def calculate_hs_metrics(self):
jobs = []
created_interval_lists = []
for sample in self.samples:
coverage_bed = bvatools.resolve_readset_coverage_bed(sample.readsets[0])
if coverage_bed:
interval_list = re.sub("\.[^.]+$", ".interval_list", coverage_bed)
recal_file_prefix = os.path.join("alignment", sample.name, sample.name + ".sorted.dup.recal.")
job = picard.calculate_hs_metrics(recal_file_prefix + "bam", recal_file_prefix + "onTarget.tsv", interval_list)
if not interval_list in created_interval_lists:
job = concat_jobs([tools.bed2interval_list(None, coverage_bed, interval_list), job])
created_interval_lists.append(interval_list)
job.name = "calculate_hs_metrics." + sample.name
jobs.append(job)
return jobs
def metrics(self):
jobs = []
for sample in self.samples:
recal_file_prefix = os.path.join("alignment", sample.name, sample.name + ".sorted.dup.recal.")
input = recal_file_prefix + "bam"
job = picard.collect_multiple_metrics(input, recal_file_prefix + "all.metrics")
job.name = "collect_multiple_metrics." + sample.name
jobs.append(job)
# Compute genome coverage
job = gatk.depth_of_coverage(input, recal_file_prefix + "all.coverage")
job.name = "genome_coverage." + sample.name
jobs.append(job)
# Compute CCDS coverage
job = gatk.depth_of_coverage(input, recal_file_prefix + "all.coverage", config.param('metrics', 'coverageTargets'))
job.name = "target_coverage." + sample.name
jobs.append(job)
job = bvatools.depth_of_coverage(input, recal_file_prefix + "coverage.tsv", bvatools.resolve_readset_coverage_bed(sample.readsets[0]))
job.name = "depth_of_coverage." + sample.name
jobs.append(job)
job = igvtools.compute_tdf(input, input + ".tdf")
job.name = "igvtools." + sample.name
jobs.append(job)
job = samtools.flagstat(input, recal_file_prefix + "bam.flagstat")
job.name = "flagstat." + sample.name
jobs.append(job)
return jobs
