Nutmeg and SPICE: Models and Data for Biomolecular Machine Learning

doi:10.1021/acs.jctc.4c00794

Nutmeg and SPICE: Models and Data for Biomolecular Machine Learning

2024-09-25

2

PRE

AI

P

Peter Eastman *

B

Benjamin P. Pritchard

DOI：10.1021/acs.jctc.4c00794

Original

Original request for help

Save

Abstract

En 中文

We describe version 2 of the SPICE data set, a collection of quantum chemistry calculations for training machine learning potentials. It expands on the original data set by adding much more sampling of chemical space and more data on noncovalent interactions. We train a set of potential energy functions called Nutmeg on it. They are based on the TensorNet architecture. They use a novel mechanism to improve performance on charged and polar molecules, injecting precomputed partial charges into the model to provide a reference for the large-scale charge distribution. Evaluation of the new models shows that they do an excellent job of reproducing energy differences between conformations even on highly charged molecules or ones that are significantly larger than the molecules in the training set. They also produce stable molecular dynamics trajectories and are fast enough to be useful for routine simulation of small molecules.

Keywords:

PARAMETERS

ACCURACY

WATER

Nutmeg and SPICE: Models and Data for Biomolecular Machine Learning

Abstract

Journal

Organization