Modeling antimicrobial effect of different grape pomace and extracts on S. aureus and E. coli in vegetable soup using artificial neural network and fuzzy logic system

In this study, antibacterial activities of the grape pomace powders (GPP) and grape pomace extracts (GPE) of Turkeys' five different grape varieties (Emir, Gamay, Kalecik Karasi, Narince and Okuzgozu grape varieties) were determined against Staphylococcus aureus and Escherichia coli in vegetable soup at 2, 5 and 10% concentrations. Antibacterial effects of GPE were more effective than those of the GPP against the bacteria. The bacterial counts decreased with increasing extract concentration. The highest antibacterial activity of the GPE was for Gamay and Kalecik Karasi against both bacteria. The 10% concentration of Gamay GPP completely inhibited S. aureus at the end of 120th h. Compared to E. coli, S. aureus was more sensitive to all GPP and GPE. This bacterium was inhibited by 10% concentrations of all the extracts at the initial storage time. Data were analyzed to predict antibacterial effects of the GPP and GPE against both bacteria by adaptive neuro fuzzy inference system (ANFIS) and artificial neural network (ANN) and multiple linear regression (MLR) models. As a result, ANFIS model was found to be better than the ANN and MLR for predicting antibacterial effects of GPP and GPE against S. aureus and E. coli in the soup. (C) 2011 Elsevier Ltd. All rights reserved.