def set_family_info(self):
"""
Extract the relevant genotype filters, as well all labels
for each family in the database.
"""
self.families = families = Family.from_cursor(self.gq.c).values()
args = self.args
self.family_ids = []
self.family_masks = []
kwargs = {'only_affected': not getattr(self.args, "allow_unaffected", False),
'min_gq': args.min_gq}
if self.model == "mendel_violations":
kwargs = {'only_affected': self.args.only_affected}
if self.model != "comp_het" and self.model != "mendel_violations":
kwargs['strict'] = not self.args.lenient
elif self.model == "comp_het":
kwargs['pattern_only'] = self.args.pattern_only
requested_fams = None if not args.families else set(args.families.split(","))
for family in families:
if requested_fams is None or family.family_id in requested_fams:
# e.g. family.auto_rec(gt_ll, min_depth)
family_filter = getattr(family, self.model)(gt_ll=self.args.gt_phred_ll,
min_depth=self.args.min_sample_depth,
**kwargs)
else:
family_filter = 'False'
self.family_masks.append(family_filter)
self.family_ids.append(family.family_id)
def set_family_info(self):
"""
Extract the relevant genotype filters, as well all labels
for each family in the database.
"""
self.families = families = Family.from_cursor(self.gq.c).values()
args = self.args
self.family_ids = []
self.family_masks = []
kwargs = {
'only_affected': not getattr(self.args, "allow_unaffected", False)
}
if self.model == "mendel_violations":
kwargs = {'only_affected': self.args.only_affected}
if self.model != "comp_het" and self.model != "mendel_violations":
kwargs['strict'] = not self.args.lenient
elif self.model == "comp_het":
kwargs['pattern_only'] = self.args.pattern_only
requested_fams = None if not args.families else set(
args.families.split(","))
for family in families:
if requested_fams is None or family.family_id in requested_fams:
# e.g. family.auto_rec(gt_ll, min_depth)
family_filter = getattr(family, self.model)(
gt_ll=self.args.gt_phred_ll,
min_depth=self.args.min_sample_depth,
**kwargs)
else:
family_filter = 'False'
self.family_masks.append(family_filter)
self.family_ids.append(family.family_id)
def get_families(db, selected_families=None):
"""
Query the samples table to return a list of Family
objects that each contain all of the Subjects in a Family.
"""
conn = sqlite3.connect(db)
conn.isolation_level = None
conn.row_factory = sqlite3.Row
c = conn.cursor()
families_dict = Family.from_cursor(c)
# if the user has specified a set of selected families
# to which the analysis should be restricted, then
# first sanity check that the family ids they specified are valid.
if selected_families is not None:
for family in selected_families.split(','):
if family not in families_dict:
sys.exit("ERROR: family \"%s\" is not a valid family_id\n" % family)
families = []
for fam in families_dict:
if selected_families is None or fam in selected_families:
families.append(families_dict[fam])
return families
def get_families(db, selected_families=None):
"""
Query the samples table to return a list of Family
objects that each contain all of the Subjects in a Family.
"""
conn = sqlite3.connect(db)
conn.isolation_level = None
conn.row_factory = sqlite3.Row
c = conn.cursor()
families_dict = Family.from_cursor(c)
# if the user has specified a set of selected families
# to which the analysis should be restricted, then
# first sanity check that the family ids they specified are valid.
if selected_families is not None:
for family in selected_families.split(','):
if family not in families_dict:
sys.exit("ERROR: family \"%s\" is not a valid family_id\n" %
family)
families = []
for fam in families_dict:
if selected_families is None or fam in selected_families:
families.append(families_dict[fam])
return families
def candidates(self):
args = self.args
self.gq._connect_to_database()
fams = self.fams = Family.from_cursor(self.gq.conn)
if args.families:
fams = {f: fam for f, fam in fams.items() if f in set(args.families.split(","))}
for grp, li in self.gen_candidates('gene'):
samples_w_hetpair = defaultdict(list)
sites, strs = [], []
for row in li:
gt_types, gt_bases, gt_phases = row['gt_types'], row['gts'], row['gt_phases']
site = Site(row)
site.gt_phases, site.gt_bases, site.gt_types = gt_phases, gt_bases, gt_types
sites.append((str(site), site))
for family_id, fam in fams.items():
