Biochemical and pharmacological development of steroidal inhibitors of aromatase

Prediction of aromatase inhibitory activity using the efficient linear method (ELM)

“…The steroidal and non-steroidal AIs exerted their inhibitory activity via a distinct mechanism. Steroidal AIs competitively and covalently bind the active site of the aromatase enzyme in an irreversible manner (Brueggemeier et al, 1990[ 4 ]), whereas non-steroidal AIs coordinates with the heme iron (Fe) atom of the enzyme thereby giving rise to reversible inhibition (Graves and Salhanick, 1979[ 10 ]). For the steroidal type, atom-centered-fragments, edge adjacency indices and 2D autocorrelation descriptors were highlighted as informative descriptors with large usage values.…”

“…Previously, support vector machine (SVM) had been successfully used to model a wide variety of biological activity. In fact, such SVM-based model is well recognized as one of the most powerful learning approach outperforming other learning methods such as artificial neural networks (ANN) and multiple linear regression (MLR) (Attar and Bulun, 2006[ 2 ]; Brueggemeier et al, 2005[ 3 ], 1990[ 4 ]). The limitation of this model is its low interpretability whereby prediction is performed in a black-box manner, i.e., practitioners may not gain insights into which molecular descriptors highly influenced the activity/inactivity of chemical compounds.…”

Investigation of aromatase inhibitory activity of metal complexes of 8-hydroxyquinoline and uracil derivatives

“…Steroidal aromatase inhibitors competitively and covalently bind to aromatase enzyme in an irreversible fashion 22. Non-steroidal aromatase inhibitors contain azoles as the privileged structure1 for coordinating with heme iron (Fe) atom of aromatase enzyme leading to reversible enzyme inhibition 23.…”