Comparison of the current model with previous models (N_m > 200).

Method	N m	N de	R 2	R cv5 2	R T 2	AD	Reference
GA-NNa	207	9	0.86	—	0.79	No	58
ANNb	296	12	0.84	—	0.77	No	17
Boosting	1272	30	0.97	0.83	0.81	Yes	3
RFc	15 791	8	0.52	—	—	No	15
Dual-RBF	1827	50	0.91	0.77	0.76	Yes	Present

^aGenetic algorithm-neural network.

^bArtificial neural networks.

^cRandom forest; N_m is the number of compounds in the modeling set, and N_de is the number of descriptors used in each model.