search this blog

Loading...

Monday, December 16, 2013

Cluster analysis of West Eurasia: 13 clusters from 18 dimensions


I ran a quick Mclust analysis to get a better idea of the substructures in my recently updated dataset of West Eurasian samples. Mclust found that the optimal outcome was produced with 18 dimensions of genetic variation and 13 clusters, the latter of which are superimposed on a two dimensional MDS plot below. I chose the labels for the clusters myself and flipped the canvass to fit geography.


Here you can see the 13 clusters superimposed on all possible combinations of the 18 dimensions. Clicking on the image will take you to a 10.3MB PDF file.



It's interesting to note the presence of the very tight Jewish cluster, which includes Ashkenazi, Sephardic and Moroccan Jews. The Basques and Sardinians also cluster together, despite being clearly distinct from each other in the fist two dimensions. This is fascinating because these two groups have been mentioned a few times now in various studies and presentations as being the best modern proxies for Europe's Neolithic farmers.

The widespread Central and Eastern European cluster mostly includes individuals from populations that aren't easily characterized in these sorts of tests, and that's basically because they're of mixed origin. Indeed, I suspect things would look somewhat different in that part of the plot if I had more sizable samples from Germany, Scandinavia, Poland and nearby areas.

Mclust can produce many more clusters than just 13 from the same data, but as per above, I wanted to see what would happen if it was asked to come up with the optimal solution. For more on this type of analysis check out the articles here, here and here.


Update 17/12/2013: On a related note, here's an Mclust analysis of West, Central and South Asia. The optimal result was obtained with 10 dimensions and 14 clusters. Please note that although some of the clusters have the same names as in the analysis above, they aren't the same clusters.




See also...

Principal component analysis (PCA) of West Eurasia

Multidimensional views of South Asia, West Asia and Eastern Europe

Eurogenes' North Euro clusters - phase 2, final results

66 comments:

Onur said...

David, could you post a spreadsheet that shows which populations and/or individuals belong to which clusters?

Davidski said...

Try this...

https://docs.google.com/spreadsheet/ccc?key=0Ato3EYTdM8lQdFJaSktGWVJFdjJmeWgzX0pROXlCekE&usp=sharing

Onur said...

Thanks.

Maju said...

I've been looking at the somewhat striking Basque-Sardinian "cluster" (I don't have time nor processor power to do the same with all groupings, although it'd be interesting) in all PC axes (i.e. linearly, as McClust surely does, because it is where the real info is) and there is only one of them in which they do cluster, which is PC2, where both pile up at the positive extreme (Sardinians and Basques vs. Volga-Ural and N. Caucasus).

There are some other occasions in which they "cluster" but in most of them they do so within much larger pan-European or, more commonly, pan-WEA accumulations in the "neutral zone" (which is quite meaningless: neither this nor that). There are many other occasions in which they do not cluster either, not even in the "neutral zone".

I could only find another instance of specific clustering of Basques and Sardinians, in PC5. This axis is defined by N. Caucasus in the positive pole and Volga-Ural in the negative, with Basque and Sardinians both tending markedly towards this negative "Volga-Uralic" and "anti-Caucasic" pole.

There is also one instance of opposite polarity between Basques and Sardinians, component 11. In this case most other populations sit neutral between them, except for Volga-Ural, who tend to the positive pole, which I'm not sure if it is the Sardinian or the Basque one.

So I'd say that most of the reasons to cluster Basques and Sardinians are in the PC2, which essentially opposes SW Europe (Sardinians and Basques) vs. Eastern Europe or even Eastern WEA (Volga-Ural and secondarily N. Caucasus, mainstream Eastern Europe, Near Eastern...). Excluding Basques and Sardinians, this SW pole is defined by Iberians and NW Africans.

Naturally the PC1 illustrates the extremes of Europe vs West Asia, polarized into two endogamous populations: Finnish and Bedouins but less dramatically by the Eastern Europe (Balto-Slavic-plus) vs Arabia axis.

Even if under-represented, North African specificness shows up in the PC3 and PC4 axes and, quite curiously, in these it is not confronted directly with Europe but with "greater" West Asia: N. Caucasus and Bedouins representing the opposites of North Africans better than anyone else.

Otherwise Basques and Sardinians do not seem particularly close and can even show diametrically opposite polarities. I wonder what these kind of PCAs would say if we'd focus only on Western Europe, excluding the East and the transmediterranean regions.

On final note: I do not think that autosomal clusters are really that meaningful, because they are so dependent on so many factors (sampling strategy particularly) and also because there is a whole other dimension to clustering, which is clinality. This clinality is quite apparent all the time in these kind of analysis and over-emphasizing clusters (sometimes a bit accidental) may confuse more than it explains in some cases.

Also I would like to insist that sampling strategies should be very careful with apportioning to realistic population approximates. Maybe a logarithmic pseudo-proportion would be quite adequate. What is clear is that we cannot compare one million Lithuanians with more than 100 million ethnic Russians as if they were the same thing. There should be at least 10x more Russians than Lithuanians to make a meaningful comparison. In this sense I'd recommend to discard the Bedouin sample because it tends to cope the axes with their endogamous peculiarities too much.

