mims-harvard / TDC

Therapeutics Commons (TDC-2): Multimodal Foundation for Therapeutic Science
https://tdcommons.ai
MIT License
977 stars 171 forks source link

New PROTACs dataset #119

Open kexinhuang12345 opened 2 years ago

kexinhuang12345 commented 2 years ago

Describe the problem PROTACs are a class of therapeutics that flag target proteins to a cell's waste disposal system, instead of the traditional binding to targets. It has been very promising due to its ability to tackle undruggable targets. However, problems still exist, such as designing the PROTACs, measure the degradability, understand the mechanism. A recent paper https://www.biorxiv.org/content/10.1101/2021.09.27.462040v1 has a nice dataset that predicts degradability given a proten PDB. The dataset is linked here: http://mapd.cistrome.org/

Describe the solution you'd like

from tdc.single_pred import PROTACs_Degrade
data = PROTACs_Degrade(name = 'MAPD',  path = './data')

Additional context N/A

kexinhuang12345 commented 2 years ago

this seems relevant: https://protacpedia.weizmann.ac.il/ptcb/download

amva13 commented 4 months ago

@marinkaz any thoughts on this task as well? my understanding is rational design of protein degraders is an exciting frontier task. https://www.sciencedirect.com/science/article/pii/S2211383524001357 https://www.nature.com/articles/s42256-022-00527-y

amva13 commented 4 months ago

@kexinhuang12345 will help contribute it.

amva13 commented 3 months ago

This for the linker design task @amva13 @marinkaz