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

Example 1 with VCFFilterHeaderLine

use of htsjdk.variant.vcf.VCFFilterHeaderLine in project gatk by broadinstitute.

the class FilterMutectCalls method onTraversalStart.

@Override
public void onTraversalStart() {
    final VCFHeader inputHeader = getHeaderForVariants();
    final Set<VCFHeaderLine> headerLines = new HashSet<>(inputHeader.getMetaDataInSortedOrder());
    Mutect2FilteringEngine.M_2_FILTER_NAMES.stream().map(GATKVCFHeaderLines::getFilterLine).forEach(headerLines::add);
    headerLines.add(new VCFFilterHeaderLine(Mutect2FilteringEngine.ARTIFACT_IN_NORMAL_FILTER_NAME, "artifact_in_normal"));
    headerLines.add(new VCFFilterHeaderLine(Mutect2FilteringEngine.MEDIAN_BASE_QUALITY_DIFFERENCE_FILTER_NAME, "ref - alt median base quality"));
    headerLines.add(new VCFFilterHeaderLine(Mutect2FilteringEngine.MEDIAN_MAPPING_QUALITY_DIFFERENCE_FILTER_NAME, "ref - alt median mapping quality"));
    headerLines.add(new VCFFilterHeaderLine(Mutect2FilteringEngine.MEDIAN_CLIPPING_DIFFERENCE_FILTER_NAME, "ref - alt median clipping"));
    headerLines.add(new VCFFilterHeaderLine(Mutect2FilteringEngine.MEDIAN_FRAGMENT_LENGTH_DIFFERENCE_FILTER_NAME, "abs(ref - alt) median fragment length"));
    headerLines.add(new VCFFilterHeaderLine(Mutect2FilteringEngine.READ_POSITION_FILTER_NAME, "median distance of alt variants from end of reads"));
    headerLines.add(new VCFFilterHeaderLine(Mutect2FilteringEngine.CONTAMINATION_FILTER_NAME, "contamination"));
    headerLines.addAll(getDefaultToolVCFHeaderLines());
    final VCFHeader vcfHeader = new VCFHeader(headerLines, inputHeader.getGenotypeSamples());
    vcfWriter = createVCFWriter(new File(outputVcf));
    vcfWriter.writeHeader(vcfHeader);
}

Example 2 with VCFFilterHeaderLine

use of htsjdk.variant.vcf.VCFFilterHeaderLine in project gatk by broadinstitute.

the class FilterVcf method doWork.

@Override
protected Object doWork() {
    IOUtil.assertFileIsReadable(INPUT);
    IOUtil.assertFileIsWritable(OUTPUT);
    final List<VariantFilter> variantFilters = CollectionUtil.makeList(new AlleleBalanceFilter(MIN_AB), new FisherStrandFilter(MAX_FS), new QdFilter(MIN_QD));
    final List<GenotypeFilter> genotypeFilters = CollectionUtil.makeList(new GenotypeQualityFilter(MIN_GQ), new DepthFilter(MIN_DP));
    try (final VCFFileReader in = new VCFFileReader(INPUT, false)) {
        final FilterApplyingVariantIterator iterator = new FilterApplyingVariantIterator(in.iterator(), variantFilters, genotypeFilters);
        try (final VariantContextWriter out = new VariantContextWriterBuilder().setOutputFile(OUTPUT).build()) {
            final VCFHeader header = in.getFileHeader();
            header.addMetaDataLine(new VCFFilterHeaderLine("AllGtsFiltered", "Site filtered out because all genotypes are filtered out."));
            header.addMetaDataLine(new VCFFormatHeaderLine("FT", VCFHeaderLineCount.UNBOUNDED, VCFHeaderLineType.String, "Genotype filters."));
            for (final VariantFilter filter : variantFilters) {
                for (final VCFFilterHeaderLine line : filter.headerLines()) {
                    header.addMetaDataLine(line);
                }
            }
            out.writeHeader(in.getFileHeader());
            while (iterator.hasNext()) {
                out.add(iterator.next());
            }
        }
    }
    return null;
}

Example 3 with VCFFilterHeaderLine

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

the class HaloplexParasite method doVcfToVcf.

