The major concerns for current TB treatment are its latency, co-infection with HIV, poor patient compliance, and drug resistance issues. Therefore, it is an imperative need to develop novel anti-tubercular drugs that can be equally effective against Mycobacterium tuberculosis and drug resistant strains, and shorten the duration of therapy. The antitubercular activity of pyridine containing compounds is being investigated from early days. We have optimized pharmacophoric requirements with 2D & 3D QSAR studies using V Life Molecular Design Suite software. Statistical methods like multiple linear regression (MLR), principle component regression (PCR), and partial least square regression (PLSR) technique were used as model building methods. In 2D QSAR, we found that partial least square regression method showed best statistical result when compared with other methods. The 3D QSAR model was developed by kohonen Nearest Neighbour Molecular Field Analysis (kNN MFA) method using simulated annealing as a variable selection method. Generation of rigorously validated QSAR model is important to ensure that the model have acceptable predictive power. With the help of QSAR results optimization of pharmacophore was carried out to identify key structural fragments required around pharmacophore for antitubercular activity. Results of QSAR studies yielded good, reliable models having scientifically acceptable, predictive ability i.e. the satisfactory statistical values of r2 greater than 0.7 and q2 greater than 0.5 etc.
Loading....