Prediction of Japanese green tea ranking by gas chromatography/mass spectrometry-based hydrophilic metabolite fingerprinting

Abstract

An innovative technique for green tea's quality determination was developed by means of metabolomics. Gas-chromatography coupled with time-of-flight mass spectrometry and multivariate data analysis was employed to evaluate the quality of green tea. Alteration of green tea varieties and manufacturing processes effects a variation in green tea metabolites, which leads to a classification of the green tea's grade. Therefore, metabolic fingerprinting of green tea samples of different qualities was studied. A set of ranked green tea samples from a Japanese commercial tea contest was analyzed with the aim of creating a reliable quality-prediction model. Several multivariate algorithms were performed. Among those, the partial least-squares projections to latent structures (PLS) analysis with the spectral filtering technique, orthogonal signal correction (OCS), was found to be the most practical approach. In addition, metabolites that play an important role in green tea's grade classification were identified.