Examples with GeneCopyNumber com.hartwig.hmftools.common.gene.GeneCopyNumber used on opensource projects

Example 6 with GeneCopyNumber

use of com.hartwig.hmftools.common.gene.GeneCopyNumber in project hmftools by hartwigmedical.

the class BachelorEligibility method fromMap.

static BachelorEligibility fromMap(final Map<String, Program> input) {
    final BachelorEligibility result = new BachelorEligibility();
    for (final Program program : input.values()) {
        final Multimap<String, String> geneToEnsemblMap = HashMultimap.create();
        program.getPanel().stream().map(ProgramPanel::getGene).flatMap(Collection::stream).forEach(g -> geneToEnsemblMap.put(g.getName(), g.getEnsembl()));
        // NOTE: copy number and SVs are untested/unverified for now, but leave in support for them
        // process copy number sections
        final List<Predicate<GeneCopyNumber>> cnvPredicates = Lists.newArrayList();
        for (final ProgramPanel panel : program.getPanel()) {
            final List<GeneIdentifier> genes = panel.getGene();
            if (panel.getEffect().contains(OtherEffect.HOMOZYGOUS_DELETION)) {
                final Predicate<GeneCopyNumber> geneCopyNumberPredicate = cnv -> genes.stream().anyMatch(g -> g.getEnsembl().equals(cnv.transcriptID()));
                // TODO: we are matching on transcript ID here but we only have canonical transcripts in our panel file
                cnvPredicates.add(geneCopyNumberPredicate);
            }
        }
        // process structural variant disruptions
        final List<Predicate<HmfGenomeRegion>> disruptionPredicates = Lists.newArrayList();
        for (final ProgramPanel panel : program.getPanel()) {
            final List<GeneIdentifier> genes = panel.getGene();
            if (panel.getEffect().contains(OtherEffect.GENE_DISRUPTION)) {
                final Predicate<HmfGenomeRegion> disruptionPredicate = sv -> genes.stream().anyMatch(g -> g.getEnsembl().equals(sv.transcriptID()));
                // TODO: we are matching on transcript ID here but we only have canonical transcripts in our panel file
                disruptionPredicates.add(disruptionPredicate);
            }
        }
        // process variants from vcf
        final List<Predicate<VariantModel>> panelPredicates = Lists.newArrayList();
        List<String> requiredEffects = Lists.newArrayList();
        List<String> panelTranscripts = Lists.newArrayList();
        for (final ProgramPanel panel : program.getPanel()) {
            final List<GeneIdentifier> genes = panel.getGene();
            // take up a collection of the effects to search for
            requiredEffects = panel.getSnpEffect().stream().map(SnpEffect::value).collect(Collectors.toList());
            panelTranscripts = genes.stream().map(GeneIdentifier::getEnsembl).collect(Collectors.toList());
            final List<String> effects = requiredEffects;
            final Predicate<VariantModel> panelPredicate = v -> genes.stream().anyMatch(p -> v.sampleAnnotations().stream().anyMatch(a -> a.featureID().equals(p.getEnsembl()) && effects.stream().anyMatch(x -> a.effects().contains(x))));
            panelPredicates.add(panelPredicate);
            // update query targets
            for (final GeneIdentifier g : genes) {
                final HmfGenomeRegion region = allTranscriptsMap.get(g.getEnsembl());
                if (region == null) {
                    final HmfGenomeRegion namedRegion = allGenesMap.get(g.getName());
                    if (namedRegion == null) {
                        LOGGER.warn("Program {} gene {} non-canonical transcript {} couldn't find region, transcript will be skipped", program.getName(), g.getName(), g.getEnsembl());
                    // just skip this gene for now
                    } else {
                        result.variantLocationsToQuery.add(namedRegion);
                    }
                } else {
                    result.variantLocationsToQuery.add(region);
                }
            }
        }
        final Predicate<VariantModel> inPanel = v -> panelPredicates.stream().anyMatch(p -> p.test(v));
        final Predicate<VariantModel> inBlacklist = new BlacklistPredicate(geneToEnsemblMap.values(), program.getBlacklist());
        final Predicate<VariantModel> inWhitelist = new WhitelistPredicate(geneToEnsemblMap, program.getWhitelist());
        final Predicate<VariantModel> snvPredicate = v -> inPanel.test(v) ? !inBlacklist.test(v) : inWhitelist.test(v);
        final Predicate<GeneCopyNumber> copyNumberPredicate = cnv -> cnvPredicates.stream().anyMatch(p -> p.test(cnv)) && cnv.minCopyNumber() < MAX_COPY_NUMBER_FOR_LOSS;
        final Predicate<HmfGenomeRegion> disruptionPredicate = disruption -> disruptionPredicates.stream().anyMatch(p -> p.test(disruption));
        BachelorProgram bachelorProgram = new BachelorProgram(program.getName(), snvPredicate, copyNumberPredicate, disruptionPredicate, requiredEffects, panelTranscripts);
        result.programs.add(bachelorProgram);
    }
    return result;
}

Example 7 with GeneCopyNumber

use of com.hartwig.hmftools.common.gene.GeneCopyNumber in project hmftools by hartwigmedical.

the class PatientReporter method analyseGenomeData.

