Razib Khan One-stop-shopping for all of my content

February 25, 2011

Brazilians, more European than not?

Filed under: Brazil,Genetics,Genomics — Razib Khan @ 1:30 am

Credit: Dragon Horse

The Pith: Brazil is often portrayed as the second largest black nation in the world, after Nigeria. But it turns out that the majority of the ancestors for non-white Brazilians are European.

One of the more popular sources of search engine traffic to this website has to do with the population genomics of Latin America. For example, my post showing that Argentina is not quite as European a country as it likes to consider itself is regularly cited in online arguments (people of various “persuasions” are invested in the racial status of the Argentine people). But last week in PLoS ONE a paper looking at the patterns of ancestry in the Brazilian population came to a somewhat inverse conclusion as to the self-conception or perception of the preponderant racial identity of that nation. Let me quote from the conclusion of the paper:

Among the actions of the State in the sphere of race relations are initiatives aimed at strengthening racial identity, especially “Black identity” encompassing the sum of those self-categorized as Brown or Black in the censuses and government surveys. The argument that ...

January 10, 2011

Ancestry in the Americas

ResearchBlogging.orgThe populations of the African Diaspora have a particular interest in the new genomics, and its relationship to ancestry. Unlike other post-Columbian Diasporas they have sketchy, at best, knowledge of the regions from which their ancestors arrived. This probably explains the popularity of Roots and Henry Louis Gates Jr.’s various genealogical projects which have utilized cutting edge genomics. It may seem silly to hang one’s hat on one maternal lineage, but perhaps it seems silly if you are relatively assured of the broad outlines of your own genealogy. The fact that I am U2b is not very interesting to me, but I also happen to know that my maternal grandmother’s mother’s family were long resident in their region of Bengal (and, that her father was a migrant from northwest India). It would be a different matter if my ancestors had been enslaved and dispossessed of their heritage.

A new paper in PLoS ONE surveys the paternal (NYR), maternal (mtDNA), and autsomal (using 175 ancestrally informative markers), heritage of a range of African origin populations from across the Americans. Dissecting the Within-Africa Ancestry of Populations of African Descent in the Americas:

Our analysis revealed that both continental admixture and within-Africa admixture may be critical to achieving an adequate understanding of the ancestry of African-descended Americans. Whilecontinental ancestry reflects gender-specific admixture processes influenced by different socio-historical practices in the Americas, the within-Africa maternal ancestry reflects the diverse colonial histories of the slave trade. We have confirmed that there is a genetic thread connecting Africa and the Americas, where each colonial system supplied their colonies in the Americas with slaves from African colonies they controlled or that were available for them at the time. This historical connection is reflected in different relative contributions from populations of W/WC/SW/SE Africa to geographically distinct Africa-derived populations of the Americas, adding to the complexity of genomic ancestry in groups ostensibly united by the same demographic label.

There isn’t anything too surprising here. Blacks from Brazil have much more ancestry from the former Portuguese colonies of Angola and Mozambique. As we should expect. Because the New World African Diaspora dates to only the past 350-150 years even mtDNA should be a good snapshot of the genetic variation. And, because of the ability to construct clean genealogies due to lack of recombination, mtDNA can be even more informative than total genome surveys in terms of elucidating fine-grained geographical patterns. The map below illustrates the mtDNA results well:

But I was more interested in some patterns comparing the various African and European identified populations. In the figure which I stitched together below you have maternal ancestry on the left, paternal in the middle, and total genome ancestry to the right. The first panel has self-identified blacks from various nations, and the second has self-identified whites from Brazil and Philadelphia, USA. These results need to be taken with a grain of salt, because the samples from each nation need not necessarily be representative of the ethnicity for that whole nation (e.g., blacks in coastal Georgia have far less white ancestry than those in urban northern cities). But they give a rough picture of the differences and similarities.

Here the numbers with margins of error as well:

Self-identified black Self-identified white
African error European error Native error African error European error Native error
Brazil mtDNA 85 3 1 1 14 3 31 6 38 5 31 5

