Ejemplo n.º 1
0
  /**
   * Creates an empty GATKSAMRecord with the read's header, read group and mate information, but
   * empty (not-null) fields: - Cigar String - Read Bases - Base Qualities
   *
   * <p>Use this method if you want to create a new empty GATKSAMRecord based on another
   * GATKSAMRecord
   *
   * @param read a read to copy the header from
   * @return a read with no bases but safe for the GATK
   */
  public static GATKSAMRecord emptyRead(GATKSAMRecord read) {
    final GATKSAMRecord emptyRead = new GATKSAMRecord(read.getHeader());
    emptyRead.setReferenceIndex(read.getReferenceIndex());
    emptyRead.setAlignmentStart(0);
    emptyRead.setMappingQuality(0);
    // setting read indexing bin last
    emptyRead.setFlags(read.getFlags());
    emptyRead.setMateReferenceIndex(read.getMateReferenceIndex());
    emptyRead.setMateAlignmentStart(read.getMateAlignmentStart());
    emptyRead.setInferredInsertSize(read.getInferredInsertSize());

    emptyRead.setCigarString("");
    emptyRead.setReadBases(new byte[0]);
    emptyRead.setBaseQualities(new byte[0]);

    SAMReadGroupRecord samRG = read.getReadGroup();
    emptyRead.clearAttributes();
    if (samRG != null) {
      GATKSAMReadGroupRecord rg = new GATKSAMReadGroupRecord(samRG);
      emptyRead.setReadGroup(rg);
    }

    GATKBin.setReadIndexingBin(emptyRead, 0);

    return emptyRead;
  }
Ejemplo n.º 2
0
  // copied from LocusViewTemplate
  protected GATKSAMRecord buildSAMRecord(
      final String readName, final String contig, final int alignmentStart) {
    GATKSAMRecord record = new GATKSAMRecord(header);

    record.setReadName(readName);
    record.setReferenceIndex(dictionary.getSequenceIndex(contig));
    record.setAlignmentStart(alignmentStart);

    record.setCigarString("1M");
    record.setReadString("A");
    record.setBaseQualityString("A");
    record.setReadGroup(readGroup);

    return record;
  }
  @Test(enabled = true)
  public void testReadWithNsRefAfterDeletion() throws FileNotFoundException {

    final IndexedFastaSequenceFile seq =
        new CachingIndexedFastaSequenceFile(new File(b37KGReference));
    final SAMFileHeader header =
        ArtificialSAMUtils.createArtificialSamHeader(seq.getSequenceDictionary());
    final int readLength = 76;

    final GATKSAMRecord read =
        ArtificialSAMUtils.createArtificialRead(header, "myRead", 0, 8975, readLength);
    read.setReadBases(Utils.dupBytes((byte) 'A', readLength));
    read.setBaseQualities(Utils.dupBytes((byte) 30, readLength));
    read.setCigarString("3M414N1D73M");

    final int result =
        ReadUtils.getReadCoordinateForReferenceCoordinateUpToEndOfRead(
            read, 9393, ReadUtils.ClippingTail.LEFT_TAIL);
    Assert.assertEquals(result, 3);
  }
  @DataProvider(name = "HasWellDefinedFragmentSizeData")
  public Object[][] makeHasWellDefinedFragmentSizeData() throws Exception {
    final List<Object[]> tests = new LinkedList<Object[]>();

    // setup a basic read that will work
    final SAMFileHeader header = ArtificialSAMUtils.createArtificialSamHeader();
    final GATKSAMRecord read = ArtificialSAMUtils.createArtificialRead(header, "read1", 0, 10, 10);
    read.setReadPairedFlag(true);
    read.setProperPairFlag(true);
    read.setReadUnmappedFlag(false);
    read.setMateUnmappedFlag(false);
    read.setAlignmentStart(100);
    read.setCigarString("50M");
    read.setMateAlignmentStart(130);
    read.setInferredInsertSize(80);
    read.setFirstOfPairFlag(true);
    read.setReadNegativeStrandFlag(false);
    read.setMateNegativeStrandFlag(true);

    tests.add(new Object[] {"basic case", read.clone(), true});

    {
      final GATKSAMRecord bad1 = (GATKSAMRecord) read.clone();
      bad1.setReadPairedFlag(false);
      tests.add(new Object[] {"not paired", bad1, false});
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setProperPairFlag(false);
      // we currently don't require the proper pair flag to be set
      tests.add(new Object[] {"not proper pair", bad, true});
      //            tests.add( new Object[]{ "not proper pair", bad, false });
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setReadUnmappedFlag(true);
      tests.add(new Object[] {"read is unmapped", bad, false});
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setMateUnmappedFlag(true);
      tests.add(new Object[] {"mate is unmapped", bad, false});
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setMateNegativeStrandFlag(false);
      tests.add(new Object[] {"read and mate both on positive strand", bad, false});
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setReadNegativeStrandFlag(true);
      tests.add(new Object[] {"read and mate both on negative strand", bad, false});
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setInferredInsertSize(0);
      tests.add(new Object[] {"insert size is 0", bad, false});
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setAlignmentStart(1000);
      tests.add(new Object[] {"positve read starts after mate end", bad, false});
    }

    {
      final GATKSAMRecord bad = (GATKSAMRecord) read.clone();
      bad.setReadNegativeStrandFlag(true);
      bad.setMateNegativeStrandFlag(false);
      bad.setMateAlignmentStart(1000);
      tests.add(new Object[] {"negative strand read ends before mate starts", bad, false});
    }

    return tests.toArray(new Object[][] {});
  }