def next(self):
next_record = None
used_records = [False for i in range(0, len(self.next_records))]
for vcf_index in range(0, len(self.next_records)):
if self.next_records[vcf_index] is None:
try:
self.next_records[vcf_index] = self.vcf_list[vcf_index].next()
except StopIteration:
continue
if next_record is None or (self.ascii_sort and next_record > self.next_records[vcf_index]) or\
(not self.ascii_sort and (next_record.chrnumber, next_record.POS) >
(self.next_records[vcf_index].chrnumber, self.next_records[vcf_index].POS)):
if next_record is not None:
used_records = [False for i in range(0, len(self.next_records))]
next_record = self.next_records[vcf_index]
used_records[vcf_index] = True
elif (next_record.CHROM, next_record.POS) == \
(self.next_records[vcf_index].CHROM, self.next_records[vcf_index].POS):
if next_record.merge(self.next_records[vcf_index]):
used_records[vcf_index] = True
if not any(record is not None for record in self.next_records):
raise StopIteration
for i in range(0, len(used_records)):
if used_records[i]:
self.next_records[i] = None
for sample in self.samples:
if next_record.get_sample(sample) is None:
samp_fmt = self._parse_sample_format('DP:FREQ')
my_freqs = [None]
my_freqs.extend([None for alt in next_record.ALT])
samp_data = [None, my_freqs]
next_record.add_call(_Call(next_record, sample, samp_fmt(*samp_data)))
next_record._sort_samples(sort=self.samples)
return next_record
def _parse_samples(self, samples, samp_fmt, site):
'''Parse a sample entry according to the format specified in the FORMAT
column.
NOTE: this method has a cython equivalent and care must be taken
to keep the two methods equivalent
'''
# check whether we already know how to parse this format
if samp_fmt not in self._format_cache:
self._format_cache[samp_fmt] = self._parse_sample_format(samp_fmt)
samp_fmt = self._format_cache[samp_fmt]
if cparse:
return cparse.parse_samples(
self.samples, samples, samp_fmt, samp_fmt._types, samp_fmt._nums, site)
samp_data = []
_map = self._map
nfields = len(samp_fmt._fields)
for name, sample in itertools.izip(self.samples, samples):
# parse the data for this sample
sampdat = [None] * nfields
for i, vals in enumerate(sample.split(':')):
# short circuit the most common
if samp_fmt._fields[i] == 'GT':
sampdat[i] = vals
continue
elif vals == ".":
sampdat[i] = None
continue
entry_num = samp_fmt._nums[i]
entry_type = samp_fmt._types[i]
# we don't need to split single entries
if entry_num == 1 or ',' not in vals:
if entry_type == 'Integer':
try:
sampdat[i] = int(vals)
except ValueError:
sampdat[i] = float(vals)
elif entry_type == 'Float':
sampdat[i] = float(vals)
else:
sampdat[i] = vals
if entry_num != 1:
sampdat[i] = (sampdat[i])
continue
vals = vals.split(',')
if entry_type == 'Integer':
try:
sampdat[i] = _map(int, vals)
except ValueError:
sampdat[i] = _map(float, vals)
elif entry_type == 'Float' or entry_type == 'Numeric':
sampdat[i] = _map(float, vals)
else:
sampdat[i] = vals
# create a call object
call = _Call(site, name, samp_fmt(*sampdat))
samp_data.append(call)
return samp_data
def _parse_samples(self, samples, samp_fmt, site):
'''Parse a sample entry according to the format specified in the FORMAT
column.
NOTE: this method has a cython equivalent and care must be taken
to keep the two methods equivalent
'''
# check whether we already know how to parse this format
if samp_fmt not in self._format_cache:
self._format_cache[samp_fmt] = self._parse_sample_format(samp_fmt)
samp_fmt = self._format_cache[samp_fmt]
if cparse:
return cparse.parse_samples(
self.samples, samples, samp_fmt, samp_fmt._types, samp_fmt._nums, site)
samp_data = []
_map = self._map
nfields = len(samp_fmt._fields)
for name, sample in itertools.izip(self.samples, samples):
# parse the data for this sample
sampdat = [None] * nfields
for i, vals in enumerate(sample.split(':')):
# short circuit the most common
if vals == '.' or vals == './.':
sampdat[i] = None
continue
entry_num = samp_fmt._nums[i]
entry_type = samp_fmt._types[i]
# we don't need to split single entries
if entry_num == 1 or ',' not in vals:
if entry_type == 'Integer':
try:
sampdat[i] = int(vals)
except ValueError:
sampdat[i] = float(vals)
elif entry_type == 'Float':
sampdat[i] = float(vals)
else:
sampdat[i] = vals
if entry_num != 1:
sampdat[i] = (sampdat[i])
continue
vals = vals.split(',')
if entry_type == 'Integer':
try:
sampdat[i] = _map(int, vals)
except ValueError:
sampdat[i] = _map(float, vals)
elif entry_type == 'Float' or entry_type == 'Numeric':
sampdat[i] = _map(float, vals)
else:
sampdat[i] = vals
# create a call object
call = _Call(site, name, samp_fmt(*sampdat))
samp_data.append(call)
return samp_data
def _parse_samples(self, samples, samp_fmt, site, EntryDbID):
'''Parse a sample entry according to the format specified in the FORMAT
column.
