基于代谢网络拓扑结构的基元模块发现方法

A method to discover elementary modules of metabolic networks based on topology

目的为提取出物种代谢网络具有的基本结构模块即基元模块,提出一种基于拓扑结构的发现代谢网络基元模块方法。方法首先利用网络模块化方法将不同物种的代谢网络划分为模块集合,然后通过合并相似的模块以及分解较大的代表模块等操作,最终获得基元模块集合。同时,利用超几何假设检验的方法以确定模块所具有的代谢功能,并在此基础上验证发现基元模块方法的合理性。结果对KEGG数据库中的1452个物种代谢网络经上述方法处理后,最终得到11446个基元模块,并且这些基元模块能够很好地表示代谢网络。结论提出了一种基于代谢网络拓扑结构的发现基元模块的方法,该方法能够有效提取出组成物种代谢网络的基元模块。

Objective In order to extract the elementary modules of organisms＇ metabolic networks,we propose a method of discovering elementary modules of metabolic networks based on topology. Methods We first decompose organisms＇ metabolic networks into a collection of modules using method of network modularity. Then we get the ultimate elementary modules through merging similar modules and breaking down larger representative modules that can be totally represented by other smaller modules. Meanwhile, we identify metabolic functions involved in module by hypergeometrie hypothesis testing and validate our method. Results Applying our method to 1452 organisms＇ metabolic network from KEGG, we finally acquire 11446 elementary modules. In addition, these elementary modules can excellently represent organisms＇ metabolic network. Conclusions We propose a method to discover elementary modules of metabolic networks based on topology, which can efficiently extract the elementary modules that compose organisms＇ metabolic network.