public static int getMeanRepresentativeReadCount(GATKSAMRecord read) {
if (!read.isReducedRead()) return 1;
// compute mean representative read counts
final byte[] counts = read.getReducedReadCounts();
return (int) Math.round((double) MathUtils.sum(counts) / counts.length);
}
private Double scoreIndelsAgainstHaplotypes(final ReadBackedPileup pileup) {
final ArrayList<double[]> haplotypeScores = new ArrayList<double[]>();
final HashMap<PileupElement, LinkedHashMap<Allele, Double>> indelLikelihoodMap =
IndelGenotypeLikelihoodsCalculationModel.getIndelLikelihoodMap();
if (indelLikelihoodMap == null) return null;
for (final PileupElement p : pileup) {
if (indelLikelihoodMap.containsKey(p)) {
// retrieve likelihood information corresponding to this read
LinkedHashMap<Allele, Double> el = indelLikelihoodMap.get(p);
// Score all the reads in the pileup, even the filtered ones
final double[] scores = new double[el.size()];
int i = 0;
for (Allele a : el.keySet()) {
scores[i++] = -el.get(a);
if (DEBUG) {
System.out.printf(" vs. haplotype %d = %f%n", i - 1, scores[i - 1]);
}
}
haplotypeScores.add(scores);
}
}
// indel likelihoods are strict log-probs, not phred scored
double overallScore = 0.0;
for (final double[] readHaplotypeScores : haplotypeScores) {
overallScore += MathUtils.arrayMin(readHaplotypeScores);
}
return overallScore;
}
// calculate the haplotype scores by walking over all reads and comparing them to the haplotypes
private double scoreReadsAgainstHaplotypes(
final List<Haplotype> haplotypes,
final ReadBackedPileup pileup,
final int contextSize,
final int locus) {
if (DEBUG) System.out.printf("HAP1: %s%n", haplotypes.get(0));
if (DEBUG) System.out.printf("HAP2: %s%n", haplotypes.get(1));
final ArrayList<double[]> haplotypeScores = new ArrayList<double[]>();
for (final PileupElement p : pileup) {
// Score all the reads in the pileup, even the filtered ones
final double[] scores = new double[haplotypes.size()];
for (int i = 0; i < haplotypes.size(); i++) {
final Haplotype haplotype = haplotypes.get(i);
final double score = scoreReadAgainstHaplotype(p, contextSize, haplotype, locus);
scores[i] = score;
if (DEBUG) {
System.out.printf(" vs. haplotype %d = %f%n", i, score);
}
}
haplotypeScores.add(scores);
}
double overallScore = 0.0;
for (final double[] readHaplotypeScores : haplotypeScores) {
overallScore += MathUtils.arrayMin(readHaplotypeScores);
}
return overallScore;
}
/**
* Returns the coverage distribution of a list of reads within the desired region.
*
* <p>See getCoverageDistributionOfRead for information on how the coverage is calculated.
*
* @param list the list of reads covering the region
* @param startLocation the first reference coordinate of the region (inclusive)
* @param stopLocation the last reference coordinate of the region (inclusive)
* @return an array with the coverage of each position from startLocation to stopLocation
*/
public static int[] getCoverageDistributionOfReads(
List<GATKSAMRecord> list, int startLocation, int stopLocation) {
int[] totalCoverage = new int[stopLocation - startLocation + 1];
for (GATKSAMRecord read : list) {
int[] readCoverage = getCoverageDistributionOfRead(read, startLocation, stopLocation);
totalCoverage = MathUtils.addArrays(totalCoverage, readCoverage);
}
return totalCoverage;
}
private void writeSampleLikelihoods(
StringBuffer out, VariantContext vc, double[] log10Likelihoods) {
if (VQSRCalibrator != null) {
log10Likelihoods =
VQSRCalibrator.includeErrorRateInLikelihoods(VQSLOD_KEY, vc, log10Likelihoods);
}
double[] normalizedLikelihoods = MathUtils.normalizeFromLog10(log10Likelihoods);
// see if we need to randomly mask out genotype in this position.
