def infect(self):
'''
Calculates whether or not the contact has covid
'''
if self.index_case.symptomatic:
infection_scale = 1
else:
infection_scale = self.parameters['epi_params']['asymp_factor']
# if a member of the household
if self.is_household:
self.has_covid = utils.bernoulli(
infection_scale * self.parameters['epi_params']['sar']['household'],
self.rng
)
if self.has_covid:
# day infected is sample from the infection profile
self.day_infected = utils.draw_from_pdf(self.rng, self.infectious_period)
else:
self.day_infected = np.nan
# if not a member of the household, then
# the probability the contact has covid is a function of
# the day they came into contact, and of the infectiosness
# distribution
else:
viral_load = self.infectious_period[self.day_exposed]
self.has_covid = utils.bernoulli(
viral_load * infection_scale * self.parameters['epi_params']['sar']['other'] * self.parameters['epi_params']['max_infectious_day'],
self.rng
)
if self.has_covid:
self.day_infected = self.day_exposed
else:
self.day_infected = np.nan
# establish whether case is symptomatic or not
if self.has_covid:
self.symptomatic = utils.bernoulli(
self.parameters['epi_params']['p_symp'],
self.rng
)
if self.symptomatic:
self.day_symptomatic = utils.draw_from_pdf(
self.rng, self.incubation_period
)
else:
self.day_symptomatic = np.nan
else:
self.symptomatic = False
self.day_symptomatic = np.nan
def trace(self):
'''
Calculate whether the contact was traced or not
'''
# if the index case enters contacts, then establish whether
# contact was successfull contacted and when
if self.index_case.enters_contacts and self.has_covid:
# if contact is a household member
# assume tracing is successful
# and contact is traced on same day as contacts entered
if self.is_household:
self.traced = True
self.day_traced = self.index_case.day_contacts_entered
else:
self.traced = utils.bernoulli(
self.parameters['trace_params']['p_traced'],
self.rng
)
# assume tracing occurs a maximum of 3 days after contacts
# were entered by the index case
# TODO: change this to a data-driven approach
if self.traced:
self.day_traced = self.index_case.day_contacts_entered + \
self.rng.integers(1, 3)
else:
self.day_traced = np.nan
else:
self.traced = False
self.day_traced = np.nan
def init_case(self):
# is symptomatic?
self.symptomatic = utils.bernoulli(
self.parameters['epi_params']['p_symp'], self.rng)
# how long in days infectious
self.infectious_length = self.parameters['epi_params'][
'max_infectious_day']
# when symptoms occur
if self.symptomatic:
self.day_symptom_onset = utils.draw_from_pdf(
self.rng, self.incubation_period)
else:
self.day_symptom_onset = np.nan
def report_contacts(self):
# does the cases supply contacts?
# assume that contacts are only reported if the case is symptomatic
if self.symptomatic:
self.enters_contacts = utils.bernoulli(
self.parameters['trace_params']['p_contacts_entered'],
self.rng)
# day contacts entered after infection
# assume that it is a maximum of 5 days
# after symptomatic onset
self.day_contacts_entered = \
utils.draw_from_pdf(self.rng, self.incubation_period) + \
self.rng.integers(0, 5)
else:
self.enters_contacts = False
self.day_contacts_entered = np.nan
def test_bernoulli():
assert utils.bernoulli(1, RNG)
def isolate(self):
'''
Calculate whether (and when) the contact isolated
'''
# isolate on symptoms
if self.symptomatic:
isolate_on_symptoms = utils.bernoulli(
self.parameters['isolate_params']['p_symp_isolate'],
self.rng
)
if isolate_on_symptoms:
# assume isolates on symptom onset
day_isolate_on_symptoms = self.day_symptomatic
else:
day_isolate_on_symptoms = False
else:
isolate_on_symptoms = False
day_isolate_on_symptoms = np.nan
# isolate on trace
if self.traced:
isolate_on_trace = utils.bernoulli(
self.parameters['isolate_params']['p_trace_isolate'],
self.rng
)
if isolate_on_trace:
day_isolate_on_trace = self.day_traced
else:
day_isolate_on_trace = np.nan
else:
isolate_on_trace = False
day_isolate_on_trace = np.nan
# isolate on test
if self.tested:
isolate_on_test = utils.bernoulli(
self.parameters['isolate_params']['p_test_isolate'],
self.rng
)
if isolate_on_test:
# TODO: change to day test result communicated
day_isolate_on_test = self.day_tested
else:
day_isolate_on_test = np.nan
else:
isolate_on_test = False
day_isolate_on_test = np.nan
if isolate_on_symptoms or isolate_on_test or isolate_on_trace:
self.isolated = True
self.day_isolated = np.nanmin([
day_isolate_on_test,
day_isolate_on_symptoms,
day_isolate_on_trace
])
else:
self.isolated = False
self.day_isolated = np.nan
def test(self):
'''
Calculate whether (and when) the contact was tested
'''
# is testing done on symptoms?
if self.symptomatic:
test_on_symptoms = utils.bernoulli(
self.parameters['test_params']['p_symp_test'],
self.rng
)
if test_on_symptoms:
# day tested is a maximum of 3 days after symptom onset
# TODO: inform the testing date using data
day_test_on_symptoms = self.day_symptomatic \
+ self.rng.integers(1, 3)
else:
test_on_symptoms = False
if not test_on_symptoms:
day_test_on_symptoms = np.nan
# is testing done on tracing?
if self.traced and self.has_covid:
test_on_tracing = utils.bernoulli(
self.parameters['test_params']['p_trace_test'],
self.rng
)
if test_on_tracing:
# is done after the index case has been contacted
# plus the number of days in communicating result
# plus the number of days it takes for an individual to test
# TODO: drive this result with data
day_test_on_tracing = self.index_case.day_contacts_entered \
+ self.rng.random.rand_int(1,5)
else:
day_test_on_tracing = np.nan
else:
test_on_tracing = False
day_test_on_tracing = np.nan
# is random asymptomatic (i.e. mass testing) done?
if self.has_covid:
test_on_mass = utils.bernoulli(
self.parameters['test_params']['p_mass_test'],
self.rng
)
if test_on_mass:
# day tested on mass is drawn from a uniform distribution
# covering the infectious period
day_test_on_mass = self.rng.integers(1, 14)
else:
day_test_on_mass = np.nan
else:
test_on_mass = False
day_test_on_mass = np.nan
# contact is tested whether any of the above are true
if test_on_symptoms or test_on_mass or test_on_tracing:
self.tested = True
self.day_tested = np.nanmin([
day_test_on_symptoms,
day_test_on_tracing,
day_test_on_mass
])
else:
self.tested = False
self.day_tested = np.nan