def _lane_stats(cur_name, work_dir):
"""Parse metrics information from files in the working directory.
"""
parser = PicardMetricsParser()
metrics_files = glob.glob(os.path.join(work_dir, "%s*metrics" % cur_name))
metrics = parser.extract_metrics(metrics_files)
return metrics
def _lane_stats(cur_name, work_dir):
"""Parse metrics information from files in the working directory.
"""
parser = PicardMetricsParser()
metrics_files = glob.glob(os.path.join(work_dir, "%s*metrics" % cur_name))
metrics = parser.extract_metrics(metrics_files)
return metrics
def _bowtie_for_innerdist(start,
fastq_file,
pair_file,
ref_file,
out_base,
out_dir,
config,
remove_workdir=False):
work_dir = os.path.join(out_dir, "innerdist_estimate")
if os.path.exists(work_dir):
shutil.rmtree(work_dir)
safe_makedir(work_dir)
extra_args = ["-s", str(start), "-u", "250000"]
bowtie_runner = _select_bowtie_version(config)
out_sam = bowtie_runner.align(fastq_file, pair_file, ref_file, out_base,
work_dir, config, extra_args)
runner = broad.runner_from_config(config)
metrics_file = runner.run_fn("picard_insert_metrics", out_sam)
if not file_exists(metrics_file):
return None, None
parser = PicardMetricsParser()
with open(metrics_file) as metrics_handle:
insert_metrics = parser._parse_insert_metrics(metrics_handle)
avg_read_length = _calculate_average_read_length(out_sam)
mean_insert = int(float(insert_metrics["MEAN_INSERT_SIZE"])) - int(
2 * avg_read_length)
std_deviation = int(float(insert_metrics["STANDARD_DEVIATION"]))
return mean_insert, std_deviation
def main(run_name,
work_dir=None,
config_file=None,
ref_file=None,
bait_file=None,
target_file=None):
if work_dir is None:
work_dir = os.getcwd()
parser = PicardMetricsParser()
base_bams = _get_base_bams(work_dir, run_name)
samples = [_get_sample_name(b) for (_, _, b) in base_bams]
metrics = [
lane_stats(l, b, f, run_name, parser, config_file, ref_file, bait_file,
target_file) for (l, b, f) in base_bams
]
header_counts = _get_header_counts(samples, metrics)
header = [m for m in WANT_METRICS if header_counts[m] > 0]
out_file = "%s-summary.csv" % (run_name)
with open(out_file, "w") as out_handle:
writer = csv.writer(out_handle)
writer.writerow(["sample"] + header)
for i, sample in enumerate(samples):
info = [metrics[i].get(m, "") for m in header]
writer.writerow([sample] + info)
def _bowtie_for_innerdist(start, fastq_file, pair_file, ref_file, out_base,
out_dir, config, remove_workdir=False):
work_dir = os.path.join(out_dir, "innerdist_estimate")
if os.path.exists(work_dir):
shutil.rmtree(work_dir)
safe_makedir(work_dir)
extra_args = ["-s", str(start), "-u", "250000"]
bowtie_runner = _select_bowtie_version(config)
out_sam = bowtie_runner.align(fastq_file, pair_file, ref_file, out_base,
work_dir, config, extra_args)
runner = broad.runner_from_config(config)
metrics_file = runner.run_fn("picard_insert_metrics", out_sam)
if not file_exists(metrics_file):
return None, None
parser = PicardMetricsParser()
with open(metrics_file) as metrics_handle:
insert_metrics = parser._parse_insert_metrics(metrics_handle)
avg_read_length = _calculate_average_read_length(out_sam)
mean_insert = int(float(insert_metrics["MEAN_INSERT_SIZE"])) - int(2 * avg_read_length)
std_deviation = int(float(insert_metrics["STANDARD_DEVIATION"]))
return mean_insert, std_deviation
def __init__(self):
PicardMetricsParser.__init__(self)
def __init__(self):
PicardMetricsParser.__init__(self)
