Optimization of fermentation medium for β-fructofuranosidase production from Arthrobacter sp 10138 using artificial neural network and genetic algorithms

In this paper, the optimization of medium components was reported for the production of beta-fructofuranosidase (FFase) from Arthrobacter sp. 10138. Experiments were conducted using the uniform design (UD), and the data were used to build an artificial neural network model. The concentrations of six medium components (sucrose, beef extract, yeast extract, (NH4)(2)HPO4, KH2PO4 and MgSO4) served as inputs to the neural network model, and the FFase activity served as outputs of the model. Using the genetic algorithms (GA), the input space of the neural network model was optimized to find out the optimum values for maximum FFase activity. Maximum FFase activity of 318.5U/mL was obtained at the GA-optimized concentrations of medium components (sucrose 33.0 g/L; beef extract 3.0 g/L; yeast extract 2.0 g/L; (NH4)(2)HPO4 4.0 g/L; KH2PO4 0.5 g/L and MgSO4 center dot 7H(2)O 0.2 g/L). The FFase activity obtained by the ANN-GA was 15.6% higher than the maximum activity of FFase obtained by UD experiments.