World Families Forums - Another Possible Iberian R-L21 Haplotype Cluster?

Welcome, Guest. Please login or register.
July 31, 2014, 10:25:39 PM
Home Help Search Login Register

+  World Families Forums
|-+  General Forums - Note: You must Be Logged In to post. Anyone can browse.
| |-+  R1b General (Moderator: rms2)
| | |-+  Another Possible Iberian R-L21 Haplotype Cluster?
« previous next »
Pages: [1] 2 3 4 Go Down Print
Author Topic: Another Possible Iberian R-L21 Haplotype Cluster?  (Read 7854 times)
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« on: June 19, 2010, 06:53:30 PM »

Okay, maybe I am going cluster crazy, and perhaps four off-modal markers are too few, but check this out:

http://tinyurl.com/2cjs3e6

Enter the Captcha codes at the bottom and click on "Show comparative Y-DNA results".

Costa and Vargas are both confirmed L21+.

I know I have a couple of 25-marker haplotypes in there, so we can't be sure about them.

If this is a cluster, the off-modal values are:

19=15

459b=9

YCAIIb=19

456=15


Too bad they don't all have 67 markers.
« Last Edit: June 19, 2010, 11:05:31 PM by rms2 » Logged

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #1 on: June 19, 2010, 08:14:46 PM »

Okay, maybe I am going cluster crazy, and perhaps four off-modal markers are too few, but check this out:

http://tinyurl.com/2cjs3e6

Enter the Captcha codes at the bottom and click on "Comparative Y-DNA Results".

Costa and Vargas are noth confirmed L21+.

I know I have a couple of 25-marker haplotypes in there, so we can't be sure about them.

If this is a cluster, the off-modal values are:

19=15

459b=9

YCAIIb=19

456=15


Too bad they don't all have 67 markers
.

I updated that link above to include one more likely haplotype (one that has been tested M269+ but negative for all the major subclades EXCEPT P312 and L21).
Logged

Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #2 on: June 19, 2010, 08:44:20 PM »

Okay, maybe I am going cluster crazy, and perhaps four off-modal markers are too few, but check this out:

http://tinyurl.com/2cjs3e6

Enter the Captcha codes at the bottom and click on "Comparative Y-DNA Results".

Costa and Vargas are noth confirmed L21+.

I know I have a couple of 25-marker haplotypes in there, so we can't be sure about them.

If this is a cluster, the off-modal values are:

19=15

459b=9

YCAIIb=19

456=15


Too bad they don't all have 67 markers.
I think you have found another one.  456=15 is not a big deal as there are a ton of those but not too many YCAII=19,19 folks and not too many 19=15 folks and when you add 459b=9 you have a fairly limited group.

I have a speculative variety in the spreadsheet that I labeled R-L21-9919 because of the signature YCAII=19,19 459=9,9 which overlays what you are describing.  

There is a non-Spaniard variety of YCAII=19,19 459=9,9 folks that are modal at 459=9,10, but they also have 640=12.  A lot of English in this group. Who knows, maybe these are sub-branches of the same variety?
« Last Edit: June 19, 2010, 08:48:47 PM by Mikewww » Logged

R1b-L21>L513(DF1)>L705.2
Mike Walsh
Guru
*****
Offline Offline

Posts: 2963


WWW
« Reply #3 on: June 19, 2010, 08:50:10 PM »

Okay, maybe I am going cluster crazy...
You're not crazy... well, at least you are not any crazier than I am.
Logged

R1b-L21>L513(DF1)>L705.2
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #4 on: June 19, 2010, 09:00:56 PM »

Okay, maybe I am going cluster crazy...
You're not crazy... well, at least you are not any crazier than I am.

Some folks would say we are both nuts! ;-)

Anyway, thanks for your positive comments on this possible cluster.

If Lopez de Prado Lopez is in it, then I think it must be fairly old, since his genetic distance from some of the other guys is substantial. He is all tested up at what used to be the old R1b1b2* level: negative for everything beyond M269 except P312 and L21. That makes him an excellent candidate for L21 testing, since we know he isn't M153, SRY2627, U106 or U152. He is also a Spanish citizen, which is an added plus.

These sorts of combinations seem to be standing out among the Spanish and Portuguese, but I couldn't find anything similar among French R-L21 guys, who all seem to be pretty distant from each other (and from everyone else, too).
Logged

alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #5 on: June 20, 2010, 10:48:25 AM »

Okay, maybe I am going cluster crazy...
You're not crazy... well, at least you are not any crazier than I am.

Some folks would say we are both nuts! ;-)

Anyway, thanks for your positive comments on this possible cluster.

If Lopez de Prado Lopez is in it, then I think it must be fairly old, since his genetic distance from some of the other guys is substantial. He is all tested up at what used to be the old R1b1b2* level: negative for everything beyond M269 except P312 and L21. That makes him an excellent candidate for L21 testing, since we know he isn't M153, SRY2627, U106 or U152. He is also a Spanish citizen, which is an added plus.

These sorts of combinations seem to be standing out among the Spanish and Portuguese, but I couldn't find anything similar among French R-L21 guys, who all seem to be pretty distant from each other (and from everyone else, too).

I wonder what that is telling us about French L21?   You would think if they are so varied that they do no cluster or match (despite so much L21 in France) that it would have a really high diversity or variance compared with other areas where a lot of it may fit into a few more modern clusters.   
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #6 on: June 20, 2010, 02:26:33 PM »

I wonder what that is telling us about French L21?   You would think if they are so varied that they do no cluster or match (despite so much L21 in France) that it would have a really high diversity or variance compared with other areas where a lot of it may fit into a few more modern clusters.   

Well, the French do have higher L21 haplotype variance than everyone else. Maybe it would be startlingly higher if we had a lot more French L21 haplotypes, especially 67-marker haplotypes, to work with.

Then again, if we had more French L21 haplotypes, we might actually be able to find a cluster or two.
Logged

NealtheRed
Old Hand
****
Offline Offline

Posts: 930


« Reply #7 on: June 20, 2010, 04:46:18 PM »

I wonder what that is telling us about French L21?   You would think if they are so varied that they do no cluster or match (despite so much L21 in France) that it would have a really high diversity or variance compared with other areas where a lot of it may fit into a few more modern clusters.   

Well, the French do have higher L21 haplotype variance than everyone else. Maybe it would be startlingly higher if we had a lot more French L21 haplotypes, especially 67-marker haplotypes, to work with.

Then again, if we had more French L21 haplotypes, we might actually be able to find a cluster or two.

So there is more haplotype variance between French samples than Spanish ones, correct? It is interesting we are finding more Spanish "clusters" within L21.

It is always the outlier that gets noticed.
Logged

Y-DNA: R-Z255 (L159.2+) - Downing (Irish Sea)


MTDNA: HV4a1 - Centrella (Avellino, Italy)


Ysearch: 4PSCK



rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #8 on: June 20, 2010, 05:12:22 PM »



So there is more haplotype variance between French samples than Spanish ones, correct? It is interesting we are finding more Spanish "clusters" within L21.

It is always the outlier that gets noticed.

I'm not sure I would characterize Iberian L21 or these clusters within Iberian L21 as outliers. L21 is pretty western, and Spain is right next door to France, where thus far the oldest L21 haplotypes are found.

I think maybe we overlooked Iberia for a long time, possibly because so many of our Spanish and Portuguese surname results came in as "Colonial". But the thing is, we were always getting them, little by little. I don't think there ever was a time, except maybe right at first, when we had zero Iberian L21.

Honestly, it wouldn't surprise me if L21 turned out to be a lot bigger in Iberia than we thought. The problem is going to be ferreting it out from among all that R-P312* and R-SRY2627.

Of course, finding a couple of clusters helps.
Logged

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #9 on: June 20, 2010, 11:02:32 PM »

Personally, I do not see L21 in Iberia as a big deal. Yes, indeed, Iberia is currently viewed as the P312 turf, and France as the L21 land. However, it is not a black-and-white matter.

The whole historical picture which is emerging now, is that P312 had arrived to Iberia around 4800-4500 years before present, went through a severe bottleneck, resurfaced again in Iberia as a small population around 3900 ybp, between those dates split off L21 who moved up North to France and further on all over Europe. As a result of this interplay P312 and l21 have identical ancestral (base) 67-marker haplotypes, and timespans to common ancestors of P312 and L21 are the same, within margin of error.

If P312 firmly stayed in Iberia and L21 jumped to Europe without traces in Iberia, they would have been cleanly separated - P312 will be exclusively Iberian, and L21 exclusively continental. However, life is not so black and white. P312 were moving from Iberia up North (Bell Beakers), hence, their descendants live on the continent. Some L21 left in Iberia, some of them went back on different occasions. As a results, some P312 are in Europe up to Scandinavia, some L21 are in Iberia as well. It is life.

         
Logged

R1a1

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #10 on: June 20, 2010, 11:53:26 PM »


If this is a cluster, the off-modal values are:

19=15

459b=9

YCAIIb=19

456=15



These 8 haplotypes certainly form a separate branch, with a base 25 marker haplotype:

13 24 15 11 11 14 12 12 12 13 13 29 -- 17 9 9 11 11 25 15 19 29 15 15 17 17

All 8 haplotypes collectively have only 16 mutations from the base haplotype, which gives 1150+/-310 years from their common ancestor. Now, let's see how they are related to L21 - quantitatively. Their base haplotype deviates from the L21 base haplotype by 8 mutations on 37 markers. This translates to 2450 years between their common ancestors. Since a common ancestor of L21 lived 3725 years before present (plus-minus margin of error), and these 8 haplotypes common ancestor lived 1150 years ago (plus-minus margin of error, see above), THEIR common ancestor lived (2450+1150+3725)/2 ~3700 years ago. Obviously, this is L21 itself.

In other words, these 8 individuals belong to a branch which descended from L21.
« Last Edit: June 20, 2010, 11:56:16 PM by aklyosov » Logged

R1a1

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #11 on: June 21, 2010, 05:37:09 PM »


These 8 haplotypes certainly form a separate branch, with a base 25 marker haplotype:

13 24 15 11 11 14 12 12 12 13 13 29 -- 17 9 9 11 11 25 15 19 29 15 15 17 17

All 8 haplotypes collectively have only 16 mutations from the base haplotype, which gives 1150+/-310 years from their common ancestor. Now, let's see how they are related to L21 - quantitatively. Their base haplotype deviates from the L21 base haplotype by 8 mutations on 37 markers. This translates to 2450 years between their common ancestors. Since a common ancestor of L21 lived 3725 years before present (plus-minus margin of error), and these 8 haplotypes common ancestor lived 1150 years ago (plus-minus margin of error, see above), THEIR common ancestor lived (2450+1150+3725)/2 ~3700 years ago. Obviously, this is L21 itself.

In other words, these 8 individuals belong to a branch which descended from L21.

