PDA

View Full Version : "I2a" and "I2b" are about to mean something different...



sparkey
02-05-11, 17:57
Ken Nordtvedt has the latest on Haplogroup I2 clade names (http://archiver.rootsweb.ancestry.com/th/read/Y-DNA-HAPLOGROUP-I/2011-04/1304173354).

Basically, a new SNP (L460) is going to force us to completely redo the I2 tree structure, and also force us to reconsider the relationship of I2* to I2a and I2b. Before, we had assumed that I2* was more closely related to I2b than to I2a due to STR similarities (not to mention geographic similarity), but it now appears that I2a and I2b are more closely related to one another than I2b is to I2*. Meanwhile, the tiny I2c-ADR clade, which is L460- alongside I2*, gets to be the new I2a or I2b.

Although ISOGG will probably do it different, here are some of Nordtvedt's new clade name suggestions:
I2a2a => I2a1b1
I2b1 => I2a2a
I2c-ADR => I2b

This will also have an effect on how we think of the way that I2 branched originally, as it now appears that rather than having a single branching event circa 20,000 years ago, there were two branching events with little time to separate them about 20,000 years ago... first current I2a/current I2b from I2*/current I2c-ADR, and then current I2a from current I2b.

Taranis
02-05-11, 20:57
Interesting. I wonder if it's a coincidence that these projected branching events seem to coincide roughly with the last glacial maximum.

sparkey
02-05-11, 21:10
Interesting. I wonder if it's a coincidence that these projected branching events seem to coincide roughly with the last glacial maximum.

I doubt it's entirely a coincidence, especially because Nordtvedt also suggests that the I1/I2 MRCA did not live too long before the first split within I2. This could mean that there were three relatively quick branching events one after the other circa 20,000 years ago. I can imagine that after the LGM, these branches were the ones that were successful and grew out of the previously very small European human population at the time. But we still need to account for the unusually recent MRCA's present among all of Haplogroup I's clades in that analysis.

Taranis
02-05-11, 21:16
I doubt it's entirely a coincidence, especially because Nordtvedt also suggests that the I1/I2 MRCA did not live too long before the first split within I2. This could mean that there were three relatively quick branching events one after the other circa 20,000 years ago. I can imagine that after the LGM, these branches were the ones that were successful and grew out of the previously very small European human population at the time. But we still need to account for the unusually recent MRCA's present among all of Haplogroup I's clades in that analysis.

How do you mean about unusually recent MRCAs? Do you mean that there's a considerable bottleneck inside Haplogroup I2, too (in addition to Haplogroup I1)?

sparkey
02-05-11, 21:50
How do you mean about unusually recent MRCAs? Do you mean that there's a considerable bottleneck inside Haplogroup I2, too (in addition to Haplogroup I1)?

Yes, the subclade MRCAs are all relatively young. All Haplogroup I people are descended directly on the paternal line from about 11 men who lived 8,500 years ago. They are:

1. Ancestor of I1
2. Ancestor of current I2c-ADR
3. Ancestor of current I2*
4. Ancestor of current I2b1 (British/Roots branch)
5. Ancestor of current I2b1 (Continental branch)
6. Ancestor of current I2b2
7. Ancestor of current I2a-Western/Alps
8. Ancestor of current I2a*-F
9. Ancestor of current I2a1
10. Ancestor of current I2a2-Isles
11. Ancestor of current I2a2-Din

Many of those MRCAs are significantly more recent than that, too... I picked "8500" so I could keep current I2a1 as a single entity, as it seems to be the most ancient, generally. So, I2's bottlenecking is clearly less severe than I1's, but it still seems to have bottlenecked in a way.

how yes no 2
02-05-11, 21:59
Yes, the subclade MRCAs are all relatively young. All Haplogroup I people are descended directly on the paternal line from about 11 men who lived 8,500 years ago. ...
Many of those MRCAs are significantly more recent than that, too... I picked "8500" so I could keep current I2a1 as a single entity, as it seems to be the most ancient, generally. So, I2's bottlenecking is clearly less severe than I1's, but it still seems to have bottlenecked in a way.

that is completely unexpected...
what could have happened 8500 years ago?

I could find only this... don't know if this could have triggered volcanic winter in northern hemisphere..


