AI-Guided Molecular Simulations in VR: Exploring Strategies for Imitation Learning in Hyperdimensional Molecular Systems

TL;DR

This work addresses the challenge of sampling high-dimensional molecular conformations by leveraging immersive iMD-VR to gather expert demonstrations for imitation learning. It surveys IL methods from robotics and multi-agent systems and demonstrates a proof-of-concept where a CNN trained on iMD-VR data learns to thread a methane molecule through a carbon nanotube. The results indicate that simple BC can capture key manipulation dynamics, and the paper discusses avenues for combining IL with IRL and adversarial IL to improve robustness and generalization. By outlining practical applications in drug design and materials science, along with citizen-science data collection, the work highlights how AI agents could augment human expertise in navigating vast molecular conformational spaces.

Abstract

Molecular dynamics (MD) simulations are a crucial computational tool for researchers to understand and engineer molecular structure and function in areas such as drug discovery, protein engineering, and material design. Despite their utility, MD simulations are expensive, owing to the high dimensionality of molecular systems. Interactive molecular dynamics in virtual reality (iMD-VR) has recently emerged as a "human-in-the-loop" strategy for efficiently navigating hyper-dimensional molecular systems. By providing an immersive 3D environment that enables visualization and manipulation of real-time molecular simulations running on high-performance computing architectures, iMD-VR enables researchers to reach out and guide molecular conformational dynamics, in order to efficiently explore complex, high-dimensional molecular systems. Moreover, iMD-VR simulations generate rich datasets that capture human experts' spatial insight regarding molecular structure and function. This paper explores the use of researcher-generated iMD-VR datasets to train AI agents via imitation learning (IL). IL enables agents to mimic complex behaviours from expert demonstrations, circumventing the need for explicit programming or intricate reward design. In this article, we review IL across robotics and Multi-agents systems domains which are comparable to iMD-VR, and discuss how iMD-VR recordings could be used to train IL models to interact with MD simulations. We then illustrate the applications of these ideas through a proof-of-principle study where iMD-VR data was used to train a CNN network on a simple molecular manipulation task; namely, threading a small molecule through a nanotube pore. Finally, we outline future research directions and potential challenges of using AI agents to augment human expertise in navigating vast molecular conformational spaces.

Figures (8)

• Figure 1: Showing an iMD-VR user docking and undocking the drug oseltamivir from the H7N9 neuraminidase protein (see text for further detail) oconnor_interactive_2019
• Figure 2: Trajectory depicting threading of a Methane molecule through a carbon Nanotube
• Figure 3: Knot tying task in 17 Alanine molecule
• Figure 4: Process of BC
• Figure 5: Process of Inverse Reinforcement Learning (IRL)