Contents Türkic Genetics
Contents Amerin Genetics
Klyosov A. Türkic DNA genealogy
Klyosov A.The 3 R's in R1 Haplogroup
Ogur and Oguz
Alans and Ases
Russian Version needs a translation
|Scythian-Iranian theory||Ossetian Genetics||Scythian language - Sources||Etruscan Genetics|
|Türkic Genetics (Papas and Mamas)|
| Hun Circle Kurgan
A Western Eurasian Male Is Found in 2000-Year-Old Elite Xiongnu Cemetery in Northeast Mongolia
Anatole A. Klyosov
http://www.lulu.com/items/volume_68/8657000/8657872/1/print/8657872.pdf Anatole A. Klyosov genetic analysis pp.623-634
This posting attempts a critical review of the first Y-DNA and mtDNA study of the “Xiongnu” people, and a major contribution to the science of Turkology. The criticism is addressed to the historical aspects of the article, which technical results are admirable and significant, but were substantially complemented by A.Klyosov's analysis of R1a1 development, see The 3 R's in R1 Haplogroup and the following comments. To dispel the unfair claim of the authors, here is the statement of Lamberg-Karlovsky, an unbiased arbiter on the Kurgan problem, his opinion (2005) is the opposite of what is asserted in this 2010 publication: "Given the increasingly large number of divisions and subdivisions of the generic Andronovo culture(s), with evidence for "no one group having undue prestige over the others," there is neither reason nor evidence to believe that they all shared an Indo-Iranian language. From the common roots of the millennia-long Andronovo culture(s) [and before that the related Timber Grave culture(s)], processes of both convergence and divergence [archaeologically indicated by the eastward migrations of the Andronovo culture(s)] allow for the presence of not only the Indo-Iranian languages but for other language families as well, that is, Altaic and Uralic. Clearly, the convergence of cultures, that is, the assimilation of local populations by an in-coming peoples, is very poorly developed within the archaeological discipline." A 2002 Lamberg-Karlovsky's statement reads: "There is, however, no compelling archaeological evidence that they (Andronovo and Bactrian Margiana archaeological complexes ) had a common ancestor or that either is Indo-Iranian." The signal attributions are misplaced: Europeans are not from Europe, Indo-Europeans are neither Indian nor European. The content of the present study is exceedingly valuable; its spin method is an echo of the times discarded, and does a great disservice to the authors. A prejudiced analysis is not an analysis, but willful machinations, and nowadays dubious references to authorities do not create facts on the ground.
The posting excluded main portions of the work that do not cause
objections, for a complete article please refer to the above link. From some omitted
paragraphs, only a pertinent material is cited, to provide a contextual background for the general comments.
The title is clearly misleading, obviously intended to stir up Eurocentric emotions: the
Western Eurasian the authors call or Baikalian or Scythian, a far cry from Western lore. It
appears that the author's loose use of the word "Indo-European" is predominantly in a geographical sense, but
in places the authors clearly apply it to the Caucasoid phenotype. The use of the terms "European"
and "Indo-European" instead of the "Caucasoid" tends to cloud the subject, and was denoted in
this posting by replacing the inappropriate term with a
For a professional analysis by Anatole A. Klyosov of the authors' verbal equilibristic, go to http://www.lulu.com/items/volume_68/8657000/8657872/1/print/8657872.pdf (Proceedings of the Russian Academy of DNA Genealogy, vol. 3, No. 4, 2010, pp. 623-634), or read a mirror below. In addition to constructive criticism that debunks the baseless enthusiastic references to Western Eurasian and Indo-European, the quantitative analysis allows to discern separate migration waves diverging from the main trunk, and thus receive a timing sequence of events, which puts the “Western Eurasian and Indo-European” squarely many millennia before the Western Eurasians and Indo-Europeans came to being.
The posting's comments and explanations added to the text of the article are shown in parentheses in blue font, (blue italics) or blue boxes. Highliting of main terms and points are posting additions.
Anatole A. Klyosov
A comment on the Paper:
A Western Eurasian Male Is Found in 2000-Year-Old Elite Xiongnu Cemetery in Northeast Mongolia
The cited paper is of the highest interest and importance for further revealing the history of humankind in Eurasia in general, and the bearers of haplogroup R1a in particular. However, it is open for some critique, starting with the very title of the paper. First, it was not a “Western Eurasian Male” which was unearthed in northeastern Mongolia. It was a male of the R1a1 haplogroup, who equally likely could have belonged to Eastern Eurasia, or Southern Siberia, or adjacent regions. Since his haplotype was not presented in the paper, there was no way of assigning of it to be necessarily Western Eurasian.
Second, the authors of the cited paper repeatedly made the statement
which belongs to false stereotypes, that “R1a1 is considered as an
Indo-European marker…”, and, to make it even less accurate, if not to say worse, the
authors continued, regarding R1a1, “supporting Kurgan expansion hypothesis” (Introduction). In fact, both R1a1 and R1b1 were on the Russian Plain in
times of the so-called Kurgan culture, with R1a1 migrating generally East and
South, and R1b1 migrating West and South, R1b1 apparently a thousand years earlier compared to R1a1. Hence, it is a common mistake to assign only R1a1 to
the “Kurgan culture”, which, supposedly, was moving mainly towards Europe, which is in the opposite direction compared to that of R1a1 tribes.
The authors rightly state – “R1a is the most common haplogroup in Europe” (Discussion), however, this does not make Europe the only source of R1a haplogroup. R1a belongs to Eastern Eurasia no less (and probably more) than it does to Europe; R1a was apparently originated in Eastern Asia (see below) therefore it just cannot possibly be an Indo-European marker. They precede Indo-Europeans by thousands and thousands of years. Allow me to explain.
To my knowledge, there were six series of data published in the literature which indicated “non-Indo-European” R1a haplotypes of ancient common ancestors in East Asia. They are as follows:
(1) The most ancient source of R1a haplotypes appeared to be provided by people who now live in North China, which geographically is South Siberia. In an article by Bittles et al (2007) titled “Physical anthropology and ethnicity in Asia: the transition from anthropometry to genome-based studies” a list of frequencies of haplogroup R1a1 is given for a number of Northern Chinese populations (ethnic communities Hui, Bolan, Dongxiang, and Sala, with R1a1 shares of 18%, 25%, 32% and 22%, respectively), however, haplotypes were not provided. The corresponding author, Professor Alan H. Bittles, kindly sent me a list of 31 five marker haplotypes (in the format of DYS19, 388, 389-1, 389-2, 393), which was published in (Klyosov, 2009a). The respective haplotype tree is shown in Figure 1.
These 31 haplotypes contain 99 mutations from the base haplotype (in the FTDNA format)
which gives 0.639±0.085 mutations per marker, or 21,000±3,000 years to a common ancestor. Details of calculations and the methodology of DNA genealogy in general are given in (Klyosov, 2009a,b,c). Briefly, the
emerging science of DNA genealogy is essentially a merge of DNA sequencing and chemical kinetics. DNA sequencing provides a pattern of alleles (STR’s,
short tandem repeats in certain loci of the Y chromosome) and haplogroup identification (via SNPs, which are single nucleotide polymorphisms).
Chemical kinetics makes possible a quantitative analysis of these STR patterns in
terms of separation of the haplotype tree into different lineages, and application
of the methodology of chemical kinetics to dynamics of accumulation of mutations
in the loci (employing the “logarithmic” approach to describe the kinetics
of disappearance of ancestral haplotypes and corrections for reverse
mutations, permutation methods which do not consider alleged ancestral, or “base” haplotypes, etc.).
The 31-haplotype tree was composed from data provided by Dr. A.H.Bitttles and collected in ethnic communities Hui, Bolan, Dongxiang, and Sala
(Bittles et al, 2007) (no haplotypes were provided in the referenced article)
Since haplotypes which descended from such an ancient common ancestor
have many mutations, this makes their base (ancestral) haplotypes rather
uncertain. The permutational method (Klyosov, 2009b) was employed for the Chinese
set of haplotypes. The permutational method does not need a base haplotype, it
does not require a correction for back mutations, and it does need a
correction for a possible asymmetry of mutations. For the given dataset it has resulted in 19,625±2,800 years to a common ancestor of the ancient South Siberian haplotypes. This is within the margin of error with that calculated by
the linear method (Klyosov, 2009a), as shown above.
Contemporary “Indo-European” R1a1 haplotypes have an ancestral haplotype of the Russian Plain of about 5,000 years before present, which in the 5 marker format given above can be presented as follows (Klyosov, 2009a):
The mutational difference between the two is 2.65 (considering fractions of the averaged alleles in both base haplotypes) and corresponds to 15,700 years of a chronological difference between them (see Klyosov, 2009a), that is about 20,700 years before present for a common ancestor of the South Siberian series of haplotypes (Klyosov, 2009a). Hence, we arrived again at the same figure.
It is likely that haplogroup R1a1 appeared in South Siberia around 20 thousand years ago, and its bearers split. One migration group headed West, and arrived at the Balkans around 12 thousand years ago (Klyosov, 2009a). Descendants of another group continue to live in South Siberia (Northern China) since then, some 20 thousand years later.
(2) R1a1 haplotypes of three different tribal population of Andra Pradesh, South India (tribes Naikpod, Andh, and Pardhan) listed in (Thanseem et al, 2006) shown in a haplotype tree in Figure 2, and gave 7,125±950 years to a common ancestor (Klyosov, 2009a). The base (ancestral) haplotype of those populations in the FTDNA format is as follows:
It differs from the “Indo-European” Indian haplotype
by four mutations on six markers, which corresponds to 11,850 years between their common ancestors, and places their common ancestor to approximately 11,500 years bp.
Apparently, the bearers of R1a haplotypes made their way from South
Siberia to South India some 12,000 ago, and those haplotypes were again quite
different compared to the Aryan ones, or the “Indo-European” haplotypes.
The 46-haplotype tree was composed from data listed in (Thanseem et al, 2006). The designations of haplotypes are those used in the cited article.
(3) 110 of 10-marker R1a1 haplotypes of various Indian populations, both tribal and Dravidian and Indo-European castes, listed in (Sengupta et al, 2006) shown in a haplotype tree in Figure 3, gave 5,275±600 years to a common ancestor (Klyosov, 2009a). The base (ancestral) haplotype of those populations in the FTDNA format is as follows:
It differs from the “Indo-European” Indian haplotype
by just 0.5 mutations on nine markers (DYS19 is in fact equals to 15.5 in
the Indian haplotypes), which makes them practically identical. However, in
this haplotype series two different populations, the “Indo-European” one and
the “South-Indian” one were mixed, therefore an “intermediate”, apparently phantom “common ancestor” was artificially created with an intermediate TMRCA, between 4,050±500 (the “Indo-European” common ancestor of R1a in India) and 7,125±950 years (the Dravidian R1a common ancestor) before
The 110-haplotype tree was composed from data listed in (Sengupta et al, 2006).
The article contains 114 Indian R1a1 haplotypes, however, four of them were incomplete.
(4) Pakistani R1a1 haplotypes listed in the Sengupta (2006) paper (Figure 4).
Analysis of the 42 haplotypes gave 7,025±890 years to a common ancestor (Klyosov, 2009a). This value fits within margin of error to the “South-Indian” figure of 7,125±950 ybp for their common ancestor. The base haplotype was as follows:
It differs from the “Indo-European” Indian haplotype by two mutations on
The 42-haplotype tree was composed from data listed in (Sengupta et al, 2006).
Finally, ten Central Asian 10 marker R1a1 haplotypes listed in the same
paper by Sengupta (2006) contain only 25 mutations, which gives 4050±900 years to
a common ancestor (Klyosov, 2009a). It is the same value that we found for
the “Indo-European” Indian haplotypes.
The above suggests that there are two different subsets of Indian R1a1 haplotypes. One was brought by European bearers known as the Aryans, seemingly on their way through Central Asia, in the 2nd millennium BC. Another, much more ancient, made its way from South Siberia apparently through China, and arrived in India some time earlier than 7 thousand years ago. The “Indo- European” haplotypes came to India with the Aryans (R1a1) about 3500 years before present from the Russian Plain, where their common ancestor lived 4750±500 years before present (Klyosov, 2009a).
(5) Haplotypes excavated from a 2000-year old necropolis in the Egyin Gol Valley in Mongolia (Keyser-Tracqui et al, 2003). Haplogroups of those haplotypes were not determined, however, out of 27 ten marker haplotypes, seven apparently were N1c, two R1b, and 17 haplotypes were likely of R1a1 haplogroup. Of the latter, 134 markers have been typed, and they contained 72 mutations total (there was no such analysis in the cited paper) from the base haplotype (in the FYDNA format plus two alleles of YCAII):
It differs by as many as nine mutations from the R1a1 haplotype of the Russian Plain in the same format:
72 mutations per 134 markers in the 10 marker haplotypes give 72/134/0.0017 = 316 generations (without a correction for back mutations), or 456 generations with the correction, that is 11,400 years to a common ancestor counting from the time of the burial (0.0017 is the mutation rate constant for the haplotypes shown in the cited paper, as described in Klyosov, 2009b). Thus, a common ancestor of the excavated R1a1 haplotypes lived 13400 years before present. Besides, the nine mutation difference between his ten marker haplotype and that of the Russian plain 4800 years ago places THEIR common ancestor to approximately 21,800 years before present. This again fits to 21000±3000 years ago for a common ancestor of R1a1 haplotypes in South Siberia as shown above.
The cited paper (Kim et al, 2010) states – “Although nearly 11% of ancient skeletons in Egyin Gol Xiongnu graves have shown European haplotypes, the later study revealed that there was no R1a1 (Keyser-Tracqui et al, 2009)”. Indeed, the last paper says – “The additional analysis performed on Xiongnu specimens revealed that …none of the specimens from the Egyin Gol valley bore this (R1a1) haplogroup”. However, the last paper did not elaborate which haplogroups those 17 haplotypes (as well as any other haplotypes in the preceding work) belonged? They must belong to some haplogroup, mustn’t they? For instance, the following haplotype (grave 70)
13-25-16-11-11-14-X-X-X-X-X-11-31 – 19-23
(in the FTDNA format of the first 12 markers, plus YCAII)? If it is not
R1a1, then what other haplogroup could it be? Another 16 haplotypes described above,
are similar in kind, with their base haplotype given above.
(6) Altai R1a1 haplotypes listed in (Underhill et al, 2009). Twelve haplotypes out of 13 listed showed a rather recent base haplotype (in the FTDNA format plus DYS461):
These 12 haplotypes have only 7 mutations per 120 markers from the above base haplotype, which gives 7/120/0.0018 = 32 generations (without the correction for back mutations), or 33 generations with the correction, that is 825±320 years to a common ancestor. This is only the 12th century, plus-minus a few centuries.
However, this base haplotype differs by three mutations from the base haplotype of the Russian Plain, which in the same format is
Besides, one more Altaian haplotype from the list
differs by 7 mutations from other
Altaian haplotypes, and by 4 mutations
from the Russian Plain base haplotype. All of this places their common
ancestor at approximately 5900 years before present. All these features, namely,
recent common ancestors along with significantly different base haplotypes
indicate severe bottlenecks of the respective populations. Clearly, a common
ancestor of the Altaian haplotypes could not be an “Indo-European” one, since the
“Indo- European” carriers of R1a1 haplotypes appeared on the Russian Plain only
about 4800 years before present. Approximately 3500 years bp they moved to India and Iran (Klyosov, 2009a).
1) It was not a “Western Eurasian Male” which was unearthed in northeastern Mongolia. It was a male of the R1a1 haplogroup, who equally likely could have belonged to Eastern Eurasia, or Southern Siberia, or adjacent regions.
2) The statement by the authors that “R1a1 is considered as an Indo-European marker…” is inaccurate. It can equally be considered as a South Siberian marker.
This assignment depends on the context of the research. Then, the authors employ a linguistic term, which is not very appropriate in the given context; proto-Indo-Europeans would be a more proper term, though, also being linguistic, it is not very adequate for ancient tribes and their haplogroups. What is the most important, only consideration of (extended) haplotypes can help to make an assignment where the R1a1 haplotypes belong to, in terms of tribes, their locations, direction of their movements, and their apparent linguistic affiliation.
3) To say that “R1a1… is supporting the Kurgan expansion hypothesis” is incorrect. R1b1 arguably “supports” the “Kurgan expansion hypothesis”, however, in the westward direction, which is opposite to the overall direction of R1a1 migrations, eastward. Besides, the authors state elsewhere in their paper (Discussion) that the “Kurgan people … of the Volga steppe region, infiltrated Europe between the middle of the fifth and the second millennium BC”. In the middle of the fifth millennium BC, that is about 6500 years before present only R1b1 people were in the Volga steppe region (with a common ancestor of R1b1 there of 6775±830 years bp, see Klyosov, 2009d). R1a1 appeared on the Russian Plain only 4750±500 ybp (Klyosov, 2009a). This is another reason that the socalled “Kurgan people” could not have been R1a1, at least not 6500 years bp and much later. They could not have brought the “Indo-European language” to Europe, since they brought it FROM Europe and moved in the opposite direction. This “Kurgan theory” was effectively dismissed by linguists as long as 30-20 years ago, at least regarding the “origin” and spread of the IE language, however, the authors continue to employ it in their considerations. As D.
Anthony stated, “The Kurgan culture was so broadly defined that almost
any culture with burial mounds, or even… without them could be included” (Anthony, 2007). As we know now, both R1a1 and R1b1 were actually
included, albeit bearers of the two haplogroups were moving in different directions
and in different times. V. Safronov, one of the well known archaeologists, wrote
more than 20 years ago – “M. Gimbutas has composed a “stew” from different archeological cultures, different by origins and by chronology. Her
hypothesis was taken for granted by many European archaeologists, which have chosen
to ignore outrageous chronological errors, lack of serious justifications,
and many more inaccuracies with respect to archaeological sources, leaving aside
her misunderstanding of the very methodology of the problem (Safronov, 1989).
4) The statement “West Eurasian male likely represents a Bronze Age migration from the Black Sea region” has no ground without considering haplotypes of the excavated West Eurasian male, which have not been presented in the paper. It may have equally likely represented a Paleolithic R1a1 haplotype from South Siberia.
5) The concluding statement in the cited paper – “We showed for the first time that an Indo-European with paternal R1a1 … was present in the Xiongnu Empire of ancient Mongolia” does not have any ground on the reason explained above. To justify the above statement an actual haplotype should have been presented. It was not.
Anthony DW. 2007. The Horse, the Wheel, and Language (Princeton University Press, Princeton and Oxford), p. 307.
Safronov VA. 1989. Indo-European Homelands. Gorky, 242 pp.
Kijeong Kim et al.
A Western Eurasian Male Is Found in 2000-Year-Old Elite Xiongnu Cemetery in Northeast Mongolia
ABSTRACT We analyzed mitochondrial DNA (mtDNA), Y-chromosome single nucleotide polymorphisms (Y-SNP), and autosomal short tandem repeats (STR) of three skeletons found in a 2,000-year-old
Am J Phys Anthropol 000:000–000, 2010. © 2010 Wiley-Liss, Inc.
DN Duurlig Nars
When the Pontic-Caspian Steppe in the West of South Siberia was dominated by the Scythians, a branch of Iranians, the eastern portion of South Siberia was ruled by scattered tribes of horse-riding nomadic pastoralists. The dominant group of the tribes was the
The Kurgan expansion hypothesis explains the (false stereotype)
Ancient DNA (aDNA) study methods are being applied to a wide variety of anthropological questions (Kaestle and Horsburgh, 2002). The analysis of aDNA is difficult due to the inability to amplify the material to a significant degree and due to the contamination with modern DNA (Pa¨a¨bo et al., 2004; Mulligan, 2005; Hunter, 2006). Several optimized aDNA study methods have now been introduced (Rohland and Hofreiter, 2007; Meyer et al., 2008). Recently, we have developed a further improved method (Kim et al., 2008). To study the origin of the human remains in the elite
MATERIALS AND METHODS
In 1974, Mongolian archeologists discovered a
Korean–Mongol archeologists excavated one T-shaped (Tomb No. 2) and two rectangular tombs (Tomb No. 3 and 4). According to archeological evidence, Tomb No. 2 with a large dromos entryway was used between the 1st c. B.C. and the 1st c. A.D. and Tomb No. 3 between the 3rd c. B.C. and the 2nd c. B.C. Tombs were separated by piled-up stones. Most of them were disturbed by pillaging at some point in the past. Despite pillaging, many valuable funeral goods remained there. Further analysis is in progress. The main funeral goods are a chariot, a golden necklace, a silver spoon, a wooden parapet, a red-walled coffin decorated with thin golden plates, and bronze and iron artifacts. The golden necklace was only the second ones to be found in
The numbers in the lower inlet figure indicate the numbers of tombs from which the ancient skeletal remains of this study were excavated.
DN, Duurlig Nars.
Ancient human samples
The associated culture, the time period, the estimated age, and the morphological sex of the specimens studied are presented in Table 1. Metric skull traits are presented (Fig. S1 and Table S1). All procedures were carried out by Asians in a restricted clean room dedicated for ancient human DNA study and under sterile conditions, using UV-irradiated gowns with head and leg covers, latex gloves, and face- and mouth-masks. Autoclaved or presterilized materials were used. Work places, containers, and pipette surfaces were cleaned by bleach and UV irradiation at 254 nm. All steps were carried out in a fume hood and a laminar airflow clean bench. Extraction blanks and template blanks were included through the entire process. DNA extraction, polymerized chain reaction (PCR) preparation, and post-PCR works were performed in separate rooms. Sterile aerosol-barrier pipettes and pipette tips were used throughout all the manipulations. Different bones or different parts of the bone were analyzed for DNA information. We followed a published protocol (Kemp and Smith, 2005) to eliminate any potential DNA contamination on the bones. In brief, bone surfaces were removed. Bone fragments were immersed in bleach and irradiated with UV. Bone powders were made with Mixer Mill MM301 (Retsch). Ancient genomic DNA extraction was carefully carried out according to a published protocol (Kim et al., 2008). In brief, bone powders were incubated in an extraction buffer. After incubation of the supernatant with silica particles, the silica extracts were purified using ion-exchange columns with QBT, QF, and QC buffers (Qiagen). Elutes were concentrated with Amicon Ultra-15 Centrifugal Filter Unit with Ultracel-30 membrane (Millipore) and Microcon YM-30 Centrifugal Filter Unit (Millipore).
HV1, HV2, and several coding regions were analyzed. PCR was performed in a 20-ll reaction mixture containing 2-ll template DNA, 2-ll 103 PCR buffer (ABI), 0.2 mM dNTP mix, 2 mM MgCl2, 1 lM each primer (Table 2), 1 mg/ml BSA (NEB), and 0.8 U AmpliTaq Gold polymerase (ABI). A GeneAmp1 PCR system 9700 (ABI) was used with the following conditions: 958C for 10 min, 40 cycles of 30 s at 958C, 1 min at the annealing temperature, 1 min at 728C, followed by a final extension step of 7 min at 728C. Low-quality amplicons were reamplified in a nested PCR reaction under the same PCR cycling condition as above but with the first PCR product diluted 1:50 and 25 cycles. PCR products were purified using Qiaquick PCR purification kit (Qiagen) and bidirectionally sequenced. Sequencing results were analyzed using SeqManTM II software (DNASTAR). Sequences were compared to revised Cambridge reference sequence (rCRS) (Andrews et al., 1999). Haplogroups of aDNA were initially assigned with HV1 and HV2 polymorphisms using the well-established web-based programs, Haplogroup Prediction Tool and mtDNAmanager. We further analyzed diagnostic coding region SNPs for corresponding haplogroups based on a recently updated global human mtDNA tree (van Oven and Kayser, 2009). Coding region SNPs with no homoplasy (Behar et al., 2007) were used. Possible abnormal mutations were checked with GenBank database and Genographic project database.
The most closely related haplotypes of the three aDNAs among 8,424 mtDNA sequences from the Gen- Bank database were investigated by constructing a Neighbor-joining (NJ) tree (Saitou and Nei, 1987) with concatenated HV2 (nucleotide position, NP 47-360) and HV1 (NP 16,024-16,380) sequences by using Clustal W 2.0.11 (Larkin et al., 2007). The NJ tree was drawn using MEGA version 4.1 (Kumar et al., 2008) (Fig. S2). mtDNA sequences proximate to the aDNA sequences were further analyzed by reconstructing a bootstrapped NJ tree calculated from 1,000 resamplings of the alignment data (see Fig. 2). Proximate mtDNA sequences were compared to rCRS. Sequence similarities to aDNA sequences were determined (Tables S2–S5).
Real-time PCR for quantification of ancient mtDNA molecules
Real-time PCR assay was carried out to quantify the number of mtDNA molecules in the aDNA extracts of MNX3 European male and MNX4 female samples as follows. Two different size fragments (440 and 221 bp) within the HV1 mtDNA region were assayed using primer sequences: F15971/R16410 for the 440-bp fragment (Table 2); and F16190 (50-CCC CAT GCT TAC AAG CAA GT-30)/R16410 for the 221-bp fragment. Real-time PCR amplification was performed in 20 ll with 2 ll of extracts, 0.5 lM each primer, 0.8 mg/ml BSA, and Light- Cycler FastStart DNA MasterPLUS SYBR Green I kit (Roche) according to the manufacturer’s instructions. Thermal-cycling conditions were preincubated for 15 min at 958C, 42 cycles of 958C for 10 s, 608C for 3 s, and 728C for 30 s, and a melting curve cycle (10 s at 958C, 60 s at 658C, and temperature increase from 65 to 958C with a temperature transition rate of 0.18C/s). SYBRgreen uptake in double-stranded DNA was measured on a LightCycler 2.0 system (Roche). Product specificity was controlled using the melting-curve analysis of the Roche LightCycler software 4.05 (Roche). Ten-fold serial dilutions (from 5 3 105 to 50 copies) of purified 440-bp PCR product quantified by NanoPhotometer (IMPLEN) were included in each experiment to generate standard curve. At least two ‘‘no-template-control’’ were included with each experiment. The real-time PCR experiment was performed in at least duplicate. Analysis of the data was performed using LightCycler software 4.05 to generate individual standard curves from each experiment and to calculate the DNA amount from each unknown sample. Standard curves showing the correlation coefficient of the trendline higher than 0.95 were used.
Y-SNP analysis and autosomal STR analysis
A set of 16 biallelic markers for Mongolia were studied based on the global Y haplogroup distribution and recently revised markers (Table 3) (Jobling and Tyler- Smith, 2003; Karafet et al., 2008). All the known primers except for one were redesigned for amplicons between 100 and 230 bp. The target markers were amplified in monoplex or multiplex reactions. The markers in multiplex reactions were RPS4Y711, M174 and M231 for multiplex I, M207, M304, and M242 for multiplex II, and M38, M217, snd M210 for multiplex III. PCR was performed in a 10-ll final volume composed of 13 PCR Gold Buffer (ABI), 2 mM MgCl2, 0.2 mM dNTP, 1 mg/ml BSA (NEB), 0.4 U AmpliTaq Gold DNA polymerase (ABI), 0.75 lM primer, and 4 ll aDNA template. Thermal cycling consisted of a first denaturation step at 958C for 11 min followed by 45 cycles of 948C for 30 s, 608C for 1 min, and 728C for 1 min with a final extension at 728C for 7 min. Each marker was amplified again in a nested PCR reaction under the same cycling condition as above but with a 1:50 dilution of the first PCR product and 25 cycles. Amplicons were sequenced in forward or reverse direction. Y haplogroup was determined with recently revised Y haplogroup tree (Karafet et al., 2008).
For the autosomal STR analysis, triplicate amplifications were performed with AmpFlSTR1 MinifilerTM PCR Amplification Kit (ABI). PCR was performed according to manufacturer’s protocol but with 32 amplification cycles. PCR products were analyzed with the ABI Prism 310 automatic sequencer (ABI). More than six repeats were done to identify reproducibility (Schmerer et al., 1999). Genetic relationships and ethnic affiliations were investigated with the DNAVIEW software.
based on the concatenated partial HV1 and HV2 control region mtDNA sequences.
The classical-stringent standards for the authentication of aDNA were followed. mtDNA quantification of aDNA extracts determined by duplicate real-time PCR revealed that two MNX3 extracts carried on average 2,550–5,500 and 92,500–122,000 copies per gram of bone powders for 440 and 221-bp mtDNA HV1 fragment, respectively (Table S6 and Fig. S3); two MNX4 extracts had on average 1,250-1,522 and 34,750–35,750 copies per gram of bone powders for 440 and 221-bp mtDNA HV1 fragment, respectively. A horse mtDNA fragment was amplified from an associated horse remain only with the primers for horse mtDNA amplification but was not with the primers for human mtDNA (Fig. S4). These results strongly demonstrate that the potential human DNA contamination has been successfully resolved in this study, and also the genuine target DNA was successfully obtained. The cloning of amplified ancient mtDNA HV1 fragments from each of the MNX3 (conditionally) West Eurasian male and MNX4 female samples showed no significant sequence variation among the 10 independent clones of each sample except poly-C heteroplasmy compared to amplicon-derived sequences from the MNX3 (conditionally) West Eurasian male sample (Table S7); this kind of heteroplasmy does not affect the haplogroup determination in this study and has been reported to be observed normally and frequently (Santos et al., 2008). None of the samples shared the same mtDNA or STR genotype with any other sample or with laboratory workers and archeologists (Table S8). We tested two or more separate extracts from one or two different bones. Extraction and template blanks were included in every PCR procedure and no positive was detectable. aDNA samples were reproducible in all the analyses. Results were consistent between haplogroup and subhaplogroup markers in mtDNA and YSNP analyses. There were no unreported, ambiguous, or heteroplasmic variations in any of the sample data. Independent replication for the mtDNA, Y-SNP and autosomal STR analysis of the ancient samples, MNX3 Eurasian male, and MNX4 female was all successfully done in a second laboratory (Lee’s Laboratory, Department of Life Science, College of Natural Science, Chung-Ang University). These findings are in favor of the authenticity of the amplified products. The second-laboratory replication of MNX2 male sample demonstrated only a small fragment of mtDNA HV1 (221 bp). PCR amplifications for the large mtDNA fragments (440 and 378 bp), Y-SNP, and autosomal STR analysis were failed. The partial HV1 sequencing data obtained from the second laboratory were consistent with those from the two independent extracts from the first laboratory. It is unlikely that there were potential contamination problems in the second laboratory. We presume that the replication could not be completed due to the lack of available aDNA extracts. The genetic data of MNX2 male should be interpreted carefully with respect to authenticity.
MtDNA sequences of HV1 (NP 15,991–16,390 or NP 15,997–16,380), HV2 (NP 35–369 or NP 47–361), and coding regions, NP 12,266–12,434, NP 15,831–15,966, NP 2,956–3,217, and NP 5,080–5,305, were reproducibly obtained and are available at GenBank (accession numbers: GQ145583–GQ145594). Compared to rCRS, MNX3 (conditionally) West Eurasian male had six (HV1) and five (HV2) variable NPs; MNX4 female had two (HV1) and four (HV2) variable NPs; MNX2 male had three (HV1) and four (HV2) variable NPs (Table 4). The HV1 sequences of the three ancient samples were different from one another. The HV2 sequences of MNX 2 male and MNX4 female were identical. The HV1-based Haplogroup Prediction Tool determined MNX3 male as haplogroup U, and MNX 2 male and MNX4 female as D. mtDNAmanager determined MNX3 (conditionally) West Eurasian male as haplogroup U2e, MNX2 male and MNX4 female as D4/G. Coding gene motifs confirmed the predicted haplogroups. MNX3 (conditionally) West Eurasian male carried haplogroup U coding gene mutations A ? G transition at NP 12,308 and G ? A transition at NP 12,372, and U2e coding gene mutation A ? G transition at NP 15,907. MNX3 carried U2e1 defining mutations T ? C transition at NP 217 on HV2 (Table 4) (Palanichamy et al., 2004; Achilli et al., 2005; van Oven and Kayser, 2009). MNX3 (conditionally) West Eurasian male determined with a NJ phylogenetic tree was a close neighbor to U2e1 (Figs. 2 and S2).
The most similar sequences to MNX3 (conditionally) West Eurasian male were a Hutterite origin DNA (GenBank No. FJ348188.1) with 1-bp differences from each of HV1 and HV2 of MNX3 and a Finnish origin DNA (AY339545.1) with 2 bp and no difference at HV1 and HV2, respectively (Tables S2 and S5). The origins of sequences that showed 3-bp differences in HV1 and HV2 from MNX3 (conditionally) West Eurasian male sequence were Spain, Hungary, Italy, and two unidentified ones. One sequence (Serial No. 3,890) of unknown origin among the 50,033 HV1 sequences in the Genographic project database was identical to the MNX3 (conditionally) West Eurasian male HV1 sequence. A search of 19,626 haplotypes from a global mtDNA control region database from the Armed Forces DNA Identification Laboratory found no exact matches. Two samples differed at one and two polymorphisms: a sample from UAE lacked the 16093C variant, and a sample from Afghanistan matched the MNX3 (conditionally) West Eurasian male and contained the 309.1C and 309.2C length polymorphisms from HV2 (Table S5).
MNX2 male and MNX4 female carried haplogroup D coding gene mutation C ? A transversion at NP 5178 and D4 coding gene mutation G ? A transition at NP 3010 (Table 4) (Kong et al., 2006; van Oven and Kayser, 2009). NJ tree analysis showed that these two ancient samples were most closely related to D4 people primarily composed of East Asians (Figs. 2 and S2 and Tables S3 and S4). Twenty-nine haplotypes were identical to MNX4 female, but none was identical to MNX2 male (Figs. 2 and S2 and Table S5).
Haplogroup Prediction Tool and mtDNAmanager correctly and consistently determined haplogroup U samples with very few incorrect assignments (Table S2). However, Haplogroup Prediction Tool resulted in incorrect determinations for haplogroup D (Tables S3 and S4). These results show that haplogroup determination with HV1 sequence variation (Haplogroup Prediction Tool) has a limitation and mtDNAmanager is preferable for Asian mtDNA and indicate that haplogroup determination, based only on control region variations, has a potential risk of haplogroup misdiagnosis; the examination of diagnostic coding region mutations should be considered for the reliable and precise haplogroup determination.
We analyzed a set of 16 Y-SNP biallelic markers for the two ancient male-determined samples. MNX3 (conditionally) West Eurasian male was initially determined as haplogroup R (M207). Subsequent analyses revealed MNX3 (conditionally) West Eurasian male as R1 (M173), R1a (SRY10831.2), and R1a1 (M17) (Fig. S5). The MNX2 male haplogroup was determined as C (RPS4Y711) and C3 (M217) (Fig. S6).
Autosomal STR analysis
Autosomal STR typing showed complete allelic profiles (Table 5). Kinship analysis with DNAVIEW excluded a close parentage relationship among three ancient subjects. Comparing MNX2 male with MNX3 (conditionally) West Eurasian male and MNX4 female shows the affirmative evidence that they are not closely related. A likelihood ratio (LR [ 10) supports unrelated versus half-sibling for MNX2 male versus either of the others. In this respect, half-sib is a reasonably representative weak relationship; it is the same calculation as for uncle/nephew or grandparent/ grandchild. Furthermore, the evidence supporting unrelated versus closely related (sibling or parent/child) is very strong (LR[500). As between MNX3 (conditionally) West Eurasian male and MNX4 female, autosomal DNA gives strong evidence that they are not immediate relatives, but only weak evidence against a half-sibling (uncle etc.) relationship and no evidence either way as regards being cousins. Autosomal DNA can also be used as evidence of population origin using the principle that the likelihood is greater for a population in which the alleles of a profile are more common (Brenner, 2006). Calculation by DNA VIEW shows that the autosomal profile of MNX3 (conditionally) West Eurasian male is 14 times more probable from a Brahmin Indian than from a modern Caucasian (Table 6).
calculated by DNA VIEW software with the autosomal STR data of Mongolian ancient samples
The MNX3 (conditionally) West Eurasian male has R1a1 of Y-SNP and U2e1 mtDNA. R1a is the most common haplogroup in Europe (Malyarchuk et al., 2004; Kayser et al., 2005; Wetton et al., 2005; Fechner et al., 2008; Volgyi et al., 2009). It shows decreasing frequencies from North to South Europe (Wells, 2007) and from Central toward South Asia (Wells et al., 2001). R1a1 in Nepal (Gayden et al., 2007) and India (Sahoo and Kashyap, 2006; Thanseem et al., 2006; Sharma et al., 2009) has been suggested to be associated with migration of Indo-European people from Central Asia (Cordaux et al., 2004). The mtDNA haplogroup U2e has been found in most Central Asian populations (Comas et al., 2004). This haplogroup shows high frequencies in Turkmenistan, Tajikistan, and Kalash in Pakistan (Table 7) (Quintana-Murci et al., 2004). U2e is present in West Eurasia at 1% on average (Richard et al., 2007). The ancestral mtDNA haplogroup U2 is subdivided into U2e and U2i; U2e is known as a characteristic European haplogroup, and U2i as an indigenous haplogroup in India (Kivisild et al., 1999; Bermisheva et al., 2002; Basu et al., 2003; Quintana- Murci et al., 2004; Maji et al., 2008; Malyarchuk et al., 2008). U2e is also found in India but exclusively in caste populations, especially in the upper caste with a high frequency (Table 7) (Basu et al., 2003; Maji et al., 2008).
Kurgan people were nomadic peoples of the Volga steppe region, infiltrating Europe between the middle of the fifth and the second millennium BC (Kurgan people continued “infiltration” in the first millennium BC under the names Scythians ans Sarnats, in the first millennium AD under the names Alans, Huns, Avars, Bulgars, Türks, Khazars, and so on). Kurgan people expansion would have resulted in the spread of the Indo- European language (Gimbutas, 1970) (Linguistic association of Gimbutas Kurgan people with Indo- European languages is, in the author's lingo “identification of the two groups is not certain”). The Kurgan culture is divided into different subcultures and thought to succeed to the following cultures (Gimbutas, 1970; Hemphill and Mallory, 2004). Historical records and archaeology attest that Kurgan nomadic groups moved across Eurasia from North of the Black sea through Central and Inner Asia, to northeast Asia in a matter of centuries (Mair, 2005). Carriers of the Kurgan culture, believed (believed as matter of faith) to be Indo-European speakers, were also carriers of the R1a1 haplogroup (Keyser-Tracqui et al., 2009). R1a1 has thus been considered a marker of Indo-European contribution (Zerjal et al., 1999; Kharkov et al., 2004) (Indo-European is a linguistic category, while Zerjal and Kharkov et al. are biologists). R1a was found in Eulau, Germany of the Corded Ware Culture (Haak et al., 2008). R1a1 was predominant in the Krasnoyarsk area in southern central Siberia with the Andronovo, the Karasuk, the Tagar, and the Tachtyk cultures (Keyser-Tracqui et al., 2009).
