PDA

View Full Version : R1b subclade prediction



Tomenable
07-11-2016, 07:24 PM
I upgraded Y12 markers to Y37 markers and have just received new STR results.

Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?

If not then I will order SNP Pack.

Tomenable
07-11-2016, 07:37 PM
Also my new matches after receiving these Y37 results are interesting:

https://s32.postimg.org/bbuanjamt/matches.png

I have a 99.44% probability that I share a common paternal ancestor with Mr. E. Chisholm and a 99.44% probability that I share such an ancestor with Mr. W. J. Chisholm within the last 24 generations (which if I'm not mistaken, is ca. 600 years).

And I have a 99.36% probability that I share a common paternal ancestor with one Mr. Oliver, and also a 98.03% probability that I share a common paternal ancestor with another Mr. Oliver - in both cases, as well within the last 24 generations / 600 years.

Apparently all of my 37 marker matches are Scotsmen:

https://en.wikipedia.org/wiki/Clan_Chisholm

https://en.wikipedia.org/wiki/Oliver_(Scottish_surname)

Tomenable
07-11-2016, 07:40 PM
Scots in Poland or Poles in Scotland?:

http://www.krakowpost.com/7881/2014/03/scots-in-poland-poles-in-scotland

Tomenable
07-11-2016, 07:47 PM
STRs (in FTDNA order):

Panel 1 - 13 24 14 10 11-14 12 12 11 13 13 29
Panel 2 - 17 9-9 11 11 26 15 19 32 15-15-17-17
Panel 3 - 10 11 19-23 15 15 18 17 36-37 12 12

TigerMW
07-11-2016, 08:08 PM
I upgraded Y12 markers to Y37 markers and have just received new STR results.

Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?

If not then I will order SNP Pack.

This probably won't sound appealing but you might consider upgrading to 67 STRs. In the R1b project we actually recommend 111 now but consider 67 minimal for predicting haplogroups. There are a few STR signatures that usually appear consistent at only 37 STRs, i.e. NW Irish, Irish III (off the top of my head.) However, these signatures probably only very reliable predictors for less than 5% (at most) R1b haplotypes.

It looks like your matches screen has no closer GDs at 37 than 3. That's not that good at 37 STRs. If you had a couple of 37/37 type matches or every 36/37 all with the same haplogroup labels that might be telling. If not, 67 STRs might tell you something and worst case is it sets you up well in the matching database.

Tomenable
07-11-2016, 08:12 PM
Thanks Mikewww.

I will probably order R1b SNP Pack, instead of more STRs. What do you think about this?


This probably won't sound appealing but you might consider upgrading to 67 STRs. In the R1b project we actually recommend 111 now but consider 67 minimal for predicting haplogroups.

But can 37 markers at least tell me if I'm R1b-L51+, and if so, then if I'm P312 or U106?

Tomenable
07-11-2016, 08:17 PM
My matches at 12 markers from Poland and Czech Republic were mostly R1b-P312-U152-L2.

So I thought that I'm probably L2, but now I'm not so convinced about it anymore.

Tomenable
07-11-2016, 08:29 PM
It looks like your matches screen has no closer GDs at 37 than 3. That's not that good at 37 STRs.

But probability of sharing an ancestor with them within last 24 generations, is still given as close to 100%.

Is this a reliable estimate of probability, or is it possible that in fact we don't share an ancestor?

sciencediver
07-11-2016, 08:37 PM
What subclade do the Chisholms belong to?

R.Rocca
07-11-2016, 08:41 PM
STRs (in FTDNA order):

Panel 1 - 13 24 14 10 11-14 12 12 11 13 13 29
Panel 2 - 17 9-9 11 11 26 15 19 32 15-15-17-17
Panel 3 - 10 11 19-23 15 15 18 17 36-37 12 12

Although this doesn't mean much, you are less likely to be L23+Z2103 due to your first value (13 instead of 12) and more likely to be P312+ instead of U106+ due to your second value (24 instead of 23). But again, there are thousand of examples that make this kind of analysis meaningless. Unlike Mike, I have very little desire to upgrade to 111, especially if not looking for a more recent ancestor. If I were you I would get the 67 marker test and then the SNP pack.

Tomenable
07-11-2016, 09:06 PM
What subclade do the Chisholms belong to?

They probably belong to R1b-L21, and I probably as well belong to R1b-L21.

In "Y-DNA - Haplogroup Origins", both of my matches at 25 markers are L21:

(both S1026 and A600 are subclades of L21):

https://s32.postimg.org/a211s6fc5/L21.png

Tomenable
07-11-2016, 09:10 PM
Chisholm Clan has its own FTDNA Project:

https://www.familytreedna.com/public/Chisholm/

mouse
07-11-2016, 09:21 PM
I upgraded Y12 markers to Y37 markers and have just received new STR results.

Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?

If not then I will order SNP Pack.

Take the Big Y test in the next sale. You will get all of your YSNPs plus 400 YSTR markers.

Gravetto-Danubian
07-11-2016, 09:23 PM
I'd just go straight for BigY

Tomenable
07-11-2016, 09:26 PM
Take the Big Y test in the next sale. You will get all of your YSNPs plus 400 YSTR markers.
I'd just go straight for BigY

The next sale = probably as late as June 2017 (Father's Day 2017) ??? :)

Well, I'm not in a hurry! Also, I do have this stereotypical Scottish frugality.

Tomenable
07-11-2016, 09:29 PM
There is even such a joke:

"- Who is a Poznaniak?" (= a Pole from the region of Poznan)

"- A Scotsman expelled from Scotland for excessive frugality".

Tomenable
07-11-2016, 09:37 PM
I think my curiosity is shifting towards autosomal ancestry recently.

But FTDNA offers no autosomal tests, right?

What do you recommend then, 23andMe?

rms2
07-11-2016, 10:32 PM
I think my curiosity is shifting towards autosomal ancestry recently.

But FTDNA offers no autosomal tests, right?

What do you recommend then, 23andMe?

FTDNA's Family Finder test is an autosomal dna test.

R.Rocca
07-11-2016, 10:37 PM
They probably belong to R1b-L21, and I probably as well belong to R1b-L21.

In "Y-DNA - Haplogroup Origins", both of my matches at 25 markers are L21:

(both S1026 and A600 are subclades of L21):

https://s32.postimg.org/a211s6fc5/L21.png

I'll caution that none of my 25 marker matches belong to my subclade. So, you may likely get a surprise if you test via SNP Pack.

ADW_1981
07-11-2016, 10:59 PM
You are probably L21 as you say (GD of 4 at 37 markers) and share a recent common ancestor one of with the Chisholm clan lineages. You are closer to Brits than I am...lol.

http://www.ancestry.ca/name-origin?surname=Chisholm

There might be some benefit to upgrading to 67 to see if those matches fall away, or simply doing an a la carte L21 SNP test for less if you are more intrigued to confirm the L21 result.

GoldenHind
07-11-2016, 11:07 PM
I upgraded Y12 markers to Y37 markers and have just received new STR results.

Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?



If not then I will order SNP Pack.

The value of 37 markers in R1b is dependent on how many off modal markers you have. The closer you are to the modal, the less valuable they are. It is entirely possible to have matches at 37 markers with different R1b subclades.

You can try plugging your markers into one of several subclade predictors, such as:

http://members.bex.net/jtcullen515/haplotest.htm

They were primarily designed for HG I, but work reasonably well with R1b. However for R1b, it will just identify how close you are to Nordtvedt's R1b varieties, most of which are connected with specific R1b subclades (for instance Frisian = U106).

My suggestion is doing the R1b M343 background test next. This will definitely tell you whether you are P312 or U106, and almost certainly identify your subclade as well. 67 markers is really essential for R1b, but if you are curious about deep ancestry, I think identifying one's subclade is more important. You can always upgrade to 67 markers later.

Incidentally there is an Oliver (the surname is much more common in England than in Scotland) who has tested R1b-P312>DF99, and some Chisholms whose markers suggest they are likely to be DF99. However your modal 13 at 389-1 makes it highly unlikely, though not impossible, that you will fall into that subclade. However DF99 has been found in someone with ancestry from southeast Poland.

miiser
07-11-2016, 11:44 PM
I upgraded Y12 markers to Y37 markers and have just received new STR results.

Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?

If not then I will order SNP Pack.

Since no one else has, I will at least attempt to answer the actual question that you asked at the start of this thread. Yes, there are several predictors, and they often have a pretty good confidence level at 37 markers. ISOGG has a collection of links: http://isogg.org/wiki/Y-DNA_tools#Y-STR_haplogroup_prediction_tools

Of course, the confidence level depends on how deeply into the branching one attempts to predict. One can predict R1b versus R1a with pretty good confidence. Z253 versus Z255, for example, not so much.

FTDNA used to make haplogroup predictions from 37 markers until they realized they could make more money selling SNP packs instead. The predictions from 37 markers weren't 100% accurate, but they were close.

