Search in sources :

Example 1 with HMMSegmentProcessor

use of org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor in project gatk by broadinstitute.

the class CoverageModelEMWorkspace method getCopyRatioSegmentsSpark.

/**
     * Fetch copy ratio segments from compute blocks (Spark implementation)
     *
     * @return a list of {@link CopyRatioHMMResults}
     */
private List<List<HiddenStateSegmentRecord<STATE, Target>>> getCopyRatioSegmentsSpark() {
    /* local final member variables for lambda capture */
    final List<Target> processedTargetList = new ArrayList<>();
    processedTargetList.addAll(this.processedTargetList);
    final List<SexGenotypeData> processedSampleSexGenotypeData = new ArrayList<>();
    processedSampleSexGenotypeData.addAll(this.processedSampleSexGenotypeData);
    final List<String> processedSampleNameList = new ArrayList<>();
    processedSampleNameList.addAll(this.processedSampleNameList);
    final INDArray sampleReadDepths = Transforms.exp(sampleMeanLogReadDepths, true);
    final CopyRatioExpectationsCalculator<CoverageModelCopyRatioEmissionData, STATE> copyRatioExpectationsCalculator = this.copyRatioExpectationsCalculator;
    final BiFunction<SexGenotypeData, Target, STATE> referenceStateFactory = this.referenceStateFactory;
    return fetchCopyRatioEmissionDataSpark().mapPartitionsToPair(it -> {
        final List<Tuple2<Integer, CopyRatioHMMResults<CoverageModelCopyRatioEmissionData, STATE>>> newPartitionData = new ArrayList<>();
        while (it.hasNext()) {
            final Tuple2<Integer, List<CoverageModelCopyRatioEmissionData>> prevDatum = it.next();
            final int sampleIndex = prevDatum._1;
            final CopyRatioCallingMetadata copyRatioCallingMetadata = CopyRatioCallingMetadata.builder().sampleName(processedSampleNameList.get(sampleIndex)).sampleSexGenotypeData(processedSampleSexGenotypeData.get(sampleIndex)).sampleCoverageDepth(sampleReadDepths.getDouble(sampleIndex)).emissionCalculationStrategy(EmissionCalculationStrategy.HYBRID_POISSON_GAUSSIAN).build();
            newPartitionData.add(new Tuple2<>(sampleIndex, copyRatioExpectationsCalculator.getCopyRatioHMMResults(copyRatioCallingMetadata, processedTargetList, prevDatum._2)));
        }
        return newPartitionData.iterator();
    }, true).mapPartitionsToPair(it -> {
        final List<Tuple2<Integer, List<HiddenStateSegmentRecord<STATE, Target>>>> newPartitionData = new ArrayList<>();
        while (it.hasNext()) {
            final Tuple2<Integer, CopyRatioHMMResults<CoverageModelCopyRatioEmissionData, STATE>> prevDatum = it.next();
            final int sampleIndex = prevDatum._1;
            final CopyRatioHMMResults<CoverageModelCopyRatioEmissionData, STATE> result = prevDatum._2;
            final HMMSegmentProcessor<CoverageModelCopyRatioEmissionData, STATE, Target> processor = new HMMSegmentProcessor<>(Collections.singletonList(result.getMetaData().getSampleName()), Collections.singletonList(result.getMetaData().getSampleSexGenotypeData()), referenceStateFactory, Collections.singletonList(new HashedListTargetCollection<>(processedTargetList)), Collections.singletonList(result.getForwardBackwardResult()), Collections.singletonList(result.getViterbiResult()));
            newPartitionData.add(new Tuple2<>(sampleIndex, processor.getSegmentsAsList()));
        }
        return newPartitionData.iterator();
    }).collect().stream().sorted(Comparator.comparingInt(t -> t._1)).map(t -> t._2).collect(Collectors.toList());
}
