Differential Gene Expression Analysis

Overview

Participants will learn how to use DESeq2 to identify differentially expressed genes and become familiar with design matrices. The session will also introduce concepts essential for working with real-world data, such as incorporating covariates and interaction terms into models. Additionally, participants will learn how to apply Over-Representation analysis (ORA) and Gene Set Enrichment Analysis (GSEA) for uncovering biologically meaningful patterns. Hands-on examples and exercises will help reinforce concepts and guide participants through a typical RNA-seq workflow.

Requirements

• Familiarity with R
• Access to an IDE (Rstudio is recommended)

Sofware

• R (>=4.0.0)
• Rstudio

In R, install the following libraries (in case they are not already installed):

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install(c("DESeq2", "vsn", "EnhancedVolcano", "fgsea"))
install.packages(c("hexbin", "pheatmap", "RColorBrewer", "ggplot2", "ggbeeswarm", "ashr", "knit", "GSA", "enrichR"))

Outline

The code can be run in Google Colab.

Alternatively, you can also run locally. These are the notebooks we will cover during our workshop:

• 01 DESeq2
• 02 GSEA and ORA

References

This workshop uses processed RNA-seq counts from this study.

This workshop used material adapted from the following sources:

• DESeq2 vignette
• RNA-seq workflow
• fgsea vignette
• enrichR vignette
• MiCM intro to RNA

Workshop created as part of the McGill Initiative in Computational Medicine.