PDA

View Full Version : Analysis of over 2,000 HVR1 profiles From SouthWest Asia



Krefter
04-16-2015, 04:30 AM
I got the HVR1 data from Table S1. of the 2013 study “Y-Chromosome and mtDNA Genetics Reveal Significant Contrasts in Affinities of Modern Middle Eastern Populations with European and African Populations (http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0054616)”.

I plan on doing this same-type of analysis with as many modern pops as possible. The online sources we have available, like Wikpedia, are pretty much useless.

Here's the link to my analysis.

https://docs.google.com/document/d/1BxqemNoVOk1XVP73awfpAL9oG6c8F5w6m9s-hACH3Ws/edit?usp=sharing

I broke down the haplotypes as much as they possibly could be to understand the maternal gene pool of SouthWest Asia. Most of what I learned is self-explainable if you look at the haplogroup frequencies and the haplotypes. I don't see a need for a lengthy description. Later I will do more work on this.

In terms of variation in SouthWest Asia: Lebanon, Palestine, Syria, and Jordan are practically indistinguishable from each other. Yeman and Saudi Arabia are united in differences from those more northwestern countries. So the split in SouthWest Asia seems to be NorthWestern vs SouthEastern.

Yeman and Saudi Arabia have more of the African haplogroup L(xM, N), especially Yeman with 38.3%. Yeman and Saudi Arabia have much less H and HV(xH), much more R0a(and more diversity in R0), J, and N1a1a. There's also a higher frequency and variety of M in Saudi Arabia and Yeman.

Because Neolithic Central Europeans were mostly of ancient West Asian decent(50-70% ENF) but separated by over 8,000 years from modern SouthWest Asians, I compared the two(not in documents or spreadsheets yet).

It is quite obvious the two are not closely related maternally. Neolithic Central Europeans trace most of the maternal lineages to Stone age West Asian-East Mediterranean women, who's lineages didn't do well in SouthWest Asia.

Here's my comparison of the two. I'll do a more formal comparison with a spreadsheet sometime in the future.

https://docs.google.com/document/d/1L8YOd8ntIAAjUIAfd8aCMBar7mTKi2qIixinnjUA94k/edit

Overall IMO SouthWest Asians are very native to SouthWest Asia. There aren't signs in autosomal DNA of any significant gene flow from other regions into SouthWest Asia besides ANE. It's probably safe to assume most of their maternal lineages have been evolving in the broad region of the Middle East for 10,000s of years.

"ENF" ancestry might be very old and very pre-Neolithic. ENF-rich people may have already had alot of regional(ethnic, etc.) mtDNA-diversity 9,000 years ago when Neolithic Central European's ENF ancestors left. This can explain why modern SouthWest Asian mtDNA is so differnt from EEF, and why they have so much diversity in "West Eurasian" haplogroups.

A 2001 study found R0a1a and R0a2c in Upper Palaeolithic Morocoo dating to 10,000BC. It could very well be contamination. Although this is what I would expect.

The package of "Near Eastern" lineages which arrived in Europe with farming, had been evolving in the Middle East for 1,000s of years before farming existed. They did not expand with farming like in Europe. So, I expect Middle Eastern hunter gatherers to display, T, J, R0, N1, etc.

Generalissimo
04-16-2015, 05:17 AM
Can you locate a specific source for the Neolithic mtDNA lineages in the Yamnaya and Corded Ware?

Krefter
04-16-2015, 05:38 AM
Can you locate a specific source for the Neolithic mtDNA lineages in the Yamnaya and Corded Ware?

Maybe with samples from all over West Eurasia at better coverage a pattern related to where Yamna's ENF ancestors came from would pop up. It didn't come from anywhere near Hungary that's for sure. SouthWest Asians appear to be more related mtDNA wise to the ENF ancestors of Yamna than EEF is.

Because of this we can exclude all of Europe to as far east As Croatia. So, I guess the Caucasus, Black sea area, and Central Asia. That would make sense considering the Teal-Georgian-Armenian-like ancestors of Yamna.

palamede
04-16-2015, 11:55 AM
.....
Because Neolithic Central Europeans were mostly of ancient West Asian decent(50-70% ENF) but separated by over 8,000 years from modern SouthWest Asians, I compared the two(not in documents or spreadsheets yet).

