Recent natural selection in human populations

A bunch of high-powered genomics people have just come out with a paper in Genome Research surveying the human genome for recent signs of natural selection (which you can assess by looking at the patterns of DNA variation around various genes).  You can get the original paper (reference and abstract below) here as a pdf file, or read a story about it in the April 2 New York Times.  The survey was based on DNA data from 938 people from 53 populations, making this the most extensive survey of selection in human groups to date.

I’m leaving shortly (see post below by Matthew Cobb, my replacement), but just a few conclusions from this survey:

1.  Geographically adjacent populations tend to have similar histories of selection: Europe, Central Asia, and the Middle East tend to share their genetic complements compared to populations elsewhere in the world.  This presumably reflects the history of migration and intermarriage between the former three areas.

2.  Genes that seem indubitably under selection include not only loci involved in pigmentation, but some loci of unknown function (e.g., the gene called C21orf34, which resides on our 21st chromosome).  As I note in WEIT, human skin pigmentation is one of the few “racial” traits whose difference between human populations is pretty clearly connected with natural selection.  We previously had a good “intuitive” explanation for this difference (in sunnier climes, darker pigmentation is selected to prevent melanomas, while in northern, less sunny areas, the skin becomes less pigmented to allow production of vitamin D), but now we have more direct evidence from looking at the genes that themselves produce pigment.  However, most of the differences in appearance between groups don’t have an evolutionary explanation so far: they may have changed via sexual selection. Note that in the chart below, the SLC genes are pigmentation alleles, with a pretty strong signal of selection in Europe, the Middle East, and Asia.

3.  A gene that is a risk factor for diabetes and celiac disase also shows signs of being positively selected in some populations.  Why?  Perhaps because, as the authors theorize, the “risk allele” is involved in some other, positively selected trait, or perhaps it is genetically linked to such traits by being nearby on the chromosome.

4.  As mentioned in an earlier post, a pygmy and nearby non-pygmy African population differ in their insulin growth factor genes, which may have been selected to have alleles for smaller height in pygmies. (See earlier post for possible reasons for this).  These data come from only two populations, however, and it would be interesting to see if pygmy populations in Southeast Asia and South America have similar forms of the IGF gene.

Note that much of this selection has to be fairly recent, since populations outside of Africa have differentiated only within the last 60,000-100,000 years.


This chart shows the sites along the genome (listed at the left) at which natural selection has occurred in the genome of eight regional groups (shown at top). These are 1) Biaka pygmies, 2) Bantu-speaking Africans, 3) Western Europeans, 4) Middle Easterners, 5) South Asians (people of Pakistan and India),

6) East Asians, 7) Oceanians and 8. Native Americans. The colored bars show the degree of selection at each site, with yellow denoting a signal of clear but moderate statistical significance and red denoting high statistical significance. (Photo and caption courtesy of The New York Times)

Signals of recent positive selection in a worldwide sample of human populations

Abstract: Genome-wide scans for recent positive selection in humans have yielded insight into the mechanisms underlying the extensive phenotypic diversity in our species, but have focused on a limited number of populations. Here, we present an analysis of recent selection in a global sample of 53 populations, using genotype data from the Human Genome Diversity-CEPH Panel. We refine the geographic distributions of known selective sweeps, and find extensive overlap between these distributions for populations in the same continental region but limited overlap between populations outside these groupings. We present several examples of previously unrecognized candidate targets of selection, including signals at a number of genes in the NRG–ERBB4 developmental pathway in non-African populations. Analysis of recently identified genes involved in complex diseases suggests that there has been selection on loci involved in susceptibility to type II diabetes. Finally, we search for local adaptation between geographically close populations, and highlight several examples.

Joseph K. Pickrell et al., Genome Res. May 2009 ; published ahead of print March 23, 2009,


  1. Ashwin
    Posted April 6, 2009 at 2:32 pm | Permalink

    Very interesting post, thank you, and also thanks for the link to the paper. There are other good examples of genes under recent selection, for example, the Vitamin D-receptor gene has been positively selected for in African populations. This may explain why Africans have a lower rate of osteoporosis than Caucasians, and other races.

  2. newenglandbob
    Posted April 6, 2009 at 7:29 pm | Permalink

    We are here reading this stuff. Some of us are awed at this kind of information but are unqualified to comment on it.

%d bloggers like this: