Advanced Analytical Sensory Correlation – Towards a Better Molecular Understanding of Coffee Flavor

The present study aimed at better understanding the link between the sensory profile of espresso coffees and their molecular aroma and taste composition. Twelve coffee blends were assessed by instrumental analysis and sensory profiling. The results were statistically correlated using a knowledge-based standardization and normalization of both datasets that selectively extracts differences in the quality of samples, while reducing the impact of variations on the overall intensity of coffees. Several of the 42 aroma and 12 taste compounds analyzed for this study exhibited a good correlation with specific sensory descriptors and may be used as chemical markers. In addition, a robust mathematical model could be developed that predicts the sensory profile of espresso coffees from instrumental data. Such model is a very useful tool for future development of coffee blends with tailored flavor profiles. The results represent significant progress in correlating sensory with instrumental data, exemplified on one of the most complex aromas, i.e., coffee.