1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
|
/*
* Copyright (C) 2009-2010 Institute for Computational Biomedicine,
* Weill Medical College of Cornell University
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
package org.campagnelab.goby.modes;
import com.martiansoftware.jsap.JSAPException;
import com.martiansoftware.jsap.JSAPResult;
import it.unimi.dsi.lang.MutableString;
import org.campagnelab.goby.algorithmic.data.WeightsInfo;
import org.campagnelab.goby.algorithmic.algorithm.*;
import org.campagnelab.goby.counts.CountWriterHelper2;
import org.campagnelab.goby.counts.CountsArchiveWriter;
import org.campagnelab.goby.counts.CountsWriterHelperI;
import org.campagnelab.goby.counts.CountsWriterI;
import org.campagnelab.goby.algorithmic.dsv.DiscoverVariantPositionData;
import org.campagnelab.goby.modes.dsv.IterateSortedAlignmentsListImpl;
import org.campagnelab.goby.util.Timer;
import edu.cornell.med.icb.identifier.DoubleIndexedIdentifier;
import it.unimi.dsi.fastutil.ints.IntOpenHashSet;
import it.unimi.dsi.fastutil.ints.IntSet;
import it.unimi.dsi.fastutil.objects.ObjectOpenHashSet;
import it.unimi.dsi.fastutil.objects.ObjectSet;
import it.unimi.dsi.logging.ProgressLogger;
import org.campagnelab.goby.alignments.*;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.IOException;
import java.util.Arrays;
import static org.campagnelab.goby.counts.CountsArchiveWriter.COUNT_ARCHIVE_MODIFIER_DEFAULT;
/**
* Converts a compact alignment to a compressed count archive.
*
* @author Fabien Campagne
*/
public class CompactAlignmentToCountsMode extends AbstractGobyMode {
/**
* The mode name.
*/
private static final String MODE_NAME = "alignment-to-counts";
/**
* The overridden short mode name.
*/
private static final String SHORT_MODE_NAME = "altc";
/**
* The mode description help text.
*/
private static final String MODE_DESCRIPTION =
"Converts a compact alignment to counts.";
/**
* The output file.
*/
private String outputFile;
private boolean filterByReferenceNames;
private ObjectSet<String> includeReferenceNames = new ObjectOpenHashSet<String>();
private boolean fullGenomeAlignment;
private String[] basenames;
private String optionalOutputFile;
private boolean accumulatePeakHistogram;
private int focusOnStrand;
private String countArchiveModifier = COUNT_ARCHIVE_MODIFIER_DEFAULT;
private WeightParameters weightParams;
private static final Logger LOG = LoggerFactory.getLogger(CompactAlignmentToCountsMode.class);
private boolean verbose;
private boolean filterAmbiguousReads;
@Override
public String getModeName() {
return MODE_NAME;
}
@Override
public String getShortModeName() {
return SHORT_MODE_NAME;
}
@Override
public String getModeDescription() {
return MODE_DESCRIPTION;
}
/**
* Configure.
*
* @param args command line arguments
* @return this object for chaining
* @throws java.io.IOException error parsing
* @throws com.martiansoftware.jsap.JSAPException error parsing
*/
@Override
public AbstractCommandLineMode configure(final String[] args) throws IOException, JSAPException {
final JSAPResult jsapResult = parseJsapArguments(args);
final String[] inputFiles = jsapResult.getStringArray("input");
final ObjectSet<String> basenameSet = new ObjectOpenHashSet<String>();
for (final String inputFile : inputFiles) {
basenameSet.add(AlignmentReaderImpl.getBasename(inputFile));
}
basenames = basenameSet.toArray(new String[basenameSet.size()]);
optionalOutputFile = jsapResult.getString("output");
final String includeReferenceNameCommas = jsapResult.getString("include-reference-names");
if (includeReferenceNameCommas != null) {
includeReferenceNames = new ObjectOpenHashSet<String>();
includeReferenceNames.addAll(Arrays.asList(includeReferenceNameCommas.split("[,]")));
System.out.println("Will write counts for the following sequences:");
for (final String name : includeReferenceNames) {
System.out.println(name);
}
filterByReferenceNames = true;
}
accumulatePeakHistogram = !jsapResult.getBoolean("start-only", false);
if (!accumulatePeakHistogram) {
final String strandChoice = jsapResult.getString("strand-choice");
if ("POSITIVE".equalsIgnoreCase(strandChoice) || "FORWARD".equalsIgnoreCase(strandChoice)) {
focusOnStrand = ComputeStartCount.POSITIVE_STRAND_ONLY;
countArchiveModifier = "forward-strand-starts";
} else if ("NEGATIVE".equalsIgnoreCase(strandChoice) || "REVERSE".equalsIgnoreCase(strandChoice)) {
focusOnStrand = ComputeStartCount.REVERSE_STRAND_ONLY;
countArchiveModifier = "reverse-strand-starts";
} else if ("BOTH".equalsIgnoreCase(strandChoice) || "EITHER".equalsIgnoreCase(strandChoice)) {
focusOnStrand = ComputeStartCount.BOTH_STRAND;
countArchiveModifier = "either-strand-starts";
} else {
System.err.println("strand choice must be one of {forward, positive, reverse, negative, both, either}.");
System.exit(2);
}
}
weightParams = CompactAlignmentToAnnotationCountsMode.configureWeights(jsapResult);
filterAmbiguousReads = jsapResult.getBoolean("filter-ambiguous-reads");
if (filterAmbiguousReads) {
System.out.println("Ambiguous reads will not be considered when estimating count statistics.");
}
return this;
}
/**
* Run the mode.
*
* @throws java.io.IOException error reading / writing
*/
@Override
public void execute() throws IOException {
final ProgressLogger progress = new ProgressLogger(LOG);
progress.expectedUpdates = basenames.length;
progress.itemsName = "basenames";
progress.start();
if (basenames.length > 1) {
verbose = true;
}
for (final String basename : basenames) {
if (optionalOutputFile == null) {
outputFile = basename;
}
processFullGenomeAlignment(basename);
progress.info = "last processed: " + basename;
progress.lightUpdate();
}
progress.done();
}
private void processFullGenomeAlignment(final String basename) throws IOException {
final AlignmentReaderFactory factory = filterAmbiguousReads ? new NonAmbiguousAlignmentReaderFactory() :
new DefaultAlignmentReaderFactory();
System.out.println("Processing " + basename);
// don't use the factory here, we just need to read the header and AlignmentReaderImpl will be faster:
AlignmentReader reader = new AlignmentReaderImpl(basename);
reader.readHeader();
/*if (reader.isSorted() && reader.getNumberOfAlignedReads() > 50000000) {
// if the alignment is sorted and has more than 50 million aligned reads, we switch to the more scalable
// count production method. The method scales to any size alignment, but is about 3-4 times slower than
// directly loading the entire alignment in memory..
processSortedAlignmentFullGenome(basename, reader,factory);
return;
} */
final int numberOfReferences = reader.getNumberOfTargets();
final DoubleIndexedIdentifier referenceIds = new DoubleIndexedIdentifier(reader.getTargetIdentifiers());
reader.close();
//System.out.println(String.format("Alignment contains %d reference sequences", numberOfReferences));
final ComputeCountInterface[] algs = new ComputeCountInterface[numberOfReferences];
final CountsArchiveWriter countArchive;
countArchive = new CountsArchiveWriter(basename, countArchiveModifier);
// be verbose if only one basename to process.