I'm not sure which pops. compose the Volga-Ural sample, because Volga-Finns are in another bloc but they may be also distorting things. I'd really like these analysis to emphasize a realistic demographic weight, otherwise small odd populations seem to distort things too much.

אחיקם גדליהו ישעיהו בן הלל said...

Fascinating, it makes sense that West Asia must've been more Sardinian-like at some point.

The NW African cluster's position also hints to such an outcome.

Once more, we shouldn't jump onto conclusions based on modern-day population clusters, this makes an even stronger case for genome-wide studies of ancient human remains.

Seinundzeit said...

Thank you very much for the spreadsheet!

Kapak said...

To me this seems to suggest North-West Europeans are a 50-50 split mixture of Mesolithic Europeans (represented by Balto-Slavic and Volga Finnic) and of Sardinian+Basque like populations.

Davidski said...

Something like that, but also throw in the Baltic Finns (minus their Siberian ancestry) into that mix, and reduce the Sardinian input.

The Mesolithic vs. Neolithic dichotomy in the modern European genome is better shown in the PDF on the plots made up of dimensions 1&3, 1&4, 1&5 and 1&6.

Also, check out how the Northwest European and Volga-Ural clusters are stuck together on plot 1&7. I don't know what that means though.

Ryan said...

Davidski, awesome work as usual. Thank you for the update. Is there any way that you could please post a spreadsheet with individual coordinates, so that one could see where they cluster? Thanks again!

Davidski said...

Try this...

https://docs.google.com/spreadsheet/ccc?key=0Ato3EYTdM8lQdFJaSktGWVJFdjJmeWgzX0pROXlCekE&usp=sharing

Davidski said...

I updated the entry with an Mclust analysis of West, Central and South Asia. The relevant table of results can be found here...

https://docs.google.com/spreadsheet/ccc?key=0Ato3EYTdM8lQdE8xQ2N2VDBFUUQzS2RmRkhBVmNuZWc&usp=sharing

About Time said...

Absolutely, yes.

Seinundzeit said...

Thanks!

MfA said...

Thanks David, So West Asian cluster is basically Kurdish/Iranian/Turkish.. I see there is no any leaning to Baloch/Makrani cluster from these groups(all scores zero).. Although it is insignificant, there is Central-Asian(Afghan/Tadjik/Turkmen) signature on all of them including Armenians and Georgians..

Onur said...

MfA,

That "West Asian" cluster is formed not because of a special genetic connection between Turks and Kurds/Iranians, but because of the fact that the small ASI (=Ancestral South Indian) admixture plus the much smaller Mongoloid admixture in Kurds/Iranians and the small Mongoloid admixture in Turks push them both from the spaces occupied by the "Near Eastern" cluster a little bit, and by coincidence in similar levels in most cases, towards the spaces occupied by the significantly ASI-admixed populations and make most of them by chance occupy nearby spaces. The Mongoloid and ASI admixtures behave similarly because ASI is significantly closer to Mongoloids than to Caucasoids. BTW, a notable proportion of Turks are in the "Near Eastern" cluster rather than the "West Asian" cluster, almost certainly because that those Turks have smaller (or no) Mongoloid admixture than have the rest of Turks.

Davidski said...

I haven't looked at the 10 dimensions closely, but from what I've seen before, East Eurasian and South Central Asian admixture isn't the only factor that differentiates Iranians and many Turks from more typically Near Eastern samples. They also have higher levels of North European (or rather Eastern European or Volga-Ural) ancestry.

Onur said...

It is already known from autosomal analyses that Balkan Turks have significantly higher levels of Northern European ancestry than Anatolian Turks, no doubt due to their high levels of Balkan ancestry. As for Anatolian Turks, Armenians are not the best extant proxy for the pre-Turkic Anatolia, Anatolian Greeks are. So you should compare Anatolian Turks with Anatolian Greeks. As far as I can see from their rather limited number of publically available autosomal analyses, Anatolian Greeks have no less North European ancestry than Anatolian Turks, so Anatolian Turks' levels of North European ancestry do not seem to be unusual for their geography.

Iranians and Kurds too do not seem to have that high levels of North European ancestry for their geography, especially taking into account the levels of their eastern neighbors.

Also, higher North European ancestry would not push a Near Eastern population towards South Asians. The "West Asian" cluster probably has nothing to do with North European ancestry.

Davidski said...

Admixture from the Volga-Ural region isn't the main factor that differentiates the so called West Asian and Near Eastern clusters here, but it's one of the factors. Admittedly, it's a much more pronounced factor for the differentiation between the North Caucasian/Near Eastern clusters. See PCA 1&3 here...

https://drive.google.com/file/d/0B9o3EYTdM8lQeEUzeWZ2TWpZbzA/edit?usp=sharing

Onur said...

I think the very small (about 1% on average) "Volga-Uralic-looking ancestry" in Turks that shows up in some ADMIXTURE analyses is not actually from the Volga-Ural region but is just an artifact of the ADMIXTURE software assigning a small part of the North European and Mongoloid ancestries of Turks together to the "Volga-Ural" component when such a component exists, as the "Volga-Ural" component is usually a mixture of North European and Mongoloid ancestries. Anyway, most of the North European ancestry in Turks is never assigned to the "Volga-Ural" component and is assigned to the regular North European components instead (this is true for Anatolian Turks too, and not just for Balkan Turks). Also, an extremely small part of the Mongoloid ancestry of Turks is assigned to the "Volga-Ural" component when such a component exists and the overwhelming majority of the Mongoloid ancestry of Turks is instead assigned to the regular Mongoloid components.

