Publikace na Přírodovědecká fakulta |
2017
Abstrakt
The CORAL software based on the optimal descriptors calculated with simplified molecular input-line entry system was used to build up quantitatively structure-activity relationship for the minimal inhibition concentrations against (i) Mycobacterium tuberculosis H37RV ATCC 27294 and (ii) M. tuberculosis drug resistant clinical isolate. The predictive potential of these models has been checked up with three random splits into the training, calibration, and validation sets.
Although all models are quantitative, it has been found that splitting have apparent influence on the statistical quality of these models. Thus, the thesis "QSAR is a random event" is confirmed in this work.