def test_load_par_variant(mongo_adapter, par_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(par_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
mongo_variant = db.variant.find_one()
assert mongo_variant == None
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['families'] == [family_id]
assert mongo_variant['observations'] == 1
assert mongo_variant['homozygote'] == 0
assert mongo_variant['hemizygote'] == 0
def test_load_par_variant(mongo_adapter, par_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(par_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
mongo_variant = db.variant.find_one()
assert mongo_variant == None
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['families'] == [family_id]
assert mongo_variant['observations'] == 1
assert mongo_variant['homozygote'] == 0
assert mongo_variant['hemizygote'] == 0
def test_delete_variants(mongo_adapter, het_variant, case_obj, ind_positions):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
ind_positions=ind_positions,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['families'] == [family_id]
delete_variants(adapter=mongo_adapter,
family_id=family_id,
ind_positions=ind_positions,
individuals=individuals,
vcf=vcf)
mongo_variant = db.variant.find_one()
assert mongo_variant == None
def frequency(context, bulk, remove):
"""Upload or remove variants to/from frequency database."""
loqusdb_auth = context.obj.get("loqusdb", {})
adapter = LoqusdbAdapter()
adapter.connect(**loqusdb_auth)
mip_case = context.obj["case"]
if mip_case.is_old:
logger.warn("unsupported MIP version for frequency: %s", mip_case.version)
else:
vcf_path = mip_case.research_vcf
case_group_id = group_id(mip_case)
# we only care about "rare" variants
ped_path = mip_case.configs.pedigree
try:
affected_inds = affected_individuals(ped_path)
except FrequencyError as error:
logger.warn("LOQUSDB: {}".format(error.message))
if "affected_inds" in locals():
with open(vcf_path, "r") as vcf_stream:
try:
if remove:
count = delete_variants(adapter, vcf_stream, case_group_id, affected_inds)
logger.info("removed {} variants".format(count))
else:
load_variants(
adapter, case_group_id, affected_inds, vcf_stream, bulk_insert=bulk, vcf_path=vcf_path
)
except CaseError as error:
logger.warn("LOQUSDB: {}".format(error.message))
def test_load_par_variant(mongo_adapter, par_variant, case_obj, ind_positions):
## GIVEN an adapter and a vcf with one PAR variant
db = mongo_adapter.db
vcf = []
vcf.append(par_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
assert db.variant.find_one() == None
## WHEN loading the variant into the database
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
ind_positions=ind_positions,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
## THEN assert that the variant is loaded correct
assert mongo_variant['families'] == [family_id]
assert mongo_variant['observations'] == 1
assert mongo_variant['homozygote'] == 0
assert mongo_variant['hemizygote'] == 0
def test_delete_variants(mongo_adapter, het_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['families'] == [family_id]
delete_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf
)
mongo_variant = db.variant.find_one()
assert mongo_variant == None
def test_load_same_variant_different_case(mongo_adapter, het_variant, case_obj,
ind_positions):
## GIVEN an adapter and a vcf
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
## WHEN loading the variant into the database
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
ind_positions=ind_positions,
vcf=vcf,
)
family_id_2 = '2'
load_variants(
adapter=mongo_adapter,
family_id=family_id_2,
individuals=individuals,
ind_positions=ind_positions,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['observations'] == 2
assert mongo_variant['families'] == [family_id, family_id_2]
# # This test works when using a real mongo adapter but not with mongomock yet...
# # def test_load_same_variant_many_cases(real_mongo_adapter, family_variant):
# # """docstring for test_load_variants"""
# # db = real_mongo_adapter.db
# #
# # vcf = []
# # vcf.append(family_variant)
# # individuals=['proband', 'mother', 'father']
# #
# # for i in range(40):
# # family_id = str(i)
# #
# # load_variants(
# # adapter=real_mongo_adapter,
# # family_id=family_id,
# # individuals=individuals,
# # vcf=vcf,
# # )
# #
# # mongo_variant = db.variant.find_one()
# #
# # assert mongo_variant['observations'] == 40
# # assert len(mongo_variant['families']) == 20
def test_load_variants_skip_case_id(mongo_adapter, het_variant, case_obj,
ind_positions):
## GIVEN an adapter and a vcf with tho heterygote variants
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
## WHEN loading the variants into the database
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
ind_positions=ind_positions,
vcf=vcf,
skip_case_id=True,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
## THEN assert that the variant is loaded correct
assert mongo_variant.get('families') == None
def test_load_two_variants(mongo_adapter, het_variant, case_obj,
ind_positions):
## GIVEN an adapter and a vcf with tho heterygote variants
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
## WHEN loading the variants into the database
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
ind_positions=ind_positions,
vcf=vcf,
gq_treshold=20,
)
## THEN assert that the variant is loaded correct
mongo_variant = db.variant.find_one()
assert mongo_variant['observations'] == 2
def test_load_same_variant_different_case(mongo_adapter, het_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf,
)
family_id_2 = '2'
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id_2,
individuals=individuals,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['observations'] == 2
assert mongo_variant['families'] == [family_id, family_id_2]
