QSAR study of oxazolidinone antibacterial agents using artificial neural networks
Chengyu Zou, Lu Zhou · Molecular Simulation · 2007
The oxazolidinones antibacterial agents have been studied for their quantitative structure-activity relationships (QSAR). Molecules were represented by constitutional, topostructural, chemical and quantum chemical descriptors. Partial least square (PLS) regression was used to model the relationships between molecular descriptors and biological activity of molecules. The predictive ability of the acquired models was evaluated by the activity prediction of the prediction set compounds. Artificial neural network (ANN) was also employed to model the nonlinear structure-activity relationships. The results showed that the linear model does not perform as well as the nonlinear model in terms of predictive ability.