The Secrets of Central Asia: Chapter II - The Nomads of West Siberia [Review]

Introduction
Molodin et al. have conveniently released an exciting paper just days ago, revealing the convergence and possible origins of maternal lines in several West Siberian sites across different points of time.

The authors made the following conclusions based on the data they had gathered;

"We therefore consider the appearance of the Haplogroup T-lineage as the most likely genetic marker of the Andronovo migration wave to the region....
Apparently, the Andronovo group... assimilated the aboriginal... population, from which it obtained these East-Eurasian mtDNA haplogroups. Obviously, there was reciprocal genetic contact between the migrant and indigenous groups in the region.
...These [autochthonous] components were represented by the Eastern Eurasian haplogroups A, C and Z, and the Western Eurasian haplogroup U5a. On the other hand, the results also reveal some changes in the mtDNA pool structure throughout the Bronze Age. Some of these changes, which point to migration waves to the West Siberian forest steppe zone, are in agreement with the archaeological and anthropological evidence. The most relevant migration waves occurred during the Middle Bronze Age (represented by the migration of the Andronovo culture, probably marked by Haplogroup-T lineages) and the transition from the Bronze to the Iron Age (represented by the migration from the south, marked by the U1a, U3 and H haplogroup lineages)."

[PDF]

In this blog entry, these conclusions reached are scrutinised together with the deeper ancestral associations of these haplogroup lineages with modern (and other ancient) populations.


The Original Paper's Findings
A total of 92 ancient DNA (aDNA) haplotypes in the form of mitochondrial DNA (mtDNA) were retrieved from five sites stratified across seven distinct archaeological periods in a fixed portion of West Siberia known as the Baraba forest-steppe, lying between the network formed between the Irtysh and Ob rivers. These haplotypes were obtained from Hypervariable Region 1 (HVR1) of mtDNA and are included in the original study (shown as Table 3). 

Sampling Sites in Babara Forest-Steppe

As no burial remains have been found dating to the Pleistocene (11th-12th millenium BC) in or around the Baraba forest-steppe, which is the earliest period where anatomically modern humans reached this region, the ultimate origins of the Early Bronze Age lineages are left open to interpretation. Nonetheless, below is a summary of each archaeological culture showcased in the paper, as well as relevant extracts from the literature. [1]


Ust-Tartas (4000-3000 B.C.)

The inhabitants of the earliest grave-containing Baraba prehistoric culture appeared to be Caucasoid-Mongoloid hybrids based on anthropological data whose distribution spanned the swathe of forest from Karelia and the Baltic through to the Ural region. Numerous Russian sources have previously described this concept as the Northern Eurasian Anthropological Formation (e.g. Bunak V.V.). Additionally, a comparison with the nearby Comb-pit Ware culture revealed enough anthropological similarities to suggest the individuals of Ust-Tartas were likely to be autochthonous and not recent migrants.

Extent of the N. Eurasian Anthropological Formation

Of the 18 mtDNA haplotypes retrieved, East Eurasian lineages (A, C, D, Z) comprised a slight majority (11/18). The authors noted "widely distributed root haplotypes" for Haplogroups C and D, which presumably indicates greater antiquity of both in the region. The two individuals belonging to haplogroup A "[represent] a subcluster that is apparently characteristic of West Siberia and the Volga-Ural Region". There was surprise at the presence of Haplogroup Z based on its' absence in modern inhabitants of West Siberians, a topic explored later in this entry.

The seven West Eurasian mtDNA Haplogroups belonged entirely to U, comprising of U2e, U4* and U5a1. The authors recalled the findings of several other recent studies on ancient DNA, stating it likely belonged to "Eastern, Central and Northern European hunter-gatherer groups". [1] 

Besides affirming previous literature concerning the migration corridor between East Europe to East Asia, the haplotypes also complement the anthropological data concerning their status as Mongoloid-Caucasoid hybrids.

Odinovo (3000 B.C.) and Krotovo (Early, 2000 B.C.)