NRY 51 4 47 4 2 2 0 1 97 2 2 2

AIMs 58 30 12 20 64 17
USA mtDNA 93 1 5 1 2 1 6 5 92 7 2 1

NRY 77 2 25 2 0 0 2 1 97 2 1 1

AIMs 83 1.6 15 2 2 1 2 2 95 3 3 3

Similarities: a strong bias toward more male European ancestors, and more female African and Native ancestresses. This is what you’d expect. But look at the Brazil black sample. The disjunction between nearly total lack of female European ancestry and a substantial proportion of male European ancestry is rather striking. I think this is partly a function of the strong male bias of the Portuguese settler community. A process of “whitening” of the original mulatto and mestizo population of the colony probably occurred with each successive generation of male immigrants, who married into the hybrid communities and shifted the total genome content, though not the maternal background. The biggest difference of course is the much more clear and distinct difference between the races in the USA as opposed to Brazil. A minority of Brazilians identify as black, while a plural majority identify as white (the remainder are “brown” or mixed-race). And yet Brazilian blacks are ancestrally much whiter than American blacks, while Brazil whites are ancestrally much less white than American whites. Additionally, while Brazil perceives itself as a mulatto nation, there is a substantial Amerindian substrate which spans the black, white, and brown, populations.

Finally, going back to the main theme of the paper, African population structure, one of the main rationales is for purposes of medical research where stratification of ancestry within an aggregate pool may lead to spurious associations. In other words, mixing blacks with ancestry from Angola with those from ancestry from Nigeria may result in false positives in GWAS due to differences in predispositions of traits between these populations, which also track their genetic differences. The authors bring up the fact that African populations are the most genetically diverse in the world. But let me resurrect a comment from geneticist Nick Patterson:

I want to comment on Africa’s genetic diversity. There is a great deal of genetic diversity within many African populations. For instance more diversity within Yoruba than within Europe. But many African populations are quite similar. For instance divergence between Kikuyu (Kenya) and Yoruba (Nigeria) is just a little more than (Spain, Holland). This is of course caused by the Bantu expansion as Razib explained.

In other words, the great African genetic diversity is more a function of the intra-population variance, than inter-population variance. This sort of work is important for reasons of intellectual interest, but may not be quite so important for GWAS.

Citation: Stefflova K, Dulik MC, Barnholtz-Sloan JS, Pai AA, & Walker AH (2011). Dissecting the Within-Africa Ancestry of Populations of African Descent in the Americas PLoS ONE : 10.1371/journal.pone.0014495

December 5, 2010

BRICs in charts

Filed under: Brazil,BRIC,BRICs,China,Data Analysis,Indian,Russia — Razib Khan @ 3:45 pm

The term “BRICs” gets thrown around a lot these days. At least it gets thrown around by people who perceive themselves to be savvy and worldly. In case you aren’t savvy and worldly, BRICs just means Brazil, Russia, India and China. The huge rising economies of the past generation, and next generation. Here’s a summary from Wikipedia:

The BRIC thesis recognizes that Brazil, Russia, India and China…have changed their political systems to embrace global capitalism. Goldman Sachs predicts that China and India, respectively, will become the dominant global suppliers of manufactured goods and services, while Brazil and Russia will become similarly dominant as suppliers of raw materials. It should be noted that of the four countries, Brazil remains the only nation that has the capacity to continue all elements, meaning manufacturing, services, and resource supplying simultaneously. Cooperation is thus hypothesized to be a logical next step among the BRICs because Brazil and Russia together form the logical commodity suppliers to India and China. Thus, the BRICs have the potential to form a powerful economic bloc to the exclusion of the modern-day states currently of “Group of Eight” status. Brazil is dominant in soy and iron ore while Russia has enormous supplies of oil and natural gas. Goldman Sachs’ thesis thus documents how commodities, work, technology, and companies have diffused outward from the United States across the world.

But there are big quantitative differences between these nations as well. Below the fold are some charts which I think illustrate those differences.

June 27, 2010

Amerindians of Brazil more numerous than you think

Filed under: Amerindian,Brazil,Culture,Genetics,Genomics,Mixed-Race — Razib Khan @ 8:09 am

adriana-limaA few months ago I was thinking a fair amount about the Neandertals. One issue which became more stark to me due to that particular finding, that a few percent of the human genome seems to have derived from Neandertal populations, is the reality that genetic distinctiveness can persist long after cultural coherency is no longer a reality. That made me reconsider one of the facts of contemporary scholarship, that the Amerindian populations of the two most populous nations of the New World, the United States of America and Brazil, have disappeared or been totally marginalized demographically.

I’ve observed before it looks like that about 15-20% of the ancestry of the Argentine population is Amerindian, despite the nation’s proud identity as a European settler offshoot (i.e., more like the United States or Australia, than Mexico, which has an explicit hybrid identity). But I realized that Brazil was perhaps the bigger catch.

