private Haplotype getHaplotypeFromRead(
final PileupElement p, final int contextSize, final int locus) {
final GATKSAMRecord read = p.getRead();
int readOffsetFromPileup = p.getOffset();
final byte[] haplotypeBases = new byte[contextSize];
Arrays.fill(haplotypeBases, (byte) REGEXP_WILDCARD);
final double[] baseQualities = new double[contextSize];
Arrays.fill(baseQualities, 0.0);
byte[] readBases = read.getReadBases();
readBases =
AlignmentUtils.readToAlignmentByteArray(
read.getCigar(), readBases); // Adjust the read bases based on the Cigar string
byte[] readQuals = read.getBaseQualities();
readQuals =
AlignmentUtils.readToAlignmentByteArray(
read.getCigar(),
readQuals); // Shift the location of the qual scores based on the Cigar string
readOffsetFromPileup =
AlignmentUtils.calcAlignmentByteArrayOffset(
read.getCigar(), p, read.getAlignmentStart(), locus);
final int baseOffsetStart = readOffsetFromPileup - (contextSize - 1) / 2;
for (int i = 0; i < contextSize; i++) {
final int baseOffset = i + baseOffsetStart;
if (baseOffset < 0) {
continue;
}
if (baseOffset >= readBases.length) {
break;
}
if (readQuals[baseOffset] == PileupElement.DELETION_BASE) {
readQuals[baseOffset] = PileupElement.DELETION_QUAL;
}
if (!BaseUtils.isRegularBase(readBases[baseOffset])) {
readBases[baseOffset] = (byte) REGEXP_WILDCARD;
readQuals[baseOffset] = (byte) 0;
} // N's shouldn't be treated as distinct bases
readQuals[baseOffset] = (byte) Math.min((int) readQuals[baseOffset], p.getMappingQual());
if (((int) readQuals[baseOffset]) < 5) {
readQuals[baseOffset] = (byte) 0;
} // quals less than 5 are used as codes and don't have actual probabilistic meaning behind
// them
haplotypeBases[i] = readBases[baseOffset];
baseQualities[i] = (double) readQuals[baseOffset];
}
return new Haplotype(haplotypeBases, baseQualities);
}
/**
* Hard clips away soft clipped bases that are below the given quality threshold
*
* @param read the read
* @param minQual the mininum base quality score to revert the base (inclusive)
* @return a new read without low quality soft clipped bases
*/
public static GATKSAMRecord hardClipLowQualitySoftClips(GATKSAMRecord read, byte minQual) {
int nLeadingSoftClips = read.getAlignmentStart() - read.getSoftStart();
if (read.isEmpty() || nLeadingSoftClips > read.getReadLength())
return GATKSAMRecord.emptyRead(read);
byte[] quals = read.getBaseQualities(EventType.BASE_SUBSTITUTION);
int left = -1;
if (nLeadingSoftClips > 0) {
for (int i = nLeadingSoftClips - 1; i >= 0; i--) {
if (quals[i] >= minQual) left = i;
else break;
}
}
int right = -1;
int nTailingSoftClips = read.getSoftEnd() - read.getAlignmentEnd();
if (nTailingSoftClips > 0) {
for (int i = read.getReadLength() - nTailingSoftClips; i < read.getReadLength(); i++) {
if (quals[i] >= minQual) right = i;
else break;
}
}
GATKSAMRecord clippedRead = read;
if (right >= 0
&& right + 1
< clippedRead
.getReadLength()) // only clip if there are softclipped bases (right >= 0) and the
// first high quality soft clip is not the last base (right+1 <
// readlength)
clippedRead =
hardClipByReadCoordinates(
clippedRead,
right + 1,
clippedRead.getReadLength()
- 1); // first we hard clip the low quality soft clips on the right tail
if (left >= 0
&& left - 1
> 0) // only clip if there are softclipped bases (left >= 0) and the first high quality
// soft clip is not the last base (left-1 > 0)
clippedRead =
hardClipByReadCoordinates(
clippedRead,
0,
left - 1); // then we hard clip the low quality soft clips on the left tail
return clippedRead;
}
/**
* Clips any contiguous tail (left, right or both) with base quality lower than lowQual using the
* desired algorithm.
