Application du criblage virtuel, du « drug-likness » au modèle QSAR dans une série hétérocyclique bioactive

Abstract

The work done in this thesis deals with a general research that aims to discuss the activity of a series of thirty-three of 1,2- diazole derivatives as inhibitors of ENR that may be potential agents for treating mycobacterium tuberculosis. To begin with, various theoretical calculation methods were used to carry out our work: PM3, ab initio HF and DFT / B3LYP. These methods were used to determine the structural and electronic parameters associated with the studied molecules. Then, a qualitative study of the structure-activity relationship (SAR) was also performed for a bioactive series of pyrazole derivatives using different MPO methods. Finally, a quantitative study was also performed to predict the biological activity of the compounds studied and its derivatives by suggesting the best QSAR model. The model prediction obtained was confirmed by the LOO cross validation method.