Thursday, May 5, 2022

Unraveling the origins of the Pannonian Avars, one sample at a time.

This blog entry will be all about the Pannonian Avars, a group of steppe nomads that in my opinion received a bit of the short end of the stick in terms of historiography. Their appearance was sudden, their Khaganate only lasted for two centuries and no subsequent populations were significantly derived from them, which all goes for the Huns as well yet they are probably the most infamous and iconic nomads from a western perspective.

Ancient genomes reveal origin and rapid trans-Eurasian migration of 7th century Avar elites

Guido Alberto Gnecchi-Ruscone, Anna Szécsényi-Nagy, István Koncz, Gergely Csiky, Zsófia Rácz, A.B. Rohrlach, Guido Brandt, Nadin Rohland, Veronika Csáky, Olivia Cheronet, Bea Szeifert, Tibor Ákos Rácz, András Benedek, Zsolt Bernert, Norbert Berta, Szabolcs Czifra, János Dani, Zoltán Farkas, Tamara Hága, Tamás Hajdu, Mónika Jászberényi, Viktória Kisjuhász, Barbara Kolozsi, Péter Major, Antónia Marcsik, Bernadett Ny. Kovacsóczy, Csilla Balogh, Gabriella M. Lezsák, János Gábor Ódor, Márta Szelekovszky, Tamás Szeniczey, Judit Tárnoki, Zoltán Tóth, Eszter K. Tutkovics, Balázs G. Mende, Patrick Geary, Walter Pohl, Tivadar Vida, Ron Pinhasi, David Reich, Zuzana Hofmanová, Choongwon Jeong, Johannes Krause,


The Avars settled the Carpathian Basin in 567/68 CE, establishing an empire lasting over 200 years. Who they were and where they came from is highly debated. Contemporaries have disagreed about whether they were, as they claimed, the direct successors of the Mongolian Steppe Rouran empire that was destroyed by the Turks in ∼550 CE. Here, we analyze new genome-wide data from 66 pre-Avar and Avar-period Carpathian Basin individuals, including the 8 richest Avar-period burials and further elite sites from Avar’s empire core region. Our results provide support for a rapid long-distance trans-Eurasian migration of Avar-period elites. These individuals carried Northeast Asian ancestry matching the profile of preceding Mongolian Steppe populations, particularly a genome available from the Rouran period. Some of the later elite individuals carried an additional non-local ancestry component broadly matching the steppe, which could point to a later migration or reflect greater genetic diversity within the initial migrant population.

This article came out just after the one with the new data from Xinjiang, but the 202 ancient samples they provided kept me busy for a while. And then I got quite busy with preparing for a fight (which will take place this sunday by the way, K1 ruleset), and well a few weeks later here we are!

On a first glance, there isn’t much I disagree with, so kudos to the authors because that rarely happens. I have my minor disagreements, but the main body of the article holds up quite well I think. I’m assuming that you have already read the article, so I will jump right into the part where I analyze these samples myself.

All I’m showing you here will be through the wonderful magic of Vahaduo Global25. You can find the relevant information here and here:

You can easily recreate the models I posted by simply copying and pasting the references, but it is still good practice to get a bit more familiar with the program before dumping all sorts of references in there. If you were able to make some better models, absolutely do share them here because I’m not going to pretend mine are some infallible models with a 100% validity that fell out of the sky.

Well, let’s dive in!

Upper Xiajiadian culture

To begin, let’s start with the Upper Xiajiadian culture, a bronze age material culture from Northeast China dating to roughly 1000-600 BC. The story of the Upper Xiajiadian culture starts with the earlier Lower Xiajidian culture, a society of millet farmers in northeastern China. In the period after there were significant changes in both the cultural expression of the peoples as well as the subsistence economy, with less of a focus on fixed settlements and a shift towards mobile pastoralism. It has been proposed that these results came about through a migration of nomadic pastoralists that established themselves as the elites in the Xiajiadian society. However other proposals link the economic shift to climatic changes.

The upper Xiajiadian culture has frequently connected to the Donghu, an iron age steppe nomad confederation that broke off into the Xianbei and Wuhuan confederations, which are both assumed to have been primarily comprised of (Pre-Proto/Para-)Mongolic speakers. Now whether this identification of the UXJD with the Donghu is correct or not is not the main important element here, it is the genomic ancestry of UXJD peoples, which were featured in the article “Ancient genomes from northern China suggest links between subsistence changes and human migration”.

Target: CHN_Western_Liao_River_BA:91KLH11

Distance: 3.3703% / 0.03370301

74.4 CHN_Western_Liao_River_LN

18.8 CHN_Amur_River_EN

6.8 MNG_East_N

Target: CHN_Western_Liao_River_BA:91KLH18

Distance: 2.2658% / 0.02265754

78.8 CHN_Western_Liao_River_LN

15.8 CHN_Amur_River_EN

5.4 MNG_East_N

The first two samples are rather similar to the Late neolithic UXJD samples, although they are seemingly more northern-shifted. The northern shift could have occurred from those pastoral newcomers I talked about earlier, but the samples were also geographically more northern than the LXJD samples, thus it might also be that we are just looking at clinal variation.

The interesting sample is the outlier of the bunch, with the ID 91KLM2.

Target: CHN_Western_Liao_River_BA_o:91KLM2

Distance: 5.3871% / 0.05387061

61.0 CHN_Amur_River_EN

39.0 MNG_East_N

0.0 CHN_Western_Liao_River_LN

0.0 MNG_Ulaanzukh_LBA_2

This sample was a male and carried C2b1a1b1, a Y-chromosome haplogroup frequently found amongst Mongolic peoples today. This sample works extremely well as a genetic proxy for Mongolic populations or people with ancestry of Mongolic peoples. It does not work as a sole proxy, which you will see below.

You thought the sword was cool? Check out this pan:

Yeah you saw that right.

The Xianbei

The legendary Xianbei do not need much of an introduction. They arose after the defeat of the Donghu by the Xiongnu, and then gradually grew into one of the major political forces on the eastern steppes. The Xianbei confederation was primarily composed of (Pre/Proto/Para) Mongolic peoples, however peoples of different ethnolinguistic origins certainly became part of the Xianbei. Several scholars have suggested that the Tuoba may have had Turkic elements for example.

As far as ancient Xianbei samples go, we do not have much but what we have is already quite informative. From the same article containing the UXJD samples, there were two Xianbei samples found near the Hulun Lake in Inner Mongolia, but unfortunately only one made the cut for  Global25.


Target: CHN_Amur_River_Xianbei_IA:MGS-M7R

Distance: 3.0831% / 0.03083059

62.0 CHN_Western_Liao_River_BA_o

23.4 MNG_Ulaanzukh_LBA_2

11.4 RUS_Krasnoyarsk_BA

3.2 Saka_Tian_Shan

Y-DNA: C2b1a1b1b

mtDNA: C4a1a4a

The other sample, MGS-M6, also had  Y-chromosome haplogroup C2b1a1b1.

