broad

install and use CRISPR-broad. Optimized for the HT-HPC.

Installation

1. set your working directory as your scratch directory

cd /scratch/$USER

$USER adds your user name..

2. clone broad

git clone https://github.com/BiCro-HT/broad.git

3. create a project inside broad

cd broad
bash install.sh <project_name>

Important

Make sure to change <project_name> with the desired name here and for the rest of instructions.

Downloading a reference genome

by default an example fasta file inside CRISPR-broad is used as reference genome. If you want to use a reference genome, you can download hg38 or T2T (human genomes) or mm10 (mus musculus) using;

bash download_ref_genome.sh hg38 # human genome
bash download_ref_genome.sh T2T # human genome telomere to telomere
bash download_ref_genome.sh mm10 # mus musculus genome

For more info on reference genomes

Note

You only need to download a ref genome once.

Usage

make sure to go to installation directory, use the following;

cd /scratch/$USER/broad/

you may use Job or Interactive run. For real work definetely use the JOB

JOB

Send the work as a sbatch job.

sbatch run_sample.sh <project_name> <reference_genome>

make sure you downloaded the reference genome after the download..

sbatch run_sample.sh <project_name> <reference_genome> # hg38 or T2T or mm10

Interactive

Just as an option but highly discouraged

srun --nodes=1 --cpus-per-task=4 --mem-per-cpu=4GB --partition=cpu-interactive --pty /bin/bash
bash run_interactive.sh <project_name> <reference_genome>

TEST RUN from zero with small example genome

For a test run check TESTRUN.md