Optimization of fermentation medium for nisin production from Lactococcus lactis subsp. lactis using response surface methodology (RSM) combined with artificial neural network-genetic algorithm (ANN-GA)

摘要: Nisin is a bacteriocin approved in more than 50 countries as safe natural food preservative. Response surface methodology (RSM) combined with artificial neural network-genetic algorithm (ANN-GA) was employed to optimize the fermentation medium for nisin production. Plackett-Burman design (PBD) used identifying significant components medium. After that, path of steepest ascent method (PSA) approach their optimal concentrations. Sequentially, Box-Behnken experiments were implemented further optimization. RSM ANNGA analysis data. Specially, model determining individual effect and mutual interaction tested variables on titer (NT), an ANN NT prediction, GA search optimum solutions based model. As obtained by ANN-GA located at verge test region, Box- Behnken statistical results implemented. using data locate solution which follow (g/l): Glucose (GLU) 15.92, peptone (PEP) 30.57, yeast extraction powder (YEP) 39.07, NaCl 5.25, KH 2 PO 4 10.00, MgSO ·7H O 0.20, expected 22216 IU/ml. The validation triplicate average 21423 IU/ml, 2.13 times higher that without methods 8.34 Key words : methodology, network, genetic algorithm, titer.