Tutorial: R and RStudio setup

This is a short, practical guide for new lab members setting up R and RStudio on their own laptop (MacBook first).

Official references:

• CRAN macOS R installers: https://cran.r-project.org/bin/macosx/
• RStudio Desktop download: https://posit.co/download/rstudio-desktop/

1) Check your Mac architecture

uname -m

• arm64 → Apple Silicon (M1/M2/M3/M4 and newer)
• x86_64 → Intel Mac

2) Install R (from CRAN)

1. Open CRAN macOS page: https://cran.r-project.org/bin/macosx/
2. Download the correct installer for your architecture:
   • Apple Silicon: R-*-arm64.pkg
   • Intel: R-*-x86_64.pkg
3. Run the .pkg installer and finish setup.

Verify in Terminal:

R --version

3) Install RStudio Desktop

1. Download from Posit: https://posit.co/download/rstudio-desktop/
2. Install RStudio Desktop.
3. Launch RStudio and confirm it detects your R installation.

4) First-time setup in RStudio

Open RStudio Console and run:

install.packages(c("tidyverse", "data.table", "remotes"))

Set a CRAN mirror when prompted.

5) Create a project (recommended workflow)

In RStudio:

• File → New Project
• choose a dedicated folder per project
• keep scripts/data/results together

This helps reproducibility and collaboration.

6) Optional: project-local package management with renv

Inside a project, run:

install.packages("renv")
renv::init()

After installing/updating packages:

renv::snapshot()

A collaborator can restore the same package set with:

renv::restore()

7) Common issues

• RStudio says “R not found”:
  • install/reinstall R from CRAN first, then restart RStudio.
• Package install fails on macOS:
  • update to current R version first,
  • then retry package install.
• Apple Silicon users with old Intel-only package binaries:
  • update R and packages; avoid mixing very old package builds.

8) Lab recommendation

Use one RStudio Project per study and use renv for reproducibility before sharing analysis code.