Biological activities of triazine derivatives. Combining DFT and QSAR results

Abstract

In order to investigate the relationship between activities and structures, a 3D-QSAR study is applied to a set of 43 molecules based on triazines. This study was conducted using the principal component analysis (PCA) method, the multiple linear regression method (MLR) and the artificial neural network (ANN). The predicted values of activities are in good agreement with the experimental results. The artificial neural network (ANN) techniques, considering the relevant descriptors obtained from the MLR, showed a correlation coefficient of 0.9 with an 8-3-1 ANN model which is a good result. As a result of quantitative structure–activity relationships, we found that the model proposed in this study is constituted of major descriptors used to describe these molecules. The obtained results suggested that the proposed combination of several calculated parameters could be useful to predict the biological activity of triazine derivatives.

Keywords

• Biological activity
• 3D-QSAR model
• MLR
• ANN
• PCA
• DFT study