Another Xianbei period sample was featured in  the 2021 article Genomic insights into the formation of human populations in East Asia.


Target: MNG_Xianbei_IA:I13175

Distance: 2.0870% / 0.02086989

51.6 CHN_Western_Liao_River_BA_o

22.6 MNG_Ulaanzukh_LBA_2

16.2 Saka_Tian_Shan

9.6 CHN_Upper_Yellow_River_IA

Y-DNA: R-PH200

mtDNA: D4b1a2a

And interestingly this sample carried R-PH200, a Y-chromosome haplogroup also found amongst samples from the Xiongnu, later Huns as well as samples from the Shirenzigou site. PH200 is downstream of the R1b-PH155 haplogroup which was carried by the Tarim_EMBA cluster, and later alo carried in high frequencies amongst the samples of the Zaghunluq cemetery.

There are also some samples from Ancient genomic time transect from the Central Asian Steppe unravels the history of the Scythians that were labeled as “Xianbei-Hun” but this label has more to do with the time period than any ethnic affiliation of the samples themselves. Genetically they looked very Turkic, unlike the other Xianbei samples. Here is a calculation of their average:

Target: KAZ_Xianbei_Hun_Antiquity

Distance: 1.2891% / 0.01289076

48.8 Saka_Tian_Shan

45.0 MNG_Slab_Grave_EIA_1

3.4 RUS_Krasnoyarsk_BA

1.4 CHN_Upper_Yellow_River_IA

1.4 CHN_Western_Liao_River_BA_o


A Dynamic 6,000-Year Genetic History of Eurasia's Eastern Steppe, an article published last year featured several samples belonging to the Khitans.


Target: MNG_Late_Med_Khitan:ULA001

Distance: 1.8225% / 0.01822528

37.6 CHN_Western_Liao_River_BA_o

35.0 CHN_Western_Liao_River_BA

10.2 CHN_Upper_Yellow_River_IA

10.2 MNG_Slab_Grave_EIA_1

7.0 Saka_Tian_Shan


mtDNA: D4c1b1


Target: MNG_Late_Med_Khitan:ZAA005

Distance: 2.1766% / 0.02176650

39.0 CHN_Western_Liao_River_BA

38.0 CHN_Western_Liao_River_BA_o

15.0 Saka_Tian_Shan

8.0 MNG_Slab_Grave_EIA_1

0.0 CHN_Upper_Yellow_River_IA


mtDNA: Y1a


Target: MNG_Late_Med_Khitan:ZAA003

Distance: 3.1488% / 0.03148836

67.6 CHN_Upper_Yellow_River_IA

28.6 CHN_Western_Liao_River_BA_o

3.8 Saka_Tian_Shan

0.0 CHN_Western_Liao_River_BA

0.0 MNG_Slab_Grave_EIA_1


mtDNA: F2a

These look quite different and are a lot more eastern, which you can see as they either lack or hav a low amount of the ancestry represented by the iron age Slab Grave samples from  Central mongolia. The western Liao_LN/BA type ancestry might point towards Korean admixture, but it doesn’t have to be so as non-Koreanic speaking people in that area could’ve carried such ancestry also.


To round it out I will also post some calculations of later medieval Mongols and modern Mongolic peoples. Unlike the earlier clusters these actually contain quite a number of samples so I will only present the averages. 

Target: MNG_Late_Med

Distance: 1.0529% / 0.01052892

25.2 CHN_Western_Liao_River_BA

24.6 CHN_Western_Liao_River_BA_o

21.2 Saka_Tian_Shan

16.2 MNG_Slab_Grave_EIA_1

10.8 CHN_Upper_Yellow_River_IA

2.0 RUS_Krasnoyarsk_BA

Target: MNG_Late_Med

Distance: 0.8755% / 0.00875496

30.0 CHN_Western_Liao_River_BA

26.8 RUS_Late_Xiongnu

25.0 CHN_Amur_River_Xianbei_IA

9.2 KAZ_Otyrar_Antiquity

5.2 CHN_Upper_Yellow_River_IA

3.8 Saka_Tian_Shan

Target: MNG_Mongol

Distance: 1.8230% / 0.01823032

35.2 Saka_Tian_Shan

29.6 CHN_Upper_Yellow_River_IA

22.4 CHN_Western_Liao_River_BA_o

12.8 MNG_Slab_Grave_EIA_1

Target: MNG_Mongol

Distance: 1.7328% / 0.01732786

29.8 CHN_Amur_River_Xianbei_IA

29.6 CHN_Upper_Yellow_River_IA

15.0 Saka_Tian_Shan

14.0 KAZ_Otyrar_Antiquity

11.6 RUS_Late_Xiongnu

And then here I have modern Mongolic averages, modeled with the same references:


Target: Mongolian

Distance: 2.1291% / 0.02129089

31.8 CHN_Western_Liao_River_BA_o

28.4 Saka_Tian_Shan

19.6 MNG_Slab_Grave_EIA_1

14.8 CHN_Western_Liao_River_BA

4.4 CHN_Upper_Yellow_River_IA

1.0 RUS_Krasnoyarsk_BA

Target: Mongolian

Distance: 1.7830% / 0.01783017

41.6 CHN_Amur_River_Xianbei_IA

19.2 RUS_Late_Xiongnu

14.8 CHN_Western_Liao_River_BA

12.6 KAZ_Otyrar_Antiquity

8.0 Saka_Tian_Shan

3.8 CHN_Upper_Yellow_River_IA


Target: Buryat

Distance: 2.2460% / 0.02245976

43.8 CHN_Western_Liao_River_BA_o

27.8 Saka_Tian_Shan

17.6 MNG_Slab_Grave_EIA_1

5.2 RUS_Krasnoyarsk_BA

3.4 CHN_Western_Liao_River_BA

2.2 CHN_Upper_Yellow_River_IA

Target: Buryat

Distance: 1.8510% / 0.01851043

47.0 CHN_Amur_River_Xianbei_IA

35.4 RUS_Late_Xiongnu

12.6 KAZ_Otyrar_Antiquity

3.4 Saka_Tian_Shan

1.6 CHN_Western_Liao_River_BA


Target: Daur

Distance: 1.8218% / 0.01821774

60.0 CHN_Western_Liao_River_BA

23.6 CHN_Western_Liao_River_BA_o

6.0 Saka_Tian_Shan

5.6 RUS_Krasnoyarsk_BA

4.8 CHN_Upper_Yellow_River_IA

Target: Daur

Distance: 1.8554% / 0.01855393

65.0 CHN_Western_Liao_River_BA

28.4 CHN_Amur_River_Xianbei_IA

3.0 KAZ_Otyrar_Antiquity

2.4 RUS_Krasnoyarsk_BA

1.2 Saka_Tian_Shan


Now that we have the genetic structure of ancient Mongolic populations sorted out, let's have a look at what the Avars have to offer us. I will not spend much time looking at people that seem Pre-Avar in terms of genetic origins. I have decided to present these samples individually, to also present the uniparentals and the burial context of the samples.

