@Override
protected Index doInBackground() throws Exception {
int binSize = IgvTools.LINEAR_BIN_SIZE;
FeatureCodec codec =
CodecFactory.getCodec(
file.getAbsolutePath(), GenomeManager.getInstance().getCurrentGenome());
if (codec != null) {
try {
Index index = IndexFactory.createLinearIndex(file, codec, binSize);
if (index != null) {
IgvTools.writeTribbleIndex(index, idxFile.getAbsolutePath());
}
return index;
} catch (TribbleException.MalformedFeatureFile e) {
StringBuffer buf = new StringBuffer();
buf.append("<html>Files must be sorted by start position prior to indexing.<br>");
buf.append(e.getMessage());
buf.append(
"<br><br>Note: igvtools can be used to sort the file, select \"File > Run igvtools...\".");
MessageUtils.showMessage(buf.toString());
}
} else {
throw new DataLoadException("Unknown File Type", file.getAbsolutePath());
}
return null;
}
/**
* Return an parser instance appropriate the the file type. Currently the filename is used to
* determine file type, this is fragile obviously but what is the alternative?
*/
public static FeatureParser getInstanceFor(ResourceLocator locator, Genome genome) {
FeatureCodec codec = CodecFactory.getCodec(locator, genome);
if (codec != null && codec instanceof AsciiFeatureCodec) {
return new FeatureCodecParser((AsciiFeatureCodec) codec, genome);
} else {
return null;
}
}
public static boolean isIndexed(ResourceLocator locator, Genome genome) {
// Checking for the index is expensive over HTTP. First see if this is an indexable format by
// fetching the codec
String fullPath = locator.getPath();
String pathNoQuery = locator.getURLPath();
if (!CodecFactory.hasCodec(locator, genome)) {
return false;
}
String indexExtension = pathNoQuery.endsWith("gz") ? ".tbi" : ".idx";
String indexPath = fullPath + indexExtension;
if (HttpUtils.isRemoteURL(fullPath)) {
// Handle query string, if it exists
String[] toks = fullPath.split("\\?", 2);
if (toks.length == 2) {
indexPath = String.format("%s%s?%s", toks[0], indexExtension, toks[1]);
}
}
return FileUtils.resourceExists(indexPath);
}
private static FeatureCodec getCodec(String path, Genome genome) {
String tmp = getStrippedFilename(path);
return CodecFactory.getCodec(tmp, genome);
}
/**
* Switches on various attributes of locator (mainly locator path extension and whether the
* locator is indexed) to call the appropriate loading method.
*
* @param locator
* @param genome
* @return
*/
public List<Track> load(ResourceLocator locator, Genome genome) throws DataLoadException {
final String path = locator.getPath().trim();
log.info("Loading resource, path " + path);
try {
String typeString = locator.getTypeString();
if (typeString.endsWith(".tbi")) {
MessageUtils.showMessage(
"<html><b>Error:</b>File type '.tbi' is not recognized. If this is a 'tabix' index <br>"
+ " load the associated gzipped file, which should have an extension of '.gz'");
}
// This list will hold all new tracks created for this locator
List<Track> newTracks = new ArrayList<Track>();
String dbUrl = locator.getDBUrl();
LoadHandler handler = getTrackLoaderHandler(typeString);
if (dbUrl != null) {
this.loadFromDatabase(locator, newTracks, genome);
} else if (typeString.endsWith(".dbxml")) {
loadFromDBProfile(locator, newTracks);
} else if (typeString.endsWith(".gmt")) {
loadGMT(locator);
} else if (typeString.equals("das")) {
loadDASResource(locator, newTracks);
} else if (typeString.endsWith(".vcf.list")) {
loadVCFListFile(locator, newTracks, genome);
} else if (typeString.endsWith(".trio")) {
loadTrioData(locator);
} else if (typeString.endsWith("varlist")) {
VariantListManager.loadVariants(locator);
} else if (typeString.endsWith("samplepathmap")) {
VariantListManager.loadSamplePathMap(locator);
} else if (typeString.endsWith(".rnai.gct")) {
loadRnaiGctFile(locator, newTracks, genome);
} else if (typeString.endsWith(".gct")
|| typeString.endsWith("res")
|| typeString.endsWith("tab")) {
loadGctFile(locator, newTracks, genome);
