Objective: The present study was designed to study the antifungal activity of Dihydropyrimidine-4-Carbonitrile analogs against the fungi Candida albicans by a 2D quantitative structure-activity relationship (QSAR) model.
Methods: The pyrimidine derivatives were produced using lipophilic, electronic, and steric parameters by Quantitative Structure Activity-Relationships (QSAR). A relationship between dependent and independent variables (biological activities and physicochemical descriptors, respectively) was resolved statistically using regression analysis. The F value shows the level of statistical significance of the regression (r2) was used to report the fitness of data. The newly synthesized derivatives were evaluated for in vitro antifungal activity against Candida albicansby Nutrient agar and Seaboard dextrose agar media.
Results: Multiple linear regression is a method of crucial importance, it allowed us to obtain a relation between the calculated parameters and the antifungal activity; this we can interpret the variance of the activity by contribution to the calculated descriptors. Quantitative structure-activity relationship (QSAR) model showing a significant activity-descriptors relationship accuracy of 90% (R2 ≥ 0.90) and activity prediction accuracy of 81% (R²cv = 0.81). These values prove that the model obtained is reliable. Out of the three descriptors studied; log P has minimum potency, molar refractivity has more potency and heat of formation has moderate potency.
Conclusion: Important structural understanding in the pattern of potent antifungal agents by Quantitative Structure Activity-Relationships (QSAR) study. The acquired physicochemical properties (electronic, topological, and steric) show the important structural features required for antifungal activity against Candida albicans.