Early Avars - Danube-Tisza

First up are the early Avars, from the Danube-Tisza group. Many of the samples here were of the upper classes of the Avar society, as indicated by the wealth in their burials.


Target: HUN_Avar_Early_Danube-Tisza:A1801

Distance: 2.2421% / 0.02242130

71.2 RUS_Late_Xiongnu

24.8 CHN_Amur_River_Xianbei_IA

4.0 RUS_Krasnoyarsk_BA

Y-DNA: N1a1a1a1a3a2 (B219)

mtDNA: C4b6

Description: High-ranking individual who probably was a member of the Avar period elite belonging to the military retinue of the Qagan. The burial contained a gold decorated sword.


Target: HUN_Avar_Early_Danube-Tisza:A1802

Distance: 2.8259% / 0.02825868

47.0 MNG_Xianbei_IA

38.4 CHN_Amur_River_Xianbei_IA

14.6 RUS_Krasnoyarsk_BA

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: D4j5a

Description: He has the richest burial assemblage from the early Avar period with 155 preserved artefacts, so he probably was a member of the highest social rank. He was interpreted as the Qagan, but the lack of proper insignia suggests otherwise.

Images of golden artefacts from A1802’s grave. Source: Genetic insights into the social organisation of the Avar period elite in the 7th century AD Carpathian Basin | Scientific Reports


Target: HUN_Avar_Early_Danube-Tisza:A1816

Distance: 3.5287% / 0.03528709

47.8 CHN_Amur_River_Xianbei_IA

38.0 MNG_Xianbei_IA

14.2 RUS_Krasnoyarsk_BA

Y-DNA: -

mtDNA: U5a1+@16192

Description: One of the few female burials that have at least some kind of prestigious grave goods. A golden earring was present in this burial.


Target: HUN_Avar_Early_Danube-Tisza:A1817

Distance: 2.4738% / 0.02473810

39.4 CHN_Amur_River_Xianbei_IA

32.6 Iberia_Northeast_Empuries2

22.2 IRN_Ganj_Dareh_Historic

5.8 RUS_Krasnoyarsk_BA

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: H8a1

Description: Infant burial without grave goods


Target: HUN_Avar_Early_Danube-Tisza:A1818

Distance: 2.8005% / 0.02800528

61.0 RUS_Late_Xiongnu

24.4 CHN_Amur_River_Xianbei_IA

13.8 RUS_Krasnoyarsk_BA

0.8 RUS_Alan_MA

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: H8a1

Description: This male probably did not belong to the highest social rank, but the burial was still rich (silver instead of gold, simple belt etc.) The burial contained a sword, bow and elements of a decorated belt.


Target: HUN_Avar_Early_Danube-Tisza:A1819

Distance: 4.1547% / 0.04154670

56.6 CHN_Amur_River_Xianbei_IA

31.2 RUS_Late_Xiongnu

12.2 RUS_Krasnoyarsk_BA

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: Z1a1

Description: Most probably a lonely grave. The burial belongs to the group of elite burials with military character located in the Danube-Tisza Interfluve that could be interpreted as the military retinue of the Qagan. The burial was richly furnished and contained a sword covered with gold plates and two belt sets


Target: HUN_Avar_Early_Danube-Tisza:A1821

Distance: 3.0908% / 0.03090817

55.4 CHN_Amur_River_Xianbei_IA

33.0 RUS_Late_Xiongnu

11.6 RUS_Krasnoyarsk_BA

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA:: F1b1b

Description: Extremely richly furnished burial containing gold-plated swords, belt, etc.


Target: HUN_Avar_Early_Danube-Tisza:A1822

Distance: 2.6877% / 0.02687656

47.2 CHN_Amur_River_Xianbei_IA

47.2 RUS_Late_Xiongnu

5.6 MNG_Xianbei_IA

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: M7c1b2b

Description: The burial belonged to an elite male, likely belonging to the military retinue of the Qagan, and was buried with a sword, bow and multiple elements of a decorated belt made of precious metals


Target: HUN_Avar_Early_Danube-Tisza:A1823

Distance: 2.8720% / 0.02871968

39.0 CHN_Amur_River_Xianbei_IA

27.6 MNG_Xianbei_IA

21.8 RUS_Late_Xiongnu

11.6 RUS_Krasnoyarsk_BA

Y-DNA: N1a1a1a1a3a (Y16310)

mtDNA: Y1a1

Description: This individual likely belonged to a military elite, and was buried with a ring-pommel sword and a belt.


Target: HUN_Avar_Early_Danube-Tisza:I18743

Distance: 3.0787% / 0.03078684

38.8 CHN_Amur_River_Xianbei_IA

29.2 MNG_Xianbei_IA

16.8 RUS_Late_Xiongnu

15.2 RUS_Krasnoyarsk_BA

Y-DNA: -

mtDNA: G2a2

Description: This burial belonged to a young adult woman of average social standing, and she was buried with glass beads, bone box, spindle ring, iron knife, bone plates of a bow, animal bones

I made a simulated average of these samples using Genoplot, with A1817 removed as he had European admixture.

Target: HUN_Avar_Early_Danube-Tisza_SIM

Distance: 2.2221% / 0.02222108

45.2 CHN_Amur_River_Xianbei_IA

39.0 RUS_Late_Xiongnu

9.2 RUS_Krasnoyarsk_BA

6.6 MNG_Xianbei_IA

Then I  made another average, this time in addition to A1817 also subtracting A1801, A1818 and A1822, the three samples which had significantly more ancestry related to the Late_Xiongnu cluster from Jeong 2021. 

Target: HUN_Avar_Early_Danube-Tisza_SIM2

Distance: 2.5607% / 0.02560666

49.2 CHN_Amur_River_Xianbei_IA

23.4 RUS_Late_Xiongnu

15.4 MNG_Xianbei_IA

12.0 RUS_Krasnoyarsk_BA

 I think these two simulated averages cover the core ancestry of the Pannonian Avars quite well, and I will be using them as references for “Avar” ancestry in other samples. If you are interested the scaled coordinates are below:

  • HUN_Avar_Early_Danube-Tisza_SIM,0.0394587,-0.406776,0.0946992,-0.0240097,-0.0848364,-0.0484961,0.0156152,0.0271782,0.0122488,0.0135663,-0.0151202,0.0006494,-0.0012387,0.0043428,0.003574,-0.000707,0.0013329,-0.0026464,0.0027653,0.0144513,-0.0244016,-0.0031601,-0.0205824,-0.003066,0.002395

  • HUN_Avar_Early_Danube-Tisza_SIM2,0.0371822,-0.41569,0.0974857,-0.0256247,-0.0860673,-0.0524315,0.014649,0.0273833,0.0135668,0.0143965,-0.0127747,0.0005495,-0.0013378,0.0038533,0.0030312,0.0019448,0.0014777,-0.0041598,0.0040852,0.0162993,-0.0254762,-0.0035862,-0.0220408,-0.0016467,0.002415

What we can see from the Avar samples is that they are quite closely related to the two published Xianbei genomes from China and Mongolia, with a variable amount of “Late Xiongnu” which is proxying for ancestry from Turkic speaking populations. Given the consistency in profiles it is quite likely much of this Xiongnu ancestry was already carried by the Avar forebearers in Mongolia.

