Early Iranian steppe nomadic pastoralists also show Y-DNA bottlenecks and R1b-L23

New paper (behind paywall) Ancient genomes suggest the eastern Pontic-Caspian steppe as the source of western Iron Age nomads, by Krzewińska et al. Science (2018) 4(10):eaat4457.

Interesting excerpts (emphasis mine, some links to images and tables deleted for clarity):

Late Bronze Age (LBA) Srubnaya-Alakulskaya individuals carried mtDNA haplogroups associated with Europeans or West Eurasians (17) including H, J1, K1, T2, U2, U4, and U5 (table S3). In contrast, the Iron Age nomads (Cimmerians, Scythians, and Sarmatians) additionally carried mtDNA haplogroups associated with Central Asia and the Far East (A, C, D, and M). The absence of East Asian mitochondrial lineages in the more eastern and older Srubnaya-Alakulskaya population suggests that the appearance of East Asian haplogroups in the steppe populations might be associated with the Iron Age nomads, starting with the Cimmerians.


#UPDATE (5 OCT 2018): Some Y-SNP calls have been published in a Molgen thread, with:

  • Srubna samples have possibly two R1a-Z280, three R1a-Z93.
  • Cimmerians may not have R1b: cim357 is reported as R1a.
  • Some Scythians have low coverage to the point where it is difficult to assign even a reliable haplogroup (they report hg I2 for scy301, or E for scy197, probably based on some shared SNPs?), but those which can be reliably assigned seem R1b-Z2103 [hence probably the use of question marks and asterisks in the table, and the assumption of the paper that all Scythians are R1b-L23]:
    • The most recent subclade is found in scy305: R1b-Z2103>Z2106 (Z2106+, Y12538/Z8131+)
    • scy304: R1b-Z2103 (M12149/Y4371/Z8128+).
    • scy009: R1b-P312>U152>L2 (P312+, U152?, L2+)?
  • Sarmatians are apparently all R1a-Z93 (including tem002 and tem003);
  • You can read here the Excel file with (some probably as speculative as the paper’s own) results.

    About the PCA

    1. Srubnaya-Alakulskaya individuals exhibited genetic affinity to northern and northeastern present-day Europeans, and these results were also consistent with outgroup f3 statistics.
    2. The Cimmerian individuals, representing the time period of transition from Bronze to Iron Age, were not homogeneous regarding their genetic similarities to present-day populations according to the PCA. F3 statistics confirmed the heterogeneity of these individuals in comparison with present-day populations
    3. The Scythians reported in this study, from the core Scythian territory in the North Pontic steppe, showed high intragroup diversity. In the PCA, they are positioned as four visually distinct groups compared to the gradient of present-day populations:
      1. A group of three individuals (scy009, scy010, and scy303) showed genetic affinity to north European populations (…).
      2. A group of four individuals (scy192, scy197, scy300, and scy305) showed genetic similarities to southern European populations (…).
      3. A group of three individuals (scy006, scy011, and scy193) located between the genetic variation of Mordovians and populations of the North Caucasus (…). In addition, one Srubnaya-Alakulskaya individual (kzb004), the most recent Cimmerian (cim357), and all Sarmatians fell within this cluster. In contrast to the Scythians, and despite being from opposite ends of the Pontic-Caspian steppe, the five Sarmatians grouped close together in this cluster.
      4. A group of three Scythians (scy301, scy304, and scy311) formed a discrete group between the SC and SE and had genetic affinities to present-day Bulgarian, Greek, Croatian, and Turkish populations (…).
      5. Finally, one individual from a Scythian cultural context (scy332) is positioned outside of the modern West Eurasian genetic variation (Fig. 1C) but shared genetic drift with East Asian populations.
    Radiocarbon ages and geographical locations of the ancient samples used in this study. Figure panels presented (Left) Bar plot visualizing approximate timeline of presented and previously published individuals. (Right) Principal component analysis (PCA) plot visualizing 35 Bronze Age and Iron Age individuals presented in this study and in published ancient individuals (table S5) in relation to modern reference panel from the Human Origins data set (41).


    The presence of an SA component (as well as finding of metals imported from Tien Shan Mountains in Muradym 8) could therefore reflect a connection to the complex networks of the nomadic transmigration patterns characteristic of seasonal steppe population movements. These movements, although dictated by the needs of the nomads and their animals, shaped the economic and social networks linking the outskirts of the steppe and facilitated the flow of goods between settled, semi-nomadic, and nomadic peoples. In contrast, all Cimmerians carried the Siberian genetic component. Both the PCA and f4 statistics supported their closer affinities to the Bronze Age western Siberian populations (including Karasuk) than to Srubnaya. It is noteworthy that the oldest of the Cimmerians studied here (cim357) carried almost equal proportions of Asian and West Eurasian components, resembling the Pazyryks, Aldy-Bel, and Iron Age individuals from Russia and Kazakhstan (12). The second oldest Cimmerian (cim358) was also the only one with both uniparental markers pointing toward East Asia. The Q1* Y chromosome sublineage of Q-M242 is widespread among Asians and Native Americans and is thought to have originated in the Altai Mountains (24)


    In contrast to the eastern steppe Scythians (Pazyryks and Aldy-Bel) that were closely related to Yamnaya, the western North Pontic Scythians were instead more closely related to individuals from Afanasievo and Andronovo groups. Some of the Scythians of the western Pontic-Caspian steppe lacked the SA and the East Eurasian components altogether and instead were more similar to a Montenegro Iron Age individual (3), possibly indicating assimilation of the earlier local groups by the Scythians.

    Toward the end of the Scythian period (fourth century CE), a possible direct influx from the southern Ural steppe zone took place, as indicated by scy332. However, it is possible that this individual might have originated in a different nomadic group despite being found in a Scythian cultural context.

    Genetic diversity and ancestral components of Srubnaya-Alakulskaya population.(here called “Srubnaya”): (Left) Mean f3 statistics for Srubnaya and other Bronze Age populations. Srubnaya group was color-coded the same as with PCA. (Right) Pairwise mismatch estimates for Bronze Age populations.


    I am surprised to find this new R1b-L23-based bottleneck in Eastern Iranian expansions so late, but admittedly – based on data from later times in the Pontic-Caspian steppe near the Caucasus – it was always a possibility. The fact that pockets of R1b-L23 lineages remained somehow ‘hidden’ in early Indo-Iranian communities was clear already since Narasimhan et al. (2018), as I predicted could happen, and is compatible with the limited archaeological data on Sintashta-Potapovka populations outside fortified settlements. I already said that Corded Ware was out of Indo-European migrations then, this further supports it.

    Even with all these data coming just from a north-west Pontic steppe region (west of the Dnieper), these ‘Cimmerians’ – or rather the ‘Proto-Scythian’ nomadic cultures appearing before ca. 800 BC in the Pontic-Caspian steppes – are shown to be probably formed by diverse peoples from Central Asia who brought about the first waves of Siberian ancestry (and Asian lineages) seen in the western steppes. You can read about a Cimmerian-related culture, Anonino, key for the evolution of Finno-Permic peoples.

    Also interesting about the Y-DNA bottleneck seen here is the rejection of the supposed continuous western expansions of R1a-Z645 subclades with steppe tribes since the Bronze Age, and thus a clearest link of the Hungarian Árpád dynasty (of R1a-Z2123 lineage) to either the early Srubna-related expansions or – much more likely – to the actual expansions of Hungarian tribes near the Urals in historic times.

    NOTE. I will add the information of this paper to the upcoming post on Ugric and Samoyedic expansions, and the late introduction of Siberian ancestry to these peoples.

