Development and application of an ANN-perception-based autonomous control system for Escherichia coli cultivation...

Area of Science:

• Biotechnology
• Artificial Intelligence
• Microbial Fermentation

Background:

• High-density Escherichia coli fermentation faces challenges with overflow metabolism and manual control.
• Optimizing dissolved oxygen (DO) and induction is crucial for enhancing protein expression.

Purpose of the Study:

• To develop an AI-driven autonomous intelligent control system for high-density E. coli fermentation.
• To improve protein expression and process stability while reducing manual intervention.

Main Methods:

• Optimized DO-stat feeding and induction using superfolder green fluorescent protein (sfGFP).
• Developed and trained an artificial neural network (ANN) for real-time DO baseline recognition (R² = 0.998).
• Integrated ANN with feeding logic to create the NeuroStat-Ctrl system for autonomous control.

Main Results:

• Achieved a 52.85% increase in cellular specific fluorescence intensity through optimized parameters.
• NeuroStat-Ctrl system further enhanced fluorescent protein production by 5.87% compared to manual control.
• Demonstrated mitigation of feeding deviations, prevention of metabolic overflow, and enhanced process stability.

Conclusions:

• The AI-driven NeuroStat-Ctrl system enables fully autonomous, unattended E. coli fermentation.
• This system offers an efficient, standardized solution for high-throughput strain screening and process validation.
• The intelligent control enhances protein production, stability, and reproducibility in microbial fermentation.