Figure 1.
Internal, Backbone, and External Branches in a Within-Host HIV Genealogy, and Mean Nucleotide Substitution Rates for These Branches in Nine Longitudinally Sampled HIV-1 Patients
To avoid the influence of deleterious mutations segregating to external branches in intrapatient HIV genealogies, we estimate mean substitution rates for the set of internal and backbone branches. These branch sets are depicted in color in the maximum a posteriori tree for “patient 1” obtained by Bayesian relaxed-clock inference [26] (backbone, red; internal, blue; external, black). The backbone represents the central trunk of trees shaped by rapid lineage turnover and can be defined phylogenetically (see Methods). Note that the set of internal branches also includes the backbone branches. Samples for each time point are indicated by the dotted line. Mean nucleotide substitution rates and their standard deviations on internal, external, and backbone branches are shown for all longitudinally sampled HIV-1 patients. The consistently higher substitution rate on external branches might be indicative of higher mutation load on these branches.
Figure 2.
Mean Synonymous and Nonsynonymous Divergence for Internal and Backbone Branches over the Course of HIV Infection in Nine Individuals
Patients with moderate and slow disease progression, categorized by progression time less than or greater than seven years [17], are shown in pink and blue, respectively. Individual patient numbers from Shankarappa et al. [14] are shown in circles, and progression times for each patient are indicated in italics.
Figure 3.
Scatter Plots of Mean Synonymous (A) and Nonsynonymous (B) Substitution Rates on Backbone Branches as a Function of Progression Time, CD4+ T Cell Count Slope, and Log Viral Load Slope
Rates as a function of progression time and the slope of the CD4+ T cell count are shown in pink, while rates as a function of the log VL slope are shown in cyan. The progression time for patient 11 was set at the last sampling time (144 months), although the CD4+ count had not dropped below 200 cells/μl at that time [14,17]. The slope of the CD4+ T cell count and log VL were calculated based on linear regression of these parameters as a function of time. The error bars represent the standard errors of the estimates.
Figure 4.
Internal Synonymous and Nonsynonymous Divergence in Two Patients with Distinct Phenotypic Escape from Neutralizing Antibodies
Nonsynonymous and synonymous divergence is shown for complete env sequences serially sampled from patient 01–0083 (A), whose virus exhibited slow escape from neutralizing antibodies, and from patient 01–0127 (B), whose virus exhibited rapid escape from neutralizing antibodies [2]. The thick lines depict the mean divergence and the colored areas represent the 95% credibility intervals.
Table 1.
Mean Synonymous and Nonsynonymous Substitution Rates Before and After Disease Progression
Table 1.
Extended.
Table 2.
Mean Synonymous and Nonsynonymous Substitution Rates for Different HIV Variants