Merge pull request #679 from broadinstitute/eb_better_tail_merging_PT74222522
Improved tail merging: now tails can be merged to branches that are not ...
This commit is contained in:
Ryan Poplin
commit
193e389b41
|
|
@ -273,6 +273,10 @@ public class HaplotypeCaller extends ActiveRegionWalker<List<VariantContext>, In
|
|||
@Argument(fullName="dontIncreaseKmerSizesForCycles", shortName="dontIncreaseKmerSizesForCycles", doc="Should we disable the iterating over kmer sizes when graph cycles are detected?", required = false)
|
||||
protected boolean dontIncreaseKmerSizesForCycles = false;
|
||||
|
||||
@Advanced
|
||||
@Argument(fullName="allowNonUniqueKmersInRef", shortName="allowNonUniqueKmersInRef", doc="Should we allow graphs which have non-unique kmers in the reference?", required = false)
|
||||
protected boolean allowNonUniqueKmersInRef = false;
|
||||
|
||||
@Advanced
|
||||
@Argument(fullName="numPruningSamples", shortName="numPruningSamples", doc="The number of samples that must pass the minPuning factor in order for the path to be kept", required = false)
|
||||
protected int numPruningSamples = 1;
|
||||
|
|
@ -615,7 +619,7 @@ public class HaplotypeCaller extends ActiveRegionWalker<List<VariantContext>, In
|
|||
}
|
||||
|
||||
// create and setup the assembler
|
||||
assemblyEngine = new ReadThreadingAssembler(maxNumHaplotypesInPopulation, kmerSizes, dontIncreaseKmerSizesForCycles, numPruningSamples);
|
||||
assemblyEngine = new ReadThreadingAssembler(maxNumHaplotypesInPopulation, kmerSizes, dontIncreaseKmerSizesForCycles, allowNonUniqueKmersInRef, numPruningSamples);
|
||||
|
||||
assemblyEngine.setErrorCorrectKmers(errorCorrectKmers);
|
||||
assemblyEngine.setPruneFactor(MIN_PRUNE_FACTOR);
|
||||
|
|
|
|||
|
|
@ -260,15 +260,39 @@ public class BaseGraph<V extends BaseVertex, E extends BaseEdge> extends Default
|
|||
/**
|
||||
* Traverse the graph and get the next reference vertex if it exists
|
||||
* @param v the current vertex, can be null
|
||||
* @return the next reference vertex if it exists
|
||||
* @return the next reference vertex if it exists, otherwise null
|
||||
*/
|
||||
public V getNextReferenceVertex( final V v ) {
|
||||
return getNextReferenceVertex(v, false, Collections.<MultiSampleEdge>emptyList());
|
||||
}
|
||||
|
||||
/**
|
||||
* Traverse the graph and get the next reference vertex if it exists
|
||||
* @param v the current vertex, can be null
|
||||
* @param allowNonRefPaths if true, allow sub-paths that are non-reference if there is only a single outgoing edge
|
||||
* @param blacklistedEdges edges to ignore in the traversal down; useful to exclude the non-reference dangling paths
|
||||
* @return the next vertex (but not necessarily on the reference path if allowNonRefPaths is true) if it exists, otherwise null
|
||||
*/
|
||||
public V getNextReferenceVertex( final V v, final boolean allowNonRefPaths, final Collection<MultiSampleEdge> blacklistedEdges ) {
|
||||
if( v == null ) { return null; }
|
||||
for( final E edgeToTest : outgoingEdgesOf(v) ) {
|
||||
|
||||
// variable must be mutable because outgoingEdgesOf is an immutable collection
|
||||
Set<E> edges = outgoingEdgesOf(v);
|
||||
|
||||
for( final E edgeToTest : edges ) {
|
||||
if( edgeToTest.isRef() ) {
|
||||
return getEdgeTarget(edgeToTest);
|
||||
}
|
||||
