Beginning as the science of life, biology has been transformed and expanded throughout the years to give rise to fields such as ecology, molecular biology, biophysics, and biochemistry. The distinction among these disciplines is perhaps artificial as the study of any particular biological process or system often spans multiple fields. The research on energy metabolism to follow involves three fields: bioinformatics, bioengineering, and systems biology. The informatics aspect deals with the mining, organization, and management of genomic, proteomic, microarray and biochemical data for reconstructing metabolic networks. The engineering aspect employs the application of mathematics and scientific fundamentals to construct constraint-based models for examination. The systems biology paradigm allows the analysis and interpretation of the models and data from a holistic perspective. Specifically, the findings of this work center on energy metabolism as understood by resource allocation and roles or mitochondria in mammalian cells in health and diseased states. The constraint-based framework provides the facilities for biological discovery using tools from the three mentioned disciplines. This modeling approach describes a biological system as a network of components whose behaviors can be predicted upon the application of constraints. Energy metabolism of four systems - mitochondria, cardiomyocytes, hepatocytes, and fibroblasts - is investigated here. The application of linear constraints on metabolite mass, reaction reversibility and substrate utilization allows an exploration of network capability and topology. The capability and topology of the mitochondrial metabolic network are evaluated based on the theoretical energetic yield of substrates, the number of alternate pathways satisfying the same metabolic objective, and the effects of diabetic and ischemic conditions on feasible steady states. The second part of this dissertation involves the use of stable isotopes to uncover physiological steady states assumed by the cell. Nonlinear constraints are used to balance isotopomers and select for a set of flux distributions that match GC-MS data. Results from studies with cardiomyocytes and hepatocytes elucidate the paths undertaken by substrates as well as effects of media composition on intracellular flux distributions. Most significantly, results from tissue culture study identify complex II as the deficient complex in fibroblasts derived from a patient affected with Leigh's disease