Sensory omics combined with mathematical modeling for integrated analysis of retronasal Muscat flavor in table grapes.

This study focused on analyzing the aroma formation mechanism of retronasal muscat flavor in table grapes. The sensory characteristics and fragrance components of table grape juice with different intensities of Muscat were investigated using GC-Quadrupole-MS, quantitative descriptive analysis and three-alternate forced choice. Free monoterpenoids were the main contributors to the retronasal Muscat flavor. The contribution of Muscat compounds to this flavor was quantified by Stevens coefficient, the most and the least sensitive compounds to concentration changes were citronellol and linalool, respectively. To predict the Muscat flavor intensity by mathematical modeling, established a model between Muscat flavor intensity and monoterpenoids concentration, and an optimal partial least squares regression model with a linear relationship between natural logarithms was obtained. These findings provide reference for understanding the formation mechanism of specific aromas in fruits and provide a basis for the development and quality control of processed products such as Muscat flavor grape juice.