Fig. 1: Metabolite medicine.

Standard metabolomics methodologies allow to differentiate between samples, even given an identical genome, as in the displayed example. Yet, the output of such methodologies, namely the levels of diverse metabolites in each sample, requires further analysis to decipher the function of each such compound. Moreover, a standardized database that gathers high volume metabolomics data is currently not available, thus hindering the translation of individual experiments to more inclusive conclusions. The metabolite medicine concept aims to bridge this gap, allowing to obtain further information about each candidate molecule before employing biological model systems. The framework includes computational and bioinformatic analysis, the execution of which requires the establishment of an inclusive database, allowing to connect metabolites to diverse related pathologies. In parallel, the chemical composition of each compound, which dictates possible intrinsic properties, as well as its aggregation propensity and the formation of various assemblies, are to be analyzed using crystallography, microscopy, spectroscopy, and other in vitro methods. Taken together, these data can provide a more accurate representation of the metabolites involved in the studied conditions, thereby setting the basis for more focused and effective biological and clinical experiments to test and verify the metabolite functions. Moreover, this suggested infrastructure aims to aid in the identification process of multiple metabolomics analyses through the integration of structural, chemical and pathological data on each compound under a given investigated condition, thus providing an indication of its possible annotation.