PDA

View Full Version : South Asia Y-DNA Distribution



Mehrdad
01-31-2014, 06:41 PM
I thought I'd start a discussion to mirror the Assyrian Y-DNA Distribution on South Asia. Thanks Humanist for the Idea :)

I gathered the data from a link (http://www.plosone.org/article/fetchSingleRepresentation.action?uri=info:doi/10.1371/journal.pone.0050269.s005) that Parasar had included in on Zack's Harappa DNA Project (http://www.harappadna.org/2013/12/webhost-move/#comments)

1331

I was really surprised at how high J2 was in South Asia and R2 was lower. I was actually expecting the opposite. Why do you guys think this would have happened. Considering R2 may have been first into South Asia before J2.


1332

Humanist
01-31-2014, 06:45 PM
I thought I'd start a discussion to mirror the Assyrian Y-DNA Distribution on South Asia. Thanks Humanist for the Idea :)

Thanks for creating the thread. Would also love to see the same for mtDNA data. :)

newtoboard
01-31-2014, 07:16 PM
A few thoughts.

-J2 doesn't seem all that high given we know some parts of South Asia have pretty high frequencies of J2b.
-Were these charts created before the discovery of H3 because most South Asian F ended up being H2/H3. But then again I would be surprised to see a frequency of H3 at such a high level. The F here likely represents a mixture of actual F, H3, G and J imo.
-R*? Probably R2 imo.
-More C1b1 than I would have thought.

Tomasso29
01-31-2014, 10:34 PM
I was really surprised at how high J2 was in South Asia and R2 was lower. I was actually expecting the opposite. Why do you guys think this would have happened. Considering R2 may have been first into South Asia before J2.

The better question is why do we consider R2a to be an earlier arrival before J2 to South Asia? Are we talking based on STR calculations? If so, take them with a grain of salt. I think the spread of R2 indicates that it's a younger lineage than its older brother R1, and younger by a lot.

Mehrdad
01-31-2014, 10:47 PM
The better question is why do we consider R2a to be an earlier arrival before J2 to South Asia? Are we talking based on STR calculations? If so, take them with a grain of salt. I think the spread of R2 indicates that it's a younger lineage than its older brother R1, and younger by a lot.

Interesting, you're correct, I was referring to STR's. So when you said younger a lot, by how many years or millenia are we talking about here?

Tomasso29
01-31-2014, 11:01 PM
Interesting, you're correct, I was referring to STR's. So when you said younger a lot, by how many years or millenia are we talking about here?

I don't think there's an answer to that, it's all speculations based on personal opinion on how the current findings and distributions are. STR calculations get abused in these studies and people end up taking them too serious in their lineage timelines. We have to remind that STR mutations are random and can happen at any moment. Yes it's true that some markers probably mutate faster than others and you can possibly assume a mutation rate, but at the end of the day, it's still random.

A good example of this is if you take some of the older studies done on R2-M124 you would believe that Southeast India has the highest diversity which at one point was suggested to be an origin point. Well recently through FTDNA we were able to find the most useful SNP under R2-M124 in L295 (R2a1). All the Indian samples from that Southeast India that tested for this SNP in the project came back positive.

Anyways, I don't want to drift away from the subject, but my main point is we have to take caution when estimating timelines. I believe these calculations and estimates are more proper for recent ancestry and if you want to determine your connection to close matches. Other than that, they're just a grain of salt.

newtoboard
02-01-2014, 11:50 AM
The better question is why do we consider R2a to be an earlier arrival before J2 to South Asia? Are we talking based on STR calculations? If so, take them with a grain of salt. I think the spread of R2 indicates that it's a younger lineage than its older brother R1, and younger by a lot.

Well you are in disagreement with most people there. R2 likely spread south during the Paleolithic/Mesolithic because it was unlikely it was hanging out in the Central Asian desserts and unlikely it took a route to Europe or the South Caspian.

DMXX
02-01-2014, 02:32 PM
Well you are in disagreement with most people there.


In the mid-2000's, one Vincent Vizachero (vineviz (http://www.anthrogenica.com/member.php?35-vineviz)) defied the popular stance on DNA-Forums when it became clear that Y-DNA R1b's genetic pattern did not support a "Cro Magnon" era origin in the Iberian refugium. He ended up being correct a thousand times over on that.

We should all be mindful of the above, perfectly exemplifying the logical fallacy of argumentam ad populum. A majority view is not necessarily the right one. Even on this forum.



R2 likely spread south during the Paleolithic/Mesolithic because it was unlikely it was hanging out in the Central Asian desserts and unlikely it took a route to Europe or the South Caspian.

This may well be correct, but Tomasso's point was centred around primary genetic evidence. If it is to be challenged, contrary primary genetic evidence is needed. Prehistoric geography is useful in confirming postulated movements but it certainly isn't contrary evidence.

newtoboard
02-01-2014, 02:41 PM
In the mid-2000's, one Vincent Vizachero (vineviz (http://www.anthrogenica.com/member.php?35-vineviz)) defied the popular stance on DNA-Forums when it became clear that Y-DNA R1b's genetic pattern did not support a "Cro Magnon" era origin in the Iberian refugium. He ended up being correct a thousand times over on that.

We should all be mindful of the above, perfectly exemplifying the logical fallacy of argumentam ad populum. A majority view is not necessarily the right one. Even on this forum.



This may well be correct, but Tomasso's point was centred around primary genetic evidence. If it is to be challenged, contrary primary genetic evidence is needed. Prehistoric geography is useful in confirming postulated movements but it certainly isn't contrary evidence.

All I saw in his posts were his opinions and a bit of information on SE India. Which hardly proves that R2a doesn't have an ancient presence in the rest of South Asia. Nor do most of the calculated dates agree with R2 being much younger than R1.

DMXX
02-01-2014, 02:44 PM
All I saw in his posts were his opinions and a bit of information on SE India. Which hardly proves that R2a doesn't have an ancient presence in the rest of South Asia. Nor do most of the calculated dates agree with R2 being much younger than R1.

I was referring to the SE Asian origin part in my post.

And, why invest any esteem in the calculated dates when all the older studies based them off STR calculations, which Tomasso correctly described as being somewhat misguiding?

The most accurate way of gauging the age of R2a-M124 right now would be through SNP's rather than STR's. No paper I'm aware of has done so.

newtoboard
02-01-2014, 03:05 PM
I was referring to the SE Asian origin part in my post.

And, why invest any esteem in the calculated dates when all the older studies based them off STR calculations, which Tomasso correctly described as being somewhat misguiding?

The most accurate way of gauging the age of R2a-M124 right now would be through SNP's rather than STR's. No paper I'm aware of has done so.

I believe a very old date was calculated by Michal quite recently. Either way R2 might be younger than R1 but is is unlikely IMO it entered South Asia after J2 since J2 likely entered South Asia in the Neolithic or just prior to that.

Tomasso29
02-01-2014, 04:06 PM
All I saw in his posts were his opinions and a bit of information on SE India. Which hardly proves that R2a doesn't have an ancient presence in the rest of South Asia. Nor do most of the calculated dates agree with R2 being much younger than R1.

The information bit I shared about Southeast India being an origin point was just an example on how STR calculations can be misleading. It was not to argue whether R2 entered South Asia before J2 or not. Simply put, determining a haplogroup's age through STR calculation is very misleading, and if the evidence for R2 being in South Asia before J2 is based on that, my comment is to take it with a grain of salt.

As for R2 being younger than R1, I have no evidence, it's an opinion I share with the way both haplogroups have spread. I think the actual mutation for R2 may have developed anywhere from the northern parts of South Asia, to the southern parts of Central Asia, and to the eastern parts of West Asia. While it was developing I would not be surprised if R1a1a-Z40 had already existed in the same region.

newtoboard
02-01-2014, 05:32 PM
The information bit I shared about Southeast India being an origin point was just an example on how STR calculations can be misleading. It was not to argue whether R2 entered South Asia before J2 or not. Simply put, determining a haplogroup's age through STR calculation is very misleading, and if the evidence for R2 being in South Asia before J2 is based on that, my comment is to take it with a grain of salt.

As for R2 being younger than R1, I have no evidence, it's an opinion I share with the way both haplogroups have spread. I think the actual mutation for R2 may have developed anywhere from the northern parts of South Asia, to the southern parts of Central Asia, and to the eastern parts of West Asia. While it was developing I would not be surprised if R1a1a-Z40 had already existed in the same region.

What is Z40. There isn't really anything to support R1a in this region prior to the Iron Age. If this was the case you would expect at least one other SNP parallel to Z93.

Tomasso29
02-01-2014, 06:52 PM
What is Z40. There isn't really anything to support R1a in this region prior to the Iron Age. If this was the case you would expect at least one other SNP parallel to Z93.

Z94 is under Z93:

http://www.isogg.org/tree/ISOGG_HapgrpR.html

It's not out of the question to assume that a potion of R1a1a which stayed in Asia may have later developed to Z93 while others that headed off to Europe later developed to Z283. If I recall, the distribution of Z93 is mostly common in Middle Easterners, South Asians, and Central Asians. When it is found in Europe, it's mostly found among the Jewish people.

It's also interesting because I find the distribution of R2a-M124 sort of resembles the spread of R1a1a-Z94 to a lesser scale.

newtoboard
02-01-2014, 07:24 PM
Z94 is under Z93:

http://www.isogg.org/tree/ISOGG_HapgrpR.html

It's not out of the question to assume that a potion of R1a1a which stayed in Asia may have later developed to Z93 while others that headed off to Europe later developed to Z283. If I recall, the distribution of Z93 is mostly common in Middle Easterners, South Asians, and Central Asians. When it is found in Europe, it's mostly found among the Jewish people.

It's also interesting because I find the distribution of R2a-M124 sort of resembles the spread of R1a1a-Z94 to a lesser scale.

I am well aware of Z93's structure. You typed Z40 in your original post so I thought you were referring to some sort of upstream SNP.

No it is not out of the question that happened but Z93 and Z283 share a common ancestor in Z645 whose brother clade is restricted to NW Europe. Not to mention the calculated age of Z645 is an almost perfect fit for Dnieper Donets. Unless you know of a migration from South Central Asia to Eastern Europe within Z645's time frame. Hopefully someone will test those Andronovo R1a samples (who ultimately have their origins in Mesolithic Europe via the Dnieper Donets->Samara->Khvalynsk->Yamnaya->Poltavka->Sintashta->Andronovo chain) so these theories of Neolithic R1a can go away. But you are basically arguing those samples will be Z93- right now which I find extremely unlikely.

How do you know Z93 wasn't found in Europe prior to the demise of Cimmerians and Scythians? How come Jewish R1a is extremely downstream of Z93 while Europe has a fair share of Z93(xZ94) and more than Asia? The best explanation is that Z93 originated in Europe and was a minority lineage among the earliest Yamnaya waves west.

Your theory also removes any sort of lineage that could be responsible for the Indo-Europeanization of South and Central Asia. And likely requires a Z93 population to get Indo-Europeanized by some sort of R1b or Z283 population while not absorbing those lineages. Given the military advantage Central Asian R1a would have had over those two lineages due to the invention and diffusion of chariots it could have pushed those lineages out of the steppe if it wanted to. I see no reason why they would adopt their language.

