public void initialize() {
if (variants.size() != 2) {
throw new UserException.BadArgumentValue(
"variant", "this tool requires exactly 2 input variant files");
}
source1 = variants.get(0).getName();
source2 = variants.get(1).getName();
if (SET_KEY.toLowerCase().equals("null")) SET_KEY = null;
if (STATUS_KEY.toLowerCase().equals("null")) STATUS_KEY = null;
// for now, INFO and FORMAT fields are not propagated to the output VCF (so they aren't put into
// the header)
Set<VCFHeaderLine> headerLines = new HashSet<VCFHeaderLine>();
if (SET_KEY != null)
headerLines.add(
new VCFInfoHeaderLine(
SET_KEY, 1, VCFHeaderLineType.String, "Source VCF for the merged record"));
if (STATUS_KEY != null)
headerLines.add(
new VCFInfoHeaderLine(
STATUS_KEY, 1, VCFHeaderLineType.String, "Extent to which records match"));
final VCFHeader vcfHeader = new VCFHeader(headerLines, Collections.<String>emptySet());
baseWriter.writeHeader(vcfHeader);
writer = VariantContextWriterFactory.sortOnTheFly(baseWriter, ACTIVE_WINDOW);
}
public void initialize() {
vcfwriter = VariantContextWriterFactory.sortOnTheFly(baseWriter, 40, false);
variantOverlapAnnotator =
new VariantOverlapAnnotator(dbsnp.dbsnp, getToolkit().getGenomeLocParser());
}
@Override
protected int execute() throws Exception {
BasicConfigurator.configure();
logger.setLevel(Level.INFO);
final ReferenceSequenceFile ref;
try {
ref = ReferenceSequenceFileFactory.getReferenceSequenceFile(refFile);
} catch (Exception e) {
throw new UserException("Couldn't load provided reference sequence file " + refFile, e);
}
variant = parseVariantList(variant);
Comparator<Pair<Integer, File>> positionComparator = new PositionComparator();
Queue<Pair<Integer, File>> priorityQueue;
if (assumeSorted) priorityQueue = new LinkedList<>();
else priorityQueue = new PriorityQueue<>(10000, positionComparator);
FileType fileType = null;
for (File file : variant) {
// if it returns a valid type, it will be the same for all files
fileType = fileExtensionCheck(file, fileType);
if (fileType == FileType.INVALID) return 1;
if (assumeSorted) {
priorityQueue.add(new Pair<>(0, file));
} else {
if (!file.exists()) {
throw new UserException(String.format("File %s doesn't exist", file.getAbsolutePath()));
}
FeatureReader<VariantContext> reader = getFeatureReader(fileType, file);
Iterator<VariantContext> it = reader.iterator();
if (!it.hasNext()) {
System.err.println(
String.format("File %s is empty. This file will be ignored", file.getAbsolutePath()));
continue;
}
VariantContext vc = it.next();
int firstPosition = vc.getStart();
reader.close();
priorityQueue.add(new Pair<>(firstPosition, file));
}
}
FileOutputStream outputStream = new FileOutputStream(outputFile);
EnumSet<Options> options = EnumSet.of(Options.INDEX_ON_THE_FLY);
IndexCreator idxCreator =
GATKVCFUtils.makeIndexCreator(
variant_index_type, variant_index_parameter, outputFile, ref.getSequenceDictionary());
final VariantContextWriter outputWriter =
VariantContextWriterFactory.create(
outputFile, outputStream, ref.getSequenceDictionary(), idxCreator, options);
boolean firstFile = true;
int count = 0;
while (!priorityQueue.isEmpty()) {
count++;
File file = priorityQueue.remove().getSecond();
if (!file.exists()) {
throw new UserException(String.format("File %s doesn't exist", file.getAbsolutePath()));
}
FeatureReader<VariantContext> reader = getFeatureReader(fileType, file);
if (count % 10 == 0) System.out.print(count);
else System.out.print(".");
if (firstFile) {
VCFHeader header = (VCFHeader) reader.getHeader();
outputWriter.writeHeader(header);
firstFile = false;
}
Iterator<VariantContext> it = reader.iterator();
while (it.hasNext()) {
VariantContext vc = it.next();
outputWriter.add(vc);
}
reader.close();
}
System.out.println();
outputWriter.close();
return 0;
}
