Conception par modélisation moléculaire et Criblage virtuel d'une série bioactive Phtalazine

Abstract

The computer tool is currently requisite in research in biology to treat the stream of data produced and to optimize its progress. The '' docking '' is, indeed, one of the methods commonly used in pharmacochimistry to discover and to finalize of new medicines by screening of thousands of compounds for a protein target. The molecular modeling methods used in our work are: PM3, (DFT / B3LYP / 6-31 ++ G (d, p)). These methods were used to determine the physicochemical and quantique parameters associated with the studied molecules. A QSAR study was performed on 25 phthalazine derivatives. Multiple linear regression (MLR) was used to quantify the relationship between molecular descriptors and the activity anticancer (skin cancer) of phthalazine derivatives. The model predictive power obtained was confirmed by the two methods of LOO cross-validation and external validation. A strong correlation was observed between the experimental and predicted values of biological activities, indicating the validity and quality of the QSAR models obtained. Finally, a molecular doking study was also carried out for a bioactive series of phthalazine, after the calculation we search the best pose (Conformation) and the formation of the complexes more stable, the best molecules obtained are anticancer drugs.