A fuzzy based approach for prediction of biological activities of HIV -1 protease inhibitor compounds

Abstract

A neuro-fuzzy based approach for predicting the HIV-1 inhibitor compounds activity has been designed when the molecular descriptor attributes of the compound are known. Standard Fuzzy ARTMAP (FAM) is provided with 196 data sets which is divided into 176 training data and 20 test data. The normalized data is given to the FAM network and the result indicates whether the compound is a suitable inhibitor or not.The result analysis is done with/ without GA for the dataset and GA-FAMR algorithm is used to optimize the relevance's assigned to the training data and the accuracy obtained is 93.09%.