How did I2a-Din get to the Balkans?

[razor]

Point taken about the unreliability of STR dating, though...
Nordtvedt's estimate does have fairly large error bars, which is why I resist saying things like "I2a-Din's MRCA lived in 300BCE." But suppose we pushed I2a-Din's TMRCA back all the way to its parent clade (I2a1b1). Then it's still less than 6,000 years old... and still has a diversity gradient coming down from the North. So Paleolithic continuity still fails.

What I like about Nordtvedt is his strict and cautious scientific approach to all this. He operates with SNP's even more than with STR's. He keeps retesting and refining. His latest surmise for TMRCA of I2a-Din is interesting in that the more he tests and refines the "younger" the subclade gets (not by much but some). I have no idea of how Dorianfinder gets his dates and the "I don't care for N's analysis" approach is of course not particularly persuasive. It would help to have something more precise. It's Nordtvedt who discovered the British Disles older brother to Din. As for Verenic's adaptations cf. the thread here, esp. from posts 113 ff. http://dna-forums.org/index.php?/top..._1#entry221867

Unfortunately we can't get aDNA from the Slavic homeland area because of the cremation ritual. But the possibility of checking exists for some of the Wielbark and Chernyakhiv gravesites, once the Ukrainans get their act together. We could see if there is any I2a1b1 there.

[sparkey]

As for Verenic's adaptations cf. the thread here, esp. from posts 113 ff. http://dna-forums.org/index.php?/top..._1#entry221867

Vadim Verenich's analysis is great. The diversity charts are very helpful... For those who aren't DNA Forums members, his maps indicate diversity spikes outside the Balkans, namely in Ukraine/Moldova (probably close to the expansion point of I2a-Din-N) and Austria, except one around Bosnia/Serbia, which looks to me like the expansion point of I2a-Din-S. There are also interesting "isolates" in Belarus, Greece, and a few other places, and a general molecular diversity pattern that spans Eastern Europe (but not most of the Balkans outside of that Bosnia/Serbia point). He compares these maps favorably to "Slavic expansion" maps.

[Dorianfinder]

OK. I think the value of the point dries up once it's clear that I2a-Din's ancestor clades were more likely not from the Balkans, than from the Balkans. Because then we can't look and say, "Hey, its ancestor was in the Balkans 11,000 years ago, and 7,500 years ago." If that changes to "Well actually it looks more like it was somewhere in Europe, maybe close-ish to the Balkans 20,000 years ago, and maybe even further away from the Balkans 13,000 years ago" then the usefulness of the whole thing goes away.

Point taken about the unreliability of STR dating, though... Nordtvedt's estimate does have fairly large error bars, which is why I resist saying things like "I2a-Din's MRCA lived in 300BCE." But suppose we pushed I2a-Din's TMRCA back all the way to its parent clade (I2a1b1). Then it's still less than 6,000 years old... and still has a diversity gradient coming down from the North. So Paleolithic continuity still fails.

Have you managed to figure out the route-relationship between the east European 'Dinaric' forms of I2a and the north-western forms such as L161 'Isles' and the tiny 'Disles' clade? What are the frequency distributions suggesting? If the Balto-Slavic hypothesis is studied in depth I think one will find Southern European countries share the up-stream clades and the north the down-stream ones.

[sparkey]

Have you managed to figure out the route-relationship between the east European 'Dinaric' forms of I2a and the north-western forms such as L161 'Isles' and the tiny 'Disles' clade? What are the frequency distributions suggesting? If the Balto-Slavic hypothesis is studied in depth I think one will find Southern European countries share the up-stream clades and the north the down-stream ones.

The different I2a1b clades have their modern centers of diversity too dispersed to figure out a route-relationship. The modern frequency distributions aren't going to help much, either. It's quite clear that both Disles and Isles have their centers of diversity in Britain, so it's appropriate to assume that both formed there. Does that mean that Dinaric came out of Britain, since 2 clades being in Britain and 1 being out means that the highest place of diversity of I2a1b as a whole is in Britain? Not necessarily, of course... but I suspect that it may be closer to the right place than the Balkans.

