def test_one_object():
"""Test if generate pairs behave as suspected"""
objects = [1]
pairs = []
with pytest.raises(SyntaxError):
for pair in generate_pairs(objects):
pairs.append(pair)
def test_one_object():
"""Test if generate pairs behave as suspected"""
objects = [1]
pairs = []
with pytest.raises(SyntaxError):
for pair in generate_pairs(objects):
pairs.append(pair)
def test_generate_pairs():
"""Test if generate pairs behave as suspected"""
objects = [1,2]
pairs = []
for pair in generate_pairs(objects):
pairs.append(pair)
assert pairs == [(1,2)]
def test_generate_pairs():
"""Test if generate pairs behave as suspected"""
objects = [1, 2]
pairs = []
for pair in generate_pairs(objects):
pairs.append(pair)
assert pairs == [(1, 2)]
def test_generate_multiple_pairs():
"""Test if generate pairs behave as suspected"""
objects = [1,2,3,4]
pairs = []
for pair in generate_pairs(objects):
pairs.append(pair)
assert len(pairs) == 6
assert pairs[0] == (1,2)
assert pairs[1] == (1,3)
assert pairs[-1] == (3,4)
def test_generate_multiple_pairs():
"""Test if generate pairs behave as suspected"""
objects = [1, 2, 3, 4]
pairs = []
for pair in generate_pairs(objects):
pairs.append(pair)
assert len(pairs) == 6
assert pairs[0] == (1, 2)
assert pairs[1] == (1, 3)
assert pairs[-1] == (3, 4)
def check_genetic_models(variant_batch, families, phased = False,
strict = False):
"""
Check and annotate which genetic models that are followed for the variants
in a batch
See more description in header of file or documentation.
Arguments:
variant_batch (dict): A dictionary with variant ids as keys and variant
dictionaries as values
families (dict): A dictionary with family ids as keys and Family
objects as values
phased (bool): If the variants are phased
strict (bool): If the strict mode should be used when checking the
genetic models
"""
# A variant batch is a dictionary on the form
# {variant_id:variant_dict, variant_2_id:variant_dict_2, ...}
logger = logging.getLogger(__name__)
intervals = variant_batch.pop('haploblocks', {})
# We check the genetic models for one family at a time
for family_id in families:
logger.debug("Checking genetic models for family {0}".format(
family_id
))
family = families[family_id]
individuals = family.individuals
compound_candidates = []
compound_pairs = []
for variant_id in variant_batch:
inheritance_models = {
'XR' : False,
'XR_dn' : False,
'XD' : False,
'XD_dn' : False,
'AD' : False,
'AD_dn' : False,
'AR_hom' : False,
'AR_hom_dn' : False,
'AR_comp' : False,
'AR_comp_dn' : False
}
variant = variant_batch[variant_id]
# save the compound pairs for a variant in a set
if 'compounds' in variant:
variant['compounds'][family_id] = set()
else:
variant['compounds'] = {family_id : set()}
# Add information of models followed:
if 'inheritance_models' in variant:
variant['inheritance_models'][family_id] = inheritance_models
else:
variant['inheritance_models'] = {family_id: inheritance_models}
# If the variant is in a genetic region we check for compound
# candidates
if variant.get('compound_candidate',True):
if check_compound_candidate(variant, family, strict):
compound_candidates.append(variant_id)
# Only check X-linked for the variants in the X-chromosome:
# For X-linked we do not need to check the other models
if variant['CHROM'] == 'X':
if check_X_recessive(variant, family, strict):
variant['inheritance_models'][family_id]['XR'] = True
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents(
'X_recessive',
individual_id,
family,
variant
)
if check_X_dominant(variant, family, strict):
variant['inheritance_models'][family_id]['XD'] = True
for individual_id in family.individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents(
'X_dominant',
individual_id,
family,
variant
)
# If variant is not on X:
else:
# Check the dominant model:
if check_dominant(variant, family, strict):
variant['inheritance_models'][family_id]['AD'] = True
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents(
'dominant',
individual_id,
family,
variant
)
# Check the recessive model:
if check_recessive(variant, family, strict):
variant['inheritance_models'][family_id]['AR_hom'] = True
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents(
'recessive',
individual_id,
family,
variant
)
# Now check the compound models:
if len(compound_candidates) > 1:
for pair in generate_pairs(compound_candidates):
# If the variants in the pair belong to the same gene we check for compounds:
variant_1 = variant_batch[pair[0]]
variant_2 = variant_batch[pair[1]]
# Check that the pair is in the same feature:
if variant_1['annotation'].intersection(variant_2['annotation']):
if len(individuals) == 1:
variant_1['compounds'][family_id].add(pair[1])
variant_2['compounds'][family_id].add(pair[0])
variant_1['inheritance_models'][family_id]['AR_comp'] = True
variant_2['inheritance_models'][family_id]['AR_comp'] = True
# We know from check_compound_candidates that all variants are present in all affected
elif check_compounds(variant_1, variant_2, family, intervals, phased):
parents_found = False
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents('compound', individual_id, family, variant_1, variant_2)
parents_found = True
if not parents_found:
variant_1['inheritance_models'][family_id]['AR_comp'] = True
variant_2['inheritance_models'][family_id]['AR_comp'] = True
variant_1['compounds'][family_id].add(pair[1])
variant_2['compounds'][family_id].add(pair[0])
return
def check_genetic_models(variant_batch, families, phased=False, strict=False):
"""
Check and annotate which genetic models that are followed for the variants
in a batch
See more description in header of file or documentation.
