A UHPLC-HRMS based metabolomics and chemoinformatics approach to chemically distinguish 'super foods' from a variety of plant-based foods.

The aim of this study was to investigate if the declared benefits associated with superfoods are related to a specific molecular composition. For this purpose, untargeted metabolomics and molecular networking were used to obtain an overview of all features, focusing on compounds with anti-inflammatory, antioxidant or antimicrobial properties. 565 plant-based food samples were analyzed using UHPLC-HRMS and advanced data analysis tools. The molecular networking of the whole dataset allowed identification of a greater diversity of molecules, in particular, prenol lipids, isoflavonoids and isoquinolines in superfoods, when compared with non-superfood species belonging to the same botanical family. Furthermore, in silico tools were used to expand our chemical knowledge of compounds observed in superfood samples.