/**
* Returns a context identical to this with the REF and ALT alleles reverse complemented.
*
* @param vc variant context
* @return new vc
*/
public static VariantContext reverseComplement(VariantContext vc) {
// create a mapping from original allele to reverse complemented allele
HashMap<Allele, Allele> alleleMap = new HashMap<Allele, Allele>(vc.getAlleles().size());
for (Allele originalAllele : vc.getAlleles()) {
Allele newAllele;
if (originalAllele.isNoCall() || originalAllele.isNull()) newAllele = originalAllele;
else
newAllele =
Allele.create(
BaseUtils.simpleReverseComplement(originalAllele.getBases()),
originalAllele.isReference());
alleleMap.put(originalAllele, newAllele);
}
// create new Genotype objects
GenotypesContext newGenotypes = GenotypesContext.create(vc.getNSamples());
for (final Genotype genotype : vc.getGenotypes()) {
List<Allele> newAlleles = new ArrayList<Allele>();
for (Allele allele : genotype.getAlleles()) {
Allele newAllele = alleleMap.get(allele);
if (newAllele == null) newAllele = Allele.NO_CALL;
newAlleles.add(newAllele);
}
newGenotypes.add(Genotype.modifyAlleles(genotype, newAlleles));
}
return new VariantContextBuilder(vc).alleles(alleleMap.values()).genotypes(newGenotypes).make();
}
/**
* Helper method to subset a VC record, modifying some metadata stored in the INFO field (i.e. AN,
* AC, AF).
*
* @param vc the VariantContext record to subset
* @param samples the samples to extract
* @return the subsetted VariantContext
*/
private VariantContext subsetRecord(VariantContext vc, Set<String> samples) {
if (samples == null || samples.isEmpty()) return vc;
ArrayList<Genotype> genotypes = new ArrayList<Genotype>();
for (Map.Entry<String, Genotype> genotypePair : vc.getGenotypes().entrySet()) {
if (samples.contains(genotypePair.getKey())) genotypes.add(genotypePair.getValue());
}
VariantContext sub = vc.subContextFromGenotypes(genotypes, vc.getAlleles());
// if we have fewer alternate alleles in the selected VC than in the original VC, we need to
// strip out the GL/PLs (because they are no longer accurate)
if (vc.getAlleles().size() != sub.getAlleles().size())
sub = VariantContext.modifyGenotypes(sub, VariantContextUtils.stripPLs(vc.getGenotypes()));
HashMap<String, Object> attributes = new HashMap<String, Object>(sub.getAttributes());
int depth = 0;
for (String sample : sub.getSampleNames()) {
Genotype g = sub.getGenotype(sample);
if (g.isNotFiltered() && g.isCalled()) {
String dp = (String) g.getAttribute("DP");
if (dp != null
&& !dp.equals(VCFConstants.MISSING_DEPTH_v3)
&& !dp.equals(VCFConstants.MISSING_VALUE_v4)) {
depth += Integer.valueOf(dp);
}
}
}
if (KEEP_ORIGINAL_CHR_COUNTS) {
if (attributes.containsKey(VCFConstants.ALLELE_COUNT_KEY))
attributes.put("AC_Orig", attributes.get(VCFConstants.ALLELE_COUNT_KEY));
if (attributes.containsKey(VCFConstants.ALLELE_FREQUENCY_KEY))
attributes.put("AF_Orig", attributes.get(VCFConstants.ALLELE_FREQUENCY_KEY));
if (attributes.containsKey(VCFConstants.ALLELE_NUMBER_KEY))
attributes.put("AN_Orig", attributes.get(VCFConstants.ALLELE_NUMBER_KEY));
}
VariantContextUtils.calculateChromosomeCounts(sub, attributes, false);
attributes.put("DP", depth);
sub = VariantContext.modifyAttributes(sub, attributes);
return sub;
}
@Test(dataProvider = "mergeFiltered")
public void testMergeFiltered(MergeFilteredTest cfg) {
final List<String> priority = vcs2priority(cfg.inputs);
final VariantContext merged =
VariantContextUtils.simpleMerge(
genomeLocParser,
cfg.inputs,
priority,
cfg.type,
VariantContextUtils.GenotypeMergeType.PRIORITIZE,
true,
false,
"set",
false,
false);
// test alleles are equal
Assert.assertEquals(merged.getAlleles(), cfg.expected.getAlleles());
// test set field
Assert.assertEquals(merged.getAttribute("set"), cfg.setExpected);
// test filter field
Assert.assertEquals(merged.getFilters(), cfg.expected.getFilters());
}
@Test(dataProvider = "mergeAlleles")
public void testMergeAlleles(MergeAllelesTest cfg) {
final List<VariantContext> inputs = new ArrayList<VariantContext>();
int i = 0;
for (final List<Allele> alleles : cfg.inputs) {
final String name = "vcf" + ++i;
inputs.add(makeVC(name, alleles));
}
final List<String> priority = vcs2priority(inputs);
final VariantContext merged =
VariantContextUtils.simpleMerge(
genomeLocParser,
inputs,
priority,
VariantContextUtils.FilteredRecordMergeType.KEEP_IF_ANY_UNFILTERED,
VariantContextUtils.GenotypeMergeType.PRIORITIZE,
false,
false,
"set",
false,
false);
Assert.assertEquals(merged.getAlleles(), cfg.expected);
}
protected void printVerboseData(
String pos,
VariantContext vc,
double PofF,
double phredScaledConfidence,
final GenotypeLikelihoodsCalculationModel.Model model) {
Allele refAllele = null, altAllele = null;
for (Allele allele : vc.getAlleles()) {
if (allele.isReference()) refAllele = allele;
else altAllele = allele;
}
for (int i = 0; i <= N; i++) {
StringBuilder AFline = new StringBuilder("AFINFO\t");
AFline.append(pos);
AFline.append("\t");
AFline.append(refAllele);
AFline.append("\t");
if (altAllele != null) AFline.append(altAllele);
else AFline.append("N/A");
AFline.append("\t");
AFline.append(i + "/" + N + "\t");
AFline.append(String.format("%.2f\t", ((float) i) / N));
AFline.append(String.format("%.8f\t", getAlleleFrequencyPriors(model)[i]));
verboseWriter.println(AFline.toString());
}
verboseWriter.println("P(f>0) = " + PofF);
verboseWriter.println("Qscore = " + phredScaledConfidence);
verboseWriter.println();
}
protected final void printCallInfo(
final VariantContext vc,
final double[] log10AlleleFrequencyPriors,
final long runtimeNano,
final AFCalcResult result) {
printCallElement(vc, "type", "ignore", vc.getType());
int allelei = 0;
for (final Allele a : vc.getAlleles())
printCallElement(vc, "allele", allelei++, a.getDisplayString());
for (final Genotype g : vc.getGenotypes())
printCallElement(vc, "PL", g.getSampleName(), g.getLikelihoodsString());
for (int priorI = 0; priorI < log10AlleleFrequencyPriors.length; priorI++)
printCallElement(vc, "priorI", priorI, log10AlleleFrequencyPriors[priorI]);
printCallElement(vc, "runtime.nano", "ignore", runtimeNano);
printCallElement(vc, "log10PosteriorOfAFEq0", "ignore", result.getLog10PosteriorOfAFEq0());
printCallElement(vc, "log10PosteriorOfAFGt0", "ignore", result.getLog10PosteriorOfAFGT0());
for (final Allele allele : result.getAllelesUsedInGenotyping()) {
if (allele.isNonReference()) {
printCallElement(vc, "MLE", allele, result.getAlleleCountAtMLE(allele));
printCallElement(
vc, "pNonRefByAllele", allele, result.getLog10PosteriorOfAFGt0ForAllele(allele));
}
}
callReport.flush();
}
@Override
public String toString() {
return String.format(
"ExactCall %s:%d alleles=%s nSamples=%s orig.pNonRef=%.2f orig.runtime=%s",
vc.getChr(),
vc.getStart(),
vc.getAlleles(),
vc.getNSamples(),
originalCall.getLog10PosteriorOfAFGT0(),
new AutoFormattingTime(runtime / 1e9).toString());
}
/**
* Copy constructor
*
* @param other the VariantContext to copy
*/
protected VariantContext(VariantContext other) {
this(
other.getSource(),
other.getID(),
other.getChr(),
other.getStart(),
other.getEnd(),
other.getAlleles(),
other.getGenotypes(),
other.getLog10PError(),
other.getFiltersMaybeNull(),
other.getAttributes(),
other.REFERENCE_BASE_FOR_INDEL,
NO_VALIDATION);
}
private VariantCallContext generateEmptyContext(
RefMetaDataTracker tracker,
ReferenceContext ref,
Map<String, AlignmentContext> stratifiedContexts,
AlignmentContext rawContext) {
VariantContext vc;
if (UAC.GenotypingMode
== GenotypeLikelihoodsCalculationModel.GENOTYPING_MODE.GENOTYPE_GIVEN_ALLELES) {
VariantContext vcInput =
UnifiedGenotyperEngine.getVCFromAllelesRod(
tracker, ref, rawContext.getLocation(), false, logger, UAC.alleles);
if (vcInput == null) return null;
vc =
new VariantContextBuilder(
"UG_call",
ref.getLocus().getContig(),
vcInput.getStart(),
vcInput.getEnd(),
vcInput.getAlleles())
.make();
} else {
// deal with bad/non-standard reference bases
if (!Allele.acceptableAlleleBases(new byte[] {ref.getBase()})) return null;
Set<Allele> alleles = new HashSet<Allele>();
alleles.add(Allele.create(ref.getBase(), true));
vc =
new VariantContextBuilder(
"UG_call",
ref.getLocus().getContig(),
ref.getLocus().getStart(),
ref.getLocus().getStart(),
alleles)
.make();
}
if (annotationEngine != null) {
// Note: we want to use the *unfiltered* and *unBAQed* context for the annotations
final ReadBackedPileup pileup = rawContext.getBasePileup();
stratifiedContexts = AlignmentContextUtils.splitContextBySampleName(pileup);
vc = annotationEngine.annotateContext(tracker, ref, stratifiedContexts, vc);
}
return new VariantCallContext(vc, false);
}
private static AlleleMapper resolveIncompatibleAlleles(
Allele refAllele, VariantContext vc, Set<Allele> allAlleles) {
if (refAllele.equals(vc.getReference())) return new AlleleMapper(vc);
else {
// we really need to do some work. The refAllele is the longest reference allele seen at this
// start site. So imagine it is:
//
// refAllele: ACGTGA
// myRef: ACGT
// myAlt: -
//
// We need to remap all of the alleles in vc to include the extra GA so that
// myRef => refAllele and myAlt => GA
//
Allele myRef = vc.getReference();
if (refAllele.length() <= myRef.length())
throw new ReviewedStingException(
"BUG: myRef=" + myRef + " is longer than refAllele=" + refAllele);
byte[] extraBases =
Arrays.copyOfRange(refAllele.getBases(), myRef.length(), refAllele.length());
// System.out.printf("Remapping allele at %s%n", vc);
// System.out.printf("ref %s%n", refAllele);
// System.out.printf("myref %s%n", myRef );
// System.out.printf("extrabases %s%n", new String(extraBases));
Map<Allele, Allele> map = new HashMap<Allele, Allele>();
for (Allele a : vc.getAlleles()) {
if (a.isReference()) map.put(a, refAllele);
else {
Allele extended = Allele.extend(a, extraBases);
for (Allele b : allAlleles) if (extended.equals(b)) extended = b;
// System.out.printf(" Extending %s => %s%n", a, extended);
map.put(a, extended);
}
}
// debugging
// System.out.printf("mapping %s%n", map);
return new AlleleMapper(map);
}
}
@Test(dataProvider = "mergeGenotypes")
public void testMergeGenotypes(MergeGenotypesTest cfg) {
final VariantContext merged =
VariantContextUtils.simpleMerge(
genomeLocParser,
cfg.inputs,
cfg.priority,
VariantContextUtils.FilteredRecordMergeType.KEEP_IF_ANY_UNFILTERED,
VariantContextUtils.GenotypeMergeType.PRIORITIZE,
true,
false,
"set",
false,
false);
// test alleles are equal
Assert.assertEquals(merged.getAlleles(), cfg.expected.getAlleles());
// test genotypes
assertGenotypesAreMostlyEqual(merged.getGenotypes(), cfg.expected.getGenotypes());
}
public void writeBeagleOutput(
VariantContext preferredVC, VariantContext otherVC, boolean isValidationSite, double prior) {
GenomeLoc currentLoc =
VariantContextUtils.getLocation(getToolkit().getGenomeLocParser(), preferredVC);
StringBuffer beagleOut = new StringBuffer();
String marker = String.format("%s:%d ", currentLoc.getContig(), currentLoc.getStart());
beagleOut.append(marker);
if (markers != null)
markers.append(marker).append("\t").append(Integer.toString(markerCounter++)).append("\t");
for (Allele allele : preferredVC.getAlleles()) {
String bglPrintString;
if (allele.isNoCall() || allele.isNull()) bglPrintString = "-";
else bglPrintString = allele.getBaseString(); // get rid of * in case of reference allele
beagleOut.append(String.format("%s ", bglPrintString));
if (markers != null) markers.append(bglPrintString).append("\t");
}
if (markers != null) markers.append("\n");
GenotypesContext preferredGenotypes = preferredVC.getGenotypes();
GenotypesContext otherGenotypes = goodSite(otherVC) ? otherVC.getGenotypes() : null;
for (String sample : samples) {
boolean isMaleOnChrX = CHECK_IS_MALE_ON_CHR_X && getSample(sample).getGender() == Gender.MALE;
Genotype genotype;
boolean isValidation;
// use sample as key into genotypes structure
if (preferredGenotypes.containsSample(sample)) {
genotype = preferredGenotypes.get(sample);
isValidation = isValidationSite;
} else if (otherGenotypes != null && otherGenotypes.containsSample(sample)) {
genotype = otherGenotypes.get(sample);
isValidation = !isValidationSite;
} else {
// there is magically no genotype for this sample.
throw new StingException(
"Sample "
+ sample
+ " arose with no genotype in variant or validation VCF. This should never happen.");
}
/*
* Use likelihoods if: is validation, prior is negative; or: is not validation, has genotype key
*/
double[] log10Likelihoods = null;
if ((isValidation && prior < 0.0) || genotype.hasLikelihoods()) {
log10Likelihoods = genotype.getLikelihoods().getAsVector();
// see if we need to randomly mask out genotype in this position.
if (GenomeAnalysisEngine.getRandomGenerator().nextDouble() <= insertedNoCallRate) {
// we are masking out this genotype
log10Likelihoods =
isMaleOnChrX ? HAPLOID_FLAT_LOG10_LIKELIHOODS : DIPLOID_FLAT_LOG10_LIKELIHOODS;
}
if (isMaleOnChrX) {
log10Likelihoods[1] = -255; // todo -- warning this is dangerous for multi-allele case
}
}
/** otherwise, use the prior uniformly */
else if (!isValidation && genotype.isCalled() && !genotype.hasLikelihoods()) {
// hack to deal with input VCFs with no genotype likelihoods. Just assume the called
// genotype
// is confident. This is useful for Hapmap and 1KG release VCFs.
