Multivariate analysis to discriminate yeast strains with technological applications in table olive processing.

This survey uses a multivariate classification analysis to discriminate yeast strains with interesting biochemical activities for the processing of table olives among a collection of 32 isolates belonging to 16 different yeast species. Lipase, esterase and β-glucosidase activities (desirable characteristics) were quantitatively evaluated in both extracellular and cellular fractions for all isolates in different types of culture media. The study of the quantitative data by cluster and principal component analyses led to the identification of several Wickerhamomyces anomalus, Candida boidinii and Candida diddensiae isolates with promising characteristics (the best global activity levels), clearly differentiated from the rest of the yeasts. The results obtained in this work open up new alternatives to this methodology for the study, classification and selection of the most suitable yeasts to be used as starters, alone or in combination with lactic acid bacteria, during table olive processing.