use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.
the class FindAllCoverageAtPosition method doWork.
@Override
public int doWork(final List<String> args) {
final Set<Mutation> mutations = new TreeSet<>();
BufferedReader r = null;
try {
if (this.referenceFileFile != null) {
this.indexedFastaSequenceFile = new IndexedFastaSequenceFile(this.referenceFileFile);
}
mutations.addAll(Arrays.asList(this.positionStr.split("[ ]")).stream().filter(S -> !S.trim().isEmpty()).map(S -> new Mutation(S)).collect(Collectors.toSet()));
for (final String s : positionStr.split("[ ]")) {
if (s.trim().isEmpty())
continue;
mutations.add(new Mutation(s));
}
if (this.positionFile != null) {
String line;
r = IOUtils.openFileForBufferedReading(positionFile);
if (positionFile.getName().endsWith(".bed")) {
final BedLineCodec codec = new BedLineCodec();
while ((line = r.readLine()) != null) {
final BedLine bedLine = codec.decode(line);
if (bedLine == null)
continue;
for (int x = bedLine.getStart(); x <= bedLine.getEnd(); ++x) {
mutations.add(new Mutation(bedLine.getContig(), x));
}
}
} else {
while ((line = r.readLine()) != null) {
if (line.trim().isEmpty() || line.startsWith("#"))
continue;
mutations.add(new Mutation(line));
}
}
r.close();
r = null;
}
if (mutations.isEmpty()) {
LOG.fatal("undefined position \'str\'");
return -1;
}
LOG.info("number of mutations " + mutations.size());
this.out = this.openFileOrStdoutAsPrintWriter(this.outputFile);
out.print("#File");
out.print('\t');
out.print("CHROM");
out.print('\t');
out.print("POS");
if (this.indexedFastaSequenceFile != null) {
out.print('\t');
out.print("REF");
}
out.print('\t');
out.print(this.groupBy.name().toUpperCase());
out.print('\t');
out.print("DEPTH");
for (final CigarOperator op : CigarOperator.values()) {
out.print('\t');
out.print(op.name());
}
for (char c : BASES_To_PRINT) {
out.print('\t');
out.print("Base(" + c + ")");
}
out.println();
if (args.isEmpty()) {
LOG.info("Reading from stdin");
r = new BufferedReader(new InputStreamReader(stdin()));
scan(r, mutations);
r.close();
r = null;
} else {
for (final String filename : args) {
LOG.info("Reading from " + filename);
r = IOUtils.openURIForBufferedReading(filename);
scan(r, mutations);
r.close();
r = null;
}
}
this.out.flush();
return 0;
} catch (Exception err) {
LOG.severe(err);
return -1;
} finally {
CloserUtil.close(this.indexedFastaSequenceFile);
CloserUtil.close(this.out);
CloserUtil.close(r);
}
}
use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.
the class BamCmpCoverage method readBedFile.
private void readBedFile(final File bedFile) {
if (this.intervals == null) {
intervals = new IntervalTreeMap<Boolean>();
}
try {
LOG.info("Reading " + bedFile);
final BedLineCodec bedLineCodec = new BedLineCodec();
BufferedReader r = IOUtils.openFileForBufferedReading(bedFile);
String line;
while ((line = r.readLine()) != null) {
final BedLine bedLine = bedLineCodec.decode(line);
if (bedLine == null)
continue;
this.intervals.put(bedLine.toInterval(), true);
}
CloserUtil.close(r);
} catch (final IOException err) {
LOG.error(err);
throw new RuntimeException(err);
}
}
use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.
the class SamFindClippedRegions method doWork.
