RNA-seq Differential Gene Expression Workflow (Snakemake)

This repository contains a reproducible RNA-seq analysis workflow implemented using Snakemake. The pipeline performs quality control, read trimming, genome alignment, gene quantification, and differential expression analysis between two experimental conditions.

This workflow was redesigned from the Biomedical Data Analysis Master’s module of Life Science Informatics. By integrating isolated Conda environments, the analysis follows current bioinformatics best practices, ensuring the pipeline remains scalable and reproducible across different computing platforms.

---

Workflow Overview

The pipeline performs the following steps:

1. Quality control

   • Tool: FastQC
   • Input: raw FASTQ files
   • Output: HTML quality reports

2. Adapter trimming and quality filtering

   • Tool: Cutadapt
   • Removes low-quality bases and short reads

3. Genome indexing

   • Tool: STAR
   • Generates genome index from reference FASTA and GTF files

4. Read alignment

   • Tool: STAR
   • Maps reads to the reference genome

5. Gene-level quantification

   • Tool: featureCounts (Subread)
   • Produces gene count matrix

6. Differential expression analysis

   • Tool: DESeq2 (R)
   • Identifies significantly up-regulated and down-regulated genes

---

Workflow Diagram

graph LR
    A[FASTQ files] --> B[FastQC]
    B --> C[Cutadapt]
    C --> D[STAR alignment]
    D --> E[featureCounts]
    E --> F[DESeq2]
    F --> G[Results]

Loading

---

Repository Structure

.
├── Snakefile              # Pipeline logic
├── deseq_rScript.r        # Differential expression analysis
├── README.md
├── .gitignore
├── envs/                  # Conda environment definitions
│   ├── fastqc.yaml
│   ├── cutadapt.yaml
|   ├── star.yaml
|   ├── subread.yaml
│   └── deseq2.yaml
├── featureCounts_output.txt
├── deseq2_results.txt
├── deseq2_up.txt
└── deseq2_down.txt

---

Requirements

Install one of the following:

• Conda (Miniconda or Anaconda)
• Snakemake

Recommended installation:

conda install -c bioconda -c conda-forge snakemake

---

Running the Workflow

Place your RNA-seq FASTQ files in the project directory using the naming format:

sample_0.fastq.gz
sample_1.fastq.gz
...
sample_5.fastq.gz

Also provide reference files:
genome.fa
annotation.gtf

Run the workflow:

snakemake --use-conda --cores 4

Snakemake will automatically:

• create environments
• run each analysis step
• generate output files

---

Output Files

Main outputs include:

File	Description
featureCounts_output.txt	Gene count matrix
deseq2_results.txt	Full differential expression results
deseq2_up.txt	Up-regulated genes (log2FC ≥ 2)
deseq2_down.txt	Down-regulated genes (log2FC ≤ −2)
fastqc_reports/	Quality control reports

---

Reproducibility

Each step runs in an isolated Conda environment defined in: \

 envs/*.yaml 

This ensures consistent and reproducible results across systems.

---

Tools Used

• Snakemake
• FastQC
• Cutadapt
• STAR aligner
• Subread (featureCounts)
• DESeq2 (R)

---

Author

Bioinformatics workflow developed as part of RNA-seq data analysis training and portfolio project.