@Override
protected void doWork(String inputSource, VcfIterator r, VariantContextWriter w)
throws IOException {
VCFHeader header = r.getHeader();
VCFHeader h2 = new VCFHeader(header.getMetaDataInInputOrder(), header.getSampleNamesInOrder());
h2.addMetaDataLine(
new VCFInfoHeaderLine(
TAG,
VCFHeaderLineCount.UNBOUNDED,
VCFHeaderLineType.String,
"metadata added from " + TABIX + " . Format was " + FORMAT));
h2.addMetaDataLine(
new VCFHeaderLine(
getClass().getSimpleName() + "CmdLine", String.valueOf(getProgramCommandLine())));
h2.addMetaDataLine(
new VCFHeaderLine(getClass().getSimpleName() + "Version", String.valueOf(getVersion())));
h2.addMetaDataLine(
new VCFHeaderLine(
getClass().getSimpleName() + "HtsJdkVersion", HtsjdkVersion.getVersion()));
h2.addMetaDataLine(
new VCFHeaderLine(getClass().getSimpleName() + "HtsJdkHome", HtsjdkVersion.getHome()));
SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(header);
w.writeHeader(h2);
while (r.hasNext()) {
VariantContext ctx = progress.watch(r.next());
Set<String> annotations = new HashSet<String>();
CloseableIterator<BedLine> iter =
this.bedReader.iterator(ctx.getContig(), ctx.getStart() - 1, ctx.getEnd() + 1);
while (iter.hasNext()) {
BedLine bedLine = iter.next();
if (!ctx.getContig().equals(bedLine.getContig())) continue;
if (ctx.getStart() - 1 >= bedLine.getEnd()) continue;
if (ctx.getEnd() - 1 < bedLine.getStart()) continue;
String newannot = this.parsedFormat.toString(bedLine);
if (!newannot.isEmpty()) annotations.add(VCFUtils.escapeInfoField(newannot));
}
CloserUtil.close(iter);
if (annotations.isEmpty()) {
w.add(ctx);
continue;
}
VariantContextBuilder vcb = new VariantContextBuilder(ctx);
vcb.attribute(TAG, annotations.toArray());
w.add(vcb.make());
incrVariantCount();
if (checkOutputError()) break;
}
progress.finish();
}
@Override
public void execute() {
VariantContextWriter vcw =
new VariantContextWriterBuilder()
.setOutputFile(out)
.clearIndexCreator()
.setReferenceDictionary(VCF.getFileHeader().getSequenceDictionary())
.build();
VCFHeader vh = VCF.getFileHeader();
vh.addMetaDataLine(
new VCFInfoHeaderLine(
"region", 1, VCFHeaderLineType.String, "region type in which the variant occurs"));
vcw.writeHeader(VCF.getFileHeader());
IntervalTreeMap<String> itm = loadRegions();
int num = 0;
for (VariantContext vc : VCF) {
if (!vc.isFiltered()
&& vc.getGenotype(CHILD).isCalled()
&& vc.getGenotype(MOTHER).isCalled()
&& vc.getGenotype(FATHER).isCalled()
&& vc.getGenotype(CHILD).getType().equals(GenotypeType.HET)
&& vc.getGenotype(MOTHER).getType().equals(GenotypeType.HOM_REF)
&& vc.getGenotype(FATHER).getType().equals(GenotypeType.HOM_REF)
&& (!vc.getGenotype(CHILD).hasAnyAttribute("GQ")
|| (vc.getGenotype(CHILD).getGQ() > GQ_THRESHOLD
&& vc.getGenotype(MOTHER).getGQ() > GQ_THRESHOLD
&& vc.getGenotype(FATHER).getGQ() > GQ_THRESHOLD))
&& (!vc.getGenotype(CHILD).hasAnyAttribute("DP")
|| (vc.getGenotype(CHILD).getDP() > DP_THRESHOLD
&& vc.getGenotype(MOTHER).getDP() > DP_THRESHOLD
&& vc.getGenotype(FATHER).getDP() > DP_THRESHOLD))) {
vcw.add(vc);
num++;
}
}
vcw.close();
log.info("num: {}", num);
}
@Override
protected Object doWork() {
IOUtil.assertFileIsReadable(INPUT);
IOUtil.assertFileIsReadable(REFERENCE_SEQUENCE);
