ANN-based Drug-isolate-fold-change model predicting the resistance profiles of HIV-1 protease inhibitors

Abstract

: Drug resistance is a primary barrier to effective treatments of HIV/AIDS. Calculating quantitative relations between genotype and phenotype observations for each inhibitor with cell-based assays requires time and money consuming experiments. Machine learning models are good options for tackling these problems by generalizing the available data with suitable linear or nonlinear mappings. The main aim of this paper is to construct drug isolate fold change (DIF)-based artificial neural network (ANN) models for estimating the resistance potential of molecules inhibiting the HIV-1 protease (PR) enzyme. Throughout the study, seven of eight protease inhibitors (PIs) have been included in the training set and the remaining ones in the test set. Using the 7-in 1-out procedure, eight ANN models have been produced to measure the learning capacity of models from the descriptors of the inhibitors. The mean value of eight ANN models for unseen inhibitors is and 95% confidence interval (CI) is Predicting the fold change resistance for hundreds of isolates allowed for robust comparison of drug pairs. These eight models have predicted the drug resistance tendencies of each inhibitor pair with the mean 2D correlation coefficient 0.933 and 95% CI A classification problem has been created to predict the ordered relationship of the PIs and the mean accuracy, sensitivity and specificity values are obtained as 0.954, 0.791 and 0.791, respectively. The currently derived ANN models can accurately predict the drug resistance tendencies of PI pairs, and this observation could help test new inhibitors with various isolates.