Only 0.4% of Brazilians identify as Amerindian. That’s about 700,000 people. But we know that a substantial number of white, brown and black Brazilians have Amerindian ancestry. Assuming for argument’s sake that the 700,000 Amerindians have undiluted indigenous ancestry, how much of the distinctive Amerindian genome in modern Brazil is to be found in this segment of the population?

There was a paper which came out in an obscure Brazilian journal last year which can help answer this question, DNA tests probe the genomic ancestry of Brazilians. For the purposes of the paper they needed to find a small number of ancestrally informative markers which would allow them to partition the ancestries of the individuals in their data set into European, African, and Amerindian, segments. Luckily these are three very distinctive populations. They cross-checked the utility of their markers against the HGDP data set. In other words the precision and accuracy of the 40 markers they selected should be assessed by how well they can distinguish these three “pure” populations. Their Brazilian subjects consisted of self-identified whites from various regions, as well as black men from Sao Paulo. Brazil’s racial taxonomy has a brown (pardo) category which is large, but judging from the very high proportion of African ancestry among the blacks in their sample I don’t think they included self-identified mixed-race people in that group (some scholars lump mixed and black Brazilians together into the black category).

First, let’s see how well the markers allow for us to distinguish populations’ whose ancestry we’re pretty sure about.


No that bad. Observe the trailing off of Europeans and Amerindians along their axis of variation. From previous papers using SNP chips with hundreds of thousands of markers It seems that the HGDP Amerindian sample has non-trivial European ancestry, so it isn’t just a limitation of their marker set. This shouldn’t be that surprising in light of the history of Latin America, and the centuries of racial fluidity which occurred. Additionally, I assume that it would be more difficult to find markers which vary between Europeans and Amerindians than between these two groups and Africans (remember that Africans have more genetic diversity, and non-African populations can be thought of as simply one branch out of Africa).

The following chart shows the outcome from the markers across various regions of Brazil. I assume that the core American readership will be pretty ignorant of the regions, so this map will help. Southern Brazil is white. The rest of the country far less so. Gisele Bündchen and Alessandra Ambrosio are from Rio Grande do Sul in the far south. Adrianna Lima, pictured above, is from Bahia in the northeast. All of the panels are for whites (through self-identification) except for the last.


Despite the limitations of 40 markers I think the results here are probably pretty good. The South region has the whitest whites. This was an area with massive immigration, and fewer non-whites to start with. It is in the northeastern region that you see more Amerindian than black ancestry among the whites. From what I have seen in other papers the brown category in Brazil is more skewed toward European ancestry than panel F, so that’s why I assume that these are men who self-identify as black.

But for the purposes of my original question we need to assess ancestral contributions within these groups from Amerindians. The following table does just that.


The authors note a curious fact: there’s no statistically significant difference between the regions in terms of Amerindian ancestry for self-identified white Brazilians. This resembles the pattern we saw with Neandertal admixture, and I assume the explanation is the same: the integration of Amerindian ancestry occurred early in the history of the white Brazilian population and has now distributed throughout it via generations of intermarriage except for those with undiluted white immigrant heritage (e.g., Gisele, who comes from a German town). The figure for blacks from Sao Paulo is about the same as whites as well.

Let’s assume 10% Amerindian ancestry and 200 million Brazilians to make the math easier. Since it is about 10% in whites and blacks, I suspect it will be 10% in those who identify as mixed race as well. Where does that leave us? That would mean 20 million Brazilians of Amerindian ancestry! As you can see around 95% of the Amerindian genome in Brazil is found among those who do not self-identify as Amerindian.

For Americans (citizens of the USA) it is then interesting to wonder how much of the Native genome is found in whites, particularly old stock colonial descended whites, and how much in self-identified Native Americans. Looking at the genetic studies I believe that the Amerindian proportion is much lower among white Americans than 10%. Additionally, while 700,000 Brazilians identify as Amerindian, 2.4 million Americans do (though I believe that a much larger proportion of American Native Americans are of mixed ancestry than Brazilian Aboriginals). I think that the odds are that more ancestry which is pre-Columbian in the USA does reside within the white population than within the self-identified Native American tribes, but it may be a close thing. And of course there are Latino populations which need to be added into the equation, whose Amerindian ancestry is significant (in the case of Mexican Americans possibly preponderant), though it is of Mesoamerican origin.

But doing those sums is for another post.

Note: A preemptive apology to those who feel I’ve used the wrong term for any particular ethnic group. I have preferred the more Politically Correct (though less popular) term Native American in the context of the USA, but feel that Amerindian is technical or scholarly enough to evade charges of insensitivity.

Image credit: UnaFraseCelebre.com

Powered by WordPress