Folks,

This story got an unexpected (for me, at least) twist. It turned out that Irish R1b1b2 haplotypes (1036 of 67-marker haplotypes) have on their haplotype tree (see a ref. I gave elsewhere on the paper on Irish R1b1b2 haplotypes) a separate and a distinct branch with YCAIIb=19. The branch sits on the tree between M222 and Irish III (Fig. 1 in the cited paper), and contains 14 haplotypes  with a base as follows:

13 24 14 11 11 14 12 12 12 13 13 29 -- 16 9 9 11 11 25 15 19 31 15 16 16 17

As you see, it has six mutaions (in bold) compared with the Iberian R-L21 base haplotype.  While the Iberian one was derived from a common ancestor of 1150 ybp (see above), the Irish has 17 mutations in the 14 25-marker haplotypes which places its common ancestor to 700+/-180 years bp, 450 years later compared to the Iberian one. No wonder, it took time for the journey from Iberia to the Isles. Six mutations netween their base haplotypes place THEIR common ancestor to around 2800 years before present.

In reality he lived some earlier, since the haplotype has more mutations in its 37- amd 67-marker format:

13 24 14 11 11 14 12 12 12 13 13 29 -- 16 9 9 11 11 25 15 19 29 15 16 16 17 -- 11 11 19 19 17 15 19 17 38 40 12 12 -- 11 9 16 17 8 10 10 8 10 10 12 23 23 16 10 12 12 14 8 13 24 22 13 12 11 13 11 12 12 12
 
Notice 19-19 in YCAII.

Its 37-marker haplotypes give 1300+/-210, and 67-marker haplotypes give 1150+/-170 years to its common ancestor. This corrects the timespan to their common ancestor with the Iberian one from 2800 to to 3100 ybp.

However, the biggest surprise is not here. The above 67-marker base haplotype differs by 26 (!!) mutations from the L21 base haplotype. This places THEIR common ancestor to about 5200 years before present. It is NOT L21, it is either L51 (5850+/-860 ybp), or that P312 which vanished in Iberia near 5000 ybp (Bell Beakers) and resutfaced some 3900 years before present.

We can conjecture, of course, left and right, however, 26 mutations in base 67 marker haplotypes places a common R1b1b2 ancestor to more than 5000 years back. That is where YCAII 19-19 came from, in Spain and Ireland.

 
Logged

R1a1

alan trowel hands.
Guru
*****
Offline Offline

Posts: 2012


« Reply #12 on: June 21, 2010, 06:27:26 PM »



So there is more haplotype variance between French samples than Spanish ones, correct? It is interesting we are finding more Spanish "clusters" within L21.

It is always the outlier that gets noticed.

I'm not sure I would characterize Iberian L21 or these clusters within Iberian L21 as outliers. L21 is pretty western, and Spain is right next door to France, where thus far the oldest L21 haplotypes are found.

I think maybe we overlooked Iberia for a long time, possibly because so many of our Spanish and Portuguese surname results came in as "Colonial". But the thing is, we were always getting them, little by little. I don't think there ever was a time, except maybe right at first, when we had zero Iberian L21.

Honestly, it wouldn't surprise me if L21 turned out to be a lot bigger in Iberia than we thought. The problem is going to be ferreting it out from among all that R-P312* and R-SRY2627.

Of course, finding a couple of clusters helps.

I do think though that in Iberia comparing the hobby tested non-Colonial S116* to L21 (prior to recent burst of hunting for L21 through cluster matching) was a fair comparison as both need an L21 test by definition.  So, I feel at least the S116*-L21 ratio must have been representative.  After all there is no reason why the non-Colonials should be biased towards one or the other.  Given that, I do feel that the S116* paragroup seems unusually dominant over L21 compared to other areas like France, Germany, the isles etc.   It seems to me that the proportion of S116* is quite low in central Europe. It looks a small portion of R1b1b2 in Germany and the recent Santiago study of France indicated that U152 must outnumber L21 and S116* combined in Alsace, which indicates a low proportion of the latter.  S116* is also very low in Ireland among native Irish surnames.  Whatever the paragroup really means, S116* seems hugely better represented in Iberia than anywhere else.   
Logged
rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #13 on: June 22, 2010, 07:31:08 AM »


These 8 haplotypes certainly form a separate branch, with a base 25 marker haplotype:

13 24 15 11 11 14 12 12 12 13 13 29 -- 17 9 9 11 11 25 15 19 29 15 15 17 17

All 8 haplotypes collectively have only 16 mutations from the base haplotype, which gives 1150+/-310 years from their common ancestor. Now, let's see how they are related to L21 - quantitatively. Their base haplotype deviates from the L21 base haplotype by 8 mutations on 37 markers. This translates to 2450 years between their common ancestors. Since a common ancestor of L21 lived 3725 years before present (plus-minus margin of error), and these 8 haplotypes common ancestor lived 1150 years ago (plus-minus margin of error, see above), THEIR common ancestor lived (2450+1150+3725)/2 ~3700 years ago. Obviously, this is L21 itself.

In other words, these 8 individuals belong to a branch which descended from L21.

Folks,

This story got an unexpected (for me, at least) twist. It turned out that Irish R1b1b2 haplotypes (1036 of 67-marker haplotypes) have on their haplotype tree (see a ref. I gave elsewhere on the paper on Irish R1b1b2 haplotypes) a separate and a distinct branch with YCAIIb=19. The branch sits on the tree between M222 and Irish III (Fig. 1 in the cited paper), and contains 14 haplotypes  with a base as follows:

13 24 14 11 11 14 12 12 12 13 13 29 -- 16 9 9 11 11 25 15 19 31 15 16 16 17

As you see, it has six mutaions (in bold) compared with the Iberian R-L21 base haplotype.  While the Iberian one was derived from a common ancestor of 1150 ybp (see above), the Irish has 17 mutations in the 14 25-marker haplotypes which places its common ancestor to 700+/-180 years bp, 450 years later compared to the Iberian one. No wonder, it took time for the journey from Iberia to the Isles. Six mutations netween their base haplotypes place THEIR common ancestor to around 2800 years before present.

