基于图论的代谢网络中流通代谢物处理新方法

A graph-theory-based method for processing of currency metabolites in metabolic networks

摘要:

使用图论方法对基因组尺度代谢网络进行途径搜索是目前途径设计中最常用的方法之一。然而,由于流通代谢物(如H⁺、H₂O、CO₂和ATP等)的影响,图论搜索得到的途径在生物学上经常是不可行的。虽然前人提出了一些方法对流通代谢物进行处理,但均存在一些问题,目前还没有标准的处理方法。文中基于流通代谢物在反应中作为转移磷酸等功能基团转移载体的特性,提出了一种对流通代谢物成对处理的方法,并将其按照其在功能基团转移中发挥的作用进行优先级排序,以保证每个反应都有对应的代谢物连接,进而通过程序实现自动化处理。与已有其他方法处理结果的比较表明,这种流通代谢物成对排序处理的方法可以排除掉生物学不可行的途径,从而提高途径搜索的准确性,为利用图论方法进行途径设计和可视化提供更好的支撑。

Abstract:

Graph-theory-based pathway analysis is a commonly used method for pathway searching in genome-scale metabolic networks.However,such searching often results in many pathways biologically infeasible due to the presence of currency metabolites (e.g.H⁺,H₂O,CO₂,ATP etc.).Several methods have been proposed to address the problem but up to now there is no well-recognized methods for processing the currency metabolites.In this study,we proposed a new method based on the function of currency metabolites for transferring of functional groups such as phosphate.We processed most currency metabolites as pairs rather than individual metabolites,and ranked the pairs based on their importance in transferring functional groups,in order to make sure at least one main metabolite link exists for any reaction.The whole process can be done automatically by programming.Comparison with existing approaches indicates that more biologically infeasible pathways were removed by our method and the calculated pathways were more reliable,which may facilitate the graph-theory-based pathway design and visualization.