March 14, 2011

Y chromosomes of Altaian Kazakhs

This paper uses both the pedigree (genealogical or germline) and evolutionary mutation rates. Readers of the blog are aware that I've been a vehement critic of the latter on theoretical grounds since 2008, and I've started keeping track of cases where the germline rate has a better fit to the archaeological record than the evolutionary rate.

In this particular case, the difference between the two rates (about 3-fold) is especially interesting, because of the whole "Genghis Khan" theory according to which a large number of central Asian men belong to a haplotype cluster dated to around the time of the Mongol conqueror and maybe the descendants of Genghis and his close male relatives. This theory relies on the use of the germline rate: otherwise the genetic signature attributed to the Khan must be redated to a much earlier time.

The authors give a convincing argument in favor of the pedigree rate:
The difficulty in reliably determining the coalescent dates for the lineages found in Kazakh populations makes it nearly impossible to determine whether these lineages were present in ancestral nomadic steppe groups (Scythians, Xiongnu, Xianbei, Toba, and Jou-Jan) or were contributed by the descendents of Genghis Khan and the Mongol armies that, at one time, held control over the region. An important reason for caution here is the current debate about the most appropriate mutation rate for NRY coalescence estimates. The evidence provided by Zerjal et al. [14] supports the younger estimates, suggesting that the Kazakh haplotypes could be the direct result of the Mongol influence in the 13th century CE. The presence of the C3* haplotype cluster in the Kazakh also supports the genealogical assertions that (for at least some Kazakh men) there is a direct paternal connection to Genghis Khan.

If the evolutionary rate is the more accurate value for Y-STRs, then the Kazakh lineages coalesce to roughly 2,000 years ago. This date suggests a far older source for them, possibly with the westward movements of Altaic-speaking peoples around the second and first centuries BCE. In this case, we would expect to see multiple haplotype clusters exhibiting a similar pattern as the Genghis Khan cluster. However, we do not observe this pattern. As Zerjal et al. [14] pointed out, this haplotype cluster is unique. Therefore, given the evidence presented here and in Zerjal et al. [14], we believe the best interpretation of the data is that Kazakh Y-chromosome diversity was strongly influenced by the Mongols of the 13th century CE.

The younger ages of the Mongoloid lineages in this population makes good historical sense, as these are derived from tribal Turko-Mongolian tribes establishing (more recently) control over the pre-existing Iranian populations of the steppe. The gene pool of the latter has been marginalized but it maintains its genetic diversity.

The presence of haplogroup J2a here as the modal Caucasoid lineage, followed by haplogroups G1 and G2a is also quite interesting, and plausibly brings origin of the ancestors of the pre-Altaic inhabitants of the region in close proximity to the West Asian homeland of the ancestors of the Indo-Aryans.

PLoS ONE 6(3): e17548. doi:10.1371/journal.pone.0017548

Y-Chromosome Variation in Altaian Kazakhs Reveals a Common Paternal Gene Pool for Kazakhs and the Influence of Mongolian Expansions

Matthew C. Dulik et al.

Kazakh populations have traditionally lived as nomadic pastoralists that seasonally migrate across the steppe and surrounding mountain ranges in Kazakhstan and southern Siberia. To clarify their population history from a paternal perspective, we analyzed the non-recombining portion of the Y-chromosome from Kazakh populations living in southern Altai Republic, Russia, using a high-resolution analysis of 60 biallelic markers and 17 STRs. We noted distinct differences in the patterns of genetic variation between maternal and paternal genetic systems in the Altaian Kazakhs. While they possess a variety of East and West Eurasian mtDNA haplogroups, only three East Eurasian paternal haplogroups appear at significant frequencies (C3*, C3c and O3a3c*). In addition, the Y-STR data revealed low genetic diversity within these lineages. Analysis of the combined biallelic and STR data also demonstrated genetic differences among Kazakh populations from across Central Asia. The observed differences between Altaian Kazakhs and indigenous Kazakhs were not the result of admixture between Altaian Kazakhs and indigenous Altaians. Overall, the shared paternal ancestry of Kazakhs differentiates them from other Central Asian populations. In addition, all of them showed evidence of genetic influence by the 13th century CE Mongol Empire. Ultimately, the social and cultural traditions of the Kazakhs shaped their current pattern of genetic variation.



andrew said...

Clearly a typo: "I've started keeping track of cases where the germline rate has a better fit to the archaeological record than the germline rate."

Valikhan said...

"we believe the best interpretation of the data is that Kazakh Y-chromosome diversity was strongly influenced by the Mongols of the 13th century CE."

That is odd. Altaians Kazaks are mainly from Kerei and Naiman tribes. Both existed and had their own khaganates long before any Mongols came to a historical scene.