The Evolution and Medicine Review

…bridging the gap.

Out of Jerusalem

The human haplotype map (HapMap) shows that human populations differ genetically and have been subject to strong, recent positive selection: selection ‘for’ particular genetic variants. Surprisingly, patterns of inferred selection vary markedly between the three human groups analyzed thus far, one Caucasian, one African, and one Asian (Voight et al. 2006). Ethnicity, and natural selection, may thus play stronger roles in diverse human traits, including genetic susceptibility to disease, than previously believed.

One ethnic group, famous for reasons including its high incidence of otherwise-rare diseases, is the Ashkenazi. The concentration of recessive diseases in this group has usually been attributed to founder events, but this interpretation has been challenged by Cochran, Hardy and Harpending (2006), who describe evidence that some ‘disease’ alleles in the Ashkenazi have actually been selected for via benefits to heterozygotes. The benefits? Controversially, they suggest higher intelligence, strongly selected for in the context of a cultural and genetic milieu that strongly translated such intelligence into reproductive success, and led over many generations to the higher IQs considered uniquely-characteristic of this group.

So, we have a human population, largely-isolated genetically, undergoing founder events, and subject to strong selection for ‘intelligence’ over many generations – sound familiar? This is the precise scenario postulated for the origin of modern humans, on a much larger scale – just replace Abraham with Adam and Eve, and Eden may indeed coincide with Ur.

How can we critically evaluate these ideas? First, we can use the published HapMap data to assess whether Ashkenazi disease genes have indeed been subject to positive selection. Go to the Haplotter web site and in Query by Gene (for phase II) enter each gene “APC”, “BRCA1”, “GBA” and “HEXA” and click “submit”, – for each, scroll down to the yellow highlights – see the P-values under 0.05? Positive selection has also been inferred for BLM, BRCA1, FANCC, GBA and HEXA in other data sets (Pavlicek et al. 2004; Bustamante et al. 2005; Voight et al. 2006; Tang et al. 2007; Wang and Moyzis 2007). None of these analyses involve the Ashkenazi, or their specific disease alleles, but they all provide inferential evidence that these particular genes may have undergone adaptive evolutionary change in humans.

Second, we need a HapMap specific to the Ashkenazi, to rigorously evaluate the roles of selection and drift on disease-gene evolution in this population. As of February, we have it (Olshen et al. 2008). The authors report on genomic differences between the European HapMap population and the Ashkenazi that “likely… reflect the impact of both selection as well as genetic drift”. Further analyses of such data should provide crucial insights into the nature and sources of Ashkenazi (and other) genetic diseases, and the evolution of modern humans.


* Bustamante CD, et al. (2005) Natural selection on protein-coding genes in the human genome. Nature 437:1153-7.

* Cochran G, Hardy J, Harpending H. (2006) Natural history of Ashkenazi intelligence. Journal of Biosocial Science 38:659-93.

* Olshen AB, et al. (2008) Analysis of genetic variation in Ashkenazi Jews by high density SNP genotyping. BMC Genetics 9:14.

* Pavlicek A, Noskov VN, Kouprina N, Barrett JC, Jurka J, Larionov V. (2004) Evolution of the tumor suppressor BRCA1 locus in primates: implications for cancer predisposition. Human Molecular Genetics 2004 13:2737-51.

* Tang K, Thornton KR, Stoneking M. (2007) A New Approach for Using Genome Scans to Detect Recent Positive Selection in the Human Genome. PLoS Biology 5:e171.

* Voight BF, Kudaravalli S, Wen X, Pritchard JK. (2006) A map of recent positive selection in the human genome. PLoS Biology 4:e72.

* Wang ET, Moyzis RK. (2007) Genetic evidence for ongoing balanced selection at human DNA repair genes ERCC8, FANCC, and RAD51C. Mutation Research 616:165-74.


The appendix, appendicitis, and appendectomy


An Anniversary Launch for The Evolution and Medicine Review


  1. Since Cochran et al. advanced their bold and controversial hypothesis, the evidence for selection on Ashkenazi Jewish genes appears to have strengthened. As they put it, the Ashkenazic genetic diseases influence metabolic pathways that fall on only a few pages of a biochemistry textbook, an extremely unlikely convergence from drift alone. While there is as yet only weak experimental evidence for a connection to intelligence (e.g., BRCA genes influence neural growth), there is clear evidence of Ashkenazic superiority in verbal and analytical intelligence along with inferiority in spatial intelligence. And it may have resulted from selection, which dragged some bad disease in its wake through balanced polymorphisms.