double AA = (1.0 - prior) / 2.0;
double AB = (1.0 - prior) / 2.0;
double BB = (1.0 - prior) / 2.0;
if (genotype.isHomRef()) {
AA = prior;
} else if (genotype.isHet()) {
AB = prior;
} else if (genotype.isHomVar()) {
BB = prior;
}
log10Likelihoods = MathUtils.toLog10(new double[] {AA, isMaleOnChrX ? 0.0 : AB, BB});
} else {
log10Likelihoods =
isMaleOnChrX ? HAPLOID_FLAT_LOG10_LIKELIHOODS : DIPLOID_FLAT_LOG10_LIKELIHOODS;
}
writeSampleLikelihoods(beagleOut, preferredVC, log10Likelihoods);
}
beagleWriter.println(beagleOut.toString());
}
/**
* Main entry function to calculate genotypes of a given VC with corresponding GL's
*
* @param tracker Tracker
* @param refContext Reference context
* @param rawContext Raw context
* @param stratifiedContexts Stratified alignment contexts
* @param vc Input VC
* @param model GL calculation model
* @param inheritAttributesFromInputVC Output VC will contain attributes inherited from input vc
* @return VC with assigned genotypes
*/
public VariantCallContext calculateGenotypes(
final RefMetaDataTracker tracker,
final ReferenceContext refContext,
final AlignmentContext rawContext,
Map<String, AlignmentContext> stratifiedContexts,
final VariantContext vc,
final GenotypeLikelihoodsCalculationModel.Model model,
final boolean inheritAttributesFromInputVC,
final Map<String, org.broadinstitute.sting.utils.genotyper.PerReadAlleleLikelihoodMap>
perReadAlleleLikelihoodMap) {
boolean limitedContext =
tracker == null || refContext == null || rawContext == null || stratifiedContexts == null;
// initialize the data for this thread if that hasn't been done yet
if (afcm.get() == null) {
afcm.set(AFCalcFactory.createAFCalc(UAC, N, logger));
}
// estimate our confidence in a reference call and return
if (vc.getNSamples() == 0) {
if (limitedContext) return null;
return (UAC.OutputMode != OUTPUT_MODE.EMIT_ALL_SITES
? estimateReferenceConfidence(vc, stratifiedContexts, getTheta(model), false, 1.0)
: generateEmptyContext(tracker, refContext, stratifiedContexts, rawContext));
}
AFCalcResult AFresult = afcm.get().getLog10PNonRef(vc, getAlleleFrequencyPriors(model));
// is the most likely frequency conformation AC=0 for all alternate alleles?
boolean bestGuessIsRef = true;
// determine which alternate alleles have AF>0
final List<Allele> myAlleles = new ArrayList<Allele>(vc.getAlleles().size());
final List<Integer> alleleCountsofMLE = new ArrayList<Integer>(vc.getAlleles().size());
myAlleles.add(vc.getReference());
for (int i = 0; i < AFresult.getAllelesUsedInGenotyping().size(); i++) {
final Allele alternateAllele = AFresult.getAllelesUsedInGenotyping().get(i);
if (alternateAllele.isReference()) continue;
// we are non-ref if the probability of being non-ref > the emit confidence.
// the emit confidence is phred-scaled, say 30 => 10^-3.
// the posterior AF > 0 is log10: -5 => 10^-5
// we are non-ref if 10^-5 < 10^-3 => -5 < -3
final boolean isNonRef =
AFresult.isPolymorphic(alternateAllele, UAC.STANDARD_CONFIDENCE_FOR_EMITTING / -10.0);
// if the most likely AC is not 0, then this is a good alternate allele to use
if (isNonRef) {
myAlleles.add(alternateAllele);
alleleCountsofMLE.add(AFresult.getAlleleCountAtMLE(alternateAllele));
bestGuessIsRef = false;
}
// if in GENOTYPE_GIVEN_ALLELES mode, we still want to allow the use of a poor allele
else if (UAC.GenotypingMode
== GenotypeLikelihoodsCalculationModel.GENOTYPING_MODE.GENOTYPE_GIVEN_ALLELES) {
myAlleles.add(alternateAllele);
alleleCountsofMLE.add(AFresult.getAlleleCountAtMLE(alternateAllele));
}
}
final double PoFGT0 = Math.pow(10, AFresult.getLog10PosteriorOfAFGT0());
// note the math.abs is necessary because -10 * 0.0 => -0.0 which isn't nice
final double phredScaledConfidence =
Math.abs(
!bestGuessIsRef
|| UAC.GenotypingMode
== GenotypeLikelihoodsCalculationModel.GENOTYPING_MODE
.GENOTYPE_GIVEN_ALLELES
? -10 * AFresult.getLog10PosteriorOfAFEq0()
: -10 * AFresult.getLog10PosteriorOfAFGT0());
// return a null call if we don't pass the confidence cutoff or the most likely allele frequency
// is zero
if (UAC.OutputMode != OUTPUT_MODE.EMIT_ALL_SITES
&& !passesEmitThreshold(phredScaledConfidence, bestGuessIsRef)) {
// technically, at this point our confidence in a reference call isn't accurately estimated
// because it didn't take into account samples with no data, so let's get a better estimate
return limitedContext
? null
: estimateReferenceConfidence(vc, stratifiedContexts, getTheta(model), true, PoFGT0);
}
// start constructing the resulting VC