TigerMW
07-12-2016, 01:59 AM
...
FTDNA used to make haplogroup predictions from 37 markers until they realized they could make more money selling SNP packs instead.
Why make these accusations about their intentions? This draws off track of the thread and is just mud slinging. I think you are wrong on your proclamation of their intentions but to discuss that further is off track so I will leave it alone.

...
The predictions from 37 markers weren't 100% accurate, but they were close.
How close is close in your books? Are talking about predicting 4000 year-old haplogroups or 1500 year-old haplogroups? What % of haplotypes were accurate?

TigerMW
07-12-2016, 02:19 AM
Also my new matches after receiving these Y37 results are interesting:

https://s32.postimg.org/bbuanjamt/matches.png

I have a 99.44% probability that I share a common paternal ancestor with .... Chisholm and a 99.44% probability that I share such an ancestor with.... Chisholm within the last 24 generations (which if I'm not mistaken, is ca. 600 years).

And I have a 99.36% probability that I share a common paternal ancestor with one Mr. Oliver, and also a 98.03% probability that I share a common paternal ancestor with another Mr. Oliver - in both cases, as well within the last 24 generations / 600 years.

Apparently all of my 37 marker matches are Scotsmen:

https://en.wikipedia.org/wiki/Clan_Chisholm

https://en.wikipedia.org/wiki/Oliver_(Scottish_surname)

I am not on my computer with all of the spreadsheets so I'm doing this by eye (- take this with a grain of salt.)

I found the Chisholm's at https://www.familytreedna.com/public/Chisholm/default.aspx?section=yresults

I think they are red herrings (false matches). You look like a match with a lot of Scots because I can see the long (111 STR) haplotype of one of the Chisholm's and he matches well with the Scots Modal (R1b-L1065) haplotype.

You have some of the Scots Modal STR markers too but it is risky to say much at 37 STRs. The tell-tale may be your YCAII values. Scots Modal folks almost always have YCAII=19,24 or YCAII=19,22. The red flag is that you have 19,23 which is the WAMH (Western Atlantic) modal. Hence, my prediction is you are NOT Scots Modal and are no more closer related to them than 4000-4500 years ago.

You do have a very WAMH-like haplotype so you probably are P312+ or at least P311+. I just gave you the sleeves out of my vest though as far as predictions go since that we are talking about 4500-6000 year old haplogroups. Sorry that is not of much help but you can see why I like at least 67 STRs. We are only working with half a puzzle at 37.

I'll look for the Oliver's tomorrow. As has been said on this thread, an NGS test like Big Y is the best thing to do but it is expensive. The reason I like the 67 STRs is that in any case you have a good position established in the matching database for future matches. If you are lucky, you might have a new match come in at 67 that is just outside the FTDNA thresholds at 37. Even if you do Big Y, if you are willing to spend that much then why you not spend the extra to put a better (taller) flagpole in the ground to make it easier for other people to find you and your newly discovered (via Big Y) SNPs.

You could still be L1065+ / Scots Modal. I think that would be most curious as that might just mean you are early branch away from the rest of the L1065+ people. ... and from Poland no less. A lot of people are stuck on L21 (L1065 is a subset) spread all over the place from the British Isles. It could be so, but the L21 might also have originated from Central Europe.

miiser
07-12-2016, 02:46 AM
Why make these accusations about their intentions? This draws off track of the thread and is just mud slinging. I think you are wrong on your proclamation of their intentions but to discuss that further is off track so I will leave it alone.

How close is close in your books? Are talking about predicting 4000 year-old haplogroups or 1500 year-old haplogroups? What % of haplotypes were accurate?

My point is that haplogroup predictions are able to be made from 37 markers, and FTDNA used to make them until they stopped.

The M-343 SNP pack became available for purchase in July 2015, and the haplogroup predictions disappeared in July 2015. I'll let Tomenable draw his own conclusions.

It's not mud slinging to observe that a for-profit company does things for profit, and I think it would be best if you didn't personalize this thread by turning it into a personal beef against me. You are not compelled to come to FTDNA's defense every time someone supposes that their actions are motivated by profit. If you have a problem with my comment, then I suggest you take it up with a moderator rather than drag Tomenable's thread down into the mud of a pro/anti FTDNA debate (which you alone, not I, have done).

Táltos
07-12-2016, 03:44 AM
My point is that haplogroup predictions are able to be made from 37 markers, and FTDNA used to make them until they stopped.

