def __call__(self, **kw):
fasta_file = kw.get('fastafile')
background = kw.get('background') or None
assembly_id = kw.get('assembly') or None
regions_file = kw.get('regions') or None
motifs_list = kw.get('motifs')
motif_add = kw.get('customMotif')
threshold = float(kw.get('threshold') or 0)
if motifs_list is None: motifs_list = []
if isinstance(motifs_list, basestring): motifs_list = motifs_list.split("|")
if not isinstance(motifs_list, list): motifs_list = [motifs_list]
if background is None and assembly_id is None:
background = self.temporary_path(fname='background.txt')
stats = {'A': 0.25,'C': 0.25, 'G': 0.25, 'T': 0.25}
if fasta_file:
fasta_file = os.path.abspath(fasta_file)
with execution(None) as ex:
stats = fasta_composition(ex,fasta_file,frequency=True)
with open(background,"w") as bgr:
bgr.write(" ".join(["1"]+[str(stats[n]) for n in 'ACGT']))
if assembly_id is not None:
assembly = genrep.Assembly(assembly_id)
else:
if regions_file is not None:
raise ValueError("Please specify an assembly if you specify regions.")
regions_file = os.path.abspath(regions_file)
assembly = None
motifs = {}
if motif_add is not None:
mname = os.path.basename(os.path.splitext(motif_add)[0])
if mname: motifs[mname] = os.path.abspath(motif_add)
for mot in motifs_list:
gid, mname = mot.split(' ')
pwmfile = self.temporary_path()
g.get_motif_PWM(int(gid), mname, output=pwmfile)
motifs[mname] = pwmfile
if len(motifs) == 0:
raise ValueError("Please give at least one motif to scan for")
track_output = self.temporary_path(fname='motif_scan', ext="bed")
with execution(None) as ex:
save_motif_profile( ex, motifs, assembly, regions_file, fasta_file,
background=background, threshold=threshold,
output=track_output,
description=None, via='local' )
self.new_file(track_output, 'motif_track')
return self.display_time()
def __call__(self, **kw):
fasta_file = kw.get('fastafile')
background = kw.get('background') or None
assembly_id = kw.get('assembly') or None
regions_file = kw.get('regions') or None
motifs_list = kw.get('motifs')
motif_add = kw.get('customMotif')
threshold = float(kw.get('threshold') or 0)
if motifs_list is None: motifs_list = []
if isinstance(motifs_list, basestring): motifs_list = motifs_list.split("|")
if not isinstance(motifs_list, list): motifs_list = [motifs_list]
if background is None and assembly_id is None:
background = self.temporary_path(fname='background.txt')
stats = {'A': 0.25,'C': 0.25, 'G': 0.25, 'T': 0.25}
if fasta_file:
with execution(None) as ex:
stats = fasta_composition(ex,fasta_file,frequency=True)
with open(background,"w") as bgr:
bgr.write(" ".join(["1"]+[str(stats[n]) for n in 'ACGT']))
if assembly_id is not None:
assembly = genrep.Assembly(assembly_id)
else:
if regions_file is not None:
raise ValueError("Please specify an assembly if you specify regions.")
regions_file = os.path.abspath(regions_file)
assembly = None
motifs = {}
if motif_add is not None:
mname = os.path.basename(os.path.splitext(motif_add)[0])
if mname: motifs[mname] = os.path.abspath(motif_add)
for mot in motifs_list:
gid, mname = mot.split(' ')
pwmfile = self.temporary_path()
g.get_motif_PWM(int(gid), mname, output=pwmfile)
motifs[mname] = pwmfile
if len(motifs) == 0:
raise ValueError("Please give at least one motif to scan for")
track_output = self.temporary_path(fname='motif_scan', ext="bed")
with execution(None) as ex:
save_motif_profile( ex, motifs, assembly, regions_file, fasta_file,
background=background, threshold=threshold,
output=track_output,
description=None, via='local' )
self.new_file(track_output, 'motif_track')
return self.display_time()
def motif_scan( ex, bedlist, assembly, groups, via, logfile ):
logfile.write("Scanning motifs\n");logfile.flush()
motifbeds = {}
supdir = os.path.split(ex.remote_working_directory)[0]
for gid,bedfile in bedlist.iteritems():
logfile.write("\n%i: "%gid);logfile.flush()
group = groups[gid]
motifs = {}
for mot in group.get('motif',[]):
if os.path.exists(mot):
mname = os.path.basename(os.path.splitext(mot)[0])
motifs[mname] = mot
elif os.path.exists(os.path.join(supdir,mot)):
mname = os.path.basename(os.path.splitext(mot)[0])
motifs[mname] = os.path.join(supdir,mot)
else:
_gnid, mname = mot.split(' ')
motifs[mname] = _gnrp.get_motif_PWM(int(_gnid), mname, output=unique_filename_in())
logfile.write(mname+", ");logfile.flush()
_descr = set_file_descr(group['name']+'_motifs.bed',
type='bed', ucsc='1', step='motifs', groupId=gid)
_out = unique_filename_in()
_hd = "track name='%s_motifs'" %group['name']
motifbeds[gid] = save_motif_profile( ex, motifs, assembly, bedfile,
keep_max_only=True, output=_out,
header=_hd, description=_descr, via=via )
return motifbeds
def motif_scan(ex, bedlist, assembly, groups, via, logfile):
logfile.write("Scanning motifs\n")
logfile.flush()
motifbeds = {}
supdir = os.path.split(ex.remote_working_directory)[0]
for gid, bedfile in bedlist.iteritems():
logfile.write("\n%i: " % gid)
logfile.flush()
group = groups[gid]
motifs = {}
for mot in group.get('motif', []):
if os.path.exists(mot):
mname = os.path.basename(os.path.splitext(mot)[0])
motifs[mname] = mot
elif os.path.exists(os.path.join(supdir, mot)):
mname = os.path.basename(os.path.splitext(mot)[0])
motifs[mname] = os.path.join(supdir, mot)
else:
_gnid, mname = mot.split(' ')
motifs[mname] = _gnrp.get_motif_PWM(
int(_gnid), mname, output=unique_filename_in())
logfile.write(mname + ", ")
logfile.flush()
_descr = set_file_descr(group['name'] + '_motifs.bed',
type='bed',
ucsc='1',
step='motifs',
groupId=gid)
_out = unique_filename_in()
_hd = "track name='%s_motifs'" % group['name']
motifbeds[gid] = save_motif_profile(ex,
motifs,
assembly,
bedfile,
keep_max_only=True,
output=_out,
header=_hd,
description=_descr,
via=via)
return motifbeds