This file will contain brief descriptions of the sample file formats in the sample_input_files directory, or other descriptions where suitable. This will likely be deprecated by a later move to HDF5 I/O, but for development for now we need to use the data in the formats we have it (as of Thursday, May 7, 2020)
This contains epidemiological rate parameters for movement within the compartmental model split by ages. Each row represents a transition between two compartmens within a specific age.
Same as above, but this splits the asymptomatics into A (pre-symptomatic) and A_2 (never symptomatic).
Use this with human/infectious-compartments/2/data.csv
This file has the population in each region (Scottish healthboards) stratified by age and sex. Each row has the following columns:
- Health_board is the name of the geographic unit.
- Sex is male or female.
- Age is the age group (eg.: [0,17) is the age group from 0 to 17 years old)
- Total is the number of people inside the the categories described by the other columns
This contains origin-destination flow data during peacetime for health boards in Scotland, derived from the dataset wu01uk.
This is essentially a weighted edge list.
- First column is the source location label
- Second column is the destination location label
- Third column is the number of individuals undertaking that journey as reported in wu01uk
- Fourth column is a multiplier that gives control over how much dampening factors impact this edge
This comes from is from https://cmmid.github.io/topics/covid19/comix-impact-of-physical-distance-measures-on-transmission-in-the-UK.html, and includes all imputed contacts (both face-to-face conversation and physical contact). This is the entire comix matrix, for all age groups, which is not compatible with the population data we have this far. Look the human/mixing-matrix/1/data.csv below.
This is a sample simplified square matrix describing age mixing between aggregated age categories, generated from human/full-mixing-matrix/1/data.csv by summing over the columns that need to be aggregated, and then summing (weighted by the approximate population in each category) over the rows that need to be aggregated. This includes both face-to-face and physical contacts, and is not yet down-weighted to approximate infection probability. Age proportion estimated from https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/articles/overviewoftheukpopulation/august2019.
This table has three columns: Time, Movement Multiplier and Contact Multiplier. For instance, if the file looks like this:
Time,Movement_Multiplier,Contact_Multiplier
0,1.0,1.0
3,0.7,0.4
10,0.9,0.6
The movement multiplier controls how much the movements between nodes is scaled up or down, whereas the contact multiplier controls how to the within node contact is increased or slowed down over time.
The values in this table are the probability that a contact between an infectious person with a susceptible person will result in a transmission over time.
A list of compartments that are considered infectious.
Same as above, but includes the A_2 compartment
A list with seed nodes, where the outbreak will start for a given simulation. This table contains both the region ID and number of infected people.
Whether to use the stochastic or deterministic version of the model. A production run will usually be stochastic, but the deterministic run can be useful during development.
The seed to be used by the model during the stochastic run. This is ignored for deterministic runs.