Machine-learned molecular mechanics force field for the simulation of protein-ligand systems and beyond

Abstract

The development of reliable and extensible molecular mechanics (MM) force fields -- fast, empirical models characterizing the potential energy surface of molecular systems -- is indispensable for biomolecular simulation and computer-aided drug design. Here, we introduce a generalized and extensible machine-learned MM force field, \texttt{espaloma-0.3}, and an end-to-end differentiable framework using graph neural networks to overcome the limitations of traditional rule-based methods. Trained in a single GPU-day to fit a large and diverse quantum chemical dataset of over 1.1M energy and force calculations, \texttt{espaloma-0.3} reproduces quantum chemical energetic properties of chemical domains highly relevant to drug discovery, including small molecules, peptides, and nucleic acids. Moreover, this force field maintains the quantum chemical energy-minimized geometries of small molecules and preserves the condensed phase properties of peptides, self-consistently parametrizing proteins and ligands to produce stable simulations leading to highly accurate predictions of binding free energies. This methodology demonstrates significant promise as a path forward for systematically building more accurate force fields that are easily extensible to new chemical domains of interest.

References in corpus (20)

• Canonical sampling through velocity-rescaling
• Improvements to the APBS biomolecular solvation software suite
• Statistically optimal analysis of samples from multiple equilibrium states
• Building Water Models, A Different Approach
• Simplifying Graph Convolutional Networks
• Pressure control using stochastic cell rescaling
• Less is more: sampling chemical space with active learning
• Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
• Best Practices for Alchemical Free Energy Calculations
• A Survey of Uncertainty in Deep Neural Networks
• Topology Adaptive Graph Convolutional Networks
• Replica exchange and expanded ensemble simulations as Gibbs sampling: Simple improvements for enhanced mixing
• Forces are not Enough: Benchmark and Critical Evaluation for Machine Learning Force Fields with Molecular Simulations
• End-to-End Differentiable Molecular Mechanics Force Field Construction
• Differentiable sampling of molecular geometries with uncertainty-based adversarial attacks
• Machine Learning Directed Optimization of Classical Molecular Modeling Force Fields
• Denoise Pretraining on Nonequilibrium Molecules for Accurate and Transferable Neural Potentials
• On the Expressive Power of Geometric Graph Neural Networks
• Graph Nets for Partial Charge Prediction
• Stochastic Aggregation in Graph Neural Networks

Cited by in corpus (5)

• On the design space between molecular mechanics and machine learning force fields
• Grappa -- A Machine Learned Molecular Mechanics Force Field
• Scaling Graph Neural Networks to Large Proteins
• Molecular Dynamics and Machine Learning Unlock Possibilities in Beauty Design -- A Perspective
• Knowledge Distillation of a Protein Language Model Yields a Foundational Implicit Solvent Model