All the Avar samples also seem to have ancestry from a more northerly East Asian population. This ancestry currently seems best represented by Kra001/Krasnoyarsk_BA, a bronze age sample found in Krasnoyarsk Krai dating to  2135- 2336 BC. [2] Out of the modern day Mongolic speaking peoples Buryats seem to have the most of this northern ancestral component well-represented in the Avar samples, although they have a bit less than these samples seem to have. 

Kra001 currently also seems to be the closest match for the northeast Asian ancestry of Uralic people in the ancient DNA record, and on yfull it’s position is at a basal N-CTS6967, which is the ancestral subclade of N-L1026, a paternal lineage that is quite connected to the spread of Uralic languages, although it certainly is not exclusive to Uralic people by any means anymore, as these N-L1026 > N-F4205 Avar samples show.

N-F4205 is definitely a Mongolic lineage in terms of modern day distribution, reaching its highest frequency amongst certain Buryat populations. The presence of Kra001-related ancestry in these samples could point towards the origin of their patrilineal lineage.

The fact that all of the Avar core samples had this particular lineage in addition to relatively uniform autosomal ancestry, to me it suggests that these Avars were all part of a single tribe or clan,in which (nearly) all members shared a common paternal ancestor.

Early Avars - Transnitzia

While the Danube-Tisza Avar samples were quite straightforward and relatively homogenous, The Avar samples from Transnitzia were quite the different story. I will first show the samples that had ancestry from Pannonian Avars.


Target: Avar_Early_Transtisza:A1804

Distance: 1.5023% / 0.01502305

38.4 HUN_Avar_Early_Danube-Tisza_SIM2

37.0 RUS_Late_Xiongnu

10.4 CZE_Early_Slav

7.6 RUS_Alan_MA


1.0 DEU_MA_Alemannic

Y-DNA: Q1a2a1a4a~ (YP817)

mtDNA: D5b1

Description: Richly furnished grave containing a belt, with pressed mounts

, food offerings and gold sheets.

Target: Avar_Early_Transtisza:A1805

Distance: 2.8891% / 0.02889115

47.0 HUN_Avar_Early_Danube-Tisza_SIM2

22.6 RUS_Late_Xiongnu

18.4 RUS_Alan_MA


1.0 DEU_MA_Alemannic

Y-DNA: - 

mtDNA: H5a2

Description: Richly furnished burial containing a horse harness and food offerings.


Target: Avar_Early_Transtisza:A1806

Distance: 1.9701% / 0.01970087

54.4 RUS_Late_Xiongnu

39.6 HUN_Avar_Early_Danube-Tisza_SIM2

5.4 RUS_Alan_MA

0.6 CZE_Early_Slav

Y-DNA: -


Description: “probably nothing special, poorly furnished burial”

A1804 carried the y-chromosome haplogroup Q-Y2679, which is a different patrilineal lineage than most Avars carried. Q-Y2679 surprisingly is also well distributed in Scandinavia!

Well, well, well…

The origins are murky, but a basal form of Q-Y269 was carried by an Okunev sample. This could suggest that there is a  Scytho-Siberian origin for this lineage, as they occasionally carried patrilineal lineages from Paleo-Siberian populations. 

it is unclear when it entered the northern European genepool. It could have been during the later bronze age of Scandinavia contemporary to the Scythian iron age, or during the migration era with Sarmatians or Huns. Two relevant TMRCAs are those of 3100bp, splitting all the European carriers on Yfull (one Ukrainian and one Bulgarian), and 2600 BP for a clade the majority of the Scandinavian carriers are under, which could be an indication this lineage was present in Scandinavia during their bronze to iron age transition period. It cannot be ruled out that multiple, already diverged Y2679 lineages entered Scandinavia at a later period and only survived there.

It is unfortunate there wasn’t a further subclade breakdown or that this sample wasn’t uploadedon Yfull, because that could have helped out with this mystery I just stumbled upon. A1804 probably got this paternal haplogroup from the Xiongnu+European side of his ancestry.

To get back to the topic, some of the early Avar samples here had no Pannonian Avar ancestry. Here are some examples:


Target: Avar_Early_Transtisza:I16812

Distance: 2.5707% / 0.02570658

40.0 RUS_Late_Xiongnu


14.6 CZE_Early_Slav

7.8 RUS_Alan_MA

5.8 DEU_MA_Alemannic

0.0 HUN_Avar_Early_Danube-Tisza_SIM

Y-DNA: Q1a2a1~ (YP799)
mtDNA: A+152+16362
Description: High status horseman warrior, the male was laid on a lamellar armour in the grave-pit


Target: Avar_Early_Transtisza:I20800

Distance: 2.5765% / 0.02576503

48.6 RUS_Late_Xiongnu


15.6 DEU_MA_Alemannic

7.2 RUS_Alan_MA

4.0 CZE_Early_Slav

0.0 HUN_Avar_Early_Danube-Tisza_SIM

Y-DNA: R1a1a1b2a2a3b~ (YP1456, YP1710)

mtDNA: C4d

Description: Average status burial containing a ceramic vessel, iron knife and four iron buckles.


Target: Avar_Early_Transtisza:I20801

Distance: 2.1299% / 0.02129902

81.8 RUS_Late_Xiongnu

18.2 Sarmatian_RUS_Caspian_steppe

0.0 HUN_Avar_Early_Danube-Tisza_SIM

y-DNA: -
mtDNA: Z3

Description:  Average status burial containing a ceramic vessel, bone comb, honing stone, iron buckle, spindle ring, animal bones



Target: Avar_Early_Transtisza:A1807

Distance: 2.4277% / 0.02427711

56.6 RUS_Late_Xiongnu

17.0 RUS_Alan_MA

11.2 CZE_Early_Slav


5.6 Sarmatian_RUS_Caspian_steppe

0.0 HUN_Avar_Early_Danube-Tisza_SIM

Y-DNA: Q1b1a3b (Y11941)

mtDNA: C4a1a4a

Description: Richly furnished burial containing a horse harness, gold sheets

What is interesting here is that we get a glimpse into the social dynamics of the Avars and their nomadic contemporaries. Some of these individuals had high status, as shown by their burials. Artificial cranial deformation was practiced by some of these samples. One high status individual had recent Avar ancestry but with a non-Avar paternal lineage. Other higher status individuals had no Avar ancestry at all. This might be reflective of how the remnants of the Huns, and entities such as the Sabirs, Kutrigurs and Utrigurs became involved within the Avar Khaganate. 