*
* <p>This function will look for low quality tails and hard clip them away. A low quality tail
* ends when a base has base quality greater than lowQual.
*
* @param algorithm the algorithm to use (HardClip, SoftClip, Write N's,...)
* @param lowQual every base quality lower than or equal to this in the tail of the read will be
* hard clipped
* @return a new read without low quality tails
*/
private GATKSAMRecord clipLowQualEnds(ClippingRepresentation algorithm, byte lowQual) {
if (read.isEmpty()) return read;
final byte[] quals = read.getBaseQualities();
final int readLength = read.getReadLength();
int leftClipIndex = 0;
int rightClipIndex = readLength - 1;
// check how far we can clip both sides
while (rightClipIndex >= 0 && quals[rightClipIndex] <= lowQual) rightClipIndex--;
while (leftClipIndex < readLength && quals[leftClipIndex] <= lowQual) leftClipIndex++;
// if the entire read should be clipped, then return an empty read.
if (leftClipIndex > rightClipIndex) return GATKSAMRecord.emptyRead(read);
if (rightClipIndex < readLength - 1) {
this.addOp(new ClippingOp(rightClipIndex + 1, readLength - 1));
}
if (leftClipIndex > 0) {
this.addOp(new ClippingOp(0, leftClipIndex - 1));
}
return this.clipRead(algorithm);
}
@Test(enabled = false)
public void testCovariateGeneration() {
final String RGID = "id";
final int length = 10;
final RecalibrationArgumentCollection RAC = new RecalibrationArgumentCollection();
GATKSAMRecord read = ReadUtils.createRandomRead(length, false);
GATKSAMReadGroupRecord rg = new GATKSAMReadGroupRecord(RGID);
rg.setPlatform("illumina");
read.setReadGroup(rg);
final byte[] mQuals = read.getBaseQualities(EventType.BASE_SUBSTITUTION);
final byte[] iQuals = read.getBaseQualities(EventType.BASE_INSERTION);
final byte[] dQuals = read.getBaseQualities(EventType.BASE_DELETION);
ReadGroupCovariate rgCov = new ReadGroupCovariate();
QualityScoreCovariate qsCov = new QualityScoreCovariate();
ContextCovariate coCov = new ContextCovariate();
CycleCovariate cyCov = new CycleCovariate();
rgCov.initialize(RAC);
qsCov.initialize(RAC);
coCov.initialize(RAC);
cyCov.initialize(RAC);
Covariate[] requestedCovariates = new Covariate[4];
requestedCovariates[0] = rgCov;
requestedCovariates[1] = qsCov;
requestedCovariates[2] = coCov;
requestedCovariates[3] = cyCov;
ReadCovariates rc = RecalDataManager.computeCovariates(read, requestedCovariates);
// check that the length is correct
Assert.assertEquals(rc.getMismatchesKeySet().length, length);
Assert.assertEquals(rc.getInsertionsKeySet().length, length);
Assert.assertEquals(rc.getDeletionsKeySet().length, length);
for (int i = 0; i < length; i++) {
// check that read group is always the same
Assert.assertEquals(rgCov.formatKey(rc.getMismatchesKeySet(i)[0]), RGID);
Assert.assertEquals(rgCov.formatKey(rc.getInsertionsKeySet(i)[0]), RGID);
Assert.assertEquals(rgCov.formatKey(rc.getDeletionsKeySet(i)[0]), RGID);
// check quality score
Assert.assertEquals(qsCov.formatKey(rc.getMismatchesKeySet(i)[1]), "" + mQuals[i]);
Assert.assertEquals(qsCov.formatKey(rc.getInsertionsKeySet(i)[1]), "" + iQuals[i]);
Assert.assertEquals(qsCov.formatKey(rc.getDeletionsKeySet(i)[1]), "" + dQuals[i]);
// check context
Assert.assertEquals(
coCov.formatKey(rc.getMismatchesKeySet(i)[2]),
ContextCovariateUnitTest.expectedContext(read, i, RAC.MISMATCHES_CONTEXT_SIZE));
Assert.assertEquals(
