FGW-CLIP: Enhancing Enzyme Screening via Fused Gromov-Wasserstein Contrastive Learning
Keywords: Contrastive Learning, Enzyme Screening, 3D Structure, Hierarchical Classifications
Abstract: Enzymes are crucial catalysts for biochemical reactions, underpinning numerous biological processes. The efficient identification of specific enzymes from extensive protein libraries is essential for understanding and harnessing these biological reactions. While traditional computational methods for enzyme screening are time-consuming and resource-intensive, recent contrastive learning approaches have shown promise. However, these methods often overlook the inherent hierarchical classifications within enzymes and reactions, as well as the significance of molecular structure in catalysis. To address these limitations, we introduce FGW-CLIP, a novel contrastive learning framework based on optimizing the fused Gromov-Wasserstein distance. This approach incorporates multiple alignments, including representation alignment between reactions and enzymes, and internal alignment within enzyme and reaction representations. By introducing a regularization term, our method minimizes the Gromov-Wasserstein distance between enzyme and reaction spaces, enhancing information exchange within these domains. FGW-CLIP demonstrates superior performance on the widely-used EnzymeMap benchmark, significantly outperforming existing methods in enzyme virtual screening tasks. Notably, it achieves state-of-the-art results in both BEDROC and EF metrics, indicating its efficacy in identifying relevant enzymes for given reactions. These results highlight the potential of our method to advance virtual enzyme screening, offering a powerful tool for enzyme discovery and characterization.
Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 13484
Loading