Response surface and artificial neural network simulation for process design to produce L-lysine by Corynebacterium glutamicum NCIM 2168

The L-lysine is one of the most important essential amino acid used in food and pharmaceutical industries. The present investigation was conducted to optimize the L-lysine production by Corynebacterium glutamicum (NCIM 2168). The production parameters such as the temperature, pH and glucose concentration (g/l) were optimised and evaluated by simulation method to develop a suitable model. The experimental design was done using central composite design (CCD). Total 20 set of experiments were performed according to the CCD. The factors and their responses were analysed by using the statistical tools: response surface methodology (RSM) and artificial neural network (ANN) linked with genetic algorithm (GA). The predicted optimum production of L-lysine was 19.003 g/l and 28.363 g/l by CCD-RSM and ANN-GA respectively. During validation by GA under optimized conditions, the L-lysine production was found to be 27.25 +/- 1.15 g/l, which was significantly high than that obtained using CCD-RSM optimization method. The ANN coupled with GA was found to be a powerful tool for optimizing production parameters with high level of accuracy. This technique may be used for other fermentation products to optimize the important process parameters before scaling up the process to industrial level.