The M-343 SNP pack became available for purchase in July 2015, and the haplogroup predictions disappeared in July 2015. I'll let Tomenable draw his own conclusions.

It's not mud slinging to observe that a for-profit company does things for profit, and I think it would be best if you didn't personalize this thread by turning it into a personal beef against me. You are not compelled to come to FTDNA's defense every time someone supposes that their actions are motivated by profit. If you have a problem with my comment, then I suggest you take it up with a moderator rather than drag Tomenable's thread down into the mud of a pro/anti FTDNA debate (which you alone, not I, have done).

That's correct. All sides are to remain civilized and on topic.

Now as obviously you realize R1b is a rather large tree, do you think Tomeable will be satisfied with just a predictor? The SNP packs are good to drill down to a subclade. Depending on his budget, and goals then sure which choice? Big Y vs FGC? Who cares if FTDNA or any of these other companies are trying to make money? Who isn't? It is a business. All of the testing companies are.

Tomeable,
My brother at 37 markers had matches at a GD of 2. Some went on to test Big Y. Their results L245 > Y2209 > Y2225 > Y2200 > YP1003 > YP1071 > YP1010 > YP1009 | Q-YP1009. My brother was only positive for YP1003.

SNPs is where it is at for the Y chromosome today. FTDNA cannot possibly make reasonable predictions based on just STRs anymore.

MitchellSince1893
07-12-2016, 04:08 AM
Just to add to what Richard said; before I got my SNP results, I thought for sure I was going to be L21 as most of my 25 marker matches were L21. It's all guess work until you get your SNP results.

miiser
07-12-2016, 04:13 AM
That's correct. All sides are to remain civilized and on topic.

Now as obviously you realize R1b is a rather large tree, do you think Tomeable will be satisfied with just a predictor? The SNP packs are good to drill down to a subclade. Depending on his budget, and goals then sure which choice? Big Y vs FGC? Who cares if FTDNA or any of these other companies are trying to make money? Who isn't? It is a business. All of the testing companies are.

Tomeable,
My brother at 37 markers had matches at a GD of 2. Some went on to test Big Y. Their results L245 > Y2209 > Y2225 > Y2200 > YP1003 > YP1071 > YP1010 > YP1009 | Q-YP1009. My brother was only positive for YP1003.

SNPs is where it is at for the Y chromosome today. FTDNA cannot possibly make reasonable predictions based on just STRs anymore.

I would not have carried on this discussion further. But since you, as an administrator, have asked me specific questions, then I conclude that it will not be considered contentious or out of line for me to respond.

I can only guess what will satisfy Tomenable. But his specific questions in the original post were:

"Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?"

The answer to both these questions is yes. So that's the answer I gave.

37 markers will not allow one to identify their "terminal" SNP, as you point out. But this is not the question that Tomenable asked. 37 markers will almost always allow one to predict a subclade below R1b-M269 with good confidence. This is what FTDNA used to do, and this is what they stopped doing in July 2015.

I don't consider it a criticism against FTDNA if they made a decision based on profit motivations. It is simply a neutral matter of fact that for-profit corporations have profits as their primary goal in everything they do. I mentioned FTDNA's cessation of haplogroup prediction in my previous comment only as support for the argument that predictions CAN be made from 37 markers, and the decision to cease and negate such predictions below R1b-M269 was not due to a failure of the prediction methodology, but attributable instead to profit motives. As such, the decision to cease such predictions does not reflect negatively on the prediction methodology itself. The reason Tomenable's FTDNA account says nothing beyond "R1b-M269" is NOT that nothing more can be deduced. The reason is that FTDNA can make more money by not sharing what can be deduced.

If Tomenable wants to know his "terminal" SNP, he should do an SNP pack or NGS test. If he wants to identify his recent paternal relatives and estimate how closely related they are, he should upgrade to 67 or 111 markers. If he wants to identify his closest relatives, not only limited to the paternal branch, he should purchase an Ancestry DNA test.

If he wants the answer to the question he actually asked in the thread OP, the answer is: Yes, one can usually predict a subclade below R1b-M269 with good confidence based on only 37 markers.

TigerMW
07-12-2016, 04:16 AM
...
SNPs is where it is at for the Y chromosome today. FTDNA cannot possibly make reasonable predictions based on just STRs anymore.
I both agree and disagree.

I agree in terms of predicting a youthful terminal haplogoup, say of the last 400-800 years, it is clearly not in cards for 37 STRs. This is probably why FTDNA backed off to older haplogroups for predictions back in early March of 2015. Prior to that there were complaints of too aggressive of predictions which did ended up being wrong on occasions. It doesn't feel good if you spent the money and are the exception.

