comoexplore

This is a shiny app accompanied with the R package comomodels to interactively explore the model behaviors of in response to different transmission parameters.

Currently we support five sub-models:

SEIRD - basic Susceptible-Exposed-Infected-Recovered-Dead model;
SEIaImIsRD - a symptom-compartment SEIRD model that divide the infectious population (I) into asymptomatic (Ia), mild (Im) and severe (Is) groups according to the severity of symptoms;
SEIRDAge - an age-structured SEIRD model that separate the population into different age ranges.
SEIRDV - an SEIRD model that includes the vaccinated population group.
SEIRD_RU - an SEIRD model that considers two interacting communities: urban and rural.

Development

To contribute to this repository, please read DEVELOPMENT.md

How to run the app

Run the deployed app online

Click here to easily explore comomodels-explore online.

Run the app locally (linux)

In addition to the online version, comomodels-explore is also open accessible and can be run locally.

Dependencies

Packages

A number of R packages are required for the app. For the details please check DESCRIPTION and the conda environment list comomodels-explore.yml

If you have a number of dependencies not installed yet, instead of manually installing each of them, you can save time and energy by installing all dependencies in a conda environment:

Install conda

Create a package-specific conda environment:

# clone the repository with git, go to the cloned directory:
git clone git@github.com:Como-DTC-Collaboration/comomodels-explore.git
cd comomodels-explore

# create a new conda environment (here with name "comoexplore")
# with all dependencies listed the provided .yml file:
conda env create -f envs/comomodels-explore.yml

# All dependenceis required by this package are now installed in comoexplore. Now activate the environment:
conda activate comoexplore

The core package comomodels need to be manually installed in R:

devtools::install_github("Como-DTC-Collaboration/como-models")

Data
- Some models (e.g. SEIRDAge and SEIRD_RU) require the loading of extra data, such as social contact matrices between age groups and country-specific population data, etc. The default data in comomodels are in .rda format and also provided here in the ./data folder, which are fully usable as an example. Referring to the same format, Users can also create and use customised data.

Installations

Method 1 - the easiest way is to directly download and run the remote version of the app in one command in R:

# 1. use runGitHub
shiny::runGitHub("comomodels-explore", "Como-DTC-Collaboration", ref = "main")

# 2. alternatively, run a tar or zip file
shiny::runUrl("https://github.com/Como-DTC-Collaboration/comomodels-explore/archive/main.tar.gz")
shiny::runUrl("https://github.com/Como-DTC-Collaboration/comomodels-explore/archive/main.zip")

Method 2 (recommended for developers) - you can first clone the git repository, then run the app:

git clone git@github.com:Como-DTC-Collaboration/comomodels-explore.git
cd comomodels-explore/

# start an R environment
R

In the R environment, make sure your working directory is comomodels-explore and all dependencies are installed. Then run the app

setwd(".")
## if you are using an interactive R session (e.g. RStudio):
shiny::runApp(appDir = ".")
## otherwise, set `launch.browser=FALSE` and manually launch the browser in the next step:
shiny::runApp(appDir = ".", launch.browser=FALSE)

Either way, a local URL will appear on your R console, e.g.:

Listening on http://127.0.0.1:5554

The corresponding interface may automatically pop out. If not, you can manually copy-and-paste the above URL to any web server.

You can now play around with the parameters on the interface.