Abstract: In this study, six cherry wines from different producers were evaluated by sensory evaluation and 33 characteristic taste compounds were identified in these samples using high performance liquid chromatography (HPLC). Principal component analysis (PCA) and agglomerative hierarchical clustering (AHC) were used to attempt to discriminate the wines. The results demonstrated that wines from the same geographical areas were clustered into one group and samples from different geographical origins were clearly discriminated. Correlation analysis between characteristic taste compounds and sensory attributes using partial least squares regression (PLSR) indicated a significant positive association between the sweet taste and glucose, sucrose and fructose; between the astringent taste and serine and phenylalanine; and between the bitter taste and caffeic acid, glutamic acid, serine, glycine, threonine, alanine, valine, methionine, phenylalanine, isoleucine, leucine and lysine.

Key words: cherry wines, high performance liquid chromatography, taste compounds, partial least squares regression