Enhanced Sampling of Protein Conformations in AlphaFold3 with Repulsive Bias in the Diffusion Generative Model
Jun Ohnuki, Kei-ichi Okazaki · 2026
Read Enhanced Sampling of Protein Conformations in AlphaFold3 with Repulsive Bias in the Diffusion Generative Model by Jun Ohnuki, Kei-ichi Okazaki (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-13625536.txt.
Description
Jun Ohnuki and Kei-ichi Okazaki present, in twenty twenty-six, a tweak to AlphaFold3, a leading AI tool that predicts protein shapes. Proteins change shape to do their work. The tool can miss shape changes that are triggered when a drug or partner binds. The authors treat its shape-making process like a random sampling walk. They add a push that keeps predicted shapes apart from each other. That push forces the tool to search a wider range of shapes.
Their method, called AF3-ReD, found many shapes at once for motor, kinase, and transporter proteins. The shapes matched real lab structures well, down to the pose of bound partners. It even found transporter shapes that postdate the tool's training data.
Catalog record
| CREATOR | Jun Ohnuki, Kei-ichi Okazaki |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |