use of org.broadinstitute.hellbender.tools.exome.ModeledSegment in project gatk-protected by broadinstitute.
the class AlleleFractionSegmenterUnitTest method testSegmentation.
@Test
public void testSegmentation() {
final RandomGenerator rng = RandomGeneratorFactory.createRandomGenerator(new Random(563));
final List<Double> trueWeights = Arrays.asList(0.2, 0.5, 0.3);
final List<Double> trueMinorAlleleFractions = Arrays.asList(0.12, 0.32, 0.5);
final double trueMemoryLength = 1e5;
final AlleleFractionGlobalParameters trueParams = new AlleleFractionGlobalParameters(1.0, 0.01, 0.01);
final AlleleFractionHMM trueModel = new AlleleFractionHMM(trueMinorAlleleFractions, trueWeights, trueMemoryLength, AllelicPanelOfNormals.EMPTY_PON, trueParams);
// randomly set positions
final int chainLength = 10000;
final List<SimpleInterval> positions = CopyRatioSegmenterUnitTest.randomPositions("chr1", chainLength, rng, trueMemoryLength / 4);
final List<Integer> trueStates = trueModel.generateHiddenStateChain(positions);
final List<Double> truthMinorFractions = trueStates.stream().map(trueModel::getMinorAlleleFraction).collect(Collectors.toList());
final AllelicCountCollection counts = generateCounts(truthMinorFractions, positions, rng, trueParams);
final AlleleFractionSegmenter segmenter = new AlleleFractionSegmenter(10, counts, AllelicPanelOfNormals.EMPTY_PON);
final List<ModeledSegment> segments = segmenter.getModeledSegments();
final double[] segmentMinorFractions = segments.stream().flatMap(s -> Collections.nCopies((int) s.getTargetCount(), s.getSegmentMean()).stream()).mapToDouble(x -> x).toArray();
final double averageMinorFractionError = IntStream.range(0, truthMinorFractions.size()).mapToDouble(n -> Math.abs(segmentMinorFractions[n] - truthMinorFractions.get(n))).average().getAsDouble();
Assert.assertEquals(averageMinorFractionError, 0, 0.01);
}
use of org.broadinstitute.hellbender.tools.exome.ModeledSegment in project gatk-protected by broadinstitute.
the class CopyRatioSegmenterUnitTest method testChromosomesOnDifferentSegments.
@Test
public void testChromosomesOnDifferentSegments() {
final RandomGenerator rng = RandomGeneratorFactory.createRandomGenerator(new Random(563));
final double[] trueLog2CopyRatios = new double[] { -2.0, 0.0, 1.7 };
final double trueMemoryLength = 1e5;
final double trueStandardDeviation = 0.2;
// randomly set positions
final int chainLength = 100;
final List<SimpleInterval> positions = randomPositions("chr1", chainLength, rng, trueMemoryLength / 4);
positions.addAll(randomPositions("chr2", chainLength, rng, trueMemoryLength / 4));
positions.addAll(randomPositions("chr3", chainLength, rng, trueMemoryLength / 4));
//fix everything to the same state 2
final int trueState = 2;
final List<Double> data = new ArrayList<>();
for (int n = 0; n < positions.size(); n++) {
final double copyRatio = trueLog2CopyRatios[trueState];
final double observed = generateData(trueStandardDeviation, copyRatio, rng);
data.add(observed);
}
final List<Target> targets = positions.stream().map(Target::new).collect(Collectors.toList());
final ReadCountCollection rcc = new ReadCountCollection(targets, Arrays.asList("SAMPLE"), new Array2DRowRealMatrix(data.stream().mapToDouble(x -> x).toArray()));
final CopyRatioSegmenter segmenter = new CopyRatioSegmenter(10, rcc);
final List<ModeledSegment> segments = segmenter.getModeledSegments();
//check that each chromosome has at least one segment
final int numDifferentContigsInSegments = (int) segments.stream().map(ModeledSegment::getContig).distinct().count();
Assert.assertEquals(numDifferentContigsInSegments, 3);
}
use of org.broadinstitute.hellbender.tools.exome.ModeledSegment in project gatk-protected by broadinstitute.
the class CopyRatioSegmenterUnitTest method testSegmentation.
