(650a) ShReD: A Novel Metric for Determining Reciprocal Interactions Between Biochemical Network Components

Modularity analysis has gained increasing attention as a tool for the systematic reduction of highly complex biochemical networks. Modularization of complex reaction networks into simpler structural units offers a number of scientific and practical benefits. For example, modularization provides fundamental insights into the organization of biochemical pathways and simplifies the development of whole cell kinetic models. Identifying robust (e.g. evolutionarily conserved) modules could also facilitate the construction of functional building block libraries to aid in the design of synthetic organisms.

In this paper, we introduce a novel metric, termed SHortest Retroactive Distance (SHRED), for quantifying the reciprocal interactions between biochemical network components. These interactions arise not only from reversibility of reactions, but also from feedback loops and other regulatory interactions between metabolites and enzymes. Utilizing graph-based path analysis algorithms, SHRED considers both immediate neighbor (local) and distant pair (global) interactions. We combine the SHRED metric with a top-down partition algorithm to identify hierarchical relationships between modules in a biochemical network. We investigate the effect of network size and activity state on modularity using models of the EGF-MAPK signaling cascade and hepatocyte metabolic network, respectively. The signaling models were adopted from published papers, whereas the hepatocyte model, including regulatory interactions between enzymes and metabolites, were constructed based on a thorough review of the literature. Results of the signaling cascade analysis suggest that network hierarchy may not be apparent unless the model includes fine details. Comparison of the fed and fasted state hepatocyte models indicate that biochemical network hierarchy is also influenced by the activity state. On-going work further examines the role of enzyme activity, as measured by reaction flux, in determining modularity.