2024-2: 5a-Introduction - Differential Expression (DE) using DESeq2
Aim of today
Identify statistically significant (FDR < 0.05) differentially expressed genes.
Visualise results with PCA plots, heatmaps and volcano plots.
Requirements
Run your samples (FASTQ) using the nextflow nf-core/RNA-seq pipeline using ‘star_salmon’ (Task 4 session) or an alternative pipeline that generates feature counts.
Get Rstudio working for you - Option 1 below for in-class session
Installing R and Rstudio
The analysis scripts in this guide are written in R script. We will be using RStudio, a front-end gui for R, to run the analysis scripts.
You have three main options for running this analysis in RStudio:
Use QUTs rVDI virtual desktop machines
Install R and RStudio on your own PC
Use the provided PCs in the QUT computer labs
Option1: Use QUTs rVDI virtual desktop machines - PREFERRED WAY TO DO THIS FOR WORKSHOP
This is the preferred method, as R and RStudio are already installed, as are all the required R packages needed for analysis. Installing all of these can take over 30 minutes on your own PC, so using an rVDI machine saves time.
rVDI provides a virtual Windows desktop that can be run in your web browser.
To access and run an rVDI virtual desktop:
Go to https://rvdi.qut.edu.au/
Click on ‘VMware Horizon HTML Access’ on the right hand side and select R_Workshop (if you have > 1 available to you).
Log on with your QUT username and password
*NOTE: you need to be connected to the QUT network first, either being on campus or connecting remotely via VPN.
Option2: Install R and RStudio on your own PC
Go to the following page https://posit.co/download/rstudio-desktop/ and follow the instructions provided to install first R and then Rstudio.
Download and install R, following the default prompts:
https://cran.r-project.org/bin/windows/base/
Download and install RStudio, following the default prompts:
https://posit.co/download/rstudio-desktop/
Option3: Use the provided PCs in the QUT computer labs
The PCs in the computer labs already have R and RStudio installed. If using this option, you will need to install the required R packages (unlike rVDI). The code for installing these packages is in the analysis section below.
We will now perform the following tasks using Rstudio
Preparing your data for Differential Expression analysis. Two data files are needed for this analysis: a samples table and your count table
R packages
Install required R packages (only need to run once) - after installation, we only need to load the packages. NOTE: If using an rVDI virtual machine, the R packages are already installed
Load required R packages. Unlike installing the packages, this needs to be done every time you run the analysis
Import your data files (count table and samples table) into R
Checking for outliers and batch effects
PCA plot
Pairwise samples heatmap
Identify differentially expressed (DE) genes using DESeq2
Annotating your DE genes
Volcano plot
DE genes heatmap