Comparative study of μ‐stat methanol feeding control in fed‐batch fermentation of Pichia pastoris producing HBsAg: an open‐loop control versus recurrent artificial neural network‐based feedback control

Abstract

BACKGROUND In recent decades, artificial neural network (ANN) has been shown to be a robust and promising tool in monitoring and controlling bioprocess systems. In a previous study, the authors designed a highly accurate and precise recurrent neural network (RNN) for predicting the biomass amount of recombinant Pichia pastoris Mut+ producing intracellular hepatitis B surface antigen (HBsAg) during fed‐batch methanol fermentation. In the current work, the aim was to compare the production efficiency of HBsAg between conventional predefined μ‐stat methanol feeding control (open‐loop control) – already established in large‐scale production – and methanol feeding based on a μ‐stat feedback control system using RNN. For this purpose, for each methanol feeding strategy, bench‐scale, fed‐batch fermentation processes were carried out twice. RESULTS According to the results, in contrast to the established μ‐stat predefined feeding strategy (open‐loop), the deviation of specific growth rate and biomass in μ‐stat feedback control based on RNN was negligible. Also, in the suggested methanol feeding control strategy, the HBsAg titer, specific productivity and yield between the performed fed‐batch fermentations – unlike the conventional method – were approximately identical, with average values of 110.8 μg mL−1, 1.52 μg g−1 dry biomass and 0.34 μg g−1 MeOH respectively. CONCLUSION Comparing the biomass growth pattern and HBsAg production efficiency with the conventional μ‐stat predefined feeding in open‐loop control, the new proposed feeding control system illustrated significantly high process efficiency. This reliable control system based on ANN can have many applications in the biopharmaceutical industry for the control of process key parameters as well as for enhancing process efficiency. © 2019 Society of Chemical Industry

From Spatial-Temporal Multiscale Modeling to Application: Bridging the Valley of Death in Industrial Biotechnology

“…Recurrent neural network (RNN) is an artificial neural network for series data, which can transfer information between neurons and express the correlation between data while taking the time dimension into account. RNNs have been widely used in the time-series prediction of key state variables and diagnosing faults in fermentation, particularly in industrial-scale production [ 89 , 90 ]. Beiroti et al accurately predicted the biomass of recombinant Pichia pastoris Mut + , and optimized the process conditions in the methanol induction phase of the fed-batch fermentation [ 91 ].…”

History and Evolution of Modeling in Biotechnology: Modeling & Simulation, Application and Hardware Performance

“…The authors reported that the newly developed RNN was used in a PID control system for maintaining a predefined specific growth rate at a constant value by adjusting the methanol feed rate. They report that the RNN based feedback controller has a “significantly higher process efficiency” than a conventional open-loop controller with predefined feeding strategy [104] .…”

Production and Purification of Soy Leghemoglobin from Pichia pastoris Cultivated in Different Expression Media

“…Despite this limitation, ANNs have demonstrated success in predicting the behavior of diverse fermentation systems, prompting their utilization in bioprocess control applications. Recent applications of ANN models in cell biomass estimation, encompass regulating specific growth rate [30], optimizing cell biomass [31][32][33], and estimating [33,34] or tracking a predefined substrate concentration trajectory [35].…”