目的 通过重启随机游走（random walk with restart，RWR）算法解析桃红四物汤药味配伍、成分群协同抗血栓作用的生物过程网络与关键成分群，揭示其“多成分-多靶点-多途径”抗血栓作用机制。方法 结合文献和课题组前期研究基础，通过中药系统药理学数据库与分析平台（TCMSP）、本草组鉴（HERB）等4个数据库，获取桃红四物汤活性成分、成分群及对应靶标基因；在人类基因数据库（Genecards）、人类孟德尔遗传综合数据库（OMIM）等4个数据库收集血栓相关基因；首先将血栓相关基因、中药和成分群靶标基因导入Metascape平台进行生物过程富集，采用Cytoscape 3.9.1软件将血栓生物过程进行聚类，初步分析桃红四物汤中药配伍、成分群协同抗血栓作用；进一步整合RWR算法、Pearson相关性分析与Z得分显著性分析，定量分析复方、单味药与成分群对血栓生物过程网络影响的相关性及显著性，筛选成分群协同抗血栓的关键生物过程、关键成分群；最后构建桃红四物汤抗血栓关键生物过程网络，分子对接筛选核心成分与核心靶点。结果 纳入桃红四物汤9类成分群，血栓生物过程聚类为4类，桃红四物汤抗血栓生物过程显著性影响大小依次为氧化应激＞炎症反应＞细胞内信号传导＞细胞活化，氧化应激与炎症反应是其抗血栓的关键生物过程；黄酮类、有机酸类、生物碱类、苯乙醇苷类、萜类、香豆素类、苯酞类是其协同抗血栓的关键成分群；生物网络构建与分子对接筛选出10个核心靶点（TNF、HMGB1、NLRP3、PARP1、PPARA、GAPDH、MMP9、IFNG、IL17A、HRAS）、6个核心成分（川芎哚、儿茶素、绿原酸、毛蕊花糖苷、羟基红花黄色素A、芍药苷）。结论 通过整合生物过程网络和机器学习算法，定量解析桃红四物汤成分群协同抗血栓的关键生物过程和核心成分、靶点，为其临床应用和基础研究提供参考。

Objective To analyze biological process network and key component groups of the synergistic anti-thrombotic effect of the components in Taohong Siwu Decoction (桃红四物汤, TSD) using the random walk with restart (RWR) algorithm, revealing its “multi-component, multi-target, and multi-pathway” anti-thrombotic mechanisms. Methods Integrating literature and previous research, active ingredients of TSD, ingredient groups, and corresponding target genes were obtained from four databases including Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) and HERB. Thrombosis-related genes were collected from four databases such as GeneCards and OMIM. Initially, thrombosis-related genes, Chinese herbal medicine, and component group target genes were imported into the Metascape platform for biological process enrichment. Cytoscape 3.9.1 software was used for clustering thrombosis biological processes, and the synergistic anti-thrombotic effect of Chinese herbal medicine combinations and ingredient groups was preliminarily analyzed. Further, RWR algorithm, Pearson correlation analysis, and Z-score significance analysis were integrated to quantitatively analyze the correlation and significance of compound, single herbal medicine, and component group effects on thrombosis biological process networks, screening key biological processes and ingredient groups synergistically combating thrombosis. Finally, the key biological process network of TSD anti-thrombosis was constructed, and molecular docking was used to screen core components and core targets. Results The study included nine categories of ingredient groups in TSD, clustering of thrombotic biological processes into four categories. The significance of TSD impact on thrombotic biological processes is as follows: oxidative stress > inflammatory response > intracellular signaling > cell activation. Oxidative stress and inflammation were identified as key biological processes in its anti-thrombotic action. Flavonoids, organic acids, alkaloids, phenylethanoid glycosides, terpenes, coumarins, and phenanthrenes were identified as key ingredient groups in its synergistic anti-thrombotic effect. The biological network construction and molecular docking identified 10 core targets (TNF, HMGB1, NLRP3, PARP1, PPARA, GAPDH, MMP9, IFNG, IL17A, HRAS) and six core components (ligustilide, catechin, chlorogenic acid, verbascoside, hydroxysafflor yellow A, paeoniflorin). Conclusion In this paper, the key biological processes, core components and targets of the synergistic antithrombotic combination of the component groups of TSD were quantitatively analyzed by integrating bioprocess networks and machine learning algorithms, so as to provide a reference for its clinical application and basic research.

Q811.4;TP18;R285

四川省自然科学基金面上项目(2024NSFSC0696)