QSAR study of isatin(1H-indole-2,3-dione) analogues as in vitro anti-cancer agents using the statistical analysis methods and the artificial neural network [ ]


Isatin (1H-indole-2,3-dione) and its derivatives arepotent anticancer agents, these compounds inhibit cancer cells proliferation and tumorgrowth.A study of quantitative structure-activity relationship (QSAR) is applied to a set of 47 molecules derived from isatin, in order to predict the anticancer biological activity of the test compounds and find a correlation between the differentphysic-chemical parameters (descriptors) of these compounds and its biological activity, using multiple linear regression (MLR) and the artificial neural network (ANN).The topological and the electronic descriptors were computed, respectively, withACD/ChemSketchand (ChemDraw Ultra 8.0, ChemBioDraw Ultra 14.0) programs. A good correlation was found between the experimental activity and that obtained by MLR such as (R = 0.94 and R2 = 0.88), this result could be improved with ANN such as(R = 0.97 andR2 = 0.94) with an architecture ANN (5-3-1). To test the performance of the neural network and the validity of our choice of descriptors selected by MLR and trained by ANN, we used cross-validation method (CV) such as (R = 0.95 and R2 = 0.90) with the procedure leave-one-out (LOO).