A quantitative structure-activity relationship (QSAR) study has been carried out for a set of 15 quinolone derivatives to correlate and predict the mycobacteria activity against mycobacterium tuberculosis (TB) using Molecular Operating Environment (MOE) program. The statistical regression expressions were obtained using partial least squares (PLS) method which can effectively establish a correlation model between the molecular descriptors and associated properties. A good QSAR model is generated by the training dataset with squared correlation coefficient (r2) = 0.8228, cross validation coefficient (Q2) = 0.9071, standard deviation (s) = 0.16457 and correlation coefficient (r) for the external dataset is 0.9892 while r2 of predicted dataset (test set) is 0.9896. Model obtained was used to predicted the activity against mycobacterium tuberculosis (TB) for a set of 34 designed 3, 7-disubstituted quinolone derivatives. Finally, molecular docking analysis was carried out to better understand the interactions between (A) test, (B) designed compounds and the active site of the mycobacterium tuberculosis (TB) target site.