Also used : ScalarProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.ScalarProducer) Function2(org.apache.spark.api.java.function.Function2) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor) GermlinePloidyAnnotatedTargetCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.GermlinePloidyAnnotatedTargetCollection) HiddenStateSegmentRecordWriter(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecordWriter) BiFunction(java.util.function.BiFunction) GATKException(org.broadinstitute.hellbender.exceptions.GATKException) SexGenotypeData(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeData) ParamUtils(org.broadinstitute.hellbender.utils.param.ParamUtils) CallStringProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.CallStringProducer) StorageLevel(org.apache.spark.storage.StorageLevel) SynchronizedUnivariateSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.SynchronizedUnivariateSolver) CopyRatioExpectationsCalculator(org.broadinstitute.hellbender.tools.coveragemodel.interfaces.CopyRatioExpectationsCalculator) UnivariateSolverSpecifications(org.broadinstitute.hellbender.tools.coveragemodel.math.UnivariateSolverSpecifications) IndexRange(org.broadinstitute.hellbender.utils.IndexRange) Broadcast(org.apache.spark.broadcast.Broadcast) ExitStatus(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray.ExitStatus) SexGenotypeDataCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeDataCollection) HashPartitioner(org.apache.spark.HashPartitioner) Predicate(java.util.function.Predicate) GeneralLinearOperator(org.broadinstitute.hellbender.tools.coveragemodel.linalg.GeneralLinearOperator) Nd4j(org.nd4j.linalg.factory.Nd4j) INDArrayIndex(org.nd4j.linalg.indexing.INDArrayIndex) FastMath(org.apache.commons.math3.util.FastMath) org.broadinstitute.hellbender.tools.exome(org.broadinstitute.hellbender.tools.exome) Tuple2(scala.Tuple2) Collectors(java.util.stream.Collectors) Sets(com.google.common.collect.Sets) AbstractUnivariateSolver(org.apache.commons.math3.analysis.solvers.AbstractUnivariateSolver) FourierLinearOperatorNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.FourierLinearOperatorNDArray) Logger(org.apache.logging.log4j.Logger) Stream(java.util.stream.Stream) UserException(org.broadinstitute.hellbender.exceptions.UserException) UnivariateFunction(org.apache.commons.math3.analysis.UnivariateFunction) TooManyEvaluationsException(org.apache.commons.math3.exception.TooManyEvaluationsException) Utils(org.broadinstitute.hellbender.utils.Utils) Function(org.apache.spark.api.java.function.Function) DataBuffer(org.nd4j.linalg.api.buffer.DataBuffer) IntStream(java.util.stream.IntStream) java.util(java.util) NDArrayIndex(org.nd4j.linalg.indexing.NDArrayIndex) JavaSparkContext(org.apache.spark.api.java.JavaSparkContext) AlleleMetadataProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.AlleleMetadataProducer) EmissionCalculationStrategy(org.broadinstitute.hellbender.tools.coveragemodel.CoverageModelCopyRatioEmissionProbabilityCalculator.EmissionCalculationStrategy) RobustBrentSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.RobustBrentSolver) IntervalUtils(org.broadinstitute.hellbender.utils.IntervalUtils) Nonnull(javax.annotation.Nonnull) Nullable(javax.annotation.Nullable) HiddenStateSegmentRecord(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecord) ImmutableTriple(org.apache.commons.lang3.tuple.ImmutableTriple) IterativeLinearSolverNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray) GATKProtectedMathUtils(org.broadinstitute.hellbender.utils.GATKProtectedMathUtils) Nd4jIOUtils(org.broadinstitute.hellbender.tools.coveragemodel.nd4jutils.Nd4jIOUtils) IOException(java.io.IOException) JavaPairRDD(org.apache.spark.api.java.JavaPairRDD) ImmutablePair(org.apache.commons.lang3.tuple.ImmutablePair) File(java.io.File) INDArray(org.nd4j.linalg.api.ndarray.INDArray) VisibleForTesting(com.google.common.annotations.VisibleForTesting) Transforms(org.nd4j.linalg.ops.transforms.Transforms) LogManager(org.apache.logging.log4j.LogManager) NoBracketingException(org.apache.commons.math3.exception.NoBracketingException) INDArray(org.nd4j.linalg.api.ndarray.INDArray) Tuple2(scala.Tuple2) SexGenotypeData(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeData) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor)

