Optimization of Cellulase Production from Isolated Cellulolytic Bacterium: Comparison between Genetic Algorithms, Simulated Annealing, and Response Surface Methodology. Issue 1 (2nd January 2017)

• Record Type: Journal Article
• Authors: Parkhey, Piyush; Gupta, Pratima; Eswari, J. Satya
• Is Part Of: Chemical engineering communications. Volume 204:Issue 1(2017)
• Page Start: 28
• Page End: 38
• Publication Date: 2017-01-02
• Subjects: Genetic algorithm -- Process optimization -- Response surface methodology -- Simulated annealing; Chemical engineering -- Periodicals; 660.205
• DOI: 10.1080/00986445.2016.1230736
• Languages: English
• ISSNs: 0098-6445

Abstract

Abstract : The present study discusses optimization of cellulase production from isolated cellulolytic bacterium. A simulated annealing (SA) algorithm is proposed for optimization of these processes to achieve the desired production goal. The approach was compared to the use of evolutionary algorithms, i.e., genetic algorithms (GAs) and response surface methodology (RSM). Ochrobactrum haematophilum was identified as the isolated bacteria. Carboxymethyl cellulose (CMC) concentration, yeast extract, pH, and incubation temperature were the significant factors screened by Plackett–Burman design and further optimized using a central composite design. The optimum values obtained were CMC concentration = 4.76% (w/v), yeast extract = 2.03% (w/v), pH = 6.3, and temperature = 44.2°C. Carboxy methyl cellulase (CMCase) activity at these values was experimentally determined to be 3.55 ± 0.16 U/ml, which was 2.8 times than the unoptimized system (1.23 U/ml). The growth-associated and non-growth-associated Leudeking–Piret constants, α and β, were respectively determined to be 0.3943 and 0.0105. The Michaelis–Menten constants, V max and K m, were determined to be 0.67 µmol/min and 2.42 mg CMC/ml, respectively. The variable-sized SA seems to be the best alternative, outperforming the GAs, showing a fast convergence and low variability among the several runs for optimized production cellulose recovery. The SA models are found to be capable of better predictions of cellulase production. The results of the SA-based RSM model indicate that it is much more robust and accurate in estimating the values of dependent variables when compared with the GA-based RSM models and only RSM models.