Application of virtual screening to obtain new 17α-methylated steroids with potential activity anabolic and androgenic

Abstract

Computational methods for virtual screening of large databases are taken into account in the use of models
of biological structure-activity relationship (QSAR), which makes an interesting alternative for high-through put screening
(SAR) as an important tool for drug discovery. In previous builds first QSAR model for prediction of new anabolic-androgenic
a loose collection of molecules whose biological activity was previously evaluated in vivo. In this model it was used
quantum molecular descriptors and chemical-physical, coupled with chemometric techniques that help explain how the
steric, electronic and hydrophobic characteristics influence in the biological response of these compounds. In this paper
it was use the model equation above and perform a virtual screening applied to anabolic- androgenic steroids synthesized
and evaluated by the authors before, where it is shown that the model developed can be applied in the identification of
new lead compounds with best anabolic activity. Because these results are designing an appropriate sequence for synthetic
17α-methyl steroids, which yielded five new compounds with potential anabolic activity, correctly characterized by 1H,
13C NMR spectroscopy and mass spectrometry.