Autonomous artificial intelligence discovers mechanisms of molecular self-organization in virtual experiments

Abstract

Molecular self-organization driven by concerted many-body interactions produces the ordered structures that define both inanimate and living matter. Understanding the physical mechanisms that govern the formation of molecular complexes and crystals is key to controlling the assembly of nanomachines and new materials. We present an artificial intelligence (AI) agent that uses deep reinforcement learning and transition path theory to discover the mechanism of molecular self-organization phenomena from computer simulations. The agent adaptively learns how to sample complex molecular events and, on the fly, constructs quantitative mechanistic models. By using the mechanistic understanding for AI-driven sampling, the agent closes the learning cycle and overcomes time-scale gaps of many orders of magnitude. Symbolic regression condenses the mechanism into a human-interpretable form. Applied to ion association in solution, gas-hydrate crystal formation, and membrane-protein assembly, the AI agent identifies the many-body solvent motions governing the assembly process, discovers the variables of classical nucleation theory, and reveals competing assembly pathways. The mechanistic descriptions produced by the agent are predictive and transferable to close thermodynamic states and similar systems. Autonomous AI sampling has the power to discover assembly and reaction mechanisms from materials science to biology.

References in corpus (7)

• Array Programming with NumPy
• Canonical sampling through velocity-rescaling
• Deep Learning in Neural Networks: An Overview
• Accurate determination of crystal structures based on averaged local bond order parameters
• Machine learning for molecular simulation
• Machine Learning Unifies the Modelling of Materials and Molecules
• From data to noise to data: mixing physics across temperatures with generative artificial intelligence

Cited by in corpus (27)

• A unified framework for machine learning collective variables for enhanced sampling simulations: $\text{mlcolvar}$
• Computing the Committor with the Committor: an Anatomy of the Transition State Ensemble
• Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications
• Descriptors-free Collective Variables From Geometric Graph Neural Networks
• Everything everywhere all at once: a probability-based enhanced sampling approach to rare events
• Spectral Map: Embedding Slow Kinetics in Collective Variables
• Manifold Learning in Atomistic Simulations: A Conceptual Review
• Optimal reaction coordinates and kinetic rates from the projected dynamics of transition paths
• Inexact iterative numerical linear algebra for neural network-based spectral estimation and rare-event prediction
• Variational deep learning of equilibrium transition path ensembles
• Learning protein-ligand unbinding pathways via single-parameter community detection
• Flow Matching for Optimal Reaction Coordinates of Biomolecular System
• Sampling a rare protein transition with a hybrid classical-quantum computing algorithm
• Variational time reversal for free energy estimation in nonequilibrium steady states
• Unveiling interatomic distances influencing the reaction coordinates in alanine dipeptide isomerization: An explainable deep learning approach
• Investigating the hyperparameter space of deep neural network models for reaction coordinates
• A tutorial for mesoscale computer simulations of lipid membranes: tether pulling, tubulation and fluctuation
• LeaPP: Learning Pathways to Polymorphs through machine learning analysis of atomic trajectories
• Understanding Reaction Mechanisms from Start to Finish
• Estimating Reaction Barriers with Deep Reinforcement Learning
• From high-dimensional committors to reactive insights
• Deep learning of committor for ion dissociation and interpretable analysis of solvent effects using atom-centered symmetry functions
• Combining multiple interface set path ensembles with MBAR reweighting
• More sophisticated is not always better: comparison of similarity measures for unsupervised learning of pathways in biomolecular simulations
• Deep Neural Networks as Discrete Dynamical Systems: Implications for Physics-Informed Learning
• Deep learning of committor and explainable artificial intelligence analysis for identifying reaction coordinates
• Improving Estimation of the Koopman Operator with Kolmogorov-Smirnov Indicator Functions