A non-parametric analytic framework for within-host viral phylogenies and a test for HIV-1 founder multiplicity

Lewitus, Éric; Rolland, Morgane

doi:10.1093/ve/vez044

Cited by 16 publications

(28 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…S9 and S10 ) ( 33 – 35 ). We estimated phylogenetic η ( 36 , 37 ) at each internal node of the SARS-CoV-2 phylogeny reconstructed from subsampled (10%) alignments and compared the distribution of estimates through time to phylogenies simulated under models of neutral and positive time-dependent rates (b(t) = be α(t) ). Simulation analyses demonstrated that this metric was robust against sampling fraction ( SI Appendix , Fig.…”

Section: Resultsmentioning

confidence: 99%

“…Time-dependent estimates of phylogenetic diversification were measured by extracting the branches descending from each internal node (above the root) of each phylogeny and calculating the peak height (η) of the spectral density profile of the graph Laplacian of each subtree, which is a measure of the density of branching events ( 36 , 37 ). The code to perform the analysis is available for download at https://www.hivresearch.org/publication-supplements .…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A SARS-CoV-2 vaccine candidate would likely match all currently circulating variants

Dearlove

Lewitus

Bai

et al. 2020

Proc. Natl. Acad. Sci. U.S.A.

Self Cite

212

227

View full text Add to dashboard Cite

The magnitude of the COVID-19 pandemic underscores the urgency for a safe and effective vaccine. Many vaccine candidates focus on the Spike protein, as it is targeted by neutralizing antibodies and plays a key role in viral entry. Here we investigate the diversity seen in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) sequences and compare it to the sequence on which most vaccine candidates are based. Using 18,514 sequences, we perform phylogenetic, population genetics, and structural bioinformatics analyses. We find limited diversity across SARS-CoV-2 genomes: Only 11 sites show polymorphisms in >5% of sequences; yet two mutations, including the D614G mutation in Spike, have already become consensus. Because SARS-CoV-2 is being transmitted more rapidly than it evolves, the viral population is becoming more homogeneous, with a median of seven nucleotide substitutions between genomes. There is evidence of purifying selection but little evidence of diversifying selection, with substitution rates comparable across structural versus nonstructural genes. Finally, the Wuhan-Hu-1 reference sequence for the Spike protein, which is the basis for different vaccine candidates, matches optimized vaccine inserts, being identical to an ancestral sequence and one mutation away from the consensus. While the rapid spread of the D614G mutation warrants further study, our results indicate that drift and bottleneck events can explain the minimal diversity found among SARS-CoV-2 sequences. These findings suggest that a single vaccine candidate should be efficacious against currently circulating lineages.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

A SARS-CoV-2 vaccine candidate would likely match all currently circulating variants

Dearlove

Lewitus

Bai

et al. 2020

Proc. Natl. Acad. Sci. U.S.A.

Self Cite

212

227

View full text Add to dashboard Cite

show abstract

“…Using qualitative (e.g., tree topologies) and quantitative (e.g., intra-host maximum diversity) measures of sequence diversity as previously described [21,[34][35][36][37], we could distinguish two subgroups in our cohort: most participants (28/39) were infected with single HIV-1 founders whereas eleven participants (28%) had multiple HIV-1 founders. To detect infections with multiple founders, we also analyzed spectral density profiles derived from each participant's maximum-likelihood phylogeny [38] [39]. We evaluated founder multiplicity based on the principal eigenvalue test, performed on the eigenvalues calculated from the weighted graph Laplacian of the distance matrix of the phylogeny for each participant (S2 Fig).…”

Section: Hiv-1 Genomes From the First Days Of Hiv-1 Infectionmentioning

confidence: 99%

“…The profiles of eigenvalues for each participant were clustered based on Jensen-Shannon distances [65] using an unbiased hierarchical approach with bootstrap probabilities calculated at each node. We then evaluated the founder multiplicity of each participant based on the principal eigenvalue test [38,39], where the threshold between single and multiple founders was determined by the median criterion (the median principal eigenvalue plus 0.5 x standard deviation squared), the jump criterion (the position of the largest discrepancy between consecutive ranked principal eigenvalues), and the partition criterion (clustering on the principal eigenvalue by partitioning around medoids [66]). Finally, we estimated the spectral density profile summary statistics for each participant: principal eigenvalue (lambda � ), the skewness of the profile (psi), and the peak height of the profile (eta).…”

