36,200YBP European genome

[motzart]

Instead of comparing Palaeolithic genomes with modern admixtures, I think it would make more sense to look at the percentage of Palaeolithic still found in modern populations. In other words we should look at Palaeolithic genomes the other way round. What's the point of attributing modern labels to a hunter-gatherer who lived 37,000 years ago ? What's interesting is to try to find out what percentage of that ancient genome left contributions in the modern gene pool, just like for Neanderthal.

By the way, the paper says that they identified (only) 0.9 ± 0.4% Neanderthal ancestry in K14, as opposed to 2.4% ± 0.4% in La Braña, and about 2% in modern Eurasians. They estimate the age of the Neandertal admixture at approx. 54,000 years ago (so most likely in the Middle East since Cro-Magnons had not yet reached Europe back then).

I think they do the admixtures backwards too, rather than showing the ancestral genomes as a composition of modern genomes they should show the ancestral genome as homogenous and measure its contribution to modern populations. It seems to me to make a lot more sense to say that Ust Ishim or K14 left little contributions to a wide variety of modern populations than to say he was most closely related to some modern group. Considering descendants of Y DNA K are widespread in Asia it would make logical sense to see the original K group diffused into many genetically different pre-existing Asian groups and making contributions to each relative to their size.

For Neanderthal ancestry it seems like it is vastly underestimated. They make the Neanderthal ancestry judgement based on the (single?) genome of an Altai Neanderthal, considering Neanderthals had been around for hundreds of thousands of years they had to be a diverse group so making that judgement based on one genome seems wrong.

 

[Angela]

They do resemble the European venus figures. Mal'ta had different figurines added to these. Some appear dressed and others are thin stick puppets.

http://donsmaps.com/images24/maltafigurineshermitage.jpg
http://donsmaps.com/malta.html

Don compared these stick puppets to Eskimo children toys. However, I found that the Siberian Ket people actually had similar dolls they dressed up and all that were considered "household deities". The Ket are partly decendant of Mal'ta boys population, as that IBD map shows:

https://verenich.files.wordpress.com/2014/10/maltaibd.png

http://upload.wikimedia.org/wikipedia/en/1/10/Dolls_of_the_Ket_people.jpg
http://en.wikipedia.org/wiki/Ket_people


For real continuity 24.000 years is maybe too long. But traditions and stories of Tasmanian aboriginals are remarkably similar to those on the mainland, even if they have been separated for 10.000 years.

Very interesting. Thanks. I wonder if the Lares and Penates might be related to this?

http://archaeologicalmuseum.jhu.edu/wp-content/uploads/2013/04/NWK4337web.jpg

By the way, this is the Venus from the Grimaldi cave:
http://media-cache-ak0.pinimg.com/236x/2c/9b/48/2c9b4887ee8f2c4fdb4ccc9485d73978.jpg

And, of course, sculptors are still at it. This is Henry Moore's version of the Magna Mater:
http://media-cache-ec0.pinimg.com/236x/27/56/85/275685be779aef24e2601ed7250b6904.jpg

 

[LeBrok]

I still don't understand how it is possible to map the available autosomal DNA from such old samples to contemporary human populations always by 100%. Shouldn't there be always a piece of unmappable DNA, and the older the sample the more unmappable DNA? Even if that ancient individual successfully created long-lasting and flourishing offspring to this day, there was still drift, and drift means loss by replacing older nucleotides by newer nucleotides, either due to mutations or crossover. Or am I wrong? I feel this is probably a naive question, but somehow I don't see or remember the answer at this moment.
If I'm right about unmappable DNA, then this could already explain the neolithic share in the Kostenki sample.

