Examples with SAMSequenceDictionaryProgress com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress used on opensource projects

Example 21 with SAMSequenceDictionaryProgress

use of com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress in project jvarkit by lindenb.

the class VcfInjectPedigree method doVcfToVcf.

@Override
protected int doVcfToVcf(final String inputName, final VcfIterator in, final VariantContextWriter delegate) {
    final SAMSequenceDictionaryProgress progess = new SAMSequenceDictionaryProgress(in.getHeader()).logger(LOG);
    final VariantContextWriter out = this.component.open(delegate);
    out.writeHeader(in.getHeader());
    while (in.hasNext()) {
        out.add(progess.watch(in.next()));
    }
    progess.finish();
    out.close();
    return 0;
}

Example 22 with SAMSequenceDictionaryProgress

use of com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress in project jvarkit by lindenb.

the class MiniCaller method doWork.

@Override
public int doWork(final List<String> args) {
    ConcatSam.ConcatSamIterator iter = null;
    try {
        if (this.fastaFile == null) {
            LOG.error("no REF");
            return -1;
        }
        /* load faid */
        final ReferenceGenomeFactory referenceGenomeFactory = new ReferenceGenomeFactory();
        this.referenceGenome = referenceGenomeFactory.openFastaFile(this.fastaFile);
        this.dictionary = this.referenceGenome.getDictionary();
        if (this.dictionary == null) {
            LOG.error(JvarkitException.FastaDictionaryMissing.getMessage(this.fastaFile.getPath()));
        }
        /* create sam record iterator */
        iter = new ConcatSam.Factory().addInterval(this.rgnStr).setEnableUnrollList(true).open(args);
        final SAMFileHeader samFileheader = iter.getFileHeader();
        final SAMSequenceDictionary dict = samFileheader.getSequenceDictionary();
        if (dict == null) {
            LOG.error(JvarkitException.BamDictionaryMissing.getMessage(String.join(", ", args)));
            return -1;
        }
        if (!SequenceUtil.areSequenceDictionariesEqual(dict, this.dictionary)) {
            LOG.error(JvarkitException.DictionariesAreNotTheSame.getMessage(dict, this.dictionary));
            return -1;
        }
        final List<SAMReadGroupRecord> groups = samFileheader.getReadGroups();
        if (groups == null || groups.isEmpty()) {
            LOG.error("No group defined in input");
            return -1;
        }
        final Set<String> sampleSet = groups.stream().map(srgr -> this.samRecordPartition.apply(srgr, samRecordPartition.name())).collect(Collectors.toSet());
        /* create VCF metadata */
        final Set<VCFHeaderLine> metaData = new HashSet<VCFHeaderLine>();
        metaData.add(VCFStandardHeaderLines.getFormatLine(VCFConstants.GENOTYPE_KEY));
        metaData.add(VCFStandardHeaderLines.getFormatLine(VCFConstants.DEPTH_KEY));
        metaData.add(VCFStandardHeaderLines.getInfoLine(VCFConstants.DEPTH_KEY));
        metaData.add(VCFStandardHeaderLines.getInfoLine(VCFConstants.ALLELE_COUNT_KEY));
        metaData.add(VCFStandardHeaderLines.getInfoLine(VCFConstants.ALLELE_NUMBER_KEY));
        metaData.add(VCFStandardHeaderLines.getInfoLine(VCFConstants.ALLELE_FREQUENCY_KEY));
        metaData.add(new VCFFormatHeaderLine("DPG", // one value of each genotype
        VCFHeaderLineCount.G, VCFHeaderLineType.Integer, "Depth for each allele"));
        metaData.add(new VCFFormatHeaderLine("DP4", 4, VCFHeaderLineType.Integer, "Depth ReforAlt|Strand : RF,RR,AF,AR"));
        metaData.add(new VCFInfoHeaderLine("INDEL", 1, VCFHeaderLineType.Flag, "Variant is indel"));
        // addMetaData(metaData);
        final VCFHeader vcfHeader = new VCFHeader(metaData, sampleSet);
        vcfHeader.setSequenceDictionary(this.dictionary);
        /* create variant context */
        this.variantContextWriter = super.openVariantContextWriter(outputFile);
        this.variantContextWriter.writeHeader(vcfHeader);
        ReferenceContig genomicSeq = null;
        SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(this.dictionary);
        for (; ; ) {
            SAMRecord rec = null;
            if (iter.hasNext()) {
                rec = progress.watch(iter.next());
                if (rec.getReadUnmappedFlag())
                    continue;
                if (this.readFilter.filterOut(rec))
                    continue;
                /* flush buffer if needed */
                while (!this.buffer.isEmpty() && (this.buffer.get(0).tid < rec.getReferenceIndex() || (this.buffer.get(0).tid == rec.getReferenceIndex() && (this.buffer.get(0).getEnd()) < rec.getAlignmentStart()))) {
                    this.buffer.remove(0).print();
                }
                /* get genomic sequence at this position */
                if (genomicSeq == null || !genomicSeq.getContig().equals(rec.getContig())) {
