Determination of important variables for CMCase production by Plackett-Burman design
Plackett-Burman design (PBD) was conducted to determine significant variables. Eleven variables, including eight main components of fermentation medium and three dummy variables were designed for twelve run experiments. Table 2 shows the PBD experimental design and the response of CMCase were observed from 0.306 Unit/ml to 0.860 Unit/ml. For statistical analysis, regression analysis was carried out for the model. From the regression analysis, the coefficient of determination (R2) was 0.98. This indicates that the model has reliability of 98% for the data. The coefficient of determination (adjusted R2) was calculated to be 0.9268 which implies agreement between the experimental and predicted values of CMCase. The F-value and P-value (>F) of the model were 18.4034 and 0.0179, respectively. The P-value was used as an indicator of analysis for the model and is important for understranding the pattern of mutual interactions between the variables [19] and [20]. In general, P- values lower than 0.005 were reliable during optimization [21]. The coefficient of variation (C.V) was 7.10%, which supports the reliability of the model. In our experiment, three variables (corn stover, urea and ammonium sulfate) were determined as sinificant variables because the P-value was less than 0.05. Among them, urea showed the strongest significance with a P-value of 0.0045 (Table 5).