Java Code Examples for htsjdk.samtools.SAMRecord#getMateReferenceName()

public static ReadRecord from(final SAMRecord record)
{
    ReadRecord read = new ReadRecord(
            record.getReadName(), record.getReferenceName(), record.getStart(), record.getEnd(),
            record.getReadString(), record.getCigar(), record.getInferredInsertSize(), record.getFlags(),
            record.getMateReferenceName(), record.getMateAlignmentStart());

    read.setSuppAlignment(record.getStringAttribute(SUPPLEMENTARY_ATTRIBUTE));
    return read;
}
 

public static Object getValue(SAMRecord record, FIELD_TYPE field, String tagId) {
	if (field == null)
		throw new IllegalArgumentException("Record field is null.");

	switch (field) {
	case QNAME:
		return record.getReadName();
	case FLAG:
		return Integer.toString(record.getFlags());
	case RNAME:
		return record.getReferenceName();
	case POS:
		return Integer.toString(record.getAlignmentStart());
	case MAPQ:
		return Integer.toString(record.getMappingQuality());
	case CIGAR:
		return record.getCigarString();
	case RNEXT:
		return record.getMateReferenceName();
	case PNEXT:
		return Integer.toString(record.getMateAlignmentStart());
	case TLEN:
		return Integer.toString(record.getInferredInsertSize());
	case SEQ:
		return record.getReadString();
	case QUAL:
		return record.getBaseQualityString();

	case TAG:
		if (tagId == null)
			throw new IllegalArgumentException("Tag mismatch reqiues tag id. ");
		return record.getAttribute(tagId);

	default:
		throw new IllegalArgumentException("Unknown record field: " + field.name());
	}
}
 

public Object getValue(SAMRecord record) {
	switch (type) {
	case QNAME:
		return record.getReadName();
	case FLAG:
		return Integer.toString(record.getFlags());
	case RNAME:
		return record.getReferenceName();
	case POS:
		return Integer.toString(record.getAlignmentStart());
	case MAPQ:
		return Integer.toString(record.getMappingQuality());
	case CIGAR:
		return record.getCigarString();
	case RNEXT:
		return record.getMateReferenceName();
	case PNEXT:
		return Integer.toString(record.getMateAlignmentStart());
	case TLEN:
		return Integer.toString(record.getInferredInsertSize());
	case SEQ:
		return record.getReadString();
	case QUAL:
		return record.getBaseQualityString();

	case TAG:
		if (tagId == null)
			throw new IllegalArgumentException("Tag mismatch reqiues tag id. ");
		return record.getAttribute(tagId);

	default:
		throw new IllegalArgumentException("Unknown record field: " + type.name());
	}
}
 

@Override
protected int doWork() {
	IOUtil.assertFileIsReadable(INPUT);
	IOUtil.assertFileIsWritable(OUTPUT);
	buildPatterns();

	SamReader in = SamReaderFactory.makeDefault().open(INPUT);

	SAMFileHeader fileHeader = editSequenceDictionary(in.getFileHeader().clone());
	SamHeaderUtil.addPgRecord(fileHeader, this);
	SAMFileWriter out = new SAMFileWriterFactory().makeSAMOrBAMWriter(fileHeader, true, OUTPUT);
	ProgressLogger progLog=new ProgressLogger(log);

	final boolean sequencesRemoved = fileHeader.getSequenceDictionary().getSequences().size() != in.getFileHeader().getSequenceDictionary().getSequences().size();
	
	FilteredReadsMetric m = new FilteredReadsMetric();
	
