Machine learning applied for metabolic flux‐based control of micro‐aerated fermentations in bioreactors

Abstract

Various bio‐based processes depend on controlled micro‐aerobic conditions to achieve a satisfactory product yield. However, the limiting oxygen concentration varies according to the micro‐organism employed, while for industrial applications, there is no cost‐effective way of measuring it at low levels. This study proposes a machine learning procedure within a metabolic flux‐based control strategy (SUPERSYS_MCU) to address this issue. The control strategy used simulations of a genome‐scale metabolic model to generate a surrogate model in the form of an artificial neural network, to be used in a micro‐aerobic fermentation strategy (MF‐ANN). The meta‐model provided setpoints to the controller, allowing adjustment of the inlet air flow to control the oxygen uptake rate. The strategy was evaluated in micro‐aerobic batch cultures employing industrial Saccharomyces cerevisiae yeast, with defined medium and glucose as the carbon source, as a case study. The performance of the proposed control scheme was compared with a conventional fermentation and with three previously reported micro‐aeration strategies, including respiratory quotient‐based control and constant air flow rate. Due to maintenance of the oxidative balance at the anaerobiosis threshold, the MF‐ANN provided volumetric ethanol productivity of 4.16 g·L−1·h−1 and a yield of 0.48 gethanol.gsubstrate−1, which were higher than the values achieved for the other conditions studied (maximum of 3.4 g·L−1·h−1 and 0.35–0.40 gethanol·gsubstrate−1, respectively). Due to its modular character, the MF‐ANN strategy could be adapted to other micro‐aerated bioprocesses.

Genome-scale modeling of yeast metabolism: retrospectives and perspectives

“…This was demonstrated by a few studies in which GEM simulations were used to control oxygen and feed flow in micro-aerated ethanol fermentation by S. cerevisiae (Mesquita et al . 2019 , 2021 ).…”

Fermentation Process Control and Optimization

“…In the study of Mesquita et al [88] on the metabolic model of ethanol production by baker's yeast, ANN was used as a soft sensor to determine the oxygen metabolic flux, based on the carbon dioxide metabolic flux. The output of ANN is then used in the control law formulated by the authors.…”

Genome-Scale Metabolic Modeling Enables In-Depth Understanding of Big Data

“…Mesquita et al identified cost-effective ways of measuring low oxygen concentrations, creating a surrogate artificial neural network model by simulations of a GEM. This surrogate model was then used in a fermentation strategy [117]. Culley et al developed an ML-based method that integrates metabolic models with large-scale gene expression data to understand the different mechanisms of cell growth in 1143 Saccaromyces cerevisiae mutant strains [118].…”