HCAF-DTA: drug-target binding affinity prediction with cross-attention fused hypergraph neural networks

TL;DR

HCAF-DTA addresses DTA prediction by integrating hypergraph-based drug substructure representations with ESM-1b-derived spatial protein graphs and a bidirectional multi-head cross-attention fusion mechanism. The method preserves higher-order molecular features and atom–residue interaction patterns, improving predictive accuracy and robustness in cold-start scenarios. On the Davis and KIBA benchmarks, HCAF-DTA achieves state-of-the-art results across MSE, CI, and $r_m^2$, and ablation confirms that both hypergraph encoding and cross-attention contribute meaningfully to performance. This approach has practical implications for drug design, enabling more reliable affinity predictions with improved generalization to novel drugs and targets.

Abstract

Accurate prediction of the binding affinity between drugs and target proteins is a core task in computer-aided drug design. Existing deep learning methods tend to ignore the information of internal sub-structural features of drug molecules and drug-target interactions, resulting in limited prediction performance. In this paper, we propose a drug-target association prediction model HCAF-DTA based on cross-attention fusion hypergraph neural network. The model innovatively introduces hypergraph representation in the feature extraction stage: drug molecule hypergraphs are constructed based on the tree decomposition algorithm, and the sub-structural and global features extracted by fusing the hypergraph neural network with the graphical neural network through hopping connections, in which the hyper edges can efficiently characterise the functional functional groups and other key chemical features; for the protein feature extraction, a weighted graph is constructed based on the residues predicted by the ESM model contact maps to construct weighted graphs, and multilayer graph neural networks were used to capture spatial dependencies. In the prediction stage, a bidirectional multi-head cross-attention mechanism is designed to model intermolecular interactions from the dual viewpoints of atoms and amino acids, and cross-modal features with correlated information are fused by attention. Experiments on benchmark datasets such as Davis and KIBA show that HCAF-DTA outperforms state of the arts in all three performance evaluation metrics, with the MSE metrics reaching 0.198 and 0.122, respectively, with an improvement of up to 4% from the optimal baseline.

Figures (5)

• Figure 1: The architecture of our proposed HCAF-DTA mode
• Figure 2: Schematic diagram of hypergraph construction process
• Figure 3: Bi-directional multi-head cross-attention module
• Figure 4: Configuration comparison of three cold-start scenarios
• Figure 5: Performance comparison of cold-start experiments