}
|
||||
|
||||
// if we got here, then we aren't on a reference path
|
||||
if ( allowNonRefPaths ) {
|
||||
edges = new HashSet<>(edges); // edges was immutable
|
||||
edges.removeAll(blacklistedEdges);
|
||||
if ( edges.size() == 1 )
|
||||
return getEdgeTarget(edges.iterator().next());
|
||||
}
|
||||
|
||||
return null;
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -315,13 +315,13 @@ public abstract class DanglingChainMergingGraph extends BaseGraph<MultiDeBruijnV
|
|||
*/
|
||||
protected DanglingChainMergeHelper generateCigarAgainstDownwardsReferencePath(final MultiDeBruijnVertex vertex, final int pruneFactor) {
|
||||
|
||||
// find the lowest common ancestor path between vertex and the reference sink if available
|
||||
final List<MultiDeBruijnVertex> altPath = findPathUpwardsToLowestCommonAncestorOfReference(vertex, pruneFactor);
|
||||
// find the lowest common ancestor path between this vertex and the diverging master path if available
|
||||
final List<MultiDeBruijnVertex> altPath = findPathUpwardsToLowestCommonAncestor(vertex, pruneFactor);
|
||||
if ( altPath == null || isRefSource(altPath.get(0)) || altPath.size() < MIN_DANGLING_TAIL_LENGTH )
|
||||
return null;
|
||||
|
||||
// now get the reference path from the LCA
|
||||
final List<MultiDeBruijnVertex> refPath = getReferencePath(altPath.get(0), TraversalDirection.downwards);
|
||||
final List<MultiDeBruijnVertex> refPath = getReferencePath(altPath.get(0), TraversalDirection.downwards, Arrays.asList(incomingEdgeOf(altPath.get(1))));
|
||||
|
||||
// create the Smith-Waterman strings to use
|
||||
final byte[] refBases = getBasesForPath(refPath, false);
|
||||
|
|
@ -348,7 +348,7 @@ public abstract class DanglingChainMergingGraph extends BaseGraph<MultiDeBruijnV
|
|||
return null;
|
||||
|
||||
// now get the reference path from the LCA
|
||||
final List<MultiDeBruijnVertex> refPath = getReferencePath(altPath.get(0), TraversalDirection.upwards);
|
||||
final List<MultiDeBruijnVertex> refPath = getReferencePath(altPath.get(0), TraversalDirection.upwards, Collections.<MultiSampleEdge>emptyList());
|
||||
|
||||
// create the Smith-Waterman strings to use
|
||||
final byte[] refBases = getBasesForPath(refPath, true);
|
||||
|
|
@ -360,19 +360,19 @@ public abstract class DanglingChainMergingGraph extends BaseGraph<MultiDeBruijnV
|
|||
}
|
||||
|
||||
/**
|
||||
* Finds the path upwards in the graph from this vertex to the reference sequence, including the lowest common ancestor vertex.
|
||||
* Finds the path upwards in the graph from this vertex to the first diverging node, including that (lowest common ancestor) vertex.
|
||||
* Note that nodes are excluded if their pruning weight is less than the pruning factor.
|
||||
*
|
||||
* @param vertex the original vertex
|
||||
* @param pruneFactor the prune factor to use in ignoring chain pieces
|
||||
* @return the path if it can be determined or null if this vertex either doesn't merge onto the reference path or
|
||||
* @return the path if it can be determined or null if this vertex either doesn't merge onto another path or
|
||||
* has an ancestor with multiple incoming edges before hitting the reference path
|
||||
*/
|
||||
protected List<MultiDeBruijnVertex> findPathUpwardsToLowestCommonAncestorOfReference(final MultiDeBruijnVertex vertex, final int pruneFactor) {