IOUtil.assertFileIsReadable(CHAIN);
IOUtil.assertFileIsWritable(OUTPUT);
IOUtil.assertFileIsWritable(REJECT);
////////////////////////////////////////////////////////////////////////
// Setup the inputs
////////////////////////////////////////////////////////////////////////
final LiftOver liftOver = new LiftOver(CHAIN);
final VCFFileReader in = new VCFFileReader(INPUT, false);
logger.info("Loading up the target reference genome.");
final ReferenceSequenceFileWalker walker = new ReferenceSequenceFileWalker(REFERENCE_SEQUENCE);
final Map<String, byte[]> refSeqs = new HashMap<>();
for (final SAMSequenceRecord rec : walker.getSequenceDictionary().getSequences()) {
refSeqs.put(rec.getSequenceName(), walker.get(rec.getSequenceIndex()).getBases());
}
CloserUtil.close(walker);
////////////////////////////////////////////////////////////////////////
// Setup the outputs
////////////////////////////////////////////////////////////////////////
final VCFHeader inHeader = in.getFileHeader();
final VCFHeader outHeader = new VCFHeader(inHeader);
outHeader.setSequenceDictionary(walker.getSequenceDictionary());
final VariantContextWriter out =
new VariantContextWriterBuilder()
.setOption(Options.INDEX_ON_THE_FLY)
.setOutputFile(OUTPUT)
.setReferenceDictionary(walker.getSequenceDictionary())
.build();
out.writeHeader(outHeader);
final VariantContextWriter rejects =
new VariantContextWriterBuilder()
.setOutputFile(REJECT)
.unsetOption(Options.INDEX_ON_THE_FLY)
.build();
final VCFHeader rejectHeader = new VCFHeader(in.getFileHeader());
for (final VCFFilterHeaderLine line : FILTERS) rejectHeader.addMetaDataLine(line);
rejects.writeHeader(rejectHeader);
////////////////////////////////////////////////////////////////////////
// Read the input VCF, lift the records over and write to the sorting
// collection.
////////////////////////////////////////////////////////////////////////
long failedLiftover = 0, failedAlleleCheck = 0, total = 0;
logger.info("Lifting variants over and sorting.");
final SortingCollection<VariantContext> sorter =
SortingCollection.newInstance(
VariantContext.class,
new VCFRecordCodec(outHeader),
outHeader.getVCFRecordComparator(),
MAX_RECORDS_IN_RAM,
TMP_DIR);
ProgressLogger progress = new ProgressLogger(logger, 1000000, "read");
for (final VariantContext ctx : in) {
++total;
final Interval source =
new Interval(
ctx.getContig(),
ctx.getStart(),
ctx.getEnd(),
false,
ctx.getContig() + ":" + ctx.getStart() + "-" + ctx.getEnd());
final Interval target = liftOver.liftOver(source, 1.0);
if (target == null) {
rejects.add(new VariantContextBuilder(ctx).filter(FILTER_CANNOT_LIFTOVER).make());
failedLiftover++;
} else {
// Fix the alleles if we went from positive to negative strand
final List<Allele> alleles = new ArrayList<>();
for (final Allele oldAllele : ctx.getAlleles()) {
if (target.isPositiveStrand() || oldAllele.isSymbolic()) {
alleles.add(oldAllele);
} else {
alleles.add(
Allele.create(
SequenceUtil.reverseComplement(oldAllele.getBaseString()),
oldAllele.isReference()));
}
}
// Build the new variant context
final VariantContextBuilder builder =
new VariantContextBuilder(
ctx.getSource(), target.getContig(), target.getStart(), target.getEnd(), alleles);
