Skip to content

haessar/HackathonAltSplicing

Repository files navigation

HackathonAltSplicing

image

Quick Start

Clone the repository and change directory:

git clone https://github.com/haessar/HackathonAltSplicing.git
cd HackathonAltSplicing

For running Python scripts or notebooks in scripts directory, first setup a Python virtual environment (tested with v3.10) and install package dependencies:

python3 -m venv venv-name
source venv-name/bin/activate
pip install .
python3 scripts/script_name.py [args]

Isoform detection pipeline

For running the snakemake pipeline in the isoform_detection directory (corresponding to Workflow 2 in figure above), install conda (e.g. Miniforge download link) and follow the guide in docs/isoform_detection.md. Slurm bash scripts for running on a HPC system can be found in the scripts directory.

JBrowse 2 with plugin and custom tracks

For running JBrowse 2 with the plugin developed for Workflow 1, ensure Node.js is installed (e.g. sudo apt install nodejs npm) and follow the guide in jbrowse-plugin-bedfeaturecoloring/README.md. To serve a custom config.json for loading in tracks for analysis, in a new terminal:

cd analysis/jbrowse_env/
npx serve . --cors -p 3001

and navigate to http://localhost:3000/?config=http://localhost:3001/config.json in a web browser.

Contributing

🔖 Issue Labelling

I've created issues that reflect the work packages that I outlined in the introductory slides (for some of them it made sense to split them in two). Work package #1 is signified by (WP1), etc. I've used labels to designate which "theme" they belong to, whether they have a significant MARS component, and whether they are Coding or Research heavy. When determining what you want to work on, you can filter the labels that appeal to you or are a suitable match to your skillset (e.g. if you have no interest in using MARS, you could filter out the MARS label).

🔀 Branching Strategy

As there will be lots of us working feverishly in the same repo, I recommend we use feature branches for all development work rather than main.

  • Create a new branch from main for each new feature or fix (perhaps to work on a GitHub issue):

    git checkout -b your-feature-name
  • Once you're happy with changes, open a pull request to merge into main and someone can review them.

  • This will help prevent conflicts, support collaboration, and maintain a clean commit history.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 9