Abstract (eng)
The aim of this master’s thesis is to provide an in-silico-based method, mainly driven by the use of molecular docking, and QSAR modelling as an additional tool, for the analysis of the sodium/glucose co-transporter 2 (SGLT2) and the predic-tion of the activity of its potential inhibitors, which have been evolving to an im-portant contribution to the treatment of diabetes mellitus. To attain this objective, a classification model based on the Docking Scores ob-tained from a docking based virtual screening was created. Furthermore, the abil-ity of various docking programs and their scoring functions to create compound rankings correlating to the ranking by activity was assessed. Finally, as an addi-tional tool for the evaluation of results attained by the structure based approach-es, a number of machine learning based QSAR models for SGLT2 inhibitors were generated and their performances were compared. The methods developed for the analysis of the activity of potential inhibitors were subsequently applied to a number of compounds with unknown activity in order to predict their ability to inhibit SGLT2.