Untargeted metabolic profiling of volatile compounds of grape seed oil using gas chromatography coupled with mass spectrometry

Grape seed oil (GSO) is one of the most nutritious food products worldwide, which has attracted a lot of attention in the food industry [1]. In the present work, untargeted metabolomic profiling was performed using gas chromatography coupled with mass spectrometry (GC-MS) for pan-metabolome profiling of different varieties of GSO obtained from different grape genotypes. We investigated the metabolome profiles of 20 samples from five different GSO genotypes, and putatively identified 175 volatile compounds (VCs), mostly including fatty acid alkyl esters (27%), aldehydes (10%), alcohols (6%), esters (3%), and ketones (2%), which we termed as the pan-metabolome of GSO. Of the 175 molecules, 20 were present in all the GSO samples (core molecules), and 155 were present in ≥10% but 100% of the samples (accessory molecules). The Boruta feature selection method [2] was used to identify core molecules related to the differentiation of GSO samples. Of the 20 core molecules, ten molecules were consistently ranked as more important than shadow features across 70 iterations. The Kruskal–Wallis test was employed to assess the significant differences between the selected core molecules in different groups of samples. Linear discriminant analysis (LDA), partial least squares discriminant analysis (PLS-DA), support vector machine (SVM), and k-nearest neighbor (KNN) methods were used to classify different GSO genotypes using 10 selected core molecules as predictors. Classification accuracies for all models for the calibration and test sets were 1.00. The results revealed that different GSO genotypes contain metabolites that can be used as markers to predict quality and distinguish genotypes. The results of this study showed that metabolomic profiling can efficiently analyze variations in GSO metabolites and detect differences in metabolic pathways among GSO genotypes.