 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

Impact of the ‘Pinot’-family on early ripening in cool climate viticulture varieties

‘Pinot Precoce Noir’ (PPN) is an early ripening clone of ‘Pinot Noir’ (PN). The phenological differentiation is visible by an about two weeks earlier onset of veraison. It was found that the early veraison locus Ver1 on chromosome 16, previously identified in ‘Calardis Musqué’, originated from PPN. A highly correlated SSR marker, namely GF16-Ver1, was developed and tested for its ability to molecularly differentiate between PPN and PN as well as its potential to trace individual descendants.

Mining microbiome data to identify antagonists of grapevine downy mildew (Plasmopara viticola)

Vineyards are home to a myriad of microorganisms that interact with each other and with the vines. Some microorganisms are plant pathogens, such as the oomycete Plasmopara viticola, causing grapevine downy mildew. Others have a positive effect on vine health, such as disease biocontrol agents. These beneficial plant-microbe and microbe-microbe interactions have gained more attention in recent years because they could represent an alternative to the use of fungicides in viticulture.

Upscaling the integrated terroir zoning through digital soil mapping: a case study in the Designation of Origin Campo de Borja

homogeneous zones by intersecting several partial zonings of major factors that influence vineyard growth. Each of them follows specific process from their corresponding disciplines. Soil zoning specifically refers to a Soil Resource Inventory map that has traditionally been generated by conventional soil mapping methods. These methods have shortcomings in reaching fine cartographic and categorical details and involve significant expenses, which undermines their applicability. A new framework named Digital Soil Mapping has introduced quantitative models by statistical techniques to establish soil-landscape relationships and is able to provide intensive scale cartography.

In the present study, a microzoning at 1:10.000 scale is generated from an initial zoning, where the conventional soil map with polytaxic map units is replaced by a new one from digital techniques that disaggregates them. The comparison between the zonings considers a quantitative evaluation of capability for each Homogeneous Terroir Unit by means of the Viticultural Quality Index and its categorization based on its distribution by map. The spatial intersection of both maps gives rise to a confusion matrix in which the flows of class variations after the substitution are assessed.

The results show a five-fold increase in the number of Homogeneous Terroir Units identified and a larger differentiation among them, evidenced by a wider range in the capability index distribution. Both elements are accompanied by an increase in the detection of areas of higher potential within previously undervalued uniform zones.These features are a direct effect of the improvements brought by Digital Soil Mapping techniques and would verify the advantages of their implementation in the Integrated Terroir zoning. Eventually, such new highly detailed terroir units would benefit precision viticulture and sustainable management practices.

Novel protocols for variable rate vineyard management

The advent of precision viticulture (PV) has allowed to address problems related to spatial and temporal variability at the within-field scale. Nowadays, several remote and proximal sensing solutions allow description of the existing variability at different temporal and ground resolution through extremely robust soil, vigor, yield, and grape quality maps. In parallel, numerous studies have described grapevine performances within the homogeneous zones and identified soil as main driver of variability. There is a broad consensus that different vigor zones within the same plot may show differential canopy growth, yield and fruit composition, depicting diverse enological potentials and cultural needs.

Applications pratiques du zonage vitivinicole

Le zonage vitivinicole présente toute une série d’applications pratiques. Son importance est en train d’augmenter, soit en fonction des moyens techniques chaque fois plus performants, qui rendent possible le développement des zonages de plus en plus intégrées, consistants et utiles, soit en fonction d’un marché de plus en plus mondialisé. L’article situe la contribution du zonage au niveau de la production vitivinicole et du développement du territoire.