final GenomeLoc loc = genomeLocParser.createGenomeLoc(vc);
final VariantContextBuilder builder =
new VariantContextBuilder(
"UG_call", loc.getContig(), loc.getStart(), loc.getStop(), myAlleles);
builder.log10PError(phredScaledConfidence / -10.0);
if (!passesCallThreshold(phredScaledConfidence)) builder.filters(filter);
// create the genotypes
final GenotypesContext genotypes = afcm.get().subsetAlleles(vc, myAlleles, true, ploidy);
builder.genotypes(genotypes);
// print out stats if we have a writer
if (verboseWriter != null && !limitedContext)
printVerboseData(refContext.getLocus().toString(), vc, PoFGT0, phredScaledConfidence, model);
// *** note that calculating strand bias involves overwriting data structures, so we do that
// last
final HashMap<String, Object> attributes = new HashMap<String, Object>();
// inherit attributed from input vc if requested
if (inheritAttributesFromInputVC) attributes.putAll(vc.getAttributes());
// if the site was downsampled, record that fact
if (!limitedContext && rawContext.hasPileupBeenDownsampled())
attributes.put(VCFConstants.DOWNSAMPLED_KEY, true);
if (UAC.ANNOTATE_NUMBER_OF_ALLELES_DISCOVERED)
attributes.put(NUMBER_OF_DISCOVERED_ALLELES_KEY, vc.getAlternateAlleles().size());
// add the MLE AC and AF annotations
if (alleleCountsofMLE.size() > 0) {
attributes.put(VCFConstants.MLE_ALLELE_COUNT_KEY, alleleCountsofMLE);
final int AN = builder.make().getCalledChrCount();
final ArrayList<Double> MLEfrequencies = new ArrayList<Double>(alleleCountsofMLE.size());
// the MLEAC is allowed to be larger than the AN (e.g. in the case of all PLs being 0, the GT
// is ./. but the exact model may arbitrarily choose an AC>1)
for (int AC : alleleCountsofMLE) MLEfrequencies.add(Math.min(1.0, (double) AC / (double) AN));
attributes.put(VCFConstants.MLE_ALLELE_FREQUENCY_KEY, MLEfrequencies);
}
if (UAC.COMPUTE_SLOD && !limitedContext && !bestGuessIsRef) {
// final boolean DEBUG_SLOD = false;
// the overall lod
// double overallLog10PofNull = AFresult.log10AlleleFrequencyPosteriors[0];
double overallLog10PofF = AFresult.getLog10LikelihoodOfAFGT0();
// if ( DEBUG_SLOD ) System.out.println("overallLog10PofF=" + overallLog10PofF);
List<Allele> allAllelesToUse = builder.make().getAlleles();
// the forward lod
VariantContext vcForward =
calculateLikelihoods(
tracker,
refContext,
stratifiedContexts,
AlignmentContextUtils.ReadOrientation.FORWARD,
allAllelesToUse,
false,
model,
perReadAlleleLikelihoodMap);
AFresult = afcm.get().getLog10PNonRef(vcForward, getAlleleFrequencyPriors(model));
// double[] normalizedLog10Posteriors =
// MathUtils.normalizeFromLog10(AFresult.log10AlleleFrequencyPosteriors, true);
double forwardLog10PofNull = AFresult.getLog10LikelihoodOfAFEq0();
double forwardLog10PofF = AFresult.getLog10LikelihoodOfAFGT0();
// if ( DEBUG_SLOD ) System.out.println("forwardLog10PofNull=" + forwardLog10PofNull + ",
// forwardLog10PofF=" + forwardLog10PofF);
// the reverse lod
VariantContext vcReverse =
calculateLikelihoods(
tracker,
refContext,
stratifiedContexts,
AlignmentContextUtils.ReadOrientation.REVERSE,
allAllelesToUse,
false,
model,
perReadAlleleLikelihoodMap);
AFresult = afcm.get().getLog10PNonRef(vcReverse, getAlleleFrequencyPriors(model));
// normalizedLog10Posteriors =
// MathUtils.normalizeFromLog10(AFresult.log10AlleleFrequencyPosteriors, true);
double reverseLog10PofNull = AFresult.getLog10LikelihoodOfAFEq0();
double reverseLog10PofF = AFresult.getLog10LikelihoodOfAFGT0();
// if ( DEBUG_SLOD ) System.out.println("reverseLog10PofNull=" + reverseLog10PofNull + ",
// reverseLog10PofF=" + reverseLog10PofF);
double forwardLod = forwardLog10PofF + reverseLog10PofNull - overallLog10PofF;
double reverseLod = reverseLog10PofF + forwardLog10PofNull - overallLog10PofF;
// if ( DEBUG_SLOD ) System.out.println("forward lod=" + forwardLod + ", reverse lod=" +
// reverseLod);
// strand score is max bias between forward and reverse strands
double strandScore = Math.max(forwardLod, reverseLod);
// rescale by a factor of 10
strandScore *= 10.0;
// logger.debug(String.format("SLOD=%f", strandScore));
if (!Double.isNaN(strandScore)) attributes.put("SB", strandScore);
}
// finish constructing the resulting VC
builder.attributes(attributes);
VariantContext vcCall = builder.make();
// if we are subsetting alleles (either because there were too many or because some were not
// polymorphic)
// then we may need to trim the alleles (because the original VariantContext may have had to pad
// at the end).
