Example 41 with CigarOperator
use of htsjdk.samtools.CigarOperator in project jvarkit by lindenb.
the class TView method paint.
void paint(final PrintStream out) {
final Colorizer colorizer;
switch(this.formatOut) {
case html:
colorizer = new HtmlColorizer(out);
break;
case tty:
colorizer = new AnsiColorizer(out);
break;
case plain:
colorizer = new Colorizer(out);
break;
default:
throw new IllegalStateException();
}
if (interval == null) {
LOG.warn("No interval defined");
return;
}
final GenomicSequence contigSequence;
final Function<Integer, Character> refPosToBase;
if (indexedFastaSequenceFile != null) {
final SAMSequenceDictionary dict = SAMSequenceDictionaryExtractor.extractDictionary(referenceFile);
if (dict.getSequence(this.interval.getContig()) == null) {
LOG.warn("No interval with contig " + interval + " in REF");
return;
}
contigSequence = new GenomicSequence(indexedFastaSequenceFile, interval.getContig());
refPosToBase = POS -> {
if (POS < 0 || POS >= contigSequence.length())
return 'N';
return contigSequence.charAt(POS);
};
} else {
contigSequence = null;
refPosToBase = POS -> 'N';
}
/**
* test if genomic position is in interval
*/
final Predicate<Integer> testInInterval = new Predicate<Integer>() {
@Override
public boolean test(final Integer pos) {
return interval.getStart() <= pos && pos <= interval.getEnd();
}
};
final int pixelWidth = this.interval.length();
final Map<Integer, Integer> genomicpos2insertlen = new TreeMap<>();
final Map<String, List<SAMRecord>> group2record = new TreeMap<>();
for (final SamReader samReader : this.samReaders) {
SAMRecordIterator iter = samReader.query(this.interval.getContig(), this.interval.getStart(), this.interval.getEnd(), false);
while (iter.hasNext()) {
final SAMRecord rec = iter.next();
if (rec.getReadUnmappedFlag())
continue;
if (rec.getCigar() == null)
continue;
if (getRecordFilter().filterOut(rec))
continue;
if (!rec.getContig().equals(interval.getContig()))
continue;
if (right().apply(rec) < this.interval.getStart())
continue;
if (this.interval.getEnd() < left().apply(rec))
continue;
String group = this.groupBy.getPartion(rec);
if (group == null || group.isEmpty()) {
group = "undefined_" + this.groupBy.name();
}
List<SAMRecord> records = group2record.get(group);
if (records == null) {
records = new ArrayList<>();
group2record.put(group, records);
}
records.add(rec);
// loop over cigar, get the longest insert
int refpos = rec.getAlignmentStart();
for (final CigarElement ce : rec.getCigar().getCigarElements()) {
if (!this.showInsertions)
break;
final CigarOperator op = ce.getOperator();
if (op.equals(CigarOperator.I) && testInInterval.test(refpos)) {
final Integer longestInsert = genomicpos2insertlen.get(refpos);
if (longestInsert == null || longestInsert.compareTo(ce.getLength()) < 0) {
genomicpos2insertlen.put(refpos, ce.getLength());
}
}
if (op.consumesReferenceBases()) {
refpos += ce.getLength();
}
if (refpos > interval.getEnd())
break;
}
}
CloserUtil.close(iter);
CloserUtil.close(samReader);
}
/**
* compute where are the insertions
*/
// LOG.debug(genomicpos2insertlen);
final Predicate<Integer> insertIsPresentAtX = SCREENX -> {
int x = 0;
int ref = interval.getStart();
while (x < pixelWidth) {
if (x > SCREENX)
return false;
final Integer insertLen = genomicpos2insertlen.get(ref);
if (insertLen == null) {
++x;
++ref;
} else {
if (x <= SCREENX && SCREENX < x + insertLen)
return true;
// (+1) I DON'T UNDERSTAND WHY, BUT IT WORKS
x += (insertLen + 1);
++ref;
}
}
return false;
};
final Function<Character, AnsiColor> base2ansiColor = BASE -> {
switch(Character.toUpperCase(BASE)) {
case 'A':
return AnsiColor.BLUE;
case 'T':
return AnsiColor.GREEN;
case 'G':
return AnsiColor.CYAN;
case 'C':
return AnsiColor.YELLOW;
default:
return null;
}
};
/**
* print interval title
*/
out.println(interval.getContig() + ":" + interval.getStart() + "-" + interval.getEnd());
/**
* paint base position
*/
int ref = this.interval.getStart();
int x = 0;
out.print(margin("POS:"));
while (x < pixelWidth) {
if (insertIsPresentAtX.test(x)) {
colorizer.pen(AnsiColor.RED).print("^");
++x;
} else if ((ref - this.interval.getStart()) % 10 == 0) {
final String f = String.format("%d", ref);
for (int i = 0; i < f.length() && x < pixelWidth; ++i) {
colorizer.pen(AnsiColor.GREEN).print(f.charAt(i));
if (!insertIsPresentAtX.test(x))
++ref;
++x;
}
} else {
out.print(".");
++ref;
++x;
}
}
out.println();
/* paint ref base */
out.print(margin("REF:"));
ref = this.interval.getStart();
x = 0;
while (x < pixelWidth) {
if (insertIsPresentAtX.test(x)) {
colorizer.paper(AnsiColor.YELLOW).print("*");
++x;
} else {
char refBase = refPosToBase.apply(ref - 1);
colorizer.pen(base2ansiColor.apply(refBase)).print(refBase);
++ref;
++x;
}
}
out.println();
/* loop over samples **/
for (final String groupName : group2record.keySet()) {
if (this.maxReadRowPerGroup == 0)
continue;
final ConsensusBuilder consensus = new ConsensusBuilder();
int y_group = 0;
final List<List<SAMRecord>> rows = new ArrayList<>();
out.println(margin(""));
switch(this.layoutReads) {
case name:
{
rows.addAll(group2record.get(groupName).stream().sorted((R1, R2) -> R1.getReadName().compareTo(R2.getReadName())).map(R -> Collections.singletonList(R)).collect(Collectors.toList()));
break;
}
default:
{
/* pileup reads */
for (final SAMRecord rec : group2record.get(groupName)) {
int y = 0;
for (y = 0; y < rows.size(); ++y) {
final List<SAMRecord> row = rows.get(y);
final SAMRecord last = row.get(row.size() - 1);
if (right().apply(last) + this.distance_between_reads < left().apply(rec)) {
row.add(rec);
break;
}
}
if (y == rows.size()) {
final List<SAMRecord> row = new ArrayList<>();
row.add(rec);
rows.add(row);
}
}
break;
}
}
// each row is only one read, so we need to print the groupName
if (layoutReads == LayoutReads.name) {
out.print(margin(groupName));
out.println();
}
/* print each row */
for (final List<SAMRecord> row : rows) {
++y_group;
boolean print_this_line = (this.maxReadRowPerGroup < 0 || y_group <= this.maxReadRowPerGroup);
if (print_this_line) {
// each row is only one read, print the read name
