Studies reported here are aimed to investigate the important structural features that characterize the human EP3 antagonists. Based on the knowledge of low-energy conformation of the endogenous ligand, the initial hit analogs were prepared. Subsequently, a ligand-based lead optimization approach using pharmacophore model generation was utilized. A 5-point pharmacophore using a training set of 19 compounds spanning the IC50 data over 4-log order was constructed using the HypoGen module of Catalyst. Following pharmacophore customization, using a linear structure-activity regression equation, a six feature three-dimensional predictive pharmacophore model, P6, was built, which resulted in improved predictive power. The P6 model was validated using a test set of 11 compounds providing a correlation coefficient (R2) of 0.90 for predictive versus experimental EP3 IC50 values. This pharmacophore model has been expanded to include diverse chemotypes, and the predictive ability of the customized pharmacophore has been tested.
ASJC Scopus subject areas
- Chemical Engineering(all)
- Computer Science Applications
- Library and Information Sciences