@NotNull
private GenomeAnalysis analyseGenomeData(@NotNull final String sample, @NotNull final String runDirectory) throws IOException {
    LOGGER.info(" Loading somatic snv and indels...");
    final List<SomaticVariant> variants = PatientReporterHelper.loadPassedSomaticVariants(sample, runDirectory);
    LOGGER.info("  " + variants.size() + " somatic passed snps, mnps and indels loaded for sample " + sample);
    LOGGER.info(" Analyzing somatic snp/mnp and indels....");
    final VariantAnalysis variantAnalysis = variantAnalyzer().run(variants);
    LOGGER.info(" Loading purity numbers...");
    final PurityContext context = PatientReporterHelper.loadPurity(runDirectory, sample);
    if (context.status().equals(FittedPurityStatus.NO_TUMOR)) {
        LOGGER.warn("PURPLE DID NOT DETECT A TUMOR. Proceed with utmost caution!");
    }
    final FittedPurity purity = context.bestFit();
    final FittedPurityScore purityScore = context.score();
    final List<PurpleCopyNumber> purpleCopyNumbers = PatientReporterHelper.loadPurpleCopyNumbers(runDirectory, sample);
    final List<GeneCopyNumber> panelGeneCopyNumbers = PatientReporterHelper.loadPurpleGeneCopyNumbers(runDirectory, sample).stream().filter(x -> reporterData().panelGeneModel().panel().contains(x.gene())).collect(Collectors.toList());
    LOGGER.info("  " + purpleCopyNumbers.size() + " purple copy number regions loaded for sample " + sample);
    LOGGER.info(" Analyzing purple somatic copy numbers...");
    final PurpleAnalysis purpleAnalysis = ImmutablePurpleAnalysis.builder().gender(context.gender()).status(context.status()).fittedPurity(purity).fittedScorePurity(purityScore).copyNumbers(purpleCopyNumbers).panelGeneCopyNumbers(panelGeneCopyNumbers).build();
    final Path structuralVariantVCF = PatientReporterHelper.findStructuralVariantVCF(runDirectory);
    LOGGER.info(" Loading structural variants...");
    final List<StructuralVariant> structuralVariants = StructuralVariantFileLoader.fromFile(structuralVariantVCF.toString(), true);
    LOGGER.info(" Enriching structural variants with purple data.");
    final List<EnrichedStructuralVariant> enrichedStructuralVariants = purpleAnalysis.enrichStructuralVariants(structuralVariants);
    LOGGER.info(" Analysing structural variants...");
    final StructuralVariantAnalysis structuralVariantAnalysis = structuralVariantAnalyzer().run(enrichedStructuralVariants, false);
    return ImmutableGenomeAnalysis.of(sample, variantAnalysis, purpleAnalysis, structuralVariantAnalysis);
}

Example 8 with GeneCopyNumber

use of com.hartwig.hmftools.common.gene.GeneCopyNumber in project hmftools by hartwigmedical.

the class GeneCopyNumberDAO method writeCopyNumber.

void writeCopyNumber(@NotNull final String sample, @NotNull List<GeneCopyNumber> copyNumbers) {
    Timestamp timestamp = new Timestamp(new Date().getTime());
    context.delete(GENECOPYNUMBER).where(GENECOPYNUMBER.SAMPLEID.eq(sample)).execute();
    for (List<GeneCopyNumber> splitCopyNumbers : Iterables.partition(copyNumbers, DB_BATCH_INSERT_SIZE)) {
        InsertValuesStepN inserter = context.insertInto(GENECOPYNUMBER, GENECOPYNUMBER.SAMPLEID, GENECOPYNUMBER.CHROMOSOME, GENECOPYNUMBER.START, GENECOPYNUMBER.END, GENECOPYNUMBER.GENE, GENECOPYNUMBER.MINCOPYNUMBER, GENECOPYNUMBER.MAXCOPYNUMBER, GENECOPYNUMBER.SOMATICREGIONS, GENECOPYNUMBER.GERMLINEHOMREGIONS, GENECOPYNUMBER.GERMLINEHETREGIONS, GENECOPYNUMBER.TRANSCRIPTID, GENECOPYNUMBER.TRANSCRIPTVERSION, GENECOPYNUMBER.CHROMOSOMEBAND, GENECOPYNUMBER.MINREGIONS, GENECOPYNUMBER.MINREGIONSTART, GENECOPYNUMBER.MINREGIONEND, GENECOPYNUMBER.MINREGIONSTARTSUPPORT, GENECOPYNUMBER.MINREGIONENDSUPPORT, GENECOPYNUMBER.MINREGIONMETHOD, GENECOPYNUMBER.NONSENSEBIALLELICVARIANTS, GENECOPYNUMBER.NONSENSENONBIALLELICVARIANTS, GENECOPYNUMBER.NONSENSENONBIALLELICPLOIDY, GENECOPYNUMBER.SPLICEBIALLELICVARIANTS, GENECOPYNUMBER.SPLICENONBIALLELICVARIANTS, GENECOPYNUMBER.SPLICENONBIALLELICPLOIDY, GENECOPYNUMBER.MISSENSEBIALLELICVARIANTS, GENECOPYNUMBER.MISSENSENONBIALLELICVARIANTS, GENECOPYNUMBER.MISSENSENONBIALLELICPLOIDY, GENECOPYNUMBER.MINMINORALLELEPLOIDY, COPYNUMBER.MODIFIED);
        splitCopyNumbers.forEach(x -> addCopynumberRecord(timestamp, inserter, sample, x));
        inserter.execute();
    }
}