Media optimization for biosurfactant production by Rhodococcus erythropolis MTCC 2794: artificial intelligence versus a statistical approach

Abstract

This paper entails a comprehensive study on production of a biosurfactant from Rhodococcus erythropolis MTCC 2794. Two optimization techniques--(1) artificial neural network (ANN) coupled with genetic algorithm (GA) and (2) response surface methodology (RSM)--were used for media optimization in order to enhance the biosurfactant yield by Rhodococcus erythropolis MTCC 2794. ANN and RSM models were developed, incorporating the quantity of four medium components (sucrose, yeast extract, meat peptone, and toluene) as independent input variables and biosurfactant yield [calculated in terms of percent emulsification index (% EI(24))] as output variable. ANN-GA and RSM were compared for their predictive and generalization ability using a separate data set of 16 experiments, for which the average quadratic errors were approximately 3 and approximately 6%, respectively. ANN-GA was found to be more accurate and consistent in predicting optimized conditions and maximum yield than RSM. For the ANN-GA model, the values of correlation coefficient and average quadratic error were approximately 0.99 and approximately 3%, respectively. It was also shown that ANN-based models could be used accurately for sensitivity analysis. ANN-GA-optimized media gave about a 3.5-fold enhancement in biosurfactant yield.

Statistical Design, a Powerful Tool for Optimizing Biosurfactant Production: A Review

“…An interesting observation from this study was that the same factors evaluated had varying effects on the two S. marcescens strains. Bs production in SM3 (represented by a reduction in ST and increase in EI 24 ) was favored at high concentrations of glycerol and peptone, while the opposite was reported for the isogenic strain. After the application of the statistical designs, ST was reduced to 26.5 and 25.2 mNm −1 , for SM3 and SMRG5, respectively.…”

“…ANN has been compared to the use of statistical design, and ANN has been shown to be more accurate in predicting the optimum conditions for BS production as in the case of the study carried out by Pal et al [24]. These authors improved the Bs (glycolipid containing trehalose as the major charbohydrate) production in Rhodococcus erythropolis.…”

“…Different nutrient sources may regulate Bs synthesis by induction or catabolic repression. Immiscible substrates such as different oils and hydrocarbons have been used as Bs production inducers [23,24]. Catabolic repression has been observed in hydrocarbon degrading bacterial strains that do not produce surface-active molecules in the presence of carbon sources like glucose and organic acids.…”

Prediction of Biosorption of Total Chromium by Bacillus sp. Using Artificial Neural Network

“…Moreover, neural network has a distinctive ability to learn nonlinear functional relationships without the requirement for structural knowledge of the process to be modeled. Among the various ANN models, the one of our interest was the feed forward back propagation network (Imandi et al 2008;Pal et al 2009). The feed forward back propagation neural network consisting of forward three neurons corresponding to the three process variables (pH, time and initial metal ion concentration) were used in the input layer, ten in the hidden layer and one in the output layer (Q e ) of the network.…”

Estimation of biosurfactant yield produced by Klebseilla sp. FKOD36 bacteria using artificial neural network approach

“…Some of the prominent features of the ANN are; nonlinearity, easy data fitting, parallelism, learning, adaptively, constructed neuron pattern and generalization which makes it more suitable than other method. ANN also had been proven very powerful tool in optimization of various parameters like incubation, carbon and nitrogen sources, temperature and hydrocarbon sources in biosurfactant production [30][31][32][33][34]. Pal et al [30] described the production optimization of biosurfactant from Rhodococcus erythropolis MTCC 2794 using the ANN coupled with genetic algorithm (GA) and response surface methodologies.…”