An Informatics Approach to Flavonoid Database Development

The association between consumption of fruits and vegetables and reduced chronic disease risk may be due in part to their flavonoid content. A food flavonoid database describes six flavonoid classes (anthocyanins, flavans, flavanones, flavones, flavonols, and isoflavonoids) in fruits, vegetables and beverages. Developing a flavonoid database requires the use of informatics and involves nine steps. The steps include identifying all foods likely to contain flavonoids, collecting and organizing sources of existing analytical data, and evaluating the quality of each value for each compound from the various sources. The methodology for evaluating each set of analytic data uses five criteria established by the United States Department of Agricultureʹs (USDA) Food Composition and Nutrient Data Laboratories: (1) analytical method, (2) analytical quality control, (3) number of samples, (4) sample handling, and (5) sampling plan. Data ratings of acceptable quality for each component and food are aggregated. Quality indices for each food value are further reviewed. Means and other statistics are calculated. The data are harmonized and standardized by assigning appropriate USDA food codes, botanical names, chemical names, etc., and the new database is finalized. Finally, the data are used to set priorities for new analyses, methods, and applications in intake and health outcome studies. The provisional flavonoid database will highlight gaps in flavonoid food analyses and needs for additional analytical work.