input_stream, process = utils_commands.get_output_stream_from_command(
command)
tmp_bam_file = output_bam_file = tmp + '_mrk_dup.bam.tmp'
command = "%s view -bS - >%s" % (samtools_bin, tmp_bam_file)
output_stream, process = utils_commands.get_input_stream_from_command(
command)
nb_reference = 0
current_reference = None
first_reads = {}
second_reads = {}
for line in input_stream:
if line.startswith("@"):
output_stream.write(line)
continue
sam_record = Sam_record(line)
if sam_record.get_reference_name(
) != current_reference and not current_reference is None:
#process this consensus
if current_reference != '*':
nb_dups, nb_uniq = find_duplicates(first_reads, second_reads,
distance_threshold)
total_nb_uniqs += nb_uniq
total_nb_dups += nb_dups
nb_fragment += len(second_reads)
output_reads(output_stream, first_reads, second_reads)
first_reads = {}
second_reads = {}
if sam_record.is_second_read():
second_reads[sam_record.get_query_name()] = sam_record
else:
first_reads[sam_record.get_query_name()] = sam_record
#get the header
line = input_stream.readline()
while line.startswith("@"):
output_stream.write(line)
line = input_stream.readline()
while line:
read1=Sam_record(line)
line = input_stream.readline()
read2=Sam_record(line)
if read1.get_query_name() == read2.get_query_name():
if read1.is_second_read() and read2.is_first_read():
tmp = read1
read1=read2
read2=tmp
read2.set_reference_name(read1.get_reference_name())
output_stream.write(str(read1))
output_stream.write(str(read2))
else:
logging.critical("bam file is not sorted by read name")
input_stream.close()
output_stream.close()
#os.remove(output_bam_file+'.bam')
return
line = input_stream.readline()
return_code=process_input.wait()
print return_code
if return_code!=0:
sys.exit(return_code)
return_code=process_output.wait()
def process_single_samtools_run_with_read_group(bam_file, all_contigs_info, samtools_bin):
command = "%s view -h -F 132 %s" % (samtools_bin, bam_file)
open_stream, process = get_output_stream_from_command(command)
current_contig = None
sample_name, ext = os.path.splitext(bam_file)
read_groups = {}
try:
for line in open_stream:
if not line.startswith("@"):
break
if line.startswith("@RG"):
sp_line = line.strip().split()
rg_id = rg_sample = rg_library = None
for value in sp_line:
if value.startswith("ID"):
rg_id = value[3:]
elif value.startswith("SM"):
rg_sample = value[3:]
elif value.startswith("LB"):
rg_library = value[3:]
if rg_id:
if rg_sample:
read_groups[rg_id] = rg_sample
elif rg_library:
read_groups[rg_id] = rg_library
else:
read_groups[rg_id] = rg_id
all_sample_coverage = {}
all_sample_duplicate = {}
for sample in read_groups.values():
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
# process the first read
# if line.startswith("@"):
# #Still in the header. There's no read, exit
# return
sam_record = Sam_record(line.strip())
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
i = 1
# process all the others
for line in open_stream:
i += 1
if i % 1000000 == 0:
print i
sam_record = Sam_record(line.strip())
if current_contig != sam_record.get_reference_name() and current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(current_contig, all_sample_coverage.get(sample),
all_sample_duplicate.get(sample), sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
if current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(current_contig, all_sample_coverage.get(sample),
all_sample_duplicate.get(sample), sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
finally:
open_stream.close()
def process_single_samtools_run_with_read_group(bam_file, all_contigs_info,
samtools_bin):
command = "%s view -h -F 132 %s" % (samtools_bin, bam_file)
open_stream, process = get_output_stream_from_command(command)
current_contig = None
sample_name, ext = os.path.splitext(bam_file)
read_groups = {}
try:
for line in open_stream:
if not line.startswith("@"):
break
if line.startswith("@RG"):
sp_line = line.strip().split()
rg_id = rg_sample = rg_library = None
for value in sp_line:
if value.startswith("ID"):
rg_id = value[3:]
elif value.startswith("SM"):
rg_sample = value[3:]
elif value.startswith("LB"):
rg_library = value[3:]
if rg_id:
if rg_sample:
read_groups[rg_id] = rg_sample
elif rg_library:
read_groups[rg_id] = rg_library
else:
read_groups[rg_id] = rg_id
all_sample_coverage = {}
all_sample_duplicate = {}
for sample in read_groups.values():
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
# process the first read
# if line.startswith("@"):
# #Still in the header. There's no read, exit
# return
sam_record = Sam_record(line.strip())
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
i = 1
