Transferable Machine Learning Interatomic Potential for Bond Dissociation Energy Prediction of Drug-like Molecules

Abstract

We present a transferable MACE interatomic potential that is applicable to open-and closed-shell drug-like molecules containing hydrogen, carbon, and oxygen atoms. Including an accurate description of radical species extends the scope of possible applications to bond dissociation energy (BDE) prediction, for example, in the context of cytochrome P450 (CYP) metabolism. The transferability of the MACE potential was validated on the COMP6 data set, containing only closed-shell molecules, where it reaches better accuracy than the readily available general ANI-2x potential. MACE achieves similar accuracy on two CYP metabolismspecific data sets, which include open-and closed-shell structures. This model enables us to calculate the aliphatic C−H BDE, which allows us to compare reaction energies of hydrogen abstraction, which is the rate-limiting step of the aliphatic hydroxylation reaction catalyzed by CYPs. On the "CYP 3A4" data set, MACE achieves a BDE RMSE of 1.37 kcal/mol and better prediction of BDE ranks than alternatives: the semiempirical AM1 and GFN2-xTB methods and the ALFABET model that directly predicts bond dissociation enthalpies. Finally, we highlight the smoothness of the MACE potential over paths of sp 3 C−H bond elongation and show that a minimal extension is enough for the MACE model to start finding reasonable minimum energy paths of methoxy radical-mediated hydrogen abstraction. Altogether, this work lays the ground for further extensions of scope in terms of chemical elements, (CYPmediated) reaction classes and modeling the full reaction paths, not only BDEs.

Methodological and force field effects in the molecular dynamics-based prediction of binding free energies of host–guest systems

“…The parameters in eqn (8) are based on the definition of atomic classes depending on the chemical environment and atom hybridization and are derived from ab initio calculations and experimental data, 51,58 most recently in combination with machine-learning techniques. 59,60…”

Predicting bond dissociation energies of cyclic hypervalent halogen reagents using DFT calculations and graph attention network model

“…In recent years, machine learning has emerged as a promising and cost-effective alternative to traditional DFT calculations for predicting key properties of organic molecules such as BDE, nucleophilicity, and electrophilicity [ 48 – 60 ]. Recently, applications of the Elastic Net model with Avalon fingerprints [ 55 ] and the deployment of artificial neural network (ANN) models [ 57 ] with the Mordred cheminformatics package have demonstrated considerable success in predicting the BDEs of hypervalent iodine(III) reagents.…”