A QSAR study of 5α –reductase inhibitors using multivariate image analysis

Testosterone and its more potent metabolite, DHT, are essential hormones for male phenotype sexual differentiation and maturation through their actions at the androgen receptor. This hormone convert to dihydrotestosterone (DHT) by the enzyme 5α -reductase (5AR) that this conversion is related to the development of many endocrine diseases such as prostatic cancer. Also 5AR play an important role in other physiological processes. 5α -Reductase is a system of two isozymes: 5α reductase Type 1 and 5α -reductase Type 2 [1-3]. In this study, we applied quantitative structure-activity relationship (QSAR) analysis to a series of 38 compounds of 4-X-17-Y-4-Azaandrost-3-ones as 5α -Reductase Type 2 inhibitors. we investigate the effect of pixel selection by application of genetic algorithms (GA) and orthogonal signal correction (OSC). In order to find correlations between inhibitory activities and calculated predictor variables, partial least-squares regression (PLS) model were developed and the resulted model showed high prediction ability.