	for (final SAMRecord r : in) {
		progLog.record(r);
		if (!filterRead(r)) {
               String sequenceName = stripReferencePrefix(r.getReferenceName());
               String mateSequenceName = null;
               if (r.getMateReferenceIndex() != -1)
				mateSequenceName = stripReferencePrefix(r.getMateReferenceName());
		    if (sequencesRemoved || sequenceName != null) {
		        if (sequenceName == null)
					sequenceName = r.getReferenceName();
                   // Even if sequence name has not been edited, if sequences have been removed, need to set
                   // reference name again to invalidate reference index cache.
                   r.setReferenceName(sequenceName);
               }
               if (r.getMateReferenceIndex() != -1 && (sequencesRemoved || mateSequenceName != null)) {
                   if (mateSequenceName == null)
					mateSequenceName = r.getMateReferenceName();
                   // It's possible that the mate was mapped to a reference sequence that has been dropped from
                   // the sequence dictionary.  If so, set the mate to be unmapped.
                   if (fileHeader.getSequenceDictionary().getSequence(mateSequenceName) != null)
					r.setMateReferenceName(mateSequenceName);
				else {
                       r.setMateUnmappedFlag(true);
                       r.setMateReferenceIndex(SAMRecord.NO_ALIGNMENT_REFERENCE_INDEX);
                       r.setMateAlignmentStart(0);
                   }
               }
			out.addAlignment(r);
			m.READS_ACCEPTED++;
		} else {
			m.READS_REJECTED++;
		}
	}
	// write the summary if the summary file is not null.
	writeSummary(this.SUMMARY, m);
       CloserUtil.close(in);
       out.close();
       FilterProgramUtils.reportAndCheckFilterResults("reads", m.READS_ACCEPTED, m.READS_REJECTED,
			PASSING_READ_THRESHOLD, log);
	return (0);
}
 

/**
 * Find reads in a given range.
 * 
 * <p>
 * TODO add cigar to the list of returned values
 * <p>
 * TODO add pair information to the list of returned values
 * 
 * @param request
 * @return
 * @throws InterruptedException 
 */
@Override
protected void processDataRequest(DataRequest request) throws InterruptedException {
	
	// Read the given region
	CloseableIterator<SAMRecord> iterator = dataSource.query(request.start.chr, request.start.bp.intValue(), request.end.bp.intValue());
	
	// Produce results
	while (iterator.hasNext()) {

		List<Feature> responseList = new LinkedList<Feature>();

		// Split results into chunks
		for (int c = 0; c < RESULT_CHUNK_SIZE && iterator.hasNext(); c++) {
			SAMRecord record = iterator.next();

			// Region for this read
			Region recordRegion = new Region((long) record.getAlignmentStart(), (long) record.getAlignmentEnd(), request.start.chr);

			// Values for this read
			LinkedHashMap<DataType, Object> values = new LinkedHashMap<DataType, Object>();

			Feature read = new Feature(recordRegion, values);

			if (request.getRequestedContents().contains(DataType.ID)) {
				values.put(DataType.ID, record.getReadName());
			}

			if (request.getRequestedContents().contains(DataType.STRAND)) {
				values.put(DataType.STRAND, BamUtils.getStrand(record, coverageType));					
			}

			if (request.getRequestedContents().contains(DataType.QUALITY)) {
				values.put(DataType.QUALITY, record.getBaseQualityString());
			}

			if (request.getRequestedContents().contains(DataType.CIGAR)) {
				Cigar cigar = new Cigar(read, record.getCigar());
				values.put(DataType.CIGAR, cigar);
			}

			// TODO Deal with "=" and "N" in read string
			if (request.getRequestedContents().contains(DataType.SEQUENCE)) {
				String seq = record.getReadString();
				values.put(DataType.SEQUENCE, seq);
			}

			if (request.getRequestedContents().contains(DataType.MATE_POSITION)) {
				
				BpCoord mate = new BpCoord((Long)(long)record.getMateAlignmentStart(),
						new Chromosome(record.getMateReferenceName()));
				
				values.put(DataType.MATE_POSITION, mate);
			}
			
			if (request.getRequestedContents().contains(DataType.BAM_TAG_NH)) {
				Object ng = record.getAttribute("NH");
				if (ng != null) {
					values.put(DataType.BAM_TAG_NH, (Integer)record.getAttribute("NH"));
				}
			}
			
			/*
			 * NOTE! RegionContents created from the same read area has to be equal in methods equals, hash and compareTo. Primary types
			 * should be ok, but objects (including tables) has to be handled in those methods separately. Otherwise tracks keep adding
			 * the same reads to their read sets again and again.
			 */
			responseList.add(read);
		}

		// Send result			
		super.createDataResult(new DataResult(request.getStatus(), responseList));			
	}

	// We are done
	iterator.close();
}