    A few interesting lessons to be learned:

    • Remember the fantasy story about that supposed steppe nomadic pastoralist society sharing different Y-DNA lineages? You know, that Yamna culture expanding with R1b from Khvalynsk-Repin into the whole Pontic-Caspian steppes and beyond, developing R1b-dominated Afanasevo, Bell Beaker, and Poltavka, but suddenly appearing (in the middle of those expansions through the steppes) as a different culture, Corded Ware, to the north (in the east-central European forest zone) and dominated by R1a? Well, it hasn’t happened with any other steppe migration, so…maybe Proto-Indo-Europeans were that kind of especially friendly language-teaching neighbours?
    • Remember that ‘pure-R1a’ Indo-Slavonic society emerged from Sintashta ca. 2100 BC? (or even Graeco-Aryan??) Hmmmm… Another good fantasy story that didn’t happen; just like a central-east European Bronze Age Balto-Slavic R1a continuity didn’t happen, either. So, given that cultures from around Estonia are those showing the closest thing to R1a continuity in Europe until the Iron Age, I assume we have to get ready for the Gulf of Finland Balto-Slavic soon.
    • Remember that ‘pure-R1a’ expansion of Indo-Europeans based on the Tarim Basin samples? This paper means ipso facto an end to the Tarim Basin – Tocharian artificial controversy. The Pre-Tocharian expansion is represented by Afanasevo, and whether or not (Andronovo-related) groups of R1a-Z645 lineages replaced part or eventually all of its population before, during, or after the Tocharian expansion into the Tarim Basin, this does not change the origin of the language split and expansion from Yamna to Central Asia; just like this paper does not change the fact that these steppe groups were Proto-Iranian (Srubna) and Eastern Iranian (Scythian) speakers, regardless of their dominant haplogroup.
    • And, best of all, remember the Copenhagen group’s recent R1a-based “Indo-Germanic” dialect revival vs. the R1b-Tocharo-Italo-Celtic? Yep, they made that proposal, in 2018, based on the obvious Yamna—R1b-L23 association, and the desire to support Kristiansen’s model of Corded Ware – Indo-European expansion. Pepperidge Farm remembers. This new data on Early Iranians means another big NO to that imaginary R1a-based PIE society. But good try to go back to Gimbutas’ times, though.
    Olander’s (2018) tree of Indo-European languages. Presented at Languages and migrations in pre-historic Europe (7-12 Aug 2018)

    Do you smell that fresher air? It’s the Central and East European post-Communist populist and ethnonationalist bullshit (viz. pure blond R1a-based Pan-Nordicism / pro-Russian Pan-Slavism / Pan-Eurasianism, as well as Pan-Turanism and similar crap from the 19th century) going down the toilet with each new paper.

    #EDIT (5 OCT 2018): It seems I was too quick to rant about the consequences of the paper without taking into account the complexity of the data presented. Not the first time this impulsivity happens, I guess it depends on my mood and on the time I have to write a post on the specific work day…

    While the data on Srubna, Cimmerians, and Sarmatians shows clearer Y-DNA bottlenecks (of R1a-Z645 subclades) with the new data, the Scythian samples remain controversial, because of the many doubts about the haplogroups (although the most certain cases are R1b-Z2103), their actual date, and cultural attribution. However, I doubt they belong to other peoples, given the expansionist trends of steppe nomads before, during, and after Scythians (as shown in statistical analyses), so most likely they are Scythian or ‘Para-Scythian’ nomadic groups that probably came from the east, whether or not they incorporated Balkan populations. This is further supported by the remaining R1b-P312 and R1b-Z2103 populations in and around the modern Eurasian steppe region.

    Early Iron Age cultures of the Carpathian basin ca. 7-6th century BC, including steppe groups Basarabi and Scythians. Ďurkovič et al. (2018).

    You can find an interesting and detailed take on the data published (in Russian) at Vol-Vlad’s LiveJournal (you can read an automatic translation from Google). I think that post is maybe too detailed in debunking all information associated to the supposed Scythians – to the point where just a single sample seems to be an actual Scythian (?!) -, but is nevertheless interesting to read the potential pitfalls of the study.


    Mitogenomes suggest rapid expansion of domesticated horse before 3500 BC

    Open access Origin and spread of Thoroughbred racehorses inferred from complete mitochondrial genome sequences: Phylogenomic and Bayesian coalescent perspectives, by Yoon et al. PLOS One (2018).

    Abstract (emphasis mine)

    The Thoroughbred horse breed was developed primarily for racing, and has a significant contribution to the qualitative improvement of many other horse breeds. Despite the importance of Thoroughbred racehorses in historical, cultural, and economical viewpoints, there was no temporal and spatial dynamics of them using the mitogenome sequences. To explore this topic, the complete mitochondrial genome sequences of 14 Thoroughbreds and two Przewalski’s horses were determined. These sequences were analyzed together along with 151 previously published horse mitochondrial genomes from a range of breeds across the globe using a Bayesian coalescent approach as well as Bayesian inference and maximum likelihood methods. The racing horses were revealed to have multiple maternal origins and to be closely related to horses from one Asian, two Middle Eastern, and five European breeds. Thoroughbred horse breed was not directly related to the Przewalski’s horse which has been regarded as the closest taxon to the all domestic horses and the only true wild horse species left in the world. Our phylogenomic analyses also supported that there was no apparent correlation between geographic origin or breed and the evolution of global horses. The most recent common ancestor of the Thoroughbreds lived approximately 8,100–111,500 years ago, which was significantly younger than the most recent common ancestor of modern horses (0.7286 My). Bayesian skyline plot revealed that the population expansion of modern horses, including Thoroughbreds, occurred approximately 5,500–11,000 years ago, which coincide with the start of domestication. This is the first phylogenomic study on the Thoroughbred racehorse in association with its spatio-temporal dynamics. The database and genetic history information of Thoroughbred mitogenomes obtained from the present study provide useful information for future horse improvement projects, as well as for the study of horse genomics, conservation, and in association with its geographical distribution.

    Bayesian skyline plot (BSP) based on mitochondrial genome sequences from 167 modern horses.
    The dark line in the BSP represents the estimated effective population size through time. The green area represents the 95% highest posterior density confidence intervals for this estimate.

    Interesting excerpts:

    We carried out a Bayesian coalescent approach using extended mitochondrial genome sequences from 167 horses in order to further assess the timescale of horse domestication. Here, we first calculated the time of the most recent common ancestor of Thoroughbred horses. Our analysis revealed the age of the most recent common ancestor of the racing horse to be around 8,100–111,500 years old. This estimate is much younger than that of the most recent common ancestor of the global horses, which has been estimated at 0.7286 Mys old.

    Bayesian maximum clade credibility phylogenomic tree on the ground of the mitochondrial genome sequences of 167 modern horses.
    The data set (16,432 base pairs) was also analyzed phylogenetically using Bayesian inference (BI) and maximum likelihood (ML) methods which showed the same topologies. 95% Highest Posterior Density of node heights are shown by blue bars. Groups are marked by a “G”. Numbers at the nodes represent (left to right): posterior probabilities (≥0.80) for the BI tree and bootstrap values (≥70%) for the ML tree. The racing horses were revealed to have multiple maternal origins and to be closely related to horses from one Asian, two Middle Eastern, and five European breeds. Results of phylogenomic analyses also uncovered no apparent association between geographic origin or breed and heterogeneity of global horses. The most recent common ancestor of the Thoroughbreds lived approximately 8,100–111,500 years ago, which was significantly younger than the most recent common ancestor of modern horses (0.7286 My).

    On the domestication time of modern horses, there have been several publications derived from both archaeological [49–51] and molecular [11–12, 23, 48] evidences. D’Andrade [49] reported that the origin of domestic horses was around 4,000 years ago. Ludwig et al. [50] stated the domestication time to be about 5,000 years ago, while Anthony [51] noted that horse rearing by humans may have occurred approximately 6,000 years ago. Subsequently, on the basis of mitochondrial genome sequences, Lippold et al. [11] and Achilli et al. [12] postulated domestication time to be about 6,000–8,000 and 6,000–7,000 years ago, respectively. Warmuth [48] dated domestication time to 5,500 years ago based on autosomal genotype data, while Orlando et al. [23] claimed that Przewalski’s and domestic horse populations diverged 38,000–72,000 years ago based on analysis of genome sequences. In contrast to the previous hypothesized date of horse domestication, the results of our Bayesian skyline plot (BSP) analysis depict a rapid expansion of the horse population approximately 5,500–11,000 years ago, which coincides with the start of domestication.

