Quantifying chemical correlations between fruits and processed fruit products: A non-targeted analysis approach.

Juices and beverages are produced by industry for long-distance distribution and shelf-stability, providing valuable nutrients. However, their nutritional value is often underestimated due to insufficient analytical methods. We have employed non-targeted analysis through a standardized analytical protocol, taking advantage of Data Independent Acquisition (DIA) technique and a novel Chromatographic Retention Behavior (CRB) data deconvolution algorithm. After analyzing 9 fruits and their products, correlations between fruits and their juices are accurately digitalized by similarities of their LC-MS fingerprints. We also specify non-targeted molecules primarily associate with nutrient loss in these analyzed juice products, including nitrogenous nutrients, flavonoids, glycosides, and vitamins. Moreover, we unveiled previously unreported fruit-characteristic metabolites, of which reconstituted-from-concentrate (RFC) juices contain over 40% of the content found in their fresh counterparts. Conclusively, our method establishes a quantitative benchmark for rational selection of RFC juices to substitute natural fruits.

Biogenic Solution Map Based on the Definition of the Metabolic Correlation Distance between 4-Dimensional Fingerprints.

Accurate discrimination and classification of unknown species are the basis to predict its characteristics or applications to make correct decisions. However, for biogenic solutions that are ubiquitous in nature and our daily lives, direct determination of their similarities and disparities by their molecular compositions remains a scientific challenge. Here, we explore a standard and visualizable ontology, termed "biogenic solution map", that organizes multifarious classes of biogenic solutions into a map of hierarchical structures. To build the map, a novel 4-dimensional (4D) fingerprinting method based on data-independent acquisition data sets of untargeted metabolomics is developed, enabling accurate characterization of complex biogenic solutions. A generic parameter of metabolic correlation distance, calculated based on averaged similarities between 4D fingerprints of sample groups, is able to define "species", "genus", and "family" of each solution in the map. With the help of the "biogenic solution map", species of unknown biogenic solutions can be explicitly defined. Simultaneously, intrinsic correlations and subtle variations among biogenic solutions in the map are accurately illustrated. Moreover, it is worth mentioning that samples of the same analyte but prepared by alternative protocols may have significantly different metabolic compositions and could be classified into different "genera".