I-Corps: Development and application of metabolite-based flavor prediction methods

University Of Florida, Gainesville FL

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is to apply innovative statistical methods of trait prediction to improve flavor in the agricultural, beverage and food industries. In conjunction with a growing societal concern about food safety and quality, there has been an increasing interest in developing food products that have enhanced flavor. Because many of the compounds that determine flavor are also associated with health benefits, increasing their content can also have other positive consequences. The development and commercial application of methods that predict and improve the flavor properties of food products will have a significant impact in an industry that is constantly seeking to increase consumer appeal of their products. In addition, the generation of large databases that relate flavor with the presence of specific compounds will be used to identify new additives for the food industry. These favorable compounds can be potentially commercialized as a flavor enhancing additives. This I-Corps project aims at determining the commercial potential of new approaches developed to predict flavor and consumer appeal of food products using statistical methods, and identify those flavor-related compounds that are most valuable to the food industry. Food texture, flavor and odor are determined by chemical (e.g. volatiles) and physical properties, of which many can be quantified by mass spectrometry. By characterizing these compounds, innovative statistical Bayesian methods can be used to account for their combined effects in consumer appeal. As a consequence, it is possible to predict in advance how new products will perform in the market based on their individual profile. Despite the clear relevance of flavor to consumer preference and many of their well-described health benefits, many industries have been largely unable to incorporate these traits into their products. The approaches developed here will fill that gap. This innovation permits rapid, low-cost and high-throughput identification of cultivars that are most probable to be favored by consumers, without the need to establish costly, low-throughput consumer or expert flavor panels.