if (layoutReads == LayoutReads.name) {
String readName = row.get(0).getReadName();
if (row.get(0).getReadPairedFlag()) {
if (row.get(0).getFirstOfPairFlag()) {
readName += "/1";
}
if (row.get(0).getSecondOfPairFlag()) {
readName += "/2";
}
}
out.print(margin(readName));
} else {
out.print(margin(y_group == 1 ? groupName : ""));
}
}
ref = interval.getStart();
x = 0;
for (final SAMRecord rec : row) {
int readRef = left().apply(rec);
// pad before record
while (x < pixelWidth && ref < readRef && testInInterval.test(ref)) {
if (!insertIsPresentAtX.test(x))
++ref;
++x;
if (print_this_line)
out.print(' ');
consensus.add(' ');
}
int readpos = 0;
/* get read base function */
final Function<Integer, Character> baseAt = new Function<Integer, Character>() {
@Override
public Character apply(final Integer readpos) {
final byte[] readBases = rec.getReadBases();
if (readBases == SAMRecord.NULL_SEQUENCE)
return 'N';
if (readpos < 0 || readpos >= rec.getReadLength())
return '?';
return (char) readBases[readpos];
}
};
for (final CigarElement ce : rec.getCigar()) {
final CigarOperator op = ce.getOperator();
if (op.equals(CigarOperator.PADDING))
continue;
/* IN INSERTION, only print if showInsertions is true */
if (this.showInsertions && op.equals(CigarOperator.I)) {
int cigarIdx = 0;
while (x < pixelWidth && cigarIdx < ce.getLength()) {
if (testInInterval.test(readRef)) {
final char readbase = baseAt.apply(readpos);
if (print_this_line)
colorizer.paper(AnsiColor.RED).print(readbase);
consensus.add(readbase);
++x;
}
++cigarIdx;
++readpos;
}
continue;
}
int cigarIdx = 0;
while (x < pixelWidth && cigarIdx < ce.getLength()) {
colorizer.clear();
// pad before base
while (x < pixelWidth && testInInterval.test(readRef) && (insertIsPresentAtX.test(x))) {
++x;
if (print_this_line)
colorizer.paper(AnsiColor.YELLOW).print("*");
consensus.add(' ');
continue;
}
switch(op) {
case I:
{
// if visible, processed above
if (showInsertions)
throw new IllegalStateException();
readpos++;
break;
}
case P:
break;
case H:
{
if (showClip) {
if (testInInterval.test(readRef)) {
if (print_this_line)
colorizer.paper(AnsiColor.YELLOW).print('N');
// CLIPPED base not part of consensus
consensus.add(' ');
++x;
}
++readRef;
}
break;
}
case S:
{
if (showClip) {
if (testInInterval.test(readRef)) {
final char readBase = baseAt.apply(readpos);
if (print_this_line)
colorizer.paper(AnsiColor.YELLOW).print(readBase);
// CLIPPED base not part of consensus
consensus.add(' ');
++x;
}
++readpos;
++readRef;
} else {
readpos++;
}
break;
}
case D:
case N:
{
if (testInInterval.test(readRef)) {
if (print_this_line)
colorizer.paper(AnsiColor.RED).print('-');
// deletion not not part of consensus
consensus.add(' ');
++x;
}
++readRef;
break;
}
case EQ:
case M:
case X:
{
if (testInInterval.test(readRef)) {
final char refBase = Character.toUpperCase(refPosToBase.apply(readRef - 1));
char readBase = Character.toUpperCase(baseAt.apply(readpos));
consensus.add(readBase);
colorizer.pen(base2ansiColor.apply(readBase));
if (op.equals(CigarOperator.X) || (refBase != 'N' && readBase != 'N' && readBase != refBase)) {
colorizer.pen(AnsiColor.RED);
} else if (hideBases) {
if (rec.getReadNegativeStrandFlag()) {
readBase = ',';
} else {
readBase = '.';
}
}
if (showReadName) {
final String readName = rec.getReadName();
if (readpos < 0 || readpos >= readName.length()) {
readBase = '_';
} else {
readBase = readName.charAt(readpos);
}
}
if (rec.getReadNegativeStrandFlag()) {
readBase = Character.toLowerCase(readBase);
} else {
readBase = Character.toUpperCase(readBase);
}
if (print_this_line)
colorizer.print(readBase);
++x;
}
++readpos;
++readRef;
break;
}
}
++cigarIdx;
}
}
// end of loop cigar
ref = readRef;
}
// out.println( " "+ref+" "+row.get(0).getAlignmentStart()+" "+row.get(0).getCigarString()+" "+row.get(0).getReadString());
while (x < pixelWidth) {
if (print_this_line)
out.print(" ");
++x;
}
if (print_this_line)
out.println();
consensus.eol();
if (out.checkError())
break;
}
if (out.checkError())
break;
if (!this.hideConsensus && consensus.bases.stream().anyMatch(C -> C.getCoverage() > 0)) {
out.print(margin(groupName + " CONSENSUS"));
x = 0;
ref = interval.getStart();
while (x < consensus.bases.size()) {
final char refBase = Character.toUpperCase(refPosToBase.apply(ref - 1));
final char consensusBase = consensus.bases.get(x).getConsensus();
if (Character.isWhitespace(consensusBase)) {
// nothing
} else if (refBase != 'N' && consensusBase != refBase) {
colorizer.pen(AnsiColor.RED);
} else {
colorizer.pen(base2ansiColor.apply(consensusBase));
}
if (!insertIsPresentAtX.test(x))
++ref;
colorizer.print(consensusBase);
++x;
}
out.println();
}
if (this.numCoverageRows > 0) {
int minCov = consensus.bases.stream().mapToInt(C -> C.getCoverage()).min().orElse(0);
final int maxCov = consensus.bases.stream().mapToInt(C -> C.getCoverage()).max().orElse(0);
for (int y = 0; maxCov > 0 && y < this.numCoverageRows; ++y) {
if (minCov == maxCov)
minCov--;
double fract = (maxCov - minCov) / ((double) this.numCoverageRows);
int inverse_y = (this.numCoverageRows - 1) - y;
int d0 = (int) ((fract) * inverse_y);
// int d1 = (int)((fract) * (inverse_y+1));
out.print(margin(y == 0 ? groupName + " " + maxCov : (y + 1 == this.numCoverageRows ? String.valueOf(minCov) : "")));
for (x = 0; x < consensus.bases.size(); ++x) {
int depth = consensus.bases.get(x).getCoverage() - minCov;
colorizer.print(depth >= d0 ? BLACK_SQUARE : ' ');
}
out.println();
}
}
}
if (this.tabixKnownGene != null && this.indexedFastaSequenceFile != null) {
final List<KnownGene> genes = this.tabixKnownGene.getItemsInInterval(this.interval);
if (!genes.isEmpty()) {
out.println(this.knownGeneUri);
for (final KnownGene gene : genes) {
final KnownGene.CodingRNA codingRna = gene.getCodingRNA(contigSequence);
final KnownGene.Peptide peptide = codingRna.getPeptide();
out.print(margin(gene.getName()));
x = 0;
int ref0 = this.interval.getStart() - 1;
while (x < pixelWidth) {
if (insertIsPresentAtX.test(x)) {
out.print("*");
++x;
} else {
char pepChar = ' ';
if (ref0 >= gene.getTxStart() && ref0 < gene.getTxEnd()) {
pepChar = (gene.isPositiveStrand() ? '>' : '<');
