Figure 1.
Political map of Brazil showing the country regions and states.
Boundaries of regions and states are indicated by black and gray lines, respectively. The position of the eight states analyzed in the present study is indicated with a two letter code: GO (Goiás), MT (Mato Grosso), MS (Mato Grosso do Sul), PR (Paraná), RJ (Rio de Janeiro), RS (Rio Grande do Sul), SC (Santa Catarina) and SP (São Paulo). The number and sampling dates of HIV-1 subtype C pol sequences from each location included in the present study are indicated.
Figure 2.
ML tree of HIV-1 subtype C pol (~1,000 bp) sequences from Brazil (n = 209) and the central (n = 53), eastern (n = 332) and southern (n = 645) African regions.
The color of the branches represents the geographic region from where the subtype C sequences originated, according to the legend and map provide in the figure. The dotted boxes highlight the position of the Brazilian subtype C lineages (CBR-II to CBR-IV) that branched outside the major Brazilian clade (CBR-I). A close view of the minor Brazilian subtype C lineages and the most closely related African sequences is also provided. For visual clarity, the Brazilian clade CBR-I and some clades that comprised mostly sequences from central, eastern or southern Africa were collapsed into triangles. The aLRT support values are indicated only at key nodes. The tree was rooted using HIV-1 subtype A1 and D reference sequences (black branches). Horizontal branch lengths are drawn to scale with the bar at the bottom indicating nucleotide substitutions per site.
Figure 3.
Time-scaled Bayesian MCC tree of the HIV-1 CEA and CBR-I lineages.
Branches are colored according to the most probable location state of their descendent nodes. The legend for the colors is shown on the left. The dotted boxes highlight the position of the major country-specific sub-clades detected in our study. The median age (with 95% HPD interval in parentheses) and PSP values of key nodes are shown. Asterisks point to key nodes with a high (> 0.85) PP support. Horizontal branch lengths are drawn to scale with the bar at the bottom indicating years. The tree was automatically rooted under the assumption of a relaxed molecular clock.
Figure 4.
Time-scaled Bayesian MCC tree of the HIV-1 CBR-II to CBR-VI lineages and the most closely related reference sequences.
Branches are colored according to the most probable location state of their descendent nodes. The legend for the colors is shown on the left. The boxes highlight the position of the minor Brazilian HIV-1 subtype C clades. The PP support is indicated only at key nodes. The scale bar at the bottom indicates years. The tree was automatically rooted under the assumption of a relaxed molecular clock. Graphics on the right depict the PSP distributions at the first ancestral nodes of Brazilian subtype C lineages at the Bayesian MCC tree. Countries represented are AO (Angola), BI (Burundi), KE (Kenya), ZA (South Africa), ZM (Zambia) and others (from Asia and Europe).
Figure 5.
Time-scaled Bayesian MCC tree of the HIV-1 CBR-I lineage.
Branches are colored according to the most probable location state of their descendent nodes as indicated in the legend and map shown on the left. The dotted boxes highlight the position of the Brazilian sub-clade characteristic of Rio Grande do Sul (CBR-RS). The median age (with 95% HPD interval in parentheses) and the PSP values of some key nodes are shown. Key nodes with a high (> 0.85) PP support are marked with an asterisk. Horizontal branch lengths are drawn to scale with the bar at the bottom indicating years. The tree was automatically rooted under the assumption of a relaxed molecular clock.
Figure 6.
Spatiotemporal dynamics of HIV-1 CBR-1 clade dissemination in Brazil.
(A) Viral dispersal pattern between 1976 and 2010. Lines between locations represent branches in the Bayesian MCC tree along which location transitions occurs. The yellow-black color gradient of lines informs the date of the earliest viral migrations among each pair of locations. (B) Bayes factor (BF) test for significant non-zero rates. Only rates supported by a BF greater than 2 are indicated. The light-dark color gradient of lines informs the relative strength by which the rates are supported (weak-strong). The maps are based on satellite pictures made available in Google™ Earth (http://earth.google.com). (C) Major estimated viral transitions rates (q) as measured by the APES program. The arrows were colored according to the source region and the width is proportional to q. All q lower than 1.0 were excluded for clarity. RS (Rio Grande do Sul), SC (Santa Catarina), PR (Paraná), RJ (Rio de Janeiro), SP (São Paulo) and CW (Central-west Region).
Figure 7.
Association between viral migration rates and human mobility data.
Viral transition rates (q, gene flow) were plotted against: (A) routine traffic ([people × trip]/1,000); (B) routine traffic multiplied by the HIV-1 subtype C prevalence in the site of origin; (C) road distances between state’s capitals (the capital of Goiás state was used as reference for the central-west region); and (D) road distances divided by the average subtype C prevalence in the corresponding states. The linear regression line is shown in each graph. The P-value and r squared (R2) from correlations are indicated in each plot.