    Cochran et al. put this selection down to the pressure on Jews to be moneylenders and traders, but historical evidence shows that Jews pursued many other paths, including all crafts, bureaucratic offices, Jewish community offices (rabbis, teachers, kosher butchers), and proxy supervisory roles in farming and forestry operations. Any of these might require verbal and analytical intelligence.

    But it should not be missed that there was in effect a lek or tournament aspect to rabbinical study and education. Local and regional academies concentrated the brightest young men and pitted them against each other in constant intellectual competition. Promising students were boarded with and eventually married the daughters of other rabbis and prominent merchants, and the resulting offspring were raised in some of the most favorable environments that the Jewish (or for that matter European) middle class had to offer.

    So there are more reasons than one to consider the possibility of selection and (perhaps) balanced polymorphism. There is no contradiction of course between that claim and the arguments in favor of one or (probably) more bottlenecks in the history of the population. Mitochondrial DNA evidence suggests that all Ashkenazi Jewish women are descended from a very small number, possibly as few as six, mothers, although there is recent evidence that even this uniformity is stratified into subpopulations. Evidence for a Y chromosome founder effect is more equivocal.

    The trouble is that Ashkenazic Jews, who make up 80 percent of the world’s Jewish population despite being almost halved between 1939-45, do not look like they are simply “out of Jerusalem.” In fact, by all modern criteria, they are, like almost all Jews in the world, a mixture of Levantine genes with the genes of whatever local population they lived among. Abraham, if he existed, was genetically and phenotypically much more similar to the traditional Jews of Iraq or Yemen than to the Ashkenazim.

    So there have been imports to all Jewish populations, either through conversion, intermarriage, or both. How do we reconcile the bottleneck evidence with the admixture evidence in the Ashkenazi case?

    It seems to me there are two possibilities:
    1. The founding population of six mitochondrial mothers was not Jewish. This is consistent with the time of origin of the Ashkenazic Jews, over a millennium ago, when we know there were Jewish traders traveling in Europe. It is extremely unlikely that these traders were women, or that they brought wives with them from the Middle East. It is possible that they took wives from local populations, asked them to convert, and raised their children as Jews. Whether these were rabbinically kosher conversions is another matter.
    2. The mitochondrial mothers were Jewish women imported from the Middle East to start families with Jewish traders, and the subsequent admixture came from the importing of non-Jewish male genes. Given that by the time of the Ashkenazi founder population Jewish identity had long since been derived by Jewish law through the maternal line, this would be more palatable to traditional Jewish thought. It is not clear whether these male genes would have been imported voluntarily. In fact, the original switch in Jewish law from patrilineal to matrilineal descent may have been a response to large numbers of rapes of Jewish women during ancient Israel’s wars. Certainly there were many such rapes during attacks on Ashkenazic communities throughout medieval and modern European history.

    However, the consideration that some imports of male genes may have been voluntary points to another possible contributor to Ashkenazic Jewish intelligence, encapsulated in an old Jewish saying about intermarriage: “Our imports are better than our exports.” At present, all such hypotheses remain in the realm of interesting conjecture.

    [Editors note: Konner is the author of Unsettled: An Anthropology of the Jews]

  2. Henry Harpending

    Figure 5 of the Olshen et al. paper cited by Crespi plots linkage disequilibrium against distance between SNPS for the northern European and Ashkenazi samples. The lines are practically indistinguishable. This ought to kill forever the idea that there was any bottleneck at all in Ashkenazi history.

    The mtDNA evidence that Mel mentions is IMHO flaky: I can’t see it in the Behar et al. paper with my eyes. There does seem to be a bit of restricted diversity in mtDNAs in Ashkenazi, but the haplogroup Behar et al. cite as Middle Eastern is not found in the Middle East but is found in Spain and the neighborhood: it is probably Roman.

Leave a Reply

Powered by WordPress & Theme by Anders Norén