int pepIdx = peptide.convertGenomicToPeptideCoordinate(ref0);
if (pepIdx != -1) {
final String aa3 = GeneticCode.aminoAcidTo3Letters(peptide.charAt(pepIdx));
final int[] offset = peptide.convertToGenomicCoordinates(pepIdx);
if (offset != null && offset.length == 3 && aa3 != null && aa3.length() == 3) {
if (offset[0] == ref0)
pepChar = aa3.charAt(0);
else if (offset[1] == ref0)
pepChar = aa3.charAt(1);
else if (offset[2] == ref0)
pepChar = aa3.charAt(2);
else
pepChar = '?';
} else {
pepChar = '?';
}
}
}
out.print(pepChar);
++ref0;
++x;
}
}
while (x < pixelWidth) {
out.print(" ");
++x;
}
out.println();
}
}
out.println();
}
/**
* variant section
*/
if (!this.vcfReaders.isEmpty() && !out.checkError()) {
final Function<GenotypeType, Character> gTypeToSymbol = new Function<GenotypeType, Character>() {
@Override
public Character apply(final GenotypeType gt) {
switch(gt) {
case NO_CALL:
return '?';
case HOM_REF:
return '0';
case HET:
return '1';
case HOM_VAR:
return '2';
case MIXED:
return 'm';
case UNAVAILABLE:
return 'u';
default:
return '.';
}
}
};
out.println();
for (final VcfSource r : this.vcfReaders) {
if (out.checkError())
break;
final VCFHeader header = r.vcfFileReader.getFileHeader();
final CloseableIterator<VariantContext> iter = r.vcfFileReader.query(this.interval.getContig(), interval.getStart(), interval.getEnd());
final List<VariantContext> variants = new ArrayList<>();
while (iter.hasNext()) {
variants.add(iter.next());
}
iter.close();
if (variants.isEmpty())
continue;
out.println(r.vcfFile.getPath());
if (header.hasGenotypingData()) {
for (final String sample : header.getSampleNamesInOrder()) {
if (!variants.stream().map(V -> V.getGenotype(sample)).filter(G -> !hideNoCall || (hideNoCall && !G.isNoCall())).filter(G -> !hideHomRef || (hideHomRef && !G.isHomRef())).findAny().isPresent()) {
continue;
}
out.print(margin(sample));
ref = this.interval.getStart();
x = 0;
while (x < pixelWidth) {
if (insertIsPresentAtX.test(x)) {
out.print("*");
++x;
} else {
char refBase = ' ';
for (final VariantContext ctx : variants) {
if (ctx.getStart() == ref) {
final Genotype g = ctx.getGenotype(sample);
if (g.isNoCall() && this.hideNoCall)
continue;
if (g.isHomRef() && this.hideHomRef)
continue;
refBase = gTypeToSymbol.apply(g.getType());
break;
}
}
out.print(refBase);
++ref;
++x;
}
}
out.println();
}
} else // no genotype
{
for (final VariantContext ctx : variants) {
out.print(margin(String.valueOf(ctx.getStart()) + ":" + ctx.getReference().getDisplayString() + "/" + ctx.getAlternateAlleles().stream().map(A -> A.getDisplayString()).collect(Collectors.joining(","))));
ref = this.interval.getStart();
x = 0;
while (x < pixelWidth) {
if (insertIsPresentAtX.test(x)) {
out.print("*");
++x;
} else {
out.print(ctx.getStart() == ref ? '+' : ' ');
++ref;
++x;
}
}
out.println();
}
}
}
}
}
Also used :
Genotype(htsjdk.variant.variantcontext.Genotype)
CloseableIterator(htsjdk.samtools.util.CloseableIterator)
Parameter(com.beust.jcommander.Parameter)
VCFFileReader(htsjdk.variant.vcf.VCFFileReader)
VCFHeader(htsjdk.variant.vcf.VCFHeader)
CigarElement(htsjdk.samtools.CigarElement)
CigarOperator(htsjdk.samtools.CigarOperator)
SAMRecordPartition(com.github.lindenb.jvarkit.util.samtools.SAMRecordPartition)
GenomicSequence(com.github.lindenb.jvarkit.util.picard.GenomicSequence)
Function(java.util.function.Function)
ValidationStringency(htsjdk.samtools.ValidationStringency)
ArrayList(java.util.ArrayList)
HashSet(java.util.HashSet)
Interval(htsjdk.samtools.util.Interval)
Map(java.util.Map)
XMLStreamException(javax.xml.stream.XMLStreamException)
IOUtils(com.github.lindenb.jvarkit.io.IOUtils)
Launcher(com.github.lindenb.jvarkit.util.jcommander.Launcher)
XMLStreamWriter(javax.xml.stream.XMLStreamWriter)
SAMSequenceDictionaryExtractor(htsjdk.variant.utils.SAMSequenceDictionaryExtractor)
GeneticCode(com.github.lindenb.jvarkit.util.bio.GeneticCode)
CloserUtil(htsjdk.samtools.util.CloserUtil)
PrintStream(java.io.PrintStream)
Counter(com.github.lindenb.jvarkit.util.Counter)
Predicate(java.util.function.Predicate)
SAMRecordIterator(htsjdk.samtools.SAMRecordIterator)
Logger(com.github.lindenb.jvarkit.util.log.Logger)
GenotypeType(htsjdk.variant.variantcontext.GenotypeType)
SAMSequenceDictionary(htsjdk.samtools.SAMSequenceDictionary)
Set(java.util.Set)
SamFilterParser(com.github.lindenb.jvarkit.util.bio.samfilter.SamFilterParser)
IOException(java.io.IOException)
SamReader(htsjdk.samtools.SamReader)
Collectors(java.util.stream.Collectors)
KnownGene(com.github.lindenb.jvarkit.util.ucsc.KnownGene)
TabixKnownGeneFileReader(com.github.lindenb.jvarkit.util.ucsc.TabixKnownGeneFileReader)
File(java.io.File)
SAMRecord(htsjdk.samtools.SAMRecord)
XMLOutputFactory(javax.xml.stream.XMLOutputFactory)
SamRecordFilter(htsjdk.samtools.filter.SamRecordFilter)
List(java.util.List)
SamInputResource(htsjdk.samtools.SamInputResource)
TreeMap(java.util.TreeMap)
IndexedFastaSequenceFile(htsjdk.samtools.reference.IndexedFastaSequenceFile)
Closeable(java.io.Closeable)
VariantContext(htsjdk.variant.variantcontext.VariantContext)
Collections(java.util.Collections)
SamReaderFactory(htsjdk.samtools.SamReaderFactory)
SAMRecordIterator(htsjdk.samtools.SAMRecordIterator)
ArrayList(java.util.ArrayList)
VariantContext(htsjdk.variant.variantcontext.VariantContext)
SAMSequenceDictionary(htsjdk.samtools.SAMSequenceDictionary)
Predicate(java.util.function.Predicate)
SamReader(htsjdk.samtools.SamReader)
Function(java.util.function.Function)
ArrayList(java.util.ArrayList)
List(java.util.List)
VCFHeader(htsjdk.variant.vcf.VCFHeader)
GenomicSequence(com.github.lindenb.jvarkit.util.picard.GenomicSequence)
Genotype(htsjdk.variant.variantcontext.Genotype)
CigarOperator(htsjdk.samtools.CigarOperator)
TreeMap(java.util.TreeMap)
CigarElement(htsjdk.samtools.CigarElement)
SAMRecord(htsjdk.samtools.SAMRecord)
GenotypeType(htsjdk.variant.variantcontext.GenotypeType)
KnownGene(com.github.lindenb.jvarkit.util.ucsc.KnownGene)
Example 42 with CigarOperator
use of htsjdk.samtools.CigarOperator in project jvarkit by lindenb.
the class VCFCombineTwoSnvs method doVcfToVcf.