In reality he lived some earlier, since the haplotype has more mutations in its 37- amd 67-marker format:

13 24 14 11 11 14 12 12 12 13 13 29 -- 16 9 9 11 11 25 15 19 29 15 16 16 17 -- 11 11 19 19 17 15 19 17 38 40 12 12 -- 11 9 16 17 8 10 10 8 10 10 12 23 23 16 10 12 12 14 8 13 24 22 13 12 11 13 11 12 12 12
 
Notice 19-19 in YCAII.

Its 37-marker haplotypes give 1300+/-210, and 67-marker haplotypes give 1150+/-170 years to its common ancestor. This corrects the timespan to their common ancestor with the Iberian one from 2800 to to 3100 ybp.

I'm a little confused.

Are you saying the Irish group with YCAIIb=19 probably originated in Iberia 2800-3100 years ago?

However, the biggest surprise is not here. The above 67-marker base haplotype differs by 26 (!!) mutations from the L21 base haplotype. This places THEIR common ancestor to about 5200 years before present. It is NOT L21, it is either L51 (5850+/-860 ybp), or that P312 which vanished in Iberia near 5000 ybp (Bell Beakers) and resutfaced some 3900 years before present.

We can conjecture, of course, left and right, however, 26 mutations in base 67 marker haplotypes places a common R1b1b2 ancestor to more than 5000 years back. That is where YCAII 19-19 came from, in Spain and Ireland.

This is where you really lose me (by that I mean I do not understand).

How can an L21+ cluster have a MRCA with the rest of L21 who was not himself L21? Wouldn't that entail two completely separate and independent origins and thus different branches of L21? Shouldn't all of L21 descend from the same, original L21 MRCA?
« Last Edit: June 22, 2010, 07:35:10 AM by rms2 » Logged

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #14 on: June 22, 2010, 01:11:32 PM »


I'm a little confused.

Are you saying the Irish group with YCAIIb=19 probably originated in Iberia 2800-3100 years ago?

This is where you really lose me (by that I mean I do not understand).

How can an L21+ cluster have a MRCA with the rest of L21 who was not himself L21? Wouldn't that entail two completely separate and independent origins and thus different branches of L21? Shouldn't all of L21 descend from the same, original L21 MRCA?

Your questions are quite reasonable. What I gave yesterday was a quick calculation, with round up alleles. Since the results obtained were a little strange, I have recalculated those things, and the common ancestor for both the Irish and the Iberian branches moved closer to the bottom of L21.

Let's go step by step. The format of free communications did not excourage me to go into details yesterday. So, please be patient. 

Let's specify several basic statements:

1. Hapogroup/subclade ladder (R1b1b2 in this particular case) present a series of subclades, some of them are parallel (such as P312 and U106), some are consecutive (such as L51 -->P310 --> P312 --> L21 --> L226.

2. When we have two branches of haplotypes with YCAII = 19-19 in each of their base haplotypes, they likely to descend from one common ancestor with YCAII = 19-19. On the other hand, they could have formed independently. In order to figure out which case is more likely, we look at other alleles/markers in those haplotypes, and make sure that they both belong to the same haplogroup (R1b1b2 in this particualr case). 

3. Mutational distances between two (or several) base haplotypes defines when THEIR common ancestor lived.

4. The "common ancestor" for a dataset refers only to haplotypes in the dataset.  For example, hundreds of present day P312 haplotypes coalesce to a common ancestor who lived 3900 years before present. It does not mean that a common ancestor of P312 in the past lived necessarily 3900 ybp. It could have been, say, 4900 ybp, but a population bottleneck moved it up, by 1000 years in this case. In other words, we see only common ancestor, closest to us. Only that, who survived and gave survived descendants.

5. Comparisons of base haplotypes from different populations and mutational differences between them can help us to move "deeper" in time and calculate more ancient times for common ancestors, which cannot be seen from nowdays populations.

To explain further, here are couple of examples.

Example 1.
If we see a branch of R1b1b2 haplotypes with a common ancestor of 1150 years bp (in Iberia), and another branch of R1b1b2 with a common ancestor of 700 years ago (in Ireland), and a mutational difference between their 67-marker haplotypes is only 2 mutations, we can safely say that the Irish branch descended from the Iberial one. (2/67 translates into 350 year difference, and (350+1150+700)/2 = 1100 years to THEIR common ancestor. This is very likely the Iberian common ancestor.

Though, this was not the case with the Iberian and Irish branches with YCAII=19-19.

If the mutational difference between them is, say, 26 mutations, which corresponds to 5475 years between their common ancestors, clearly, they could not have possibly be descended from each other. THEIR common ancestor lived (5475+1150+700)/2 = 3700 years ago.

This is closer to the actual picture, but, still, is a theoretical example.

Now let's move to other FACTS and (justified) calculations, and then to interpretation, who their common ancestor might have been.

In the preceding message both the Iberian R-L21 branch and the Irish R1b1b2 branch (no subclade was identified, except that it was R1b1b2) both had YCAII = 19-19, so we can guess that both came from one common ancestor. This is further justified by only 6 mutations between their base haplotypes in the first 25 markers, and 13 mutations in the 37 markers (this translates to 3750 and 4225 years between them). This gives 2800 or 3000 years to THEIR common ancestor. This is well within L21 subclade.

Therefore, looking at those Iberian and Irish haplotypes, dates to their common ancestors, and mutational differences between them, it is reasonable to conclude that both descended from L21, one stayed in Iberia (with a common ancestor 1150 ybp), another moved to Ireland with L21, and settled with its common ancestor 700 ybp.

