search this blog

Thursday, July 5, 2018

SMBE 2018 abstracts

Abstracts from the upcoming SMBE 2018 conference (8-12 July) are now available HERE. Below are a few that I found interesting. Emphasis is mine. Feel free to post your own favorites in the comments.

The first Epipaleolithic Genome from Anatolia suggests a limited role of demic diffusion in the Advent of Farming in Anatolia

Feldman et al.

Anatolia was home to some of the earliest farming communities, which in the following millennia expanded into Europe and largely replaced local hunter-gatherers. The lack of genetic data from pre-farming Anatolians has so far limited demographic investigations of the Anatolian Neolithisation process. In particular, it has been unclear whether farming was adopted by indigenous hunter-gatherers in Central Anatolia or imported by settlers from earlier farming centers. Here we present the first genome-wide data from an Anatolian Epipaleolithic hunter-gatherer who lived ~15,000 years ago, as well as from Early Neolithic individuals from Anatolia and the Levant. By using a comparative dataset of modern and ancient genomes, we estimate that the earliest Anatolian farmers derive over 90 percent of their ancestry from the local Epipaleolithic population, indicating a high degree of genetic continuity throughout the Neolithic transition. In addition, we detect two distinct waves of gene flow during the Neolithic transition: an earlier one related to Iranian/Caucasus ancestry and a later one linked to the Levant. Finally, we observe a genetic link between Epipaleolithic Near-Easterners and post-glacial European hunter-gatherers that suggests a bidirectional genetic exchange between Europe and the Near East predating 15,000 years ago. Our results suggest that the Neolithisation model in Central Anatolia was demographically similar to the one previously observed in the southern Levant and in the southern Caucasus-Iran highlands, further supporting the limited role of demic diffusion during the early spread of agriculture in the Near East, in contrast to the later Neolithisation of Europe.


Demographic processes in Estonia from Bronze Age through Iron Age to Medieval times

Metspalu et al.

N3 and R1a are the two most common Y chromosome haplogroups among modern Estonians. R1a appears with Corded Ware culture but the arrival of hg N has not been determined. To this end we have extracted and studied aDNA from teeth of 18 individuals bracketing the changes in the material culture in the end of the Bronze and early Iron Age. We find N3 in Iron Age but not in Bronze Age. Due to the small sample size we cannot refute the existence of hg N in the latter. In genome wide analyses the Bronze Age and especially Iron Age samples appear very similar to modern Estonians implying population continuity. Christianization (13 cc AD) established a new elite of West European origin, which presumably had an impact on the genetic structure of the local population. To investigate this we extracted DNA from teeth of 35 individuals, who have been uncovered from both rural (considered local Estonian population) and town (likely of West European origin) cemeteries of Estonia. We compared the low coverage genomes with each other and with relevant modern and ancient Estonian and other European populations. We find that there is a clear discontinuity between the elite and common people, where the former group genetically with modern German samples and the latter with modern Estonians. We do find three individuals of mixed genetic ancestry. But importantly we do not see a steady shift of either local population strata, which suggests limited contact between the elite and the common people.


Genetic transition in the Swiss Late Neolithic and Early Bronze Age

Furtwaengler et al.

Recent studies have shown that the beginning of the Neolithic period as well as final stages of the Neolithic were marked by major genetic turnovers in European populations.The transition from hunter-gatherers to agriculturalists and farmers/farming in the 6 th millennium BP coincided with a human migration from the Near East. In the 3 rd millennium BP a second migration into Central Europe occurred originating from the Pontic steppe linked to the spread of the Corded Ware Complex ranging as far southwest as modern day Switzerland. These genetic processes are well studied for example for the Middle-Elbe-Saale region in Eastern Germany, however, little is known from the regions that connect Central and Southern Europe. Here, we investigate genomic data from 69 individuals from the Swiss Plateau and Southern Germany that span the transition of the Neolithic to the Bronze Age (5500 to 4000 BP). Our results show a similar genetic process as reported for the Middle-Elbe-Saale region suggesting that the migration from the Pontic steppe reached all the way into the Swiss plateau. The high quality of the ancient genomic data also allowed an analysis of core families within multiple burials, the determination and qualification of different ancestry components and the determination of the migration route taken by the ancestors of the Late Neolithic populations in this region. This study presents the first comprehensive genome wide dataset from Holocene individuals from the Swiss plateau and provides the first glimpse into the genetic history of this genetically and linguistically diverse region.


Genome-Wide Ancient DNA Portrays the Forming of the Finnish Population Along a 1400-Year Transect

Majander et al.

