THE USE OF THE ORDERED ORTHOGONALIZED MULTIVARIATE LINEAR-REGRESSION IN A STRUCTURE-ACTIVITY STUDY OF COUMARIN AND FLAVONOID DERIVATIVES ASINHIBITORS OF ALDOSE REDUCTASE

The relationship between molecular descriptors and the inhibitory acti vity of aldose reductase (AR) for a series of coumarin and flavonoid d erivatives has been investigated using a novel multivariate linear reg ression based on the ordered orthogonalized descriptor set. First, sta rting from the set of 31 descriptors we produced absolutely the best n onorthogonalized QSAR models with I descriptors (I = 1-7). These model s are always better than the models that the most authors achieve by t he use of the stepwise inclusion-exclusion procedure. In the next step we realized all possible orthogonalization orderings of a given set o f N descriptors (there are N! of these). The key result is that some o rthogonalization orderings lead to QSAR models with I ordered orthogon alized descriptors that have higher values of both the correlation coe fficient R and cross-validated correlation coefficient R-cv than the c orresponding models with the same number of nonorthogonalized descript ors. In order to achieve the highest possible reliability in predictin g the inhibition we produced several models that were obtained applyin g the ordered orthogonalization procedure on one set with five (N = 5) and on two sets with seven (N = 7) descriptors. Then the inhibitory a ctivity for 34 coumarins and 30 flavonoids was predicted, and several compounds were detected with a very high inhibitory activity.