Lin Yitong, He Jiaying, Chen Jiahe, Zhu Xinnan, Zheng Jianwei, Bo Tao
College of Computer Science and Technology, Zhejiang University of Technology , 288 Liuhe Road, Xihu District, Hangzhou, Zhejiang Province, 310023, China.
Key Laboratory of Endocrine Glucose & Lipids Metabolism, Department of Endocrinology, , Shandong Provincial Hospital Affiliated to Shandong First Medical University , 324 Jingwu Road, Huaiyin District, Jinan, Shandong Province, 250021, China.
Bioinformatics. 2025 Jul 1;41(7). doi: 10.1093/bioinformatics/btaf408.
Biomedical knowledge graphs (KGs) are crucial for drug discovery and disease understanding, yet their completion and reasoning are challenging. Knowledge embedding (KE) methods capture global semantics but struggle with dynamic structural integration, while graph neural networks (GNNs) excel locally but often lack semantic understanding. Even ensemble approaches, including those leveraging language models, often fail to achieve a deep, adaptive, and synergistic co-evolution between semantic comprehension and structural learning. Addressing this critical gap in fostering continuous, reciprocal refinement between these two aspects in complex biomedical KGs is paramount.
We introduce BioGraphFusion, a novel framework for deeply synergistic semantic and structural learning. BioGraphFusion establishes a global semantic foundation via tensor decomposition, guiding an LSTM-driven mechanism to dynamically refine relation embeddings during graph propagation. This fosters adaptive interplay between semantic understanding and structural learning, further enhanced by query-guided subgraph construction and a hybrid scoring mechanism. Experiments across three key biomedical tasks demonstrate BioGraphFusion's superior performance over state-of-the-art KE, GNN, and ensemble models. A case study on cutaneous malignant melanoma 1 highlights its ability to unveil biologically meaningful pathways.
Source code and all data underlying this article are freely available in the GitHub repository at https://github.com/Y-TARL/BioGraphFusion.
生物医学知识图谱对于药物发现和疾病理解至关重要，但其完善和推理具有挑战性。知识嵌入（KE）方法能够捕捉全局语义，但在动态结构整合方面存在困难，而图神经网络（GNN）在局部表现出色，但往往缺乏语义理解。即使是包括利用语言模型的集成方法，也常常无法在语义理解和结构学习之间实现深度、自适应和协同的共同进化。弥合复杂生物医学知识图谱中这两个方面之间持续、相互优化的关键差距至关重要。
我们引入了BioGraphFusion，这是一个用于深度协同语义和结构学习的新颖框架。BioGraphFusion通过张量分解建立全局语义基础，引导由长短期记忆网络（LSTM）驱动的机制在图传播过程中动态优化关系嵌入。这促进了语义理解和结构学习之间的自适应相互作用，通过查询引导的子图构建和混合评分机制进一步增强。在三个关键生物医学任务上的实验表明，BioGraphFusion比现有最先进的KE、GNN和集成模型具有更优的性能。一项关于皮肤恶性黑色素瘤1的案例研究突出了其揭示生物学上有意义的通路的能力。
本文的源代码和所有基础数据可在GitHub仓库（https://github.com/Y-TARL/BioGraphFusion）上免费获取。