However, these dates are based on their 25- and 37-marker haplotypes, and only on the comparison of the two. If we look at L21 as their supposed parent (ancestral) haplotype, we see a bit different pattern. The Irish (19-19) differs from its parent L21 by only 1 mutation in the first 12 markers, 2 mutations in the first 25 markers, and 8 mutations in 37 markers. The latter figure translates into 2450 years between the Iberian branch and L21, which poins at THEIR common ancestor of (2450+1150+3725)/2 = 3700 ybp, obviously, it is L21 itself. No problems, it is what the doctor ordered. However, the Irish branch differs from L21 by 26 mutations in 67 markers, which gives (5475+700+3725)/2 = 4950 years from THEIR common ancestor.

Now, I admit that these calculations were preliminary ones. If to look at the mutational difference more closely (which I normally do), and not at thye round up alleles, but at the average ones, with decimals, there are not 26 mutations, but "only" 23.85 mutations. This translates into 4925 years, and THEIR common ancestor lived (4925+700+3725)/2 = 4675 years before present. Considering margins of error, this is close to L21 common ancestor.

It might also indicate that the common ancestor lived a few centuries earlier compared that we see from L21 haplotypes. In a grand scheme of things it is not very significant.

Example 2. It shows that we sometimes look at things too directly. Suppose P312 landed in Iberia 4800 years ago. E3b chased them and killed so many that P312 went through a deep populational bottleneck. Survived P312 fled up North as Bell Beakers, starting at 4800 ybp, and became L21 as soon as in France. They carried with them an ancient R1b1b2 language, which was a blend of Hamitic language and the old non-IE language of R1b1b2. P312 in Iberia recovered 3900 ybp after the bottleneck.

What do we see now? We are scratching our head. We know that Bell Beaker culture was between 4800 and 3300, but see P312 in Iberia only from 3900 ybp. We say - "it is not them". We see L21 in France older than P312 in Iberia, and we ask - "how could it be, that the son is older than the father?". We discover the Hamitic language in European North and scratch our head - "how could it be, when P312 in Iberia do not show that language?". Where it came from? Why in the European North?

The reason is simple - there was life before population bottlenecks.                         

Anatole Klyosov
Logged

R1a1

Maliclavelli
Guru
*****
Offline Offline

Posts: 2147


« Reply #15 on: June 23, 2010, 03:34:58 AM »

Klyosov writes: "The "common ancestor" for a dataset refers only to haplotypes in the dataset. 
For example, hundreds of present day P312 haplotypes coalesce to a common ancestor who lived
3900 years before present. It does not mean that a common ancestor of P312 in the past lived
necessarily 3900 ybp. It could have been, say, 4900 ybp, but a population bottleneck moved it
up, by 1000 years in this case. In other words, we see only common ancestor, closest to us. Only
that, who survived and gave survived descendants".

This is the only reasonable thing you say. Apart your calculation of the MRCA we could discuss,
the same is worth for all clades upstream, and R1b1* and its subclades in this way can be more
more ancient than you and others (friends and enemies) are thinking.
This puts in game my theory of the mutations around the modal: any clade is one of the
numerous possible and if it has some mutations in very low mutating markers and the others into
the modal, it does mean that the others have mutated many times around the modal itself.
This was, is and will be my thinking.

Gioiello Tognoni del Badia, R1b1b2a (S136+). K1a1b1 (9932A)
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #16 on: June 23, 2010, 07:57:00 AM »

This is the only reasonable thing you say. ...This was, is and will be my thinking.

:-)))))))))))))

Funny. Greetings, my good old friend Gioiello, you have not been changed re. you superdirect comments and sweeping evaluations.  

However, as you well know, I do not accept that kind of "critique" which does not contain any substance.  Are the mutations rate constants which I employ are "unreasonable"? Is the concept of base haplotypes "unreasonable"? Is the concept of mutational difference between base haplotypes "unreasonable"?. Is the method of translation of those mutational differences into generations and years to a common ancestor "unreasonable"? Are those timespans to the common ancestors of said subclades "unreasonable"? Are calculations of margins of error "unreasonable"? Yes, I got it, it is above your pay. Well, if you cannot evaluate it - quantitatively, I mean, why to say that it is all "unreasonable"? (that is what you have essentially said).

O.K., I am just smiling. It is fun to get negative comments from folks who do not get it albeit have a burning desire to say something negative. Please continue.

Back to your comments.    

R1b1* and its subclades in this way can be more more ancient than you and others (friends and enemies) are thinking.

I do not know where you have assigned me, to which battalion, friend or foe. My guess is that still friendly one, since you said the other day that I am truly Toscanian. I hope, Toscanians are not enemies to you. Having said that, I have to tell that I did not mention R1b1* "age" in here. I think it is ancient enough. R1b seemingly appeared around 16,000 years before present (in Asia), and the "age" of R1b1 should be close to it. Is it ancient enough to you? R1b1/R1b1b2-M269 on the Russian Plain were, it seems, around 8,000-6,000 ybp. Is it ancient? To me, yes, in a way. Not as ancient as the Universe, but still...

I have not seen R1b1b2 and its subclades in Europe older than 4800-4500 years bp. Have you? I would appreciate DATA, not your theory of mutations recycling around the "modal" (I do not know what does it mean in your "theory").    

Let's see it again:

This puts in game my theory of the mutations around the modal: any clade is one of the
numerous possible and if it has some mutations in very low mutating markers and the others into the modal, it does mean that the others have mutated many times around the modal itself.
This was, is and will be my thinking.

Well, the last phrase is funny indeed. By saying it, you effectively removed yourself from science.  Science is always changeable in its frontiers. It constantly adjust itself to newly acquired knowledge. What you have said, it is a position of a fanatic, not a scientist. I hope you realize it, at least now, after my "description".