Example 2 with HMMSegmentProcessor

use of org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor in project gatk by broadinstitute.

the class CoverageModelEMWorkspace method getCopyRatioSegmentsLocal.

private List<List<HiddenStateSegmentRecord<STATE, Target>>> getCopyRatioSegmentsLocal() {
    final List<List<CoverageModelCopyRatioEmissionData>> copyRatioEmissionData = fetchCopyRatioEmissionDataLocal();
    final INDArray sampleReadDepths = Transforms.exp(sampleMeanLogReadDepths, true);
    return sampleIndexStream().mapToObj(si -> {
        final CopyRatioCallingMetadata metadata = CopyRatioCallingMetadata.builder().sampleName(processedSampleNameList.get(si)).sampleSexGenotypeData(processedSampleSexGenotypeData.get(si)).sampleCoverageDepth(sampleReadDepths.getDouble(si)).emissionCalculationStrategy(EmissionCalculationStrategy.HYBRID_POISSON_GAUSSIAN).build();
        return copyRatioExpectationsCalculator.getCopyRatioHMMResults(metadata, processedTargetList, copyRatioEmissionData.get(si));
    }).map(result -> {
        final HMMSegmentProcessor<CoverageModelCopyRatioEmissionData, STATE, Target> processor = new HMMSegmentProcessor<>(Collections.singletonList(result.getMetaData().getSampleName()), Collections.singletonList(result.getMetaData().getSampleSexGenotypeData()), referenceStateFactory, Collections.singletonList(new HashedListTargetCollection<>(processedTargetList)), Collections.singletonList(result.getForwardBackwardResult()), Collections.singletonList(result.getViterbiResult()));
        return processor.getSegmentsAsList();
    }).collect(Collectors.toList());
}
Also used : ScalarProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.ScalarProducer) Function2(org.apache.spark.api.java.function.Function2) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor) GermlinePloidyAnnotatedTargetCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.GermlinePloidyAnnotatedTargetCollection) HiddenStateSegmentRecordWriter(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecordWriter) BiFunction(java.util.function.BiFunction) GATKException(org.broadinstitute.hellbender.exceptions.GATKException) SexGenotypeData(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeData) ParamUtils(org.broadinstitute.hellbender.utils.param.ParamUtils) CallStringProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.CallStringProducer) StorageLevel(org.apache.spark.storage.StorageLevel) SynchronizedUnivariateSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.SynchronizedUnivariateSolver) CopyRatioExpectationsCalculator(org.broadinstitute.hellbender.tools.coveragemodel.interfaces.CopyRatioExpectationsCalculator) UnivariateSolverSpecifications(org.broadinstitute.hellbender.tools.coveragemodel.math.UnivariateSolverSpecifications) IndexRange(org.broadinstitute.hellbender.utils.IndexRange) Broadcast(org.apache.spark.broadcast.Broadcast) ExitStatus(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray.ExitStatus) SexGenotypeDataCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeDataCollection) HashPartitioner(org.apache.spark.HashPartitioner) Predicate(java.util.function.Predicate) GeneralLinearOperator(org.broadinstitute.hellbender.tools.coveragemodel.linalg.GeneralLinearOperator) Nd4j(org.nd4j.linalg.factory.Nd4j) INDArrayIndex(org.nd4j.linalg.indexing.INDArrayIndex) FastMath(org.apache.commons.math3.util.FastMath) org.broadinstitute.hellbender.tools.exome(org.broadinstitute.hellbender.tools.exome) Tuple2(scala.Tuple2) Collectors(java.util.stream.Collectors) Sets(com.google.common.collect.Sets) AbstractUnivariateSolver(org.apache.commons.math3.analysis.solvers.AbstractUnivariateSolver) FourierLinearOperatorNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.FourierLinearOperatorNDArray) Logger(org.apache.logging.log4j.Logger) Stream(java.util.stream.Stream) UserException(org.broadinstitute.hellbender.exceptions.UserException) UnivariateFunction(org.apache.commons.math3.analysis.UnivariateFunction) TooManyEvaluationsException(org.apache.commons.math3.exception.TooManyEvaluationsException) Utils(org.broadinstitute.hellbender.utils.Utils) Function(org.apache.spark.api.java.function.Function) DataBuffer(org.nd4j.linalg.api.buffer.DataBuffer) IntStream(java.util.stream.IntStream) java.util(java.util) NDArrayIndex(org.nd4j.linalg.indexing.NDArrayIndex) JavaSparkContext(org.apache.spark.api.java.JavaSparkContext) AlleleMetadataProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.AlleleMetadataProducer) EmissionCalculationStrategy(org.broadinstitute.hellbender.tools.coveragemodel.CoverageModelCopyRatioEmissionProbabilityCalculator.EmissionCalculationStrategy) RobustBrentSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.RobustBrentSolver) IntervalUtils(org.broadinstitute.hellbender.utils.IntervalUtils) Nonnull(javax.annotation.Nonnull) Nullable(javax.annotation.Nullable) HiddenStateSegmentRecord(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecord) ImmutableTriple(org.apache.commons.lang3.tuple.ImmutableTriple) IterativeLinearSolverNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray) GATKProtectedMathUtils(org.broadinstitute.hellbender.utils.GATKProtectedMathUtils) Nd4jIOUtils(org.broadinstitute.hellbender.tools.coveragemodel.nd4jutils.Nd4jIOUtils) IOException(java.io.IOException) JavaPairRDD(org.apache.spark.api.java.JavaPairRDD) ImmutablePair(org.apache.commons.lang3.tuple.ImmutablePair) File(java.io.File) INDArray(org.nd4j.linalg.api.ndarray.INDArray) VisibleForTesting(com.google.common.annotations.VisibleForTesting) Transforms(org.nd4j.linalg.ops.transforms.Transforms) LogManager(org.apache.logging.log4j.LogManager) NoBracketingException(org.apache.commons.math3.exception.NoBracketingException) INDArray(org.nd4j.linalg.api.ndarray.INDArray) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor)

Example 3 with HMMSegmentProcessor

use of org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor in project gatk-protected by broadinstitute.

the class CoverageModelEMWorkspace method getCopyRatioSegmentsSpark.

/**
     * Fetch copy ratio segments from compute blocks (Spark implementation)
     *
     * @return a list of {@link CopyRatioHMMResults}
     */
private List<List<HiddenStateSegmentRecord<STATE, Target>>> getCopyRatioSegmentsSpark() {
    /* local final member variables for lambda capture */
    final List<Target> processedTargetList = new ArrayList<>();
    processedTargetList.addAll(this.processedTargetList);
    final List<SexGenotypeData> processedSampleSexGenotypeData = new ArrayList<>();
    processedSampleSexGenotypeData.addAll(this.processedSampleSexGenotypeData);
    final List<String> processedSampleNameList = new ArrayList<>();
    processedSampleNameList.addAll(this.processedSampleNameList);
    final INDArray sampleReadDepths = Transforms.exp(sampleMeanLogReadDepths, true);
    final CopyRatioExpectationsCalculator<CoverageModelCopyRatioEmissionData, STATE> copyRatioExpectationsCalculator = this.copyRatioExpectationsCalculator;
    final BiFunction<SexGenotypeData, Target, STATE> referenceStateFactory = this.referenceStateFactory;
    return fetchCopyRatioEmissionDataSpark().mapPartitionsToPair(it -> {
        final List<Tuple2<Integer, CopyRatioHMMResults<CoverageModelCopyRatioEmissionData, STATE>>> newPartitionData = new ArrayList<>();
        while (it.hasNext()) {
            final Tuple2<Integer, List<CoverageModelCopyRatioEmissionData>> prevDatum = it.next();
            final int sampleIndex = prevDatum._1;
            final CopyRatioCallingMetadata copyRatioCallingMetadata = CopyRatioCallingMetadata.builder().sampleName(processedSampleNameList.get(sampleIndex)).sampleSexGenotypeData(processedSampleSexGenotypeData.get(sampleIndex)).sampleCoverageDepth(sampleReadDepths.getDouble(sampleIndex)).emissionCalculationStrategy(EmissionCalculationStrategy.HYBRID_POISSON_GAUSSIAN).build();
            newPartitionData.add(new Tuple2<>(sampleIndex, copyRatioExpectationsCalculator.getCopyRatioHMMResults(copyRatioCallingMetadata, processedTargetList, prevDatum._2)));
        }
        return newPartitionData.iterator();
    }, true).mapPartitionsToPair(it -> {
        final List<Tuple2<Integer, List<HiddenStateSegmentRecord<STATE, Target>>>> newPartitionData = new ArrayList<>();
        while (it.hasNext()) {
            final Tuple2<Integer, CopyRatioHMMResults<CoverageModelCopyRatioEmissionData, STATE>> prevDatum = it.next();
            final int sampleIndex = prevDatum._1;
            final CopyRatioHMMResults<CoverageModelCopyRatioEmissionData, STATE> result = prevDatum._2;
            final HMMSegmentProcessor<CoverageModelCopyRatioEmissionData, STATE, Target> processor = new HMMSegmentProcessor<>(Collections.singletonList(result.getMetaData().getSampleName()), Collections.singletonList(result.getMetaData().getSampleSexGenotypeData()), referenceStateFactory, Collections.singletonList(new HashedListTargetCollection<>(processedTargetList)), Collections.singletonList(result.getForwardBackwardResult()), Collections.singletonList(result.getViterbiResult()));
            newPartitionData.add(new Tuple2<>(sampleIndex, processor.getSegmentsAsList()));
        }
        return newPartitionData.iterator();
    }).collect().stream().sorted(Comparator.comparingInt(t -> t._1)).map(t -> t._2).collect(Collectors.toList());
}
Also used : ScalarProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.ScalarProducer) Function2(org.apache.spark.api.java.function.Function2) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor) GermlinePloidyAnnotatedTargetCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.GermlinePloidyAnnotatedTargetCollection) HiddenStateSegmentRecordWriter(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecordWriter) BiFunction(java.util.function.BiFunction) GATKException(org.broadinstitute.hellbender.exceptions.GATKException) SexGenotypeData(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeData) ParamUtils(org.broadinstitute.hellbender.utils.param.ParamUtils) CallStringProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.CallStringProducer) StorageLevel(org.apache.spark.storage.StorageLevel) SynchronizedUnivariateSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.SynchronizedUnivariateSolver) CopyRatioExpectationsCalculator(org.broadinstitute.hellbender.tools.coveragemodel.interfaces.CopyRatioExpectationsCalculator) UnivariateSolverSpecifications(org.broadinstitute.hellbender.tools.coveragemodel.math.UnivariateSolverSpecifications) IndexRange(org.broadinstitute.hellbender.utils.IndexRange) Broadcast(org.apache.spark.broadcast.Broadcast) ExitStatus(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray.ExitStatus) SexGenotypeDataCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeDataCollection) HashPartitioner(org.apache.spark.HashPartitioner) Predicate(java.util.function.Predicate) GeneralLinearOperator(org.broadinstitute.hellbender.tools.coveragemodel.linalg.GeneralLinearOperator) Nd4j(org.nd4j.linalg.factory.Nd4j) INDArrayIndex(org.nd4j.linalg.indexing.INDArrayIndex) FastMath(org.apache.commons.math3.util.FastMath) org.broadinstitute.hellbender.tools.exome(org.broadinstitute.hellbender.tools.exome) Tuple2(scala.Tuple2) Collectors(java.util.stream.Collectors) Sets(com.google.common.collect.Sets) AbstractUnivariateSolver(org.apache.commons.math3.analysis.solvers.AbstractUnivariateSolver) FourierLinearOperatorNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.FourierLinearOperatorNDArray) Logger(org.apache.logging.log4j.Logger) Stream(java.util.stream.Stream) UserException(org.broadinstitute.hellbender.exceptions.UserException) UnivariateFunction(org.apache.commons.math3.analysis.UnivariateFunction) TooManyEvaluationsException(org.apache.commons.math3.exception.TooManyEvaluationsException) Utils(org.broadinstitute.hellbender.utils.Utils) Function(org.apache.spark.api.java.function.Function) DataBuffer(org.nd4j.linalg.api.buffer.DataBuffer) IntStream(java.util.stream.IntStream) java.util(java.util) NDArrayIndex(org.nd4j.linalg.indexing.NDArrayIndex) JavaSparkContext(org.apache.spark.api.java.JavaSparkContext) AlleleMetadataProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.AlleleMetadataProducer) EmissionCalculationStrategy(org.broadinstitute.hellbender.tools.coveragemodel.CoverageModelCopyRatioEmissionProbabilityCalculator.EmissionCalculationStrategy) RobustBrentSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.RobustBrentSolver) IntervalUtils(org.broadinstitute.hellbender.utils.IntervalUtils) Nonnull(javax.annotation.Nonnull) Nullable(javax.annotation.Nullable) HiddenStateSegmentRecord(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecord) ImmutableTriple(org.apache.commons.lang3.tuple.ImmutableTriple) IterativeLinearSolverNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray) GATKProtectedMathUtils(org.broadinstitute.hellbender.utils.GATKProtectedMathUtils) Nd4jIOUtils(org.broadinstitute.hellbender.tools.coveragemodel.nd4jutils.Nd4jIOUtils) IOException(java.io.IOException) JavaPairRDD(org.apache.spark.api.java.JavaPairRDD) ImmutablePair(org.apache.commons.lang3.tuple.ImmutablePair) File(java.io.File) INDArray(org.nd4j.linalg.api.ndarray.INDArray) VisibleForTesting(com.google.common.annotations.VisibleForTesting) Transforms(org.nd4j.linalg.ops.transforms.Transforms) LogManager(org.apache.logging.log4j.LogManager) NoBracketingException(org.apache.commons.math3.exception.NoBracketingException) INDArray(org.nd4j.linalg.api.ndarray.INDArray) Tuple2(scala.Tuple2) SexGenotypeData(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeData) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor)

Example 4 with HMMSegmentProcessor

use of org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor in project gatk-protected by broadinstitute.

the class CoverageModelEMWorkspace method getCopyRatioSegmentsLocal.

private List<List<HiddenStateSegmentRecord<STATE, Target>>> getCopyRatioSegmentsLocal() {
    final List<List<CoverageModelCopyRatioEmissionData>> copyRatioEmissionData = fetchCopyRatioEmissionDataLocal();
    final INDArray sampleReadDepths = Transforms.exp(sampleMeanLogReadDepths, true);
    return sampleIndexStream().mapToObj(si -> {
        final CopyRatioCallingMetadata metadata = CopyRatioCallingMetadata.builder().sampleName(processedSampleNameList.get(si)).sampleSexGenotypeData(processedSampleSexGenotypeData.get(si)).sampleCoverageDepth(sampleReadDepths.getDouble(si)).emissionCalculationStrategy(EmissionCalculationStrategy.HYBRID_POISSON_GAUSSIAN).build();
        return copyRatioExpectationsCalculator.getCopyRatioHMMResults(metadata, processedTargetList, copyRatioEmissionData.get(si));
    }).map(result -> {
        final HMMSegmentProcessor<CoverageModelCopyRatioEmissionData, STATE, Target> processor = new HMMSegmentProcessor<>(Collections.singletonList(result.getMetaData().getSampleName()), Collections.singletonList(result.getMetaData().getSampleSexGenotypeData()), referenceStateFactory, Collections.singletonList(new HashedListTargetCollection<>(processedTargetList)), Collections.singletonList(result.getForwardBackwardResult()), Collections.singletonList(result.getViterbiResult()));
        return processor.getSegmentsAsList();
    }).collect(Collectors.toList());
}
Also used : ScalarProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.ScalarProducer) Function2(org.apache.spark.api.java.function.Function2) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor) GermlinePloidyAnnotatedTargetCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.GermlinePloidyAnnotatedTargetCollection) HiddenStateSegmentRecordWriter(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecordWriter) BiFunction(java.util.function.BiFunction) GATKException(org.broadinstitute.hellbender.exceptions.GATKException) SexGenotypeData(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeData) ParamUtils(org.broadinstitute.hellbender.utils.param.ParamUtils) CallStringProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.CallStringProducer) StorageLevel(org.apache.spark.storage.StorageLevel) SynchronizedUnivariateSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.SynchronizedUnivariateSolver) CopyRatioExpectationsCalculator(org.broadinstitute.hellbender.tools.coveragemodel.interfaces.CopyRatioExpectationsCalculator) UnivariateSolverSpecifications(org.broadinstitute.hellbender.tools.coveragemodel.math.UnivariateSolverSpecifications) IndexRange(org.broadinstitute.hellbender.utils.IndexRange) Broadcast(org.apache.spark.broadcast.Broadcast) ExitStatus(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray.ExitStatus) SexGenotypeDataCollection(org.broadinstitute.hellbender.tools.exome.sexgenotyper.SexGenotypeDataCollection) HashPartitioner(org.apache.spark.HashPartitioner) Predicate(java.util.function.Predicate) GeneralLinearOperator(org.broadinstitute.hellbender.tools.coveragemodel.linalg.GeneralLinearOperator) Nd4j(org.nd4j.linalg.factory.Nd4j) INDArrayIndex(org.nd4j.linalg.indexing.INDArrayIndex) FastMath(org.apache.commons.math3.util.FastMath) org.broadinstitute.hellbender.tools.exome(org.broadinstitute.hellbender.tools.exome) Tuple2(scala