NMR offers tremendous advantages in the analysis of molecularly complex compounds, such as crude biological extracts, supramacromolecular complexes, geochemical samples, and natural and artificial materials. Herein, I introduce the recent applications of several NMR approaches for evaluating homeostatic plasticity for human and environmental health using a data science approach. Further challenges in addressing the macromolecular complexity include determining the supramolecular structures, composition, and interactions of plant biomass, soil humic substances, aqueous particulate organic matter, and natural and artificial materials. Because solution and solid-state NMR can produce numerical matrix data (e.g., chemical shifts versus intensity) as well as time-domain NMR data (e.g., time versus relaxation decay) with high reproducibility and inter-institution convertibility, further data science approaches, such as multivariate analysis, machine learning, and optimization, are desired. Therefore, I also introduce informatics techniques for data pretreatment prior to solid-state NMR, feature extraction from heterogeneously measured spectroscopic data, and extraction of submerged information using data science approaches.