/**
* Note: This is an exact copy of the method in ExpressionFileParser. Refactor to merge these two
* parsers, or share a common base class.
*
* @param comment
* @param dataset
*/
private void parseDirective(String comment, IGVDataset dataset) {
String tmp = comment.substring(1, comment.length());
if (tmp.startsWith("track")) {
ParsingUtils.parseTrackLine(tmp, dataset.getTrackProperties());
} else if (tmp.startsWith("columns")) {
parseColumnLine(tmp);
} else {
String[] tokens = tmp.split("=");
if (tokens.length != 2) {
return;
}
String key = tokens[0].trim().toLowerCase();
if (key.equals("name")) {
dataset.setName(tokens[1].trim());
} else if (key.equals("type")) {
try {
dataset.setTrackType(TrackType.valueOf(tokens[1].trim().toUpperCase()));
} catch (Exception exception) {
// Ignore
}
} else if (key.equals("coords")) {
startBase = Integer.parseInt(tokens[1].trim());
}
}
}
public void loadTDFFile(ResourceLocator locator, List<Track> newTracks, Genome genome) {
log.debug("Loading TDF file " + locator.getPath());
TDFReader reader = TDFReader.getReader(locator);
TrackType type = reader.getTrackType();
TrackProperties props = null;
String trackLine = reader.getTrackLine();
if (trackLine != null && trackLine.length() > 0) {
props = new TrackProperties();
ParsingUtils.parseTrackLine(trackLine, props);
}
// In case of conflict between the resource locator display name and the track properties name,
// use the resource locator
String name = locator.getName();
if (name != null && props != null) {
props.setName(name);
}
if (name == null) {
name = props == null ? locator.getTrackName() : props.getName();
}
int trackNumber = 0;
String path = locator.getPath();
boolean multiTrack = reader.getTrackNames().length > 1;
for (String heading : reader.getTrackNames()) {
String trackId = multiTrack ? path + "_" + heading : path;
String trackName = multiTrack ? heading : name;
final DataSource dataSource =
locator.getPath().endsWith(".counts")
? new GobyCountArchiveDataSource(locator)
: new TDFDataSource(reader, trackNumber, heading, genome);
DataSourceTrack track = new DataSourceTrack(locator, trackId, trackName, dataSource);
String displayName = (name == null || multiTrack) ? heading : name;
track.setName(displayName);
track.setTrackType(type);
if (props != null) {
track.setProperties(props);
}
newTracks.add(track);
trackNumber++;
}
}
private void loadEwigIBFFile(ResourceLocator locator, List<Track> newTracks, Genome genome) {
TDFReader reader = TDFReader.getReader(locator.getPath());
TrackProperties props = null;
String trackLine = reader.getTrackLine();
if (trackLine != null && trackLine.length() > 0) {
props = new TrackProperties();
ParsingUtils.parseTrackLine(trackLine, props);
}
EWigTrack track = new EWigTrack(locator, genome);
if (props != null) {
track.setProperties(props);
}
track.setName(locator.getTrackName());
newTracks.add(track);
}
/**
* Load all features in this file.
*
* @param reader
* @param maxLines
* @return
*/
public List<org.broad.tribble.Feature> loadFeatures(BufferedReader reader, int maxLines) {
List<org.broad.tribble.Feature> features = new ArrayList<org.broad.tribble.Feature>();
String nextLine = null;
int nLines = 0;
try {
while ((nextLine = reader.readLine()) != null) {
nextLine = nextLine.trim();
if (nextLine.length() == 0) continue;
nLines++;
if ((maxLines > 0) && (nLines > maxLines)) {
break;
}
try {
if (nextLine.startsWith("#")) {
if (nextLine.startsWith("#type")) {
String[] tokens = Globals.equalPattern.split(nextLine);
if (tokens.length > 1) {
try {
// TODO: type is not currently used, is there any reason to keep this?
TrackType type = TrackType.valueOf(tokens[1]);
} catch (Exception e) {
log.error("Error converting track type: " + tokens[1]);
}
}
} else if (nextLine.startsWith("#track")) {
TrackProperties tp = new TrackProperties();
ParsingUtils.parseTrackLine(nextLine, tp);
setTrackProperties(tp);
if (tp.isGffTags()) {
gffTags = true;
}
} else if (nextLine.startsWith("#coords")) {
try {
String[] tokens = Globals.equalPattern.split(nextLine);
startBase = Integer.parseInt(tokens[1]);
} catch (Exception e) {
log.error("Error parsing coords line: " + nextLine, e);
}
} else if (nextLine.startsWith("#gffTags")) {
gffTags = true;
}
} else {
Feature feature = parseLine(nextLine);
if (feature != null) {
features.add(feature);
}
}
} catch (NumberFormatException e) {
// Expected condition -- for example comments. don't log as it slows down
// the parsing and is not useful information.
}
}
} catch (java.io.EOFException e) {
// This exception is due to a known bug with java zip library. Not
// in general a real error, and nothing we can do about it in any
// event.
return features;
} catch (Exception e) {
if (nextLine != null && nLines != 0) {
throw new ParserException(e.getMessage(), e, nLines, nextLine);
} else {
throw new RuntimeException(e);
}
}
// TODO -- why is this test here? This will break igvtools processing of expression files
// if (IGV.hasInstance() || Globals.isTesting()) {
FeatureDB.addFeatures(features);
// }
return features;
}
/**
* 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());
}
}