Efficient Discovery of Potent AKR1B10 Inhibitors Using Co-folding Model Boltz‑2
Shinya Kawano, Akira Ikari, Satoshi Endo, Kentaro Tomii · 2026
Read Efficient Discovery of Potent AKR1B10 Inhibitors Using Co-folding Model Boltz‑2 by Shinya Kawano, Akira Ikari, Satoshi Endo, Kentaro Tomii (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13625949.txt.
Description
Shinya Kawano and three co-authors let an artificial intelligence model that folds proteins around drug candidates pre-screen a chemical library for them, in this 2026 study. Their target is an enzyme called AKR1B10, overproduced in several cancers and therefore an attractive drug target. Predicting which molecules will bind and block it remains one of drug discovery's bottlenecks.
Starting from 867 candidates in a diverse library, the co-folding model Boltz-2 scored how each molecule would sit against the enzyme. The team then picked forty compounds, favoring chemical variety, for real laboratory testing. Twenty-eight of the forty, seventy percent, genuinely inhibited the enzyme at useful strength, and several were potent at below one micromole with brand-new chemical skeletons. A hit rate like that turns AI screening from a curiosity into a working instrument.
Catalog record
| CREATOR | Shinya Kawano, Akira Ikari, Satoshi Endo, Kentaro Tomii |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |