Examples with GATKRead org.broadinstitute.hellbender.utils.read.GATKRead used on opensource projects

Example 6 with GATKRead

use of org.broadinstitute.hellbender.utils.read.GATKRead in project gatk by broadinstitute.

the class CollectMultipleMetricsSpark method runTool.

@Override
protected void runTool(final JavaSparkContext ctx) {
    final JavaRDD<GATKRead> unFilteredReads = getUnfilteredReads();
    List<SparkCollectorProvider> collectorsToRun = getCollectorsToRun();
    if (collectorsToRun.size() > 1) {
        // if there is more than one collector to run, cache the
        // unfiltered RDD so we don't recompute it
        unFilteredReads.cache();
    }
    for (final SparkCollectorProvider provider : collectorsToRun) {
        MetricsCollectorSpark<? extends MetricsArgumentCollection> metricsCollector = provider.createCollector(outputBaseName, metricAccumulationLevel.accumulationLevels, getDefaultHeaders(), getHeaderForReads());
        validateCollector(metricsCollector, collectorsToRun.get(collectorsToRun.indexOf(provider)).getClass().getName());
        // Execute the collector's lifecycle
        //Bypass the framework merging of command line filters and just apply the default
        //ones specified by the collector
        ReadFilter readFilter = ReadFilter.fromList(metricsCollector.getDefaultReadFilters(), getHeaderForReads());
        metricsCollector.collectMetrics(unFilteredReads.filter(r -> readFilter.test(r)), getHeaderForReads());
        metricsCollector.saveMetrics(getReadSourceName(), getAuthHolder());
    }
}

Example 7 with GATKRead

use of org.broadinstitute.hellbender.utils.read.GATKRead in project gatk by broadinstitute.

the class CountReadsSpark method runTool.

@Override
protected void runTool(final JavaSparkContext ctx) {
    final JavaRDD<GATKRead> reads = getReads();
    final long count = reads.count();
    System.out.println(count);
    if (out != null) {
        try (final PrintStream ps = new PrintStream(BucketUtils.createFile(out))) {
            ps.print(count);
        }
    }
}

Example 8 with GATKRead

use of org.broadinstitute.hellbender.utils.read.GATKRead in project gatk by broadinstitute.

the class FlagStatSpark method runTool.

@Override
protected void runTool(final JavaSparkContext ctx) {
    final JavaRDD<GATKRead> reads = getReads();
    final FlagStatus result = reads.aggregate(new FlagStatus(), FlagStatus::add, FlagStatus::merge);
    System.out.println(result);
    if (out != null) {
        try (final PrintStream ps = new PrintStream(BucketUtils.createFile(out))) {
            ps.print(result);
        }
    }
}

Example 9 with GATKRead

use of org.broadinstitute.hellbender.utils.read.GATKRead in project gatk by broadinstitute.

the class ReadsPipelineSpark method runTool.

@Override
protected void runTool(final JavaSparkContext ctx) {
    if (joinStrategy == JoinStrategy.BROADCAST && !getReference().isCompatibleWithSparkBroadcast()) {
        throw new UserException.Require2BitReferenceForBroadcast();
    }
    //TOOO: should this use getUnfilteredReads? getReads will apply default and command line filters
    final JavaRDD<GATKRead> initialReads = getReads();
    final JavaRDD<GATKRead> markedReadsWithOD = MarkDuplicatesSpark.mark(initialReads, getHeaderForReads(), duplicatesScoringStrategy, new OpticalDuplicateFinder(), getRecommendedNumReducers());
    final JavaRDD<GATKRead> markedReads = MarkDuplicatesSpark.cleanupTemporaryAttributes(markedReadsWithOD);
    // The markedReads have already had the WellformedReadFilter applied to them, which
    // is all the filtering that MarkDupes and ApplyBQSR want. BQSR itself wants additional
    // filtering performed, so we do that here.
    //NOTE: this doesn't honor enabled/disabled commandline filters
    final ReadFilter bqsrReadFilter = ReadFilter.fromList(BaseRecalibrator.getBQSRSpecificReadFilterList(), getHeaderForReads());
    final JavaRDD<GATKRead> markedFilteredReadsForBQSR = markedReads.filter(read -> bqsrReadFilter.test(read));
    VariantsSparkSource variantsSparkSource = new VariantsSparkSource(ctx);
    JavaRDD<GATKVariant> bqsrKnownVariants = variantsSparkSource.getParallelVariants(baseRecalibrationKnownVariants, getIntervals());
    JavaPairRDD<GATKRead, ReadContextData> rddReadContext = AddContextDataToReadSpark.add(ctx, markedFilteredReadsForBQSR, getReference(), bqsrKnownVariants, joinStrategy, getReferenceSequenceDictionary(), readShardSize, readShardPadding);
    final RecalibrationReport bqsrReport = BaseRecalibratorSparkFn.apply(rddReadContext, getHeaderForReads(), getReferenceSequenceDictionary(), bqsrArgs);
    final Broadcast<RecalibrationReport> reportBroadcast = ctx.broadcast(bqsrReport);
    final JavaRDD<GATKRead> finalReads = ApplyBQSRSparkFn.apply(markedReads, reportBroadcast, getHeaderForReads(), applyBqsrArgs.toApplyBQSRArgumentCollection(bqsrArgs.PRESERVE_QSCORES_LESS_THAN));
    writeReads(ctx, output, finalReads);
}

Example 10 with GATKRead

use of org.broadinstitute.hellbender.utils.read.GATKRead in project gatk by broadinstitute.

the class ReadsDownsamplingIterator method next.

@Override
public GATKRead next() {
    if (nextRead == null) {
        throw new NoSuchElementException("next() called when there are no more items");
    }
    final GATKRead toReturn = nextRead;
    advanceToNextRead();
    return toReturn;
}