Examples with SAMRecordCoordinateComparator htsjdk.samtools.SAMRecordCoordinateComparator used on opensource projects

Example 6 with SAMRecordCoordinateComparator

use of htsjdk.samtools.SAMRecordCoordinateComparator in project gridss by PapenfussLab.

the class AutoClosingMergedIteratorTest method should_terminate_async_reader_thread_for_underlying_stream.

@Test
public void should_terminate_async_reader_thread_for_underlying_stream() throws InterruptedException {
    List<SAMRecord> list = ImmutableList.of(Read(0, 1, 1), Read(0, 1, 1), Read(0, 1, 1), Read(0, 1, 1), Read(0, 2, 1));
    AsyncBufferedIterator<SAMRecord> abi = new AsyncBufferedIterator<SAMRecord>(list.iterator(), 1, 1);
    AutoClosingMergedIterator<SAMRecord> merged = new AutoClosingMergedIterator<SAMRecord>(ImmutableList.of(abi), new SAMRecordCoordinateComparator());
    merged.next();
    Thread.sleep(50);
    assertNotNull(AsyncBufferedIteratorTest.getThreadWithName(abi.getBackgroundThreadName()));
    merged.close();
    Thread.sleep(50);
    assertNull(AsyncBufferedIteratorTest.getThreadWithName(abi.getBackgroundThreadName()));
}

Example 7 with SAMRecordCoordinateComparator

use of htsjdk.samtools.SAMRecordCoordinateComparator in project jvarkit by lindenb.

the class GcPercentAndDepth method doWork.

