Related Experiment Video
Updated: May 14, 2025
19:57
An Affordable HIV-1 Drug Resistance Monitoring Method for Resource Limited Settings
Published on: March 30, 2014
18.4K
Threshold sensitivity analysis for HIV-1 transmission cluster detection using different genomic regions and subtypes
Billal M Obeng1, Roger D Kouyos2, Katharina Kusejko2
1The Kirby Institute, UNSW Sydney, Sydney, Australia.
Virology
|May 6, 2025
Summary
HIV-1 cluster analysis helps track transmission, but genome region choice impacts results. The pol region provides robust phylogenetic clustering, ideal for near real-time detection.
Area of Science:
- Molecular epidemiology
- Virology
- Bioinformatics
Background:
- HIV-1 cluster analysis is crucial for understanding transmission dynamics and informing prevention strategies.
- Phylogenetic clustering outputs are sensitive to the specific HIV-1 genome regions analyzed.
- Optimizing cluster analysis requires evaluating different genetic distance thresholds and data sources.
Purpose of the Study:
- To conduct a sensitivity analysis of HIV-1 cluster detection using varying genetic distance thresholds and genome regions.
- To identify the most cohesive clustering outputs across different data sources and sub-genomic regions.
- To determine the optimal genetic distance threshold for reliable HIV-1 cluster analysis and near real-time detection.
Main Methods:
- Utilized full-length subtype B (Swiss HIV Cohort Study) and subtype C (South Africa) HIV-1 sequences.
- Divided datasets into sub-genomic regions (gag, pol, env), with pol further subdivided (pr-rt, rt-int, pr-rt-int).
- Performed cluster analyses using ClusterPicker with distance thresholds (0.5%-4.5%) and branch support (70%, 90%, 99%), comparing tree topologies and cluster similarity.
Main Results:
- Phylogenies using the pol region (pr-rt-int, rt-int) exhibited more robust tree topologies than gag and env.
- Cluster composition varied with increasing genetic distance thresholds but was unaffected by branch support.
- Optimal cluster identity and similarity were observed around a 2.5% (±0.5%) genetic distance for all sub-genomic regions and subtypes.
Conclusions:
- Sensitivity analysis is valuable for selecting appropriate genetic distance thresholds in HIV-1 cluster analysis.
- The pol region is suitable for robust HIV-1 phylogenetic clustering and can be employed for near real-time detection.
- Consistent clustering results across subtypes B and C highlight the generalizability of findings regarding the pol region and optimal distance threshold.
