A hybrid system of Artificial Neural Networks (ANN) and Genetic Algorithm (GA) was used to optimize the medium
constituents to enhance the lipase production by Yarrowia lipolytica NCIM 3589 in solid state fermentation (SSF) using
niger seed oil cake as the substrate. Different microbial metabolism regulating fermentation variables (initial moisture content of
the substrate, glucose and urea) were used to construct a ?3?5?1? topology of the ANN for identifying the nonlinear relationship
between fermentation variables and enzyme yield. ANN predicted values were further optimized for lipase production using GA.
The goodness of the neural network prediction (coefficient of R2) was found to be 0.988. Improved enzyme yield could be achieved
by this microbial strain in wide nutrient concentration range and each selected variable concentration depends on rest of the
variable concentration. The usage of ANN?GA hybrid methodology has resulted in a significant improvement (25 %) in the lipase
yield. The present study helps to optimize enzyme production and its regulation pattern by combinatorial influence of different
medium constituents. Further, the information obtained in this study signifies its importance during scale-up studies. MATLAB
7.0 was used for implementing the ANN?GA and the MATLAB program developed for data fitting is included as the appendix.
Peer Reviewed Journals
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals