Prediction of anti-inflammatory activity of anthranylic acids using StructuralMolecular Fragment and topochemical models

摘要

The method of Substructural Molecular Fragments based on the representation of the molecular graph by ensembles of fragments and involving calculations of those contributions to a given property. We also use the relationship between the topochemical indices, Wiener’s index : defined as the sum of all distance between unordered pairs of vertices, Zagreb group parameter M1 and M2: defined as the summation of the squares of chemical degrees over all the vertices an adjacency and eccentric connectivity index : defined as the summation of the product of chemical eccentricity and the chemical degree of each vertex with anthranylic acids has been investigated. A data set comprising of 100 analogues of anthranylic acids was selected for the present study. The values of the Wiener’s index, Zagreb group parameter, and eccentric connectivity index were computed for each of the 100 analogues using an in-house computer program and suitable models were developed after identification of the active ranges. For the first model, the predicted values for the biological activity of the structures in the prediction set are pertinent: the plot of Acal vs. Aobs showed a correlation R2= 0.9175. Subsequently for the second model, each compound was assigned a biological activity using these models, which was then compared with the reported antiflammatory activity. Accuracy of prediction was found to be, ≈86% using models based upon topochemical descriptors.