    It seems that we will not have an update on horse aDNA from the ISBA 8, so we will have to make do with this for the moment.


    Global demographic history inferred from mitogenomes

    Open access Global demographic history of human populations inferred from whole mitochondrial genomes, by Miller, Manica, and Amos, Royal Society Open Science (2018).

    Relevant excerpts (emphasis mine):


    The Phase 3 sequence data from 20 populations, comprising five populations for each of the four main geographical regions of Europe, East Asia, South Asia and Africa, were downloaded from the 1000 Genomes Project website (www.1000genomes.org/data, [8]), including whole mitochondrial genome data for 1999 individuals. We decided not to analyse populations from the Americas due to the region’s complex history of admixture [13,14].

    The European populations were as follows: Finnish sampled in Finland (FIN); European Caucasians resident in Utah, USA (CEU); British in England and Scotland (GBR); an Iberian population from Spain (IBS) and Toscani from Italy (TSI). Representing East Asia were the Han Chinese in Beijing (CHB); Southern Han Chinese (CHS); Dai Chinese from Xishuangbanna, China (CDX); Kinh population from Ho Chi Minh City, Vietnam (KHV) and Japanese from Tokyo (JPT). The South Asian populations were Punjabi Indians from Lahore, Pakistan (PJL); Gujarati Indians in Houston, USA (GIH) as well as Indian Telugu sampled in the UK (ITU); Bengali from Bangladesh (BEB) and Sri Lankan Tamil from the UK (STU). (…)


    We analysed our mtDNA data with the extended Bayesian skyline plot (EBSP) method, a Bayesian, non-parametric technique for inferring past population size fluctuations from genetic data. Building on the previous Bayesian skyline plot (BSP) approach, EBSP uses a piecewise-linear model and Markov chain Monte Carlo (MCMC) methods to reconstruct a populations’ demographic history [17] and is implemented in the software package BEAST v. 2.3.2 [11]. Alignments for each of the 20 populations were loaded separately into the Bayesian Evolutionary Analysis Utility tool (BEAUti v. 2.3.2) in NEXUS format.

    Relationship between profile similarity and genetic distance, measured as Fst. Comparisons between regions, circles, are colour-coded: black ¼ AFR-EA; yellow ¼ AFR-EUR; blue ¼ AFR-SA; orange ¼ EUR-EA; green ¼ EA-SA; red ¼ EUR-SA. Comparisons within regions, squares, are coded: peach ¼ EUR; pink ¼ EA; dark blue ¼ EA; light blue ¼ AFR. Profile similarity is calculated as inferred size difference summed over 20 evenly spaced intervals (see Material and methods).

    Regional demographic histories


    The five European profiles are presented in figure 2. The four southerly populations all show profiles with a stable size up to approximately 14 ka followed by a sudden, rapid increase that becomes progressively less steep towards the present. There is also a north-south trend, with confidence intervals becoming broader towards the north, particularly for the oldest time-points. The Finnish population profile appears rather different, but this is to be expected both because it is so far north and because previous studies have identified Finns as a strong genetic outlier in Europe [19–22].

    Inferred demographic histories of five European populations. Dotted line is the median estimate of Ne and the thin grey lines show the boundary of the 95% CPD interval. The x-axis represents time from the present in years and all plots are on the same scale. Map shows origins of sampled populations.

    South Asia:

    The five profiles for South Asia are shown in figure 3. All populations reveal a period of rapid growth approximately 45–40 ka which then slows. Near the present the two southerly populations, GIH and STU both show evidence of a decline. However, this may be due to these samples being drawn from populations no longer living on the subcontinent, with the downward trend capturing a bottleneck associated with moving to Europe/America, perhaps accentuated by the tendency for immigrant populations to group by region, religion and race [23].

    Inferred South Asian population demographic histories. Dotted line is the median Ne estimate and the thin grey lines show the boundary of the 95% CPD intervals. The x-axis represents time from the present in thousands of years and all plots are on the same scale. The map shows location of sampled populations.


    Mitogenomes from Avar nomadic elite show Inner Asian origin


    Inner Asian maternal genetic origin of the Avar period nomadic elite in the 7th century AD Carpathian Basin, by Csáky et al. bioRxiv (2018).

    Abstract (emphasis mine):

    After 568 AD the nomadic Avars settled in the Carpathian Basin and founded their empire, which was an important force in Central Europe until the beginning of the 9th century AD. The Avar elite was probably of Inner Asian origin; its identification with the Rourans (who ruled the region of today’s Mongolia and North China in the 4th-6th centuries AD) is widely accepted in the historical research.

    Here, we study the whole mitochondrial genomes of twenty-three 7th century and two 8th century AD individuals from a well-characterised Avar elite group of burials excavated in Hungary. Most of them were buried with high value prestige artefacts and their skulls showed Mongoloid morphological traits.

    The majority (64%) of the studied samples’ mitochondrial DNA variability belongs to Asian haplogroups (C, D, F, M, R, Y and Z). This Avar elite group shows affinities to several ancient and modern Inner Asian populations.

    The genetic results verify the historical thesis on the Inner Asian origin of the Avar elite, as not only a military retinue consisting of armed men, but an endogamous group of families migrated. This correlates well with records on historical nomadic societies where maternal lineages were as important as paternal descent.

    MDS with 23 ancient populations. The Multidimensional Scaling plot is based on linearised Slatkin FST values that were calculated based on whole mitochondrial sequences (stress value is 0.1581). The MDS plot shows the connection of the Avars (AVAR) to the Central-Asian populations of the Late Iron Age (C-ASIA_LIAge) and Medieval period (C-ASIA_Medieval) along coordinate 1 and coordinate 2, which is caused by non-significant genetic distances between these populations. The European ancient populations are situated on the left part of the plot, where the Iberian (IB_EBRAge), Central-European (C-EU_BRAge) and British (BRIT_BRAge) populations from Early Bronze Age and Bronze Age are clustered along coordinate 2, while the Neolithic populations from Germany (GER_Neo), Hungary (HUN_Neo), Near-East (TUR_ _Neo) and Baltic region (BALT_Neo) are located on the skirt of the plot along coordinate 1. The linearised Slatkin FST values, abbreviations and references are presented in Table S4.

    Interesting excerpts:

    The mitochondrial genome sequences can be assigned to a wide range of the Eurasian haplogroups with dominance of the Asian lineages, which represent 64% of the variability: four samples belong to Asian macrohaplogroup C (two C4a1a4, one C4a1a4a and one C4b6); five samples to macrohaplogroup D (one by one D4i2, D4j, D4j12, D4j5a, D5b1), and three individuals to F (two F1b1b and one F1b1f). Each haplogroup M7c1b2b, R2, Y1a1 and Z1a1 is represented by one individual. One further haplogroup, M7 (probably M7c1b2b), was detected (sample AC20); however, the poor quality of its sequence data (2.19x average coverage) did not allow further analysis of this sample.

    European lineages (occurring mainly among females) are represented by the following haplogroups: H (one H5a2 and one H8a1), one J1b1a1, three T1a (two T1a1 and one T1a1b), one U5a1 and one U5b1b (Table S1).

    We detected two identical F1b1f haplotypes (AC11 female and AC12 male) and two identical C4a1a4 haplotypes (AC13 and AC15 males) from the same cemetery of Kunszállás; these matches indicate the maternal kinship of these individuals. There is no chronological difference between the female and the male from Grave 30 and 32 (AC11 and AC12), but the two males buried in Grave 28 and 52 (AC13 and AC15) are not contemporaries; they lived at least 2-3 generations apart.

