scholarly journals Multilabel classification for exploiting cross-resistance information in HIV-1 drug resistance prediction

2013 ◽  
Vol 29 (16) ◽  
pp. 1946-1952 ◽  
Author(s):  
Dominik Heider ◽  
Robin Senge ◽  
Weiwei Cheng ◽  
Eyke Hüllermeier
Keyword(s):  
Drug Resistance ◽  
Cross Resistance ◽  
Multilabel Classification ◽  
Drug Resistance Prediction ◽  
Resistance Prediction ◽  
Hiv 1

Algorithm evolution for drug resistance prediction: comparison of systems for HIV-1 genotyping

2015 ◽  
Vol 20 (6) ◽  
pp. 661-665 ◽  
Author(s):  
Sarah Wagner ◽  
◽  
Mario Kurz ◽  
Thomas Klimkait
Keyword(s):  
Drug Resistance ◽  
Drug Resistance Prediction ◽  
Resistance Prediction ◽  
Hiv 1

scholarly journals Drug Resistance Prediction Using Deep Learning Techniques on HIV-1 Sequence Data

Viruses ◽  
2020 ◽  
Vol 12 (5) ◽  
pp. 560
Author(s):  
Margaret C. Steiner ◽  
Keylie M. Gibson ◽  
Keith A. Crandall
Keyword(s):  
Neural Network ◽  
Drug Resistance ◽  
Deep Learning ◽  
Sequence Data ◽  
Classification Performance ◽  
Resistance Mutations ◽  
Evolutionary Mechanisms ◽  
Drug Resistance Prediction ◽  
Resistance Prediction ◽  
Hiv 1

The fast replication rate and lack of repair mechanisms of human immunodeficiency virus (HIV) contribute to its high mutation frequency, with some mutations resulting in the evolution of resistance to antiretroviral therapies (ART). As such, studying HIV drug resistance allows for real-time evaluation of evolutionary mechanisms. Characterizing the biological process of drug resistance is also critically important for sustained effectiveness of ART. Investigating the link between “black box” deep learning methods applied to this problem and evolutionary principles governing drug resistance has been overlooked to date. Here, we utilized publicly available HIV-1 sequence data and drug resistance assay results for 18 ART drugs to evaluate the performance of three architectures (multilayer perceptron, bidirectional recurrent neural network, and convolutional neural network) for drug resistance prediction, jointly with biological analysis. We identified convolutional neural networks as the best performing architecture and displayed a correspondence between the importance of biologically relevant features in the classifier and overall performance. Our results suggest that the high classification performance of deep learning models is indeed dependent on drug resistance mutations (DRMs). These models heavily weighted several features that are not known DRM locations, indicating the utility of model interpretability to address causal relationships in viral genotype-phenotype data.

Impact of genotypic diversity on selection of subtype-specific drug resistance profiles during raltegravir-based therapy in individuals infected with B and BF recombinant HIV-1 strains

2020 ◽  
Vol 75 (6) ◽  
pp. 1567-1574
Author(s):  
Daniela Sánchez ◽  
Solange Arazi Caillaud ◽  
Ines Zapiola ◽  
Silvina Fernandez Giuliano ◽  
Rosa Bologna ◽  
...  
Keyword(s):  
Drug Resistance ◽  
Current Knowledge ◽  
Phylogenetic Analyses ◽  
Genotypic Diversity ◽  
Resistance Testing ◽  
Cross Resistance ◽  
Resistance Mutations ◽  
Middle Income ◽  
Subtype B ◽  
Hiv 1

Abstract Background Current knowledge on HIV-1 resistance to integrase inhibitors (INIs) is based mostly on subtype B strains. This contrasts with the increasing use of INIs in low- and middle-income countries, where non-B subtypes predominate. Materials and methods HIV-1 drug resistance genotyping was performed in 30 HIV-1-infected individuals undergoing virological failure to raltegravir. Drug resistance mutations (DRMs) and HIV-1 subtype were characterized using Stanford HIVdb and phylogenetic analyses. Results Of the 30 integrase (IN) sequences, 14 were characterized as subtype F (47%), 8 as subtype B (27%), 7 as BF recombinants (23%) and 1 as a putative CRF05_DF (3%). In 25 cases (83%), protease and reverse transcriptase (PR-RT) sequences from the same individuals confirmed the presence of different BF recombinants. Stanford HIVdb genotyping was concordant with phylogenetic inference in 70% of IN and 60% of PR-RT sequences. INI DRMs differed between B and F IN subtypes, with Q148K/R/H, G140S and E138K/A being more prevalent in subtype B (63% versus 0%, P = 0.0021; 50% versus 0%, P = 0.0096; and 50% versus 0%, P = 0.0096, respectively). These differences were independent of the time on raltegravir therapy or viral load at the time of genotyping. INI DRMs in subtype F IN genomes predicted a lower level of resistance to raltegravir and no cross-resistance to second-generation INIs. Conclusions Alternative resistance pathways to raltegravir develop in subtypes B and F IN genomes, with implications for clinical practice. Evaluating the role of HIV-1 subtype in development and persistence of mutations that confer resistance to INIs will be important to improve algorithms for resistance testing and optimize the use of INIs.

Sign in / Sign up

Export Citation Format

Share Document