Testing the neutral theory of molecular evolution using genomic data: a comparison of the human and bovine transcriptome

Despite growing evidence of rapid evolution in protein coding genes, the contribution of positive selection to intra- and interspecific differences in protein coding regions of the genome is unclear. We attempted to see if genes coding for secreted proteins and genes with narrow expression, specifically those preferentially expressed in the mammary gland, have diverged at a faster rate between domestic cattle (Bos taurus) and humans (Homo sapiens) than other genes and whether positive selection is responsible. Using a large data set, we identified groups of genes based on secretion and expression patterns and compared them for the rate of nonsynonymous (dN) and synonymous (dS) substitutions per site and the number of radical (Dr) and conservative (Dc) amino acid substitutions. We found evidence of rapid evolution in genes with narrow expression, especially for those expressed in the liver and mammary gland and for genes coding for secreted proteins. We compared common human polymorphism data with human-cattle divergence and found that genes with high evolutionary rates in human-cattle divergence also had a large number of common human polymorphisms. This argues against positive selection causing rapid divergence in these groups of genes. In most cases dN/dS ratios were lower in human-cattle divergence than in common human polymorphism presumably due to differences in the effectiveness of purifying selection between long-term divergence and short-term polymorphism.     

Sex-linked mammalian sperm proteins evolve faster than autosomal ones

X-linked genes can evolve slower or faster depending on whether most recessive, or at least partially recessive alleles are deleterious or beneficial due to their hemizygous expression in males. Molecular studies of X chromosome divergence have provided conflicting evidence for both a higher and lower rate of nucleotide substitution at both synonymous and nonsynonymous sites, depending on the nucleotide sites sampled. Using human and mouse orthologous genes, we tested the hypothesis that genes encoding male-specific sperm proteins are evolving faster on the X chromosome compared with autosomes. X-linked sperm proteins have an average nonsynonymous mutation rate almost twice as high as sperm genes found on autosomes, unlike other tissue-specific genes, where no significant difference in the nonsynonymous mutation rate between the X chromosome and autosomes was found. However, no difference was found in the average synonymous mutation rate of X-linked versus autosomal sperm proteins, which along with corresponding higher values of Ka/Ks in X-linked sperm proteins suggest that differences in selective forces and not mutation rates are the underlying cause of higher X-linked mammalian sperm protein divergence.     

Parallel adaptive divergence among geographically diverse human populations

Few genetic differences between human populations conform to the classic model of positive selection, in which a newly arisen mutation rapidly approaches fixation in one lineage, suggesting that adaptation more commonly occurs via moderate changes in standing variation at many loci. Detecting and characterizing this type of complex selection requires integrating individually ambiguous signatures across genomically and geographically extensive data. Here, we develop a novel approach to test the hypothesis that selection has favored modest divergence at particular loci multiple times in independent human populations. We find an excess of SNPs showing non-neutral parallel divergence, enriched for genic and nonsynonymous polymorphisms in genes encompassing diverse and often disease related functions. Repeated parallel evolution in the same direction suggests common selective pressures in disparate habitats. We test our method with extensive coalescent simulations and show that it is robust to a wide range of demographic events. Our results demonstrate phylogenetically orthogonal patterns of local adaptation caused by subtle shifts at many widespread polymorphisms that likely underlie substantial phenotypic diversity.     

High proportions of deleterious polymorphisms in constrained human genes

Previous studies on human mitochondrial genomes showed that the ratio of intra-specific diversities at nonsynonymous-to-synonymous positions was two to ten times higher than the ratio of interspecific divergences at these positions, suggesting an excess of slightly deleterious nonsynonymous polymorphisms. However, such an overabundance of nonsynonymous single nucleotide polymorphisms (SNPs) was not found in human nuclear genomes. Here, genome-wide estimates using >14,000 human-chimp nuclear genes and 1 million SNPs from four human genomes showed a significant proportion of deleterious nonsynonymous SNPs (～ 15%). Importantly, this study reveals a negative correlation between the magnitude of selection pressure and the proportion of deleterious SNPs on human genes. The proportion of deleterious amino acid replacement polymorphisms is 3.5 times higher in genes under high purifying selection compared with that in less constrained genes (28% vs. 8%). These results are explained by differences in the extent of contribution of mildly deleterious mutations to diversity and substitution.     