builder.id(ctx.getID());
builder.attributes(ctx.getAttributes());
builder.genotypes(ctx.getGenotypes());
builder.filters(ctx.getFilters());
builder.log10PError(ctx.getLog10PError());
// Check that the reference allele still agrees with the reference sequence
boolean mismatchesReference = false;
for (final Allele allele : builder.getAlleles()) {
if (allele.isReference()) {
final byte[] ref = refSeqs.get(target.getContig());
final String refString =
StringUtil.bytesToString(ref, target.getStart() - 1, target.length());
if (!refString.equalsIgnoreCase(allele.getBaseString())) {
mismatchesReference = true;
}
break;
}
}
if (mismatchesReference) {
rejects.add(new VariantContextBuilder(ctx).filter(FILTER_MISMATCHING_REF_ALLELE).make());
failedAlleleCheck++;
} else {
sorter.add(builder.make());
}
}
progress.record(ctx.getContig(), ctx.getStart());
}
final NumberFormat pfmt = new DecimalFormat("0.0000%");
final String pct = pfmt.format((failedLiftover + failedAlleleCheck) / (double) total);
logger.info("Processed ", total, " variants.");
logger.info(Long.toString(failedLiftover), " variants failed to liftover.");
logger.info(
Long.toString(failedAlleleCheck),
" variants lifted over but had mismatching reference alleles after lift over.");
logger.info(pct, " of variants were not successfully lifted over and written to the output.");
rejects.close();
in.close();
////////////////////////////////////////////////////////////////////////
// Write the sorted outputs to the final output file
////////////////////////////////////////////////////////////////////////
sorter.doneAdding();
progress = new ProgressLogger(logger, 1000000, "written");
logger.info("Writing out sorted records to final VCF.");
for (final VariantContext ctx : sorter) {
out.add(ctx);
progress.record(ctx.getContig(), ctx.getStart());
}
out.close();
sorter.cleanup();
return null;
}
@Override
protected void doWork(String source, VcfIterator in, VariantContextWriter out)
throws IOException {
try {
VCFHeader header = in.getHeader();
VCFHeader h2 = new VCFHeader(header);
h2.addMetaDataLine(
new VCFHeaderLine(
getClass().getSimpleName() + "CmdLine", String.valueOf(getProgramCommandLine())));
h2.addMetaDataLine(
new VCFHeaderLine(getClass().getSimpleName() + "Version", String.valueOf(getVersion())));
h2.addMetaDataLine(
new VCFHeaderLine(
getClass().getSimpleName() + "HtsJdkVersion", HtsjdkVersion.getVersion()));
h2.addMetaDataLine(
new VCFHeaderLine(getClass().getSimpleName() + "HtsJdkHome", HtsjdkVersion.getHome()));
out.writeHeader(h2);
final VepPredictionParser vepParser = new VepPredictionParser(header);
final SnpEffPredictionParser snpEffparser = new SnpEffPredictionParser(header);
final MyPredictionParser myPredParser = new MyPredictionParser(header);
SAMSequenceDictionaryProgress progress =
new SAMSequenceDictionaryProgress(header.getSequenceDictionary());
while (in.hasNext()) {
this.checkKnimeCancelled();
VariantContext ctx = progress.watch(in.next());
boolean keep = false;
for (SnpEffPredictionParser.SnpEffPrediction pred : snpEffparser.getPredictions(ctx)) {
if (hasUserTem(pred.getSOTerms())) {
keep = true;
break;
}
}
if (!keep) {
for (VepPredictionParser.VepPrediction pred : vepParser.getPredictions(ctx)) {