Alan admixed outliers

I noticed that several samples from both the early and late Avar sites seemed to consistently model as primarily being composed of Xiongnu and Sarmatian/Alanic ancestry, and often in near one-to-one ratios. These samples do not seem to have ancestry from the Pannonian Avars.


Target: Avar_Early_Transtisza:I16744

Distance: 1.9357% / 0.01935711

39.6 RUS_Late_Xiongnu

32.4 RUS_Alan_MA

12.4 Sarmatian_RUS_Caspian_steppe


7.6 CZE_Early_Slav

Y-DNA: - 

mtDNA: I4a

Description: Silver earring, golden bead relatively richly furnished


Target: Avar_Late_Transtisza:I20798

Distance: 2.8990% / 0.02898960

43.8 RUS_Late_Xiongnu

40.6 RUS_Alan_MA


7.0 CZE_Early_Slav

Y-DNA: - 

mtDNA: I4a

Description: ceramic vessel, silver earring, beads average furnished


Target: Avar_Early_South_Transdanubia:A1824

Distance: 2.0163% / 0.02016317

58.4 RUS_Alan_MA

28.0 RUS_Late_Xiongnu

13.6 Iberia_Northeast_Empuries2

Y-DNA: R1a1a1b1a2a (S205)
mtDNA: K1a17a
Description: The burial probably belonged to a highest ranking member of the local society, the grave goods suggest eastern contacts. The burial was richly furnished and had a belt, sword, and bow as grave goods.

I wonder if these are migrants from old Great Bulgaria, The combination of Late_Xiongnu and Alan_MA could point towards that direction.  The Alans that remained in the southern russian steppes adjacent to the Caucasus mountains during the Hunnic migrations likely would have joined or were assimilated into the early Bulgar populations that migrated westwards. There was a period of depopulation in the foothills Caucasus around the 8th century AD, and the later Alan presence associated with the Saltovo-Mayaki culture in steppe regions around the Don (also in Tatarstan!) were Alans from the Caucasus mountains which were relocated by the Khazars.

A1824 might have been a Caucasian Alan with Turkic ancestry, considering that the majority of his ancestry is Alan related. The burial being one of the highest ranking ones is certainly interesting too.  The rest are probably Bulgars.


Later Avars

The early Avars gave us a glimpse into the genomic ancestry of the Avar core as well as the other nomadic populations within the Avar Khaganate, in their early days. But fortunately we also have plenty of samples from the later Avar period, which I will also present here. I have opted to primarily present the individuals from the Danube-Tisza group, as I think these give the best view on the Avar core population.



Target: Avar_Middle_Danube-Tisza:A1812

Distance: 2.4284% / 0.02428426

92.2 HUN_Avar_Early_Danube-Tisza_SIM

6.8 RUS_Late_Xiongnu

1.0 Iberia_Northeast_Empuries2

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: F1b1f

Description: Richly furnished burial which contained gold earrings, golden hair clips, gold-foiled strap-ends with interlace pattern made of silver and a sword decorated with gold plates


Target: Avar_Middle_Danube-Tisza:I18742

Distance: 2.5477% / 0.02547727

95.6 HUN_Avar_Early_Danube-Tisza_SIM2

4.4 CZE_Early_Slav

Y-DNA: -

mtDNA: HV2a3

Description: Average burial, contained a male hair ornament in secondary position, fragments of iron objects. Originally assumed to have been a male.

Middle - late


Target: Avar_Middle-Late_Danube-Tisza:I18223

Distance: 1.9648% / 0.01964799

93.4 HUN_Avar_Early_Danube-Tisza_SIM

6.6 RUS_Alan_MA

Y-DNA: -

mtDNA: C4a1a4

Description: Rich burial iron clamps (coffin), pair of golden earring, spindle, iron knife, iron object


Target: Avar_Middle-Late_Danube-Tisza:I18224

Distance: 2.4271% / 0.02427121

89.2 HUN_Avar_Early_Danube-Tisza_SIM

9.6 ITA_Rome_Late_Antiquity

1.2 CZE_Early_Slav

Y-DNA: N1a1a1a1a3a (Y16310, Y16319)

Description: Poor burial, originally assumed to have been a female


Target: Avar_Middle-Late_Danube-Tisza:I18744

Distance: 2.9950% / 0.02994997

58.6 HUN_Avar_Early_Danube-Tisza_SIM


15.0 RUS_Alan_MA

6.8 DEU_MA_Alemannic

Y-DNA: N1a1a1a1a3a (Y16310, Y16319)

Description: Poor burial containing iron artefacts. The individual was a 11-12 year old male, and was sampled due to his pathological alterations.

As can be seen, there was a good amount of genetic continuity going on between the early and middle/late Avar samples, although they all seem to have a bit of ancestry that doesn’t come from the Avar core populations. While you have two samples that can be modeled as  being ~90% derived from the Avar core population, you also have one sample that has ~60%. The average of the Avar samples came out looking like this:

Target: HUN_Avar_Middle-Late_Danube-Tisza

Distance: 1.3305% / 0.01330494

71.8 HUN_Avar_Early_Danube-Tisza_SIM

9.6 RUS_Late_Xiongnu

8.4 RUS_Alan_MA


2.6 CZE_Early_Slav


These are the samples from the later period of the Avars, until their collapse in the late 8th/early 9th century AD. The average of all Danube-Tisza late Avars  was very similar to that of the Middle-late period, coming out like this:

Target: HUN_Avar_Late_Danube-Tisza

Distance: 0.8619% / 0.00861864

70.4 HUN_Avar_Early_Danube-Tisza_SIM

10.2 RUS_Late_Xiongnu

9.4 RUS_Alan_MA

7.8 ITA_Rome_Late_Antiquity

2.2 CZE_Early_Slav

Amongst the Late Danube-Tisza group you had samples with a strong genetic continuity of their early Avar predecessors. Interestingly both of these were women. 


Target: Avar_Late_Danube-Tisza:A1810

Distance: 2.2231% / 0.02223052

84.4 HUN_Avar_Early_Danube-Tisza_SIM


4.8 CZE_Early_Slav

3.4 RUS_Alan_MA

mtDNA: F1b1f

Description: This sample originates from the grave of an adult female. Sheep meat was placed into her grave. 


Target: Avar_Late_Danube-Tisza:A1811

Distance: 2.6369% / 0.02636864

98.2 HUN_Avar_Early_Danube-Tisza_SIM

1.6 DEU_MA_Alemannic

0.2 RUS_Alan_MA

Y-DNA: -
mtDNA: F1b1f

Age: This sample originates from the grave of an elderly female of poor/average standing. The burial contained goat/sheep bones.

This sample might be 100% continuous even, I’m not sure if that 1.8% is legit rather than noise. On the flip side you also had individuals which had acquired more of a mixed profile, such as this individual:


Target: Avar_Late_Danube-Tisza:A1809

Distance: 1.7518% / 0.01751821

62.0 RUS_Late_Xiongnu

24.8 HUN_Avar_Early_Danube-Tisza_SIM


2.4 RUS_Alan_MA

mtDNA: T1a1

Description: This sample originates from the grave of a mature female of poor/average standing. She was buried with sheep meat (bones) and wore some beads.

