|
Post by Admin on Jan 26, 2022 5:01:28 GMT
The European Huns had Xiongnu ancestry Despite the paucity of Hun period samples, we can discern a “Hun-cline” along the PC1 axes (Fig 2a). Two individuals, MSG-1 and VZ-12673 (the same sample as HUN00113, resequenced with higher coverage) site at the extreme eastern pole of the cline, close to modern Kalmyks and Mongols. On PC50 clustering they tightly cluster together with two other Hun period samples; Kurayly_Hun_380CE (KRY001)13 and a Tian Shan Hun outlier (DA127)12 (Supplementary Table 3). As latter samples also form a genetic clade with VZ12673 (see below) we grouped these four samples under the name of Hun_Asia_Core (Fig. 2), though analyzed the new samples separately. The Hun_Asia_Core also clusters with numerous Xiongnu, Medieval Mongol, Turk14 and Xianbei13 genomes from Mongolia as well as several Avar samples from this study. ADMIXTURE confirmed the similarity of Hun_Asia_Core individuals, and showed prevailing east Eurasian Nganasan and Han components with no traces of WHG (Fig. 2b and Supplementary Table 4), implying that these individuals represent immigrants with no European background. Outgroup f3-statistics indicated common ancestry of MSG-1 and VZ-12673 (Extended Data Fig. 2), as both individuals shared highest drift with Neolithic farmers from the Wuzhuangguoliang site in northern China15, earlyXiongnu_rest, Ulaanzukh and SlabGrave samples from Mongolia14, pointing to a likely Mongolian origin and early Xiongnu affinity of these individuals. Distal qpAdm modeling from pre-Iron Age sources indicated major (70-94%) Wuzhuangguoliang and minor (6-30%) Mongolian Bronze Age ancestries in MSG-1 and VZ12673, while proximal modeling from post-Bronze Age sources gave two types of alternative models representing two different time periods (Supplementary Table 5a and 5b). The best Pvalue models showed major late Xiongnu (with Han admixture) and minor ScythoSiberian/Xianbei ancestries, while alternative models indicated 78-100% Kazakhstan_OutTianShanHun or Kurayly_Hun_380CE and 0-12% Xiongnu/Xianbei/Han ancestries. In latter models VZ-12673 formed a clade with both published Hun_Asia_Core samples. In conclusion, our Hun_Asia_Core individuals could be equally modelled from earlier Xiongnu and later Hun age genomes. The two other Hun period samples KMT-2785 and ASZK-1 were located in the middle of our PCA clines (Fig. 2 and Extended Data Fig 1a), and accordingly they could be modelled from European and Asian ancestors. The best passing models for KMT-2785 predicted 76% Late Xiongnu and 24% local EU_Core, while alternative model showed 86% Sarmatian12 and 14% Xiongnu ancestries (Supplementary Table 5c). Both models implicate Sarmatians as in the Late Xiongnus of the first model 46-52% Sarmatian and 48-54% Ulaanzuukh_SlabGrave components had been predicted14. The ASZK-1 genome formed a clade with Sarmatians in nearly all models. The rest of the Hun period samples map to the northern half of the EU cline, nevertheless two of these (SEI-1 and SEI-5) could be modelled from ~70% EU_Core and 30% Sarmatian components. The prevalent Sarmatian ancestry in 4 Hun period samples, implies significant Sarmatian influence on European Huns. CSB-3 was modelled as ~80% EU_Core and 20% Scytho-Siberian, while SEI-6 formed a clade with the Ukraine_Chernyakhiv16 (Eastern Germanic/Goth) genomes. The SZLA-646 outlier individual at the top of the EU-cline formed a clade with Lithuania_Late_Antiquity12 and England_Saxon17 individuals. The last two individuals presumably also belonged to Germanic groups allied with the Huns. Out of the 6 individuals in the Hun-cline (including DA127 and KRY001) four carried the R1a1a1b2 (R1a-Z93) Y-chromosomal haplogroup (Y-Hg), and one carried Q (Supplementary Table 1a), indicating that these Hg-s could be common among the European Huns, most likely inherited from Xiongnus18. Considering all published post-Xiongnu Hun era genomes12,13, we counted 10/23 R1a-Z93 and 9/23 Q Hgs, supporting this observation.
|
|
|
Post by Admin on Jan 26, 2022 20:17:48 GMT
Huns and Avars had related ancestry Our Avar period samples also form a characteristic PCA “Avar-cline” on Fig. 2, extending from Europe to Asia. PC50 clustering identified a single genetic cluster at the Asian extreme of the cline with 12 samples, derived from 8 different cemeteries, which we termed Avar_Asia_Core (Fig. 2, Supplementary Table 3). 10/12 samples of Avar_Asia_Core were assigned to the early Avar period, 4 of them belonging to the elite, and 9/12 were males.
