htsjdk.samtools.SamInputResource Java Examples

private SamReader createSamReader(SeekableStream in, SeekableStream inIndex,
		ValidationStringency stringency, boolean useIntelInflater) {
	SamReaderFactory readerFactory = SamReaderFactory.makeDefault()
			.setOption(SamReaderFactory.Option.CACHE_FILE_BASED_INDEXES, true)
			.setOption(SamReaderFactory.Option.EAGERLY_DECODE, false)
			.setUseAsyncIo(false);
	if (stringency != null) {
		readerFactory.validationStringency(stringency);
	}
	SamInputResource resource = SamInputResource.of(in);
	if (inIndex != null) {
		resource.index(inIndex);
	}
	if (useIntelInflater) {
		readerFactory.inflaterFactory(IntelGKLAccessor.newInflatorFactor());
	}
	return readerFactory.open(resource);
}
 

/** Does not close the stream. */
public static SAMFileHeader readSAMHeaderFrom(
	final InputStream in, final Configuration conf)
{
	final ValidationStringency
		stringency = getValidationStringency(conf);
	SamReaderFactory readerFactory = SamReaderFactory.makeDefault()
			.setOption(SamReaderFactory.Option.EAGERLY_DECODE, false)
			.setUseAsyncIo(false);
	if (stringency != null) {
		readerFactory.validationStringency(stringency);
	}

	final ReferenceSource refSource = getReferenceSource(conf);
	if (null != refSource) {
		readerFactory.referenceSource(refSource);
	}
	return readerFactory.open(SamInputResource.of(in)).getFileHeader();
}
 

private static SamReader openBAMReader(SamInputResource resource, ValidationStringency stringency, boolean includeFileSource, long offset) throws IOException {
  SamReaderFactory samReaderFactory = SamReaderFactory
      .makeDefault()
      .validationStringency(stringency)
      .enable(SamReaderFactory.Option.CACHE_FILE_BASED_INDEXES);
  if (includeFileSource) {
    samReaderFactory.enable(SamReaderFactory.Option.INCLUDE_SOURCE_IN_RECORDS);
  }
  if (offset == 0) {
    return samReaderFactory.open(resource);
  }
  LOG.info("Initializing seeking reader with the offset of " + offset);
  SeekingBAMFileReader primitiveReader = new SeekingBAMFileReader(resource,
      false,
      stringency,
      DefaultSAMRecordFactory.getInstance(),
      offset);
  final SeekingReaderAdapter reader =
      new SeekingReaderAdapter(primitiveReader, resource);
  samReaderFactory.reapplyOptions(reader);
  return reader;
}
 

private static void sliceFromURLs(@NotNull URL indexUrl, @NotNull URL bamUrl, @NotNull CommandLine cmd) throws IOException {
    File indexFile = downloadIndex(indexUrl);
    indexFile.deleteOnExit();

    SamReader reader = createFromCommandLine(cmd).open(SamInputResource.of(bamUrl).index(indexFile));

    BAMIndex bamIndex;
    if (indexFile.getPath().contains(".crai")) {
        SeekableStream craiIndex = CRAIIndex.openCraiFileAsBaiStream(indexFile, reader.getFileHeader().getSequenceDictionary());
        bamIndex = new DiskBasedBAMFileIndex(craiIndex, reader.getFileHeader().getSequenceDictionary());
    } else {
        bamIndex = new DiskBasedBAMFileIndex(indexFile, reader.getFileHeader().getSequenceDictionary(), false);
    }

    Optional<Pair<QueryInterval[], BAMFileSpan>> queryIntervalsAndSpan = queryIntervalsAndSpan(reader, bamIndex, cmd);
    Optional<Chunk> unmappedChunk = getUnmappedChunk(bamIndex, HttpUtils.getHeaderField(bamUrl, "Content-Length"), cmd);
    List<Chunk> sliceChunks = sliceChunks(queryIntervalsAndSpan, unmappedChunk);
    SamReader cachingReader = createCachingReader(indexFile, bamUrl, cmd, sliceChunks);

