class Genome_2bit:
def __init__(self,genome_2bit_file,verbose=False):
self.genome=TwoBitFile(open(genome_2bit_file,'rb'))
self.chr_len=dict()
self.verbose=verbose
for chr_id in self.genome.keys():
self.chr_len[chr_id]=len(self.genome[chr_id])
if verbose:
print 'Genome initializated'
def extract_sequence(self,coordinate,mask_repetitive=False):
if mask_repetitive:
seq= ''.join([mask(c) for c in self.genome[coordinate.chr_id][coordinate.bpstart-1:coordinate.bpend]]).lower()
else:
seq= self.genome[coordinate.chr_id][coordinate.bpstart-1:coordinate.bpend].lower()
if coordinate.strand=='-':
return Sequence.reverse_complement(seq)
else:
return seq
def estimate_background(self):
counting={'a':.0,'c':.0,'g':.0,'t':.0}
all=0.0
for chr_id in self.genome.keys():
if self.verbose:
start_time = time.time()
print 'Counting on:',chr_id
for nt in counting.keys():
count_nt=self.genome[chr_id][:].lower().count(nt)
counting[nt]+=count_nt
all+=count_nt
print 'elapsed:',time.time() - start_time
if self.verbose:
print counting
for nt in counting.keys():
counting[nt]/=all
return counting
def write_meme_background(self,filename):
counting=self.estimate_background()
with open(filename,'w+') as outfile:
for nt in counting.keys():
outfile.write('%s\t%2.4f\n' % (nt,counting[nt]))
def write_chr_len(self,filename):
with open(filename,'w+') as outfile:
for chr_id in self.genome.keys():
outfile.write('%s\t%s\n' % (chr_id,self.chr_len[chr_id]) )
def __init__(self,genome_2bit_file,verbose=False):
self.genome=TwoBitFile(open(genome_2bit_file,'rb'))
self.chr_len=dict()
self.verbose=verbose
for chr_id in self.genome.keys():
self.chr_len[chr_id]=len(self.genome[chr_id])
if verbose:
print 'Genome initializated'
class Genome_2bit:
def __init__(self,genome_2bit_file,verbose=False):
self.genome=TwoBitFile(open(genome_2bit_file,'rb'))
self.chr_len=dict()
self.verbose=verbose
for chr_id in self.genome.keys():
self.chr_len[chr_id]=len(self.genome[chr_id])
if verbose:
print 'Genome initializated'
def estimate_background(self):
counting={'a':.0,'c':.0,'g':.0,'t':.0}
all=0.0
for chr_id in self.genome.keys():
if self.verbose:
start_time = time.time()
print 'Counting on:',chr_id
for nt in counting.keys():
count_nt=self.genome[chr_id][:].lower().count(nt)
counting[nt]+=count_nt
all+=count_nt
print 'elapsed:',time.time() - start_time
if self.verbose:
print counting
for nt in counting.keys():
counting[nt]/=all
return counting
def write_meme_background(self,filename):
counting=self.estimate_background()
with open(filename,'w+') as outfile:
for nt in counting.keys():
outfile.write('%s\t%2.4f\n' % (nt.upper(),counting[nt]))
def write_chr_len(self,filename):
with open(filename,'w+') as outfile:
for chr_id in self.genome.keys():
outfile.write('%s\t%s\n' % (chr_id,self.chr_len[chr_id]) )
class Genome_2bit:
def __init__(self, genome_2bit_file, verbose=False):
self.genome = TwoBitFile(open(genome_2bit_file, 'rb'))
self.chr_len = dict()
self.verbose = verbose
for chr_id in self.genome.keys():
self.chr_len[chr_id] = len(self.genome[chr_id])
if verbose:
print 'Genome initializated'
def estimate_background(self):
counting = {'a': .0, 'c': .0, 'g': .0, 't': .0}
all = 0.0
for chr_id in self.genome.keys():
if self.verbose:
start_time = time.time()
print 'Counting on:', chr_id
for nt in counting.keys():
count_nt = self.genome[chr_id][:].lower().count(nt)
counting[nt] += count_nt
all += count_nt
if self.verbose:
print counting
for nt in counting.keys():
counting[nt] /= all
return counting
def write_meme_background(self, filename):
counting = self.estimate_background()
with open(filename, 'w+') as outfile:
for nt in counting.keys():
outfile.write('%s\t%2.4f\n' % (nt.upper(), counting[nt]))
def write_chr_len(self, filename):
with open(filename, 'w+') as outfile:
for chr_id in self.genome.keys():
outfile.write('%s\t%s\n' % (chr_id, self.chr_len[chr_id]))
def _get_reference_data(twobit_file_name, chrom, low, high):
# Read and return reference data.
try:
with open(twobit_file_name, 'rb') as f:
twobit = TwoBitFile(f)
if chrom in twobit:
seq_data = twobit[chrom].get(int(low), int(high))
return GenomeRegion(chrom=chrom, start=low, end=high, sequence=seq_data)
except OSError as e:
raise e()
def reference(self, trans, dbkey, chrom, low, high):
"""
Return reference data for a build.
"""
self.check_and_reload()
# If there is no dbkey owner, default to current user.
dbkey_owner, dbkey = decode_dbkey(dbkey)
if dbkey_owner:
dbkey_user = trans.sa_session.query(
trans.app.model.User).filter_by(username=dbkey_owner).first()
else:
dbkey_user = trans.user
if not self.has_reference_data(dbkey, dbkey_user):
return None
#
# Get twobit file with reference data.
#
twobit_file_name = None
if dbkey in self.genomes:
# Built-in twobit.
twobit_file_name = self.genomes[dbkey].twobit_file
else:
user_keys = loads(dbkey_user.preferences['dbkeys'])
dbkey_attributes = user_keys[dbkey]
fasta_dataset = trans.sa_session.query(
trans.app.model.HistoryDatasetAssociation).get(
dbkey_attributes['fasta'])
msg = fasta_dataset.convert_dataset(trans, 'twobit')
if msg:
return msg
else:
twobit_dataset = fasta_dataset.get_converted_dataset(
trans, 'twobit')
twobit_file_name = twobit_dataset.file_name
# Read and return reference data.
try:
twobit = TwoBitFile(open(twobit_file_name))
if chrom in twobit:
seq_data = twobit[chrom].get(int(low), int(high))
return GenomeRegion(chrom=chrom,
start=low,
end=high,
sequence=seq_data)
except IOError:
return None
