Summary and Schedule
Author: Manveer Chauhan, Clark Lab, The University
of Melbourne.
Contributors: Vini Salazar, Susan (Jiayi) Si, Raymond
(Bibo) Zhao, Emma Gail, Melbourne Bioinformatics.
Last updated November 2025
Overview
Topic
Skill level
Content:
Section 1: Setup, Quality Control and Sample Integration.
Section 2: Differential Gene Expression when dealing with two
treatment conditions.
Section 3: Differential Expression using a pseudobulk approach
and DESeq2.
| Setup Instructions | Download files required for the lesson | |
| Duration: 00h 00m | 1. Introduction | How to perform differential expression analysis with Seurat? |
| Duration: 00h 12m | 2. Setup, Quality Control and Sample Integration |
How do we identify and remove low-quality cells in scRNA-seq
data? What signs suggest batch effects between treatment conditions? When and why do we need to integrate datasets before downstream analysis? How do Harmony and Seurat CCA compare in aligning similar cell types? |
| Duration: 00h 24m | 3. Differential Gene Expression when dealing with two treatment conditions |
How do conserved markers help us label clusters reliably across
conditions? What exactly do avg_log2FC, pct.1, pct.2, and p_val_adj mean in FindMarkers? Why must DE be run within a cell type (e.g., CD16 Mono_STIM vs CD16 Mono_CTRL) rather than “all cells”? |
| Duration: 00h 36m | 4. Differential Expression using a pseudobulk approach and DESeq2 |
Why aggregate to pseudobulk instead of testing each cell? How does donor act as the biological replicate in DESeq2? When do pseudobulk and single-cell DE agree or disagree, and why (e.g., zero inflation, composition, variance models)? What minimum sampling per group (cells per donor, donors per group) is sensible before aggregating? |
| Duration: 00h 48m | Finish |
The actual schedule may vary slightly depending on the topics and exercises chosen by the instructor.
Some knowledge of R and scRNA-seq analysis is assumed. We recommend reviewing the following materials before the starting the materials: - Introduction to R - 10X single-cell RNA-seq analysis in R
Please see our Intro to R and RNA-seq: counts to genes workshops if you need a refresher.
Data Sets
Click here to download the RData object that will be used for this workshop.
Software Setup
RStudio Setup
We use RStudio for coding in R.
Click here and follow the instructions to install RStudio Desktop in your system.
R packages
Run the code block below to install the packages needed for this workshop.
To check if installed properly, load each package in one at a time
using the library() function.
R
install.packages('Seurat')
if (!requireNamespace("remotes", quietly = TRUE)) {
install.packages("remotes")
}
remotes::install_github("satijalab/seurat-data", quiet = TRUE)
if (!require("BiocManager", quietly = TRUE)) {
install.packages("BiocManager")
}
BiocManager::install("DESeq2")
BiocManager::install("multtest") # dependency commonly missing
BiocManager::install(c("SingleR", "celldex"))
install.packages("harmony") # dependency needed for harmony analysis
install.packages("tidyverse")
install.packages("pheatmap")
install.packages("metap")
install.packages("ggplot2")