countArchive.setVerbose(verbose);
// CountsWriter writers[] = new CountsWriter[numberOfReferences];
final IntSet referencesToProcess = new IntOpenHashSet();
WeightsInfo weights = null;
if (weightParams.useWeights) {
weights = CompactAlignmentToAnnotationCountsMode.loadWeights(basename, weightParams.useWeights,
weightParams.weightId);
if (weights != null) {
System.err.println("Weights have been provided and loaded and will be used to reweight counts.");
}
}
// create count writers, one for each reference sequence in the alignment:
for (int referenceIndex = 0; referenceIndex < numberOfReferences; referenceIndex++) {
final MutableString id = referenceIds.getId(referenceIndex);
if (id != null) {
final String referenceName = id.toString();
if (filterByReferenceNames) {
if (includeReferenceNames.contains(referenceName)) {
// subset of reference names selected by the command line:
referencesToProcess.add(referenceIndex);
}
} else {
// process each sequence:
referencesToProcess.add(referenceIndex);
}
if (referencesToProcess.contains(referenceIndex)) {
if (accumulatePeakHistogram) {
final ComputeCountInterface algo = new ComputeCount();
algs[referenceIndex] = chooseAlgorithm(weightParams, weights, algo);
} else {
algs[referenceIndex] = new ComputeStartCount(focusOnStrand);
}
algs[referenceIndex].startPopulating();
}
}
}
final AlignmentReader referenceReader = new AlignmentReaderImpl(basename);
referenceReader.readHeader();
// read the alignment:
// System.out.println("Loading the alignment..");
for (final Alignments.AlignmentEntry alignmentEntry : referenceReader) {
final int referenceIndex = alignmentEntry.getTargetIndex();
if (referencesToProcess.contains(referenceIndex)) {
final int startPosition = alignmentEntry.getPosition();
final int alignmentLength = alignmentEntry.getTargetAlignedLength();
for (int i = 0; i < alignmentEntry.getMultiplicity(); ++i) {
algs[referenceIndex].populate(startPosition, startPosition + alignmentLength,
!alignmentEntry.getMatchingReverseStrand(), alignmentEntry.getQueryIndex());
}
}
}
reader.close();
final Timer timer = new Timer();
timer.start();
for (final int referenceIndex : referencesToProcess) {
final String chromosomeName = referenceIds.getId(referenceIndex).toString();
if (verbose) {
System.out.println("Writing counts for reference " + chromosomeName);
}
algs[referenceIndex].accumulate();
final CountsWriterI countsWriterI = countArchive.newCountWriter(referenceIndex, chromosomeName);
algs[referenceIndex].baseCount(countsWriterI);
countArchive.returnWriter(countsWriterI);
algs[referenceIndex] = null;
// Runtime.getRuntime().gc();
}
countArchive.close();
timer.stop();
System.out.println(timer);
}
private void processSortedAlignmentFullGenome(final String basename, AlignmentReader reader, AlignmentReaderFactory factory) throws IOException {
reader.close();
System.out.println("Processing in large alignment mode: " + basename);
final CountsArchiveWriter countArchive = new CountsArchiveWriter(basename, countArchiveModifier);
final IterateForCounts sortedPositionIterator = new IterateForCounts(countArchive, LOG);
sortedPositionIterator.setAlignmentReaderFactory(factory);
sortedPositionIterator.iterate(basename);
sortedPositionIterator.finishWriter();
}
private class IterateForCounts extends IterateSortedAlignmentsListImpl {
CountsWriterI writerI;
CountsArchiveWriter archiveWriter;
CountsWriterHelperI helperI;
private IterateForCounts(CountsArchiveWriter archiveWriter, Logger LOG) {
this.archiveWriter = archiveWriter;
}
int lastReferenceIndex = -1;
int lastPosition = -1;
@Override
public void processPositions(int referenceIndex, int position, DiscoverVariantPositionData positionBaseInfos) {
try {
if (referenceIndex != lastReferenceIndex) {
// System.out.println("Switching reference, last position was "+lastPosition);
if (writerI != null) {
finishWriter();
}
writerI = archiveWriter.newCountWriter(referenceIndex, getReferenceId(referenceIndex).toString());
helperI = new CountWriterHelper2(writerI);
lastReferenceIndex = referenceIndex;
lastPosition = -1;
}
if (position > lastPosition) {
helperI.appendCountAtPosition(positionBaseInfos.size(), position);
lastPosition = position;
}
} catch (IOException e) {
LOG.error("cannot return counts writer to archive", e);
}
}
public void finishWriter() throws IOException {
helperI.close();
archiveWriter.returnWriter(writerI);
writerI = null;
}
}
private ComputeCountInterface chooseAlgorithm(final WeightParameters weightParams, final WeightsInfo weights, ComputeCountInterface algo) {
if (weightParams.useWeights) {
if (!weightParams.adjustGcBias) {
// weights only:
algo = new ComputeWeightCount(weights);
} else {
// use weights to reweight with formula:
final FormulaWeightCount algo1 = new FormulaWeightCount(weights);
algo1.setFormulaChoice(FormulaWeightAnnotationCount.FormulaChoice.valueOf(weightParams.formulaChoice));
algo = algo1;
}
}
return algo;
}
/**
* Main method.
*
* @param args command line args.
* @throws com.martiansoftware.jsap.JSAPException error parsing
* @throws java.io.IOException error parsing or executing.
*/
public static void main(final String[] args) throws JSAPException, IOException {
new CompactAlignmentToCountsMode().configure(args).execute();
}
}
|