基于多层次结构化网络整体分析技术的生脉方药效成分群虚拟筛选与实验验证研究

How to fast and accurately identify effective components of Chinese medicine is a critical scientific problem that needs to be solved to modernize Chinese medicine. It is thus our goal in this proposal to develop a holistic network analysis based virtual screening strategy to identify effective components using multi-level hierarchical network analysis approach followed by experimental validations. Multi-level hierarchical network in the form of “component-target-pathway-gene-disease” will be firstly constructed by integrating high throughput omics data, literature information and disease knowledge base. Holistic network analysis approaches will then be developed to quantitatively evaluate the network influence scores of chemical components on the disease network and the interaction strength between the components. The scores will be used for further fast virtual screening of effective components with respect to a specific clinical indication for a Chinese medicine. In our preliminary studies, we have investigated extensively the chemical substituents, tissue distribution, PK, PD and mechanism of actions of Shengmai formula using proteomics and metabolomics approaches. In addition, we have also developed a prototype multi-level network analysis algorithm and tool with successful applications in drug reposition for complex diseases such as cardiovascular diseases. This proposal will follow up on this to build multi-level hierarchical network for Shengmai formula against ischemic heart diseases and screen its effective components followed by validation in cellular and animal disease models. The validation results will be fed back to the computational models to optimize the parameters for best performances. This study integrates macro-level holistic network analysis with micro-level molecular interaction analysis to explore the strategy for effective components identification that is fitful to Chinese medicine. It is therefore of great importance to the innovation and development of Chinese medicine theory.

如何快速准确发现药效成分群是中药现代研究中的一个关键科学问题。本研究拟发展基于多层次结构化网络整体分析技术的药效成分群虚拟筛选与实验验证方法。项目通过融合高通量组学数据、文本信息、疾病知识库等信息，构建“成分-靶点-通路-基因-疾病”多层次结构化网络并建立网络整体影响力分析等关键技术，定量评价中药成分对疾病网络的作用强度和成分间相互作用关系，据此建立中药药效成分群快速虚拟筛选技术。前期工作研究了生脉方的化学物质、体内过程及基于蛋白和代谢组学的药理作用机制，同时初步建立了多层次结构化网络分析技术并成功应用于心血管等疾病的药物重定位研究。本研究将在此基础上建立生脉方治疗缺血性心脏病的多层次网络以筛选药效成分群，并通过细胞和动物模型进行实验验证，对虚拟筛选模型进行参数修正和优化。本项目结合宏观整体药效评价和微观药效物质辨析探索适合中药特点的药效物质发现方法，对中医药理论的创新发展具有重要意义。

项目针对如何快速准确发现中药复方中药效成分群的关键科学问题，构建了“成分-靶点-通路-疾病”多层次结构化网络整体分析方法，根据中药成分对疾病网络的影响强度，建立中药药效成分群快速虚拟筛选技术。技术应用于生脉方等中药复方的成分-靶点网络分析，通过虚拟筛选发现了有效成分群，并通过细胞实验验证确定了相关药效成分，其中生脉方的药效成分信息进一步应用于工业生产过程中的质量传递分析，优化了生脉注射液质量标志物和过程质量控制标准。项目发表论文12篇，其中sci论文8篇，中文核心4篇，申报发明专利1项，建立企业过程质量控制标准3项。项目建立了一套整合中药复方成分解析、药理研究、质量控制、工业生产的技术平台，为建立面向临床疗效的中药质量控制方法提供了一种快速、精准的质量标志物辨识手段。该方法被成功应用于生脉注射液标准化建设，具有较大的推广应用前景。