Extended Data Table 2 HIV-1 evolutionary rate calculated after removing hypermutated haplotypes

From: Persistent HIV-1 replication maintains the tissue reservoir during therapy

  1. We used a linear regression model to estimate the evolutionary rate for the Gag or Pol regions (protease (Pol1) and reverse transcriptase (Pol2)) of HIV-1 within lymph node or peripheral blood of each subject. We calculated the slope (μ) of the linear regression between time and the direct pairwise genetic distances (number of substitutions per site per month) from the most common haplotype at day 0 for each subject. Haplotypes found to harbour G-to-A hypermutated sequences were removed from the analysis to limit the effect of inactivating mutations on the estimates. There is very strong evidence, now presented in numerous studies17, that acute HIV-1 infections are largely founded by a single (or at best a few) viral strains; this initial bottleneck is followed by rapid population diversification in the absence of treatment. The structures of our intra-host phylogenies at day 0 support this pattern. Even if multiple populations are transmitted and are able to establish infection and co-circulate, Bayesian phylodynamics models can properly account for this and estimate evolutionary rates accurately.
Back to article page