Authors
- Shunfan Li China University of Geoscience, Wuhan, China
- Jiangkai Long China University of Geoscience, Wuhan, China
- Xin Zou China University of Geoscience, Wuhan, China
- Chang Tang Huazhong University of Science and Technology
- Yuanyuan Liu China University of Geoscience, Wuhan, China
- Xiao He China University of Geoscience, Wuhan, China
- Xuesong Yan China University of Geoscience, Wuhan, China
DOI:
https://doi.org/10.1609/aaai.v40i1.37031Abstract
Protein-Ligand Affinity (PLA) prediction quantifies the interaction strength to guide rational drug design. Existing approaches typically analyze interaction at a single granularity and overlook tightly coupled relationships between protein and ligand in both structure and functionality, consequently yielding suboptimal representations, leading to significant performance drops in real-world scenarios. To address this problem, we propose PLA-MGRA, a minimalist and effective PLA prediction framework. Specifically, PLA-MGRA captures both fine-grained atomic details and coarse grained functional semantics within the 3D structure of protein–ligand complexes, through multi-granularity learning. To further parse the coupled protein–ligand relationships, we design relation-aware learning to enhance the binding nature of representations. Extensive experiments demonstrate that our method achieves state-of-the-art performance on multiple protein–ligand affinity prediction benchmarks, while also offering generalizability and interpretability.
How to Cite
Li, S., Long, J., Zou, X., Tang, C., Liu, Y., He, X., & Yan, X. (2026). PLA-MGRA: Multi-Granularity and Relation-Aware Learning for Efficient and Generalizable Protein-Ligand Binding Affinity Prediction. Proceedings of the AAAI Conference on Artificial Intelligence, 40(1), 659–667. https://doi.org/10.1609/aaai.v40i1.37031
Issue
Section
AAAI Technical Track on Application Domains I