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

Author(s)

Mesquita, Thiago J. B. ^[1] ; Campani, Gilson ^[2] ; Giordano, Roberto C. ^[1] ; Zangirolami, Teresa C. ^[1] ; Horta, Antonio C. L. ^[1]

Affiliation

^[1] Fed Univ Sao Carlos PPGEQ UFSCar, Grad Program Chem Engn, Rodovia Washington Luis, Km 235, BR-13565905 Sao Carlos, SP - Brazil

^[2] Univ Fed Lavras, Dept Engn, Lavras, MG - Brazil

• Document type: Journal article
• Source: Biotechnology and Bioengineering; v. 118, n. 5 MAR 2021.
• Web of Science Citations: 1

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 center dot L-1 center dot h(-1) and a yield of 0.48 g(ethanol).g(substrate)(-1), which were higher than the values achieved for the other conditions studied (maximum of 3.4 g center dot L-1 center dot h(-1) and 0.35-0.40 g(ethanol)center dot g(substrate)(-1), respectively). Due to its modular character, the MF-ANN strategy could be adapted to other micro-aerated bioprocesses. (AU)

• FAPESP's process: 16/10636-8 - From the cell factory to the Biodiesel-Bioethanol integrated biorefinery: a systems approach applied to complex problems in micro and macroscales
• Grantee: Roberto de Campos Giordano
• Support Opportunities: Program for Research on Bioenergy (BIOEN) - Thematic Grants