Spectrofluorometric analysis combined with machine learning for geographical and varietal authentication, and prediction of phenolic compound concentrations in red wine

Authentication of selected white wines by geographical origin using ICP spectrometric and chemometric analysis

An important aspect of assessing the authenticity of wines is its geographical origin. The aim of the work is to authenticate by geographical origin according to the data of the ICP-spectrometric and chemometric analysis of elemental "images" of wines produced from white grape varieties Chardonnay, Riesling and Muscat grown in four regions of the Krasnodar Territory, Russia. The difference in the contents of Al, Ba, Ca and Rb in wines was found depending on the variety, and Al, Ba, Rb, Fe, Li, Sr - depending on the region of grape growth. Different models of the experimental data processing were used for attribution of the produced varieties of wine to the area of the grape's growth. The criterion for the quality of the constructed models was the accuracy of the attribution of a wine variety to the area of the grape's growth (%). Analysis of the elemental analysis data of 153 wine samples showed that in terms of attribution accuracy, automated neural networks (100 %) are preferred among machine learning methods, followed by support vector machines (98.69 %) and general discriminant analysis (94.77 %). The applied mathematical models enabled the revealing of the cluster structure of the analyzed wine varieties and their attribution to the area of a grape growth with high accuracy. Sr, Li and Fe concentrations in wines were found as the dominating predictors in the constructed models for definition of the geographical origin of wines. The combination of ICP-spectrometric analysis data with the capabilities of statistical modeling of machine learning methods focused on large-dimensional data made it possible to successfully solve small-dimensional problems of the definition of the geographical origin of wines by their elemental composition and variety.

Machine Learning Model Stability for Sub-Regional Classification of Barossa Valley Shiraz Wine Using A-TEEM Spectroscopy

With a view to maintaining the reputation of wine-producing regions among consumers, minimising economic losses caused by wine fraud, and achieving the purpose of data-driven terroir classification, the use of an absorbance-transmission and fluorescence excitation-emission matrix (A-TEEM) technique has shown great potential based on the molecular fingerprinting of a sample. The effects of changes in wine composition due to ageing and the stability of A-TEEM models over time had not been addressed, however, and the classification of wine blends required investigation. Thus, A-TEEM data were combined with an extreme gradient boosting discriminant analysis (XGBDA) algorithm to build classification models based on a range of Shiraz research wines (n = 217) from five Barossa Valley sub-regions over four vintages that had aged in bottle for several years. This spectral fingerprinting and machine learning approach revealed a 100% class prediction accuracy based on cross-validation (CV) model results for vintage year and 98.8% for unknown sample prediction accuracy when splitting the wine samples into training and test sets to obtain the classification models. The modelling and prediction of sub-regional production area showed a class CV prediction accuracy of 99.5% and an unknown sample prediction accuracy of 93.8% when modelling with the split dataset. Inputting a sub-set of the current A-TEEM data into the models generated previously for these Barossa sub-region wines yielded a 100% accurate prediction of vintage year for 2018-2020 wines, 92% accuracy for sub-region for 2018 wines, and 91% accuracy for sub-region using 2021 wine spectral data that were not included in the original modelling. Satisfactory results were also obtained from the modelling and prediction of blended samples for the vintages and sub-regions, which is of significance when considering the practice of wine blending.

Utilization of multiple-dilution fluorescence fingerprint facilitates prediction of chemical attributes in spice extracts

Fluorescence Fingerprint (FF) is a powerful tool for rapid quality assessment of various foods and plant-derived products. However, the conventional utilization of FFs measured at a single dilution level (DL) to substitute chemical analyses is extremely challenging, especially for multicomponent materials like spice extracts because fluorescence intensity and concentration widely differ between components, with complex phenomena like inner filter effects. Here, we proposed a new strategy to use the meta-data comprised of FFs measured at multiple DLs with machine learning to estimate common chemical attributes including total polyphenol and flavonoid contents, and antioxidant abilities. This strategy achieved more consistently satisfactory performance in estimation of all chemical attributes of spice extracts compared to using a single DL. Hence, the workflow employed in this study is expected to serve as an alternative method to quickly evaluate the chemical quality of spice extracts, as well as other plant products and food materials.

