Examples with SeqGraph org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.SeqGraph used on opensource projects

Example 1 with SeqGraph

use of org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.SeqGraph in project gatk-protected by broadinstitute.

the class AssemblyResultSetUnitTest method trimmingData.

@DataProvider(name = "trimmingData")
public Iterator<Object[]> trimmingData() {
    final AssemblyRegion activeRegion = new AssemblyRegion(new SimpleInterval("1", 1000, 1100), 25, header);
    final int length = activeRegion.getExtendedSpan().size();
    // keep it prepoducible by fixing the seed to lucky 13.
    final RandomDNA rnd = new RandomDNA(13);
    final AssemblyRegionTestDataSet actd = new AssemblyRegionTestDataSet(10, new String(rnd.nextBases(length)), new String[] { "Civar:*1T*" }, new String[0], new byte[0], new byte[0], new byte[0]);
    final List<Haplotype> haplotypes = actd.haplotypeList();
    for (final Haplotype h : haplotypes) h.setGenomeLocation(activeRegion.getExtendedSpan());
    final ReadThreadingGraph rtg = new ReadThreadingGraph(10);
    for (final Haplotype h : haplotypes) rtg.addSequence("seq-" + Math.abs(h.hashCode()), h.getBases(), h.isReference());
    final SeqGraph seqGraph = rtg.toSequenceGraph();
    final AssemblyResult ar = new AssemblyResult(AssemblyResult.Status.ASSEMBLED_SOME_VARIATION, seqGraph, rtg);
    final Map<Haplotype, AssemblyResult> result = new HashMap<>();
    for (final Haplotype h : haplotypes) result.put(h, ar);
    return Collections.singleton(new Object[] { result, activeRegion }).iterator();
}

Example 2 with SeqGraph

use of org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.SeqGraph in project gatk by broadinstitute.

the class ReadThreadingGraphUnitTest method testDanglingHeads.

@Test(dataProvider = "DanglingHeads")
public void testDanglingHeads(final String ref, final String alt, final String cigar, final boolean shouldBeMerged) {
    final int kmerSize = 5;
    // create the graph and populate it
    final ReadThreadingGraph rtgraph = new ReadThreadingGraph(kmerSize);
    rtgraph.addSequence("ref", ref.getBytes(), true);
    final GATKRead read = ArtificialReadUtils.createArtificialRead(alt.getBytes(), Utils.dupBytes((byte) 30, alt.length()), alt.length() + "M");
    final SAMFileHeader header = ArtificialReadUtils.createArtificialSamHeader();
    rtgraph.addRead(read, header);
    rtgraph.setMaxMismatchesInDanglingHead(10);
    rtgraph.buildGraphIfNecessary();
    // confirm that we have just a single dangling head
    MultiDeBruijnVertex altSource = null;
    for (final MultiDeBruijnVertex v : rtgraph.vertexSet()) {
        if (rtgraph.isSource(v) && !rtgraph.isReferenceNode(v)) {
            Assert.assertTrue(altSource == null, "We found more than one non-reference source");
            altSource = v;
        }
    }
    Assert.assertTrue(altSource != null, "We did not find a non-reference source");
    // confirm that the SW alignment agrees with our expectations
    final ReadThreadingGraph.DanglingChainMergeHelper result = rtgraph.generateCigarAgainstUpwardsReferencePath(altSource, 0, 1);
    if (result == null) {
        Assert.assertFalse(shouldBeMerged);
        return;
    }
    Assert.assertTrue(cigar.equals(result.cigar.toString()), "SW generated cigar = " + result.cigar.toString());
    // confirm that the tail merging works as expected
    final int mergeResult = rtgraph.mergeDanglingHead(result);
    Assert.assertTrue(mergeResult > 0 || !shouldBeMerged);
    // confirm that we created the appropriate bubble in the graph only if expected
    rtgraph.cleanNonRefPaths();
    final SeqGraph seqGraph = rtgraph.toSequenceGraph();
    final List<KBestHaplotype> paths = new KBestHaplotypeFinder(seqGraph, seqGraph.getReferenceSourceVertex(), seqGraph.getReferenceSinkVertex());
    Assert.assertEquals(paths.size(), shouldBeMerged ? 2 : 1);
}

Example 3 with SeqGraph

use of org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.SeqGraph in project gatk by broadinstitute.

the class ReadThreadingAssemblerUnitTest method testMismatchInFirstKmer.

@Test
public void testMismatchInFirstKmer() {
    final TestAssembler assembler = new TestAssembler(3);
    final String ref = "ACAACTGA";
    final String alt = "AGCTGA";
    assembler.addSequence(ref.getBytes(), true);
    assembler.addSequence(alt.getBytes(), false);
    final SeqGraph graph = assembler.assemble();
    graph.simplifyGraph();
    graph.removeSingletonOrphanVertices();
    final Set<SeqVertex> sources = graph.getSources();
    final Set<SeqVertex> sinks = graph.getSinks();
    Assert.assertEquals(sources.size(), 1);
    Assert.assertEquals(sinks.size(), 1);
    Assert.assertNotNull(graph.getReferenceSourceVertex());
    Assert.assertNotNull(graph.getReferenceSinkVertex());
    final List<KBestHaplotype> paths = new KBestHaplotypeFinder(graph);
    Assert.assertEquals(paths.size(), 1);
}

Example 4 with SeqGraph

use of org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.SeqGraph in project gatk by broadinstitute.

the class ReadThreadingAssemblerUnitTest method assertLinearGraph.

private void assertLinearGraph(final TestAssembler assembler, final String seq) {
    final SeqGraph graph = assembler.assemble();
    graph.simplifyGraph();
    Assert.assertEquals(graph.vertexSet().size(), 1);
    Assert.assertEquals(graph.vertexSet().iterator().next().getSequenceString(), seq);
}

Example 5 with SeqGraph

use of org.broadinstitute.hellbender.tools.walkers.haplotypecaller.graphs.SeqGraph in project gatk-protected by broadinstitute.

the class ReadThreadingGraphUnitTest method testNsInReadsAreNotUsedForGraph.

@Test(enabled = !DEBUG)
public void testNsInReadsAreNotUsedForGraph() {
    final int length = 100;
    final byte[] ref = Utils.dupBytes((byte) 'A', length);
    final ReadThreadingGraph rtgraph = new ReadThreadingGraph(25);
    rtgraph.addSequence("ref", ref, true);
    // add reads with Ns at any position
    final SAMFileHeader header = ArtificialReadUtils.createArtificialSamHeader();
    for (int i = 0; i < length; i++) {
        final byte[] bases = ref.clone();
        bases[i] = 'N';
        final GATKRead read = ArtificialReadUtils.createArtificialRead(bases, Utils.dupBytes((byte) 30, length), length + "M");
        rtgraph.addRead(read, header);
    }
    rtgraph.buildGraphIfNecessary();
    final SeqGraph graph = rtgraph.toSequenceGraph();
    Assert.assertEquals(new KBestHaplotypeFinder(graph, graph.getReferenceSourceVertex(), graph.getReferenceSinkVertex()).size(), 1);
}