|
||||
protected List<MultiDeBruijnVertex> findPathUpwardsToLowestCommonAncestor(final MultiDeBruijnVertex vertex, final int pruneFactor) {
|
||||
final LinkedList<MultiDeBruijnVertex> path = new LinkedList<>();
|
||||
|
||||
MultiDeBruijnVertex v = vertex;
|
||||
while ( ! isReferenceNode(v) && inDegreeOf(v) == 1 ) {
|
||||
while ( inDegreeOf(v) == 1 && outDegreeOf(v) < 2 ) {
|
||||
final MultiSampleEdge edge = incomingEdgeOf(v);
|
||||
// if it has too low a weight, don't use it (or previous vertexes) for the path
|
||||
if ( edge.getPruningMultiplicity() < pruneFactor )
|
||||
|
|
@ -384,7 +384,7 @@ public abstract class DanglingChainMergingGraph extends BaseGraph<MultiDeBruijnV
|
|||
}
|
||||
path.addFirst(v);
|
||||
|
||||
return isReferenceNode(v) ? path : null;
|
||||
return outDegreeOf(v) > 1 ? path : null;
|
||||
}
|
||||
|
||||
/**
|
||||
|
|
@ -426,17 +426,19 @@ public abstract class DanglingChainMergingGraph extends BaseGraph<MultiDeBruijnV
|
|||
*
|
||||
* @param start the reference vertex to start from
|
||||
* @param direction describes which direction to move in the graph (i.e. down to the reference sink or up to the source)
|
||||
* @param blacklistedEdges edges to ignore in the traversal down; useful to exclude the non-reference dangling paths
|
||||
* @return the path (non-null, non-empty)
|
||||
*/
|
||||
protected List<MultiDeBruijnVertex> getReferencePath(final MultiDeBruijnVertex start, final TraversalDirection direction) {
|
||||
if ( ! isReferenceNode(start) ) throw new IllegalArgumentException("Cannot construct the reference path from a vertex that is not on that path");
|
||||
protected List<MultiDeBruijnVertex> getReferencePath(final MultiDeBruijnVertex start,
|
||||
final TraversalDirection direction,
|
||||
final Collection<MultiSampleEdge> blacklistedEdges) {
|
||||
|
||||
final List<MultiDeBruijnVertex> path = new ArrayList<>();
|
||||
|
||||
MultiDeBruijnVertex v = start;
|
||||
while ( v != null ) {
|
||||
path.add(v);
|
||||
v = (direction == TraversalDirection.downwards ? getNextReferenceVertex(v) : getPrevReferenceVertex(v));
|
||||
v = (direction == TraversalDirection.downwards ? getNextReferenceVertex(v, true, blacklistedEdges) : getPrevReferenceVertex(v));
|
||||
}
|
||||
|
||||
return path;
|
||||
|
|
|
|||
|
|
@ -72,6 +72,7 @@ public class ReadThreadingAssembler extends LocalAssemblyEngine {
|
|||
private final int maxAllowedPathsForReadThreadingAssembler;
|
||||
|
||||
private final boolean dontIncreaseKmerSizesForCycles;
|
||||
private final boolean allowNonUniqueKmersInRef;
|
||||
private final int numPruningSamples;
|
||||
protected boolean removePathsNotConnectedToRef = true;
|
||||
private boolean justReturnRawGraph = false;
|
||||
|
|
@ -81,16 +82,17 @@ public class ReadThreadingAssembler extends LocalAssemblyEngine {
|
|||
this(DEFAULT_NUM_PATHS_PER_GRAPH, Arrays.asList(25));
|
||||
}
|
||||
|
||||
public ReadThreadingAssembler(final int maxAllowedPathsForReadThreadingAssembler, final List<Integer> kmerSizes, final boolean dontIncreaseKmerSizesForCycles, final int numPruningSamples) {
|
||||
public ReadThreadingAssembler(final int maxAllowedPathsForReadThreadingAssembler, final List<Integer> kmerSizes, final boolean dontIncreaseKmerSizesForCycles, final boolean allowNonUniqueKmersInRef, final int numPruningSamples) {