Even 111 STRs is highly questionable for prediction purposes. The Royal Stewarts have a good example of this. Maybe at 300 or 400 STRs, that's another thought, but we aren't there yet.

STRs have multiple uses so I would not totally characterize that SNPs are where it is at. Essentially, SNPs provide the framework for the tree, but we can not suppose everyone can afford Next Generation Sequencing (i.e. Big Y or FGC).
STR predictions can be a useful shortcut to the right SNPs to test for in the case NGS is not in the cards for an individual.

Likewise, even if NGS is in the cards ($) for an individual, we can't expect it to be in the cards for every potentially related individual. This is where a matching database becomes very important. NGS tests don't always test every individual for every region whereas STR panels as well as SNP packs and panels do reliably hit what they say they will. (barring beta runs.)

Another use of STRs is the last mile. They can be differentiators at the tips of the SNP branch twigs. Dr. Maurice Gleeson and others have shown us this. A change on the 111 STR panel happens about every 3 father-son transmissions. This is additive with what SNPs depict for branching.

I'm afraid we really need the 300-400 STRs though. Patterns of mutations are always better than a singular isolated mutation examination, which is what I did for the YCAII=19,23 versus 19,24 in Tometable's case.

TigerMW
07-12-2016, 04:29 AM
...
If he wants the answer to the question he actually asked in the thread OP, the answer is: Yes, one can usually predict a subclade below R1b-M269 with good confidence based on only 37 markers.
Your statement needs further qualification. What ages of haplogroup are you saying you can predict with only 37 STRs a good deal of the time with good confidence or whatever you are saying? If it is just "below M269" that is not much of a bar to reach and not practically doing better than FTDNA's conservative prediction.

FTDNA has already given us the R-M269 prediction. That's a real haplogroup with a TMRCA probably 8,000 years old or so, give or take a couple of thousand.

I went a little more aggressive and feel I can predict R-P311 with good confidence but that is not error free and as I said that is the sleeves out of my vest as P311 covers over half of Europe. If we had DYS492 value (in the 38-67 STR panel) we could probably cut this in half and predict P312 versus U106 but that is not error free either. We could be lucky though and match something like the L1065 Scots, L1066 Irish IV, Z209 North-Sourth folks, etc.

Even if we knew Tometable was P312+ versus U106+ we are still 4500-6000 years ago. How much does that help? These are massive haplogroups. SNP testing is warranted for sure, its just a question of which path to get there.

Táltos
07-12-2016, 04:59 AM
I would not have carried on this discussion further. But since you, as an administrator, have asked me specific questions, then I conclude that it will not be considered contentious or out of line for me to respond.

I can only guess what will satisfy Tomenable. But his specific questions in the original post were:

"Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?"

The answer to both these questions is yes. So that's the answer I gave.

37 markers will not allow one to identify their "terminal" SNP, as you point out. But this is not the question that Tomenable asked. 37 markers will almost always allow one to predict a subclade below R1b-M269 with good confidence. This is what FTDNA used to do, and this is what they stopped doing in July 2015.

I don't consider it a criticism against FTDNA if they made a decision based on profit motivations. It is simply a neutral matter of fact that for-profit corporations have profits as their primary goal in everything they do. I mentioned FTDNA's cessation of haplogroup prediction in my previous comment only as support for the argument that predictions CAN be made from 37 markers, and the decision to cease and negate such predictions below R1b-M269 was not due to a failure of the prediction methodology, but attributable instead to profit motives. As such, the decision to cease such predictions does not reflect negatively on the prediction methodology itself. The reason Tomenable's FTDNA account says nothing beyond "R1b-M269" is NOT that nothing more can be deduced. The reason is that FTDNA can make more money by not sharing what can be deduced.

If Tomenable wants to know his "terminal" SNP, he should do an SNP pack or NGS test. If he wants to identify his recent paternal relatives and estimate how closely related they are, he should upgrade to 67 or 111 markers. If he wants to identify his closest relatives, not only limited to the paternal branch, he should purchase an Ancestry DNA test.

If he wants the answer to the question he actually asked in the thread OP, the answer is: Yes, one can usually predict a subclade below R1b-M269 with good confidence based on only 37 markers.

You are right. He only asked for a predictor. Common sense says in the end he will want a more definitive answer.

