def coverage(data):
AVERAGE_REGION_STRING_LENGTH = 100
bed_file = dd.get_coverage_experimental(data)
if not bed_file:
return data
work_dir = os.path.join(dd.get_work_dir(data), "report", "coverage")
batch_size = max_command_length() / AVERAGE_REGION_STRING_LENGTH
with chdir(work_dir):
in_bam = data['work_bam']
sample = dd.get_sample_name(data)
logger.debug("doing coverage for %s" % sample)
region_bed = pybedtools.BedTool(bed_file)
parse_file = os.path.join(sample + "_coverage.bed")
parse_total_file = os.path.join(sample + "_cov_total.tsv")
if not file_exists(parse_file):
total_cov = cov_class(0, None, sample)
with file_transaction(parse_file) as out_tx:
with open(out_tx, 'w') as out_handle:
HEADER = ["#chrom", "start", "end", "region", "reads",
"strand", "size", "sample", "mean", "sd", "cutoff10",
"cutoff20", "cutoff4", "cutoff50"]
out_handle.write("\t".join(HEADER) + "\n")
with tmpfile() as tx_tmp_file:
lcount = 0
for chunk in robust_partition_all(batch_size, region_bed):
coord_batch = []
line_batch = ""
for line in chunk:
lcount += 1
chrom = line.chrom
start = max(line.start, 0)
end = line.end
coords = "%s:%s-%s" % (chrom, start, end)
coord_batch.append(coords)
line_batch += str(line)
if not coord_batch:
continue
region_file = pybedtools.BedTool(line_batch,
from_string=True).saveas().fn
coord_string = " ".join(coord_batch)
cmd = ("samtools view -b {in_bam} {coord_string} | "
"bedtools coverage -a {region_file} -b - "
"-hist > {tx_tmp_file}")
_silence_run(cmd.format(**locals()))
total_cov = _get_exome_coverage_stats(os.path.abspath(tx_tmp_file), sample, out_tx, total_cov)
logger.debug("Processed %d regions." % lcount)
total_cov.write_coverage(parse_total_file)
data['coverage'] = os.path.abspath(parse_file)
return data
def priority_coverage(data):
AVERAGE_REGION_STRING_LENGTH = 100
bed_file = dd.get_svprioritize(data)
if not bed_file or not file_exists(bed_file):
return data
work_dir = os.path.join(dd.get_work_dir(data), "report", "coverage")
batch_size = max_command_length() / AVERAGE_REGION_STRING_LENGTH
sample = dd.get_sample_name(data)
out_file = os.path.join(work_dir, sample + "_priority_depth.bed")
if file_exists(out_file):
data['priority_coverage'] = os.path.abspath(out_file)
return data
with chdir(work_dir):
in_bam = dd.get_align_bam(data) or dd.get_work_bam(data)
logger.debug("Calculating priority coverage for %s" % sample)
region_bed = pybedtools.BedTool(bed_file)
with file_transaction(out_file) as tx_out_file:
lcount = 0
for chunk in robust_partition_all(batch_size, region_bed):
coord_batch = []
line_batch = ""
for line in chunk:
lcount += 1
chrom = line.chrom
start = max(line.start, 0)
end = line.end
coords = "%s:%s-%s" % (chrom, start, end)
coord_batch.append(coords)
line_batch += "%s\t%s\t%s\n" % (chrom, start, end)
if not coord_batch:
continue
region_file = pybedtools.BedTool(line_batch,
from_string=True).saveas().fn
coord_string = " ".join(coord_batch)
awk_string = r"""'BEGIN {OFS="\t"} {print $1,$2+$5,$2+$5,$4,$6"\t%s"}'""" % sample
samtools = config_utils.get_program("samtools", data["config"])
bedtools = config_utils.get_program("bedtools", data["config"])
cmd = (
"{samtools} view -b {in_bam} {coord_string} | "
"{bedtools} coverage -sorted -d -a {region_file} -b - | "
"awk {awk_string} >> {tx_out_file}")
_silence_run(cmd.format(**locals()))
data['priority_coverage'] = os.path.abspath(out_file)
return data
def priority_coverage(data):
AVERAGE_REGION_STRING_LENGTH = 100
bed_file = dd.get_svprioritize(data)
if not bed_file or not file_exists(bed_file):
return data
work_dir = os.path.join(dd.get_work_dir(data), "report", "coverage")
batch_size = max_command_length() / AVERAGE_REGION_STRING_LENGTH
sample = dd.get_sample_name(data)
out_file = os.path.join(work_dir, sample + "_priority_depth.bed")
if file_exists(out_file):
data['priority_coverage'] = os.path.abspath(out_file)
return data
with chdir(work_dir):
in_bam = dd.get_align_bam(data) or dd.get_work_bam(data)
logger.debug("Calculating priority coverage for %s" % sample)
region_bed = pybedtools.BedTool(bed_file)
with file_transaction(out_file) as tx_out_file:
lcount = 0
for chunk in robust_partition_all(batch_size, region_bed):
coord_batch = []
line_batch = ""
for line in chunk:
lcount += 1
chrom = line.chrom
start = max(line.start, 0)
end = line.end
coords = "%s:%s-%s" % (chrom, start, end)
coord_batch.append(coords)
line_batch += "%s\t%s\t%s\n" % (chrom, start, end)
if not coord_batch:
continue
region_file = pybedtools.BedTool(line_batch,
from_string=True).saveas().fn
coord_string = " ".join(coord_batch)
awk_string = r"""'BEGIN {OFS="\t"} {print $1,$2+$5,$2+$5,$4,$6"\t%s"}'""" % sample
samtools = config_utils.get_program("samtools", data["config"])
bedtools = config_utils.get_program("bedtools", data["config"])
cmd = ("{samtools} view -b {in_bam} {coord_string} | "
"{bedtools} coverage -sorted -d -a {region_file} -b - | "
"awk {awk_string} >> {tx_out_file}")
_silence_run(cmd.format(**locals()))
data['priority_coverage'] = os.path.abspath(out_file)
return data