public Map<String, Object> annotate(
final RefMetaDataTracker tracker,
final AnnotatorCompatible walker,
final ReferenceContext ref,
final Map<String, AlignmentContext> stratifiedContexts,
final VariantContext vc,
final Map<String, PerReadAlleleLikelihoodMap> stratifiedPerReadAlleleLikelihoodMap) {
int run;
if (vc.isMixed()) {
Map<String, Object> map = new HashMap<String, Object>();
map.put(getKeyNames().get(0), String.format("%s", "MIXED"));
return map;
} else if (vc.isIndel()) {
String type = "";
if (!vc.isBiallelic()) type = "MULTIALLELIC_INDEL";
else {
if (vc.isSimpleInsertion()) type = "INS.";
else if (vc.isSimpleDeletion()) type = "DEL.";
else type = "OTHER.";
ArrayList<Integer> inds = IndelUtils.findEventClassificationIndex(vc, ref);
for (int k : inds) {
type = type + IndelUtils.getIndelClassificationName(k) + ".";
}
}
Map<String, Object> map = new HashMap<String, Object>();
map.put(getKeyNames().get(0), String.format("%s", type));
return map;
} else {
return null;
}
}
public static BaseUtils.BaseSubstitutionType getSNPSubstitutionType(VariantContext context) {
if (!context.isSNP() || !context.isBiallelic())
throw new IllegalStateException(
"Requested SNP substitution type for bialleic non-SNP " + context);
return BaseUtils.SNPSubstitutionType(
context.getReference().getBases()[0], context.getAlternateAllele(0).getBases()[0]);
}
private Map<String, Object> calculateIC(final VariantContext vc) {
final GenotypesContext genotypes =
(founderIds == null || founderIds.isEmpty())
? vc.getGenotypes()
: vc.getGenotypes(founderIds);
if (genotypes == null || genotypes.size() < MIN_SAMPLES) return null;
int idxAA = 0, idxAB = 1, idxBB = 2;
if (!vc.isBiallelic()) {
// for non-bliallelic case, do test with most common alt allele.
// Get then corresponding indeces in GL vectors to retrieve GL of AA,AB and BB.
int[] idxVector = vc.getGLIndecesOfAlternateAllele(vc.getAltAlleleWithHighestAlleleCount());
idxAA = idxVector[0];
idxAB = idxVector[1];
idxBB = idxVector[2];
}
double refCount = 0.0;
double hetCount = 0.0;
double homCount = 0.0;
int N = 0; // number of samples that have likelihoods
for (final Genotype g : genotypes) {
if (g.isNoCall() || !g.hasLikelihoods()) continue;
if (g.getPloidy() != 2) // only work for diploid samples
continue;
N++;
final double[] normalizedLikelihoods =
MathUtils.normalizeFromLog10(g.getLikelihoods().getAsVector());
refCount += normalizedLikelihoods[idxAA];
hetCount += normalizedLikelihoods[idxAB];
homCount += normalizedLikelihoods[idxBB];
}
if (N < MIN_SAMPLES) {
return null;
}
final double p =
(2.0 * refCount + hetCount)
/ (2.0 * (refCount + hetCount + homCount)); // expected reference allele frequency
final double q = 1.0 - p; // expected alternative allele frequency
final double F = 1.0 - (hetCount / (2.0 * p * q * (double) N)); // inbreeding coefficient
Map<String, Object> map = new HashMap<String, Object>();
map.put(getKeyNames().get(0), String.format("%.4f", F));
return map;
}
public void update2(
VariantContext eval,
VariantContext comp,
RefMetaDataTracker tracker,
ReferenceContext ref,
AlignmentContext context) {
if (eval == null || (getWalker().ignoreAC0Sites() && eval.isMonomorphicInSamples())) return;
final Type type = getType(eval);
if (type == null) return;
TypeSampleMap titvTable = null;
// update DP, if possible
if (eval.hasAttribute(VCFConstants.DEPTH_KEY)) depthPerSample.inc(type, ALL);
// update counts
allVariantCounts.inc(type, ALL);