The Finnish population has long been a subject of interest for the fields of medical and population genetics, due to its isolation-affected genetic structure and the associated unique set of inherited diseases. Recent advances in ancient DNA techniques now enable the in-depth investigation of Finland's demographic past: the impact of migrations, trade and altering livelihood practices. Here we analyse genome-wide data from over 30 individuals, representing ten archaeological burial sites from southern Finland, that span from the 5th to 19th century. We find the historical individuals to differ genetically from Finns today. Comparing them with surrounding ancient and modern populations, we detect a transition from genotypes generally connected with prehistoric hunter-gatherers, and specifically resembling those of the contemporary Saami people, into a more East-Central European composition, associated with the established agricultural lifestyle. Starting from the Iron Age and continuing through the Early Medieval period, this transition dates remarkably late compared to the respective changes in most regions of Europe. Our results suggest a population shift, presumably related to Baltic and Slavic influences, also manifested in the archaeological record of the local artefacts from the late Iron Age. Our observations also agree with the archaeological models of relatively recent and gradual adoption of farming in Finland.


Population migration and dairy pastoralism on the Bronze Age Mongolian steppe

Warinner et al.

The steppe belt that extends across Eurasia was the primary corridor of Eneolithic and Bronze Age migrations that reshaped the genetics of Europe and Asia and dispersed the Indo-European language family. Beginning in the Eneolithic, a new and highly mobile pastoralist society formed on the Western Steppe. These steppe herders expanded both westwards, contributing to the Corded Ware culture of Eastern and Central Europe, and eastwards, contributing to the mobile pastoralist Afanasevo, Sintashta, Andronovo, and Okunevo cultures in Central Asia. The eastern extent of this Western Steppe herder expansion is not well defined. Here we investigate genome-wide ancestry data obtained from 20 Late Bronze Age (16th-9th century BCE) khirigsuur burials from Khovsgol, Mongolia and further investigate evidence for dairy pastoralism by LC-MS/MS analysis of dental calculus. Overall, we observe limited Western Steppe gene flow into Late Bronze Age Mongolia, but adoption of Western ruminant dairying by ca. 1500 BCE.


The Transition to Farming in Eneolithic (Copper Age) Ukraine was Largely Driven by Population Replacement

Schmidt et al.

The transition to a farming-based economy during the Neolithic happened relatively late in southeastern Europe. Material changes occurred through pottery manufacture, but it wasn't until the sixth millennium BCE that farming was adopted by the Cucuteni-Trypillian archaeological complex (4800-3000 BCE). In many parts of Europe, early farmers who were descended from Anatolian migrants slowly admixed with local hunter-gatherers over the course of the Neolithic. In Eastern Europe and the Balkans, this process may have been more complex since early farmers would likely have admixed with local groups prior to spreading into continental Europe. Studies from the Baltic and Estonia suggest little genetic input from early farmers or continuous admixture with hunter-gatherers. Here, we investigate the impact of Trypillian migrations into Ukraine through the analyses of 19 ancient genomes (0.6 to 2.1X coverage) from the site of Verteba Cave. Ceramic typology and radiocarbon dating of the cave indicate continuous occupation from the Mesolithic to the Medieval Period, with peak occupation coinciding with the middle to late Tripolye. We show that the Trypillians replaced local Ukrainian Neolithic cultures. Also, hunter-gatherers contributed very little ancestry to the Trypillians, who are genetically indistinct from early Neolithic farmers. The one exception is a female that has mostly steppe-related ancestry. Direct radiocarbon dating of this individual places her in the the Middle Bronze Age (3545 years before present). Her lack of farmer ancestry suggests abrupt population replacement resulting perhaps from inter-group hostilities or plague that spread through Europe during the Late Neolithic.

See also...

Ahead of the pack

Genetic borders are usually linguistic borders too

Yamnaya isn't from Iran just like R1a isn't from India


Samuel Andrews said...

So the EEF signal is at least 15,000 years old. They didn't move deep into Europe until 7,500ya. That is enough time for diversity to develop in Anatolia before moving into Europe.

Farmer waves into Europe could have brought a "diverse" collection of "distinct" farmer populations.

Davidski said...

In regards to that first abstract about the Epipaleolithic Anatolian genome, I said ages ago that western Anatolian Neolithic farmers had some Caucasus-related ancestry, but it was only minor.

CHG admixture in early western Anatolian farmers

So that was spot on.

Davidski said...

By the way, this is also one of the talks at SMBE 2018.

Population Dynamics at Late Chalcolithic and Early Bronze Age Arslantepe, Anatolia

But I've already blogged about it here...

How relevant is Arslantepe to the PIE debate?

Arza said...

Finally some news about Tollense:

Reliable Inference of Genetic Diversity within and between Ancient and Modern Genomes

Vivian Link et al.

Comparing ancient and modern genomes provides insights into many aspects of population history. However, such comparison are complicated by particularities of ancient DNA (aDNA) such as post-mortem damage (PMD) and low endogenous DNA content resulting in low sequencing depth. In addition, modern data may be a poor reference for ancient diversity, and the analyses of aDNA should thus not rely on modern data to avoid biases.

Here we introduce a probabilistic framework that accounts for these biases to accurately infer within and between individual genetic diversity even from genomes with median depth <1x, as we show using simulations and by downsampling. We achieve this accuracy by explicitly modeling PMD, and by carefully recalibrating base quality scores with a new method that does not rely on modern data, but exploits homozygous regions in the genome. Importantly, our method also allows for an unbiased clustering of heterochronous individuals using Multi-Dimensional Scaling, rather than by projecting ancient individuals onto a PCA spanned by modern diversity. Finally, our framework also allows for genotype calling, which we found to be unbiased and more accurate for aDNA than GATK.

To illustrate the power of our framework, we studied the diversity among soldiers from a colossal Bronze-age battlefield in northern Europe at the banks of the Tollense River in northern Germany. Our findings suggest that these soldiers, while from a large geographic area, likely represented a single ethnic group.

Samuel Andrews said...

Congrats on figuring that out David and the European farmer admixture in Yamnaya. Using G25, it looks like EEF's eastern ancestry is as basal or more basal than Iranian farmers.

Samuel Andrews said...

From the Furtwaengler et al abstract
"This study presents the first comprehensive genome wide dataset from Holocene individuals from the Swiss plateau and provides the first glimpse into the genetic history of this *genetically and linguistically diverse region*."

I see papers use the same exciting descriptives for multiple places. Like, going back this far in time most regions will seem diverse. Ancient DNA has raised the bar for how complex a place's history has to be for it to be considered diverse when compared to the rest of the world.

Another thing. We have confirmation of strong genetic continuation in Anatolia between 15ky and 7ky, in Russia from 11ky to 6ky, in Ukraine from 11ky to 6ky, in Sweden from 9ky to 5ky, in western Europe from 14ky to 7ky, in the Caucasus from 13ky to 8ky.

For 4,000-7,000 year periods in the Mesolithic to early Neolithic we see good amounts of continuation in many places. Maybe, the admixture events seen from 6000 BC onwards are out of the ordinary.....

AWood said...

At some point there has to be an upper bound to the accidental discovery of agriculture. Obviously the process was not "demic" since it had to start somewhere. If the paleolithic Anatolians look similar to the neolithic/agriculturalist ones from a genetic standpoint, this shouldn't come as a surprise. It's not like it was brought by aliens.

Ric Hern said...

This is particularly interesting :

"Finally, we observe a genetic link between Epipaleolithic Near-Easterners and post-glacial European hunter-gatherers that suggests a bidirectional genetic exchange between Europe and the Near East predating 15,000 years ago."

Wonder if R1b(V88)s Ancestors were among them ? And I wonder if Villabruna was a sign of a Migration who fanned out in different directions with some migrating into Anatolia Pre-15 000 years ago ?

And this:

"By using a comparative dataset of modern and ancient genomes, we estimate that the earliest Anatolian farmers derive over 90 percent of their ancestry from the local Epipaleolithic population, indicating a high degree of genetic continuity throughout the Neolithic transition."

Wonder if the migration from Iran displaced some Locals ? Maybe R1b(V88) who ended up in the Sahara ?

Ric Hern said...

So it looks like Trypillians were issolated for a fairly long time ? Maybe a later migration from Anatolia hugging the Black Sea Coast ?

Ric Hern said...

@ Davidski

Looks like most of your predictions were in or near the Bullseye. Congratulations.

Matt said...

The new goat paper is worth checking out -

Main figures of interest:

Goat pops of Europe (Ireland+France), China and Africa (Morocco+Togo) today reflect the early Neolithic populations of domesticated goats in Anatolia/Balkans, Iran and the Levant respectively.