@Override
protected int doVcfToVcf(final String inputName, File saveAs) {
BufferedReader bufferedReader = null;
htsjdk.variant.variantcontext.writer.VariantContextWriter w = null;
SortingCollection<CombinedMutation> mutations = null;
CloseableIterator<Variant> varIter = null;
CloseableIterator<CombinedMutation> mutIter = null;
Map<String, SamReader> sample2samReader = new HashMap<>();
try {
bufferedReader = inputName == null ? IOUtils.openStreamForBufferedReader(stdin()) : IOUtils.openURIForBufferedReading(inputName);
final VCFUtils.CodecAndHeader cah = VCFUtils.parseHeader(bufferedReader);
/* get VCF header */
final VCFHeader header = cah.header;
final Set<String> sampleNamesInOrder = new HashSet<>(header.getSampleNamesInOrder());
LOG.info("opening REF:" + referenceFile);
this.indexedFastaSequenceFile = new IndexedFastaSequenceFile(this.referenceFile);
final SAMSequenceDictionary dict = this.indexedFastaSequenceFile.getSequenceDictionary();
if (dict == null)
throw new IOException("dictionary missing");
if (this.bamIn != null) {
/**
* unroll and open bam file
*/
for (final File bamFile : IOUtils.unrollFileCollection(Collections.singletonList(this.bamIn))) {
LOG.info("opening BAM :" + this.bamIn);
final SamReader samReader = SamReaderFactory.makeDefault().referenceSequence(this.referenceFile).validationStringency(ValidationStringency.LENIENT).open(this.bamIn);
if (!samReader.hasIndex()) {
throw new IOException("Sam file is NOT indexed: " + bamFile);
}
final SAMFileHeader samHeader = samReader.getFileHeader();
if (samHeader.getSequenceDictionary() == null || !SequenceUtil.areSequenceDictionariesEqual(dict, samReader.getFileHeader().getSequenceDictionary())) {
throw new IOException(bamFile + " and REF don't have the same Sequence Dictionary.");
}
/* get sample name */
String sampleName = null;
for (final SAMReadGroupRecord rg : samHeader.getReadGroups()) {
if (rg.getSample() == null)
continue;
if (sampleName != null && !sampleName.equals(rg.getSample())) {
samReader.close();
throw new IOException(bamFile + " Contains two samples " + sampleName + " " + rg.getSample());
}
sampleName = rg.getSample();
}
if (sampleName == null) {
samReader.close();
LOG.warn("no sample in " + bamFile);
continue;
}
if (!sampleNamesInOrder.contains(sampleName)) {
samReader.close();
LOG.warn("no sample " + sampleName + " in vcf");
continue;
}
sample2samReader.put(sampleName, samReader);
}
}
loadKnownGenesFromUri();
this.variants = SortingCollection.newInstance(Variant.class, new VariantCodec(), new VariantComparator(dict), this.writingSortingCollection.getMaxRecordsInRam(), this.writingSortingCollection.getTmpPaths());
this.variants.setDestructiveIteration(true);
SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(header);
String vcfLine = null;
while ((vcfLine = bufferedReader.readLine()) != null) {
final VariantContext ctx = progress.watch(cah.codec.decode(vcfLine));
/* discard non SNV variant */
if (!ctx.isVariant() || ctx.isIndel()) {
continue;
}
/* find the overlapping genes : extend the interval of the variant to include the stop codon */
final Collection<KnownGene> genes = new ArrayList<>();
for (List<KnownGene> lkg : this.knownGenes.getOverlapping(new Interval(ctx.getContig(), Math.max(1, ctx.getStart() - 3), ctx.getEnd() + 3))) {
genes.addAll(lkg);
}
final List<Allele> alternateAlleles = ctx.getAlternateAlleles();
/* loop over overlapping genes */
for (final KnownGene kg : genes) {
/* loop over available alleles */
for (int allele_idx = 0; allele_idx < alternateAlleles.size(); ++allele_idx) {
final Allele alt = alternateAlleles.get(allele_idx);
challenge(ctx, alt, kg, vcfLine);
}
}
}
progress.finish();
this.variants.doneAdding();
mutations = SortingCollection.newInstance(CombinedMutation.class, new MutationCodec(), new MutationComparator(dict), this.writingSortingCollection.getMaxRecordsInRam(), this.writingSortingCollection.getTmpPaths());
mutations.setDestructiveIteration(true);
final VCFFilterHeaderLine vcfFilterHeaderLine = new VCFFilterHeaderLine("TwoHaplotypes", "(number of reads carrying both mutation) < (reads carrying variant 1 + reads carrying variant 2) ");
varIter = this.variants.iterator();
progress = new SAMSequenceDictionaryProgress(header);
final ArrayList<Variant> buffer = new ArrayList<>();
for (; ; ) {
Variant variant = null;
if (varIter.hasNext()) {
variant = varIter.next();
progress.watch(variant.contig, variant.genomicPosition1);
}
if (variant == null || !(!buffer.isEmpty() && buffer.get(0).contig.equals(variant.contig) && buffer.get(0).transcriptName.equals(variant.transcriptName))) {
if (!buffer.isEmpty()) {
for (int i = 0; i < buffer.size(); ++i) {
final Variant v1 = buffer.get(i);
for (int j = i + 1; j < buffer.size(); ++j) {
final Variant v2 = buffer.get(j);
if (v1.codonStart() != v2.codonStart())
continue;
if (v1.positionInCodon() == v2.positionInCodon())
continue;
if (!v1.wildCodon.equals(v2.wildCodon)) {
throw new IllegalStateException();
}
final StringBuilder combinedCodon = new StringBuilder(v1.wildCodon);
combinedCodon.setCharAt(v1.positionInCodon(), v1.mutCodon.charAt(v1.positionInCodon()));
combinedCodon.setCharAt(v2.positionInCodon(), v2.mutCodon.charAt(v2.positionInCodon()));
final String pwild = new ProteinCharSequence(v1.wildCodon).getString();
final String p1 = new ProteinCharSequence(v1.mutCodon).getString();
final String p2 = new ProteinCharSequence(v2.mutCodon).getString();
final String pCombined = new ProteinCharSequence(combinedCodon).getString();
final String combinedSO;
final String combinedType;
/* both AA are synonymous, while combined is not */
if (!pCombined.equals(pwild) && p1.equals(pwild) && p2.equals(pwild)) {
combinedType = "combined_is_nonsynonymous";
if (pCombined.equals("*")) {
/* http://www.sequenceontology.org/browser/current_svn/term/SO:0001587 */
combinedSO = "stop_gained";
} else if (pwild.equals("*")) {
/* http://www.sequenceontology.org/browser/current_svn/term/SO:0002012 */
combinedSO = "stop_lost";
} else {
/* http://www.sequenceontology.org/miso/current_svn/term/SO:0001992 */
combinedSO = "nonsynonymous_variant";
}
} else if (!pCombined.equals(p1) && !pCombined.equals(p2) && !pCombined.equals(pwild)) {
combinedType = "combined_is_new";
if (pCombined.equals("*")) {
/* http://www.sequenceontology.org/browser/current_svn/term/SO:0001587 */
combinedSO = "stop_gained";
