Application of ligand- and receptor-based approaches for prediction of the HIV-RT inhibitory activity of fullerene derivatives

Fullerene and its derivatives have potential to be utilized in many biomedical applications. In the present study, we investigated the role of fullerene derivatives as inhibitors of HIV-RT by combined protein–ligand docking approach and QSAR methods. The study shows the best predictive QSAR model that represents a two-variable model. It has a good ratio of the number of descriptors and predictive ability. The main contributions to the inhibitory activity are provided by signal JhetZ descriptor and μ (dipole moment, as a measure of the polarity of a compound). The developed GA-MLRA-based model demonstrates a good performance, confirmed by statistics $$\left( {R²_{\text{training}} = 0.867,\;Q² = 0.788,\;R²_{\text{test}} = 0.902} \right).$$ The structure–activity analysis of these fullerene analogues allowed us to design and suggest for synthesis a set of new potentially active fullerenes. Finally, the molecular docking analysis was carried out to understand the details of interactions between HIV-RT and fullerene-C60 derivatives.