Volatile-Based Prediction of Sauvignon Blanc Quality Gradings with Static Headspace–Gas Chromatography–Ion Mobility Spectrometry (SHS–GC–IMS) and Interpretable Machine Learning Techniques

Abstract

The analytical scope of static headspace–gas chromatography–ion mobility spectrometry (SHS–GC–IMS) was applied to wine aroma analysis for the first time. The method parameters were first fine-tuned to achieve optimal analytical results, before the method stability was demonstrated, in terms of repeatability and reproducibility. Succinct qualitative identification of compounds was also realized, with the identification of several volatiles that have seldom been described previously in Sauvignon Blanc wine, such as methyl acetate, ethyl formate, and amyl acetate. Using the SHS–GC–IMS data in an untargeted approach, computer modeling of large datasets was applied to link aroma chemistry via prediction models to wine sensory quality gradings. Six machine learning models were compared, and artificial neural network (ANN) returned the most promising performance with a prediction accuracy of 95.4%. Despite its inherent complexity, the ANN model offered intriguing insights on the influential volatiles that correlated well with higher and lower sensory gradings. These findings could, in the future, guide winemakers in establishing wine quality, particularly during blending operations prior to bottling.

A Validation Study on the Simultaneous Quantification of Multiple Wine Aroma Compounds with Static Headspace-Gas Chromatography-Ion Mobility Spectrometry (SHS-GC-IMS)

“…Instrumentation and method details were the same as reported in a previous publication. 11 The G.A.S.…”

“…Hence, in the current study special focus was placed on the repeatability and reproducibility of the quantification results, while employing the same data reported in the previous study. 11 A moderately aged (vintage 2018) Sauvignon Blanc wine was analyzed in quadruplicates per day for five days. The repeatability and reproducibility were calculated as intra-and inter-day variations, respectively.…”

“…Being a separation technology that has only been commercialized in recent years, gas chromatography coupled with ion mobility spectrometry (GC-IMS) has rapidly gathered attention of researchers, especially from the field of food and beverage science. 1 Multiple studies have successfully applied GC-IMS for identifying food adulteration [2][3][4] , optimizing food processing and storage conditions [5][6][7] , assigning food origins [8][9] , differentiating food quality gradings [10][11] , and detecting food spoilage [12][13] . The majority of these findings have been summarized in a review article published in 2020.…”

“…This feature also enables nontargeted analyses that overcomes the need of prior peak identification and indifferently processes all peak information, which has been proven immensely helpful in establishing prediction models. [10][11] Nevertheless, it remains that the major focus of research still centers around the qualitative and semiquantitative use of the instrument, whereas quantitative studies using GC-IMS are still quite rare. Compared to gas chromatography-mass spectrometry (GC-MS), for which well-established protocols of quantitative method development are available, consensus is still to be reached even on some fundamental aspects in quantitative GC-IMS, such as the choice of curve-fitting functions [15][16][17] , and the inclusion of internal standards during calibration [16][17][18] .…”

“…Also, the possibility of utilizing multiple ion species (e.g., monomer and dimer) will be discussed for further improving the method accuracy and sensitivity. As a preliminary trial, the method was used to establish calibration models for several volatile compounds commonly found in wine, as have been previously identified in our earlier study 11 . The analytical performance of this method was assessed using conventional ideas and some new tools that specifically tackle issues linked to GC-IMS.…”