MS-eNose and chemometrics for the geographic origin discrimination of durum wheat

Headspace solid-phase microextraction coupled with mass spectrometry-based electronic nose (HS-SPME/MS-eNose) in combination with chemometrics was developed as non-targeted analytical method to discriminate durum wheat cultivated in Italy from samples cultivated in other countries. A workflow was implemented, combining two alternative statistical approaches for variable feature reduction in combination with three alternative classifiers, i.e. Partial Least Squares Discriminant Analysis (PLS-DA), Support Vector Machines (SVM), and Artificial Neural Networks (ANN). All models yielded classification accuracy values in prediction, ranging from 88% to 92%. Moreover, ten potential volatiles markers, directly related to the geographical origin, were identified by employing the same extraction protocol coupled with gas chromatography-mass spectrometry (HS-SPME/GC-MS) analysis. The proposed methodology offers a reliable, rapid, and powerful strategy for authenticity assessment, ensuring protection for both the market and consumer.