Modeling of biological processes using self-cycling fermentation and genetic algorithms

Area of Science:

• Biotechnology
• Computational Biology
• Environmental Engineering

Background:

• Biological model evaluation is crucial for understanding complex systems.
• Traditional nonlinear regression methods face computational challenges.
• A need exists for efficient and accurate biological model characterization techniques.

Purpose of the Study:

• To develop a simplified system for evaluating biological models.
• To couple Self-cycling Fermentation (SCF) with a Genetic Algorithm (GA).
• To demonstrate the efficacy of SCF-GA for model fitting and comparison.

Main Methods:

• Utilized Self-cycling Fermentation (SCF) for system excitation and data richness.
• Employed a Genetic Algorithm (GA) as a solution scheme, avoiding calculus-based regression.
• Validated the mathematical approach with denitrifying conditions data.
• Applied SCF data for phenol removal to compare multiple biological models.

Main Results:

• The SCF-GA system provided rapid and accurate convergence for model fitting.
• Successfully fitted an established model to SCF data under denitrifying conditions.
• Effectively compared multiple models using SCF data for phenol removal.
• Demonstrated the GA's ability to overcome computational difficulties of nonlinear regression.

Conclusions:

• Self-cycling fermentation (SCF) coupled with a genetic algorithm (GA) offers a coherent and efficient system for biological model evaluation.
• This integrated approach facilitates the complete definition and characterization of biological systems.
• The SCF-GA methodology simplifies complex biological modeling tasks.