You are correct. It is impossible to recover 100% DNA from ancient bones, and further back you go the smaller it gets. For example in this research, table on page 3, look for Mean Coverage. Numbers bigger than 7 might mean almost full genome, less than 1 mean very small sections recovered. It is hard to find info about number of base pairs of the dna recovered, which would give us full picture.
More info about coverage of genome: https://www.broadinstitute.org/crd/wiki/index.php/Read_coverage

Here is a paragraph from related paper, how hard it is to recover ancient dna and keep it clean from bacterial and living human dna (researchers' dna), remove errors due to shotgun technique and due to dna degeneration since death of individual:

Finally, to further test whether our initial assumption that sequencing errors are
randomly distributed across the genome, we compared the coordinates of our 161
candidate regions with a list of ~400,000 genomic locations where duplicate Illumina
whole-genome sequence data from the same individual had discrepant genotype calls
(without standard filters and with a genotype quality cutoff of GQ ≥ 20; cf. (175)). We
filtered out all candidate regions that contained one or more of these discrepant genotype
calls. 121 regions failed this filter, with each containing an average of 97 discrepant
genotype calls (null expectation of random locations for these Illumina sequencing errors
would lead to ~1.2 per region). This left 40 candidates for ancient introgression, with an
average length of 6.0 Kb. Given the uncertainty in our assumptions regarding the
baseline distributions of sequencing errors and post-mortem DNA damage, and the small
size of the identified regions, we do not believe these observations are necessarily
inconsistent with a null model of no admixture.

 

[LeBrok]

"“In principle, you just have sex with your neighbor and they have it with their next neighbor—you don’t need to have these armies of people moving around to spread the genes.”

I'm not convinced that's true if you have lots of small, mostly stationary groups. If you imagine two populations split up into lots of clans with a fixed territory then if there's 10% bride swapping between the two opposing clans at the border and also 10% with the adjacent clan of the same population away from the border then for that second clan it becomes 10% of 10%. For the next adjacent clan three steps away from the border the chance is 10% of 10% of 10%. The fourth step away is 0.1 * 0.1 * 0.1 * 0.1 etc.


If there were two populations across north Eurasia with the border at the Urals I doubt there would be many of the eastern population's genes reaching Ireland *if* those two populations were split up into lots of small groups with fixed territories who only bride-swapped with their adjacent clans.


On the other hand if you had a region with *nomadic* HGs roaming around then maybe - so I wouldn't be surprised if the mammoth hunters were a meta population.

I think, on a scale of tens and hundreds of years people didn't mix much, as naturally they stick to their own culture, language and religion and prefer to stay separated from others. On scale of thousands of years many mixing events will happen, either migrations, conquests, or even slow but steady and cumulative gene flow.

The slow, rare mixing is interesting though. Your mathematical formula makes lot of sense in a scenario where all genes are equal on selective level. However in cases of very positive mutations, their flow from other tribes will be accelerated or even explosive. This will surely increase transfer of the rest of the genome of the founder of positive mutation. Other words, some admixture can piggyback on a positive mutation.

 

[Greying Wanderer]

Given the look of the skull in question, does the very "Australoid" or "Veddoid" looking reconstruction of the Kostenki young man make sense, or were the original Russian scientists imposing their interpretation of the find onto the remains?

http://news.sciencemag.org/sites/default/files/sn-bust.jpg

http://1.bp.blogspot.com/-7t4meQzsnBY/TrMu5mGheDI/AAAAAAAAH7I/RXb3Mrd69yY/s1600/Kostenki.jpg

FWIW, this is supposedly Sunghir man, also a UP Russian, from 28,000-30,000 BP:
http://donsmaps.com/images8/sunghir1.jpg

I find the cultural artifacts from the site fascinating as well. The most prominent are the so called "Venus" figurines:
http://donsmaps.com/kostenkivenus.html

Their resemblance to the figurines from Germany and the Balkans is quite extraordinary.

From the description it seems they were meant to be worn as amulets around the neck or the waist perhaps. Were they a form of sympathetic magic to ensure fertility? I have to say I find it off putting that often they don't have a head, although the Romans went around wearing reproductions of men's genitals, so I guess I shouldn't be complaining.