for (double likeVal : normalizedLikelihoods) {
out.append(formatter.format(likeVal));
// out.append(String.format("%5.4f ",likeVal));
}
}
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;
}
// Private function to compare 2d arrays
private boolean compareDoubleArrays(double[][] b1, double[][] b2) {
if (b1.length != b2.length) {
return false; // sanity check
}
for (int i = 0; i < b1.length; i++) {
if (b1[i].length != b2[i].length) {
return false; // sanity check
}
for (int j = 0; j < b1.length; j++) {
if (MathUtils.compareDoubles(b1[i][j], b2[i][j]) != 0
&& !Double.isInfinite(b1[i][j])
&& !Double.isInfinite(b2[i][j])) return false;
}
}
return true;
}
/**
* Read in a list of ExactCall objects from reader, keeping only those with starts in startsToKeep
* or all sites (if this is empty)
*
* @param reader a just-opened reader sitting at the start of the file
* @param startsToKeep a list of start position of the calls to keep, or empty if all calls should
* be kept
* @param parser a genome loc parser to create genome locs
* @return a list of ExactCall objects in reader
* @throws IOException
*/
public static List<ExactCall> readExactLog(
final BufferedReader reader, final List<Integer> startsToKeep, GenomeLocParser parser)
throws IOException {
if (reader == null) throw new IllegalArgumentException("reader cannot be null");
if (startsToKeep == null) throw new IllegalArgumentException("startsToKeep cannot be null");
if (parser == null) throw new IllegalArgumentException("GenomeLocParser cannot be null");
List<ExactCall> calls = new LinkedList<ExactCall>();
// skip the header line
reader.readLine();
// skip the first "type" line
reader.readLine();
while (true) {
final VariantContextBuilder builder = new VariantContextBuilder();
final List<Allele> alleles = new ArrayList<Allele>();
final List<Genotype> genotypes = new ArrayList<Genotype>();
final double[] posteriors = new double[2];
final double[] priors = MathUtils.normalizeFromLog10(new double[] {0.5, 0.5}, true);
final List<Integer> mle = new ArrayList<Integer>();
final Map<Allele, Double> log10pNonRefByAllele = new HashMap<Allele, Double>();
long runtimeNano = -1;
GenomeLoc currentLoc = null;
while (true) {
final String line = reader.readLine();
if (line == null) return calls;
final String[] parts = line.split("\t");
final GenomeLoc lineLoc = parser.parseGenomeLoc(parts[0]);
final String variable = parts[1];
final String key = parts[2];
final String value = parts[3];
if (currentLoc == null) currentLoc = lineLoc;
if (variable.equals("type")) {
if (startsToKeep.isEmpty() || startsToKeep.contains(currentLoc.getStart())) {
builder.alleles(alleles);
final int stop = currentLoc.getStart() + alleles.get(0).length() - 1;
builder.chr(currentLoc.getContig()).start(currentLoc.getStart()).stop(stop);
builder.genotypes(genotypes);
final int[] mleInts = ArrayUtils.toPrimitive(mle.toArray(new Integer[] {}));
final AFCalcResult result =
new AFCalcResult(mleInts, 1, alleles, posteriors, priors, log10pNonRefByAllele);
calls.add(new ExactCall(builder.make(), runtimeNano, result));
}
break;
} else if (variable.equals("allele")) {
final boolean isRef = key.equals("0");
alleles.add(Allele.create(value, isRef));
} else if (variable.equals("PL")) {
final GenotypeBuilder gb = new GenotypeBuilder(key);
gb.PL(GenotypeLikelihoods.fromPLField(value).getAsPLs());
genotypes.add(gb.make());
} else if (variable.equals("log10PosteriorOfAFEq0")) {
posteriors[0] = Double.valueOf(value);
} else if (variable.equals("log10PosteriorOfAFGt0")) {
posteriors[1] = Double.valueOf(value);
} else if (variable.equals("MLE")) {
mle.add(Integer.valueOf(value));
} else if (variable.equals("pNonRefByAllele")) {
final Allele a = Allele.create(key);
log10pNonRefByAllele.put(a, Double.valueOf(value));
} else if (variable.equals("runtime.nano")) {
runtimeNano = Long.valueOf(value);
} else {
// nothing to do
}
}
}
}
/**
* Converts the input VCF into a format accepted by the Beagle imputation/analysis program.
