Machine learning algorithms used in Quantitative structure-activity relationships studies as new approaches in drug discovery

Developing machine learning algorithms have become important tools in drug discovery process. Nowadays, a variety of machine learning tools are used in quantitative structure-activity relationships (QSARs) to establish QSAR models. The 2D-QSAR analysis involves the study of quantitative relationships between the molecular descriptors and biological activity by using machine learning algorithms, such as partial least squares (PLS) and artificial neural networks (ANNs). The best linear 2D-QSAR model was developed through partial least squares (PLS) gave a high predictive ability (R2 = 0.87, F=52.80, R²pred = 0.80, Q2 = 0.77). Moreover, the non-linear artificial neural networks (ANNs) was shown better performance with Levenberge Marquardt (L-M) algorithm (architecture [3-3-1]: R²=0.94, R²pred=0.81, Q2=0.86). Those results uncovered that a_nO, PEOE_VSA+6 and Vsurf_R are important descriptors on which biological activity depends. Moreover, the retained 3D-QSAR model exhibits the best results (R2 = 0.94, R²pred = 0.80, Q2 = 0.67). However, the derived contour maps obtained by 3D-QSAR analysis indicate the favorable and unfavorable regions that could improve the cytotoxic activity. As a consequence, the established QSAR models based on machine learning methods could help us to understand the structural requirements necessary to design new compounds with improved biological activity.