/**
* 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();
}
private static void mergeGenotypes(
GenotypesContext mergedGenotypes,
VariantContext oneVC,
AlleleMapper alleleMapping,
boolean uniqifySamples) {
for (Genotype g : oneVC.getGenotypes()) {
String name = mergedSampleName(oneVC.getSource(), g.getSampleName(), uniqifySamples);
if (!mergedGenotypes.containsSample(name)) {
// only add if the name is new
Genotype newG = g;
if (uniqifySamples || alleleMapping.needsRemapping()) {
final List<Allele> alleles =
alleleMapping.needsRemapping() ? alleleMapping.remap(g.getAlleles()) : g.getAlleles();
newG =
new Genotype(
name,
alleles,
g.getLog10PError(),
g.getFilters(),
g.getAttributes(),
g.isPhased());
}
mergedGenotypes.add(newG);
}
}
}
public static GenotypesContext stripPLs(GenotypesContext genotypes) {
GenotypesContext newGs = GenotypesContext.create(genotypes.size());
for (final Genotype g : genotypes) {
newGs.add(g.hasLikelihoods() ? removePLs(g) : g);
}
return newGs;
}
@Test
public void testFixReverseComplementedGenotypes() {
final Allele refA = Allele.create("A", true);
final Allele altC = Allele.create("C", false);
final GenotypesContext originalGenotypes = GenotypesContext.create(3);
originalGenotypes.add(new GenotypeBuilder("homref").alleles(Arrays.asList(refA, refA)).make());
originalGenotypes.add(new GenotypeBuilder("het").alleles(Arrays.asList(refA, altC)).make());
originalGenotypes.add(new GenotypeBuilder("homvar").alleles(Arrays.asList(altC, altC)).make());
final Allele refT = Allele.create("T", true);
final Allele altG = Allele.create("G", false);
final GenotypesContext expectedGenotypes = GenotypesContext.create(3);
expectedGenotypes.add(new GenotypeBuilder("homref").alleles(Arrays.asList(refT, refT)).make());
expectedGenotypes.add(new GenotypeBuilder("het").alleles(Arrays.asList(refT, altG)).make());
expectedGenotypes.add(new GenotypeBuilder("homvar").alleles(Arrays.asList(altG, altG)).make());
final Map<Allele, Allele> reverseComplementAlleleMap = new HashMap<Allele, Allele>(2);
reverseComplementAlleleMap.put(refA, refT);
reverseComplementAlleleMap.put(altC, altG);
final GenotypesContext actualGenotypes =
LiftoverVcf.fixGenotypes(originalGenotypes, reverseComplementAlleleMap);
for (final String sample : Arrays.asList("homref", "het", "homvar")) {
final List<Allele> expected = expectedGenotypes.get(sample).getAlleles();
final List<Allele> actual = actualGenotypes.get(sample).getAlleles();
Assert.assertEquals(expected.get(0), actual.get(0));
Assert.assertEquals(expected.get(1), actual.get(1));
}
}
public static VariantContext purgeUnallowedGenotypeAttributes(
VariantContext vc, Set<String> allowedAttributes) {
if (allowedAttributes == null) return vc;
GenotypesContext newGenotypes = GenotypesContext.create(vc.getNSamples());
for (final Genotype genotype : vc.getGenotypes()) {
Map<String, Object> attrs = new HashMap<String, Object>();
for (Map.Entry<String, Object> attr : genotype.getAttributes().entrySet()) {
if (allowedAttributes.contains(attr.getKey())) attrs.put(attr.getKey(), attr.getValue());
}
newGenotypes.add(Genotype.modifyAttributes(genotype, attrs));
}
return new VariantContextBuilder(vc).genotypes(newGenotypes).make();
}
public static VariantContextBuilder pruneVariantContext(
final VariantContextBuilder builder, Collection<String> keysToPreserve) {
final VariantContext vc = builder.make();
if (keysToPreserve == null) keysToPreserve = Collections.emptyList();
// VC info
