Novel 3-substituted-1-aryl-5-phenyl-6-anilino-pyrazolo[3,4-d]pyrimidin-4-one derivatives with anti-inflammatory\r\nactivity were subjected for the two and three dimensional quantitative structure activity relationships (2D-QSAR and 3D-QSAR)\r\nstudies using Vlife MDS 4.0 drug design module with various combinations of thermodynamic, electronic and spatial descriptors.\r\nThe 2D-QSAR models multiple regression as well as principle component regression approach and 3D-QSAR model using the k\r\nnearest neighbour (kNN) approach were developed. The best 2D-QSAR model (r2 = 0.94, Fisher test value F = 22.09, r2 se = 0.06)\r\nhas acceptable statistical quality and predictive potential as indicated by the value of cross validated squared correlation\r\ncoefficient (q2= 0.76) by genetic algorithm multiple regression. The other 2D QSAR model (r2 = 0.70, F = 12.12, r2 se = 0.14, q2 =\r\n0.18) by stepwise forward-backward principle component regression has also acceptable statistical quality. The best 3D-QSAR\r\nmodel (q2 = 0.86, predr2 = 0.023, kNN= 2) was selected by k nearest neighbour method. From these models, it is seen that chiV1,\r\nH-donar count, chi V5 contribute positively and molecular weight contribute negatively for anti-inflammatory activity. Thus\r\nthese validated models bring important structural insight to aid the design of effective anti-inflammatory agent
Loading....