Automated analysis of phylogenetic clusters

Ragonnet‐Cronin, Manon; Hodcroft, Emma B.; Hué, Stéphane; Fearnhill, Esther; Delpech, Valérie; Brown, Andrew J. Leigh; Lycett, Samantha

doi:10.1186/1471-2105-14-317

Cited by 307 publications

(335 citation statements)

References 32 publications

(55 reference statements)

Supporting

Mentioning

320

Contrasting

Order By: Relevance

“…Clusters were identified and enumerated by ClusterPicker. 43 Axis y shows the proportion of HIV-1C V1C5 sequences in clusters. Axis x shows the pairwise distance thresholds for cluster identification from 1% to 15%.…”

Section: Figmentioning

confidence: 99%

Impact of Sampling Density on the Extent of HIV Clustering

NovitskyVlad

Moyo²,

LeiQuanhong³

et al. 2014

AIDS Research and Human Retroviruses

View full text Add to dashboard Cite

Identifying and monitoring HIV clusters could be useful in tracking the leading edge of HIV transmission in epidemics. Currently, greater specificity in the definition of HIV clusters is needed to reduce confusion in the interpretation of HIV clustering results. We address sampling density as one of the key aspects of HIV cluster analysis. The proportion of viral sequences in clusters was estimated at sampling densities from 1.0% to 70%. A set of 1,248 HIV-1C env gp120 V1C5 sequences from a single community in Botswana was utilized in simulation studies. Matching numbers of HIV-1C V1C5 sequences from the LANL HIV Database were used as comparators. HIV clusters were identified by phylogenetic inference under bootstrapped maximum likelihood and pairwise distance cut-offs. Sampling density below 10% was associated with stochastic HIV clustering with broad confidence intervals. HIV clustering increased linearly at sampling density > 10%, and was accompanied by narrowing confidence intervals. Patterns of HIV clustering were similar at bootstrap thresholds 0.7 to 1.0, but the extent of HIV clustering decreased with higher bootstrap thresholds. The origin of sampling (local concentrated vs. scattered global) had a substantial impact on HIV clustering at sampling densities ‡ 10%. Pairwise distances at 10% were estimated as a threshold for cluster analysis of HIV-1 V1C5 sequences. The node bootstrap support distribution provided additional evidence for 10% sampling density as the threshold for HIV cluster analysis. The detectability of HIV clusters is substantially affected by sampling density. A minimal genotyping density of 10% and sampling density of 50-70% are suggested for HIV-1 V1C5 cluster analysis.

show abstract

Section: Figmentioning

confidence: 99%

Impact of Sampling Density on the Extent of HIV Clustering

NovitskyVlad

Moyo²,

LeiQuanhong³

et al. 2014

AIDS Research and Human Retroviruses

View full text Add to dashboard Cite

show abstract

“…Cluster Picker [12] is a Java-based program available at: http://hiv.bio.ed.ac.uk/software.html as input, Cluster Picker requires a phylogenetic tree and sequence data. Clusters are specified using user-specified genetic distance and bootstrap thresholds.…”

Section: Clusteringmentioning

confidence: 99%

“…3 However, genetic distance and branch support cutoffs markedly vary between studies and the rationale for a given cluster definition is rarely specified. [10][11][12][13][14][15] Many factors can influence the choice of genetic distance and branch support cutoff values used to define clusters, such as the spatial and temporal scale of analysis, HIV subtype, the underlying mode of transmission (e.g., heterosexual vs. injection drug use), and the viral genomic region(s) being analyzed. Branch support values may also be affected by the statistical model used to reconstruct the phylogenies (e.g., maximum likelihood, Bayesian) and the total amount of viral genetic diversity in the dataset.…”

Section: Introductionmentioning

confidence: 99%

“…Branch support values may also be affected by the statistical model used to reconstruct the phylogenies (e.g., maximum likelihood, Bayesian) and the total amount of viral genetic diversity in the dataset. 10 In this study, we systematically compare two programs used to detect HIV-1 transmission clusters, Cluster Picker 12 and HIV-TRACE, 10 using three separate HIV-1 sequence datasets from HIV-infected participants in the Rakai Community Cohort Study (RCCS). Two of these datasets were generated using next-generation sequence (NGS) methods and included (1) sequential sequences from the same person and (2) sequences from epidemiologically linked heterosexual couples.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Transmission Clusters with Cluster Picker and HIV-TRACE

