snakemake-bulkrna-seq

A snakemake workflow for bulk RNA-seq data preprocessing including sequence alignment, quality check etc.

Zhengqiao Zhao, Yuri Pritykin

October, 2022

Installation

1. The core environment can be installed using conda. The pipeline relies on hisat2, samtools and fastqc.

conda create -n snakemake_bulkRNA -c bioconda -c conda-forge snakemake hisat2 samtools fastqc gnuplot

2. We also provide an R environment for gene count matrix build and differential gene expression analysis using Rsubread and DESeq2.

conda create -n r_deseq -c bioconda -c conda-forge -c r r-tidyverse r-data.table bioconductor-deseq2 bioconductor-rtracklayer bioconductor-summarizedexperiment bioconductor-rsubread r-pheatmap

Running the pipeline

Data Requirements

The raw fastq files should be placed in one directory, e.g., raw_data. The name of the fastq files should follow this pattern: <sample_name>_R<1,2>.fastq. For example, raw_data should have the following directory structure:

raw_data
|   SAMPLE1_R1.fastq
|   SAMPLE1_R2.fastq
|   SAMPLE2_R1.fastq
|   SAMPLE2_R2.fastq
|   ...

The config file should have the exact sample names (excluding _R<1,2>.fastq suffix) and Snakemake file should have the absolute path to the raw data directory and the results directory that used to save the pipeline outputs including the alignment results.

Run the pipeline

1. modify the cluster.json file so that it is compatible with your computation cluster settings.
2. run the following command in terminal:

bash run_snakemake.sh

DAG of this snakemake workflow

The following image shows the directed acyclic graph (DAG) of jobs where the edges represent dependencies.

It can be obtained by running the following command in the snakemake conda environment:

snakemake --forceall --rulegraph | dot -Tpng > RULE_DAG.png