The outstanding information processing capacity of the brain relies on numerous molecular mechanisms. It is an extremely energy-expensive function, which involves specialization and collaboration of different cell types. A common framework to study the complex interplay of energy demand-supply in the brain is the Neuron-Glia-Vasculature (NGV) unit. Its numerous models are tailored to answer specific questions, but systematic understanding incorporating multiple aspects within one model has yet to be achieved. It is the main goal I addressed with this thesis, which consists of four parts: 1. Multifunctional Energy Sources in Brain Aging (review); 2. A Standardized Brain Molecular Atlas: A Resource for Systems Modeling and Simulation; 3. Brain Aging Mechanisms in Dynamical Model of Metabolism and Physiology; 4. Metabolic Fitness Landscape of the Neocortical Microcircuitry. In the first part, I reassess seemingly the most basic initial steps of brain energy production. To do so, I review metabolic and regulatory functions of the key brain energy substrates and highlight their involvement in aging and neurodegenerative diseases. Further in this review, I bring together computational models and experimental data to identify which knowledge gaps need to be addressed with either computational or experimental techniques. In the second part, I perform a meta-analysis of publicly available data from a large variety of sources to obtain a consistent estimate of protein and metabolite concentrations in the brain cells. The resulting integrated data is a ready-to-use resource for systems biology modeling and simulation. In the third part, I reconstruct and simulate a biophysically realistic dynamic NGV unit metabolism model driven by and constraining the neuronal electrophysiological behavior. I validate the model, estimate neuronal and astrocytic energy consumption, and perform sensitivity analysis. Next, I simulate metabolism and neuronal firing in aging by incorporating single-cell RNAseq data. I perform a number of computational experiments to study the mechanisms that affect metabolism and neuronal firing in aging. In the fourth part I integrate the model described above into the detailed model of neocortical microcircuitry to analyze the metabolism of different layers and types of neurons. By doing so I estimate the metabolic fitness landscape of the neocortical column. The concept of metabolic fitness landscape introduces a systematic measure to look at the complex interplay of energy demand-supply under multiple biological constraints, and can potentially be considered as an objective function for future brain evolution studies. To sum up, the main outcomes of my work are the data-driven biologically detailed dynamic model of the NGV metabolism coupled to neuronal firing and its applications to explore brain aging and metabolic fitness landscape. This work enables both the study of fundamental theoretical questions of brain energy metabolism and neurometabolic interplay, as well as future application for simulations of the molecular mechanisms of neurological disorders.