Simulated annealing and artificial neural network as optimization tools to enhance yields of bioemulsifier and exopolysaccharides by thermophilic Brevibacillus borstelensis

Abstract

Bioemulsifiers and exopolysaccharides are industrially important biomolecules that are produced by microbial fermentation and have applications in food, cosmetics and pharmaceutical industries. Both the compounds are produced extracellularly as secondary metabolites by microorganisms. In the current work, two statistical optimization tools, simulated annealing and artificial neural network, were used for optimized production of bioemulsifier and exopolysaccharide by thermophilic Brevibacillus borstelensis. The process variables were concentrations of glucose, monosodium glutamate, yeast extract and magnesium sulfate in production medium and concentrations of bioemulsifier, exopolysaccharide and biomass were studied as responses. The best function values for the three responses as obtained by simulated annealing were 7.14, 17.87 and 3.3 while the model generated by ANN predicted 6.1 g/l, 17.5 g/l and 2.5 g/l as the values of the three responses. The study shows the efficiency of statistical optimization tools compared to the experimental methods for optimization of bioprocess to maximize yield. These techniques also reduce the time and cost of any bioprocess. There is a rise in agricultural residues and the problems associated with them. Agricultural wastes contain high amounts of carbon and nitrogen and hence can be used as suitable substrates for the production of bioemulsifier and exopolysaccharide. The concentrations of medium components obtained from optimized media will be useful when agricultural wastes are used as substrates. This will not only lead to production of two industrially important compounds from sustainable sources but will also result in environmental remediation.