Examples with BedLineCodec com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec used on opensource projects

Example 16 with BedLineCodec

use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.

the class Biostar178713 method readBed.

private final void readBed(Collection<BedLine> bed, final LineIterator in) {
    final BedLineCodec codec = new BedLineCodec();
    codec.readActualHeader(in);
    while (in.hasNext()) {
        final BedLine line = codec.decode(in);
        if (line == null)
            continue;
        bed.add(line);
    }
    CloserUtil.close(in);
}

Example 17 with BedLineCodec

use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.

the class Biostar78285 method doWork.

@Override
public int doWork(final List<String> args) {
    if (this.gc_percent_window < 1) {
        LOG.error("Bad GC% window size:" + this.gc_percent_window);
        return -1;
    }
    final List<File> bamFiles = IOUtil.unrollFiles(args.stream().map(F -> new File(F)).collect(Collectors.toCollection(HashSet::new)), ".bam");
    SAMSequenceDictionary dict = null;
    final List<SamReader> samReaders = new ArrayList<>();
    final List<CloseableIterator<SAMRecord>> samIterators = new ArrayList<>();
    final TreeSet<String> samples = new TreeSet<>();
    final String DEFAULT_PARTITION = "UNDEFINED_PARTITION";
    IndexedFastaSequenceFile indexedFastaSequenceFile = null;
    VariantContextWriter out = null;
    try {
        final SamReaderFactory samReaderFactory = SamReaderFactory.makeDefault().validationStringency(ValidationStringency.LENIENT);
        for (final File bamFile : bamFiles) {
            LOG.info("Opening " + bamFile);
            final SamReader samReader = samReaderFactory.open(bamFile);
            samReaders.add(samReader);
            final SAMFileHeader header = samReader.getFileHeader();
            if (header == null) {
                LOG.error("No header in " + bamFile);
                return -1;
            }
            if (header.getSortOrder() != SortOrder.coordinate) {
                LOG.error("Sam file " + bamFile + " is not sorted on coordinate :" + header.getSortOrder());
                return -1;
            }
            samples.addAll(header.getReadGroups().stream().map(RG -> this.partition.apply(RG, DEFAULT_PARTITION)).collect(Collectors.toSet()));
            final SAMSequenceDictionary currDict = header.getSequenceDictionary();
            if (currDict == null) {
                LOG.error("SamFile doesn't contain a SAMSequenceDictionary : " + bamFile);
                return -1;
            }
            if (dict == null) {
                dict = currDict;
            } else if (!SequenceUtil.areSequenceDictionariesEqual(dict, currDict)) {
                LOG.error(JvarkitException.DictionariesAreNotTheSame.getMessage(dict, currDict));
                return -1;
            }
        }
        if (samReaders.isEmpty()) {
            LOG.error("no bam");
            return -1;
        }
        if (dict == null) {
            LOG.error("no dictionary");
            return -1;
        }
        final QueryInterval[] intervals;
        if (this.captureBed != null) {
            LOG.info("Opening " + this.captureBed);
            ContigNameConverter.setDefaultAliases(dict);
            final List<QueryInterval> L = new ArrayList<>();
            final BedLineCodec codec = new BedLineCodec();
            final LineIterator li = IOUtils.openFileForLineIterator(this.captureBed);
            while (li.hasNext()) {
                final BedLine bed = codec.decode(li.next());
                if (bed == null)
                    continue;
                final QueryInterval q = bed.toQueryInterval(dict);
                L.add(q);
            }
            CloserUtil.close(li);
            intervals = QueryInterval.optimizeIntervals(L.toArray(new QueryInterval[L.size()]));
        } else {
            intervals = null;
        }
        for (final SamReader samReader : samReaders) {
            LOG.info("querying " + samReader.getResourceDescription());
            final CloseableIterator<SAMRecord> iter;
            if (intervals == null) {
                iter = samReader.iterator();
            } else {
                iter = samReader.queryOverlapping(intervals);
            }
            samIterators.add(new FilterIterator<SAMRecord>(iter, R -> !R.getReadUnmappedFlag() && !filter.filterOut(R)));
        }
        if (this.refFile != null) {
            LOG.info("opening " + refFile);
            indexedFastaSequenceFile = new IndexedFastaSequenceFile(this.refFile);
            final SAMSequenceDictionary refdict = indexedFastaSequenceFile.getSequenceDictionary();
            ContigNameConverter.setDefaultAliases(refdict);
            if (refdict == null) {
                throw new JvarkitException.FastaDictionaryMissing(this.refFile);
            }
            if (!SequenceUtil.areSequenceDictionariesEqual(dict, refdict)) {
                LOG.error(JvarkitException.DictionariesAreNotTheSame.getMessage(dict, refdict));