if (hasUserTem(pred.getSOTerms())) {
keep = true;
break;
}
}
}
if (!keep) {
for (MyPredictionParser.MyPrediction pred : myPredParser.getPredictions(ctx)) {
if (hasUserTem(pred.getSOTerms())) {
keep = true;
break;
}
}
}
if (isInverseResult()) keep = !keep;
if (keep) {
incrVariantCount();
out.add(ctx);
}
if (checkOutputError()) break;
}
progress.finish();
} finally {
}
}
@Override
protected void doWork(VcfIterator r, VariantContextWriter w) throws IOException {
AbstractVCFCodec codeIn3 = VCFUtils.createDefaultVCFCodec();
String line;
StringWriter sw = new StringWriter();
LOG.info("opening tabix file: " + this.TABIX);
TabixReader tabix = new TabixReader(this.TABIX);
while ((line = tabix.readLine()) != null) {
if (!line.startsWith(VCFHeader.HEADER_INDICATOR)) {
break;
}
sw.append(line).append("\n");
}
VCFHeader header3 =
(VCFHeader)
codeIn3.readActualHeader(
new LineIteratorImpl(
LineReaderUtil.fromBufferedStream(
new ByteArrayInputStream(sw.toString().getBytes()))));
VCFHeader header1 = r.getHeader();
VCFHeader h2 =
new VCFHeader(header1.getMetaDataInInputOrder(), header1.getSampleNamesInOrder());
for (String infoId : this.INFO_IDS) {
VCFInfoHeaderLine vihl = header3.getInfoHeaderLine(infoId);
if (vihl == null) {
LOG.warn("Not INFO=" + infoId + " in " + TABIX);
continue;
}
if (h2.getInfoHeaderLine(infoId) != null) {
LOG.warn("Input already contains INFO=" + vihl);
}
h2.addMetaDataLine(vihl);
}
if (ALT_CONFLICT_FLAG != null) {
h2.addMetaDataLine(
new VCFInfoHeaderLine(
ALT_CONFLICT_FLAG,
1,
VCFHeaderLineType.Flag,
"conflict ALT allele with " + this.TABIX));
}
w.writeHeader(h2);
while (r.hasNext()) {
VariantContext ctx1 = r.next();
VariantContextBuilder vcb = new VariantContextBuilder(ctx1);
String line2;
String BEST_ID = null;
boolean best_id_match_alt = false;
List<VariantContext> variantsList = new ArrayList<VariantContext>();
int[] array = tabix.parseReg(ctx1.getChr() + ":" + (ctx1.getStart()) + "-" + (ctx1.getEnd()));
TabixReader.Iterator iter = null;
if (array != null && array.length == 3 && array[0] != -1 && array[1] >= 0 && array[2] >= 0) {
iter = tabix.query(array[0], array[1], array[2]);
} else {
LOG.info("Cannot get " + ctx1.getChr() + ":" + (ctx1.getStart()) + "-" + (ctx1.getEnd()));
}
while (iter != null && (line2 = iter.next()) != null) {
VariantContext ctx3 = codeIn3.decode(line2);
if (ctx3.getStart() != ctx1.getStart()) continue;
if (ctx3.getEnd() != ctx1.getEnd()) continue;
if (ctx1.getReference().equals(ctx3.getReference())
&& ctx1.getAlternateAlleles().equals(ctx3.getAlternateAlleles())) {
variantsList.clear();
variantsList.add(ctx3);
break;
} else {
variantsList.add(ctx3);
}
}
for (VariantContext ctx3 : variantsList) {
if (this.REF_ALLELE_MATTERS && !ctx1.getReference().equals(ctx3.getReference())) {
continue;
}
if (this.ALT_ALLELES_MATTERS
&& !ctx1.getAlternateAlleles().equals(ctx3.getAlternateAlleles())) {
continue;
}
if (ctx3.getID() != null && this.REPLACE_ID) {
if (BEST_ID != null && best_id_match_alt) {
// nothing
} else {
BEST_ID = ctx3.getID();
best_id_match_alt = ctx1.getAlternateAlleles().equals(ctx3.getAlternateAlleles());
}
}
for (String id : this.INFO_IDS) {
Object info3 = ctx3.getAttribute(id);
if (info3 == null) {
continue;
}