} else if (typeString.endsWith(".gbk") || typeString.endsWith(".gb")) {
loadGbkFile(locator, newTracks, genome);
} else if (typeString.endsWith(".cn")
|| typeString.endsWith(".xcn")
|| typeString.endsWith(".snp")
|| typeString.endsWith(".igv")
|| typeString.endsWith(".loh")) {
loadIGVFile(locator, newTracks, genome);
} else if (typeString.endsWith(".cbs")
|| typeString.endsWith(".seg")
|| typeString.endsWith("glad")
|| typeString.endsWith("birdseye_canary_calls")
|| typeString.endsWith(".seg.zip")) {
loadSegFile(locator, newTracks, genome);
} else if (typeString.endsWith(".gistic")) {
loadGisticFile(locator, newTracks);
} else if (typeString.endsWith(".gs")) {
loadRNAiGeneScoreFile(locator, newTracks, RNAIGeneScoreParser.Type.GENE_SCORE, genome);
} else if (typeString.endsWith(".riger")) {
loadRNAiGeneScoreFile(locator, newTracks, RNAIGeneScoreParser.Type.POOLED, genome);
} else if (typeString.endsWith(".hp")) {
loadRNAiHPScoreFile(locator);
} else if (typeString.contains(".tabblastn") || typeString.endsWith(".orthologs")) {
loadSyntentyMapping(locator, newTracks);
} else if (typeString.endsWith(".sam")
|| typeString.endsWith(".bam")
|| typeString.endsWith(".cram")
|| typeString.endsWith(".sam.list")
|| typeString.endsWith(".bam.list")
|| typeString.endsWith(".aligned")
|| typeString.endsWith(".sai")
|| typeString.endsWith(".bai")
|| typeString.equals("alist")
|| typeString.equals(Ga4ghAPIHelper.RESOURCE_TYPE)) {
loadAlignmentsTrack(locator, newTracks, genome);
} else if (typeString.endsWith(".wig")
|| typeString.endsWith(".bedgraph")
|| typeString.endsWith(".bdg")
|| typeString.endsWith("cpg.txt")
|| typeString.endsWith(".expr")) {
loadWigFile(locator, newTracks, genome);
} else if (typeString.endsWith("fpkm_tracking")
|| typeString.endsWith("gene_exp.diff")
|| typeString.endsWith("cds_exp.diff")) {
loadCufflinksFile(locator, newTracks, genome);
} else if (typeString.contains(".dranger")) {
loadDRangerFile(locator, newTracks, genome);
} else if (typeString.endsWith(".ewig.tdf") || (typeString.endsWith(".ewig.ibf"))) {
loadEwigIBFFile(locator, newTracks, genome);
} else if (typeString.endsWith(".bw")
|| typeString.endsWith(".bb")
|| typeString.endsWith(".bigwig")
|| typeString.endsWith(".bigbed")) {
loadBWFile(locator, newTracks, genome);
} else if (typeString.endsWith(".ibf") || typeString.endsWith(".tdf")) {
loadTDFFile(locator, newTracks, genome);
} else if (typeString.endsWith(".counts")) {
loadGobyCountsArchive(locator, newTracks, genome);
} else if (WiggleParser.isWiggle(locator)) {
loadWigFile(locator, newTracks, genome);
} else if (typeString.endsWith(".maf")) {
loadMultipleAlignmentTrack(locator, newTracks, genome);
} else if (typeString.endsWith(".maf.dict")) {
loadMultipleAlignmentTrack(locator, newTracks, genome);
} else if (typeString.contains(".peak.bin")) {
loadPeakTrack(locator, newTracks, genome);
} else if (typeString.endsWith("mage-tab")
|| ExpressionFileParser.parsableMAGE_TAB(locator)) {
locator.setDescription("MAGE_TAB");
loadGctFile(locator, newTracks, genome);
} else if (typeString.endsWith(".bp")) {
loadBasePairFile(locator, newTracks, genome);
} else if (GWASParser.isGWASFile(typeString)) {
loadGWASFile(locator, newTracks, genome);
} else if (GobyAlignmentQueryReader.supportsFileType(path)) {
loadAlignmentsTrack(locator, newTracks, genome);
} else if (typeString.endsWith(".list")) {
// This should be deprecated
loadListFile(locator, newTracks, genome);
} else if (typeString.endsWith(".smap")) {
loadSMAPFile(locator, newTracks, genome);
} else if (CodecFactory.hasCodec(locator, genome) && !forceNotTribble(typeString)) {
loadTribbleFile(locator, newTracks, genome);
} else if (handler != null) {
// Custom loader specified
log.info(String.format("Loading %s with %s", path, handler));
handler.load(path, newTracks);
} else if (AttributeManager.isSampleInfoFile(locator)) {
// This might be a sample information file.