if (myAlleles.size() != vc.getAlleles().size()
&& !limitedContext) // limitedContext callers need to handle allele trimming on their own to
// keep their perReadAlleleLikelihoodMap alleles in sync
vcCall = VariantContextUtils.reverseTrimAlleles(vcCall);
if (annotationEngine != null
&& !limitedContext) { // limitedContext callers need to handle annotations on their own by
// calling their own annotationEngine
// Note: we want to use the *unfiltered* and *unBAQed* context for the annotations
final ReadBackedPileup pileup = rawContext.getBasePileup();
stratifiedContexts = AlignmentContextUtils.splitContextBySampleName(pileup);
vcCall =
annotationEngine.annotateContext(
tracker, refContext, stratifiedContexts, vcCall, perReadAlleleLikelihoodMap);
}
return new VariantCallContext(vcCall, confidentlyCalled(phredScaledConfidence, PoFGT0));
}
public Collection<Allele> values() {
return map != null ? map.values() : vc.getAlleles();
}
public static VariantContext createVariantContextWithTrimmedAlleles(VariantContext inputVC) {
// see if we need to trim common reference base from all alleles
boolean trimVC;
// We need to trim common reference base from all alleles in all genotypes if a ref base is
// common to all alleles
Allele refAllele = inputVC.getReference();
if (!inputVC.isVariant()) trimVC = false;
else if (refAllele.isNull()) trimVC = false;
else {
trimVC =
(AbstractVCFCodec.computeForwardClipping(
new ArrayList<Allele>(inputVC.getAlternateAlleles()),
inputVC.getReference().getDisplayString())
> 0);
}
// nothing to do if we don't need to trim bases
if (trimVC) {
List<Allele> alleles = new ArrayList<Allele>();
GenotypesContext genotypes = GenotypesContext.create();
// set the reference base for indels in the attributes
Map<String, Object> attributes = new TreeMap<String, Object>(inputVC.getAttributes());
Map<Allele, Allele> originalToTrimmedAlleleMap = new HashMap<Allele, Allele>();
for (Allele a : inputVC.getAlleles()) {
if (a.isSymbolic()) {
alleles.add(a);
originalToTrimmedAlleleMap.put(a, a);
} else {
// get bases for current allele and create a new one with trimmed bases
byte[] newBases = Arrays.copyOfRange(a.getBases(), 1, a.length());
Allele trimmedAllele = Allele.create(newBases, a.isReference());
alleles.add(trimmedAllele);
originalToTrimmedAlleleMap.put(a, trimmedAllele);
}
}
// detect case where we're trimming bases but resulting vc doesn't have any null allele. In
// that case, we keep original representation
// example: mixed records such as {TA*,TGA,TG}
boolean hasNullAlleles = false;
for (Allele a : originalToTrimmedAlleleMap.values()) {
if (a.isNull()) hasNullAlleles = true;
if (a.isReference()) refAllele = a;
}
if (!hasNullAlleles) return inputVC;
// now we can recreate new genotypes with trimmed alleles
for (final Genotype genotype : inputVC.getGenotypes()) {
List<Allele> originalAlleles = genotype.getAlleles();
List<Allele> trimmedAlleles = new ArrayList<Allele>();
for (Allele a : originalAlleles) {
if (a.isCalled()) trimmedAlleles.add(originalToTrimmedAlleleMap.get(a));
else trimmedAlleles.add(Allele.NO_CALL);
}
genotypes.add(Genotype.modifyAlleles(genotype, trimmedAlleles));
}
final VariantContextBuilder builder = new VariantContextBuilder(inputVC);
return builder
.alleles(alleles)
.genotypes(genotypes)
.attributes(attributes)
.referenceBaseForIndel(new Byte(inputVC.getReference().getBases()[0]))
.make();
}
return inputVC;
}
public static VariantContext createVariantContextWithPaddedAlleles(
VariantContext inputVC, boolean refBaseShouldBeAppliedToEndOfAlleles) {
// see if we need to pad common reference base from all alleles
boolean padVC;
// We need to pad a VC with a common base if the length of the reference allele is less than the
// length of the VariantContext.
// This happens because the position of e.g. an indel is always one before the actual event (as
// per VCF convention).
long locLength = (inputVC.getEnd() - inputVC.getStart()) + 1;
if (inputVC.hasSymbolicAlleles()) padVC = true;
else if (inputVC.getReference().length() == locLength) padVC = false;
else if (inputVC.getReference().length() == locLength - 1) padVC = true;
else
throw new IllegalArgumentException(
"Badly formed variant context at location "
+ String.valueOf(inputVC.getStart())
+ " in contig "
+ inputVC.getChr()
+ ". Reference length must be at most one base shorter than location size");
// nothing to do if we don't need to pad bases
if (padVC) {
if (!inputVC.hasReferenceBaseForIndel())
throw new ReviewedStingException(
"Badly formed variant context at location "
