def __call__(self, **kw):
b2wargs = []
control = None
if kw.get('control'):
control = kw['control']
b2wargs = ["-c", str(control)]
bamfile = track(kw['sample'], format='bam')
nreads = int(kw.get('normalization') or -1)
if nreads < 0:
if control is None:
nreads = len(set((t[4] for t in bamfile.read())))
else:
b2wargs += ["-r"]
merge_strands = int(kw.get('merge_strands') or -1)
if merge_strands >= 0:
suffixes = ["merged"]
else:
suffixes = ["fwd", "rev"]
read_extension = int(kw.get('read_extension') or -1)
output = self.temporary_path(fname='density_')
with execution(None) as ex:
files = bam_to_density(ex, kw['sample'], output,
nreads=nreads, merge=merge_strands,
read_extension=read_extension,
sql=True, args=b2wargs)
for n, x in enumerate(files):
tout = track(x, format='sql', fields=['start', 'end', 'score'],
chrmeta=bamfile.chrmeta, info={'datatype': 'quantitative'})
tout.save()
self.new_file(x, 'density_' + suffixes[n])
return 1
def __call__(self, **kw):
b2wargs = []
control = None
sample = kw.get("sample")
assert os.path.exists(str(sample)), "Bam file not found: '%s'." % sample
if kw.get('control'):
control = kw['control']
b2wargs = ["-c", str(control)]
assert os.path.exists(str(control)), "Control file not found: '%s'." % control
control = os.path.abspath(control)
sample = os.path.abspath(sample)
nreads = int(kw.get('normalization') or -1)
bamfile = track(sample, format='bam')
if nreads < 0:
if control is None:
nreads = len(set((t[4] for t in bamfile.read())))
else:
b2wargs += ["-r"]
merge_strands = int(kw.get('merge_strands') or -1)
read_extension = int(kw.get('read_extension') or -1)
output = self.temporary_path(fname='density_')
format = kw.get("format", "sql")
with execution(None) as ex:
files = bam_to_density(ex, sample, output,
nreads=nreads, merge=merge_strands,
read_extension=read_extension,
sql=True, args=b2wargs)
if merge_strands >= 0:
suffixes = ["merged"]
else:
suffixes = ["fwd", "rev"]
for n, x in enumerate(files):
tsql = track(x, format='sql', fields=['start', 'end', 'score'],
chrmeta=bamfile.chrmeta,
info={'datatype': 'quantitative'})
tsql.save()
if format == "sql":
outname = x
else:
outname = os.path.splitext(x)[0]+"."+format
convert(x, outname, mode="overwrite")
self.new_file(outname, 'density_'+suffixes[n])
return self.display_time()
def __call__(self, **kw):
b2wargs = []
control = None
samples = kw.get('BamMulti', {}).get('sample', [])
if not isinstance(samples, list): samples = [samples]
samples = [os.path.abspath(s) for s in samples if os.path.exists(s)]
if kw.get('control'):
control = kw['control']
b2wargs = ["-c", str(control)]
assert os.path.exists(
str(control)), "Control file not found: '%s'." % control
control = os.path.abspath(control)
try:
nreads = int(kw.get('normalization'))
except (ValueError, TypeError):
nreads = -1
bamfiles = [track(s, format='bam') for s in samples]
if nreads < 0:
_nreads = [0] * len(samples)
if control is not None:
b2wargs += ["-r"]
else:
_nreads = [nreads for s in samples]
try:
merge_strands = int(kw.get('merge_strands'))
except (ValueError, TypeError):
merge_strands = -1
try:
read_extension = int(kw.get('read_extension'))
except (ValueError, TypeError):
read_extension = -1
output = [
self.temporary_path(fname=b.name + '_density_') for b in bamfiles
]
format = kw.get("format", "sql")
with execution(None) as ex:
files = [
bam_to_density(ex,