# if a site has been deemed "impossible", we store and then
# skip it to avoid compuational overhead on it multiple times.
impossible_sites = {}
for i, (strsite1, site1) in enumerate(sites[:-1], start=1):
if strsite1 in impossible_sites:
continue
for (strsite2, site2) in sites[i:]:
if strsite2 in impossible_sites:
continue
ch = fam.comp_het_pair(site1.gt_types, site1.gt_bases,
site2.gt_types, site2.gt_bases,
site1.gt_phases, site2.gt_phases,
ref1=site1.row['ref'],
alt1=site1.row['alt'],
ref2=site2.row['ref'],
alt2=site2.row['alt'],
allow_unaffected=args.allow_unaffected,
fast_mode=True,
pattern_only=args.pattern_only)
if ch.get('impossible') == 'site1':
impossible_sites[strsite1] = True
break
if ch.get('impossible') == 'site2':
impossible_sites[strsite2] = True
if not ch['candidate']: continue
samples_w_hetpair[(site1, site2)].append(ch)
yield grp, self.filter_candidates(samples_w_hetpair)
def candidates(self):
args = self.args
self.gq._connect_to_database()
fams = self.fams = Family.from_cursor(self.gq.c)
if args.families:
fams = {
f: fam
for f, fam in fams.items()
if f in set(args.families.split(","))
}
for grp, li in self.gen_candidates('gene'):
samples_w_hetpair = defaultdict(list)
sites = []
for row in li:
gt_types, gt_bases, gt_phases = row['gt_types'], row[
'gts'], row['gt_phases']
site = Site(row)
site.gt_phases, site.gt_bases, site.gt_types = gt_phases, gt_bases, gt_types
sites.append(site)
for i, site1 in enumerate(sites[:-1], start=1):
for site2 in sites[i:]:
for family_id, fam in fams.items():
ch = fam.comp_het_pair(
site1.gt_types,
site1.gt_bases,
site2.gt_types,
site2.gt_bases,
site1.gt_phases,
site2.gt_phases,
ref1=site1.row['ref'],
alt1=site1.row['alt'],
ref2=site2.row['ref'],
alt2=site2.row['alt'],
allow_unaffected=args.allow_unaffected,
fast_mode=True,
pattern_only=args.pattern_only)
if not ch['candidate']: continue
samples_w_hetpair[(site1, site2)].append(ch)
yield grp, self.filter_candidates(samples_w_hetpair)
def candidates(self):
args = self.args
self.gq._connect_to_database()
fams = self.fams = Family.from_cursor(self.gq.c)
if args.families:
fams = {
f: fam
for f, fam in fams.items()
if f in set(args.families.split(","))
}
for grp, li in self.gen_candidates('gene'):
samples_w_hetpair = defaultdict(list)
sites = []
for row in li:
gt_types, gt_bases, gt_phases = row['gt_types'], row[
'gts'], row['gt_phases']
site = Site(row)
site.gt_phases, site.gt_bases, site.gt_types = gt_phases, gt_bases, gt_types
sites.append(site)
for i, site1 in enumerate(sites[:-1], start=1):
for site2 in sites[i:]:
for family_id, fam in fams.items():
ch = fam.comp_het_pair(
site1.gt_types,
site1.gt_bases,
site2.gt_types,
site2.gt_bases,
site1.gt_phases,
site2.gt_phases,
ref1=site1.row['ref'],
alt1=site1.row['alt'],
ref2=site2.row['ref'],
alt2=site2.row['alt'],
allow_unaffected=args.allow_unaffected,
fast_mode=True,
pattern_only=args.pattern_only)
if not ch['candidate']: continue
samples_w_hetpair[(site1, site2)].append(ch)
yield grp, self.filter_candidates(samples_w_hetpair)
def get_families(db, selected_families=None):
"""
Query the samples table to return a list of Family
objects that each contain all of the Subjects in a Family.
"""
conn, metadata = database.get_session_metadata(db)
families_dict = Family.from_cursor(conn)
# if the user has specified a set of selected families
# to which the analysis should be restricted, then
# first sanity check that the family ids they specified are valid.
if selected_families is not None:
for family in selected_families.split(','):
if family not in families_dict:
raise ValueError("Family \"%s\" is not a valid family_id\n" % family)
families = []
for fam in families_dict:
if selected_families is None or fam in selected_families:
families.append(families_dict[fam])
return families
def get_families(db, selected_families=None):
"""
Query the samples table to return a list of Family
objects that each contain all of the Subjects in a Family.
"""
conn, metadata = database.get_session_metadata(db)
families_dict = Family.from_cursor(conn)
# if the user has specified a set of selected families
# to which the analysis should be restricted, then
# first sanity check that the family ids they specified are valid.
if selected_families is not None:
for family in selected_families.split(','):
if family not in families_dict:
raise ValueError("Family \"%s\" is not a valid family_id\n" % family)
families = []
for fam in families_dict:
if selected_families is None or fam in selected_families:
families.append(families_dict[fam])
return families
def candidates(self):
args = self.args
self.gq._connect_to_database()
fams = self.fams = Family.from_cursor(self.gq.conn)
if args.families:
fams = {
f: fam
for f, fam in fams.items()
if f in set(args.families.split(","))
}
for grp, li in self.gen_candidates('gene'):
samples_w_hetpair = defaultdict(list)
sites, strs = [], []
for row in li:
gt_types, gt_bases, gt_phases = row['gt_types'], row[
'gts'], row['gt_phases']
site = Site(row)
site.gt_phases, site.gt_bases, site.gt_types = gt_phases, gt_bases, gt_types
sites.append((str(site), site))
for family_id, fam in fams.items():
# if a site has been deemed "impossible", we store and then
# skip it to avoid compuational overhead on it multiple times.
impossible_sites = {}
for i, (strsite1, site1) in enumerate(sites[:-1], start=1):
if strsite1 in impossible_sites:
continue
for (strsite2, site2) in sites[i:]:
if strsite2 in impossible_sites:
continue
ch = fam.comp_het_pair(
site1.gt_types,
site1.gt_bases,
site2.gt_types,
site2.gt_bases,
site1.gt_phases,
site2.gt_phases,
ref1=site1.row['ref'],
alt1=site1.row['alt'],
ref2=site2.row['ref'],
alt2=site2.row['alt'],
allow_unaffected=args.allow_unaffected,
fast_mode=True,
pattern_only=args.pattern_only)
if ch.get('impossible') == 'site1':
impossible_sites[strsite1] = True
break
if ch.get('impossible') == 'site2':
impossible_sites[strsite2] = True
if not ch['candidate']: continue
samples_w_hetpair[(site1, site2)].append(ch)
yield grp, self.filter_candidates(samples_w_hetpair)