Avar_Asia_Core clusters together with Shamanka_Eneolithic and Lokomotiv_Eneolithic19 samples from the Baikal region, as well as with Mongolia_N_East, Mongolia_N_North15, Fofonovo_EN, Ulaanzuukh_SlabGrave and Xiongnu14 from Mongolia (Supplementary Table 3). This result is recapitulated in ADMIXTURE (Fig. 2b), which also shows that Nganasan and Han components predominate in Avar_Asia_Core with traces of Anat_N and ANE, while Iranian and WHG ingredients are entirely missing. It follows, that Avar_Asia_Core was derived from East Asia, most likely from present day Mongolia.
We performed two-dimensional f4-statistics to detect minor genetic differences within the Avar_Asia_Core group. Avar_Asia_Core individuals could be separated along a Bactria-Margiana Archaeological Complex (BMAC)-Steppe Middle-Late Bronze Age (Steppe_MLBA) cline (Extended Data Fig 3), with 3 individuals bearing negligible proportion of these ancestries. The Steppe_MLBA-ANE f4-statistics gave similar results. As the 3 individuals with the smallest Iranian, Steppe and ANE ancestries also visibly separated on PCA, we set apart these under the name of Avar_Asia_Core1, while the other 9 samples were regrouped as Avar_Asia_Core2 (Fig. 2).
According to outgroup f3-statistics both Avar_Asia_Core groups had highest shared drift with genomes having predominantly Ancient North-East Asian (ANA) ancestry (Extended Data Fig. 2), like earlyXiongnu_rest, Ulaanzuukh, and Slab Grave14. It is notable that from the populations with top 50 f3 values, 41 are shared with Hun_Asia_Core, moreover Avar_Asia_Core1 is the 16th in the top list of VZ-12673 and 35th in that of MSG-1, signifying common deep ancestry of European Huns and Avars.
According to distal qpAdm models Avar_Asia_Core formed a clade with the Fofonovo_EN and centralMongolia_preBA genomes (Supplementary Table 6a), both of which had been modelled from 83%–87% ANA and 12%–17% ANE14. All data consistently show that Avar_Asia_Core preserved very ancient Mongolian pre-Bronze Age genomes, with ~90% ANA ancestry.
Most proximodistal qpAdm models (defined in Methods) retained distal sources, as Avar_Asia_Core1 was modelled from 95% UstBelaya_N20 plus 5% Steppe Iron Age (Steppe_IA) and Avar_Asia_Core2 from 80-92% UstBelaya_N plus 8-20% Steppe_IA (Supplementary Table 6b). The exceptional proximal model for Avar_Asia_Core1 indicated 58% Yana_Medieval20 plus 42% Ulaanzukh, while for Avar_Asia_Core2 69% Xianbei_Hun_Berel13 plus 31% Kazakhstan_Nomad_Hun_Sarmatian12 ancestries. The latter model also points to shared ancestries between Huns and Avars.
From the 76 samples in the Avar-cline, 26 could be modelled as a simple 2-way admixture of Avar_Asia_Core and EU_Core (Supplementary Table 6c) indicating that these were admixed descendants of locals and immigrants, while further 9 samples required additional Hun and/or Iranian related sources. In the remaining 40 models Hun_Asia_Core and/or Xiongnu sources replaced Avar_Asia_Core (Supplementary Table 6d, summarized in Supplementary Table 1b). Scythian-related sources with significant Iranian ancestries, like Alan, Tian Shan Hun, Tian Shan Saka12, or Anapa (this study), were ubiquitous in the Avar-cline, but given their low proportion, qpAdm was unable to identify the exact source.
