/fch-drug-discovery

MIT LicenseMIT

Forkwell Coronavirus Hack: Drug Discovery

Fork this repository to start participating!

About

Welcome to Forkwell Coronavirus Hack!

This repository contains the hackathon kit for you to get started on solving topic 1: Drug Discovery and eventually a place where you host all your submission artifacts as your own fork.

Sponsors

Microsoft MDEC Runcloud
AWS DigitlOcean Sunway iLabs
CoronaTracker LEAD AI Geeks

Background

A few studies have shown that HIV antivirals offer promising results, but this is still an open field of discovery.

Finding a new drug or validating an existing drug are both suitable approaches.

Goal

Find a candidate drug (ligand) with a high binding affinity with the COVID-19 main protease.

  1. Use machine learning to identify a potential candidate, then use docking software to get the binding affinity between it and the main protease.
  2. Write a report that describes your process and results in detail in the form of a jupyter notebook.

Submission

You are required to submit a 15-minute presentation video of your report. However, to ensure the legitimacy of your submission, you are required to submit your project artifacts (code, datasets, documents) as a fork to this repository.

A submission form will be available on the 14th of April 2020 for submission, please join our facebook group and discord channel to keep updated with latest changes.

After creating your own fork of this repository, clone the repository:

git clone git@github.com:<your-github-username>/fch-drug-discovery

Change to the directory:

cd fch-drug-discovery

Set upstream:

git remote add upstream git@github.com:forkwell-io/fch-drug-discovery

...and start Hacking!!

Once you are ready to submit, create a pull request from your fork to us and include the link to your fork in your submission form that will be available on the 14th of April 2020.

Please remember that your code will be publicly available, open-sourced licesed and free for the internet to use. Please ensure that you don't commit any sensitive information!

Resources

Presentations

Thomas MacDougall

Virus Primer - Target a Virus

Thomas MacDougall - Graduate Student in Computer Science at the University of Montreal

slides

Databases

Name Link
CMap https://clue.io/cmap
COVID-19 main protease https://www.wwpdb.org/pdb?id=pdb_00006lu7
ChemBL https://www.ebi.ac.uk/chembl/
DrugBank https://www.drugbank.ca/
Folding@Home https://github.com/FoldingAtHome/coronavirus
Formatted ZINC Database https://github.com/molecularsets/moses
GEO Signatures of Differentially Expressed Genes for Viral Infections https://amp.pharm.mssm.edu/Harmonizome/dataset/GEO+Signatures+of+Differentially+Expressed+Genes+for+Viral+Infections
NextStrain https://nextstrain.org/
ZINC Database https://zinc.docking.org/
Dataset Search https://blog-google.cdn.ampproject.org/c/s/blog.google/products/search/discovering-millions-datasets-web/amp/

APIs

Name Link
CoronaTracker API https://api.coronatracker.com
COVID-19 Postman API https://covid-19-apis.postman.com/
COVID-19 Rapid API https://rapidapi.com/api-sports/api/covid-193?endpoint=apiendpoint_dfb9e52d-bd90-48ec-a571-8b78610a736d

Docking Tools

Name Link
AutoDock VINA http://vina.scripps.edu/
PyRX https://pyrx.sourceforge.io/

Deep Reinforcement Learning Networks

Name Link
DrugEx https://github.com/XuhanLiu/DrugEx
GENTRL https://github.com/insilicomedicine/GENTRL
ReLeaSE https://github.com/isayev/ReLeaSE

Awesome perks from our sponsors!

Deployment

E-learning

Help us organize better

Feel free to open issues if you find anything lacking and we appreciate your feedback greatly!

Community

Join the community!