} else {
/* http://www.sequenceontology.org/miso/current_svn/term/SO:0001992 */
combinedSO = "nonsynonymous_variant";
}
} else {
combinedType = null;
combinedSO = null;
}
/**
* ok, there is something interesting here ,
* create two new Mutations carrying the
* two variants
*/
if (combinedSO != null) {
/**
* grantham score is max found combined vs (p1/p2/wild)
*/
int grantham_score = GranthamScore.score(pCombined.charAt(0), pwild.charAt(0));
grantham_score = Math.max(grantham_score, GranthamScore.score(pCombined.charAt(0), p1.charAt(0)));
grantham_score = Math.max(grantham_score, GranthamScore.score(pCombined.charAt(0), p2.charAt(0)));
/**
* info that will be displayed in the vcf
*/
final Map<String, Object> info1 = v1.getInfo(v2);
final Map<String, Object> info2 = v2.getInfo(v1);
// filter for this combined: default it fails the filter
String filter = vcfFilterHeaderLine.getID();
final Map<String, Object> combinedMap = new LinkedHashMap<>();
combinedMap.put("CombinedCodon", combinedCodon);
combinedMap.put("CombinedAA", pCombined);
combinedMap.put("CombinedSO", combinedSO);
combinedMap.put("CombinedType", combinedType);
combinedMap.put("GranthamScore", grantham_score);
info1.putAll(combinedMap);
info2.putAll(combinedMap);
final Map<String, CoverageInfo> sample2coverageInfo = new HashMap<>(sample2samReader.size());
final int chromStart = Math.min(v1.genomicPosition1, v2.genomicPosition1);
final int chromEnd = Math.max(v1.genomicPosition1, v2.genomicPosition1);
/* get phasing info for each sample*/
for (final String sampleName : sample2samReader.keySet()) {
final SamReader samReader = sample2samReader.get(sampleName);
final CoverageInfo covInfo = new CoverageInfo();
sample2coverageInfo.put(sampleName, covInfo);
SAMRecordIterator iter = null;
try {
iter = samReader.query(v1.contig, chromStart, chromEnd, false);
while (iter.hasNext()) {
final SAMRecord rec = iter.next();
if (rec.getReadUnmappedFlag())
continue;
if (rec.isSecondaryOrSupplementary())
continue;
if (rec.getDuplicateReadFlag())
continue;
if (rec.getReadFailsVendorQualityCheckFlag())
continue;
// get DEPTh for variant 1
if (rec.getAlignmentStart() <= v1.genomicPosition1 && v1.genomicPosition1 <= rec.getAlignmentEnd()) {
covInfo.depth1++;
}
// get DEPTh for variant 2
if (rec.getAlignmentStart() <= v2.genomicPosition1 && v2.genomicPosition1 <= rec.getAlignmentEnd()) {
covInfo.depth2++;
}
if (rec.getAlignmentEnd() < chromEnd)
continue;
if (rec.getAlignmentStart() > chromStart)
continue;
final Cigar cigar = rec.getCigar();
if (cigar == null)
continue;
final byte[] bases = rec.getReadBases();
if (bases == null)
continue;
int refpos1 = rec.getAlignmentStart();
int readpos = 0;
boolean found_variant1_on_this_read = false;
boolean found_variant2_on_this_read = false;
/**
* loop over cigar
*/
for (final CigarElement ce : cigar.getCigarElements()) {
final CigarOperator op = ce.getOperator();
switch(op) {
case P:
continue;
case S:
case I:
readpos += ce.getLength();
break;
case D:
case N:
refpos1 += ce.getLength();
break;
case H:
continue;
case EQ:
case M:
case X:
for (int x = 0; x < ce.getLength(); ++x) {
if (refpos1 == v1.genomicPosition1 && same(bases[readpos], v1.altAllele)) {
found_variant1_on_this_read = true;
} else if (refpos1 == v2.genomicPosition1 && same(bases[readpos], v2.altAllele)) {
found_variant2_on_this_read = true;
}
refpos1++;
readpos++;
}
break;
default:
throw new IllegalStateException(op.name());
}
/* skip remaining bases after last variant */
if (refpos1 > chromEnd)
break;
}
/* sum-up what we found */
if (found_variant1_on_this_read && found_variant2_on_this_read) {
covInfo.count_reads_having_both_variants++;
} else if (!found_variant1_on_this_read && !found_variant2_on_this_read) {
covInfo.count_reads_having_no_variants++;
} else if (found_variant1_on_this_read) {
covInfo.count_reads_having_variant1++;
} else if (found_variant2_on_this_read) {
covInfo.count_reads_having_variant2++;
}
}
/* end of loop over reads */
} finally {
iter.close();
iter = null;
}
info1.put("N_READS_BOTH_VARIANTS_" + sampleName, covInfo.count_reads_having_both_variants);
info2.put("N_READS_BOTH_VARIANTS_" + sampleName, covInfo.count_reads_having_both_variants);
info1.put("N_READS_NO_VARIANTS_" + sampleName, covInfo.count_reads_having_no_variants);
info2.put("N_READS_NO_VARIANTS_" + sampleName, covInfo.count_reads_having_no_variants);
info1.put("N_READS_TOTAL_" + sampleName, covInfo.count_reads_having_both_variants + covInfo.count_reads_having_no_variants + covInfo.count_reads_having_variant1 + covInfo.count_reads_having_variant2);
info2.put("N_READS_TOTAL_" + sampleName, covInfo.count_reads_having_both_variants + covInfo.count_reads_having_no_variants + covInfo.count_reads_having_variant1 + covInfo.count_reads_having_variant2);
// count for variant 1
info1.put("N_READS_ONLY_1_" + sampleName, covInfo.count_reads_having_variant1);
info1.put("N_READS_ONLY_2_" + sampleName, covInfo.count_reads_having_variant2);
info1.put("DEPTH_1_" + sampleName, covInfo.depth1);
// inverse previous count
info2.put("N_READS_ONLY_1_" + sampleName, covInfo.count_reads_having_variant2);
info2.put("N_READS_ONLY_2_" + sampleName, covInfo.count_reads_having_variant1);
info2.put("DEPTH_2_" + sampleName, covInfo.depth2);
/* number of reads with both variant is greater than
* reads carrying only one variant: reset the filter
*/
if (2 * covInfo.count_reads_having_both_variants > (covInfo.count_reads_having_variant1 + covInfo.count_reads_having_variant2)) {
/* reset filter */
filter = VCFConstants.UNFILTERED;
info1.put("FILTER_1_" + sampleName, ".");
info2.put("FILTER_2_" + sampleName, ".");
} else {
info1.put("FILTER_1_" + sampleName, vcfFilterHeaderLine.getID());
info2.put("FILTER_2_" + sampleName, vcfFilterHeaderLine.getID());
}
}
/* end of loop over bams */
final CombinedMutation m1 = new CombinedMutation();
m1.contig = v1.contig;
m1.genomicPosition1 = v1.genomicPosition1;
m1.id = v1.id;
m1.refAllele = v1.refAllele;
m1.altAllele = v1.altAllele;
m1.vcfLine = v1.vcfLine;
m1.info = mapToString(info1);
m1.filter = filter;
m1.grantham_score = grantham_score;
m1.sorting_id = ID_GENERATOR++;
mutations.add(m1);
final CombinedMutation m2 = new CombinedMutation();
m2.contig = v2.contig;
m2.genomicPosition1 = v2.genomicPosition1;
m2.id = v2.id;
m2.refAllele = v2.refAllele;
m2.altAllele = v2.altAllele;
