No Evidence that Selection on Synonymous Codon Usage Affects Patterns of Protein Evolution in Bacteria

Bias in synonymous codon usage has been reported across all kingdoms of life. Evidence suggests that codon usage bias is often driven by selective pressures, typically for translational efficiency. These selective pressures have been shown to depress the rate at which synonymous sites evolve. We hypothesize that selection on synonymous codon use could also slow the rate of protein evolution if a non-synonymous mutation changes the codon from being preferred to unpreferred. We test this hypothesis by looking at patterns of protein evolution using polymorphism and substitution data in two bacterial species, Escherichia coli and Streptococcus pneumoniae. We find no evidence that non-synonymous mutations that change a codon from being unpreferred to preferred are more common than the opposite. Overall, selection on codon bias seems to have little influence over non-synonymous polymorphism or substitution patterns.

Limited role of generation time changes in driving the evolution of the mutation spectrum in humans

Recent studies have suggested that the human germline mutation rate and spectrum evolve rapidly. Variation in generation time has been linked to these changes, though its contribution remains unclear. We develop a framework to characterize temporal changes in polymorphisms within and between populations, while controlling for the effects of natural selection and biased gene conversion. Application to the 1000 Genomes Project dataset reveals multiple independent changes that arose after the split of continental groups, including a previously reported, transient elevation in TCC>TTC mutations in Europeans and novel signals of divergence in C>Gand T>A mutation rates among population samples. We also find a significant difference between groups sampled in and outside of Africa in old T>C polymorphisms that predate the out-of-Africa migration. This surprising signal is driven by TpG>CpG mutations and stems in part from mis-polarized CpG transitions, which are more likely to undergo recurrent mutations. Finally, by relating the mutation spectrum of polymorphisms to parental age effects on de novo mutations, we show that plausible changes in the generation time cannot explain the patterns observed for different mutation types jointly. Thus, other factors - genetic modifiers or environmental exposures - must have had a non-negligible impact on the human mutation landscape.

Linked-Read Sequencing of Eight Falcons Reveals a Unique Genomic Architecture in Flux

Falcons are diverse birds of cultural and economic importance. They have undergone major lineage-specific chromosomal rearrangements, resulting in greatly-reduced chromosome counts relative to other birds. Here, we use 10X Genomics linked reads to provide new high-contiguity genomes for two gyrfalcons, a saker falcon, a lanner falcon, three subspecies of peregrine falcons, and the common kestrel. Assisted by a transcriptome sequenced from 22 gyrfalcon tissues, we annotate these genomes for a variety of genomic features, estimate historical demography, and then investigate genomic equilibrium in the context of falcon-specific chromosomal rearrangements. We find that falcon genomes are not in AT-GC equilibrium with a bias in substitutions towards higher AT content; this bias is predominantly but not exclusively driven by hypermutability of CpG sites. Small indels and large structural variants were also biased towards insertions rather than deletions. Patterns of disequilibrium were linked to chromosomal rearrangements: falcons have lost GC content in regions that have fused to larger chromosomes from microchromosomes and gained GC content in regions of macrochromosomes that have translocated to microchromosomes. Inserted bases have accumulated on regions ancestrally belonging to microchromosomes, consistent with insertion-biased gene conversion. We also find an excess of interspersed repeats on regions of microchromosomes that have fused to macrochromosomes. Our results reveal that falcon genomes are in a state of flux. They further suggest that many of the key differences between microchromosomes and macrochromosomes are driven by differences in chromosome size, and indicate a clear role for recombination and biased-gene-conversion in determining genomic equilibrium.

Population dynamics of GC-changing mutations in humans and great apes

The nucleotide composition of the genome is a balance between origin and fixation rates of different mutations. For example, it is well-known that transitions occur more frequently than transversions, particularly at CpG sites. Differences in fixation rates of mutation types are less explored. Specifically, recombination-associated GC-biased gene conversion (gBGC) may differentially impact GC-changing mutations, due to differences in their genomic distributions and efficiency of mismatch repair mechanisms. Given that recombination evolves rapidly across species, we explore gBGC of different mutation types across human populations and great ape species. We report a stronger correlation between segregating GC frequency and recombination for transitions than for transversions. Notably, CpG transitions are most strongly affected by gBGC in humans and chimpanzees. We show that the overall strength of gBGC is generally correlated with effective population sizes in humans, with some notable exceptions, such as a stronger effect of gBGC on non-CpG transitions in populations of European descent. Furthermore, species of the Gorilla and Pongo genus have a greatly reduced gBGC effect on CpG sites. We also study the dependence of gBGC dynamics on flanking nucleotides and show that some mutation types evolve in opposition to the gBGC expectation, likely due to hypermutability of specific nucleotide contexts. Our results highlight the importance of different gBGC dynamics experienced by GC-changing mutations and their impact on nucleotide composition evolution.

Inferring Adaptive Codon Preference to Understand Sources of Selection Shaping Codon Usage Bias

Alternative synonymous codons are often used at unequal frequencies. Classically, studies of such codon usage bias (CUB) attempted to separate the impact of neutral from selective forces by assuming that deviations from a predicted neutral equilibrium capture selection. However, GC-biased gene conversion (gBGC) can also cause deviation from a neutral null. Alternatively, selection has been inferred from CUB in highly expressed genes, but the accuracy of this approach has not been extensively tested, and gBGC can interfere with such extrapolations (e.g., if expression and gene conversion rates covary). It is therefore critical to examine deviations from a mutational null in a species with no gBGC. To achieve this goal, we implement such an analysis in the highly AT rich genome of Dictyostelium discoideum, where we find no evidence of gBGC. We infer neutral CUB under mutational equilibrium to quantify "adaptive codon preference," a nontautologous genome wide quantitative measure of the relative selection strength driving CUB. We observe signatures of purifying selection consistent with selection favoring adaptive codon preference. Preferred codons are not GC rich, underscoring the independence from gBGC. Expression-associated "preference" largely matches adaptive codon preference but does not wholly capture the influence of selection shaping patterns across all genes, suggesting selective constraints associated specifically with high expression. We observe patterns consistent with effects on mRNA translation and stability shaping adaptive codon preference. Thus, our approach to quantifying adaptive codon preference provides a framework for inferring the sources of selection that shape CUB across different contexts within the genome.

Biased Gene Conversion Constrains Adaptation in Arabidopsis thaliana

Reduction of fitness due to deleterious mutations imposes a limit to adaptive evolution. By characterizing features that influence this genetic load we may better understand constraints on responses to both natural and human-mediated selection. Here, using whole-genome, transcriptome, and methylome data from >600 Arabidopsis thaliana individuals, we set out to identify important features influencing selective constraint. Our analyses reveal that multiple factors underlie the accumulation of maladaptive mutations, including gene expression level, gene network connectivity, and gene-body methylation. We then focus on a feature with major effect, nucleotide composition. The ancestral vs. derived status of segregating alleles suggests that GC-biased gene conversion, a recombination-associated process that increases the frequency of G and C nucleotides regardless of their fitness effects, shapes sequence patterns in A. thaliana Through estimation of mutational effects, we present evidence that biased gene conversion hinders the purging of deleterious mutations and contributes to a genome-wide signal of decreased efficacy of selection. By comparing these results to two outcrossing relatives, Arabidopsis lyrata and Capsella grandiflora, we find that protein evolution in A. thaliana is as strongly affected by biased gene conversion as in the outcrossing species. Last, we perform simulations to show that natural levels of outcrossing in A. thaliana are sufficient to facilitate biased gene conversion despite increased homozygosity due to selfing. Together, our results show that even predominantly selfing taxa are susceptible to biased gene conversion, suggesting that it may constitute an important constraint to adaptation among plant species.

The Interaction of Natural Selection and GC Skew May Drive the Fast Evolution of a Sand Rat Homeobox Gene

Several processes can lead to strong GC skew in localized genomic regions. In most cases, GC skew should not affect conserved amino acids because natural selection will purge deleterious alleles. However, in the gerbil subfamily of rodents, several conserved genes have undergone radical alteration in association with strong GC skew. An extreme example concerns the highly conserved homeobox gene Pdx1, which is uniquely divergent and GC rich in the sand rat Psammomys obesus and close relatives. Here, we investigate the antagonistic interplay between very rare amino acid changes driven by GC skew and the force of natural selection. Using ectopic protein expression in cell culture, pulse-chase labeling, in vitro mutagenesis, and drug treatment, we compare properties of mouse and sand rat Pdx1 proteins. We find that amino acid change driven by GC skew resulted in altered protein stability, with a significantly longer protein half-life for sand rat Pdx1. Using a reversible inhibitor of the 26S proteasome, MG132, we find that sand rat and mouse Pdx1 are both degraded through the ubiquitin proteasome pathway. However, in vitro mutagenesis reveals this pathway operates through different amino acid residues. We propose that GC skew caused loss of a key ubiquitination site, conserved through vertebrate evolution, and that sand rat Pdx1 evolved or fixed a new ubiquitination site to compensate. Our results give molecular insight into the power of natural selection in the face of maladaptive changes driven by strong GC skew.

