Automated prediction of HIV drug resistance from genotype data

Abstract

BackgroundHIV/AIDS is a serious threat to public health. The emergence of drug resistance mutations diminishes the effectiveness of drug therapy for HIV/AIDS. Developing a computational prediction of drug resistance phenotype will enable efficient and timely selection of the best treatment regimens.ResultsA unified encoding of protein sequence and structure was used as the feature vector for predicting phenotypic resistance from genotype data. Two machine learning algorithms, Random Forest and K-nearest neighbor, were used. The prediction accuracies were examined by five-fold cross-validation on the genotype-phenotype datasets. A supervised machine learning approach for automatic prediction of drug resistance was developed to handle genotype-phenotype datasets of HIV protease (PR) and reverse transcriptase (RT). It predicts the drug resistance phenotype and its relative severity from a query sequence. The accuracy of the classification was higher than 0.973 for eight PR inhibitors and 0.986 for ten RT inhibitors, respectively. The overall cross-validated regression R2-values for the severity of drug resistance were 0.772–0.953 for 8 PR inhibitors and 0.773–0.995 for 10 RT inhibitors.ConclusionsMachine learning using a unified encoding of sequence and protein structure as a feature vector provides an accurate prediction of drug resistance from genotype data. A practical webserver for clinicians has been implemented.

Analysis of drug resistance in HIV protease

“…The nearly uniform values of close to 1.0 for accuracy, PPV, recall, and F-score, show that the models reliably predict both resistant and non-resistant classes. These results compare favorably with our earlier results using non-generative machine learning algorithms [4–6, 8]. …”

A Review on Applications of Computational Methods in Drug Screening and Design

“…Based on ligand-based virtual screening, kNN can be viewed as a prolongation form chemical similarity searching to supervised learning, and the top search results predicted the best bioactivities. Weber et al [79] tried two machine learning algorithms of classification (KNN and RF) to analyze genotype-phenotype datasets of HIV protease and reverse transcriptase (RT). As a result, both algorithms had high accuracies for predicting the drug resistance for protease and RT inhibitors.…”

Improving fold resistance prediction of HIV-1 against protease and reverse transcriptase inhibitors using artificial neural networks

“…Further, the dataset was randomly divided into 5 subsets of approximately equal size, and 5 different ANNs were trained on datasets that comprised 4 of the 5 subsets, and then 5 different R 2 values were calculated; we then calculated the mean and the standard deviation of these 5 R 2 values. Regression performances were then compared against prediction models from the article published in 2016 by Shen and co-workers [ 19 ], in which regression machine learning models, namely the Random Forest and the K-nearest neighbor algorithms were used. The raw dataset used in this work and in ref.…”

“…Our method incorporated the following features: (a) The prediction algorithm used was a regression Artificial Neural Network (ANN); (b) because the great majority of publicly available data in the Stanford HIVdb is for subtype B HIV, only subtype B data was used in this database to train and test the network, so that the prediction algorithm is mainly applicable to subtype B sequence data; (c) in order to reduce data noise, various forms of data filtering, as described in the Methodology section, were used. Our regression ANN models compared favourably against recent work by Shen and co-workers [ 19 ], for which similar metrics were used. The ANN regression models were applied to the protease (PR) inhibitors fosamprenavir (FPV), atazanavir (ATV), indinavir (IDV), lopinavir (LPV), saquinavir (SQV), tipranavir (TPV), nelfinavir (NFV) and darunavir (DRV), and to the reverse transcriptase (RT) inhibitors lamivudine (3TC), abacavir (ABC), zidovudine (AZT), stavudine (D4T), didanosine (DDI), tenofovir (TDF), efavirenz (EFV), etravirine (ETR), nevirapine (NVP), rilpivirine (RPV).…”

“…As stated in [ 17 ], to date the most widely used ones are the HIVdb algorithm [ 14 ] and the support vector machine-based geno2pheno tool [ 16 , 18 ]. More recent work has applied different machine learning approaches for drug resistance prediction, for instance multi-label classification [ 17 ], K-Nearest Neighbor and Random Forests [ 19 ], sparse signal representations coupled to Delaunay triangulation [ 20 , 21 ] and Support Vector Machines variants [ 22 ], some of which are based on sequence information, while others also utilise protein structural information.…”