The rest of the Late Avar samples looked like this:


Target: Avar_Late_Danube-Tisza:A1813

Distance: 1.9577% / 0.01957736

51.2 HUN_Avar_Early_Danube-Tisza_SIM

31.0 RUS_Late_Xiongnu

16.8 RUS_Alan_MA

1.0 CZE_Early_Slav

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: C4a1a4

Description: Poor/average burial, grave goods were pottery, sheep/goat bones and a knife.


Target: Avar_Late_Danube-Tisza:A1814

Distance: 1.9581% / 0.01958080

72.0 HUN_Avar_Early_Danube-Tisza_SIM

14.4 ITA_Rome_Late_Antiquity

8.2 RUS_Alan_MA

5.4 RUS_Late_Xiongnu

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: T1a1+@152

Description: An elite male buried in a  richly furnished grave, containing a gold-foiled bronze belt.


Target: Avar_Late_Danube-Tisza:A1815

Distance: 2.2007% / 0.02200660

70.6 HUN_Avar_Early_Danube-Tisza_SIM

19.4 RUS_Alan_MA

6.4 Iberia_Northeast_Empuries2

3.6 RUS_Late_Xiongnu

Y-DNA: N1a1a1a1a3a (F4205)

mtDNA: C4a1a4

Description: This sample was of a male individual, whose grave was richly furnished (with gold-foiled cast

bronze belt set with griffin and floral ornament).


Target: Avar_Late_Danube-Tisza:I18225

Distance: 2.0962% / 0.02096155

76.2 HUN_Avar_Early_Danube-Tisza_SIM2

12.4 ITA_Rome_Late_Antiquity

11.4 CZE_Early_Slav

y-DNA: -

mtDNA: D4l2

Description: Grave of an adult female (30-39 years old). Rectangular burial pit. Traces of oval posts in the

four corners of the pit. The undisturbed skeleton lay on its back in an extended position. The burial contained a bronze earring, ceramic vessel, spindle, needle box and animal bones.


Target: Avar_Late_Danube-Tisza:I18222

Distance: 2.1965% / 0.02196514

58.6 HUN_Avar_Early_Danube-Tisza_SIM

25.8 RUS_Alan_MA

8.0 RUS_Late_Xiongnu

7.4 Iberia_Northeast_Empuries2


Y-DNA: N1a1a1a1a3a (F4205, Y16310)

mtDNA: H8c

Description”: Adult male buried in a richly furnished grave containing bone plates of a bow, wooden fragments, belt set, iron knife and a ceramic vessel.

During the later Avar period there is a good amount of continuity, although there certainly were some genetic exchanges occurring. It seems to me that there was a lot more social mobility within the various steppe nomadic populations, as we Avar ancestry mixed with Hunnic and Alanic populations, similar to what the authors of the article suggested:

During the late Avar period, we observe a shift among the elite in the Avar core area toward a more recently admixed ancestry. Even if late Avar individuals still preserve a predominant northern East Asian component, the western Eurasian source that best fits the remaining 20%–30% of their ancestry is mostly a non-local one (i.e., it does not match the gene pools of the available preceding Carpathian Basin populations). Instead, it rather matches the steppes north of the Caucasus, although the scarceness of comparative data from the steppe in the 1st millennium CE calls for a future investigation of possible better alternative sources.

However I think there definitely is an influx of European ancestry in these late Avars, although it is minor and it is quite likely that much of this was mediated by steppe nomads with a mixed genetic profile, of which there were plenty.

What I also found noteworthy is that you had relatively unmixed Avar samples until quite late, and these were found in both wealthy and poor burials. Likewise, the N-F4205 lineage so prominently featured amongst the Pannonian Avars was found in late Avar lower status burials as well. This suggests that the Avar core were not just the elites in their society, but also that commoners or lower classes could have come from the Avar core population. Not to mention that plenty of non-Avaric people also had attained high standing within the Avar society.

The Rouran connection

Now that we have gone through all the Avar samples, it is time to speculate on their origins. Many origins have been proposed for the Pannonian Avars, but the most common theory is that the Pannonian Avars were the remnants of the Rouran Khaganate elites, which fled westwards and founded a new Khaganate. This theory was first proposed by 18th century French Turkologist Joseph de Guignes, a man who certainly was ahead of his time. I found an old archived copy of his book where he proposes this connection:

Histoire générale des Huns, des Turcs, des Mogols, et des autres Tartares occidentaux, &c. avant et depuis Jésus-Christ jusqu'à présent : précédée d'une introduction contenant des tables chronol. & historiques des princes qui ont regné dans l'Asie : Guignes, Joseph de, 1721-1800

The theory was sound, but an issue always remained that we did not have textual attestations of the language of the Avars, nor did we have any from the Rouran Khaganate. However, a few years ago was an uncovery of the oldest Mongolic inscriptions, the Brāhmī Bugut and Kuis Tolgoi inscriptions, presented by the late Alexander Vovin, who unfortunately has recently passed away. Dating to 584-587 AD and 602-620 AD respectively, it is likely that the language inscribed on these stelae was also the language spoken by the Rouran Khaganate. You can read more about the inscriptions in this article by Alexander Vovin: A Sketch of the Earliest Mongolic Language: the Brāhmī Bugut and Khüis Tolgoi Inscriptions

The Brāhmī Bugut inscription

The Khüis Tolgoi inscription

Now while we don't have linguistic evidence for the Pannonian Avars at hand, we do have plenty of genetic evidence and it shows a population with a very high amount of XIanbei-related ancestry. One thing to note is that while these samples to me clearly indicate a population from the greater Mongolian plateau migrating to the west and forming the Avars, this in itself is no confirmation of the original hypothesis regarding the Avar-Rouran connection, which postulated that the appearance of the Avars was linked to the elites of the Rouran Khaganate fleeing the Gokturks when their Khaganate collapsed.

The Rouran Khaganate was founded by a confederation of tribes led by the Yujiulü clan, and as the confederation grew it also became multiethnic. So not everyone in the Rouran Khaganate were the same people, as many different tribes of Mongolic, Turkic or even other ethno linguistic origins became part of the Rouran Khaganate.

The Yujiulü clan first appeared in history within the context of the Tuoba Wei dynasty. It began with the mythological founder called Mugulü, who according to Chinese sources was a slave, although not all modern historians agree with this. Mugulü became a free man and a warrior under the Xianbei. However due to unfortunate circumstances he was set for a beheading, which Mugulu escaped by crossing the Gobi Desert and finding refuge amongst the Gaoche, together with hundred other fugitives. It was his son that founded the Rouran, but it was not until the 5th century AD that the Rouran gained independence under Yujiulü Shelun, who then proclaimed himself “Khagan”.  An interesting parallel is that both the mythological Mugulü and historical Shelun crossed the Gobi Desert and united tribes along the way.