[Dorianfinder]

The different I2a1b clades have their modern centers of diversity too dispersed to figure out a route-relationship. The modern frequency distributions aren't going to help much, either. It's quite clear that both Disles and Isles have their centers of diversity in Britain, so it's appropriate to assume that both formed there. Does that mean that Dinaric came out of Britain, since 2 clades being in Britain and 1 being out means that the highest place of diversity of I2a1b as a whole is in Britain? Not necessarily, of course... but I suspect that it may be closer to the right place than the Balkans.

Disles and Isles may be more diverse in Britain but they are parallel to Dinaric. All Disles and Isles have the SNPs common to Dinaric however the inverse is not true.

Edited: correction, downstream should be parallel. Please read #83 for explanation.

[terranova]

Disles and Isles may be more diverse in Britain but they are downstream of Dinaric.

They are not downstream of Dinaric.

Has anyone thought about Cucuteni-Trypillian culture as the location of the I2a1b1 founder? Disles could have been carried to Britain by the advancing Indo-Europeans and Dinaric eventually became a part of the Proto-Slavic community.

[Shetop]

Disles and Isles may be more diverse in Britain but they are downstream of Dinaric. All Disles and Isles have the SNPs common to Dinaric however the inverse is not true.

Then we have basic misunderstanding what I2a-Din is, because I think your interpretation is not correct. Nordtvedt gave the name "Dinaric" so you should check his work related to this issue.

[Dorianfinder]

They are not downstream of Dinaric.

Has anyone thought about Cucuteni-Trypillian culture as the location of the I2a1b1 founder? Disles could have been carried to Britain by the advancing Indo-Europeans and Dinaric eventually became a part of the Proto-Slavic community.

Post #83

My mistake, they are parallel not downstream sorry I meant that Dinaric was formed well before Disles and Isles not that it was the parent clade. I think what may have given sparkey the idea that Disles and Isles may have formed during a similar time-frame or even preceded the formation of I2a1b1a Dinaric is that I2a1b1 Disles and I2a1b2 Isles are shorter designations (written in a shorter form).

I agree with Kenneth Nordtvedt on most everything but have reservations regarding his interclade estimation method for TMRCA approximations regarding I2a-Din. I believe the mutation rate and random aspect of crossover within isolated populations with a founder-type distribution pattern (definite hotspot) are prone to younger estimates. My estimates only differ wrt I2a-Din as this clade suggests founder effect and multiple effect within several Balkan populations.

The distribution of I2a-Din is indicative of an older subclade than Disles and Isles. I strongly suspect too that Disles and Isles have a few more SNPs in common with Dinaric in effect making them likely branches of a descendant of one of Dinaric's relative clades; not Dinaric itself. I would not be surprised if within Dinaric there are a number of subclades waiting to be discovered and that within one of those lies a parent clade of Disles and Isles.

[terranova]

I meant that Dinaric was formed well before Disles and Isles

Where do you get that? Dinaric is younger than Isles for sure.
TMRCA(Din-N) = 2340 years
TMRCA(Din-S) = 2040 years
TMRCA(interclade node) = 2820 years

[sparkey]

The distribution of I2a-Din is indicative of a significantly older (approx.1000yrs) subclade than Disles and Isles. I strongly suspect too that Disles and Isles have a few more SNPs in common with Dinaric in effect making them likely branches of a descendant of one of Dinaric's relative clades; not Dinaric itself. I would not be surprised if within Dinaric there are a number of subclades waiting to be discovered and that within one of those lies a parent clade of Disles and Isles.

Well so far we have I2a-Din and I2a-Disles as L621+ L161- but I2a-Isles as L161+ L621-. The distinction between I2a-Din and I2a-Disles is L147, with I2a-Din being L147+ and I2a-Disles being L147-. The only one of these that is particularly volatile AFAIK is L147. So I2a-Din has no chance of being a parent of I2a-Isles, and only a particularly unlikely chance of being a parent of I2a-Disles (it would have to have switched L147 back to - right after it had gained it, and the existing STR diversity analysis would have to be junk).

[sparkey]

Has anyone thought about Cucuteni-Trypillian culture as the location of the I2a1b1 founder? Disles could have been carried to Britain by the advancing Indo-Europeans and Dinaric eventually became a part of the Proto-Slavic community.

It's within range, and wouldn't surprise me. I still wonder about its parent I2a1b, though, and whether it's just a coincidence that both I2a1b2-Isles and I2a1b1*-Disles ended up confined to Britain. Maybe there are other outliers like Disles that we just haven't found yet that will clue us in.

[Dorianfinder]

@ terranova

In the estimates you supply Ken Nordtvedt clearly states that he used L147+ individuals and L147+ is a downstream SNP in the I2a-Din subclade used to distinguish the Disles from the continental Dinaric. All this TMCRA suggests is the age of L147, an SNP that formed after the split of Disles from its continental I2a parent clade. In fact, L147 may not be Dinaric it is most likely only a branch within Dinaric that formed relatively recently.

[Dorianfinder]

Well so far we have I2a-Din and I2a-Disles as L621+ L161- but I2a-Isles as L161+ L621-. The distinction between I2a-Din and I2a-Disles is L147, with I2a-Din being L147+ and I2a-Disles being L147-. The only one of these that is particularly volatile AFAIK is L147. So I2a-Din has no chance of being a parent of I2a-Isles, and only a particularly unlikely chance of being a parent of I2a-Disles (it would have to have switched L147 back to - right after it had gained it, and the existing STR diversity analysis would have to be junk).

L147 is currently being used to separate Disles from Dinaric as you say, I do not see why Dinaric is younger than Disles and Isles if it carries L147 as you seem to suggest. It's a red herring really, L147 is simply an SNP that may have formed after Disles left the continent.

[terranova]

In the estimates you supply Ken Nordtvedt clearly states that he used L147+ individuals and L147+ is a downstream SNP in the I2a-Din subclade used to distinguish the Disles from the continental Dinaric. All this TMCRA suggests is the age of L147, an SNP that formed after the split of Disles from its continental I2a parent clade. In fact, L147 may not be Dinaric it is most likely only a branch within Dinaric that formed relatively recently.

No, all Dinaric is L147+ AFAIK.

[zanipolo]

With historians and genetic historians claiming that Roman-illyrian troops served in britain during the Roman occupation, does it not make sense to see if I2a-din is present there.
If it is not , then we can assume that it came to the Balkans after the Roman empire died in the west. If it is present, then this i2a-din was present in the west balkans prior to Roman occupation of Illyricum

[razor]

No, all Dinaric is L147+ AFAIK.

I notice that Ken prefers to use DYS 565 as the main distinguisher between Disles and Dinaric. It's 9 in Dinaric and 11 in Disles. Though he occasionally mentions L147 also. For more particulars cf.

http://www.familytreedna.com/public/...ection=results

[Dorianfinder]

OK. I think the value of the point dries up once it's clear that I2a-Din's ancestor clades were more likely not from the Balkans, than from the Balkans.

@ Sparkey

Please explain what ancestor clades you are referring to here. Where do you believe these ancestral clades were originally from?

[sparkey]

@ Sparkey

Please explain what ancestor clades you are referring to here. Where do you believe these ancestral clades were originally from?

I2a1b1 => evidence is against being from the Balkans because no I2a1b1 subclade has its center of diversity there (currently 1 in the British Isles and 1 in Eastern Europe outside of the Balkans)
I2a1b => evidence is against being from the Balkans because no I2a1b subclade has its center of diversity there (currently 1 in the British Isles and 1 split between the British Isles and Eastern Europe outside of the Balkans)
I2a1 => evidence is against being from the Balkans because no I2a1 subclade has its center of diversity there (currently 1 in Western Europe and 1 split between the British Isles and Eastern Europe outside of the Balkans)

etc...

[terranova]

With historians and genetic historians claiming that Roman-illyrian troops served in britain during the Roman occupation, does it not make sense to see if I2a-din is present there.
If it is not , then we can assume that it came to the Balkans after the Roman empire died in the west. If it is present, then this i2a-din was present in the west balkans prior to Roman occupation of Illyricum

Good point. I2a-din is not present in Britain, nor anywhere else where the Romans went. I2a-disles is found mainly in Scotland which was not a part of Roman Britain. I think it's highly unlikely that any of the I2a1b clades were Illyrian.

EDIT Another interesting thing is that the Arbereshe from Italy were found to have almost no I2a-din compared to Albanians who have a much higher percentage, but this I2a-din appears to be a late introduction.

[Dorianfinder]

I2a1b1 => evidence is against being from the Balkans because no I2a1b1 subclade has its center of diversity there (currently 1 in the British Isles and 1 in Eastern Europe outside of the Balkans)
I2a1b => evidence is against being from the Balkans because no I2a1b subclade has its center of diversity there (currently 1 in the British Isles and 1 split between the British Isles and Eastern Europe outside of the Balkans)
I2a1 => evidence is against being from the Balkans because no I2a1 subclade has its center of diversity there (currently 1 in Western Europe and 1 split between the British Isles and Eastern Europe outside of the Balkans)

etc...

Thanks for replying, I cannot reply to you as I do not have the comparison of variance between the various European regions in front of me. But if you have seen the figures and the variance within the Balkans is comparatively lower then why not place a reference to the figures here so that people can make an informed opinion? After all you are the one who asked the question and now claim to have STR variance figures that suggest the I2a1b1 subclades are older outside the Balkans. If you know the answer why ask the question Sparkey?

[sparkey]

Thanks for replying, I cannot reply to you as I do not have the comparison of variance between the various European regions in front of me. But if you have seen the figures and the variance within the Balkans is comparatively lower then why not place a reference to the figures here so that people can make an informed opinion? After all you are the one who asked the question and now claim to have STR variance figures that suggest the I2a1b1 subclades are older outside the Balkans. If you know the answer why ask the question Sparkey?

Razor already linked to Vadim Verenich's analysis, that is probably the best out there. Sorry that it does require DNA Forums registration to view, I have summarized it a little here already, but it's best to view the original.

For interclade analysis, Nordtvedt is the best, as usual.

Or try your hand with the raw data... the largest collection I know of is at the I2a Project.

I don't claim to have the answer to my own question... Hopefully, I've been able to communicate what I know, what I don't know, why I made the choice in my own poll that I did, and what it would take to swing me to another choice. I want to hear what others have to say. To be honest, I was expecting more to defend the Sarmatians, and fewer to defend Paleolithic continuity.

Maybe it would be best to rank the choices from most to least likely. I would put them like this:

(1) The Slavs (best based on what we know about history and I2a-Din)
--gap--
(2) The Sarmatians (also fits history, I suppose, but poorer correlation to I2a-Din so far... we'll need more Eastern I2a-Din to lend credence to this theory)
--gap--
(3) Paleolithic continuity (I think that dating it back this far is impossible, so this needs a demonstration of a recent double-bottleneck in the Balkans to explain modern I2a-Din... discovering close, native Balkan relative clades of I2a-Din would help this one)
(4) Sea Peoples (would explain the distribution relatively cleanly, but needs some actual archaeological accounting and a badly wrong date estimate based on STRs, or a double-bottleneck)
--gap--
(5) The Early Indo-Europeans (can be safely discarded IMHO... wrong everything)

[Dorianfinder]

Razor already linked to Vadim Verenich's analysis, that is probably the best out there. Sorry that it does require DNA Forums registration to view, I have summarized it a little here already, but it's best to view the original.

For interclade analysis, Nordtvedt is the best, as usual.

Or try your hand with the raw data... the largest collection I know of is at the I2a Project.

I don't claim to have the answer to my own question... Hopefully, I've been able to communicate what I know, what I don't know, why I made the choice in my own poll that I did, and what it would take to swing me to another choice. I want to hear what others have to say. To be honest, I was expecting more to defend the Sarmatians, and fewer to defend Paleolithic continuity.