@Override
protected int doVcfToVcf(String inputName, VcfIterator in, VariantContextWriter out) {
    SamReader samReader = null;
    final List<Mutation> mutations = new ArrayList<>();
    try {
        final Set<File> bamFiles = Files.lines(this.bamList.toPath()).filter(T -> !(T.isEmpty() || T.startsWith("#"))).map(T -> new File(T)).collect(Collectors.toSet());
        final VCFHeader header = new VCFHeader();
        header.setSequenceDictionary(in.getHeader().getSequenceDictionary());
        final VCFFilterHeaderLine filter = new VCFFilterHeaderLine("HALOPLEXPARASITE", "fails Parasite Haloplex Sequence");
        final VCFInfoHeaderLine infoWord = new VCFInfoHeaderLine(filter.getID(), 1, VCFHeaderLineType.String, "Parasite Haloplex Sequence (Word|Count|Fraction)");
        super.addMetaData(header);
        out.writeHeader(header);
        header.addMetaDataLine(filter);
        header.addMetaDataLine(infoWord);
        while (in.hasNext()) {
            final VariantContext ctx = in.next();
            final VariantContextBuilder vcb = new VariantContextBuilder(inputName, ctx.getContig(), ctx.getStart(), ctx.getEnd(), ctx.getAlleles());
            if (!(ctx.isIndel() || ctx.isMixed())) {
                out.add(vcb.make());
                continue;
            }
            if (!vcb.getAlleles().stream().filter(A -> !(A.isSymbolic() || A.isNoCall() || A.length() < this.minClipSize)).findAny().isPresent()) {
                out.add(vcb.make());
                continue;
            }
            final Mutation mut = new Mutation(ctx);
            mutations.add(mut);
        }
        final Counter<String> words = new Counter<>();
        for (final File bamFile : bamFiles) {
            LOG.info("Opening " + bamFile);
            samReader = createSamReaderFactory().open(bamFile);
            for (final Mutation mut : mutations) {
                // words seen in that mutation : don't use a Counter
                final Set<String> mutWords = new HashSet<>();
                /* loop over reads overlapping that mutation */
                final SAMRecordIterator sri = samReader.queryOverlapping(mut.contig, mut.start, mut.end);
                while (sri.hasNext()) {
                    final SAMRecord rec = sri.next();
                    if (rec.getReadUnmappedFlag())
                        continue;
                    if (rec.isSecondaryOrSupplementary())
                        continue;
                    if (rec.getDuplicateReadFlag())
                        continue;
                    if (rec.getReadFailsVendorQualityCheckFlag())
                        continue;
                    final Cigar cigar = rec.getCigar();
                    if (cigar.numCigarElements() == 1)
                        continue;
                    final byte[] bases = rec.getReadBases();
                    int refpos = rec.getUnclippedStart();
                    int readpos = 0;
                    /* scan cigar overlapping that mutation */
                    for (final CigarElement ce : cigar) {
                        final CigarOperator op = ce.getOperator();
                        final int ref_end = refpos + (op.consumesReferenceBases() || op.isClipping() ? ce.getLength() : 0);
                        final int read_end = readpos + (op.consumesReadBases() ? ce.getLength() : 0);
                        /* check clip is large enough */
                        if (op.equals(CigarOperator.S) && ce.getLength() >= this.minClipSize) {
                            /* check clip overlap mutation */
                            if (!(ref_end < mut.start || refpos > mut.end)) {
                                /* break read of soft clip into words */
                                for (int x = 0; x + this.minClipSize <= ce.getLength(); ++x) {
                                    final String substr = new String(bases, readpos + x, this.minClipSize);
                                    if (!substr.contains("N")) {
                                        final String revcomp = AcidNucleics.reverseComplement(substr);
                                        mutWords.add(substr);
                                        if (!revcomp.equals(substr))
                                            mutWords.add(revcomp);
                                    }
                                }
                            }
                        }
                        refpos = ref_end;
                        readpos = read_end;
                    }
                }
                sri.close();
                for (final String w : mutWords) {
                    words.incr(w);
                }
            }
            // end of for(mutations)
            samReader.close();
            samReader = null;
        }
        LOG.info("mutations:" + mutations.size());
        final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(header);
        for (final Mutation mut : mutations) {
            progress.watch(mut.contig, mut.start);
            final VariantContextBuilder vcb = new VariantContextBuilder(inputName, mut.contig, mut.start, mut.end, mut.alleles);
            String worstString = null;
            Double worstFraction = null;
            final double totalWords = words.getTotal();
            for (final Allele a : mut.alleles) {
                if (a.isSymbolic() || a.isNoCall() || a.length() < this.minClipSize)
                    continue;
                for (int x = 0; x + this.minClipSize <= a.length(); ++x) {
                    final String substr = new String(a.getBases(), x, this.minClipSize);
                    final long count = words.count(substr);
                    final double fraction = count / totalWords;
                    if (worstFraction == null || worstFraction < fraction) {
                        worstString = substr + "|" + count + "|" + fraction;
                        worstFraction = fraction;
                    }
                }
            }
            if (worstString != null) {
                vcb.attribute(infoWord.getID(), worstString);
            }
            if (worstFraction != null && worstFraction.doubleValue() >= this.tresholdFraction) {
                vcb.filter(filter.getID());
            }
            out.add(vcb.make());
        }
        progress.finish();
        return RETURN_OK;
    } catch (Exception e) {
        LOG.error(e);
        return -1;
    } finally {
        CloserUtil.close(samReader);
    }
}

Example 4 with VCFFilterHeaderLine

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

the class VcfBurdenFilterGenes method doVcfToVcf.

