Macrowine 2021
IVES 9 IVES Conference Series 9 Data fusion approaches for sensory and multimodal chemistry data applied to storage conditions

Data fusion approaches for sensory and multimodal chemistry data applied to storage conditions

Abstract

AIM: The need to combine multimodal data for complex samples is due to the different information captured in each of the techniques (modes). The aim of the study was to provide a critical evaluation of two approaches to fusing multi-modal chemistry and sensory data, namely, multiblock multiple factor analysis (MFA) and concatenation using principal component analysis (PCA).

METHODS: Wines were submitted to sensory analysis using Pivot©Profile (Thuillier et al. 2015) and chemical analysis in four modes: antioxidant measurements (AM), volatile compounds composition (VCC), ultraviolet-visible light (UV-Vis) spectrophotometry (Mafata et al. 2019), and infra-red (IR) spectroscopy. Correspondence analysis (CA), principal component analysis (PCA), and multiple factor analysis (MFA) were used to model data under the data analysis steps involving data cleaning, visualizing, modelling and evaluation (Pagès 2004). Percentage explained variation (%EV) and regression vector (RV) coefficients were used as comparative evaluation parameters between data models (Abdi 2007).

RESULTS: IR spectral data were used as an example of the assessment of the need for data cleaning/pre-processing. Similarities in MFA and high RV coefficients indicated that the raw (unprocessed data) could be used for the data fusion. High RV coefficients and MFA proximity between the antioxidants and UV-Vis measurements indicated an overlap between the type of information contained in the two. The differences between the information captured in each of the five modes can be seen in the different measurements, from the knowledge of the theory/ ontext behind the technique, and statistically. Statistically, the differences are measured and visualised by a lack of overlap (redundancy) in the MFA and its accompanying cluster analysis. 

CONCLUSIONS

The %EV when performing PCA are higher than with MFA, a consequence of fusing big data sets from various modes and not necessarily a direct result of the relationships among the data sets. Therefore, the %EV was ruled out as a reliable measure of the differences in informational value between MFA and PCA fusion strategies. RV coefficients, of which MFA were highest, were the best measurements of the performance of data fusion approaches. MFA demonstrated greater appropriateness as a statistical tool for fusing multi-modal data.

DOI:

Publication date: September 13, 2021

Issue: Macrowine 2021

Type: Article

Authors

Jeanne Brand

South African Grape and Wine Research Institute, Department of Viticulture and Oenology, Stellenbosch University, South Africa,Mpho, MAFATA, South African Grape and Wine Research Institute, Department of Viticulture and Oenology, Stellenbosch University, South Africa  Martin, KIDD, Centre for Statistical Consultation, Stellenbosch University, South Africa Andrei, MEDVEDOVICI, Faculty of Chemistry, University of Bucharest, Romania Astrid, BUICA, South African Grape and Wine Research Institute, Department of Viticulture and Oenology, Stellenbosch University, South Africa

Contact the author

Keywords

data fusion; sensory evaluation; chemical composition; white wines; storage

Citation

Related articles…

Future scenarios for viticultural climatic zoning in Europe

Climate is one of the main conditioning factors of winemaking. In this context, bioclimatic indices are a useful zoning tool, allowing the description of the suitability of a particular region

Spatial variability of grape berry maturation program at the molecular level 

The application of sensors in viticulture is a fast and efficient method to monitor grapevine vegetative, yield and quality parameters and determine their spatial intra-vineyard variability. Molecular analysis at the gene expression level can further contribute to the understanding of the observed variability by elucidating how pathways responsible for different grape quality traits behave in zones diverging for one or the other parameter. The intra-vineyard variability of a Cabernet Sauvignon vineyard was evaluated by a standard Normalized Difference Vegetation Index (NDVI) mapping approach, employing UAV platform, accompanied by detailed ground-truthing (e.g. vegetative, yield, and berry ripening compositional parameters) that was applied in 14 spots in the vineyard. Berries from different spots were additionally investigated by microarray gene expression analysis, performed at five time points from fruit set to full ripening.

Profiling and evaluating wine lees by-products from various yeast strains against grapevine pathogens

Wine lees are the sediment that settles at the bottom of wine barrels, tanks, or bottles during the winemaking process and represent the second most significant by-product of wineries.

Hormonal and associated metabolic changes in susceptible harvest-ripe grapes under asymptomatic and symptomatic Esca disease

Esca complex is a disease affecting grapevine trunks, characterized by the colonization of the wood by xylem-residing fungi (Phaeomoniella chlamydospora, Phaeoacremonium minimum and Fomitiporia mediterranea), and posing significant risks to vineyard longevity since no efficient treatment is available. Despite its prevalence, the mechanisms beyond symptomatic manifestations like interveinal chlorosis and leaf necrosis remain unclear. Preliminary findings indicated a more pronounced metabolic reprogramming in fruits compared to vegetative organs and a putative impact on wine quality by using fruits from symptomatic grapevines.

The antioxidant properties of wine lees extracts in model wine

While the ethanol and tartaric acid contained in wine lees are typically recovered by distilleries, the remaining solid fraction (yeast biomass) is usually disposed of, thus negatively affecting the overall sustainability of the wine industry.