We constructed the phylogenetic tree for Applications 1 and 2 from the full Gag DNA alignment by applying PHYML [51] with the general reversible model, optimized tree topology, maximum likelihood estimates for the base frequencies, estimated proportion of invariable sites, and four substitution rate categories with an estimated gamma distribution parameter.
For all trees, the best model according to AIC Akaike Information Criteriaa) was Le-Gascuel with empirical frequencies, estimated proportion of invariable sites, and estimated gamma shape parameter.
Maximum likelihood (ML) phylogenetic trees were constructed using PhyML (v3.0) with JTT amino acid substitution model, 1000 bootstrap replicates, estimated proportions of invariable sites, estimated gamma distribution parameters, and an optimized starting BIONJ tree [ 41, 42].
The exon trees were built using PHYML (substitution model = JTT, estimated proportion of invariable sites, four categories, estimated gamma, initial tree with BIONJ) [ 40].
The ML phylogenetic analyses used the following parameters: JTT model, estimated proportion of invariable sites, 4 rate categories, estimated gamma-distribution parameter, and BIONJ-optimized starting tree.
The ML phylogenetic analyses were conducted with the following parameters: JTT model, estimated proportion of invariable sites, 4 rate categories, estimated gamma distribution parameter, and optimized starting BIONJ tree.
Specifically, PhyML analyses were conducted with the LG model, estimated proportion of invariable sites, four rate categories, estimated gamma distribution parameter, and optimized starting BIONJ tree, both with aLRT and 100 bootstraps branch support measures.
A Maximum Likelihood tree was derived using PhyML [ 53] with the following parameters: model, HKY; transition/transversion ratio; estimated; proportion of invariable sites, estimated; number of relative substitution rate categories, 4; gamma distribution parameter, estimated; starting tree, BIONJ; optimize tree topology, yes; optimize branch lengths and rate parameters, yes.
The program FindModel (http://hcv.lanl.gov/content/sequence/findmodel/findmodel.html from the HCV sequence database) was used to determine the best phylogenetic model, and the following parameters were applied: estimated proportion of invariable sites (GTRGAMMA + I).
Phylogenies and 100 bootstrap supports for all datasets were calculated with PhyML v3.0 [ 80] with the LG [ 81] amino acid substitution model, estimated portions of invariable sites, estimated Γ distribution parameter, four rate categories, estimated amino acid frequencies, and an NJ starting tree.