Within the greater Middle East, Anatolia and the Levant shift to having more Iranian Neolithic like goat populations in the Chalcolithic-Bronze Age, while Iran has a smaller introduction of what looks like Levant Neolihic and then a reduction in relatedness to Iranian Neolithic goats again more recently.
In relation to expansions from steppes into Europe, it looks like this indicates that, if steppe groups had Eastern Neolithic goats, as which seems reasonably likely based on the Armenian record but by no means necessarily, as steppe groups could have adopted Western Neolithic goats via Ukraine, then they did not bring these into Europe in major numbers or switched breeds. That would not be too surprising as I think cows show some continuity of ancestry as well, at least, and populations expanding from steppe seem like they copied or adopted or stole pottery styles, similar weapon forms (e.g. CW battle-axes and pots with links to Funnelbeaker?), so animals could well be the same.

Mikkel said...

Fascinating stuff indeed!
By the way, I recently launched an interactive time map visualization of prehistoric migrations shown through 1768 ancient individuals (from 41 aDNA articles) from 8200 - 1 BC.
It also features mapping of almost 200 archaeological cultures (an attempt at mapping them at least, but many are still missing), and early confirmed wool and wheels, as well as more uncertain indications of these.
I have tried to colorgroup the ancient individuals according to their genetic clustering in the different articles, highlight the steppe-related R1 haplogroups (by defining SNP number).
check it out at :)

Davidski said...

@Mikkel Nørtoft

Nice one!

I almost lost hope that I'd see anything, but it's looking good.

Mikkel said...

Thanks a lot! There are even a few references to some of your posts hidden in there ;) If you have any corrections, I'm all ears :)

And yes, the data is a bit heavy, so it might take a few seconds to load.

Unknown said...

Hello, Davidsky, i tried to google the paper on Estonia, but failed, can you please provide the link to this study? I'm very interested as i partially descent from Baltic Germans.

Unknown said...

Is there any place i can read more on Estonian paper? I tried to google it, but got no result.

supernord said...

Interesting, in Schmidt et al. data from Verteba Cave is there any new data or data that has already been published?

bellbeakerblogger said...

Mikkel, the map is awesome. Ironically (maybe), you chose the color red for the steppe component which splashes across Northern Europe all at once. It's one thing to know the data, another to see it visually depicted in a timescale with actual, dated samples.
It's very powerful.

Chad said...

Epipaleolithic Anatolians and onto Barcin all have a lot of CHG/Iranian -related ancestry. Probably as much as the do Natufian-related. Maybe even more. It's only a minor addition of Iranian-related between epipaleolithic to later Neolithic.

Folker said...

By itself, the paper about Cucuteni-Trypillia is only a confirmation. We already have some results for the Verteba cave:

But the Verteba Cave is in Western Ukraine, so its population could have less admixture than in Eastern Ukraine.

Anyway, given the size of CT population, it is no surprise to find little admixture in them. With settlements of more than 10 000, with some possibly of neraly 50 000 (perhaps 15 000 houses), its overall population was probably more than 1 or 2 millions, and could be at its height reached 3/ millions, perhaps more. With a "spider web network" of settlements covering between 200 and 300 000 km², difficult to not have some homogeneity in the population.

Ric Hern said...

@ Mikkel Nørtoft

Nice Map.

Where are the Irish samples ?

Davidski said...

@Иван Иванов

The Estonian paper hasn't been published yet. You'd have to go to the SMBE talk to learn more about the results, or try e-mailing the authors for the preprint.

Matt said...

Samuel Andrews: For 4,000-7,000 year periods in the Mesolithic to early Neolithic we see good amounts of continuation in many places. Maybe, the admixture events seen from 6000 BC onwards are out of the ordinary.....

Could well be, though kinda seems hard to work out if we'll ever really know what was typical or atypical.

On this line of thought, adna has been confounding to some expectations in that we see sometimes see whole scale replacements before agriculture (e.g. quite possibly in Ice Age Europe, though that depends a lot on the parameters, but at least of Ust Ishim and Oase1), while sometimes hunter gatherer populations can be surprisingly more robust against agricultural expansions than was expected.

25% HG ancestry in north+central+west EuropeMNChl, and similar in SE Asia and Japan, and much higher in lots of other conditions, such as what looks like diffusion of pastoralism across the Caucasus by the Steppe_Eneolithic groups, or farming in Anatolia, and survival by AASI in South Asia.

It looks like the typical thing during the agricultural and pastoral expansions seems most often survival of a 10-20% (even where it looked like ANE disappeared following Yamnaya expansion, Steppe MLBA East showed that not to be so and actually to be about a combination of European and ANE geneflow, or Taforalt showed survival of the ancient North Africans until today and so on). At least on broad group scale, even if there is complete local replacement. Then there are the exceptions, like we see highlighted here.