m2.vcfLine = v2.vcfLine;
m2.info = mapToString(info2);
m2.filter = filter;
m2.grantham_score = grantham_score;
m2.sorting_id = ID_GENERATOR++;
mutations.add(m2);
}
}
}
}
buffer.clear();
if (variant == null)
break;
}
buffer.add(variant);
}
progress.finish();
mutations.doneAdding();
varIter.close();
varIter = null;
variants.cleanup();
variants = null;
final ArrayList<CombinedMutation> mBuffer = new ArrayList<>();
final VCFHeader header2 = new VCFHeader(header);
header2.addMetaDataLine(new VCFHeaderLine(getProgramName() + "AboutQUAL", "QUAL is filled with Grantham Score http://www.ncbi.nlm.nih.gov/pubmed/4843792"));
final StringBuilder infoDesc = new StringBuilder("Variant affected by two distinct mutation. Format is defined in the INFO column. ");
final VCFInfoHeaderLine infoHeaderLine = new VCFInfoHeaderLine("CodonVariant", VCFHeaderLineCount.UNBOUNDED, VCFHeaderLineType.String, infoDesc.toString());
super.addMetaData(header2);
header2.addMetaDataLine(infoHeaderLine);
if (!sample2samReader.isEmpty()) {
header2.addMetaDataLine(vcfFilterHeaderLine);
}
w = super.openVariantContextWriter(saveAs);
w.writeHeader(header2);
progress = new SAMSequenceDictionaryProgress(header);
mutIter = mutations.iterator();
for (; ; ) {
CombinedMutation mutation = null;
if (mutIter.hasNext()) {
mutation = mutIter.next();
progress.watch(mutation.contig, mutation.genomicPosition1);
}
if (mutation == null || !(!mBuffer.isEmpty() && mBuffer.get(0).contig.equals(mutation.contig) && mBuffer.get(0).genomicPosition1 == mutation.genomicPosition1 && mBuffer.get(0).refAllele.equals(mutation.refAllele))) {
if (!mBuffer.isEmpty()) {
// default grantham score used in QUAL
int grantham_score = -1;
// default filter fails
String filter = vcfFilterHeaderLine.getID();
final CombinedMutation first = mBuffer.get(0);
final Set<String> info = new HashSet<>();
final VariantContext ctx = cah.codec.decode(first.vcfLine);
final VariantContextBuilder vcb = new VariantContextBuilder(ctx);
vcb.chr(first.contig);
vcb.start(first.genomicPosition1);
vcb.stop(first.genomicPosition1 + first.refAllele.length() - 1);
if (!first.id.equals(VCFConstants.EMPTY_ID_FIELD))
vcb.id(first.id);
for (final CombinedMutation m : mBuffer) {
info.add(m.info);
grantham_score = Math.max(grantham_score, m.grantham_score);
if (VCFConstants.UNFILTERED.equals(m.filter)) {
// at least one SNP is ok one this line
filter = null;
}
}
vcb.unfiltered();
if (filter != null && !sample2samReader.isEmpty()) {
vcb.filter(filter);
} else {
vcb.passFilters();
}
vcb.attribute(infoHeaderLine.getID(), new ArrayList<String>(info));
if (grantham_score > 0) {
vcb.log10PError(grantham_score / -10.0);
} else {
vcb.log10PError(VariantContext.NO_LOG10_PERROR);
}
w.add(vcb.make());
}
mBuffer.clear();
if (mutation == null)
break;
}
mBuffer.add(mutation);
}
progress.finish();
mutIter.close();
mutations.cleanup();
mutations = null;
return RETURN_OK;
} catch (Exception err) {
LOG.error(err);
return -1;
} finally {
CloserUtil.close(this.indexedFastaSequenceFile);
CloserUtil.close(mutIter);
CloserUtil.close(varIter);
if (this.variants != null)
this.variants.cleanup();
if (mutations != null)
mutations.cleanup();
this.variants = null;
for (SamReader r : sample2samReader.values()) CloserUtil.close(r);
CloserUtil.close(w);
CloserUtil.close(bufferedReader);
}
}
Also used :
VCFHeaderLine(htsjdk.variant.vcf.VCFHeaderLine)
SAMRecordIterator(htsjdk.samtools.SAMRecordIterator)
HashMap(java.util.HashMap)
LinkedHashMap(java.util.LinkedHashMap)
ArrayList(java.util.ArrayList)
IndexedFastaSequenceFile(htsjdk.samtools.reference.IndexedFastaSequenceFile)
LinkedHashMap(java.util.LinkedHashMap)
HashSet(java.util.HashSet)
CigarOperator(htsjdk.samtools.CigarOperator)
CigarElement(htsjdk.samtools.CigarElement)
VariantContextWriter(htsjdk.variant.variantcontext.writer.VariantContextWriter)
SAMRecord(htsjdk.samtools.SAMRecord)
KnownGene(com.github.lindenb.jvarkit.util.ucsc.KnownGene)
SAMFileHeader(htsjdk.samtools.SAMFileHeader)
IndexedFastaSequenceFile(htsjdk.samtools.reference.IndexedFastaSequenceFile)
File(java.io.File)
Interval(htsjdk.samtools.util.Interval)
VCFUtils(com.github.lindenb.jvarkit.util.vcf.VCFUtils)
SAMReadGroupRecord(htsjdk.samtools.SAMReadGroupRecord)
VariantContext(htsjdk.variant.variantcontext.VariantContext)
SAMSequenceDictionary(htsjdk.samtools.SAMSequenceDictionary)
SamReader(htsjdk.samtools.SamReader)
VCFFilterHeaderLine(htsjdk.variant.vcf.VCFFilterHeaderLine)
VCFHeader(htsjdk.variant.vcf.VCFHeader)
SAMSequenceDictionaryProgress(com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress)
IOException(java.io.IOException)
VCFInfoHeaderLine(htsjdk.variant.vcf.VCFInfoHeaderLine)
IOException(java.io.IOException)
Allele(htsjdk.variant.variantcontext.Allele)
Cigar(htsjdk.samtools.Cigar)
VariantContextBuilder(htsjdk.variant.variantcontext.VariantContextBuilder)
BufferedReader(java.io.BufferedReader)
Example 43 with CigarOperator
use of htsjdk.samtools.CigarOperator in project jvarkit by lindenb.
the class CigarOpPerPositionChartFactory method build.
@Override
public StackedBarChart<String, Number> build() {
final CategoryAxis xAxis = new CategoryAxis();
xAxis.setLabel("Position in Read");
final NumberAxis yAxis = new NumberAxis();
yAxis.setLabel("Count");
final List<XYChart.Series<String, Number>> base2count = new ArrayList<>();
for (final CigarOperator cigarop : CigarOperator.values()) {
if (cigarop == CigarOperator.P)
continue;
final XYChart.Series<String, Number> serie = new XYChart.Series<String, Number>();
serie.setName(cigarop.name());
base2count.add(serie);
for (int i = 0; i < this.cigar2pos2count.size(); ++i) {
serie.getData().add(new XYChart.Data<String, Number>(String.valueOf(i + 1), this.cigar2pos2count.get(i).count(cigarop)));
}
}
final StackedBarChart<String, Number> sbc = new StackedBarChart<String, Number>(xAxis, yAxis);
sbc.setTitle(getName());
sbc.getData().addAll(base2count);
sbc.setCategoryGap(0.2);
return sbc;
}
Also used :
NumberAxis(javafx.scene.chart.NumberAxis)
ArrayList(java.util.ArrayList)
StackedBarChart(javafx.scene.chart.StackedBarChart)
CigarOperator(htsjdk.samtools.CigarOperator)
CategoryAxis(javafx.scene.chart.CategoryAxis)
XYChart(javafx.scene.chart.XYChart)
Example 44 with CigarOperator
use of htsjdk.samtools.CigarOperator in project jvarkit by lindenb.
the class Biostar170742 method doWork.
