Model-based reinforcement learning for protein backbone design

TL;DR

This paper introduces a novel application of AlphaZero, a model-based reinforcement learning algorithm, to the design of protein backbones, significantly improving the accuracy and efficiency of generating proteins with desired shapes and structural properties.

Contribution

It extends Monte Carlo tree search with new reward mechanisms and applies AlphaZero to protein backbone design, outperforming existing methods by over 100%.

Findings

AlphaZero surpasses baseline MCTS by more than 100% in top-down protein design.

Incorporating secondary objectives improves design outcomes.

The approach demonstrates promising potential for complex protein design tasks.

Abstract

Designing protein nanomaterials of predefined shape and characteristics has the potential to dramatically impact the medical industry. Machine learning (ML) has proven successful in protein design, reducing the need for expensive wet lab experiment rounds. However, challenges persist in efficiently exploring the protein fitness landscapes to identify optimal protein designs. In response, we propose the use of AlphaZero to generate protein backbones, meeting shape and structural scoring requirements. We extend an existing Monte Carlo tree search (MCTS) framework by incorporating a novel threshold-based reward and secondary objectives to improve design precision. This innovation considerably outperforms existing approaches, leading to protein backbones that better respect structural scores. The application of AlphaZero is novel in the context of protein backbone design and demonstrates promising performance. AlphaZero consistently surpasses baseline MCTS by more than 100% in top-down protein design tasks. Additionally, our application of AlphaZero with secondary objectives uncovers further promising outcomes, indicating the potential of model-based reinforcement learning (RL) in navigating the intricate and nuanced aspects of protein design