README: Bulk RNA-Seq Analytics

Overview

This repository contains a complete workflow for the analysis of bulk RNA-Seq data. It includes scripts and documentation for processing, analyzing, and visualizing high-throughput sequencing (HTS) data. The project is focused on comparing the performance of differential expression analysis pipelines (edgeR and limma-voom) using RNA-Seq data from a publicly available dataset.

Key Features

• Programmatic data acquisition from the NCBI GEO database.
• Preprocessing steps including quality control, trimming, and alignment.
• Comprehensive differential expression analysis with edgeR and limma-voom.
• Visualization of results including MA plots, MDS plots, and word clouds.
• Analysis of under-sampled and full-count datasets for comparative evaluation.

Dataset

The RNA-Seq data analyzed in this project is from:

Fu, N.Y. et al. (2015). EGF-mediated induction of Mcl-1 at the switch to lactation is essential for alveolar cell survival. Nature Cell Biology.
GEO Accession: GSE60450

Prerequisites

• Programming Languages: R (≥ 4.0), UNIX shell scripting
• R Packages:
  • edgeR
  • limma
  • Rsubread
  • org.Mm.eg.db
  • DT
  • wordcloud2
• Tools:
  • SRA Toolkit
  • FastQC
  • MultiQC
  • fastp

Installation

1. Clone the repository:

   git clone https://github.com/username/bulk-rna-seq.git
   cd bulk-rna-seq

2. Install required R packages:

   install.packages("BiocManager")
   BiocManager::install(c("edgeR", "limma", "Rsubread", "org.Mm.eg.db", "DT", "wordcloud2"))

3. Install additional tools (SRA Toolkit, fastp, FastQC, MultiQC) using your system's package manager.
4. Obtain reference genome and its associated annotation data (see below).

Outputs

• Differentially Expressed Genes (DEGs): Lists of DEGs with $p$-values and fold changes.
• Quality Control Reports: Aggregated results from FastQC and MultiQC.
• Plots:
  • MA plots and MDS plots
  • Word clouds of DEGs
  • Interactive Tables: Sortable and searchable tables of DEGs.

Project Goals

This project aims to:

• Develop a reproducible workflow for bulk RNA-Seq data analysis.
• Compare the performance of edgeR and limma-voom pipelines.
• Highlight the challenges and considerations in RNA-Seq data analysis.

Reference Genome and Annotation Data

In your project directory, create a new folder Mus_musculus.GRCm39

Navigate over to [Ensembl genome browser 113] (http://useast.ensembl.org/index.html) and Find the Page for the Mouse (GRCm39) Reference Assembly. Although there are many "images" or "version" of this assembly that include or exclude various products of the assembly process, a good starting point is often the "primary assembly" that's been "masked" for repetitive elements and low-complexity regions.

Click on the link "Download DNA sequence (FASTA)" to reach an FTP-page with genome data. According to the README file at the bottom of the file-list:

<sequence type>:
  * 'dna' - unmasked genomic DNA sequences.
  * 'dna_rm' - masked genomic DNA.  Interspersed repeats and low
     complexity regions are detected with the RepeatMasker tool and masked
     by replacing repeats with 'N's.
  * 'dna_sm' - soft-masked genomic DNA. All repeats and low complexity regions
    have been replaced with lowercased versions of their nucleic base

We will work with the masked primary assembly for this assignment, so please download the file Mus_musculus.GRCm39.dna_rm.primary_assembly.fa.gz.

Back out to the main page for "Mouse (GRCm39)", in the Gene Annotation panel on the top-right, get the associated annotation data paired to the reference genome assembly in Release 113 in GTF format by clicking "Download GTF" to obtain the file Mus_musculus.GRCm39.113.gtf.gz

Inside your project directory for this lab, move your Ensembl genome data files to the subdirectory Mus_musculus.GRCm39.

After downloading binaries or compiling, you will only need to install subread-buildindex, subread-align, and featureCounts, for example by copying these executable binaries to /usr/bin/local and possibly also re-starting your shell or terminal.

From the project directory (parent directory to where the genome data now is) do:

subread-buildindex -o Mus_musculus.GRCm39/Mus_musculus.GRCm39_subread Mus_musculus.GRCm39/Mus_musculus.GRCm39.dna_rm.primary_assembly.fa.gz