AscRtain.shiny

This pedagogical shiny app is a wrapper for functions in the https://github.com/explodecomputer/AscRtain package.

It builds on work by Groenwold, Palmer and Tilling (2019) "Conditioning on a mediator" https://osf.io/vrcuf/, and forms part of the Nature Communications paper "Collider Bias undermines our understanding of COVID-19 disease risk and severity" (2020), authored by MRC-IEU colleagues.

To run

Install dependencies

remotes::install_github("explodecomputer/AscRtain")
install.packages(c("dplyr","shiny", "shinydashboard", "shinycssloaders", "plotly", "latex2exp", "ggplot2"))

Run from github:

shiny::runGitHub("Zimbabwelsh/AscRtainShiny", subdir="app")

To deploy

Runs on the rocker/shiny-verse image.

docker-compose up -d --no-deps --build