P-40 - Harnessing Co-Fermentation: Exploring Yeast Interactions for Enhanced Fermented Beverage Production

While modern brewing and distilling operations typically operate fermentations with a single organism (monoculture), historical applications often relied on multiple organisms simultaneously. How co-fermentation contributes to fermentation kinetics and the production of secondary metabolites has not yet been fully described, limiting the ability for researchers to use predictive models to tailor fermentation products. In this study, methods are proposed to quantify and model these parameters between two microorganisms in a controlled fermentation assay. Whiskey wash was donated by a collaborating local craft distillery, as was industrial yeast. Fermentation was conducted using the American Society of Brewing Chemists (ASBC) Yeast-14 method to facilitate multiple replications with differing yeast ratios. The impacts of varying yeast ratio on the following parameters were assessed: fermentation kinetics, yeast flocculation, and volatile production. To explore modeling of fermentation parameters, five models (linear, exponential, square root, gaussian, and step) were fit to the volatile data and statistically compared. The fermentation density attenuation showed a linear (r = 0.928) relationship between yeast ratio and final alcohol concentration. Significance of mean differences between volatile metabolites were identified by one-way ANOVA and post-hoc t-tests. Differences were identified for 16 compounds, with 6 of these found at concentrations above published detection thresholds. Only one of these, 4-ethylphenol, crossed the detection threshold due to yeast ratio. Concentrations were summed across secondary metabolite functional groups to find trends. The findings of this paper suggest that different chemical groups could be modeled with respect to yeast ratio and that the behavior of different groups followed unique trends. While some exhibited linear relationships between yeast ratios (alcohols, esters, benzenes, terpenes, alkanes), others were non-linear (acids, aldehydes, ketones, phenols, others). Modelling volatile compounds in this way simplified analysis and is indicative of general flavor trends. These findings will allow researchers and industry partners to characterize and better predict how varying ratios of organisms produce volatile metabolites which contribute to flavor.