NOTE: this method has a cython equivalent and care must be taken
to keep the two methods equivalent
'''
# check whether we already know how to parse this format
# TODO at some point add DB
# TODO 1 remove print when ready
print samp_fmt
individGeno = samp_fmt.split(":")
IndividualFunctions = []
CustomGeno = []
#Supported
#TODO individual
#JULIA: AD DP, GLE, GL, EC GP, GT, FT, PL, GQ, HQ, PS, PQ
for genotype in individGeno:
if ( genotype == "AD" ):
IndividualFunctions.append(self.db.createAD)
elif (genotype == "DP" ):
IndividualFunctions.append(self.db.createDP)
elif (genotype == "EC" ):
IndividualFunctions.append(self.db.createEC)
elif (genotype == "FT" ):
IndividualFunctions.append(self.db.createFT)
elif (genotype == "GL" ):
IndividualFunctions.append(self.db.createGL)
elif (genotype == "GLE" ):
IndividualFunctions.append(self.db.createGLE)
elif (genotype == "GP" ):
IndividualFunctions.append(self.db.createGP)
elif (genotype == "GQ" ):
IndividualFunctions.append(self.db.createGQ)
elif (genotype == "GT" ):
IndividualFunctions.append(self.db.createGT)
elif (genotype == "HQ" ):
IndividualFunctions.append(self.db.createHQ)
elif (genotype == "PL" ):
IndividualFunctions.append(self.db.createPL)
elif (genotype == "PQ" ):
IndividualFunctions.append(self.db.createPQ)
elif (genotype == "PS" ):
IndividualFunctions.append(self.db.createPS)
else:
CustomGeno.append( genotype )
IndividualFunctions.append(self.db.createIndividualDefault)
if samp_fmt not in self._format_cache:
self._format_cache[samp_fmt] = self._parse_sample_format(samp_fmt)
samp_fmt = self._format_cache[samp_fmt]
if cparse:
return cparse.parse_samples(
self.samples, samples, samp_fmt, samp_fmt._types, samp_fmt._nums, site)
samp_data = []
_map = self._map
nfields = len(samp_fmt._fields)
indNumber = 0;
indId = 0
for name, sample in itertools.izip(self.samples, samples):
customCount = 0
indId = self.db.createIndividualEntry( EntryDbID, indNumber );
if indId == -1:
print "Failed to create individual entry"
indNumber += 1
# parse the data for this sample
sampdat = [None] * nfields
for i, vals in enumerate(sample.split(':')):
#TODO individ here
# short circuit the most common
#MINE
if ( IndividualFunctions[i] == self.db.createIndividualDefault ):
IndividualFunctions[i]( CustomGeno[customCount], indId, vals )
customCount += 1
else:
IndividualFunctions[i]( indId, vals )
if vals == '.' or vals == './.':
sampdat[i] = None
continue
entry_num = samp_fmt._nums[i]
entry_type = samp_fmt._types[i]
# we don't need to split single entries
if entry_num == 1 or ',' not in vals:
#TODO: add DB upload and subroutines
if entry_type == 'Integer':
sampdat[i] = int(vals)
elif entry_type == 'Float':
sampdat[i] = float(vals)
else:
sampdat[i] = vals
if entry_num != 1:
sampdat[i] = (sampdat[i])
continue
vals = vals.split(',')
if entry_type == 'Integer':
sampdat[i] = _map(int, vals)
elif entry_type == 'Float' or entry_type == 'Numeric':
sampdat[i] = _map(float, vals)
else:
sampdat[i] = vals
# create a call object
call = _Call(site, name, samp_fmt(*sampdat))
samp_data.append(call)
return samp_data