NMR offers tremendous advantages in the analyses of molecular complexity, such as crude bio-fluids, bio-extracts, and intact cells and tissues. Here we introduce recent applications of NMR approaches, as well as next generation sequencing (NGS), for the evaluation of human and environmental health (i.e., maintenance of a homeostatic state) based on metabolic and microbial profiling and data science. We describe useful databases and web tools that are used to support these studies by facilitating the characterization of metabolites from complex NMR spectra. Because the NMR spectra of metabolic mixtures can produce numerical matrix data (e.g., chemical shift versus intensity) with high reproducibility and inter-institution convertibility, advanced data science approaches, such as multivariate analysis and machine learning, are desirable; therefore, we also introduce informatics techniques derived from heterogeneously measured data, such as environmental microbiota, for the extraction of submerged information using data science approaches. We summarize recent studies of microbiomes that are based on these techniques and show that, particularly in human studies, NMR-based metabolic characterization of non-invasive samples, such as feces, can provide a large quantity of beneficial information regarding human health and disease.