Examples with ReadCountCollection org.broadinstitute.hellbender.tools.exome.ReadCountCollection used on opensource projects

Example 21 with ReadCountCollection

use of org.broadinstitute.hellbender.tools.exome.ReadCountCollection in project gatk-protected by broadinstitute.

the class TargetCoverageSexGenotypeCalculatorUnitTest method initSexGenotyper.

@BeforeClass
public static void initSexGenotyper() {
    final List<ContigGermlinePloidyAnnotation> contigPloidyAnnotsList;
    final ReadCountCollection readCounts;
    try {
        contigPloidyAnnotsList = ContigGermlinePloidyAnnotationTableReader.readContigGermlinePloidyAnnotationsFromFile(TEST_CONTIG_PLOIDY_ANNOTS_FILE);
        readCounts = ReadCountCollectionUtils.parse(TEST_RCC_FILE);
    } catch (final IOException ex) {
        throw new UserException.CouldNotReadInputFile("Could not read test resource files");
    }
    genotyper = new TargetCoverageSexGenotypeCalculator(readCounts, readCounts.targets(), contigPloidyAnnotsList, DEFAULT_MAPPING_ERROR_PROBABILITY);
}

Example 22 with ReadCountCollection

use of org.broadinstitute.hellbender.tools.exome.ReadCountCollection 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);
}

Example 23 with ReadCountCollection

use of org.broadinstitute.hellbender.tools.exome.ReadCountCollection in project gatk-protected by broadinstitute.

the class SparkGenomeReadCountsIntegrationTest method testSparkGenomeReadCounts.

@Test
public void testSparkGenomeReadCounts() throws IOException {
    final File outputFile = createTempFile(BAM_FILE.getName(), ".cov");
    final String[] arguments = { "--disableSequenceDictionaryValidation", "-" + StandardArgumentDefinitions.REFERENCE_SHORT_NAME, REFERENCE_FILE.getAbsolutePath(), "-" + StandardArgumentDefinitions.INPUT_SHORT_NAME, BAM_FILE.getAbsolutePath(), "-" + SparkGenomeReadCounts.OUTPUT_FILE_SHORT_NAME, outputFile.getAbsolutePath(), "-" + SparkGenomeReadCounts.BINSIZE_SHORT_NAME, "10000" };
    runCommandLine(arguments);
    Assert.assertTrue(outputFile.exists());
    Assert.assertTrue(outputFile.length() > 0);
    final ReadCountCollection coverage = ReadCountCollectionUtils.parse(outputFile);
    final File targetsFile = new File(outputFile.getAbsolutePath() + ".targets.tsv");
    Assert.assertTrue(targetsFile.exists());
    Assert.assertTrue(targetsFile.length() > 0);
    final List<Target> targets = TargetTableReader.readTargetFile(targetsFile);
    Assert.assertEquals(targets.size(), 8);
    Assert.assertEquals(targets.get(1).getEnd(), 16000);
    Assert.assertEquals(targets.get(5).getName(), "target_3_10001_16000");
    Assert.assertEquals(coverage.targets().size(), targets.size());
}

Example 24 with ReadCountCollection

use of org.broadinstitute.hellbender.tools.exome.ReadCountCollection in project gatk-protected by broadinstitute.

the class SparkGenomeReadCountsIntegrationTest method testSparkGenomeReadCountsSmallBins.