                return -1;
            }
        }
        out = openVariantContextWriter(this.outputFile);
        final Set<VCFHeaderLine> metaData = new HashSet<>();
        VCFStandardHeaderLines.addStandardFormatLines(metaData, true, VCFConstants.DEPTH_KEY, VCFConstants.GENOTYPE_KEY);
        VCFStandardHeaderLines.addStandardInfoLines(metaData, true, VCFConstants.DEPTH_KEY);
        metaData.add(new VCFFormatHeaderLine("DF", 1, VCFHeaderLineType.Integer, "Number of Reads on plus strand"));
        metaData.add(new VCFFormatHeaderLine("DR", 1, VCFHeaderLineType.Integer, "Number of Reads on minus strand"));
        metaData.add(new VCFInfoHeaderLine("AVG_DP", 1, VCFHeaderLineType.Float, "Mean depth"));
        metaData.add(new VCFInfoHeaderLine("MEDIAN_DP", 1, VCFHeaderLineType.Float, "Median depth"));
        metaData.add(new VCFInfoHeaderLine("MIN_DP", 1, VCFHeaderLineType.Integer, "Min depth"));
        metaData.add(new VCFInfoHeaderLine("MAX_DP", 1, VCFHeaderLineType.Integer, "Max depth"));
        metaData.add(new VCFHeaderLine(Biostar78285.class.getSimpleName() + ".SamFilter", this.filter.toString()));
        for (final Integer treshold : this.minDepthTresholds) {
            metaData.add(new VCFFilterHeaderLine("DP_LT_" + treshold, "All  genotypes have DP< " + treshold));
            metaData.add(new VCFInfoHeaderLine("NUM_DP_LT_" + treshold, 1, VCFHeaderLineType.Integer, "Number of genotypes having DP< " + treshold));
            metaData.add(new VCFInfoHeaderLine("FRACT_DP_LT_" + treshold, 1, VCFHeaderLineType.Float, "Fraction of genotypes having DP< " + treshold));
        }
        if (indexedFastaSequenceFile != null) {
            metaData.add(new VCFInfoHeaderLine("GC_PERCENT", 1, VCFHeaderLineType.Integer, "GC% window_size:" + this.gc_percent_window));
        }
        final List<Allele> refAlleles = Collections.singletonList(Allele.create("N", true));
        final List<Allele> NO_CALLS = Arrays.asList(Allele.NO_CALL, Allele.NO_CALL);
        final VCFHeader vcfHeader = new VCFHeader(metaData, samples);
        vcfHeader.setSequenceDictionary(dict);
        out.writeHeader(vcfHeader);
        final SAMRecordCoordinateComparator samRecordCoordinateComparator = new SAMRecordCoordinateComparator();
        final PeekableIterator<SAMRecord> peekIter = new PeekableIterator<>(new MergingIterator<>((R1, R2) -> samRecordCoordinateComparator.fileOrderCompare(R1, R2), samIterators));
        final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(dict);
        for (final SAMSequenceRecord ssr : dict.getSequences()) {
            final IntervalTree<Boolean> capturePos;
            if (intervals != null) {
                if (!Arrays.stream(intervals).anyMatch(I -> I.referenceIndex == ssr.getSequenceIndex())) {
                    continue;
                }
                capturePos = new IntervalTree<>();
                Arrays.stream(intervals).filter(I -> I.referenceIndex == ssr.getSequenceIndex()).forEach(I -> capturePos.put(I.start, I.end, true));
                ;
            } else {
                capturePos = null;
            }
            final GenomicSequence genomicSequence;
            if (indexedFastaSequenceFile != null && indexedFastaSequenceFile.getSequenceDictionary().getSequence(ssr.getSequenceName()) != null) {
                genomicSequence = new GenomicSequence(indexedFastaSequenceFile, ssr.getSequenceName());
            } else {
                genomicSequence = null;
            }
            final List<SAMRecord> buffer = new ArrayList<>();
            for (int ssr_pos = 1; ssr_pos <= ssr.getSequenceLength(); ++ssr_pos) {
                if (capturePos != null && !capturePos.overlappers(ssr_pos, ssr_pos).hasNext())
                    continue;
                progress.watch(ssr.getSequenceName(), ssr_pos);
                while (peekIter.hasNext()) {
                    final SAMRecord rec = peekIter.peek();
                    if (rec.getReadUnmappedFlag()) {
                        // consumme
                        peekIter.next();
                        continue;
                    }
                    if (this.filter.filterOut(rec)) {
                        // consumme
                        peekIter.next();
                        continue;
                    }
                    if (rec.getReferenceIndex() < ssr.getSequenceIndex()) {
                        throw new IllegalStateException("should not happen");
                    }
                    if (rec.getReferenceIndex() > ssr.getSequenceIndex()) {
                        break;
                    }
                    if (rec.getAlignmentEnd() < ssr_pos) {
                        throw new IllegalStateException("should not happen");
                    }
                    if (rec.getAlignmentStart() > ssr_pos) {
                        break;
                    }
                    buffer.add(peekIter.next());
                }
                int x = 0;
                while (x < buffer.size()) {
                    final SAMRecord R = buffer.get(x);