    SAMFileWriter writer = new SAMFileWriterFactory().setCreateIndex(true)
            .makeBAMWriter(reader.getFileHeader(), true, new File(cmd.getOptionValue(OUTPUT)));

    queryIntervalsAndSpan.ifPresent(pair -> {
        LOGGER.info("Slicing bam on bed regions...");
        CloseableIterator<SAMRecord> bedIterator = getIterator(cachingReader, pair.getKey(), pair.getValue().toCoordinateArray());
        writeToSlice(writer, bedIterator);
        LOGGER.info("Done writing bed slices.");
    });

    unmappedChunk.ifPresent(chunk -> {
        LOGGER.info("Slicing unmapped reads...");
        CloseableIterator<SAMRecord> unmappedIterator = cachingReader.queryUnmapped();
        writeToSlice(writer, unmappedIterator);
        LOGGER.info("Done writing unmapped reads.");
    });

    reader.close();
    writer.close();
    cachingReader.close();
}
 

@NotNull
private static SamReader createCachingReader(@NotNull File indexFile, @NotNull URL bamUrl, @NotNull CommandLine cmd,
        @NotNull List<Chunk> sliceChunks) throws IOException {
    OkHttpClient httpClient =
            SlicerHttpClient.create(Integer.parseInt(cmd.getOptionValue(MAX_CONCURRENT_REQUESTS, MAX_CONCURRENT_REQUESTS_DEFAULT)));
    int maxBufferSize = readMaxBufferSize(cmd);

    SamInputResource bamResource =
            SamInputResource.of(new CachingSeekableHTTPStream(httpClient, bamUrl, sliceChunks, maxBufferSize)).index(indexFile);
    SamReaderFactory readerFactory = createFromCommandLine(cmd);

    return readerFactory.open(bamResource);
}
 

/**
 * Entry point for specifically creating a SamReader given a pre-positioned stream, header, and known type
 * @param stream the stream to read from. Must already be performing decompression if required.
 * @param reference the SequencesReader to be used as the reference (required for CRAM files).
 * @param headerOverride the pre-determined SAM header
 * @param assumeType the type of input to assume.
 * @return the SamReader
 * @throws IOException if an I/O problem occurs opening the file
 */
public static SamReader makeSamReader(InputStream stream, SequencesReader reference, SAMFileHeader headerOverride, SamReader.Type assumeType) throws IOException {
  if (assumeType == null) {
    throw new NullPointerException();
  }
  try {
    return getSamReaderFactory(reference)
      .open(SamInputResource.of(stream).header(headerOverride).assumeType(assumeType));
  } catch (final RuntimeIOException e) {
    throw (IOException) e.getCause();
  }
}
 

private SamReader createSamReader(InputStream in, ValidationStringency stringency) {
	SamReaderFactory readerFactory = SamReaderFactory.makeDefault()
			.setOption(SamReaderFactory.Option.EAGERLY_DECODE, false)
			.setUseAsyncIo(false);
	if (stringency != null) {
		readerFactory.validationStringency(stringency);
	}
	return readerFactory.open(SamInputResource.of(in));
}
 

private static SamInputResource openBAMFile(Storage.Objects storageClient, String gcsStoragePath, SeekableStream index) throws IOException {
  SeekableGCSStream s = new SeekableGCSStream(storageClient, gcsStoragePath);
  SamInputResource samInputResource =
      SamInputResource.of(s);

  if (index != null) {
    samInputResource.index(index);
  }

  LOG.info("getReadsFromBAMFile - got input resources");
  return samInputResource;
}
 

/**
 * This is factored out of doWork only for unit testing.
 */
int writeSamText(PrintStream printStream) {
    try {
        final CloseableIterator<SAMRecord> samRecordsIterator;
        final SamReader samReader = SamReaderFactory.makeDefault()
                .referenceSequence(REFERENCE_SEQUENCE)
                .open(SamInputResource.of(INPUT));

        // if we are only using the header or we aren't using intervals, then use the reader as the iterator.
        // otherwise use the SamRecordIntervalIteratorFactory to make an interval-ed iterator
        if (HEADER_ONLY || INTERVAL_LIST == null) {
            samRecordsIterator = samReader.iterator();
        } else {
            IOUtil.assertFileIsReadable(INTERVAL_LIST);

            final List<Interval> intervals = IntervalList.fromFile(INTERVAL_LIST).uniqued().getIntervals();
            samRecordsIterator = new SamRecordIntervalIteratorFactory().makeSamRecordIntervalIterator(samReader, intervals, samReader.hasIndex());
        }
        final AsciiWriter writer = new AsciiWriter(printStream);
        final SAMFileHeader header = samReader.getFileHeader();
        if (!RECORDS_ONLY) {
            if (header.getTextHeader() != null) {
                writer.write(header.getTextHeader());
            } else {
                // Headers that are too large are not retained as text, so need to regenerate text
                new SAMTextHeaderCodec().encode(writer, header, true);
            }
        }
        if (!HEADER_ONLY) {
            while (samRecordsIterator.hasNext()) {
                final SAMRecord rec = samRecordsIterator.next();

                if (printStream.checkError()) {
                    return 1;
                }

                if (this.ALIGNMENT_STATUS == AlignmentStatus.Aligned && rec.getReadUnmappedFlag()) continue;
                if (this.ALIGNMENT_STATUS == AlignmentStatus.Unaligned && !rec.getReadUnmappedFlag()) continue;