|
||||
super(maxAllowedPathsForReadThreadingAssembler);
|
||||
this.kmerSizes = kmerSizes;
|
||||
this.maxAllowedPathsForReadThreadingAssembler = maxAllowedPathsForReadThreadingAssembler;
|
||||
this.dontIncreaseKmerSizesForCycles = dontIncreaseKmerSizesForCycles;
|
||||
this.allowNonUniqueKmersInRef = allowNonUniqueKmersInRef;
|
||||
this.numPruningSamples = numPruningSamples;
|
||||
}
|
||||
|
||||
public ReadThreadingAssembler(final int maxAllowedPathsForReadThreadingAssembler, final List<Integer> kmerSizes) {
|
||||
this(maxAllowedPathsForReadThreadingAssembler, kmerSizes, true, 1);
|
||||
protected ReadThreadingAssembler(final int maxAllowedPathsForReadThreadingAssembler, final List<Integer> kmerSizes) {
|
||||
this(maxAllowedPathsForReadThreadingAssembler, kmerSizes, true, true, 1);
|
||||
}
|
||||
|
||||
/** for testing purposes */
|
||||
|
|
@ -109,7 +111,7 @@ public class ReadThreadingAssembler extends LocalAssemblyEngine {
|
|||
|
||||
// first, try using the requested kmer sizes
|
||||
for ( final int kmerSize : kmerSizes ) {
|
||||
addResult(results, createGraph(reads, refHaplotype, kmerSize, givenHaplotypes, dontIncreaseKmerSizesForCycles));
|
||||
addResult(results, createGraph(reads, refHaplotype, kmerSize, givenHaplotypes, dontIncreaseKmerSizesForCycles, allowNonUniqueKmersInRef));
|
||||
}
|
||||
|
||||
// if none of those worked, iterate over larger sizes if allowed to do so
|
||||
|
|
@ -118,7 +120,8 @@ public class ReadThreadingAssembler extends LocalAssemblyEngine {
|
|||
int numIterations = 1;
|
||||
while ( results.isEmpty() && numIterations <= MAX_KMER_ITERATIONS_TO_ATTEMPT ) {
|
||||
// on the last attempt we will allow low complexity graphs
|
||||
addResult(results, createGraph(reads, refHaplotype, kmerSize, givenHaplotypes, numIterations == MAX_KMER_ITERATIONS_TO_ATTEMPT));
|
||||
final boolean lastAttempt = numIterations == MAX_KMER_ITERATIONS_TO_ATTEMPT;
|
||||
addResult(results, createGraph(reads, refHaplotype, kmerSize, givenHaplotypes, lastAttempt, lastAttempt));
|
||||
kmerSize += KMER_SIZE_ITERATION_INCREASE;
|
||||
numIterations++;
|
||||
}
|
||||
|
|
@ -135,18 +138,25 @@ public class ReadThreadingAssembler extends LocalAssemblyEngine {
|
|||
* @param kmerSize kmer size
|
||||
* @param activeAlleleHaplotypes the GGA haplotypes to inject into the graph
|
||||
* @param allowLowComplexityGraphs if true, do not check for low-complexity graphs
|
||||
* @param allowNonUniqueKmersInRef if true, do not fail if the reference has non-unique kmers
|
||||
* @return sequence graph or null if one could not be created (e.g. because it contains cycles or too many paths or is low complexity)
|
||||
*/
|
||||
protected AssemblyResult createGraph(final List<GATKSAMRecord> reads,
|
||||
final Haplotype refHaplotype,
|
||||
final int kmerSize,
|
||||
final List<Haplotype> activeAlleleHaplotypes,
|
||||
final boolean allowLowComplexityGraphs) {
|
||||
final boolean allowLowComplexityGraphs,
|
||||
final boolean allowNonUniqueKmersInRef) {
|
||||
if ( refHaplotype.length() < kmerSize ) {
|
||||
// happens in cases where the assembled region is just too small
|
||||
return new AssemblyResult(AssemblyResult.Status.FAILED, null);
|
||||
}
|
||||
|
||||