The real reason why R2a and R1a have a similar distribution is because R1a likely absorbed R2a males in Central Asia and they moved together.

But you are free to think what you want.

parasar
02-01-2014, 07:34 PM
A few thoughts.

-J2 doesn't seem all that high given we know some parts of South Asia have pretty high frequencies of J2b.
-Were these charts created before the discovery of H3 because most South Asian F ended up being H2/H3. But then again I would be surprised to see a frequency of H3 at such a high level. The F here likely represents a mixture of actual F, H3, G and J imo.
-R*? Probably R2 imo.
-More C1b1 than I would have thought.

Why? This is a predominantly Tamil data set after all.

newtoboard
02-01-2014, 07:40 PM
Why? This is a predominantly Tamil data set after all.

I didn't know that. Either way I thought C1b1 was strongest in the Terai and in Western India.

parasar
02-01-2014, 07:42 PM
... All the Indian samples from that Southeast India that tested for this SNP in the project came back positive.


R2-M124+ L295-
HG04158 Bengali
HG03727 Telugu

Tomasso29
02-01-2014, 08:09 PM
No it is not out of the question that happened but Z93 and Z283 share a common ancestor in Z645 whose brother clade is restricted to NW Europe. Not to mention the calculated age of Z645 is an almost perfect fit for Dnieper Donets. Unless you know of a migration from South Central Asia to Eastern Europe within Z645's time frame. Hopefully someone will test those Andronovo R1a samples (who ultimately have their origins in Mesolithic Europe via the Dnieper Donets->Samara->Khvalynsk->Yamnaya->Poltavka->Sintashta->Andronovo chain) so these theories of Neolithic R1a can go away. But you are basically arguing those samples will be Z93- right now which I find extremely unlikely.

If you're talking about L664, that may have been an even earlier migration of some of the M417's from Asia.


How do you know Z93 wasn't found in Europe prior to the demise of Cimmerians and Scythians? How come Jewish R1a is extremely downstream of Z93 while Europe has a fair share of Z93(xZ94) and more than Asia? The best explanation is that Z93 originated in Europe and was a minority lineage among the earliest Yamnaya waves west.

I don't understand what Scythians and Cimmerians have to do with this. As far as I know, the Scythians were an Iranian folk (Allegedly), and once again it's not out of the question that they migrated from Central Asia out west, not the other way around.

As for the Z93 found among a small number of Europeans, so what? A small number of R2a is found among Europeans too, try explaining that.


Your theory also removes any sort of lineage that could be responsible for the Indo-Europeanization of South and Central Asia. And likely requires a Z93 population to get Indo-Europeanized by some sort of R1b or Z283 population while not absorbing those lineages. Given the military advantage Central Asian R1a would have had over those two lineages due to the invention and diffusion of chariots it could have pushed those lineages out of the steppe if it wanted to. I see no reason why they would adopt their language.

Indo-Europeanization of South and Central Asia is a complicated matter and I think should be a separate subject. But to be clear, I think the Indo-Europeanization of South Asia was done by a group of multiple lineages, not one. R1a1a does not have to be there for the language to spread, look at the Iberian peninsula as an example.


The real reason why R2a and R1a have a similar distribution is because R1a likely absorbed R2a males in Central Asia and they moved together.

If that's the case, then you would see a similar distribution of other haplogroups too.


I am well aware of Z93's structure. You typed Z40 in your original post so I thought you were referring to some sort of upstream SNP.

That was a typo on my end.

parasar
02-01-2014, 08:14 PM
I didn't know that. Either way I thought C1b1 was strongest in the Terai and in Western India.

Regarding Western India you are likely referring to the Hapmap Gujaratis (Magoon et. al. http://biorxiv.org/highwire/filestream/220/field_highwire_adjunct_files/2/000802-3.pdf ) and for the Terai the Tharu from the "Morang district of Eastern Terai (Th-E)" (http://www.biomedcentral.com/1471-2148/9/154).

There were 12 M356+ among Hapmap Gujaratis, all from a high "ASI" south Asian type cluster (all are very close - likely related) - entitled Gujaratis-A by Zack.
None of the other Hapmap Gujaratis outside the cluster have M356.

As far the Tharus are concerned they are some the most "ASI" derived groups in India. All we can say about M356 is that it is one of the oldest in the subcontinent perhaps dating to its first occupation by modern humans.

Tomasso29
02-01-2014, 08:15 PM
R2-M124+ L295-
HG04158 Bengali
HG03727 Telugu

There is one member from the Bengal area that tested L295-, I'm not aware of the Telugu sample. For the record, I'm not saying there's no L295- in South India, the R2-Project has 3 samples from the Kerala area that are L295-, but when you look at the map from this thread (http://www.anthrogenica.com/showthread.php?1814-R2a-Map&p=25482#post25482), you can get an idea that it's highly unlikely that it originated in South India.

Imadaddin Nasimi
02-01-2014, 08:17 PM
Haplogroup L2 (L-M317 or L1b) wasn't found in India interesting.

newtoboard
02-01-2014, 08:43 PM
If you're talking about L664, that may have been an even earlier migration of some of the M417's from Asia.


L664 is younger than Z645, Z93 and Z283.



I don't understand what Scythians and Cimmerians have to do with this. As far as I know, the Scythians were an Iranian folk (Allegedly), and once again it's not out of the question that they migrated from Central Asia out west, not the other way around.

You alluded to an Asian distribution of Z93 when in ancient times it was likely distributed throughout the steppes. And Scythians ultimately derive from the European cultures between the Don and Ural regions so even if they migrated west it is still a back migration. There is no allegedly about it. It's pretty clear they were Iranian speaking unless everybody in the field of Indo-European studies is wrong and you possess some sort of Scythian writings that we don't know about.


As for the Z93 found among a small number of Europeans, so what? A small number of R2a is found among Europeans too, try explaining that.

Is European R2a consistently upstream like European Z93? Both lineages likely got there with Asian populations although there is a case to be made for Michal's idea that Z93 was distributed throughout Yamnaya. The difference is Z93 Scythian derived their ydnas from a European population. The story of the first Indo-Iranian expansion into Asia is the story of a nomadic tribe just crossing the Volga into sparsely populated land where they would have had an advantage due to a combination of a pastoralist economy, horse riding and chariots. There was no absorption of any other lineages until they started expanding north and south into South Siberia and South Central Asia respectively. R2a carriers likely were not European derived.



Indo-Europeanization of South and Central Asia is a complicated matter and I think should be a separate subject. But to be clear, I think the Indo-Europeanization of South Asia was done by a group of multiple lineages, not one. R1a1a does not have to be there for the language to spread, look at the Iberian peninsula as an example.

All other lineages besides R1a are fellow travelers and do not represent the original Proto Indo-Iranian speaking population. There is no getting around the the idea that Pre Indo-Iranian originated in Europe and Proto Indo-Iranian was spoken in Andronovo by men who were overwhelmingly R1a. So you have no point unless you are arguing Z93 represents a South Central Asian population adopting the language of and replacing the original speakers of Proto Indo -Iranian.

Iberia has R1b. Every IE language is connected to R1a or R1b.



that's the case, then you would see a similar distribution of other haplogroups too.

To some degree you do. R2a , L , G2a and J2a are all seen across Iran, Central Asia and South Asia. Frequency is a different matter, because we don't know how common each of these haplogroups was , how they were distributed in various regions of Central Asia and which portion of Central Asia's native population was assimilated into the various Indo -Iranian groups migration from Central Asia to more Southern latitudes.

newtoboard
02-01-2014, 08:43 PM
Haplogroup L2 (L-M317 or L1b) wasn't found in India interesting.

This lineage likely originated in West Asia.

parasar
02-01-2014, 08:48 PM
Haplogroup L2 (L-M317 or L1b) wasn't found in India interesting.

It has a widespread but trace presence from Europe to China to the Makran coast. Perhaps with larger samples it will be found in India.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380230/table/TB3/
http://www.yfull.com/tree/L/

newtoboard
02-01-2014, 08:50 PM
It has a widespread but trace presence from Europe to China to the Makran coast. Perhaps with larger samples it will be found in India.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380230/table/TB3/
http://www.yfull.com/tree/L/

I doubt it. I get the feeling it is found in Makran due to the Iranian heritage of the Balochi. I think that means the only place it might be found in South Asia is Sindh or Gujarat.

newtoboard
02-01-2014, 08:51 PM
It has a widespread but trace presence from Europe to China to the Makran coast. Perhaps with larger samples it will be found in India.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380230/table/TB3/
http://www.yfull.com/tree/L/

btw which chinese group has this lineage?

parasar
02-01-2014, 08:52 PM
There is one member from the Bengal area that tested L295-, I'm not aware of the Telugu sample. For the record, I'm not saying there's no L295- in South India, the R2-Project has 3 samples from the Kerala area that are L295-, but when you look at the map from this thread (http://www.anthrogenica.com/showthread.php?1814-R2a-Map&p=25482#post25482), you can get an idea that it's highly unlikely that it originated in South India.
I think R2-M124 originated where there is M479, and there is a lot of R2-M479* is southern India.

newtoboard
02-01-2014, 08:56 PM
Regarding Western India you are likely referring to the Hapmap Gujaratis (Magoon et. al. http://biorxiv.org/highwire/filestream/220/field_highwire_adjunct_files/2/000802-3.pdf ) and for the Terai the Tharu from the "Morang district of Eastern Terai (Th-E)" (http://www.biomedcentral.com/1471-2148/9/154).

There were 12 M356+ among Hapmap Gujaratis, all from a high "ASI" south Asian type cluster (all are very close - likely related) - entitled Gujaratis-A by Zack.
None of the other Hapmap Gujaratis outside the cluster have M356.

As far the Tharus are concerned they are some the most "ASI" derived groups in India. All we can say about M356 is that it is one of the oldest in the subcontinent perhaps dating to its first occupation by modern humans.

I heard Tharus originally spoke Sino Tibetan. But then again the Balti are South Asian in their ydna and it might be South Asian like Ydnas and population extended more north and east(into Tibet and Burma?) before the Sino Tibetan and Austrio-asiatic expansions. This might explain why amount Kailash was so important in South Asian mythology(and maybe related to the Kambojas).

newtoboard
02-01-2014, 09:00 PM
I think R2-M124 originated where there is M479, and there is a lot of R2-M479* is southern India.

So I guess we have three competing theorieson the origins of R2.

1. Mine that R2 originated in Central Asia and entered South Asia in the Paleolithic or Mesolithic

2. Tomasso's that it originated in South Central Asia and entered South Asia recently

3. Parsar's that it originated in South India.

Hopefully ancient DNA will sort this out eventually.

parasar
02-01-2014, 09:35 PM
I heard Tharus originally spoke Sino Tibetan. But then again the Balti are South Asian in their ydna and it might be South Asian like Ydnas and population extended more north and east(into Tibet and Burma?) before the Sino Tibetan and Austrio-asiatic expansions. This might explain why amount Kailash was so important in South Asian mythology(and maybe related to the Kambojas).

