Evolution Of Protein-coding Sequences Research Articles

Seminal fluid gene expression and reproductive fitness in Drosophila melanogaster

BackgroundThe rapid evolution of seminal fluid proteins (SFPs) has been suggested to be driven by adaptations to postcopulatory sexual selection (e.g. sperm competition). However, we have recently shown that most SFPs evolve rapidly under relaxed selective pressures. Given the role of SFPs in competition for fertilization phenotypes, like the ability to transfer and store sperm and the modulation of female receptivity and ovulation, the prevalence of selectively relaxed SFPs appears as a conundrum. One possible explanation is that selection on SFPs might be relaxed in terms of protein amino acid content, but adjustments of expression are essential for post-mating function. Interestingly, there is a general lack of systematic implementation of gene expression perturbation assays to monitor their effect on phenotypes related to sperm competition.ResultsWe successfully manipulated the expression of 16 SFP encoding genes using tissue-specific knockdowns (KDs) and determined the effect of these genes’ perturbation on three important post-mating phenotypes: female refractoriness to remating, defensive (P1), and offensive (P2) sperm competitive abilities in Drosophila melanogaster. Our analyses show that KDs of tested SFP genes do not affect female refractoriness to remating and P2, however, most gene KDs significantly decreased P1. Moreover, KDs of SFP genes that are selectively constrained in terms of protein-coding sequence evolution have lower P1 than KDs of genes evolving under relaxed selection.ConclusionsOur results suggest a more predominant role, than previously acknowledged, of variation in gene expression than coding sequence changes on sperm competitive ability in D. melanogaster.

Selection Shapes Synonymous Stop Codon Use in Mammals.

Phylogenetic models of the evolution of protein-coding sequences can provide insights into the selection pressures that have shaped them. In the application of these models synonymous nucleotide substitutions, which do not alter the encoded amino acid, are often assumed to have limited functional consequences and used as a proxy for the neutral rate of evolution. The ratio of nonsynonymous to synonymous substitution rates is then used to categorize the selective regime that applies to the protein (e.g., purifying selection, neutral evolution, diversifying selection). Here, we extend the Muse and Gaut model of codon evolution to explore the extent of purifying selection acting on substitutions between synonymous stop codons. Using a large collection of coding sequence alignments, we estimate that a high proportion (approximately 57%) of mammalian genes are affected by selection acting on stop codon preference. This proportion varies substantially by codon, with UGA stop codons far more likely to be conserved. Genes with evidence of selection acting on synonymous stop codons have distinctive characteristics, compared to unconserved genes with the same stop codon, including longer [Formula: see text] untranslated regions (UTRs) and shorter mRNA half-life. The coding regions of these genes are also much more likely to be under strong purifying selection pressure. Our results suggest that the preference for UGA stop codons found in many multicellular eukaryotes is selective rather than mutational in origin.

Evolutionary Patterns of Sex-Biased Genes in Three Species of Haplodiploid Insects.

Females and males often differ obviously in morphology and behavior, and the differences between sexes are the result of natural selection and/or sexual selection. To a great extent, the differences between the two sexes are the result of differential gene expression. In haplodiploid insects, this phenomenon is obvious, since males develop from unfertilized zygotes and females develop from fertilized zygotes. Whiteflies of the Bemisia tabaci species complex are typical haplodiploid insects, and some species of this complex are important pests of many crops worldwide. Here, we report the transcriptome profiles of males and females in three species of this whitefly complex. Between-species comparisons revealed that non-sex-biased genes display higher variation than male-biased or female-biased genes. Sex-biased genes evolve at a slow rate in protein coding sequences and gene expression and have a pattern of evolution that differs from those of social haplodiploid insects and diploid animals. Genes with high evolutionary rates are more related to non-sex-biased traits—such as nutrition, immune system, and detoxification—than to sex-biased traits, indicating that the evolution of protein coding sequences and gene expression has been mainly driven by non-sex-biased traits.