Extraordinary really...it would be nice to get inside their psyches for a second to "see" these things as they did.

Ed. Corrected link for Sunghir man

The placing of the (small) hands over the stomach strikes me as being surprisingly consistent.

I'd guess the heads/legs often broke off over time.

 

[Greying Wanderer]

I think, on a scale of tens and hundreds of years people didn't mix much, as naturally they stick to their own culture, language and religion and prefer to stay separated from others. On scale of thousands of years many mixing events will happen, either migrations, conquests, or even slow but steady and cumulative gene flow.

The slow, rare mixing is interesting though. Your mathematical formula makes lot of sense in a scenario where all genes are equal on selective level. However in cases of very positive mutations, their flow from other tribes will be accelerated or even explosive. This will surely increase transfer of the rest of the genome of the founder of positive mutation. Other words, some admixture can piggyback on a positive mutation.

good point

 

[Greying Wanderer]

I think they do the admixtures backwards too, rather than showing the ancestral genomes as a composition of modern genomes they should show the ancestral genome as homogenous and measure its contribution to modern populations. It seems to me to make a lot more sense to say that Ust Ishim or K14 left little contributions to a wide variety of modern populations than to say he was most closely related to some modern group. Considering descendants of Y DNA K are widespread in Asia it would make logical sense to see the original K group diffused into many genetically different pre-existing Asian groups and making contributions to each relative to their size.

For Neanderthal ancestry it seems like it is vastly underestimated. They make the Neanderthal ancestry judgement based on the (single?) genome of an Altai Neanderthal, considering Neanderthals had been around for hundreds of thousands of years they had to be a diverse group so making that judgement based on one genome seems wrong.

Especially if there was more than one archaic involved.

 

[arvistro]

There are statements like:
Every Human being shares 99% of its genome. Or more catchy about humans sharing 50% of genes with banana thing.
Now we are talking about say few % of Neanderthal in humans. Obviously there is a big gap between these 2 concepts. Can anyone suggest me a good read to catch up? Preferrably sort of idiot's guide.

 

[LeBrok]

There are statements like:
Every Human being shares 99% of its genome. Or more catchy about humans sharing 50% of genes with banana thing.
Now we are talking about say few % of Neanderthal in humans. Obviously there is a big gap between these 2 concepts. Can anyone suggest me a good read to catch up? Preferrably sort of idiot's guide.

I never found well written explanation about this kind of discrepancies. If people are 99.999% genetically similar, why am I only 50% like my mother and 4% like my cousin?
What I managed to gather through years, is that it all depends on resolution of genetic similarities. If we compare only genetic functions, regardless if genes, more precisely alleles (mutations of single letters) are identical, we will get very close relationships between banana, chimp and human. We have same genes, meaning genes that perform same functions.
When we increase resolution of comparison, and starting comparing genes, the alleles, to the single letters, similarities starting to drop dramatically. We might have 0% exact matches to banana, 0.5% with chimps, 4% cousins and 50% with mother.
With less resolution we are talking about genetic similarities. With more resolution we are talking about exact matching alleles.

Comparison with Neanderthal is still a bit different animal, so to speak. For this, I think, we need modern African and European genome and Neanderthal one. Then we compare genes of an European to other two, checking one by one which gene is more like Europeans. They won't match exactly to the letter, because of recent mutations from the time we all separated, but most important is which ones are more similar. More similar means more closely related.
In this case about 2.5% genes of single European resemble Neandertal alleles the most. Well, that's my guess how it is done.