Rose¹,

Lamers²,

Dollar³

et al. 2017

AIDS Research and Human Retroviruses

Self Cite

View full text Add to dashboard Cite

We compared the behavior of two approaches (Cluster Picker and HIV-TRACE) at varying genetic distances to identify transmission clusters. We used three HIV gp41 sequence datasets originating from the Rakai Community Cohort Study: (1) next-generation sequence (NGS) data from nine linked couples; (2) NGS data from longitudinal sampling of 14 individuals; and (3) Sanger consensus sequences from a cross-sectional dataset (n = 1,022) containing 91 epidemiologically linked heterosexual couples. We calculated the optimal genetic distance threshold to separate linked versus unlinked NGS datasets using a receiver operating curve analysis. We evaluated the number, size, and composition of clusters detected by Cluster Picker and HIV-TRACE at six genetic distance thresholds (1%-5.3%) on all three datasets. We further tested the effect of using all NGS, versus only a single variant for each patient/time point, for datasets (1) and (2). The optimal gp41 genetic distance threshold to distinguish linked and unlinked couples and individuals was 5.3% and 4%, respectively. HIV-TRACE tended to detect larger and fewer clusters, whereas Cluster Picker detected more clusters containing only two sequences. For NGS datasets (1) and (2), HIV-TRACE and Cluster Picker detected all linked pairs at 3% and 4% genetic distances, respectively. However, at 5.3% genetic distance, 20% of couples in dataset (3) did not cluster using either program, and for >1/3 of couples cluster assignment were discordant. We suggest caution in choosing thresholds for clustering analyses in a generalized epidemic.

show abstract

“…To explore the potential transmission linkage between Yunnan-mIDUs and Chinese IDUs in Dehong, transmission clusters of the HIV-1 sequences were identified using Cluster Picker 1.2.4 [37]. According to the Cluster Picker manual and the suggestions of previous studies, the parameter support threshold was set to 0 and the genetic distance threshold was set to 4% to identify a close transmission relationship between Yunnan-mIDUs and Dehong IDUs [22–24,38].…”

Section: Methodsmentioning

confidence: 99%

Burmese injecting drug users in Yunnan play a pivotal role in the cross-border transmission of HIV-1 in the China-Myanmar border region

Chen

Wang

et al. 2018

Virulence

View full text Add to dashboard Cite

Injecting drug users (IDUs) are the major risk group for HIV-1 infection in the China-Myanmar border area. There are a large number of Burmese IDUs living in Yunnan (Yunnan-mIDUs) who might be associated with the cross-border transmission of HIV-1. From 2010 to 2013, 617 Yunnan-mIDUs were recruited from three counties of Yunnan, 19.0% of whom were detected to be HIV-1 positive by serological testing. Partial HIV-1 p17, pol, vif-env, and env genes were amplified from the positive samples and were sequenced. Phylogenetic and HIV-1 subtyping analyses revealed that HIV-1 recombinant forms (RFs), including RF_BC (36.4%), RF_01BC (26.1%), RF_01C (9.1%) and RF_01B (1.1%), were predominant among this cohort. Of the identified HIV-1 strains, 14.8%, 9.1% and 3.4% belonged to subtype C, CRF01_AE and subtype B, respectively. Transmission cluster analysis showed that sequences from the Yunnan-mIDUs formed transmission clusters not only with those from Burmese IDUs but also with those from Chinese IDUs, indicating that Yunnan-mIDUs might acquire HIV-1 infection from or spread HIV-1 to both Burmese and Chinese IDUs. Phylogeographic analyses revealed three cross-border transmission patterns associated with Yunnan-mIDUs, in which Yunnan-mIDUs served as the crucial nodes linking the Burmese and Chinese IDUs. These results suggest that Yunnan-mIDUs are a potential viral reservoir for the diffusion of HIV-1 in Yunnan and play a pivotal role in the bidirectional cross-border transmission of HIV-1 in the China-Myanmar border region. More intervention efforts that focus on Yunnan-mIDUs are recommended in Yunnan’s campaign against HIV/AIDS.

show abstract

Automated analysis of phylogenetic clusters

Cited by 307 publications

References 32 publications

Impact of Sampling Density on the Extent of HIV Clustering

Impact of Sampling Density on the Extent of HIV Clustering

Identifying Transmission Clusters with Cluster Picker and HIV-TRACE

Burmese injecting drug users in Yunnan play a pivotal role in the cross-border transmission of HIV-1 in the China-Myanmar border region

Contact Info

Product

Resources

About