def get_sequela_data(with_survey):
sequelae = gbd.get_sequela_id_mapping()
if with_survey:
data_survey = gbd.get_survey_summary('sequela', SURVEY_LOCATION_ID)
assert len(sequelae) == len(data_survey)
sequelae = sequelae.merge(data_survey, on='sequela_id')
else:
data_survey = make_empty_survey([
'incidence_exists', 'prevalence_exists', 'birth_prevalence_exists',
'incidence_in_range', 'prevalence_in_range',
'birth_prevalence_in_range'
], sequelae.index)
sequelae = sequelae.join(data_survey)
dw = gbd.get_auxiliary_data('disability_weight', 'sequela', 'all', 1)
sequelae['disability_weight_exists'] = sequelae['healthstate_id'].apply(
lambda h: bool(h in set(dw.healthstate_id)))
return list(
zip(clean_entity_list(sequelae.sequela_name), sequelae.sequela_id,
sequelae.modelable_entity_id,
clean_entity_list(sequelae.healthstate_name),
sequelae.healthstate_id, sequelae.disability_weight_exists,
sequelae.incidence_exists, sequelae.prevalence_exists,
sequelae.birth_prevalence_exists, sequelae.incidence_in_range,
sequelae.prevalence_in_range, sequelae.birth_prevalence_in_range))
def get_sequelae():
sequelae = gbd.get_sequela_id_mapping()
sequelae = pd.DataFrame({
"sequela_name":
clean_entity_list(sequelae.sequela_name),
"sequela_id":
sequelae.sequela_id,
})
return sequelae.sort_values("sequela_id")
def get_sequelae():
sequelae = gbd.get_sequela_id_mapping()
sequelae = pd.DataFrame({
'sequela_name':
clean_entity_list(sequelae.sequela_name),
'sequela_id':
sequelae.sequela_id
})
return sequelae.sort_values('sequela_id')
def get_sequela_data() -> List:
sequelae = gbd.get_sequela_id_mapping()
return list(
zip(
clean_entity_list(sequelae.sequela_name),
sequelae.sequela_id,
sequelae.modelable_entity_id,
clean_entity_list(sequelae.healthstate_name),
sequelae.healthstate_id,
))
def get_cause_data():
sequelae = gbd.get_sequela_id_mapping().sort_values("sequela_id")
etiologies = gbd.get_rei_metadata(rei_set_id=ETIOLOGY_SET_ID)
etiologies = etiologies[etiologies["most_detailed"] == 1].sort_values(
"rei_id")
cause_etiology_map = gbd.get_cause_etiology_mapping()
cause_me_map = gbd.get_cause_me_id_mapping()
cause_me_map["cause_name"] = clean_entity_list(
cause_me_map["modelable_entity_name"])
cause_me_map = cause_me_map[["modelable_entity_id",
"cause_name"]].set_index("cause_name")
causes = gbd.get_cause_metadata(cause_set_id=CAUSE_SET_ID)
causes = pd.DataFrame({
"cause_name":
clean_entity_list(causes.cause_name),
"cause_id":
causes.cause_id,
"parent_id":
causes.parent_id,
"most_detailed":
causes.most_detailed,
"level":
causes.level,
"male":
causes.male.replace({
np.NaN: False,
1: True
}),
"female":
causes.female.replace({
np.NaN: False,
1: True
}),
"yll_only":
causes.yll_only.replace({
np.NaN: False,
1: True
}),
"yld_only":
causes.yld_only.replace({
np.NaN: False,
1: True
}),
"yll_age_start":
causes.yll_age_start.replace({np.NaN: 0}),
"yll_age_end":
causes.yll_age_end.replace({np.NaN: 95}),
"yld_age_start":
causes.yld_age_start.replace({np.NaN: 0}),
"yld_age_end":
causes.yld_age_end.replace({np.NaN: 95}),
})
causes = (causes.set_index("cause_name").join(cause_me_map).sort_values(
"cause_id").reset_index())
cause_data = []
for _, cause in causes.iterrows():
name = cause["cause_name"]
cid = cause["cause_id"]
parent = causes.set_index("cause_id").at[cause["parent_id"],
"cause_name"]
dismod_id = cause["modelable_entity_id"]
most_detailed = cause["most_detailed"]
level = cause["level"]
restrictions = make_cause_restrictions(cause)
eti_ids = cause_etiology_map[cause_etiology_map.cause_id ==
cid].rei_id.tolist()
associated_etiologies = clean_entity_list(
etiologies[etiologies.rei_id.isin(eti_ids)].rei_name)
associated_sequelae = clean_entity_list(
sequelae[sequelae.cause_id == cid].sequela_name)
sub_causes = causes[causes.parent_id == cid].cause_name.tolist()
cause_data.append((
name,
cid,
dismod_id,
most_detailed,
level,
parent,
restrictions,
associated_sequelae,
associated_etiologies,
sub_causes,
))
return cause_data
def get_cause_data(with_survey):
