CWFBind: Geometry-Awareness for Fast and Accurate Protein-Ligand Docking

2025-08-13

AI Summary

Accurate prediction of small-molecule–protein binding poses remains a central challenge in structure-based drug design. Existing graph neural network (GNN)-based deep learning docking methods often neglect 3D geometric information, leading to inaccurate binding pocket localization and distorted ligand conformations. To address these limitations, we propose GeoDock—a novel geometric deep learning framework for molecular docking. First, it introduces local curvature features into the docking model to enhance explicit perception of protein pocket geometry. Second, it designs a degree-aware weighted message-passing mechanism to improve identification accuracy of functionally critical residues. Third, it incorporates a ligand-aware dynamic radius strategy to mitigate class imbalance in pocket prediction. Evaluated on standard benchmarks including PDBbind and CrossDocked, GeoDock achieves state-of-the-art performance—significantly outperforming mainstream methods—while maintaining millisecond-level inference speed. It attains high-precision pose prediction with RMSD < 2 Å, striking an optimal balance between accuracy and computational efficiency.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurately predicting the binding conformation of small-molecule ligands to protein targets is a critical step in rational drug design. Although recent deep learning-based docking surpasses traditional methods in speed and accuracy, many approaches rely on graph representations and language model-inspired encoders while neglecting critical geometric information, resulting in inaccurate pocket localization and unrealistic binding conformations. In this study, we introduce CWFBind, a weighted, fast, and accurate docking method based on local curvature features. Specifically, we integrate local curvature descriptors during the feature extraction phase to enrich the geometric representation of both proteins and ligands, complementing existing chemical, sequence, and structural features. Furthermore, we embed degree-aware weighting mechanisms into the message passing process, enhancing the model's ability to capture spatial structural distinctions and interaction strengths. To address the class imbalance challenge in pocket prediction, CWFBind employs a ligand-aware dynamic radius strategy alongside an enhanced loss function, facilitating more precise identification of binding regions and key residues. Comprehensive experimental evaluations demonstrate that CWFBind achieves competitive performance across multiple docking benchmarks, offering a balanced trade-off between accuracy and efficiency.

Problem

• Predict protein-ligand binding conformations accurately
• Address geometric information neglect in docking methods
• Improve pocket localization and binding conformation realism

Innovation

• Integrates local curvature descriptors for geometric representation
• Embeds degree-aware weighting in message passing
• Uses ligand-aware dynamic radius for pocket prediction