def run(self, config, config_file, parallel, dirs, samples): ## Alignment and preparation requiring the entire input file (multicore cluster) with prun.start(_wres(parallel, ["aligner", "samtools", "sambamba"], (["reference", "fasta"], ["reference", "aligner"], ["files"])), samples, config, dirs, "multicore", multiplier=alignprep.parallel_multiplier(samples)) as run_parallel: with profile.report("alignment preparation", dirs): samples = run_parallel("prep_align_inputs", samples) samples = disambiguate.split(samples) with profile.report("alignment", dirs): samples = run_parallel("process_alignment", samples) samples = alignprep.merge_split_alignments(samples, run_parallel) samples = disambiguate.resolve(samples, run_parallel) with profile.report("callable regions", dirs): samples = run_parallel("postprocess_alignment", samples) samples = run_parallel("combine_sample_regions", [samples]) samples = region.clean_sample_data(samples) with profile.report("coverage", dirs): samples = coverage.summarize_samples(samples, run_parallel) ## Variant calling on sub-regions of the input file (full cluster) with prun.start(_wres(parallel, ["gatk", "picard", "variantcaller"]), samples, config, dirs, "full", multiplier=region.get_max_counts(samples), max_multicore=1) as run_parallel: with profile.report("alignment post-processing", dirs): samples = region.parallel_prep_region(samples, run_parallel) with profile.report("variant calling", dirs): samples = genotype.parallel_variantcall_region(samples, run_parallel) ## Finalize variants (per-sample cluster) with prun.start(_wres(parallel, ["gatk", "gatk-vqsr", "snpeff", "bcbio_variation"]), samples, config, dirs, "persample") as run_parallel: with profile.report("joint squaring off/backfilling", dirs): samples = joint.square_off(samples, run_parallel) with profile.report("variant post-processing", dirs): samples = run_parallel("postprocess_variants", samples) samples = run_parallel("split_variants_by_sample", samples) with profile.report("validation", dirs): samples = run_parallel("compare_to_rm", samples) samples = genotype.combine_multiple_callers(samples) ## Finalizing BAMs and population databases, handle multicore computation with prun.start(_wres(parallel, ["gemini", "samtools", "fastqc", "bamtools", "bcbio_variation", "bcbio-variation-recall"]), samples, config, dirs, "multicore2") as run_parallel: with profile.report("prepped BAM merging", dirs): samples = region.delayed_bamprep_merge(samples, run_parallel) with profile.report("ensemble calling", dirs): samples = ensemble.combine_calls_parallel(samples, run_parallel) with profile.report("validation summary", dirs): samples = validate.summarize_grading(samples) with profile.report("structural variation", dirs): samples = structural.run(samples, run_parallel) with profile.report("population database", dirs): samples = population.prep_db_parallel(samples, run_parallel) with profile.report("quality control", dirs): samples = qcsummary.generate_parallel(samples, run_parallel) with profile.report("archive", dirs): samples = archive.compress(samples, run_parallel) logger.info("Timing: finished") return samples
def run(self, config, config_file, parallel, dirs, samples): ## Alignment and preparation requiring the entire input file (multicore cluster) with prun.start(_wres(parallel, ["aligner", "samtools", "sambamba"], (["reference", "fasta"], ["reference", "aligner"], ["files"])), samples, config, dirs, "multicore", multiplier=alignprep.parallel_multiplier(samples)) as run_parallel: with profile.report("alignment preparation", dirs): samples = run_parallel("prep_align_inputs", samples) samples = disambiguate.split(samples) with profile.report("alignment", dirs): samples = run_parallel("process_alignment", samples) samples = alignprep.merge_split_alignments(samples, run_parallel) samples = disambiguate.resolve(samples, run_parallel) with profile.report("callable regions", dirs): samples = run_parallel("postprocess_alignment", samples) samples = run_parallel("combine_sample_regions", [samples]) samples = region.clean_sample_data(samples) with profile.report("coverage", dirs): samples = coverage.summarize_samples(samples, run_parallel) ## Variant calling on sub-regions of the input file (full cluster) with prun.start(_wres(parallel, ["gatk", "picard", "variantcaller"]), samples, config, dirs, "full", multiplier=region.get_max_counts(samples), max_multicore=1) as run_parallel: with profile.report("alignment post-processing", dirs): samples = region.parallel_prep_region(samples, run_parallel) with profile.report("variant calling", dirs): samples = genotype.parallel_variantcall_region(samples, run_parallel) ## Finalize variants (per-sample cluster) with prun.start(_wres(parallel, ["gatk", "gatk-vqsr", "snpeff", "bcbio_variation"]), samples, config, dirs, "persample") as run_parallel: with profile.report("variant post-processing", dirs): samples = run_parallel("postprocess_variants", samples) samples = run_parallel("split_variants_by_sample", samples) with profile.report("validation", dirs): samples = run_parallel("compare_to_rm", samples) samples = genotype.combine_multiple_callers(samples) ## Finalizing BAMs and population databases, handle multicore computation with prun.start(_wres(parallel, ["gemini", "samtools", "fastqc", "bamtools", "bcbio_variation", "bcbio-variation-recall"]), samples, config, dirs, "multicore2") as run_parallel: with profile.report("prepped BAM merging", dirs): samples = region.delayed_bamprep_merge(samples, run_parallel) with profile.report("ensemble calling", dirs): samples = ensemble.combine_calls_parallel(samples, run_parallel) with profile.report("validation summary", dirs): samples = validate.summarize_grading(samples) with profile.report("structural variation", dirs): samples = structural.run(samples, run_parallel) with profile.report("population database", dirs): samples = population.prep_db_parallel(samples, run_parallel) with profile.report("quality control", dirs): samples = qcsummary.generate_parallel(samples, run_parallel) with profile.report("archive", dirs): samples = archive.compress(samples, run_parallel) logger.info("Timing: finished") return samples
def run(self, config, config_file, run_parallel, parallel, dirs, samples): ## Alignment and preparation requiring the entire input file (multicore cluster) with global_parallel(parallel, "multicore", ["process_alignment", "postprocess_alignment"], samples, dirs, config, multiplier=alignprep.parallel_multiplier(samples)) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: alignment") samples = run_parallel("prep_align_inputs", samples) samples = disambiguate.split(samples) samples = run_parallel("process_alignment", samples) samples = alignprep.merge_split_alignments(samples, run_parallel) samples = disambiguate.resolve(samples, run_parallel) samples = run_parallel("postprocess_alignment", samples) regions = callable.combine_sample_regions(samples) samples = region.add_region_info(samples, regions) samples = region.clean_sample_data(samples) logger.info("Timing: coverage") samples = coverage.summarize_samples(samples, run_parallel) ## Variant calling on sub-regions of the input file (full cluster) with global_parallel(parallel, "full", ["piped_bamprep", "variantcall_sample"], samples, dirs, config, multiplier=len(regions["analysis"]), max_multicore=1) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: alignment post-processing") samples = region.parallel_prep_region(samples, regions, run_parallel) logger.info("Timing: variant calling") samples = region.parallel_variantcall_region(samples, run_parallel) ## Finalize variants (per-sample cluster) with global_parallel(parallel, "persample", ["postprocess_variants"], samples, dirs, config) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: variant post-processing") samples = run_parallel("postprocess_variants", samples) logger.info("Timing: validation") samples = run_parallel("compare_to_rm", samples) samples = combine_multiple_callers(samples) logger.info("Timing: ensemble calling") samples = ensemble.combine_calls_parallel(samples, run_parallel) samples = validate.summarize_grading(samples) ## Finalizing BAMs and population databases, handle multicore computation with global_parallel(parallel, "multicore2", ["prep_gemini_db", "delayed_bam_merge"], samples, dirs, config) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: prepped BAM merging") samples = region.delayed_bamprep_merge(samples, run_parallel) logger.info("Timing: structural variation") samples = structural.run(samples, run_parallel) logger.info("Timing: population database") samples = population.prep_db_parallel(samples, run_parallel) logger.info("Timing: quality control") samples = qcsummary.generate_parallel(samples, run_parallel) logger.info("Timing: finished") return samples
def run(self, config, config_file, run_parallel, parallel, dirs, lane_items): ## Alignment and preparation requiring the entire input file (multicore cluster) with global_parallel(parallel, "multicore", ["align_prep_full"], lane_items, dirs, config) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: alignment") samples = run_parallel( "align_prep_full", [list(x) + [config_file] for x in lane_items]) regions = callable.combine_sample_regions(samples) samples = region.add_region_info(samples, regions) samples = region.clean_sample_data(samples) logger.info("Timing: coverage") samples = coverage.summarize_samples(samples, run_parallel) ## Variant calling on sub-regions of the input file (full cluster) with global_parallel(parallel, "full", ["piped_bamprep", "variantcall_sample"], samples, dirs, config, multiplier=len(regions["analysis"])) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: alignment post-processing") samples = region.parallel_prep_region(samples, regions, run_parallel) logger.info("Timing: variant calling") samples = region.parallel_variantcall_region(samples, run_parallel) ## Finalize variants (per-sample