Just north of my hometown there are many Tharu villages in Champaran and Nepal. They now speak the local dialects but perhaps in the past spoke something different. They must have been living there a long time to have developed malaria resistance, but even now look pretty much as El Beruni described them a 1000 years back as living in our area - "Taru ...people of very black color and flat-nosed like the Turks.”

newtoboard
02-01-2014, 09:45 PM
Just north of my hometown there are many Tharu villages in Champaran and Nepal. They now speak the local dialects but perhaps in the past spoke something different. They must have been living there a long time to have developed malaria resistance, but even now look pretty much as El Beruni described them a 1000 years back as living in our area - "Taru ...people of very black color and flat-nosed like the Turks.”

Do you have a breakdown of their ydna distribution? Maybe they didn't spread Sino Tibetan but some sort of Paleo South Asian language.

parasar
02-01-2014, 09:47 PM
So I guess we have three competing theorieson the origins of R2.

1. Mine that R2 originated in Central Asia and entered South Asia in the Paleolithic or Mesolithic

2. Tomasso's that it originated in South Central Asia and entered South Asia recently

3. Parsar's that it originated in South India.

Hopefully ancient DNA will sort this out eventually.

No I don't think that it originated in South India, but that it potentially could have as the precursors (PxQxR, R-M207, and R-M479) are present there. Right now, I'm not sure how P got to Inner Asia (Mal'ta, Baikal). If it was through South Asia then yes a Southern Indian origin is possible. If it came from the Americas or eastern Siberia, then PxQxR, R-M207, and R-M479 could have entered India from the north.

parasar
02-01-2014, 10:25 PM
Do you have a breakdown of their ydna distribution? Maybe they didn't spread Sino Tibetan but some sort of Paleo South Asian language.
The H-Terai are essentially madhesiyas like us as can be seen from their 69.2% R1a1 and minimal O.
http://1.bp.blogspot.com/_Ish7688voT0/Sk9-TO7DceI/AAAAAAAABwY/9QUGqTOlQfA/s1600/tharus.jpg

parasar
02-02-2014, 12:32 AM
Regarding Western India you are likely referring to the Hapmap Gujaratis (Magoon et. al. http://biorxiv.org/highwire/filestream/220/field_highwire_adjunct_files/2/000802-3.pdf ) and for the Terai the Tharu from the "Morang district of Eastern Terai (Th-E)" (http://www.biomedcentral.com/1471-2148/9/154).
...

Correction on my mistake in the post: Not the Tharu from Eastern Terai but H-Te, please see table above in #34.

Tomasso29
02-02-2014, 02:12 AM
L664 is younger than Z645, Z93 and Z283.

If that's an STR based estimation, then it means very little to me.


You alluded to an Asian distribution of Z93 when in ancient times it was likely distributed throughout the steppes. And Scythians ultimately derive from the European cultures between the Don and Ural regions so even if they migrated west it is still a back migration. There is no allegedly about it. It's pretty clear they were Iranian speaking unless everybody in the field of Indo-European studies is wrong and you possess some sort of Scythian writings that we don't know about.

I say allegedly because there's no actual proof that they were Iranian speaking. Now I am in agreement with you that they were probably Iranian speaking and they may have left their mark on the modern Ossetians. But nothing suggests that they originated in Europe. In fact, the oldest historical record on them mentions that the Scythians were a people that left Asia due to wars with other tribes in the region. Also you have to consider that the later period Scythians were more of an umbrella of groups and not really an ethnic group.


Is European R2a consistently upstream like European Z93? Both lineages likely got there with Asian populations although there is a case to be made for Michal's idea that Z93 was distributed throughout Yamnaya. The difference is Z93 Scythian derived their ydnas from a European population. The story of the first Indo-Iranian expansion into Asia is the story of a nomadic tribe just crossing the Volga into sparsely populated land where they would have had an advantage due to a combination of a pastoralist economy, horse riding and chariots. There was no absorption of any other lineages until they started expanding north and south into South Siberia and South Central Asia respectively. R2a carriers likely were not European derived.

I said that because I'm trying to prove to you that Z93 is not common among actual European people in the same sense that M124 is not common. Meaning to say that it originated among them is highly unlikely. For the record, along with the Jewish mix, there's still a chance that some Scythians did indeed inject Z93 among the modern European populations, but these numbers are so small and comparable to Europeans carrying other haplogroups such as Q1a or C3. Sounds familiar? I think you have an idea where such lineages can come from.


All other lineages besides R1a are fellow travelers and do not represent the original Proto Indo-Iranian speaking population. There is no getting around the the idea that Pre Indo-Iranian originated in Europe and Proto Indo-Iranian was spoken in Andronovo by men who were overwhelmingly R1a. So you have no point unless you are arguing Z93 represents a South Central Asian population adopting the language of and replacing the original speakers of Proto Indo -Iranian.
Iberia has R1b. Every IE language is connected to R1a or R1b.

I said to leave the topic of Indo-European languages out because it will make things a bigger mess. Are you now saying that R1b were also Indo-Europeans? Since you believe that the Proto-IE came from the Steppes, were R1b Steppe people too? Because even your own sources will not agree with you on that. Please separate the two subjects because they're not the same. There may be some correlation but by the time these languages spread and carried full force, DNA was out of the question.


To some degree you do. R2a , L , G2a and J2a are all seen across Iran, Central Asia and South Asia. Frequency is a different matter, because we don't know how common each of these haplogroups was , how they were distributed in various regions of Central Asia and which portion of Central Asia's native population was assimilated into the various Indo -Iranian groups migration from Central Asia to more Southern latitudes.

You're right, there was plenty of migrations in our history that happened across the region and may have carried lineages east to west and back. But we're talking about a period beyond our historical scope.

Tomasso29
02-02-2014, 02:29 AM
I think R2-M124 originated where there is M479, and there is a lot of R2-M479* is southern India.

Let's be clear that just because we say L295-, that does not mean anything. It just means that there are people that are M124 with no further SNPs identified. The heavy number of L295+ in South India may indicate that L295 most likely originated somewhere very near. While M124 in general may have a little further north origin along with M479. It is true that M479 exists in South India, but it also exists in the northern parts and all the way up to Central Asia. More studies are obviously needed to identify whether we have other brother lineages to R2a under M479.

everest59
02-02-2014, 03:09 AM
The H-Terai are essentially madhesiyas like us as can be seen from their 69.2% R1a1 and minimal O.
http://1.bp.blogspot.com/_Ish7688voT0/Sk9-TO7DceI/AAAAAAAABwY/9QUGqTOlQfA/s1600/tharus.jpg

This is a good chart right here.
7.7% in Terai is a big number. So these are mostly Madhesi samples?
Whereas the Tharus have some as well, but the count is pretty low at 1.3%.
What's H-ND?

parasar
02-02-2014, 04:12 AM
This is a good chart right here.
7.7% in Terai is a big number. So these are mostly Madhesi samples?
Whereas the Tharus have some as well, but the count is pretty low at 1.3%.
What's H-ND?

H-ND Hindus from New Delhi.

Hindus from Terai - they will be akin to Madhesis even if not recent transplants. The samples are from Chitwan district upriver (http://en.wikipedia.org/wiki/Gandaki_River *) form my village. That area was at one time ruled by the Karnata-Mithila kings of Simraongarh.

*The terrain these antecedent rivers pass through from Tibet to Ganges are fascinating in their variety.

Gandak
http://upload.wikimedia.org/wikipedia/commons/8/8f/Ghaghra-River.png

http://upload.wikimedia.org/wikipedia/commons/6/61/Kali-gandaki.jpg
http://upload.wikimedia.org/wikipedia/commons/thumb/0/00/GandakiKagbeni.JPG/800px-GandakiKagbeni.JPG
http://upload.wikimedia.org/wikipedia/commons/thumb/5/5f/Kaligandaki_ghasa.jpg/800px-Kaligandaki_ghasa.jpg
http://upload.wikimedia.org/wikipedia/commons/thumb/7/71/GandakiPineForest.JPG/800px-GandakiPineForest.JPG
http://upload.wikimedia.org/wikipedia/commons/1/18/Rhinos_leofleck.jpg
http://thewanderers.in/images/Sonepur84.jpg

everest59
02-02-2014, 04:24 AM
Yeah, I have been to Chitwan as a kid. There is a very nice national park there. I have also been to some places in Terai like Janakpur.
These days Chitwan is very diverse. Mixture of Madhesis, Pahadiyas, Gurungs, of course Tharus, Newars,etc.

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

parasar
02-02-2014, 05:22 AM
Yeah, I have been to Chitwan as a kid. There is a very nice national park there. I have also been to some places in Terai like Janakpur.
These days Chitwan is very diverse. Mixture of Madhesis, Pahadiyas, Gurungs, of course Tharus, Newars,etc.

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

And the Valmiki National Park is essentially a continuation on the south side of the border in Champaran.

Many marriages in our family were with people from the Terai. We essentially had four family lines that inter-married:
Jaitheria http://en.wikipedia.org/wiki/Bettiah_Raj who supposedly had some connections to the Sahis of Kabul through "Gorakh Rai, a scion of the Brahmin Shahi"
Bhagocchia http://en.wikipedia.org/wiki/Hathwa_Raj who still continue carry the Sahi name eg Fateh Sahi
Donwar - they have Karnataka and Terai connections.
Eksariya who also have Nepal conections - some of these also have Sahi titles http://books.google.com/books?id=Ck4jmD7H34UC&pg=PA266

Dr_McNinja
02-02-2014, 06:35 AM
^ Those are interesting. The histories about Jatts say that Gills were descended from the kings of Mithila and that the Shahis were Gills too. On the other hand, there's still a gotra of Jatts named Shahi and Sahi. I wonder why there is this persistent connection between ancient Afghanistan and Nepal/Bihar/Mithila in the stories.

The haplogroups don't reveal much. I'm the only J-M241 Gill I've seen so far (and I keep getting placed near that area, DNA-Tribes even said Tharus were my highest match in their database), the others are J2a and mostly L1c (and one J1c3). J2a and L1c are present in both Nepal according to that other study and Afghanistan (so far most of the Popalzai Durranis I've seen are L1c-M357). Still, too few to reveal much. I'm waiting for some more South Asian J and L users to get FTDNA marker tests. There was that one guy from Mumbai with the surname Pradhan, but he only purchased Geno 2.0.

newtoboard
02-02-2014, 11:44 AM
If that's an STR based estimation, then it means very little to me.

It isn't.




I say allegedly because there's no actual proof that they were Iranian speaking. Now I am in agreement with you that they were probably Iranian speaking and they may have left their mark on the modern Ossetians. But nothing suggests that they originated in Europe. In fact, the oldest historical record on them mentions that the Scythians were a people that left Asia due to wars with other tribes in the region. Also you have to consider that the later period Scythians were more of an umbrella of groups and not really an ethnic group.

Scythian a likely just represent the steppe Andronovo population that did not migrate South. This population ultimately has its origins in the Poltavka culture of Europe just a while back and moved into likely unoccupied space on the Asian steppe. So it's not like crossing the Ural river suddenly made Scythians resemble Asians autosomally.




I said that because I'm trying to prove to you that Z93 is not common among actual European people in the same sense that M124 is not common. Meaning to say that it originated among them is highly unlikely. For the record, along with the Jewish mix, there's still a chance that some Scythians did indeed inject Z93 among the modern European populations, but these numbers are so small and comparable to Europeans carrying other haplogroups such as Q1a or C3. Sounds familiar? I think you have an idea where such lineages can come from.

