def __scan_sfs(self):
for group, loci_input in enumerate(self.file.data):
if group >= self.limit:
break
# text sfs file does not have any haplotype pools
loci_input['pool'] = [[]]
if self.chip_file:
cdata = self.chip_file.getdata(loci_input['name'])
if cdata is None or (not is_within(cdata['num_variants'],
self.args.def_valid_locus)):
continue
loci_input['missing'] = [
False if x in cdata['pos'] else True
for x in loci_input['pos']
]
assert len(loci_input['missing']) == len(loci_input['maf'])
else:
loci_input['missing'] = None
if is_within(loci_input['num_variants'],
self.args.def_valid_locus):
if self.data_buffer is None:
self.result.append(
Calculator(self.args, self.unknown_args,
loci_input).run())
else:
self.data_buffer.append(loci_input)
self.pbar.update(group + 1)
def __scan_gdat(self):
'''scan gdat file'''
maf = 'maf'
pos = 'position'
function_score = 'annotation'
# Allow for customized key names in gdat file
try:
for x, y in zip(getColumn(self.args.data[:-5] + '.key', 1),
getColumn(self.args.data[:-5] + '.key', 2)):
if x == 'maf':
maf = y
if x == 'position':
pos = y
if x == 'annotation':
function_score = y
except:
pass
#
for group, item in enumerate(self.groups):
if group >= self.limit:
break
data = self.file.getdata(item)
if self.args.resampling:
data.decompress()
else:
data['haplotype'] = [[]]
try:
loci_input = {
'pool': data['haplotype'],
'name': item,
'maf': list(data[maf]),
'pos': list(data[pos]),
'function_score': list(data[function_score])
}
except KeyError as e:
env.logger.error('Column name {} not found. Please provide [{}.key] file to overwrite column naming conventions.'.\
format(e, self.args.data[:-5]))
continue
loci_input['num_variants'] = len(loci_input['maf'])
if self.chip_file:
cdata = self.chip_file.getdata(item)
if cdata is None or (not is_within(cdata['num_variants'],
self.args.def_valid_locus)):
continue
loci_input['missing'] = [
False if x in cdata['pos'] else True
for x in loci_input['pos']
]
else:
loci_input['missing'] = None
if is_within(loci_input['num_variants'],
self.args.def_valid_locus):
if self.data_buffer is None:
self.result.append(
Calculator(self.args, self.unknown_args,
loci_input).run())
else:
self.data_buffer.append(loci_input)
self.pbar.update(group + 1)
def __scan_gdat(self):
"""scan gdat file"""
maf = "maf"
pos = "position"
function_score = "annotation"
# Allow for customized key names in gdat file
try:
for x, y in zip(getColumn(self.args.data[:-5] + ".key", 1), getColumn(self.args.data[:-5] + ".key", 2)):
if x == "maf":
maf = y
if x == "position":
pos = y
if x == "annotation":
function_score = y
except:
pass
#
for group, item in enumerate(self.groups):
if group >= self.limit:
break
data = self.file.getdata(item)
if self.args.resampling:
data.decompress()
else:
data["haplotype"] = [[]]
try:
loci_input = {
"pool": data["haplotype"],
"name": item,
"maf": list(data[maf]),
"pos": list(data[pos]),
"function_score": list(data[function_score]),
}
except KeyError as e:
env.logger.error(
"Column name {} not found. Please provide [{}.key] file to overwrite column naming conventions.".format(
e, self.args.data[:-5]
)
)
continue
loci_input["num_variants"] = len(loci_input["maf"])
if self.chip_file:
cdata = self.chip_file.getdata(item)
if cdata is None or (not is_within(cdata["num_variants"], self.args.def_valid_locus)):
continue
loci_input["missing"] = [False if x in cdata["pos"] else True for x in loci_input["pos"]]
else:
loci_input["missing"] = None
if is_within(loci_input["num_variants"], self.args.def_valid_locus):
if self.data_buffer is None:
self.result.append(Calculator(self.args, self.unknown_args, loci_input).run())
else:
self.data_buffer.append(loci_input)
self.pbar.update(group + 1)
def __scan_gdat(self):
'''scan gdat file'''
maf = 'maf'
pos = 'position'
function_score = 'annotation'
# Allow for customized key names in gdat file
try:
for x, y in zip(getColumn(self.args.data[:-5] + '.key', 1),
getColumn(self.args.data[:-5] + '.key', 2)):
if x == 'maf':
maf = y
if x == 'position':
pos = y
if x == 'annotation':
function_score = y
except:
pass
#
for group, item in enumerate(self.groups):
if group >= self.limit:
break
data = self.file.getdata(item)
if self.args.resampling:
data.decompress()
else:
data['haplotype'] = [[]]
try:
loci_input = {'pool':data['haplotype'], 'name':item,
'maf':list(data[maf]), 'pos':list(data[pos]),
'function_score':list(data[function_score])}
except KeyError as e:
env.logger.error('Column name {} not found. Please provide [{}.key] file to overwrite column naming conventions.'.\
format(e, self.args.data[:-5]))
continue
loci_input['num_variants'] = len(loci_input['maf'])
if self.chip_file:
cdata = self.chip_file.getdata(item)
