EnzyPGM: Pocket-conditioned Generative Model for Substrate-specific Enzyme Design

TL;DR

EnzyPGM is a novel generative framework that designs substrate-specific enzymes by modeling pocket-substrate interactions and incorporating enzyme function priors, advancing enzyme engineering capabilities.

Contribution

It introduces a unified model with bi-scale attention and function fusion modules, along with a new enzyme-pocket dataset, enabling more accurate substrate-specific enzyme design.

Findings

Achieves state-of-the-art performance on EnzyPock dataset.

Reduces average binding energy by 0.47 kcal/mol compared to previous models.

Effectively models pocket-substrate interactions for enzyme design.

Abstract

Designing enzymes with substrate-binding pockets is a critical challenge in protein engineering, as catalytic activity depends on the precise interaction between pockets and substrates. Currently, generative models dominate functional protein design but cannot model pocket-substrate interactions, which limits the generation of enzymes with precise catalytic environments. To address this issue, we propose EnzyPGM, a unified framework that jointly generates enzymes and substrate-binding pockets conditioned on functional priors and substrates, with a particular focus on learning accurate pocket-substrate interactions. At its core, EnzyPGM includes two main modules: a Residue-atom Bi-scale Attention (RBA) that jointly models intra-residue dependencies and fine-grained interactions between pocket residues and substrate atoms, and a Residue Function Fusion (RFF) that incorporates enzyme function priors into residue representations. Also, we curate EnzyPock, an enzyme-pocket dataset comprising 83,062 enzyme-substrate pairs across 1,036 four-level enzyme families. Extensive experiments demonstrate that EnzyPGM achieves state-of-the-art performance on EnzyPock. Notably, EnzyPGM reduces the average binding energy of 0.47 kcal/mol over EnzyGen, showing its superior performance on substrate-specific enzyme design. The code and dataset will be released later.