It is quite obvious the two are not closely related maternally. Neolithic Central Europeans trace most of the maternal lineages to Stone age West Asian-East Mediterranean women, who's lineages didn't do well in SouthWest Asia.

.....

Don't forget Stuttgart the prototype of ENF Early Neolithic Farmer is only 30% of SW Asian origin (Eurogenes K15 East Med 26.5% Red Sea 3.5%. The remaining part is mainly from European origin 47% West Med (Mesolithic of South Europe and partly in Norther Africa) due to the mixture of Near Easter migrants with local mesolithicsin Balkans and Danubian basun. The remaining 22% Atlantic is local in South Germany.

Generalissimo
04-16-2015, 01:56 PM
Stuttgart is almost 100% Near Eastern.

Of course, the Near East that Stuttgart's ancestors came from no longer exists, but it'll be interesting to see how some people cope with the idea that Neolithic farmers from the ancient northern Near East were more European-like than many present-day Europeans.

I expect a long hangover and lots of vitriol on various forums, including this one. All we need is a single Neolithic genome from the Near East to get the ball rolling. Bring it on I say.

Arbogan
04-16-2015, 02:43 PM
Stuttgart is almost 100% Near Eastern.

Of course, the Near East that Stuttgart's ancestors came from no longer exists, but it'll be interesting to see how some people cope with the idea that Neolithic farmers from the ancient northern Near East were more European-like than many present-day Europeans.

I expect a long hangover and lots of vitriol on various forums, including this one. All we need is a single Neolithic genome from the Near East to get the ball rolling. Bring it on I say.

According to what? What would cause such a massive genetic over turn? Arent Stuttgart half whg? You'd think that if this is true some pockets of Stuttgart like groups would survive. Like the basques. Its more likely that european neolithics were a subset of neolithics. Than somehow neolithics in near east were smallowed by sea of south west asian arabs or other immigration or disappeared into a vacuum.

Generalissimo
04-16-2015, 02:55 PM
According to what? What would cause such a massive genetic over turn? Arent Stuttgart half whg? You'd think that if this is true some pockets of Stuttgart like groups would survive. Like the basques. Its more likely that european neolithics were a subset of neolithics. Than somehow neolithics in near east were smallowed by sea of south west asian arabs or other immigration or dropped off the face of the earth.

Some good questions there. I'm sure they'll be answered in time, when enough ancient genomes are sequenced.

Ignis90
04-16-2015, 03:38 PM
Some good questions there. I'm sure they'll be answered in time, when enough ancient genomes are sequenced.

Is it possible to try a PCA of West Eurasia (like the good-old ones) in order to see if some Neolithic farmers like Stuttgart can cluster close to NorthWest Africans? It wasn't possible with the K8 ones since I clustered with [ANE-admixed] Sephardi Jews. Isn't there a way to neutralize the non-Eurasian part of North Africans so it doesn't skew the results?

Arbogan
04-16-2015, 06:37 PM
Is it possible to try a PCA of West Eurasia (like the good-old ones) in order to see if some Neolithic farmers like Stuttgart can cluster close to NorthWest Africans? It wasn't possible with the K8 ones since I clustered with [ANE-admixed] Sephardi Jews. Isn't there a way to neutralize the non-Eurasian part of North Africans so it doesn't skew the results?
If you took away the sub Saharan snp frequencies in north west africans. You'd be close to sardinians and hence stuttgart.

Krefter
04-16-2015, 08:52 PM
According to what? What would cause such a massive genetic over turn? Arent Stuttgart half whg? You'd think that if this is true some pockets of Stuttgart like groups would survive. Like the basques. Its more likely that european neolithics were a subset of neolithics. Than somehow neolithics in near east were smallowed by sea of south west asian arabs or other immigration or disappeared into a vacuum.

I think what he was saying is that WHG/UHG was very high in some parts of West Asia during the Neolithic and before, and ANE was absent. The fact Neolithic Europeans fit better as the non-ANE ancestors of modern West Asians than anyone around today is very telling. It probably means they're our best proxy of Neolithic west Asians.

J Man
04-16-2015, 08:57 PM
I think that it is very likely that the ENF autosomal type component will be of pre-Neolithic origins in the Near/Middle East. Still though it will most likely be the main autosomal component of the very first Neolithic farmers of the Near/Middle East as they descended from Near/Middle Eastern hunter-gatherers.