However this thread should not deteriorate into a back and forth between yourself and Mike over FTDNA. This has happened in the past in numerous threads. It will not be tolerated from either side of the coin.

miiser
07-12-2016, 05:05 AM
Your statement needs further qualification. What ages of haplogroup are you saying you can predict with only 37 STRs a good deal of the time with good confidence or whatever you are saying?

I don't address the question of what age of haplogroups may be reliably predicted, because this is not a question that can be simply answered. Tomenable didn't ask about a specific time range of branching. He asked whether ANYTHING below R1b-M269 can be predicted, and this is the question I answered. How deep a prediction can reach, and how confidently, depends on the selectivity of the haplogroup - that is, how unique the haplotype is for a particular haplogroup, how heavily sampled the haplogroup is, and how closely one's haplotype matches the haplogroup. But, undoubtedly, the reach of the prediction is nearly always somewhat lower than R1b-M269 with a confidence over 95%.

Tomenable can gain useful information from haplogroup predictors, beyond what FTDNA has told him, using only his 37 marker test without any additional testing. This is my main point, a point which was neglected by most other posters in the thread.

Tomenable
07-12-2016, 07:15 AM
I've used Jim Cullen's Predictor and I'm getting mostly "R1b-Ub[iquitous]" and "R1b", but also these results:

R1b-C.Europe
R1b-S.Irish
R1b-IrishIII
R1b-Irish/Continental
R1b-Leinster(Irish)
R1b-North/South 2
R1b-U106(S21)->S26
R1b-U106->Scottish
R1b-S28(U152)

I noticed that each time I run it, slightly different percentages show up, so I ran it 15 times, the results:

(apart from percentages for "R1b-Ub" and for "R1b")

1) R1b-C.Europe =>11% R1b-S.Irish =>8% R1b-S26 =>8% R1b-S28 =>5% R1b-S21-Scottish =>3% R1b-Irish/Continental =>1%
2) R1b-S.Irish =>13% R1b-C.Europe =>8% R1b-S26 =>8% R1b-S28 =>2% R1b-Leinster =>1%
3) R1b-S26 =>10% R1b-C.Europe =>9% R1b-S.Irish =>6% R1b-IrishIII =>1% R1b-S28 =>1%
4) R1b-S.Irish =>12% R1b-S26 =>8% R1b-C.Europe =>7% R1b-S21-Scottish =>2% R1b-IrishIII =>2% R1b-North/South 2 =>2%
5) R1b-C.Europe =>15% R1b-S.Irish =>7% R1b-S26 =>7% R1b-S28 =>4% R1b-S21-Scottish =>1%
6) R1b-C.Europe =>8% R1b-S.Irish =>8% R1b-S26 =>8% R1b-S28 =>2%
7) R1b-C.Europe =>12% R1b-S.Irish =>11% R1b-S26 =>7% R1b-S21-Scottish =>1% R1b-S28 =>1%
8) R1b-C.Europe =>11% R1b-S.Irish =>9% R1b-S26 =>8% R1b-S28 =>4% R1b-Leinster =>3%
9) R1b-S.Irish =>11% R1b-C.Europe =>8% R1b-S28 =>3% R1b-S21-Scottish =>2% R1b-S26 =>2%
10) R1b-S26 =>9% R1b-S.Irish =>8% R1b-C.Europe =>7% R1b-S21-Scottish =>2% R1b-North/South 2 =>2% R1b-S28 =>2%
11) R1b-S.Irish =>8% R1b-C.Europe =>7% R1b-S26 =>6% R1b-S28 =>5% R1b-S21-Scottish =>1% R1b-IrishIII =>1% R1b-North/South 2 =>1%
12) R1b-C.Europe =>14% R1b-S.Irish =>7% R1b-S26 =>7% R1b-S28 =>2% R1b-S21-Scottish =>1%
13) R1b-S.Irish =>11% R1b-S26 =>7% R1b-C.Europe =>5% R1b-S28 =>4%
14) R1b-C.Europe =>9% R1b-S.Irish =>7% R1b-S26 =>5% R1b-IrishIII =>2% R1b-S28 =>2% R1b-S21-Scottish =>1%
15) R1b-C.Europe =>6% R1b-S.Irish =>6% R1b-S26 =>6% R1b-S28 =>3% R1b-IrishIII =>1%

Tomenable
07-12-2016, 07:17 AM
What subclade is "R1b-C.Europe" ??? I guess this category includes lineages from more than one subclade?

MacUalraig
07-12-2016, 07:49 AM
I upgraded Y12 markers to Y37 markers and have just received new STR results.