Arguments:
variant_batch (dict): A dictionary with variant ids as keys and variant
dictionaries as values
families (dict): A dictionary with family ids as keys and Family
objects as values
phased (bool): If the variants are phased
strict (bool): If the strict mode should be used when checking the
genetic models
"""
# A variant batch is a dictionary on the form
# {variant_id:variant_dict, variant_2_id:variant_dict_2, ...}
logger = logging.getLogger(__name__)
intervals = variant_batch.pop('haploblocks', {})
# We check the genetic models for one family at a time
for family_id in families:
logger.debug(
"Checking genetic models for family {0}".format(family_id))
family = families[family_id]
individuals = family.individuals
compound_candidates = []
compound_pairs = []
for variant_id in variant_batch:
inheritance_models = {
'XR': False,
'XR_dn': False,
'XD': False,
'XD_dn': False,
'AD': False,
'AD_dn': False,
'AR_hom': False,
'AR_hom_dn': False,
'AR_comp': False,
'AR_comp_dn': False
}
variant = variant_batch[variant_id]
# save the compound pairs for a variant in a set
if 'compounds' in variant:
variant['compounds'][family_id] = set()
else:
variant['compounds'] = {family_id: set()}
# Add information of models followed:
if 'inheritance_models' in variant:
variant['inheritance_models'][family_id] = inheritance_models
else:
variant['inheritance_models'] = {family_id: inheritance_models}
# If the variant is in a genetic region we check for compound
# candidates
if variant.get('compound_candidate', True):
if check_compound_candidate(variant, family, strict):
compound_candidates.append(variant_id)
# Only check X-linked for the variants in the X-chromosome:
# For X-linked we do not need to check the other models
if variant['CHROM'] == 'X':
if check_X_recessive(variant, family, strict):
variant['inheritance_models'][family_id]['XR'] = True
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents('X_recessive',
individual_id,
family,
variant,
strict=strict)
if check_X_dominant(variant, family, strict):
variant['inheritance_models'][family_id]['XD'] = True
for individual_id in family.individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents('X_dominant',
individual_id,
family,
variant,
strict=strict)
# If variant is not on X:
else:
# Check the dominant model:
if check_dominant(variant, family, strict):
variant['inheritance_models'][family_id]['AD'] = True
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents('dominant',
individual_id,
family,
variant,
strict=strict)
# Check the recessive model:
if check_recessive(variant, family, strict):
variant['inheritance_models'][family_id]['AR_hom'] = True
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents('recessive',
individual_id,
family,
variant,
strict=strict)
# Now check the compound models:
if len(compound_candidates) > 1:
for pair in generate_pairs(compound_candidates):
# If the variants in the pair belong to the same gene we check for compounds:
variant_1 = variant_batch[pair[0]]
variant_2 = variant_batch[pair[1]]
# Check that the pair is in the same feature:
if variant_1['annotation'].intersection(
variant_2['annotation']):
if len(individuals) == 1:
variant_1['compounds'][family_id].add(pair[1])
variant_2['compounds'][family_id].add(pair[0])
variant_1['inheritance_models'][family_id][
'AR_comp'] = True
variant_2['inheritance_models'][family_id][
'AR_comp'] = True
# We know from check_compound_candidates that all variants are present in all affected
elif check_compounds(variant_1, variant_2, family,
intervals, phased):
parents_found = False
for individual_id in individuals:
individual = individuals[individual_id]
if individual.has_parents:
check_parents(model='compound',
individual_id=individual_id,
family=family,
variant=variant_1,
variant_2=variant_2,
strict=strict)
parents_found = True
if not parents_found:
variant_1['inheritance_models'][family_id][
'AR_comp'] = True
variant_2['inheritance_models'][family_id][
'AR_comp'] = True
if (variant_1['inheritance_models'][family_id]
['AR_comp'] or variant_1['inheritance_models']
[family_id]['AR_comp_dn']):
variant_1['compounds'][family_id].add(pair[1])
if (variant_2['inheritance_models'][family_id]
['AR_comp'] or variant_2['inheritance_models']
[family_id]['AR_comp_dn']):
variant_2['compounds'][family_id].add(pair[0])
return