From history we know that the Rouran were quite involved in the Heqin policy, and they should’ve had input from the northern Wei dynasty and the northern Yan dynasty through marriages. One dynasty was lead by the Tuoba Wei, the other by remnants of the Xiongnu, but both dynasties and upper class would have been culturally quite Sinicized and due to intermarriages could’ve carried a decent amount of Han-related ancestry If the Avars were the descendants of the Rouran elite, their ancestry should reflect these marriage practices.

Such southern affinities seem to be lacking in the early Avar samples, who on the other hand had a genetic profile that seems to suggest a geographically more northern origin with ancestry a-typical for people around or below the Gobi desert. The one Y-chromosome lineage we have from a Rouran context was C-F3830, whereas the Avars are under N-F4205. The C-F380 subclade has also been found in Xianbei, Xiwei and Mongol period samples in Mongolia and Inner Mongolia. [1] Although it should be pointed out that this ‘Rouran’ sample TL1 actually seems to postdate the Rouran khaganate as it was dated to 590-655 AD. It would be quite likely that the Yujiulü lineage also fell under this clade, and all the thirteen rulers of the Rouran khaganate had the same patrilineal lineage.

One of the samples from Damgaard 2018 also was under N-F4205, specifically the N-F22331 subclade, which Avar sample A1814 falls under as well. And as you can see here his profile was very similar to the Pannonian Avars:

Target: KAZ_Nomad_HP:DA95

Distance: 1.9499% / 0.01949946

29.8 CHN_Amur_River_Xianbei_IA

29.0 MNG_Xianbei_IA

17.4 RUS_Late_Xiongnu

15.2 Saka_Tian_Shan

8.6 RUS_Krasnoyarsk_BA

Thi sample might have some consequences for the potential Rouran-Avar connection. It's certain that this individual at least shares a paternal ancestor with the Avars going back to at least 178  BC. Going by his autosomal ancestry, there probably was more ancestor sharing than that single paternal ancestor too.

DA95 was from a period contemporary to the foundation of the Rouran Khaganate, dating to 331- 354 AD and was found beyond what would have been the borders of the Rouran khaganate, being buried on the right banks of the Irtysh river, north of Pavlodar, Kazakhstan. The ventures of the ROuran Khaganate in Central Asia that I am aware of postdate the Rouran campaigns in Central Asia, such as those against the Wusun.

I can’t say what the context is behind that sample being buried around the Irtysh river, but it could be an indication that populations similar to him were already on the move towards the west before the collapse of the Rouran Khaganate. Although looking at the genomic profile of the Avar samples, they do not exactly strike me as people who moved from east to west over a period of two centuries.

I think it is quite likely that the Pannonian Avars would have been part of the Rouran Khaganate given the territory the Khaganate encompassed, but this would not be the same thing as them being “the Rouran” elite or core. Given the wealth of the early Avars, it is possible they were a prominent tribe within the Rouran Khaganate however, but if so they would likely have been situated on the northern territories of the Rouran Khaganate. The steppes around Lake Baikal could make sense for a geographic origin, as this would border the Siberian forest regions where you’d find populations with Krasnoyarsk_BA related ancestry as well as paternal lineages under N-L1026.

It is possible that the migration of the Avars was linked to the demise of the Rouran Khaganate. The Rouran Khaganate ultimately collapsed in 550 AD, but it did not fall in a day and collapsed due to a series of military defeats and internal struggles. The collapse of the Rouran is also what lead to the first Turkic Khaganate. The Pannonian Avars are first attested in 552 AD, and while it certainly is possible they only migrated westwards around 550 AD, it may also be that the Avar core had started their migration earlier. This will currently remain an unsolved matter as ancient genetics cannot clarify this matter.

The authors of this article seem to suggest that the genetic evidence provided show that the Avars were descendants of “the Rouran”:

What happened then is described in reports based on diplomatic exchanges between the Turks and Byzantium. The Turks denied that the European Avars were directly descended from the Rouran, which would have given their khagans the ancient legitimacy of the Rouran khaganate. They claimed that those who had fled Turkish expansion in the 550s were a mixed people called Warchonites, composed of groups of former Rouran subjects, mainly Ogurs who had lived in western Central Asia. They had only adopted the prestigious name Avars to frighten those whom they encountered on their westward migration. Indeed, we know from Chinese sources that many among the Rouran had been killed by Turks and Chinese, and others had fled eastward, and reputedly ended up in Korea. Still, it is not unlikely that we can assume that groups from the Western Eurasian steppes took part in the Avar migration, and that at least the Avar khagans and their core group were actually descended from the Rouran. 

In the more loose interpretation of the Rouran-Avar connection there certainly is a connection to be made in that their origins geographically lie within the Rouran khaganate and that these people would have been Mongolic speaking like the Rouran probably were.  But I don’t think you can make claims beyond that. 

The original proposal, where the Avars were the Rouran elites fleeing the Gokturks to set up a new Khaganate in the west, does not seem substantiated by the genetic data we have here if you consider the finer details. And if so then the anger the Gokturks had towards the Avar rulers calling themselves Khagan is perhaps more reasonable than typically depicted.

Funnily enough this mirrors my stance on another of De Guignes’ theories, linking the Xiongnu to the European Huns. To me it seems pretty clear that the European Huns were primarily comprised of an Oghuric Turkic speaking population, and such a population would undoubtedly have been part of the Xiongnu Empire and identify as Huns due to that, however I don’t think there is sufficient evidence to suggest that there was some kind of singular “elite” of the Xiongnu which migrated westwards after the collapse of the Northern Xiongnu to become the Huns of Attila. However given the lack of sufficient data in regards to the early Xiongnu confederation as well as the European Huns this still has a possibility to be correct, whereas with the Avars we now have quite substantial genomic data. It sure helps that the core was so homogenous.

Avar-Hungarian continuity hypothesis

Another theory related to the Pannonian Avars was the  Avar-Hungarian continuity hypothesis proposed by Hungarian historian  Gyula László in the 1970s. I won;t go into the finer details of this hypothesis, but László used a combination of toponyms/hydronyms, history and archaeology to come to the conclusion that the Magyars were not Hungarian conquerors, but Turkic conquerors. With the real Hungarians having arrived in the 7th century AD, represented by the later Avar culture. Supporters of this theory have also provided other suggestions such as the Hungarians being subjugated tribes that came in with the Pannonian Avars as they migrated westwards in the 6th century AD.

As mentioned before, the N-F4205 lineage is downstream of N-L1026, a lineage commonly associated with Uralic speakers. And they seemingly also seem to carry a northeast Asian genetic component that is similar to the northeast ancestry common amongst Uralic people, in addition to their ancestry from the eastern steppes. Now, for the people that think this gives credence to the Avar continuity hypothesis, keep a couple of things in mind:

  • All of the Avars here have the same patrilineal lineage of N-F4205. Going by yfull, this lineage seemingly separated from lineages ancestral to currently Uralic speaking populations around 2700 BC.