J Man
04-16-2015, 09:05 PM
I think what he was saying is that WHG/UHG was very high in some parts of West Asia during the Neolithic and before, and ANE was absent. The fact Neolithic Europeans fit better as the non-ANE ancestors of modern West Asians than anyone around today is very telling. It probably means they're our best proxy of Neolithic west Asians.

That is possible. Hopefully soonish ancient DNA will tell us if that is true or not.

Arbogan
04-16-2015, 09:08 PM
I think what he was saying is that WHG/UHG was very high in some parts of West Asia during the Neolithic and before, and ANE was absent. The fact Neolithic Europeans fit better as the non-ANE ancestors of modern West Asians than anyone around today is very telling. It probably means they're our best proxy of Neolithic west Asians.

Uhm no. They fit better as ancestors for mediterranean. Not for our ancestors. Also we have no way of knowing whether stuttgart is part WHG or something else. What we know so far is small fraction of variation. The neolithic groups responsible for dispersions. I have a hard-time imaging stuttgart as related to gökili-tepe or chogamish. ANE or not, it's unlikely they were the only people present.

Krefter
04-16-2015, 09:13 PM
Uhm no. They fit better as ancestors for mediterranean. Not for our ancestors. Also we have no way of knowing whether stuttgart is part WHG or something else. What we know so far is small fraction of variation. The neolithic groups responsible for dispersions. I have a hard-time imaging stuttgart as related to gökili-tepe or chogamish. ANE or not, it's unlikely they were the only people present.

In formal stats like D-stats and F-stats Neolithic Europeans fit as an ancestor for west Asians. West Asians fit best as EEF+MA1.

http://eurogenes.blogspot.com/2014/11/3-pop-analysis-featuring-19-ancient.html

This is probably because EEF lacks ANE. Neolithic Euro's ancestors had just recently left the Middle East, and their distant cousins are the main ancestors of modern ones.

Arbogan
04-16-2015, 09:24 PM
In formal stats like D-stats and F-stats Neolithic Europeans fit as an ancestor for west Asians. West Asians fit best as EEF+MA1.

http://eurogenes.blogspot.com/2014/11/3-pop-analysis-featuring-19-ancient.html

This is probably because EEF lacks ANE. Neolithic Euro's ancestors had just recently left the Middle East, and their distant cousins are the main ancestors of modern ones.

Those fits are hypotheticals, shouldn't be taken literally. And if this was the case UHG/WHG would be widespread and occur at at very least 10% in west-asia. Besides turks and turkmen aren't good proxys for near-eastern neolithics, because of ionians, and sucessive waves from europe, while turkmen have quite alot of steppe ancestry. Run me these D statistics or an assyrian, or an iraqi-jew. You'll see this hypothesis makes no sense based on the absenceof WHG/UHG found in stuttgart.

edit:Ok looked at the sheet. There is still something amiss. How do you explain the lack of WHG/UHG? What these are describing is ANE influence. But nothing about WHG.

Krefter
04-16-2015, 09:47 PM
Those fits are hypotheticals, shouldn't be taken literally. And if this was the case UHG/WHG would be widespread and occur at at very least 10% in west-asia. Besides turks and turkmen aren't good proxys for near-eastern neolithics, because of ionians, and sucessive waves from europe, while turkmen have quite alot of steppe ancestry. Run me these D statistics or an assyrian, or an iraqi-jew. You'll see this hypothesis makes no sense based on the absenceof WHG/UHG found in stuttgart.

edit:Ok looked at the sheet. There is still something amiss. How do you explain the lack of WHG/UHG? What these are describing is ANE influence. But nothing about WHG.

A lack of ANE is probably why EEF fits as an ancestor of West Asians. WHG-Basal Eurasian levels can be thrown around each way, but ANE makes the biggest difference. West Asians have WHG/UHG-ancestry just at a much smaller amount than EEF did. There's unipaternal connections between European-WHG and modern West Asians, with Y DNA IJ and mtDNA U.

Arbogan
04-16-2015, 09:56 PM
A lack of ANE is probably why EEF fits as an ancestor of West Asians. WHG-Basal Eurasian levels can be thrown around each way, but ANE makes the biggest difference. West Asians have WHG/UHG-ancestry .