Object info1 = ctx1.getAttribute(id);
if (info1 != null && !this.REPLACE_INFO_FIELD) {
continue;
}
vcb.attribute(id, info3);
}
if (ALT_CONFLICT_FLAG != null
&& !ctx1.getAlternateAlleles().equals(ctx3.getAlternateAlleles())) {
vcb.attribute(ALT_CONFLICT_FLAG, true);
}
}
if (BEST_ID != null) {
vcb.id(BEST_ID);
}
w.add(vcb.make());
}
tabix.close();
}
@Override
protected void doWork(VcfIterator r, VariantContextWriter w) throws IOException {
long nChanged = 0L;
final String TAG = "INDELFIXED";
VCFHeader header = r.getHeader();
VCFHeader h2 = new VCFHeader(header.getMetaDataInInputOrder(), header.getSampleNamesInOrder());
h2.addMetaDataLine(
new VCFInfoHeaderLine(TAG, 1, VCFHeaderLineType.String, "Fix Indels for @SolenaLS."));
w.writeHeader(h2);
final Pattern dna = Pattern.compile("[ATGCatgc]+");
while (r.hasNext()) {
VariantContext ctx = r.next();
VariantContextBuilder b = new VariantContextBuilder(ctx);
List<Allele> alleles = ctx.getAlternateAlleles();
if (alleles.size() != 1
|| !dna.matcher(ctx.getReference().getBaseString()).matches()
|| !dna.matcher(alleles.get(0).getBaseString()).matches()) {
w.add(ctx);
continue;
}
StringBuffer ref = new StringBuffer(ctx.getReference().getBaseString().toUpperCase());
StringBuffer alt = new StringBuffer(alleles.get(0).getBaseString().toUpperCase());
int start = ctx.getStart();
int end = ctx.getEnd();
boolean changed = false;
/** ** we trim on the right side *** */
// REF=TGCTGCGGGGGCCGCTGCGGGGG ALT=TGCTGCGGGGG
while (alt.length() > 1
&& alt.length() < ref.length()
&& ref.charAt(ref.length() - 1) == alt.charAt(alt.length() - 1)) {
changed = true;
ref.setLength(ref.length() - 1);
alt.deleteCharAt(alt.length() - 1);
end--;
}
// REF=TGCTGCGGGGG ALT= TGCTGCGGGGGCCGCTGCGGGGG
while (ref.length() > 1
&& alt.length() > ref.length()
&& ref.charAt(ref.length() - 1) == alt.charAt(alt.length() - 1)) {
changed = true;
ref.setLength(ref.length() - 1);
alt.deleteCharAt(alt.length() - 1);
end--;
}
/** ** we trim on the left side *** */
// REF=TGCTGCGGGGGCCGCTGCGGGGG ALT=TGCTGCGGGGG
while (alt.length() > 1 && alt.length() < ref.length() && ref.charAt(0) == alt.charAt(0)) {
changed = true;
ref.deleteCharAt(0);
alt.deleteCharAt(0);
start++;
}
// REF=TGCTGCGGGGG ALT= TGCTGCGGGGGCCGCTGCGGGGG
while (ref.length() > 1 && alt.length() > ref.length() && ref.charAt(0) == alt.charAt(0)) {
changed = true;
ref.deleteCharAt(0);
alt.deleteCharAt(0);
start++;
}
if (!changed) {
w.add(ctx);
continue;
}
/*
LOG.info(line);
LOG.info("ctx.getStart() "+ctx.getStart());
LOG.info("ctx.getEnd() "+ ctx.getEnd());
LOG.info("start " + start);
LOG.info("end "+end);
LOG.info("ref " + ref.toString());
LOG.info("alt "+alt.toString());
*/
Allele newRef = Allele.create(ref.toString(), true);
Allele newAlt = Allele.create(alt.toString(), false);
Allele newalleles[] = new Allele[] {newRef, newAlt};
b.attribute(
TAG,
ctx.getReference().getBaseString()
+ "|"
+ alleles.get(0).getBaseString()
+ "|"
+ ctx.getStart());
b.start(start);
b.stop(end);
b.alleles(Arrays.asList(newalleles));
nChanged++;
VariantContext ctx2 = b.make();
try {
w.add(ctx2);
} catch (TribbleException err) {
error(err, "Cannot convert new context:" + ctx2 + " old context:" + ctx);
w.add(ctx);
}
}
info("indels changed:" + nChanged);
}
