Study of the structure‐activity relationship for theoretical molecular descriptors using density functional theory and chemometric methods in cannabinoid metabolites
T.B. Silva, Mariano Alves Pereira, V.R.S. Malta, Edson S. Bento, Miguel A. San‐Miguel, Roberta Lourênço Ziolli, João B. L. Martins, Andre Sih, Carlton Anthony Taft · International Journal of Quantum Chemistry · 2008
Abstract A set of 30 cannabinoid metabolites has been investigated from a combination of electronic and chemometric methods. Density functional calculations have been carried out to obtain optimized geometries, energies, and selected molecular properties. These molecular descriptors take into account steric effects, electronic properties, and chemical reactivity. The use of statistical methods including principal component analysis (PCA), hierarchical cluster analysis (HCA) and nonhierarchical cluster analysis (K‐means), nearest neighbor (KNN) and artificial neural networks (ANN) has enabled to classify the compounds into psychoactive, moderately psychoactive and psychoinactive groups in good agreement with experimental evidences. © 2008 Wiley Periodicals, Inc. Int J Quantum Chem, 2008