def get_driver_df(driver_dict):
"""Function to process driver annotation results dictionary.
It collapses info from different sources into one representation.
Then creates dataframe from updated dictionary."""
for key, annotation in driver_dict.items():
role = set()
var_type = set()
source = []
source_pmid = []
reference = []
reference_source = set()
tumor = []
db_tumor = []
for driver_info in annotation:
# fn updating global role_set
helper.populate_set(driver_info, "driver_role", role)
# fn updating global variant type
helper.populate_set(driver_info, "var_type", var_type)
# fn updating global source_list
helper.populate_list(driver_info, "source_name", source)
# fn updating global source_pmid_list
helper.populate_list(driver_info, "source_pmid", source_pmid)
# fn updating global reference_id list
helper.populate_list(driver_info, "reference_id", reference)
# fn updating global reference_source list
helper.populate_set(driver_info, "reference_source",
reference_source)
# fn updating global tumor_list
helper.populate_list(driver_info, "tumor_list", tumor)
# fn updating global db disease keywords
helper.populate_list(driver_info, "db_tumor_repr", db_tumor)
driver_role = assign_driver_role(role)
agg_var_type = "|".join(list(var_type))
agg_source = "|".join(source)
agg_pmid = "|".join(source_pmid)
agg_ref = "|".join(reference)
agg_ref_source = "|".join(list(reference_source))
agg_tumor = "|".join([t for t in tumor if t != ""])
agg_db_tumor = "|".join([t for t in db_tumor if t != ""])
# replace the value of the dict. collapse list into one dict
driver_dict[key] = {
"driver_role": driver_role,
"source_name": agg_source,
"source_pmid": agg_pmid,
"reference_id": agg_ref,
"reference_source": agg_ref_source,
"tumor_list": agg_tumor,
"db_tumor_repr": agg_db_tumor,
"var_type": agg_var_type
}
# driver information dictionary to dataframe
driver_dataframe = dict_to_dataframe(driver_dict)
return driver_dataframe
# collapse info from sources into one representation
temp = {}
for key, annotation in cnv_mechanistic_drug_targets.items():
temp[key] = []
for db_id, info in annotation.items():
if len(info) == 1:
temp[key].append(info[0])
elif db_id != "null":
approval = set()
drug = set()
source = []
source_pmid = []
reference = []
reference_source = set()
for drug_info in info:
helper.populate_set(drug_info, "approval_status", approval)
helper.populate_set(drug_info, "drug_name", drug)
helper.populate_list(drug_info, "source_name", source)
helper.populate_list(drug_info, "source_pmid", source_pmid)
helper.populate_list(drug_info, "reference_id", reference)
helper.populate_set(drug_info, "reference_source",
reference_source)
approval_status = mechanistic.assign_approval_status(approval)
name = list(drug)[0]
agg_source = "|".join(source)
agg_pmid = "|".join(source_pmid)
agg_ref = "|".join(reference)
agg_ref_source = "|".join(list(reference_source))
temp[key].append({