Finding Branched Pathways In Metabolic Networks Via Atom Group Tracking

The development of metabolic database makes it possible to explore the metabolic network further and the large volume of metabolic data has created a need for computational tools to identify biologically relevant pathways.Branched metabolic pathways consist of multiple linear pathways and depict a more comprehensive picture of metabolism in comparison to linear pathways.Although a number of methods have been developed to find branched metabolic pathways,they require users to define the tracking atoms.What’s more,to reduce the searching space,these methods only merge the linear pathways that do not contain overlapping conserved atoms in the branched and target compounds to produce branched pathways.This may result in failing to predict the branched pathways that contain overlapping conserved atoms in the branched and target compounds.On the other hand,existing branched finding methods usually rank the resulting pathways by using the characteristics that are biologically related,such as chemical similarity,Gibbs free energy and so on.Few efforts have been made in ranking branched metabolic pathways via specific reactions.To analyze and study the problems and defects of the methods mentioned above,the main contents and contributions of this paper are as follows:(1)In this paper,we present a pathfinding method called BPFinder for finding branched metabolic pathways by atom group tracking.BPFinder enumerates linear metabolic pathways by tracking the movements of atom groups in metabolic networks and merges the linear atom group conserving pathways into branched pathways.Two merging rules based on the structure of conserved atom groups are proposed to accurately merge the branched compounds of linear pathways to identify branched pathways.(2)In this paper,we present a method for ranking resulting pathways.In this method,high-frequency reaction segments of 63 known metabolic networks are extracted,and then the scores to evaluate reaction segments being specific reactions by computing the information entropy and mutual information of the high-frequency reaction segments are used to compute the branched pathway score.Experimental results show that BPFinder is more capable of recovering known branched metabolic pathways as compared to other existing methods,and is able to return biologically relevant branched pathways and discover alternative branched pathways of biochemical interest.The experimental results also show that adding reaction segment score improves the pathway ranking results of BPFinder,and makes it easy for users to find the branched pathways which contain specific reactions.