school.epi.abm

The goal of school.epi.abm is to implement an agent-based model of COVID infection and mitigation dynamics in K-6 schools, as described in https://doi.org/10.3389/fpubh.2023.856940.

Installation

You can install the development version of school.epi.abm like so:

# install.packages("devtools") # if not already installed
devtools::install_github("UCLA-PHP/school.epi.abm")

Loading the package

After instalation, you can load the package with the library() function:

library(school.epi.abm)

Examples

The graphical user interface (GUI) for the model can be run using the following command:

shiny::runApp()

The GUI is also deployed at https://agent-based-models.shinyapps.io/RegionalCOVIDSchoolSimulation/.

The model can also be run directly from the R command line using the following commands:

library(school.epi.abm)

simulation_outputs = run_simulation(
  n_schools = 10,
  verbose = TRUE)

We can summarize and visualize the results as follows:

average_class_data_by_day = summarize_records(simulation_outputs$class_records)

plot1a = plot1_plotly(
  class_records,
  plot3_ymax = 6)

print(plot1a)

analyze_results(simulation_outputs)

The subfolder inst/extdata contains the analysis script ABM results loop.R, which can be used to loop over a table of input conditions (specified in that script as the tibble conditions_to_cross) and reproduce the tables in the corresponding article (https://doi.org/10.3389/fpubh.2023.856940).