Fig. 8

Schematics of Phylogenetic/coevolution analyses. (a) Full length subtype A1 and subtype D envelope amino acid sequences downloaded from the Los Alamos National Laboratory HIV sequence database were filtered and resulted in data sets containing 144 and 97 unique sequences of subtype A1 and subtype D, respectively. Subtype specific alignments were performed using MUSCLE and then trimmed to the positions containing the C1–C5 regions and the ecto-gp41 domain. (b) Maximum likelihood trees were generated for each region using the algorithms implemented in PHYML with 100 bootstrap replicates, and the tree topology search was performed using Nearest Neighbor Interchanges (NNIs). The phylogeny of the regions was estimated using the rtREV model of amino acid substitution. The resulting maximum likelihood trees were used as input to the Mirrortree web server. (c) Pairwise distance matrices of each region were built with MEGA, which used a JTT matrix-based model and a uniform substitution rate among sites, to calculate the number of amino acid substitutions per site/residue from between each sequence in the alignment. Correlation between each conserved region in gp120 or the gp41 domain were then measured by Pearson’s product-moment correlation coefficient, and significance estimated with Fisher’s exact test