Энергии граничных молекулярных орбиталей как фактор прогноза антигликирующей активности N-гидроксибензоил-производных тимина и урацила

Dependences of the antiglycation activity of N -hydroxybenzoyl-substituted thymine and uracil derivatives on their boundary molecular orbital energies (E LUMO, E HOMO ) and HOMO-LUMO gap (Δ EHOMO-LUMO ) were studied. Changes in the structure of compounds (nitrogenous base type, number of substitutions, hydroxy group position in the benzoyl radical) were analyzed to reveal the relationships between the activity and quantum chemical parameters using methods of the artificial neural network modeling carried out with the aid of StatSoft software. It is established that the dependence of antiglycation activity on E LUMO , E HOMO , and Δ EHOMO-LUMO values can be modeled with high precision by a correlation neural network using multiple range correlation analysis with additional information on the variable fragments of molecular structure so that that this activity of N -hydroxybenzoyl-substituted thymine and uracil can be predicted based on the corresponding molecular orbital energies.