sequelae = gbd.get_sequela_id_mapping().sort_values('sequela_id')
etiologies = gbd.get_rei_metadata(rei_set_id=ETIOLOGY_SET_ID)
etiologies = etiologies[etiologies['most_detailed'] == 1].sort_values(
'rei_id')
cause_etiology_map = gbd.get_cause_etiology_mapping()
cause_me_map = gbd.get_cause_me_id_mapping()
cause_me_map['cause_name'] = clean_entity_list(
cause_me_map['modelable_entity_name'])
cause_me_map = cause_me_map[['modelable_entity_id',
'cause_name']].set_index('cause_name')
causes = gbd.get_cause_metadata(cause_set_id=CAUSE_SET_ID)
causes = pd.DataFrame({
'cause_name':
clean_entity_list(causes.cause_name),
'cause_id':
causes.cause_id,
'parent_id':
causes.parent_id,
'most_detailed':
causes.most_detailed,
'level':
causes.level,
'male':
causes.male.replace({
np.NaN: False,
1: True
}),
'female':
causes.female.replace({
np.NaN: False,
1: True
}),
'yll_only':
causes.yll_only.replace({
np.NaN: False,
1: True
}),
'yld_only':
causes.yld_only.replace({
np.NaN: False,
1: True
}),
'yll_age_start':
causes.yll_age_start.replace({np.NaN: 0}),
'yll_age_end':
causes.yll_age_end,
'yld_age_start':
causes.yld_age_start.replace({np.NaN: 0}),
'yld_age_end':
causes.yld_age_end
})
if with_survey:
data_survey = gbd.get_survey_summary('cause', SURVEY_LOCATION_ID)
assert len(causes) == len(data_survey)
causes = causes.merge(data_survey, on='cause_id')
else:
data_survey = make_empty_survey([
'prevalence_exists', 'incidence_exists', 'remission_exists',
'deaths_exists', 'birth_prevalence_exists', 'prevalence_in_range',
'incidence_in_range', 'remission_in_range', 'deaths_in_range',
'birth_prevalence_in_range', 'prevalence_consistent',
'incidence_consistent', 'deaths_consistent',
'birth_prevalence_consistent', 'prevalence_aggregates',
'incidence_aggregates', 'deaths_aggregates',
'birth_prevalence_aggregates', 'violated_restrictions'
],
index=causes.index)
causes = causes.join(data_survey)
causes = causes.set_index('cause_name').join(cause_me_map).sort_values(
'cause_id').reset_index()
cause_data = []
for _, cause in causes.iterrows():
name = cause['cause_name']
cid = cause['cause_id']
parent = causes.set_index('cause_id').at[cause['parent_id'],
'cause_name']
dismod_id = cause['modelable_entity_id']
most_detailed = cause['most_detailed']
level = cause['level']
restrictions = make_cause_restrictions(cause)
prev_exists = cause['prevalence_exists']
inc_exists = cause['incidence_exists']
remission_exists = cause['remission_exists']
deaths_exists = cause['deaths_exists']
birth_prevalence_exists = cause['birth_prevalence_exists']
prev_in_range = cause['prevalence_in_range']
inc_in_range = cause['incidence_in_range']
remission_in_range = cause['remission_in_range']
deaths_in_range = cause['deaths_in_range']
birth_prev_in_range = cause['birth_prevalence_in_range']
prev_consistent = cause['prevalence_consistent']
inc_consistent = cause['incidence_consistent']
deaths_consistent = cause['deaths_consistent']
birth_prev_consistent = cause['birth_prevalence_consistent']
prev_aggregates = cause['prevalence_aggregates']
inc_aggregates = cause['incidence_aggregates']
deaths_aggregates = cause['deaths_aggregates']
birth_prev_aggregates = cause['birth_prevalence_aggregates']
eti_ids = cause_etiology_map[cause_etiology_map.cause_id ==
cid].rei_id.tolist()
associated_etiologies = clean_entity_list(
etiologies[etiologies.rei_id.isin(eti_ids)].rei_name)
associated_sequelae = clean_entity_list(
sequelae[sequelae.cause_id == cid].sequela_name)
sub_causes = causes[causes.parent_id == cid].cause_name.tolist()
cause_data.append(
(name, cid, dismod_id, most_detailed, level, parent, restrictions,
prev_exists, inc_exists, remission_exists, deaths_exists,
birth_prevalence_exists, prev_in_range, inc_in_range,
remission_in_range, deaths_in_range, birth_prev_in_range,
prev_consistent, inc_consistent, deaths_consistent,
birth_prev_consistent, prev_aggregates, inc_aggregates,
deaths_aggregates, birth_prev_aggregates, associated_sequelae,
associated_etiologies, sub_causes))
return cause_data