Mammalian housekeeping genes evolve more slowly than tissue-specific genes

Do housekeeping genes, which are turned on most of the time in almost every tissue, evolve more slowly than genes that are turned on only at specific developmental times or tissues? Recent large-scale gene expression studies enable us to have a better definition of housekeeping genes and to address the above question in detail. In this study, we examined 1581 human-mouse orthologous gene pairs for their patterns of sequence evolution, contrasting housekeeping genes with tissue-specific genes. Our results show that, in comparison to tissue-specific genes, housekeeping genes on average evolve more slowly and are under stronger selective constraints as reflected by significantly smaller values of Ka/Ks. Besides stronger purifying selection, we explored several other factors that can possibly slow down nonsynonymous rates in housekeeping genes. Although mutational bias might slightly slow the nonsynonymous rates in housekeeping genes, it is unlikely to be the major cause of the rate difference between the two types of genes. The codon usage pattern of housekeeping genes does not seem to differ from that of tissue-specific genes. Moreover, contrary to the old textbook concept, we found that approximately 74% of the housekeeping genes in our study belong to multigene families, not significantly different from that of the tissue-specific genes ( approximately 70%). Therefore, the stronger selective constraints on housekeeping genes are not due to a lower degree of genetic redundancy.     

Human-chimpanzee DNA sequence variation in the four major genes of the renin angiotensin system

The renin angiotensin system (RAS) is involved in blood pressure control and water/sodium metabolism. The genes encoding the proteins of this system are candidate genes for essential hypertension. The RAS involves four main molecules: angiotensinogen, renin, angiotensin I-converting enzyme, and the angiotensin II type 1 receptor (encoded by the genes AGT, REN, DCP1, and AGTR1, respectively). We performed a molecular screening over 17,037 bp of the coding and 5' and 3' untranslated regions of these genes, from three to six common chimpanzees. We identified 44 single-nucleotide polymorphisms (SNPs) in chimpanzee samples, including 18 coding-region SNPs, 5 of which led to an amino acid replacement. We observed common and different features at various sites (synonymous, nonsynonymous, and noncoding) within and between the four chimpanzee genes: (1) the nucleotide diversity at noncoding sites was similar; (2) the nucleotide diversity at nonsynonymous sites was low, probably reflecting purifying selection, except for the AGT gene; (3) the nucleotide diversity at synonymous sites, which was dependent on the G+C content at the third position of the codon, was high, except for the AGTR1 gene. Comparison of the chimpanzee SNPs with those previously reported for humans identified 119 sites with fixed differences (including 62 coding sites, 17 of which resulted in amino acid differences between the species). Analysis of polymorphism within species and divergence between species shed light on the evolutionary constraints on these genes. In particular, comparison of the pattern of mutation at polymorphic and fixed sites between humans and chimpanzees suggested that the high G+C content of the DCP1 gene was maintained by positive selection at its silent sites. Finally, we propose 68 ancestral alleles for the human RAS genes and discuss the implications for their use in future hypertension-susceptibility association studies.     

