Wine metrics: Unraveling the chemical matrix responsible for wine flavour and aroma
Stellenbosch University. Department of Viticulture and Oenology
To deliver on industry-relevant outputs, research in the DVO/IWBT investigates factors, from environmental conditions to human intervention, that impact the biological networks that produce sensory active substances in wine. Very large data sets (popularly known as “ Big Data”) are generated in many of these research programs, including in viticulture, oenology, wine biology, wine chemistry and sensory science. Most of the useful information in these data sets is not found within individual sets, but emerges from the alignmemt of many different data sets to meaningfully link environmental factors and human intervention to the chemical and sensory attributes of a wine. These data sets however are inherently of very different scales and sometimes of a very different nature (from numerical to text), and specific data analysis tools need to be developed and / or implemented to extract meaningful information. The project was designed to improve and support the larger Winetech-funded research endeavour, and not to deliver freestanding outputs.
The project developed new and used existing methods and combined bioinformatics, network analysis, multivariate statistics and lattice analysis. The project also focused on tools for better visualisation of such data. These methods are the primary outputs of the project.
The lead scientist, Dr Dan Jacobson, resigned in 2015, and some objectives which required his specific skills could not be fully completed. Prof. Hugh Patterton was appointed in his place and was able to provide the necessary input and support to deliver on most aspects of the project. The project also benefitted from association with the Department of Information Technology (Profs Bruce Watson and Arina Britz), which helped to develop entirely novel approaches of data mining and text data mining through a lattice-based method derived from work in the fields of Artificial Intelligence (Valente et al. 2018).
Highlights include (non-exhaustive list):
– Providing methodology for metagenomics and RNAseq analysis to better describe vine and wine-related microbial ecosystems and interactions between species (Dr. E Setati and Prof FF Bauer) (Setati et al. 2015; Bagheri et al. 2017).
– Development of a lattice-based analysis of unstructured text data leading to a analysis of the full sensory space of SA Sauvignon blanc and Chenin blanc wines (Valente et al. 2018).
– Training of post-graduate students in application of new bioinformatics tools and multivariate statistical tools. These tools are applied in and essential for nearly all projects.
– Network analysis of transcriptional and chemical data sets (Fairbairn et al. 2017).
– Analysis of ecological interactions within the wine microbial ecosystem (Shekhawat et al., 2016; Bagheri et al. 2017).
– Improved use of spectrophotometric methods for grape and wine analysis (Moore et al., 2015)
The project has successfully supported a large number of Winetech-funded projects and a better exploitation of very large data sets generated by current winetech-funded research projects. Most importantly, it has provided a sound data analytical base for wine science in SA.
Jacobson, D, Monforte, A R and Ferreira, A C S. 2013. Chemonics, our initial foray into systems chemistry. Paper presented at the 13th Scandinavian Symposium on Chemometrics. 17-20 June, Djurhamn, Sweden.
Vendrame, M, Jacobson, D, Du Toit, M and Comi, G. 2013. Preliminary expression analysis of stress related genes in Lactobacillus paracasei: Strains under winemaking conditions. Paper presented at the 5th Congress of European Microbiologists. 21-25 July, Leipzig, Germany.
Jacobson, D and Weighill, D. 2013 Adventures in network modelling: Theory, construction, annotation and visualisation. Paper presented at the 13th Scandinavian Symposium on Chemometrics. 17-20 July, Stockholm, Sweden.
Jacobson, D and Weighill, D. 2013. Network comparison through network topology mining: Cross-omics comparison and new models for evolution. Paper presented at the 13th Scandinavian Symposium on Chemometrics. 17-20 July, Stockholm, Sweden.
Jacobson, D, Young, P, Alexandersson, E and Vivier, M. 2014. Gene sets, strolling through a (random) forest. Paper presented at the 14th Conference on Chemometrics in Analytical Chemistry. 9-13 June 2014. Richmond, Virginia, United States.
Weighill, D and Jacobson, D. 3-way networks: New tools for comparative genomics. Paper presented at the 14th Conference on Chemometrics in Analytical Chemistry. 9-13 June 2014. Richmond, Virginia, United States.
