IBM/regression-transformer

Regression Transformer (2023; Nature Machine Intelligence)

45
/ 100
Emerging

This project helps scientists in chemistry and materials science explore new molecules, proteins, and chemical reactions with desired properties. You input an existing chemical structure (like a small molecule or protein) and specify target property values (e.g., higher solubility, better stability). The system then suggests new chemical structures that are similar to your input but are predicted to have the desired properties. This is for researchers who need to design novel materials or compounds by fine-tuning specific characteristics.

159 stars. No commits in the last 6 months.

Use this if you need to generate new chemical compounds, proteins, or polymers that are optimized for specific properties like solubility, synthesizability, or stability, starting from a known structure.

Not ideal if you need to predict properties for existing structures without generating new ones, or if your domain is outside of chemistry and materials science.

drug-discovery materials-science molecular-design protein-engineering organic-chemistry
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

159

Forks

23

Language

Python

License

MIT

Last pushed

Sep 17, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/IBM/regression-transformer"

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