UniZyme: A Unified Protein Cleavage Site Predictor Enhanced with Enzyme Active-Site Knowledge

TL;DR

UniZyme is a novel unified model that predicts protein cleavage sites across diverse enzymes by integrating enzyme active-site knowledge, improving generalization and accuracy for biological and medical applications.

Contribution

The paper introduces UniZyme, a biochemically-informed model that generalizes cleavage site prediction across multiple enzymes, including unseen ones, by leveraging shared enzyme knowledge.

Findings

Achieves high accuracy across various proteolytic enzymes

Successfully predicts cleavage sites for previously unseen enzymes

Demonstrates the effectiveness of enzyme active-site knowledge integration

Abstract

Enzyme-catalyzed protein cleavage is essential for many biological functions. Accurate prediction of cleavage sites can facilitate various applications such as drug development, enzyme design, and a deeper understanding of biological mechanisms. However, most existing models are restricted to an individual enzyme, which neglects shared knowledge of enzymes and fails to generalize to novel enzymes. Thus, we introduce a unified protein cleavage site predictor named UniZyme, which can generalize across diverse enzymes. To enhance the enzyme encoding for the protein cleavage site prediction, UniZyme employs a novel biochemically-informed model architecture along with active-site knowledge of proteolytic enzymes. Extensive experiments demonstrate that UniZyme achieves high accuracy in predicting cleavage sites across a range of proteolytic enzymes, including unseen enzymes. The code is available in https://github.com/Ao-LiChen/UniZyme