If European Z93 derived from Scythians it would be Z94+ more often than it is. So it makes more sense to view as it as a minority lineage among the first steppe waves west. The age of Z93 is 5700 years which puts its origins likely in the eastern portion of the Yamnaya horizon.



I said to leave the topic of Indo-European languages out because it will make things a bigger mess. Are you now saying that R1b were also Indo-Europeans? Since you believe that the Proto-IE came from the Steppes, were R1b Steppe people too? Because even your own sources will not agree with you on that. Please separate the two subjects because they're not the same. There may be some correlation but by the time these languages spread and carried full force, DNA was out of the question.


Who are my sources? I don't see how IE languages can be explained without R1b also being on the steppe or somewhere nearby in the Carpathian basin/NE Balkans. It wouldn't surprise me if some R1b originated on the steppe or either migrated there with dairy farmers from NW Anatolia. Sorry if I have moved beyond the idea that R1b is some sort of Neolithic West Asian lineage. That may turn out to be the case but it's pretty clear R1b and R1a were located by the time of PIE were likely neighbors.

newtoboard
02-02-2014, 11:47 AM
^ Those are interesting. The histories about Jatts say that Gills were descended from the kings of Mithila and that the Shahis were Gills too. On the other hand, there's still a gotra of Jatts named Shahi and Sahi. I wonder why there is this persistent connection between ancient Afghanistan and Nepal/Bihar/Mithila in the stories.

The haplogroups don't reveal much. I'm the only J-M241 Gill I've seen so far (and I keep getting placed near that area, DNA-Tribes even said Tharus were my highest match in their database), the others are J2a and mostly L1c (and one J1c3). J2a and L1c are present in both Nepal according to that other study and Afghanistan (so far most of the Popalzai Durranis I've seen are L1c-M357). Still, too few to reveal much. I'm waiting for some more South Asian J and L users to get FTDNA marker tests. There was that one guy from Mumbai with the surname Pradhan, but he only purchased Geno 2.0.

Maybe the stories have to do with the Kambojas? Weren't they distributed from Tajikistan to Burma and Bhutan at one point?

Dr_McNinja
02-02-2014, 01:32 PM
Maybe the stories have to do with the Kambojas? Weren't they distributed from Tajikistan to Burma and Bhutan at one point?I think their history is definitely related in that it shows the spread/route of such peoples but I don't think it refers to them specifically because they are still a distinct tribe/gotra in Punjab at least. There are plenty of stories devoted specifically to them already. Whenever these stories spread, the Kamboj would have already been there and known to them.

I think, as that study implied, Nepal/Bihar and that general area was a main stop for incoming people.

The route could be from west of the Indus (Afghanistan) towards Nepal, with too many migrations/invasions making Punjab an unstable home until a steady Jat presence developed there a few thousand years ago. From there in North India near the Himalayas people could stop and back-migrate towards the Indus (or southwest towards Gujarat) or on towards the Bengal area and from there to the southeast. The distribution and Y-STR variation for R1a and J-M241 seem to imply this as well. The only other people to maintain a presence in the Punjab since before then were the Kamboj. Otherwise, who were the ancient "Punjabis"? Anyone from neighboring areas who happened to be in the area at that time, whether from the north, south, east, or west. That's why "Punjabi" didn't even become an identity until modern times.

The Shahis of Afghanistan (the non-Turkic ones) were either a Brahmin or Kshatriya dynasty of Afghanistan and the stories could indicate a link to North India (i.e, further east than Punjab). The various high caste clans within the region just probably made up stories reflecting this link. The Gill->Mithila link in the story probably reflects an ancient migratory pattern of back-migration. The story of Gills mentions Gilan in Iran (as well as Greece though I don't know if that was a new addition), north Pakistan (Gilgit/Baltistan), Mithila, then Punjab/Haryana/Rajasthan.

However, this might only explain some of the haplogroup presence/movement. Some of these haplogroups (including J-M241) are well represented among the Kalash and tribals. Whether this was a relic of this mythical migratory route or pre-existing is up for debate. At this point I think some of the haplogroups were introduced long ago and then autosomal admixture varied with repeated waves of migrants (with the newer haplogroups being the same as the existing ones or somehow not out-competing them).

J-M241 in Gill Jatts probably has one of two possible origins. Either it was a rare strain left in Punjab since the early days of J-M241 in the subcontinent or it came from the Nepal area. A third origin for J-M241 among Punjabis in general is from Gujarat (where it could have reached either directly early on, from the north, or have come from the south like in one big circle). This makes me wonder as to whether R1a and L1c exhibit signs of any kind of migratory pattern within the subcontinent. Some might have stayed in the northwest from early on, other strains might have made the full circuit of the subcontinent, who knows. Maybe my uncle's data (R1a) will shed some light since his seems unusual among Northwest Indians. Could either be more Central Asian or more South Indian. If the latter, it could show that some strains of R1a have made quite a journey.

If we got more people to test their Y-STR markers at FTDNA we could maybe learn more. There might be enough R1a data to hazard a few guesses though but I haven't really read up on that haplogroup.

Dr_McNinja
02-02-2014, 01:48 PM
Also, the Sandhu story is that they were around the Indus and what is now Sindh ("Sindhu"), then some of them retreated into Gujarat when the Arabs took Sindh and continued to fight them off. I read on one of those Jatt history groups that a few of the gotras like Sandhu claim an Eastern origin as well but I don't know if there's any truth behind that:


Sandhu jats are among the oldest and largest Jat
> clans.
> According to the British Gazzettes of Punjabs
> Districts written in 1870-80s that recorded first
> hand the ancient histories of the clans first hand
> from the word of mouth --and Rose book ---record
> that Sandhu Jats migrated from Ghazni in India.
> They( sandhu old men) emphasised to the British that
> the Ghazni they came from was not the one in
> Afganistan but the one in Dekhan in the East in
> India beyond Delhi.Actually this Ghazni of Sandhus
> appear close to what rose as the Kingdom of Gohad
> Jats which sounds just like Ghor in Afghanistan.
> Gohad jats conquered Gwalior which after Mughals in
> 1700s and stayed a Jat State other than Bharatpur.(
> The ruling Bharatpur JAts are SINSINWAR from the
> village of Sinsini-very big clan--sounds similiar to
> Sansi of Ranjit Singh does-nt it but rose to power
> almst 100 years before Sukherchakias)Sindhia
> Marathas could take Gwalior from Gohad only in 1890s
> only to loose it again. It was the British who
> finally betrayed Gohad Jats and gave Sindhia
> Traitors
> Gwalior in 1802.I do not know the Goths of Gohad
> Jats but some one must go there to find out.I read
> it all in "Jat Ithiyas by Thakur Desraj" written in
> 1927 or 1936.Very authoritative sources. Brij jats
> write Thakur because they became Khatriyas by
> Bhahubal meaning by Force of the arms defeating
> Rajputs and Mughals)
>
>
> Even Dhillons have emigrated from the region of
> Muzzfarnagar in the Yamuna region of UP in India(
> Rose book and Britiah Gazzettes) to the Central
> Districts of PUnjab. The districts of Muzzfarnagar,
> Baghpat,Meerut,Mathura and the lands between Yamuna
> and Ganga were the Hardcore belt of Jat settlements
> that no Turk,Afghan,or other invading horde could
> challenge.

everest59
02-02-2014, 03:57 PM
And the Valmiki National Park is essentially a continuation on the south side of the border in Champaran.

Many marriages in our family were with people from the Terai. We essentially had four family lines that inter-married:
Jaitheria http://en.wikipedia.org/wiki/Bettiah_Raj who supposedly had some connections to the Sahis of Kabul through "Gorakh Rai, a scion of the Brahmin Shahi"
Bhagocchia http://en.wikipedia.org/wiki/Hathwa_Raj who still continue carry the Sahi name eg Fateh Sahi
Donwar - they have Karnataka and Terai connections.
Eksariya who also have Nepal conections - some of these also have Sahi titles http://books.google.com/books?id=Ck4jmD7H34UC&pg=PA266

I see. Nepal has an open border with India, so I would imagine there would be a lot of movements in the region.

Do you know where in Chitwan the samples were collected? The Hindu terai samples were also from Chitwan, correct? I found a nice image that breaks down Chitwan demographically:
http://un.org.np/sites/default/files/Chitwan_Caste_Ethnicity_A3.pdf

It seems a lot of regions is dominated by hill brahmins. Do you know if they specifically collected Madhesi samples? The elevated C5 and some C3 seems to point to some Pahadi samples.

These also seem to be upper caste samples due to super high R1a.
My thinking here is, they collected mainly hill brahmin samples.

parasar
02-02-2014, 04:12 PM
Maybe the stories have to do with the Kambojas? Weren't they distributed from Tajikistan to Burma and Bhutan at one point?
Kambhoj is a generic term for mountain peoples - blanket bearing - kambal bhoj.
Kamboh is the term for people from coastal Gujarat who were displaced north by the Arab invasions.

Tomasso29
02-02-2014, 05:43 PM
It isn't.

Then what is it based on?


Scythian a likely just represent the steppe Andronovo population that did not migrate South. This population ultimately has its origins in the Poltavka culture of Europe just a while back and moved into likely unoccupied space on the Asian steppe. So it's not like crossing the Ural river suddenly made Scythians resemble Asians autosomally.

If European Z93 derived from Scythians it would be Z94+ more often than it is. So it makes more sense to view as it as a minority lineage among the first steppe waves west. The age of Z93 is 5700 years which puts its origins likely in the eastern portion of the Yamnaya horizon.

Mixing Andronovo with the Scythians is not proper imo. There may be some connections but we're talking about two different periods in history here. As for the European Z93, I'm sorry but did those Scythian skeletons even test positive for Z94? Let's hold judgment until they actually test them for those SNPs.


Who are my sources? I don't see how IE languages can be explained without R1b also being on the steppe or somewhere nearby in the Carpathian basin/NE Balkans. It wouldn't surprise me if some R1b originated on the steppe or either migrated there with dairy farmers from NW Anatolia. Sorry if I have moved beyond the idea that R1b is some sort of Neolithic West Asian lineage. That may turn out to be the case but it's pretty clear R1b and R1a were located by the time of PIE were likely neighbors.

When I said your sources I meant it in the sense that you believe in the Steppe theory. Those who believe in the Steppe/R1a theory also believe that R1b was not part of this IE expansion. Perhaps in Europe R1b did spread the IE languages to the western parts, but now you get why I don't mix the two subjects together? Because it comes to a point where Y-DNA should not even be taken into consideration. Also here's an interesting thought, why R1b and not J2 for example? If anything, J2 would probably be more qualified based in its spread to both, Europe and Asia. I'm not saying J2 is the lineage that spread IE languages, what I am saying is if one wants to speak of such expansions, don't mix it with DNA.

newtoboard
02-02-2014, 06:01 PM
Then what is it based on?

On SNP calculations.


Mixing Andronovo with the Scythians is not proper imo. There may be some connections but we're talking about two different periods in history here. As for the European Z93, I'm sorry but did those Scythian skeletons even test positive for Z94? Let's hold judgment until they actually test them for those SNPs.

Well Z94 likely originated right after Z93 ( as in about a 100 years or less after) so it is likely that most Asian Scythians were Z94+. I will hold off on judgement till those Scythian samples are tested.




