Model Complexity has a Significant Effect on the Numerical Value and Interpretation of Metabolic Sensitivity Coefficients

Abstract

Logarithmic sensitivity coefficients were promulgated for the analysis of metabolic regulation about 20 years ago. Interest in their use has risen significantly since their introduction. However, no comprehensive evaluation of the utility of these metabolic sensitivity coefficients is available for realistic metabolic models. In this study, logarithmic sensitivity coefficients calculated from three progressively simpler metabolic models of red blood cell metabolism were compared. Two simpler models were obtained from a comprehensive red cell model by first removing volume regulation, and second by removing two pathways. The comparisons of sensitivity coefficients obtained for these three models showed that model complexity has significant effects on the numerical values and interpretation of the metabolic sensitivity coefficients. Additionally, it was found that the physiochemical volume regulatory mechanism, namely electroneutrality and osmotic balances, play an important role in the red cell metabolic flux control. In general, there is no proportionate relationship between sensitivity coefficients calculated from the three different red cell metabolic models or other simple red cell models reported in the early literature. Some sensitivity coefficients determined by different models even have opposite signs. Thus, analysis of incomplete metabolic models can be seriously misleading and produce inappropriate indicators of the characteristics of a full model for the same metabolic network.