The validated 3D QSAR pharmacophore model Hypo1 was used as a 3D structural query for retrieving potent compounds from NCI database and Maybridge database having 238819 molecules and 2000 molecules respectively .A total of 8833 compounds were showing good mapping with Hypo1 using fast and flexible search method. Out of 8833 compounds 8530 compounds were from NCI and 333 compounds were from Maybridge database. Out of these 8833 molecules, selected only 2033 molecules were selected having their IC50<1 µM for study. These hit compounds were further screened by using Lipinski’s rule of five, to evaluate them drug similarity, and a total of 1613 molecules passed this evaluative process. These 1613 molecules were further subsequent for the ADMET studies. Only 842 molecules were passed from the ADMET filtration. Further we passed these filtered molecules for the docking analysis.
3.6 Molecular Docking study
For further refining the retrieved hits and evaluating the binding mode between compounds and proteins, all compounds and compound_1 were docked into the binding site of PDK1 (PDB entry:1UU7) by using LigandFit  docking method implemented in Discovery Studio 2.5 program package. Before docking the all the molecules, compound_1(most active compound of training set ) was docked in to the active site of PDK1.Compound_1 has shown the docking energy of - 64.5 kcal/mol and RMSD value of 0.841.It showed the hydrogen bond interactions with important residues of like Lys111, Asp 230, Ala 162 & Tyr 161shown in Figure.7 (a).This depict that LigandFit docking method reproduced the original binding mode, so for the further docking study we used the LigandFit docking method.
Finally 7 compounds having high docking energy, having different scaffolds, better hydrogen bond interactions with active sites residues and compounds having lower estimated activity(<=0.19µM) were selected....