Detour matrix-based adjacent path eccentric distance sum indices for (Q)SAR/QSPR. Part II: application in development of models for COX-2 inhibitory activity of indomethacin derivatives.

Abstract

In present study, adjacent path eccentric distance sum indices proposed in Part-I of the manuscript were successfully utilised for the development of models for cycloxygenase-2 (COX-2) inhibitory activity. Values of diverse molecular descriptors (MDs) for each of 38 indomethacin analogues involved in the dataset were computed. A total of 55 diverse MDs were ultimately shortlisted for further analysis. The suitable models were developed using decision tree (DT), random forest (RF) and moving average analysis (MAA). The DT identified the proposed topological index (TI)-(A)ξ(3)(PDS) as one of the important indices. The accuracy of prediction of DT, RF and MAA-based models varied from 81.58% to 97.37%. The statistical significance of proposed models was assessed through inter-correlation analysis, sensitivity, specificity, non-error rate and Mathews correlation coefficient. Proposed models offer vast potential for providing lead structures for the development of potent anti-inflammatory agents devoid of COX-1 side effects.