When I said your sources I meant it in the sense that you believe in the Steppe theory. Those who believe in the Steppe/R1a theory also believe that R1b was not part of this IE expansion. Perhaps in Europe R1b did spread the IE languages to the western parts, but now you get why I don't mix the two subjects together? Because it comes to a point where Y-DNA should not even be taken into consideration. Also here's an interesting thought, why R1b and not J2 for example? If anything, J2 would probably be more qualified based in its spread to both, Europe and Asia. I'm not saying J2 is the lineage that spread IE languages, what I am saying is if one wants to speak of such expansions, don't mix it with DNA.

I didn't realize there was one steppe theory. Thanks for letting me know everybody who believes in a steppe origin for PIE all hold the exact same views the details. I'll be sure to let everyone know.

Why would J2 be more qualified? R1b is more numerous in Western Europeans and R1a is significantly more numerous in South/Central Asia and well represented in Iran as well. Where is the ancient J2? Nonein Andronovo, the Tarim mummies , Tagar? What about Corded Ware? What about Bell Beaker? Nope. Just a bunch of R1a and R1b. J2 is nonexistent in most parts of the IE speaking world. And the idea that J2 was even an important lineage as a fellow traveler with Indo-Iranian expansions is an internet myth. A minor one at best IMO. Pamiri Tajiks and Pashtuns have significantly more R1a than J2 and plenty of Indo-Aryan groups have ridiculously high amounts of R1a in comparison to low J2a. The idea came about because of the the high frequencies of J2a in Iranians, when likely most of it was Lready there in the Neolithic and before, and relative high frequency in some Indo-Aryan groups but that frequency is only high when compared to South Indians.

parasar
02-02-2014, 06:05 PM
...

The Shahis of Afghanistan (the non-Turkic ones) were either a Brahmin or Kshatriya dynasty of Afghanistan and the stories could indicate a link to North India (i.e, further east than Punjab) ...

Kshatriya lines were either dead or had become mercantile by the Sahi period. These were Brahman (later Rajputs) of the type seen all over India (from Sindh Chach to Kabul Sahi to Bengal Sen). The Rajput moniker gained currency later, especially after they lost power in Kabul, Ohind, and Lahore to the Ghaznavids and could not be called Raja anymore but only sons of Rajas (rajaputra).

In this Ghaznavid dismantling of the Sahi connection, I have a comment on what toast mentioned on the muhajir thread:

i think those sentiments were embedded there from 500 years before when pashtuns came in the armies of mahmud ghazni, sher suri, ghori, and durranis to wage holy jihad against india's hindus, killing, looting, and destroying, all the hindu inhabitants and idols. true the turkics were the most rabidly anti-hindu, but pashtuns did assist them in carrying out genocide, so they were anything but innocent bystanders


While it is true the Ghanavids had a religious zeal (destruction of the 'Mecca' idol at Somnath), they were equal opportunity looters as well as employers. Beruni mentions many 'Hindus' in their employ as well as soldiers in Merv and Samarkand from as far as from Karnataka - "the Karnˆat .a, used in Karnˆat .ade´sa, whence those troops come which in the [Ghaznavid] armies are known as Kannara"

Plus at that time many of those later entitled Pathan had not yet converted:
'Utbi, Abu Nasr Muhammad ibn 'Abd al-Jabbar
Translated from the Persian Version of the Contemporary Arabic Chronicle of AL UTBI

And the Sultán, informed of this, forthwith arose, with a body of his own special slaves, and came to their aid, and step by step drove those wretches far from the defile, and their places of asylum and (repositories) of wealth, until he had scattered them all from the protection of their narrow passes and the benefit of their difficult ground, and opened a way for his infantry, and made a road to arrive close to the stronghold-nest of the king and chief, whose name was Ibn-Súrí (son of Suri), and by a ravine (or village) named Ahingiran (?) came over to the sides of his fortress. He then issued out with ten thousand men, and drew up in line of battle against the Sultán, and by his opportunities of entrenching himself behind walls, and by reason of the aid he derived from his strong places of retreat and deep ditches, resisted half the day. Thus they continued striking on both sides in stubborn fight, and confused shootings and blows. But the Sultán ordered that they should turn their backs, as though his army were yielding and descending. These doomed ones were deluded with this deception, and the Hindú no longer held firm (to his ground) ...

Mustawfī, Ḥamd-Allāh ibn Abī Bakr Qazvīnī fl. ca. 740/1339-40

Sultán Mahmúd now went to fight with the Ghorians, who were infidels at that time. Súrí, their chief, was killed in this war, and his son was taken prisoner; but dreading the Sultán's vengeance, he killed himself by sucking poison which he had kept under the stone of his ring. The country of Ghor was annexed to that of the Sultán, and the population thereof converted to Islám. He now attacked the fort of Bhím , where was a temple of the Hindus.”


When the Pathans came to power they exacted a measure of revenge on Ghazni (as described by the later Turko-Mongol Babar):

Ghazni is but a poor, mean place, and I have always wondered how its princes, who possessed also Hindustan and Khorasan, could have chosen such a wretched country for the seat of their government, in preference to Khorasan ... [B]Alaeddin Jehansoz Ghuri, when he subdued this country, broke down the mound, burned and destroyed many of the tombs of the royal family of the Sultan, ruined and burned the city of Ghazni, and plundered and massacred the inhabitants. In short, there was no act of desolation and destruction from which he refrained. Ever since that time ... had remained in a state of ruin ...

So in a way both the Rajput (Sahi) and the Pathan (Sur/Ghurid) phenomenon had its genesis from the times of the Ghaznavids.

parasar
02-02-2014, 06:27 PM
Let's be clear that just because we say L295-, that does not mean anything. It just means that there are people that are M124 with no further SNPs identified. The heavy number of L295+ in South India may indicate that L295 most likely originated somewhere very near. While M124 in general may have a little further north origin along with M479. It is true that M479 exists in South India, but it also exists in the northern parts and all the way up to Central Asia. More studies are obviously needed to identify whether we have other brother lineages to R2a under M479.

Not to forget Portugal, Spain, Russia-Bashkir, Italy, and Caucasus-Osset.
Column E http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3039512/bin/ejhg2010146x5.xls

Tomasso29
02-02-2014, 06:28 PM
On SNP calculations.

What exactly is "SNP Calculation"?


Well Z94 likely originated right after Z93 ( as in about a 100 years or less after) so it is likely that most Asian Scythians were Z94+. I will hold off on judgement till those Scythian samples are tested.

It's not likely, it's for sure that Z94 came after Z93, but come again? Where on earth did you come up with that 100 timeline?


I didn't realize there was one steppe theory. Thanks for letting me know everybody who believes in a steppe origin for PIE all hold the exact same views the details. I'll be sure to let everyone know.

Why would J2 be more qualified? R1b is more numerous in Western Europeans and R1a is significantly more numerous in South/Central Asia and well represented in Iran as well. Where is the ancient J2? Nonein Andronovo, the Tarim mummies , Tagar? What about Corded Ware? What about Bell Beaker? Nope. Just a bunch of R1a and R1b. J2 is nonexistent in most parts of the IE speaking world. And the idea that J2 was even an important lineage as a fellow traveler with Indo-Iranian expansions is an internet myth. A minor one at best IMO. Pamiri Tajiks and Pashtuns have significantly more R1a than J2 and plenty of Indo-Aryan groups have ridiculously high amounts of R1a in comparison to low J2a. The idea came about because of the the high frequencies of J2a in Iranians, when likely most of it was Lready there in the Neolithic and before, and relative high frequency in some Indo-Aryan groups but that frequency is only high when compared to South Indians.

J2 is very common in the southern parts of Europe, parts of Central, South, and West Asia. Given that the IE languages are spoken all across these places, it would be absurd not to consider it as at least one of the lineages. Regardless, I had already mentioned that I'm not going to mix the two subjects together because ultimately, you will never find the right answer since the two subjects are not the same.

Humanist
02-02-2014, 06:37 PM
Not to forget Portugal, Spain, Russia-Bashkir, Italy, and Caucasus-Osset.
Column E http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3039512/bin/ejhg2010146x5.xls

OT. What study is that? Atypical Armenian population. Sample size is small, 26, but only one Armenian (3.8%) tested R-M269.

parasar
02-02-2014, 06:46 PM
OT. What study is that? Atypical Armenian population. Sample size is small, 26, but only one Armenian (3.8%) tested R-M269.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3039512/

Humanist
02-02-2014, 06:50 PM
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3039512/

Thanks. Interesting. A significant paper as far as R-M269 is concerned (Myres et al.). So, it has the Armenians at 3.8%, and has no data on Iranian minority populations, Assyrians, Alawites, and Druze. The populations with among the highest frequencies in the region.

newtoboard
02-02-2014, 06:56 PM
What exactly is "SNP Calculation"?

Calculations from the FGC analysis.




It's not likely, it's for sure that Z94 came after Z93, but come again? Where on earth did you come up with that 100 timeline?

I said its likely that Z94 originated right after Z93. I didn't argue that Z94 came first. I'd didn't come up with that but I think it was posted in calculations from the FGC analysis thread.




J2 is very common in the southern parts of Europe, parts of Central, South, and West Asia. Given that the IE languages are spoken all across these places, it would be absurd not to consider it as at least one of the lineages. Regardless, I had already mentioned that I'm not going to mix the two subjects together because ultimately, you will never find the right answer since the two subjects are not the same.

And your Indo Europeanization of Northern Europe happened from these mythical J2 IE speakers who despite managing to impose their language on Northern Europe didn't leave a trace in certain regions. News to me that Corded Ware and late Bell Beaker weren't IE speaking. Like I said where is the ancient J2 in IE associated cultures? Where is the J2 hiding? Pretty naive argument. Especially since parts of the regions mentioned above were Indo-Europeanized quite recently in comparison to Northwestern, Central and Eastern Europe. These subjects are ultimately mixed because the expansion of R1 clades and Indo-European are quite recent and related. Nor is the expansion of R1 very similar to expansion of other clades. Metallurgy, pastoral economies and the invention of the chariot allowed R1 clades to expand in a way like nothing else before.

Tomasso29
02-02-2014, 07:54 PM
Calculations from the FGC analysis.

I said its likely that Z94 originated right after Z93. I didn't argue that Z94 came first. I'd didn't come up with that but I think it was posted in calculations from the FGC analysis thread.

Walk me through this FGC analysis, how do they determine the age? For example when one calculates an age of a haplogroup through STR's, they use a mutation rate per marker based on mutation rate estimates through educated guesses. But as I mentioned that the mutation rate is still random which only makes it useful with high matches in recent times. Mind you SNP mutation is also random and rare might I add, explain to me how you can estimate the age? The other thing is, you're talking as if those Scythian skeletons are already Z94+.


And your Indo Europeanization of Northern Europe happened from these mythical J2 IE speakers who despite managing to impose their language on Northern Europe didn't leave a trace in certain regions. News to me that Corded Ware and late Bell Beaker weren't IE speaking. Like I said where is the ancient J2 in IE associated cultures? Where is the J2 hiding? Pretty naive argument. Especially since parts of the regions mentioned above were Indo-Europeanized quite recently in comparison to Northwestern, Central and Eastern Europe. These subjects are ultimately mixed because the expansion of R1 clades and Indo-European are quite recent and related. Nor is the expansion of R1 very similar to expansion of other clades. Metallurgy, pastoral economies and the invention of the chariot allowed R1 clades to expand in a way like nothing else before.

I never said J2 people spread the IE languages to Northern Europe. I said by the time the IE expansion was happening DNA was out of the question because there seems to be way too many lineages overlapping one another. My interpretation for the expansion is a domino effect over a few thousand years of migrations, invasions, and mixing. Again, this is why I don't mix the two subject.

newtoboard
02-02-2014, 08:18 PM
There is an entire thread on the FGC analysis.

Yes but the majority of IE expansions ultimately stem from Surovovo, Yamna/Repin and Corded ware. Which will likely show a mixture of R1a and R1b and if other lineages are present I, C and G are more likely than J2 IMO. J2 is likely a late Neolithic lineage across Southern Europe, West Asia, Central Asia and South Asia and has nothing to do with the origin of IE language or their expansions except maybe as a small fellow traveler lineage among a minority of IE groups.

parasar
02-02-2014, 08:21 PM
Walk me through this FGC analysis, how do they determine the age? For example when one calculates an age of a haplogroup through STR's, they use a mutation rate per marker based on mutation rate estimates through educated guesses. But as I mentioned that the mutation rate is still random which only makes it useful with high matches in recent times. Mind you SNP mutation is also random and rare might I add, explain to me how you can estimate the age? The other thing is, you're talking as if those Scythian skeletons are already Z94+.

...

While it is true that SNP mutation rate is not fixed, an SNP happens so rarely that there is almost no error from reverse mutations as there is from STRs. Plus we now have ancient dna SNP so we can have calibration points and do not have to depend just on pre-historical markers. Now with full genome data tests growing we can have rates from averages of many hundred+ year old confirmed pedigrees.

Francalacci et al. used Sardinia: "The tree was calibrated with archaeological data from the initial expansion of the Sardinian population ~7700 years ago ... rate is equivalent to 0.53 × 10−9 bp−1 year−1."

Poznik et al. used Americas: "Using entry to the Americas as a calibration point, we estimate a mutation rate of 0.82 × 10−9 per base pair (bp) per year [95% confidence interval (CI): 0.72 × 10−9 to 0.92 × 10−9/bp/year]"

Scozzari et al. used direct measurement using autosomes and have a time for P as 44000ybp:
http://4.bp.blogspot.com/-AFN4jjwuNBc/Us15U6bBVVI/AAAAAAAAJdc/uY8iTtxO0Mo/s1600/Scozzari.png

Based on the above, from a few R2 full genomes we could get a good ballpark age.

parasar
02-02-2014, 08:36 PM
^ Those are interesting. The histories about Jatts say that Gills were descended from the kings of Mithila and that the Shahis were Gills too. On the other hand, there's still a gotra of Jatts named Shahi and Sahi. I wonder why there is this persistent connection between ancient Afghanistan and Nepal/Bihar/Mithila in the stories.

The haplogroups don't reveal much. I'm the only J-M241 Gill I've seen so far (and I keep getting placed near that area, DNA-Tribes even said Tharus were my highest match in their database), the others are J2a and mostly L1c (and one J1c3). J2a and L1c are present in both Nepal according to that other study and Afghanistan (so far most of the Popalzai Durranis I've seen are L1c-M357). Still, too few to reveal much. I'm waiting for some more South Asian J and L users to get FTDNA marker tests. There was that one guy from Mumbai with the surname Pradhan, but he only purchased Geno 2.0.

J-M241 is clearly a very old lineage so its pan-Indian distribution makes sense. L1c on the other hand, while also old, looks have undergone a rapid expansion recently in the NW part of the subcontinent.

If the connection is to both Nepal and Mithila and recent enough to still remains in peoples memories, it may have to with the period just prior to the Muslim-Turkic invasions. The kingdom of Mithila and Nepal had managed to survive the Muslim-Turks even while the rest of Bihar was ravaged (The reason I say Muslim-Turk is that there had been supposedly prior pre-Islamic Turkic invasions from the north-east). A Jat friend of mine (of the Bhangoo clan) had mentioned that they had migrated from Nepal, but did not know when.

Tomasso29
02-03-2014, 12:26 AM
While it is true that SNP mutation rate is not fixed, an SNP happens so rarely that there is almost no error from reverse mutations as there is from STRs. Plus we now have ancient dna SNP so we can have calibration points and do not have to depend just on pre-historical markers. Now with full genome data tests growing we can have rates from averages of many hundred+ year old confirmed pedigrees.

Francalacci et al. used Sardinia: "The tree was calibrated with archaeological data from the initial expansion of the Sardinian population ~7700 years ago ... rate is equivalent to 0.53 × 10−9 bp−1 year−1."

Poznik et al. used Americas: "Using entry to the Americas as a calibration point, we estimate a mutation rate of 0.82 × 10−9 per base pair (bp) per year [95% confidence interval (CI): 0.72 × 10−9 to 0.92 × 10−9/bp/year]"

Based on the above, from a few R2 full genomes we could get a good ballpark age.

Calculating a SNP's age through ancient DNA can be useful but still not accurate because while you can certainly confirm that it existed at the time when that ancient sample lived, you still have to estimate the timeline prior to that. The fact that you have conflicting formulas to do such calculations says a lot.

parasar
02-03-2014, 02:21 PM
Calculating a SNP's age through ancient DNA can be useful but still not accurate because while you can certainly confirm that it existed at the time when that ancient sample lived, you still have to estimate the timeline prior to that. The fact that you have conflicting formulas to do such calculations says a lot.

One of the reasons rates conflict is that there is no consistent methodology applied. No one for example knows the exact time when OoA happened, or Sardinia was occupied, or humans reached the Americas. None of the studied have consistent base coverage. Once we have known pedigree data using the same testing metrics, scan quality, base coverage etc., we will have pretty accurate SNP rate numbers. I'm already quite confident of a subclade's relative age related to a superclade. For example if someone gives an accurate estimate for the root say 150000 years for 'Adam' with say 2000 mutations after that, a ratio of the mutations shared by the subclade can give us the relative age of a subclade.

This can be then verified with known pedigrees to confirm. For example many Icelanders know the exact person they descend from listed in the records, as do many Americans, and domesday descendants. Those mutations shared only within the pedigree group and no one else, will give us a lower absolute limit. This has still to happen, but for R1b subclades it will happen soon.

Tomasso29
02-03-2014, 04:08 PM
One of the reasons rates conflict is that there is no consistent methodology applied. No one for example knows the exact time when OoA happened, or Sardinia was occupied, or humans reached the Americas. None of the studied have consistent base coverage. Once we have known pedigree data using the same testing metrics, scan quality, base coverage etc., we will have pretty accurate SNP rate numbers. I'm already quite confident of a subclade's relative age related to a superclade. For example if someone gives an accurate estimate for the root say 150000 years for 'Adam' with say 2000 mutations after that, a ratio of the mutations shared by the subclade can give us the relative age of a subclade.

This can be then verified with known pedigrees to confirm. For example many Icelanders know the exact person they descend from listed in the records, as do many Americans, and domesday descendants. Those mutations shared only within the pedigree group and no one else, will give us a lower absolute limit. This has still to happen, but for R1b subclades it will happen soon.

Which is the exact reason why I don't take these calculations too seriously.

Sapporo
02-03-2014, 06:40 PM
Just shared with a Haryanvi Jatt (Beniwal) who is Y-DNA J2 and mt-DNA HV2 (same as me). :)

Mehrdad
02-03-2014, 07:01 PM
hi Sapporo, has there been a wide study done on Jatts Y and mtDNA's? It would be really interesting to compare Jatt's and their neighboring populations.

Dr_McNinja
02-03-2014, 10:07 PM
Just shared with a Haryanvi Jatt (Beniwal) who is Y-DNA J2 and mt-DNA HV2 (same as me). :)That's just plain J-M172 I take it? I'll add it to the spreadsheet.

Mehrdad, you might find it interesting: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdG8yX1hkLXNVbFVsR1JuOURBdnpad EE&usp=drive_web#gid=0

everest59
02-05-2014, 02:26 AM
That's just plain J-M172 I take it? I'll add it to the spreadsheet.

Mehrdad, you might find it interesting: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdG8yX1hkLXNVbFVsR1JuOURBdnpad EE&usp=drive_web#gid=0

I just noticed that one of my DNA relatives is Y-DNA J2B2* and mtdna R. He is from Gorkha, Nepal.

Dr_McNinja
02-05-2014, 02:42 AM
I just noticed that one of my DNA relatives is Y-DNA J2B2* and mtdna R. He is from Gorkha, Nepal.Do you know his surname/clan/caste/tribal name?

everest59
02-05-2014, 09:48 AM
Do you know his surname/clan/caste/tribal name?

Yeah, his surname is Pokharel. He is a Nepali Brahmin.

Dr_McNinja
02-05-2014, 01:40 PM
Just shared with a Haryanvi Jatt (Beniwal) who is Y-DNA J2 and mt-DNA HV2 (same as me). :)I ran his gedmatch kit through the gamut of the calculators and added him to the admixture spreadsheet here:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=0

Also another Punjabi Jatt (Dhesi) right above him, both have interesting results, especially in a couple of the Eurogenes calculators where the Haryana Jatt winds up with noticeably more South Asian/South Indian than the Punjabi Jatts.

He also wound up with consistently low mideastern/bedouin across most of the calculators. Which makes me wonder what the source of that little bit is in the rest of us.

- - - Updated - - -

^ Just got it, when the calculator has a separate Caucasian component, his West Asian winds up being less than mine and my mother's and his South Asian/South Indian winds up being more than most of the Punjabi Sikh Jatts. Where there is no Caucasian component, his West Asian winds up being higher than me and my mom and his South Asian/South Indian is in line with the other Punjabi Sikh Jatts. The exception is Harappa where his S-Indian is similar to the Punjabi Sikh Jatts and Baloch is a little low, near me and my mom.

Dr_McNinja
02-05-2014, 02:12 PM
^ Looking over it again, the Harappa calculator is definitely not accurate for the two reasons I've mentioned previously. The population Zack is using for Caucasus is biased towards West Caucasian (I think, don't recall which population it was exactly), which boosts my Caucasian numbers to near the other Jatts when, if he had used North Caucasian populations, it would otherwise be noticeably lower and everyone else's would go up relative to mine.

And it's missing an Atlantic component (so the Mediterranean acts as Atlantic-Mediterranean).

This leads me to believe the Baloch and S-Indian numbers would also change by a little if these two changes were implemented, perhaps only noticeable in results like the new Haryana Jatt's. Still not sure why his HarappaWorld Caucasian number is low when it's just fine in the other calculators with a more robust group of Caucasian populations. Could be that some of his Caucasian is being captured as Baloch.

I'm also beginning to think that the calculator effect is showing up heavily in results similar to mine and some of the Brahmins. I think we should have results similar to Punjabis overall but instead it pushes our European components westward and South Asian components south/east which cancels each other out and still averages out to the same area, near Punjab usually. It's like we represent a different group of North Indian/Punjabi samples which don't have any corollary in the reference populations. Everyone knows that's been a major flaw of these studies, using the results from studies/projects like 1000 Genomes as if they're comprehensive when there are probably many populations which aren't described well by their selection. What I'd suggest in addition to the above two changes would be having two South Asian components, or maybe even more, representing tribals, South Indians, and North Indians. Similar to how Dr. McDonald uses South Indian to represent tribals and North Indian to represent high castes (perhaps North Indian high caste, South Indian high caste, Southeast Indian Tribals, and North Indian Tribals (i.e, near Nepal)?).

If I could figure out how to use linux/admixture, I'd try testing it out myself but I have no idea how to get around in it. -_-

EDIT: One of the things Dr. McDonald noted with my mom's raw data was that the European was "very very weak and almost is S. Asian." (his exact words). This might explain why I have higher S-Asian than her and my dad's phased raw data (and lower European, but higher Western European proportionately). This could be addressed with better reference populations (more North Indian Brahmins and tribals).

parasar
08-22-2014, 07:18 PM
Any haplogroup data on those Rana Tharus who claim descent from Rajasthan or something?

Wish we could get a deep clade Y-DNA test on one of the Tharu J2b2-M241.


Just north of my hometown there are many Tharu villages in Champaran and Nepal. They now speak the local dialects but perhaps in the past spoke something different. They must have been living there a long time to have developed malaria resistance, but even now look pretty much as El Beruni described them a 1000 years back as living in our area - "Taru ...people of very black color and flat-nosed like the Turks.”

Regarding the Rana (they don't like to be called Tharu), in the British period they claimed a Kshatriya origin from Rajputana.
The royal house of Nepal makes that claim (descent from Udaiput Gehlot line) too (in the period a Rajputana claim was prestigious).

But as descent form an iconic ancient line is even more prestigious, the Udaipur family claims a descent from the region of Ayodhya (Rama)!

None of these claims has much merit.
http://www.macalester.edu/~guneratne/Teaching/tharus.html

The two main and culturally contrasted communities are the Dangauras and the Ranas. 'Dangaura' refers to the Tharus who claim Dang as their original home (which includes the Deokhuri Valley, Dang denoting a 'country' larger than the Inner Terai valley of the same name), and 'Rana' to the Tharus of the far western Terai who claim to have a Rajput origin. We do not know exactly when this last appellation and the royal pedigree attached to it became an ethnic label. In fact 'Rana Tharu' is an anthropological creation, since the Ranas do not want to be called Tharu, preferring Rana or Rana Thakur. I am of the opinion that the process of 'kshatrisation' attached to this ethnonym could be a relatively recent phenomena, linked to a general tendency of lower groups to raise their status particularly marked in colonial British India...

other Tharu subgroups live in Western Terai: the Katharyas, mostly concentrated in India, south of Dangaura habitat, and in Kailali district, have clothes and houses quite similar to their western neighbours the Ranas...

Some refer to themselves as Katharya (or Kacharya?), some are migrants of Eastern Terai or Koshila Tharus, intermixed with others minorities like the Danuwars. The Danuwars, whose main homeland is the Sindhuli Inner Terai valley around the Kamala river, are culturally and sociologically close to the Tharus, with whom they can intermarry. We do not know much about this part of Terai, but the Tharus living there are probably related to those living in the Indian Terai of Bihar, especially in Champaran...

the oldest reference we have concerning the Tharus refers precisely to the eastern Terai. In his book on India written in 1033, the Muslim scholar Alberuni mentioned the 'Taru, people of very black colour and flat nose like the Turks' who lived in Tilwat (Tirhut or Mithila). Later on, in the 13th century, a Persian historian cited the 'th'rw' near by the Meche and the Kooch but further east in Kamrup or north Bengal. We can therefore presume that people called Tharu have been living in the eastern Terai for at least a millennium...

the area of Bhojpuri influence extends much more westward for Tharu than for Indian Bhojpuri speaking groups, a fact which could confirm a western migration (especially from Bihar) of Tharus otherwise noted. But regarding linguistic transformations or a hypothetical substratum of Tharu languages, more research is necessary. More generally, we need a better knowledge of the enigmatic old processes of Austro-Asiatic and Tibeto-Burman reciprocal influence in north-east India to understand the linguistic genesis of the distant past in the far eastern Terai.

The Indo-Nepalese political boundary does have a meaning. First, the political destinies of India and Nepal have split, especially since the colonial period; and second, border regulations have strengthened the barrier. Today the Tharus, except partly the Ranas who have maintained matrimonial and economic relations in spite of the border, are different on either side of the frontier, influenced by a different socio-political milieu. The Nepalese Tharus are more numerous and have remained generally more isolated until the middle of our century; moreover, their caste status and their relations with other castes have evolved differently ; sanskritisation or 'kshatrisation' have played a stronger role on the Indian side; finally political and agrarian conditions, especially in the Inner Terai during the post-Unification period, have diverged over time.

Mehrdad
08-22-2014, 07:28 PM
parasar - what's your opinion on the migration of R1a1 in Nepal and the surrounding areas? Were they within the last 2000 years or perhaps earlier?

parasar
08-23-2014, 04:33 PM
parasar - what's your opinion on the migration of R1a1 in Nepal and the surrounding areas? Were they within the last 2000 years or perhaps earlier?

IMO, definitely earlier.
The earliest written records are from the Asokan period, and our caste, the bulk of whom are R1a1, is mentioned in them.

The very early settlements would perhaps go back to 3000bc along rivers banks (diara) and the Himalayan foot-hills (terai) where the forest cover was less dense.
By 800bc the area has to have been well established for major empires to arise there.

Megasthenes personal eyewitness account of the region looks correct and even if partly embellished it is difficult to discard.
Even in country like India where history has been problematic, folk remember events going back a few hundred years and Megasthenes does not report any such story of a recent migration prevalent in the region.
http://www.sdstate.edu/projectsouthasia/upload/Megasthene-Indika.pdf

Dr_McNinja
08-28-2014, 11:20 AM
BigY on sale for $495 (and Y Prime is $589) so hopefully more South Asians sign up for these next gen sequencing tests. FTDNA also has a sale on all Y-DNA products.

Dr_McNinja
08-28-2014, 04:46 PM
IMO, definitely earlier.
The earliest written records are from the Asokan period, and our caste, the bulk of whom are R1a1, is mentioned in them.

The very early settlements would perhaps go back to 3000bc along rivers banks (diara) and the Himalayan foot-hills (terai) where the forest cover was less dense.
By 800bc the area has to have been well established for major empires to arise there.

Megasthenes personal eyewitness account of the region looks correct and even if partly embellished it is difficult to discard.
Even in country like India where history has been problematic, folk remember events going back a few hundred years and Megasthenes does not report any such story of a recent migration prevalent in the region.
http://www.sdstate.edu/projectsouthasia/upload/Megasthene-Indika.pdfSince you're a project admin I don't suppose you could reach out to the J haplogroup admins and have them recommend their Saudi/Arab J-M241 users try FTDNA's L283 test and YSEQ.net tests for Z2432 and Z2433? I think they'd take you seriously.

Ebizur
08-29-2014, 10:19 PM
IMO, definitely earlier.
The earliest written records are from the Asokan period, and our caste, the bulk of whom are R1a1, is mentioned in them.

The very early settlements would perhaps go back to 3000bc along rivers banks (diara) and the Himalayan foot-hills (terai) where the forest cover was less dense.
By 800bc the area has to have been well established for major empires to arise there.

Megasthenes personal eyewitness account of the region looks correct and even if partly embellished it is difficult to discard.
Even in country like India where history has been problematic, folk remember events going back a few hundred years and Megasthenes does not report any such story of a recent migration prevalent in the region.
http://www.sdstate.edu/projectsouthasia/upload/Megasthene-Indika.pdfThe autosomal analysis of Chaubey et al. (2014) does suggest that their sample of Tharus of Uttarakhand (probably the so-called Rana Tharus) represents a much more homogeneous and inbred population (i.e. probably an older ethnic group) than the Tibetans of Spiti in Himachal Pradesh. Although the Tharus of Uttarakhand are genetically about halfway between mainstream South Asians (Aryans and Dravidians of India) and mainstream East Asians (Altaics, Sino-Tibetans, and Hmong-Miens), they also appear to be very homogeneous and closely related to one another (except one individual who appears to have recent admixture from a more typical Indian ethnic group). The people of Spiti, on the other hand, are strung out along a line between Pakistanis and northernmost Indians on one side and East Asian Altaics and Sino-Tibetans on the other. This pattern suggests either a recent ethnogenesis of the Spiti Tibetans or else continuous/ongoing gene flow from both northern South Asians and northern East Asians.

However, the authors have claimed on the basis of their fineSTRUCTURE analysis that the Spiti people are actually the nearest outgroup to the Uttarakhand Tharus, sharing much more recent ancestry than the Uttarakhand Tharus share with the Uttar Pradesh Tharus. Perhaps the Spiti people were originally like the Uttarakhand Tharus, midway between mainstream South Asians and mainstream East Asians, but Spiti has been affected to a greater degree by recent exogamy involving both South Asian and East Asian immigrants. I suppose the other explanation would be that Uttarakhand Tharus are an extremely inbred bunch of descendants of people who were close relatives of the diverse ancestors who have contributed to the formation of the modern population of Spiti. For whatever reason, the (apparently more exogamous or more recently formed) population of Spiti has ended up speaking a Tibetan dialect, whereas the (apparently more endogamous or more anciently formed) Uttarakhand Tharus have ended up speaking an Aryan dialect.

The Tharus of Uttar Pradesh seem to be genetically less of a distinct ethnic group than the Tharus of Uttarakhand. About one half of the Uttar Pradesh Tharus appear to be indistinguishable from the general Aryan and Dravidian population of India. The other half of the Uttar Pradesh Tharus are a bit eccentric, but still more similar to Aryans, Dravidians, Mundas, and Nihals than to the Uttarakhand Tharus, Spitians, or any other population.

Ebizur
09-01-2014, 05:59 AM
A few more tidbits from the coancestry matrix (in the form of a heatmap and dendrogram produced by fineSTRUCTURE & ChromoPainter) of Chaubey et al. (2014):

1) The Uttarakhand Tharus share the most among themselves.
2) After themselves, the Uttarakhand Tharus share the most with the least Aryan-admixed subset of the Spiti Tibetans and with the Nyishi and Ao tribals of Northeast India. The amount of sharing with the most Aryan-admixed and moderately Aryan-admixed subsets of the Spiti Tibetans is much lower, which suggests that the affinity between the Uttarakhand Tharus and the Spiti Tibetans should be attributed to their Tibeto-Burman element rather than to their Aryan element. However, that Tibeto-Burman element is not specifically Spitian nor Bodish, but rather something shared more generally among sub-Himalayan Tibeto-Burman peoples.
3) After the Nyishi, Ao, and less admixed Spiti Tibetans, the Uttarakhand Tharus share the most with Hakka (a subgroup of Han Chinese people), Palaung (an Austroasiatic-speaking people located in Shan State of eastern Myanmar as well as southwestern Yunnan and northern Thailand), a subset of the Tai Lue (a subgroup of Tai/Dai people), and a subset of the Hmong. The Palaung, like their linguistic relatives, the Wa, seem to contain a strong Tibeto-Burman and/or Sinitic element. The link between the Tai Lue and the Hmong might also be Tibeto-Burman or Chinese people.
4) Only at this point do the Uttarakhand Tharus share a more general affinity with the rest of the Hmong, Tai-Kadai peoples, Austroasiatic peoples (of both Southeast Asia and India, including the Mon, Blang, Santhal, Kharia, etc.), a subset of the Malay (perhaps Peninsulars), and Dravidian and Aryan tribes of India (Mala, Kurumba, Bhil, and some of the Uttar Pradesh Tharu).

