Adaptation to Human Populations Is Revealed by Within-Host Polymorphisms in HIV-1 and Hepatitis C Virus
Figure 3
Factors Influencing Within-Host Polymorphisms and the Global Frequency of Escape Variants
(A) A contour plot depicting the mean effect of selection and transmission rate on displacing the frequency of detectable polymorphisms from the neutral expectation (Δfpoly, refer to the color key), as estimated from simulations. (The expectation E[fpoly] is jointly determined by the forward and back mutation rates, μ and ν, and population size, N.) The x-axis corresponds to the log-transformed transmission rate, log10k. The y-axis represents the mean log-transformed selection coefficient, E(log10s) = qlog10(sesc) + (1 − q)log10(srev).
(B) A 10-fold disparity in selection intensities sesc = 0.02, srev = 0.002) causes π˄ to substantially exceed q with increasing transmission rate, k. Each set of points represents mean estimates of π˄ from simulations (with virus population size N = 5,000 and μ = ν = 10−4). Dashed lines indicate predicted values from the deterministic model, which performs poorly when k is too high (i.e., when transmissions occur rapidly, allele frequencies are almost always near zero or one where stochastic variation is greatest [31]). The typical range of q is indicated by the shaded plot region.