Topological and topographical indices for prediction of solvent accessible surface area in hydrocarbons

Abstract

The topological and topographical indices are broadly used tools in molecular modeling studies. Its computer easiness and high descriptive capacity of the chemical structure confers it a great applicability in the field of chemical-physical and biological properties prediction. On the other hand, the solvent accessible surface area is a theoretical chemical-physical parameter that is calculated, among other, for the delphi method. It is broadly used in relationship studies between the chemical structure and the biological activity, solvatation and lipophilicity studies. A regression model that structural features to the solvent accessible surface area for a sample of 205 aliphatic, aromatic and aliciclic hydrocarbons is presented. As independent variables, different topological and topographical indices were used. The sample was divided in samples of training and validation using cluster analysis, with 163 and 42 compounds respectively. The obtained models for the samples of training and validation shows a correlation coeficient of 0 .997 and 0 .998 with a standard deviation of 7 .575 and 7 .595A2, respectively. This model offers a new alternative for the calculation of the solvent accessible surface area in hydrocarbons as a useful tool for molecular modeling studies.