April 30, 2008

mtDNA of admixed Colombians

See also: Genomic admixture of Latin American Mestizos

Am J Hum Biol. 2008 Apr 28 [Epub ahead of print]

The mtDNA ancestry of admixed Colombian populations.

Salas A, Acosta A, Alvarez-Iglesias V, Cerezo M, Phillips C, Lareu MV, Carracedo A.

A total of 185 individuals from Colombia were sequenced for the first hypervariable region (HVS-I) of the mitochondrial DNA (mtDNA) genome, and a subset of these individuals were additionally genotyped for the second hypervariable segment (HVS-II). These individuals were collected according to their "self-reported ethnicity" in Colombia, comprising "Mestizos," "Mulatos," and "Afro-Colombians." We used databases containing more than 4,300 Native American lineages, 6,800 Africans, and 15,600 Europeans for population comparisons and phylogeographic inferences. We observe that Mulatos and Afro-Colombians have a dominant African mtDNA component, whereas Mestizos carry predominantly Native American haplotypes. All the populations analyzed have high diversity indices and there are no signatures of dramatic genetic drift episodes. Central and South America are the main candidate source populations of the Colombian Native American lineages, whereas west-central, southwest, and southeast Africa are the main original mtDNA sources for the African Colombian mtDNAs. We found that our results differ from those obtained in other studies for the same "population groups" in terms of haplogroup frequencies. This observation leads us to conclude that (i) self-reported ancestry is not a reliable proxy to indicate an individual's "ethinicity" in Colombia, (ii) our results do not support the use of outmoded race descriptions (Mestizos, Mulatos, etc.) mainly because these labels do not correspond to any genetically homogeneous population group, and (iii) studies relying on these terms to describe the population group of the individual, which then treat them as genetically homogeneous, carry a high risk of type I error (false positives) in medical studies in this country and of misinterpretation of the frequency of observed variation in forensic casework.



Maju said...

Well, I find as totally unsurprising that there is almost not European mtDNA. It's widely known that Spanish colonization in America managed to bring very very few European women. Almost all colonists were men. So whatever the actual admixture, you will find nearly zero European mtDNA because the European input was exclusively male mediated.

Other studies have found large ammounts of European Y-DNA and varied apportions of autosomal and X-DNA.

Additionally (even if maybe less important), it's important to realize that Latin American racial categories are sometimes just phenotypical and class perceptions, partly indepependent of real ancestry; plus often whitewashed because of historical racism towards non-whites (i.e. mestizos and mulattos would be half-whites, at least in theory, and therefore higher in the social scale). In some places people of nearly 100% native ancestry but of Spanish culture would be considered mestizos, while in others tribals with high degree of European admixture are considered pure "Indians". It's not like geneticists were there all the time since 1492 to make sure nobody could escape their "true" racial classification. It was always a social and largely subjective classification. In places like Brazil full syblings can well belong to totally different official "races", depending on their phenotypes only and the judgement of the census bureaucrat.

Maju said...

Is the term "Afro-Colombians" equivalent with the term "Zambo"?

Probably a US-influenced neologism to mean (roughly) black people. According to Wikipedia, the Colombian census does not include racial definitions of any sort but estimates are that 14% are mulatto, 4% black and 3% zambo (and you are right about the meaning of this term).