Invariant features of metabolic networks: a data analysis application on scaling properties of biochemical pathways

The network metaphor is currently one of the most common general paradigms in biological sciences: this paradigm spans different scales of definition going from gene regulation to protein-protein interaction studies and metabolic regulation networks. Generally, the networks are defined by the nature of the connected elements (nodes) and their relative relations (edges). In this paper we demonstrate how the same biochemical regulation network can assume different shapes in terms of both constituting elements and intervening relations while remaining recognizable as a specific entity. This behaviour can be explained by the general scaling properties of biological networks and points to regulation pathways as emergent features of biochemical systems posited at a different hierarchical level with respect to the intervening metabolites. (C) 2004 Elsevier B.V. All rights reserved.