*
* <p>
*
* <h2>Input</h2>
*
* <p>A VCF with variants to convert to Beagle format
*
* <h2>Outputs</h2>
*
* <p>A single text file which can be fed to Beagle
*
* <p>Optional: A file with a list of markers
*
* <h2>Examples</h2>
*
* <pre>
* java -Xmx2g -jar dist/GenomeAnalysisTK.jar -L 20 \
* -R reffile.fasta -T ProduceBeagleInput \
* -V path_to_input_vcf/inputvcf.vcf -o path_to_beagle_output/beagle_output
* </pre>
*/
public class ProduceBeagleInputWalker extends RodWalker<Integer, Integer> {
@ArgumentCollection
protected StandardVariantContextInputArgumentCollection variantCollection =
new StandardVariantContextInputArgumentCollection();
@Hidden
@Input(
fullName = "validation",
shortName = "validation",
doc = "Validation VCF file",
required = false)
public RodBinding<VariantContext> validation;
@Output(doc = "File to which BEAGLE input should be written", required = true)
protected PrintStream beagleWriter = null;
@Hidden
@Output(
doc = "File to which BEAGLE markers should be written",
shortName = "markers",
fullName = "markers",
required = false)
protected PrintStream markers = null;
int markerCounter = 1;
@Hidden
@Input(doc = "VQSqual calibration file", shortName = "cc", required = false)
protected File VQSRCalibrationFile = null;
protected VQSRCalibrationCurve VQSRCalibrator = null;
@Hidden
@Argument(doc = "VQSqual key", shortName = "vqskey", required = false)
protected String VQSLOD_KEY = "VQSqual";
@Hidden
@Argument(
fullName = "inserted_nocall_rate",
shortName = "nc_rate",
doc = "Rate (0-1) at which genotype no-calls will be randomly inserted, for testing",
required = false)
public double insertedNoCallRate = 0;
@Hidden
@Argument(
fullName = "validation_genotype_ptrue",
shortName = "valp",
doc = "Flat probability to assign to validation genotypes. Will override GL field.",
required = false)
public double validationPrior = -1.0;
@Hidden
@Argument(
fullName = "validation_bootstrap",
shortName = "bs",
doc = "Proportion of records to be used in bootstrap set",
required = false)
public double bootstrap = 0.0;
@Hidden
@Argument(
fullName = "bootstrap_vcf",
shortName = "bvcf",
doc = "Output a VCF with the records used for bootstrapping filtered out",
required = false)
VCFWriter bootstrapVCFOutput = null;
/**
* If sample gender is known, this flag should be set to true to ensure that Beagle treats male
* Chr X properly.
*/
@Argument(
fullName = "checkIsMaleOnChrX",
shortName = "checkIsMaleOnChrX",
doc =
"Set to true when Beagle-ing chrX and want to ensure male samples don't have heterozygous calls.",
required = false)
public boolean CHECK_IS_MALE_ON_CHR_X = false;
@Hidden
@Argument(
fullName = "variant_genotype_ptrue",
shortName = "varp",
doc =
"Flat probability prior to assign to variant (not validation) genotypes. Does not override GL field.",
required = false)
public double variantPrior = 0.96;
private Set<String> samples = null;
private Set<String> BOOTSTRAP_FILTER = new HashSet<String>(Arrays.asList("bootstrap"));
private int bootstrapSetSize = 0;
private int testSetSize = 0;
private CachingFormatter formatter = new CachingFormatter("%5.4f ", 100000);
private int certainFPs = 0;
public void initialize() {
samples =
SampleUtils.getSampleListWithVCFHeader(
getToolkit(), Arrays.asList(variantCollection.variants.getName()));
beagleWriter.print("marker alleleA alleleB");
for (String sample : samples)
beagleWriter.print(String.format(" %s %s %s", sample, sample, sample));
beagleWriter.println();
if (bootstrapVCFOutput != null) {
initializeVcfWriter();
}
if (VQSRCalibrationFile != null) {
VQSRCalibrator = VQSRCalibrationCurve.readFromFile(VQSRCalibrationFile);
logger.info("Read calibration curve");
VQSRCalibrator.printInfo(logger);
}
}
public Integer map(RefMetaDataTracker tracker, ReferenceContext ref, AlignmentContext context) {
if (tracker != null) {
GenomeLoc loc = context.getLocation();
VariantContext variant_eval = tracker.getFirstValue(variantCollection.variants, loc);
VariantContext validation_eval = tracker.getFirstValue(validation, loc);
if (goodSite(variant_eval, validation_eval)) {
if (useValidation(validation_eval, ref)) {
writeBeagleOutput(validation_eval, variant_eval, true, validationPrior);
return 1;
} else {
if (goodSite(variant_eval)) {
writeBeagleOutput(variant_eval, validation_eval, false, variantPrior);
return 1;
} else { // todo -- if the variant site is bad, validation is good, but not in bootstrap
// set -- what do?