The Red Queen model of recombination hot-spot evolution: a theoretical investigation

In humans and many other species, recombination events cluster in narrow and short-lived hot spots distributed across the genome, whose location is determined by the Zn-finger protein PRDM9. To explain these fast evolutionary dynamics, an intra-genomic Red Queen model has been proposed, based on the interplay between two antagonistic forces: biased gene conversion, mediated by double-strand breaks, resulting in hot-spot extinction, followed by positive selection favouring new PRDM9 alleles recognizing new sequence motifs. Thus far, however, this Red Queen model has not been formalized as a quantitative population-genetic model, fully accounting for the intricate interplay between biased gene conversion, mutation, selection, demography and genetic diversity at the PRDM9 locus. Here, we explore the population genetics of the Red Queen model of recombination. A Wright–Fisher simulator was implemented, allowing exploration of the behaviour of the model (mean equilibrium recombination rate, diversity at the PRDM9 locus or turnover rate) as a function of the parameters (effective population size, mutation and erosion rates). In a second step, analytical results based on self-consistent mean-field approximations were derived, reproducing the scaling relations observed in the simulations. Empirical fit of the model to current data from the mouse suggests both a high mutation rate at PRDM9 and strong biased gene conversion on its targets.

This article is part of the themed issue ‘Evolutionary causes and consequences of recombination rate variation in sexual organisms’.

Impact of Recombination on the Base Composition of Bacteria and Archaea

The mutational process in bacteria is biased toward A and T, and most species are GC-rich relative to the mutational input to their genome. It has been proposed that the shift in base composition is an adaptive process-that natural selection operates to increase GC-contents-and there is experimental evidence that bacterial strains with GC-rich versions of genes have higher growth rates than those strains with AT-rich versions expressing identical proteins. Alternatively, a nonadaptive process, GC-biased gene conversion (gBGC), could also increase the GC-content of DNA due to the mechanistic bias of gene conversion events during recombination. To determine what role recombination plays in the base composition of bacterial genomes, we compared the spectrum of nucleotide polymorphisms introduced by recombination in all microbial species represented by large numbers of sequenced strains. We found that recombinant alleles are consistently biased toward A and T, and that the magnitude of AT-bias introduced by recombination is similar to that of mutations. These results indicate that recombination alone, without the intervention of selection, is unlikely to counteract the AT-enrichment of bacterial genomes.

Recombination, meiotic expression and human codon usage

Synonymous codon usage (SCU) varies widely among human genes. In particular, genes involved in different functional categories display a distinct codon usage, which was interpreted as evidence that SCU is adaptively constrained to optimize translation efficiency in distinct cellular states. We demonstrate here that SCU is not driven by constraints on tRNA abundance, but by large-scale variation in GC-content, caused by meiotic recombination, via the non-adaptive process of GC-biased gene conversion (gBGC). Expression in meiotic cells is associated with a strong decrease in recombination within genes. Differences in SCU among functional categories reflect differences in levels of meiotic transcription, which is linked to variation in recombination and therefore in gBGC. Overall, the gBGC model explains 70% of the variance in SCU among genes. We argue that the strong heterogeneity of SCU induced by gBGC in mammalian genomes precludes any optimization of the tRNA pool to the demand in codon usage.

Analytical Biases Associated with GC-Content in Molecular Evolution

Molecular evolution is being revolutionized by high-throughput sequencing allowing an increased amount of genome-wide data available for multiple species. While base composition summarized by GC-content is one of the first metrics measured in genomes, its genomic distribution is a frequently neglected feature in downstream analyses based on DNA sequence comparisons. Here, we show how base composition heterogeneity among loci and taxa can bias common molecular evolution analyses such as phylogenetic tree reconstruction, detection of natural selection and estimation of codon usage. We then discuss the biological, technical and methodological causes of these GC-associated biases and suggest approaches to overcome them.