cluster) with global_parallel(parallel, "persample", ["postprocess_variants"], samples, dirs, config) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: variant post-processing") samples = run_parallel("postprocess_variants", samples) logger.info("Timing: validation") samples = run_parallel("compare_to_rm", samples) samples = combine_multiple_callers(samples) logger.info("Timing: ensemble calling") samples = ensemble.combine_calls_parallel(samples, run_parallel) samples = validate.summarize_grading(samples) logger.info("Timing: quality control") samples = qcsummary.generate_parallel(samples, run_parallel) ## Finalizing BAMs and population databases, handle multicore computation with global_parallel(parallel, "multicore2", ["prep_gemini_db", "delayed_bam_merge"], samples, dirs, config) as parallel: run_parallel = parallel_runner(parallel, dirs, config) logger.info("Timing: prepped BAM merging") samples = region.delayed_bamprep_merge(samples, run_parallel) logger.info("Timing: population database") samples = population.prep_db_parallel(samples, run_parallel) logger.info("Timing: finished") return samples
def run(self, config, run_info_yaml, parallel, dirs, samples): ## Alignment and preparation requiring the entire input file (multicore cluster) with prun.start(_wres(parallel, ["aligner", "samtools", "sambamba"], (["reference", "fasta"], ["reference", "aligner"], ["files"])), samples, config, dirs, "multicore", multiplier=alignprep.parallel_multiplier(samples)) as run_parallel: with profile.report("organize samples", dirs): samples = run_parallel("organize_samples", [[dirs, config, run_info_yaml, [x[0]["description"] for x in samples]]]) with profile.report("alignment preparation", dirs): samples = run_parallel("prep_align_inputs", samples) samples = run_parallel("disambiguate_split", [samples]) with profile.report("alignment", dirs): samples = run_parallel("process_alignment", samples) samples = disambiguate.resolve(samples, run_parallel) samples = alignprep.merge_split_alignments(samples, run_parallel) with profile.report("callable regions", dirs): samples = run_parallel("prep_samples", [samples]) samples = run_parallel("postprocess_alignment", samples) samples = run_parallel("combine_sample_regions", [samples]) samples = region.clean_sample_data(samples) with profile.report("coverage", dirs): samples = coverage.summarize_samples(samples, run_parallel) with profile.report("structural variation initial", dirs): samples = structural.run(samples, run_parallel, initial_only=True) ## Variant calling on sub-regions of the input file (full cluster) with prun.start(_wres(parallel, ["gatk", "picard", "variantcaller"]), samples, config, dirs, "full", multiplier=region.get_max_counts(samples), max_multicore=1) as run_parallel: with profile.report("alignment post-processing", dirs): samples = region.parallel_prep_region(samples, run_parallel) with profile.report("variant calling", dirs): samples = genotype.parallel_variantcall_region(samples, run_parallel) ## Finalize variants, BAMs and population databases (per-sample multicore cluster) with prun.start(_wres(parallel, ["gatk", "gatk-vqsr", "snpeff", "bcbio_variation", "gemini", "samtools", "fastqc", "bamtools", "bcbio-variation-recall", "qsignature"]), samples, config, dirs, "multicore2") as run_parallel: with profile.report("joint squaring off/backfilling", dirs): samples = joint.square_off(samples, run_parallel) with profile.report("variant post-processing", dirs): samples = run_parallel("postprocess_variants", samples) samples = run_parallel("split_variants_by_sample", samples) with profile.report("prepped BAM merging", dirs): samples = region.delayed_bamprep_merge(samples, run_parallel) with profile.report("validation", dirs): samples = run_parallel("compare_to_rm", samples) samples = genotype.combine_multiple_callers(samples) with profile.report("ensemble calling", dirs): samples = ensemble.combine_calls_parallel(samples, run_parallel) with profile.report("validation summary", dirs): samples = validate.summarize_grading(samples) with profile.report("structural variation final", dirs): samples = structural.run(samples, run_parallel) with profile.report("heterogeneity", dirs): samples = heterogeneity.run(samples, run_parallel) with profile.report("population database", dirs): samples = population.prep_db_parallel(samples, run_parallel) with profile.report("quality control", dirs): samples = qcsummary.generate_parallel(samples, run_parallel) with prun.start(_wres(parallel, ["gatk"], ensure_mem = {"gatk": 8}), samples, config, dirs, "coverage") as run_parallel: with profile.report("report", dirs): samples = qcsummary.report_summary(samples, run_parallel) with profile.report("archive", dirs): samples = archive.compress(samples, run_parallel) with profile.report("upload", dirs): samples = run_parallel("upload_samples", samples) for sample in samples: run_parallel("upload_samples_project", [sample]) logger.info("Timing: finished") return samples