Xiongnu/Hun-related ancestries were more common in certain cemeteries, for example it was detected in most samples from Hortobágy-Árkus (ARK), Szegvár-Oromdűlő (SZOD), Makó-Mikócsa-halom (MM) and Szarvas-Grexa (SZRV). Y-chromosomal data seem to corroborate this conclusion, as 8/10 males from ARK carried Y-Hg Q, while 2/10 R1a-Z94, 3/3 males from SZRV carried R1a-Z94 and 2/2 males from MM carried Hg Q (Supplementary Table 1a).
The Conquerors have Ugric, Sarmatian and Hun ancestry The Conquest period samples also form a characteristic genetic “Conq-cline” on PCA (Fig. 2). It is positioned north of the he Avar-cline, whilst only reaching the midpoint of the P1 axis. PC50 clustering identified a single genetic cluster at the Asian extreme of the cline (Supplementary Table 3) with 12 samples, derived from 9 different cemeteries, which we termed Conq_Asia_Core. This genetic group consists of 6 males and 6 females and 11 of the 12 individuals belonged to the Conqueror elite according to archaeological evaluation.
The PCA position of Conq_Asia_Core corresponds to modern Bashkirs and Volga Tatars (Fig. 2a) and they cluster together with a wide range of eastern Scythians, western Xiongnus and Tian Shan Huns12, which is also supported by ADMIXTURE (Fig. 2b).
Two-dimensional f4-statistics detected slight genetic differences between Conq_Asia_Core individuals (Extended Data Fig 4), obtained via multiple gene flow, as they had different proportion of ancestry related to Miao (a modern Chinese group) and Ulaanzuukh_SlabGrave (ANA)14. Besides, individuals were arranged linearly along the Miao-ANA cline, suggesting that these ancestries covary in the Conqueror group, thus could have arrived together, most likely from present day Mongolia. As four individuals with highest Miao and ANA ancestries also had shifted PCA locations, we set these apart under the name of Conq_Asia_Core2, while the rest were regrouped as Conq_Asia_Core1 (Fig. 2).
Admixture f3-statistics indicated that the main admixture sources of Conq_Asia_Core1 were Steppe_MLBA populations and ancestors of modern Nganasans (Extended Data Fig. 5). Outgroup f3-statistics revealed that Conq_Asia_Core1 shared highest drift with modern Siberian populations speaking Uralic languages; Nganasan (Samoyedic), Mansi (Ugric), Selkup (Samoyedic) and Enets (Samoyedic) (Extended Data Fig. 5), implicating that Conq_Asia_Core shared evolutionary past with language relatives of modern Hungarians. For this reason we co-analyzed Mansis, the closest language relatives of Hungarians with Conq_Asia_Core.
From pre-Iron Age sources Mansis could be qpAdm modelled from 48% Mezhovskaya10, 44% Nganasan and 8% Botai19, while Conq_Asia_Core1 from 52% Mezhovskaya, 13% Nganasan, 20% Altai_MLBA_o13 and 15% Mongolia_LBA_CenterWest_4D15 (Supplementary Table 7a and 7b) confirming shared late Bronze Age ancestries of these groups, but also signifying that the Nganasan-like ancestry was largely replaced in Conq_Asia_Core by a Scytho-Siberian-like ancestry including BMAC13,15 derived from the Altai-Mongolia region.
From proximal sources Conq_Asia_Core1 could be consistently modelled from 50% Mansi, 35% Early/Late Sarmatian and 15% Scytho-Siberian-outlier/Xiongnu/Hun ancestries, and Conq_Asia_Core2 had comparable models with shifted proportions (Supplementary Table 7c). As the source populations in these models defined inconsistent time periods, we performed DATES analysis21 to clarify admixture time.
DATES revealed that the Mansi-Sarmatian admixture happened around 643-431 BCE, apparently corresponding to the early Sarmatian period, while the Mansi-Scythian/Hun admixture was dated around 217-315 CE, consistent with the post-Xiongnu, early Hun period rather than the Iron Age (Extended Data Fig 6).
Most individuals of the Conqueror cline proved to be admixed descendants of the immigrants and locals, as 31 samples could be modelled as two-way admixtures of Conq_Asia_Core and EU_Core (Supplementary Table 7d, summarized in Supplementary Table 1b).