# # This test works when using a real mongo adapter but not with mongomock yet...
# # def test_load_same_variant_many_cases(real_mongo_adapter, family_variant):
# # """docstring for test_load_variants"""
# # db = real_mongo_adapter.db
# #
# # vcf = []
# # vcf.append(family_variant)
# # individuals=['proband', 'mother', 'father']
# #
# # for i in range(40):
# # family_id = str(i)
# #
# # load_variants(
# # adapter=real_mongo_adapter,
# # family_id=family_id,
# # individuals=individuals,
# # vcf=vcf,
# # )
# #
# # mongo_variant = db.variant.find_one()
# #
# # assert mongo_variant['observations'] == 40
# # assert len(mongo_variant['families']) == 20
def upload(adapter, vcf_path, affected, case_id, force=False, bulk=True):
with open(vcf_path, 'r') as vcf_stream:
try:
if force:
count = delete_variants(adapter, vcf_stream,
case_id, affected)
logger.info("removed {} variants".format(count))
load_variants(adapter, case_id, affected, vcf_stream,
bulk_insert=bulk, vcf_path=vcf_path)
except CaseError as error:
logger.warn("LOQUSDB: {}".format(error.message))
def test_delete_variant(mongo_adapter, het_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id_1 = case_obj.family_id
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id_1,
individuals=individuals,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['families'] == [family_id_1]
family_id_2 = '2'
load_variants(
adapter=mongo_adapter,
family_id=family_id_2,
individuals=individuals,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['families'] == [family_id_1, family_id_2]
assert mongo_variant['observations'] == 2
delete_variants(adapter=mongo_adapter,
family_id=family_id_2,
individuals=individuals,
vcf=vcf)
mongo_variant = db.variant.find_one()
assert mongo_variant['families'] == [family_id_1]
assert mongo_variant['observations'] == 1
def test_load_variants_skip_case_id(mongo_adapter, het_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf,
skip_case_id=True,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant.get('families') == None
def test_load_two_variants(mongo_adapter, het_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant['observations'] == 2
def test_load_variants_skip_case_id(mongo_adapter, het_variant, case_obj):
"""docstring for test_load_variants"""
db = mongo_adapter.db
vcf = []
vcf.append(het_variant)
family_id = case_obj.family_id
individuals = case_obj.individuals
load_variants(
adapter=mongo_adapter,
family_id=family_id,
individuals=individuals,
vcf=vcf,
skip_case_id=True,
gq_treshold=20,
)
mongo_variant = db.variant.find_one()
assert mongo_variant.get('families') == None
def load(ctx, variant_file, family_file, family_type, bulk_insert):
"""Load the variants of a case
The loading is based on if the variant is seen in a ny affected individual
in the family.
"""
if not family_file:
logger.error("Please provide a family file")
ctx.abort()
if variant_file == '-':
logger.info("Parsing variants from stdin")
variant_file = get_vcf_handle(fsock=sys.stdin)
else:
logger.info("Start parsing variants from stdin")
variant_path = os.path.abspath(variant_file)
variant_file = get_vcf_handle(infile=variant_file)
try:
family = get_family(family_lines=family_file, family_type=family_type)
except SyntaxError as error:
logger.warning(error.message)
ctx.abort()
if not family.affected_individuals:
logger.error("No affected individuals could be found in ped file")
ctx.abort()
logger.info("Found affected individuals in ped file: {0}"
.format(', '.join(family.affected_individuals)))
adapter = ctx.obj['adapter']
try:
load_variants(adapter, family.family_id, family.affected_individuals,
variant_file, bulk_insert=bulk_insert,
vcf_path=variant_path)
except CaseError as error:
logger.error(error.message)
ctx.abort()