On a deeper comprehension of EFMs biological significance by using clustering techniques

Abstract

Background: The study of structural properties of a metabolic network can be approached by analyzing its Elementary Flux Modes (EFMs). Even in cases in which this set can be fully computed, its large cardinality makes it difficult to interpret the information contained in it. Results: This paper presents a proposal to improve the study of structural properties of a network by using clustering techniques in its set of EFMs. It is shown how some properties of this set such as their length distribution or the reaction participation can be better elucidated after a clustering process and how it allows for a better comprehension of the possible behaviours of the network. Conclusions: Our clustering approach can help in the extraction of relevant biological significance from the set of EFMs and can be applied to different problems related to the structural properties of the network under study