The remaining samples mostly belonged to the elite, many projecting with the Avar-cline (Fig. 1), of which 5 could be modelled from Conq_Asia_Core with Hun and Iranian associated additional sources. 17 outlier individuals lacked Conq_Asia_Core ancestry, which was replaced with Avar_Asia_Core or Xiongnu/Hun-related sources, accompanied by Iranian associated 3rd sources (Supplementary Table 7e). This result was again in line with Y-Hg data, as nearly all Conquest period males with R1a-Z94 or Q Hgs belonged to the last category (Supplementary Table 1a).
|
|
|
Post by Admin on Jan 26, 2022 20:55:22 GMT
Discussion The genomic history of Huns Avars and Conquerors revealed in this study reconciles with historical, archaeological and linguistic sources (summarized in Fig. 3). Our data shows that the leader strata of both European Huns and Avars originated from the area of the former Xiongnu Empire, from present day Mongolia, and both groups can be traced back to early Xiongnu ancestors. Northern Xiongnus were expelled from Mongolia in the second century CE, and during their westward migration Sarmatians were one of the largest groups they confronted. Sergey Botalov presumed the formation of a Hun-Sarmatian mixed culture in the Ural region before the appearance of Huns in Europe22, which fits the significant Sarmatian ancestry detected in our Hun samples, though this ancestry had been present in late Xiongnus as well14. Thus our data are in accordance with the Xiongnu ancestry of European Huns, claimed by several historians23,24. We also detected Goth- or other German-type genomes among our Hun period samples, again consistent with historical sources23. Fig. 3 Summary map. a, Proto-Ugric peoples emerged from the admixture of Mezhovskaya and Nganasan populations in the late Bronze. b, 1. During the Iron Age Mansis separated. 2. proto-Conquerors admixed with Early Sarmatians 643-431 BCE and 3. with early Huns 217-315 CE. c, By the 5th century the Xiongnu descent Hun Empire occupied Eastern Europe incorporating its population, and the Rouran Khaganate emerged on the former Xiongnu territory. d, By the middle 6th century the Avar Khaganate occupied the territory of the former Hun Empire incorporating its populations. 4. By the 10th century Conquerors associated with the remnants of both empires during their migration and within the Carpathian Basin. Our data are compatible with the Rouran origin of Avar elite25, though the single low coverage Rouran genome26 provided a poor fit in the qpAm models (Supplementary Table 6b). The elite preserved very ancient east Asian genomes with undisputable origin, as had been also inferred from Y-Hg data27,28, however just half of the Avar-cline individuals had Avar_Asia_Core ancestry, implicating diverse origin of the Avar population. Our models indicate that the Avars incorporated groups with Xiongnu/Hun_Asia_Core and Iranian ancestries, presumably the remnants of the European Huns and Alans or other Iranian peoples on the Pontic Steppe, as suggested by Kim 201323. People with different origin were seemingly distinguished, as samples with Hun-related genomes were buried in separate cemeteries. The Conquerors, who arrived in the Carpathian Basin after the Avars, had distinct genomic background with elevated levels of western Eurasian admixture. They carried very similar genomes to modern Bashkirs and Tatars, in agreement with our previous results from uniparental markers28,29. Their genomes were shaped by several admixture events, of which the most fundamental was the Mezhovskaya-Nganasan admixture around the late Bronze Age, leading to the formation of a “proto-Ugric” gene pool. This was part of a general demographic process, when most Steppe_MLBA populations received an eastern Khovsgol related Siberian influx together with a BMAC influx13, and ANA related admixture became ubiquitous on the eastern Steppe21 establishing the Scytho-Siberian gene pool. Consequently proto-Ugric groups could be part of the early Scytho-Siberian societies of the late Bronze Age-early Iron Age steppe-forest zone in the northern Kazakhstan region, in the proximity of the Mezhovskaya territory. Our data support linguistic models, which predicted that Conquerors and Mansis had a common early history4,30. Then