Abstract: In order to evaluate the potential of instrumental analysis to replace sensory analysis for food aroma evaluation by combining instrumental and sensory analysis results, the aroma characteristics of nine cherry cultivars were analyzed by gas chromatography-mass spectrometry (GC-MS) and sensory evaluation. The Pearson correlation coefficients between the GC-MS and sensory data were calculated to build an aroma network. According to the Pearson correlation coefficient between the sensory evaluation and GC-MS results, partial least square regression (PLSR) was used to verify the reliability of the established aroma network. The results of Pearson correlation analysis and the aroma network diagram showed that the sensory attributes of cherry aroma were significantly related to a variety of aroma substances identified by GC-MS analysis. Similarly, PLSR analysis showed that the citrus-like, lemon-like, almond-like, rosy and wine-like cherry aroma attributes could be well explained by the results of GC-MS analysis. It was proved that the aroma network was feasible to combine the sensory evaluation results of aroma with the results of GC-MS. It is considered that GC-MS analysis, an objective and easy-to-standardize aroma evaluation method, has the potential to replace sensory analysis in the study of fruit aroma.

Key words: cherry; aroma substances; gas chromatography-mass spectrometry; sensory evaluation; aroma network; partial least squares regression