In summary, it seems that the non-Aryan element in the Uttarakhand Tharus is what is responsible for their especially close relationship with the Spiti Tibetans, and also that that non-Aryan element is of a generalized Tibeto-Burman affinity, with slightly more distant connections to the Sinitics (in agreement with the Sino-Tibetan hypothesis), and, finally, to the "haplogroup O-M175 peoples" in general (but not so much Austronesians). This is quite consistent with the Y-DNA evidence, which shows a lot of Tibeto-Burman-like Y-DNA in the modern Tharus, especially the ones in Nepal and Uttarakhand. It also appears that the excess of sharing with Sino-Tibetan folks should not be ascribable to low genetic diversity of Sino-Tibetans relative to South Asians (Aryans, Dravidians, and Kolarians), so the most likely hypothesis for the origin of the Uttarakhand (Rana?) Tharus is that they consist of a South Asian substrate to which a Tibeto-Burman superstrate has (relatively recently) been added.

Mehrdad
09-03-2014, 07:04 PM
IMO, definitely earlier.
The earliest written records are from the Asokan period, and our caste, the bulk of whom are R1a1, is mentioned in them.

The very early settlements would perhaps go back to 3000bc along rivers banks (diara) and the Himalayan foot-hills (terai) where the forest cover was less dense.
By 800bc the area has to have been well established for major empires to arise there.

Megasthenes personal eyewitness account of the region looks correct and even if partly embellished it is difficult to discard.
Even in country like India where history has been problematic, folk remember events going back a few hundred years and Megasthenes does not report any such story of a recent migration prevalent in the region.
http://www.sdstate.edu/projectsouthasia/upload/Megasthene-Indika.pdf

I'm wondering whether R1a1 in the middle east may have come from South Asia during the Harappan period 3000-1500 BCE. This would throw a ranch into the whole R1a1 and Indo-European migration into the Indian sub-continent theory. Thoughts?!

Dr_McNinja
09-04-2014, 05:13 PM
Parasar how common is R1* among South Asians? I just shared with someone on my 23andMe Relative Finder who was R1. The mutations listed in their Haplotree tool are the basic R1 defining mutations. No R1a or R1b. Their paternal lineage is from Hyderabad, India. Interestingly their mother was from Pakistan/Afghanistan border area (Pathan) so they're 93.2% South Asian, 4.2% Middle Eastern, 1.1% European, 0.2% East Asian, 1.3% Unassigned (7.7% Unassigned on Standard). They're my highest match by far among the V4 individuals. The chromosome which the segment we share is on was number 20 and Ancestry Comp labels half that chromosome entirely Middle Eastern for them, so I'm not sure whether our match is South Asian or Mideast/Caucasus according to 23andMe (they match me on my mother's side). Their mtDNA hg is M4b.

MasterRoshi
09-07-2014, 03:45 AM
I say allegedly because there's no actual proof that they were Iranian speaking. Now I am in agreement with you that they were probably Iranian speaking and they may have left their mark on the modern Ossetians. But nothing suggests that they originated in Europe. In fact, the oldest historical record on them mentions that the Scythians were a people that left Asia due to wars with other tribes in the region. Also you have to consider that the later period Scythians were more of an umbrella of groups and not really an ethnic group.



Thats new to me. Ancient Khotanese Scythian scripts in Western Chima have been analysed and classified clearly as Iranic. The Scythians deserve the classification of an "Umbrella" group as much as any other ancient people who have build up Empires which span lands of other tribes. Sure Scythian conuquered non Iranic land but quite surely these were not considered "Scythians by any means. There was a Scythian ethnic group which Heredotus described as the Royal Scythians in Northern Caucasus and Eastern Ukraine. And the Sakae in Central Asia. Most of the other conquered tribes were Iranic speakers anyways like the Massagaeta, Dahae, Cimmerians and Heredotus and other ancient historians clearly distinguish between Scythian and the other tribes.

DMXX
09-07-2014, 07:34 AM
... Most of the other conquered tribes were Iranic speakers anyways like the Massagaeta, Dahae, Cimmerians ....

One little tidbit folks might find interesting. The etymology of the Massagetae has been broken down by at least one linguist to mean "Great Scythians" (Mass-saka-ta). There are many strange cognate claims online (such as a purported connection between the Goths, Massagetae, Jatts etc.) but this certainly seems to be the most intuitive.

MasterRoshi
09-07-2014, 11:30 AM
One little tidbit folks might find interesting. The etymology of the Massagetae has been broken down by at least one linguist to mean "Great Scythians" (Mass-saka-ta). There are many strange cognate claims online (such as a purported connection between the Goths, Massagetae, Jatts etc.) but this certainly seems to be the most intuitive.

Well there is a second explanation of the name. What I personally find quite interesting is that ancient historians like Heredotus break up the meaning of Massagetae to this Massa meaning "Great" and Getae, is the name of several Thracian tribes.

Why I find this explanation so fascinating is the recent results of ancient Thracian analyses, which turned out having Haplogroups typical for populations of Indo_Iranian speakers and populations of the Fertile Crescent.

Like mtDNA HV, U2e, U3b.
And Ydna H1b1* (Can't be contaminated with some worker of Roma ancestry because they are usually H1a*). I remember a neolithic sample from Syria turning out as some sort of H* too.

Also the autosomal data shows strong evidence of Indo_Iranian and Mesopotamian admixture. with Caucasus_Gedrosia reaching in some individuals levels of up to 50% and beeing the strongest! This exceeds levels found in modern Balkanians by far.


I find this quite fascinating because I always had this impression and always told that the closest tribe to ancient Indo_Iranians (Beside maybe Tocharians) would be the Thracians. There are also Scythian tribes found among Thracians like the Agathyrsi. And than this connection of Massagetae and the Thracians by Heredotus. Couldn't it be that the Massagetae crossed from East of the Caspian through Western Asia into the Balkans?

parasar
09-07-2014, 02:58 PM
Parasar how common is R1* among South Asians? I just shared with someone on my 23andMe Relative Finder who was R1. The mutations listed in their Haplotree tool are the basic R1 defining mutations. No R1a or R1b. Their paternal lineage is from Hyderabad, India. Interestingly their mother was from Pakistan/Afghanistan border area (Pathan) so they're 93.2% South Asian, 4.2% Middle Eastern, 1.1% European, 0.2% East Asian, 1.3% Unassigned (7.7% Unassigned on Standard). They're my highest match by far among the V4 individuals. The chromosome which the segment we share is on was number 20 and Ancestry Comp labels half that chromosome entirely Middle Eastern for them, so I'm not sure whether our match is South Asian or Mideast/Caucasus according to 23andMe (they match me on my mother's side). Their mtDNA hg is M4b.

R1* is not common anywhere. Essentially R1* would mean M173+ M420- M343-
I'm not sure if V4 includes M343 and M420
They should check their values at rs9786184 (M343), rs17250535 (M420) and rs2534636 (SRY1532.2) .

M420 R1a rs17250535 T->A
SRY1532.2 R1a1 rs2534636 C->T
M343 R1b rs9786184 C->A
http://www.isogg.org/tree/ISOGG_YDNA_SNP_Index.html

vettor
09-28-2014, 06:31 PM
With the new finding of X ( ydna ) in southern Indian, I presume then that since X is the parent of N and O haplogroups that the theory of a coastal route from west to east by the K group is the logical path to create P ( r and Q ) in south-east asia.

The only large amount of O subcaldes I can find in this area is in northern Burma.

Maybe the path of O and N was a path completely by passing south east asia and heading for central and eastern china

Dr_McNinja
09-28-2014, 10:44 PM
From my spreadsheet, all the Jatt Y-DNA haplogroups collected so far:

F - 3

R1a1a - 27 (of which 10 are L657+ and of those, 2 are Y7+)

R2 - 3

L1c-M357 - 14
L1 (L1a) - 1

J1c3 (J1e) - 2 (1 of which is J-Z1853)
J2 - 1
J2a1c (J2a1-M68) - 1
J2a1j (J2a2a/J2a3-P279) - 2
J2b2 - 3

G1 - 1
G2a5 - 1

H - 1
H1a - 3

Q1 - 1
Q1a - 1

This includes multiple members of families.

Both G individuals are Western Punjabi/Pahari Jatts (HRP0402 and HRP0283).

Haplogroup C seems to be conspicuously absent.

Dr_McNinja
09-28-2014, 10:49 PM
Out of the fewer Brahmins I've got:

J2b2 - 4
R2 - 1
R1a1a - 7 (1 of which is Parasar, Y9+, R-Y2392)
R1b1 - 1
C5a - 2
L1c-M357 - 1
Q1 - 1

Humanist
09-28-2014, 10:53 PM
From my spreadsheet, all the Jatt Y-DNA haplogroups collected so far:

G1 - 1
G2a5 - 1


This includes multiple members of families.

Both G individuals are Western Punjabi/Pahari Jatts (HRP0402 and HRP0283).

Always nice to see G1 represented. :)

Dr_McNinja
09-28-2014, 11:04 PM
Pashtun:

R1a1a - 10 (of which 5 are R-Z2124 and 1 is R-Z94)
R1b1a2 - 2

J1 - 1

L1c-M357 - 9

O3a3c1 - 1

G2c - 1
G2c1 - 1

bored
09-28-2014, 11:12 PM
Out of the fewer Brahmins I've got:

J2b2 - 4
R2 - 1
R1a1a - 7 (1 of which is Parasar, Y9+, R-Y2392)
R1b1 - 1
C5a - 2
L1c-M357 - 1
Q1 - 1

I think R2 is fairly common though, right? More or less common than J2B2? I have seen everyone from Pashtuns to Tamils with R2.

kenji.aryan
09-28-2014, 11:36 PM
I think R2 is fairly common though, right? More or less common than J2B2? I have seen everyone from Pashtuns to Tamils with R2.

http://s23.postimg.org/rwz0deliv/hh_001.jpg (http://postimg.org/image/rwz0deliv/)ttp://s23.postimg.org/rwz0deliv/hh_001.jpg][/url]



http://s15.postimg.org/j13kst8xz/df_001.jpg (http://postimg.org/image/j13kst8xz/)

bored
11-14-2014, 02:06 AM
It seems like my Y-DNA R2 is particularly common among Punjabi Brahmins..

http://www.nature.com/jhg/journal/v54/n1/fig_tab/jhg20082t1.html#figure-title

2948

BMG
04-17-2015, 04:42 AM
I came across a gedmatch kit with ydna E1b1b1a2 . His name is given as Reddy. So I guess him to be from AP .Has anybody came across any E1b among non-muslim south Asians

paulgill
04-17-2015, 06:39 AM
I came across a gedmatch kit with ydna E1b1b1a2 . His name is given as Reddy. So I guess him to be from AP .Has anybody came across any E1b among non-muslim south AsiansThe Greek E1b1b1a has been found among Afghans, Tajiks and Brahmins. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0050269