Absorbance-Transmittance Excitation Emission Matrix Method for Quantification of Major Cannabinoids and Corresponding Acids: A Rapid Alternative to Chromatography for Rapid Chemotype Discrimination of Cannabis sativa Varieties

Background: Phytocannabinoids naturally occur in the cannabis plant (Cannabis sativa), and Δ9-tetrahydrocannabinol (THC) and cannabidiol (CBD) predominate. There is a need for rapid inexpensive methods to quantify total THC (for statutory definition) and THC-CBD ratio (for classification into three chemotypes). This study explores the capabilities of a spectroscopic technique that combines ultraviolet-visible and fluorescence, absorbance-transmittance excitation emission matrix (A-TEEM). Methods: The A-TEEM technique classifies 49 dry flower extracts into three C. sativa chemotypes, and quantifies the total THC-CBD ratio, using validated gas chromatography (GC)-flame ionization (FID) and High-Performance Liquid Chromatography (HPLC) methods for reference. Multivariate methods used are principal components analysis for a chemotype classification, extreme gradient boost (XGB) discriminant analysis (DA) to classify unknown samples by chemotype, and XGB regression to quantify total THC and CBD content using GC-FID and HPLC data on the same samples. Results: The A-TEEM technique provides robust classification of C. sativa samples, predicting chemotype classification, defined by THC-CBD content, of unknown samples with 100% accuracy. In addition, A-TEEM can quantify total THC and CBD levels relevant to statutory determination, with limit of quantifications (LOQs) of 0.061% (THC) and 0.059% (CBD), and high cross-validation (>0.99) and prediction (>0.99), using a GC-FID method for reference data; and LOQs of 0.026% (THC) and 0.080% (CBD) with high cross-validation (>0.98) and prediction (>0.98), using an HPLC method for reference data. A-TEEM is highly predictive in separately quantifying acid and neutral forms of THC and CBD with HPLC reference data. Conclusions: The A-TEEM technique provides a sensitive method for the qualitative and quantitative characterization of the major cannabinoids in solution, with LOQs comparable with GC-FID and HPLC, and high values of cross-validation and prediction. As a spectroscopic technique, it is rapid, with data acquisition <45 sec per measurement; sample preparation is simple, requiring only solvent extraction. A-TEEM has the sensitivity to resolve and quantify cannabinoids in solution based on their unique spectral characteristics. Discrimination of legal and illegal chemotypes can be rapidly verified using XGB DA, and quantitation of statutory levels of total THC and total CBD comparable with GC-FID and HPLC can be obtained using XBD regression.

Can Rare Earth Elements Be Considered as Markers of the Varietal and Geographical Origin of Wines?

The possibility of establishing the varietal and territorial affiliation of wines by the content of rare earth elements (REE) in them was studied. ICP-OES and ICP-MS with subsequent chemometric processing of the results were applied to determine the elemental image of soils containing negligible REE amounts, grapes grown on these soils, and wine materials of Cabernet Sauvignon, Merlot, and Moldova varieties produced from these grapes. To stabilize and clarify wine materials, the traditional processing of wine materials with various types of bentonite clays (BT) was used, which turned out to be a source of REE in the wine material. Discriminant analysis revealed that the processed wine materials were homogeneous within one denomination and that those of different denominations were heterogeneous with respect to the content of REE. It was found that REE in wine materials were transferred from BT during the processing, and thus they can poorly characterize the geographical origin and varietal affiliation of wines. Analysis of these wine materials according to the intrinsic concentrations of macro- and microelements showed that they formed clusters according to their varietal affiliation. In terms of their influence on the varietal image of wine materials, REE are significantly inferior to macro- and microelements, but they enhance their influence to a certain extent when used together.