    Ward type clustering of 44 ancient populations. The Ward type clustering shows separation of Asian and European populations. The Avar elite group (AVAR) is situated on an Asian branch and clustered together with Central Asian populations from Late Iron Age (C-ASIA_LIAge) and Medieval period (C-ASIA_Medieval), furthermore with Xiongnu period population from Mongolia (MON_Xiongnu) and Scythians from the Altai region (E-EU_IAge_Scyth). P values are given in percent as red numbers on the dendogram, where red rectangles indicate clusters with significant p values. The abbreviations and references are presented in Table S2.

    The Avar period elite shows the lowest and non-significant genetic distances to ancient Central Asian populations dated to the Late Iron Age (Hunnic) and to the Medieval period, which is displayed on the ancient MDS plot (Fig. 4); these connections are also reflected on the haplogroup based Ward-type clustering tree (Fig. 3). Building of these large Central Asian sample pools is enabled by the small number of samples per cultural/ethnic group. Further mitogenomic data from Inner Asia are needed to specify the ancient genetic connections; however, genomic analyses are also set back by the state of archaeological research, i.e. the lack of human remains from the 4th-5th century Mongolia, which would be a particularly important region in the study of the Avar elite’s origin.

    The investigated elite group from the Avar period elite also shows low genetic distances and phylogenetic connections to several Central and Inner Asian modern populations. Our results indicate that the source population of the elite group of the Avar Qaganate might have existed in Inner Asia (region of today’s Mongolia and North China) and the studied stratum of the Avars moved from there westwards towards Europe. Further genetic connections of the Avars to modern populations living to East and North of Inner Asia (Yakuts, Buryats, Tungus) probably indicate common source populations.

    MDS with the 44 modern populations and the Avar elite group. The Multidimensional Scaling plot is displayed based on linearised Slatkin FST values calculated based on whole mitochondrial sequences (stress value is 0.0677). The MDS plot shows differentiation of European, Near-Eastern, Central- and East-Asian populations along coordinates 1 and 2. The Avar elite (AVAR) is located on the Asian part of plot and clustered with Uyghurs from Northwest-China (NW-CHIN_UYG) and Han Chinese (CHIN), as well as with Burusho and Hazara populations from the Central-Asian Highland (Pakistan). The linearised Slatkin FST values, abbreviations and references are presented in Table S5.

    Sadly, no Y-DNA is available from this paper, although haplogroups Q, C2, or R1b (xM269) are probably to be expected, given the reported mtDNA. A replacement of the male population with subsequent migrations is obvious from the current distribution of Y-DNA haplogroups in the Carpathian Basin.

    Hungarians and Corded Ware

    Ancient Hungarians are important to understand the evolution, not only of Ugric, but also of Finno-Ugric peoples and their origin, since they show a genetic picture before more recent population expansions, genetic drift, and bottlenecks in eastern Europe.

    By now it is evident that the migration of Magyar clans from their homeland in the Cis-Urals region (from the 4th century AD on) happened after the first waves of late and gradual expansion of N1c subclades among Finno-Ugric peoples, but before the bottlenecks seen in modern populations of eastern Europe.

    In Ob-Ugric peoples, from the scarce data found in Pimenoff et al. (2018), we can see how Siberian N subclades expanded further after the separation of Magyars, evidenced by the inverted proportion of haplogroups R1a and N in modern Khantys and Mansis compared to Hungarians, and the diversity of N subclades compared to modern Fennic peoples.

    Similarly to Hungarians, the situation of modern Estonians (where R1a and N subclades show approximately the same proportion, ca. 33%) is probably closer to Fennic peoples in Antiquity, not having undergone the latest strong founder effect evident in modern Finns after their expansion to the north.

    Hungarian expansion from the 4th to the 10th century AD.

    Modern Hungary

    This is data from recent papers, summed up in Wikipedia:

    • In Semino et al. (2001) they found among 45 Palóc from Budapest and northern Hungary: 60% R1a, 13% R1b, 11% I, 9% E, 2% G, 2% J2.
    • In Csányi et al. (2008) Among 100 Hungarian men, 90 of whom from the Great Hungarian Plain: 30% R1a, 15% R1b, 13% I2a1, 13% J2, 9% E1b1b1a, 8% I1, 3% G2, 3% J1, 3% I*, 1% E*, 1% F*, 1% K*. Among 97 Székelys, in Romania: 20% R1b, 19% R1a, 17% I1, 11% J2, 10% J1, 8% E1b1b1a, 5% I2a1, 5% G2, 3% P*, 1% E*, 1% N.
    • In Pamjav et al. (2011), among 230 samples expected to include 6-8% Gypsy peoples: 26% R1a, 20% I2a, 19% R1b, 7% I, 6% J2, 5% H, 5% G2a, 5% E1b1b1a1, 3% J1, <1% N, <1% R2.
    • In Pamjav et al. (2017), from the Bodrogköz population: R1a-M458 (20.4%), I2a1-P37 (19%), R1b-M343 (15%), R1a-Z280 (14.3%), E1b-M78 (10.2%), and N1c-Tat (6.2%).

    NOTE. The N1c-Tat found in Bodrogköz belongs to the N1c-VL29 subgroup, more frequent among Balto-Slavic peoples, which may suggest (yet again) an initial stage of the expansion of N subclades among Finno-Ugric peoples by the time of the Hungarian migration.

    This is the data from FTDNA group on Hungary (copied from a Wikipedia summary of 2017 data):

    • 26.1% R1a (15% Z280, 6.5% M458, 0.9% Z93=>S23201, 3.7% unknown)
    • 19.2% R1b (6% L11-P312/U106, 5.3% P312, 4.2% L23/Z2103, 3.7% U106)
    • 16.9% I2 (15.2% CTS10228, 1.4% M223, 0.5% L38)
    • 8.3% I1
    • 8.1% J2 (5.3% M410, 2.8% M102)
    • 6.9% E1b1b1 (6% V13, 0.3% V22, 0.3% M123, 0.3% M81)
    • 6.9% G2a
    • 3.2% N (1.4% Z9136, 0.5% M2019/VL67, 0.5% Y7310, 0.9% Z16981)- note: only unrelated males are sampled
    • 2.3% Q (1.2% YP789, 0.9% M346, 0.2% M242)
    • 0.9% T
    • 0.5% J1
    • 0.2% L
    • 0.2% C

    R1a-Z280 stands out in FDNA (which we have to assume has no geographic preference among modern Hungarians), while R1a-M458 is prevalent in the north, which probably points to its relationship with (at least West) Slavic populations.

    Ancient Hungarians

    We already knew that Hungarians show similarities with Srubna and Hunnic peoples, and this paper shows a good reason for the similarities with the Huns.

    Also, recent population movements in the region (before the Avars) probably increased the proportion of R1b-L23 and I1 subclades (related to Roman and Germanic peoples) as well as possibly R1a-Z283 (mainly M458, related to the expansion of Slavs). From Understanding 6th-century barbarian social organization and migration through paleogenomics, by Amorim et al. (2018):

    Y-chromosome haplogroup attribution for 37 medieval and 1 Bronze age individuals.

    NOTE. The sample SZ15, of haplogroup R1a1a1b1a3a (S200), belongs to the Germanic branch Z284, which has a completely different history with its integration into the Nordic Bronze Age community.

    Interesting is the Szólád Bronze Age sample of R1a1a1b2a2a (Z2123) subclade (ca. 2100-1700 BC), which is possibly the same haplogroup found in King Béla III [Z93+ (80.6%), Z2123+ (10.8%)]*. Nevertheless, Z2123 refers to an upper clade, found also in East Andronovo sites in Narasimhan et al. (2018), as well as in the modern population of the Tarim Basin.

    NOTE. For more on the analysis of probability of the actual subclade, see here.

