Table 3.

Statistical information of the 3D-QSAR models.

Fields	NC	R2	S	Q2	SLOO	$R_{\text{test}}^2$	Stest	%S	%E
S	8	0.98	0.13	0.64	0.54	0.34	0.81	1
E	4	0.81	0.39	0.54	0.62	0.59	0.63	–	1
SE	7	0.97	0.15	0.70	0.50	0.61	0.62	0.88	0.12

NC is the number of components; S is the standard deviation of the fitted activity of the training set; R², Q², and $R_{test}^{\mathit{\text{2}}}$ are the coefficients of correlation of the training set, LOO cross validation, and test set, respectively; S_LOO is the standard deviation of the LOO cross validation, and S_test is the standard deviation of the test predictions. %S and %E are the relative contributions of the steric (S) and the electrostatic (E) fields, respectively.