基于链接聚类的代谢网络社团结构研究

Research on community structure in metabolic network based on link clustering

社团结构分析有助于识别代谢网络中的功能模块,有助于理解代谢网络的结构和功能关系,是代谢网络研究领域的一个重要研究课题。然而,当前的社团结构分析方法均依赖于对网络中的节点进行聚类分析,导致每个节点只能属于某一个社团。采用了一种对复杂网络中的链接进行聚类分析的方法,对高质量金黄色葡萄球菌代谢网络模型的巨强连通体进行了社团结构分析,得到了10个具有生物学意义的功能模块,结果表明链接聚类可用于识别新陈代谢网络中的功能社团。

Community structure can helpful for identifying functional modules in metabolic network,understanding the structure and function of metabolic network,and thus being an important subject in metabolic network study.However,current com- munity structure methods is mainly conducted by nodes clustering,which results each node only belong to a single community.This paper engages a links clustering based method for analyzing the giant strong component of S.aureus metabolic network from published high-quality models,and obtains 10 functional modules with better biological significance,which suggest that links clustering can identify functional communities in metabolic network.