A scan for positively selected genes in the genomes of humans and chimpanzees

Since the divergence of humans and chimpanzees about 5 million years ago, these species have undergone a remarkable evolution with drastic divergence in anatomy and cognitive abilities. At the molecular level, despite the small overall magnitude of DNA sequence divergence, we might expect such evolutionary changes to leave a noticeable signature throughout the genome. We here compare 13,731 annotated genes from humans to their chimpanzee orthologs to identify genes that show evidence of positive selection. Many of the genes that present a signature of positive selection tend to be involved in sensory perception or immune defenses. However, the group of genes that show the strongest evidence for positive selection also includes a surprising number of genes involved in tumor suppression and apoptosis, and of genes involved in spermatogenesis. We hypothesize that positive selection in some of these genes may be driven by genomic conflict due to apoptosis during spermatogenesis. Genes with maximal expression in the brain show little or no evidence for positive selection, while genes with maximal expression in the testis tend to be enriched with positively selected genes. Genes on the X chromosome also tend to show an elevated tendency for positive selection. We also present polymorphism data from 20 Caucasian Americans and 19 African Americans for the 50 annotated genes showing the strongest evidence for positive selection. The polymorphism analysis further supports the presence of positive selection in these genes by showing an excess of high-frequency derived nonsynonymous mutations.     

Synonymous and nonsynonymous polymorphisms versus divergences in bacterial genomes

Comparison of the ratio of nonsynonymous to synonymous polymorphisms within species with the ratio of nonsynonymous to synonymous substitutions between species has been widely used as a supposed indicator of positive Darwinian selection, with the ratio of these 2 ratios being designated as a neutrality index (NI). Comparison of genome-wide polymorphism within 12 species of bacteria with divergence from an outgroup species showed substantial differences in NI among taxa. A low level of nonsynonymous polymorphism at a locus was the best predictor of NI < 1, rather than a high level of nonsynonymous substitution between species. Moreover, genes with NI < 1 showed a strong tendency toward the occurrence of rare nonsynonymous polymorphisms, as expected under the action of ongoing purifying selection. Thus, our results are more consistent with the hypothesis that a high relative rate of between-species nonsynonymous substitution reflects mainly the action of purifying selection within species to eliminate slightly deleterious mutations rather than positive selection between species. This conclusion is consistent with previous results highlighting an important role of slightly deleterious variants in bacterial evolution and suggests caution in the use of the McDonald-Kreitman test and related statistics as tests of positive selection.     

A catalog of neutral and deleterious polymorphism in yeast

The abundance and identity of functional variation segregating in natural populations is paramount to dissecting the molecular basis of quantitative traits as well as human genetic diseases. Genome sequencing of multiple organisms of the same species provides an efficient means of cataloging rearrangements, insertion, or deletion polymorphisms (InDels) and single-nucleotide polymorphisms (SNPs). While inbreeding depression and heterosis imply that a substantial amount of polymorphism is deleterious, distinguishing deleterious from neutral polymorphism remains a significant challenge. To identify deleterious and neutral DNA sequence variation within Saccharomyces cerevisiae, we sequenced the genome of a vineyard and oak tree strain and compared them to a reference genome. Among these three strains, 6% of the genome is variable, mostly attributable to variation in genome content that results from large InDels. Out of the 88,000 polymorphisms identified, 93% are SNPs and a small but significant fraction can be attributed to recent interspecific introgression and ectopic gene conversion. In comparison to the reference genome, there is substantial evidence for functional variation in gene content and structure that results from large InDels, frame-shifts, and polymorphic start and stop codons. Comparison of polymorphism to divergence reveals scant evidence for positive selection but an abundance of evidence for deleterious SNPs. We estimate that 12% of coding and 7% of noncoding SNPs are deleterious. Based on divergence among 11 yeast species, we identified 1,666 nonsynonymous SNPs that disrupt conserved amino acids and 1,863 noncoding SNPs that disrupt conserved noncoding motifs. The deleterious coding SNPs include those known to affect quantitative traits, and a subset of the deleterious noncoding SNPs occurs in the promoters of genes that show allele-specific expression, implying that some cis-regulatory SNPs are deleterious. Our results show that the genome sequences of both closely and distantly related species provide a means of identifying deleterious polymorphisms that disrupt functionally conserved coding and noncoding sequences.     