Mansis migrated northward, probably during the Iron Age, and in isolation they preserved their Bronze-Age genomes. In contrast the Conquerors stayed at the steppe-forest zone and admixed with Iranian speaking early Sarmatians, also attested by the presence of Iranian loanwords in the Hungarian language30. This admixture likely happened when Sarmatians rose to power and started to integrate their neighboring tribes before they occupied the Pontic-Caspian Steppe. All analysis congruently indicated, that the ancestors of Conquerors further admixed with a group from Mongolia, carrying Han-ANA related ancestry, which could be identified with early European Huns, compelling reconsideration of written historical sources about the Hun-Hungarian relations. It is to be examined, how this genetic link is related to reports in medieval Hungarian chronicles about the Hun ancestry of the Conqueror elite, which according to the current state of historiography is not sufficiently supported31. This admixture could happen before the Huns arrived to the Volga region and integrated local tribes east of the Urals, including Sarmatians and the ancestors of Conquerors. These data are compatible with a Conqueror homeland around the Ural region, in the vicinity of early Sarmatians, along the migration route of the Huns, as had been surmised from the phylogenetic connections between the Conquerors and individuals of the Kushnarenkovo-Karayakupovo culture in the Trans-Uralic Uyelgi cemetery32. Recently a Nganasan-like shared Siberian genetic ancestry was detected in all Uralic-speaking populations, Hungarians being an exception33. Our data fills this gap, as Conq_Asia_Core has high Nganasan ancestry, notwithstanding this is negligible in modern Hungarians, partly because of the substantially smaller number of immigrants compared to the local population. The large number of genetic outliers with Hun_Asia_Core ancestry in both Avars and Conquerors testify that these successive nomadic groups were indeed assembled from overlaping populations. Additional Information www.biorxiv.org/content/10.1101/2022.01.19.476915v1.full
|
|
|
Post by Admin on Feb 7, 2022 21:09:02 GMT
Interdisciplinary analyses of Bronze Age communities from Western Hungary reveal complex population histories Dániel Gerber1,2,3, Bea Szeifert1,2,3, Orsolya Székely1
1) Institute of Archaeogenomics, Research Centre for the Humanities, Eötvös Loránd Research Network (ELKH); Tóth Kálmán utca 4., 1097 Budapest, Hungary 2) Department of Genetics, ELTE Eötvös Loránd University; Pázmány Péter sétány 1/C, 1117 Budapest, Hungary 3) Doctoral School of Biology, Institute of Biology, ELTE Eötvös Loránd University, Pázmány Péter sétány 1/C. 1117 Budapest, Hungary
Abstract In this study we report 20 ancient shotgun genomes from present-day Western Hungary (3530 – 1620 cal BCE), mainly from previously understudied Baden, Somogyvár-Vinkovci, Kisapostag, and Encrusted Pottery archaeological cultures. Besides analysing archaeological, anthropological and genetic data, 14C and strontium isotope measurements complemented reconstructing the dynamics of the communities discovered at the site Balatonkeresztúr. Our results indicate the appearance of an outstandingly high Mesolithic hunter-gatherer ancestry in the largest proportion (up to ~46%) among Kisapostag associated individuals, despite this component being thought to be highly diluted by the Early Bronze Age. We show that hunter-gatherer ancestry was likely derived from a previously unrecognised source in Eastern Europe that contributed mostly to prehistoric populations in Central Europe and the Baltic region. We revealed a patrilocal residence system and local female exogamy for this Kisapostag population that was also the genetic basis of the succeeding community of the Encrusted Pottery culture, represented by a mass grave that likely resulted from an epidemic. We also created a bioinformatic pipeline dedicated for archaeogenetic data processing. By developing and applying analytical methods for analysing genetic variants we found carriers of aneuploidy and inheritable genetic diseases. Furthermore, based on genetic and anthropological data, we present here the first female facial reconstruction from the Bronze Age Carpathian Basin.