                    genomicSeq = this.referenceGenome.getContig(rec.getContig());
                }
                final Cigar cigar = rec.getCigar();
                if (cigar == null)
                    continue;
                int readPos = 0;
                // 0 based-reference
                int refPos0 = rec.getAlignmentStart() - 1;
                final byte[] bases = rec.getReadBases();
                final byte[] quals = rec.getBaseQualities();
                final String sampleName = this.samRecordPartition.getPartion(rec, samRecordPartition.name());
                for (final CigarElement ce : cigar.getCigarElements()) {
                    final CigarOperator op = ce.getOperator();
                    switch(op) {
                        case P:
                            break;
                        case H:
                            break;
                        case S:
                            readPos += ce.getLength();
                            break;
                        // go
                        case N:
                        case D:
                            {
                                if (// we need base before deletion
                                refPos0 > 0) {
                                    char refBase = genomicSeq.charAt(refPos0 - 1);
                                    /* we use base before deletion */
                                    final StringBuilder sb = new StringBuilder(ce.getLength());
                                    sb.append(refBase);
                                    for (int i = 0; i < ce.getLength(); ++i) {
                                        sb.append(genomicSeq.charAt(refPos0 + i));
                                    }
                                    findContext(rec.getReferenceIndex(), // we use base *before deletion */
                                    refPos0 - 1, Allele.create(sb.toString(), true)).getSample(sampleName).getAllele(Allele.create(String.valueOf(refBase), false)).incr(rec.getReadNegativeStrandFlag());
                                }
                                refPos0 += ce.getLength();
                                break;
                            }
                        case I:
                            {
                                if (refPos0 > 0) {
                                    // float qual=0;
                                    char refBase = Character.toUpperCase(genomicSeq.charAt(refPos0 - 1));
                                    final StringBuilder sb = new StringBuilder(1 + ce.getLength());
                                    sb.append(refBase);
                                    for (int i = 0; i < ce.getLength(); ++i) {
                                        sb.append((char) bases[readPos + i]);
                                    // qual+=(readPos + i < quals.length?quals[ readPos + i]:0);
                                    }
                                    findContext(rec.getReferenceIndex(), // we use base *before deletion */
                                    refPos0 - 1, Allele.create(String.valueOf(refBase), true)).getSample(sampleName).getAllele(Allele.create(sb.toString().toUpperCase(), false)).incr(rec.getReadNegativeStrandFlag());
                                }
                                readPos += ce.getLength();
                                break;
                            }
                        case EQ:
                        case M:
                        case X:
                            {
                                for (int i = 0; i < ce.getLength(); ++i) {
                                    findContext(rec.getReferenceIndex(), refPos0 + i, Allele.create(String.valueOf(genomicSeq.charAt(refPos0 + i)), true)).getSample(sampleName).getAllele(Allele.create(String.valueOf((char) bases[readPos + i]), false)).incr(rec.getReadNegativeStrandFlag());
                                }
                                readPos += ce.getLength();
                                refPos0 += ce.getLength();
                                break;
                            }
                        default:
                            throw new IllegalStateException("Case statement didn't deal with cigar op: " + op);
                    }
                }
            } else {
                break;
            }
        }
        while (!buffer.isEmpty()) buffer.remove(0).print();
        progress.finish();
        iter.close();
        iter = null;
        this.variantContextWriter.close();
        this.variantContextWriter = null;
        return RETURN_OK;
    } catch (Exception e) {
        LOG.error(e);
        return -1;
    } finally {
        CloserUtil.close(iter);
        CloserUtil.close(this.referenceGenome);
        CloserUtil.close(this.variantContextWriter);
    }
}