AttributeManager.getInstance().loadSampleInfo(locator);
} else {
MessageUtils.showMessage("<html>Unknown file type: " + path + "<br>Check file extension");
}
// Track line
TrackProperties tp = null;
String trackLine = locator.getTrackLine();
if (trackLine != null) {
tp = new TrackProperties();
ParsingUtils.parseTrackLine(trackLine, tp);
}
for (Track track : newTracks) {
if (locator.getFeatureInfoURL() != null) {
track.setUrl(locator.getFeatureInfoURL());
}
if (tp != null) {
track.setProperties(tp);
}
if (locator.getColor() != null) {
track.setColor(locator.getColor());
}
if (locator.getSampleId() != null) {
track.setSampleId(locator.getSampleId());
}
}
return newTracks;
} catch (Exception e) {
if (!NOLogExceptions.contains(e.getClass())) {
log.error(e.getMessage(), e);
}
throw new DataLoadException(e.getMessage());
}
}
/**
* Create an index for an alignment or feature file The output index will have the same base name
* is the input file, although it may be in a different directory. An appropriate index extension
* (.sai, .idx, etc.) will be appended.
*
* @param ifile
* @param typeString
* @param outputDir
* @param indexType
* @param binSize
* @throws IOException
*/
public String doIndex(
String ifile, String typeString, String outputDir, int indexType, int binSize)
throws IOException {
File inputFile = new File(ifile);
if (outputDir == null) {
outputDir = inputFile.getParent();
}
String outputFileName = (new File(outputDir, inputFile.getName())).getAbsolutePath();
if (typeString.endsWith("gz")) {
System.out.println("Cannot index a gzipped file");
throw new PreprocessingException("Cannot index a gzipped file");
}
if (typeString.endsWith("bam")) {
String msg =
"Cannot index a BAM file. Use the samtools package for sorting and indexing BAM files.";
System.out.println(msg);
throw new PreprocessingException(msg);
}
// We have different naming conventions for different index files
if (typeString.endsWith("sam") && !outputFileName.endsWith(".sai")) {
outputFileName += ".sai";
} else if (typeString.endsWith("bam") && !outputFileName.endsWith(".bai")) {
outputFileName += ".bai";
} else if (typeString.endsWith("fa") && !outputFileName.endsWith(".fai")) {
outputFileName += ".fai";
} else if (typeString.endsWith("fasta") && !outputFileName.endsWith(".fai")) {
outputFileName += ".fai";
} else if (!typeString.endsWith("sam")
&& !typeString.endsWith("bam")
&& !outputFileName.endsWith(".idx")) {
outputFileName += ".idx";
}
File outputFile = new File(outputFileName);
// Sam/FASTA files are special
try {
if (typeString.endsWith("sam")) {
AlignmentIndexer indexer = AlignmentIndexer.getInstance(inputFile, null, null);
indexer.createSamIndex(outputFile);
return outputFileName;
} else if (typeString.equals(".fa") || typeString.equals(".fasta")) {
FastaUtils.createIndexFile(inputFile.getAbsolutePath(), outputFileName);
return outputFileName;
}
} catch (Exception e) {
e.printStackTrace();
// Delete output file as it is probably corrupt
if (outputFile.exists()) {
outputFile.delete();
}
}
Genome genome = null; // <= don't do chromosome conversion
FeatureCodec codec = CodecFactory.getCodec(ifile, genome);
if (codec != null) {
try {
createTribbleIndex(ifile, outputFile, indexType, binSize, codec);
} catch (TribbleException.MalformedFeatureFile e) {
StringBuffer buf = new StringBuffer();
buf.append("<html>Files must be sorted by start position prior to indexing.<br>");
buf.append(e.getMessage());
buf.append(
"<br><br>Note: igvtools can be used to sort the file, select \"File > Run igvtools...\".");
MessageUtils.showMessage(buf.toString());
}
} else {
throw new DataLoadException("Unknown File Type", ifile);
}
System.out.flush();
return outputFileName;
}
public static boolean canParse(ResourceLocator locator) {
return CodecFactory.getCodec(locator, null) != null;
}