As for Iranians and Kurds, they have virtually no "Volga-Uralic-looking ancestry".

Davidski said...

I usually see more pronounced levels of Volga-Ural and similar components in the Near East, and these are clearly higher in Kurds, Iranians and many Turks than in Assyrians, by a few per cent in most cases. Also, groups from the Volga-Ural like the Chuvash and Mari only carry around 40% of the Volga-Ural component, so to estimate admixture from the Volga-Ural region, you need to look at other components too, like North Sea, Baltic and even Siberian. These are again higher among Kurds, Iranians and many Turks, than Assyrians, usually by a few per cent.

Onur said...

I think you and I are talking about different things when we say "the "Volga-Ural" component". What I mean is actually a more Uralic component than a Volga-Uralic or eastern European component thus a more Mongoloid-admixed component. The "Volga-Ural" component you refer to seems to be a more Caucasoidized version of the "Volga-Ural" component to which I refer (explaining the former component's higher levels in West Eurasia), so, like the latter component, the former component's relatively elevated levels in Iranians/Kurds and many Turks is likely due to the ADMIXTURE software artifact effect I referred to in my previous comment, the only difference from the latter component being the former component's higher Caucasoid/Mongoloid ratio hence higher levels in West Eurasia.

Davidski said...

No, what I mean is that Kurds, Iranians and some Turks have higher levels of North European ancestry from the Volga-Ural region than Assyrians, Armenians and other Turks. This is supported by all the data I've seen.

Onur said...

What exactly do you mean by "North European ancestry from the Volga-Ural region"? North European ancestry is very widespread in West Eurasia and beyond, so its source cannot be confined to the Volga-Ural region, unless you are talking about a specific type of North European ancestry rather than North European ancestry as a whole.

Davidski said...

Northern European genetic ancestry peaks in a wide area from the Atlantic to the Urals (north of the Pyrenees, Alps, Black Sea and Caucasus).

But the population that lived in the Volga-Ural region in prehistoric times was even more Northern European genetically than the present inhabitants there, who are partly of recent Asian ancestry due to the Turkic expansion.

So a migration from the pre-Turkic Volga-Ural region of, say, the early Iranians would certainly have taken a lot of North European DNA to West, Central and South Asia.

Onur said...

So you are talking about North European ancestry as a whole. I already presented my explanation for its levels in Iranians/Kurds and Turks when I wrote above:

"It is already known from autosomal analyses that Balkan Turks have significantly higher levels of Northern European ancestry than Anatolian Turks, no doubt due to their high levels of Balkan ancestry. As for Anatolian Turks, Armenians are not the best extant proxy for the pre-Turkic Anatolia, Anatolian Greeks are. So you should compare Anatolian Turks with Anatolian Greeks. As far as I can see from their rather limited number of publically available autosomal analyses, Anatolian Greeks have no less North European ancestry than Anatolian Turks, so Anatolian Turks' levels of North European ancestry do not seem to be unusual for their geography.

Iranians and Kurds too do not seem to have that high levels of North European ancestry for their geography, especially taking into account the levels of their eastern neighbors."

Your Volga-Ural scenario may be correct for Iranians and Kurds, but apparently not for Turks. Turks with Armenian levels of North European ancestry seem to be confined to the parts of the Turkish areas geographically in ancient Armenian or Pontic Greek lands such as NE Anatolia. North European ancestry in Turks increases as one moves to more western areas and peaks in Balkan Turks, likely correlating with the increase of North European ancestry in Greeks as one moves west from Pontus to the Balkans.

Nirjhar007 said...

The Iranian aryans had there homeland say very close to South Asia so Having the South Indian Component is fine as they left After 2000 B.C.
Good Day.

Davidski said...

Eastern Europe isn't all that close to South Asia.

Onur said...

David, could you post the name of population corresponding to each population symbol used in the MDS plots?

Davidski said...

The MDS doesn't show symbols for each of the ethnic groups on the plot. It just shows symbols for each cluster, and I've already labeled these clusters.

Volodymyr Lutsyk said...

Could you, please, tell me which part of Ukraine UA4 comes from?

Davidski said...

I forgot, but maybe he'll tell you if he sees this post.

Ponto said...

One of the Greeks came out mostly in the Jewish cluster. Is that a mistake? With that sort of result that would mean that person would be wholly Jewish.

Davidski said...

It just means he was placed near European and Moroccan Jews in most of the dimensions that were considered. At least two people who are only partly Jewish also showed 100% membership in the so called Jewish cluster.

When I use more dimensions I get two different Jewish clusters; Ashkenazi and Sephardic/Moroccan Jewish. These clusters don't include any non-Jews.

But resolving fine scale ancestry like that wasn't the point of the experiment. I wanted the program to show me the optimal solution for the entire dataset. The fact that this resulted in European Jews falling into a cluster with a Greek using 18 dimensions is very interesting, and probably gives a strong hint about their origins.

