Metabolic network-based interpretation of gene expression data elucidates human cellular metabolism

Research into human metabolism is expanding rapidly due to the emergence of metabolism as a key factor in common diseases. Mathematical modeling of human cellular metabolism has traditionally been performed via kinetic approaches whose applicability for large-scale systems is limited by lack of kinetic constants data. An alternative computational approach bypassing this hurdle called constraint-based modeling (CBM) serves to analyze the function of large-scale metabolic networks by solely relying on simple physical-chemical constraints. However, while extensive research has been performed on constraint-based modeling of microbial metabolism, large-scale modeling of human metabolism is still in its infancy. Utilizing constraint-based modeling to model human cellular metabolism is significantly more complicated than modeling microbial metabolism as in multi-cellular organisms the metabolic behavior varies across cell-types and tissues. It is further complicated due to lack of data on cell type- and tissue-specific metabolite uptake from the surrounding micro-environments and tissue-specific metabolic objective functions. To overcome these problems, several studies suggested CBM methods that integrate metabolic networks with gene expression data that is easily measurable under various conditions. This paper, reviews three CBM methods for analyzing and predicting metabolic states based on gene expression data. These methods lay the foundation for studying normal and abnormal human cellular metabolism in tissue-specific manner.