    Bronze Age R1a-Z93 samples of central-east Europe – like the Balkans BA sample (ca. 1750-1625 BC) from Merichleri, of R1a1a1b2 subclade – correspond most likely to the expansion of Iranian-speaking peoples in the early 2nd millennium BC, probably to the westward expansion of the Srubna culture.

    The specific subclade of King Béla III, on the other hand, probably corresponds to the more recent expansion of Magyar tribes settled in the region during the 9th century AD, so the specific subclade must have separated from those found in central-east Europe and in Andronovo during the Corded Ware expansion.

    Modified image, from Underhill et al. (2015). Spatial frequency distributions of Z282 (green) and Z93 (blue) affiliated haplogroups. Notice the potential Finno-Ugric-associated distribution of Z282 (including M558, a Z280 subclade) according to ancient maps; the northern Eurasian finds of Z2125 (upper clade of Z2123); and the potential of M458 subclades representing a west-east expansion of Balto-Slavic as a western outgroup of an original Fenno-Ugric population, equivalent to Z284 in Scandinavia.

    The study by Csányi et al. (2008), where the Tat C allele was found in 2 of 4 ancient samples, showed thus a potential 50:50 relationship of N1c in ancient Magyars, which is striking given the modern 1-3% a mere 1,000 years later, without any relevant population movement in between. This result remains to be reproduced with the current technology.

    In fact, recent studies of ancient Magyars, from the 10th to the 12th century, have not shown any N1c sample, and have confirmed instead the ancient presence of R1a (two other samples, interred near Béla III), R1b (four samples), I2a (two samples) J1, and E1b, a mixed genetic picture which is more in line with what is expected.

    So the question that I recently posed about east Corded Ware groups remains open: were Proto-Ugric peoples mainly of R1a-Z282 or R1a-Z93 subclades? Without ancient DNA from Middle Dnieper, Fatyanovo, Afanasevo, and the succeeding cultures (like Netted Ware) in north-eastern Europe, it is difficult to say.

    It is very likely that they are going to show mainly a mixture of both R1a-Z282 and R1a-Z93 lineages, with later populations showing a higher proportion of R1a-Z280 subclades. Whether this mixture happened already during the Corded Ware period, or is the result of later developments, is still unknown. What is certain is that Hungarian N1a1a1a-L708 subclades belong to more recent additions of Siberian haplogroups to the Ugric stock, probably during the Iron Age, just centuries before the Magyar expansion.


    Y-DNA haplogroups of Tuvinian tribes show little effect of the Mongol expansion


    Open access Estimating the impact of the Mongol expansion upon the gene pool of Tuvans, by Balanovskaya et al., Vavilov Journal of genetics and breeding (2018), 22(5):611-619.

    Abstract (emphasis mine):

    With a view to trace the Mongol expansion in Tuvinian gene pool we studied two largest Tuvinian clans – those in which, according to data of humanities, one could expect the highest Central Asian ancestry, connected with the Mongol expansion. Thus, the results of Central Asian ancestry in these two clans component may be used as upper limit of the Mongol influence upon the Tuvinian gene pool in a whole. According to the data of 59 Y-chromosomal SNP markers, the haplogroup spectra in these Tuvinian tribal groups (Mongush, N = 64, and Oorzhak, N = 27) were similar. On average, two-thirds of their gene pools (63 %) are composed by North Eurasian haplogroups (N*, N1a2, N3a, Q) connected with autochtonous populations of modern area of Tuvans. The Central Asian haplogroups (C2, O2) composed less then fifth part (17 %) of gene pools of the clans studied. The opposite ratio was revealed in Mongols: there were 10 % North Eurasian haplogroups and 75 % Central Asian haplogroups in their gene pool. All the results derived – “genetic portraits”, the matrix of genetic distances, the dendrogram and the multidimensional scaling plot, which mirror the genetic connections between Tuvinian clans and populations of South Siberia and East Asia, demonstrated the prominent similarity of the Tuvinian gene pools with populations from and Khakassia and Altai. It could be therefore assumed that Tuvinian clans Mongush and Oorzhak originated from autochtonous people (supposedly, from the local Samoyed and Kets substrata). The minor component of Central Asian haplogroups in the gene pool of these clans allowed to suppose that Mongol expansion did not have a significant influence upon the Tuvinan gene pool at a whole.


    Interesting excerpts:

    Haplogroup C2 peaks in Central Asia (Wells et al., 2001; Zerial et al., 2003), though its variants are abundant in other peoples of Siberia and Far East. For instance, in one of Buryat clans, namely Ekhirids, hg C2 frequency is 88 % (Y-base); in Kazakhs from different regions of Kazakhstan, total occurrence of hg C2 variants averages between 17 and 81 % (Abilev et al., 2012; Zhabagin et al., 2013, 2014, 2017), in populations of the Amur River (such as Nanais, Negidals, Nivkhs, Ulchs) – between 40 and 65 %, in Evenks – up to 68 % (Y-base), in Kyrgyz people of Pamir-Alay – up to 22 %, correspondingly; of all Turkic peoples of Altai, relatively high hg C2 frequency (16 %) is detected only in Telengits (Balanovskaya et al., 2014; Balaganskaya et al., 2011a, 2016). In Tuvinian clans under the study, hg C2 frequency is rather low – 19 % in Mongush and 11 % in Oorzhak, while in Mongols it makes up almost two thirds of the entire gene pool an comprises different genetic lines (subhaplogroups).

    Y-chromosomal haplogroup spectra in gene pools of Tuvinian Oorzhak and Mongush clans and of the neighboring populations of South Siberia and Central Asia.

    Haplogroup N is abundant all over North Eurasia from Scandinavia to Far East (Rootsi et al., 2007). The study on whole Y-chromosome sequencing conducted with participation of our group (Ilumäe et al., 2016) subdivided this haplogroup into several branches with their regional distribution. In gene pools of the Tuvans involved, hg N was represented by two sub-clades, namely N1a2 and N3a.

    Sub-clade N1a2 peaks in populations of West Siberia (in Nganasans, frequency is 92 %) and South Siberia (in Khakas 34 %, in Tofalars 25 %) (Y-base). In Tuvans, N1a2 occurrence is nearly 16 % in Mongush and 15 % in Oorzhak clans, respectively, while in Mongols, the frequency is three times less (5 %). Hg N1a2 is supposed to display the impact of the Samoyedic component to the gene pool of Tuvinian clans (Kharkov et al., 2013).

    Sub-clade N3a is major in the Oorzhak clan comprising almost half of the gene pool (45 %); it is represented by two sub-clades, namely N3a* and N3a5. The same sub-branches are specific to the Mongush clan as well, though with lower frequencies: N3a* – 9 % and N3a5 – 14 % (see Table). In Khori-Buryats from the Transbaikal region, a high frequency is observed – 82 % (Kharkov et al., 2014), while in Mongols, N3a5 occurs rather rarely (6 %). Hg N3a* was detected in populations of South Siberia only, and was widely spread in Khakas-Sagays and Shors (up to 40 %) (Ilumäe et al., 2016) (Y-base).

    Map of distribution of Samoyedic languages (red) in the XVII century (approximate; hatching) and in the end of XX century (continuous background). Modified from Wikipedia, with the Tuva region labelled.

    Within the pan-Eurasian haplogroup R1a1a, two large genetic lines (sub-haplogroups) are identified: “European” (marker M458) and “Asian” (marker Z93) the latter almost never occurring in Europe (Balanovsky, 2015) but abundant in South Siberia and northern Hindustan. In the Altai-Sayan region, high frequencies of the “Asian” branch are spread in many peoples – Shors, Tubalars, Altai-Kizhi people, Telengits, Sagays, Kyzyl Khakas, Koibals, Teleuts (Y-base) (Kharkov et al., 2009). Hg R1a1a comprises perceptible parts of gene pools of Tuvinian clans (19 % in Mongush, and 15 % in Oorzhak), though its occurrence in Mongols is much lower (6 %). Those results also count in favor of the hypothesis of autochtonous component dominance even in the gene pools of clans potentially most influenced by Mongolian ancestry. If we add R1a1a variants to the “North Eurasian” haplogroups, the “not-Central Asian” component will compose average four fifth of the entire gene pools for Tuvinian clans (in Mongush 77 %, and in Oorzhak 81 %), being only 16 % in Mongols. Such data are definitely contrary to the hypothesis of a crucial influence of the Mongol expansion upon the development of Tuvinian gene pool.

