def set_read1_consensus_to_read2(input_stream, output_stream):
#get the header
line = input_stream.readline()
while line.startswith("@"):
output_stream.write(line)
line = input_stream.readline()
prev_read=Sam_record(line)
for line in input_stream:
read=Sam_record(line)
if prev_read and read.get_query_name() == prev_read.get_query_name():
if read.is_second_read() and prev_read.is_first_read():
read1=prev_read
read2=read
else:
read2=prev_read
read1=read
if not read1.is_unmapped():
read2.set_reference_name(read1.get_reference_name())
read2.set_unmapped_flag(False)
read2.set_position(1)
read2.set_cigar_string("%sM"%len(read2.get_query_sequence()))
output_stream.write(str(read1))
output_stream.write(str(read2))
prev_read=None
elif prev_read:
output_stream.write(str(prev_read))
prev_read=read
else:
prev_read=read
def load_from_sites_generator(stream):
all_unmatched_read1={}
all_unmatched_read2={}
count_line=0
for line in stream:
count_line+=1
if count_line%10000==0:
sys.stderr.write('%s %s %s\n'%(count_line, len(all_unmatched_read1), len(all_unmatched_read2)))
sam_record = Sam_record(line)
if sam_record.is_first_read():
sam_record_r1 = sam_record
sam_record_r2 = all_unmatched_read2.pop(sam_record.get_query_name(),None)
if not sam_record_r2:
all_unmatched_read1[sam_record.get_query_name()]=sam_record
else:
sam_record_r2 = sam_record
sam_record_r1 = all_unmatched_read1.pop(sam_record.get_query_name(),None)
if not sam_record_r1:
all_unmatched_read2[sam_record.get_query_name()]=sam_record
if sam_record_r1 and sam_record_r2:
yield ((sam_record_r1,sam_record_r2))
command ="%s view -bS - | %s sort - %s"%(samtools_bin, samtools_bin, output_bam_file)
logging.info(command)
output_stream,process_output= utils_commands.get_input_stream_from_command(command)
#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()
