Impacts and shortcomings of genetic clustering methods for infectious disease outbreaks

Poon, Art F. Y.

doi:10.1093/ve/vew031

Cited by 82 publications

(87 citation statements)

References 61 publications

(90 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yet, drug resistance rates are stabilizing or decreasing due to more efficacious and tolerable regimens, particularly in developed countries, resulting in a nearly 10 normal life expectancy for treated HIV-1 infected individuals [6,7].…”

Section: Introductionmentioning

confidence: 99%

The impact of HIV-1 within-host evolution on transmission dynamics

Theys

Libin

Pineda-Peña

et al. 2017

Preprint

View full text Add to dashboard Cite

The adaptive potential of HIV-1 is a vital mechanism to evade host immune responses and antiviral treatment. However, high evolutionary rates during persistent infection can impair transmission efficiency and alter disease progression in the new host, resulting in a delicate trade-off between within-host virulence and between-host infectiousness. This trade-off is visible in the disparity in evolutionary rates at within-host and between-host levels, and preferential transmission of ancestral donor viruses. Understanding the impact of withinhost evolution for epidemiological studies is essential for the design of preventive and therapeutic measures. Herein, we review recent theoretical and experimental work that generated new insights into the complex link between within-host evolution and between-host fitness, revealing temporal and selective processes underlying the structure and dynamics of HIV-1 transmission.$ Contributed equally

show abstract

Section: Introductionmentioning

confidence: 99%

The impact of HIV-1 within-host evolution on transmission dynamics

Theys

Libin

Pineda-Peña

et al. 2017

Preprint

View full text Add to dashboard Cite

show abstract

“…However, it is important to consider the distribution 612 of sampling times, which can drive clustering artificially. This is especially pertinent for transmission 613 dynamic studies, where clustering is often driven by heterogeneity in sampling rates across 614 subpopulations rather than heterogeneity in transmission rates (Poon 2016;McCloskey and Poon 615 2017). PhyCLIP can be applied to time-resolved phylogenies in heterochronous datasets.…”

mentioning

confidence: 99%

Phylogenetic Clustering by Linear Integer Programming (PhyCLIP)

Han

Parker

Scholer

et al. 2018

Preprint

View full text Add to dashboard Cite

249/250 words)20 Sub-species nomenclature systems of pathogens are increasingly based on sequence data. The use of 21 phylogenetics to identify and differentiate between clusters of genetically similar pathogens is 22 particularly prevalent in virology from the nomenclature of human papillomaviruses to highly pathogenic 23 avian influenza (HPAI) H5Nx viruses. These nomenclature systems rely on absolute genetic distance 24 thresholds to define the maximum genetic divergence tolerated between viruses designated as closely 25 related. However, the phylogenetic clustering methods used in these nomenclature systems are limited 26 by the arbitrariness of setting intra-and inter-cluster diversity thresholds. The lack of a consensus 27 ground truth to define well-delineated, meaningful phylogenetic subpopulations amplifies the difficulties 28 in identifying an informative distance threshold. Consequently, phylogenetic clustering often becomes 29 an exploratory, ad-hoc exercise. 30Phylogenetic Clustering by Linear Integer Programming (PhyCLIP) was developed to provide a 31 statistically-principled phylogenetic clustering framework that negates the need for an arbitrarily-defined 32 distance threshold. Using the pairwise patristic distance distributions of an input phylogeny, PhyCLIP 33 parameterises the intra-and inter-cluster divergence limits as statistical bounds in an integer linear 34 programming model which is subsequently optimised to cluster as many sequences as possible. When 35 applied to the haemagglutinin phylogeny of HPAI H5Nx viruses, PhyCLIP was not only able to 36 recapitulate the current WHO/OIE/FAO H5 nomenclature system but also further delineated informative 37 higher resolution clusters that capture geographically-distinct subpopulations of viruses. PhyCLIP is 38 pathogen-agnostic and can be generalised to a wide variety of research questions concerning the 39 identification of biologically informative clusters in pathogen phylogenies. PhyCLIP is freely available at 40

show abstract

“…However, HIV-TRACE still requires users to 460 specify an arbitrary absolute distance threshold. Additionally, while it performed better than 461 other phylogenetic clustering method, HIV-TRACE did not preclude problems with bias 462 towards higher sampling rates (Poon 2016). 463 464 Despite the limitations discussed above, clustering results generated by Phydelity for the 465 simulation and empirical datasets in this study demonstrate its superior performance over 466 current widely used phylogenetic clustering methods.…”

Section: Computational Performance 373mentioning

confidence: 86%

Inferring putative transmission clusters with Phydelity

Han

Parker

Maurer‐Stroh

et al. 2018

Preprint

View full text Add to dashboard Cite

Current phylogenetic clustering approaches for identifying pathogen transmission 18 clusters are centrally limited by their dependency on arbitrarily-defined genetic distance 19 thresholds for within-cluster divergence. Incomplete knowledge of a pathogen's underlying 20 dynamics often reduces the choice of distance threshold to an exploratory, ad-hoc exercise 21 that is difficult to standardise across studies. Phydelity is a new tool for the identification of 22 transmission clusters in pathogen phylogenies. It identifies groups of sequences that are more 23 closely-related than the ensemble distribution of the phylogeny under a statistically-24 principled and phylogeny-informed framework, without the introduction of arbitrary distance 25 thresholds. Relative to other distance threshold-based and model-based methods, Phydelity 26 outputs clusters with higher purity and lower probability of misclassification in simulated 27 phylogenies. Applying Phydelity to empirical datasets of hepatitis B and C virus infections 28showed that Phydelity identified clusters with better correspondence to individuals that are 29 more likely to be linked by rapid transmission events relative to other widely-used 30 phylogenetic clustering methods without the need for parameter calibration. Phydelity is 31 generalisable to any pathogen and can be used to reliably identify putative direct transmission 32 events. Phydelity is freely available at https://github.com/alvinxhan/Phydelity. 33

show abstract

Impacts and shortcomings of genetic clustering methods for infectious disease outbreaks

Cited by 82 publications

References 61 publications

The impact of HIV-1 within-host evolution on transmission dynamics

The impact of HIV-1 within-host evolution on transmission dynamics

Phylogenetic Clustering by Linear Integer Programming (PhyCLIP)

Inferring putative transmission clusters with Phydelity

Contact Info

Product

Resources

About