@Test
public void testSparkGenomeReadCountsSmallBins() throws IOException {
    final File outputFile = createTempFile(BAM_FILE.getName(), ".cov");
    final String[] arguments = { "--disableSequenceDictionaryValidation", "-" + StandardArgumentDefinitions.REFERENCE_SHORT_NAME, REFERENCE_FILE.getAbsolutePath(), "-" + StandardArgumentDefinitions.INPUT_SHORT_NAME, BAM_FILE.getAbsolutePath(), "-" + SparkGenomeReadCounts.OUTPUT_FILE_SHORT_NAME, outputFile.getAbsolutePath(), "-" + SparkGenomeReadCounts.BINSIZE_SHORT_NAME, "2000" };
    runCommandLine(arguments);
    Assert.assertTrue(outputFile.exists());
    Assert.assertTrue(outputFile.length() > 0);
    // Proportional Coverage
    final ReadCountCollection proportionalCoverage = ReadCountCollectionUtils.parse(outputFile);
    Assert.assertTrue(proportionalCoverage.records().stream().anyMatch(t -> Math.abs(t.getDouble(0)) > 1e-10));
    // The reads are all in three bins of contig 3 with values {.5, .25, .25}
    Assert.assertTrue(proportionalCoverage.records().stream().filter(t -> t.getContig().equals("3")).anyMatch(t -> Math.abs(t.getDouble(0)) > .2));
    Assert.assertTrue(Math.abs(proportionalCoverage.records().stream().filter(t -> t.getContig().equals("3")).mapToDouble(t -> t.getDouble(0)).sum() - 1.0) < 1e-10);
    // raw coverage
    final ReadCountCollection coverage = ReadCountCollectionUtils.parse(new File(outputFile.getAbsolutePath() + SparkGenomeReadCounts.RAW_COV_OUTPUT_EXTENSION));
    Assert.assertTrue(coverage.records().stream().anyMatch(t -> Math.abs(t.getDouble(0)) > 1e-10));
    // The reads are all in three bins of contig 3 with values
    Assert.assertEquals(coverage.records().stream().filter(t -> t.getContig().equals("3")).filter(t -> Math.abs(t.getDouble(0)) >= 1).count(), 3);
    final File targetsFile = new File(outputFile.getAbsolutePath() + ".targets.tsv");
    Assert.assertTrue(targetsFile.exists());
    Assert.assertTrue(targetsFile.length() > 0);
    final List<Target> targets = TargetTableReader.readTargetFile(targetsFile);
    // 4 is the number of contigs in the fasta file
    Assert.assertEquals(targets.size(), 16000 / 2000 * 4);
    Assert.assertEquals(targets.get(1).getEnd(), 4000);
    Assert.assertEquals(targets.get(2).getName(), "target_1_4001_6000");
    Assert.assertEquals(targets.get(8).getName(), "target_2_1_2000");
    Assert.assertEquals(targets.get(17).getName(), "target_3_2001_4000");
    Assert.assertEquals(proportionalCoverage.targets().size(), targets.size());
}

Example 25 with ReadCountCollection

use of org.broadinstitute.hellbender.tools.exome.ReadCountCollection in project gatk-protected by broadinstitute.

the class SparkGenomeReadCountsIntegrationTest method testSparkGenomeReadCountsTwoIntervals.

@Test
public void testSparkGenomeReadCountsTwoIntervals() {
    final File outputFile = createTempFile(BAM_FILE.getName(), ".cov");
    final String[] arguments = { "--disableSequenceDictionaryValidation", "-" + StandardArgumentDefinitions.REFERENCE_SHORT_NAME, REFERENCE_FILE.getAbsolutePath(), "-" + StandardArgumentDefinitions.INPUT_SHORT_NAME, BAM_FILE.getAbsolutePath(), "-" + SparkGenomeReadCounts.OUTPUT_FILE_SHORT_NAME, outputFile.getAbsolutePath(), "-" + SparkGenomeReadCounts.BINSIZE_SHORT_NAME, "10000", "-L", "1", "-L", "2" };
    runCommandLine(arguments);
    final ReadCountCollection proportionalCoverage = loadReadCountCollection(outputFile);
    Assert.assertTrue(proportionalCoverage.records().stream().anyMatch(t -> t.getContig().equals("1")));
    Assert.assertTrue(proportionalCoverage.records().stream().anyMatch(t -> t.getContig().equals("2")));
    Assert.assertTrue(proportionalCoverage.records().stream().noneMatch(t -> t.getContig().equals("3") || t.getContig().equals("4")));
    // raw coverage
    final ReadCountCollection rawCoverage = loadReadCountCollection(new File(outputFile.getAbsolutePath() + SparkGenomeReadCounts.RAW_COV_OUTPUT_EXTENSION));
    Assert.assertTrue(rawCoverage.records().stream().anyMatch(t -> t.getContig().equals("1")));
    Assert.assertTrue(rawCoverage.records().stream().anyMatch(t -> t.getContig().equals("2")));
    Assert.assertTrue(rawCoverage.records().stream().noneMatch(t -> t.getContig().equals("3") || t.getContig().equals("4")));
    final File targetsFile = new File(outputFile.getAbsolutePath() + ".targets.tsv");
    final List<Target> targets = TargetTableReader.readTargetFile(targetsFile);
    Assert.assertTrue(targets.stream().allMatch(t -> (t.getContig().equals("1")) || (t.getContig().equals("2"))));
}