Skip to contents

Installation

Source: INSTALLATION.md

Prerequisites

SelectSim requires R ≥ 3.5 and includes compiled C++ code via Rcpp and RcppArmadillo. Before installing, make sure you have a working C++ compiler:

  • macOS — Install Xcode Command Line Tools: xcode-select --install
  • Linux — Install build tools, e.g. on Ubuntu/Debian: sudo apt install build-essential
  • Windows — Install Rtools and ensure it is on your PATH

Option 1 — Install from GitHub (standard)

Install directly from GitHub using pak:

# install.packages("pak")
pak::pak("CSOgroup/SelectSim")

Or using devtools (legacy):

# install.packages("devtools")
devtools::install_github("CSOgroup/SelectSim", dependencies = TRUE, build_vignettes = TRUE)

Option 2 — Install with micromamba (recommended)

micromamba provides a fully isolated, reproducible R environment. This is the recommended approach if you are on a shared server, HPC cluster, or want to avoid conflicts with your existing R installation.

Step 1 — Install micromamba

Follow the official micromamba installation guide for your platform.

Step 2 — Create and activate an R environment

micromamba create -n r_env
micromamba activate r_env

Step 3 — Install R and required system dependencies

micromamba install \
  conda-forge::r-base \
  conda-forge::r-essentials \
  conda-forge::r-devtools \
  conda-forge::r-pak \
  conda-forge::cmake \
  conda-forge::r-rcppparallel \
  conda-forge::r-rfast

Step 4 — Install SelectSim inside R

# install.packages("pak")
pak::pak("CSOgroup/SelectSim")

Option 3 — Restore the exact development environment

A fully pinned conda environment file is provided at dev/environment.yml. This reproduces the exact software stack (R version, all system libraries) used during development.

micromamba create -f dev/environment.yml
micromamba activate r_env

Then install SelectSim:

pak::pak("CSOgroup/SelectSim")

Note: The environment file pins exact package versions and is Linux-specific (linux-64). It may not work on macOS or Windows.

Verifying the installation

After installing, confirm everything works with a quick test run:

library(SelectSim)
data(luad_run_data, package = "SelectSim")

result <- selectX(
  M                = luad_run_data$M,
  sample.class     = luad_run_data$sample.class,
  alteration.class = luad_run_data$alteration.class,
  n.cores          = 1,
  min.freq         = 10,
  n.permut         = 10    # small number for a quick smoke test
)

A successful run returns a list with a result data frame of evolutionary dependencies. For a full analysis, use n.permut = 1000 or higher.

Troubleshooting

rfast fails to install rfast requires a Fortran compiler. On macOS, install gfortran from mac.r-project.org/tools. On Linux, install via your package manager (e.g. sudo apt install gfortran). With micromamba, conda-forge::r-rfast handles this automatically.

Vignettes not building Ensure pandoc is installed (pandoc --version in a terminal). Install via brew install pandoc (macOS), sudo apt install pandoc (Linux), or through the micromamba environment (conda-forge::pandoc).

Compilation errors on Windows Ensure Rtools is installed and its bin/ directory is on your PATH. Run pkgbuild::check_build_tools() in R to verify your toolchain is detected correctly.

Vignettes missing after install pak::pak() does not build vignettes by default. If you need the vignettes locally, use:

devtools::install_github("CSOgroup/SelectSim", dependencies = TRUE, build_vignettes = TRUE)