Both of these cultures, regardless of stage, represent a fairly linear continuity from the populations and traditions of the Ust-Tartas culture before them.

The Odinovo culture succeeds Ust-Tartas, although it is viewed as a synthesis between it and the Comb-Pit Ware archaeologically. Anthropological kinship between it and contemporary Baraba findings also confirm the autochthonous nature of Odinovo. However, it differs from its' antecedents in grave objects, funeral rites and the presence of bronze artefacts belonging to the Seima-Turbino cultural phenomenon, a short-lived (2200-1700 B.C.) but "striking" package of metallurgical  goods originating around the Sayan-Altai region in South Siberia that was oriented westwards towards Europe. [2]

In turn, the Krotovo culture is partially derived from Odinovo, although it isn't without its' own influences from adjoining regions. As well as "strikingly different" funeral rites, [1] new archaeological features, including items fashioned out of chalcedony, jaspilite and enstatite, point toward interactions of some degree with the Petrovo culture found further south in Kazakhstan, where the nearest deposits of these materials lie. It is worth noting the physical type of the Krotovo people revealed no significant changes, remaining in-line with the previous autochthonous type.

A total of 16 mtDNA haplotypes were recovered from both Odinovo and the Early Krotovo stage. The spectrum of mtDNA Haplogroups remain unaltered from the Ust-Tartas samples, supporting the archaeological record of continuity. 

The paper goes on to elaborate on the discrepancy between the mtDNA results and the archaeological features of Krotovo by stating "our data did not allow us to detect any Central Asian genetic influence". [1] Several possible explanations which may be considered;

1. New material items from Petrovo accompanied a male-mediated migration towards Krotovo, resulting in some level of cultural assimilation
2. In support of the above, the Petrovo culture natives may have themselves been a southward extension of the "Northern Eurasian Anthropological Formation" and belong to the same basic physical type as Ust-Tartas, Odinovo and Krotovo individuals further north, making any inter-culture interactions difficult to infer
3. Some mode of transmission between Krotovo and Petrovo took place (trade, "package diffusion")

Further information is needed to ascertain which is more probable, including (but not restricted to) Y-Chromosomal data from all concerned cultures for evidence of (dis)continuity between Odinovo and Krotovo through southern influence, as well as anthropological data from Petrovo to determine if they were indeed of the same basic physical type.

The summation of the evidence provided, however, indicates material items from further south were brought northwards into the Baraba forest-steppe after 2000 B.C., but these cultural changes do not reflect in the native maternal lineages, implying less overt processes (or male-mediated migration) were causative.

Krotovo (Late, 1750 B.C.) and Andronovo (1500 B.C.)

The next significant period of Baraban history comes with the arrival of semi-nomadic pastoralists whose origins lay further to the west. We are, of course, referring to the founders of the Andronovo archaeological complex, whose Indo-European language, culture and even ideology had eventually infiltrated deep into the Iranian plateau and Indian subcontinent through their utilisation of both horse and chariot. [3]

Schematic Tree of mtDNA Haplogroups  Found

Within Baraba, despite the Krotovo population coexisting with these newcomers for a length of time (presumably due to their occupancy of different pastoralist niches), we see evidence of a shift from Seima-Turbino to Andronovo with regard to their material traditions. Andronovan dominance is also reflected in the eventual northward displacement of some Krotovo natives based on archaeological data. [1] However, cranioanalysis presents a more complicated picture; the presence of an "autochthonous Mongoloid" variant not typically seen in the Baraba steppe-forest, differing from the hybrid type seen for hundreds of years prior, may suggest the two were not in direct contact and Andronovan influence was exerted by proxy of other native groups who were displaced northwards and east following their assimilation. This is anecdotally supported by Keyser et al.'s discovery of one Andronovo male (specimen S07) from near Krasnoyarsk in South Siberia carrying Y-DNA Haplogroup C*. [4] It is worth stating the physical type of those from Andronovo are commonly described as "Variants of three proto-Europoid types " with minor Mongoloid. [1]

40 mtDNA haplotypes from Late Krotovo (1750 B.C) and Andronovo (1500 B.C.) sites and time periods were taken.  As expected, the same spectrum of mixed West-East Eurasian lineages made an appearance, except for the strong introduction of one new Haplogroup.

In both Late Krotovo and Andronovo, Haplogroup T reaches a stable frequency of 15% in both despite being completely absent in 34 earlier haplotypes. The authors cite this as direct genetic evidence of Andronovan influence on Late Krotovo and postulate this lineage was, as a result, a major contingent in the Andronovo culture's spread.

All of these events precede the Irmen culture (1400-900 B.C.), the eventual successor to Andronovo. Those Irmen individuals found in the Baraba region were found to be predominantly Caucasoid and practiced a mixed economy of agriculture and animal husbandry. Only data from the Late stage (900-800 B.C.) was considered in the study.

Baraba (Late, 1000 B.C.)

The Late Baraba culture is a consequence of a Krotovo-modified Andronovo successor (known as Suzgan) interacting with the Irmen culture (described above). This was a particularly tumultuous period in West Siberian prehistory with tribes continuously coalescing unto one another, forming new identities in the process.

Anthropological data from the Late Baraba culture painted a far more diverse picture than over the previous three millennia. The authors noted that, contrary to the general insignificance of gender on physical type, the men were found to be more similar to a "Southern Eurasian Anthropological Formation", whereas females were closer to the Andronovan derivatives in North Kazakhstan. 

Only five mtDNA haplotypes were recovered from this period. Haplogroups A and C once again were represented, as was U5b and T, indicating the previous assimilation events had been maintained uptil this point. 

Irmen (Late, 900-800 B.C.)

From 1000 B.C. onwards, a complex set of migrations took place in West Siberia between the cultures formed by this point. Archaeologists attribute this to ecological changes involving climatic cooling across the region. 

The last sampled site is the Late Irmen culture, which is a continuation of the Irmen culture proper described earlier in this entry. The intricate interactions between cultures of this period are evident through multi-plural settlements in the archaeological record here.

The final 14 mtDNA haplotypes were, unexpectedly, a complete departure from the partial continuity that we have seen since Ust-Tartas uptil Late Baraba. Almost all belonged to West Eurasian lineages, such as Haplogroups J, K and W. The study had suggested the ultimate origins of these lineages came from further south, in the vicinity of West Kazakhstan and West Central Asia (Turkmenistan and Uzbekistan likely implied). This suggestion will be assessed in detail later in this entry.


Confirmation of the 'Migration Corridor'?
It is remarkable to finally find genetic evidence of the migration corridor, an archaeological concept mentioned several times in Vaêdhya, firmly imprint it in such a definitive way (visit North European Component Variation within the Eurasian Heartland for additional information). 

As it stands, we can now safely conclude that prehistoric hybridisation between hunter-gatherer Paleo-European populations and those from along the East across the Eurasian steppe. The crossover of both along opposing ends of this corridor has been supplemented with aDNA and anthropological evidence, with the finding of a near-equal hybrid population midway between the two poles all but confirming what the raw results have already revealed. Therefore, the connection between Northeast Europe and East Asia through the Eurasian steppe (even before Proto-Indo-European's formation) can no longer be considered a hypothesis, but a verified reality of demic prehistory. If supported with autosomal DNA (auDNA) from similar gravesites, it will drastically alter our perception of the migrations that happened afterwards, as well as doing away with over-simplified models of how certain languages and cultures permeated across Eurasia.

Afanasievo: Without a trail?

It is interesting to note that, despite covering over 3,000 years of prehistory, there is yet to be a trace of the Afanasievo culture, the earliest known offshoot of Yamnaya in the east, across this territory. Under the Eurasian steppe theory, the Afanasievo culture is connected with pastoral nomads who spoke an early (proto) form of the Tocharian branch, an extinct Centum Indo-European language which subverts the Centum:Satem isogloss in Eurasia. [5] The only attested connection between Afanasievo and the Baraba forest-steppe is through interactions between its' successor culture, the Karasuk, with the easternmost of the early Irmen. [1]

The question that persists is thus; where is the Afanasievo trail from Yamanaya through to the Urals and their final archaeological seat in South Siberia? Why have none of the Baraba forest-steppe cultures shown any indication of influence, be it cultural or anthropological, of Caucasoid pastoral nomads before those of Andronovo? 

To arrive at one likely answer, Frachetti's Pastoralist Landscapes and Social Interaction in Bronze Age Eurasia clarifies the material culture and mode of living in Central Asia during the Bronze Age;

"The calibrated C14 dates of Afanas'evo material are generally slightly earlier than those taken from Yamnaya contexts in the western steppe, which complicates a diffusionist explanation of the emergence of pastoralists in the eastern steppe. Although their origins may be obscure, communities associated with Afanas'evo materials still represent the earliest mobile pastoralists east of the Ural Mountains... [their] incipient strategy of cattle and sheep/goat herding, supplemented by hunting and fishing.
The Afanas'evo subsistence economy might best be characterized as a mixed or transitional form between hunting/fishing and localized pastoralism, arising from local antecedents or combining native strategies with diffused domestic innovations among local populations.
...Perhaps the strongest evidence that divides the Yamnaya and Afanas'evo pastoralists in the mid-fourth millenium BCE is the discontinuity of pastoral economic strategies among societies living between these territories." [6]

If the Afanasievo culture was itself a combination of local hunting strategies and farming practices with their origins further west in the Yamnaya despite differing with contemporary societies above the Black and Caspian seas, one can postulate the Afanasievo people would have likely intermingled with native cultures in South Siberia whilst retaining their core pastoral attributes, and such an event would have occurred some time earlier. 

The Afanasievo bearers needn't travel through the Baraba forest-steppe neither; the maps shown in Chernykh's The “Steppe Belt” of stockbreeding cultures in Eurasia during the Early Metal Age, for instance, show a straight trajectory from the Urals to the Sayan-Altai region out of clarity rather than a factual basis. Little is currently known about the journey taken by these nomads, but the findings of this paper do help in confirming the founders of Afanasievo did not stray along the northern rim of the forest-steppe towards South Siberia.

References
1. Molodin VI, Pilipenko AS, Romaschenko AG, Zhuravlev AA, Trapezov RO. Human migrations in the southern region of the West Siberian Plain during the Bronze Age: Archaeological, palaeogenetic and anthropological data. 2012. Retrieved from here: http://www.degruyter.com/dg/viewbookchapter.fullcontentlink:pdfeventlink/contentUri?t:ac=books$002f9783110266306$002f9783110266306.93$002f9783110266306.93.xml [Last Accessed 17th July 2012]

2. Chernykh E. The “Steppe Belt” of stockbreeding cultures in Eurasia during the Early Metal Age. Trabajos De Prehistoria. 2008;65:73-93.

3. Kuz'mina EE. The Origin of the Indo-Iranians. Koninklijke Brill NV, Leiden, The Netherlands. 2007.

4. Keyser C, Bouakaze C, Crubézy E, Nikolaev VG, Montagnon D. Ancient DNA provides new insights into the history of south Siberian Kurgan people. Hum Genet. 2009;126:395–410.

5. Anthony DW. The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian Steppes Shaped the Modern World. Princeton University Press. 2007.

6. Frachetti MD. Pastoralist Landscapes and Social Interaction in Bronze Age Eurasia. University of California Press, Ltd. 2008.

Secrets of Central Asia: Chapter I - The Pokrovsk Man [Review]

The first of a series focused entirely on ancient and prehistoric Central Asian ancient DNA (aDNA), this entry covers the furthering of an investigation into frozen remains found in a remote part of Siberian Russia.

Pokrovsk, Sakha Republic, Russia

Introduction
In 2006, Amory et al. tested bone fragments of a grave found near Pokrovsk, a locale the Russian federal republic of Sakha (Yakutia) with the intention of discerning the remain's origins. [1] Amory et al. briefly elaborate on the purported archaeological history of Siberia, where an autochthonous hunter-gatherer population was either subjugated or partially displaced by expanding Tungus-Manchurian nomadic tribes, before the movement of Yakut herdsmen northwards into their present demographic range as a result of Mongolian domination in the region between the sixth and thirteenth centuries. The Abstract of the paper below:

"The Yakuts, Middle Age Turkic speakers (15th–16th centuries), are widely accepted as the first settlers of the Altai-Baikal area in eastern Siberia. They are supposed to have introduced horses and developed metallurgy in this geographic area during the 15th or 16th century a.d. The analysis of the Siberian grave of Pokrovsk, recently discovered near the Lena River (61_29_ N) and dated by accelerator mass spectrometry from 2,400 to 2,200 years b.p., may provide new elements to test this hypothesis. The exceptional combination of various artifacts and the mitochondrial DNA data extracted from the bone remains of the Pokrovsk man might prove the existence of previous contacts between autochthonous hunters of Oriental Siberia and the nomadic horse breeders from the Altai-Baikal area (Mongolia and Buryatia). Indeed, the stone arrowhead and the harpoons relate this Pokrovsk man to the traditional hunters of the Taiga. Some artifacts made of horse bone and the pieces of armor, however, are related to the tribes of Mongolia and Buryatia of the Xiongnu period (3rd century b.c.). This affinity has been confirmed by the match of the mitochondrial haplotype of this subject with a woman of the Egyin Gol necropolis (Mongolia, 2nd/3rd century a.d.) as well as with two modern Buryats. This result allows us to postulate that contacts between southern steppe populations and Siberian tribes occurred before the 15th century."

[Link]


Grave Features
The Pokrovsk grave is located at the top of a glacial terrace near the Lena-Pokrovsk river junction. Radiocarbon dating places the site at approximately 2390-2190YBP. The physical type of The Pokrovsk Man was found to be gracile skeletally with a brachycephalic skull. The physical type was found to be Mongoloid, although the authors note it was "less accentuated" than that of Middle-Age Yakuts. It was also noted that the torus mandibularis, a normal variational bony protuberance located within the interior aspect of the mandible, was absent, despite it occurring commonly in East Asian and Native American populations. [2] Several material items were observed in the grave, including bone tools, harpoon heads, reindeer bone armour and flint arrowheads connected to archaic Siberian culture. However, other goods, including an iron arrowhead, are reputedly of South Siberian built. [1]


Methods
DNA extraction from bone by technique outlined in Keyser-Tracqui & Ludes’  Methods for the study of ancient DNA. [3] Autosomal DNA (auDNA) was retrieved from Profiler+ Multiplex kit (nine Short Tandem Repeat’s, or STR’s). Y-Chromosomal DNA (Y-DNA) tested using Powerplex Y System (eleven STR’s) as well as a Single-Nucleotide Polymorphism (SNP) on the TAT locus. Finally, a 421 base pair (bp) segment on the sample’s mitochondrial DNA (mtDNA) at the first hypervariable segment (HVS1) was tested (position 15989→16410) and compared with the Cambridge Reference Sequence (CRS). 

Consensus data obtained directly from paper; auDNA analysis was achieved through popSTR, an online research processing engine which displays auDNA STR allele frequencies within different populations. [8]
mtDNA and Y-DNA analysis would have ideally been conducted through ySearch, mitosearch, the SMGF, supplementary data from relevant scientific literature as well as online DNA projects.

Allelic Frequencies

auDNA Analysis

Nine auDNA STR’s were retrieved from the Pokrovsk Man's remains. Unfortunately, the utilisation of STR's is questionable given they have a large margin of error and lack of population specificity due to the presence of multiple alleles within a single population, as well as heavy inter-population overlapping. This investigative tool has largely been made redundant by SNP testing, which employ thousands of markers rather than a few. Nonetheless, processing of these results will still be attempted.