if ( !allowNonUniqueKmersInRef && !ReadThreadingGraph.determineNonUniqueKmers(new SequenceForKmers("ref", refHaplotype.getBases(), 0, refHaplotype.getBases().length, 1, true), kmerSize).isEmpty() ) {
|
||||
if ( debug ) logger.info("Not using kmer size of " + kmerSize + " in read threading assembler because reference contains non-unique kmers");
|
||||
return null;
|
||||
}
|
||||
|
||||
final ReadThreadingGraph rtgraph = new ReadThreadingGraph(kmerSize, debugGraphTransformations, minBaseQualityToUseInAssembly, numPruningSamples);
|
||||
|
||||
rtgraph.setThreadingStartOnlyAtExistingVertex(!recoverDanglingHeads);
|
||||
|
|
|
|||
|
|
@ -435,7 +435,7 @@ public class ReadThreadingGraph extends DanglingChainMergingGraph implements Kme
|
|||
* @param kmerSize the size of the kmers
|
||||
* @return a non-null collection of non-unique kmers in sequence
|
||||
*/
|
||||
private Collection<Kmer> determineNonUniqueKmers(final SequenceForKmers seqForKmers, final int kmerSize) {
|
||||
static protected Collection<Kmer> determineNonUniqueKmers(final SequenceForKmers seqForKmers, final int kmerSize) {
|
||||
// count up occurrences of kmers within each read
|
||||
final KMerCounter counter = new KMerCounter(kmerSize);
|
||||
final int stopPosition = seqForKmers.stop - kmerSize;
|
||||
|
|
|
|||
|
|
@ -94,7 +94,7 @@ public class HaplotypeCallerComplexAndSymbolicVariantsIntegrationTest extends Wa
|
|||
@Test
|
||||
public void testHaplotypeCallerMultiSampleGGAMultiAllelic() {
|
||||
HCTestComplexGGA(NA12878_CHR20_BAM, "-L 20:133041-133161 -L 20:300207-300337",
|
||||
"b63bf0f2848f28e52fd5c7621bbd27c7");
|
||||
"3c02e454fe6d0e296effd99fbd3cba4c");
|
||||
}
|
||||
|
||||
private void HCTestComplexConsensusMode(String bam, String args, String md5) {
|
||||
|
|
@ -106,7 +106,7 @@ public class HaplotypeCallerComplexAndSymbolicVariantsIntegrationTest extends Wa
|
|||
@Test
|
||||
public void testHaplotypeCallerMultiSampleConsensusModeComplex() {
|
||||
HCTestComplexGGA(NA12878_CHR20_BAM, "-L 20:119673-119823 -L 20:121408-121538 -L 20:133041-133161 -L 20:300207-300337",
|
||||
"972a726b0ba476f6215bf162439db2cd");
|
||||
"f0560ba2ca20e5202181d0e57b1ee8cf");
|
||||
}
|
||||
|
||||
}
|
||||
|
|
|
|||
|
|
@ -67,9 +67,9 @@ public class HaplotypeCallerGVCFIntegrationTest extends WalkerTest {
|
|||
final String WExIntervals = "-L 20:10,000,000-10,100,000 -isr INTERSECTION -L " + hg19Chr20Intervals;
|
||||
|
||||
// this functionality can be adapted to provide input data for whatever you might want in your data
|
||||
tests.add(new Object[]{NA12878_PCRFREE, ReferenceConfidenceMode.NONE, PCRFreeIntervals, "662481d96a58c5475dc4752466a8d3b2"});
|
||||
tests.add(new Object[]{NA12878_PCRFREE, ReferenceConfidenceMode.BP_RESOLUTION, PCRFreeIntervals, "d9f7fd119eec40b4610cb4aae1cbed75"});
|
||||
tests.add(new Object[]{NA12878_PCRFREE, ReferenceConfidenceMode.GVCF, PCRFreeIntervals, "cbf6d876045051a68aca784491cca6cf"});
|
||||
tests.add(new Object[]{NA12878_PCRFREE, ReferenceConfidenceMode.NONE, PCRFreeIntervals, "f321fadf3d6b4608536fba1015e9693a"});
|
||||
tests.add(new Object[]{NA12878_PCRFREE, ReferenceConfidenceMode.BP_RESOLUTION, PCRFreeIntervals, "8d5626262b255ce6d4a53394ea8cd30f"});
|
||||
tests.add(new Object[]{NA12878_PCRFREE, ReferenceConfidenceMode.GVCF, PCRFreeIntervals, "126c94d06e00c67380051c4924054841"});
|
||||
