Background: Tuberculosis is one of the leading causes of deaths due to infectious disease worldwide. There is an urgent need for developing new drugs due to the rising incidents of drug resistance. Previously, triazole molecules showing antitubercular activity, were reported. Various computational tools pave the way for a rational approach to understanding the structural importance of these compounds in inhibiting the growth of Mycobacterium Tuberculosis. Objective: The aim of this study is to develop and compare two different QSAR models based on a set of previously reported triazole molecules and use the best one for gaining structural insights into those molecules. Methods: In this current study, two separate models were made with CoMFA and CoMSIA descriptors based on a dataset of triazole molecules showing antitubercular activity. Several one dimensional (1D) descriptors were added to each of the models and the validation results and contour data generated from them were compared. The best model was analysed to give a detailed understanding of the triazole molecules and their role in the antitubercular activity. Results: The r2, q2, predicted r2 and SEP (Standard error of prediction) for the CoMFA model were 0.866, 0.573, 0.119 and 0.736 respectively and for the CoMSIA model, the r2, q2, predicted r2 and SEP were calculated to be 0.998, 0.634, 0.013 and 0.869 respectively. Although both the QSAR models produced acceptable internal and external validation scores, but the CoMSIA results were significantly better. The CoMSIA contours also provided a better match than CoMFA with most of the features of the active compound 30b. Hence, the CoMSIA model was chosen and its contours were explored for gaining structural insights into the triazole molecules. Conclusion: The CoMSIA contours helped us understand the role of several atoms and groups of the triazole molecules in their biological activity. The possibilities for substitution in the triazole compounds that would enhance the activity were also analyzed. Thus, this study paves the way for designing new antitubercular drugs in future.
All Science Journal Classification (ASJC) codes
- Molecular Medicine
- Drug Discovery