+ inputVC.getChr()
+ ":"
+ inputVC.getStart()
+ "; no padded reference base is available.");
Byte refByte = inputVC.getReferenceBaseForIndel();
List<Allele> alleles = new ArrayList<Allele>();
for (Allele a : inputVC.getAlleles()) {
// get bases for current allele and create a new one with trimmed bases
if (a.isSymbolic()) {
alleles.add(a);
} else {
String newBases;
if (refBaseShouldBeAppliedToEndOfAlleles)
newBases = a.getBaseString() + new String(new byte[] {refByte});
else newBases = new String(new byte[] {refByte}) + a.getBaseString();
alleles.add(Allele.create(newBases, a.isReference()));
}
}
// now we can recreate new genotypes with trimmed alleles
GenotypesContext genotypes = GenotypesContext.create(inputVC.getNSamples());
for (final Genotype g : inputVC.getGenotypes()) {
List<Allele> inAlleles = g.getAlleles();
List<Allele> newGenotypeAlleles = new ArrayList<Allele>(g.getAlleles().size());
for (Allele a : inAlleles) {
if (a.isCalled()) {
if (a.isSymbolic()) {
newGenotypeAlleles.add(a);
} else {
String newBases;
if (refBaseShouldBeAppliedToEndOfAlleles)
newBases = a.getBaseString() + new String(new byte[] {refByte});
else newBases = new String(new byte[] {refByte}) + a.getBaseString();
newGenotypeAlleles.add(Allele.create(newBases, a.isReference()));
}
} else {
// add no-call allele
newGenotypeAlleles.add(Allele.NO_CALL);
}
}
genotypes.add(
new Genotype(
g.getSampleName(),
newGenotypeAlleles,
g.getLog10PError(),
g.getFilters(),
g.getAttributes(),
g.isPhased()));
}
return new VariantContextBuilder(inputVC).alleles(alleles).genotypes(genotypes).make();
} else return inputVC;
}
public Allele getLikelihoods(
RefMetaDataTracker tracker,
ReferenceContext ref,
Map<String, AlignmentContext> contexts,
AlignmentContextUtils.ReadOrientation contextType,
GenotypePriors priors,
Map<String, MultiallelicGenotypeLikelihoods> GLs,
Allele alternateAlleleToUse,
boolean useBAQedPileup) {
if (tracker == null) return null;
GenomeLoc loc = ref.getLocus();
Allele refAllele, altAllele;
VariantContext vc = null;
if (!ref.getLocus().equals(lastSiteVisited)) {
// starting a new site: clear allele list
alleleList.clear();
lastSiteVisited = ref.getLocus();
indelLikelihoodMap.set(new HashMap<PileupElement, LinkedHashMap<Allele, Double>>());
haplotypeMap.clear();
if (getAlleleListFromVCF) {
for (final VariantContext vc_input : tracker.getValues(UAC.alleles, loc)) {
if (vc_input != null
&& allowableTypes.contains(vc_input.getType())
&& ref.getLocus().getStart() == vc_input.getStart()) {
vc = vc_input;
break;
}
}
// ignore places where we don't have a variant
if (vc == null) return null;
alleleList.clear();
if (ignoreSNPAllelesWhenGenotypingIndels) {
// if there's an allele that has same length as the reference (i.e. a SNP or MNP), ignore
// it and don't genotype it
for (Allele a : vc.getAlleles())
if (a.isNonReference() && a.getBases().length == vc.getReference().getBases().length)
continue;
else alleleList.add(a);
} else {
for (Allele a : vc.getAlleles()) alleleList.add(a);
}
} else {
alleleList = computeConsensusAlleles(ref, contexts, contextType);
if (alleleList.isEmpty()) return null;
}
}
// protect against having an indel too close to the edge of a contig
if (loc.getStart() <= HAPLOTYPE_SIZE) return null;
// check if there is enough reference window to create haplotypes (can be an issue at end of
// contigs)
if (ref.getWindow().getStop() < loc.getStop() + HAPLOTYPE_SIZE) return null;
if (!(priors instanceof DiploidIndelGenotypePriors))
throw new StingException(
"Only diploid-based Indel priors are supported in the DINDEL GL model");
if (alleleList.isEmpty()) return null;
refAllele = alleleList.get(0);
altAllele = alleleList.get(1);
// look for alt allele that has biggest length distance to ref allele
int maxLenDiff = 0;
for (Allele a : alleleList) {
if (a.isNonReference()) {
int lenDiff = Math.abs(a.getBaseString().length() - refAllele.getBaseString().length());
if (lenDiff > maxLenDiff) {
maxLenDiff = lenDiff;
altAllele = a;
}
}
}
final int eventLength = altAllele.getBaseString().length() - refAllele.getBaseString().length();
final int hsize = (int) ref.getWindow().size() - Math.abs(eventLength) - 1;
final int numPrefBases = ref.getLocus().getStart() - ref.getWindow().getStart() + 1;
haplotypeMap =
Haplotype.makeHaplotypeListFromAlleles(
alleleList, loc.getStart(), ref, hsize, numPrefBases);
// For each sample, get genotype likelihoods based on pileup
// compute prior likelihoods on haplotypes, and initialize haplotype likelihood matrix with
// them.