                    if (R.getReferenceIndex() != ssr.getSequenceIndex() || R.getAlignmentEnd() < ssr_pos) {
                        buffer.remove(x);
                    } else {
                        x++;
                    }
                }
                final Map<String, PosInfo> count = samples.stream().map(S -> new PosInfo(S)).collect(Collectors.toMap(P -> P.sample, Function.identity()));
                for (final SAMRecord rec : buffer) {
                    if (rec.getReferenceIndex() != ssr.getSequenceIndex())
                        throw new IllegalStateException("should not happen");
                    if (rec.getAlignmentEnd() < ssr_pos)
                        continue;
                    if (rec.getAlignmentStart() > ssr_pos)
                        continue;
                    final Cigar cigar = rec.getCigar();
                    if (cigar == null)
                        continue;
                    int refpos = rec.getAlignmentStart();
                    final String sample = this.partition.getPartion(rec, DEFAULT_PARTITION);
                    for (final CigarElement ce : cigar.getCigarElements()) {
                        if (refpos > ssr_pos)
                            break;
                        final CigarOperator op = ce.getOperator();
                        if (op.consumesReferenceBases()) {
                            if (op.consumesReadBases()) {
                                if (refpos <= ssr_pos && ssr_pos <= refpos + ce.getLength()) {
                                    final PosInfo posInfo = count.get(sample);
                                    if (posInfo != null) {
                                        posInfo.dp++;
                                        if (rec.getReadNegativeStrandFlag()) {
                                            posInfo.negative_strand++;
                                        }
                                    }
                                    break;
                                }
                            }
                            refpos += ce.getLength();
                        }
                    }
                }
                final VariantContextBuilder vcb = new VariantContextBuilder();
                final Set<String> filters = new HashSet<>();
                vcb.chr(ssr.getSequenceName());
                vcb.start(ssr_pos);
                vcb.stop(ssr_pos);
                if (genomicSequence == null) {
                    vcb.alleles(refAlleles);
                } else {
                    vcb.alleles(Collections.singletonList(Allele.create((byte) genomicSequence.charAt(ssr_pos - 1), true)));
                    final GenomicSequence.GCPercent gcp = genomicSequence.getGCPercent(Math.max((ssr_pos - 1) - this.gc_percent_window, 0), Math.min(ssr_pos + this.gc_percent_window, ssr.getSequenceLength()));
                    if (!gcp.isEmpty()) {
                        vcb.attribute("GC_PERCENT", gcp.getGCPercentAsInteger());
                    }
                }
                vcb.attribute(VCFConstants.DEPTH_KEY, (int) count.values().stream().mapToInt(S -> S.dp).sum());
                vcb.genotypes(count.values().stream().map(C -> new GenotypeBuilder(C.sample, NO_CALLS).DP((int) C.dp).attribute("DR", C.negative_strand).attribute("DF", C.dp - C.negative_strand).make()).collect(Collectors.toList()));
                for (final Integer treshold : this.minDepthTresholds) {
                    final int count_lt = (int) count.values().stream().filter(S -> S.dp < treshold).count();
                    if (count_lt == samples.size()) {
                        filters.add("DP_LT_" + treshold);
                    }
                    vcb.attribute("NUM_DP_LT_" + treshold, count_lt);
                    if (!samples.isEmpty()) {
                        vcb.attribute("FRACT_DP_LT_" + treshold, count_lt / (float) samples.size());
                    }
                }
                if (!samples.isEmpty()) {
                    final int[] array = count.values().stream().mapToInt(S -> S.dp).toArray();
                    vcb.attribute("AVG_DP", Percentile.average().evaluate(array));
                    vcb.attribute("MEDIAN_DP", Percentile.median().evaluate(array));
                    vcb.attribute("MIN_DP", (int) Percentile.min().evaluate(array));
                    vcb.attribute("MAX_DP", (int) Percentile.max().evaluate(array));
                }
                if (filters.isEmpty()) {
                    vcb.passFilters();
                } else {
                    vcb.filters(filters);
                }
                out.add(vcb.make());
            }
        }
        progress.finish();
        peekIter.close();
        out.close();
        out = null;
        return 0;
    } catch (final Exception err) {
        LOG.error(err);
        return -1;
    } finally {
        CloserUtil.close(out);
        CloserUtil.close(samIterators);
        CloserUtil.close(samReaders);
        CloserUtil.close(indexedFastaSequenceFile);
    }
}

