Development of a computational approach to predict blood-brain permeability on anti-viral Nucleoside Analogues
Z. Bayat, Jebrail Movaffagh, S. Noruzi · Russian Journal of Physical Chemistry A · 2011
The usefulness of the quantum chemical descriptors, calculated at the level of the DFT and HF theories using 6-31G* basis set for QSAR study of anti-viral Nucleoside Analogues drugs was examined. The QSAR model developed contributed to a mechanistic understanding of the investigated biological effects. Biological activities contain the logarithm of the ratio of the steady-state concentration of a compound in the brain to in the blood, logBB. A multi-parametric equation containing maximum five descriptors at HF/6-31G* and four descriptors at B3LYP/6-31G* method with good statistical qualities (R max = 0.988, R max 2 = 0.977 at HF/6-31G* and R max = 0.996, R max 2 = 0.991 at B3LYP/6-31G*) was obtained by Multiple Linear Regression (MLR) using stepwise method. Also, GA-MLR regression (GA—genetic algorithm) was used to model the structure—activity relationships.