Thursday, January 15, 2009

Polygyny and X-chromosome diversity

Last year, a team of University of Arizona researchers found evidence of widespread polygyny in five different human populations: Biaka (Central African Republic), Mandenka (Senegal), San (Namibia), Basques (France), Han (China), and Melanesians (Papua New Guinea). In short, the maternally inherited X chromosome was genetically more diverse than the chromosomes inherited by both sexes (autosomes) (Hammer et al., 2008). So more women than men seem to have contributed to the gene pool. Surprisingly, there was little difference in this respect between the Mandenka (known to be highly polygynous) and the Basques and the Han (among whom the incidence of polygyny is much lower).

I was frankly skeptical. For one thing, maternally inherited genetic diversity reflects not only the number of women who contribute to the gene pool but also their own genetic diversity. If these women are drawn from a larger geographic area than the men are, the female gene pool will be more genetically diverse than the male gene pool. This is often the case. In a patriarchal society, land ownership is vested in the man’s lineage, so women are usually the ones who move to their mate’s community when they get married. We see this ‘patrilocality’ even in societies where land ownership is matrilineal. Among the Iroquois, wives were often abducted from other tribes through warfare.

In any case, the above findings have now been challenged. Another study has found much less maternally inherited genetic diversity in East Asians and Europeans than in West Africans (Keinan et al., 2008).

So what gives? The methodology is similar in both studies. John Hawks points out that the second study scales X-chromosome diversity to the human-macaque divergence whereas the first study uses the human-orangutan divergence. While this might explain differences in calculation of mutation rate and hence X-chromosome diversity, I don’t see how it could explain why one study found geographic differences (i.e., African versus non-African) and the other did not.

I suspect that the key difference is that the first study just did not have enough resolution to pick out these geographic differences, i.e., its dataset was too small. The second study used 130,000 loci (SNPs) whereas the first one used 40.


Please check out my latest article: “Sexual selection and human geographic variation” in The Journal of Social, Evolutionary & Cultural Psychology.

References

Frost, P. (2008). Sexual selection and human geographic variation, Special Issue: Proceedings of the 2nd Annual Meeting of the NorthEastern Evolutionary Psychology Society. Journal of Social, Evolutionary, and Cultural Psychology, 2(4), pp. 169-191.
http://www.jsecjournal.com/articles/volume2/issue4/NEEPSfrost.pdf

Hammer, M.F., Mendez, F.L., Cox, M.P., Woerner, A.E., & Wall, J.D. (2008). Sex-biased evolutionary forces shape genomic patterns of human diversity. PLoS Genet, 4(9), e1000202. doi:10.1371/journal.pgen.1000202

Keinan, A., Mullikin, J.C., Patterson, N., & Reich, D. (2008). Accelerated genetic drift on chromosome X during the human dispersal out of Africa. Nature Genetics, early view December 2008; doi:10.1038/ng.303

7 comments:

Tod said...

The Evolution of Human Sociality
By Stephen K. Sanderson states
"Of the 752 societies in the Ethnographic Atlas for which data are available , 188 practice cross - cousin marriage. Most of these (154 or 82%) practice bilateral cross-cousin marriage, 30 (16%) practice matrilateral cross cousin marriage, and only 4 (2%) practice patrilateral cross-cousin marriage"


"In many societies the preferred marriage for a boy is with his mother's brother's daughter — that ... cross-cousin marriage" (Homans and Schneider, 1955 ). ..."

JSTOR: Cousin Marriage in China
"Hsu (1945: 84), for example, finds that, with few exceptions, in all parts of China which he knows matrilateral cross-cousin marriage is "the preferred..."

Family and Relationships in Senegal

" UNCLE (MATERNAL)

The Maternal Uncle, or nijaay in Wolof, often helps to settle family disputes. He may also be expected to help his sisters in times of need when the parents are no longer able. Among many groups in Senegal, a prefered marriage for a man is to his nijaay's daughter."

