R package installation
Objectives
-
Learn how to run R efficiently on Tufts HPC using both graphical and command-line interfaces.
-
Understand R package management, including installation and configuration on the HPC environment.
-
Develop skills to run R scripts in batch mode, submit jobs to the queue, and manage multiple jobs effectively.
How to run R on Tufts HPC
1. Open OnDemand Rstudio app
1.1 Log in to Open OnDemand with your UTLN
https://ondemand.pax.tufts.edu/
1.2 You will see RStudio Pax under Interactive Apps and other topic-specific Rstudio apps under Bioinformatics Apps
1.3 Fill in the parameters according to your needs. You can start with default value.
1.4 Launch the job.
2. Command line interface
2.1 Make sure you are on the compute mode
srun -p interactive -n 1 --time=02:00:00 --mem 4g --pty bash
2.2 Load R modules
module load R/4.4.1
2.3 Launch R interactive session
R
R package installation
Pre-installed R packages
To provide convenience to users, we pre-installed many commonly used R packages used in bioinformatics analysis. Before you try to install packages, load the module for R, start R and check if the package is already installed.
library(packageName)
Get the list of installed R packages
installed.packages()
Install R packages
install.packages("packageName")
Rstudio
.libPaths()
This shows the folders where the packages are installed.
> .libPaths()
[1] "/home/user/R/x86_64-pc-linux-gnu-library/4.4" # Personal library
[2] "/usr/local/lib/R/site-library" # System-wide library
[3] "/usr/lib/R/library" # Default R library
You can change the libpath to folder under lab storage or research projects.
.libPaths("your_path_to_install_packages")
For R/4.4.1, try not to change libpath, just install to your default $HOME directory.
R modules
r base
R/4.0.0
R/4.1.1
R/4.3.0
R/4.4.0
R/4.4.1 # Recommend to use this version
r-bioinformatics
$ module av r-bioinformatics
------- /cluster/tufts/biocontainers/modules ---------------
r-bioinformatics/4.3.2 r-bioinformatics/4.4.0 (D)
Where:
D: Default Module
r-scrnaseq
$ module av r-scrnaseq
------- /cluster/tufts/biocontainers/modules -----------------------------
r-scrnaseq/4.2.3 r-scrnaseq/4.3.1 r-scrnaseq/4.3.2 r-scrnaseq/4.4.0 (D)
Where:
D: Default Module
TIPS
For new projects, it's recommended to use the latest version, R/4.4.1, as it comes with the most comprehensive set of installed packages (1300+).
For older scripts, either those you created some time ago or inherited from a colleague, be mindful of the R version and libPaths() used in those scripts. Do not run it without paying attention to the version.
Bioconductor
Bioconductor is both an open source project and repository for R packages related to the analysis of biological data, primarily bioinformatics and computational biology, and as such it is a great place to search for -omics packages and pipelines.
Do not run the belew code to install bioconductor core packages
We have pre-installed bioconductor core packages for users.
if (!require("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install(version = "3.xx")
Installation with BiocManager
To install specific bioconductor packages, use BiocManager to install.
Ex: Let's install adverSCarial
BiocManager::install("adverSCarial")
devtools
devtools is an R package designed to simplify the process of developing, testing, and installing R packages from various sources.
devtools has been pre-installed in our R modules. You can load it to install packages require it.
library(devtools)
Installing Packages from GitHub
Ex: Install an R package STew used to jointly characterize the gene expression variation and spatial information.
.libPaths("/home/user/R/x86_64-pc-linux-gnu-library/4.4")
devtools::install_github("fanzhanglab/STew")
Notice that for many R package, especially those process scRNA-Seq or spatial transcriptomics data, it often require dependencies or version-specific packages.