A high dimensional QSAR study on the aldose reductase inhibitory activity of some flavones: topological descriptors in modeling the activity.

The quantitative structure-activity relationships (QSAR) of the Aldose Reductase (AR) inhibitory activity of 48 flavones were studied using Free-Wilson, Combinatorial Protocol in Multiple Linear Regression (CP-MLR), and Partial Least Squares (PLS) procedures. For the latter two procedures 152 Molconn-Z parameters and six indicators corresponding to the hydroxyls of flavones were used as molecular descriptors. Independently, all procedures suggested the significance of hydroxyls in modulating the activity of these compounds. The CP-MLR procedure identified 26 descriptors to model the activity. They suggested that structures rich in aromatic CH fragments, with a limited number of aliphatic fragments such as -CH2-, -CH<, and free hydroxyls at 7-, 3'-, and 4'-positions of the 2-arylbenzpyran-4-one core would be preferred for the activity. The PLS analysis agreed with the information content and the relative significance of the descriptors identified in the CP-MLR for modeling the activity. The study offers the scope to modulate the inhibitory activity of these compounds.