Our research shows that the MNX3 ancient human sample was not that of an East Asian. There are previous reports of ancient mtDNA haplogroup U; U5 in China 1,400 years ago and U2 in Mongolia 2,000 years ago (Keyser-Tracqui et al., 2003; Xie et al., 2007). mtDNA haplogrouping of Egyin Gol only with HV1 was an incomplete analysis. Although nearly 11% of ancient skeletons in Egyin Gol
MNX2 male and MNX4 female belonged to D4 mtDNA. D is common in Northeast and Central Asia (Yao et al., 2002; Comas et al., 2004; Gokcumen et al., 2008). D4 is more common in Turkmenistan (Quintana-Murci et al., 2004) and Northeast Asia (Yao et al., 2002; Maruyama et al., 2003; Cheng et al., 2008; Jin et al., 2009) ( Northeast Asian Tungus people - Koreans, Japanese, Manju, and Aleuts and Amerindians).
MNX2 male showed C3 of Y-SNP. C is one of major groups in Asia. The C frequency is higher in Northeast Asia (Jin et al., 2003; Yoshida and Kubo, 2008). Buryat shows the highest frequency of C3 (Derenko et al., 2007). The C3 frequency is also high in Mongolia and Evenks (Derenko et al., 2007).
Different mtDNA and Y-SNP haplogroup and sequences excluded immediate kinship among the three aDNA samples. Although the both belong to D4, MNX4 female is different at one site (NP 16,311) from MNX2 male. NP 16,311 is considered as a mutational hotspot (Malyarchuk and Rogozin, 2004). Autosomal STR analysis with DNAVIEW supports unrelatedness among the three subjects. The kit for degraded DNA has five loci in common with the AmpFl/STR profiler Plus Kit: D13S317, D7S820, D21S11, D18S51, FGA, and amelogenin. No matched case was found between our data and data from ancient Kurgan sites, Egyin Gol, and modern Mongolians (Keyser-Tracqui et al., 2006, 2009).
In conclusion, paternal, maternal, and biparental genetic analyses were done on three
We gratefully acknowledge Ariunaa Togloom, and Sung-Hoon Kang for their technical assistance, Qing-Peng Kong for helpful advice in some East Asian mtDNA haplogroup assignments, and Peter Underhill for valuable discussion. This work was supported by National Research Foundation of Korea Grant funded by the Korean Government (2009-0073831), and by a grant from the National Museum of Korea. We also thank Mike Coble for his generous mtDNA haplotype analysis using AFDIL mtDNA database supported by NIJ grant No. 2005-DN-R-086.
URLs GenBank database: http://www.ncbi.nlm.nih.gov/ Genbank/ Genographic project database: http://www.nationalgeographic. com/genographic Haplogroup Predection Tool: http://nnhgtool.nationalgeographic. com/classify MtDNAmanager: http://mtmanager.yonsei.ac.kr DNAVIEW software: http://www.dna-view.com/dnaview. htm Clustal W 2.0.11: ftp://ftp.ebi.ac.uk/pub/software/clustalw2/ 2.0.11/ MEGA software version 4.1: http://www.megasoftware. net/mega41.html
Achilli A, Rengo C, Battaglia V, Pala M, Olivieri A, Fornarino S, Magri C, Scozzari R, Babudri N, Santachiara-Benerecetti AS, Bandelt HJ, Semino O, Torroni A. 2005. Saami and Berbers—an unexpected mitochondrial DNA link. Am J Hum Genet 76:883–886.
Gokcumen O, Dulik MC, Pai AA, Zhadanov SI, Rubinstein S, Osipova LP, Andreenkov OV, Tabikhanova LE, Gubina MA, Labuda D, Schurr TG. 2008. Genetic variation in the enigmatic Altaian Kazakhs of South-Central Russia: insights into Turkic population history. Am J Phys Anthropol 136:278–293.
Richard C, Pennarun E, Kivisild T, Tambets K, Tolk HV, Metspalu E, Reidla M, Chevalier S, Giraudet S, Lauc LB, Pericic M, Rudan P, Claustres M, Journel H, Dorval I, Muller C, Villems R, Chaventre A, Moisan JP. 2007. An mtDNA perspective of French genetic variation. Ann Hum Biol 34:68–79.
Professor Victor H. Mair comments (Bio articleVictor H. Mair)
Dr. Anatole A. Klyosov [PhD ( 1972), DSc (1977), Professor (1978), Professor of Biochemistry at Harvard Medical School (1990)] sent his comments privately to the local co-author of the above K.Kim et al. article, Prof. Victor H. Mair, Professor of Philology (Chinese Language and Literature) at the University of Pennsylvania, Philadelphia, United States. Somehow Prof. V.H. Mair did not welcome the professional comments, and rejected any discussion on the subject. Instead, Prof. of Philology V.H. Mair offered Prof. of Biochemistry Dr. A.A. Klyosov to read his books. In response, Dr. A.A. Klyosov noted that although he had no lesser grounds for that kind of suggestions, he did not offer Prof. of Philology V.H. Mair to read his books,. That was the end of the professorial discussion, and Prof. Victor H. Mair comments. The discussion was followed by Dr. A.A. Klyosov's open publication of the comments posted above. Probably, such outcome is a blessing for everybody, since it provided us with penetrating graphical story on the Hunnic genetic trunks dispersed in time and space juxtaposed against the original informative testing, accompanied by tendentious, misleading, and shallowish composition.
|Scythian-Iranian theoryy||Ossetian Genetics||Scythian language - Sources||Etruscan Genetics|
Contents Türkic Genetics
Contents Amerin Genetics
Klyosov A. Türkic DNA genealogy
Klyosov A.The 3 R's in R1 Haplogroup
Ogur and Oguz
Alans and Ases
Eastern Hun Genetics 3 c. BC
Eastern Hun mtDNA Genetics 5 and 1cc. BC
Eastern Hun Y-DNA Genetics ca. 0 AD
Vikings - Scandinavian Huns
Tarim genetics, Uigurs, and IE exuberance
Modern Uigur Y-DNA