Combining pharmacophore and protein modeling to predict CYP450 inhibitors and substrates

This chapter discusses experience in homology modeling of cytochrome P450 (CYPs) 2C8, 2C9, 2C18, and CYP2C19 based on the rabbit CYP2C5 crystal structure. A substrate selectivity analysis for the CYP2C subfamily is also discussed in the chapter and highlights the amino acids responsible for the selectivity. Generation of a three dimension-quantitative structure–activity relationship (QSAR) model for a diverse set of CYP2C9 inhibitors taking into account important parameters, such as mechanism of inhibition and stereochemistry, is described in the chapter. Basic validation of the QSAR models involves cross validation using the “leave one out” (L.O.O.) technique or different percentages of elements of the original training set and trying to predict their biological effect by the model generated with the remaining compounds. This method evaluates the predictive power of the model inside the set defined to build it but it could give an overly optimistic view of the performance of the model.