Frankly, I do not understand what you have written above the last funny statement. Please help me out.

You said "mutations around the modal". What does it mean? First, define "modal" for me, please. Often there are many "modals" on a haplotype tree (in a dataset, if you wish). "Modal" in my "book" is just a series of identical haplotypes in a population. For instance, if you are prolific enough, a bunch of your children have your "modal". A series of haplotypes with null-mutation have a "modal" on their branch of the tree. The "Cohen Modal Haplotype" is just a modal of the branch with TMRCA of 1000 ybp on the J1 haplotype tree. R1b1b2 have a "modal" of around 4000 ybp, but many "modals" of their subclades and other branches. Therefore, a "modal" is typically an ancestral haplotype for a subpopulation, young or old. Each haplogroup has many modals.

Is that what you mean?

Now, mutations from each "modal" occur "up" and "down", from each marker. Some faster, some slower. It is a statistical process. Hence, some of them come back, we call it "reverse" mutations, or "back" mutations.
  
Now, after these definitions let's look again at your writing:

>any clade is one of the numerous possible...

(that is fine, no problem)

>... and if it has some mutations in very low mutating markers and the others into the modal...

(well, it "has some mutations" in ALL mutating markers, and they ALL are mutating on their time scale. Why "very low mutating markers", where they came from? Where did you lose "fast mutating markers")  

>... it does mean that the others have mutated many times around the modal itself.

(there is something strange here. Please define "around" here).

What is "around"? ALL mutations are AROUND in a way. How else? They do not jump on some other haplotypes. 17-->18 is "around"? How about 23-->21? Is it "around"? How about 21-->23? Still "around"?

O.K., let's settle that ALL mutations in ANY haplotype, including the ancestral one ("modal"), are "around". Then what we get from your wording? Where is substance? What does it change, if to call all mutations as "around"? Does it change any calculations? Any estimates?

My friend Gioiello, you need to come back to your drawing board. Please think twice before you say something (anything) about haplotypes and their mutations. And, please, remember, that sometimes you talk to professionals in the field. It does not mean that professionals do not make mistakes, however, a likelihood of that is much lower compared to that for lay folks. This is a "rule of thumb".

Love,

Anatole Klyosov
« Last Edit: June 23, 2010, 08:05:17 AM by aklyosov » Logged

R1a1

Maliclavelli
Guru
*****
Offline Offline

Posts: 2147


« Reply #17 on: June 23, 2010, 09:10:10 AM »

Anatole,
I have always considered you a friend, being a Russian, and Russians have a tender heart.
I don't consider a friend Vizachero, with whom you too have had some problems in
the last times and I don't consider a friend Mayka, who caused my banishment from Forums-dna
(Russians and Poles, even though they speak a similar language and have similar haplogroups
aren't the same). But these are positions of mine, personal and criticizable.

About the "mutations around the modal" I wrote a letter to you in the past and you should know what
I mean.
How is it possible that DYS426, a very slow mutating marker, has changed many times from the modal
R1b1* and subclades till R1b1b2a1b etc. and other fast mutating markers are in the modal the same?
I think that they have changed many times around the modal, i.e., for instance DYS391, that we find
above all with two values, 10 or 11, can have changed: 10 to 11 to 10 to 11 to 10 etc., but in your
calculation you count it only for one.
I have also hypothesized the mutation for the tangent, when a mutation, by chance, goes only in one
direction, and those values are the most interesting.

Gioiello
« Last Edit: June 23, 2010, 01:44:51 PM by Maliclavelli » Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #18 on: June 23, 2010, 10:26:46 AM »


About the "mutations around the modal" I wrote a letter to you in the past and you should know what I mean.

Dear Gioiello,

I was never able to understand it, and I do not understand it now. That is why I hoped that some day you can explain what do you mean by "mutations around the modal", and this day apparently came. However, your explanation is not satisfactory. In fact, it is not an explanation at all. It is some conjecture based on some illusion(s). Let take a close look at your "explanation", and I will explain you what your illusion is based on. 

(see the next message)
Logged

R1a1

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #19 on: June 23, 2010, 11:04:02 AM »

How is it possible that DYS426, a very slow mutating marker, has changed many times from the modal R1b1* and subclades till R1b1b2a1b etc. and other fast mutating markers are in the modal the same?
I think that they have changed many times around the modal, i.e., for instance DYS391, that we find above all with two values, 10 or 11, can have changed: 10 to 11 to 10 to 11 to 10 etc., but in your calculation you count it only for one.
I have also hypothesized the mutation for the tangent, when a mutation, by chance, goes only in one direction, and those values are the most interesting.

This is your first and foremost illusion. DYS426 is a marker which mutates with its pace, acording to its mutation rate constant, which is (John Chandler's data) 0.00009 mutations per generation. It means that it mutates once per every 11 thousand birth of boys.  For a comparison, in the USA in 2008 there were 4,247,000 births, and about half of them, that is 2 million, were boys. In other words, DYS426 mutated in them - only in 2008 - about 180 times. Each of their male descendants wil carry a mutated DYS426. Can you imagine, how many of those mutations of DYS426 were in the past thousands of years? Was it "changed many times"? 

Another example. DYS388 is a very slow marker as well. However, almost 20% of R1a1 in Ireland have DYS388=10. Do you think it was "changed many times"? No, it changed only once in the past, all others are inherited. That is why DYS388=10 form just one branch on the tree, and I do not "overcount" them. I count them only once. Apparently, you have a vague idea how mutations should be counted.

So, they have not "changed many times around the modal", they changed with their own pace in each case. They go in a direction determined by blind statistics. They do not go "in one direction", their assymmetry is typically just an apparent one. It just reflect branches, as in the DYS388=10 case. It was mutated from 12 to 11 then to 10, all "11" died out, and 12 and 10 have proliferated.