tests.add(new Object[]{NA12878_WEx, ReferenceConfidenceMode.NONE, WExIntervals, "d5c07fa3edca496a84fd17cecad06230"});
|
||||
tests.add(new Object[]{NA12878_WEx, ReferenceConfidenceMode.BP_RESOLUTION, WExIntervals, "330685c734e277d70a44637de85ad54d"});
|
||||
tests.add(new Object[]{NA12878_WEx, ReferenceConfidenceMode.GVCF, WExIntervals, "e6ec90da053a612f0c615e221eb34baa"});
|
||||
|
|
@ -137,7 +137,7 @@ public class HaplotypeCallerGVCFIntegrationTest extends WalkerTest {
|
|||
public void testWrongGVCFNonVariantRecordOrderBugFix() {
|
||||
final String commandLine = String.format("-T HaplotypeCaller --pcr_indel_model NONE -R %s -I %s -L %s -ERC GVCF --no_cmdline_in_header -variant_index_type %s -variant_index_parameter %d",
|
||||
b37KGReference, WRONG_GVCF_RECORD_ORDER_BUGFIX_BAM, WRONG_GVCF_RECORD_ORDER_BUGFIX_INTERVALS, HaplotypeCaller.OPTIMAL_GVCF_INDEX_TYPE, HaplotypeCaller.OPTIMAL_GVCF_INDEX_PARAMETER);
|
||||
final WalkerTestSpec spec = new WalkerTestSpec(commandLine + " -o %s", Arrays.asList("5fbc88cf1136212afac1fd0b7e0e8ce8"));
|
||||
final WalkerTestSpec spec = new WalkerTestSpec(commandLine + " -o %s", Arrays.asList("9fecd24420579f321157e5605b6cd7f2"));
|
||||
spec.disableShadowBCF();
|
||||
executeTest("testMissingGVCFIndexingStrategyException", spec);
|
||||
}
|
||||
|
|
@ -149,7 +149,7 @@ public class HaplotypeCallerGVCFIntegrationTest extends WalkerTest {
|
|||
public void testNoCallGVCFMissingPLsBugFix() {
|
||||
final String commandLine = String.format("-T HaplotypeCaller --pcr_indel_model NONE -R %s -I %s -L %s -ERC GVCF --no_cmdline_in_header -variant_index_type %s -variant_index_parameter %d",
|
||||
b37KGReference, NOCALL_GVCF_BUGFIX_BAM, NOCALL_GVCF_BUGFIX_INTERVALS, HaplotypeCaller.OPTIMAL_GVCF_INDEX_TYPE, HaplotypeCaller.OPTIMAL_GVCF_INDEX_PARAMETER);
|
||||
final WalkerTestSpec spec = new WalkerTestSpec(commandLine + " -o %s", Arrays.asList("1b77bc92962fa10b5eec86fe9400c528"));
|
||||
final WalkerTestSpec spec = new WalkerTestSpec(commandLine + " -o %s", Arrays.asList("95bae0b4a8fe9bfcf71762704b22a76e"));
|
||||
spec.disableShadowBCF();
|
||||
executeTest("testNoCallGVCFMissingPLsBugFix", spec);
|
||||
}
|
||||
|
|
|
|||
|
|
@ -84,7 +84,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
|
||||
@Test
|
||||
public void testHaplotypeCallerMultiSample() {
|
||||
HCTest(CEUTRIO_BAM, "", "b66f8d08e0b94ddc52bd06dadbb9b299");
|
||||
HCTest(CEUTRIO_BAM, "", "60e2f0c3ce33a05c060035d86bc79543");
|
||||
}
|
||||
|
||||
@Test
|
||||
|
|
@ -94,17 +94,17 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
|
||||
@Test
|
||||
public void testHaplotypeCallerMinBaseQuality() {
|
||||
HCTest(NA12878_BAM, "-mbq 15", "f753c9e284eafb1424b7f9d88193fdee");
|
||||
HCTest(NA12878_BAM, "-mbq 15", "d063c0e5af1fd413be0500609ae36d46");
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testHaplotypeCallerGraphBasedSingleSample() {
|
||||
HCTest(NA12878_BAM, "-likelihoodEngine GraphBased", "5a49b5b98247070e8de637a706b02db9");
|
||||
HCTest(NA12878_BAM, "-likelihoodEngine GraphBased", "07910f50710349eacd2560452fac3e8d");
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testHaplotypeCallerGraphBasedMultiSample() {
|
||||
HCTest(CEUTRIO_BAM, "-likelihoodEngine GraphBased", "f200260127d0f63cc61ce5f2287bd5a0");
|
||||
HCTest(CEUTRIO_BAM, "-likelihoodEngine GraphBased", "c5ef449a46b80b69dde87aa52041fe50");
