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 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 test_comp_het_one_parent_2kids():
"""
test that we cant have a candidate when a parent is HOM_REF at both sites.
"""
mom._i = 0
kid._i = 1
kid2._i = 2
kid.dad = None
kid.mom = None
kid.mom = mom
efam = EvalFamily(Family([mom, kid, kid2], '2kids'))
efam.gt_types = [Family.HOM_REF, Family.HET, Family.HET]
res = efam.comp_het_pair([Family.HOM_REF, Family.HET, Family.HET],
["T/T", "T/C", "T/C"],
[Family.HOM_REF, Family.HET, Family.HET],
["A/A", "A/C", "A/C"], [False] * 3, [False] * 3,
"T",
"C",
"A",
"C",
fast_mode=False,
allow_unaffected=True)
assert not res['candidate'], res
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 test_comp_het_singleton():
kid = Sample('kid', affected=True)
efam = EvalFamily(Family([kid], 'singleton'))
efam.gt_types = [Family.HET]
res = efam.comp_het_pair([Family.HET], ["A/C"], [Family.HET], ["A/C"],
[False], [False], "A", "C", "A", "C")
assert res['candidate']
assert res['priority'] == 2, res
def make_fam2():
# 1 affected kid, parent, grandparent
fam = Family.from_ped("""\
#family_id sample_id paternal_id maternal_id sex phenotype
1 dad 0 0 1 1
1 mom grandpa grandma 2 2
1 kid dad mom 1 2
1 kid2 dad mom 1 1
1 grandma 0 0 2 2
1 grandpa 0 0 1 1""")
return fam
def make_fam1():
# only 1 affected kid.
fam = Family.from_ped("""\
#family_id sample_id paternal_id maternal_id sex phenotype
1 dad 0 0 1 1
1 mom grandpa grandma 2 1
1 kid dad mom 1 2
1 kid2 dad mom 1 1
1 grandma 0 0 2 1
1 grandpa 0 0 1 1""")
return fam
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 make_fam2():
# 1 affected kid, parent, grandparent
fam = Family.from_ped("""\
#family_id sample_id paternal_id maternal_id sex phenotype
1 dad 0 0 1 1
1 mom grandpa grandma 2 2
1 kid dad mom 1 2
1 kid2 dad mom 1 1
1 grandma 0 0 2 2
1 grandpa 0 0 1 1""")
return fam
def make_fam(n_affecteds, n_unaffecteds, n_unknowns, id="xxx"):
samples = []
for i in range(n_affecteds):
samples.append(
Sample('affected_%d' % i,
affected=True,
sex=random.randint(1, 2),
name='affected_%d' % i))
for i in range(n_unaffecteds):
samples.append(
Sample('unaffected_%d' % i,
affected=False,
sex=random.randint(1, 2),
name='affected_%d' % i))
for i in range(n_unknowns):
samples.append(
Sample('unknown_%d' % i,
affected=None,
sex=random.randint(1, 2),
name='affected_%d' % i))
for i in range(int((n_affecteds + n_affecteds + n_unknowns) / 2)):
sample = random.choice(samples)
if random.random() < 0.9:
try:
sample.dad = random.choice([
s for s in samples if not s == sample and s.sex == 'male'
])
except IndexError:
pass
if random.random() < 0.9:
try:
sample.mom = random.choice([
s for s in samples if not s == sample and s.sex == 'female'
])
except IndexError:
pass
fam = EvalFamily(Family(samples, 'fam_%s' % id))
fam.gt_types = [random.randrange(0, 4) for _ in range(len(samples))]
fam.gt_depths = [random.randrange(0, 100) for _ in range(len(samples))]
fam.gt_phred_ll_homref = [
random.randrange(0, 100) for _ in range(len(samples))
]
fam.gt_phred_ll_het = [
random.randrange(0, 100) for _ in range(len(samples))
]
fam.gt_phred_ll_homalt = [
random.randrange(0, 100) for _ in range(len(samples))
]
fam.gt_quals = [random.randrange(5, 100) for _ in range(len(samples))]
return fam
def make_fam1():