// initialize the GenotypeLikelihoods
GLs.clear();
for (Map.Entry<String, AlignmentContext> sample : contexts.entrySet()) {
AlignmentContext context = AlignmentContextUtils.stratify(sample.getValue(), contextType);
ReadBackedPileup pileup = null;
if (context.hasExtendedEventPileup()) pileup = context.getExtendedEventPileup();
else if (context.hasBasePileup()) pileup = context.getBasePileup();
if (pileup != null) {
final double[] genotypeLikelihoods =
pairModel.computeReadHaplotypeLikelihoods(
pileup, haplotypeMap, ref, eventLength, getIndelLikelihoodMap());
GLs.put(
sample.getKey(),
new MultiallelicGenotypeLikelihoods(
sample.getKey(), alleleList, genotypeLikelihoods, getFilteredDepth(pileup)));
if (DEBUG) {
System.out.format("Sample:%s Alleles:%s GL:", sample.getKey(), alleleList.toString());
for (int k = 0; k < genotypeLikelihoods.length; k++)
System.out.format("%1.4f ", genotypeLikelihoods[k]);
System.out.println();
}
}
}
return refAllele;
}
/**
* add a record to the file
*
* @param vc the Variant Context object
* @param refBase the ref base used for indels
* @param refBaseShouldBeAppliedToEndOfAlleles *** THIS SHOULD BE FALSE EXCEPT FOR AN INDEL AT THE
* EXTREME BEGINNING OF A CONTIG (WHERE THERE IS NO PREVIOUS BASE, SO WE USE THE BASE AFTER
* THE EVENT INSTEAD)
*/
public void add(VariantContext vc, byte refBase, boolean refBaseShouldBeAppliedToEndOfAlleles) {
if (mHeader == null)
throw new IllegalStateException(
"The VCF Header must be written before records can be added: " + locationString());
if (doNotWriteGenotypes) vc = VariantContext.modifyGenotypes(vc, null);
try {
vc =
VariantContext.createVariantContextWithPaddedAlleles(
vc, refBase, refBaseShouldBeAppliedToEndOfAlleles);
// if we are doing on the fly indexing, add the record ***before*** we write any bytes
if (indexer != null) indexer.addFeature(vc, positionalStream.getPosition());
Map<Allele, String> alleleMap = new HashMap<Allele, String>(vc.getAlleles().size());
alleleMap.put(Allele.NO_CALL, VCFConstants.EMPTY_ALLELE); // convenience for lookup
// CHROM
mWriter.write(vc.getChr());
mWriter.write(VCFConstants.FIELD_SEPARATOR);
// POS
mWriter.write(String.valueOf(vc.getStart()));
mWriter.write(VCFConstants.FIELD_SEPARATOR);
// ID
String ID = vc.hasID() ? vc.getID() : VCFConstants.EMPTY_ID_FIELD;
mWriter.write(ID);
mWriter.write(VCFConstants.FIELD_SEPARATOR);
// REF
alleleMap.put(vc.getReference(), "0");
String refString = vc.getReference().getDisplayString();
mWriter.write(refString);
mWriter.write(VCFConstants.FIELD_SEPARATOR);
// ALT
if (vc.isVariant()) {
Allele altAllele = vc.getAlternateAllele(0);
alleleMap.put(altAllele, "1");
String alt = altAllele.getDisplayString();
mWriter.write(alt);
for (int i = 1; i < vc.getAlternateAlleles().size(); i++) {
altAllele = vc.getAlternateAllele(i);
alleleMap.put(altAllele, String.valueOf(i + 1));
alt = altAllele.getDisplayString();
mWriter.write(",");
mWriter.write(alt);
}
} else {
mWriter.write(VCFConstants.EMPTY_ALTERNATE_ALLELE_FIELD);
}
mWriter.write(VCFConstants.FIELD_SEPARATOR);
// QUAL
if (!vc.hasNegLog10PError()) mWriter.write(VCFConstants.MISSING_VALUE_v4);
else mWriter.write(getQualValue(vc.getPhredScaledQual()));
mWriter.write(VCFConstants.FIELD_SEPARATOR);
// FILTER
String filters =
vc.isFiltered()
? ParsingUtils.join(";", ParsingUtils.sortList(vc.getFilters()))
: (filtersWereAppliedToContext || vc.filtersWereApplied()
? VCFConstants.PASSES_FILTERS_v4
: VCFConstants.UNFILTERED);
mWriter.write(filters);
mWriter.write(VCFConstants.FIELD_SEPARATOR);
// INFO
Map<String, String> infoFields = new TreeMap<String, String>();
for (Map.Entry<String, Object> field : vc.getAttributes().entrySet()) {
String key = field.getKey();
if (key.equals(VariantContext.ID_KEY)
|| key.equals(VariantContext.REFERENCE_BASE_FOR_INDEL_KEY)
|| key.equals(VariantContext.UNPARSED_GENOTYPE_MAP_KEY)
|| key.equals(VariantContext.UNPARSED_GENOTYPE_PARSER_KEY)) continue;
String outputValue = formatVCFField(field.getValue());
if (outputValue != null) infoFields.put(key, outputValue);
}
writeInfoString(infoFields);
// FORMAT
if (vc.hasAttribute(VariantContext.UNPARSED_GENOTYPE_MAP_KEY)) {
mWriter.write(VCFConstants.FIELD_SEPARATOR);
mWriter.write(vc.getAttributeAsString(VariantContext.UNPARSED_GENOTYPE_MAP_KEY, ""));
} else {
List<String> genotypeAttributeKeys = new ArrayList<String>();
if (vc.hasGenotypes()) {
genotypeAttributeKeys.addAll(calcVCFGenotypeKeys(vc));
} else if (mHeader.hasGenotypingData()) {
// this needs to be done in case all samples are no-calls
genotypeAttributeKeys.add(VCFConstants.GENOTYPE_KEY);
}
if (genotypeAttributeKeys.size() > 0) {
String genotypeFormatString =
ParsingUtils.join(VCFConstants.GENOTYPE_FIELD_SEPARATOR, genotypeAttributeKeys);
mWriter.write(VCFConstants.FIELD_SEPARATOR);
mWriter.write(genotypeFormatString);
addGenotypeData(vc, alleleMap, genotypeAttributeKeys);
}
}
mWriter.write("\n");
mWriter.flush(); // necessary so that writing to an output stream will work
} catch (IOException e) {
throw new RuntimeException("Unable to write the VCF object to " + locationString());
}
}