PoSE: visualization of patterns of sequence evolution using PAML and MATLAB

BackgroundDetermining patterns of nucleotide and amino acid substitution is the first step during sequence evolution analysis. However, it is not easy to visualize the different phylogenetic signatures imprinted in aligned nucleotide and amino acid sequences.ResultsHere we present PoSE (Pattern of Sequence Evolution), a reliable resource for unveiling the evolutionary history of sequence alignments and for graphically displaying their contents. Substitutions are displayed by category (transitions and transversions), codon position, and phenotypic effect (synonymous and nonsynonymous). Visualization is accomplished using MATLAB scripts wrapped around PAML (Phylogenetic Analysis by Maximum Likelihood), implemented in an easy-to-use graphical user interface. The application displays inferred substitutions estimated by baseml or codeml, two programs included in the PAML software package. PoSE organizes patterns of substitution in eleven plots, including estimated non-synonymous/synonymous ratios (dN/dS) along the sequence alignment. In addition, PoSE provides visualization and annotation of patterns of amino acid substitutions along groups of related sequences that can be graphically inspected in a phylogenetic tree window.ConclusionsPoSE is a useful tool to help determine major patterns during sequence evolution of protein-coding sequences, hypervariable regions, or changes in dN/dS ratios. PoSE is publicly available at https://github.com/CDCgov/PoSE

Multiple Modes of Adaptation: Regulatory and Structural Evolution in a Small Heat Shock Protein Gene.

Thermal tolerance is a key determinant of species distribution. Despite much study, the genetic basis of adaptive evolution of thermal tolerance, including the relative contributions of transcriptional regulation versus protein evolution, remains unclear. Populations of the intertidal copepod Tigriopus californicus are adapted to local thermal regimes across their broad geographic range. Upon thermal stress, adults from a heat tolerant southern population, San Diego (SD), upregulate several heat shock proteins (HSPs) to higher levels than those from a less tolerant northern population, Santa Cruz (SC). Suppression of a specific HSP, HSPB1, significantly reduces T. californicus survival following acute heat stress. Sequencing of HSPB1 revealed population specific nucleotide substitutions in both promoter and coding regions of the gene. HSPB1 promoters from heat tolerant populations contain two canonical heat shock elements (HSEs), the binding sites for heat shock transcription factor (HSF), whereas less tolerant populations have mutations in these conserved motifs. Allele specific expression of HSPB1 in F1 hybrids between tolerant and less tolerant populations showed significantly biased expression favoring alleles from tolerant populations and supporting the adaptive divergence in these cis-regulatory variants. The functional impact of population-specific nonsynonymous substitutions in HSPB1 coding sequences was tested by assessing the thermal stabilization properties of SD versus SC HSPB1 protein variants. Recombinant HSPB1 from the southern SD population showed greater capacity for protecting protein structure under elevated temperature. Our results indicate that both regulatory and protein coding sequence evolution within a single gene appear to contribute to thermal tolerance phenotypes and local adaptation among conspecific populations.

Adaptive sequence evolution is driven by biotic stress in a pair of orchid species (Dactylorhiza) with distinct ecological optima.

The orchid family is the largest in the angiosperms, but little is known about the molecular basis of the significant variation they exhibit. We investigate here the transcriptomic divergence between two European terrestrial orchids, Dactylorhiza incarnata and Dactylorhiza fuchsii, and integrate these results in the context of their distinct ecologies that we also document. Clear signals of lineage‐specific adaptive evolution of protein‐coding sequences are identified, notably targeting elements of biotic defence, including both physical and chemical adaptations in the context of divergent pools of pathogens and herbivores. In turn, a substantial regulatory divergence between the two species appears linked to adaptation/acclimation to abiotic conditions. Several of the pathways affected by differential expression are also targeted by deviating post‐transcriptional regulation via sRNAs. Finally, D. incarnata appears to suffer from insufficient sRNA control over the activity of RNA‐dependent DNA polymerase, resulting in increased activity of class I transposable elements and, over time, in larger genome size than that of D. fuchsii. The extensive molecular divergence between the two species suggests significant genomic and transcriptomic shock in their hybrids and offers insights into the difficulty of coexistence at the homoploid level. Altogether, biological response to selection, accumulated during the history of these orchids, appears governed by their microenvironmental context, in which biotic and abiotic pressures act synergistically to shape transcriptome structure, expression and regulation.

Long-Term Balancing Selection in LAD1 Maintains a Missense Trans-Species Polymorphism in Humans, Chimpanzees, and Bonobos.

