Libraries

Poll 0.1: What is the first element in your .libPaths() ?

We will use a couple of libraries in this course. We have to make sure these are installed - in a suitable location defined by .libPaths().

These installations can take a while. So it’s best to tackle this first. The following code chunk will checks if a required library is available and will attempt to install it if this is not the case.

# 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")

After running the above chunk, make sure the libraries can be loaded:

library("DESeq2")
library("tidyverse")
library("pheatmap")
library("ggrepel")
library("EnsDb.Hsapiens.v75")
library("UpSetR")
library("ashr")

What you will (hopefully) learn

After the course you will be able to

Perform a full bioinformatics workflow (DESeq2): data input \(\to\) output
Understand the theory behind RNA-seq data generation and analysis
Perform exploratory analysis: quality controls, filters
Visualize data and results with modern tools
Apply complex (multifactorial) experimental designs
Obtain and visualise differential genes from simple or complex comparison setups
Share workflows with your colleagues.

Credits

There is lots of useful material out there. This course is heavily influenced by those tutorials:

We will also be using publicly available data, which has been altered for educational purposes (and should thus be treated as such).

Overview

1. Day I: Getting and understanding the data

Warmup & R-epetition
Data, Metadata & Design
Data Exploration
Quality Control

2. Day II: Data Transformations and Model

Quality Controls & Filtering
Data Transformations and Normalization
Size Factors
Modeling Count Data: NB and Over-dispersion
Running DESeq2

3. Day III: Differentially Expressed Genes

Multifactorial Designs & Complex Contrasts
Hypothesis tests: Wald test & LRT
Inspection & Visualization
Gene Annotations
Exporting results

4. Day IV: On your own

Run a multifactorial DESeq2 Analysis. We’ll be around to help.

Recap

The Rstudio GUI.

An Rscript in the Rstudio GUI

An R notebook in the Rstudio GUI

Poll 0.2: How will you keep track of the code, course material and notes during this course ?

Set up our project

Workbench users

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

If you are working in workbench, start a new session.

In your session (local or workbench):

file -> new project -> ‘Version Control’ -> git

Repository URL: https://github.com/maxplanck-ie/Rdeseq2
Project directory name: chose a folder name

create the project as a subdirectory under your group volume (e.g. /data/PI/group/MYFOLDER), or a processing volume (/data/processing/MYFOLDER), or the scratch space (/scratch/local/MYFOLDER)
Remember to choose version 4.2.3

Local users

Go to https://github.com/maxplanck-ie/Rdeseq2 , click on the green ‘< > Code’ dropdown button and select ‘Download ZIP’. Save the file to an appropriate location and extract its content.

Now open your Rstudio instances, and set the working directory to the folder you have just extracted.

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

How it should look like:

An new R notebook in our freshly cloned project

DESeq2 Analysis with R: outline

Bioinfo-Core @ MPI-IE

Sat May 25 12:13:00 2024