南方医科大学学报

南方医科大学学报 ›› 2026, Vol. 46 ›› Issue (8): 1947-1955.doi: 10.12122/j.issn.1673-4254.2026.08.22

• • 上一篇    

基于知识感知图谱的药物重定向模型预测三阴性乳腺癌治疗药物及验证

吴迪恒1,2(), 徐展发1, 李镱3, 张明旭1,2, 林嘉泽1,2, 吕祎君2, 官道刚3(), 秦耿耿1,2()   

  1. 1.南方医科大学，生物医学工程学院，广东 广州 510515
    2.南方医科大学，南方医院影像诊断科，广东 广州 510515
    3.南方医科大学，基础医学院生物化学与分子生物学系，广东 广州 510515
  • 收稿日期:2025-12-03 出版日期:2026-08-20 发布日期:2026-08-01
  • 通讯作者: 官道刚,秦耿耿 E-mail:804352722@qq.com;guandg0929@smu.edu.cn;zealotq@smu.edu.cn
  • 作者简介:吴迪恒，在读硕士研究生，E-mail: 804352722@qq.com
  • 基金资助:
    国家自然科学基金(3257071033);广东省自然科学基金(414050003969);广东省研究生教育创新计划项目(2024JGXM_029)

Prediction and verification of therapeutic drugs for triple-negative breast cancer using a knowledge graph-based drug repurposing model

Diheng WU1,2(), Zhanfa XU1, Yi LI3, Mingxu ZHANG1,2, Jiaze LIN1,2, Yijun LÜ2, Daogang GUAN3(), Genggeng QIN1,2()   

  1. 1.School of Biomedical Engineering, School of Basic Medical Sciences, Southern Medical University, Guangzhou 510515, China
    2.Department of Imaging Diagnosis, Nanfang Hospital, School of Basic Medical Sciences, Southern Medical University, Guangzhou 510515, China
    3.Department of Biochemistry and Molecular Biology, School of Basic Medical Sciences, Southern Medical University, Guangzhou 510515, China
  • Received:2025-12-03 Online:2026-08-20 Published:2026-08-01
  • Contact: Daogang GUAN, Genggeng QIN E-mail:804352722@qq.com;guandg0929@smu.edu.cn;zealotq@smu.edu.cn
  • Supported by:
    National Natural Science Foundation of China(3257071033)

RichHTML

2

PDF

36

摘要：

目的 构建一个基于知识图谱的药物重定向模型用于预测三阴性乳腺癌（TNBC）潜在治疗药物。 方法 基于BindingDB构建大规模药物-靶点相互作用（DTI）数据集（涵盖多种亲和力指标，规模达数万条），并引入STRING数据库中百万级蛋白质互作网络作为外部知识。提出知识感知图谱神经网络（KGNN）模型，融合图卷积网络提取的药物分子图特征、ProtBERT编码的蛋白质序列特征以及PPI网络信息，通过多头注意力机制和门控融合模块建模药物-靶点交互关系。采用均方误差（MSE）、皮尔逊相关系数（PCC）和一致性指数（CI）进行评估，并通过消融实验与冷启动实验验证模型有效性与泛化能力。进一步结合TCGA筛选TNBC相关基因，预测候选药物并通过分子对接模拟进行验证。 结果 在DTI亲和力预测任务中，KGNN性能（MSE=3.2697，PCC=0.8037，CI=0.7862）优于KronRLS、SimBoost、DeepDTA、FusionDTA和GraphDTA基准模型。消融实验证实各模块（如多头注意力MSE上升5.60%，PPI融合MSE上升10.82%）对性能提升的关键作用。冷启动场景下，KGNN在未见药物/靶点条件下仍优于对比模型，展示强泛化能力。TNBC候选药物预测分数与分子对接模拟结果一致（PCC>0.85），注意力可视化揭示药效热点。 结论 KGNN模型可有效预测DTI，有助于药物重定向，为多靶点结合新药设计提供导向，减少新药筛选空间与实验验证成本，具备良好的生物应用前景。

关键词: 药物重定向, 药物-靶点结合亲和力, 知识图谱, 注意力机制, 知识感知图神经网络, 三阴性乳腺癌

Abstract:

Objective To construct a knowledge graph-based drug repurposing model for predicting potential therapeutic drugs for triple-negative breast cancer (TNBC). Methods Drug-target interaction (DTI) affinity data were collected from the BindingDB database and filtered (including data of Kd, Ki, EC50 and IC50). The proposed KGNN model integrates graph convolutional network (GCN)‑extracted drug molecular graph features, ProtBERT-pretrained protein sequence representations, and STRING-derived protein-protein interaction (PPI) knowledge graphs. Multi-head attention mechanisms and gated fusion modules were used to model interaction dependencies. Model performance was evaluated using mean squared error (MSE), Pearson correlation coefficient (PCC), and concordance index (CI). Ablation studies were performed to assess module contributions, and cold-start experiments were conducted to test generalization ability of the model. Using data from TCGA, 1340 TNBC-associated pathogenic genes were screened by bioinformatics analyses and mapped to targets using UniProt. KGNN was applied to predict the candidate drugs, which were validated through molecular docking and molecular dynamics simulations. Results In the DTI affinity prediction task, KGNN outperformed the benchmark models including KronRLS, SimBoost, DeepDTA, FusionDTA, and GraphDTA (MSE=3.2697, PCC=0.8037, and CI=0.7862). Ablation studies confirmed the critical roles of the modules for enhancing model performance (multi-head attention increased MSE by 5.60%; PPI fusion increased MSE by 10.82%). In cold-start scenarios, KGNN maintained superior performance over the comparators in unseen drug/target settings, demonstrating robust generalization. The TNBC candidate drug predictions well aligned with docking affinities and dynamics simulations (Pearson correlation coefficient>0.85), while attention visualization highlighted the efficacy hotspots. Conclusion The KGNN model can effectively predict drug-target interactions to facilitate drug repurposing and the design of multi-target drugs while reducing the screening space and experimental validation costs.

Key words: drug repurposing, drug-target binding affinity, knowledge graph, attention mechanism, knowledge-aware graph neural network, triple-negative breast cancer