@Override
public int doWork(final List<String> args) {
if (this.faidx == null) {
LOG.error("Reference sequence was not defined");
return -1;
}
PrintStream out = null;
SamReader sfr = null;
SAMRecordIterator iter = null;
GenomicSequence genomicSequence = null;
IndexedFastaSequenceFile indexedFastaSequenceFile = null;
try {
indexedFastaSequenceFile = new IndexedFastaSequenceFile(this.faidx);
long align_id = 0;
sfr = openSamReader(oneFileOrNull(args));
out = super.openFileOrStdoutAsPrintStream(this.outputFile);
final StringBuilder refseq = new StringBuilder();
final StringBuilder readseq = new StringBuilder();
final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(sfr.getFileHeader());
iter = sfr.iterator();
while (iter.hasNext()) {
final SAMRecord rec = progress.watch(iter.next());
if (rec.getReadUnmappedFlag())
continue;
final Cigar cigar = rec.getCigar();
if (cigar == null)
continue;
final byte[] readbases = rec.getReadBases();
if (readbases == null)
continue;
if (genomicSequence == null || !rec.getReferenceName().equals(genomicSequence.getChrom())) {
genomicSequence = new GenomicSequence(indexedFastaSequenceFile, rec.getReferenceName());
}
int refpos1 = rec.getAlignmentStart();
int readpos = 0;
refseq.setLength(0);
readseq.setLength(0);
for (final CigarElement ce : cigar.getCigarElements()) {
final CigarOperator op = ce.getOperator();
if (op.equals(CigarOperator.S)) {
readpos += ce.getLength();
continue;
}
if (op.equals(CigarOperator.H)) {
continue;
}
for (int i = 0; i < ce.getLength(); ++i) {
if (op.consumesReferenceBases() && op.consumesReadBases()) {
refseq.append(genomicSequence.charAt(refpos1 - 1));
readseq.append((char) readbases[readpos]);
readpos++;
refpos1++;
} else if (op.consumesReferenceBases()) {
refseq.append(genomicSequence.charAt(refpos1 - 1));
readseq.append('-');
refpos1++;
} else if (op.consumesReadBases()) {
refseq.append('-');
readseq.append((char) readbases[readpos]);
readpos++;
}
}
}
out.print(align_id);
out.print(' ');
out.print(rec.getReferenceName());
out.print(' ');
out.print(rec.getAlignmentStart());
out.print(' ');
out.print(rec.getAlignmentEnd());
out.print(' ');
out.print(rec.getReadName());
if (rec.getReadPairedFlag()) {
if (rec.getFirstOfPairFlag()) {
out.print("/1");
} else if (rec.getSecondOfPairFlag()) {
out.print("/2");
}
}
out.print(' ');
out.print(1 + rec.getAlignmentStart() - rec.getUnclippedStart());
out.print(' ');
out.print(rec.getReadLength() - (rec.getUnclippedEnd() - rec.getAlignmentEnd()));
out.print(' ');
out.print(rec.getReadNegativeStrandFlag() ? "-" : "+");
out.print(' ');
out.print(rec.getMappingQuality());
out.println();
out.println(refseq);
out.println(readseq);
out.println();
++align_id;
}
progress.finish();
iter.close();
out.flush();
LOG.info("done");
return RETURN_OK;
} catch (Exception err) {
LOG.error(err);
return -1;
} finally {
CloserUtil.close(out);
CloserUtil.close(iter);
CloserUtil.close(sfr);
CloserUtil.close(indexedFastaSequenceFile);
}
}
Also used :
PrintStream(java.io.PrintStream)
SAMRecordIterator(htsjdk.samtools.SAMRecordIterator)
SAMSequenceDictionaryProgress(com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress)
GenomicSequence(com.github.lindenb.jvarkit.util.picard.GenomicSequence)
CigarOperator(htsjdk.samtools.CigarOperator)
CigarElement(htsjdk.samtools.CigarElement)
IndexedFastaSequenceFile(htsjdk.samtools.reference.IndexedFastaSequenceFile)
SamReader(htsjdk.samtools.SamReader)
Cigar(htsjdk.samtools.Cigar)
SAMRecord(htsjdk.samtools.SAMRecord)
Example 45 with CigarOperator
use of htsjdk.samtools.CigarOperator in project jvarkit by lindenb.
the class Biostar214299 method doWork.
@Override
public int doWork(final List<String> args) {
if (this.positionFile == null) {
LOG.error("position File is not defined.");
return -1;
}
final String UNAFFECTED_SAMPLE = "UNAFFECTED";
final String AMBIGOUS_SAMPLE = "AMBIGOUS";
final String UNMAPPED = "UNMAPPED";
SamReader sfr = null;
SAMFileWriter sfw = null;
final IntervalTreeMap<Position> positionsTreeMap = new IntervalTreeMap<>();
final Set<String> samples = new HashSet<>();
try {
sfr = openSamReader(oneFileOrNull(args));
final SAMFileHeader header = sfr.getFileHeader();
final SAMSequenceDictionary dict = header.getSequenceDictionary();
if (dict == null) {
LOG.error("Dictionary missing in input sam");
return -1;
}
try (BufferedReader br = IOUtils.openFileForBufferedReading(this.positionFile)) {
String line;
while ((line = br.readLine()) != null) {
if (line.trim().isEmpty() || line.startsWith("#"))
continue;
final String[] tokens = line.split("[\t]");
if (tokens.length < 4) {
LOG.error("Not enough columns in " + line);
return -1;
}
final String contig = tokens[0];
if (dict.getSequence(contig) == null) {
LOG.error("No such contig in input's sam dictionary: " + contig);
return -1;
}
final int refpos = Integer.parseInt(tokens[1]);
final Interval interval = new Interval(contig, refpos, refpos);
Position position = positionsTreeMap.get(interval);
if (position == null) {
position = new Position();
// position.contig = contig;
position.refpos = refpos;
positionsTreeMap.put(interval, position);
}
final String bases = tokens[2].toUpperCase();
if (bases.length() != 1 || !bases.matches("[ATGC]")) {
LOG.error("in " + line + " bases should be one letter an ATGC");
return -1;
}
if (position.base2sample.containsKey(bases.charAt(0))) {
LOG.error("in " + line + " bases already defined for this position");
return -1;
}
final String sampleName = tokens[3].trim();
if (sampleName.isEmpty()) {
LOG.error("sample name cannot be empty");
return -1;
}
samples.add(sampleName);
position.base2sample.put(bases.charAt(0), sampleName);
}
} catch (final IOException err) {
LOG.error(err);
return -1;
}
if (samples.contains(UNAFFECTED_SAMPLE)) {
LOG.error("Sample cannot be named " + UNAFFECTED_SAMPLE);
return -1;
}
if (samples.contains(AMBIGOUS_SAMPLE)) {
LOG.error("Sample cannot be named " + AMBIGOUS_SAMPLE);
return -1;
}
if (samples.contains(UNMAPPED)) {
LOG.error("Sample cannot be named " + UNMAPPED);
return -1;
}
samples.add(UNAFFECTED_SAMPLE);