    I found interesting the high proportion of R1a-Z93 subclades among Sagays in Khakhasia, which stem from a local Samoyed substratum, as described by the paper…

    Featured Image: Map of Uralic and Altaic languages, from Wikipedia.


    Origin of horse domestication likely on the North Caspian steppes

    Open access Late Quaternary horses in Eurasia in the face of climate and vegetation change, by Leonardi et al. Science Advances (2008) 4(7):eaar5589.

    Interesting excerpts (emphasis mine):

    Here, we compiled an extensive continental-scale database, consisting of 3070 radiocarbon dates associated to horse paleontological and archeological finds across the whole of Eurasia, that has been analyzed in association with coarse-scale paleoclimatic reconstructions. We further collected the number of identified specimens (NISP) frequency data for horses versus other ungulates in 1120 archeological layers in Europe (…) This ma.ssive amount of data allowed us to track,with unprecedented details, how the geographic distribution of the species changed through time

    Geographic range through time

    For most analyses, the data have been divided into climatic periods: pre-LGM(older than 27 ka B.P.), LGM(27 to 18 ka B.P.), Late Glacial (18 to 11.7 ka B.P.), Preboreal (11.7 to 10.6 ka B.P.), Boreal (10.6 to 9.1 ka B.P.), Early Atlantic (9.1 to 7.5 ka B.P.), Late Atlantic (7.5 to 5.5 ka B.P.), and Recent (younger than 5.5 ka B.P.) (Fig. 1, A and B). The spatial and temporal distribution of horse remains compiled in our database reveals a strong imbalance in Eurasia (Fig. 1, A and B).

    We found a common trend in both regions for a high number of occurrences at the end of the Pleistocene (with a decrease during the LGM, only visible in Europe), followed by a drastic reduction in the Early and Middle Holocene, and a relative increase toward more recent times. These included both the Early Atlantic in Europe, which started ~9.1 ka B.P., and the time range after 5.5 ka B.P. for Asia. The horse fossil record appears ubiquitous throughout Europe in the Late Pleistocene, while in the Early and Middle Holocene the finds are concentrated in central-western Europe and Iberia. From 7.5 ka B.P., the number of finds increases markedly, and the geographical distribution extends toward the east and southeast.

    Horse occurrences through time. (A) Horse occurrences through time. Histograms showing the number of horse observations in Europe (left panel) and Asia (right panel) for each time bin (top) and for climatic period (bottom). Only time bins with more than 10 observations (black horizontal line) have been considered for the SDM analyses. From 22 ka B.P. backward (gray vertical line), time bins cover 2 ka following the available paleoclimatic reconstructions. The central map shows the boundaries considered while defining European and Asian regions, with the black line representing the Urals. The zoomed area shows the geographical resolution of the climatic reconstructions, with each pixel representing a grid cell. (B) Geographic distribution of horse occurrences. Maps showing horse occurrences for each climatic period in Europe (left) and Asia (right).

    Different Asian and European niches

    This analysis revealed that, in both continents, horses occupied only a portion of the climatic space available. The range covered by random locations shows that the paleoecological conditions present in Europe were only a subset of those found in Asia. However, European horses occupied a much wider climatic space than in Asia, with only limited overlap between the two ranges.

    Horses conquered temperate environments from a European source

    There is no evidence of climatic barriers between those two populations through time because the forecasts from Europe and Asia always overlap in central Eurasia, except 5 ka B.P. (figs. S3 and S4). An alternative explanation is the role of the Urals as a potential constraint for the dispersal of horses between Europe and north central Asia.

    Climatic suitability. (A) Cumulative climatic suitability for the past 44 ka based on simulation on the European (left), Eurasian (middle), and Asian (right) data sets. To correct for sampling bias in the Eurasian data set, for each time slice, all estimates and projections for Eurasia are performed considering 100 random resampling of European occurrences in the same number as Asian occurrences. The darker the colors, themore stable the climatic suitability for horses (climatic niche = p-Hor) through time. (B) Projection of climatic suitability across Eurasia in different climatic periods based on occurrences in Europe (left), Eurasia (middle), and Asia (right). Because of the scarcity of data available for Asia, no models for the Holocene have been possible for both Asia and Eurasia, with the exception of 5 and 3 ka B.P. (both included in the “Recent” period).

    Climatic and habitat association patterns for horses in Europe support increasing habitat fragmentation

    The decrease of horse remains in Europe is not characterized by a geographic reduction in the overall extent of the area occupied by the species but in a drop of frequencies in a geographic extent that does not vary much between the Late Glacial and the Early Atlantic (Figs. 1B and 4B). This pattern is more likely to result from habitat fragmentation than from a geographic shift in the climatic range suitable for the species, as observed for many animals during the LGM (23).

    In the whole period ranging from the Preboreal (11.7 to 10.6 ka B.P.) to the Late Atlantic (7.5 to 5.5 ka B.P.), the total amount of land space most and likely suitable to horses is wider than in the Late Glacial, and only between 8 to 7 ka ago the European range appears patchy and fragmented (Fig. 4C). When comparing each of four successive time bins during the Holocene (8, 7, 6, and 5 ka B.P., respectively) (Fig. 4E), the difference in successive p-Hor values in Europe shows that the suitability for the species in Iberia, northeastern France, Italy, the Balkans, and eastern Europe steadily increased, while in Central Europe strong differences can be observed between neighboring regions.

    Analyses of the European data set and biomefrequency. (A) Distribution through time of the frequency of horse remains in Europe calculated as NISP of horses versus other ungulates. (B) Density of horse remains through time in Europe, calculated as NISP of horses versus other ungulates. The numbers at the bottom of each bar represent the number of observations falling in each class, from 0 to >5%. (C) Climatic suitability for horses in Europe between 10 and 3 ka B.P. (D) Climatic suitability per time period. Percentage of land cells in Europe with a value of suitability for horses (p-Hor) > 0.5 and p-Hor > 0.8. (E) Holocene climatic amelioration. Difference in p-Hor in Europe comparing five successive time bins during the Holocene: 9, 8, 7, 6, and 5 ka B.P. Eachmap shows the difference in themore recent distribution compared to the previous one. (F) Environmental reconstructions in themacro area surrounding horse finds in Europe (left) and Asia (right) per climatic period. The lighter the color, the less forested is the region. The numbers at the bottom of the bars show the number of occurrences in closed environments over all the observations. The dotted line represents a frequency of 0.5.

    Taken at face value, this pattern would suggest that horses were not restricted to open environments but could equally well inhabit closed, forested environments, as previously suggested (18). However, as others recently emphasized (19), the faunal associations inHolocene sites from Europe suggest a different pattern. The PCAs based on faunal assemblages (figs. S1 and S2) separate on the second principal component sites characterized by ungulates associated to forested areas (red deer, wild boar, and roe deer) and all other animals, associated to semi-open and open environments, including horses for most records.

    Together, the contrast between the reconstructed microscale and macroscale vegetable coverage in Europe, the increase of horses in mainly forested macroregions, and the spatial pattern of extinction suggest that, from the beginning of the Holocene, the suitable environment became more and more patchy, with open areas increasingly fragmented by forests, where wild populations of horses could have survived in isolation until one or several waves of arrivals of domestic horses, leading to either local admixture or a full replacement of the preexisting local populations.