Maybe it would be best to rank the choices from most to least likely. I would put them like this:

(1) The Slavs (best based on what we know about history and I2a-Din)
--gap--
(2) The Sarmatians (also fits history, I suppose, but poorer correlation to I2a-Din so far... we'll need more Eastern I2a-Din to lend credence to this theory)
--gap--
(3) Paleolithic continuity (I think that dating it back this far is impossible, so this needs a demonstration of a recent double-bottleneck in the Balkans to explain modern I2a-Din... discovering close, native Balkan relative clades of I2a-Din would help this one)
(4) Sea Peoples (would explain the distribution relatively cleanly, but needs some actual archaeological accounting and a badly wrong date estimate based on STRs, or a double-bottleneck)
--gap--
(5) The Early Indo-Europeans (can be safely discarded IMHO... wrong everything)

I had a look at the data and find the difference in variance a moot point considering its spread and the poor interclade definition as I have mentioned previously. You are not wrong for looking at the variance but I feel you have been swayed somewhat by the pan-slavists who frequent forum discussions. I would be very cautious linking the Sarmatians and/or Slavs with Dinaric for a couple of reasons.

1. The distribution within the Balkans is reminiscent of a much older migration, that is if one did take place, than the Slavic expansion of the 6th-7th cent. AD.
2. Montenegrins, Croatians, Serbians and Bosniaks are three distinct peoples, despite what some panslavists want us to believe. I believe the Serb and Bosniak communities are early Balkan.
3. Greek and Armenian I2a puts the Slav and Sarmatian theories in the realm of fantasy.
4. No links have been found linking Disles and Isles to Sarmatia or the Baltoslavic regions.

I agree wholeheartedly regarding the Sea Peoples and believe we simply do not know enough about them, however it remains a possibility and is not mutually exclusive to a Balkan Paleolithic continuity thesis, in fact they fit rather well.

[Haustor]

I voted for Other.
I think that I2a-Dinarics got to Balkan as Ostrogoths. This would also explain presence of I2* in Georgia, Armenia and Turkey as those would be people who did not complete their voyage to Balkan and NW Italy. TMRCA for Dinarics would be older as it probably belonged to clan chiefs while TMRCA for Ia*s would be younger branches that remained behind.
Second option would be Gepids who lived little bit north of Balkan and have joined Huns later on. This aliance with Huns allowed them to spread into areas where I2a-Ds are found today.
This option may not be liked by many and is based on observation that there is not much (if any) genetic footprint left by Huns or Magyars in Europe which lead me to believe that people in old times might have commited infanticide against Asian babies while Gepid's babies lived on.

I do not think Dinarics lived in Balkan during Roman Empire as their dna were not found in graves of Roman soldiers from Illyria (graves from Wales seem to indicate hg E as Illyrian).
They are too young for Paleolithic continuity and even if they were not again they were not part of Roman armies so they would have not lived in Balkan.

I think that early Indo-Europeans were more of R1a and R1b folks.
I do not think we were Sea people as we don't have much of seafaring experience. During medivial times Bosnian Kingdom waged wars against both Raguscan and Venetian and took most of land in east Adriatic but those Bosnians had no navy at all.

There is very little R1a in parts of Balkan for it to be mainly Slavic migration.
Don't know much about Sarmatians to make any comments.

[Shetop]

1. The distribution within the Balkans is reminiscent of a much older migration, that is if one did take place, than the Slavic expansion of the 6th-7th cent. AD.

Why is that? What about distribution within the Balkans is reminiscent of a much older migration?

[Dorianfinder]

Why is that? What about distribution within the Balkans is reminiscent of a much older migration?

The frequency in isolated populations of primarily non-Slavic communities strongly suggests that it would have had to have been introduced in pre-Slavic times. Armenian levels of I2a and Greek I2a1* suggests that I2a may have been part of an early Illyro/Phrygian exchange, Phrygian being the Armenian I2a so prominent today.

The King paper showed significantly higher levels of I2a in the Peloponnese compared to North Greece. The King paper found I2a2a (Former I2b1 in the Y2010 tree) at relatively low frequencies throughout Greece from the North to the isolated Lasithi Plateau on Crete. The Battaglia study found a sturdy 17.5% of Greek Macedonian men to be I2a1*. Despite the complex variation of Turkey's genetic make-up and the relevant insignificance of Armenian I2a within the broader Anatolian context, it still reaches 4% in the Turkish population.