Accurate Quantification of Quality Compounds and Varietal Classification from Grape Extracts using the Absorbance-Transmittance Fluorescence Excitation Emission Matrix (A-TEEM) Method and Machine Learning

Abstract

Rapid and accurate quantification of grape berry phenolics, anthocyanins and tannins, and identification of grape varieties are both important for effective quality control of harvesting and initial processing for wine making. Current reference technologies including High Performance Liquid Chromatography (HPLC) can be rate limiting and too complex and expensive for effective field operations. Secondary calibrated techniques including UV-VIS and Near and Mid Infrared spectroscopy are insensitive to specific quality compounds and unable to make accurate varietal assignments. In this paper we analyze robotically prepared grape extracts from several key varieties (n=Calibration/p=Prediction samples) including Cabernet sauvignon (64/10), Grenache (16/4), Malbec (14/4), Merlot (56/10), Petit syrah (52/10), Pinot noir (54/8), Syrah (20/2), Terlodego (14/2) and Zinfandel (62/12). Key phenolic and anthocyanin parameters measured by HPLC included Catechin, Epicatechin, Quercetin Glycosides, Malvidin 3-glucoside, Total Anthocyanins and Polymeric Tannins. Separate samples diluted 150 fold in 50% EtOH pH 2 were analyzed in parallel using the A-TEEM method following Multiblock Data Fusion of the absorbance and unfolded EEM data. A-TEEM chemical data were calibrated (n=390) using Extreme Gradient Boost (XGB) Regression and evaluated based on the Root Mean Square Error of the Prediction (RMSEP), the Relative Error of Prediction (REP%) and Coefficient of Variation (R2P) of the Prediction data (n=62). The regression results yielded an average REP% value of 6.0±2.4% and R2P of 0.941±0.024. While we consider the REP% values to be in the acceptable range at <10% we acknowledge that both the grape extraction method repeatability and HPLC reference method repeatability likely contributed the major sources of variation; e.g., A-TEEM sample REP%=1.31 for Polymeric Tannins. Varietal classification was analyzed using XGB discrimination analysis of the Multiblock data and evaluated based on the Prediction data. The classification results yielded 100% True Positive and True negative results for the Prediction Data for all varieties. We conclude that the A-TEEM method requires a minimum of sample preparation and rapid acquisition times (<1 min) and can serve as an accurate secondary method for both grape composition and varietal identification. Importantly, the application of the regression and classification models can be effectively automated for operators.