Onur said...

OK. Thanks.

spagetiMeatball said...

David, what were the Volga-Ural populations like before the asian admixture? This "north european/mesolithic european" cluster is basically the "Hyperborean" population that Greg Cochran is talking about? He says they were very diverged from other global populations like east asians. A significant portion of european ancestry is from this population, whatever you want to call it, but europeans are obviously a west eurasian population on global PCA plots, and not somewhere floating all on their on. But how can that be if they carry ancestry from a population so diverged from other west eurasians?

Sorry if these questions seem naive, but I am an engineer by education/profession, with a small interest in anthropology; I am a caveman with this genetics stuff.

Davidski said...

Yes, a significant portion of West Eurasian ancestry seems to come from a northern clade that isn't closely related to the East Asian clade, and not even all that close to the other main West Eurasian clade, which is probably best represented today by Bedouins minus their Sub-Saharan admix.

But the reason this "Hyperborean" clade doesn't look so distinct today is because it's extinct, and most West Eurasians only carry varying degrees of admixture from it. East Baltic populations seem to have the most.

As for the Volga-Ural region, yes, it was probably populated by this Hyperborean population of foragers, just like the rest of Europe, during the Upper Paleolithic and Mesolithic. But I think that Siberian admixture affected this region from the Mesolithic onwards. Then Near Eastern admixture arrived during the Neolithic, and then some more Siberian and also East Asian admixture during the Turkic expansion.

So the Volga-Ural region is basically like Europe (ie. Hyperborean forager + Near Eastern farmer), but with significantly more Siberian and East Asian admixture.

spagetiMeatball said...

Interesting. Thanks for the reply.

One thing that is striking about the eurasian population rearrangement from the neolithic revolution onwards, that east asians have escaped being admixed with anything. Or mixing with anybody. What I mean is that japanese, chinese, koreans, and south-east asians appear to be derived almost exclusively from the populations that started the neolithic/agricultural revolution in east asia. On all admixture tests, they come up something like 97% of that original population. What happened to the other HGs in asia?
Any bloggers talking at all about east asian prehistory?

Davidski said...

The Tianyuan Cave sample does suggest that East Asians are today pretty much like they were 40,000 years ago, at least in terms of autosomal DNA. But there might be some surprises when we see more data.

Unfortunately, I don't know anyone who's focusing on East Asians specifically.

Unknown said...

Very interesting stuff. I notice i show about 96.3% membership to the NW European, and 3.7% membership to Central & Eastern European groups (Uk42), but none to any other. I had thought i had a small amount (3-6%) Italian ancestry, although southern European, or any other nearby populations, don't show up here. Is this just due to the nature of the program equating my pull away from the NW cluster as being due to membership of another nearby cluster, or would you think it was representing actual ancestry from that Central European cluster?

Thanks.

Davidski said...

Yes, it might be because Central Europe is on the way from the UK to Italy. Also, keep in mind that some Italians are in the Central European cluster.

I probably need a lot more samples to resolve the issue.

jackson_montgomery_devoni said...

CA1 reporting in. I get like 99% in the Northwest European cluster that you linked above Davidski. That really does not make much sense given what I know about my ancestry.

Davidski said...

Well, you're in that cluster because you're 50% British. The algorithm should've picked up your Italian and Finnish ancestries too, but it's not very good at that sort of thing. Maybe in this case it needs many more dimensions than 18?

Fanty said...

From what I ahve seen on Mclust ever, its the normal case that it puts someone almost 100% into one cluster. Its pretty rare that it actually shows "mix" in clusters.

I recall that someone with one Polish and one Southern Italian parent had been 100% in the Romanian cluster in a Mclust experiment of Dienekes.

I would say, you beeing in the North-Western European cluster and getting a tiny bit of central does mean, you are placed somewhere, where the NW Euro cluster overlapts with the central one.

Nirjhar007 said...

See Kiddo,
Eastern Europe has nothing to do with the aryans as Aryans are quite an Ancient groups like the Jew folks whom maintain there traditions like we do and Since Jews didn't forget where their homeland was it is impossible that did happen to the Aryans..
Unfortunately for our Balto-Slav brothers they do not have any i mean any Ancient documentations about their whereabouts like the Aryans did.
According to the Indian Aryans of Rigveda their homeland was the Sapta Sindhu (Hapta Hindu Of Avestans) Area from Panjab in Middle to the west Afghanistan and Ganges in the East and their foremost northern known region was Shortugai settlement beside the Aryan named river Vaksh See the wiki of both...
From shortugai they also ventured north to BMAC+ Andronovo for trade of what the Harappans were famous of....
About PIE the thing is that it is a Story told by Our Great Great Great Great Great Grandmother but with time her successors created and new versions of the Story so when look for the original one (PIE) ALL WE CAN DO IS take together the common materials of the stories(IE Language Branches) TO RECONSTRUCT, but it would be never equal to the Original though from the Elderly folk like the Aryans we get the story most closest to the original so we should listen to them....
About R1a1a Like the M582 tells Most branches have their own specific mutations likeZ93+,Z280,M458,M582 etc but they all are not capable to show any Migration patterns whatsoever,we even don't know what was the SNP of the Andronovo folks or more importantly the Harappans of Farmana though in the Farmana case from the crack as i told you there is Sorrow on your Fate Davidski:(...
Good Day.
P.S. I'm lacking time today otherwise i would have explained much wider, that i do on Monday, See you,
N.

