simulate-rnaseq

Simple workflow to simulate RNA-seq reads and perform DE-analysis.

Workflow simulates RNA-seq reads using Polyester R package https://github.com/alyssafrazee/polyester, quantifies expression with Salmon https://combine-lab.github.io/salmon/ and performs DE-analysis using DESeq2 R package https://bioconductor.org/packages/release/bioc/html/DESeq2.html.

Human protein coding transcript sequences are downloaded from https://www.gencodegenes.org/human/. Transcript sequences can be customised by directly editing Snakefile.

Install

To install clone this repository and install dependencies (assumes that conda/miniconda is already installed):

conda env create -f environment.yml

This command creates conda environment called "simulaternaseq".

Running

Activate conda environment:

conda activate simulaternaseq

For test run replace "-j" flag with "-n":

snakemake -j --use-conda

To generate snakemake run report:

snakemake --report output/report.html

Some outputs of interest

File output/diffexp/group1-vs-group2.diffexp.tsv contains differential expression results, which can be compared to ground truth in output/simulated_reads/sim_tx_info.txt.

Acknowledgements

The workflow borrowed bits and pieces from rna-seq-star-deseq2 https://github.com/snakemake-workflows/rna-seq-star-deseq2.