[Paper Review] Polyply: a python suite for facilitating simulations of (bio-)macromolecules and nanomaterials

Fabian Grünewald, Riccardo Alessandri+4 more|arXiv (Cornell University)|May 12, 2021

Block Copolymer Self-Assembly4 citations

TL;DR

Polyply is a Python software suite that accelerates molecular dynamics simulations of complex (bio-)macromolecules and nanomaterials by combining a multi-scale graph matching algorithm for rapid parameterization and a generic multi-scale random walk protocol for efficient system setup. It enables high-throughput simulations of diverse systems, including multi-lamellar block copolymers and phase-separated polymer-lipid vesicles, with validated accuracy across multiple force fields.

ABSTRACT

Molecular dynamics simulations play an increasingly important role in the rational design of (nano)-materials and in the study of biomacromolecules. However, generating input files and realistic starting coordinates for these simulations is a major bottleneck, especially for high throughput protocols and for complex multi-component systems. To eliminate this bottleneck, we present the polyply software suite that leverages 1) a multi-scale graph matching algorithm designed to generate parameters quickly and for arbitrarily complex polymeric topologies, and 2) a generic multi-scale random walk protocol capable of setting up complex systems efficiently and independent of the target force-field or model resolution. We benchmark quality and performance of the approach by creating melt simulations of six different polymers using two force-fields with different resolution. We further demonstrate the power of our approach by setting up a multi lamellar microphase-separated block copolymer system for next generation batteries, and by generating a liquid-liquid phase separated polyethylene oxide-dextran system inside a lipid vesicle, featuring both branching and molecular weight distribution of the dextran component.

Motivation & Objective

• To address the bottleneck in generating input files and realistic starting coordinates for molecular dynamics simulations of complex multi-component systems.
• To enable high-throughput simulation workflows for (bio-)macromolecules and nanomaterials with minimal manual intervention.
• To support arbitrary polymeric topologies and diverse force fields through a flexible, extensible software framework.
• To facilitate the setup of complex systems such as phase-separated or multi-lamellar structures with minimal user-defined parameters.

Proposed method

• The suite employs a multi-scale graph matching algorithm to automatically assign force field parameters to complex polymeric topologies, enabling rapid parameterization.
• It uses a generic multi-scale random walk protocol to generate initial configurations that are physically realistic and independent of the target force field or resolution level.
• The framework supports both coarse-grained and all-atom representations, allowing seamless transitions across resolution scales.
• It integrates with established molecular dynamics engines and supports multiple force fields, including those with varying levels of detail.
• The system setup pipeline is modular and extensible, enabling application to diverse materials such as block copolymers, branched polymers, and biomolecular assemblies.
• The software is implemented in Python, ensuring accessibility, reproducibility, and integration with existing scientific computing workflows.

Experimental results

Research questions

• RQ1 Can a graph matching algorithm efficiently parameterize arbitrarily complex polymeric topologies across different force fields?
• RQ2 How well does the multi-scale random walk protocol generate physically realistic initial configurations for complex multi-component systems?
• RQ3 Can the framework enable high-throughput simulation of challenging systems such as microphase-separated block copolymers or liquid-liquid phase-separated systems in vesicles?
• RQ4 What is the performance and accuracy of Polyply in generating melt simulations of polymers with different force fields?
• RQ5 To what extent can the software reduce manual effort in preparing complex simulation inputs without compromising system quality?

Key findings

• Polyply successfully generated melt simulations of six different polymers using two distinct force fields, demonstrating robustness and transferability across parameter sets.
• The multi-scale graph matching algorithm enabled rapid and accurate parameter assignment for complex polymeric topologies, including branched and multi-block architectures.
• The generic multi-scale random walk protocol produced physically stable and realistic initial configurations for complex systems, including multi-lamellar block copolymer structures.
• The software enabled the setup of a liquid-liquid phase-separated polyethylene oxide-dextran system within a lipid vesicle, featuring molecular weight distribution and branching in dextran.
• Benchmarking showed that Polyply maintains high performance and accuracy, supporting high-throughput simulation workflows for (bio-)macromolecular and nanomaterial systems.