Identification of Oak-Barrel and Stainless Steel Tanks with Oak Chips Aged Wines in Ningxia Based on Three-Dimensional Fluorescence Spectroscopy Combined with Chemometrics

With the increased incidence of wine fraud, a fast and reliable method for wine certification has become a necessary prerequisite for the vigorous development of the global wine industry. In this study, a classification strategy based on three-dimensional fluorescence spectroscopy combined with chemometrics was proposed for oak-barrel and stainless steel tanks with oak chips aged wines. Principal component analysis (PCA), partial least squares analysis (PLS-DA), and Fisher discriminant analysis (FDA) were used to distinguish and evaluate the data matrix of the three-dimensional fluorescence spectra of wines. The results showed that FDA was superior to PCA and PLS-DA in classifying oak-barrel and stainless steel tanks with oak chips aged wines. As a general conclusion, three-dimensional fluorescence spectroscopy can provide valuable fingerprint information for the identification of oak-barrel and stainless steel tanks with oak chips aged wines, while the study will provide some theoretical references and standards for the quality control and quality assessment of oak-barrel aged wines.

Authentication of Polish Red Wines Produced from Zweigelt and Rondo Grape Varieties Based on Volatile Compounds Analysis in Combination with Machine Learning Algorithms: Hotrienol as a Marker of the Zweigelt Variety

The aim of this study was to determine volatile compounds in red wines of Zweigelt and Rondo varieties using HS-SPME/GC-MS and to find a marker and/or a classification model for the assessment of varietal authenticity. The wines were produced by using five commercial yeast strains and two types of malolactic fermentation. Sixty-seven volatile compounds were tentatively identified in the test wines; they represented several classes: 9 acids, 24 alcohols, 2 aldehydes, 19 esters, 2 furan compounds, 2 ketones, 1 sulfur compound and 8 terpenes. 3,7-dimethyl-1,5,7-octatrien-3-ol (hotrienol) was found to be a variety marker for Zweigelt wines, since it was detected in all the Zweigelt wines, but was not present in the Rondo wines at all. The relative concentrations of volatiles were used as an input data set, divided into two subsets (training and testing), to the support vector machine (SVM) and k-nearest neighbor (kNN) algorithms. Both machine learning methods yielded models with the highest possible classification accuracy (100%) when the relative concentrations of all the test compounds or alcohols alone were used as input data. An evaluation of the importance value of subsets consisting of six volatile compounds with the highest potential to distinguish between the Zweigelt and Rondo varieties revealed that SVM and kNN yielded the best classification models (F-score of 1, accuracy of 100%) when 3-ethyl-4-methylpentan-1-ol or 3,7-dimethyl-1,5,7-octatrien-3-ol (hotrienol) or subsets containing one or both of them were used. Moreover, the best SVM model (F-score of 1) was built with a subset containing 2-phenylethyl acetate and 3-(methylsulfanyl)propan-1-ol.

Use of Machine Learning with Fused Spectral Data for Prediction of Product Sensory Characteristics: The Case of Grape to Wine

Generations of sensors have been developed for predicting food sensory profiles to circumvent the use of a human sensory panel, but a technology that can rapidly predict a suite of sensory attributes from one spectral measurement remains unavailable. Using spectra from grape extracts, this novel study aimed to address this challenge by exploring the use of a machine learning algorithm, extreme gradient boosting (XGBoost), to predict twenty-two wine sensory attribute scores from five sensory stimuli: aroma, colour, taste, flavour, and mouthfeel. Two datasets were obtained from absorbance-transmission and fluorescence excitation-emission matrix (A-TEEM) spectroscopy with different fusion methods: variable-level data fusion of absorbance and fluorescence spectral fingerprints, and feature-level data fusion of A-TEEM and CIELAB datasets. The results for externally validated models showed slightly better performance using only A-TEEM data, predicting five out of twenty-two wine sensory attributes with R² values above 0.7 and fifteen with R² values above 0.5. Considering the complex biotransformation involved in processing grapes to wine, the ability to predict sensory properties based on underlying chemical composition in this way suggests that the approach could be more broadly applicable to the agri-food sector and other transformed foodstuffs to predict a product's sensory characteristics from raw material spectral attributes.