That's mostly replacement, but 10-20% is still much more than some people were thinking less than 10 years ago in terms of demic advance models of agriculture, where Neolithic populations were thought to have on the order of >100x the population of HGs!

Wondering afresh about Mathieson's tweet that he suspects Greece and the Southern Balkans to have mostly been a continuation of the Anatolian population (; if so then that probably will end up somewhat revising downwards the estimates of Neolithic era replacement in SE Europe where replacement seemed strongest across Europe.

Samuel Andrews said...



Early Neolithic farmers in Europe have less IranNeo-like ancestry than Barcin. IberiaEN has the most. IberiaN's fit improves when artificially reduce Wezmeh_Cave ancestry from BarcinN. IberiaN's fit stopped improving after I reduced 8% Wezmeh_Cave from BarcinN.

Barcin minus 8 percent Wezh (IranNeo)-86.2

When, IranNeo-type ancestry in Barcin is taken away signifcant ElMiron admixture pops up in several Iberian farmer populations.

Davidski said...


Global25 coordinates for some of the Guanches and one of the Moroccan Neolithic samples are available here. Apparently they work fine, but try them out yourselves and let me know what you think.

Davidski said...

@Chad Rohlfsen

Epipaleolithic Anatolians and onto Barcin all have a lot of CHG/Iranian-related ancestry.

They have a lot of shared ancestry, but the point is that admixture from populations very similar to CHG/Iran_N is minimal in Barcin. Almost non-existent in fact.

Using the Natufians and/or Levant_N to estimate CHG/Iran_N admixture in Barcin and other Anatolians produces spuriously high estimates, because the Natufians and Levant_N have ancestry from the Levant that is largely missing in the Anatolians.

Samuel Andrews said...

mtDNA K1 is the only recent mtDNA link between Euro HGs and Anatolian farmes. In Mesolithic Europe, only BalkanHGs carried K1 at a high frequency. About 20% of IranGateHGs and BarcinN carry K1.

Most modern K1 (almost all is K1a), in both Europe and Middle East, probably derives from EEF-like pops. K1a's first expansion points date mostly near 15ky.

I think a population with high frequencies mtDNA K1 & no basal could be the source of the link between Paleolithic Anatolia & Europe.

Anthony Hanken said...

The Estonia and Finland preprints seem to finally give answers to the entrance of Balto-Finnic speakers in the Baltic.

Balto-Finnic language/Tarand graves/N1c are brought to Estonia from the Volga during the Iron Age and are autosomaly very close to modern Estonians.

Finland remains almost completly Sami like, even in the south until at least the 5th century. Modern Finns are a result of these Estonian Balto-Finns pushing north gradually settling and mixing with the Proto-Sami as well as Germanic groups. This occurred gradually and as late as the Middle Ages.

Samuel Andrews said...


That makes sense to me. One hole in that narrative is that Bronze age Latvia shows close ties to modern Balts, east Slavs, and west slavs. Not just in autosomal DNA but also mtDNA. The Baltic-like makeup of Iron age N1c Estonians could be caused by huge amounts of local Baltic ancestry not Estonian-like people coming from the Volga.

Queequeg said...

@Samuel Andrews and re "The Baltic-like makeup of Iron age N1c Estonians could be caused by huge amounts of local Baltic ancestry not Estonian-like people coming from the Volga."

Very good point. According to Saag et al, for instance, the whole Baltic area is being shifted towards (either) more EHG (or even WSHG?) in the late Bronze Age/early Iron Age, maybe related to Sintasha outlier type of groups but any ways most probably related to Uralic speaking immigration into area.

Davidski said...

The movement of anything derived from or even closely related to Sintashta into the East Baltic would bring with it at least some R1a-Z93. There aren't any signs of that in modern or ancient DNA.

What seems to have happened is that what are now Estonia and southern Finland were the meeting point between three or four populations similar to today's Balts, Saami, Volga Finns and Scandinavians, and out of this mixture came Baltic Finns.

I think it'll be possible to confirm this once the ancient data from the Estonian paper are released.

Mikkel said...

Thank you, @bellbeakerblogger.
Yes, I think the data is quite difficult to ignore when you see it visualized in "real-time" :)

Shaikorth said...

Doesn't look like that N1c1 Baltic_IA sample has any Sintashta proper, and outlier ancestry maybe in diluted form. Tarand graves are from LBA Central Russia, sampling around Moscow should resolve the non-Baltic ancestry but here goes with what we have:


Srubnaya, 0
Sintashta_o3, 0
Glazkova_BA, 0
Karasuk, 0
Karasuk_o, 0

Replacing BA Lithuanian with modern Lithuanian, everything's the same except a bit more Latvia_MN:

rest 0

Matt said...

Re: CHG/IranN "in" Anatolia deeper in history, in terms of ultimate origins of the Anatolian Epipaleolithic, I suspect, as you try and go deeper and deeper beyond the range of the late Upper Paleolithic, trying to build qpGraphs of relations between different Near Eastern groups might devolve into incoherence.

My guess is as probably it won't be in a place of periodic, substantial pulse admixture between long separated, clearly defined populations at all, and something more like continuous, but very low, geneflow over very long time scales, on clines. The assumptions of qpGraph will be violated and it won't be a good way to describe what is going on and as well you'll get multiple plausible viable models.

So trying to talk about whether ultimately in some sense IranN has AnatoliaN ancestry or AnatoliaN ultimately has IranN will be difficult to meaningfully do.

I could be wrong! Anyway the important thing here for the topic of hand is really whether the formation of agricultural groups in Anatolia involved any kind of pulse of admixture from societies from the zone of the Fertile Crescent, who were probably on some sort of cline between the LevantN and IranN (and perhaps also the Natufians, and not yet sampled more Basal Eurasian variants of the IranN samples). And it seems from this that to a very high degree, it did not.

Matt said...

Btw @Chad, from what I remember when you were doing models of the big picture of human population history, you ended up needing lots of different sized edges from Neanderthal into Ust Ishim, Upper Paleolithic Europeans, etc. in order to explain the structure.

Now as Paabo's new paper on the advantages for accurately measuring Neanderthal ancestry of using Altai and Vindija together in a model, and in their model a single pulse from a Neanderthal population which forms a clade within Vindija, my question would be: Do you have any plans to try and reconstruct your qpGraphs to take advantage of this?

E.g using one of your models ( as a base, something like: for a simplified phylogeny in which a population that is a clade with Vindija with Altai as outgroup mixes equally into an OoA human population (from which descends Basal, East and West).

@Davidski, if you have the data in one dataset to test that qpGraph, I'd be interested to see if it works and where the failure stats are.

PF said...

The 15K year old Anatolian is super exciting.

So it's looking like there was some sort of early expansion of a Basal-rich population(s?), pre-dating the LGM (?), that when mixed with distinct HGs helped establish the main poles of ancestry we see in the Near East already by the start of the Neolithic. Roughly speaking, Basal + western UHG = proto-Anatolia/Levant; Basal + eastern UHG = proto-Iranian/CHG?

Anthony Hanken said...

If Baltic BA ancestry makes up the majority of modern and IA Estonian autosomal and MtDNA components that must mean N1c was spread by an elite which left little autosomal ancestry in modern Estonians correct?

Another scenario is that The N1c carriers were already Corded Ware like prior to reaching the Baltic.

Davidski said...

@Anthony Haken

I think you rather mean Estonian-like.

Yes, but on closer inspection probably with some key differences.

AF said...

I've noticed using Davids PCA spreadsheet with nMonte that Eastern Euros and especially Balts get a rather bad fit when using the CWC samples + extra farmer + extra local HG. Not sure why though? Are Balts actually not derived of the Yamnaya-like populations?

Arza said...

@ AF

It's due to admixture in Baltic_BA from yet not sampled WHG(-like) HG population.

qpWave Narva + IIRC Baltic_LN model for Baltic_BA has failed. (Mittnik et al., preprint)

Interesting is that Varna outlier (Varna_o:ANI163) also shows this type of ancestry in Global 25.

Arza said...
page 8

Baltic_BA Baltic_LN 1.96E-11 rejected
Baltic_BA Baltic_LN Narva 1.18E-255 1.14E-02 rejected rejected
Baltic_BA Baltic_LN CordedWare_Central 1.63E-13 2.03E-02 rejected rejected

Baltic_LN_Spiginas2 CordedWare_Central 8.08E-07 rejected
Baltic_LN_Spiginas2 Steppe_EMBA 4.68E-16 rejected
Baltic_LN_Spiginas2 Narva 8.44E-68 rejected
Baltic_LN_Spiginas2 CordedWare_Central Narva 1.38E-242 1.80E-01 rejected not rejected

On the other hand Spiginas2, that also has this kind of ancestry and generally is similar to Baltic_BA, can be modelled as CordedWare_Central + Narva.

BUT... Spiginas2 can be approximated in G25 as e.g. Baltic_BA + Ukraine_Eneolithic mix. In other words it's shifted towards "classic" HG to the point, where standard models are starting to work.