Would it affect maternally inherited genetic diversity in some way that accounts for the Hammer et al. result for West Africa if over generations a man's wife (or even wives) were related to him though his mother?

n/a said...

Note that neither study actually supports Peter.

The newer study offers no genetic evidence of polygyny in the sub-Saharan sample (the reported ratio of X to autosomal diversity being about what you'd expect with a 1:1 sex ratio). Additionally, the authors attribute the apparent depressed X diversity in non-Africans to some demographic or other process affecting the "out-of-Africa" population rather than later (post-split) European and Asian populations.

Tod said...

SNPwatch:Shuffling SNPs Have Opposite Effect in Men and Women
A report released today in the online edition of Science magazine demonstrates that two SNPs influence the amount of chromosome shuffling that takes place during the production of sperm and eggs. Surprisingly, the versions of the SNPs that increase the rate of shuffling in men actually decrease it in women, and vice versa....Augustine Kong and co-workers conducted a genome-wide search in more than 6000 people of European ancestry to find variants that correlated with recombination rate. This rate was determined by looking at the chromosomes of both subjects and their children.

The researchers found that men who had at least one G at rs3796619 and at least one A at rs1670533 had higher rates of recombination. In women, the opposite was true: at least one A at rs3796619 and at least one G at rs1670533 was associated with high rates of recombination. Customers can look up their data on both of these SNPs in 23andMe’s Genome Explorer.

The report’s authors suggest that the opposite effect of these SNPs in men compared to women may help to keep the pace of change in the genome relatively constant. By balancing high rates of recombination in one sex with low rates in the opposite sex, nature may have found a way to maintain stability in the human genome and ensure the evolutionary success of the species

A report released today in the online edition of Science magazine demonstrates that two SNPs influence the amount of chromosome shuffling that takes place during the production of sperm and eggs. Surprisingly, the versions of the SNPs that increase the rate of shuffling in men actually decrease it in women, and vice versa....Augustine Kong and co-workers conducted a genome-wide search in more than 6000 people of European ancestry to find variants that correlated with recombination rate. This rate was determined by looking at the chromosomes of both subjects and their children.

The researchers found that men who had at least one G at rs3796619 and at least one A at rs1670533 had higher rates of recombination. In women, the opposite was true: at least one A at rs3796619 and at least one G at rs1670533 was associated with high rates of recombination. Customers can look up their data on both of these SNPs in 23andMe’s Genome Explorer.

The report’s authors suggest that the opposite effect of these SNPs in men compared to women may help to keep the pace of change in the genome relatively constant. By balancing high rates of recombination in one sex with low rates in the opposite sex, nature may have found a way to maintain stability in the human genome and ensure the evolutionary success of the species

Tod said...

Polygynists and their wives in sub-Saharan Africa: an analysis of five Demographic and Health Surveys.
Author: Timaeus IM; Reynar A
Source: POPULATION STUDIES. 1998 Jul;52(2):145-62.
Abstract: This study identifies the characteristics of first- and higher-order wives in polygynous unions and describes the determinants, prevalence, and intensity of polygyny. Data came from 1993-96 Demographic and Health Surveys (DHS) from Ghana, Kenya, Senegal, Uganda, and Zambia. Polygyny is most common in western Africa; however, it is in decline, especially in Kenya. Prevalence is low in urban areas. Polygyny is very common in Senegal, high in Uganda and Ghana, and fairly common in Kenya and Zambia. The percent distribution of currently-married women is given by age, marital duration, marriage age, number of unions, infertility status, level of education, and husband's education, occupation, wealth, residence, and religion for each country. Prevalence is affected more by socioeconomic and other characteristics than by intensity. Grouping first and higher order wives separately is key. In all countries, divorcees and widows who married single men were unlikely to engage in polygyny. Only in Senegal did polygyny accelerate at higher marital durations. Polygyny involves childless women. Lower prevalence in Kenya is due to better educated women and low prevalence in Zambia is due to a small agricultural sector; low prevalence in both countries is due to a high degree of socioeconomic change. In Kenya, the decline is associated with changes in the composition of the population. Individual level characteristics explained only some of the ethnic variations in polygyny. Socioeconomic factors explained the prevalence and intensity, but factors and their impact varied between countries. This study clarifies social conditions that support or depress polygyny and reveals that declines in prevalence may occur with a rise in those with over 3 wives.
Language: English