if cdata is None or (not is_within(cdata['num_variants'], self.args.def_valid_locus)):
continue
loci_input['missing'] = [False if x in cdata['pos'] else True for x in loci_input['pos']]
else:
loci_input['missing'] = None
if is_within(loci_input['num_variants'], self.args.def_valid_locus):
if self.data_buffer is None:
self.result.append(Calculator(self.args, self.unknown_args,loci_input).run())
else:
self.data_buffer.append(loci_input)
self.pbar.update(group + 1)
def __update_direction(self):
d = ['d'] * self.data['num_variants']
if self.args.def_protective is not None:
upper = max(self.args.def_protective)
d = ['p' if is_within(y, self.args.def_protective) else 'd' for y in self.data['function_score']]
# adjust wrt function for neutral sites
d = ['n' if y == 's' else x for x, y in zip(d, self.data['function_class'])]
return d
def __update_function(self):
f = ['ns'] * self.data['num_variants']
if self.args.def_neutral is not None:
f = [
's' if is_within(y, self.args.def_neutral) else 'ns'
for y in self.data['function_score']
]
return f
def __update_direction(self):
d = ["d"] * self.data["num_variants"]
if self.args.def_protective is not None:
upper = max(self.args.def_protective)
d = ["p" if is_within(y, self.args.def_protective) else "d" for y in self.data["function_score"]]
# adjust wrt function for neutral sites
d = ["n" if y == "s" else x for x, y in zip(d, self.data["function_class"])]
return d
def __scan_sfs(self):
for group, loci_input in enumerate(self.file.data):
if group >= self.limit:
break
# text sfs file does not have any haplotype pools
loci_input['pool'] = [[]]
if self.chip_file:
cdata = self.chip_file.getdata(loci_input['name'])
if cdata is None or (not is_within(cdata['num_variants'], self.args.def_valid_locus)):
continue
loci_input['missing'] = [False if x in cdata['pos'] else True for x in loci_input['pos']]
assert len(loci_input['missing']) == len(loci_input['maf'])
else:
loci_input['missing'] = None
if is_within(loci_input['num_variants'], self.args.def_valid_locus):
if self.data_buffer is None:
self.result.append(Calculator(self.args, self.unknown_args,loci_input).run())
else:
self.data_buffer.append(loci_input)
self.pbar.update(group + 1)
def __update_direction_random(self):
'''based on self.swap['direction']'''
d = copy.deepcopy(self.swap['direction'])
if self.args.proportion_detrimental is not None:
# mark some as neutral
p = 1 - self.args.proportion_detrimental
d = ['n' if (rng.random() < p and x == 'd' and not is_within(y, self.args.def_disruptive)) else x
for x,y in zip(d, self.data['function_score'])]
if self.args.proportion_protective is not None:
# mark some as neutral
p = 1 - self.args.proportion_protective
d = ['n' if (rng.random() < p and x == 'p') else x for x in d]
return d
def __update_direction(self):
d = ['d'] * self.data['num_variants']
if self.args.def_protective is not None:
upper = max(self.args.def_protective)
d = [
'p' if is_within(y, self.args.def_protective) else 'd'
for y in self.data['function_score']
]
# adjust wrt function for neutral sites
d = [
'n' if y == 's' else x
for x, y in zip(d, self.data['function_class'])
]
return d
def __update_direction_random(self):
"""based on self.swap['direction']"""
d = copy.deepcopy(self.swap["direction"])
if self.args.proportion_detrimental is not None:
# mark some as neutral
p = 1 - self.args.proportion_detrimental
d = [
"n" if (rng.random() < p and x == "d" and not is_within(y, self.args.def_disruptive)) else x
for x, y in zip(d, self.data["function_score"])
]
if self.args.proportion_protective is not None:
# mark some as neutral
p = 1 - self.args.proportion_protective
d = ["n" if (rng.random() < p and x == "p") else x for x in d]
return d
def __update_direction_random(self):
'''based on self.swap['direction']'''
d = copy.deepcopy(self.swap['direction'])
if self.args.proportion_detrimental is not None:
# mark some as neutral
p = 1 - self.args.proportion_detrimental
d = [
'n' if (rng.random() < p and x == 'd'
and not is_within(y, self.args.def_disruptive)) else x
for x, y in zip(d, self.data['function_score'])
]
if self.args.proportion_protective is not None:
# mark some as neutral
p = 1 - self.args.proportion_protective
d = ['n' if (rng.random() < p and x == 'p') else x for x in d]
return d
def __update_function(self):
f = ['ns'] * self.data['num_variants']
if self.args.def_neutral is not None:
f = ['s' if is_within(y, self.args.def_neutral) else 'ns' for y in self.data['function_score']]
return f
def __update_function(self):
f = ["ns"] * self.data["num_variants"]
if self.args.def_neutral is not None:
f = ["s" if is_within(y, self.args.def_neutral) else "ns" for y in self.data["function_score"]]
return f
