VALIDATION OF COMPUTATIONAL METHODS APPLIED IN MOLECULAR MODELING OF CAFFEINE WITH EPITHELIAL ANTICANCER ACTIVITY: THEORETICAL STUDY OF GEOMETRIC, THERMOCHEMICAL AND SPECTROMETRIC DATA

Models validation in QSAR, pharmacophore, docking, and others, can ensure the accuracy and reliability of future predictions in design and selection of molecules with biological activity. In these study, the caffeine molecule was optimized using Hartree-Fock (HF) and Density Functional Theory (DFT/B3LYP) methods, with seven basis sets. Linear correlation data, errors and RMSD values between theoretical and experimental data allowed us to classify the methods and basis sets for evaluating their correspondence with experimental data (geometric parameters, heat capacity, as well as Infrared, Raman and NMR spectra). The HF method has shown the highest correspondence with the experimental data, occupying the top-five rank of the general classification. The HF/6-31G** method was the best one classified and it can be used to model the biological activity of the caffeine molecule.