Characterization of mono-varietal oils from Molise (Gentile di Larino and Oliva nera di Colletorto) through an advanced statistical procedure applied to data acquired by sensory, conventional, gas chromatographic and nuclear magnetic resonance analysis

Monovarietal olive oils from the Molise region (Central Italy), obtained from Gentile di Larino and Oliva nera di Colletorto cultivars, were characterized by sensory and chemical analyses. The influences of the genetic factor (cultivar) on olive oil composition was investigated by gaschromatographic analysis of fatty chains, spectrophotometric analysis of phenols, nuclear magnetic resonance analysis (NMR) as well as through quality parameters such as free acidity, peroxide value and UV spectrophotometric indices. To avoid misinterpretation, we used general linear model analysis (GLM) to adjust the result obtained from the analysis of variance (ANOVA). In this way, the genetic effect was corrected for the crop year effect, which in turn might influence the variable under investigation. The weight of two factors was evaluated by variance component analysis (VCA). Finally, we performed multivariate statistical analysis such as principal component analysis (PCA). All samples were grouped on the basis of their cultivar. The value of free acidity, sn 1,2 diglycerides, saturated fatty chains, total phenols, green and bitter attributes were higher in the olive oils obtained from Gentile di Larino, whereas the levels of unsaturated fatty chains, palmitoleic acid (C16:1), heptadecanoic acid (C17:0), heptadecenoic acid (C17:1) as well as K270 parameter were higher in the oil obtained from Oliva nera di Colletorto.