@Test
public void testSegmentation() {
final RandomGenerator rng = RandomGeneratorFactory.createRandomGenerator(new Random(563));
final List<Double> trueWeights = Arrays.asList(0.2, 0.5, 0.3);
final List<Double> trueLog2CopyRatios = Arrays.asList(-2.0, 0.0, 1.4);
final double trueMemoryLength = 1e5;
final double trueStandardDeviation = 0.2;
final CopyRatioHMM trueModel = new CopyRatioHMM(trueLog2CopyRatios, trueWeights, trueMemoryLength, trueStandardDeviation);
final int chainLength = 10000;
final List<SimpleInterval> positions = randomPositions("chr1", chainLength, rng, trueMemoryLength / 4);
final List<Integer> trueStates = trueModel.generateHiddenStateChain(positions);
final List<Double> trueLog2CopyRatioSequence = trueStates.stream().map(n -> trueLog2CopyRatios.get(n)).collect(Collectors.toList());
final List<Double> data = trueLog2CopyRatioSequence.stream().map(cr -> generateData(trueStandardDeviation, cr, rng)).collect(Collectors.toList());
final List<Target> targets = positions.stream().map(Target::new).collect(Collectors.toList());
final ReadCountCollection rcc = new ReadCountCollection(targets, Arrays.asList("SAMPLE"), new Array2DRowRealMatrix(data.stream().mapToDouble(x -> x).toArray()));
final CopyRatioSegmenter segmenter = new CopyRatioSegmenter(10, rcc);
final List<ModeledSegment> segments = segmenter.getModeledSegments();
final double[] segmentCopyRatios = segments.stream().flatMap(s -> Collections.nCopies((int) s.getTargetCount(), s.getSegmentMeanInLog2CRSpace()).stream()).mapToDouble(x -> x).toArray();
final double averageCopyRatioError = IntStream.range(0, trueLog2CopyRatioSequence.size()).mapToDouble(n -> Math.abs(segmentCopyRatios[n] - trueLog2CopyRatioSequence.get(n))).average().getAsDouble();
Assert.assertEquals(averageCopyRatioError, 0, 0.025);
}
use of org.broadinstitute.hellbender.tools.exome.ModeledSegment in project gatk-protected by broadinstitute.
the class PerformCopyRatioSegmentationIntegrationTest method testCommandLine.
// checks that segmentation output is created -- only the unit test checks correctness of results
@Test
public void testCommandLine() throws IOException {
final File tnCoverageFile = LOG2_TN_COVERAGE_FILE;
final File outputSegmentFile = createTempFile("segments", ".seg");
final int initialNumStates = 10;
final String[] arguments = { "-" + ExomeStandardArgumentDefinitions.TANGENT_NORMALIZED_COUNTS_FILE_SHORT_NAME, tnCoverageFile.getAbsolutePath(), "-" + PerformCopyRatioSegmentation.INITIAL_NUM_STATES_SHORT_NAME, Integer.toString(initialNumStates), "-" + ExomeStandardArgumentDefinitions.SEGMENT_FILE_SHORT_NAME, outputSegmentFile.getAbsolutePath() };
runCommandLine(arguments);
final List<ModeledSegment> segments = SegmentUtils.readModeledSegmentsFromSegmentFile(outputSegmentFile);
}
use of org.broadinstitute.hellbender.tools.exome.ModeledSegment in project gatk by broadinstitute.
the class PerformCopyRatioSegmentation method doWork.
@Override
public Object doWork() {
final String sampleName = ReadCountCollectionUtils.getSampleNameForCLIsFromReadCountsFile(new File(coverageFile));
final ReadCountCollection rcc;
try {
rcc = ReadCountCollectionUtils.parse(new File(coverageFile));
} catch (final IOException ex) {
throw new UserException.BadInput("could not read input file");
}
final CopyRatioSegmenter segmenter = new CopyRatioSegmenter(initialNumStates, rcc);
final List<ModeledSegment> segments = segmenter.getModeledSegments();
SegmentUtils.writeModeledSegmentFile(outputSegmentsFile, segments, sampleName, false);
return "SUCCESS";
}
Aggregations