Tendency Stoichiometric Modeling of Metabolic Pathways

Cells are complex systems whose function results from the concerted actions of networks of biochemical reactions. Important types of biochemical networks involve interactions between proteins that exist as both structural and functional building blocks. Quantitative description of the concentration changes due to active pathways in a protein interaction network is a challenging task. Due to the high complexity of those systems and the partial understanding that stems from the limited experimental data, emerges the need for a modeling methodology. So far only the composition changes of pairs of proteins were inter-related to develop some understanding of the active reaction stoichiometries (A. Jayaraman et al., 2005). In this research work a novel approach is introduced for the systematic analysis of the stoichiometric inter-relationship of all proteins measured, not just two at a time. The proposed approach involves the singular value decomposition of the concentration change over time data in a batch culture. The right hand side singular vectors that correspond to the leading singular values define the abstract stoichiometric space of the active pathways. The comparisons between the abstract stoichiometric spaces, obtained under different experimental conditions, reveal which pathways are active. Tendency modeling has been previously applied to simpler reactions (C. Filippi et al., 1986) and here it is tested and expanded for the protein expression data describing the long-term (several days) inflammatory response of liver cells stimulated by combinations of cytokines.