Balancing selection maintains advantageous genetic and phenotypic diversity in populations. When selection acts for long evolutionary periods selected polymorphisms may survive species splits and segregate in present-day populations of different species. Here, we investigate the role of long-term balancing selection in the evolution of protein-coding sequences in the Homo-Pan clade. We sequenced the exome of 20 humans, 20 chimpanzees, and 20 bonobos and detected eight coding trans-species polymorphisms (trSNPs) that are shared among the three species and have segregated for approximately 14 My of independent evolution. Although the majority of these trSNPs were found in three genes of the major histocompatibility locus cluster, we also uncovered one coding trSNP (rs12088790) in the gene LAD1. All these trSNPs show clustering of sequences by allele rather than by species and also exhibit other signatures of long-term balancing selection, such as segregating at intermediate frequency and lying in a locus with high genetic diversity. Here, we focus on the trSNP in LAD1, a gene that encodes for Ladinin-1, a collagenous anchoring filament protein of basement membrane that is responsible for maintaining cohesion at the dermal-epidermal junction; the gene is also an autoantigen responsible for linear IgA disease. This trSNP results in a missense change (Leucine257Proline) and, besides altering the protein sequence, is associated with changes in gene expression of LAD1.

Genomic, Transcriptomic, and Phenomic Variation Reveals the Complex Adaptation of Modern Maize Breeding

The temperate-tropical division of early maize germplasms to different agricultural environments was arguably the greatest adaptation process associated with the success and near ubiquitous importance of global maize production. Deciphering this history is challenging, but new insight has been gained from examining 558 529 single nucleotide polymorphisms, expression data of 28 769 genes, and 662 traits collected from 368 diverse temperate and tropical maize inbred lines in this study. This is a new attempt to systematically exploit the mechanisms of the adaptation process in maize. Our results indicate that divergence between tropical and temperate lines apparently occurred 3400–6700 years ago. Seven hundred and one genomic selection signals and transcriptomic variants including 2700 differentially expressed individual genes and 389 rewired co-expression network genes were identified. These candidate signals were found to be functionally related to stress responses, and most were associated with directionally selected traits, which may have been an advantage under widely varying environmental conditions faced by maize as it was migrated away from its domestication center. Our study also clearly indicates that such stress adaptation could involve evolution of protein-coding sequences as well as transcriptome-level regulatory changes. The latter process may be a more flexible and dynamic way for maize to adapt to environmental changes along its short evolutionary history.

Unraveling patterns of site-to-site synonymous rates variation and associated gene properties of protein domains and families.

In protein-coding genes, synonymous mutations are often thought not to affect fitness and therefore are not subject to natural selection. Yet increasingly, cases of non-neutral evolution at certain synonymous sites were reported over the last decade. To evaluate the extent and the nature of site-specific selection on synonymous codons, we computed the site-to-site synonymous rate variation (SRV) and identified gene properties that make SRV more likely in a large database of protein-coding gene families and protein domains. To our knowledge, this is the first study that explores the determinants and patterns of the SRV in real data. We show that the SRV is widespread in the evolution of protein-coding sequences, putting in doubt the validity of the synonymous rate as a standard neutral proxy. While protein domains rarely undergo adaptive evolution, the SRV appears to play important role in optimizing the domain function at the level of DNA. In contrast, protein families are more likely to evolve by positive selection, but are less likely to exhibit SRV. Stronger SRV was detected in genes with stronger codon bias and tRNA reusage, those coding for proteins with larger number of interactions or forming larger number of structures, located in intracellular components and those involved in typically conserved complex processes and functions. Genes with extreme SRV show higher expression levels in nearly all tissues. This indicates that codon bias in a gene, which often correlates with gene expression, may often be a site-specific phenomenon regulating the speed of translation along the sequence, consistent with the co-translational folding hypothesis. Strikingly, genes with SRV were strongly overrepresented for metabolic pathways and those associated with several genetic diseases, particularly cancers and diabetes.

Evolution at Two Levels in Fire Ants: The Relationship between Patterns of Gene Expression and Protein Sequence Evolution

Variation in protein sequence and gene expression each contribute to phenotypic diversity, and may be subject to similar selective pressures. Eusocial insects are particularly useful for investigating the evolutionary link between protein sequence and condition-dependent patterns of gene expression because gene expression plays a central role in determining differences between eusocial insect sexes and castes. We investigated the relationship between protein coding sequence evolution and gene expression patterns in the fire ants Solenopsis invicta, S. richteri, and their hybrids to gain greater insight into how selection jointly operates on gene expression and coding sequence. We found that genes with high expression variability within castes and sexes were frequently differentially expressed between castes and sexes, as well as between species and hybrids. These results indicate that genes showing high variation in expression in one context also tend to show high variation in expression in other contexts. Our analyses further revealed that variation in both intra- and interspecific gene expression was positively associated with rate of protein sequence evolution in Solenopsis. This suggests that selective constraints on a gene operate both at the level of protein sequence and at the level of gene expression regulation. Overall, our study provides one of the strongest demonstrations that selective constraints mediate both protein sequence evolution and gene expression variability across different biological contexts and timescales.