Direct Determination of the Mutation Rate in the Bumblebee Reveals Evidence for Weak Recombination-Associated Mutation and an Approximate Rate Constancy in Insects

Accurate knowledge of the mutation rate provides a base line for inferring expected rates of evolution, for testing evolutionary hypotheses and for estimation of key parameters. Advances in sequencing technology now permit direct estimates of the mutation rate from sequencing of close relatives. Within insects there have been three prior such estimates, two in nonsocial insects (Drosophila: 2.8 × 10-9 per bp per haploid genome per generation; Heliconius: 2.9 × 10-9) and one in a social species, the honeybee (3.4 × 10-9). Might the honeybee's rate be ∼20% higher because it has an exceptionally high recombination rate and recombination may be directly or indirectly mutagenic? To address this possibility, we provide a direct estimate of the mutation rate in the bumblebee (Bombus terrestris), this being a close relative of the honeybee but with a much lower recombination rate. We confirm that the crossover rate of the bumblebee is indeed much lower than honeybees (8.7 cM/Mb vs. 37 cM/Mb). Importantly, we find no significant difference in the mutation rates: we estimate for bumblebees a rate of 3.6 × 10-9 per haploid genome per generation (95% confidence intervals 2.38 × 10-9 and 5.37 × 10-9) which is just 5% higher than the estimate that of honeybees. Both genomes have approximately one new mutation per haploid genome per generation. While we find evidence for a direct coupling between recombination and mutation (also seen in honeybees), the effect is so weak as to leave almost no footprint on any between-species differences. The similarity in mutation rates suggests an approximate constancy of the mutation rate in insects.

Open chromatin reveals the functional maize genome

Cellular processes mediated through nuclear DNA must contend with chromatin. Chromatin structural assays can efficiently integrate information across diverse regulatory elements, revealing the functional noncoding genome. In this study, we use a differential nuclease sensitivity assay based on micrococcal nuclease (MNase) digestion to discover open chromatin regions in the maize genome. We find that maize MNase-hypersensitive (MNase HS) regions localize around active genes and within recombination hotspots, focusing biased gene conversion at their flanks. Although MNase HS regions map to less than 1% of the genome, they consistently explain a remarkably large amount (∼40%) of heritable phenotypic variance in diverse complex traits. MNase HS regions are therefore on par with coding sequences as annotations that demarcate the functional parts of the maize genome. These results imply that less than 3% of the maize genome (coding and MNase HS regions) may give rise to the overwhelming majority of phenotypic variation, greatly narrowing the scope of the functional genome.

The Spontaneous Mutation Rate in the Fission Yeast Schizosaccharomyces pombe

The rate at which new mutations arise in the genome is a key factor in the evolution and adaptation of species. Here we describe the rate and spectrum of spontaneous mutations for the fission yeast Schizosaccharomyces pombe, a key model organism with many similarities to higher eukaryotes. We undertook an ∼1700-generation mutation accumulation (MA) experiment with a haploid S. pombe, generating 422 single-base substitutions and 119 insertion-deletion mutations (indels) across the 96 replicates. This equates to a base-substitution mutation rate of 2.00 × 10(-10) mutations per site per generation, similar to that reported for the distantly related budding yeast Saccharomyces cerevisiae. However, these two yeast species differ dramatically in their spectrum of base substitutions, the types of indels (S. pombe is more prone to insertions), and the pattern of selection required to counteract a strong AT-biased mutation rate. Overall, our results indicate that GC-biased gene conversion does not play a major role in shaping the nucleotide composition of the S. pombe genome and suggest that the mechanisms of DNA maintenance may have diverged significantly between fission and budding yeasts. Unexpectedly, CpG sites appear to be excessively liable to mutation in both species despite the likely absence of DNA methylation.

Crossovers are associated with mutation and biased gene conversion at recombination hotspots

Meiosis is a potentially important source of germline mutations, as sites of meiotic recombination experience recurrent double-strand breaks (DSBs). However, evidence for a local mutagenic effect of recombination from population sequence data has been equivocal, likely because mutation is only one of several forces shaping sequence variation. By sequencing large numbers of single crossover molecules obtained from human sperm for two recombination hotspots, we find direct evidence that recombination is mutagenic: Crossovers carry more de novo mutations than nonrecombinant DNA molecules analyzed for the same donors and hotspots. The observed mutations were primarily CG to TA transitions, with a higher frequency of transitions at CpG than non-CpGs sites. This enrichment of mutations at CpG sites at hotspots could predominate in methylated regions involving frequent single-stranded DNA processing as part of DSB repair. In addition, our data set provides evidence that GC alleles are preferentially transmitted during crossing over, opposing mutation, and shows that GC-biased gene conversion (gBGC) predominates over mutation in the sequence evolution of hotspots. These findings are consistent with the idea that gBGC could be an adaptation to counteract the mutational load of recombination.

