isayevlab/geom-drugs-3dgen-evaluation

A refined evaluation pipeline for 3D molecular generative models trained on GEOM-Drugs.

33
/ 100
Emerging

This project helps computational chemists and cheminformaticians rigorously evaluate the chemical accuracy of 3D molecular generative models. It takes outputs from these models (molecular structures in SDF files) and assesses them for chemical stability, geometric accuracy (bond lengths, angles, torsions), and energy improvements after quantum chemical optimization. The result is a comprehensive set of metrics to benchmark and compare different generative models.

No commits in the last 6 months.

Use this if you are developing or using 3D molecular generative models and need a robust, chemically aware pipeline to benchmark their outputs for stability and geometric realism.

Not ideal if you are looking for a tool to generate molecules, or if you only need simple 2D molecular property calculations.

computational-chemistry drug-discovery cheminformatics molecular-modeling materials-science
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 15 / 25
Community 11 / 25

How are scores calculated?

Stars

12

Forks

2

Language

Python

License

MIT

Last pushed

Sep 10, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/generative-ai/isayevlab/geom-drugs-3dgen-evaluation"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.