GHF-ACL: A novel contrastive learning framework with multi-order graph structures for herb-disease association prediction

Predicting Herb–Disease Associations (HDA) is pivotal for modernizing Traditional Chinese Medicine (TCM); however, this is impeded by data heterogeneity and the complex, multi-component mechanisms of herbal medicines. Existing drug–disease prediction models often struggle to capture high-order structural patterns and resolve semantic inconsistencies intrinsic to herbs. To overcome these limitations, we present HData, a standardized benchmark dataset that integrates herbal medicinal properties, chemical compositions, and disease associations. We further propose GHF-ACL, a novel multi-order graph contrastive learning framework designed for HDA prediction. Specifically, GHF-ACL explicitly models low-order functional similarities via a herb–disease similarity graph while capturing high-order component interactions through a herb–chemical hypergraph. Furthermore, an adaptive gating-guided structural interaction module aligns heterogeneous graph representations into a unified latent space, and hierarchical contrastive learning enforces consistency across structural views. Extensive experiments on five datasets demonstrate that GHF-ACL achieves superior or competitive performance over six state-of-the-art models across most metrics, with significant improvements over the best-performing baseline model in AUPR (+4.8% on LRSSL, + 3.81% on Cdata), F1 score, and Recall. These results underscore the model’s superior capability in detecting true positive associations within imbalanced biomedical data. By synergizing multi-view graph modeling, semantic fusion, and contrastive regularization, this work establishes a unified framework for HDA prediction, offering valuable insights for computational TCM and data-driven drug discovery.Author summary: Traditional Chinese Medicine has been used for centuries, but understanding how it works at a scientific level is still difficult. One reason is that herbs are naturally complex. Instead of relying on a single active substance, most herbs contain many components that work together to influence the body. This makes them hard to study using modern computer tools that are designed for simpler drugs. In our work, we aim to make this complexity easier to understand using artificial intelligence. We first collected and organized scattered information about herbs and diseases into a single, standardized data resource. We then built a computer model that looks not only at similarities between herbs and diseases, but also at how different components within an herb may work together. When we tested our approach, it was more successful than existing methods at identifying meaningful connections between herbs and diseases. As an example, our model highlighted a possible role for the herb Niuxi in treating skin cancers, which aligns with recent biological findings. Overall, we hope our study provides a clearer way to study traditional herbal medicine and helps connect ancient medical knowledge with modern, data-driven drug discovery.