Interaction between Selection and Biased Gene Conversion in Mammalian Protein-Coding Sequence Evolution Revealed by a Phylogenetic Covariance Analysis

According to the nearly-neutral model, variation in long-term effective population size among species should result in correlated variation in the ratio of nonsynonymous over synonymous substitution rates (dN/dS). Previous empirical investigations in mammals have been consistent with this prediction, suggesting an important role for nearly-neutral effects on protein-coding sequence evolution. GC-biased gene conversion (gBGC), on the other hand, is increasingly recognized as a major evolutionary force shaping genome nucleotide composition. When sufficiently strong compared with random drift, gBGC may significantly interfere with a nearly-neutral regime and impact dN/dS in a complex manner. Here, we investigate the phylogenetic correlations between dN/dS, the equilibrium GC composition (GC*), and several life-history and karyotypic traits in placental mammals. We show that the equilibrium GC composition decreases with body mass and increases with the number of chromosomes, suggesting a modulation of the strength of biased gene conversion due to changes in effective population size and genome-wide recombination rate. The variation in dN/dS is complex and only partially fits the prediction of the nearly-neutral theory. However, specifically restricting estimation of the dN/dS ratio on GC-conservative transversions, which are immune from gBGC, results in correlations that are more compatible with a nearly-neutral interpretation. Our investigation indicates the presence of complex interactions between selection and biased gene conversion and suggests that further mechanistic development is warranted, to tease out mutation, selection, drift, and conversion.

Towards a More Nuanced Understanding of the Relationship between Sex-Biased Gene Expression and Rates of Protein-Coding Sequence Evolution

Genes that are differentially expressed between the sexes (sex-biased genes) are among the fastest evolving genes in animal genomes. The majority of sex-biased expression is attributable to genes that are primarily expressed in sex-limited reproductive tissues, and these reproductive genes are often rapidly evolving because of intra- and intersexual selection pressures. Additionally, studies of multiple taxa have revealed that genes with sex-biased expression are also expressed in a limited number of tissues. This is worth noting because narrowly expressed genes are known to evolve faster than broadly expressed genes. Therefore, it is not clear whether sex-biased genes are rapidly evolving because they have sexually dimorphic expression, because they are expressed in sex-limited reproductive tissues, or because they are narrowly expressed. To determine the extend to which other confounding variables can explain the rapid evolution of sex-biased genes, I analyzed the rates of evolution of sex-biased genes in Drosophila melanogaster and Mus musculus in light of tissue-specific measures of expression. I find that genes with sex-biased expression in somatic tissues shared by both sexes are often evolving faster than non-sex-biased genes, but this is best explained by the narrow expression profiles of sex-biased genes. Sex-biased genes in sex-limited tissues in D. melanogaster, however, evolve faster than other narrowly expressed genes. Therefore, the rapid evolution of sex-biased genes is limited only to those genes primarily expressed in sex-limited reproductive tissues.

Genome Wide Analyses Reveal Little Evidence for Adaptive Evolution in Many Plant Species

The relative contribution of advantageous and neutral mutations to the evolutionary process is a central problem in evolutionary biology. Current estimates suggest that whereas Drosophila, mice, and bacteria have undergone extensive adaptive evolution, hominids show little or no evidence of adaptive evolution in protein-coding sequences. This may be a consequence of differences in effective population size. To study the matter further, we have investigated whether plants show evidence of adaptive evolution using an extension of the McDonald-Kreitman test that explicitly models slightly deleterious mutations by estimating the distribution of fitness effects of new mutations. We apply this method to data from nine pairs of species. Altogether more than 2,400 loci with an average length of approximately 280 nucleotides were analyzed. We observe very similar results in all species; we find little evidence of adaptive amino acid substitution in any comparison except sunflowers. This may be because many plant species have modest effective population sizes.

Many-core algorithms for statistical phylogenetics

Statistical phylogenetics is computationally intensive, resulting in considerable attention meted on techniques for parallelization. Codon-based models allow for independent rates of synonymous and replacement substitutions and have the potential to more adequately model the process of protein-coding sequence evolution with a resulting increase in phylogenetic accuracy. Unfortunately, due to the high number of codon states, computational burden has largely thwarted phylogenetic reconstruction under codon models, particularly at the genomic-scale. Here, we describe novel algorithms and methods for evaluating phylogenies under arbitrary molecular evolutionary models on graphics processing units (GPUs), making use of the large number of processing cores to efficiently parallelize calculations even for large state-size models. We implement the approach in an existing Bayesian framework and apply the algorithms to estimating the phylogeny of 62 complete mitochondrial genomes of carnivores under a 60-state codon model. We see a near 90-fold speed increase over an optimized CPU-based computation and a >140-fold increase over the currently available implementation, making this the first practical use of codon models for phylogenetic inference over whole mitochondrial or microorganism genomes. Source code provided in BEAGLE: Broad-platform Evolutionary Analysis General Likelihood Evaluator, a cross-platform/processor library for phylogenetic likelihood computation (http://beagle-lib.googlecode.com/). We employ a BEAGLE-implementation using the Bayesian phylogenetics framework BEAST (http://beast.bio.ed.ac.uk/).

Fast optimization of statistical potentials for structurally constrained phylogenetic models

BackgroundStatistical approaches for protein design are relevant in the field of molecular evolutionary studies. In recent years, new, so-called structurally constrained (SC) models of protein-coding sequence evolution have been proposed, which use statistical potentials to assess sequence-structure compatibility. In a previous work, we defined a statistical framework for optimizing knowledge-based potentials especially suited to SC models. Our method used the maximum likelihood principle and provided what we call the joint potentials. However, the method required numerical estimations by the use of computationally heavy Markov Chain Monte Carlo sampling algorithms.ResultsHere, we develop an alternative optimization procedure, based on a leave-one-out argument coupled to fast gradient descent algorithms. We assess that the leave-one-out potential yields very similar results to the joint approach developed previously, both in terms of the resulting potential parameters, and by Bayes factor evaluation in a phylogenetic context. On the other hand, the leave-one-out approach results in a considerable computational benefit (up to a 1,000 fold decrease in computational time for the optimization procedure).ConclusionDue to its computational speed, the optimization method we propose offers an attractive alternative for the design and empirical evaluation of alternative forms of potentials, using large data sets and high-dimensional parameterizations.

The Influence of Demography and Weak Selection on the McDonald-Kreitman Test: An Empirical Study in Drosophila

The McDonald-Kreitman (MK) test, which compares the ratio of polymorphism to divergence at nonsynonymous and synonymous sites, is frequently used to detect adaptive evolution in protein-coding sequences. Because the two classes of sites share a common evolutionary history, the MK test is thought to be robust to most demographic factors. However, weak selection on nonsynonymous sites can bias the MK test, especially when a species' effective population size has not been constant. Here, we present an empirical analysis of the influence of demography on the MK test by comparing test results for a common set of 136 genes, including a set of sex-biased genes that shows a strong signal of adaptive evolution, in two Drosophila melanogaster populations: an ancestral population from Africa and a derived population from Europe. The latter has undergone a relatively recent bottleneck, which has reduced its effective population size. We find that the MK test has less power to detect positive selection in the European population for two reasons. First, the overall reduced level of standing variation decreases the statistical power of the test. Second, the segregation of slightly deleterious nonsynonymous mutations biases the MK test away from detecting positive selection. The latter effect is stronger for X-linked genes, which have experienced the greatest reduction in effective population size outside of Africa, and also leads to the underestimation of rates of adaptive protein evolution by multilocus implementations of the MK test. Interestingly, a subset of autosomal female-biased genes shows an increased signal of adaptive evolution in the European population. This is inconsistent with currently accepted demographic scenarios and may reflect female-specific changes in selective constraint following the colonization of non-African habitats.

The Impact of the Nucleosome Code on Protein-Coding Sequence Evolution in Yeast

Coding sequence evolution was once thought to be the result of selection on optimal protein function alone. Selection can, however, also act at the RNA level, for example, to facilitate rapid translation or ensure correct splicing. Here, we ask whether the way DNA works also imposes constraints on coding sequence evolution. We identify nucleosome positioning as a likely candidate to set up such a DNA-level selective regime and use high-resolution microarray data in yeast to compare the evolution of coding sequence bound to or free from nucleosomes. Controlling for gene expression and intra-gene location, we find a nucleosome-free “linker” sequence to evolve on average 5–6% slower at synonymous sites. A reduced rate of evolution in linker is especially evident at the 5′ end of genes, where the effect extends to non-synonymous substitution rates. This is consistent with regular nucleosome architecture in this region being important in the context of gene expression control. As predicted, codons likely to generate a sequence unfavourable to nucleosome formation are enriched in linker sequence. Amino acid content is likewise skewed as a function of nucleosome occupancy. We conclude that selection operating on DNA to maintain correct positioning of nucleosomes impacts codon choice, amino acid choice, and synonymous and non-synonymous rates of evolution in coding sequence. The results support the exclusion model for nucleosome positioning and provide an alternative interpretation for runs of rare codons. As the intimate association of histones and DNA is a universal characteristic of genic sequence in eukaryotes, selection on coding sequence composition imposed by nucleosome positioning should be phylogenetically widespread.

Modelling the evolution of protein coding sequences sampled from Measurably Evolving Populations.

Models of nucleotide or amino acid sequence evolution that implement homogeneous and stationary Markov processes of substitutions are mathematically convenient but are unlikely to represent the true complexity of evolution. With the large amounts of data that next generation sequencing promises, appropriate models of evolution are important, particularly when data are collected from ancient and sub-fossil remains, where changes in evolutionary parameters are the norm and not the exception. In this paper, we describe a new codon-based model of evolution that applies to Measurably Evolving Populations (MEPs). A MEP is defined as a population from which it is possible to detect a statistically significant accumulation of substitutions when sequences are obtained at different times. The new model of codon evolution permits changes to the substitution process, including changes to the intensity of selection and the proportions of sites undergoing different selective pressures. In our serial model of codon evolution, changes in the selective regime occur simultaneously across all lineages. Different regions of the protein may also evolve under distinct selective patterns. We illustrate the application of the new model to a dataset of HIV-1 sequences obtained from an infected individual before and after the commencement of antiretroviral therapy.

Radical amino acid change versus positive selection in the evolution of viral envelope proteins

To detect positive selection in protein-coding sequence evolution, the ratio of the nonsynonymous to synonymous substitution rate ( K A/ K S) is commonly used. When this ratio is higher than 1, positive selection on nonsynonymous changes is considered to have occurred. However, the question of what kinds of amino acid change are likely to be involved in positive selection has not been well studied, though intuitively it seems that radical changes frequently occur in positively selected changes. To address this question, we examined chemically radical and conservative replacements in the evolution of hepatitis C virus (HCV) protein sequences. In the envelope region, 34 positively and 440 negatively selected sites were identified by the K A/ K S ratio. Radical and conservative changes were compared between the two types of selected sites using two methods. First, the numbers of radical and conservative replacements were counted at the positively and negatively selected sites according to three kinds of chemical classifications. In all three classifications, the resulting ratios of the two numbers were not statistically different for the two types of selected sites ( P > 0.05). Second, the distribution of chemical changes was compared between the two types of selected sites using two kinds of chemical distances. The distributions of the two chemical distances were not statistically different for the two types of selected sites ( P > 0.05). These results indicate that the ratio of chemically radical and conservative changes is similar for positively and negatively selected sites in the envelope protein of HCV or, in other words, there is no correlation between radical change and positive selection in the evolution of this protein.

Adaptation to Different Human Populations by HIV-1 Revealed by Codon-Based Analyses

Several codon-based methods are available for detecting adaptive evolution in protein-coding sequences, but to date none specifically identify sites that are selected differentially in two populations, although such comparisons between populations have been historically useful in identifying the action of natural selection. We have developed two fixed effects maximum likelihood methods: one for identifying codon positions showing selection patterns that persist in a population and another for detecting whether selection is operating differentially on individual codons of a gene sampled from two different populations. Applying these methods to two HIV populations infecting genetically distinct human hosts, we have found that few of the positively selected amino acid sites persist in the population; the other changes are detected only at the tips of the phylogenetic tree and appear deleterious in the long term. Additionally, we have identified seven amino acid sites in protease and reverse transcriptase that are selected differentially in the two samples, demonstrating specific population-level adaptation of HIV to human populations.