final Map<String, Object> attributes = subsetAttributes(vc.commonInfo, keysToPreserve);
// Genotypes
final GenotypesContext genotypes = GenotypesContext.create(vc.getNSamples());
for (final Genotype g : vc.getGenotypes()) {
Map<String, Object> genotypeAttributes = subsetAttributes(g.commonInfo, keysToPreserve);
genotypes.add(
new Genotype(
g.getSampleName(),
g.getAlleles(),
g.getLog10PError(),
g.getFilters(),
genotypeAttributes,
g.isPhased()));
}
return builder.genotypes(genotypes).attributes(attributes);
}
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;
}
static VariantContext reallyMergeIntoMNP(
VariantContext vc1, VariantContext vc2, ReferenceSequenceFile referenceFile) {
int startInter = vc1.getEnd() + 1;
int endInter = vc2.getStart() - 1;
byte[] intermediateBases = null;
if (startInter <= endInter) {
intermediateBases =
referenceFile.getSubsequenceAt(vc1.getChr(), startInter, endInter).getBases();
StringUtil.toUpperCase(intermediateBases);
}
MergedAllelesData mergeData =
new MergedAllelesData(
intermediateBases, vc1, vc2); // ensures that the reference allele is added
GenotypesContext mergedGenotypes = GenotypesContext.create();
for (final Genotype gt1 : vc1.getGenotypes()) {
Genotype gt2 = vc2.getGenotype(gt1.getSampleName());
List<Allele> site1Alleles = gt1.getAlleles();
List<Allele> site2Alleles = gt2.getAlleles();
List<Allele> mergedAllelesForSample = new LinkedList<Allele>();
/* NOTE: Since merged alleles are added to mergedAllelesForSample in the SAME order as in the input VC records,
we preserve phase information (if any) relative to whatever precedes vc1:
*/
Iterator<Allele> all2It = site2Alleles.iterator();
for (Allele all1 : site1Alleles) {
Allele all2 = all2It.next(); // this is OK, since allSamplesAreMergeable()
Allele mergedAllele = mergeData.ensureMergedAllele(all1, all2);
mergedAllelesForSample.add(mergedAllele);
}
double mergedGQ = Math.max(gt1.getLog10PError(), gt2.getLog10PError());
Set<String> mergedGtFilters =
new HashSet<
String>(); // Since gt1 and gt2 were unfiltered, the Genotype remains unfiltered
Map<String, Object> mergedGtAttribs = new HashMap<String, Object>();
PhaseAndQuality phaseQual = calcPhaseForMergedGenotypes(gt1, gt2);
if (phaseQual.PQ != null) mergedGtAttribs.put(ReadBackedPhasingWalker.PQ_KEY, phaseQual.PQ);
Genotype mergedGt =
new Genotype(
gt1.getSampleName(),
mergedAllelesForSample,
mergedGQ,
mergedGtFilters,
mergedGtAttribs,
phaseQual.isPhased);
mergedGenotypes.add(mergedGt);
}
String mergedName = mergeVariantContextNames(vc1.getSource(), vc2.getSource());
double mergedLog10PError = Math.min(vc1.getLog10PError(), vc2.getLog10PError());
Set<String> mergedFilters =
new HashSet<
String>(); // Since vc1 and vc2 were unfiltered, the merged record remains unfiltered
Map<String, Object> mergedAttribs = mergeVariantContextAttributes(vc1, vc2);
// ids
List<String> mergedIDs = new ArrayList<String>();
if (vc1.hasID()) mergedIDs.add(vc1.getID());
if (vc2.hasID()) mergedIDs.add(vc2.getID());
String mergedID =
mergedIDs.isEmpty()
? VCFConstants.EMPTY_ID_FIELD
: Utils.join(VCFConstants.ID_FIELD_SEPARATOR, mergedIDs);
VariantContextBuilder mergedBuilder =
new VariantContextBuilder(
mergedName,
vc1.getChr(),
vc1.getStart(),
vc2.getEnd(),
mergeData.getAllMergedAlleles())
.id(mergedID)
.genotypes(mergedGenotypes)
.log10PError(mergedLog10PError)
.filters(mergedFilters)
.attributes(mergedAttribs);
VariantContextUtils.calculateChromosomeCounts(mergedBuilder, true);
return mergedBuilder.make();
}