Modeling the sake brewing characteristics of rice from brown rice metabolites

Sake, soy sauce, miso (Japanese bean paste), and beer are made from grains. The characteristics of the grain significantly affect the quality of the final product. Many studies have been performed to evaluate the sake-brewing characteristics of rice. However, current rice analysis methods are time and labor intensive and require large samples. We developed a novel method for predicting the brewing characteristics of sake rice using <1 g of sample. Brown rice extracts were analyzed by ultrahigh-performance liquid chromatography quadrupole time-of-flight mass spectrometry, and mass chromatogram data were used as explanatory variables. The objective variables were the physical and chemical properties of the rice, the enzymatic activity of the rice-koji, the fermentation properties of the sake mash, the standard analytical values of the sake, and the flavor component concentrations in the sake. Prediction models were developed using the orthogonal projections to latent structures method. The prediction performances of the models were verified, and 32 out of the 54 objective variables were used in well-performing models. In conclusion, we developed a method for predicting the rice properties and brewing characteristics from results acquired by analyzing <1 g of brown rice. The method is a powerful tool for breeding new sake rice cultivars for good brewing characteristics in early generations and will improve our understanding of fluctuations in the brewing characteristics of sake rice before each sake brewing season starts.