Do you know any predictor which will allow me to roughly predict my subclade?

I guess that something more than "R1b-M269" can be deduced from 37 markers?

If not then I will order SNP Pack.

If you want to know what SNPs you are positive for (aka your haplogroup), test SNPs.

Broadly speaking Chisholm is from the far north of Scotland and Oliver is an English name. There is a bit of overlap in their distribution in the Borders though. But no point in even worrying about such things unless you have proved the match is real which you can only do with SNPs.

Tomenable
07-12-2016, 07:53 AM
The problem is that those Chisholms/Olivers who match me on STRs, also did not test their SNPs.

Their haplogroup is just listed as "R1b-M269" as well. No terminal SNPs are given.

MacUalraig
07-12-2016, 08:21 AM
The problem is that those Chisholms/Olivers who match me on STRs, also did not test their SNPs.

Their haplogroup is just listed as "R1b-M269" as well. No terminal SNPs are given.

I know, but don't drop down to their level of technology otherwise we never make progress. Modernise your testing. Then you can do accurate matching with those who are enlightened enough to do likewise.

R.Rocca
07-12-2016, 12:00 PM
I've used Jim Cullen's Predictor and I'm getting mostly "R1b-Ub[iquitous]" and "R1b", but also these results:
...

So like I said, you could be anything. Join the ranks of the known, and do the SNP Pack! :biggrin1:

Joe B
07-12-2016, 04:27 PM
The problem is that those Chisholms/Olivers who match me on STRs, also did not test their SNPs.

Their haplogroup is just listed as "R1b-M269" as well. No terminal SNPs are given.
Time to be the leader. One way or another, find your haplogroup by SNP testing.

Pribislav
07-12-2016, 07:31 PM
I agree only SNP testing can give definitive answer, but I have run your STRs in NevGen out of curiosity, since R1b level was recently upgraded and now supports 168 subclades. The most probable result was DF27>ZZ12> ZZ39 , though I must add probability of unsupported subclade was very high (29.85%), but this is because this level of prediction requires 67 or more STRs for optimal results. The number in brackets shows percentage if supported clades account for 100%.

1. R1b DF27>ZZ12> ZZ39 51,74 % (73,8%)
2. R1b L21>DF13> FGC5494 4,04 % (5,8%)
3. R1b P312>DF99 3,28 % (4,7%)
4. R1b L21>FGC11134>> S1121 2,35 % (3,3%)
5. R1b U106>Z381> Z156>DF96 1,97 % (2,8%)
6. R1b L21>DF13> ZZ10>Z253> BY60 1,64 % (2,3%)
7. R1b L21>DF13> Z39589>DF49> ZP20 1,04 % (1,5%)

R.Rocca
07-12-2016, 07:50 PM
I agree only SNP testing can give definitive answer, but I have run your STRs in NevGen out of curiosity, since R1b level was recently upgraded and now supports 168 subclades. The most probable result was DF27>ZZ12> ZZ39 , though I must add probability of unsupported subclade was very high (29.85%), but this is because this level of prediction requires 67 or more STRs for optimal results. The number in brackets shows percentage if supported clades account for 100%.

1. R1b DF27>ZZ12> ZZ39 51,74 % (73,8%)
2. R1b L21>DF13> FGC5494 4,04 % (5,8%)
3. R1b P312>DF99 3,28 % (4,7%)
4. R1b L21>FGC11134>> S1121 2,35 % (3,3%)
5. R1b U106>Z381> Z156>DF96 1,97 % (2,8%)
6. R1b L21>DF13> ZZ10>Z253> BY60 1,64 % (2,3%)
7. R1b L21>DF13> Z39589>DF49> ZP20 1,04 % (1,5%)

Mine predicts me at 99.9% certainty... and it is wrong.

TigerMW
07-12-2016, 11:27 PM
The problem is that those Chisholms/Olivers who match me on STRs, also did not test their SNPs.

Their haplogroup is just listed as "R1b-M269" as well. No terminal SNPs are given.

Okay, I waved my magic wand and have a prediction. ;) This is just fun so consider it half-cocked but it does demonstrate how having more Y STRs can help with shortcuts in SNP testing. This does not mean I recommend against Next Gen Sequencing. That is the way to go for those who can afford it.

My half-cocked prediction is that you are S1026+ and possibly of this subclade:

R1b-L21>DF13>S1026>Z16886

Let me re-iterate this is truly half-cocked but if you went to 67 Y STRs I might be willing to put money on it (not the farm, though.)