Too small to be attributed the EFFs. who are like 40-50% WHG. ANE admixture is tops 20% how would that reduce to miniscule levels. Even if you add the south-west-asian element. It still wouldn't lower it to the levels found in west-asians. More likely, that they received it from external sources like admixture with post-neolithic europeans or steppe people.

Krefter
04-16-2015, 10:06 PM
Too small to be attributed the EFFs. who are like 40-50% WHG.

What matters is that they didn't have African, ANE, and South Asian. Those are more differnt to Neolithic West Asians than WHG is.

Arbogan
04-16-2015, 10:11 PM
What matters is that they didn't have African, ANE, and South Asian. Those are more differnt to Neolithic West Asians than WHG is.

Again african admixture is something like 0-2% in non-semitic speaking west-asians. ANE is 20% and south-asian(Which itself is half neolithic-like) admixture is neglible in everyone save for iranics. We still get to the point where the supposed "erosion" of WHG doesn't explain the low WHG. Even in some places which have been directly in-contact with WHG- rich populations. More likely the ENF, in EFF, is what is signalling the MA1+EFF. There would be a 40-50% overlap in that case.

Krefter
04-16-2015, 10:20 PM
Again african admixture is something like 0-2% in non-semitic speaking west-asians. ANE is 20% and south-asian(Which itself is half neolithic) admixture is neglible in everyone save for iranics. We still get to the point where the supposed "errosion" of WHG doesn't explain the low WHG. Even in some places which have been directly in-contact with WHG- rich populations.

I never said there was erosion of WHG. IN some parts of west Asia, like Turkey, WHG may have been very high. My point was that EEF being a simple Basal Eurasian-WHG mix, makes them a good ancestor proxy for west Asians.

Agamemnon
04-17-2015, 12:06 AM
Theoretically, the fact that haplogroups I and J share a common ancestor (IJ) should make an association with WHG-like ancestry all the more likely... But I'm afraid things aren't as clear cut, if anything they might get much more complicated once we get actual results from the Near East.

Arbogan
04-17-2015, 01:34 AM
Theoretically, the fact that haplogroups I and J share a common ancestor (IJ) should make an association with WHG-like ancestry all the more likely... But I'm afraid things aren't as clear cut, if anything they might get much more complicated once we get actual results from the Near East.

If there is an association between real genetic association between (beyond sharing ancestors, god knows when)IJ, it's certainly not being expressed in autosomal terms.

J Man
04-17-2015, 09:51 PM
Theoretically, the fact that haplogroups I and J share a common ancestor (IJ) should make an association with WHG-like ancestry all the more likely... But I'm afraid things aren't as clear cut, if anything they might get much more complicated once we get actual results from the Near East.

Yes I have thought for a while now that the WHG like ancestry that so many think will be present among the early Neolithic farmers of the Near East may come from Y-DNA haplogroup J and mtDNA haplogroup U ancestors as Y-DNA haplogroup J is distantly related to European haplogroup I and mtDNA haplogroup U types from the Near East are distantly related to European mtDNA haplogroup U types (U5, U4 and U2e). Of course this could all be wrong as well though. Hopefully in the not to distant future ancient DNA from the Mesolithic and Neolithic Near East will clear all of this up.

Krefter
04-17-2015, 10:06 PM
If there is an association between real genetic association between (beyond sharing ancestors, god knows when)IJ, it's certainly not being expressed in autosomal terms.

UHG/WHG exists in the middle east. No test has been designed to test how much though. All so far have been designed to measure WHG in Euros, and all of west Asian's WHG is expressed via ENF.

Arbogan
04-17-2015, 10:27 PM
UHG/WHG exists in the middle east. No test has been designed to test how much though. All so far have been designed to measure WHG in Euros, and all of west Asian's WHG is expressed via ENF.

A tiny amount, and not enough to attribute it ENF. It could just as possible be a result of back immigration as much as you can assume1 it's native. Like i've said. This entire thing is in a too preliminary stage for us to presume that EFF, are the main west-eurasian neolithic group.

Krefter
04-18-2015, 02:42 AM
This is the next study I'll work on. They have much better coverage than the last one. I should be done in the next few weeks. There are less than half as many haplotypes to work with.

http://www.sciencedirect.com/science/article/pii/S0002929707643523