coCov.formatKey(rc.getInsertionsKeySet(i)[2]),
ContextCovariateUnitTest.expectedContext(read, i, RAC.INDELS_CONTEXT_SIZE));
Assert.assertEquals(
coCov.formatKey(rc.getDeletionsKeySet(i)[2]),
ContextCovariateUnitTest.expectedContext(read, i, RAC.INDELS_CONTEXT_SIZE));
// check cycle
Assert.assertEquals(cyCov.formatKey(rc.getMismatchesKeySet(i)[3]), "" + (i + 1));
Assert.assertEquals(cyCov.formatKey(rc.getInsertionsKeySet(i)[3]), "" + (i + 1));
Assert.assertEquals(cyCov.formatKey(rc.getDeletionsKeySet(i)[3]), "" + (i + 1));
}
}
private double scoreReadAgainstHaplotype(
final PileupElement p, final int contextSize, final Haplotype haplotype, final int locus) {
double expected = 0.0;
double mismatches = 0.0;
// What's the expected mismatch rate under the model that this read is actually sampled from
// this haplotype? Let's assume the consensus base c is a random choice one of A, C, G, or T,
// and that
// the observed base is actually from a c with an error rate e. Since e is the rate at which
// we'd
// see a miscalled c, the expected mismatch rate is really e. So the expected number of
// mismatches
// is just sum_i e_i for i from 1..n for n sites
//
// Now, what's the probabilistic sum of mismatches? Suppose that the base b is equal to c.
// Well, it could
// actually be a miscall in a matching direction, which would happen at a e / 3 rate. If b !=
// c, then
// the chance that it is actually a mismatch is 1 - e, since any of the other 3 options would be
// a mismatch.
// so the probability-weighted mismatch rate is sum_i ( matched ? e_i / 3 : 1 - e_i ) for i = 1
// ... n
final byte[] haplotypeBases = haplotype.getBases();
final GATKSAMRecord read = p.getRead();
byte[] readBases = read.getReadBases();
readBases =
AlignmentUtils.readToAlignmentByteArray(
p.getRead().getCigar(), readBases); // Adjust the read bases based on the Cigar string
byte[] readQuals = read.getBaseQualities();
readQuals =
AlignmentUtils.readToAlignmentByteArray(
p.getRead().getCigar(),
readQuals); // Shift the location of the qual scores based on the Cigar string
int readOffsetFromPileup = p.getOffset();
readOffsetFromPileup =
AlignmentUtils.calcAlignmentByteArrayOffset(
p.getRead().getCigar(), p, read.getAlignmentStart(), locus);
final int baseOffsetStart = readOffsetFromPileup - (contextSize - 1) / 2;
for (int i = 0; i < contextSize; i++) {
final int baseOffset = i + baseOffsetStart;
if (baseOffset < 0) {
continue;
}
if (baseOffset >= readBases.length) {
break;
}
final byte haplotypeBase = haplotypeBases[i];
final byte readBase = readBases[baseOffset];
final boolean matched =
(readBase == haplotypeBase || haplotypeBase == (byte) REGEXP_WILDCARD);
byte qual = readQuals[baseOffset];
if (qual == PileupElement.DELETION_BASE) {
qual = PileupElement.DELETION_QUAL;
} // calcAlignmentByteArrayOffset fills the readQuals array with DELETION_BASE at deletions
qual = (byte) Math.min((int) qual, p.getMappingQual());
if (((int) qual)
>= 5) { // quals less than 5 are used as codes and don't have actual probabilistic meaning
// behind them
final double e = QualityUtils.qualToErrorProb(qual);
expected += e;
mismatches += matched ? e : 1.0 - e / 3.0;
}
// a more sophisticated calculation would include the reference quality, but it's nice to
// actually penalize
// the mismatching of poorly determined regions of the consensus
}
return mismatches - expected;
}