# only 1 affected kid.
fam = Family.from_ped("""\
#family_id sample_id paternal_id maternal_id sex phenotype
1 dad 0 0 1 1
1 mom grandpa grandma 2 1
1 kid dad mom 1 2
1 kid2 dad mom 1 1
1 grandma 0 0 2 1
1 grandpa 0 0 1 1""")
return fam
def test_comp_het_all_hets():
efam = EvalFamily(Family([dad, mom, kid], 'triox'))
efam.gt_types = [Family.HET] * 3
res = efam.comp_het_pair([Family.HET] * 3, ["A/C"] * 3, [Family.HET] * 3,
["A/C"] * 3, [False] * 3, [False] * 3, "A", "C",
"A", "C")
assert res['candidate']
assert res['priority'] == 3
def test_x_rec():
mom = Sample('mom_1239NIH', affected=False, sex='female')
dad = Sample('dad_1240NIH', affected=False, sex='male')
kid_aff = Sample('kidaff_1238NIH', affected=True, sex='female')
kid_aff.mom = mom
kid_aff.dad = dad
efam = EvalFamily(Family([dad, mom, kid_aff], 'oler-trio'))
# mom should be a carrier
efam.gt_types = [Family.HOM_REF, Family.HOM_REF, Family.HOM_ALT]
assert efam.x_rec()
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 test_comp_het_one_parent():
mom._i = 0
kid._i = 1
kid.dad = None
kid.mom = None
efam = EvalFamily(Family([mom, kid], 'pair_mom'))
efam.gt_types = [Family.HET] * 2
res = efam.comp_het_pair([Family.HET] * 2, ["A/C"] * 2, [Family.HET] * 2,
["A/C"] * 2, [False] * 2, [False] * 2, "A", "C",
"A", "C")
assert res['candidate']
assert res['priority'] == 3, res['priority']
res = efam.comp_het_pair([Family.HOM_REF, Family.HET] * 2, ["A/A", "A/C"],
[Family.HET, Family.HET], ["A/C"] * 2,
[False] * 2, [False] * 2, "A", "C", "A", "C")
assert res['candidate']
assert res['priority'] == 2, res['priority']
res = efam.comp_het_pair([Family.HOM_REF, Family.HOM_REF] * 2,
["A/A", "A/A"], [Family.HET, Family.HET],
["A/C"] * 2, [False] * 2, [False] * 2, "A", "C",
"A", "C")
assert not res['candidate']
def test_x_dom_parents():
mom = Sample('mom', affected=False, sex='female')
dad = Sample('dad', affected=False, sex='male')
kid = Sample('kid', affected=True, sex='female')
kid.mom, kid.dad = mom, dad
efam = EvalFamily(Family([dad, mom, kid], 'trio'))
efam.gt_types = [Family.HOM_REF, Family.HOM_REF, Family.HET]
# neither parent is het
assert not efam.x_dom()
# neither parent is affected
efam.gt_types = [Family.HET, Family.HOM_REF, Family.HET]
assert not efam.x_dom()
dad.affected = True
assert efam.x_dom()
# for male, only mom must be affected
kid.sex = 'male'
assert not efam.x_dom()
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)
from __future__ import print_function
import sys
from inheritance import Sample, Family, EvalFamily
mom = Sample('mom', affected=False)
dad = Sample('dad', affected=False)
kid = Sample('kid', affected=True)
kid.mom, kid.dad = mom, dad
fam = Family([mom, dad, kid], 'a')
def make_fam1():
# only 1 affected kid.
fam = Family.from_ped("""\
#family_id sample_id paternal_id maternal_id sex phenotype
1 dad 0 0 1 1
1 mom grandpa grandma 2 1
1 kid dad mom 1 2
1 kid2 dad mom 1 1
1 grandma 0 0 2 1
1 grandpa 0 0 1 1""")
return fam
def make_fam2():
# 1 affected kid, parent, grandparent
fam = Family.from_ped("""\
#family_id sample_id paternal_id maternal_id sex phenotype
1 dad 0 0 1 1
