The COVID19 Campus Model is a tool to guide mitigation strategies and resource needs for universities considering in-person instruction for Fall 2020. We developed a susceptible-exposed-infectious-recovered (SEIR) type of deterministic compartmental model that captures transmission process, allowing for estimates of direct and indirect (transmission-mediated) effects of control strategies.
Code in this repository was used for the simulations and results presented our paper. An interactive software tool is available for additional simulations. Links below:
| File | Description |
|---|---|
| Transmission scenarios | Explores different transmission scenarios |
| Screening scenarios | Explores a range of screening intervals and with an outcome of total and cumulative cases in staff and students |
| Testing scenarios | Explores a range of testing interval (time from symptom-onset to testing and subsequent isolation) |
| Screen and test scenarios | Explores combinations of screening and testing intervals |
| File | Description |
|---|---|
| Dependencies | Loads relevant packages |
| Model function | Loads function for transmission model |
| Parameters & initials | Loads paramemeters, initial conditions and control settings |
| Distribution optimizer | Loads function used to optimize beta distribution for the probabilistic senstivity analysis (PSA) |
| PSA parameters | Loads parameters and distributions used for the PSA |
| PSA plot function | Loads plot function used for the PSA |
| File | Description |
|---|---|
| Plots | Folder containing figures from model outputs |
The application is available online.
In order to run the app from your computer you require the following libraries:
install.packages(c(
"shiny",
"shinydashboard",
"plotly",
"EpiModel",
"lhs",
"tidyr",
"purrr",
"glue",
"kableExtra",
"ggplot2"
))Alternatively if you use renv to isolate your libraries and ensure reproducibility, a "renv.lock" file is present at the root of this project.
renv::restore()To launch the app, either run from the R console:
shiny::runApp("shiny/app")Or using RStudio, open either "shiny/app/ui.R" or "shiny/app/server.R" and click the "Run App" button on the top right corner of the editor.