基于多源生物网络融合的肥胖与疾病关联研究

In today's society, the impact of obesity on health has become a major concern all over the world. Thus, research on obesity and its associated diseases aims to provide an important theoretical basis for disease prevention and treatment, in which how to integrate multi-source information to systematically predict the association between obesity and disease and to detect key pathogenic genes is the key problem. This project is based on text mining and multi-source information fusion to carry out research on the association of obesity diseases. Some main tasks include: build a network of obesity and disease genes based on fusion of multi-source protein interactions, gene transcription, and KEGG database; with the integration of gene expression profile data, and obesity and disease genes network, design a module mining method based on graph theory to identify the active network module, and adopt the WGCNA algorithm to verify the performance of the module mining algorithm; based on the topological features, design a random network and shortest path based method to calculate accessibility between obesity and disease genes thereby rank genes; design neighborhood search and network centrality based algorithms to search for key driver genes that can be possible drug targets for mediating the relationship between disease and obesity. The in-depth study of this project will develop new approaches for the early prevention and treatment of obesity-related diseases.

当今社会，肥胖对健康的影响已经成为了全世界范围都关注的焦点问题，对肥胖及其关联疾病的研究为疾病的预防和治疗提供重要的理论基础。如何融合多源信息系统性的预测肥胖和疾病关联、挖掘关键致病基因等是研究该领域的关键问题。本项目以文本挖掘及多源信息融合为基础，开展肥胖疾病关联研究：融合多源蛋白质相互作用、基因转录以及KEGG数据库等信息构建肥胖与疾病基因网络；整合基因表达谱数据和肥胖与疾病基因网络，设计基于图论的模块挖掘方法识别活性网络模块，并结合WGCNA算法验证模块挖掘算法的性能；根据构建的肥胖与疾病基因网络的拓扑特性以及基因功能富集设计基于全局网络分析的随机游走和最短路径方法来计算肥胖与疾病基因之间的可达性来排序基因；设计邻域搜索和网络节点拓扑中心算法搜索调停疾病与肥胖间关系和可能的药物靶点的关键驱动基因。本项目的深入研究将为肥胖关联疾病的早期预防和治疗开辟全新的途径。

肥胖对健康的影响已经成为了全世界范围都关注的焦点问题，对肥胖及其关联疾病的研究为疾病的预防和治疗提供重要的理论基础。如何融合多源信息系统性的预测肥胖和疾病关联、挖掘关键致病基因等是研究该领域的关键问题。本项目以文本挖掘及多源信息融合为基础，开展肥胖疾病关联研究：融合多源蛋白质相互作用、基因转录以及KEGG数据库等信息构建肥胖与疾病基因网络；整合基因表达谱数据和肥胖与疾病基因网络，设计基于图论的模块挖掘方法识别活性网络模块，并结合WGCNA算法验证模块挖掘算法的性能；根据构建的肥胖与疾病基因网络的拓扑特性以及基因功能富集设计基于全局网络分析的随机游走和最短路径方法来计算肥胖与疾病基因之间的可达性来排序基因；设计邻域搜索和网络节点拓扑中心算法搜索调停疾病与肥胖间关系和可能的药物靶点的关键驱动基因。本项目的深入研究将为肥胖关联疾病的早期预防和治疗开辟全新的途径。本项目已在生物信息杂志上发表SCI论文21篇，培养博士生2名，硕士生9名。