Skip to content

Minimal Tutorial

Jump to bottom
Ken Nakatsu edited this page Aug 3, 2023 · 4 revisions

Overview

Running sRNAfrag on an HPC allows for users to run it on a large number of samples. In this example, I will ran sRNAfrag on >200 small RNA-seq samples.

  1. Load modules, and clone the github repo and install.
module load git
module load Anaconda
module load R_4.2.1
git clone https://github.com/kenminsoo/sRNAfrag

cd sRNAfrag/install
sh easy_install.sh
  1. Build a bowtie-index for my genome of interest. Obtain a gtf file to label for out-of-space maps.

The ITAS publication has a nice summary of transcripts that are not snoRNA, which is what I wanted for this example.

https://github.com/EpiEpiMSU/ITAS/tree/main/Integrated_annotation/human

cd my/dir/somewhere/ref_genome
bowtie-build hg38.fa hg38_index // add path to the config file

// download annotation from ITAS, add to config

  1. Add directories to config file. Change any adapters if needed.

  2. Run the pipeline

python sRNA_fragment.py
  • TIP: Keep the working directory the first time you run sRNAfrag if you are processing reads. Then
cd working
mkdir ../processed
mv *.fastq.gz ../processed

This will prevent the need to rerun all modules again, and will speed up your analysis time.

Clone this wiki locally