Example 23 with SAMSequenceDictionaryProgress

use of com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress in project jvarkit by lindenb.

the class CaseControlPlot method doWork.

@Override
public int doWork(final List<String> args) {
    ArchiveFactory archiveFactory = null;
    VcfIterator in = null;
    VariantContextWriter teeVariantWriter = null;
    final List<CaseControlExtractor> excractors = new ArrayList<>();
    try {
        in = super.openVcfIterator(oneFileOrNull(args));
        final VCFHeader header = in.getHeader();
        excractors.addAll(parseConfigFile(header));
        final Pedigree pedigree;
        if (this.pedigreeFile != null) {
            pedigree = Pedigree.newParser().parse(this.pedigreeFile);
        } else {
            pedigree = Pedigree.newParser().parse(header);
        }
        if (pedigree == null || pedigree.isEmpty()) {
            LOG.error("No pedigree defined , or it is empty");
            return -1;
        }
        final Set<Pedigree.Person> casepersons = pedigree.getPersons().stream().filter(F -> F.isAffected() && header.getSampleNamesInOrder().indexOf(F.getId()) != -1).collect(Collectors.toSet());
        if (casepersons.isEmpty()) {
            LOG.error("No Affected individuals in pedigree/header");
            return -1;
        }
        final Set<Pedigree.Person> controlpersons = pedigree.getPersons().stream().filter(F -> F.isUnaffected() && header.getSampleNamesInOrder().indexOf(F.getId()) != -1).collect(Collectors.toSet());
        if (controlpersons.isEmpty()) {
            LOG.error("No Unaffected individuals in pedigree/header");
            return -1;
        }
        if (this.teeToStdout) {
            teeVariantWriter = super.openVariantContextWriter(null);
            teeVariantWriter.writeHeader(header);
        }
        archiveFactory = ArchiveFactory.open(this.outputDirOrZip);
        for (final CaseControlExtractor extractor : excractors) {
            extractor.pw = archiveFactory.openWriter(extractor.name);
        }
        final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(header);
        while (in.hasNext()) {
            final VariantContext ctx = progress.watch(in.next());
            if (teeVariantWriter != null)
                teeVariantWriter.add(ctx);
            for (final CaseControlExtractor handler : excractors) {
                handler.visit(ctx, casepersons, controlpersons);
            }
        }
        for (final CaseControlExtractor handler : excractors) {
            handler.close();
        }
        progress.finish();
        if (teeVariantWriter != null) {
            teeVariantWriter.close();
            teeVariantWriter = null;
        }
        in.close();
        in = null;
        archiveFactory.close();
        archiveFactory = null;
        return RETURN_OK;
    } catch (final Exception e) {
        LOG.error(e);
        return -1;
    } finally {
        CloserUtil.close(in);
        CloserUtil.close(archiveFactory);
        CloserUtil.close(teeVariantWriter);
        for (final CaseControlExtractor handler : excractors) {
            handler.close();
        }
    }
}

Example 24 with SAMSequenceDictionaryProgress

use of com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress 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 25 with SAMSequenceDictionaryProgress

use of com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress in project jvarkit by lindenb.

the class VcfBurdenFisherH method doVcfToVcf.

@Override
protected int doVcfToVcf(final String inputName, final VcfIterator r, final VariantContextWriter delegate) {
    final VariantContextWriter w = this.component.open(delegate);
    w.writeHeader(r.getHeader());
    final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(r.getHeader()).logger(LOG);
    while (r.hasNext()) {
        w.add(progress.watch(r.next()));
    }
    progress.finish();
    w.close();
    return 0;
}