Predicting networking couples for metabolic pathways of Arabidopsis

Abstract

Given an enzyme-compound couple, how can we identify whether it belongs to a networking couple or non-networking couple? This is very important for investigating the metabolic pathways. To address this problem, a novel approach was developed that is featured by using the knowledge of gene ontology (GO), chemical functional group (FunG), and pseudo amino acid composition (PseAA) to represent the samples of enzyme-compound couples. Two basic identifiers were formulated: one is called "GOFunG", and the other, "PseAA-FunG". The prediction was operated by fusing these two basic identifiers into one. As a showcase, the metabolic pathways were investigated for Arabidopsis thaliana, a small flowering plant widely used as a model organism for studies of the cellular and molecular biology of flowering plants. The average overall success rate via the jackknife cross-validation tests for the 72 metabolic pathways in the Arabidopsis system was over 95%, suggesting that the current approach might become a very useful tool for studying metabolic pathways and many other problems in the cellular networking related areas.