ResPocket: A Multi-Scale Feature Fusion Method for Improving Protein Binding Site Detection

Xiaoli Lin, Wenli Yang, Yaoyu Chen, Xiongwei Liao, Zimo Wu, Xiaolong ZhangBIBM 2024 · OpenReview

Abstract

Accurate detection of protein binding sites is critical for facilitating drug design. Most existing models rely on features at a single scale, leading to the loss of crucial global or local structural information. Therefore, this paper proposes a novel 3D model (ResPocket), to fully capture the multi-scale features of proteins. First, Residual Block (ResB) based on U-Net is used to capture features from different scales, including global and local structural information. Then, an Enhanced Attention Module (EAM) is designed to encode semantic dependencies by effectively capturing global context information. In addition, ResPocket focuses on learning local information about protein pockets with Self-Guided Mask Block (SGMB), to increase the robustness of the overall model and reduce its dependence on external mask guidance. Experimental results show that ResPocket improves by an average of 4.89% on DCA Top-n, by an average of 1.48% and 2.25% on DCC and DVO metrics, respectively.
BibTeX
@inproceedings{lin2024respocket,
  author = {Xiaoli Lin and Wenli Yang and Yaoyu Chen and Xiongwei Liao and Zimo Wu and Xiaolong Zhang},
  title = {{ResPocket: A Multi-Scale Feature Fusion Method for Improving Protein Binding Site Detection}},
  booktitle = {BIBM},
  year = {2024},
  url = {https://openreview.net/forum?id=mPSsu6y3J3},
}

Loading related content...

Figure