    Our data show that, up to 5.5 ka ago, horse finds do not show association with species characteristic of forested areas such as wild boar and roe deer. We infer that the open and semi-open habitats occupied by horses on a narrow geographic scale appear less and less frequent at a macroenvironmental scale, supporting the possibility of increasing fragmentation of open habitats. This event is also likely to have led to an intensification of genetic isolation for the remaining horse populations, a pattern that still needs to be tested on genomic data.

    The suitability of both Iberia and eastern Europe appears constant throughout the entire post-LGM period, in line with these regions being hotspots of genetic diversity and, possibly, the refugia sources for the recolonization of the continent (11). While the Pontic-Caspian region appears not suitable for European horses around the time when horses where first domesticated some 5.5 ka ago (6), part of this region appears suitable for the Asian horses (with the Caspian Sea as the westernmost boundary). This may suggest that horse domestication started from a population background related to an Asian ancestry and that the further spread of the domesticated horses in Europe involved either adaptation to novel niches (possibly through selective breeding) or the application of domestication techniques to local horse populations pre-adapted to these environmental conditions. Testing this scenario will require mapping the genetic structure of the Eurasian horse population within the fifth to third millennium BCE.

    Some remarks

    Cultural-anthropological research and archaeological remains (see here), genetics (see here and here), and now also thorough palaeoclimatic and archaeological models point to the North Caspian region, settled by the Khvalynsk culture, as the most likely earliest origin of horse domestication. The paper also supports the favorable conditions of western Europe up to Iberia for the introduction of a horse-riding culture.

    I intended to write a post about the myth of Corded Ware horse riders, but for the moment I haven’t found the time. Not that Corded Ware pastoralists didn’t have horses, or could not ride them: they were a highly mobile culture of pastoralists stemming from eastern Poland / western Ukraine, so they must have known horses, like many other European cultures of the late 4th / early 3rd millennium influenced by expanding Yamna settlers. But it just cannot be said to have formed an essential part of their culture, as it was for Khvalynsk-Novodanilovka, and especially Yamna and later East Bell Beaker, Sintashta, etc.

    A mere look at these maps suffices to assess the limited role of the horse in north-eastern Europe, the only region where groups of late Corded Ware-derived cultures survived the expansion of Yamna, and especially East Bell Beakers after ca. 2500 BC, which transformed Western, Northern, and Central Europe, and even East Europe reaching the modern Baltic countries, Belarus, and Romania. Even Trzciniec was born out of the influence from expanding Bell Beakers into earlier Corded Ware territory, although the later (Iron Age) relevance of this culture was probably quite limited.

    As you can imagine, without horses and horse symbolism, horse riding, carts, and intensive cattle-breeding (associated with Yamna and the broad, east-central European grasslands typical of steppe regions), there can be no Proto-Indo-European, whose reconstructed vocabulary is particulary rich in horse-related words, and whose reconstructed culture, society, and religion cannot be understood without the domesticated horse. In forest regions to the north-east and eastern Europe, there was apparently little space for horses, but plenty of room for other ungulates and thus hunting, and indeed Uralic languages

    In the upcoming months we will see R1a-fans associating Proto-Indo-Europeans more and more with wool, and sheep, and corded ware, and forest regions, until the proposed homeland shifts to the Baltic and Finland, instead of dat boring horse-riding people of the steppes…No wait, it’s already happening.

    NOTE. Also open access is the recent Horse Y chromosome assembly displays unique evolutionary features and putative stallion fertility genes, by Janečka et al. Nature Communications (2018).


    Close inbreeding and low genetic diversity in Inner Asian human populations despite geographical exogamy


    Open access Close inbreeding and low genetic diversity in Inner Asian human populations despite geographical exogamy, by Marchi et al. Scientific Reports (2018) 8:9397.

    Abstract (emphasis mine):

    When closely related individuals mate, they produce inbred offspring, which often have lower fitness than outbred ones. Geographical exogamy, by favouring matings between distant individuals, is thought to be an inbreeding avoidance mechanism; however, no data has clearly tested this prediction. Here, we took advantage of the diversity of matrimonial systems in humans to explore the impact of geographical exogamy on genetic diversity and inbreeding. We collected ethno-demographic data for 1,344 individuals in 16 populations from two Inner Asian cultural groups with contrasting dispersal behaviours (Turko-Mongols and Indo-Iranians) and genotyped genome-wide single nucleotide polymorphisms in 503 individuals. We estimated the population exogamy rate and confirmed the expected dispersal differences: Turko-Mongols are geographically more exogamous than Indo-Iranians. Unexpectedly, across populations, exogamy patterns correlated neither with the proportion of inbred individuals nor with their genetic diversity. Even more surprisingly, among Turko-Mongols, descendants from exogamous couples were significantly more inbred than descendants from endogamous couples, except for large distances (>40 km). Overall, 37% of the descendants from exogamous couples were closely inbred. This suggests that in Inner Asia, geographical exogamy is neither efficient in increasing genetic diversity nor in avoiding inbreeding, which might be due to kinship endogamy despite the occurrence of dispersal.

    Interesting excerpts:

    Two cultural groups, which matrimonial systems are reported to differ, coexist in Inner Asia: Turko-Mongols are described as mainly exogamous while Indo-Iranians are thought to be mainly endogamous45. However, it is not always clear if exogamy refers to clan (ethnic) or village (geographical) exogamy. Here, we used a dataset of 16 populations representing 11 different ethnic groups from both cultural groups and we quantified geographical exogamy rates and distances in each population. Using an empirical threshold of 4 km, we confirmed that matrimonial behaviours differ as described in the literature, even though we found some exceptions: three Turko-Mongol populations (out of 14) have less than 50% exogamy, whereas one Indo-Iranian population (out of four) has more than 50% exogamy.(…).

    Geographical distances between the birth places of couples in Turko-Mongols and Indo-Iranians. The geographical distances are plotted in log scale (km). Their densities are represented by population (dashed lines) or for the Indo-Iranian and Turko-Mongol groups (solid lines). We represented the average distances within couples per population using a Kernel’s density estimate implemented in R with a smoothing bandwidth of 0.2. See Supplementary Table 1B for population codes.

    An additional important result of our study is that geographical distances are not negatively correlated with inbreeding, as could have been expected under an isolation-by-distance model65. Interestingly, a recent study based on a large genealogical dataset, collected across Western Europe and North America, and including birth places information, similarly found an absence of correlation between relatedness and the distance between couples, for the cohorts born before 185066. Our analyses within present-day Turko-Mongols reveal more specifically that the structure of the relationship between geographical distance and mating choice inbreeding is not linear, but rather tends to be bell-shaped, and thus cannot be correctly assessed with a single correlation test. Indeed, descendants from parents born 4 to 40 km apart are more inbred than descendants from endogamous couples (≤4 km) or from long-range exogamous ones (>40 km). As a consequence, close inbreeding exists despite geographical exogamy, and about a third of descendants from exogamous couples are inbred.

    These results, in addition to those obtained by [Kaplanis et al. 2018]66, highlight the importance of using geographic distances rather than exogamy rates to characterize the impact of exogamy on inbreeding, as already described when studying patrilocality67. Indeed, when we compare mating choice inbreeding patterns for descendants from exogamous and endogamous couples defined for thresholds of 4, 10, 20 and 30 km, we find no significant differences (for number and total length of class C-ROHs and F-Median coefficient: MWU test p-values > 0.1). We only detect significantly lower values in descendants from exogamous couples for larger distances above 40 and 50 km (p-values < 0.03).

    Genetic diversity (A) and inbreeding patterns (B,C) within populations. Grey lines in (B) represent inbreeding values corresponding to second-cousins and first-cousins. The grey line in (C) represents the homozygosity population baseline expected under panmixia. The number of samples per population is indicated between parentheses. See Supplementary Table 1B for population codes.

