About

This repository implements an RNA-seq pipeline, which is able to:

1. map reads in .fastq.gz format with HISAT2
2. quantify gene expression
3. produce quality control plots with R

This pipeline should be run in our ngs container

Quick start

1. Make sure docker (https://www.docker.com) is installed on your system and is running.

2. Download source code of this repository containing the rnaseq pipeline. Alternatively, if git is installed on your system, clone it with the command:

git clone 'https://github.com/BioinfoSupport/rnaseq.git' my_new_project

3. Run the container: docker compose up -d

4. Connect to RStudio GUI running within the container at URL http://localhost:8787

5. Run app.R to download FASTQ files from iGE3 genomic platform and run quantification. Alternatively copy your .fastq.gz files into subfolder data/fastq/.

6. Run notebooks in src/ to generate QC reports

Directory structure

data/
  | ref/      folder containing reference genome subdirectories. 
  | | Dd+Mm/  a reference genome folder for _Dictyostelium discoideum_ and _Mycobacterium
  | |         marinum_. Example genome folders can be found in our [`genomes` repository]
  | |         (https://github.com/BioinfoSupport/genomes/releases). If you are using a
  | |         known reference genome, it will be downloaded automatically from this repository.
  | fastq/    folder containing sequenced reads (.fastq.gz)
  | | test/   example FASTQ reads
src/
  | 00_qc_genome.Rmd   Notebook to compute statistics on a reference genome
  | 01_qc_mapping.Rmd  Notebook to extract mapping statistics
  | 02_DESeq.Rmd       Notebook with an example usage of the pipeline with DESeq2  
.local/       hidden folder with pipeline-specific scripts

Useful commands

# Run the container 
docker compose up -d

# Run tests
docker compose exec rnaseq make TESTS

# Map and quantify all .fastq.gz files located in data/fastq/pilot
docker compose exec rnaseq make data/fastq/pilot/RNASEQ.ALL

# Run alignment and quantification on FASTQ in folder data/fastq on genome Dd+Mm (with automatic download of the genome)
docker compose exec rnaseq make GENOME=Dd+Mm data/fastq/test/RNASEQ.ALL

# Get a Bash in the container
docker compose exec rnaseq bash

# Stop the container
docker compose down

# Run container at the commandline without compose 
docker run --rm -v ./:/cwd --workdir /cwd unigebsp/ngs make -n

# Download fastq from ige3 plateform to data/fastq
wget -P ./data/fastq --content-disposition --trust-server-names -i 'https://data.ige3.genomics.unige.ch/dataset/download/xxxxxxx.txt'

Running the pipeline on a HPC cluster

singularity exec 'docker://unigebsp/ngs' make GENOME=Dd+Mm data/fastq/test/all

Conventions

• Use uppercase for .PHONY rules (e.g. %.fastq.gz.ALL, %.bam.ALL, %.BWAMEM, %.HT2)

• Try to add .ALL prefix to .PHONY rules that are fast to compute