Reactive machine learning potential for accelerating transition state search in organic synthesis

Kaipai Ren, Kun Tang, Yujing Zhao, Lei Zhang, Jian Du, Qingwei Meng, Qilei Liu · 2026

Read Reactive machine learning potential for accelerating transition state search in organic synthesis by Kaipai Ren, Kun Tang, Yujing Zhao, Lei Zhang, Jian Du, Qingwei Meng, Qilei Liu (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-13376211.txt.

Description

Kaipai Ren and six co-authors build a reactive machine learning potential, DeePEST-OS, that speeds the search for transition states in organic synthesis nearly ten thousand times over quantum chemistry. Transition states set how fast a reaction runs, and finding them is the slow step in reaction kinetics. Their 2026 model trains on about 75,000 reactions generated by a low-cost data strategy. Physical priors from semi-empirical quantum chemistry mix with an equivariant message passing network inside it.

On unseen reactions, predicted transition state geometries landed within a root-mean-square deviation of 0.12 angstroms. Energy barriers missed by a mean of 0.60 kilocalories per mole. Demonstrated uses span transition state conformer screening, barrier prediction for retrosynthesis of complex pharmaceuticals, and an experimentally validated selectivity prediction in a Diels-Alder reaction. The team pitches the model as a practical tool for accelerating reaction studies.

Catalog record

CREATORKaipai Ren, Kun Tang, Yujing Zhao, Lei Zhang, Jian Du, Qingwei Meng, Qilei Liu
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE73 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13376211

Browse the full library · About the Tower of Babel