Significance Here we present a genomic time transect study from the Carpathian Basin (3530 – 1620 cal BCE), that sheds light on local and interregional population processes. We not only discovered long-distance mobility to provide detailed analysis of yet understudied Bronze Age communities, but we also recovered a previously hidden remnant hunter-gatherer genetic ancestry and its contribution to various populations in Eastern and Central Europe. We integrated 14C and strontium isotope measurements to the interdisciplinary interpretation of a site with 19 individuals analysed, where patrilocal social organisation and several health-related genetic traits were detected. Furthermore, we developed new methods and method standards for computational analyses of archaic DNA, implemented to our newly developed and freely available bioinformatic pipeline.
|
|
|
Post by Admin on Feb 8, 2022 2:00:38 GMT
Introduction A number of studies addressed population historical questions in Prehistoric Europe by recovering major events connected to the pre-Neolithic hunter-gatherers (HG)1–3, their assimilation to early European farmers during the Neolithic era2,4–6, and the appearance, expansion and admixture of steppe ancestry during the Eneolithic / Late Copper Age to the dawn of Early Bronze Age4,7,8. While some of these studies are essential for understanding the foundation of the European gene pool, studies are sparse in the literature that uncover regional interactions or social stratification via kinship9–11. Additionally, except for a few well-known markers in most archaic studies – e.g. basic pigmentation markers or lactose intolerance analysed large-scale in Mathieson et al. 201512 – no deeper analyses have been made e.g. on clinical variants. Our study aimed to make a transect analysis on a single site concerning understudied archaeological assemblies, as well as introducing the PAPline (Performing Archaeogenetic Pipeline, Supplementary Information section 6), a new bioinformatic pipeline for archaic DNA analysis. We analysed the archaeological finds from Balatonkeresztúr-Réti-dűlő site in Western Hungary, where - among others - Bronze Age assemblies were found during roadwork in 2003. Three Bronze Age archaeological horizons can be distinguished, from the Somogyvár-Vinkovci culture (~2500-2200 BCE), Kisapostag culture (~2200–1900 BCE) and to the Encrusted pottery culture (~1900–1450 BCE) that were named into Bk-I, II and III phases in this study, respectively (Table 1, Supplementary Information section 1, and Fig. S.1.2.1). In order to provide additional proxy to population ancestry of the region one further Late Copper Age individual from a multiple grave of the Baden culture (3600-2800 BCE) excavated at site Balatonlelle-Rádpuszta, ~30 km away from Balatonkeresztúr was added to our dataset.
Results We shotgun sequenced genomes of 20 individuals with 0.008 to 0.17x coverage. We also sequenced reads of a capture set consisting 3000 nuclear SNPs (single nucleotide polymorphisms), and whole mitochondrial DNAs (mtDNAs) of all individuals. The shotgun and the capture sequenced samples ultimately resulted in an average ~104k SNPs/individuals using the 1240k SNP panel for genotype calling12, see Materials and Methods and Supplementary Tables S4 and S7. We utilised STR analysis of the Y chromosome to recover paternal kinship patterns. Furthermore, we reconstructed the face of individual S13 (Bk-II), where all known biological and archaeological details were considered, see Supplementary Information section 4. The bioarchaeological analyses were completed with radiocarbon and strontium isotope analyses, the latter can be used to trace individual mobility.
Archaeological and anthropological evaluation of samples Bk-I contained the remains of a single male individual having a very long (ultradolichocran) skull type which differentiates it from most individuals found at Bk-II and Bk-III that have a very short (brachycranic) skull type13 (Table 1). In Bk-II and Bk-III male dominance (~78%) suggest distinctive funeral treatment for males and females. Bk-II is represented by 3 juveniles (16-19 years olds) and 7 adults (30+ years olds) distributed into two grave groups of A and B (Table 1, Supplementary Information Fig. S.1.2.1), and one child grave (individual S10) far from the others. Most of the burials contained no remaining grave goods except for small copper jewellery in S10 and S13. Radiocarbon dates place these inhumations to ca. 2200-1770 cal BCE, however, with Bayesian analysis using the OxCal software the timespan of the Bk-II burials can be reduced to ca. 2120-1900 cal BCE (95.4% CI), with two graves (individuals S10 and S11) possibly being slightly earlier (Supplementary Information section 1.8). The absence of children from the site is a common phenomenon that can be traced back to different preservation dynamics or burial practises to adults14, who the reason for the absence of young adults (~20-30 year olds) is unknown. Bk-III is represented by a single mass grave of 8 skeletal remains of all ages that turned out to be an unusual find in a period when the cremation practises and single inhumations were common, from ca. 1870-1620 cal BCE (95.4% CI). For details, see Supplementary Information section 1.
|
|