@Override
protected int doVcfToVcf(final String inputName, final VcfIterator in, final VariantContextWriter out) {
    final VCFHeader header = in.getHeader();
    try {
        final VCFHeader h2 = addMetaData(new VCFHeader(header));
        final VCFFilterHeaderLine filterControlsHeader;
        if (!StringUtil.isBlank(this.filterTag)) {
            filterControlsHeader = new VCFFilterHeaderLine(this.filterTag.trim(), "Genes not in list " + this.geneFile);
            h2.addMetaDataLine(filterControlsHeader);
        } else {
            filterControlsHeader = null;
        }
        final List<String> lookColumns = Arrays.asList("CCDS", "Feature", "ENSP", "Gene", "HGNC", "HGNC_ID", "SYMBOL", "RefSeq");
        final VepPredictionParser vepParser = new VepPredictionParserFactory(header).get();
        final AnnPredictionParser annParser = new AnnPredictionParserFactory(header).get();
        final SAMSequenceDictionaryProgress progess = new SAMSequenceDictionaryProgress(header.getSequenceDictionary()).logger(LOG);
        out.writeHeader(h2);
        while (in.hasNext() && !out.checkError()) {
            final VariantContext ctx = progess.watch(in.next());
            boolean keep = false;
            final VariantContextBuilder vcb = new VariantContextBuilder(ctx);
            // not just set FILTER ?
            if (filterControlsHeader == null) {
                vcb.rmAttribute(vepParser.getTag());
                vcb.rmAttribute(annParser.getTag());
            }
            final List<String> newVepList = new ArrayList<>();
            for (final String predStr : ctx.getAttributeAsList(vepParser.getTag()).stream().map(O -> String.class.cast(O)).collect(Collectors.toList())) {
                final VepPredictionParser.VepPrediction pred = vepParser.parseOnePrediction(ctx, predStr);
                for (final String col : lookColumns) {
                    final String token = pred.getByCol(col);
                    if (!StringUtil.isBlank(token) && this.geneNames.contains(token)) {
                        newVepList.add(predStr);
                        keep = true;
                        break;
                    }
                }
            }
            final List<String> newEffList = new ArrayList<>();
            for (final String predStr : ctx.getAttributeAsList(annParser.getTag()).stream().map(O -> String.class.cast(O)).collect(Collectors.toList())) {
                final AnnPredictionParser.AnnPrediction pred = annParser.parseOnePrediction(predStr);
                final String token = pred.getGeneName();
                if (!StringUtil.isBlank(token) && this.geneNames.contains(token)) {
                    newEffList.add(predStr);
                    keep = true;
                    break;
                }
            }
            // not just set FILTER ?
            if (filterControlsHeader == null) {
                if (!newVepList.isEmpty())
                    vcb.attribute(vepParser.getTag(), newVepList);
                if (!newEffList.isEmpty())
                    vcb.attribute(annParser.getTag(), newEffList);
            }
            if (filterControlsHeader != null) {
                if (!keep) {
                    vcb.filter(filterControlsHeader.getID());
                } else if (!ctx.isFiltered()) {
                    vcb.passFilters();
                }
                out.add(vcb.make());
            } else {
                if (keep)
                    out.add(vcb.make());
            }
        }
        progess.finish();
        return RETURN_OK;
    } catch (final Exception err) {
        LOG.error(err);
        return -1;
    } finally {
        CloserUtil.close(in);
    }
}

Example 5 with VCFFilterHeaderLine

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

the class VcfToSql method read.

