A Simple yet Effective DDG Predictor is An Unsupervised Antibody Optimizer and Explainer

TL;DR

This paper introduces Light-DDG, a lightweight, unsupervised predictor for protein mutation effects that improves efficiency and interpretability in antibody optimization using a structure-aware Transformer and large-scale mutation data.

Contribution

The paper presents a novel, efficient DDG predictor based on a Transformer architecture, enhanced by knowledge distillation and large-scale pre-training, along with a mutation explainer for antibody optimization.

Findings

Light-DDG achieves competitive accuracy with less computational cost.

The mutation explainer effectively identifies mutation preferences.

The approach enables rapid antibody candidate screening and optimization.

Abstract

The proteins that exist today have been optimized over billions of years of natural evolution, during which nature creates random mutations and selects them. The discovery of functionally promising mutations is challenged by the limited evolutionary accessible regions, i.e., only a small region on the fitness landscape is beneficial. There have been numerous priors used to constrain protein evolution to regions of landscapes with high-fitness variants, among which the change in binding free energy (DDG) of protein complexes upon mutations is one of the most commonly used priors. However, the huge mutation space poses two challenges: (1) how to improve the efficiency of DDG prediction for fast mutation screening; and (2) how to explain mutation preferences and efficiently explore accessible evolutionary regions. To address these challenges, we propose a lightweight DDG predictor (Light-DDG), which adopts a structure-aware Transformer as the backbone and enhances it by knowledge distilled from existing powerful but computationally heavy DDG predictors. Additionally, we augmented, annotated, and released a large-scale dataset containing millions of mutation data for pre-training Light-DDG. We find that such a simple yet effective Light-DDG can serve as a good unsupervised antibody optimizer and explainer. For the target antibody, we propose a novel Mutation Explainer to learn mutation preferences, which accounts for the marginal benefit of each mutation per residue. To further explore accessible evolutionary regions, we conduct preference-guided antibody optimization and evaluate antibody candidates quickly using Light-DDG to identify desirable mutations.