It is very basic in DNA genealogy.

Regards,

Anatole Klyosov
Logged

R1a1

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #20 on: June 23, 2010, 11:45:32 AM »

. . . We discover the Hamitic language in European North and scratch our head - "how could it be, when P312 in Iberia do not show that language?". Where it came from? Why in the European North?

The reason is simple - there was life before population bottlenecks.                         

Anatole Klyosov


I appreciated and understood the rest of your explanation (in that earlier post above, before your exchange with Gioiello).

But what do you mean "Hamitic language"? Where is there a Hamitic language (now classed together with Semitic languages in the "Afro-Asiatic" language family) connected with R1b1b2 of any kind, especially in Northern Europe?

Basque is the language of a small minority group and cannot be connected to R1b1b2 or to P312 as a whole and is certainly not "Hamitic". It cannot be convincingly linked to any known family of languages, but seems to perhaps have some remote ties to some of the Caucasian languages, like Chechen.

You once mentioned that Iberian cattle came from North Africa. Do you have a link to the study that showed that? I could only find one such study, and it found only a single T1, African-type cow (dated to 1800 BC) among a group of Bronze Age cattle remains. The rest were all of T3 European type. One T1 cow among a bunch of T3 cows doesn't seem to argue for a North African route to Iberia for P312.
Logged

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #21 on: June 23, 2010, 12:21:12 PM »


But what do you mean "Hamitic language"? Where is there a Hamitic language (now classed together with Semitic languages in the "Afro-Asiatic" language family) connected with R1b1b2 of any kind, especially in Northern Europe?

I gave it as an example of how things can be complicated. You can use a XYZ language in place of Hamitic. I did not intend to discuss specifically Hamitic language, since it is not a linguistic Forum. Linguists know that an ancient Celtic language, including that in the North of Europe, contains a Hamitic component. This, actually, is one more argument that R1b1b2 came across Gibraltar from North Africa. As you see, I have many cards up my sleeve. As you see, you are just confirming my statement that things (about R1b1b2 in this particular case) are complicated to be considered as black and white.  We normally see "above" population bottlenecks, but there was life below them. And we see sometimes reflections of that life.


Basque is the language of a small minority group and cannot be connected to R1b1b2 or to P312 as a whole and is certainly not "Hamitic". It cannot be convincingly linked to any known family of languages, but seems to perhaps have some remote ties to some of the Caucasian languages, like Chechen.

I have never said that Basque is Hamitic language. On the contrary, I gave an example how it might be that Basque is not Hamitic, but ancient Celtic could carry a component of Hamitic.

As to the rest, I am a bit surprised how certain you are re. things, which nobody in the words knows for sure. How can you say that Basque language "cannot be connected to R1b1b2"?? Are you expert in Basque language? Are you really familiar with its linguistic features, so poorly understood even among specialists? How it "cannot be connected to R1b1b2" when 93% of the Basques ARE R1b1b2? Of course Baqsque language has some roots in Caucasian languages, and this fits to what I have written here about a route of the (future) Basques via the Caucasus to the middle East and - across North Africa - to Iberia.          


You once mentioned that Iberian cattle came from North Africa. Do you have a link to the study that showed that? I could only find one such study, and it found only a single T1, African-type cow (dated to 1800 BC) among a group of Bronze Age cattle remains. The rest were all of T3 European type. One T1 cow among a bunch of T3 cows doesn't seem to argue for a North African route to Iberia for P312.

I do not have it right now. However, I have never based the "African-Iberian" concept on just one cow. Your last sentence implies that I did exactly that.

I have a reasonable suggestion. When you CAN present an alternative concept, please do it. This will be a constructive approach. Just to criticise separate "moments" not in their entirety, but picking them separately, often in a distorted way, and without presenting an alternative consideration, would not do us any good.

It is a complex issue, and nobody in the world has a definite answer which connects all "the dots". I try to do it, putting together various observations, facts and interpretations, along with my own calculations. Some of those fragments might fall out in the future, but a result of new and good explanations, based on facts, missed now, or just unknown as yet. But they cannot fall out just because someone does not "like it". It is not enough.

I hope you support such a suggestion. I have heard enough arguments such as "Earth is flat, otherwise people would fall from the globe".

  
« Last Edit: June 23, 2010, 12:24:56 PM by aklyosov » Logged

R1a1

rms2
Board Moderator
Guru
*****
Offline Offline

Posts: 5023


« Reply #22 on: June 23, 2010, 01:13:21 PM »

. . .

I have never said that Basque is Hamitic language. On the contrary, I gave an example how it might be that Basque is not Hamitic, but ancient Celtic could carry a component of Hamitic.

As to the rest, I am a bit surprised how certain you are re. things, which nobody in the words knows for sure. How can you say that Basque language "cannot be connected to R1b1b2"?? Are you expert in Basque language? Are you really familiar with its linguistic features, so poorly understood even among specialists? How it "cannot be connected to R1b1b2" when 93% of the Basques ARE R1b1b2? Of course Baqsque language has some roots in Caucasian languages, and this fits to what I have written here about a route of the (future) Basques via the Caucasus to the middle East and - across North Africa - to Iberia.

Basque cannot be connected to R1b1b2 as a whole in the same way that bedwetting cannot be connected to R1b1b2 as a whole, even if you found that 93% of bedwetters were R1b1b2. That is because by far most men are not bedwetters.

Basques number about 3 million people. If, for the sake of argument, we agree that about half of them are male, that is 1.5 million. Even if 100% of them were R1b1b2, that would mean you had found 1.5 million modern men whose ancestors might have spoken the Basque language.

