Improving protein-ligand complex generation with force field guidance

Helen Lai, Tingyu Wang, Hassan Sirelkhatim, Joe Eaton, Howard Huang, Brad Rees, Ola Engkvist, Jon Paul Janet, Xiaoyun Wang, Alessandro Tibo · 2026

Read Improving protein-ligand complex generation with force field guidance by Helen Lai, Tingyu Wang, Hassan Sirelkhatim, Joe Eaton, Howard Huang, Brad Rees, Ola Engkvist, Jon Paul Janet, Xiaoyun Wang, Alessandro Tibo (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-13137607.txt.

Description

Lai and co-authors improve AI generation of drug like molecules bound to proteins by steering the process with a physics based energy score. Generative models often propose poses that break physical rules. Their 2026 method feeds a standard force field into the sampling step, nudging molecules toward stable, plausible shapes without retraining the model.

The team tested the guidance on two leading model types using a public dataset of protein ligand pairs. Guided runs improved interaction energy and cut strain energy by up to three quarters. Over a thousand generated ligands scored better in docking than the natural ligands they were modeled on. The plug in design works across model types, the authors note, while keeping chemical variety.

Catalog record

CREATORHelen Lai, Tingyu Wang, Hassan Sirelkhatim, Joe Eaton, Howard Huang, Brad Rees, Ola Engkvist, Jon Paul Janet, Xiaoyun Wang, Alessandro Tibo
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13137607

Browse the full library · About the Tower of Babel