The evolution of genomic GC content undergoes a rapid reversal within the genus Plasmodium

The genome of the malarial parasite Plasmodium falciparum is extremely AT rich. This bias toward a low GC content is a characteristic of several, but not all, species within the genus Plasmodium. We compared 4283 orthologous pairs of protein-coding sequences between Plasmodium falciparum and the less AT-biased Plasmodium vivax. Our results indicate that the common ancestor of these two species was also extremely AT rich. This means that, although there was a strong bias toward A+T during the early evolution of the ancestral Plasmodium lineage, there was a subsequent reversal of this trend during the more recent evolution of some species, such as P. vivax. Moreover, we show that not only is the P. vivax genome losing its AT richness, it is actually gaining a very significant degree of GC richness. This example illustrates the potential volatility of nucleotide content during the course of molecular evolution. Such reversible fluxes in nucleotide content within lineages could have important implications for phylogenetic reconstruction based on molecular sequence data.

Germline methylation patterns determine the distribution of recombination events in the dog genome

The positive-regulatory domain containing nine gene, PRDM9, which strongly associates with the location of recombination events in several vertebrates, is inferred to be inactive in the dog genome. Here, we address several questions regarding the control of recombination and its influence on genome evolution in dogs. First, we address whether the association between CpG islands (CGIs) and recombination hotspots is generated by lack of methylation, GC-biased gene conversion (gBGC), or both. Using a genome-wide dog single nucleotide polymorphism data set and comparisons of the dog genome with related species, we show that recombination-associated CGIs have low CpG mutation rates, and that CpG mutation rate is negatively correlated with recombination rate genome wide, indicating that nonmethylation attracts the recombination machinery. We next use a neighbor-dependent model of nucleotide substitution to disentangle the effects of CpG mutability and gBGC and analyze the effects that loss of PRDM9 has on these rates. We infer that methylation patterns have been stable during canid genome evolution, but that dog CGIs have experienced a drastic increase in substitution rate due to gBGC, consistent with increased levels of recombination in these regions. We also show that gBGC is likely to have generated many new CGIs in the dog genome, but these mostly occur away from genes, whereas the number of CGIs in gene promoter regions has not increased greatly in recent evolutionary history. Recombination has a major impact on the distribution of CGIs that are detected in the dog genome due to the interaction between methylation and gBGC. The results indicate that germline methylation patterns are the main determinant of recombination rates in the absence of PRDM9.

Linking great apes genome evolution across time scales using polymorphism-aware phylogenetic models

The genomes of related species contain valuable information on the history of the considered taxa. Great apes in particular exhibit variation of evolutionary patterns along their genomes. However, the great ape data also bring new challenges, such as the presence of incomplete lineage sorting and ancestral shared polymorphisms. Previous methods for genome-scale analysis are restricted to very few individuals or cannot disentangle the contribution of mutation rates and fixation biases. This represents a limitation both for the understanding of these forces as well as for the detection of regions affected by selection. Here, we present a new model designed to estimate mutation rates and fixation biases from genetic variation within and between species. We relax the assumption of instantaneous substitutions, modeling substitutions as mutational events followed by a gradual fixation. Hence, we straightforwardly account for shared ancestral polymorphisms and incomplete lineage sorting. We analyze genome-wide synonymous site alignments of human, chimpanzee, and two orangutan species. From each taxon, we include data from several individuals. We estimate mutation rates and GC-biased gene conversion intensity. We find that both mutation rates and biased gene conversion vary with GC content. We also find lineage-specific differences, with weaker fixation biases in orangutan species, suggesting a reduced historical effective population size. Finally, our results are consistent with directional selection acting on coding sequences in relation to exonic splicing enhancers.

