Dunni3/keypoint-diffusion
A diffusion model for structure-based drug design with faster inference from learned representations of protein structure.
This project helps medicinal chemists and drug designers accelerate the crucial step of generating new potential drug molecules that can bind effectively to a specific protein target. You provide the protein's structure (PDB or mmCIF file) and the desired binding site location, and it outputs a set of optimized 3D molecular structures designed to fit that site. It's a tool for researchers working on novel drug discovery.
No commits in the last 6 months.
Use this if you need to rapidly generate candidate small molecules that are predicted to bind to a specific pocket on a protein target, without manually designing each one.
Not ideal if you need to optimize existing molecules or predict binding affinity rather than generating new molecular structures.
Stars
31
Forks
3
Language
Python
License
MIT
Category
Last pushed
Dec 18, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Dunni3/keypoint-diffusion"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Compare
Higher-rated alternatives
gcorso/DiffDock
Implementation of DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking
arneschneuing/DiffSBDD
A Euclidean diffusion model for structure-based drug design.
gcorso/torsional-diffusion
Implementation of Torsional Diffusion for Molecular Conformer Generation (NeurIPS 2022)
igashov/DiffLinker
DiffLinker: Equivariant 3D-Conditional Diffusion Model for Molecular Linker Design
BioinfoMachineLearning/bio-diffusion
A geometry-complete diffusion generative model (GCDM) for 3D molecule generation and...