Response surface methodology and artificial neural network based media optimization for pullulan production in Aureobasidium pullulans
Year: 2024
Type: article
Abstract: The selection and optimization of carbon and nitrogen sources are essential for enhancing pullulan production in Aureobasidium pullulans . In this study, combinations of carbon (sucrose, fructose, glucose) and nitrogen sources ((NH 4 ) 2 SO 4 , urea, NaNO 3 ) were screened, where sucrose and NaNO 3 offered the highest pullulan yield (9.33 g L −1 ). Plackett–Burman design of experiment identified KH 2 PO 4 , NaCl, and sucrose as significant factors, which were further optimized using a central composite design. A hyperparameter-optimized artificial neural network (ANN) model with a 3-6-2-1 architecture demonstrated superior predictive accuracy (R 2 : 0.96) and generalizability (R 2 CV : 0.74) over a reduced quadratic model (R 2 : 0.82). The predicted pullulan yield (31.9 g L −1 ) under ANN model optimized conditions (sucrose: 79.9 g L −1 , KH 2 PO 4 : 0.25 g L −1 , NaCl: 4.3 g L −1 ) closely matched with the observed yield (30.17 g L −1 ), while quadratic model showed a significant deviation (39.7 g L −1 vs. 21.0 g L −1 ), highlighting the reliability of the ANN model.
Institution: Karunya University
Language: English
Subfield: Food Science
Domain: Life Sciences
SDG: Zero hunger
Open Access status: hybrid
Estimated APC paid: $4,260
OpenAlex ID: W4404655171