6440±25 BC: Kurile volcano on Russia's Kamchatka Peninsula has VEI 7 eruption. It is one of the largest of the Holocene epoch
http://en.wikipedia.org/wiki/7th_millennium_BC

Taranis
02-05-11, 22:02
Yes, the subclade MRCAs are all relatively young. All Haplogroup I people are descended directly on the paternal line from about 11 men who lived 8,500 years ago. They are:

1. Ancestor of I1
2. Ancestor of current I2c-ADR
3. Ancestor of current I2*
4. Ancestor of current I2b1 (British/Roots branch)
5. Ancestor of current I2b1 (Continental branch)
6. Ancestor of current I2b2
7. Ancestor of current I2a-Western/Alps
8. Ancestor of current I2a*-F
9. Ancestor of current I2a1
10. Ancestor of current I2a2-Isles
11. Ancestor of current I2a2-Din

Many of those MRCAs are significantly more recent than that, too... I picked "8500" so I could keep current I2a1 as a single entity, as it seems to be the most ancient, generally. So, I2's bottlenecking is clearly less severe than I1's, but it still seems to have bottlenecked in a way.

Wow... I wonder what the cause of this bottleneck effect is. Do you think it is likely that all eleven bottlenecks occured around the same time, or did they occur at different times?

sparkey
02-05-11, 22:21
I'm not the speculation expert here, but I can think of 3 possibilities:

(1) Haplogroup I had a decent tree structure at the beginning of the Neolithic (ca. 8000 years ago in Europe, what do you know) but got displaced over time by Neolithic and Bronze Age migrations. Modern high concentrations of certain clades, like I1 and I2a-Din, are due to recent expansions.
(2) Haplogroup I had bottlenecked already before the migration of other haplogroups into Europe due to some unknown reason. Expansions could have occurred later as well.
(3) Nordtvedt's clade MRCA estimates are flawed and they really date back farther.

It's also possible that different clades had different reasons for bottlenecks. For example, it could be (2) for I1 (1) for I2, I really don't know. The differences in MRCAs might suggest different bottleneck times. Current I2a1 is a lot older than current I2c-ADR, for example. But the different MRCA ages could also just reflect their relative concentrations at the beginning of the bottleneck, with older MRCA = more individuals carrying that clade at the beginning of the Neolithic.

sparkey
02-05-11, 22:28
Jean Manco speculates (http://www.buildinghistory.org/distantpast/haplogroupi.shtml):


The haplogroup may date deep into the distant European past, but it seems that most of the hunters and foragers who carried it have no direct descents in the male line today. That doesn't mean that none of their genes survive in Europeans. It just means that the direct son to son to son line died out or "daughtered out" at some point. So the subclades that remain lead back to just a handful of Mesolithic men.

Where were those Mesolithic handful? One clue noted in several scholarly papers is that Haplogroup I appears most diverse in south-eastern Europe. So my instinct was to place Haplogroup I among hunter-gatherers in the Carpathian Basin, some of whom later turned to farming. As farmers, they lived on the fringes of the European steppe, the presumed home of Proto-Indo-Euopean. So they would be unlikely to travel east with steppe nomads. But they could form mixed farming-herding groups on the rivers north and west from the steppe and the lush lands west of the Black Sea, and so be associated with Indo-European movements in those directions.


Rather to my surprise, this picture is still holding up quite well. That doesn't mean that alternative scenarios can be ruled out. We have too little ancient Y-DNA to make firm pronouncements. Yet most of the existing Haplogroup I subclades can be explained as the result of known or suspected migrations from South-Eastern Europe, mostly long after farming had taken over from fishing and hunting.

zanipolo
03-05-11, 10:04
does this site help

http://familypedia.wikia.com/wiki/Haplogroup_I_%28Y-DNA%29

zanipolo
07-05-11, 12:05
data in link for R1a , R1b, I and N in eastern europe and the balkans

note , the last graph, the slovenians seems to have eual representation of R1a, Rib and I

http://www.jogg.info/41/Wiik2.pdf

Maciamo
07-05-11, 12:31
does this site help

http://familypedia.wikia.com/wiki/Haplogroup_I_%28Y-DNA%29

Very outdated. This nomenclature (I1a, I1b1, I1b2) was abandoned over 2 years ago.

sparkey
11-07-11, 19:56
ISOGG has been updated. ISOGG has followed the Nordtvedt nomenclature as far as I can tell. So, it's official:

I2a1 => I2a1a
I2a2a => I2a1b1
I2a-Din => I2a1b1a
I2b1 => I2a2a
I2b2 => I2a2b
I2c-ADR => I2b

Mine (I2*) is still I2* for obvious reasons... although a couple of new SNPs (L596 and L597) found in someone on the same branch of I2* as me may give me a subclade yet.

Knovas
12-07-11, 12:21
Oh, I read about this yesterday too. So mine it seems to be now I2a1a, but I have an asterisc. I wonder if 23andme will change this, and what will happen with my asterisc now. I supose it will remain.

sparkey
12-07-11, 17:15
Oh, I read about this yesterday too. So mine it seems to be now I2a1a, but I have an asterisc. I wonder if 23andme will change this, and what will happen with my asterisc now. I supose it will remain.

If you're negative for the SNPs that define subclades of I2a1a, then you should retain an asterisk. You probably already know that you're L160-, so not I2a1a1. I know that they've found SNPs that define another subclade of I2a1a, namely L247 and L277, have you tested for either of those, or do you have an STR pattern that's similar to people who have those? See Bernie Cullen on that (http://webcache.googleusercontent.com/search?q=cache:TvrO0ewB5w0J:dna-forums.org/index.php%3F/topic/15203-2-new-snps-found-in-1821-branch-of-i-m26/+http://dna-forums.org/index.php%3F/topic/15203-2-new-snps-found-in-1821-branch-of-i-m26/&cd=1&hl=en&ct=clnk&gl=us&client=firefox-a&source=www.google.com). If you do have one of those SNPs, you'd become I2a1a2, otherwise you'll stay I2a1a* (which definitely still exists).

Knovas
12-07-11, 17:23
Thanks Sparkey. Time ago, I cheked some SNP's posted by Bernie, and I think I was TT for each, what meant ancestral if I remember well. So I think I'll be I2a1a*

I tried other's perhaps, but I don't remember to get any special genotype.

sparkey
04-08-11, 17:57
Another ISOGG update: L596 and L597 define a new I2 subclade, I2c. I2c probably totally replaces all of the remaining I2*, with the only cluster still needing testing being I2*-C, which itself was probably covered by an outlier designated "CAB" (as in, it probably belongs in C, but could be in A, or doubtfully B). So we can now say that the most common I subclade in Asia and Crete is "I2c." Nordtvedt definitely thinks that I2b and I2c are more closely related to each other than either is to I2a, so they may yet collapse into a new I2b, but no SNP confirms that yet.

zanipolo
21-08-11, 07:29
with the recent Haplogroup I changes, does I2a2a become a gothic marker since its in france, spain and italy ( meditterrean area only )

and the old "illyrian" marker was I2a-Din and now becomes I2a1b1a , is this correct.?

zanipolo
21-08-11, 07:36
with the recent Haplogroup I changes, does I2a1a become a gothic marker since its in france, spain and italy ( meditterrean area only )

and the old "illyrian" marker was I2a-Din and now becomes I2a1b1a , is this correct.?

sparkey
22-08-11, 02:14
with the recent Haplogroup I changes, does I2a1a become a gothic marker since its in france, spain and italy ( meditterrean area only )

I2a1a (I2a1 in the old nomenclature) is a Stone Age European relic that expanded all over Europe starting in the Neolithic, largely alongside Haplogroup G2. I doubt Goths had much I2a1a, they probably had more I2a2 (I2b in the old nomenclature) and I1 as far as Haplogroup I goes.


and the old "illyrian" marker was I2a-Din and now becomes I2a1b1a , is this correct.?

Yes, although we can still call it "I2a-Din" if we want, that's just a nickname. Its close cousin I2a-Disles is I2a1b1*, the difference being that I2a-Din is L147.5+.

zanipolo
22-08-11, 10:37
with comments by KN on the 1st of July 2011
[[ The two P37+ are consistent with I2a1 M26+. Note the ( - 12 ) for DYS385
> which I interpret to mean they only saw an electropherogram peak at 12
> repeats. I2a1 has a 12,12 modality at DYS385, and other STR repeats are
> consistent with the modalities of today's I2a1. Balkan I2a2a is modal 14-15
> at DYS385 and probably too young a clade to have populated France 5000 years
> ago. I2a1 M26+ on the other hand is one of the older (if not the oldest)
> clades in haplogroup I as measured by population variance.
>
> Today's I2a1 M26+ population is highly concentrated in far western Europe.
> It basically is the anti-R1a in European frequency --- rarely seen east of a
> line going north from Venice, Italy. Although very strong in Sardinia and
> Iberia, I2a1 M26+ continues at decent frequency up the Atlantic seaboard and
> into the British Isles, but hardly makes an appearance in Scandinavia. Very
> interesting to see it in SW France 5000 years ago.
>
> There are several snps now discovered downstream of M26. One of them, L160,
> nicely divides I2a1. It would be nice if these researchers have saved some
> dna and could run L160 on it?
>
> KN ]]

and also that the I2a1a in Basque lands is 8000 BC, sardinia is 5000BC and North Italy is 2000BC , does that indicate that the movement of these genes was from West to East , or am I reading something different

sparkey
22-08-11, 17:23
and also that the I2a1a in Basque lands is 8000 BC, sardinia is 5000BC and North Italy is 2000BC , does that indicate that the movement of these genes was from West to East , or am I reading something different

I think that sounds about right, at least that it was situated in the West (probably Iberia or a part of France) when it started expanding. That probably explains why it correlates poorly with G2a (which came from the East) in Iberia but well elsewhere.

Knovas
22-08-11, 17:34
Probalby I2a1a originated in the Pyrenees, doesn't matter if was in the Basque region or not. It's easy from there to spread and, except for Sardinia (in the middle) its location is predominately Western.

adamo
25-04-13, 08:15
I2a and I2b diverged ( split off) somewhere in the northern Balkans, I2b would begin heading towards Germany whereas I2a would head towards Ukraine/Moldova. In fact, I am currently under the impression that I2a first arrived near Moldova and then later spread to Bosnia and Herzegovina/ southern Balkans, but probably some piece of evidence will be suggested later that will, once more, alter my opinion. Unless I2b split from I2a near the Bosnian refugium and headed to Germany and later I2a spread from the southern Balkans towards Moldova lollll.....a complicated question this is

gyms
26-04-13, 14:00
Four new Dinaric SNPs are now available for order in FTDNA's Advanced menu.

Two of them, CTS11768 and CTS4002, apparently define the clade as CTS10936 also reportedly does. All three were derived in Dinaric Geno 2.0 results but ancestral in Disles, the nearest other clade.

Two of them, CTS10228 and CTS5966, apparently split the Dinaric clade. They were ancestral in one south-central Polish Dinaric, but derived in about seven other Dinarics across Central-Eastern Europe. The one Dinaric that tested negative for CTS10228 and CTS5966 has ordered Y-DNA67, but no marker results have arrived yet.(Igmayka)

sparkey
29-04-13, 16:59
I2a and I2b diverged ( split off) somewhere in the northern Balkans, I2b would begin heading towards Germany whereas I2a would head towards Ukraine/Moldova. In fact, I am currently under the impression that I2a first arrived near Moldova and then later spread to Bosnia and Herzegovina/ southern Balkans, but probably some piece of evidence will be suggested later that will, once more, alter my opinion. Unless I2b split from I2a near the Bosnian refugium and headed to Germany and later I2a spread from the southern Balkans towards Moldova lollll.....a complicated question this is

Your suggestions would make more sense if the Dinaric branch of I2a and I2a were equivalent, but they're not. I2a-Din is far downstream of I2a itself (assuming you define I2a as P37.2+ here). It's therefore not correct to try to extract where I2-P37.2 and I2-L35 split by taking the Dinaric branch of I2-P37.2 as representing I2-P37.2 as a whole.

Instead, you need to walk back along the I2-P37.2 tree until you get to the root. At that point, you will have extrapolated the approximate geographic branching location of the two major branches using a formula that looks like: (((Dinaric+Disles)+Isles)+NF+F)+(M26+(Western+Alpi ne)). To me, it looks like the greatest diversity of the (((Dinaric+Disles)+Isles)+NF+F) branch is France, considering that 2 of the 3 components are in France, and the third has a huge geographic range that includes France. The (M26+(Western+Alpine)) branch similarly has the greatest diversity in France or Germany with some certainty. So, a wiser guess would probably be France or somewhere nearby for the P37.2 branch. Doing a similar calculation for L35+ gets you to a similar area, maybe Germany or nearby. So, unless better data comes in to contradict this, we're probably looking a branching location in France, Germany, or nearby, rather than the Balkans.

zanipolo
01-07-13, 12:44
http://img841.imageshack.us/img841/8444/qihw.jpg (http://imageshack.us/photo/my-images/841/qihw.jpg/)

Uploaded with ImageShack.us (http://imageshack.us)

latest map of where I2a-din began

sparkey
02-07-13, 01:04
latest map of where I2a-din began

What is the original source? It's a bit difficult to read, but looks within reason if I'm reading it correctly.