Skip to content

CLAUDE.md

Latest commit

 

History

History
80 lines (63 loc) · 2.95 KB

CLAUDE.md

File metadata and controls

80 lines (63 loc) · 2.95 KB

CLAUDE.md

Project Overview

EpiModel-Gallery is a collection of template examples for extending the EpiModel R package to model infectious disease dynamics over networks. Examples focus on stochastic network models using ERGMs (exponential random graph models). EpiModel itself is documented at https://github.com/statnet/EpiModel.

Repository Structure

Each subdirectory is a self-contained example named YYYY-MM-Description/ containing:

  • README.md — explanation of the model, authors, and details
  • model.R — main script for network estimation and epidemic simulation
  • module-fx.R — custom module functions plugged into control.net()

Top-level files:

  • DESCRIPTION — R package metadata (used for CI dependency management)
  • test.sh — test runner that executes each example's model.R
  • EpiModel-Gallery.Rproj — RStudio project config

Gallery Examples

Directory Model
2018-08-AddingAnExposedState SEIR/SEIRS — adding exposed (latent) state to SIR
2018-08-ObservedNetworkData Epidemics over observed (census) dynamic networks
2018-08-SIwithVitalDynamics SI with aging, births, and deaths
2018-08-TestAndTreatIntervention SIS with testing and treatment for bacterial STIs
2018-09-CompetingStrains SIS with two competing pathogen strains
2018-09-SocialDiffusion Information/behavior diffusion using SI framework
2018-10-SEIRwithAONVax SEIR with all-or-nothing vaccination and vital dynamics
2018-11-Syphilis Multi-stage syphilis model with diagnosis/treatment
2018-12-SEIRSwithLeakyVax SEIRS with leaky vaccination and vital dynamics
2019-03-HIV HIV with acute/chronic/AIDS stages and ART treatment
2021-10-CostEffectivenessAnalysis SI with cost-effectiveness (costs, QALYs, discounting)
2022-12-Multinets Multiple interacting networks with shared node set

Language and Dependencies

  • Language: R (>= 4.0)
  • Primary dependency: EpiModel (>= 2.4.0)
  • Other imports: dplyr, networkDynamicData
  • Suggested: testthat, dampack
  • Code style: 2-space indentation, UTF-8 encoding

Testing

Run all examples:

bash test.sh

This iterates through each subdirectory (excluding renv/), runs Rscript <dir>/model.R with an error handler, and reports pass/fail with timing.

Unit Test Pattern

Every model.R must include these lines near the top (after loading EpiModel):

# Standard Gallery unit test lines
rm(list = ls())
eval(parse(text = print(commandArgs(TRUE)[1])))

And use conditional parameters for interactive vs. CI runs:

if (interactive()) {
  nsims <- 5
  ncores <- 5
  nsteps <- 800
} else {
  nsims <- 2
  ncores <- 2
  nsteps <- 200
}

Contributing a New Example

  1. Create a new subdirectory named YYYY-MM-Description/
  2. Add three files: README.md, model.R, module-fx.R
  3. Include the standard unit test lines in model.R
  4. Ensure bash test.sh passes before submitting