Keeping track of secondary infections

[mghosh00]

Is your feature request related to a problem? Please describe. We wish to record the number of people each person infects over the course of the epidemic (each time they are re-infected) so that we can find the R_t value at each time step. The dataframes produced will be further used in the sampling package EpiOS.

Describe the solution you'd like I will change the Person class to record a list of secondary infections each person has during the epidemic. I will also update the simulation to record and output a dataframe containing this information.

[mghosh00]

Classes and functions to update:

core

• [x] Person.__init__ (infection_start_time, secondary_infections)
• [x] Person.increment_secondary_infections new
• [x] tests

output

• [x] AgeStratifiedNewCasesWriter.write
• [x] NewCasesWriter.write
• [x] tests

property

• [x] PersonalFOI.person_susc
• [x] tests

routine

• [x] Simulation.configure (add boolean to inf_history_params)
• [x] Simulation.run_sweeps
• [x] Simulation.write_to_rt_file new
• [x] tests

sweep

• [x] PlaceSweep.__call__
• [x] HouseholdSweep.__call__
• [x] SpatialSweep.__call__
• [x] HostProgressionSweep.set_infectiousness
• [x] HostProgressionSweep._updates_infectiousness
• [x] tests

[mghosh00]

Structure of output dataframe

Here is a template for what the output dataframe will look like (maybe not including R_t but is there to illustrate):

time	person A	person B	person C	person D	...	person Z	R_t
0.0	nan	nan	5	nan	...	0	2.5
1.0	3	nan	nan	nan	...	nan	3.0
2.0	nan	nan	nan	nan	...	nan	0.0
3.0	nan	nan	nan	3	...	nan	3.0
4.0	nan	6	nan	nan	...	2	4.0

In this scenario, here are the lists of days in which each person becomes infected: A: 1.0 - goes on to infect 3 others B: 4.0 - goes on to infect 6 others C: 0.0 - goes on to infect 5 others D: 3.0 - goes on to infect 3 others E: 0.0 - goes on to infect 0 others, reinfected at 4.0 goes on to infect 2 others

[mghosh00]

Tracking the serial interval of the epidemic

We wish to also take into account the serial interval. For every infection event, the time between the infector's infection_time and the infectee's infection_time will be recorded, and back-tracked to the timestep for which the infector was infected. Then we record a histogram of the these times for each timestep.

Example:	time	person A	person B	person C	person D	person E	person F
0.0	gets infected
1.0	infects B, C	gets infected by A	gets infected by A
2.0	infects D	gets infected by B
3.0	infects E	infects F	gets infected by A	gets infected by C

In this scenario, the serial interval vectors look like: 0.0: [1, 1, 3] - A infects B (1), C (1) and E (3) 1.0: [1, 2] - B infects D (1), C infects F (2) 2.0: [] - D infects no-one 3.0: [] - E and F infect no-one

[mghosh00]

Classes and functions to update:

core

• [x] Person.__init__ (serial_intervals)
• [x] Person.set_exposure_period new
• [x] Person.store_serial_interval new
• [x] tests

routine

• [x] Simulation.configure (add boolean to inf_history_params)
• [x] Simulation.run_sweeps
• [x] Simulation.write_to_serial_file new
• [x] tests

sweep

• [x] PlaceSweep.__call__
• [x] HouseholdSweep.__call__
• [x] SpatialSweep.__call__
• [x] HostProgressionSweep.update_time_status_change
• [x] tests