private VariantContext getDbsnp(String rsID) {
if (dbsnpIterator == null) {
if (dbsnp == null)
throw new UserException.BadInput(
"No dbSNP rod was provided, but one is needed to decipher the correct indel alleles from the HapMap records");
RMDTrackBuilder builder =
new RMDTrackBuilder(
getToolkit().getReferenceDataSource().getReference().getSequenceDictionary(),
getToolkit().getGenomeLocParser(),
getToolkit().getArguments().unsafe,
getToolkit().getArguments().disableAutoIndexCreationAndLockingWhenReadingRods,
null);
dbsnpIterator =
builder
.createInstanceOfTrack(VCFCodec.class, new File(dbsnp.dbsnp.getSource()))
.getIterator();
// Note that we should really use some sort of seekable iterator here so that the search
// doesn't take forever
// (but it's complicated because the hapmap location doesn't match the dbsnp location, so we
// don't know where to seek to)
}
while (dbsnpIterator.hasNext()) {
GATKFeature feature = dbsnpIterator.next();
VariantContext vc = (VariantContext) feature.getUnderlyingObject();
if (vc.getID().equals(rsID)) return vc;
}
return null;
}
@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 int doWork(String[] args) {
boolean compressed = false;
int maxRecordsInRAM = 100000;
long count = -1L;
File fileout = null;
com.github.lindenb.jvarkit.util.cli.GetOpt opt =
new com.github.lindenb.jvarkit.util.cli.GetOpt();
int c;
while ((c = opt.getopt(args, getGetOptDefault() + "o:n:N:T:b")) != -1) {
switch (c) {
case 'b':
compressed = true;
break;
case 'N':
maxRecordsInRAM = Integer.parseInt(opt.getOptArg());
break;
case 'n':
count = Long.parseLong(opt.getOptArg());
break;
case 'o':
fileout = new File(opt.getOptArg());
break;
case 'T':
this.addTmpDirectory(new File(opt.getOptArg()));
break;
default:
{
switch (handleOtherOptions(c, opt, null)) {
case EXIT_FAILURE:
return -1;
case EXIT_SUCCESS:
return 0;
default:
break;
}
}
}
}
if (count < -1L) // -1 == infinite
{
error("Bad count:" + count);
return -1;
}
SamReader samReader = null;
SAMRecordIterator iter = null;
SAMFileWriter samWriter = null;
Random random = new Random();
CloseableIterator<RandSamRecord> iter2 = null;
try {
SamFileReaderFactory.setDefaultValidationStringency(ValidationStringency.SILENT);
if (opt.getOptInd() == args.length) {
info("Reading from stdin");
samReader = SamFileReaderFactory.mewInstance().openStdin();
} else if (opt.getOptInd() + 1 == args.length) {
File filename = new File(args[opt.getOptInd()]);
info("Reading from " + filename);
samReader = SamFileReaderFactory.mewInstance().open(filename);
} else {
error("Illegal number of arguments.");
return -1;
}
SAMFileHeader header = samReader.getFileHeader();
header = header.clone();
header.setSortOrder(SortOrder.unsorted);
header.addComment("Processed with " + getProgramName() + " : " + getProgramCommandLine());
SAMFileWriterFactory sfw = new SAMFileWriterFactory();
sfw.setCreateIndex(false);
sfw.setCreateMd5File(false);
if (fileout == null) {
if (compressed) {
samWriter = sfw.makeBAMWriter(header, true, System.out);
} else {
samWriter = sfw.makeSAMWriter(header, true, System.out);
}
} else {
samWriter = sfw.makeSAMOrBAMWriter(header, true, fileout);
this.addTmpDirectory(fileout);
}
iter = samReader.iterator();
SAMSequenceDictionaryProgress progress =
new SAMSequenceDictionaryProgress(samReader.getFileHeader().getSequenceDictionary());
SortingCollection<RandSamRecord> sorter =
SortingCollection.newInstance(
RandSamRecord.class,
new RandSamRecordCodec(header),
new RandSamRecordComparator(),
maxRecordsInRAM,
getTmpDirectories());
sorter.setDestructiveIteration(true);
while (iter.hasNext()) {
RandSamRecord r = new RandSamRecord();
r.rand_index = random.nextInt();
r.samRecord = progress.watch(iter.next());
sorter.add(r);
}
iter.close();
iter = null;
sorter.doneAdding();
iter2 = sorter.iterator();
if (count == -1) {
while (iter2.hasNext()) {
samWriter.addAlignment(iter2.next().samRecord);
}
} else {
while (iter2.hasNext() && count > 0) {
samWriter.addAlignment(iter2.next().samRecord);
count--;
}
}
iter2.close();
iter2 = null;
sorter.cleanup();
progress.finish();
} catch (Exception e) {
error(e);
return -1;
} finally {
CloserUtil.close(iter);
CloserUtil.close(iter2);
CloserUtil.close(samReader);
CloserUtil.close(samWriter);
}
return 0;
}
@Override
protected int doWork() {
if (DELIM.length() != 1) {
LOG.error("DELIM must have length==1 . Got " + DELIM.length());
return -1;
}
InputStream in = System.in;
SortingCollection<Cell> sorter = null;
final Comparator<Cell> comparator =
new Comparator<Biostar84786.Cell>() {
@Override
public int compare(final Cell o1, final Cell o2) {
int i;
i = (o1.col < o2.col ? -1 : o1.col > o2.col ? 1 : 0);
if (i != 0) return i;
i = (o1.row < o2.row ? -1 : o1.row > o2.row ? 1 : 0);
if (i != 0) return i;
return o1.content.compareTo(o2.content);
}
};
try {
final char delimiter = DELIM.charAt(0);
sorter =
SortingCollection.newInstance(
Cell.class, new CellCodec(), comparator, super.MAX_RECORDS_IN_RAM);
sorter.setDestructiveIteration(true);
if (IN != null) {
LOG.info("opening " + IN);
in = IOUtils.openFileForReading(IN);
}
long row = 0L;
long col = 0L;
StringBuilder b = new StringBuilder();
for (; ; ) {
int c = in.read();
if (c == '\n' || c == -1) {
sorter.add(new Cell(row, col, b));
row++;
col = 0;
b.setLength(0);
if (c == -1) break;
if (row % 10000 == 0) LOG.info("row:" + row);
} else if (c == delimiter) {
sorter.add(new Cell(row, col, b));
b.setLength(0);
col++;
} else {
b.append((char) c);
}
}
sorter.doneAdding();
if (IN != null) in.close();
in = null;
CloseableIterator<Cell> iter = sorter.iterator();
long curr_col = -1L;
long x = 0L;
for (; ; ) {
if (!iter.hasNext()) {
System.out.println();
break;
}
Cell c = iter.next();
if (c.col != curr_col) {
if (curr_col != -1L) System.out.println();
x = 0L;
curr_col = c.col;
}
if (x > 0L) System.out.print(DELIM);
System.out.print(c.content);
x++;
}
iter.close();
LOG.info("Done.");
} catch (Exception e) {
e.printStackTrace();
LOG.error(e, "BOUM");
return -1;
} finally {
if (sorter != null) sorter.cleanup();
if (in != null) CloserUtil.close(in);
}
return 0;
}