/*private static boolean closeTo(int pos1,int pos2, int max)
{
return Math.abs(pos2-pos1)<=max;
}*/
/*
private static boolean same(char c1,char c2)
{
if(c1=='N' || c2=='N') return false;
return Character.toUpperCase(c1)==Character.toUpperCase(c2);
}*/
@Override
public int doWork(List<String> args) {
int readLength = 150;
if (args.isEmpty()) {
LOG.error("illegal.number.of.arguments");
return -1;
}
List<Input> inputs = new ArrayList<Input>();
VariantContextWriter w = null;
// SAMFileWriter w=null;
try {
SAMSequenceDictionary dict = null;
/* create input, collect sample names */
Map<String, Input> sample2input = new HashMap<String, Input>();
for (final String filename : args) {
Input input = new Input(new File(filename));
// input.index=inputs.size();
inputs.add(input);
if (sample2input.containsKey(input.sampleName)) {
LOG.error("Duplicate sample " + input.sampleName + " in " + input.bamFile + " and " + sample2input.get(input.sampleName).bamFile);
return -1;
}
sample2input.put(input.sampleName, input);
if (dict == null) {
dict = input.header.getSequenceDictionary();
} else if (!SequenceUtil.areSequenceDictionariesEqual(dict, input.header.getSequenceDictionary())) {
LOG.error("Found more than one dictint sequence dictionary");
return -1;
}
}
LOG.info("Sample N= " + sample2input.size());
/* create merged iterator */
List<SAMFileHeader> headers = new ArrayList<SAMFileHeader>(sample2input.size());
for (Input input : inputs) headers.add(input.header);
SamFileHeaderMerger headerMerger = new SamFileHeaderMerger(SortOrder.coordinate, headers, true);
List<SamReader> readers = new ArrayList<SamReader>(sample2input.size());
for (Input input : inputs) readers.add(input.samFileReaderScan);
MergingSamRecordIterator merginIter = new MergingSamRecordIterator(headerMerger, readers, true);
Allele reference_allele = Allele.create("N", true);
Allele[] alternate_alleles = new Allele[] { Allele.create("<CLIP5>", false), Allele.create("<CLIP3>", false) };
Set<VCFHeaderLine> vcfHeaderLines = new HashSet<VCFHeaderLine>();
for (Allele alt : alternate_alleles) {
vcfHeaderLines.add(new VCFSimpleHeaderLine("<ID=" + alt.getDisplayString() + ",Description=\"StructVar\">", VCFHeaderVersion.VCF4_1, VCFConstants.ALT_HEADER_START.substring(2), Arrays.asList("ID", "Description")));
}
vcfHeaderLines.add(new VCFInfoHeaderLine("COUNT_SAMPLES", 1, VCFHeaderLineType.Integer, "Number of samples with depth>=" + this.min_depth));
vcfHeaderLines.add(new VCFInfoHeaderLine(VCFConstants.DEPTH_KEY, 1, VCFHeaderLineType.Integer, "Approximate read depth."));
vcfHeaderLines.add(new VCFFormatHeaderLine(VCFConstants.GENOTYPE_KEY, 1, VCFHeaderLineType.String, "Genotype"));
vcfHeaderLines.add(new VCFFormatHeaderLine(VCFConstants.DEPTH_KEY, 1, VCFHeaderLineType.Integer, "Approximate read depth"));
vcfHeaderLines.add(new VCFHeaderLine(getClass().getSimpleName() + "CmdLine", String.valueOf(getProgramCommandLine())));
vcfHeaderLines.add(new VCFHeaderLine(getClass().getSimpleName() + "Version", String.valueOf(getVersion())));
for (int side = 0; side < 2; ++side) {
vcfHeaderLines.add(new VCFFormatHeaderLine("CN" + (side == 0 ? 5 : 3), 1, VCFHeaderLineType.Integer, "count clipped in " + (side == 0 ? 5 : 3) + "'"));
}
if (dict != null) {
vcfHeaderLines.addAll(VCFUtils.samSequenceDictToVCFContigHeaderLine(dict));
}
VCFHeader vcfHeader = new VCFHeader(vcfHeaderLines, sample2input.keySet());
w = VCFUtils.createVariantContextWriterToStdout();
w.writeHeader(vcfHeader);
final IntervalTreeMap<Boolean> intervals = new IntervalTreeMap<>();
// w=swf.make(header, System.out);
SAMSequenceDictionaryProgress progress = new SAMSequenceDictionaryProgress(dict);
if (bedFile != null) {
final BedLineCodec bedLineCodec = new BedLineCodec();
