This repository is developed for analyzing NGS datasets, specifically the ones related with genome-wide DNA Damage and Repair.

Custom setup assumes that you have the necessary programs installed and they are exectuble in your $PATH.

The list of the necessary programs are:

• Bowtie2
• cutadapt
• sra toolkit
• samtools
• bedtools
• ucsc tools

To place the source code of this repository in your path, please follow these commands:

cd ~
git clone git@bitbucket.org:adebali/NGStoolkit.git
cd NGStoolkit
bash setup.sh

Make sure that you have Docker and docker-compose are installed. The recommended mamory for the pipeline is at least 6GB. Allow docker to use that much memory from the Docker settings.

docker-compose -p ngs up --build -d

docker-compose maps the data directory to /data in the container. All the programs should be installed for the docker container.

Go into the container with docker-compose run main bash

Now it is time to build a reference genome index for the alignment program. Here we use Bowtie2. Download and prepare the reference genome with /NGStoolkit/Docker/prepareReferenceGenome.sh. This will take some time, feel free to have a cup of coffee.

After the reference genome index is built successfully, we can run our pipeline cd data && /NGStoolkit/stable/XR-seq-basics.sh. Here we go!

• Download and prepare the reference genome with docker-compose run main /NGStoolkit/Docker/prepareReferenceGenome.sh
• Run the pipeline with docker-compose run main /NGStoolkit/stable/XR-seq-basics.sh

• Move your .fastq file into the data directory.
• Edit the XR-seq-basic.sh and repalce the SAMPLE variable with the base sample name in your file. For example if you file is named as myFile.fastq the base name will be myFile.
• If you want to retrieve the existing data set from SRA please see the fastq-dump command and replace the SRA acccession number with the one of interest. If you use youw own file please comment out that two lines in XR-seq-basics.sh.

• Ogun Adebali

This project is licensed under the MIT License