Predicting the influence of multiple components on microbial inhibition using a logistic response model - a novel approach
- Equal contributors
1 Department of Plant Science, Faculty of Natural and Agriculture Sciences, University of Pretoria, Pretoria 0002, South Africa
2 Department of Statistics, Faculty of Natural and Agriculture Sciences, University of Pretoria, Pretoria 0002, South Africa
3 Department of Paraclinical Sciences, Phytomedicine Programme, Faculty of Veterinary Sciences, University of Pretoria, Pretoria 0002, South Africa
BMC Complementary and Alternative Medicine 2014, 14:190 doi:10.1186/1472-6882-14-190Published: 13 June 2014
There are several synergistic methods available. However, there is a vast discrepancy in the interpretation of the synergistic results. Also, these synergistic methods do not assess the influence the tested components (drugs, plant and natural extracts), have upon one another, when more than two components are combined.
A modified checkerboard method was used to evaluate the synergistic potential of Heteropyxis natalensis, Melaleuca alternifolia, Mentha piperita and the green tea extract known as TEAVIGO™. The synergistic combination was tested against the oral pathogens, Streptococcus mutans, Prevotella intermedia and Candida albicans. Inhibition data obtained from the checkerboard method, in the form of binary code, was used to compute a logistic response model with statistically significant results (p < 0.05). This information was used to construct a novel predictive inhibition model.
Based on the predictive inhibition model for each microorganism, the oral pathogens tested were successfully inhibited (at 100% probability) with their respective synergistic combinations. The predictive inhibition model also provided information on the influence that different components have upon one another, and on the overall probability of inhibition.
Using the logistic response model negates the need to ‘calculate’ synergism as the results are statistically significant. In successfully determining the influence multiple components have upon one another and their effect on microbial inhibition, a novel predictive model was established. This ability to screen multiple components may have far reaching effects in ethnopharmacology, agriculture and pharmaceuticals.