Rapid geographical origin identification and functional compound content prediction of Chrysanthemum ("Gongju") using excitation-emission matrix fluorescence spectroscopy coupled with chemometrics

Abstract

Gongju (Chrysanthemum morifolium Ramat.), an edible flower widely consumed in East Asia, shows quality variations depending on its geographical origins and functional compound contents. This study proposes a "two-in-one" strategy that utilizes excitation-emission matrix (EEM) fluorescence spectroscopy combined with chemometrics to simultaneously identify the geographical origin of Gongju and quantify its functional compounds. Briefly, 660 Gongju samples from 11 regions were analyzed. Their spectral characteristics were preliminarily characterized via the alternating trilinear decomposition (ATLD) algorithm, which showed valuable chemical information. Partial least squares discriminant analysis with Savitzky-Golay (SG) smoothing achieved 98.5 % accuracy for geographical authentication. Ultra-performance liquid chromatography triple-quadrupole mass spectrometry (UPLC-TQ-MS/MS) provided reference values for 12 functional compounds. Random forest regression with SG preprocessing emerged as the optimal quantification model, providing relative prediction deviation values > 2.63 for all compounds. This study establishes an effective framework for comprehensively assessing Gongju quality and holds promise for broader food authentication applications.

Keywords: Chemometrics; Excitation–emission matrix fluorescence; Geographical authentication; Gongju; Quantitative analysis.