Guide for the Differential Expression Analysis of RNAseq data using DESeq2

---

Developed by:

David R. Requena Anicama, Ph.D.

• Author's name: David Requena | Google Scholar | ORCID: 0000-0002-5968-1133

Daniel F. Guevara Díaz, B.Sc.(s)

• Author's name: Daniel F. Guevara-Díaz | Google Scholar | ORCID: 0009-0001-2786-8729

---

Description

This guide provides a comprehensive methodology for performing Differential Expression Analysis (DEA) to identify genes significantly associated with specific conditions or diseases using RNA-Seq data. The process is streamlined with the use of our custom library, OmicsKit, to automate various steps and optimize the analysis workflow. The steps include:

1. Set up

   • Install and/or call the required libraries
   • Include OmicsKit library
   • Manage sample metadata
   • Create the DESeq2 object

2. Exploring confounders

   • PCA plot
   • tSNE plot
   • UMAP plot
   • HeatMap of Samples

3. Data Analysis

   • Retrieve comparisons
   • Obtain mutually exclusive cases
   • Annotation and output tables
   • Apply detectability filter

4. Plotting

   • Histogram of p-values
   • Dispersion Estimates
   • MA Plot
   • Volcano Plot
   • HeatMap Samples vs genes
   • Box-Scatter-Violin (BSV) plots

5. Pathway Enrichment

   • Overrepresentation Analysis (ORA) and Gene Set Enrichment (GSE)
   • Manage results from Gene Set Enrichment Analysis (GSEA) results' data frames
   • Create balloon plots and concurrency networks of gene sets/pathways

Requirements

To run the analysis script, the following input files are required:

1. Mandatory Files:

   • A metadata table containing sample information (e.g., case/control status, gender).
   • A table of raw gene counts by sample.

2. Optional Files:

   • A table specifying genes to be excluded from the analysis (e.g., ribosomal genes).
   • A table listing genes of interest for creating individual gene-specific plots.