private void read(File filename) throws IOException {
    /* insert ATGC */
    this.alleleTable.insert(outputWriter, null, "A");
    this.alleleTable.insert(outputWriter, null, "C");
    this.alleleTable.insert(outputWriter, null, "G");
    this.alleleTable.insert(outputWriter, null, "T");
    /* insert this sample */
    this.vcfFileTable.insert(outputWriter, null, filename);
    final SelectStmt vcffile_id = new SelectStmt(this.vcfFileTable);
    final Map<String, SelectStmt> sample2sampleid = new HashMap<String, SelectStmt>();
    final Map<String, SelectStmt> filter2filterid = new HashMap<String, SelectStmt>();
    final Map<String, SelectStmt> chrom2chromId = new HashMap<String, SelectStmt>();
    final VcfIterator r = VCFUtils.createVcfIteratorFromFile(filename);
    final VCFHeader header = r.getHeader();
    /* parse samples */
    for (final String sampleName : header.getSampleNamesInOrder()) {
        this.sampleTable.insert(outputWriter, null, sampleName);
        SelectStmt sample_id = new SelectStmt(this.sampleTable, "name", sampleName);
        sample2sampleid.put(sampleName, sample_id);
        this.sample2fileTable.insert(outputWriter, null, vcffile_id, sample_id);
    }
    /* parse filters */
    for (final VCFFilterHeaderLine filter : header.getFilterLines()) {
        this.filterTable.insert(outputWriter, null, vcffile_id, filter.getID(), filter.getValue());
        filter2filterid.put(filter.getID(), new SelectStmt(this.filterTable, "name", filter.getID()));
    }
    filter2filterid.put(VCFConstants.PASSES_FILTERS_v4, new SelectStmt(this.filterTable, "name", VCFConstants.PASSES_FILTERS_v4));
    final SAMSequenceDictionary dict = header.getSequenceDictionary();
    if (dict == null) {
        throw new RuntimeException("dictionary missing in VCF");
    }
    /* parse sequence dict */
    for (final SAMSequenceRecord ssr : dict.getSequences()) {
        this.chromosomeTable.insert(outputWriter, null, vcffile_id, ssr.getSequenceName(), ssr.getSequenceLength());
        chrom2chromId.put(ssr.getSequenceName(), new SelectStmt(this.chromosomeTable, "name", ssr.getSequenceName()));
    }
    VepPredictionParser vepPredictionParser = new VepPredictionParserFactory(header).get();
    SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(dict);
    int nVariants = 0;
    while (r.hasNext()) {
        if (this.outputWriter.checkError())
            break;
        VariantContext var = progress.watch(r.next());
        ++nVariants;
        /* insert ref allele */
        this.alleleTable.insert(outputWriter, null, var.getReference().getBaseString());
        /* insert variant */
        this.variantTable.insert(outputWriter, null, vcffile_id, nVariants, chrom2chromId.get(var.getContig()), var.getStart(), (var.hasID() ? var.getID() : null), new SelectStmt(this.alleleTable, "bases", var.getReference().getBaseString()), (var.hasLog10PError() ? var.getPhredScaledQual() : null));
        SelectStmt variant_id = new SelectStmt(variantTable);
        /* insert alternate alleles */
        for (Allele alt : var.getAlternateAlleles()) {
            /* insert alt allele */
            this.alleleTable.insert(outputWriter, null, alt.getBaseString());
            this.variant2altTable.insert(outputWriter, null, variant_id, new SelectStmt(this.alleleTable, "bases", alt.getBaseString()));
        }
        /* insert filters */
        for (final String filter : var.getFilters()) {
            if (filter2filterid.get(filter) == null) {
                throw new IOException("VCF Error: filter " + filter + " is not defined in the VCF header.");
            }
            this.variant2filters.insert(outputWriter, null, variant_id, filter2filterid.get(filter));
        }
        if (!this.ignore_info) {
            for (final VepPrediction pred : vepPredictionParser.getPredictions(var)) {
            /*
					vepPrediction.insert(
							outputWriter,
							null,
							variant_id,
							pred.getEnsemblGene(),
							pred.getEnsemblTranscript(),
							pred.getEnsemblProtein(),
							pred.getSymbol()
							);
					SelectStmt pred_id = new SelectStmt(vepPrediction);
			
					for(SequenceOntologyTree.Term t: pred.getSOTerms())
						{
						String term=t.getAcn().replace(':', '_');
						soTermTable.insert(
								outputWriter,
								null,
								term,
								t.getAcn()
								);//for bioportal compatibility
						SelectStmt term_id = new SelectStmt(soTermTable,"acn",term);
						
						vepPrediction2so.insert(
							outputWriter,
							null,
							pred_id,
							term_id
							);
						}
					*/
            }
        }
        /* insert genotypes */
        for (final String sampleName : sample2sampleid.keySet()) {
            final Genotype g = var.getGenotype(sampleName);
            if (!g.isAvailable() || g.isNoCall())
                continue;
            genotypeTable.insert(outputWriter, null, variant_id, sample2sampleid.get(sampleName), g.isCalled() ? new SelectStmt(this.alleleTable, "bases", g.getAllele(0).getBaseString()) : null, g.isCalled() ? new SelectStmt(this.alleleTable, "bases", g.getAllele(1).getBaseString()) : null, g.hasDP() ? g.getDP() : null, g.hasGQ() ? g.getGQ() : null);
        }
    }
    r.close();
}