// type specific calculations
if (type == Type.SNP && eval.isBiallelic()) {
titvTable =
VariantContextUtils.isTransition(eval) ? transitionsPerSample : transversionsPerSample;
titvTable.inc(type, ALL);
}
// novelty calculation
if (comp != null || (type == Type.CNV && overlapsKnownCNV(eval)))
knownVariantCounts.inc(type, ALL);
// per sample metrics
for (final Genotype g : eval.getGenotypes()) {
if (!g.isNoCall() && !g.isHomRef()) {
countsPerSample.inc(type, g.getSampleName());
// update transition / transversion ratio
if (titvTable != null) titvTable.inc(type, g.getSampleName());
if (g.hasDP()) depthPerSample.inc(type, g.getSampleName());
}
}
}
public static boolean canBeOutputToBeagle(VariantContext v) {
return v != null && !v.isFiltered() && v.isBiallelic() && v.hasGenotypes();
}
/**
* Subset VC record if necessary and emit the modified record (provided it satisfies criteria for
* printing)
*
* @param tracker the ROD tracker
* @param ref reference information
* @param context alignment info
* @return 1 if the record was printed to the output file, 0 if otherwise
*/
@Override
public Integer map(RefMetaDataTracker tracker, ReferenceContext ref, AlignmentContext context) {
if (tracker == null) return 0;
Collection<VariantContext> vcs =
tracker.getValues(variantCollection.variants, context.getLocation());
if (vcs == null || vcs.size() == 0) {
return 0;
}
for (VariantContext vc : vcs) {
if (MENDELIAN_VIOLATIONS) {
boolean foundMV = false;
for (MendelianViolation mv : mvSet) {
if (mv.isViolation(vc)) {
foundMV = true;
// System.out.println(vc.toString());
if (outMVFile != null)
outMVFileStream.format(
"MV@%s:%d. REF=%s, ALT=%s, AC=%d, momID=%s, dadID=%s, childID=%s, momG=%s, momGL=%s, dadG=%s, dadGL=%s, "
+ "childG=%s childGL=%s\n",
vc.getChr(),
vc.getStart(),
vc.getReference().getDisplayString(),
vc.getAlternateAllele(0).getDisplayString(),
vc.getChromosomeCount(vc.getAlternateAllele(0)),
mv.getSampleMom(),
mv.getSampleDad(),
mv.getSampleChild(),
vc.getGenotype(mv.getSampleMom()).toBriefString(),
vc.getGenotype(mv.getSampleMom()).getLikelihoods().getAsString(),
vc.getGenotype(mv.getSampleDad()).toBriefString(),
vc.getGenotype(mv.getSampleMom()).getLikelihoods().getAsString(),
vc.getGenotype(mv.getSampleChild()).toBriefString(),
vc.getGenotype(mv.getSampleChild()).getLikelihoods().getAsString());
}
}
if (!foundMV) break;
}
if (DISCORDANCE_ONLY) {
Collection<VariantContext> compVCs =
tracker.getValues(discordanceTrack, context.getLocation());
if (!isDiscordant(vc, compVCs)) return 0;
}
if (CONCORDANCE_ONLY) {
Collection<VariantContext> compVCs =
tracker.getValues(concordanceTrack, context.getLocation());
if (!isConcordant(vc, compVCs)) return 0;
}
if (alleleRestriction.equals(NumberAlleleRestriction.BIALLELIC) && !vc.isBiallelic())
continue;
if (alleleRestriction.equals(NumberAlleleRestriction.MULTIALLELIC) && vc.isBiallelic())
continue;
if (!selectedTypes.contains(vc.getType())) continue;
VariantContext sub = subsetRecord(vc, samples);
if ((sub.isPolymorphic() || !EXCLUDE_NON_VARIANTS)
&& (!sub.isFiltered() || !EXCLUDE_FILTERED)) {
for (VariantContextUtils.JexlVCMatchExp jexl : jexls) {
if (!VariantContextUtils.match(sub, jexl)) {
return 0;
}
}
if (SELECT_RANDOM_NUMBER) {
randomlyAddVariant(++variantNumber, sub, ref.getBase());
} else if (!SELECT_RANDOM_FRACTION
|| (GenomeAnalysisEngine.getRandomGenerator().nextDouble() < fractionRandom)) {
vcfWriter.add(sub);
}
}
}
return 1;
}