return 0;
}
}
} else {
return 0;
}
} else {
return 0;
}
}
public boolean goodSite(VariantContext a, VariantContext b) {
return goodSite(a) || goodSite(b);
}
public boolean goodSite(VariantContext v) {
if (canBeOutputToBeagle(v)) {
if (VQSRCalibrator != null && VQSRCalibrator.certainFalsePositive(VQSLOD_KEY, v)) {
certainFPs++;
return false;
} else {
return true;
}
} else {
return false;
}
}
public static boolean canBeOutputToBeagle(VariantContext v) {
return v != null && !v.isFiltered() && v.isBiallelic() && v.hasGenotypes();
}
public boolean useValidation(VariantContext validation, ReferenceContext ref) {
if (goodSite(validation)) {
// if using record keeps us below expected proportion, use it
logger.debug(
String.format(
"boot: %d, test: %d, total: %d",
bootstrapSetSize, testSetSize, bootstrapSetSize + testSetSize + 1));
if ((bootstrapSetSize + 1.0) / (1.0 + bootstrapSetSize + testSetSize) <= bootstrap) {
if (bootstrapVCFOutput != null) {
bootstrapVCFOutput.add(
new VariantContextBuilder(validation).filters(BOOTSTRAP_FILTER).make());
}
bootstrapSetSize++;
return true;
} else {
if (bootstrapVCFOutput != null) {
bootstrapVCFOutput.add(validation);
}
testSetSize++;
return false;
}
} else {
if (validation != null && bootstrapVCFOutput != null) {
bootstrapVCFOutput.add(validation);
}
return false;
}
}
private static final double[] HAPLOID_FLAT_LOG10_LIKELIHOODS =
MathUtils.toLog10(new double[] {0.5, 0.0, 0.5});
private static final double[] DIPLOID_FLAT_LOG10_LIKELIHOODS =
MathUtils.toLog10(new double[] {0.33, 0.33, 0.33});
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());
}
private void writeSampleLikelihoods(
StringBuffer out, VariantContext vc, double[] log10Likelihoods) {
if (VQSRCalibrator != null) {
log10Likelihoods =
VQSRCalibrator.includeErrorRateInLikelihoods(VQSLOD_KEY, vc, log10Likelihoods);
}
double[] normalizedLikelihoods = MathUtils.normalizeFromLog10(log10Likelihoods);
// see if we need to randomly mask out genotype in this position.