The allelic frequencies per worldwide regional groups for the retrieved STR’s are shown opposite. All markers from the Profiler+ Multiplex were utilised in the subsequent popSTR search. The sample populations are largely derived from the HGDP-CEPH Human Genome Diversity Cell Line Panel. [4]

African frequencies of the Pokrovsk alleles are generally lower relative to Eurasian, Oceanian and American regional groups. This warrants the exclusion of such values from the analysis hereon due to their uninformative nature, apart from confirming the Pokrovsk Man had no recent African ancestry, which is in accordance with anthropological, historical and linguistic data from Siberia. Allele frequencies of remaining regions are shown in the Data Sink.

To elucidate the regional affinities of the Pokrovsk Man, averages for the alleles across the given regions were taken and ranked in order of descending magnitude (found again in the Data Sink).

The results indicate his affinity was greatest to the Americas, followed by East Asia and Europe (discussed later) in joint position, ending with the Middle-East and South-Central Asia. The discrepancy between the American and East Asian scores are explained by the East Asia regional group being constituted largely of ethnic groups from East Asia proper and Southeast Asia, such as the She, Naxi and Japanese. The Yakuts, who are the only sample population located in Siberia, are a part of this group, reducing the specificity further. However, the greater score to native American and East Asian populations than others is still consistent with both geographic position and the known demic expansions into of both regions.

The decreased allelic frequency average of South-Central Asians and Middle-Easterners with the Pokrovsk Man supports the above further. However, the Middle-Eastern group did not include populations from West Asia or the Caucasus, such as Anatolian Turks, Iranians or Georgians. Additionally, the lack of North-Central Asian ethnic groups such as the Kazakh, Tatars or Altaians may affect the results further.

It would have been preferable if auDNA SNP’s were obtained instead and compared with specific sample populations - Better yet if IBD segment analysis was also undertaken. SNP analysis could have been possible in 2006, given the HGDP-CEPH samples were made available at least four years prior, [4] which would have opened the door to analysis far deeper than the extent undertaken by Amory et al. or even this investigation.

The authors greatly limited the extent of their own investigation, noting the Pokrovsk Man showed identical matches with Buryats, West Siberians, Altaian Mansis, ancient and modern Yakuts, one Evenk and an Egyin Gol necropolis female [5] in their private haplotype database.


mtDNA Analysis
Of the ten loci tested, only three yielded consistent nucleotide variations (16223T-16362C-16368C). The mitosearch 1-step matches with a known maternal ancestor location were considered only (Data Sink). These results not only confirm Amory et al.'s conclusion the Pokrovsk Man belonged to mtDNA Haplogroup D, but the bulk of the distribution within Asia is expected based on modern samples. [6]

Unfortunately, once more, the scope of the initial investigation has hindered any further analysis, as the lack of testing regions beyond HVS1 cannot elucidate the extent of mitochondrial sharing outside of the data showcased here.


Y-DNA Analysis
None of the eleven Y-DNA STR's provided a successful return. The only SNP tested for was TAT, where a T→C mutation is considered equivalent to the M46 marker, which is defined as Haplogroup N1c under the current International Society of Genetic Genealogy (ISOGG) nomenclature. [7]

As the Pokrovsk Man yielded a T allele at this locus, his Y-DNA Haplogroup could not have been N1c-M46. However, this does not rule out him belonging to a lineage upstream of N1c-M46.


European Affinities & Conclusion
Despite the great limitations, several invaluable inferences can be made from the data presented in the furthering of Amory et al.'s Early influence of the steppe tribes in the peopling of Siberia which cannot be reasonably excluded as anomalous without also discarding conclusions made from other sources.