Davidski said...

Nirjhar007, you're a lot of fun, but sadly, your time is almost up. The first paper on R1a using full sequencing was just released, and more are on the way soon. Here's the first phylo tree from the full sequencing...

http://img854.imageshack.us/img854/9505/8q7z.png

Already very suggestive. But please don't jump off a bridge or take up crack when you see the next one, because that'll be the killer.

spagetiMeatball said...

I don't get why people are so obsessed with things that happened 5,000 years ago. Rather than looking and planning for the future, people seem to have a mental obsessive compulsive disorder with defending their past or something. It's one thing when it is something historical within the last 2,000 years, but going further back is clearly a mark of insanity.

Another thing is, why are Indians (or anybody else, for that matter) so upset if their parental markers came from somewhere far away? Logically you can't be upset that your potential great-great-great-to-the-power-of-umpteenth grandfather was killed and cuckolded, because you are descended from the guys who kill and cuckolded him. You wouldn't be here if that far-off event didn't happen.

Ultimately, humans are just machines for transferring genes (in a sense), so seeing people get upset when told that they share alleles or haplotypes with someone outside their ethnic group is fun.

Maju said...

Why "2000 years"? Why not 20, 200 or 20,000? It's totally arbitrary. Many people would not even care for anything that happened before they were born or even later if they have not experienced the suffering personally. On the other hand, some others are more concerned about the past because, you know, "for what they were we are, for what we are they will be". And it's not just or even mostly about mere genes (which get diluted with time and may not be that important anyhow): it's about culture and identity, which are ideas (memes if you wish).

In my personal experience, I am genetically speaking the partial product of fascism. If the Spanish Civil War and WWII would never have happened, I would not exist. And honestly I would prefer that because non-existence is trivial (you don't exist: you don't care) and fascism and war horrible. I'm not as selfish as to think that my ego is above everything else. My ego is just an accident, probably other egos would do roughly the same in different conditions or even better in better conditions. Humans are essentially interchangeable, most depending on social and personal circumstances, rather than those silly genes.

I am more concerned about the present and the future than the past, however I do have a passion for unveiling the secrets of the past and I do realize that what happened long ago has effects today. And unveiling it may have emotional and socio-political effects, at least at the personal level, because emotions are important, very important, and also the past contextualizes the present, it largely explains it.

With different tempos and some different details in Europe and Southern Asia our ancestors experienced the violence and disruption of Indoeuropean expansion. And paradoxically some of our ancestors belonged to those violent thugs we would personally kill gladly with our own hands if possible. All that has emotional and rational implications that are hard to manage, depending maybe on your personality and other factors (such as identity, ideology, etc.)

Naturally one has to grow up and accept the past for what it was (or seems to have been). But often we do it only reluctantly, painfully, because we do not like it anyhow. Sometimes denial is part of this process.

spagetiMeatball said...

Yeah, but they're clearly not upset because of some universal injustice in most cases.

Grey said...

"What I mean is that japanese, chinese, koreans, and south-east asians appear to be derived almost exclusively from the populations that started the neolithic/agricultural revolution in east asia."

Maybe they managed to fight off any HG-to-herder tribes until they'd developed a large enough population density for any herder conquest to be no more than elite replacement e.g. Mongols.

http://en.wikipedia.org/wiki/Xirong

Matt said...

With East Asia, we should be mindful that the hunter-gatherers of Southeast Asia before the Neolithic were not necessarily like the Great Andamanese or Papuans.

They might have been more like the Han, Malays and Thai - there are plenty of relict hunter-gatherers in South East Asia, such as the Mentawai, Iban, Dayaks, etc. who are more or less East Asian, in terms of phenotype.
When it comes to the groups that get labelled "Negrito", we might have a slightly distorted impression of how widespread they were before the Neolithic, because their presentation of unusual mixes of derived East Asian features and generalized tropical morphology combined with their restricted geographical spread ends up making them of outsize interest to genetic and physical anthropology.

Most admixture projects seem quite light on South East Asia hunter gather samples, both because it's difficult to crowdsource new samples, because the existing samples (the HUGO pan-asian samples) have little overlap with most other panels and scanty coverage, and they're mostly West Eurasian focused.

This said, for what it's worth, there is a South East Asian cluster in the Harappa Project (South Asian focused) that peaks in the hunter-gatherer Iban, and which reduces along a cline to Southern China (Malays, Thais, Southern Han Chinese).

The unusual element in East Asian contexts may not be the degree to which HGs mixed with farmers, which may have been typical, but the degree to which the HGs and farmers were related already, due to pre-Neolithic changes.

In connection with this see - http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0025835 - Major Population Expansion of East Asians Began before Neolithic Time: Evidence of mtDNA Genomes

http://dienekes.blogspot.co.uk/2013/09/pre-farming-population-expansions-aime.html - Human genetic data reveal contrasting demographic patterns between sedentary and nomadic populations that predate the emergence of farming

There's a similar story with the Jomon-like people in Northeast Asia at least - http://dienekes.blogspot.co.uk/2013/07/smbe-2013-abstracts.html - (1) Sanganji Jomon was very similar with modern East Asians when we compared the worldwide populations in the PCA plot; (2) when only East Asians were compared, Sanganji Jomon was distant from both modern Northeast and Southeast Asians, indicating that Sanganji Jomon people were already isolated from other continental populations for a long time.

