def bowtie(fastq, outfile, config):
"""
Use bowtie to map `fastq`, saving the SAM file as `outfile`. Ensures that
'--sam' is in the parameters.
"""
index = config['index']
params = config['bowtie params'].split()
if ('--sam' not in params) and ('-S' not in params):
params.append('-S')
cmds = ['bowtie']
cmds.extend(params)
cmds.append(index)
cmds.append(fastq)
print outfile
logfn = outfile + '.log'
p = subprocess.Popen(cmds,
stdout=open(outfile, 'w'),
stderr=open(logfn, 'w'),
bufsize=1)
stdout, stderr = p.communicate()
return Result(infiles=fastq,
outfiles=outfile,
cmds=' '.join(cmds),
log=logfn)
def bam2sam(bam, sam):
cmds = ['samtools', 'view', '-h', bam]
p = subprocess.Popen(cmds,
stdout=open(sam, 'w'),
stderr=subprocess.PIPE,
bufsize=1)
stdout, stderr = p.communicate()
return Result(sam, bam, stderr=stderr, cmds=' '.join(cmds))
def clip(fastq, clipped_fastq, config):
adapter = config['adapter']
clipping_report = clipped_fastq + '.clipping_report'
if adapter is None:
fout = open(clipped_fastq, 'w')
fout.write(open(fastq).read())
fout.close()
fout = open(clipping_report, 'w')
fout.write('No adapter specified; %s is a copy of %s' %
(clipped_fastq, fastq))
fout.close()
return Result(fastq, (clipped_fastq, clipping_report))
cmds = [
'fastx_clipper',
'-i',
fastq,
'-o',
clipped_fastq,
'-n', # *keep* Ns
'-a',
adapter,
'-v', # report to stdout
]
p = subprocess.Popen(cmds,
stdout=open(clipping_report, 'w'),
stderr=subprocess.PIPE,
bufsize=1)
stdout, stderr = p.communicate()
failed = False
if p.returncode or not os.path.exists(clipped_fastq):
failed = True
return Result(fastq, (clipped_fastq, clipping_report),
stdout=stdout,
stderr=stderr,
failed=failed,
cmds=' '.join(cmds))
def count(samfile, countfile, config):
cmds = ['htseq-count']
cmds += config['htseq params'].split()
cmds += [samfile, config['gff']]
p = subprocess.Popen(cmds,
stdout=open(countfile, 'w'),
stderr=subprocess.PIPE,
bufsize=1)
stdout, stderr = p.communicate()
failed = p.returncode
return Result(infiles=samfile,
outfiles=countfile,
stderr=stderr,
failed=failed,
cmds=' '.join(cmds))
def filter(sam, outfile, config):
bam = tempfile.mktemp()
filtered_bam = tempfile.mktemp()
result0 = sam2bam(sam, bam)
cmds1 = ['intersectBed', '-abam', bam, '-b', config['filter bed'], '-v']
p = subprocess.Popen(cmds1,
stdout=open(filtered_bam, 'w'),
stderr=subprocess.PIPE,
bufsize=1)
stdout1, stderr1 = p.communicate()
result2 = bam2sam(filtered_bam, outfile)
#os.unlink(bam)
#os.unlink(filtered_bam)
failed = p.returncode
cmds = '\n'.join([result0.cmds, ' '.join(cmds1), result2.cmds])
stderr = '\n'.join([result0.stderr, stderr1, result2.stderr])
return Result(sam, outfile, failed=failed, cmds=cmds, stderr=stderr)
' ',
progressbar.ETA(),
' ',
progressbar.AdaptiveETA(),
])
bar.start()
bar_worker = BarWorker(bar, tasks, task_count)
bar_worker.start()
for i in range(num_workers):
tasks.put(None)
tasks.join()
result_dict = {}
while task_count:
r = Result(results.get())
permutations = (result_dict.get(r.args['theta_name'], {}).get(
r.args['main_method_name'], {}).get(r.args['line_search_name'],
[]))
existing = next((p for p in permutations if p.is_same(r)), None)
if existing:
existing += r
else:
(result_dict.setdefault(r.args['theta_name'], {}).setdefault(
r.args['main_method_name'],
{}).setdefault(r.args['line_search_name'], [])).append(r)
task_count -= 1
for th in result_dict.values():
for mm in th.values():
for ls in mm.values():