  • The N-F4025 has a gap between its foundation and most recent common ancestor of a few thousand years, which suggests it was a minor lineage until it had its shining moment. After its shining moment it seemingly was only carried by Mongolic populations and later also by Turkic populations.

  • Hungarian is a Ugric language closely related to Khanty and Mansi, whose Uralic associated paternal lineages fall under completely different branches of N-L1026, as well as N-P43.[3] [4]

  • The traditional continuity theory never suggested that the Avar elites were of Hungarian origin, and one prominent version was that the later Avar culture corresponded to the coming of the Hungarians. Yet we can see there is a strong genetic continuity between earlier and later Avar elites in both autosomal ancestry and paternal lineages.

  • The genetic samples of the Hungarian conquerors on the other hand definitely have a genetic affinity to Siberian Ugric speaking people, and were argued to have been derived from a combination of Mansi-like peoples, Sarmatians and early Turkic peoples, entirely in line with the linguistic data of the Hungarian language. [5]

Is it possible that the long-distant paternal ancestor of the Pannonian Avar samples was Uralic speaking or spoke a language related to Uralic? Yes, it certainly is. It is also possible that the paternal lineage broke off quite a bit before Proto-Uralic itself had developed, and that this lineage then was assimilated into an eastern steppe population that spoke something akin to Pre-Proto-Mongolic, which given the data shown is a scenario that is more likely.

If the argument is that due to the consistent signal of paternal lineages and autosomal ancestry that are not of (Pre-Proto-)Mongolic origin the Avars should be linguistically tied to that component rather than the 80-90% autosomal ancestry clearly tied to Mongolic populations, that’s fine too. However in such a situation their language would not be Hungarian, and quite likely not even Proto-Uralic but rather something like Pre-Proto-Uralic based on the arguments given above.

Speaking of Hungarians, this pre-print just came out:

Tracing genetic connections of ancient Hungarians to the 6-14th century populations of the Volga-Ural region

Thus to summarize it all,

We have gotten a lot closer to the centuries old riddle “who were the Pannonian Avars?”. De Guignes was more right than he was wrong, although I certainly think you could argue de Guignes was not entirely correct in his theory of linking the Avars to the Rouran elites. Given the autosomal ancestry of the Pannonian Avars, they likely originated in a more northern area of the Mongolian plateau, and given that practically all Avars were under the same patrilineal lineage in my eyes it is quite likely they could’ve originated from the same tribe.

After their arrival in the west they quickly established themselves as a dominant political entity, however this core population was not limited to just the elite class of the Avar society as shown by lower class samples genetically identical to the elite samples. The integration of other nomadic populations seems to have been an important element too as quite some non-Avaric samples had prominent burials.

While there definitely was intermixing going on, it is interesting that there was a relative degree of continuity between the earlier and later Avar period, and it seems like most of the mixing occurred between steppe nomads, although many of these could’ve had admixture from diverse non-steppe populations. Despite the early non-Avar nomadic samples with elite burials, the late Avar samples were still absolutely dominated by their original paternal lineage under N-F2405.


  1. Li J, Zhang Y, Zhao Y, Chen Y, Ochir A, Sarenbilige, Zhu H, Zhou H. The genome of an ancient Rouran individual reveals an important paternal lineage in the Donghu population. Am J Phys Anthropol. 2018 Aug;166(4):895-905. doi:10.1002/ajpa.23491. Epub 2018 Apr 21. PMID: 29681138.

  2. Kılınç, G. M., Kashuba, N., Koptekin, D., Bergfeldt, N., Dönertaş, H. M., Rodríguez-Varela, R., Shergin, D., Ivanov, G., Kichigin, D., Pestereva, K., Volkov, D., Mandryka, P., Kharinskii, A., Tishkin, A., Ineshin, E., Kovychev, E., Stepanov, A., Dalén, L., Günther, T., Kırdök, E., … Götherström, A. (2021). Human population dynamics and Yersinia pestis in ancient northeast Asia. Science advances, 7(2), eabc4587.

  3. Post, H., Németh, E., Klima, L. et al. Y-chromosomal connection between Hungarians and geographically distant populations of the Ural Mountain region and West Siberia. Sci Rep 9, 7786 (2019).

  4. Pimenoff, V., Comas, D., Palo, J. et al. Northwest Siberian Khanty and Mansi in the junction of West and East Eurasian gene pools as revealed by uniparental markers. Eur J Hum Genet 16, 1254–1264 (2008).

  5. Maróti, Z., Neparáczki, E., Schütz, O., Maár, K., Varga, G. I. B., Kovács, B., Kalmár, T., Nyerki, E., Nagy, I., Latinovics, D., Tihanyi, B., Marcsik, A., Pálfi, G., Bernert, Z., Gallina, Z., Horváth, C., Varga, S., Költő, L., Raskó, I., … Török, T. (2022, January 1). Whole genome analysis sheds light on the genetic origin of Huns, Avars and conquering Hungarians. bioRxiv. Retrieved May 6, 2022, from 


  1. Copper Axe, the absence of descent Han related ancestry in any extant Xiongnu samples is also interesting given that there were many marriage policies between Han empire and Xiongnu between 200 BC and 135 BC before Wu took aggressive policies against Xiongnu.

    1. We actually have quite a bunch of Xiongnu related samples that had some Han related ancestry, in addition to two samples that probably were Han (captives, defectives?). Keep in mind that many of the samples we have are just from your average people, who wouldn't have been involved in the Heqin policies anyways.

  2. This comment has been removed by the author.

  3. Decent effort mate.
    Minor comment - ''N-F4205 is definitely a Mongolic lineage in terms of modern day distribution,''. I would say that N-F4205 is in fact uncommon in Mongols (who are predominanlty Y-hg C), apart from aforementioned Buryat, meaning they are recent migrants to Mongolia.
    Overall, whilst i applaud their efforts, one of the points the authors try to force (Avars = Rouran) isn't supported. So the real question is where ''lineage (Avar y-hg N) was assimilated into an eastern steppe population''. Could be outside Mongolia, given that NEA ancestry was already in east-central Kazakhstan by 250 BC (Damgaard 2018)

    1. Thanks bro! What I meant was that if you look at the carriers of N-F4205 today, it is mostly people who either are Mongolic or were of Avar/Mongol descent. As far as the region of assimilation, I'd say around lake Baikal would make more sense. It would also explain why eastern Buryats have the most of it and you have the perfect border region between forested areas and the steppes. The NEA ancestry in Kazakhstan you are referring to I assume is from the Korgantas period? These populations had a bit of a different profile compared to the Avars, far more similar to the Slab Grave samples from Mongolia.

      "one of the points the authors try to force (Avars = Rouran) isn't supported" Yeah I think both the historians and the geneticists of this article should have thought a bit harder about this link.

  4. Also for the people wondering how my fight went, I won in the fist round by TKO after dropping my opponent with a pushkick to the face and messing up his nose. I'm the baddest blogger around guys!