The mid-depth method and HIV-1: A practical approach for testing hypotheses of viral epidemic history

We introduce the mid-depth method, a practical approach for testing hypothe ses of demographic history using genealogies reconstructed from sequence da ta. The relative positions of internal nodes within a genealogy contain inf ormation about past population dynamics. We explain how this information ca n be used to (1) test the null hypothesis of constant population size and ( 2) estimate the growth rate and current population size of an exponentially growing population. Simulation tests indicate that, as expected, estimates of exponential growth rates are sometimes biased. The mid-depth method is computationally rapid and does not require knowledge of the sample's mutati on rate. However, it does assume that the reconstructed genealogy is correc t and is therefore best suited to the analysis of variation-rich viral data sets. When applied to HIV-1 sequence data, the mid-depth method provides p hylogenetic evidence of different exponential growth rates for subtypes A a nd B. We posit that this difference in growth rate reflects the different t ransmission routes and epidemiological histories of the two subtypes.