Example 18 with BedLineCodec

use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.

the class KnownGenesToBedTest method test.

@Test(dataProvider = "src1")
public void test(final String kgfile) throws IOException {
    File out = super.createTmpFile(".bed");
    Assert.assertEquals(new KnownGenesToBed().instanceMain(new String[] { "-o", out.getPath(), kgfile }), 0);
    Assert.assertTrue(out.exists());
    final BedLineCodec codec = new BedLineCodec();
    Assert.assertTrue(IOUtil.slurpLines(out).stream().map(L -> codec.decode(L)).filter(L -> L != null).count() > 0L);
}

Example 19 with BedLineCodec

use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.

the class BamStats05 method readBedFile.

private Map<String, List<Interval>> readBedFile(final File bedFile) throws IOException {
    final Map<String, List<Interval>> gene2interval = new TreeMap<String, List<Interval>>();
    BufferedReader bedIn = null;
    try {
        bedIn = IOUtils.openFileForBufferedReading(bedFile);
        final BedLineCodec codec = new BedLineCodec();
        String line = null;
        while ((line = bedIn.readLine()) != null) {
            if (line.isEmpty() || line.startsWith("#"))
                continue;
            final BedLine bedLine = codec.decode(line);
            if (bedLine == null)
                continue;
            if (bedLine.getColumnCount() < 4) {
                throw new IOException("bad bed line in " + line + " " + bedFile);
            }
            final String chrom = bedLine.getContig();
            final int chromStart1 = bedLine.getStart();
            final int chromEnd1 = bedLine.getEnd();
            final String gene = bedLine.get(3);
            if (gene.isEmpty())
                throw new IOException("bad bed gene in " + line + " " + bedFile);
            List<Interval> intervals = gene2interval.get(gene);
            if (intervals == null) {
                intervals = new ArrayList<>();
                gene2interval.put(gene, intervals);
            } else if (!intervals.get(0).getContig().equals(chrom)) {
                throw new IOException("more than one chromosome for gene:" + gene);
            } else {
                for (final Interval interval : intervals) {
                    if (interval.getEnd() < chromStart1)
                        continue;
                    if (interval.getStart() > chromEnd1)
                        continue;
                    throw new IOException("overlapping region: " + line + " and " + interval);
                }
            }
            intervals.add(new Interval(chrom, chromStart1, chromEnd1));
        }
        bedIn.close();
        return gene2interval;
    } finally {
        CloserUtil.close(bedIn);
    }
}

Example 20 with BedLineCodec

use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.

the class Biostar105754 method run.

private void run(final BufferedReader r) throws IOException {
    final BedLineCodec codec = new BedLineCodec();
    String line;
    while ((line = r.readLine()) != null && !this.out.checkError()) {
        final BedLine bedLine = codec.decode(line);
        if (bedLine == null) {
            continue;
        }
        final String chrom = bedLine.getContig();
        int chromStart0 = bedLine.getStart() - 1;
        int chromEnd0 = bedLine.getEnd();
        if (chrom.isEmpty() || chromStart0 < 0L || chromEnd0 < chromStart0) {
            System.err.println("Bad BED line: " + line);
            continue;
        }
        // extends bed area until something was found
        int chromStart = chromStart0;
        int chromEnd = chromEnd0;
        for (; ; ) {
            BigWigIterator iter = this.bbFileReader.getBigWigIterator(chrom, chromStart, chrom, chromEnd, false);
            if (iter != null) {
                WigItem best = null;
                while (iter.hasNext()) {
                    WigItem wigItem = iter.next();
                    if (best == null || distance(chromStart, chromEnd, best.getStartBase(), best.getEndBase()) > distance(chromStart, chromEnd, wigItem.getStartBase(), wigItem.getEndBase())) {
                        best = wigItem;
                    }
                }
                if (best != null) {
                    this.out.print(best.getChromosome());
                    this.out.print("\t");
                    this.out.print(best.getStartBase());
                    this.out.print("\t");
                    this.out.print(best.getEndBase());
                    this.out.print("\t");
                    this.out.print(best.getWigValue());
                    this.out.print("\t");
                    this.out.print(line);
                    this.out.println();
                    break;
                }
            }
            // extend bed area
            long start2 = chromStart - EXTEND_SHIFT;
            long end2 = chromEnd + EXTEND_SHIFT;
            if (start2 < 0)
                start2 = 0;
            if (end2 > MAX_CHROM_END)
                end2 = MAX_CHROM_END;
            // too wide, break loop
            if (start2 == 0 && end2 == MAX_CHROM_END) {
                LOG.warn("no data found for\t" + line);
                break;
            }
            chromStart = (int) start2;
            chromEnd = (int) end2;
        }
    }
}