Predicting anticonvulsant activity of benzamides/benzylamines: computational approach using topological descriptors

The relationship of Wiener's index (a distance-based topological descriptor), Zagreb group parameter (an adjacency-based topological descriptor) and eccentric connectivity index (an adjacency-cum-distance-based topological descriptor) with the anticonvulsant activity of a series of substituted benazamides/benzylamines has been investigated. A training set comprising 41 analogues of substituted benazamides/benzylamines was selected for the present investigations. The values of the Wiener's index, Zagreb group parameter and eccentric connectivity index and of each of 41 analogues comprising the data set were computed and active ranges were identified. Subsequently, a biological activity was assigned to each analogue involved in the data set which was then compared with the reported anticonvulsant activity. An exceptionally high accuracy of predictions ranging from a minimum of approximately 88% for the Zagreb group parameter to a maximum of approximately 97% for Wiener's index were obtained.