Effects of natural selection on interpopulation divergence at polymorphic sites in human protein-coding Loci

To develop new strategies for searching for genetic associations with complex human diseases, we analyzed 2784 single-nucleotide polymorphisms (SNPs) in 396 protein-coding genes involved in biological processes relevant to cancer and other complex diseases, with respect to gene diversity within samples of individuals representing the three major historic human populations (African, European, and Asian) and with respect to interpopulation genetic distance. Reduced levels of both intrapopulation gene diversity and interpopulation genetic distance were seen in the case of SNPs located within the 5'-UTR and at nonsynonymous SNPs, causing radical changes to protein structure. Reduction of gene diversity at SNP loci in these categories was evidence of purifying selection acting at these sites, which in turn causes a reduction in interpopulation divergence. By contrast, a small number of SNP sites in these categories revealed unusually high genetic distances between the two most diverged populations (African and Asian); these loci may have historically been subject to divergent selection pressures.     

On the Pattern of Polymorphisms at Major Histocompatibility Complex Loci

The pattern of polymorphisms at major histocompatibility complex loci was studied by computer simulations and by DNA sequence analysis. Two types of selection, overdominance plus short-term selection and maternal–fetal incompatibility, were simulated for a gene family with intra- and interlocus gene conversion. Both types of selection were found to be consistent with the observed patterns of polymorphisms. It was also found that the more interlocus conversion occurs, the higher the divergence becomes at both nonsynonymous and synonymous sites. The ratio of nonsynonymous-to-synonymous divergence among alleles decreases as the interlocus conversion rate increases. These results agree with the interpretation that the rate of interlocus conversion is lower in human genes than in genes of other nonprimate mammals. This is because, in the latter, synonymous divergence at the ARS (antigen recognition site) is often higher than that at the non-ARS, whereas in the former, this is not so. Also, the ratio of nonsynonymous to synonymous substitutions at the ARS tends to be higher in human genes than in other mammalian genes. The main difference between overdominance plus short-term selection and maternal–fetal interaction is that the number of alleles and heterozygosity per locus are higher in the latter than in the former under the presumed selection intensities. However, the average divergence among alleles tends to be lower in the latter than in the former under similar conditions. Received: 30 September 1997 / Accepted: 15 December 1997     

Inference of expression-dependent negative selection based on polymorphism and divergence in the human genome

There is a mounting evidence for the correlation between the gene expression pattern and sequence divergence. However, little is known about the relationship between the gene expression pattern and polymorphism. We compiled the gene expression, polymorphism, and divergence data from the public databases of the human genome. The ratios of nonsynonymous (A) to synonymous (S) substitutions in polymorphism and divergence in the human genome were strongly influenced by the expression pattern and breadth of genes and showed strong correlations. Among the tissues we analyzed, the brain-expressed genes have the smallest and the liver-expressed genes have the largest proportion of amino acid changes both in polymorphism and divergence. The analysis implies that negative selection is the primary factor affecting expression-dependent gene evolution and the prevalent but nonuniform distribution of slightly deleterious mutations in the genome. Although the genes under relaxed negative selection evolved faster than the other genes, these genes are even more liable to slightly deleterious mutations in the population. On the other hand, nonneutral mutations in the highly conservative genes, such as brain-expressed and housekeeping genes, are largely deleterious and eliminated before they enter the population.     

A role for selection in regulating the evolutionary emergence of disease-causing and other coding CAG repeats in humans and mice

The evolutionary expansion of CAG repeats in human triplet expansion disease genes is intriguing because of their deleterious phenotype. In the past, this expansion has been suggested to reflect a broad genomewide expansion of repeats, which would imply that mutational and evolutionary processes acting on repeats differ between species. Here, we tested this hypothesis by analyzing repeat- and flanking-sequence evolution in 28 repeat-containing genes that had been sequenced in humans and mice and by considering overall lengths and distributions of CAG repeats in the two species. We found no evidence that these repeats were longer in humans than in mice. We also found no evidence for preferential accumulation of CAG repeats in the human genome relative to mice from an analysis of the lengths of repeats identified in sequence databases. We then investigated whether sequence properties, such as base and amino acid composition and base substitution rates, showed any relationship to repeat evolution. We found that repeat-containing genes were enriched in certain amino acids, presumably as the result of selection, but that this did not reflect underlying biases in base composition. We also found that regions near repeats showed higher nonsynonymous substitution rates than the remainder of the gene and lower nonsynonymous rates in genes that contained a repeat in both the human and the mouse. Higher rates of nonsynonymous mutation in the neighborhood of repeats presumably reflect weaker purifying selection acting in these regions of the proteins, while the very low rate of nonsynonymous mutation in proteins containing a CAG repeat in both species presumably reflects a high level of purifying selection. Based on these observations, we propose that the mutational processes giving rise to polyglutamine repeats in human and murine proteins do not differ. Instead, we propose that the evolution of polyglutamine repeats in proteins results from an interplay between mutational processes and selection.     

Pattern of Nucleotide Substitutions in Growth Hormone-Prolactin Gene Family: A Paradigm for Evolution by Gene Duplication

The growth hormone-prolactin gene family in mammals is an interesting example of evolution by gene duplication. Divergence among members of duplicated gene families and among species was examined by using reported gene sequences of growth hormone, prolactin and their receptors. Sequence divergence among species was found to show a general tendency in which a generation-time effect is pronounced for synonymous substitutions but not so for nonsynonymous substitutions. Divergence among duplicated genes is characterized by the relatively high rate of nonsynonymous substitutions, i.e., the rate is close to that of synonymous ones. In view of the stage- and tissue-specific expression of duplicated genes, some of the amino acid substitutions among duplicated genes is likely to be caused by positive Darwinian selection.     

Shifting paradigm of association studies: value of rare single-nucleotide polymorphisms

Currently, single-nucleotide polymorphisms (SNPs) with minor allele frequency (MAF) of >5% are preferentially used in case-control association studies of common human diseases. Recent technological developments enable inexpensive and accurate genotyping of a large number of SNPs in thousands of cases and controls, which can provide adequate statistical power to analyze SNPs with MAF <5%. Our purpose was to determine whether evaluating rare SNPs in case-control association studies could help identify causal SNPs for common diseases. We suggest that slightly deleterious SNPs (sdSNPs) subjected to weak purifying selection are major players in genetic control of susceptibility to common diseases. We compared the distribution of MAFs of synonymous SNPs with that of nonsynonymous SNPs (1) predicted to be benign, (2) predicted to be possibly damaging, and (3) predicted to be probably damaging by PolyPhen. Our sources of data were the International HapMap Project, ENCODE, and the SeattleSNPs project. We found that the MAF distribution of possibly and probably damaging SNPs was shifted toward rare SNPs compared with the MAF distribution of benign and synonymous SNPs that are not likely to be functional. We also found an inverse relationship between MAF and the proportion of nsSNPs predicted to be protein disturbing. On the basis of this relationship, we estimated the joint probability that a SNP is functional and would be detected as significant in a case-control study. Our analysis suggests that including rare SNPs in genotyping platforms will advance identification of causal SNPs in case-control association studies, particularly as sample sizes increase.     

Positive and negative selection on the human genome.

The distinction between deleterious, neutral, and adaptive mutations is a fundamental problem in the study of molecular evolution. Two significant quantities are the fraction of DNA variation in natural populations that is deleterious and destined to be eliminated and the fraction of fixed differences between species driven by positive Darwinian selection. We estimate these quantities using the large number of human genes for which there are polymorphism and divergence data. The fraction of amino acid mutations that is neutral is estimated to be 0.20 from the ratio of common amino acid (A) to synonymous (S) single nucleotide polymorphisms (SNPs) at frequencies of > or =15%. Among the 80% of amino acid mutations that are deleterious at least 20% of them are only slightly deleterious and often attain frequencies of 1-10%. We estimate that these slightly deleterious mutations comprise at least 3% of amino acid SNPs in the average individual or at least 300 per diploid genome. This estimate is not sensitive to human population history. The A/S ratio of fixed differences is greater than that of common SNPs and suggests that a large fraction of protein divergence is adaptive and driven by positive Darwinian selection.     

Rates of protein evolution are positively correlated with developmental timing of expression during mouse spermatogenesis

Male reproductive genes often evolve very rapidly, and sexual selection is thought to be a primary force driving this divergence. We investigated the molecular evolution of 987 genes expressed at different times during mouse spermatogenesis to determine if the rate of evolution and the intensity of positive selection vary across stages of male gamete development. Using mouse-rat orthologs, we found that rates of protein evolution were positively correlated with the developmental timing of expression. Genes expressed early in spermatogenesis had rates of divergence similar to the genome median, while genes expressed after the onset of meiosis were found to evolve much more quickly. Rates of protein evolution were fastest for genes expressed during the dramatic morphogenesis of round spermatids into spermatozoa. Late-expressed genes were also more likely to be specific to the male germline. To test for evidence of positive selection, we analyzed the ratio of nonsynonymous to synonymous changes using a maximum likelihood framework in comparisons among mouse, rat, and human. Many genes showed evidence of positive selection, and most of these genes were expressed late in spermatogenesis and were testis specific. Overall, these data suggest that the intensity of positive selection associated with the evolution of male gametes varies considerably across development and acts primarily on phenotypes that develop late in spermatogenesis.     

Ethnic differences in the linkage disequilibrium and distribution of single-nucleotide polymorphisms in 35 candidate genes for cardiovascular diseases

Single-nucleotide polymorphisms (SNPs) are commonly used to study genetics for common diseases and predict pharmacological response. The selection of likely informative SNPs in association studies depends on their allele frequencies and on the linkage disequilibrium (LD) between SNPs, both of which may show interethnic differences. Among three populations consisting of 207 Chinese, 858 French, and 395 Spanish, we compared the allele frequency distributions of 64 intragenic SNPs of 35 candidate genes for cardiovascular diseases. Twenty-eight of these SNPs from 12 genes were also examined for intragenic LD. About 20% of SNPs were restricted to Europeans, being monomorphic in Chinese, among them mostly nonsynonymous coding SNPs and noncoding SNPs. Only 1.6% of SNPs were specific in Chinese, commensurate with the detection of these SNPs almost exclusively in Caucasians. Similarly, these SNPs were more often rare (<0.1 minor allele frequency) in Chinese (44.3%) than in Europeans (31.1%). The variant allele frequencies and intermarker LDs in terms of D' and Delta(2) were highly correlated between French and Spanish populations (r = 0.98-0.99, p < 0.001). However, only moderate correlations of allele frequencies and D' were found between the Chinese and the European populations (r = 0.7 and 0.3, respectively) despite a high correlation of Delta(2) values (r = 0.8). These results suggest that ethnic considerations are important in the selection of SNPs for association studies of candidate genes, as this may affect the power of the study as well as the likelihood of asking relevant questions and getting medically meaningful answers.     

Conditional neutrality at two adjacent NBS-LRR disease resistance loci in natural populations of Arabidopsis lyrata

We examined patterns of nucleotide diversity at a genomic region containing two linked candidate disease resistance (NBS-LRR) genes in seven populations of the outcrossing plant Arabidopsis lyrata. In comparison with two adjacent control genes and neutral reference genes across the genome, the NBS-LRR genes exhibited elevated nonsynonymous variation and a large number of major-effect polymorphisms causing early stop codons and/or frameshift mutations. In contrast, analysis of synonymous diversity provided no evidence that the region was subject to long-term balancing selection or recent selective sweeps in any of the seven populations surveyed. Also in contrast with earlier surveys of one of these R genes, there was no evidence that the resistance genes or the major-effect mutations were subject to elevated differentiation between populations. We suggest that conditional neutrality in the absence of the corresponding pathogen, rather than long-term balancing selection or local adaptation, may in some circumstances be a significant cause of elevated functional polymorphism at R genes. In contrast with the R genes, analysis of diversity and differentiation at the flanking FERONIA locus showed high population divergence, suggesting local adaptation on this locus controlling male-female signalling during fertilization.