                if (this.PF_STATUS == PfStatus.PF && rec.getReadFailsVendorQualityCheckFlag()) continue;
                if (this.PF_STATUS == PfStatus.NonPF && !rec.getReadFailsVendorQualityCheckFlag()) continue;
                writer.write(rec.getSAMString());
            }
        }
        writer.flush();
        if (printStream.checkError()) {
            return 1;
        }
        CloserUtil.close(writer);
        CloserUtil.close(samRecordsIterator);
        return 0;
    } catch (IOException e) {
        throw new PicardException("Exception writing SAM text", e);
    }
}
 

protected int doWork() {
    IOUtil.assertInputIsValid(INPUT);
    IOUtil.assertFileIsWritable(OUTPUT);

    final SamReader in = SamReaderFactory.makeDefault()
        .referenceSequence(REFERENCE_SEQUENCE)
        .open(SamInputResource.of(INPUT));

    // create the read-group we'll be using
    final SAMReadGroupRecord rg = new SAMReadGroupRecord(RGID);
    rg.setLibrary(RGLB);
    rg.setPlatform(RGPL);
    rg.setSample(RGSM);
    rg.setPlatformUnit(RGPU);
    if (RGCN != null) rg.setSequencingCenter(RGCN);
    if (RGDS != null) rg.setDescription(RGDS);
    if (RGDT != null) rg.setRunDate(RGDT);
    if (RGPI != null) rg.setPredictedMedianInsertSize(RGPI);
    if (RGPG != null) rg.setProgramGroup(RGPG);
    if (RGPM != null) rg.setPlatformModel(RGPM);
    if (RGKS != null) rg.setKeySequence(RGKS);
    if (RGFO != null) rg.setFlowOrder(RGFO);

    log.info(String.format("Created read-group ID=%s PL=%s LB=%s SM=%s%n", rg.getId(), rg.getPlatform(), rg.getLibrary(), rg.getSample()));

    // create the new header and output file
    final SAMFileHeader inHeader = in.getFileHeader();
    final SAMFileHeader outHeader = inHeader.clone();
    outHeader.setReadGroups(Collections.singletonList(rg));
    if (SORT_ORDER != null) outHeader.setSortOrder(SORT_ORDER);

    final SAMFileWriter outWriter = new SAMFileWriterFactory().makeSAMOrBAMWriter(outHeader,
            outHeader.getSortOrder() == inHeader.getSortOrder(),
            OUTPUT);

    final ProgressLogger progress = new ProgressLogger(log);
    for (final SAMRecord read : in) {
        read.setAttribute(SAMTag.RG.name(), RGID);
        outWriter.addAlignment(read);
        progress.record(read);
    }

    // cleanup
    CloserUtil.close(in);
    outWriter.close();
    return 0;
}
 

public SeekingReaderAdapter(SeekingBAMFileReader reader, SamInputResource resource){
    super(reader, resource);
    underlyingReader = reader;
}
 

/**
 * Initialize this data source with multiple SAM/BAM/CRAM files, explicit indices for those files,
 * and a custom SamReaderFactory.
 *
 * @param samPaths paths to SAM/BAM/CRAM files, not null
 * @param samIndices indices for all of the SAM/BAM/CRAM files, in the same order as samPaths. May be null,
 *                   in which case index paths are inferred automatically.
 * @param customSamReaderFactory SamReaderFactory to use, if null a default factory with no reference and validation
 *                               stringency SILENT is used.
 * @param cloudWrapper caching/prefetching wrapper for the data, if on Google Cloud.
 * @param cloudIndexWrapper caching/prefetching wrapper for the index, if on Google Cloud.
 */
public ReadsPathDataSource( final List<Path> samPaths, final List<Path> samIndices,
                           SamReaderFactory customSamReaderFactory,
                           Function<SeekableByteChannel, SeekableByteChannel> cloudWrapper,
                           Function<SeekableByteChannel, SeekableByteChannel> cloudIndexWrapper ) {
    Utils.nonNull(samPaths);
    Utils.nonEmpty(samPaths, "ReadsPathDataSource cannot be created from empty file list");

    if ( samIndices != null && samPaths.size() != samIndices.size() ) {
        throw new UserException(String.format("Must have the same number of BAM/CRAM/SAM paths and indices. Saw %d BAM/CRAM/SAMs but %d indices",
                                              samPaths.size(), samIndices.size()));
    }

    readers = new LinkedHashMap<>(samPaths.size() * 2);
    backingPaths = new LinkedHashMap<>(samPaths.size() * 2);
    indicesAvailable = true;

    final SamReaderFactory samReaderFactory =
            customSamReaderFactory == null ?
                SamReaderFactory.makeDefault().validationStringency(ReadConstants.DEFAULT_READ_VALIDATION_STRINGENCY) :
                customSamReaderFactory;

    int samCount = 0;
    for ( final Path samPath : samPaths ) {
        // Ensure each file can be read
        try {
            IOUtil.assertFileIsReadable(samPath);
        }
        catch ( SAMException|IllegalArgumentException e ) {
            throw new UserException.CouldNotReadInputFile(samPath.toString(), e);
        }

