Determination of the chain termination rate constants of the radical chain oxidation of organic compounds on antioxidant molecules by the QSPR method

Abstract

A quantitative analysis of the structure-antioxidant activity relationship was performed for 128 derivatives of phenols, amines, uracil, benzopyrane, and benzofuran using the GUSAR 2013 program. Nine statistically signifi cant QSAR consensus models characterized by a high accuracy of prediction of the chain termination rate constant of oxidation on the antioxidant molecules were constructed. The results of the structural analysis performed in the GUSAR 2013 program are in good agreement with the literature data.

QSAR Assessing the Efficiency of Antioxidants in the Termination of Radical-Chain Oxidation Processes of Organic Compounds

“…The structures of the phenol, aminophenol and uracil derivatives selected for the QSAR modeling in GUSAR 2013 [ 35 , 36 , 37 , 42 , 43 , 44 , 45 , 46 , 47 ] are shown in Figure 1 . A complete list of compounds with their experimental k 7 values is presented in Supplementary Materials (Table S2) .…”

“…The QSAR models M1–M9 for quantitative prediction of the antioxidant activity of uracil, phenol, and aminophenol derivatives were based on two types of substructural descriptors of atomic neighborhoods: QNA (Quantitative Neighborhood of Atoms) and MNA (Multilevel Neighborhoods of Atoms) [ 35 , 36 , 37 , 42 , 43 , 44 , 45 , 46 , 47 ]. These descriptors are automatically computed in the GUSAR2013 program based on the structural formulas of chemical compounds, taking into account the valence and partial atomic charges.…”

“…Let’s briefly explain the ideology of calculating descriptors in the program GUSAR2013. The ideology of calculating QNA and MNA descriptors is described in detail Supplementary Materials and papers [ 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 ]. Additionally, three descriptors of the whole molecule (topological length, topological volume, and lipophilicity) were used to enhance the descriptive and predictive ability of the models.…”

“…It is important to note that in the final QSAR models, the independent variables include the mean values of the individual two-dimensional Chebyshev polynomials from the P and Q values calculated for all atoms in the molecule. Thus, the regression equations constructed in the GUSAR 2013 program take into account both the specificity and physicochemical properties of each atom entering the training set [ 41 , 43 , 44 , 45 , 46 ]. However, QNA descriptors cannot be physically interpreted due to the peculiarities of their calculation.…”

“…This program for the formation of independent descriptors and their selection at the stage of the formation of regression models uses unique approaches, which we describe in detail in Supplementary Materials . The program has demonstrated its efficiency for modeling various types of biological activity [ 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 ]. It is important to note that GUSAR 2013 allows prompt processing (in a matter of seconds) various structural and physicochemical descriptors, automatic selecting the most significant of them, and develop statistically significant QSAR consensus models.…”