Less is more in mammalian phylogenomics: AT-rich genes minimize tree conflicts and unravel the root of placental mammals

Despite the rapid increase of size in phylogenomic data sets, a number of important nodes on animal phylogeny are still unresolved. Among these, the rooting of the placental mammal tree is still a controversial issue. One difficulty lies in the pervasive phylogenetic conflicts among genes, with each one telling its own story, which may be reliable or not. Here, we identified a simple criterion, that is, the GC content, which substantially helps in determining which gene trees best reflect the species tree. We assessed the ability of 13,111 coding sequence alignments to correctly reconstruct the placental phylogeny. We found that GC-rich genes induced a higher amount of conflict among gene trees and performed worse than AT-rich genes in retrieving well-supported, consensual nodes on the placental tree. We interpret this GC effect mainly as a consequence of genome-wide variations in recombination rate. Indeed, recombination is known to drive GC-content evolution through GC-biased gene conversion and might be problematic for phylogenetic reconstruction, for instance, in an incomplete lineage sorting context. When we focused on the AT-richest fraction of the data set, the resolution level of the placental phylogeny was greatly increased, and a strong support was obtained in favor of an Afrotheria rooting, that is, Afrotheria as the sister group of all other placentals. We show that in mammals most conflicts among gene trees, which have so far hampered the resolution of the placental tree, are concentrated in the GC-rich regions of the genome. We argue that the GC content-because it is a reliable indicator of the long-term recombination rate-is an informative criterion that could help in identifying the most reliable molecular markers for species tree inference.

Evolution of recombination and genome structure in eusocial insects

Eusocial Hymenoptera, such as the European honey bee, Apis mellifera, have the highest recombination rates of multicellular animals.(1) Recently, we showed(2) that a side-effect of recombination in the honey bee, GC biased gene conversion (bGC), helps maintain the unusual bimodal GC-content distribution of the bee genome by increasing GC-content in high recombination areas while low recombination areas are losing GC-content because of biased AT mutations and low rates of bGC. Although the very high recombination rate of A. mellifera makes GC-content evolution easier to study, the pattern is consistent with results found in many other species including mammals and yeast.(3) Also consistent across phyla is the association of higher genetic diversity and divergence with high GC and high recombination areas.(4) (,) (5) Finally, we showed that genes overexpressed in the brains of workers cluster in GC-rich genomic areas with the highest rates of recombination and molecular evolution.(2) In this Addendum we present a conceptual model of how eusociality and high recombination rates may co-evolve.

GC-biased gene conversion in yeast is specifically associated with crossovers: molecular mechanisms and evolutionary significance

GC-biased gene conversion (gBGC) is a process associated with recombination that favors the transmission of GC alleles over AT alleles during meiosis. gBGC plays a major role in genome evolution in many eukaryotes. However, the molecular mechanisms of gBGC are still unknown. Different steps of the recombination process could potentially cause gBGC: the formation of double-strand breaks (DSBs), the invasion of the homologous or sister chromatid, and the repair of mismatches in heteroduplexes. To investigate these models, we analyzed a genome-wide data set of crossovers (COs) and noncrossovers (NCOs) in Saccharomyces cerevisiae. We demonstrate that the overtransmission of GC alleles is specific to COs and that it occurs among conversion tracts in which all alleles are converted from the same donor haplotype. Thus, gBGC results from a process that leads to long-patch repair. We show that gBGC is associated with longer tracts and that it is driven by the nature (GC or AT) of the alleles located at the extremities of the tract. These observations invalidate the hypotheses that gBGC is due to the base excision repair machinery or to a bias in DSB formation and suggest that in S. cerevisiae, gBGC is caused by the mismatch repair (MMR) system. We propose that the presence of nicks on both DNA strands during CO resolution could be the cause of the bias in MMR activity. Our observations are consistent with the hypothesis that gBGC is a nonadaptive consequence of a selective pressure to limit the mutation rate in mitotic cells.

Long-range bidirectional strand asymmetries originate at CpG islands in the human genome

In the human genome, CpG islands (CGIs), which are GC- and CpG-rich sequences, are associated with transcription starting sites (TSSs); in addition, there is evidence that CGIs harbor origins of bidirectional replication (OBRs) and are preferred sites for heteroduplex formation during recombination. Transcription, replication, and recombination processes are known to induce specific mutational patterns in various genomes, and therefore, these patterns are expected to be found around CGIs. We use triple alignments of human, chimp, and macaque to compute the rates of nucleotide substitutions in up to 1 Mbps long intergenic regions on both sides of CGIs. Our analysis revealed that around a CGI there is an asymmetry between complementary substitution rates that is similar to the one that found around the OBR in bacteria. We hypothesize that these asymmetries are induced by differences in the replication of the leading and lagging strand and that a significant number of CGIs overlap OBRs. Within CGIs, we observed a mutational signature of GC-biased gene conversion that is associated with recombination. We suggest that recombination has played a major role in the creation of CGIs.