.Tuple2) Collectors(java.util.stream.Collectors) Sets(com.google.common.collect.Sets) AbstractUnivariateSolver(org.apache.commons.math3.analysis.solvers.AbstractUnivariateSolver) FourierLinearOperatorNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.FourierLinearOperatorNDArray) Logger(org.apache.logging.log4j.Logger) Stream(java.util.stream.Stream) UserException(org.broadinstitute.hellbender.exceptions.UserException) UnivariateFunction(org.apache.commons.math3.analysis.UnivariateFunction) TooManyEvaluationsException(org.apache.commons.math3.exception.TooManyEvaluationsException) Utils(org.broadinstitute.hellbender.utils.Utils) Function(org.apache.spark.api.java.function.Function) DataBuffer(org.nd4j.linalg.api.buffer.DataBuffer) IntStream(java.util.stream.IntStream) java.util(java.util) NDArrayIndex(org.nd4j.linalg.indexing.NDArrayIndex) JavaSparkContext(org.apache.spark.api.java.JavaSparkContext) AlleleMetadataProducer(org.broadinstitute.hellbender.utils.hmm.interfaces.AlleleMetadataProducer) EmissionCalculationStrategy(org.broadinstitute.hellbender.tools.coveragemodel.CoverageModelCopyRatioEmissionProbabilityCalculator.EmissionCalculationStrategy) RobustBrentSolver(org.broadinstitute.hellbender.tools.coveragemodel.math.RobustBrentSolver) IntervalUtils(org.broadinstitute.hellbender.utils.IntervalUtils) Nonnull(javax.annotation.Nonnull) Nullable(javax.annotation.Nullable) HiddenStateSegmentRecord(org.broadinstitute.hellbender.utils.hmm.segmentation.HiddenStateSegmentRecord) ImmutableTriple(org.apache.commons.lang3.tuple.ImmutableTriple) IterativeLinearSolverNDArray(org.broadinstitute.hellbender.tools.coveragemodel.linalg.IterativeLinearSolverNDArray) GATKProtectedMathUtils(org.broadinstitute.hellbender.utils.GATKProtectedMathUtils) Nd4jIOUtils(org.broadinstitute.hellbender.tools.coveragemodel.nd4jutils.Nd4jIOUtils) IOException(java.io.IOException) JavaPairRDD(org.apache.spark.api.java.JavaPairRDD) ImmutablePair(org.apache.commons.lang3.tuple.ImmutablePair) File(java.io.File) INDArray(org.nd4j.linalg.api.ndarray.INDArray) VisibleForTesting(com.google.common.annotations.VisibleForTesting) Transforms(org.nd4j.linalg.ops.transforms.Transforms) LogManager(org.apache.logging.log4j.LogManager) NoBracketingException(org.apache.commons.math3.exception.NoBracketingException) INDArray(org.nd4j.linalg.api.ndarray.INDArray) HMMSegmentProcessor(org.broadinstitute.hellbender.utils.hmm.segmentation.HMMSegmentProcessor)

Aggregations

VisibleForTesting (com.google.common.annotations.VisibleForTesting)4 Sets (com.google.common.collect.Sets)4 File (java.io.File)4 IOException (java.io.IOException)4 java.util (java.util)4 BiFunction (java.util.function.BiFunction)4 Predicate (java.util.function.Predicate)4 Collectors (java.util.stream.Collectors)4 IntStream (java.util.stream.IntStream)4 Stream (java.util.stream.Stream)4 Nonnull (javax.annotation.Nonnull)4 Nullable (javax.annotation.Nullable)4 ImmutablePair (org.apache.commons.lang3.tuple.ImmutablePair)4 ImmutableTriple (org.apache.commons.lang3.tuple.ImmutableTriple)4 UnivariateFunction (org.apache.commons.math3.analysis.UnivariateFunction)4 AbstractUnivariateSolver (org.apache.commons.math3.analysis.solvers.AbstractUnivariateSolver)4 NoBracketingException (org.apache.commons.math3.exception.NoBracketingException)4 TooManyEvaluationsException (org.apache.commons.math3.exception.TooManyEvaluationsException)4 FastMath (org.apache.commons.math3.util.FastMath)4 LogManager (org.apache.logging.log4j.LogManager)4