So in West Eurasia there might have been a more stable pre-Neolithic structure, disrupted by the Neolithic while in E Eurasia, a pre-agricultural sweep may have had more effect, and the Neolithic may have had less effect.

Grey said...

Yes, seems to me there are at least three possibilities for first farmer expansion depending on terrain and circumstances.

1. Farmer expansion into suitable territory spawning a HG to herder transition on adjacent marginal terrain around the periphery but where the farmer population remains dominant long enough for their population density to reach a point where any subsequent herder conquest is only elite replacement.

2. Same as above but where something happens to weaken the farmer population before they are overwhelmingly demographically dominant and the peripheral herder population over-runs the farmer territory.

3. Farmers expand to a bottle-neck where their numbers advantage over HGs is neutralized e.g. a region where the final farming region in an expansion chain is fairly marginal while the adjacent HG region is unusually productive such that a stalemate ensues long enough - probably 1000s of years imo - for pans not people to operate. In this 3rd case the HGs would become the farmers.

I wonder about this 3rd option happening in China with farmers expanding along the silk route and getting blocked by the wetlands of the Wei valley so the farmers are stuck in the Tarim basin long enough for their farming techniques to spread to the HGs in the wetlands.

I wonder if that may have happened at the Nile also for the same reason.

Grey said...

"happening in China"

meant northern China, not necessarily the same in the south.

Nirjhar007 said...

@David
''Already very suggestive. But please don't jump off a bridge or take up crack when you see the next one, because that'll be the killer.
LOL, Thanks David for giving me such a joyful start for the day, However your ideology seems to be of a school boy Since you are trying to compare a Conclusive aDNA Data with modern indirect Biased research, but that is obvious since it is the concern of your childish Balto-Slavic collective ego though it is a LOST case but still venerable dude;)....

Nirjhar007 said...

2Meatball
''I don't get why people are so obsessed with things that happened 5,000 years ago. Rather than looking and planning for the future, people seem to have a mental obsessive compulsive disorder with defending their past or something. It's one thing when it is something historical within the last 2,000 years, but going further back is clearly a mark of insanity.
''
See boy there are some races still intact in current earth whom still live there religion well alive like they lived 4000 Y ago but i'm currious about you what were your ancestors doing 4kyo?...

Nirjhar007 said...

@Maju, Sorry i forgot about you....
''With different tempos and some different details in Europe and Southern Asia our ancestors experienced the violence and disruption of Indoeuropean expansion. And paradoxically some of our ancestors belonged to those violent thugs we would personally kill gladly with our own hands if possible. All that has emotional and rational implications that are hard to manage, depending maybe on your personality and other factors (such as identity, ideology, etc.)

Naturally one has to grow up and accept the past for what it was (or seems to have been). But often we do it only reluctantly, painfully, because we do not like it anyhow. Sometimes denial is part of this process.''
See, You can not be delusional with your personal Manifestations on the society by applying it to the dispersal of a large and vivid language group, there is not a single fact that you can show that would suggest your ideology as the happening as there is always a way way standard ideology more in agreement with Historic ethics,traditions, archaeology etc, however talking about ''Denial'' after the Thrashing you got here http://new-indology.blogspot.in/2013/02/indo-iranians-new-perspectives.html
There is no doubt which person is living with the denial....
Nirjhar.
P.S. I f you think that the 1700 B.C. broken Skull of a child of SSC is the sign of ''White Devils invasion to the poor and civilzed Harappan people'' then the truth is it was the time when the SSC was in a transitional phase and climatic catastrophe was on, leading to a Social distress....
N.

Davidski said...

That New Indology blog is like a mental asylum.

Maju said...

Hey, denial is perfectly respectable, Nirjhar. One day, hopefully you get over that phase and move on. But then of course, there's people who get stuck in stubborn denial forever: it absorbs them and destroys them, intellectually and emotionally.

In any case, regardless of your personal destiny, which is ultimately your personal problem, science does not advance through denial: that's invariably a wall to tear down and move on.

"P.S. I f you think that the 1700 B.C. broken Skull of a child of SSC is the sign of ''White Devils invasion to the poor and civilzed Harappan people''"

I think that is part of the evidence. There's quite more. Also, unlike you, I don't study Indoeuropeans focused on South Asia: I do it globally. The main problem with Indian continuity, as well as with European continuity freaks is that they think locally and ethnocentrically, obviating the rest of the World.

Yes there are European IE continuity ideologues too, and, like yourself, they take the part for the whole all the time. Personally I think that it's a waste of time even discussing with them and be sure that I think the same of Indian IE continuity ones and therefore I would send your comments to spam (no hard feelings) if you ever posted to my blog. Because it's a waste of time, sincerely.

So I'm making an exception here in the almost certainly unfounded hope that you might reconsider, if not now, maybe in the future, and that way saving yourself from wasting your life in the pursuit of nonsense.