Section: Identification Of Infections With Multiple Hiv-1 Foundersmentioning

confidence: 99%

Molecular dating and viral load growth rates suggested that the eclipse phase lasted about a week in HIV-1 infected adults in East Africa and Thailand

et al. 2020

Self Cite

View full text Add to dashboard Cite

Most HIV-1 infected individuals do not know their infection dates. Precise infection timing is crucial information for studies that document transmission networks or drug levels at infection. To improve infection timing, we used the prospective RV217 cohort where the window when plasma viremia becomes detectable is narrow: the last negative visit occurred a median of four days before the first detectable HIV-1 viremia with an RNA test, referred below as diagnosis. We sequenced 1,280 HIV-1 genomes from 39 participants at a median of 4, 32 and 170 days post-diagnosis. HIV-1 infections were dated by using sequencebased methods and a viral load regression method. Bayesian coalescent and viral load regression estimated that infections occurred a median of 6 days prior to diagnosis (IQR: 9-3 and 11-4 days prior, respectively). Poisson-Fitter, which analyzes the distribution of hamming distances among sequences, estimated a median of 7 days prior to diagnosis (IQR: 15-4 days) based on sequences sampled 4 days post-diagnosis, but it did not yield plausible results using sequences sampled at 32 days. Fourteen participants reported a high-risk exposure event at a median of 8 days prior to diagnosis (IQR: 12 to 6 days prior). These different methods concurred that HIV-1 infection occurred about a week before detectable viremia, corresponding to 20 days (IQR: 34-15 days) before peak viral load.

show abstract

“…25 It is particularly unclear in this scenario how to distinguish closely-related founder variants that diverged in the donor population from stochastic early mutations separating variants that diverged in the recipient population. A variety of founder identification methods of varying complexity exist, [26][27][28] and do not uniformly agree. Both APOBEC and multivariant transmission are examples of processes that inflate the average pairwise distance be- tween sequences and cause violations of the Poisson assumptions that Poisson-Fitter relies upon.…”

Section: Introductionmentioning

confidence: 99%

Estimating the timing of HIV infection from unmutated sequences

Pankow

Christian

Smith

et al. 2020

Preprint

View full text Add to dashboard Cite

For HIV, the time since infection can be estimated from sequence data for acutely infected samples. One popular approach relies on the star-like nature of phylogenies generated under exponential population growth, and the resulting Poisson distribution of mutations away from the founding variant. However, real-world complications, such as APOBEC hypermutation and multiple-founder transmission, present a challenge to this approach, requiring data curation to remove these signals before reasonable timing estimates may be obtained. Here we suggest a simple alternative approach that derives the timing estimate not from the entire mutational spectrum but from the proportion of sequences that have no mutations. This can be approximated quickly and is robust to phenomena such as multiple founder transmission and APOBEC hypermutation. Our approach is Bayesian, and we adopt a conjugate prior to obtain closed form posterior distributions at negligible computational expense. Using real data and simulations, we show that this approach provides accurate timing estimates and credible intervals without the inconvenience of data curation and is robust to complicating phenomena that can mislead existing approaches or cause them to fail entirely. For immediate use we provide an implementation via Google Sheets, which offers bulk analysis of multiple datasets, as well as more detailed individual-donor analyses. For inclusion in data processing pipelines we provide implementations in three languages: Julia, R, and Python.

show abstract

A non-parametric analytic framework for within-host viral phylogenies and a test for HIV-1 founder multiplicity

Cited by 16 publications

References 58 publications

A SARS-CoV-2 vaccine candidate would likely match all currently circulating variants

A SARS-CoV-2 vaccine candidate would likely match all currently circulating variants

Molecular dating and viral load growth rates suggested that the eclipse phase lasted about a week in HIV-1 infected adults in East Africa and Thailand

Estimating the timing of HIV infection from unmutated sequences

Contact Info

Product

Resources

About