HIV-2 Diversity Displays Two Clades within group A with Distinct Geographical Distribution and Evolution
This dataset contains the xml and MCC trees files supporting a phylogeographic analysis of early HIV-2 dispersal within Western Africa. All methodology and results are presented in the following work submitted for publication: Title: HIV-2 Diversity Displays Two Clades within group A with Distinct Geographical Distribution and Evolution Authors: Benoit Visseaux, Mélanie Bertine, Quentin Le Hingrat, Valentine Ferré, Charlotte Charpentier, Fidéline Collin, Florence Damond, Sophie Matheron, Stéphane Hué, Diane Descamps on behalf of the French ANRS CO5 HIV-2 cohort
Steps to reproduce
Phylogeographic analyses were performed separately for HIV-2 group A and B, including all pol sequences with available country of birth and sampling date from the ANRS CO5 HIV-2 cohort or available country and date of sampling from publicly available sequence databases. A preliminary analysis using TempEst was conducted to check the absence of sequence presenting obvious incorrect sampling date or immanent abnormalities in comparison with the global phylogenetic signal of the dataset. Phylogeographic reconstructions were performed using BEAST 1.8.4 with the Gaussian Markov Random Field (GMRF) Bayesian skyride coalescent model, the GTR-G model of nucleotide substitutions, a discrete symmetric trait substitution model and with either a strict or relaxed molecular clock model. The best-fitting molecular clock model was identified by marginal likelihood estimation. We ran two sets of 50 million generations each, sampling every 10 000th generations. Parameter estimations were deemed satisfactory when their Effective Sampling Size (ESS) was above 200 in Tracer 1.6. After discarding 10% burn-in, combined files were produced using LogCombiner. The maximum clade credibility (MCC) tree generated with TreeAnnotator was then annotated in FigTree v1.4.3.