# process all the others
for line in open_stream:
i += 1
if i % 1000000 == 0:
print i
sam_record = Sam_record(line.strip())
if current_contig != sam_record.get_reference_name(
) and current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(
current_contig,
all_sample_coverage.get(sample),
all_sample_duplicate.get(sample),
sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)] += 1
all_sample_coverage[read_groups.get(rg_id)] += 1
if current_contig != None:
for sample in read_groups.values():
all_contigs_info.add_values(current_contig,
all_sample_coverage.get(sample),
all_sample_duplicate.get(sample),
sample=sample)
all_sample_coverage[sample] = 0
all_sample_duplicate[sample] = 0
finally:
open_stream.close()
command = "%s view -h %s"%(samtools_bin, bam_file)
input_stream, process = utils_commands.get_output_stream_from_command(command)
tmp_bam_file = output_bam_file = tmp + '_mrk_dup.bam.tmp'
command = "%s view -bS - >%s"%(samtools_bin, tmp_bam_file)
output_stream, process = utils_commands.get_input_stream_from_command(command)
nb_reference=0
current_reference=None
first_reads={}
second_reads={}
for line in input_stream:
if line.startswith("@"):
output_stream.write(line)
continue
sam_record = Sam_record(line)
if sam_record.get_reference_name()!=current_reference and not current_reference is None:
#process this consensus
if current_reference!='*':
nb_dups, nb_uniq = find_duplicates(first_reads,second_reads, distance_threshold)
total_nb_uniqs+=nb_uniq
total_nb_dups+=nb_dups
nb_fragment+=len(second_reads)
output_reads(output_stream, first_reads, second_reads)
first_reads={}
second_reads={}
if sam_record.is_second_read():
second_reads[sam_record.get_query_name()]=sam_record
else:
first_reads[sam_record.get_query_name()]=sam_record
nb_reference+=1
def process_single_samtools_run_with_read_group(bam_file,all_contigs_info,samtools_bin):
command="%s view -h -F 132 %s"%(samtools_bin, bam_file)
open_stream, process = get_output_stream_from_command(command)
current_contig=None
sample_name, ext = os.path.splitext(bam_file)
read_groups={}
try:
for line in open_stream:
if not line.startswith("@"):
break
if line.startswith("@RG"):
sp_line = line.strip().split()
rg_id=rg_sample=rg_library=None
for value in sp_line:
if value.startswith("ID"):
rg_id=value[3:]
elif value.startswith("SM"):
rg_sample=value[3:]
elif value.startswith("LB"):
rg_library=value[3:]
if rg_id:
if rg_sample:
read_groups[rg_id]=rg_sample
elif rg_library:
read_groups[rg_id]=rg_library
else:
read_groups[rg_id]=rg_id
all_sample_coverage={}
all_sample_coverage_reads = {}
all_sample_duplicate={}
for sample in read_groups.values():
all_sample_coverage[sample]=Counter()
all_sample_duplicate[sample]=Counter()
all_sample_coverage_reads[sample] = defaultdict(Counter)
#process the first read
sam_record = Sam_record(line.strip())
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
read_sequence = sam_record.get_query_sequence()
loci = get_loci_from_read(sam_record)
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage_reads[read_groups.get(rg_id)][str(loci)][read_sequence] +=1
i=1
#process all the others
for line in open_stream:
i+=1
if i%1000000==0:
print i
sam_record = Sam_record(line.strip())
if current_contig != sam_record.get_reference_name() and current_contig != None:
for sample in read_groups.values():
for loci in all_sample_coverage.get(sample):
alleles = all_sample_coverage_reads[sample].get(loci)
all_contigs_info.add_values(current_contig, loci, all_sample_coverage.get(sample).get(loci, 0),
all_sample_duplicate.get(sample).get(loci, 0), alleles=alleles,
sample=sample)
all_sample_coverage[sample]=Counter()
all_sample_duplicate[sample]=Counter()
all_sample_coverage_reads[sample] = defaultdict(Counter)
current_contig = sam_record.get_reference_name()
if not sam_record.is_unmapped():
rg_id = sam_record.get_tag("RG")
loci = get_loci_from_read(sam_record)
read_sequence = sam_record.get_query_sequence()
if sam_record.is_duplicate_read():
all_sample_duplicate[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage[read_groups.get(rg_id)][str(loci)]+=1
all_sample_coverage_reads[read_groups.get(rg_id)][str(loci)][read_sequence] +=1
if current_contig != None:
for sample in read_groups.values():
for loci in all_sample_coverage.get(sample):
alleles = all_sample_coverage_reads[sample].get(loci)
all_contigs_info.add_values(current_contig, loci, all_sample_coverage.get(sample).get(loci, 0),
all_sample_duplicate.get(sample).get(loci, 0), alleles=alleles,
sample=sample)
all_sample_coverage[sample]=Counter()
all_sample_duplicate[sample]=Counter()
all_sample_coverage_reads[sample] = defaultdict(Counter)
finally:
open_stream.close()
