Skip to contents

cobrar installtation instructions to reconfigure R environment in Windows

Source: README_WIN.md

R Environment Configuration for Package Building (Windows, Rtools + R 4.4.3)

[!IMPORTANT] This guide describes how to configure your R environment to build R packages successfully on Windows using R 4.4.3 and Rtools44 (UCRT64 toolchain). Has also been tested in R 4.4.1. If you your system is using a newer Rtools version, remember to replace all “Rtools44” with e.g. “Rtools45”.

The changes involve:

  • Installation of the libraries libSBML (to read and write files in SBML which is the current model standard) and GLPK (linear problem optimization).

  • Modification of the R environment to indicate the location of the libraries so that they can be used for package compilation.

0. Prerequisite: Install Rtools44

Before making any configuration changes, ensure that Rtools44 is installed on your system. Rtools44 provides glpk and enables installation of libSBML

Download and install from: https://cran.r-project.org/bin/windows/Rtools/

[!NOTE] During installation, make sure to install the UCRT64 toolchain and remember the installation path (usually C:/rtools44).

1. Use Rtools to install libSBML and GLPK

After installing Rtools, open an Rtools Bash terminal , you can launch this APP from the Windows start menu). Within the terminal, run the following command to install libSBML and GLPK

pacman -S mingw-w64-ucrt-x86_64-libsbml mingw-w64-ucrt-x86_64-glpk

2. Create and/or Edit the .Renviron File

In your R home folder, create (or edit if it already exists) a text file named .Renviron.

Location: The R home folder is the folder where R starts by defaulf. It can be obtained after starting R by typing

The typical location is

C:/Users/USERNAME/Documents/.Renviron

(Make sure that you replace your USERNAME by the relevant information in your system)

Contents: The file should define a PATH variable with information on the location of the R installation and of Rtools. It should also contain the location of a Makevars.win file (that will be generated in the followint steps)

PATH=C:/Program Files/R/R-4.4.3/bin/x64;${PATH};C:/rtools44/ucrt64/bin
R_MAKEVARS_USER=C:/Users/USERNAME/Documents/.R/Makevars.win

Make sure to replace the R version and the USERNAME by the relevant information in your system.

💡 Notes:

  • Always use forward slashes (/) instead of backslashes (\) in R path settings, even on Windows.
  • The PATH variable ensures R and Rtools are both available in your environment.
  • The R_MAKEVARS_USER variable tells R where to find your custom Makevars.win file.

3. Create or Edit the Makevars.win File

Inside your R home folder check for a folder called .R . If it is not present, create one. Inside the .R folder create or edit a text file named Makevars.win:

Location:

C:/Users/UNSERNAME/Documents/.R/Makevars.win

(Make sure that you replace your USERNAME by the relevant information in your system)

Contents:

UCRT64 = C:/rtools44/ucrt64
PKG_CPPFLAGS += -I"$(UCRT64)/include"
PKG_LIBS     += -L"$(UCRT64)/lib"
PKG_CPPFLAGS += -I"$(UCRT64)/include/sbml/packages"
PKG_LIBS     += -L"$(UCRT64)/lib"
LIBRARY_PATH += -I"$(UCRT64)/include/sbml/packages"

💡 Explanation:

  • UCRT64 defines the path to the Rtools64 UCRT directory.

  • PKG_CPPFLAGS adds include paths for headers (including the required SBML packages).

  • PKG_LIBS adds library search paths.

  • LIBRARY_PATH extends the linker search path to find additional libraries.

4. Purpose of These Files

  • .Renviron — defines environment variables for R at startup, making sure it knows where to find compilers and tools.
  • Makevars.win — controls compiler and linker settings when you build R packages from source. Together, they ensure R uses the correct build toolchain (Rtools44 / UCRT64) and library paths.

5. Restart R and Verify the Setup

Before installing the package, make sure to restart R to ensure that the updated environment variables are loaded.You can confirm your environment is correctly configured by running the following in R:

Sys.getenv("PATH")
Sys.getenv("R_MAKEVARS_USER")

The location of the rtools should be shown in the PATH (among many other locations) and the R_MAKEVARS_USER should show the location of the Makevars.win

6. Install the cobrar Package

Once your environment is configured and verified, you can install the COBRA-R package directly from GitHub using the remotes package.

Note: You may need to install the remotes package first:

install.packages("remotes")

Once the remotes packages is installed you can just get cobrar by running

remotes::install_github("Waschina/cobrar")

After the installation is completed you can just load the package using

If it loads without errors, your configuration is correct and ready for use.