PF said...

I predicted that some level of interbreeding with archaics everywhere modern humans expanded should be the default assumption. (Looking back it's funny how recently most people were arguing against any interbreeding with Neanderthals, even though we actually had those skeletons and knew that our ranges intersected in space and time.)

Detecting unknown introgressed archaic haplotypes in modern and ancient human genome sequences

Skov et al.

Introgression of archaic haplotypes into human populations is an already known phenomenon, with some haplotypes even providing a selective advantage such as adaptation to living in high altitudes or haplotypes carrying alleles of genes involved in the immune-system.
Most published methods for identification of archaic haplotypes rely on ancient DNA samples from the archaic population, to compare the modern samples with. Here we use a novel method for identifying regions of archaic introgression in ancient and modern human genomes. The method is a Hidden Markov Model that classify the genome into human any number of archaic states.
Not relying on a reference genome allows us to identify the introgressed haplotypes much more accurate. This allows us to calculate the length distribution and then estimate the admixture times with archaic hominins for different present day and ancient human populations.
Because our method does not rely on an archaic reference genome we are able to reconstruct the several MB of the unknown introgressing hominin genomes. These include unknown archaic populations from Africa, East and South East Asia along with a highly divergent archaic population into Melanesian populations.

PF said...

Surprised this one about "mother's tongues" hasn't been discussed here yet. ;)

Mother' tongues? A global study of sex-biased genetic and linguistic transmission after Out-of-Africa

Creanza et al.

Worldwide patterns of genetic variation are driven by human demographic history during and after Out-of-Africa. To test whether this demographic history has left similar signatures on languages to those it has left on genes, we analyzed phonemes from 2,082 languages and microsatellite polymorphisms from 246 populations. Globally, populations that were geographically closer tended to be more similar genetically and linguistically. Our analysis suggests that two processes influence this pattern: vertical transmission of both genes and languages after Out-of-Africa, and linguistic borrowing (often coupled with genetic admixture) when neighboring populations speak very different languages. To understand whether sex-biased patterns in human history affect genetic and cultural evolution, we merged genetic, linguistic, and ethnographic data to perform two new studies. First, we address the hypothesis that children preferentially learn language from their mothers by comparing linguistic variation to mitochondrial and Y-chromosome variation separately. We show that long-range patterns of correlation between linguistic and genetic distance are present for both maternally inherited genes (mtDNA) and paternally inherited genes (Y-chromosome). However, mitochondrial genetic variation also significantly correlates with short-range patterns of linguistic variation (calculated with a different metric), whereas Y-chromosome variation does not. In addition, we gathered ethnographic data to categorize 182 genotyped populations by kinship system and postmarital-residence pattern. With this annotated database, we observe measurable evolutionary effects of matrilineal kinship and matrilocal residence; our results suggest that matriliny and matrilocality strengthen the association between languages and geography. These analyses shed new light on genetic and cultural population histories after Out-of-Africa.

PF said...

A few others of interest but I'll just post one more. Looking forward to more details in the full papers.

Ancient Genomic Diversity Reveals Differences in Cultural Practices and Cultural Barriers between Prehistoric Farmers and Hunter-gatherers in Europe

Hofmanova et al.

Humans differ from most other species in that we create our own ecological niche. Culture has thus shaped human genetic variation over millennia. While surprisingly little is known about prehistoric cultural practices, there is vested hope that patterns of ancient genetic diversity will elucidate how past societies were organized and interacted with each other. Yet such inferences remain challenging due to generally low numbers of individuals and especially the lack of population-level samples. Here we present novel samples from the region of the Danube Gorges (Balkans), located in the heart of the migration corridor through which farming was brought from Anatolia to Central Europe. Our archaeologically well-defined samples (~10000-5500 calBC) represent multiple closeby-settlements of a sedentary society before and during Neolithisation. Contrasting population-genomic and cultural affinities of our samples revealed that settlements differed strikingly in their interaction with immigrating farmers: while some exhibited strong barriers to gene flow, others incorporated multiple individuals of genetic ancestry common to Aegean farmers. To elucidate important aspects of social practices before, during and after this demographic shift, we accurately inferred within and between individual genetic diversity of our population sample by sequencing either whole genomes or many putatively neutral regions, and by using novel methods that account for post-mortem damage and the heterochronous nature of our reference panel. Notably, we found a lower within-individual diversity as well as a lower X to autosomes diversity in hunter-gatherers than farmers prior to their contact, consistent with an elevated population size and stronger patrilocality in farmers.