The auDNA results, though derived from STR data, fully agree with the SNP-based analysis of the Eurogenes Project by David W. in a previous run (described in an earlier Vaêdhya entry), as modern Siberian populations show trace values of various European or Caucasian ADMIXTURE components at the least with an absence of Southwest or South Asian specific components, whilst being predominantly Siberian and East Asian.

The European affinity in this investigation coming third may form a convenient explanation for why the Pokrovsk Man's features were less Mongoloid anthropometrically than Middle-Age Yakuts. It may suggest a West Eurasian physical element existed prior to the tribal and political upheavals that resulted in the Yakut settlement deeper into this portion of Siberia. Although the origins of this element were not elaborated upon, there may also be a connection with the postulated "migration corridor" covered previously and described in Malyarchuk et al.'s On the Origin of Mongoloid Component in the Mitochondrial Gene Pool of Slavs. [10]

This result supplements the picture of a West Eurasian genetic component of ambiguous origins being brought towards Siberia, challenging one interpretation of West Eurasian physical influence in the region stopping abruptly at Lake Baikal. [9] Instead, the totality of the evidence presented raises the possibility of this influence extending itself beyond the lake and manifesting itself simply as a "reduction" of Mongoloid cranial characteristics, which the Pokrovsk Man demonstrated, whose anthropometric configuration may well have been an artefact of this.

Unfortunately, the mtDNA and Y-DNA results were far too non-specific to merit further analyses. Their generality, however, do pose several questions; what subtype of mtDNA Haplogroup D did the Pokrovsk Man belong to? If he was not Y-DNA Haplogroup N1c-M46, what was he?

The material goods found in the Pokrovsk Man's gravesite may point us in the direction of the orientation his apparent European affinities came from. As South Siberia was the source of his iron and horse-derived goods, could he also have inherited West Eurasian genes from there? Were the benefactors ancient, or prehistoric?


Acknowledgements
Pokrovsk map from WolframAlpha.


References
1. Amory S, Crubézy E, Keyser C, Alekseev AN, Ludes B. Early influence of the steppe tribes in the peopling of Siberia. Hum Biol. 2006;78:531-49.

2. Apinhasmit W, Jainkittivong A, Swasdison S. Torus Palatinus and Torus Mandibularis in a Thai population. ScienceAsia. 2002;28:105-111.

3. Keyser-Tracqui C, Ludes B. Methods for the study of ancient DNA. Meth. Mol. Biol. 2005;297:253–264.

4. Rosenberg NA. Standardized subsets of the HGDP-CEPH Human Genome Diversity Cell Line Panel, accounting for atypical and duplicated samples and pairs of close relatives. Ann Hum Genet. 2006;70:841-7.

5. Keyser-Tracqui C,Crubézy E, Ludes B. Nuclear and Mitochondrial DNA Analysis of a 2,000-Year-Old Necropolis in the Egyin Gol Valley of Mongolia. Am J Hum Genet. 2003;73:247–260.

6. Mishmar D, Ruiz-Pesini E, Golik P, Macaulay V, Clark AG. Natural selection shaped regional mtDNA variation in humans. Proc Natl Acad Sci. 2003;00:171-6.

7. Zerjal T, Dashnyam B, Pandya A, Kayser M, Roewer L. Genetic relationships of Asians and Northern Europeans, revealed by Y-chromosomal DNA analysis. Am J Hum Genet. 1997;60:1174–1183.

8. Amigo J, Phillips C, Salas T, Fernández Formoso L, Carracedo A. pop.STR - An online population frequency browser for established and new forensic STRs. Forensic Sci. Int. Gene. Suppl. 2009.

9. Mooder KP, Schurr TG, Bamforth FJ, Bazaliiski VI, Savel'ev NA. Population affinities of Neolithic Siberians: A snapshot from prehistoric Lake Baikal. Am J Phys Anthropol. 2006;129:349-61

10. Maliarchuk BA, Perkova MA, Derenko MV. Origin of the Mongoloid component in the mitochondrial gene pool of Slavs. Genetika. 2008;44:401-6.