@Override
public int doWork(final List<String> args) {
    if (this.windowSize <= 0) {
        LOG.error("Bad window size.");
        return -1;
    }
    if (this.windowStep <= 0) {
        LOG.error("Bad window step.");
        return -1;
    }
    if (this.refFile == null) {
        LOG.error("Undefined REF File");
        return -1;
    }
    if (args.isEmpty()) {
        LOG.error("Illegal Number of arguments.");
        return -1;
    }
    ReferenceGenome indexedFastaSequenceFile = null;
    List<SamReader> readers = new ArrayList<SamReader>();
    PrintWriter out = null;
    try {
        LOG.info("Loading " + this.refFile);
        indexedFastaSequenceFile = new ReferenceGenomeFactory().openFastaFile(this.refFile);
        this.samSequenceDictionary = indexedFastaSequenceFile.getDictionary();
        if (this.samSequenceDictionary == null) {
            LOG.error("Cannot get sequence dictionary for " + this.refFile);
            return -1;
        }
        out = super.openFileOrStdoutAsPrintWriter(outPutFile);
        Set<String> all_samples = new TreeSet<String>();
        /* create input, collect sample names */
        for (int optind = 0; optind < args.size(); ++optind) {
            LOG.info("Opening " + args.get(optind));
            final SamReader samFileReaderScan = super.openSamReader(args.get(optind));
            readers.add(samFileReaderScan);
            final SAMFileHeader header = samFileReaderScan.getFileHeader();
            if (!SequenceUtil.areSequenceDictionariesEqual(this.samSequenceDictionary, header.getSequenceDictionary())) {
                LOG.error(JvarkitException.DictionariesAreNotTheSame.getMessage(this.samSequenceDictionary, header.getSequenceDictionary()));
                return -1;
            }
            for (final SAMReadGroupRecord g : header.getReadGroups()) {
                final String sample = this.partition.apply(g);
                if (StringUtil.isBlank(sample)) {
                    LOG.warning("Read group " + g.getId() + " has no sample in merged dictionary");
                    continue;
                }
                all_samples.add(sample);
            }
        }
        LOG.info("N " + this.partition.name() + "=" + all_samples.size());
        /* print header */
        out.print("#");
        if (!this.hide_genomic_index) {
            out.print("id");
            out.print("\t");
        }
        out.print("chrom");
        out.print("\t");
        out.print("start");
        out.print("\t");
        out.print("end");
        out.print("\t");
        out.print("GCPercent");
        for (final String sample : all_samples) {
            out.print("\t");
            out.print(sample);
        }
        out.println();
        final List<RegionCaptured> regionsCaptured = new ArrayList<RegionCaptured>();
        if (bedFile != null) {
            LOG.info("Reading BED:" + bedFile);
            final BedLineCodec bedLineCodec = new BedLineCodec();
            BufferedReader r = IOUtils.openFileForBufferedReading(bedFile);
            r.lines().filter(L -> !L.startsWith("#")).filter(L -> !StringUtil.isBlank(L)).map(L -> bedLineCodec.decode(L)).filter(B -> B != null).forEach(B -> {
                final SAMSequenceRecord ssr = this.samSequenceDictionary.getSequence(B.getContig());
                if (ssr == null) {
                    LOG.warning("Cannot resolve " + B.getContig());
                    return;
                }
                final RegionCaptured roi = new RegionCaptured(ssr, B.getStart() - 1, B.getEnd());
                regionsCaptured.add(roi);
            });
            CloserUtil.close(r);
            LOG.info("end Reading BED:" + bedFile);
            Collections.sort(regionsCaptured);
        } else {
            LOG.info("No capture, peeking everything");
            for (final SAMSequenceRecord ssr : this.samSequenceDictionary.getSequences()) {
                final RegionCaptured roi = new RegionCaptured(ssr, 0, ssr.getSequenceLength());
                regionsCaptured.add(roi);
            }
        }
        final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(this.samSequenceDictionary).logger(LOG);
        ReferenceContig genomicSequence = null;
        for (final RegionCaptured roi : regionsCaptured) {
            if (genomicSequence == null || !genomicSequence.hasName(roi.getContig())) {
                genomicSequence = indexedFastaSequenceFile.getContig(roi.getContig());
                if (genomicSequence == null) {
                    LOG.error(JvarkitException.ContigNotFoundInDictionary.getMessage(roi.getContig(), this.samSequenceDictionary));
                    return -1;
                }
            }
            Map<String, int[]> sample2depth = new HashMap<String, int[]>();
            Map<String, Double> sample2meanDepth = new HashMap<String, Double>();
            for (final String sample : all_samples) {
                int[] depth = new int[roi.length()];
                Arrays.fill(depth, 0);
                sample2depth.put(sample, depth);
            }
            List<CloseableIterator<SAMRecord>> iterators = new ArrayList<CloseableIterator<SAMRecord>>();
            for (final SamReader r : readers) {
                iterators.add(r.query(roi.getContig(), roi.getStart(), roi.getEnd(), false));
            }
            final MergingIterator<SAMRecord> merginIter = new MergingIterator<>(new SAMRecordCoordinateComparator(), iterators);
            while (merginIter.hasNext()) {
                final SAMRecord rec = merginIter.next();
                if (rec.getReadUnmappedFlag())
                    continue;
                if (this.filter.filterOut(rec))
                    continue;
                final String sample = this.partition.getPartion(rec, null);
                if (sample == null)
                    continue;
                final int[] depth = sample2depth.get(sample);
                if (depth == null)
                    continue;
                final Cigar cigar = rec.getCigar();
                if (cigar == null)
                    continue;
                int refpos1 = rec.getAlignmentStart();
                for (final CigarElement ce : cigar.getCigarElements()) {
                    final CigarOperator op = ce.getOperator();
                    if (!op.consumesReferenceBases())
                        continue;
                    if (op.consumesReadBases()) {
                        for (int i = 0; i < ce.getLength(); ++i) {
                            if (refpos1 + i < roi.getStart())
                                continue;
                            if (refpos1 + i > roi.getEnd())
                                break;
                            depth[refpos1 + i - roi.getStart()]++;
                        }
                    }
                    refpos1 += ce.getLength();
                }
            }
            merginIter.close();
            for (final RegionCaptured.SlidingWindow win : roi) {
                double total = 0f;
                int countN = 0;
                for (int pos1 = win.getStart(); pos1 <= win.getEnd(); ++pos1) {
                    switch(genomicSequence.charAt(pos1 - 1)) {
                        case 'c':
                        case 'C':
                        case 'g':
                        case 'G':
                        case 's':
                        case 'S':
                            {
                                total++;
                                break;
                            }
                        case 'n':
                        case 'N':
                            countN++;
                            break;
                        default:
                            break;
                    }
                }
                if (skip_if_contains_N && countN > 0)
                    continue;
                double GCPercent = total / (double) win.length();
                int max_depth_for_win = 0;
                sample2meanDepth.clear();
                for (final String sample : all_samples) {
                    int[] depth = sample2depth.get(sample);
                    double sum = 0;
                    for (int pos = win.getStart(); pos < win.getEnd() && (pos - roi.getStart()) < depth.length; ++pos) {
                        sum += depth[pos - roi.getStart()];
                    }
                    double mean = (sum / (double) depth.length);
                    max_depth_for_win = Math.max(max_depth_for_win, (int) mean);
                    sample2meanDepth.put(sample, mean);
                }
                if (max_depth_for_win < this.min_depth)
                    continue;
                if (!this.hide_genomic_index) {
                    out.print(win.getGenomicIndex());
                    out.print("\t");
                }
                out.print(win.getContig());
                out.print("\t");
                out.print(win.getStart() - 1);
                out.print("\t");
                out.print(win.getEnd());
                out.print("\t");
                out.printf("%.2f", GCPercent);
                for (String sample : all_samples) {
                    out.print("\t");
                    out.printf("%.2f", (double) sample2meanDepth.get(sample));
                }
                out.println();
            }
        }
        progress.finish();
        out.flush();
        return 0;
    } catch (Exception err) {
        LOG.error(err);
        return -1;
    } finally {
        for (SamReader r : readers) CloserUtil.close(r);
        CloserUtil.close(indexedFastaSequenceFile);
        CloserUtil.close(out);
    }
}

Example 8 with SAMRecordCoordinateComparator

use of htsjdk.samtools.SAMRecordCoordinateComparator 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);
    }
}