def assembly_worker(input_queue,
locus_id_value_obj,
gene_id_value_obj,
tss_id_value_obj,
t_id_value_obj,
worker_prefix,
config):
# setup output files
gtf_fileh = None
bed_fileh = None
bedgraph_filehs = [None, None, None]
if config.create_gtf:
gtf_fileh = open(worker_prefix + ".gtf", "w")
if config.create_bed:
bed_fileh = open(worker_prefix + ".bed", "w")
if config.create_bedgraph:
for strand in xrange(0,3):
filename = '%s_%s.bedgraph' % (worker_prefix,
STRAND_NAMES[strand])
fileh = open(filename, 'w')
bedgraph_filehs[strand] = fileh
# process input
while True:
lines = input_queue.get()
if len(lines) == 0:
break
transcripts = transcripts_from_gtf_lines(lines)
# conserve memory
del lines
# assign scores to each transcript
for t in transcripts:
if config.scoring_mode == "unweighted":
t.score = 1.0
elif config.scoring_mode == "gtf_attr":
score = t.attrs.get(config.gtf_score_attr, '0')
t.score = float_check_nan(score)
# assemble
assemble_locus(transcripts,
locus_id_value_obj,
gene_id_value_obj,
tss_id_value_obj,
t_id_value_obj,
config,
gtf_fileh,
bed_fileh,
bedgraph_filehs)
input_queue.task_done()
# cleanup output files
if config.create_bed:
bed_fileh.close()
if config.create_gtf:
gtf_fileh.close()
if config.create_bedgraph:
for fileh in bedgraph_filehs:
fileh.close()
input_queue.task_done()
def assembly_worker(input_queue,
locus_id_value_obj,
gene_id_value_obj,
tss_id_value_obj,
t_id_value_obj,
worker_prefix,
config):
# setup output files
gtf_fileh = None
bed_fileh = None
bedgraph_filehs = [None, None, None]
if config.create_gtf:
gtf_fileh = open(worker_prefix + ".gtf", "w")
if config.create_bed:
bed_fileh = open(worker_prefix + ".bed", "w")
if config.create_bedgraph:
for strand in xrange(0,3):
filename = '%s_%s.bedgraph' % (worker_prefix,
STRAND_NAMES[strand])
fileh = open(filename, 'w')
bedgraph_filehs[strand] = fileh
# process input
while True:
lines = input_queue.get()
if len(lines) == 0:
break
transcripts = transcripts_from_gtf_lines(lines)
# assign scores to each transcript
for t in transcripts:
if config.scoring_mode == "unweighted":
t.score = 1.0
elif config.scoring_mode == "gtf_attr":
score = t.attrs.get(config.gtf_score_attr, '0')
t.score = float_check_nan(score)
# assemble
assemble_locus(transcripts,
locus_id_value_obj,
gene_id_value_obj,
tss_id_value_obj,
t_id_value_obj,
config,
gtf_fileh,
bed_fileh,
bedgraph_filehs)
input_queue.task_done()
# cleanup output files
if config.create_bed:
bed_fileh.close()
if config.create_gtf:
gtf_fileh.close()
if config.create_bedgraph:
for fileh in bedgraph_filehs:
fileh.close()
input_queue.task_done()
def annotate_gtf_worker(input_queue, gtf_file, gtf_sample_attr):
fileh = open(gtf_file, 'w')
while True:
lines = input_queue.get()
if len(lines) == 0:
break
transcripts = transcripts_from_gtf_lines(lines)
annotate_locus(transcripts, gtf_sample_attr)
for t in transcripts:
for f in t.to_gtf_features():
print >>fileh, str(f)
input_queue.task_done()
# explicitly delete large objects
del lines
del transcripts
fileh.close()
input_queue.task_done()
def annotate_gtf_worker(input_queue, gtf_file, gtf_sample_attr):
fileh = open(gtf_file, 'w')
while True:
lines = input_queue.get()
if len(lines) == 0:
break
transcripts = transcripts_from_gtf_lines(lines)
annotate_locus(transcripts, gtf_sample_attr)
for t in transcripts:
for f in t.to_gtf_features():
print >> fileh, str(f)
input_queue.task_done()
# explicitly delete large objects
del lines
del transcripts
fileh.close()
input_queue.task_done()