But how many more millions of R1b1b2 men are not Basque and have no provable connection to the Basques and their language?

So, what is the reason to associate R1b1b2 as a whole with the Basques and their odd language?  


I do not have it right now. However, I have never based the "African-Iberian" concept on just one cow. Your last sentence implies that I did exactly that.

I have a reasonable suggestion. When you CAN present an alternative concept, please do it. This will be a constructive approach. Just to criticise separate "moments" not in their entirety, but picking them separately, often in a distorted way, and without presenting an alternative consideration, would not do us any good.

It is a complex issue, and nobody in the world has a definite answer which connects all "the dots". I try to do it, putting together various observations, facts and interpretations, along with my own calculations. Some of those fragments might fall out in the future, but a result of new and good explanations, based on facts, missed now, or just unknown as yet. But they cannot fall out just because someone does not "like it". It is not enough.

I hope you support such a suggestion. I have heard enough arguments such as "Earth is flat, otherwise people would fall from the globe".

You think I am criticizing you, when I am really just asking questions. I wasn't offering an alternative scenario. I was merely trying to understand the one you have presented.

I wasn't trying to distort anything. I merely asked about aspects of what you presented that raised questions in my mind.

I have read a lot about the Celts and Celtic languages, but I cannot recall anyone suggesting there is a Hamitic element in the Celtic languages. But maybe I missed that. After all, one person cannot have read all there is to read.

I did not mean to imply that you were basing your Iberian cattle idea on one African cow. I merely asked if you could provide me with a link to a study that showed that Iberian cattle came from Africa. I mentioned the study that I mentioned because it tended to show that Iberian cattle were far more European than African.

I think I will leave off asking you any more questions. You seem to take things too personally and to read antagonism into things where no antagonism or criticism is intended.
« Last Edit: June 23, 2010, 01:16:34 PM by rms2 » Logged

Maliclavelli
Guru
*****
Offline Offline

Posts: 2147


« Reply #23 on: June 23, 2010, 01:59:15 PM »

Klyosov writes: "This is your first and foremost illusion. DYS426 is a marker
which mutates with its pace, acording to its mutation rate constant, which is
(John Chandler's data) 0.00009 mutations per generation. It means that it mutates
once per every 11 thousand birth of boys.  For a comparison, in the USA in 2008
there were 4,247,000 births, and about half of them, that is 2 million, were boys.
In other words, DYS426 mutated in them - only in 2008 - about 180 times. Each of
their male descendants wil carry a mutated DYS426. Can you imagine, how many of
those mutations of DYS426 were in the past thousands of years? Was it "changed
many times"? 

Anatole, you are a mathematician. If every 11 thousands of masculin births there
is a mutation in DYS426 (rate 0.00009), how many mutations there are in a marker
which has a rate of 0.002? And how are the possibilities that the only survivor of
that mutated marker has the other marker not changed? And we are speaking at least
of 25, 37, 67 markers.
 
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

aklyosov
Member
**
Offline Offline

Posts: 45


« Reply #24 on: June 23, 2010, 02:00:46 PM »


Basque cannot be connected to R1b1b2 as a whole...

Basques number about 3 million people. If, for the sake of argument, we agree that about half of them are male, that is 1.5 million. Even if 100% of them were R1b1b2, that would mean you had found 1.5 million modern men whose ancestors might have spoken the Basque language.

But how many more millions of R1b1b2 men are not Basque and have no provable connection to the Basques and their language?

So, what is the reason to associate R1b1b2 as a whole with the Basques and their odd language?  

:-))))))

No problem with that. If you mean that not all R1b1b2 are the Basques, I would not argue.

However, if you mean that the Basques are not "connected" to R1b1b2 (which, probably is not what you mean), I would disagree. That "as a whole" does not make any sense to me.

The Basques have their history, and that history and their language have a deep meaning in the history of R1b1b2. As a "whole" or not.

There were a number of pre-IE languages in Europe, associated with R1b1b2 (presumably). There is a strong reason to believe that the R1b1b2 language 4500-4000 years back was not Indo-European language.  The Basque was one of them, or the first one which set foot in Europe. Nobody in the world knows where the Basques language came from, though it is known that the Basque language is a agglutinative language, similar (in that regard) with Finno-Ugric language, some Caucasian languages, Turkic languages,  some Native American languages, etc. It all points to Altaian group of languages, where R1b has probably emerged many thousands years ago. That is how some dots are being connected.    

It does not make any difference how many Basques are around.


You think I am criticizing you...

Not at all. You have not been personal at all. Again, it was not what I meant. All I meant that a productive discussion needs alternative suggestions with their justification. If one cannot offer an alternative explanation of the puzzle, it is not a constructive discussion. That is why to say that "the Basques are not connected to R1b1b2 as a whole" is not constructive. It does not solve any question, it does not offer any hypothesis, it does not explain anything.

Hence, my reaction. I want to find answers to the principal question - what is history of R1b haplogroup, and the Basques hold an important key to that (among, of course, a number of other puzzles, such as what was the initial R1b language, and how it was transforming to the IE language, which R1b1b2 acquired, apparently, only some 3000 years before present).  There are many puzzles, and all of them need considerations and alternative hypotheses. Based on tangible things.

If you (or anyone) want to ask questions, fine. However, I would prefer alternative hypotheses. Or to see DATA, based on which questions are asked. This would be the best. In this case it would be a two-way street.      

Anatole Klyosov
Logged

R1a1

Pages: [1] 2 3 4 Go Up Print 
« previous next »
Jump to:  


SEO light theme by © Mustang forums. Powered by SMF 1.1.13 | SMF © 2006-2011, Simple Machines LLC

Page created in 0.161 seconds with 19 queries.