Functional analysis of toxins by clustering of their electrostatic potentials

Abstract

This is the first time, to our knowledge, that a structural description of the function of protein toxins has been addressed from a new perspective living behind the traditional approach of using common characteristic like α-helixes, β-sheets, and others, and instead using the electrostatic surface potential (ESP). These potentials have been normally used for 3D visual representation of charge distribution, qualitative interpretations of electrophilic and nucleophilic reactions, and for molecular interactions because of the considerable computational time and effort it takes to calculate them. We calculate the ESP for 16 proteins toxins and compared the 3D shapes of their potentials using the Hausdorff-Gromov distance; then we ran k-means cluster analysis to determine the relation among the shapes of the ESPs. Our results show that the analysis was able to cluster toxins depending on the shape of their ESP and that this clustering is related to the toxins function. There were only 4 toxins that clustered in a different group according to their function, and one that did not cluster with any other group. There was no evidence that taxonomy has a relation with the clusters found.