Enantioselective Amino Acid Profile Improves Metabolomics-based Sensory Prediction of Japanese Sake

Japanese sake reportedly contains several D-amino acids that could contribute to its characteristic flavor. In this study, component profiling of Japanese sake using combined data of gas chromatography/mass spectrometry (GC/MS)-based metabolomics and liquid chromatography/mass spectrometry (LC/MS)-based enantioselective amino acid analysis, and orthogonal partial least squares (OPLS) regression analysis was conducted. GC/MS-based metabolomics is typically the first choice in constructing taste prediction models because it can perform a comprehensive analysis of hydrophilic compounds to obtain informative, explanatory variables. Here, enantioselective amino acid profiles obtained by LC/MS were also used as explanatory variables instead of non-enantioselective amino acid data, which are part of the GC/MS results. The evaluation of prediction models by linearity (R-2), predictability (Q(2)), and root mean square error estimation (RMSEE) indicated that enantioselective amino acid profiles improved metabolomics-based sensory prediction.