Start Rstudio @ workbench

Task: Sign into workbench and start a new “Rstudio Pro” session (You may want to give it a suitable name, e.g. “DESeq2”)

Lecture Material

Task: Create a new project and clone the datafiles we will be using from github.

In your Rstudio session:

  • File -> New Project -> ‘Version Control’ -> git
  • create the project as a subdirectory under your group volume
    • e.g. /data/PI/group/MYFOLDER, or /scratch/local/MYFOLDER
  • checkout the 2025 branch of the repository (Git > 2025)

Verification

If it worked, you should have a new session, and in your console you will retrieve your project directory:

getwd()

and you should be able to see the data folder:

list.files(‘data’)

which should contain these folders:

  • genesets
  • mpp
  • myeloma

Task: Find the file rmd/packages.Rmd and open it in Rstudio.

This contains the content of the current lecture in R-markdown format (text and code cells)

Packages

Generalities

Many useful functions are already available with R: base packages.

Most analysis tools come with software packages that will need to be installed before use.

For R, two of the most common and reliable package repositories are:

Reminder

  1. All packages come for a specific R-version.
  2. All packages have specific package version, and can come from different sources (CRAN, Bioconductor, github, …).
  3. Most packages have dependencies on other packages that should also be installed
  4. Installation can be very time consuming but needs to be done only once.

Poll 0.1: Which R-version are you running?

In the following we will use R-version: 4.2.3 Please make sure to select this version in the drop-down menu (upper right)

DESeq2 and Dependencies

In this course, we will use several additional R packages and their dependencies.

We have to make sure these are installed - in a suitable location defined by .libPaths().

Poll 0.2: What is the first element in your .libPaths() ?

Install packages

The following code-cell installs the packages we need in this course - for each user.

Task: Find the corresponding line in your markdown document and “Run the chunk” by pressing “Play button” or “Shift-Command-Enter”

Be patient - these installations can take a while.

# Bioconductor
if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
if (!require("DESeq2", quietly = TRUE))
    BiocManager::install("DESeq2")
if (!require("EnsDb.Hsapiens.v75", quietly = TRUE))
    BiocManager::install("EnsDb.Hsapiens.v75")

# CRAN
if (!require("tidyverse", quietly = TRUE))
    install.packages("tidyverse")
if (!require("pheatmap", quietly = TRUE))
    install.packages("pheatmap")
if (!require("ggrepel", quietly = TRUE))
    install.packages("ggrepel")
if (!require("UpSetR", quietly = TRUE))
    install.packages("UpSetR")
if (!require("ashr", quietly = TRUE))
    install.packages("ashr")

Use packages

Task: Load the following packages: DESeq2, tidyverse, pheatmap

library(DESeq2)
library(tidyverse)
library(pheatmap)

Poll 1.2: Did you manage to load DESeq2, tidyverse & pheatmap ?

Reminder: Loading new packages enables new functionalities for a given R-session.

This may result in multiple (sometimes redundant) functions with the same name: e.g. sd()

In case of doubt, use package name explicitly: e.g. stats::sd()

Task: Use sessionInfo() to inspect which packages you have just added an their versions.