Matt said...

@PF, the Hofmanova abstract is interesting on the cultural angle;

Notably, we found a lower within-individual diversity as well as a lower X to autosomes diversity in hunter-gatherers than farmers prior to their contact, consistent with an elevated population size and stronger patrilocality in farmers..

In a sense it's "dog bites man", because patrilocality is the dominant form of organisation ("around 70 percent of the world's cultures that have been described ethnographically" in wiki) at all levels of social organisation and in cultures both in contact with the West Eurasian "oikumene" and well outside it (e.g. Native Americans, New Guineans, etc.).

There have been suggestions that the cultures of Neolithic Southeast Europe - "Old Europe" - had a matrilocal form of social organisation.
So this is some nice, relatively solid evidence, evidence that is more direct than archaeological inference on symbolism, that this is not the case.

Though strontium isotopes already seem to have showed this, at least for Central Europe (rather than "Old Europe"). From a link from Maju's blog back in 2012 - - "By using strontium isotopic data from more than 300 early Neolithic human skeletons, we find significantly less variance in geographic signatures among males than we find among females, and less variance among burials with ground stone adzes than burials without such adzes. From this, in context with other available evidence, we infer differential land use in early Neolithic central Europe within a patrilocal kinship system.". (Maju's commentary on this referring to "dismantling the idea of Gimbutas of a matrilocal "Old Europe", which was already very much discredited" in characteristically direct fashion).

But it's nice to have two lines of evidence that are "hard" and direct confirming that supposed matrilocality of Neolithic European cultures is not the case. It's great to have these multiple lines of evidence.

(Not that matrilocality implies any form of matriarchal or egalitarian society though either; the Chaco Canyon society adna findings seem to imply a social structure around a matrilineage - But at the political level, same old male focused chiefdoms. No sex egalitarianism.

Incidentally, this article on Chaco Canyon also in mentioning that the Lycians, speakers of an Anatolian language, were probably matrilocal. But that again, this did not change the fact that on the political level they were male dominated societies.)

Matt said...

Was hoping for a release of the Maykop paper and data following SMBE2018, no such luck.

Looking at abstract from SMBE's version against differences from the published paper. Differences in bold.

"Into the great wide open: the genomic history of the Greater Caucasus region"

"The Caucasus mountains, bound by the Black and Caspian seas, connect the Near East and the Eurasian steppes. Recent archaeogenetics studies have described the formation of 'steppe ancestry' ultimately as a mixture of Eastern European and Caucasian hunter gatherers.

However, it remains unclear when this ancestry arose and whether cultural innovations originating in the Near East had facilitated the opening of the steppe environment for pastoralist economies

(original abstract states: "whether it was related to a horizon of cultural innovations in the 4th millennium BCE that subsequently facilitated the advance of pastoral societies likely linked to the dispersal of Indo-European languages.")

To test whether this also involved gene flow, we generated genome-wide SNP data from 50 prehistoric individuals along a 3000-year transect through time in the North Caucasus region, ranging from the Eneolithic (6300 yBP) to the Late Bronze Age (3400 yBP).

(45 individuals stated to be in preprint, 45 used in D-stats and 49 in PCA/admixture, out of 58 samples, so 50 may simply be revision).

We observe a genetic separation between the groups in the northern foothills and south of the Caucasus, and those of the bordering steppe regions in the north. We coin these 'mountain' and 'steppe' Caucasus groups, according to vegetation zones and characteristics of the associated archaeological cultures.

(E.g. terminology possibly changed from Caucasus_Eneolithic and Steppe_Eneolithic to Mountain_Caucasus_Eneolithic and Steppe_Caucasus_Eneolithic or some such to clarify).

Furthermore, 'Steppe Majkop' individuals harbor a distinct ancestry component that relates them to Upper Paleolithic Siberians and Native Americans. In contrast, genomic ancestry profiles of groups from the northern foothills are similar to those in ancient Georgia and Armenia. This suggests that the Caucasus Mountains are not an insurmountable barrier to human movement, and further permits the detection of periods of genetic continuity as well as occasional gene flow.

(Have samples been published yet for ancient Georgia? Or is this CHG?)

Intriguingly, individuals associated with Yamnaya and subsequent pastoralist cultures show subtle evidence for Anatolian Neolithic-farming-related ancestry, possibly from different sources in the western and southern contact zones."

Davidski said...


I'm pretty sure that the abstract to the SMBE 2018 was submitted a lot earlier than the preprint to the BioRxiv.