| 摘要: |
| 目的:通过整合空间代谢组学、单细胞转录组学及机器学习算法，系统解析黄芪赤风自拟方靶向“线粒体-代谢轴”改善肌少症的多层级调控机制。方法：在中医药系统药理学数据库与分析平台（Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform，TCMSP）、在线人类孟德尔遗传数据库（Online Mendelian Inheritance in Man，OMIM）等数据库上筛选黄芪赤风自拟方活性成分和肌少症相关靶点，结合多组学数据（GSE1428等数据集）进行差异基因识别，通过12种机器学习算法筛选出5个核心基因。结果:功能富集分析表明，这些基因参与氧化磷酸化、嘌呤代谢及转化生长因子-β（Transforming Growth Factor-Beta，TGF-β）通路，并介导线粒体生物合成与能量代谢调控。免疫浸润分析揭示肌少症中T细胞亚群（CD4记忆T细胞）与巨噬细胞M2的异常浸润，且脂素1（Lipin 1，LPIN1）与免疫微环境显著相关。分子对接验证了维生素E与核心靶点的高亲和力，提示其潜在治疗价值。结论：研究构建了药物-成分-线粒体-代谢基因网络，阐明了黄芪赤风自拟方多靶点协同机制，为中医治疗肌少症提供了“数据-算法-机制”三位一体的新范式，为精准介入策略奠定了理论基础。 |
| 关键词: 肌少症 黄芪赤风自拟方 线粒体-代谢轴 多组学 机器学习 |
| DOI： |
|
|
| Mechanism of self-made Huangqi Chifeng prescription improving sarcopenia by regulating the mitochondria-metabolism axis: A study based on multi-omics and machine learning |
| LIU Xiaoming,LIU Hui,KUANG Yingyan,HU Shaowen |
| (Baoan Hospital of Traditional Chinese Medicine，Shenzhen 518100，Guangdong，China;The Seventh Clinical Medical School of Guangzhou University of Chinese Medicine，Shenzhen 518100，Guangdong，China;Zhongshan Rehabilitation Hospital，Zhongshan 528400，Guangdong，China) |
| Abstract: |
| Objective: To systematically analyze the multi-level regulatory mechanism of self-made Huangqi Chifeng Decoction in improving sarcopenia by targeting the “mitochondria-metabolism axis” based on the integration of spatial metabolomics，single-cell transcriptomics，and machine learning algorithms. Methods: Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform and Online Mendelian Inheritance in Man were used to obtain the targets associated with the active components of self-made Huangqi Chifeng Decoction and the disease sarcopenia，and multi-omics data （including GSE1428 dataset） were used to identify differentially expressed genes. A total of 12 machine learning algorithms were used to obtain 5 core genes. Results: The functional enrichment analysis showed that these genes participated in oxidative phosphorylation，purine metabolism，and the transforming growth factor-β pathway and mediated the regulation of mitochondrial biogenesis and energy metabolism. The immune infiltration analysis revealed abnormal infiltration of T-cell subsets （CD4 memory T cells） and macrophage M2 in sarcopenia，and lipin 1 was significantly associated with the immune microenvironment. Molecular docking validated the high affinity between vitamin E and core targets，suggesting that it had a certain therapeutic value. Conclusion: This study constructs a drug-component-mitochondria-metabolic gene network，clarifies the multi-target synergistic mechanism of self-made Huangqi Chifeng Decoction，and establishes a novel “data-algorithm-mechanism” paradigm for traditional Chinese medicine treatment of sarcopenia，which lays a theoretical foundation for precise intervention strategies. |
| Key words: sarcopenia self-made Huangqi Chifeng Decoction mitochondria-metabolism axis multi-omics machine learning |
