Abstract
Lipases are a class of triacylglycerol hydrolases which have found a lot of applications as a result of their unique characteristics such as stability, specificity, economic attractiveness etc. This study examined the effect of some microbial stimulants (olive oil, MgSO4 and KH2PO4) on the production of lipase from waste cooking oil (WCO). The fermentation experiments were planned using a three-variable Box-Behnken design and the impact of the stimulants was optimized with response surface methodology (RSM) and artificial neural network (ANN). The results revealed that intermediate concentrations of olive oil, MgSO4 and KH2PO4 were needed to maximize lipase activity. The ANN model predicted an optimal lipase activity of 177.19 U/mL and this was obtained at olive oil, MgSO4 and KH2PO4 concentration of 0.58, 0.04 and 0.22 w/w% respectively while the RSM model predicted an optimal lipase activity of 176.52 U/mL at olive oil, MgSO4 and KH2PO4 concentration of 0.63, 0.05 and 0.25 w/w% respectively. The ANN model was superior to the RSM model in predicting lipase production and this was reflected by better statistical metrics. Thus, biological stimulants can facilitate the fermentation process for optimal lipase production from WCO.
Conflict of interest
The authors have no conflicts of interest to declare that are relevant to the content of this article.
Data availability statement
All data generated or analyzed during this study are included in this published article.
List of Symbols | ||
bi | = | Single effect coefficient |
bij | = | Interaction effect coefficient |
bo | = | Offset term |
ei | = | Experimental error term |
k | = | Number of input variables |
n | = | Number of experimental runs |
R | = | Correlation coefficient |
R2 | = | Coefficient of determination |
= | Average experimental values | |
= | Experimental values | |
= | Average estimated values | |
= | Estimated values | |
X1 | = | MgSO4 |
X2 | = | Olive oil |
X3 | = | KH2PO4 |
Xi | = | Independent variables |
Xj | = | Independent variables |
Y | = | Dependent variable (lipase activity) |
List of abbreviations | ||
AAD | = | Average absolute deviation |
ANN | = | Artificial neural networks |
ANOVA | = | Analysis of variance |
BBD | = | Box-Behnken design |
BBP | = | Batch back propagation |
CCD | = | Central composite design |
CV | = | Coefficient of variation |
DOE | = | Design of experiments |
GA | = | Generic algorithm |
IBP | = | Incremental back propagation |
LM | = | Levenberg-Marquadt |
MAE | = | Mean absolute error |
MFFF | = | Multilayer full feed forward |
MNFF | = | Multilayer normal feed forward |
MSE | = | Mean square error |
OFAT | = | One-factor-at-a-time |
PDA | = | Potato dextrose agar |
QP | = | Quick propagation |
RMSE | = | Root mean square error |
RSM | = | Response surface methodology |
SEP | = | Standard error of prediction |
WCO | = | Waste cooking oil |