Chemometrics-assisted excitation-emission matrix fluorescence spectroscopy for rapid identification of commercial reconstituted and sweetened grape juices

As a common fruit juice, grape juice is delicious and nutritious, making it very popular among consumers. However, some illegal manufacturers used shoddy products to lower costs and obtain high profits, which seriously threatens the health and interests of consumers. Hence, this paper proposed excitation-emission matrix (EEM) fluorescence spectroscopy combined with chemometric methods for the rapid identification and classification of commercial grape juices. Spectral characterization of different samples was achieved using the alternating trilinear decomposition (ATLD) algorithm, and chemically meaningful information was obtained and analyzed. Although both reconstituted and sweetened grape juices contain methyl anthranilate (MA) and 2'-aminoacetophenone (o-AAP), the content of MA in sweetened grape juice far exceeds that in reconstituted grape juice, and the MA in sweetened grape juice mainly comes from artificially added grape essence. Then two chemometric methods of hierarchical cluster analysis (HCA) and partial least squares discriminant analysis (PLS-DA) were used for the classification of reconstituted and sweetened grape juices. The results showed that the supervised classification model had a higher correct classification rate (CCR) than the unsupervised classification model, with PLS-DA obtaining 100% CCRs in both training and prediction sets. Therefore, the proposed strategy can be used as a powerful analytical method for the identification and classification of reconstituted and sweetened grape juices and provides a reliable scientific means for ensuring the authenticity and safety of the juice market.

White wine phenolics: current methods of analysis

White wine phenolic analyses are less common in the literature than analyses of red wine phenolics. Analytical techniques for white wine phenolic analyses using spectrophotometric, chromatographic, spectroscopic, and electrochemical methods are reported. The interest of research in this area combined with the advances in technology aimed at the winemaking industry are promoting the establishment of novel approaches for identifying, quantifying, and classifying phenolic compounds in white wine. This review article provides an overview of the current research into white wine phenolics through a critical discussion of the analytical methods employed. © 2022 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Could Collected Chemical Parameters Be Utilized to Build Soft Sensors Capable of Predicting the Provenance, Vintages, and Price Points of New Zealand Pinot Noir Wines Simultaneously?

Soft sensors work as predictive frameworks encapsulating a set of easy-to-collect input data and a machine learning method (ML) to predict highly related variables that are difficult to measure. The machine learning method could provide a prediction of complex unknown relations between the input data and desired output parameters. Recently, soft sensors have been applicable in predicting the prices and vintages of New Zealand Pinot noir wines based on chemical parameters. However, the previous sample size did not adequately represent the diversity of provenances, vintages, and price points across commercially available New Zealand Pinot noir wines. Consequently, a representative sample of 39 commercially available New Zealand Pinot noir wines from diverse provenances, vintages, and price points were selected. Literature has shown that wine phenolic compounds strongly correlated with wine provenances, vintages and price points, which could be used as input data for developing soft sensors. Due to the significance of these phenolic compounds, chemical parameters, including phenolic compounds and pH, were collected using UV-Vis visible spectrophotometry and a pH meter. The soft sensor utilising Naive Bayes (belongs to ML) was designed to predict Pinot noir wines' provenances (regions of origin) based on six chemical parameters with the prediction accuracy of over 75%. Soft sensors based on decision trees (within ML) could predict Pinot noir wines' vintages and price points with prediction accuracies of over 75% based on six chemical parameters. These predictions were based on the same collected six chemical parameters as aforementioned.

Near-infrared spectroscopy analysis of wines through bottles to assess quality traits and provenance

Due to increased fraud rates through counterfeiting and adulteration of quality wines, it is important to develop novel non-destructive techniques to assess wine quality and provenance. Therefore, our research group developed a novel method using near-infrared (NIR) spectroscopy (1596-2396 nm) coupled with machine learning (ML) modeling to assess wine vintages and quality traits based on the intensity of sensory descriptors through the bottle. These were developed using samples from an Australian vineyard for Shirazwines. Models resulted in high accuracy 97% for classification (vintages) and R=0.95 regression (sensory quality traits). The proposed method will allow to assess authenticity and sensory quality traits of any wines in the market without the need to open the bottles, which is rapid, accurate, effective, and convenient. Furthermore, currently, there are low-cost NIR devices available in the market with the required spectral range and sensitivity, which can be affordable for winemakers and retailers that can be used with the ML models proposed here.

