Novel descriptor generation method for prediction of HIV-1 PR inhibitory of fullerene (C60) analogues
Eslam Pourbasheer
Center of Excellence in Electrochemistry
Faculty of Chemistry
University of Tehran, Tehran, Iran
Abstract:
For the first time a new descriptor generation method has developed for prediction of HIV-1 PR inhibitory of novel fullerene (C60) analogues by quantitative structure activity relationship (QSAR). For generation of descriptors, only a part of molecular structure has been used instead of complete molecular structure. This method takes much lower time in generation of descriptors, comparing with that of general method. The whole data set was divided into a training set and a test set on the basis of K-means clustering technique. A variable selection method of genetic algorithm (GA) was employed to select optimal subset of descriptors. The results of obtained models by new method and general method are compared. The models were validated using leave one out (LOO), leave many out (LMO) crossvalidation, bootstrapping, Y- randomization test and applicability domain (AD). Comparison of the two methods showed that the new method was superior to general method in predicting the HIV-1 PR inhibitory of fullerenes.
