def validate_mutation_1(uniprot_id, mutation):
"""Select Provean; assert length > 0
"""
logger.debug(helper.underline("Validating that we have provean..."))
sql_query = """\
select 1
from {db_schema}.provean
where uniprot_id = '{uniprot_id}' and
provean_supset_filename is not null;
""".format(
uniprot_id=uniprot_id, db_schema=conf.CONFIGS["db_schema"]
)
logger.debug(sql_query)
df1 = pd.read_sql_query(sql_query, conf.CONFIGS["engine"])
logger.debug(df1.head(2))
#
logger.debug(helper.underline("And that we have at least one domain with a template..."))
sql_query = """\
select 1
from {db_schema}.uniprot_domain
join {db_schema}.uniprot_domain_template using (uniprot_domain_id)
where uniprot_id = '{uniprot_id}';
""".format(
uniprot_id=uniprot_id, db_schema=conf.CONFIGS["db_schema"]
)
logger.debug(sql_query)
df2 = pd.read_sql_query(sql_query, conf.CONFIGS["engine"])
logger.debug(df2.head(2))
assert len(df1) >= 1 or len(df2) == 0
def validate_mutation_5(uniprot_id, mutation):
"""Select domain pairs where we don't have mutatons even though we should; assert length 0
"""
logger.debug(helper.underline("Validating that we have domain pair mutations..."))
sql_query = """\
SELECT *
FROM {db_schema}.uniprot_domain_pair udp
JOIN {db_schema}.uniprot_domain ud1 on (ud1.uniprot_domain_id = udp.uniprot_domain_id_1)
JOIN {db_schema}.uniprot_domain ud2 on (ud2.uniprot_domain_id = udp.uniprot_domain_id_2)
JOIN {db_schema}.uniprot_domain_pair_template udpt using (uniprot_domain_pair_id)
JOIN {db_schema}.uniprot_domain_pair_model udpm using (uniprot_domain_pair_id)
LEFT JOIN {db_schema}.uniprot_domain_pair_mutation udpmut ON
(udpmut.uniprot_domain_pair_id = udp.uniprot_domain_pair_id AND
udpmut.uniprot_id = '{uniprot_id}' and udpmut.mutation = '{mutation}')
WHERE
((ud1.uniprot_id = '{uniprot_id}' AND
{db_schema}.mutation_in_interface('{mutation}', udpm.interacting_aa_1))
OR (ud2.uniprot_id = '{uniprot_id}' AND
{db_schema}.mutation_in_interface('{mutation}', udpm.interacting_aa_2))) AND
udpm.model_errors IS NULL AND -- we allow for some model errors
model_filename_wt IS NULL;
""".format(
uniprot_id=uniprot_id, mutation=mutation, db_schema=conf.CONFIGS["db_schema"]
)
logger.debug(sql_query)
df = pd.read_sql_query(sql_query, conf.CONFIGS["engine"])
assert len(df) == 0
def validate_mutation_2(uniprot_id, mutation):
"""Select domains without models; assert length 0
"""
logger.debug(helper.underline("Validating that we have domain models..."))
sql_query = """\
select *
from {db_schema}.uniprot_domain
join {db_schema}.uniprot_domain_template using (uniprot_domain_id)
left join {db_schema}.uniprot_domain_model using (uniprot_domain_id)
where uniprot_id = '{uniprot_id}' and
model_filename is null and
model_errors is null;
""".format(
uniprot_id=uniprot_id, db_schema=conf.CONFIGS["db_schema"]
)
logger.debug(sql_query)
df = pd.read_sql_query(sql_query, conf.CONFIGS["engine"])
assert len(df) == 0
def validate_mutation_3(uniprot_id, mutation):
"""Select interfaces without models; assert length 0
"""
logger.debug(helper.underline("Validating that we have domain pair models..."))
sql_query = """\
select *
from {db_schema}.uniprot_domain_pair udp
join {db_schema}.uniprot_domain ud1 on (ud1.uniprot_domain_id = udp.uniprot_domain_id_1)
join {db_schema}.uniprot_domain ud2 on (ud2.uniprot_domain_id = udp.uniprot_domain_id_2)
join {db_schema}.uniprot_domain_pair_template udpt using (uniprot_domain_pair_id)
left join {db_schema}.uniprot_domain_pair_model udpm using (uniprot_domain_pair_id)
where (ud1.uniprot_id = '{uniprot_id}' or ud2.uniprot_id = '{uniprot_id}') and
model_filename is null and model_errors is null;
""".format(
uniprot_id=uniprot_id, db_schema=conf.CONFIGS["db_schema"]
)
logger.debug(sql_query)
df = pd.read_sql_query(sql_query, conf.CONFIGS["engine"])
assert len(df) == 0
def validate_mutation_4(uniprot_id, mutation):
"""Select domains where we don't have mutatons even though we should; assert length 0
"""
logger.debug(helper.underline("Validating that we have domain mutations..."))
sql_query = """\
SELECT *
FROM {db_schema}.uniprot_domain ud
JOIN {db_schema}.uniprot_domain_template using (uniprot_domain_id)
JOIN {db_schema}.uniprot_domain_model udm using (uniprot_domain_id)
LEFT JOIN {db_schema}.uniprot_domain_mutation udmut ON
(udmut.uniprot_domain_id = ud.uniprot_domain_id AND mutation = '{mutation}')
WHERE
ud.uniprot_id = '{uniprot_id}' AND
{db_schema}.mutation_in_domain('{mutation}', model_domain_def) AND
udm.model_errors is null AND -- we allow for some model errors
model_filename_wt is null;
""".format(
uniprot_id=uniprot_id, mutation=mutation, db_schema=conf.CONFIGS["db_schema"]
)
logger.debug(sql_query)
df = pd.read_sql_query(sql_query, conf.CONFIGS["engine"])
assert len(df) == 0