Comparative molecular field analysis and comparative molecular similarity indices analysis of human thymidine kinase 1 substrates

Thymidine kinase 1 (TK1) is a key target for antiviral and anticancer chemotherapy. Three-dimensional quantitative structure-activity relationship (3D-QSAR) using comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) techniques was applied to analyze the phosphorylation capacity of a series of 31 TK1 substrates. The optimal predictive CoMFA model with 26 molecules provided the following values: cross-validated r(2) (q(2)) = 0.651, non-cross-validated r(2) = 0.980, standard error of estimate (s) = 0.207, F = 129.3. For the optimal CoMSIA model the following values were found: q(2) = 0.619, r(2) = 0.994, s = 0.104, F = 372.2. The CoMSIA model includes steric, electrostatic, and hydrogen bond donor fields. The predictive capacity of both models was successfully validated by calculating known phosphorylation rates of five TK1 substrates that were not included in the training set. Contour maps obtained from CoMFA and CoMSIA models correlated with the experimentally developed SAR. (C) 2004 Elsevier Ltd. All rights reserved.