for (double likeVal : normalizedLikelihoods) {
out.append(formatter.format(likeVal));
// out.append(String.format("%5.4f ",likeVal));
}
}
public Integer reduceInit() {
return 0; // Nothing to do here
}
public Integer reduce(Integer value, Integer sum) {
return value + sum; // count up the sites
}
public void onTraversalDone(Integer includedSites) {
logger.info("Sites included in beagle likelihoods file : " + includedSites);
logger.info(
String.format(
"Certain false positive found from recalibration curve : %d (%.2f%%)",
certainFPs, (100.0 * certainFPs) / (Math.max(certainFPs + includedSites, 1))));
}
private void initializeVcfWriter() {
final List<String> inputNames = Arrays.asList(validation.getName());
// setup the header fields
Set<VCFHeaderLine> hInfo = new HashSet<VCFHeaderLine>();
hInfo.addAll(VCFUtils.getHeaderFields(getToolkit(), inputNames));
hInfo.add(
new VCFFilterHeaderLine(
"bootstrap",
"This site used for genotype bootstrapping with ProduceBeagleInputWalker"));
bootstrapVCFOutput.writeHeader(
new VCFHeader(hInfo, SampleUtils.getUniqueSamplesFromRods(getToolkit(), inputNames)));
}
public static class CachingFormatter {
private int maxCacheSize = 0;
private String format;
private LRUCache<Double, String> cache;
public String getFormat() {
return format;
}
public String format(double value) {
String f = cache.get(value);
if (f == null) {
f = String.format(format, value);
cache.put(value, f);
// if ( cache.usedEntries() < maxCacheSize ) {
// System.out.printf("CACHE size %d%n", cache.usedEntries());
// } else {
// System.out.printf("CACHE is full %f%n", value);
// }
// }
// } else {
// System.out.printf("CACHE hit %f%n", value);
// }
}
return f;
}
public CachingFormatter(String format, int maxCacheSize) {
this.maxCacheSize = maxCacheSize;
this.format = format;
this.cache = new LRUCache<Double, String>(maxCacheSize);
}
}
/**
* An LRU cache, based on <code>LinkedHashMap</code>.
*
* <p>This cache has a fixed maximum number of elements (<code>cacheSize</code>). If the cache is
* full and another entry is added, the LRU (least recently used) entry is dropped.
*
* <p>This class is thread-safe. All methods of this class are synchronized.
*
* <p>Author: Christian d'Heureuse, Inventec Informatik AG, Zurich, Switzerland<br>
* Multi-licensed: EPL / LGPL / GPL / AL / BSD.
*/
public static class LRUCache<K, V> {
private static final float hashTableLoadFactor = 0.75f;
private LinkedHashMap<K, V> map;
private int cacheSize;
/**
* Creates a new LRU cache.
*
* @param cacheSize the maximum number of entries that will be kept in this cache.
*/
public LRUCache(int cacheSize) {
this.cacheSize = cacheSize;
int hashTableCapacity = (int) Math.ceil(cacheSize / hashTableLoadFactor) + 1;
map =
new LinkedHashMap<K, V>(hashTableCapacity, hashTableLoadFactor, true) {
// (an anonymous inner class)
private static final long serialVersionUID = 1;
@Override
protected boolean removeEldestEntry(Map.Entry<K, V> eldest) {
return size() > LRUCache.this.cacheSize;
}
};
}
/**
* Retrieves an entry from the cache.<br>
* The retrieved entry becomes the MRU (most recently used) entry.
*
* @param key the key whose associated value is to be returned.
* @return the value associated to this key, or null if no value with this key exists in the
* cache.
*/
public synchronized V get(K key) {
return map.get(key);
}
/**
* Adds an entry to this cache. The new entry becomes the MRU (most recently used) entry. If an
* entry with the specified key already exists in the cache, it is replaced by the new entry. If
* the cache is full, the LRU (least recently used) entry is removed from the cache.
*
* @param key the key with which the specified value is to be associated.
* @param value a value to be associated with the specified key.
*/
public synchronized void put(K key, V value) {
map.put(key, value);
}
/** Clears the cache. */
public synchronized void clear() {
map.clear();
}
/**
* Returns the number of used entries in the cache.
*
* @return the number of entries currently in the cache.
*/
public synchronized int usedEntries() {
return map.size();
}
/**
* Returns a <code>Collection</code> that contains a copy of all cache entries.
*
* @return a <code>Collection</code> with a copy of the cache content.
*/
public synchronized Collection<Map.Entry<K, V>> getAll() {
return new ArrayList<Map.Entry<K, V>>(map.entrySet());
}
} // end class LRUCache
}
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());
}
public final List<Sample> parse(
Reader reader, EnumSet<MissingPedField> missingFields, SampleDB sampleDB) {
final List<String> lines = new XReadLines(reader).readLines();
// What are the record offsets?
