In this study, quantitative structure-activity relationship studies which make use of molecular dynamics trajectories were performed on a set of 54 glucokinase protein activators. The conformations obtained by molecular dynamics simulation were superimposed according to the twelve alignments tested in a virtual three-dimensional box comprised of 2 Å cells. The models were generated by the technique that combines genetic algorithms and partial least squares. The best alignment models generated with a determination coefficient (r(2)) between 0.674 and 0.743 and cross-validation (q(2)) between 0.509 and 0.610, indicating good predictive capacity. The 4D-QSAR models developed in this study suggest novel molecular regions to be explored in the search for better glucokinase activators.

Department of Chemistry Federal University of Lavras P O Box 3037 Lavras 37200 000 Brazil

Department of Chemistry Federal University of Lavras P O Box 3037 Lavras 37200 000 Brazil Center for Basic and Applied Research Faculty of Informatics and Management University of Hradec Kralove Hradec Kralove Czech Republic

Citace poskytuje Crossref.org