|
||||
}
|
||||
|
||||
@Test(enabled = false) // can't annotate the rsID's yet
|
||||
|
|
@ -115,7 +115,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
@Test
|
||||
public void testHaplotypeCallerMultiSampleGGA() {
|
||||
HCTest(CEUTRIO_BAM, "--max_alternate_alleles 3 -gt_mode GENOTYPE_GIVEN_ALLELES -alleles " + validationDataLocation + "combined.phase1.chr20.raw.indels.sites.vcf",
|
||||
"4b1271aaa70f46ed6c987cdf4610b7f0");
|
||||
"b61e0bdf0e3180cb4f5abd3491b05aa6");
|
||||
}
|
||||
|
||||
@Test
|
||||
|
|
@ -131,7 +131,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
|
||||
@Test
|
||||
public void testHaplotypeCallerSingleSampleIndelQualityScores() {
|
||||
HCTestIndelQualityScores(NA12878_RECALIBRATED_BAM, "", "575850d9a8195000a2ef747f838aaf77");
|
||||
HCTestIndelQualityScores(NA12878_RECALIBRATED_BAM, "", "ec90508248b82800eb2596348a7cf85b");
|
||||
}
|
||||
|
||||
private void HCTestNearbySmallIntervals(String bam, String args, String md5) {
|
||||
|
|
@ -199,7 +199,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
|
||||
@Test
|
||||
public void HCTestDanglingTailMergingForDeletions() throws IOException {
|
||||
final String base = String.format("-T HaplotypeCaller --disableDithering --pcr_indel_model NONE -R %s -I %s", REF, NA12878_BAM) + " --no_cmdline_in_header -o %s -L 20:10130740-10130800";
|
||||
final String base = String.format("-T HaplotypeCaller --disableDithering --pcr_indel_model NONE -R %s -I %s", REF, NA12878_BAM) + " --no_cmdline_in_header -o %s -L 20:10130740-10130800 --allowNonUniqueKmersInRef";
|
||||
final WalkerTestSpec spec = new WalkerTestSpec(base, 1, Arrays.asList(""));
|
||||
final File outputVCF = executeTest("HCTestDanglingTailMergingForDeletions", spec).getFirst().get(0);
|
||||
|
||||
|
|
@ -227,7 +227,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
public void HCTestDBSNPAnnotationWGS() {
|
||||
WalkerTest.WalkerTestSpec spec = new WalkerTest.WalkerTestSpec(
|
||||
"-T HaplotypeCaller --disableDithering --pcr_indel_model NONE -R " + b37KGReference + " --no_cmdline_in_header -I " + NA12878_PCRFREE + " -o %s -L 20:10,000,000-10,100,000 -D " + b37dbSNP132, 1,
|
||||
Arrays.asList("cf314806fcb0847e2ed99c7013e5d55d"));
|
||||
Arrays.asList("10a7a3d49cb6cb172ae8404996c524df"));
|
||||
executeTest("HC calling with dbSNP ID annotation on WGS intervals", spec);
|
||||
}
|
||||
|
||||
|
|
@ -244,7 +244,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
public void HCTestDBSNPAnnotationWGSGraphBased() {
|
||||
WalkerTest.WalkerTestSpec spec = new WalkerTest.WalkerTestSpec(
|
||||
"-T HaplotypeCaller -likelihoodEngine GraphBased --disableDithering --pcr_indel_model NONE -R " + b37KGReference + " --no_cmdline_in_header -I " + NA12878_PCRFREE + " -o %s -L 20:10,000,000-10,100,000 -D " + b37dbSNP132, 1,
|
||||
Arrays.asList("0e68c82163d4fee30b4135d4ebd1a3ba"));
|
||||
Arrays.asList("14384fe06359bf35e11d3802217e1a23"));
|
||||
executeTest("HC calling with dbSNP ID annotation on WGS intervals", spec);
|
||||
}
|
||||
|
||||
|
|
@ -276,7 +276,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
public void HCTestAggressivePcrIndelModelWGS() {
|
||||
WalkerTest.WalkerTestSpec spec = new WalkerTest.WalkerTestSpec(
|
||||
