* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download "Positive Selection on Genes in Humans as Compared to
Ridge (biology) wikipedia , lookup
Biology and consumer behaviour wikipedia , lookup
Metagenomics wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Gene expression profiling wikipedia , lookup
Public health genomics wikipedia , lookup
Pathogenomics wikipedia , lookup
Human–animal hybrid wikipedia , lookup
Minimal genome wikipedia , lookup
The Selfish Gene wikipedia , lookup
Polymorphism (biology) wikipedia , lookup
Genome (book) wikipedia , lookup
Site-specific recombinase technology wikipedia , lookup
Human genome wikipedia , lookup
History of genetic engineering wikipedia , lookup
Human Genome Project wikipedia , lookup
Designer baby wikipedia , lookup
Human genetic variation wikipedia , lookup
Adaptive evolution in the human genome wikipedia , lookup
Population genetics wikipedia , lookup
Genome evolution wikipedia , lookup
Positive Selection on Genes in Humans as Compared to Chimpanzees Advanced article Article Contents . Introduction . Positive Selection in Human . Genome-wide Scans for Positive Selection in Human and Chimpanzee . Implications Margaret A Bakewell, University of Michigan, Ann Arbor, Michigan, USA Jianzhi Zhang, University of Michigan, Ann Arbor, Michigan, USA Online posting date: 15th July 2008 Positive selection has been identified and extensively studied in many human genes. In contrast, positive selection in the closest human relative, chimpanzee, has been largely unstudied until recently. With the complete sequencing of the human and chimpanzee genomes, a comparison of the types and numbers of genes that underwent positive selection in the two species is possible. The types of genes under positive selection in human and chimpanzee are different with regard to gene function, but no particular type accounts for a large part of this difference. However, significantly more genes experienced positive selection in the chimpanzee lineage than in the human lineage. Synonymous and nonsynonymous mutations Introduction Selective forces Positive selection occurs when an allele increases in frequency in a population because it confers a survival or reproductive advantage to individuals who possess it. Although positive selection makes it more likely that beneficial mutations will reach high frequencies or become fixed in a population, some advantageous mutations are lost due to the random process of genetic drift. In contrast, many mutations produce deleterious alleles. Those that are strongly deleterious, conferring early death or sterility on the individuals who carry them, are rapidly purged from a population by the action of purifying selection. Others with an only slightly deleterious effect may remain in a population or even increase in frequency by genetic drift or linkage to beneficial alleles at nearby loci. However, many mutations are neither beneficial nor deleterious. These are said to be selectively neutral or nearly neutral. Neutral alleles increase or decrease in frequency within a population by genetic drift. See also: Molecular Evolution: Neutral Theory ELS subject area: Evolution and Diversity of Life How to cite: Bakewell, Margaret A; and, Zhang, Jianzhi (July 2008) Positive Selection on Genes in Humans as Compared to Chimpanzees. In: Encyclopedia of Life Sciences (ELS). John Wiley & Sons, Ltd: Chichester. DOI: 10.1002/9780470015902.a0020856 Mutations that occur in protein-coding deoxyribonucleic acid (DNA) sequences are termed synonymous if they result in no change to the amino acid sequence of the encoded protein due to the degenerate nature of the genetic code. Synonymous mutations are largely selectively neutral. Nonsynonymous mutations that result in a change of amino acid may also be neutral if the change does not affect the function of the encoded protein. If a nonsynonymous mutation does affect the protein function, it is most likely deleterious and subject to purifying selection, but it may be beneficial and therefore subject to positive selection. The processes of mutation, drift and selection combine to determine which alleles are present in a population and which allele will get fixed in the population. See also: Synonymous and Nonsynonymous Rates Statistical tests of positive selection and their application to human and chimp The study of molecular evolution has produced several statistical tests which can be used to identify the signature of positive selection in DNA sequences. Generally, these tests compare differences in DNA or protein sequences between species (interspecific variation) and/or between individuals or populations within a species (intraspecific variation). For human, high-quality completed genome sequences are available. Over 99% of the human genome has been sequenced to an accuracy of less than 1 error per 100 000 nucleotides (International Human Genome Sequencing Consortium, 2004). In addition, intraspecific polymorphism data are available encompassing many individuals from diverse human populations. One major source of such data is the ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 1 Positive Selection on Human Genes Compared to Chimpanzees HapMap project, which collected over 3.1 million single nucleotide polymorphisms (SNPs) from 270 individuals from 4 global populations, estimated to be over one quarter of the common variation in those populations (International HapMap Consortium, 2007). In contrast, the published chimpanzee genome sequence was completed with a quality level of less than 1 error per 1000 bases for only approximately 98% of the genome (Chimpanzee Sequencing and Analysis Consortium, 2005). Additional sequencing efforts have improved the quality of the chimpanzee genome sequence, but not to a level comparable with the human genome. Although a few polymorphism data are available for chimpanzees, no attempt has been made to survey genomewide polymorphism in a large sample of chimpanzees. Therefore, although the full array of tests for positive selection involving both inter- and intraspecific datasets can be applied to human, genome-wide scans for positive selection in chimpanzee are limited to the use of interspecific divergence data, and even these are hampered by the lower quality of the chimp genome sequence. Interspecific divergence data can be used to compare the rate of nonsynonymous substitution with the rate of synonymous substitution. The ratio (o) of nonsynonymous to synonymous substitution rates is expected to equal 1 if sequences are evolving neutrally. A deficiency of nonsynonymous substitutions (o51) indicates the action of purifying selection, where deleterious nonsynonymous mutations are eliminated from the population. An excess of nonsynonymous substitutions (o41) indicates the action of positive selection on favourable nonsynonymous mutations. The o for a given gene can easily be calculated from aligned DNA sequences of coding regions. Since positive selection may act on only a few sites within a protein, whereas the majority of sites are evolving neutrally or under purifying selection, the overall signal of positive selection may be difficult to detect with this approach. A maximum likelihood approach based on estimation of o for different codon sites within a sequence can be applied to detect positive selection that occurs at a limited number of sites within the alignment. Such a test can also be used to identify specific sites within specific branches of a phylogenetic tree that have been subject to positive selection. Briefly, the likelihood of the observed sequences is found under a null model where all codon sites are evolving either neutrally (o 5 1) or under purifying selection (o51). The likelihood of observing the sequences is also computed under an alternative model where some of the codons evolve under positive selection (o41). Positive selection is inferred if the likelihood of the observed sequences under the alternative model is significantly higher than the likelihood with the null model. Positive selection on a particular branch (called the foreground branch) is discovered by limiting the sites evolving under positive selection to that branch. All sites in the remaining branches (background branches) remain under purifying selection or neutral evolution. The direct comparison between human and chimpanzee sequences can identify differences between them, but to distinguish those changes that occurred on the human branch from those that occurred on the chimp branch it is 2 necessary to compare to an outgroup genome. The availability of additional genome sequences such as mouse (Mus musculus) and rhesus macaque (Macaca mulatta) makes it possible to infer ancestral states and therefore to discover which of the differences between human and chimpanzee occurred in the human lineage and which in the chimpanzee lineage. See also: Neutrality and Selection in Molecular Evolution: Statistical Tests Positive Selection in Human Insight into human evolution and disease The study of positive selection in the human lineage is motivated by several factors. Many complex human-specific traits such as advanced cognitive function and speech and language faculties are presumably adaptive and thus may have arisen through the action of positive selection. In this case, identification of genes under positive selection may help to elucidate the genetic basis for such traits and help us to understand more about the origin of the human species. Genes involved in these adaptive human traits may also be implicated in disorders related to them, such as microcephaly and speech deficits. The identification of genes involved can therefore provide insight into the pathology of these disorders. The present environment of humans is significantly different from that in which our ancestors lived. Therefore, traits that were adaptive in the past may now be deleterious. One example might be the so-called thrifty genes, very beneficial to a nutrient-limited population, but deleterious to individuals who carry them in an environment where excess calories are readily available and thus become susceptible to metabolic disorders like diabetes. Genes implicated in alcoholism and hypertension have also been identified as past targets of positive selection (Sabeti et al., 2006). Again, identification of positively selected genes (PSGs) involved in such newly maladaptive processes can inform biomedical research into prevention and treatment of such diseases. Examples of positive selection in human from the candidate gene approach Before the availability of genome sequence, positive selection could be identified in candidate genes. Dozens of genes were identified as having undergone positive selection in the human lineage since its divergence from chimpanzee (reviewed in Sabeti et al., 2006). Many of these genes are involved in processes such as immunity and reproduction, categories that tend to evolve quickly in many lineages. For example, APOBEC3G, an antiviral enzyme, appears to have evolved under positive selection in humans as well as in other primates, presumably as a response to the evolution of viral pathogens to evade immune responses (Zhang and Webb, 2004). An example of a reproductive gene evolving under positive selection in humans is PRM1, a gene involved in sperm competition (Rooney and Zhang, 1999). Indeed, functionally diverse sperm-expressed genes ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net Positive Selection on Human Genes Compared to Chimpanzees including PRM1 are a group of rapidly evolving genes in mammals (Torgerson et al., 2002). Of particular interest in the study of human evolution are those traits that are uniquely human, which have arisen since the divergence of humans and our closest living relatives, the chimpanzees. Examples of positive selection in genes involved in brain anatomy (ASPM, Microcephalin) and speech (FOXP2) confirm that such traits may have been shaped at least in part by positive selection. Although the candidate gene approach has yielded many interesting targets of selection, analysis of the genome-wide distribution of the test statistics for positive selection has revealed that some of the genes previously thought to be positively selected may not actually be unusual compared to the range of variation present in the entire genome. For example, CCR5D32, a relatively common allele found in European populations that confers significant resistance to HIV (human immunodeficiency virus) infection was thought to have rapidly increased in frequency due to positive selection, based on the finding of the young age of the advantageous allele (Stephens et al., 1998). However, upon comparison with the range of normal variation in the human genome, the sequences associated with this gene are consistent with neutral evolution (Sabeti et al., 2005). Hypothesis of accelerated positive selection in human Humans appear to have undergone much more dramatic morphological, physiological and behavioural change than chimpanzee since divergence from their common ancestor. For example, fossil data show that human brain size tripled in the last 2–3 million years. It is thus possible that more genes have been subject to positive selection in the human lineage than in the chimpanzee lineage. The knowledge from candidate gene studies that many genes have evolved under positive selection in humans and that some of these genes are involved in the traits that are the hallmarks of our species reinforces this belief. But, a rigorous test of the hypothesis that either all genes or some particular categories of genes experienced heightened rates of positive selection in the human lineage requires an unbiased comparison with a closely related species. If there has been any general or category-specific acceleration of positive selection in the human lineage, a genome-wide scan for positive selection in humans and chimpanzees should reveal an excess of PSGs in the human lineage. Genome-wide Scans for Positive Selection in Human and Chimpanzee Pioneering genome-wide scans for positive selection in human and chimp The first genome-wide scan for PSGs in human and chimpanzee was performed by Clark et al. (2003), who examined approximately 8000 orthologous genes in human and chimpanzee, using mouse as the outgroup. The authors used a likelihood-ratio test to detect positive selection and ranked all the genes according to their p-value. Since genes evolving neutrally are expected to have o values near 1, this process potentially mixes PSGs with those under relaxed constraint. Furthermore, the models used in this maximum likelihood analysis have subsequently been improved to reduce the false detection of positive selection (Zhang et al., 2005). With these caveats, their analysis showed that genes involved in sensory perception, especially olfaction, are enriched in the group of genes that are likely candidates for positive selection in human, whereas signal transduction was the most enriched category in chimpanzee. Another genome-wide scan for positive selection was undertaken by Arbiza et al. (2006) using an improved branch-site test (Zhang et al., 2005). In this study, mouse, rat and dog were used as outgroup species. Over 9600 genes were tested for positive selection in the human and chimp branches. The authors identified 108 genes as positively selected in humans and 577 in chimpanzee after multiple testing correction. The analysis of the Gene Ontology (GO) functions of these PSGs showed that G-protein-coupled receptors and sensory perception were most enriched in human PSGs, whereas cellular protein metabolism was enriched for chimp PSGs. Caveats for the initial comparisons The interesting findings of these initial forays into genomewide comparison of positive selection in human and chimp must be viewed with caution due to some deficiencies of their design. As mentioned earlier, the first study (Clark et al., 2003) used a test for positive selection that has been shown to produce an unacceptably high level of false positives (Zhang, 2004). This shortcoming potentially mixes genes that have experienced a relaxation of constraint with those truly under positive selection. In addition, these pioneering studies used the mouse genome as the outgroup to distinguish human-specific from chimpanzee-specific PSGs. Primates and rodents diverged approximately 100 million years ago, so the inference of ancestral state within primates based on rodent sequences is potentially inaccurate. The genome of the rhesus macaque which has subsequently become available (Rhesus Macaque Genome Sequencing and Analysis Consortium, 2007) allows for more accurate determination of ancestral state, as the macaque monkey diverged from humans and apes only approximately 25 million years ago. Finally, the chimpanzee genome sequence has a lower quality than the human genome sequence. This is of critical importance for detection of positive selection because sequencing errors are expected to occur randomly with respect to synonymous and nonsynonymous sites. Thus, the expected o for sequencing errors is 1. In contrast, the expected o for most sites in protein-coding DNA is 51. Therefore, high rates of sequencing error, as seen in the chimpanzee genome compared to the human genome, ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 3 Positive Selection on Human Genes Compared to Chimpanzees inflate o and can lead to false detection of positive selection. This prevented the previous studies from making a robust quantitative comparison of the prevalence of positive selection in human and chimpanzee. examined (6.1%). The proportion of human PSGs associated with human diseases is also greater than the fraction of chimp PSGs underlying human diseases, whereas the proportion of chimp PSGs and chimp non-PSGs associated with human disease are not different. Population size affects the strength of selection Unbiased comparison finds more PSGs in chimpanzee than in human To address the caveats noted and to make an impartial comparison of positive selection in human and chimpanzee, Bakewell et al. (2007) performed a scan of nearly 14 000 genes (greater than 50% of the genes in the primate genome) using the improved branch-site test of positive selection (Zhang et al., 2005). Rhesus macaque was used as the outgroup. To ensure that chimpanzee sequence quality did not affect the results, codons containing low-quality bases in the chimp sequence were eliminated. The results of this study show that significantly more genes evolved under positive selection in chimpanzee than in human (Table 1), contrary to the common belief that an enhancement of positive selection in the human lineage led to our dramatic phenotypic differentiation from the human–chimp common ancestor. The excess of PSGs in chimpanzee is true at the nominal significance level of 5% as well as after multiple testing corrections. Seven genes are PSGs in both human and chimpanzee, more than would be expected by chance, indicating some common targets of selection in the two lineages. A randomization test revealed that there are significant differences in the distribution of unshared human and chimp PSGs among biological process and molecular function categories as annotated in the PANTHER database. However, the categories contributing most to the difference are categories such as protein metabolism and ion transport that are not obviously linked to the major phenotypic differences between human and chimp (Figure 1). An analysis of tissue group of highest expression reveals no significant differences of PSG distribution between human and chimp. Human PSGs are more likely than non-PSGs to be associated with human genetic disease: 9.7% of human PSGs are disease-associated, significantly greater than that among the non-PSGs Although chimpanzee has more PSGs than human, the overall o in human is greater for the entire dataset as well as for the 13 508 non-PSGs. Most of these genes are under negative selection as shown by their o551. The higher overall o in human therefore reflects weaker purifying selection. Population genetic data indicate that the long-term effective population size of humans is 3–5-fold smaller than that of chimpanzees. The last common ancestor of human and chimp had an effective population size comparable to that of the chimpanzee lineage. Population genetic theories predict that both purifying selection and positive selection are less effective in a smaller population. Both the greater number of PSGs in chimp and the overall higher o in human are consistent with this prediction. See also: Effective Population Size; Human and Chimpanzee Nucleotide Diversity Implications Patterns of evolution and laws of population genetics apply to human evolution Genes found to be under positive selection in humans by candidate gene approaches are often involved in processes that are also under positive selection in other lineages, such as immunity and reproduction. This demonstrates that humans have been subject to some of the same evolutionary pressures as other taxa. The finding that fewer genes have undergone positive selection in the human lineage than in the chimpanzee lineage while more nonsynonymous changes have been tolerated in human genes is consistent with the population genetic theory, which predicts weaker natural selection in smaller populations. This Table 1 Genic positive selection in human and chimp lineages since their split Comparison Chimp Human Chimp/human ratio p-value Number of PSGs Number of PSGs after Bonferroni correction Number of PSGs at 5% false discovery rate Number of synonymous changes in all genes Number of nonsynonymous changes in all genes Mean o of all genes Mean o of 13 508 non-PSGs 233 21 59 29 644 17 701 0.245 0.238 154 2 2 30 083 19 000 0.259 0.252 1.51 10.5 29.5 0.985 .0932 0.946 0.944 50.0001 50.0001 50.0001 40.05 50.0001 50.0001 50.0001 Source: Reproduced from Bakewell et al. (2007). 4 ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 0.03 0.03 0.02 0.02 Frequency Frequency Positive Selection on Human Genes Compared to Chimpanzees Observed 0.01 0 Observed 0.01 0 90 130 170 210 2 (a) 90 250 170 130 210 250 2 (b) # of PSGs PANTHER category Biological Process Protein metabolism and modification mRNA transcription Anion transport Phosphate transport # of genes Human Chimp 2 2815 9 40 9.09 1144 171 5 6 25 1 6.50 6.29 80 4 0 6.15 Proteolysis 938 2 16 6.04 Ectoderm development mRNA transcription regulation 604 891 8 3 3 17 5.11 4.99 Stress response Fatty acid metabolism 780 169 2 3 14 0 4.85 4.61 Ion transport G-protein mediated signaling 578 682 12 14 7 9 4.49 4.44 Molecular function Lyase 153 6 0 9.22 Phosphatase Nucleic acid binding 194 2597 5 13 0 46 7.69 7.46 Nuclease 2285 12 42 6.68 Transferase Other transporters 1296 208 5 6 25 2 6.50 4.24 (c) Figure 1 Functional differences between human and chimp unshared positively selected genes (PSGs). Human and chimp PSGs show a significantly larger difference in distribution across (a) biological process groups and (b) molecular function groups than by chance (P 5 0.84% and 0.26%, respectively; one-tail randomization test). The 373 unshared human and chimp PSGs were randomly divided into 147 human PSGs and 226 chimp PSGs and w2 was computed. This procedure was repeated 10 000 times to obtain the null distribution of w2. The bars show the frequency distribution of the w2 values in the random divisions and the arrows show the observed w2 values. Here the randomization test is superior to the standard w2 test because the functional groups are not independent from one another and a single gene may belong to more than one group. Similar results are obtained when the seven shared PSGs are included. (c) Biological process and molecular function groups show the greatest differences between human and chimp unshared PSGs, as ranked by individual w2 values. Shown here are the groups that each contributes at least 2% of the total w2 of all groups. Groups with a higher frequency of human PSGs than chimp PSGs are in red, whereas those with a higher frequency of chimp PSGs than human PSGs are in blue. Reproduced from Bakewell et al. (2007). finding demonstrates that the laws of population genetics apply equally to humans despite the dramatic morphological and behavioural changes that have occurred in the human lineage. Many adaptive traits may be unrecognized The biological categories in which human and chimpanzee PSGs differ do not match up well with the expected categories based on the most well-known differences between human and chimpanzee, such as brain size. This suggests that there are adaptive differences between the two species of which we are not aware, or that the genetic basis for the known differences may involve unexpected processes. This is especially true for chimpanzee where the adaptive traits and differences since the human–chimp common ancestor have not been well studied. Indeed, a comprehensive catalogue of differences in ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 5 Positive Selection on Human Genes Compared to Chimpanzees development, anatomy, behaviour, physiology and disease susceptibility between humans and the great apes is a goal, the achievement of which would greatly enhance our ability to robustly define ‘humanness’ (Varki and Altheide, 2005). Given that chimpanzees have more PSGs than humans, it is probably true that they have many unknown adaptations. The catalogue of positive selection is incomplete Genome-wide studies of positive selection using tests that depend only on interspecific divergence data are limited in their power to detect recent or ongoing selection. Proper investigation of selection in these time frames requires the use of polymorphism data. Beneficial alleles in human populations during and after the out-of-Africa migration as they moved into new environments and encountered new pathogens may still have to be fixed and thus be undetectable by studies such as the ones described earlier. See also: Human Evolution: Radiations in the Last 300 000 Years; Modern Human Origins: The ‘Out of Africa’ Debate There is not necessarily a direct correspondence between the amount of molecular change and the amount of phenotypic change. A single nucleotide change can produce pleiotropic effects, as for example with the sickle-cell mutation. The effects of single or few mutations can produce phenotypes with very different fitnesses and be acted on by natural selection, yet are undetectable to a test of positive selection based on o due to their minor impact to the overall substitution rate of the gene. Single genes can have large effects, and it may be that the seemingly large phenotypic changes in modern humans are due to a few changes of large effect rather than an overall acceleration of positive selection. This article focuses on positive selection in genes, but positive selection can also act on regulatory regions of the genome. Because of the very high level of identity (approximately 99%) between human and chimpanzee coding sequences, King and Wilson (1975) proposed that the important differences between human and chimp are likely to be in gene expression rather than in protein sequence. The study of the noncoding regions of the genome will identify many additional candidates for positive selection in human and in chimpanzee (Haygood et al., 2007). References Arbiza L, Dopazo J and Dopazo H (2006) Positive selection, relaxation, and acceleration in the evolution of the human and chimp genome. PLOS Computational Biology 2: e38. Bakewell MA, Shi P and Zhang J (2007) More genes underwent positive selection in chimpanzee evolution than in human evolution. Proceedings of the National Academy of Sciences of the USA 104: 7489–7494. 6 Chimpanzee Sequencing and Analysis Consortium (2005) Initial sequence of the chimpanzee genome and comparison with the human genome. Nature 437: 69–87. Clark AG, Glanowski S, Nielsen R et al. (2003) Inferring nonneutral evolution from human–chimp–mouse orthologous gene trios. Science 302: 1960–1963. Haygood R, Fedrigo O, Hanson B, Yokoyama KD and Wray G A (2007) Promoter regions of many neural- and nutrition-related genes have experienced positive selection during human evolution. Nature Genetics 39: 1140–1144. International HapMap Consortium (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449: 851–861. International Human Genome Sequencing Consortium (2004) Finishing the euchromatic sequence of the human genome. Nature 431: 931–945. King MC and Wilson AC (1975) Evolution at two levels in humans and chimpanzees. Science 188: 107–116. Rhesus Macaque Genome Sequencing and Analysis Consortium (2007) Evolutionary and biomedical insights from the rhesus macaque genome. Science 316: 222–234. Rooney AP and Zhang J (1999) Rapid evolution of a primate sperm protein: relaxation of functional constraint or positive Darwinian selection? Molecular Biology and Evolution 16: 706–710. Sabeti PC, Schaffner SF, Fry B et al. (2006) Positive natural selection in the human lineage. Science 312: 1614–1620. Sabeti PC, Walsh E, Schaffner SF et al. (2005) The case for selection at CCR5-Delta32. PLoS Biology 3: e378. Stephens JC, Reich DE, Goldstein DB et al. (1998) Dating the origin of the CCR5-Delta32 AIDS-resistance allele by the coalescence of haplotypes. American Journal of Human Genetics 62: 1507–1515. Torgerson DG, Kulathinal RJ and Singh RS (2002) Mammalian sperm proteins are rapidly evolving: evidence of positive selection in functionally diverse genes. Molecular Biology and Evolution 19: 1973–1980. Varki A and Altheide TK (2005) Comparing the human and chimpanzee genomes: searching for needles in a haystack. Genome Research 15: 1746–1758. Zhang J (2004) Frequent false detection of positive selection by the likelihood method with branch-site models. Molecular Biology and Evolution 21: 1332–1339. Zhang J, Nielsen R and Yang Z (2005) Evaluation of an improved branch-site likelihood method for detecting positive selection at the molecular level. Molecular Biology and Evolution 22: 2472–2479. Zhang J and Webb DM (2004) Rapid evolution of primate antiviral enzyme APOBEC3G. Human Molecular Genetics 13: 1785–1791. Further Reading Chen FC and Li WH (2001) Genomic divergences between humans and other hominoids and the effective population size of the common ancestor of humans and chimpanzees. American Journal of Human Genetics 68: 444–456. Goodman M, Porter CA, Czelusniak J et al. (1998) Toward a phylogenetic classification of primates based on DNA evidence ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net Positive Selection on Human Genes Compared to Chimpanzees complemented by fossil evidence. Molecular Phylogenetics and Evolution 9: 585–598. Kimura M (1983) The Neutral Theory of Molecular Evolution. Cambridge: Cambridge University Press. Nei M and Kumar S (2000) Molecular Evolution and Phylogenetics. New York: Oxford University Press. Nielsen R (2005) Molecular signatures of natural selection. Annual Review of Genetics 39: 197–218. Olson MV and Varki A (2003) Sequencing the chimpanzee genome: insights into human evolution and disease. Nature Reviews. Genetics 4: 20–28. Shi P, Bakewell MA and Zhang J (2006) Did brain-specific genes evolve faster in humans than in chimpanzees? Trends in Genetics 22: 608–613. Takahata N, Satta Y and Klein J (1995) Divergence time and population size in the lineage leading to modern humans. Theoretical Population Biology 48: 198–221. Taudien S, Ebersberger I, Glockner G and Platzer M (2006) Should the draft chimpanzee sequence be finished? Trends in Genetics 22: 122–125. Wall JD (2003) Estimating ancestral population sizes and divergence times. Genetics 163: 395–404. ENCYCLOPEDIA OF LIFE SCIENCES & 2008, John Wiley & Sons, Ltd. www.els.net 7