final int familyPos = missingFields.contains(MissingPedField.NO_FAMILY_ID) ? -1 : 0;
final int samplePos = familyPos + 1;
final int paternalPos = missingFields.contains(MissingPedField.NO_PARENTS) ? -1 : samplePos + 1;
final int maternalPos =
missingFields.contains(MissingPedField.NO_PARENTS) ? -1 : paternalPos + 1;
final int sexPos =
missingFields.contains(MissingPedField.NO_SEX) ? -1 : Math.max(maternalPos, samplePos) + 1;
final int phenotypePos =
missingFields.contains(MissingPedField.NO_PHENOTYPE)
? -1
: Math.max(sexPos, Math.max(maternalPos, samplePos)) + 1;
final int nExpectedFields =
MathUtils.arrayMaxInt(
Arrays.asList(samplePos, paternalPos, maternalPos, sexPos, phenotypePos))
+ 1;
// go through once and determine properties
int lineNo = 1;
boolean isQT = false;
final List<String[]> splits = new ArrayList<String[]>(lines.size());
for (final String line : lines) {
if (line.startsWith(commentMarker)) continue;
if (line.trim().equals("")) continue;
final String[] parts = line.split("\\s+");
if (parts.length != nExpectedFields)
throw new UserException.MalformedFile(
reader.toString(), "Bad PED line " + lineNo + ": wrong number of fields");
if (phenotypePos != -1) {
isQT = isQT || !CATAGORICAL_TRAIT_VALUES.contains(parts[phenotypePos]);
}
splits.add(parts);
lineNo++;
}
logger.info("Phenotype is other? " + isQT);
// now go through and parse each record
lineNo = 1;
final List<Sample> samples = new ArrayList<Sample>(splits.size());
for (final String[] parts : splits) {
String familyID = null, individualID, paternalID = null, maternalID = null;
Gender sex = Gender.UNKNOWN;
String quantitativePhenotype = Sample.UNSET_QT;
Affection affection = Affection.UNKNOWN;
if (familyPos != -1) familyID = maybeMissing(parts[familyPos]);
individualID = parts[samplePos];
if (paternalPos != -1) paternalID = maybeMissing(parts[paternalPos]);
if (maternalPos != -1) maternalID = maybeMissing(parts[maternalPos]);
if (sexPos != -1) {
if (parts[sexPos].equals(SEX_MALE)) sex = Gender.MALE;
else if (parts[sexPos].equals(SEX_FEMALE)) sex = Gender.FEMALE;
else sex = Gender.UNKNOWN;
}
if (phenotypePos != -1) {
if (isQT) {
if (parts[phenotypePos].equals(MISSING_VALUE1)) affection = Affection.UNKNOWN;
else {
affection = Affection.OTHER;
quantitativePhenotype = parts[phenotypePos];
}
} else {
if (parts[phenotypePos].equals(MISSING_VALUE1)) affection = Affection.UNKNOWN;
else if (parts[phenotypePos].equals(MISSING_VALUE2)) affection = Affection.UNKNOWN;
else if (parts[phenotypePos].equals(PHENOTYPE_UNAFFECTED))
affection = Affection.UNAFFECTED;
else if (parts[phenotypePos].equals(PHENOTYPE_AFFECTED)) affection = Affection.AFFECTED;
else
throw new ReviewedStingException(
"Unexpected phenotype type " + parts[phenotypePos] + " at line " + lineNo);
}
}
final Sample s =
new Sample(
individualID,
sampleDB,
familyID,
paternalID,
maternalID,
sex,
affection,
quantitativePhenotype);
samples.add(s);
sampleDB.addSample(s);
lineNo++;
}
for (final Sample sample : new ArrayList<Sample>(samples)) {
Sample dad =
maybeAddImplicitSample(
sampleDB, sample.getPaternalID(), sample.getFamilyID(), Gender.MALE);
if (dad != null) samples.add(dad);
Sample mom =
maybeAddImplicitSample(
sampleDB, sample.getMaternalID(), sample.getFamilyID(), Gender.FEMALE);
if (mom != null) samples.add(mom);
}
return samples;
}
private final double getRefBinomialProb(final int depth) {
if (depth < binomialProbabilityDepthCache.length) return binomialProbabilityDepthCache[depth];
else return MathUtils.binomialProbability(0, depth, 0.5);
}
static {
for (int i = 1; i < binomialProbabilityDepthCache.length; i++) {
binomialProbabilityDepthCache[i] = MathUtils.binomialProbability(0, i, 0.5);
}
}