bartongroup / MG_sulfRNAseq

Re-analysis of public data to compare to COVID RNA-seq results (Chalmers)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Re-analysis of L-SFN-treated PBMC mRNAseq data

Software to accompany the manuscript of Long et al., "SFX-01 in hospitalised patients with community-acquired pneumonia during the COVID19 pandemic: a double-blind, randomised, placebo-controlled trial", ERJ Open Research (2024).

Usage

On a Linux cluster

Create and activate a conda environment

cd rna_seq
conda env create -f env.yaml
conda activate sulfrnaseq

Download FASTQ files

./get_fastq.sh

Run snakemake

./run_snake.sh

This will trim adapters, perform quality control, download genome files, map reads to the reference and count reads per gene.

In RStudio

Once snakemake is finished, we suggest using RStudio. If this is done on a different machinge (I run RStudio on a laptop), some data need to be copied over (see ./scripts/get_data.sh and ./scripts/rsync_include.txt). Once in RStudio, start in the top project directory. The first step is to create environment using renv:

install.packages("renv")
renv::restore()

This will install all necessary packages. Run the targets pipeline.

targets::tar_make()

This will carry out all the calculations, create figures (some as targets, some in ./fig directory) and output TSV files in directory ./tab.

About

Re-analysis of public data to compare to COVID RNA-seq results (Chalmers)


Languages

Language:R 91.0%Language:Python 8.1%Language:Shell 0.8%