Over the past decade, the advances in human brain magnetic resonance imaging (MRI) have significantly improved our ability to gain insightful information about the structure and function of the brain. One of the MRI imaging modalities that still awaits more comprehensive data mining is magnetic resonance spectroscopy (MRS). MRS provides information on the functional status of the brain tissue and can detect metabolic abnormalities that precede structural changes. The chemical specificity of proton MRS ((1)H-MRS) allows detection of several biomarkers that are specific for neurons (N-acetyl aspartate, NAA) and astrocytes (myoinositol (mI) and choline (Cho)), the two most abundant cell types present in the brain tissue. However, apart from a dozen metabolites, current methodologies utilized for MRS analysis do not allow further biomarker discoveries. Herein, we introduce a bioinformatics approach to MRS data processing and discuss possible discoveries that such approach may provide. Specifically, we describe the methodology for neural stem/progenitor cell (NPC) detection in vitro and in vivo, utilizing metabolomic profiling and singular value decomposition analyses.