scholarly journals HIV-1 drug resistance profiling using amino acid sequence space cartography

2021 ◽  
Author(s):  
Karina Pikalyova ◽  
Alexey Orlov ◽  
Arkadii Lin ◽  
Olga Tarasova ◽  
Gilles Marcou ◽  
...  
Keyword(s):  
Drug Resistance ◽  
Computational Methods ◽  
Sequence Data ◽  
High Accuracy ◽  
Generative Topographic Mapping ◽  
Hiv Drug Resistance ◽  
Treatment Regimens ◽  
Depth Analysis ◽  
Drug Resistance Prediction ◽  
Resistance Prediction

Motivation: Human immunodeficiency virus (HIV) drug resistance is a global healthcare issue. The emergence of drug resistance demands treatment adaptation. Computational methods predicting the drug resistance profile from genomic data of HIV isolates are advantageous for monitoring drug resistance in patients. Yet, the currently existing computational methods for drug resistance prediction are either not suitable for complex mutational patterns in emerging HIV strains or lack interpretability of prediction results which is of paramount importance in clinical practice. Hence, to overcome these limitations, new approaches for the HIV drug resistance prediction combining high accuracy and interpretability are required. Results: In this work, a new methodology for the analysis of protein sequence data based on the application of generative topographic mapping was developed and applied for HIV drug resistance profiling. It allowed achieving high accuracy of resistance predictions and intuitive interpretation of prediction results. The developed approach was successfully applied for the prediction of HIV re-sistance towards protease, reverse-transcriptase and integrase inhibitors and in-depth analysis of HIV resistance-inducing mutation patterns. Hence, it can serve as an efficient and interpretable tool to suggest optimal treatment regimens. Availability: https://github.com/karinapikalyova/ISIDASeq

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.

scholarly journals Evaluation of whole-genome sequence data analysis approaches for short- and long-read sequencing of Mycobacterium tuberculosis

2021 ◽  
Vol 7 (11) ◽  
Author(s):  
Nilay Peker ◽  
Leonard Schuele ◽  
Nienke Kok ◽  
Miguel Terrazos ◽  
Stefan M. Neuenschwander ◽  
...  
Keyword(s):  
Drug Resistance ◽  
Mycobacterium Tuberculosis ◽  
Core Genome ◽  
Sequence Data ◽  
Complete Agreement ◽  
Whole Genome ◽  
First Line ◽  
Long Read ◽  
Drug Resistance Prediction ◽  
Resistance Prediction

Whole-genome sequencing (WGS) of Mycobacterium tuberculosis (MTB) isolates can be used to get an accurate diagnosis, to guide clinical decision making, to control tuberculosis (TB) and for outbreak investigations. We evaluated the performance of long-read (LR) and/or short-read (SR) sequencing for anti-TB drug-resistance prediction using the TBProfiler and Mykrobe tools, the fraction of genome recovery, assembly accuracies and the robustness of two typing approaches based on core-genome SNP (cgSNP) typing and core-genome multi-locus sequence typing (cgMLST). Most of the discrepancies between phenotypic drug-susceptibility testing (DST) and drug-resistance prediction were observed for the first-line drugs rifampicin, isoniazid, pyrazinamide and ethambutol, mainly with LR sequence data. Resistance prediction to second-line drugs made by both TBProfiler and Mykrobe tools with SR- and LR-sequence data were in complete agreement with phenotypic DST except for one isolate. The SR assemblies were more accurate than the LR assemblies, having significantly (P<0.05) fewer indels and mismatches per 100 kbp. However, the hybrid and LR assemblies had slightly higher genome fractions. For LR assemblies, Canu followed by Racon, and Medaka polishing was the most accurate approach. The cgSNP approach, based on either reads or assemblies, was more robust than the cgMLST approach, especially for LR sequence data. In conclusion, anti-TB drug-resistance prediction, particularly with only LR sequence data, remains challenging, especially for first-line drugs. In addition, SR assemblies appear more accurate than LR ones, and reproducible phylogeny can be achieved using cgSNP approaches.

scholarly journals From Sequence Data to Patient Result: A Solution for HIV Drug Resistance Genotyping With Exatype, End to End Software for Pol-HIV-1 Sanger Based Sequence Analysis and Patient HIV Drug Resistance Result Generation

2020 ◽  
Vol 19 ◽  
pp. 232595822096268
Author(s):  
Leonard Kingwara ◽  
Muthoni Karanja ◽  
Catherine Ngugi ◽  
Geoffrey Kangogo ◽  
Kipkerich Bera ◽  
...  
Keyword(s):  
Drug Resistance ◽  
Sequence Analysis ◽  
Standard Method ◽  
Sequence Data ◽  
Scale Up ◽  
Reference Laboratory ◽  
Hiv Viral Load ◽  
Hiv Drug Resistance ◽  
Base Calling ◽  
Hands On

Introduction: With the rapid scale-up of antiretroviral therapy (ART) to treat HIV infection, there are ongoing concerns regarding probable emergence and transmission of HIV drug resistance (HIVDR) mutations. This scale-up has to lead to an increased need for routine HIVDR testing to inform the clinical decision on a regimen switch. Although the majority of wet laboratory processes are standardized, slow, labor-intensive data transfer and subjective manual sequence interpretation steps are still required to finalize and release patient results. We thus set out to validate the applicability of a software package to generate HIVDR patient results from raw sequence data independently. Methods: We assessed the performance characteristics of Hyrax Bioscience’s Exatype (a sequence data to patient result, fully automated sequence analysis software, which consolidates RECall, MEGA X and the Stanford HIV database) against the standard method (RECall and Stanford database). Exatype is a web-based HIV Drug resistance bioinformatic pipeline available at sanger. exatype.com . To validate the exatype, we used a test set of 135 remnant HIV viral load samples at the National HIV Reference Laboratory (NHRL). Result: We analyzed, and successfully generated results of 126 sequences out of 135 specimens by both Standard and Exatype software. Result production using Exatype required minimal hands-on time in comparison to the Standard (6 computation-hours using the standard method versus 1.5 Exatype computation-hours). Concordance between the 2 systems was 99.8% for 311,227 bases compared. 99.7% of the 0.2% discordant bases, were attributed to nucleotide mixtures as a result of the sequence editing in Recall. Both methods identified similar (99.1%) critical antiretroviral resistance-associated mutations resulting in a 99.2% concordance of resistance susceptibility interpretations. The Base-calling comparison between the 2 methods had Cohen’s kappa (0.97 to 0.99), implying an almost perfect agreement with minimal base calling variation. On a predefined dataset, RECall editing displayed the highest probability to score mixtures accurately 1 vs. 0.71 and the lowest chance to inaccurately assign mixtures to pure nucleotides (0.002–0.0008). This advantage is attributable to the manual sequence editing in RECall. Conclusion: The reduction in hands-on time needed is a benefit when using the Exatype HIV DR sequence analysis platform and result generation tool. There is a minimal difference in base calling between Exatype and standard methods. Although the discrepancy has minimal impact on drug resistance interpretation, allowance of sequence editing in Exatype as RECall can significantly improve its performance.

Sign in / Sign up

Export Citation Format

Share Document