基于靶点网络的中药复方协同整合作用机制研究

发布时间：2018-08-08 15:34

【摘要】：中药是祖国医药宝库的重要组成部分,其治疗疾病的系统观和整体观被科学界日益重视。系统阐明中药整体性作用机制对促进中医药现代化和现代医药发展有着重要的意义。中药发挥整体性药效的微观体现是中药多成分作用于多个靶点形成复杂网络,并在不同的通路以及功能模块中产生协同性、整合性作用。揭示这一问题的科学本质已成为当今中医药学研究的迫切任务。为此,本研究从中药复方的作用靶点出发,从靶点的协同整合作用角度,开展复方整体性作用机制研究。主要包括:结合计算机技术和生物大数据,开发一个符合中药特点的直接靶点预测算法,以此发现中药复方作用靶点;基于上述靶点预测方法,从靶点的作用通路和功能角度阐明复方的多通路、多层次协同整合作用;此外,还在更高尺度的蛋白互作网络上探索了复方靶点网络的功能模块及其对疾病的多环节调节作用机制。具体内容包括:(1)为了识别中药复方的直接作用靶点,本研究基于大规模的配体-靶点真实结合数据,提出“配体群系综特征能反映其直接靶点的结合信息”的假设,构建了加权系综相似度算法(WES)来预测药物的直接靶点。WES算法的操作流程如下:a)通过不同数据源的检索,构建了98327个配体-靶点直接结合数据集;b)在系综框架下鉴定与药理学性质高度相关的关键配体结构参数;c)通过整体相似度(系综相似度)而非单个配体的相似度来判断一个药物能否与某一靶点结合;d)运用贝叶斯网络和多维核将标准化的系综相似度整合起来做出理论预测。WES模型外部测试和实验评价均有较高的准确率,分别为70%和71%。这些结果表明WES方法为药物重定位和药物开发提供了一个可靠的生物信息学模型。(2)基于上述靶点预测方法,为从靶点的作用通路和功能角度阐明复方的多通路、多层次协同整合作用,本研究以黄芪栀子复方(黄芩、秦皮、栀子、大黄、牡丹皮、生地黄、诃子、石榴皮、厚朴和枳翘)为范例进行说明。主要结果如下:a)通过类药性评价,获取了复方的212个潜在活性成分;b)以潜在活性成分为探针,借助WES算法预测得到了这些化合物靶向的122个蛋白靶点,并且通过进一步的体外实验验证了化合物和靶点的直接互作关系;c)化合物-靶点网络和靶点-通路网络关联分析阐明了疾病宏观表型与微观分子(化合物、蛋白、通路)之间的复杂联系,并筛选出了黄芩栀子复方靶点关联的多条通路,如Toll-like受体信号通路和Jak-STAT信号通路;d)整合并分析靶点通路发现复方靶点的多条通路通过抑制炎症、抗病毒感染和增强免疫协同发挥效果;e)复方靶点的分子功能、细胞组分及参与的生物学过程分析,揭示了复方的多层次调节作用。(3)此外,本研究以二陈汤为例,在更高尺度的蛋白互作网络上探索了复方靶点网络的功能模块及其对疾病的多环节调节作用机制。主要内容包括:a)通过类药性评价筛选出了二陈汤中的126个潜在活性成分;b)借助WES模型推断化合物-靶点关联,本研究获取了该复方的448个靶点,直观体现了化合物是如何靶向特定的靶蛋白并发挥作用从而形成复方靶点网络的;c)整合靶点互作网络和药物靶点网络,采用模块化分析方法,鉴别了11个具有特定生物学功能的模块,阐明了复方靶点网络中功能模块协同调节机制。综上所述,本研究围绕中药复方靶点以及其作用网络对中药复方协同整合作用进行了研究。构建了一个符合中药特点的直接靶点预测模型,该模型能够快速有效的识别中药复方的靶向蛋白,为中药复方机制研究和新药研发提供了新工具。在此基础上,本研究以黄芪栀子复方和二陈汤为范例开展复方作用靶点预测和分析,分别从靶点的功能、作用通路和靶点网络的功能模块角度,阐明复方的多通路、多层次、多环节协同整合作用,为中医药整体机制认识提供了新的借鉴。
[Abstract]:Traditional Chinese medicine is an important part of the treasure house of Chinese medicine. The systematic view and overall view of its treatment of diseases have been paid more and more attention by the scientific community. It is of great significance to systematically clarify the overall mechanism of the action of traditional Chinese medicine to promote the modernization of traditional Chinese Medicine and the development of modern medicine. The target points form complex networks and produce synergistic and integrated roles in different channels and functional modules. To reveal the scientific essence of this problem has become an urgent task in the study of traditional Chinese medicine. It mainly includes: developing a direct target prediction algorithm that conforms to the characteristics of traditional Chinese medicine in combination with computer technology and biological large data to find the target target of the Chinese medicine compound. Based on the target prediction method mentioned above, the multi pathway and multi level synergistic integration function of the compound are clarified from the target point of action path and function. In addition, the target point prediction method is also used. On the higher scale protein interaction network, the functional modules of the compound target network and the multi link regulation mechanism of the disease are explored. The specific contents include: (1) in order to identify the direct target of the compound Chinese medicine compound, this study is based on the real binding of the ligand target point on a large scale, and suggests that the characteristics of the ligand group ensemble can reflect its direct effect. A weighted ensemble similarity algorithm (WES) is constructed to predict the direct target.WES algorithm of the drug as follows: a) through the retrieval of different data sources, 98327 ligand target direct binding data sets are constructed, and b) identifies the key ligand structure highly related to the pharmacological properties under the ensemble framework. Parameters; c) through the overall similarity degree (ensemble similarity) rather than the similarity of single ligands to judge whether a drug can combine with a target; d) using the Bayesian network and multidimensional kernel to integrate the standardized ensemble similarity to make a theoretical prediction of the.WES model external test and experimental evaluation have a higher accuracy rate of 70% and 7, respectively. 1%. these results show that the WES method provides a reliable bioinformatics model for drug relocation and drug development. (2) based on the target prediction method mentioned above, the multi pathway and multi-level synergistic integration of the compound is clarified for the functional pathway and function angle of the target. Peony skin, rehmannia, pomegranate, pomegranate skin, Magnolia officinalis and trifoliate orange are illustrated as examples. The main results are as follows: a) 212 potential active components of the compound were obtained through the evaluation of drug resistance; b) was used as a probe to predict the target of these compounds with the help of WES algorithm, and through further in vitro The experiment verified the direct interaction of compounds and targets; c) compound target network and target point network association analysis elucidated the complex relationship between the macroscopic phenotype of the disease and the micromolecule (compound, protein, pathway), and screened the multiple pathways associated with the target of the Cape Gardenia compound, such as the Toll-like receptor signaling pathway and the Jak -STAT signal pathway; d) integration and analysis of target pathways to find multiple pathways of compound targets by inhibiting inflammation, antiviral infection and enhancing immune synergy; E) molecular function of compound targets, cell components and biological process analysis, revealing the multilevel regulation of compound compound. (3) in addition, this study was two Chen Tangwei. In the higher scale protein interaction network, the functional modules of the compound target network and the multi link regulation mechanism of the disease are explored. The main contents include: a) screening out 126 potential active components of two Chen Tangzhong through the evaluation of drug resistance; b) the compound target association was broken by the WES model. This study obtained the compound. The 448 target points directly reflect how compounds target specific target proteins and play a role in the formation of compound target network. C) integration of target interaction network and drug target network, and using modular analysis method to identify 11 modules with specific biological functions, and clarify the function module synergy in the compound target network. In summary, this study focuses on the synergistic integration of Chinese medicine compound target and its function network. A direct target prediction model conforms to the characteristics of traditional Chinese medicine. The model can quickly and effectively identify the target egg white of Chinese medicine compound, and provide the research of Chinese medicine compound mechanism and the research and development of new drugs. On this basis, on this basis, the target prediction and analysis of compound action target were carried out with the compound of Radix Astragali gardenia and two Chen Tang as an example. From the function of the target, the function pathway and the function module of the target network, the multi-channel, multi-level and multi link synergistic integration of the compound was clarified to provide the understanding of the overall mechanism of traditional Chinese medicine. A new reference.
【学位授予单位】：西北农林科技大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：R285

【相似文献】