Jacobson, D, Monforte, A R, Ferreira, A C S. 2013. Untangling the chemistry of port wine aging with the use of GC-FID, multivariate statistics, and network reconstruction, Journal of Agricultural and Food Chemistry, Mnth Mar v. 60 (10) (p. 2513-2521)
Alexandersson, E, Jacobson, D, Vivier, M A, Weckwerth, W, Andreasson, E. 2014. Fieldomics – understanding large-scale molecular data from field crops, Frontiers in Plant Science, Mnth Jun v. 5 (p. 286)
Bengtsson, T, Weighill, D, Proux-Wera, E. 2014. Proteomics and transcriptomics of the BABA induced resistance response in potato using a novel functional annotation approach, BMC Genomics, v. 15 (p. 315)
Fairbairn, S, Smit, A, Jacobson, D, Prior, B, Bauer, F F. 2014. Environmental stress and aroma production during wine fermentation, South African Journal of Enology and Viticulture, v. 35 (2) (p. 168-177)
Silva Ferreria, A C, Monforte, A R, Teixeira, C S S, Martins, R, Bauer, F F, Fairbairn, S. 2014. Monitoring alcoholic fermentation: An untargeted approach, Journal of Agricultural and Food Chemistry, Mnth Jun v. 62
Van Wyngaard, E, Brand, J, Jacobson, D, Du Toit, W J. 2014. Sensory interaction between 3-mercaptohexan-1-ol and 2-isobutyl-3-methoxypyrazine in de-aromatised Sauvignon blanc wine, Australian Journal of Grape and Wine Research, v. 20 (p. 178-185)
Moore, J P, Zhang, S-L, Nieuwoudt, H H, Divol, B, Trygg, J, Bauer, F F. 2015. A multivariate approach using ATR mid-infrared spectroscopy to measure the surface mannoproteins and ß-glucans of yeast cell walls during wine fermentations. [Online] Journal of Agricultural and Food Chemistry, Mnth October v. 63 (45) (p. 10054-10063)
Setati, M E, Jacobson, D, Bauer, F F. 2015. Sequence-based analysis of the Vitis vinifera L. cv Cabernet Sauvignon grape must mycobiome. [Online] Frontiers in Microbiology, Mnth November v. 6 (p. 1358)
Chidi, B S, Rossouw, D, Bauer, F F. 2016. Identifying and assessing the impact of wine acid related genes in yeast. [Online] Current Genetics, Mnth February v. 62 (1) (p. 149-164)
Bagheri, B., F.F. Bauer & M.E.Setati. 2015. The diversity and dynamics of indigenous yeast communities in grape must from vineyards employing different agronomic practices and their influence on wine fermentation. South African Journal for Enology and Viticulture 36:243-251.
Rossouw, D., B. Bagheri, M.E. Setati & F.F. Bauer. 2015. Co-Flocculation of yeast species, a new mechanism to govern population dynamics in microbial ecosystems. PLoS ONE 10:e0136249.
Chidi, B.S., D. Rossouw, A. Buica & F.F. Bauer. 2015. Determining the impact of industrial wine yeast strains on organic acid production. South African Journal of Viticulture and Enology 36:316-327.
Rossouw, D. & F.F Bauer. 2016. Exploring the phenotypic space of non- Saccharomyces yeast biodiversity. Food Microbiology 55:32-46.
Shekhawat, K., F.F. Bauer, M.E. Setati. 2016. Impact of oxygenation on the performance of three non-Saccharomyces yeasts in co-fermentation with Saccharomyces cerevisiae. Applied Microbiology and Biotechnology. 101:2479-2489.
Bagheri, B., F.F. Bauer & M.E. Setati. 2017. The Impact of Saccharomyces cerevisiae on a Wine Yeast Consortium in Natural and Inoculated Fermentations. Frontiers in Microbiology 8:1988
Fairbairn S, McKinnon A, Musarurwa H, Ferreira AC, Bauer FF. 2017. The impact of single amino acids on growth and volatile aroma production by Saccharomyces cerevisiae strains. Frontiers in Microbiology 19: 2554.
Valente, C.C., F.F. Bauer, F. Venter, B. Watson & H.H. Nieuwoudt. 2018. Modelling the sensory space of varietal wines: Mining of large, unstructured text data and visualisation of style patterns. Scientific Reports. 8: 4987