        Function<SeekableByteChannel, SeekableByteChannel> wrapper =
            (BucketUtils.isEligibleForPrefetching(samPath)
                ? cloudWrapper
                : Function.identity());
        // if samIndices==null then we'll guess the index name from the file name.
        // If the file's on the cloud, then the search will only consider locations that are also
        // in the cloud.
        Function<SeekableByteChannel, SeekableByteChannel> indexWrapper =
            ((samIndices != null && BucketUtils.isEligibleForPrefetching(samIndices.get(samCount))
             || (samIndices == null && BucketUtils.isEligibleForPrefetching(samPath)))
                ? cloudIndexWrapper
                : Function.identity());

        SamReader reader;
        if ( samIndices == null ) {
            reader = samReaderFactory.open(samPath, wrapper, indexWrapper);
        }
        else {
            final SamInputResource samResource = SamInputResource.of(samPath, wrapper);
            Path indexPath = samIndices.get(samCount);
            samResource.index(indexPath, indexWrapper);
            reader = samReaderFactory.open(samResource);
        }

        // Ensure that each file has an index
        if ( ! reader.hasIndex() ) {
            indicesAvailable = false;
        }

        readers.put(reader, null);
        backingPaths.put(reader, samPath);
        ++samCount;
    }

    // Prepare a header merger only if we have multiple readers
    headerMerger = samPaths.size() > 1 ? createHeaderMerger() : null;
}
 

public static void main(String[] args) throws IOException {
	Params params = new Params();
	JCommander jc = new JCommander(params);
	jc.parse(args);

	Log.setGlobalLogLevel(params.logLevel);

	if (args.length == 0 || params.help) {
		usage(jc);
		System.exit(1);
	}

	if (params.reference == null) {
		System.out.println("Reference file not found, will try downloading...");
	}

	ReferenceSource referenceSource = null;
	if (params.reference != null) {
		System.setProperty("reference", params.reference.getAbsolutePath());
		referenceSource = new ReferenceSource(params.reference);
	} else {
		String prop = System.getProperty("reference");
		if (prop != null) {
			referenceSource = new ReferenceSource(new File(prop));
		}
	}

	SamReaderFactory factory = SamReaderFactory.make().validationStringency(params.validationLevel);
	SamInputResource r;
	if ("file".equalsIgnoreCase(params.url.getProtocol()))
		r = SamInputResource.of(params.url.getPath());
	else
		r = SamInputResource.of(params.url);
	SamReader reader = factory.open(r);
	SAMRecordIterator iterator = reader.iterator();

	SAMFileWriterFactory writerFactory = new SAMFileWriterFactory();
	SAMFileWriter writer = null;
	OutputStream os = new BufferedOutputStream(new FileOutputStream(params.outputFile));
	switch (params.outputFormat) {
	case BAM:
		writer = writerFactory.makeBAMWriter(reader.getFileHeader(),
				reader.getFileHeader().getSortOrder() == SortOrder.coordinate, os);
		break;
	case CRAM:
		writer = writerFactory.makeCRAMWriter(reader.getFileHeader(), os, params.reference);
		break;

	default:
		System.out.println("Unknown output format: " + params.outputFormat);
		System.exit(1);
	}

	while (iterator.hasNext()) {
		writer.addAlignment(iterator.next());
	}
	writer.close();
	reader.close();
}
 

/**
 * Entry point for specifically creating a SamReader given a provided stream and header, but let
 * htsjdk decide the underlying format (including working out whether the input is compressed).
 * @param stream the stream to read from
 * @param reference the SequencesReader to be used as the reference (required for CRAM files).
 * @param headerOverride the pre-determined SAM header (or null to use the header from the stream)
 * @return the SamReader
 * @throws IOException if an I/O problem occurs opening the file
 */
public static SamReader makeSamReader(InputStream stream, SequencesReader reference, SAMFileHeader headerOverride) throws IOException {
  try {
    return getSamReaderFactory(reference).open(SamInputResource.of(stream).header(headerOverride));
  } catch (final RuntimeIOException e) {
    throw (IOException) e.getCause();
  }
}