    Our results also challenge the intuition that exogamy necessarily increases the genetic diversity within a population and therefore reduces drift inbreeding. Indeed, we found that Turko-Mongol populations have a lower genetic diversity (as measured by the mean haplotypic heterozygosity) and more intermediate ROHs associated with drift inbreeding than those of Indo-Iranians despite higher exogamous rates. (…)

    Overall, this research sheds light on mating choice preferences: we showed that two thirds of partners that have not dispersed did mate with unrelated individuals, and that drift and mating choice inbreeding is variable, even among close-by populations. We also provide new insights into the relationship between dispersal and inbreeding in humans, based on genetic data, and demonstrate that geographical exogamy is not necessarily negatively associated with mating choice inbreeding, but rather can have a more complex non-linear relationship. Contrary to the common situation in many animals, this finding suggests that Inner Asian human populations who practise exogamy at small geographical scales might be focused on alliance strategies that result in kinship endogamy. (…)


    Yamna/Afanasevo elite males dominated by R1b-L23, Okunevo brings ancient Siberian/Asian population


    Open access paper New genetic evidence of affinities and discontinuities between bronze age Siberian populations, by Hollard et al., Am J Phys Anthropol. (2018) 00:1–11.

    NOTE. This seems to be a peer-reviewed paper based on a more precise re-examination of the samples from Hollard’s PhD thesis, Peuplement du sud de la Sibérie et de l’Altaï à l’âge du Bronze : apport de la paléogénétique (2014).

    Interesting excerpts:

    Afanasevo and Yamna

    The Afanasievo culture is the earliest known archaeological culture of southern Siberia, occupying the Minusinsk-Altai region during the Eneolithic era 3600/3300 BC to 2500 BC (Svyatko et al., 2009; Vadetskaya et al., 2014). Archeological data showed that the Afanasievo culture had strong affinities with the Yamnaya and pre-Yamnaya Eneolithic cultures in the West (Grushin et al., 2009). This suggests a Yamnaya migration into western Altai and into Afanasievo. Note that, in most current publications, “the Yamnaya culture” combines the so-called “classical Yamnaya culture” of the Early Bronze Age and archeological sites of the preceding Repin culture in the middle reaches of the Don and Volga rivers. In the present article we conventionally use the term Yamnaya in the same sense, in which case the beginning of the “Yamnaya culture” can be dated after the middle of the 4th millennium BC, when the Afanasievo culture appeared in the Altai.

    Because of numerous traits attributed to early Indo-Europeans and cultural relations with Kurgan steppe cultures, members of the Afanasievo culture are believed to have been Indo-European speakers (Mallory and Mair, 2000). In a recent whole-genome sequencing study, Allentoft et al. (2015) concluded that Eastern Yamnaya individuals and Afanasievo individuals were genetically indistinguishable. Moreover, this study and one published concurrently by Haak et al. (2015) analyzed 11 Eastern Yamnaya males and showed that all of them belonged to the R1b1a1a (formerly R1b1a) (…)

    Early Chalcolithic migrations ca. 3300-2600 BC.

    Published works indicate that R1b was a predominant haplogroup from the late Neolithic to the early Bronze Age, notably in the Bell Beaker and Yamnaya cultures (Allentoft et al., 2015; Haak et al., 2015; Lee et al., 2012; Mathieson et al., 2015). Nearly 100% of the Afanasievo men we typed belonged to the R1b1a1a subhaplogroup and, for at least three of them, more precisely to the L23 (xM412) subclade. (…)

    (…) our results therefore support the hypothesis of a genetic link between Afanasievo and Yamnaya. This also suggests that R1b was indeed dominant in the early Bronze Age Siberian steppe, at least in individuals that were buried in kurgans (possibly an elite part of the population). The geographical and temporal distribution of subhaplogroup R1b1a1a supports the hypothesis of population expansion from West to East in the Eurasian steppe during this period. It should however be noted that the Yamnaya burials from which the samples for DNA analysis were obtained (Allentoft et al., 2015; Haak et al., 2015; Mathieson et al., 2015) were dated within the limits of the Afanasievo period. Ancestors of both East Yamnaya and Afanasievo populations must therefore be sought in the context of earlier Eneolithic cultures in Eastern Europe. Sufficient Y-chromosomal data from such Eneolithic populations is, unfortunately, not yet available.

    Mitochondrial- (A) and Y- (B) haplogroup distribution in studied populations

    Okunevo and paternal lineage shift in South Siberia

    Results obtained in the current study, from more than a dozen Okunevo individuals belonging to the earliest stage of Okunevo culture, that is the Uibat period (2500–2200 BC) (Lazaretov, 1997), suggest a discontinuity in the genetic pool between Afanasievo and Okunevo cultures. Although Y-chromosomal data obtained for bearers of the Okunevo culture showed that one individual carried haplogroup R1b, most Okunevo Y-haplogroups are representative of an Asian component represented by paternal lineages Q and NO1.

    Okunevo carrier of Y-haplogroup Q1b1a-L54, which also supports this hypothesis (L54 being a marker of the lineage from which M3, the main Ameridian lineage, arose). Okunevo people could therefore be a remnant paleo-Siberian population with possible Afanasievo input, as suggested by the presence of the R1b1a1a2a subhaplogroup in one individual.

    Late Chalcolithic migrations ca. 2600-2250 BC.

    Replacement of Asian Indo-European elite lineages by R1a

    Published genetic data from the late Bronze Age Andronovo culture from the Minusinsk Basin (Keyser et al., 2009), the Sintashta culture from Russia (Allentoft et al., 2015) and the Srubnaya culture from the region of Samara (Mathieson et al., 2015), show that males did not belong to Y-haplogroup R1b but mostly to R1a clades: there appears to have been a change in the dominant Y-chromosomal haplogroup between the early and the late Bronze Age in these regions. Moreover, as described in Allentoft et al. (2015), the Andronovo and Sintashta peoples were closely related to each other but clearly distinct from both Yamnaya and Afanasievo. Although these results do not imply that Y-haplogroup R1b was entirely absent in these later populations, they could correspond to a replacement of the elite between these two main periods and therefore a difference in the haplogroups of the men that were preferentially buried.

    Early Bronze Age migrations ca. 2250-1750 BC.

    Afanasevo and the Tarim Basin

    The discovery, in the Tarim Basin, of well-preserved mummies from the Bronze Age allows for the construction of two hypotheses regarding the peopling of the Xinjiang province at this period. The “steppe hypothesis,” argues for a link with nomadic steppe herders (Hemphill and Mallory, 2004), possibly represented in this case by Afanasievo populations and their descendants (Mallory and Mair, 2000). However, newly published cultural data from the burial grounds of Gumugou (Wang, 2014) and Xiaohe (Xinjiang, 2003, 2007) shows material culture and burial rites incompatible with the Afanasievo culture. The earliest 14C date for Tarim Basin burials would place them at the turn of the 2nd millenium BC (Wang, 2013), 500 years after the Afanasievo period.

    Instead, early Gumugou and Xiaohe burial grounds were contemporary with the start of the Andronovo period. Likewise, the Bronze Age population of the Xinjiang at Gumugou/Qäwrighul is not phenotypically closest to Afanasievo but to the Andronovo (Fedorovo) group of northeastern Kazakhstan and western Altai (Kozintsev, 2009). Our investigations demonstrate that Y-chromosomal lineage composition is also compatible with the notion that the ancient Tarim population was genetically distinct from the Afanasievo population. The only Y-haplogroup found by Li et al. (2010) in the Bronze Age Tarim Basin population was Y-haplogroup R1a, which suggests a proximity of this population with Andronovo groups rather than Afanasievo groups.

    I don’t think these finds are much of a surprise based on what we already know, or need much explanation…

    I would add that, once again, we have more proof that the movement of Okunevo and related ancient Siberian migrants from Central or North Asia will not be able to explain the presence of Uralic languages spread over North-East Europe and Scandinavia already during the Bronze Age.

    Also interesting is to read in more peer-reviewed papers the idea of Late Indo-European speakers clearly linked to the expansion of patrilineally-related elite males marked by haplogroup R1b-L23, most likely since Eneolithic Khvalynsk/Repin cultures.