypes);
                out.add(vcb.make());
            }
        }
        if (failed != null) {
            failed.close();
            failed = null;
        }
        return RETURN_OK;
    } catch (final Exception err) {
        LOG.error(err);
        return -1;
    } finally {
        CloserUtil.close(failed);
    }
}

Example 65 with VCFHeader

use of htsjdk.variant.vcf.VCFHeader in project jvarkit by lindenb.

the class LumpyMoreSamples method doWork.

@Override
public int doWork(final List<String> args) {
    VcfIterator r = null;
    VariantContextWriter vcw = null;
    final Map<String, SamReader> sample2samreaders = new HashMap<>();
    try {
        r = super.openVcfIterator(oneFileOrNull(args));
        final VCFHeader headerIn = r.getHeader();
        final SAMSequenceDictionary dict = headerIn.getSequenceDictionary();
        if (dict == null) {
            LOG.error(JvarkitException.VcfDictionaryMissing.getMessage("input vcf"));
            return -1;
        }
        final SamReaderFactory samReaderFactory = SamReaderFactory.makeDefault().validationStringency(ValidationStringency.LENIENT);
        IOUtil.slurpLines(this.bamFileList).stream().forEach(F -> {
            if (F.trim().isEmpty())
                return;
            final SamReader sr = samReaderFactory.open(SamInputResource.of(F));
            final SAMFileHeader samHeader = sr.getFileHeader();
            final SAMSequenceDictionary dict2 = samHeader.getSequenceDictionary();
            if (dict2 == null) {
                throw new JvarkitException.BamDictionaryMissing(F);
            }
            if (!SequenceUtil.areSequenceDictionariesEqual(dict, dict2)) {
                throw new JvarkitException.DictionariesAreNotTheSame(dict, dict2);
            }
            for (final SAMReadGroupRecord rg : samHeader.getReadGroups()) {
                final String sample = rg.getSample();
                if (StringUtil.isBlank(sample))
                    continue;
                final SamReader reader = sample2samreaders.get(sample);
                if (reader == null) {
                    sample2samreaders.put(sample, reader);
                } else if (reader == sr) {
                    continue;
                } else {
                    throw new JvarkitException.UserError("more than one sample per bam:" + F);
                }
            }
        });
        final Set<String> inVcfSampleNames = new HashSet<>(headerIn.getSampleNamesInOrder());
        final Set<String> outVcfSampleNames = new HashSet<>(inVcfSampleNames);
        outVcfSampleNames.addAll(sample2samreaders.keySet());
        final VCFHeader headerOut = new VCFHeader(headerIn.getMetaDataInInputOrder(), outVcfSampleNames);
        final VCFFormatHeaderLine SU2 = new VCFFormatHeaderLine("SU2", 1, VCFHeaderLineType.Integer, "Number of pieces of evidence supporting the variant");
        final VCFFormatHeaderLine PE2 = new VCFFormatHeaderLine("PE2", 1, VCFHeaderLineType.Integer, "Number of split reads supporting the variant");
        final VCFFormatHeaderLine SR2 = new VCFFormatHeaderLine("SR2", 1, VCFHeaderLineType.Integer, "Number of paired-end reads supporting the variant");
        headerOut.addMetaDataLine(SU2);
        headerOut.addMetaDataLine(PE2);
        headerOut.addMetaDataLine(SR2);
        vcw = super.openVariantContextWriter(this.outputFile);
        vcw.writeHeader(headerOut);
        while (r.hasNext()) {
            final VariantContext ctx = r.next();
            final StructuralVariantType sttype = ctx.getStructuralVariantType();
            if (sttype == null)
                continue;
            final int tid = dict.getSequenceIndex(ctx.getContig());
            final Map<String, Genotype> genotypeMap = new HashMap<>();
            ctx.getGenotypes().stream().forEach(G -> genotypeMap.put(G.getSampleName(), G));
            for (final String sample : sample2samreaders.keySet()) {
                final SamReader samReader = sample2samreaders.get(sample);
                final SupportingReads sr = new SupportingReads();
                switch(sttype) {
                    case DEL:
                        {
                            int pos = ctx.getStart();
                            int[] ci = confidenceIntervalPos(ctx);
                            final QueryInterval left = new QueryInterval(tid, pos - ci[0], pos + ci[1]);
                            int end = ctx.getEnd();
                            ci = confidenceIntervalEnd(ctx);
                            final QueryInterval right = new QueryInterval(tid, end - ci[0], end + ci[1]);
                            for (final SAMRecord rec : extractSupportingReads(ctx, sample, samReader, new QueryInterval[] { left, right })) {
                                final Cigar cigar = rec.getCigar();
                                if (cigar.isLeftClipped()) {
                                    final QueryInterval qi = new QueryInterval(tid, rec.getUnclippedStart(), rec.getStart() - 1);
                                    if (qi.overlaps(left)) {
                                        sr.splitReads++;
                                        if (rec.getReadPairedFlag())
                                            sr.pairedReads++;
                                    }
                                }
                                if (cigar.isRightClipped()) {
                                    final QueryInterval qi = new QueryInterval(tid, rec.getEnd() + 1, rec.getUnclippedEnd());
                                    if (qi.overlaps(right)) {
                                        sr.splitReads++;
                                        if (rec.getReadPairedFlag())
                                            sr.pairedReads++;
                                    }
                                }
                            }
                            break;
                        }
                    default:
                        break;
                }
                final GenotypeBuilder gb;
                if (genotypeMap.containsKey(sample)) {
                    gb = new GenotypeBuilder(genotypeMap.get(sample));
                } else {
                    gb = new GenotypeBuilder(sample, Arrays.asList(Allele.NO_CALL, Allele.NO_CALL));
                }
                gb.attribute(SR2.getID(), sr.splitReads);
                gb.attribute(PE2.getID(), sr.pairedReads);
                gb.attribute(SU2.getID(), 0);
                genotypeMap.put(sample, gb.make());
            }
            final VariantContextBuilder vcb = new VariantContextBuilder(ctx);
            // add missing samples.
            for (final String sampleName : outVcfSampleNames) {
                if (genotypeMap.containsKey(sampleName))
                    continue;
                genotypeMap.put(sampleName, new GenotypeBuilder(sampleName, Arrays.asList(Allele.NO_CALL, Allele.NO_CALL)).make());
            }
            vcb.genotypes(genotypeMap.values());
            vcw.add(vcb.make());
        }
        r.close();
        r = null;
        sample2samreaders.values().stream().forEach(R -> CloserUtil.close(R));
        LOG.info("done");
        return 0;
    } catch (final Exception err) {
        LOG.error(err);
        return -1;
    } finally {
        CloserUtil.close(r);
    }
}