I won't discuss this further. Have a nice day, year and life.

Nirjhar007 said...

@David
''That New Indology blog is like a mental asylum.''
Yes it is for patients like you and for us is a door way to Future....

Nirjhar007 said...

''Hey, denial is perfectly respectable, Nirjhar. One day, hopefully you get over that phase and move on. But then of course, there's people who get stuck in stubborn denial forever: it absorbs them and destroys them, intellectually and emotionally.

In any case, regardless of your personal destiny, which is ultimately your personal problem, science does not advance through denial: that's invariably a wall to tear down and move on. ''
Hey do not get me wrong as you seems to be an educated person but you are delusional on personal manifestations and you think putting it to the Dispersal Of language group by the mirror of current Happenings is nothing but a Creationist and ultrapathetic approach and of course Unscientific....
''I think that is part of the evidence. There's quite more. Also, unlike you, I don't study Indoeuropeans focused on South Asia: I do it globally. The main problem with Indian continuity, as well as with European continuity freaks is that they think locally and ethnocentrically, obviating the rest of the World.''
No you are missing the point first we have to know the factual history of say Every IE Language group by starting with the Root then we put the facts on global perspective that is the Scientific fact rather we create a global Theory and try to apply on the Language groups forcefully or say by hook or crook, that is exactly what is happening except for example the scholars blog i show...
Look i will have absolutely no problem accepting your proposals but i believe truth can not be brought out by personal deed as it always presents by own and if you think you truly you have the evidence to prove your mettle then just go to the New Indology blog and show it why so shy?...

Maju said...

I've visited your blog, Nirhjar and while there are occasional things that are interesting, most is not. At least not for me.

Indo-Iranian is just one of several IE branches, a numerically important one and IMO the closest one to the original steppe thing, but just one. Have you ever wondered, let alone researched or studied how may IE languages spread in Europe and some other regions of Asia (namely Anatolia and East Turkestan)? Surely not, at least not in any depth. So don't tell me stories about pondering each of the branches: you are pondering only one (and in a very arbitrary manner, BTW).

Naturally we all have our preferential focus an in my case is Europe. I was not particularly interested in Europeans to begin with but rather pre-Indoeuropeans. My whole personal quest to understand so many things about the past begins, other than with a generic curiosity for history and prehistory, with the personal need to unravel the origins of my people, which is not Indoeuropean but an unlikely survivor of indoeuropeanization, just as Dravidians or Balochis may be in your part of the World (mutatis mutandi).

Initially, as I was learning all the involved archaeology (rather than linguistics or mythology, what is at best a side line of research), I tended to perceive IEs as more or less like Gengis Khan Mongols or Attilla's Huns (and admittedly they may have some similitudes, after all they were steppe peoples, the first notorious ones of them) but I say this meaning that I thought they were rather Asiatic. Eventually I had to reckon that this was not their main characteristic and that they were essentially European or West Eurasian in stock (excepted lesser Siberian influences). I also had to admit to some extent that the almost Manichean dichotomy inferred by Gimbutas (who was right in the fundamentals but not in everything) between an idyllic "Old Europe" and the disruptive Patriarchal steppe savages was, at least largely, an ideological fantasy: that most likely the peoples of "Old Europe" also had their social conflicts and were most likely also Patriarchal, even if maybe in a less extremist manner. I had to understand that the Indoeuropean invaders got largely "Europeanized" by their complex interaction with the previous peoples. I had to understand that these previous peoples were also, at least partly, invaders and that the peoples who arrived before them, in the Paleolithic, were as well (where are the Neanderthals now?)

So, in brief: the reality of the past is too complex and very rarely to our like to become emotionally too entangled with it. That I tell you and to everyone else, because a lot of people falls in the trap of ethnocentric emotional adherence to whatever they feel fits their narrative. Me too. But it doesn't work: you discover dark secrets in the "good guys" and nice qualities in the "bad ones", whoever they are for each one.

But what is clear is that, very especially since the Metal Ages, there has been growing social hierarchization and militarization and oppression and exploitation and wars and invasions and injustice and machiavellianism. And I do not condone a single bit of them, but I can't but study and describe them as (pre-)historian, and not fix them, not in the past certainly - maybe in the future if anything.

Well, back to central matter: can you explain me how civilized Indoeuropeans spawned out of Pakistan and India and spread so vigorously and aggressively to West Eurasia? Can you explain me how a civilization conquers the steppe before modern conditions of urban-industrial superiority? Can you even propose a model that makes the slightest archaeological sense for such expansion? Because if the proto-Scythians did not invade India, then Indians must have invaded Scythia and everything else.

I am positive that you don't have the slightest idea on how to globalize your regional pseudo-analysis. Such a problematic is outside your radar, so parochial!

Nirjhar007 said...

Maju,As i have time today i reply with patience....
''I've visited your blog, Nirhjar and while there are occasional things that are interesting, most is not. At least not for me. ''
What do mean by my blog is it the New Indology blog you are saying about or mine blog? In the New Indology you have some unfinished business with Giacomo-
http://new-indology.blogspot.in/2013/02/indo-iranians-new-perspectives.html
Go answer him as not answering to some one is not considered noble in our Tradition....
''Indo-Iranian is just one of several IE branches, a numerically important one and IMO the closest one to the original steppe thing, but just one. Have you ever wondered, let alone researched or studied how may IE languages spread in Europe and some other regions of Asia (namely Anatolia and East Turkestan)? Surely not, at least not in any depth. So don't tell me stories about pondering each of the branches: you are pondering only one (and in a very arbitrary manner, BTW''
Yes the case of the aryans in significant but for the Anatolian dudes are concerned their language is extremely polluted and talking about Linguistics it is indeed our current topic In the New Indology Blog, Feel free to join the discussion...
http://new-indology.blogspot.in/2013/07/indo-european-linguistics-indo-iranian.html

Nirjhar007 said...

''Naturally we all have our preferential focus an in my case is Europe. I was not particularly interested in Europeans to begin with but rather pre-Indoeuropeans. My whole personal quest to understand so many things about the past begins, other than with a generic curiosity for history and prehistory, with the personal need to unravel the origins of my people, which is not Indoeuropean but an unlikely survivor of indoeuropeanization, just as Dravidians or Balochis may be in your part of the World (mutatis mutandi). ''
According to the earliest account of Dravidians the Sangam literature they are the people of the South and Aryas of the north compared to out siders like the Greeks-
http://controversialhistory.blogspot.in/2007/07/who-are-aryans-let-us-see-from.html#.UrurjdIW1m8
And Balochis have experienced many admixtures since even middle ages though the Baloch component is significant.....
My focus is not just the Aryans but the whole but the problem is there are many things to correct on the Identity of the aryans, so it is my Priority...

Nirjhar007 said...

''Initially, as I was learning all the involved archaeology (rather than linguistics or mythology, what is at best a side line of research), I tended to perceive IEs as more or less like Gengis Khan Mongols or Attilla's Huns (and admittedly they may have some similitudes, after all they were steppe peoples, the first notorious ones of them) but I say this meaning that I thought they were rather Asiatic. Eventually I had to reckon that this was not their main characteristic and that they were essentially European or West Eurasian in stock (excepted lesser Siberian influences). I also had to admit to some extent that the almost Manichean dichotomy inferred by Gimbutas (who was right in the fundamentals but not in everything) between an idyllic "Old Europe" and the disruptive Patriarchal steppe savages was, at least largely, an ideological fantasy: that most likely the peoples of "Old Europe" also had their social conflicts and were most likely also Patriarchal, even if maybe in a less extremist manner. I had to understand that the Indoeuropean invaders got largely "Europeanized" by their complex interaction with the previous peoples. I had to understand that these previous peoples were also, at least partly, invaders and that the peoples who arrived before them, in the Paleolithic, were as well (where are the Neanderthals now?)
''
The problem as i told you is a deep one you can say it is quite like the Anthropic principle, we see what we want to see it is the Paradox we face but with true heart in case of aryans if you go deep to the topic you do find them invading but not the Subcontinent but the rest of the India via expansion, And aryans were semi Farming(See RV 4.57)+Semi herding people still pretty much the life style of modern Rural India Where Herding is also important with Farming, but their tradition do go back to the Chalcolithics the only out come in a scientific attitude we are getting now is the ''IE People were present more Anciently than thought in most of the hot spots like Eastern Europe, SC Asia and Near East.....''

Nirjhar007 said...

''So, in brief: the reality of the past is too complex and very rarely to our like to become emotionally too entangled with it. That I tell you and to everyone else, because a lot of people falls in the trap of ethnocentric emotional adherence to whatever they feel fits their narrative. Me too. But it doesn't work: you discover dark secrets in the "good guys" and nice qualities in the "bad ones", whoever they are for each one. ''
Don't worry I'm not an emotional person and i have also in my to get the secrets unveiled and i'm always very curious...
''But what is clear is that, very especially since the Metal Ages, there has been growing social hierarchization and militarization and oppression and exploitation and wars and invasions and injustice and machiavellianism. And I do not condone a single bit of them, but I can't but study and describe them as (pre-)historian, and not fix them, not in the past certainly - maybe in the future if anything. ''
Well past is the pillar of Present which is the Platform of the future...
The main thing is to study with open heart and Neutrality which is the key...

Nirjhar007 said...

''Well, back to central matter: can you explain me how civilized Indoeuropeans spawned out of Pakistan and India and spread so vigorously and aggressively to West Eurasia? Can you explain me how a civilization conquers the steppe before modern conditions of urban-industrial superiority? Can you even propose a model that makes the slightest archaeological sense for such expansion? Because if the proto-Scythians did not invade India, then Indians must have invaded Scythia and everything else.
Telling the Truth i'm more interested in 'What it is' rather than creating or Following with advocating for a theory but hey there is an Expert on the matter He is available for any one, just show more patience and come again to the Blog on Indology, i hope you will get satisfactory and Neutral Scientific answers....
Wishing a happy 2014 and Joyful times....
Nirjhar.

Maju said...

"In the New Indology you have some unfinished business with Giacomo-"...

I was not subscribed for some reason. Possibly because I did not expect a serious discussion. I "answered" (he was not making any questions, mind you) and subscribed now but I still have to see a meaningful explanation that makes any sense other than IA invasion model at Cemetery H, with whatever nuances.