Application of an artificial neural network coupled to a genetic algorithm for the production of polyphenols in Parachlorella kessleri grown under mixotrophic conditions

Fernández Izquierdo et al. (2024) used ANN-GA to optimize light conditions for polyphenol production in Parachlorella kessleri, identifying white light at 1000 lx and 12:12 light-dark cycle as optimal, yielding 28 mg GAE/g DW.

“For instance, Fernández Izquierdo et al. (2024) [28] employed an ANN coupled with a genetic algorithm (ANN-GA) to optimize light conditions for polyphenol production in Parachlorella kessleri. The model identified white light at an intensity of 1000 lx and a 12:12 light–dark cycle as optimal, contributing to a polyphenol yield of 28 mg gallic acid equivalents per gram dry weight (GAE/g DW),”

The ANN model demonstrated superior computational efficiency and better generalization across light intensities and photoperiods than mechanistic and hybrid models.

“Furthermore, the ANN model demonstrated superior computational efficiency, requiring significantly less time for training and prediction compared to the mechanistic and hybrid models.”

AI-driven models can enhance lipid and high-value compound yields by more than 20-43% compared to traditional methods.

“Key case studies demonstrate that AI-driven models can increase biomass productivity by up to 15–57%, improve CO2 biofixation efficiency, and enhance lipid and high-value compound yields by more than 20–43% compared to traditional methods.”