"-T HaplotypeCaller --disableDithering --pcr_indel_model AGGRESSIVE -R " + b37KGReference + " --no_cmdline_in_header -I " + NA12878_BAM + " -o %s -L 20:10,000,000-10,300,000", 1,
|
||||
Arrays.asList("be7bf9e8f78d6a9ee569b750054991ac"));
|
||||
Arrays.asList("daf2a533d83f1a5fd7d0e5a92d67fd64"));
|
||||
executeTest("HC calling with aggressive indel error modeling on WGS intervals", spec);
|
||||
}
|
||||
|
||||
|
|
@ -284,7 +284,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
public void HCTestConservativePcrIndelModelWGS() {
|
||||
WalkerTest.WalkerTestSpec spec = new WalkerTest.WalkerTestSpec(
|
||||
"-T HaplotypeCaller --disableDithering --pcr_indel_model CONSERVATIVE -R " + b37KGReference + " --no_cmdline_in_header -I " + NA12878_BAM + " -o %s -L 20:10,000,000-10,300,000", 1,
|
||||
Arrays.asList("01b716d421dc33f7a6bc8c387b7e3322"));
|
||||
Arrays.asList("69bccf72c900361775d076d7d131955d"));
|
||||
executeTest("HC calling with conservative indel error modeling on WGS intervals", spec);
|
||||
}
|
||||
|
||||
|
|
@ -313,7 +313,7 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
public void testMissingKeyAlternativeHaplotypesBugFix() {
|
||||
final String commandLine = String.format("-T HaplotypeCaller -R %s -I %s -L %s --no_cmdline_in_header ",
|
||||
b37KGReferenceWithDecoy, privateTestDir + "lost-alt-key-hap.bam", privateTestDir + "lost-alt-key-hap.interval_list");
|
||||
final WalkerTestSpec spec = new WalkerTestSpec(commandLine + " -o %s", Arrays.asList("cd33c69a34a66b5baca72841fe8a2bc0"));
|
||||
final WalkerTestSpec spec = new WalkerTestSpec(commandLine + " -o %s", Arrays.asList("4322083ad7decca26a3257be2f1d0347"));
|
||||
spec.disableShadowBCF();
|
||||
executeTest("testMissingKeyAlternativeHaplotypesBugFix", spec);
|
||||
}
|
||||
|
|
@ -336,9 +336,10 @@ public class HaplotypeCallerIntegrationTest extends WalkerTest {
|
|||
// but please make sure that both outputs get the same variant,
|
||||
// alleles all with DBSNP ids
|
||||
// We test here that change in active region size does not have an effect in placement of indels.
|
||||
final WalkerTestSpec shortSpec = new WalkerTestSpec(commandLineShortInterval + " -o %s",Arrays.asList("0f87fd95350fade475fe3c5df8dd9e08"));
|
||||
final String md5 = "54c2cea6e2266fe3ff05c1b5d58bb609";
|
||||
final WalkerTestSpec shortSpec = new WalkerTestSpec(commandLineShortInterval + " -o %s",Arrays.asList(md5));
|
||||
executeTest("testDifferentIndelLocationsDueToSWExactDoubleComparisonsFix::shortInterval",shortSpec);
|
||||
final WalkerTestSpec longSpec = new WalkerTestSpec(commandLineLongInterval + " -o %s",Arrays.asList("b3a2853d51a7817999a3be8e72d9ac23"));
|
||||
final WalkerTestSpec longSpec = new WalkerTestSpec(commandLineLongInterval + " -o %s",Arrays.asList(md5));
|
||||
executeTest("testDifferentIndelLocationsDueToSWExactDoubleComparisonsFix::longInterval",longSpec);
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -60,7 +60,7 @@ public class HaplotypeCallerParallelIntegrationTest extends WalkerTest {
|
|||
List<Object[]> tests = new ArrayList<>();
|
||||
|
||||
for ( final int nct : Arrays.asList(1, 2, 4) ) {
|
||||
tests.add(new Object[]{nct, "a2718251ffae9db885b7f74b33dd5b57"});
|
||||
tests.add(new Object[]{nct, "af29c396835b86d88f366629884e6ba4"});
|
||||
}
|
||||
|
||||
return tests.toArray(new Object[][]{});
|
||||
|
|
|
|||
Loading…
Reference in New Issue