LOG.info("Reading " + bedFile);
BufferedReader r = IOUtils.openFileForBufferedReading(bedFile);
String line;
while ((line = r.readLine()) != null) {
BedLine bedLine = bedLineCodec.decode(line);
if (bedLine == null)
continue;
if (dict != null && dict.getSequence(bedLine.getContig()) == null) {
LOG.warning("undefined chromosome in " + bedFile + " " + line);
continue;
}
intervals.put(bedLine.toInterval(), true);
}
CloserUtil.close(r);
}
LinkedList<SAMRecord> buffer = new LinkedList<SAMRecord>();
final Predicate<SAMRecord> filterSamRecords = new Predicate<SAMRecord>() {
@Override
public boolean test(SAMRecord rec) {
if (rec.getReadUnmappedFlag())
return false;
if (rec.isSecondaryOrSupplementary())
return false;
if (rec.getDuplicateReadFlag())
return false;
if (rec.getReadFailsVendorQualityCheckFlag())
return false;
Cigar cigar = rec.getCigar();
if (cigar == null || cigar.numCigarElements() < 2)
return false;
boolean found_S = false;
for (int side = 0; side < 2; ++side) {
CigarElement ce = cigar.getCigarElement(side == 0 ? 0 : cigar.numCigarElements() - 1);
// read must be clipped on 5' or 3' with a good length
if (!ce.getOperator().equals(CigarOperator.S))
continue;
found_S = true;
break;
}
if (!found_S)
return false;
SAMReadGroupRecord g = rec.getReadGroup();
if (g == null || g.getSample() == null || g.getSample().isEmpty())
return false;
return true;
}
};
final FilteringIterator<SAMRecord> forwardIterator = new FilteringIterator<SAMRecord>(merginIter, filterSamRecords);
for (; ; ) {
SAMRecord rec = null;
if (forwardIterator.hasNext()) {
rec = forwardIterator.next();
progress.watch(rec);
if (intervals != null && !intervals.containsOverlapping(new Interval(rec.getReferenceName(), rec.getAlignmentStart(), rec.getAlignmentEnd())))
continue;
}
// need to flush buffer ?
if (rec == null || (!buffer.isEmpty() && !buffer.getLast().getReferenceIndex().equals(rec.getReferenceIndex())) || (!buffer.isEmpty() && buffer.getLast().getUnclippedEnd() + readLength < rec.getUnclippedStart())) {
if (!buffer.isEmpty()) {
int chromStart = buffer.getFirst().getUnclippedStart();
int chromEnd = buffer.getFirst().getUnclippedEnd();
for (SAMRecord sam : buffer) {
chromStart = Math.min(chromStart, sam.getUnclippedStart());
chromEnd = Math.max(chromEnd, sam.getUnclippedEnd());
}
final int winShift = 5;
for (int pos = chromStart; pos + winShift <= chromEnd; pos += winShift) {
int[] count_big_clip = new int[] { 0, 0 };
// int max_depth[]=new int[]{0,0};
List<Genotype> genotypes = new ArrayList<Genotype>();
Set<Allele> all_alleles = new HashSet<Allele>();
all_alleles.add(reference_allele);
boolean found_one_depth_ok = false;
int sum_depth = 0;
int samples_with_high_depth = 0;
for (String sample : sample2input.keySet()) {
GenotypeBuilder gb = new GenotypeBuilder(sample);
int[] count_clipped = new int[] { 0, 0 };
Set<Allele> sample_alleles = new HashSet<Allele>(3);
for (int side = 0; side < 2; ++side) {
for (SAMRecord sam : buffer) {
if (!sam.getReadGroup().getSample().equals(sample))
continue;
Cigar cigar = sam.getCigar();
CigarElement ce = cigar.getCigarElement(side == 0 ? 0 : cigar.numCigarElements() - 1);
if (!ce.getOperator().equals(CigarOperator.S))
continue;
int clipStart = (side == 0 ? sam.getUnclippedStart() : sam.getAlignmentEnd() + 1);
int clipEnd = (side == 0 ? sam.getAlignmentStart() - 1 : sam.getUnclippedEnd());
if ((pos + winShift < clipStart || pos > clipEnd))
continue;
count_clipped[side]++;
if (ce.getLength() >= this.min_clip_length) {
count_big_clip[side]++;
}
sample_alleles.add(alternate_alleles[side]);
gb.attribute("CN" + (side == 0 ? 5 : 3), count_clipped[side]);
}
}
// if(!(found_one_big_clip[0] || found_one_big_clip[1])) continue;
if (count_clipped[0] + count_clipped[1] == 0)
continue;
if ((count_clipped[0] + count_clipped[1]) > min_depth) {
found_one_depth_ok = true;
++samples_with_high_depth;
}
sum_depth += (count_clipped[0] + count_clipped[1]);
gb.alleles(new ArrayList<Allele>(sample_alleles));
all_alleles.addAll(sample_alleles);
gb.DP(count_clipped[0] + count_clipped[1]);
genotypes.add(gb.make());
}
if (all_alleles.size() == 1) {
// all homozygotes
continue;
}
if (!found_one_depth_ok) {
continue;
}
if (!(count_big_clip[0] >= 1 || count_big_clip[1] >= 1)) {
continue;
}
Map<String, Object> atts = new HashMap<String, Object>();
atts.put("COUNT_SAMPLES", samples_with_high_depth);
atts.put(VCFConstants.DEPTH_KEY, sum_depth);
VariantContextBuilder vcb = new VariantContextBuilder();
vcb.chr(buffer.getFirst().getReferenceName());
vcb.start(pos);
vcb.stop(pos + winShift);
vcb.alleles(all_alleles);
vcb.attributes(atts);
vcb.genotypes(genotypes);
w.add(vcb.make());
}
buffer.clear();
}
if (rec == null) {
break;
}
}
buffer.add(rec);
}
merginIter.close();
progress.finish();
return 0;
} catch (Exception err) {
LOG.error(err);
return -1;
} finally {
for (Input input : inputs) {
CloserUtil.close(input);
}
}
}
use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.
the class PcrSliceReads method doWork.
@Override
public int doWork(List<String> args) {
if (bedFile == null) {
LOG.error("undefined bed file");
return -1;
}
BufferedReader r = null;
SamReader samReader = null;
try {
samReader = super.openSamReader(oneFileOrNull(args));
final BedLineCodec codec = new BedLineCodec();
r = IOUtils.openFileForBufferedReading(bedFile);
String line;
while ((line = r.readLine()) != null) {
final BedLine bed = codec.decode(line);
if (bed == null)
continue;
final String chrom = bed.getContig();
int chromStart1 = bed.getStart();
int chromEnd1 = bed.getEnd();
if (chromStart1 < 1 || chromStart1 > chromEnd1) {
LOG.error("Bad bed line " + line);
return -1;
}
final String name = bed.get(3).trim();
if (name == null || name.isEmpty()) {
LOG.error("Bad bed line (name missing) in " + line);
return -1;
}
final Interval i = new Interval(chrom, chromStart1, chromEnd1, false, name);
this.bedIntervals.put(i, i);
}
return run(samReader);
} catch (Exception e) {
LOG.error(e);
return -1;
} finally {
CloserUtil.close(r);
CloserUtil.close(samReader);
}
}
use of com.github.lindenb.jvarkit.util.bio.bed.BedLineCodec in project jvarkit by lindenb.
the class PcrClipReads method doWork.
@Override
public int doWork(final List<String> args) {
if (this.bedFile == null) {
LOG.error("undefined bed file ");
return -1;
}
BufferedReader r = null;
SamReader samReader = null;
try {
final String inputName = oneFileOrNull(args);
samReader = openSamReader(inputName);
final SAMFileHeader header = samReader.getFileHeader();
if (header == null) {
LOG.error("No SAM header in input");
return -1;
}
final SAMSequenceDictionary dict = header.getSequenceDictionary();
LOG.info("reading bed File " + this.bedFile);
r = IOUtils.openFileForBufferedReading(this.bedFile);
String line;
final BedLineCodec codec = new BedLineCodec();
while ((line = r.readLine()) != null) {
final BedLine bedLine = codec.decode(line);
if (bedLine == null) {
LOG.warn("Ignoring bed line " + line);
continue;
}
if (dict != null) {
if (dict.getSequenceIndex(bedLine.getContig()) == -1) {
LOG.warn("unknown contig in " + bedLine + ". Available are " + dict.getSequences().stream().map(SSR -> SSR.getSequenceName()).collect(Collectors.joining(", ")) + ". Skipping.");
continue;
}
}
final Interval i = bedLine.toInterval();
this.bedIntervals.put(i, i);
}
CloserUtil.close(r);
r = null;
return run(samReader);
} catch (final Exception err) {
LOG.error(err);
return -1;
} finally {
CloserUtil.close(r);
CloserUtil.close(samReader);
this.bedIntervals.clear();
;
}
}
Aggregations