Detecting Ongoing Natural Selection In Human Populations Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

In search of natural selection in human populations. Evolution is the change in the allele frequencies over time due to processes such as mutation, drift, migration and natural selection. The search for the footprints of selection and adaptation in humans has always been a widely discussed topic in the field of evolutionary biology as well as in the related disciplines of human genetics and biological anthropology. Although these subject areas address different questions, they come together on the topic of human population genetics, where data on genetic variation is analysed to determine past demographic events and shed more light on the topic of human evolution.

Traditionally, local adaptations in a human population were identified by hypothesizing an environmental variable that would be acting on an observed phenotypic distribution as a selective pressure. This was pertinent in the case of the malaria hypothesis explaining the geographic distribution of the thalassemia and sickle cell alleles and the resistance they conferred to individuals against the malaria parasite (Allison 1954, Brown 1981). Additionally, there have been studies looking at human adaptations in hypoxic, high-altitude environments (Beall at al. 1997), evolution of increased skin pigmentation in human populations exposed to increased solar radiation (Jablonski & Chaplin 2000) and adaptations of the human body mass in response to temperature variance (Katzmarzyk & Leonard 1998) all focusing on observational phenotypes as a response to a hypothesized environmental stress. In these studies, genetic basis of adaptation is implied if there is evidence for inheritance of the phenotype as opposed to processes such as acclimatization which imply phenotypic plasticity. Moreover, the case for natural selection is further strengthened if a strong correlation between the particular phenotype and fitness is established (e.g., Beall et al. 2004).

Although these approaches directly test a given hypothesis concerning natural selection on a genotypic variant, it’s not easy to implement them on a large scale. They require collection of phenotypic data from large samples to achieve statistical significance which is often expensive, time consuming and in some cases impractical. The recent development of technologies and resources in bioinformatics for studying genetic variation enabled investigators to scan the entire genome for signals of natural selection. The rationale behind this approach is the observed differences between regions of DNA which are under positive selection and the ones which are being mostly shaped by genetic drift. Being a purely genotypic approach, it allows researchers to look at genomic data, without collecting phenotypic information, and detect adaptive changes due to significantly smaller selection coefficients which were previously undetectable (Gillespie 1991).

By drawing on data collected regarding human adaptation from the fields of human anthropology, human genetics and evolutionary biology, it may soon be possible to reconstruct the course of natural selection in human evolution and also infer the role of disease in human adaptations. Three studies reporting selection are discussed in this essay. These studies are focusing on the ongoing selection as indicated by variation that is still segregating in humans, rather than mutations that have been fixed in the human population. The aim of this essay is to present a sound case in favour of the currently ongoing selection in the human population and also demonstrate the utility of traditional phenotypic and new genomic techniques in the detection of natural selection.

Detecting Natural Selection

The frequency change of a beneficial allele over time in a given population is significantly different than that of a neutral allele under natural selection. During the initial phases when a beneficial mutation emerges in the population, given that the population is sufficiently large to counter loss by drift, there is a rapid increase in the frequency of this beneficial mutation. Additionally due to the increasing probability of recombination between shorter distances of a beneficial allele, the selection histories of nearby neutral alleles are also strongly correlated. Due to this effect, natural selection generates a local reduction in variation along that genomic region which is tightly linked to the site under selection. In contrast, neutrally evolving genomic regions far away from the alleles under selection do not exhibit such linkages and their pattern of variation is only shaped by genetic drift. Therefore, observing differing patterns of allele variation in a genomic region forms the basis of detecting the footprints of natural selection; which could be strengthened by collecting information on the function of that region and consequently identifying the selective pressure acting on it.

For a short time when a new beneficial allele is introduced into the population, it will be associated with the particular genomic background it appeared on. As this allele quickly increases in frequency, the neutral alleles linked to the selected site will also tend to increase in frequency. Due to the rapid increase in frequency of the beneficial allele, recombination will not have had sufficient time to break associations between it and nearby neutral alleles. This would lead to the emergence of a local pattern of identical haplotypes which is also known as extended haplotype homozygosity (EHH) (Sabeti et al. 2002). If EHH persists in the population at intermediate or high frequencies, it is referred to as a partial or incomplete selective sweep. When a beneficial allele undergoes fixation, all variation present near the selected site will also be fixed and only new neutral mutations arising between the time of emergence and fixation of the beneficial mutation will segregate in the population at that region in low frequencies. Therefore, the pattern expected near a fixed beneficial allele will be a reduction in polymorphism. On the other hand, further away from the site under selection, recombination will break the linkages between a beneficial allele and other neutral alleles. This will result in an increased frequency of derived alleles and because such alleles tend to be rare under neutrality, the occurrence of multiple high frequency derived alleles within a small region, when compared to other populations, would constitute a strong signal of selection in that population (Harris and Diogo 2006).

When selection acts on an allele which is beneficial for only a subgroup of the population, its allele frequency will differ across the population to a greater extent than expected of neutrally evolving variants in the population. Several approaches have been devised to detect such local adaptations in a population. One such approach is using the fixation index FST and its related formulations to detect population differentiation. FST expresses the differences in allele frequencies between two populations (Weir 1996).Variants with unusually large FST values are interpreted as being significantly divergent at that locus from the rest of the population and hence selection is implied (Lewontin & Krakauer 1973).

Additionally if selection is varying across a spatial cline then the frequency of the beneficial allele will vary across the population. Therefore, a strong positive correlation between the intensity of the hypothesized selective pressure and the frequency variation of the beneficial allele will imply spatially varying selection. This approach is particularly appropriate when looking to find adaptations to environmental selective pressures such as amount of solar radiation and temperature (Jablonski & Chaplin 2000, Katzmarzyk & Leonard 1998).

Furthermore by comparing the ratio of synonymous to non-synonymous mutations in coding regions within and between populations can indicate the action of natural selection on the genetic variation of that population. For example, if a gene indeed is evolving neutrally, the ratios of synonymous to non-synonymous mutations within species and among species will be the same. However, if natural selection is affecting a locus, it will generate an excess of non-synonymous relative to synonymous mutations among species. On the other hand, if an excess of non-synonymous mutations are observed within a species (in comparison to among species), one can possibly interpret is as diversifying selection maintaining a higher number of non-synonymous variants in the population.

Natural Selection in the Human Population

During the course of human evolution, we have tremendously expanded our ecological niche to include varying degrees of environment â€" from the heat of the deserts to the cold extremes of the arctic and many more. We have adapted to this huge variation by behavioural and biological responses (Trinkaus, 2005). The essential role of the physical environment in shaping the human phenotypic diversity is clearly evident from the strong correlations between traits such as body mass and temperature (Leonard et al. 2005), basal metabolic rate and latitude (Henry & Rees 1991) and skin pigmentation and solar radiation (Jablonski & Chaplin 2000, Roberts & Kahlon 1976). In general, phenotypic and genotypic studies have provided strong evidence in establishing climate as a strong selective pressure on humans.

Hypoxic environments

Human populations indigenous to high-altitude environments provide a unique opportunity to study natural selection as historically these societies have depended on behavioural and biological responses to environmental stresses due to lack of technology to create suitable microclimates. Two recent studies by Beall et al. (2001) and Brutsaert et al. (2005) have focused on the high-altitude inhabiting populations of Tibetan and Andean plateaus respectively. The indigenous populations of these plateaus have a continuous history occupation of their habitat for over 20,000 years in Tibet and 10,000 years in the Andes (Aldenderfer, 2003). These recent studies have identified unique physiological and behavioural adaptations thought to offset the high-altitude environmental stress, such as the relative hypoventilation (Brutsaert et al., 2005) of the Andean high-altitude natives and the increased nitric oxide exhaled by Tibetan high-altitude natives (Beall et al., 2001). Theoretically if different distinguishing characteristics are present between two high-altitude native groups then it may be inferred that the adaptations are the result of natural selection acting on the original variation present in their respective ancestral populations. In these recent studies different adaptations were found in regard to oxygen delivery and utilization in the two native populations, offering evidence of natural selection resulting in different adaptive results (Beall et al., 2001, Brutsaert et al., 2005).

The stress of low oxygen at high-altitude is clear. Individuals have to adapt to this stress to offset the decreased level of oxygen in every breath and maintain a continuous supply of oxygen to essential oxygen utilizing processes such as the mitochondrial electron exchange. Both indigenous populations have had sufficient opportunity for natural selection to improve oxygen delivery or use. Adaptations to other environmental stresses, such as Plasmodium falciparum or diet containing lactose, have certainly arisen in similar lengths of time (Wiesenfeld, 1967; Tishkoff et al., 2007). However, the oxygen delivery has been conceptualized to be a more complex, integrated process involving many physiological systems including the pulmonary and cardiovascular systems, increasing the predicted time for natural selection to become detectable.

Studies show that the elevated haemoglobin concentration of the Andean high-altitude natives can be mimicked by European natives inhabiting altitudes higher than 1500m and haemoglobin levels normalise at sea level (Okin et al., 1966; Faura et al., 1969; Hochachka et al., 1996b; Ward et al., 2000; Beall, 2001b). Since the elevated haemoglobin is a reversible trait and exhibited in non-indigenous population, it may be interpreted as being an ancestral trait not unique to the high-altitude natives, discounting its emergence as a unique adaptation with genetic basis due to natural selection.

Taking into account for the immigration and mixing of populations with the advent of modern transport, Brutsaert et al. (2005) went on to study the adaptations to hypoxic environments with taking into account the degree of Native American ancestry of individuals being observed. They found that individuals with higher Native American ancestry (>85%) had a lower hypoxic ventilatory response and low ventilation during exercise when compared to control populations (Brutsaert et al., 2005). This study, taking into consideration the interbreeding among different populations, implies a unique high-altitude gene pool due to variations at unknown loci.

The use of traditional phenotypic approaches in these studies have presented evidence for the possibility of natural selection acting on these indigenous population leading to the emergence of a unique high-altitude genetic background, which supports hypoxic environments with adaptations such as increased oxygen saturation of haemoglobin and haemoglobin concentration, lower hypoxic ventilatory response, and increased exhalation of nitric oxide.

Perception of smell and taste

The human faculty of smell and taste has undoubtedly played a major role in shaping human evolution. The ability to smell has been often linked to mate recognition and predation. Taste may play a bigger role in identifying palatable nutrition sources. There is a growing body of data on the role of natural selection on the genes responsible for taste and smell.

A recent genomic study identified increased ratio of non-synonymous to synonymous substitutions at olfactory receptor gene loci, and low level of polymorphisms and an excess of rare variants in regions encompassing it (Gilad et al. 2003). Another study identified a strong signal of recent natural selection within a cluster of olfactory receptor genes (Williamson et al. 2007).

Geneticists have been using phenylthiocarbamide (PTC) based on the variable ability of individuals to taste it for the better part of the century, but the genetic basis of this variability to taste PTC was recently elucidated (Drayna et al. 2003). Two high frequency haplotypes differentiating on the basis of three non-synonymous variations were identified. Additionally a strong signal of natural selection was identified at twenty-one genes involved in taste reception by having significantly higher non-synonymous to synonymous substitutions (Kim et al. 2005).

Climatic Stress

The sodium-retention hypothesis identifies heat stress as a selective pressure in humans (Denton 1982). It states that selection for high sodium retention in hot and humid environments is strong as salts important for maintaining temperature homeostasis are lost more quickly in this type of environment (Gleibermann 1973). This hypothesis predicted a strong correlation between genetic variation allowing for adaptations to heat stress, and temperature and humidity. In agreement with this prediction, strong correlations were noted between the variation pattern of the CYP3A5 gene (involved in the activation of cortisol in the kidney to allow for sodium retention), AGT gene (involved in the renin-angiotensin pathway increasing blood pressure) and latitude, which was hypothesized as an indirect indicator of temperature (Thompson et al. 2004,Young et al. 2005). In addition, genome-wide studies by Voight et al. (2006) and Tang et al. (2007) identified strong signals of natural selection at the CYP3A5 locus which also showed a strong correlation with climatic variables.

Additionally, Mishmar et al. (2003) and Ruiz-Pesini et al. (2004), in agreement with selection for cold tolerance, duly noted an increase in the frequencies of non-synonymous mutation in the mitochondrial genes with increasing distance from Africa. Also recently strong correlation with climatic variables, and signals of natural selection were found in genes previously associated with thermogenesis pathway (Hancock et al. 2008).


As demonstrated from the first example, human populations inhabiting hypoxic environments have unique biological adaptations like elevated resting ventilation and haemoglobin concentration that appear to offset the stress of low-oxygen at those altitudes. This represents the effect of natural selection in favouring more effective adaptive response. Although the adaptive phenotype to hypoxic condition was identified, it is proving difficult to elucidate the genetic basis of this adaptation due to the complexity of the system regulating the phenotype. In identifying adaptive responses in complex systems lies the advantage of the traditional phenotypic approaches. Since acclimatization could not account for the phenotype, natural selection acting on the genetic level of the population was inferred. This shows that the conclusion was based on the exclusion of alternative possibilities as opposed to evidence of association.

The other two examples presented in this essay used the genomic approach, looking at sequence variation data to find evidence of natural selection in humans. Undoubtedly the advantage of this approach is its capability of detecting adaptations at the genetic sequence level, without having to collect a lot of phenotypic data over time. More so, with the recent drive to sequence complete genomes (1000 genomes project) and the availability of complete genomes of various organisms from several different populations have allowed such studies to be undertaken much more readily. The continued collection of data from such projects will be valuable for performing more comprehensive genome scans to substantiate ongoing natural selection. Concurrently, more intensive studies to test specific hypotheses by using genome-wide information will continue to identify and enhance our understanding of selective pressures in humans.