2D QSAR modeling and preliminary database searching for dopamine transporter inhibitors using genetic algorithm variable selection of Molconn Z descriptors
Bt. Hoffman et al., 2D QSAR modeling and preliminary database searching for dopamine transporter inhibitors using genetic algorithm variable selection of Molconn Z descriptors, J MED CHEM, 43(22), 2000, pp. 4151-4159
In light, of the chronic problem of abuse of the controlled substance cocai
ne, we have investigated novel approaches toward both understanding the act
ivity of inhibitors of the dopamine transporter (DAT) and identifying novel
inhibitors that may be of therapeutic potential. Our most recent studies t
oward these ends have made use of two-dimensional (2D) quantitative structu
re-activity relationship (QSAR) methods in order to develop predictive mode
ls that correlate structural features of DAT ligands to their biological ac
tivities. Specifically, we have adapted the method of genetic algorithms-pa
rtial least squares (GA-PLS) (Cho et al. J. Comput.-Aided Mel. DES., submit
ted) to the task of variable selection of the descriptors generated by the
software Molconn Z. As the successor to the program Molconn X, which genera
ted 462 descriptors, Molconn Z provides 749 chemical descriptors. By employ
ing genetic algorithms in optimizing the inclusion of predictive descriptor
s, we have successfully developed a robust model of the DAT affinities of 7
0 structurally diverse DAT ligands. This model, with an exceptional q(2) va
lue of 0.85, is nearly 25% more accurate in predictive value than a compara
ble model derived from Molconn X-derived descriptors (q(2) = 0.69). Utilizi
ng activity-shuffling validation methods, we have demonstrated the robustne
ss of both this DAT inhibitor model and our QSAR method. Moreover, we have
extended this method to the analysis of dopamine D-1 antagonist affinity an
d serotonin ligand activity, illustrating the significant improvement in q(
2) for a variety of data sets. Finally, we have employed our method in perf
orming a search of the National Cancer Institute database based upon activi
ty predictions from our DAT model. We report the preliminary results of thi
s search, which has yielded five compounds suitable for lead development as
novel DAT inhibitors.