s,
output[n],
nreads=_nreads[n],
merge=merge_strands,
read_extension=read_extension,
sql=True,
args=b2wargs) for n, s in enumerate(samples)
]
info = {'datatype': 'quantitative', 'read_extension': read_extension}
if merge_strands >= 0:
suffixes = ["merged"]
info['shift'] = merge_strands
else:
suffixes = ["fwd", "rev"]
chrmeta = bamfiles[0].chrmeta
for suf in suffixes:
all_s_files = [
x for y in files for x in y if x.endswith(suf + ".sql")
]
if len(all_s_files) > 1:
x = self.temporary_path(fname="Density_average_" + suf +
".sql")
tsql = track(x,
fields=['start', 'end', 'score'],
chrmeta=chrmeta,
info={'datatype': 'quantitative'})
insql = []
for f in all_s_files:
t = track(f, format='sql', chrmeta=chrmeta)
t.save()
insql.append(t)
for c in tsql.chrmeta:
tsql.write(merge_scores([t.read(c) for t in insql]),
chrom=c)
else:
x = all_s_files[0]
tsql = track(x,
format='sql',
fields=['start', 'end', 'score'],
chrmeta=chrmeta,
info=info)
tsql.save()
if format in [None, "sql"]:
outname = x
else:
outname = os.path.splitext(x)[0] + "." + format
convert(x, outname, mode="overwrite")
self.new_file(outname, 'density_' + suf)
return self.display_time()
def __call__(self, **kw):
b2wargs = []
control = None
samples = kw.get('BamMulti',{}).get('sample', [])
if not isinstance(samples, list): samples = [samples]
samples = [os.path.abspath(s) for s in samples if os.path.exists(s)]
if kw.get('control'):
control = kw['control']
b2wargs = ["-c", str(control)]
assert os.path.exists(str(control)), "Control file not found: '%s'." % control
control = os.path.abspath(control)
try:
nreads = int(kw.get('normalization'))
except (ValueError, TypeError):
nreads = -1
bamfiles = [track(s, format='bam') for s in samples]
if nreads < 0:
_nreads = [0]*len(samples)
if control is not None:
b2wargs += ["-r"]
else:
_nreads = [nreads for s in samples]
try:
merge_strands = int(kw.get('merge_strands'))
except (ValueError, TypeError):
merge_strands = -1
try:
read_extension = int(kw.get('read_extension'))
except (ValueError, TypeError):
read_extension = -1
output = [self.temporary_path(fname=b.name+'_density_') for b in bamfiles]
format = kw.get("format", "sql")
with execution(None) as ex:
files = [bam_to_density( ex, s, output[n], nreads=_nreads[n],
merge=merge_strands,
read_extension=read_extension,
sql=True, args=b2wargs )
for n,s in enumerate(samples)]
info = {'datatype': 'quantitative', 'read_extension': read_extension}
if merge_strands >= 0:
suffixes = ["merged"]
info['shift'] = merge_strands
else:
suffixes = ["fwd", "rev"]
chrmeta = bamfiles[0].chrmeta
for suf in suffixes:
all_s_files = [x for y in files for x in y if x.endswith(suf+".sql")]
if len(all_s_files) > 1:
x = self.temporary_path(fname="Density_average_"+suf+".sql")
tsql = track( x, fields=['start', 'end', 'score'],
chrmeta=chrmeta, info={'datatype': 'quantitative'} )
insql = []
for f in all_s_files:
t = track(f, format='sql', chrmeta=chrmeta)
t.save()
insql.append(t)
for c in tsql.chrmeta:
tsql.write(merge_scores([t.read(c) for t in insql]),chrom=c)
else:
x = all_s_files[0]
tsql = track( x, format='sql', fields=['start', 'end', 'score'],
chrmeta=chrmeta, info=info )
tsql.save()
if format in [None,"sql"]:
outname = x
else:
outname = os.path.splitext(x)[0]+"."+format
convert(x, outname, mode="overwrite")
self.new_file(outname, 'density_'+suf)
return self.display_time()