Augmenting Control over Exploration Space in Molecular Dynamics Simulators to Streamline De Novo Analysis through Generative Control Policies

06/26/2023

∙

This study introduces the P5 model - a foundational method that utilizes reinforcement learning (RL) to augment control, effectiveness, and scalability in molecular dynamics simulations (MD). Our innovative strategy optimizes the sampling of target polymer chain conformations, marking an efficiency improvement of over 37.1 inductive bias, modulating Brownian forces to steer the system towards the preferred state, thereby expanding the exploration of the configuration space beyond what traditional MD allows. This broadened exploration generates a more varied set of conformations and targets specific properties, a feature pivotal for progress in polymer development, drug discovery, and material design. Our technique offers significant advantages when investigating new systems with limited prior knowledge, opening up new methodologies for tackling complex simulation problems with generative techniques.

READ FULL TEXT

Augmenting Control over Exploration Space in Molecular Dynamics Simulators to Streamline De Novo Analysis through Generative Control Policies

Curiosity in exploring chemical space: Intrinsic rewards for deep molecular reinforcement learning

Molecular Design in Synthetically Accessible Chemical Space via Deep Reinforcement Learning

sMolBoxes: Dataflow Model for Molecular Dynamics Exploration

Reinforced dynamics for enhanced sampling in large atomic and molecular systems. I. Basic Methodology

Hit and Lead Discovery with Explorative RL and Fragment-based Molecule Generation

Data-efficient, Explainable and Safe Payload Manipulation: An Illustration of the Advantages of Physical Priors in Model-Predictive Control

MDsrv – visual sharing and analysis of molecular dynamics simulations

Augmenting Control over Exploration Space in Molecular Dynamics Simulators to Streamline De Novo Analysis through Generative Control Policies

Related Research

Curiosity in exploring chemical space: Intrinsic rewards for deep molecular reinforcement learning

Molecular Design in Synthetically Accessible Chemical Space via Deep Reinforcement Learning

sMolBoxes: Dataflow Model for Molecular Dynamics Exploration

Reinforced dynamics for enhanced sampling in large atomic and molecular systems. I. Basic Methodology

Hit and Lead Discovery with Explorative RL and Fragment-based Molecule Generation

Data-efficient, Explainable and Safe Payload Manipulation: An Illustration of the Advantages of Physical Priors in Model-Predictive Control

MDsrv – visual sharing and analysis of molecular dynamics simulations