Monitoring biotechnological processes through quantitative image analysis: Application to 2-phenylethanol production by Yarrowia lipolytica

This is the authors' abstract. We don't add key points for this paper.

Quantitative image analysis (QIA) is a simple and automated tool for process monitoring that, when combined with chemometric techniques, enables the association of changes in microbiota morphology to various operational parameters. To that effect, principal component analysis, multilinear regression, and ordinary least squares methods were applied to the obtained dataset of the biotransformation conditions for Y. lipolytica through the monitor of yeast morphology, substrates (glycerol, L-phenylalanine - L-Phe) consumption and metabolites (2-phenylethanol – 2-PE) production was developed. Glycerol and L-Phe were successfully monitored by the proposed approach, though with a lower monitoring ability for 2-PE, and mostly related to yeast and cluster size and proportion, yeasts contents and cluster morphology. The chemometric approach also allowed to identify significant morphological modifications related with the change in the stirring speed in the experiments at 600 rpm, 600/400 rpm (600 rpm for 24 h, and 400 rpm until the end of the experiment) and in pH from 5.5 to 7.5. This work demonstrated, for the first time, that QIA combined with chemometric analysis can be considered a valuable tool to monitor biotechnological processes, namely the 2-PE production by Y. lipolytica, by analyzing yeast and cluster morphology.