samples.add(AMBIGOUS_SAMPLE);
samples.add(UNMAPPED);
final SAMFileHeader newHeader = new SAMFileHeader();
newHeader.setSortOrder(header.getSortOrder());
newHeader.setSequenceDictionary(dict);
newHeader.addComment("generated with " + getProgramName() + " " + getVersion() + " Pierre Lindenbaum : " + getProgramCommandLine());
/* create groups */
for (final String sample : samples) {
final SAMReadGroupRecord rg = new SAMReadGroupRecord(sample);
rg.setSample(sample);
rg.setLibrary(sample);
newHeader.addReadGroup(rg);
}
sfw = this.writingBamArgs.openSAMFileWriter(this.outputFile, newHeader, true);
final SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(header).logger(LOG);
final SAMRecordIterator iter = sfr.iterator();
while (iter.hasNext()) {
final SAMRecord rec = progress.watch(iter.next());
rec.setAttribute("RG", null);
if (rec.getReadUnmappedFlag()) {
rec.setAttribute("RG", UNMAPPED);
sfw.addAlignment(rec);
continue;
}
final Cigar cigar = rec.getCigar();
final Collection<Position> snps = positionsTreeMap.getContained(new Interval(rec.getContig(), rec.getUnclippedStart(), rec.getUnclippedEnd()));
if (snps == null || snps.isEmpty()) {
rec.setAttribute("RG", UNAFFECTED_SAMPLE);
sfw.addAlignment(rec);
continue;
}
final Map<Integer, Position> index2pos = snps.stream().collect(Collectors.toMap(P -> P.refpos, P -> P));
final Set<String> selectedSamples = new HashSet<>();
final byte[] bases = rec.getReadBases();
if (bases == null || bases.equals(SAMRecord.NULL_SEQUENCE)) {
LOG.error("Bases missing in read " + rec);
return -1;
}
int refPos1 = rec.getUnclippedStart();
int readPos0 = 0;
for (final CigarElement ce : cigar.getCigarElements()) {
final CigarOperator op = ce.getOperator();
final boolean consummeReadBaseOrSoftClip = op.consumesReadBases() || op.equals(CigarOperator.S);
if (op.consumesReferenceBases() && consummeReadBaseOrSoftClip) {
for (int i = 0; i < ce.getLength(); ++i) {
final int nowRefPos1 = (refPos1 + i);
final int nowReadPos0 = (readPos0 + i);
final Position position = index2pos.get(nowRefPos1);
if (position == null)
continue;
if (nowReadPos0 >= bases.length)
continue;
final char base = (char) Character.toUpperCase(bases[nowReadPos0]);
final String sample = position.base2sample.get(base);
if (sample == null)
continue;
selectedSamples.add(sample);
index2pos.remove(nowRefPos1);
if (index2pos.isEmpty())
break;
}
}
if (op.consumesReferenceBases())
refPos1 += ce.getLength();
if (consummeReadBaseOrSoftClip || op.equals(CigarOperator.H)) {
readPos0 += ce.getLength();
}
}
if (selectedSamples.isEmpty()) {
rec.setAttribute("RG", UNAFFECTED_SAMPLE);
} else if (selectedSamples.size() == 1) {
rec.setAttribute("RG", selectedSamples.iterator().next());
} else {
rec.setAttribute("RG", AMBIGOUS_SAMPLE);
}
sfw.addAlignment(rec);
}
progress.finish();
LOG.info("done");
return RETURN_OK;
} catch (final Exception err) {
LOG.error(err);
return -1;
} finally {
CloserUtil.close(sfr);
CloserUtil.close(sfw);
}
}
Also used :
Cigar(htsjdk.samtools.Cigar)
Program(com.github.lindenb.jvarkit.util.jcommander.Program)
Parameter(com.beust.jcommander.Parameter)
CigarElement(htsjdk.samtools.CigarElement)
SAMSequenceDictionaryProgress(com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress)
CigarOperator(htsjdk.samtools.CigarOperator)
HashMap(java.util.HashMap)
Term(com.github.lindenb.semontology.Term)
SAMFileHeader(htsjdk.samtools.SAMFileHeader)
ParametersDelegate(com.beust.jcommander.ParametersDelegate)
HashSet(java.util.HashSet)
Interval(htsjdk.samtools.util.Interval)
Map(java.util.Map)
IOUtils(com.github.lindenb.jvarkit.io.IOUtils)
Launcher(com.github.lindenb.jvarkit.util.jcommander.Launcher)
CloserUtil(htsjdk.samtools.util.CloserUtil)
IntervalTreeMap(htsjdk.samtools.util.IntervalTreeMap)
SAMRecordIterator(htsjdk.samtools.SAMRecordIterator)
Collection(java.util.Collection)
Logger(com.github.lindenb.jvarkit.util.log.Logger)
SAMSequenceDictionary(htsjdk.samtools.SAMSequenceDictionary)
Set(java.util.Set)
IOException(java.io.IOException)
SAMFileWriter(htsjdk.samtools.SAMFileWriter)
SamReader(htsjdk.samtools.SamReader)
Collectors(java.util.stream.Collectors)
File(java.io.File)
SAMRecord(htsjdk.samtools.SAMRecord)
List(java.util.List)
SAMReadGroupRecord(htsjdk.samtools.SAMReadGroupRecord)
BufferedReader(java.io.BufferedReader)
SAMRecordIterator(htsjdk.samtools.SAMRecordIterator)
SAMReadGroupRecord(htsjdk.samtools.SAMReadGroupRecord)
SAMSequenceDictionary(htsjdk.samtools.SAMSequenceDictionary)
SamReader(htsjdk.samtools.SamReader)
HashSet(java.util.HashSet)
SAMSequenceDictionaryProgress(com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress)
SAMFileWriter(htsjdk.samtools.SAMFileWriter)
IOException(java.io.IOException)
CigarOperator(htsjdk.samtools.CigarOperator)
CigarElement(htsjdk.samtools.CigarElement)
IOException(java.io.IOException)
Cigar(htsjdk.samtools.Cigar)
SAMRecord(htsjdk.samtools.SAMRecord)
BufferedReader(java.io.BufferedReader)
SAMFileHeader(htsjdk.samtools.SAMFileHeader)
IntervalTreeMap(htsjdk.samtools.util.IntervalTreeMap)
Interval(htsjdk.samtools.util.Interval)
Aggregations
CigarOperator (htsjdk.samtools.CigarOperator)48
CigarElement (htsjdk.samtools.CigarElement)31
Cigar (htsjdk.samtools.Cigar)24
SAMRecord (htsjdk.samtools.SAMRecord)22
ArrayList (java.util.ArrayList)17
SAMRecordIterator (htsjdk.samtools.SAMRecordIterator)15
SamReader (htsjdk.samtools.SamReader)15
SAMSequenceDictionaryProgress (com.github.lindenb.jvarkit.util.picard.SAMSequenceDictionaryProgress)13
SAMFileHeader (htsjdk.samtools.SAMFileHeader)12
File (java.io.File)12
SAMSequenceDictionary (htsjdk.samtools.SAMSequenceDictionary)11
List (java.util.List)11
Parameter (com.beust.jcommander.Parameter)9
Launcher (com.github.lindenb.jvarkit.util.jcommander.Launcher)9
Logger (com.github.lindenb.jvarkit.util.log.Logger)9
SAMReadGroupRecord (htsjdk.samtools.SAMReadGroupRecord)9
CloserUtil (htsjdk.samtools.util.CloserUtil)9
Program (com.github.lindenb.jvarkit.util.jcommander.Program)8
IndexedFastaSequenceFile (htsjdk.samtools.reference.IndexedFastaSequenceFile)8
HashMap (java.util.HashMap)8