 

[LeBrok]

You are correct. It is impossible to recover 100% DNA from ancient bones, and further back you go the smaller it gets. For example in this research, table on page 3, look for Mean Coverage. Numbers bigger than 7 might mean almost full genome, less than 1 mean very small sections recovered. It is hard to find info about number of base pairs of the dna recovered, which would give us full picture.
More info about coverage of genome: https://www.broadinstitute.org/crd/wiki/index.php/Read_coverage

Here is a paragraph from related paper, how hard it is to recover ancient dna and keep it clean from bacterial and living human dna (researchers' dna), remove errors due to shotgun technique and due to dna degeneration since death of individual:

Adding some more.

http://www.eva.mpg.de/neandertal/index.html

DNA sequences were generated on the Illumina HiSeq platform and constitute an average 50-fold coverage of the genome. 99.9% of the 1.7GB of uniquely mappable DNA sequences in the human genome are covered at least ten times.

What exactly does it mean? Expert opinion very welcome!

What I think it means:
1. They "covered", produced enough DNA (through multiplication/copying of DNA) to run their full tests 50 times.
2. 50 runs are enough to cover whole human genome 10 times.
3. "99.9% Uniquely mappable" means without mistakes, exact matches.

At first glance "99.9% of the 1.7GB of uniquely mappable DNA sequences in the human genome" seems like they recovered 99.9% of full genome, but is it the case? Perhaps it means that whatever they recovered they could read with 99.9% precision?

However, they had run the test 50 times, if enough genetic material was collected, they should have been able to reconstruct the full DNA.
I think that they are not sure about how close to the whole genome they got, that's why the wording is not direct, like " we recovered" 99.9% of genome".

 

[epoch]

The IBD map of Ust-Ishim.

The links with Saami and Estonians is clear. The link with Siberians is clear. The very slight link with French and Italians is interesting. The Bulgarian uptick is interesting. Does this seem related to Y-DNA N? Not everywhere, but it seems related nevertheless. It also seems related to Turkics

http://en.wikipedia.org/wiki/Y-DNA_haplogroups_in_Central_and_North_Asian_populations
[url]http://www.eupedia.com/europe/european_y-dna_haplogroups.shtml
[/URL]

 

[Alan]

A Russian called Vadim has a page up with maps comparing IBD's of ancient examples to modern populations. You especially want to check the one of Mal'ta. Although it wasn't the best example according to him.

https://verenich.files.wordpress.com/2014/10/maltaibd.png

Google Translate does a good enough job on this page to roughly make out what it means:

http://verenich.wordpress.com/2014/10/27/сравнение-двух-древних-европейцев-и-о/

It is not only not good, the coloring/shading is so completely wrong.


Just as example

Russian ~20%
Swedish~18%
French ~13,5%
Lezgian ~27%
Iranian = ~19,5%
Turkish = ~16%
Kurdish= ~18%
Tadjik = ~28%
Pathan = ~32.5%
Krygyz = ~18%
Italian = ~10%
Polish = ~19%


https://docs.google.com/spreadsheet...SUrLf8R62NHPbMRySUJ2J48_Q/edit#gid=1410860471


Now compare these results with the map.

 

[JS Bach]

From what I gather, the IBD map may well be more accurate, but I think the admixture calculators are also interesting. In Eurogenes15 it breaks down as:

Population

[TD="bgcolor: #ffffff, align: right"][/TD]

[TD="bgcolor: #ffffff"]North_Sea[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]Atlantic[/TD]
[TD="bgcolor: #F0F0FF, align: right"]11.24%[/TD]

[TD="bgcolor: #ffffff"]Baltic[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]Eastern_Euro[/TD]
[TD="bgcolor: #FBFBFF, align: right"]3.08%[/TD]

[TD="bgcolor: #ffffff"]West_Med[/TD]
[TD="bgcolor: #F8F8FF, align: right"]4.82%[/TD]

[TD="bgcolor: #ffffff"]West_Asian[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]East_Med[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]Red_Sea[/TD]
[TD="bgcolor: #FAFAFF, align: right"]3.36%[/TD]

[TD="bgcolor: #ffffff"]South_Asian[/TD]
[TD="bgcolor: #D7D7FF, align: right"]30.76%[/TD]

[TD="bgcolor: #ffffff"]Southeast_Asian[/TD]
[TD="bgcolor: #EBEBFF, align: right"]15.25%[/TD]

[TD="bgcolor: #ffffff"]Siberian[/TD]
[TD="bgcolor: #FCFCFF, align: right"]2.02%[/TD]

[TD="bgcolor: #ffffff"]Amerindian[/TD]
[TD="bgcolor: #FCFCFF, align: right"]2.20%[/TD]

[TD="bgcolor: #ffffff"]Oceanian[/TD]
[TD="bgcolor: #F0F0FF, align: right"]10.96%[/TD]

[TD="bgcolor: #ffffff"]Northeast_African[/TD]
[TD="bgcolor: #F2F2FF, align: right"]10.08%[/TD]

[TD="bgcolor: #ffffff"]Sub-Saharan[/TD]
[TD="bgcolor: #F7F7FF, align: right"]6.22%[/TD]

Of note, he has 11.24% Atlantic, but has none of the North_Sea, Baltic, West_Asian or East_Med components. I wonder how it got the extreme West Eurasian Atlantic component, but missed those North European, Caucasian and Near Eastern components? Could it be that Y-dna R perhaps made it to the Basque region early? I've noticed in runs of Eurogenes15 with other Ancient dna that the Atlantic component is a popular component, suggesting it's been there for a while. If it's related to Y-dna I then how come the Y-dna J-heavy Caucasian and Near Eastern components are zero? Or maybe it's related to Y-dna C, which might have piggy-backed on Y-dna K and then had a branch go over to where La Brana was? Or maybe it's a Solutrean Q? Who knows?

 

[JS Bach]

Looking at some ancient genomes through Eurogenes15, it's striking how Mal'ta lacks the Atlantic component that Ust'-Ishim has:

[TD="class: xl85"]Eurogenes15[/TD]
[TD="class: xl86"] Ust'-Ishim [/TD]
[TD="class: xl87"]Mal'ta[/TD]

[TD="class: xl65, width: 122"]North_Sea[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl67, width: 69"]15.91%[/TD]

[TD="class: xl65, width: 122"]Atlantic[/TD]
[TD="class: xl68, width: 69"]11.24%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Baltic[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl70, width: 69"]6.54%[/TD]

[TD="class: xl65, width: 122"]Eastern_Euro[/TD]
[TD="class: xl71, width: 69"]3.08%[/TD]
[TD="class: xl72, width: 69"]38.02%[/TD]

[TD="class: xl65, width: 122"]West_Med[/TD]
[TD="class: xl73, width: 69"]4.82%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]West_Asian[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]East_Med[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Red_Sea[/TD]
[TD="class: xl74, width: 69"]3.36%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]South_Asian[/TD]
[TD="class: xl75, width: 69"]30.76%[/TD]
[TD="class: xl76, width: 69"]20.31%[/TD]

[TD="class: xl65, width: 122"]Southeast_Asian[/TD]
[TD="class: xl77, width: 69"]15.25%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Siberian[/TD]
[TD="class: xl78, width: 69"]2.02%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Amerindian[/TD]
[TD="class: xl78, width: 69"]2.20%[/TD]
[TD="class: xl79, width: 69"]18.62%[/TD]

[TD="class: xl65, width: 122"]Oceanian[/TD]
[TD="class: xl68, width: 69"]10.96%[/TD]
[TD="class: xl80, width: 69"]0.12%[/TD]

[TD="class: xl65, width: 122"]Northeast_African[/TD]
[TD="class: xl81, width: 69"]10.08%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl82, width: 122"]Sub-Saharan[/TD]
[TD="class: xl83, width: 69"]6.22%[/TD]
[TD="class: xl84, width: 69"]0.47%[/TD]

Seeing as Mal'ta was Y-dna R, this leads me away from thinking the Atlantic component is associated with R. Although I understand how weak correlations between uni-parental and autosomal markers are these days, maybe in ancient times those associations were stronger.

 

[Alan]

Looking at some ancient genomes through Eurogenes15, it's striking how Mal'ta lacks the Atlantic component that Ust'-Ishim has:

[TD="class: xl85"]Eurogenes15[/TD]
[TD="class: xl86"] Ust'-Ishim [/TD]
[TD="class: xl87"]Mal'ta[/TD]

[TD="class: xl65"]North_Sea[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl67, width: 69"]15.91%[/TD]

[TD="class: xl65, width: 122"]Atlantic[/TD]
[TD="class: xl68, width: 69"]11.24%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Baltic[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl70, width: 69"]6.54%[/TD]

[TD="class: xl65, width: 122"]Eastern_Euro[/TD]
[TD="class: xl71, width: 69"]3.08%[/TD]
[TD="class: xl72, width: 69"]38.02%[/TD]

[TD="class: xl65, width: 122"]West_Med[/TD]
[TD="class: xl73, width: 69"]4.82%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]West_Asian[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]East_Med[/TD]
[TD="class: xl66, width: 69"]-[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Red_Sea[/TD]
[TD="class: xl74, width: 69"]3.36%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]South_Asian[/TD]
[TD="class: xl75, width: 69"]30.76%[/TD]
[TD="class: xl76, width: 69"]20.31%[/TD]

[TD="class: xl65, width: 122"]Southeast_Asian[/TD]
[TD="class: xl77, width: 69"]15.25%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Siberian[/TD]
[TD="class: xl78, width: 69"]2.02%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl65, width: 122"]Amerindian[/TD]
[TD="class: xl78, width: 69"]2.20%[/TD]
[TD="class: xl79, width: 69"]18.62%[/TD]

[TD="class: xl65, width: 122"]Oceanian[/TD]
[TD="class: xl68, width: 69"]10.96%[/TD]
[TD="class: xl80, width: 69"]0.12%[/TD]

[TD="class: xl65, width: 122"]Northeast_African[/TD]
[TD="class: xl81, width: 69"]10.08%[/TD]
[TD="class: xl69, width: 69"]-[/TD]

[TD="class: xl82, width: 122"]Sub-Saharan[/TD]
[TD="class: xl83, width: 69"]6.22%[/TD]
[TD="class: xl84, width: 69"]0.47%[/TD]

Seeing as Mal'ta was Y-dna R, this leads me away from thinking the Atlantic component is associated with R. Although I understand how weak correlations between uni-parental and autosomal markers are these days, maybe in ancient times those associations were stronger.

In this calculator most of the Gedrosia get eaten up by "South Asian" and Caucasus by "Eastern Euro" I remember very well.

 

[LeBrok]

Interesting. They really went separate ways, only mating together in South Asia.

 

[epoch]

@Alan

That was an IBD map, not just ANE admixture. IBD's are series of SNP's, each in itself not necessarily unique to an admixture, recognizable in offspring. The order is what makes them unique, if I understand it all well, not the mutations of drifts. IBD should be markers of decent.

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

Could the difference be that the original ANE population was far larger than the tribe the Mal'ta boy lived in?

 

[ElHorsto]

@Alan

That was an IBD map, not just ANE admixture. IBD's are series of SNP's, each in itself not necessarily unique to an admixture, recognizable in offspring. The order is what makes them unique, if I understand it all well, not the mutations of drifts. IBD should be markers of decent.

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

Could the difference be that the original ANE population was far larger than the tribe the Mal'ta boy lived in?

I think you are right. IBD patterns are more fragile, somewhat like haplogroups, while admixtures are more robust by relying on independent SNP's and are thus more continuously distributed. Thus ANE admixture could have been much more widespread than the Mal'ta tribe already back then.

 

[Greying Wanderer]

From what I gather, the IBD map may well be more accurate, but I think the admixture calculators are also interesting. In Eurogenes15 it breaks down as:

Population

[TD="bgcolor: #ffffff, align: right"][/TD]

[TD="bgcolor: #ffffff"]North_Sea[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]Atlantic[/TD]
[TD="bgcolor: #F0F0FF, align: right"]11.24%[/TD]

[TD="bgcolor: #ffffff"]Baltic[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]Eastern_Euro[/TD]
[TD="bgcolor: #FBFBFF, align: right"]3.08%[/TD]

[TD="bgcolor: #ffffff"]West_Med[/TD]
[TD="bgcolor: #F8F8FF, align: right"]4.82%[/TD]

[TD="bgcolor: #ffffff"]West_Asian[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]East_Med[/TD]
[TD="bgcolor: #ffffff, align: right"]-[/TD]

[TD="bgcolor: #ffffff"]Red_Sea[/TD]
[TD="bgcolor: #FAFAFF, align: right"]3.36%[/TD]

[TD="bgcolor: #ffffff"]South_Asian[/TD]
[TD="bgcolor: #D7D7FF, align: right"]30.76%[/TD]

[TD="bgcolor: #ffffff"]Southeast_Asian[/TD]
[TD="bgcolor: #EBEBFF, align: right"]15.25%[/TD]

[TD="bgcolor: #ffffff"]Siberian[/TD]
[TD="bgcolor: #FCFCFF, align: right"]2.02%[/TD]

[TD="bgcolor: #ffffff"]Amerindian[/TD]
[TD="bgcolor: #FCFCFF, align: right"]2.20%[/TD]

[TD="bgcolor: #ffffff"]Oceanian[/TD]
[TD="bgcolor: #F0F0FF, align: right"]10.96%[/TD]

[TD="bgcolor: #ffffff"]Northeast_African[/TD]
[TD="bgcolor: #F2F2FF, align: right"]10.08%[/TD]

[TD="bgcolor: #ffffff"]Sub-Saharan[/TD]
[TD="bgcolor: #F7F7FF, align: right"]6.22%[/TD]

Of note, he has 11.24% Atlantic, but has none of the North_Sea, Baltic, West_Asian or East_Med components. I wonder how it got the extreme West Eurasian Atlantic component, but missed those North European, Caucasian and Near Eastern components? Could it be that Y-dna R perhaps made it to the Basque region early? I've noticed in runs of Eurogenes15 with other Ancient dna that the Atlantic component is a popular component, suggesting it's been there for a while. If it's related to Y-dna I then how come the Y-dna J-heavy Caucasian and Near Eastern components are zero? Or maybe it's related to Y-dna C, which might have piggy-backed on Y-dna K and then had a branch go over to where La Brana was? Or maybe it's a Solutrean Q? Who knows?

I tend to think it's layers of expansions with later ones largely but not completely covering over the earlier ones leaving behind a collection of autosomal signals from each layer, particularly in refuge zones.

So in my model I imagine a population coming out of the African tropical zone into the rest of Africa and across Eurasia, coastal dominated. So that's the first layer, layer B.

Then somewhere within its range (S/SE Asia maybe) a segment of the B population develops a major advantage of some kind becoming population C and that population then expands out from its source region *over the top* of the B layer.

So now you have a C layer over most of the globe with B mostly surviving as an autosomal component within C. The proportion that survives varies from place to place with higher quantities in certain refuge zones (mountains, swamps etc).

If that was correct then it seems plausible to me that if the C layer was out of S/SE Asia then the B layer might have survived best in refuge zones in the farthest periphery e.g. the Atlantic coast. These zones might include mountainous areas along the coast like Atlantic Iberia or Brittany but might include bits of the Baltic coast also. This above average underlying signal would pull out the "Atlantic" component.

So in this model Ust-Ishim would represent a part of the C layer population with a higher than average amount of the B layer.

(This would explain the very high S/SE Asian/Oceanian components (the C layer) with Atlantic (above average amount of the B layer.)

(This would imply the "North European" and "Caucasian/Near Eastern/West Asian" components were other separate layers - possibly connected to I and J in some way.)

.

The combination of the B and C layers might then create two "basal" signals: a widespread one where the C layer is dominant and there is only a small autosomal B signal and a more western coastal orientated basal where the B layer is a significant component.

The test would be looking for a slightly different "basal" along the Atlantic coast and maybe parts of the med. coast also.)

 

[Angela]

I don't know if everyone has seen Martin Sikorra's post on Razib Khan's blog. He's one of the study authors.

http://www.unz.com/gnxp/

Hi Razib,
after reading your post it I thought it would not hurt to chime in with a bit of perspective from my side, as I don’t entirely agree with some of your criticisms. Some of the reactions to our paper have caught me a little by surprise, but in retrospect it probably reflects the complexity of the story, which is something I also struggled with (and still am!).

Part of the confusion seems to be that it is assumed that since we find that K14 somehow relates to all three European ancestral proposed by Lazaridis et al., that it necessarily also has contributed these components to modern Europeans. In your post you also seem to imply that, i.e we don’t “acknowledge the possibility that K14 did not leave modern descendants, and was part of an early population which did not end up flourishing”. I actually agree with the early population part, and we also acknowledge that in our suggested model in Figure 2, which does not have a K14-related population directly contributing to modern Europeans. What one can say with reasonable certainty though is that K14 does share substantial amount of ancestry with Mesolithic Hunter-gatherers (and therefore modern Europeans by extension), but at the same time appears less close to East Asians than all Western Eurasians, so things are complex.

Therefore if you take the Lazaridis et al. model as a backbone, you need some extra gene flow to account for that, be it from Basal Eurasian into K14, or some sort of basal gene flow between East Asia and early West Eurasians, post-K14 but pre-ANE/HG split. While we don’t have the resolution to be sure, our results do suggest that K14 was close to or a already somewhat down the HG branch of the ANE/HG split, which implies that those proposed components would not only have to be already somewhat differentiated by 36 kya, but also already have had mixed to a certain extent.

Regarding your take on the PCA results, I would disagree and say that these are very much what you would expect for an individual of that age. K14 is after all ~36,000 years closer to the East Asia / West Eurasia split, so it lacks a substantial amount of drift on the European branch. It is nevertheless shifted towards Europe on PC1 from the origin as expected (a bit more so than MA1 actually). Pontus Skoglund had a nice recent paper in MBE that demonstrates the same effect (see Figure 9 in doi:10.1093/molbev/msu1920). As you say, using modern variation to infer affinities of ancient samples has limitations, and PCs are often hard to interpret. In the same spirit I would also not interpret the different admixture components in K14 as itself being admixed with all those components, but rather reflecting ancestral relationship with modern populations represented by these components. The same is obviously true for the “Middle East” component, but it still implies that K14 somehow relates ancestrally to those populations whereas all other HGs including MA1 do not.

Overall, I do think that migrations played an important role, e.g. I don’t think that “Basal Eurasian” came with K14 to Central Europe or was already present back then in another way, that seems pretty clear. I would also not say that our results are necessarily a refutation of the Lazaridis et al model, but I do think they show that it seems to have been already quite complicated in the Upper Paleolithic. If you need a new migration/component for every new individual, to me this questions at least to a some extent whether one can really talk about three or any other number of discrete ancestral populations for all modern Europeans. Personally I would expect ancient samples from the Caucasus or Central Asia to yet again spring some surprises. The cool thing is that we’ll probably know soon, since many groups are adding more and more samples to the picture.
Anyways, I just wanted to share my thoughts, hope this clears up things a bit.

Btw regrading your subsequent ANE post, I can confirm that those are the Kalash. Interesting also that the correspondingly the Kalash ADMIXTURE component shows up in MA1, but is almost absent in K14 (see our Figure S20).