Characterization and Classification of Direct and Commercial Strawberry Beverages Using Absorbance–Transmission and Fluorescence Excitation–Emission Matrix Technique

The subject of this study was to characterize the absorption and fluorescence spectra of various types of strawberry beverages and to test the possibility of distinguishing between direct juices and pasteurized commercial products on the basis of their spectral properties. An absorbance and transmission excitation–emission matrix (A-TEEM^TM) technique was used for the acquisition of spectra. The obtained spectra were analyzed using chemometric methods. The principal component analysis (PCA) revealed differences in both the absorption spectra and excitation–emission matrices (EEMs) of two groups of juices. The parallel factor analysis (PARAFAC) enabled the extraction and characterization of excitation and emission profiles and the relative contribution of four fluorescent components of juices, which were related to various groups of polyphenols and nonenzymatic browning products. Partial least squares–discriminant analysis (PLS-DA) models enabled 100% correct class assignment using the absorption spectra in the visible region, unfolded EEMs, and set of emission spectra with excitation at wavelengths of 275, 305, and 365 nm. The analysis of variable importance in projection (VIP) suggested that the polyphenols and nonenzymatic browning products may contribute significantly to the differentiation of commercial and direct juices. The results of the research may contribute to the development of fast methods to test the quality and authenticity of direct and processed strawberry juices.

Multispectral fluorescence sensitivity to acidic and polyphenolic changes in Chardonnay wines - The case study of malolactic fermentation

In this study, stationary and time-resolvedfluorescence signatures, were statistically and chemometrically analyzed among three typologies of Chardonnay wines (A, B and C) with the objectives to evaluate their sensitivity to acidic and polyphenolic changes. For that purpose, a dataset was built using Excitation Emission Matrices of fluorescence (N = 103) decomposed by a Parallel Factor Analysis (PARAFAC), andfluorescence decays (N = 22), mathematically fitted, using the conventional exponential modeling and the phasor plot representation. Wine PARAFAC component C4 coupledwith its phasor plot g and s values enable the description of malolactic fermentation (MLF) occurrence in Chardonnay wines. Such proxies reflect wine concentration modifications in total acidity, malic/lactic and phenol acids.Lower g values among fresh MLF + wines compared to MLF- wines are explained by a quenching effect on wine fluorophores by both organic and phenolic acids.The combination of multispectral fluorescence parametersopens a novel routinely implementable methodology to diagnose fermentative processes.

The application of right-angle fluorescence spectroscopy as a tool to distinguish five autochthonous commercial Greek white wines

White wine is among the most widely consumed alcoholic beverages. Varietal discrimination of wines has received increasing attention. Today's consumers require a sense of authenticity and are deterred by falsehood or misrepresentation in product marketing. However, wine can involve various types of frauds, which directly affects the distribution of wine in national and international markets. Right-angle fluorescence spectroscopy is a simple and rapid analytical technique that in combination with chemometric algorithms, constitutes a novel method for wine authentication. In this study, the stepwise-Linear Discriminant Analysis algorithm was applied in three representative spectral regions related to phenolic compounds for the purpose of distinguishing white wines according to the grape variety. The wavelength at 310 nm attributed to the hydroxycinnamic acids and stilbene provided a higher classification rate (95.5%) than the λex 280 and 295 nm regions (79.8%), suggesting that these compounds are highly related to the botanical origin of samples. The chemometric models were validated utilizing cross-validation and an external validation set to enhance the robustness of the proposed methodology. The above-mentioned methodology constitutes a powerful tool for the varietal discrimination of white wines and can be used in industrial setting. The ultimate goal of this study is to contribute to the efforts towards the authentication of Greek white wine which will eventually support producers and suppliers to remain competitive and simultaneously protect the consumers from fraudulent practices.

A Review of Wine Authentication Using Spectroscopic Approaches in Combination with Chemometrics

In a global context where trading of wines involves considerable economic value, the requirement to guarantee wine authenticity can never be underestimated. With the ever-increasing advancements in analytical platforms, research into spectroscopic methods is thriving as they offer a powerful tool for rapid wine authentication. In particular, spectroscopic techniques have been identified as a user-friendly and economical alternative to traditional analyses involving more complex instrumentation that may not readily be deployable in an industry setting. Chemometrics plays an indispensable role in the interpretation and modelling of spectral data and is frequently used in conjunction with spectroscopy for sample classification. Considering the variety of available techniques under the banner of spectroscopy, this review aims to provide an update on the most popular spectroscopic approaches and chemometric data analysis procedures that are applicable to wine authentication.