New paper in Chemical Science: enzyme catalysis with EMLE/MM

Our paper Simulating enzyme catalysis with electrostatically embedded machine learning potentials is now published in Chemical Science.

We show that reactive machine-learned potentials trained on gas-phase data can accurately capture enzyme catalytic effects when coupled to the protein environment via electrostatic machine learning embedding (EMLE). The approach is validated on two systems — the natural Diels-Alderase AbyU and the chorismate-to-prephenate reaction — and agrees with high-level QM/MM reference calculations while being far more computationally efficient.

Work by Valentin Gradisteanu, Elliot W. Chan, Lester Hedges, Meritxell Malagarriga Perez, Rolf David, Miguel de la Puente, Damien Laage, Iñaki Tuñón, Marc W. van der Kamp, and Kirill Zinovjev.