----------
Subsistence strategies and economy in the
Magdalenian of the Paris Basin, France

Reindeer (Rangifer tarandus) can be hunted in a variety of ways (Burch 1972; Spiess 1979). One is
the opportunistic encounter of single animals or small groups on their summer or winter ranges, such as any other territorial game.[...} reindeer as a resource for hunters depends on how they position themselves in the landscape relative to the migration pattern of the reindeer
herds. Further, the economic organisation of subsistence for the acquisition, processing and
consumption of meat will be largely determined by that landscape position. For example, a group of hunter/gatherers such as the Netsilik Eskimo (Balikci 1970) on the northern tundra has access to herds during the summer. At that time, the herds are broken down into small bull or cow/calf bands which are widely dispersed so that hunter/prey encounters tend to be opportunistic and unpredictable over time and space. An appropriate strategy for increasing the probability of encounter is the dispersal of hunters in small bands and individual hunts over a large
territory, precluding larger scale cooperative hunting. The likelihood that all hunters from a
given camp would be simultaneously successful is small, so that successive reciprocity in game
sharing provides a buffer against variable success rates. Further, the quantity of game brought into a
camp at any one time should be relatively small, and meat can be consumed immediately.


At some times of year lone hunters looking for prey would cover long distances. Once a kill had been made the hunter would be going straight home with the prized food resources, the temptation would be to carry the absolute maximum. Female reindeer start at 130lb - males 220lb, allowing for removal of the least useful parts of the carcass the successful hunter in a time of shortage would often be carrying > 100lb across the steppe-tunda. Ihis would place tremendous stress on the lower body and must have resulted in many broken ankles ect. far from home with fatal consequences. Women may have been at a great disadvantage inasmuch as they could carry far less and they are far more vulnerable to such injuries

In view of the anatomical, physiological and ergonomic differences between the sexes, gender is the single most important factor16. Jones showed that 37% of male US Army recruits experienced a lower limb injury during a 12-week basic training course12 whereas female injury rates may be as high as 60%1.

There are many reasons why female recruits are at greater risk of injury; the gynaecoid pelvis may produce a relative and transient biomechanical disadvantage when marching, running or carrying weight at the extremes of endurance17,18.

Peter Frost said...

Tod,

Inbreeding on the maternal side would tend to depress X-chromosome diversity and lead us to underestimate the incidence of polygyny.

Your second post underscores the problems in calibrating genetic diversity. How much X-chromosome diversity corresponds to a 1:1 breedign ratio? I don't see any easy of solving this problem.

The data seem to be better at showing relative differences in the polygyny rate among human populations.

n/a,

Since the two studies contradict each other, it would be difficult for both of them to "support Peter", wouldn't it? ;-)

But, seriously, both studies are weak when it comes to calculating the mutation rate on the X chromosome, which in turn is used to calculate the absolute ratio of breeding females to breeding males. By changing the assumptions that underlie this calculation, one can easily change the breeding ratio.

This kind of study is much better at showing the relative differences among human populations in terms of polygyny rate.

Jenea said...

It's super the things you're saying, I see just how it also comes in. Oh, nearly forgot, I have a resource just like this called tablet android barato

vasya said...

While this might explain differences in calculation of mutation rate and hence X-chromosome diversity, I don’t see how it could explain compras en china why one study found geographic differences (i.e., African versus non-African) and the other did not.