EnzymeHunter: Achieving fine-grained enzyme function prediction with a hierarchically aware contrastive learning framework

Accurate enzyme function annotation is a grand challenge due to the vast number of uncharacterized proteins and the difficulty of distinguishing subtle functions. We introduce EnzymeHunter, a deep-learning framework that achieves fine-grained prediction via a hierarchically aware contrastive learning strategy. By integrating sequence and structural information and using the Enzyme Commission (EC) hierarchy to guide its loss function, our model learns a functionally coherent embedding space where distances reflect precise levels of catalytic similarity. EnzymeHunter significantly outperforms state-of-the-art models, particularly in challenging scenarios, achieving fine-grained precision down to the fourth EC level, maintaining robust performance in low-homology cases, and accurately predicting rare enzyme classes. In a proteome-wide application to Thermus thermophilus , EnzymeHunter discovered novel catalytic functions, one of which was subsequently validated by an independent UniProt update. Furthermore, our model is interpretable, with predictions guided by learned attention on mechanistically critical functional sites.