There is an STR Signature that I track in the L21 project. I labeled as an z56513 variety before we discovered that it is under S1026. This is the signature or list of off-modals from the P312 and L21 modal haplotypes (which essentially are WAMH).

DYS391=10 (not 11)
DYS439=11 (not 10) frequently
DYS447=26 (not 25)
DYS449=32 (not 29)
YCAII=19,24 (not 19,23)
Low CDYs=36/37=37

DYS511=12 (not 10)
DYS565=13 (not 12)
DYS446=14 (not 13) frequently

The first five above are in 1-37 STRs and I think you match all but the pesky YCAII=19,24. That could make this half-cocked.

They last three STRs are in the 38-67 STR panel and are fairly slow moving making this is a pretty good pattern (signature). Of course we don't know if you have those which is why this is purely for fun as is my half-cocked prediction.

Here are folks that do have them that have tested out to 67 STRs. I don't see the kit #s in the matching myFTDNA screen for you but I think the Chisholm and Oliver you see are in this list below according to my three-way bank-shot approach here (... to be close to fellows who are close to a confirmed S1026+ guy.)

f249178 Chisholm
f362733 Chisholm
f190831 Hall (S1026+ confirmed)
f141532 Trimble
f104145 Oliver
f24686 Tolliver

These folks are all GD's of 5 at 67 STRs to each other. That's not bad and is why I think this is a good STR signature.

The good news is S1026 is in the R1b-M343 Backbone SNP Pack so if you go that route and my magic is working you are okay.

Menchaca
07-13-2016, 12:38 AM
According to yfull, S1026 is approximately 4,300 years old.

In my case, my haplogroup (FGC20764) is also around 4,000 years old and have just found a BigY match with whom I share 15 novel SNP's (Alex Williamson has not finished reviewing the othe guy's data). If we calculate 100 or 150 years per mutation that would bring us 1,500 to ~2,200 years closer to the present.

The thing is that the STR GD between us is around 25 at 67 markers. What Rachel Unkefer told me is that GD´s are probably not very meaningful beyond 1,000 years into the past.

What I am trying to say is that a GD of 5 at 67 STRs in the above case described by Mike may be an artifact.

Of course, I'm really new at this and probably don't know what I'm writing about.

Regards,

Menchaca

TigerMW
07-13-2016, 03:45 AM
According to yfull, S1026 is approximately 4,300 years old.

In my case, my haplogroup (FGC20764) is also around 4,000 years old and have just found a BigY match with whom I share 15 novel SNP's (Alex Williamson has not finished reviewing the othe guy's data). If we calculate 100 or 150 years per mutation that would bring us 1,500 to ~2,200 years closer to the present.

The thing is that the STR GD between us is around 25 at 67 markers. What Rachel Unkefer told me is that GD´s are probably not very meaningful beyond 1,000 years into the past.

What I am trying to say is that a GD of 5 at 67 STRs in the above case described by Mike may be an artifact.

Of course, I'm really new at this and probably don't know what I'm writing about.

Regards,

Menchaca
Yes, S1026 appears to be relatively old. There are other STR signatures in it and the GDs are much larger than just the handful for the subgroup I cited that Tomenable might be in. I actually think this group of DYS565=13 guys is Z16886+ which is younger. Still, we don't know how much and this is speculative anyway. The low GDs leads to the speculation that there is an undiscovered block of SNPs for S1026+ DYS565=13 people. This is why I like NGS discovery testing like Big Y. You need two people out of each STR cluster to do discovery testing.

BTW, the unproven periphery of this STR cluster includes a fellow at GD=10 from the others who is named Wahab. I don't know what that means but S1026 does not appear to be a tightly geographically knit group.

Tomenable
08-07-2016, 09:32 PM
Just ordered SNP Pack and also Family Finder. ;)

How long will it take until I get the results back?

lgmayka
08-08-2016, 01:23 AM
How long will it take until I get the results back?
This FTDNA page (https://www.familytreedna.com/learn/testing-process/return-kit-long-will-results-take/)gives typical test processing times:
Family Finder, 4-6 weeks
Big Y, 8-10 weeks
All others, 6-8 weeks

Of course, your mileage may vary. :)

Tomenable
08-30-2016, 09:02 PM
I just got my R1b SNP Pack results. Very interesting - it says that I am R1b-L617:

https://www.yfull.com/tree/R-L617/

If this is correct then I am under R1b-DF27 - not under R1b-L21 - after all.

But I remember that such a possibility was also considered in this thread. :)

Dewsloth
08-30-2016, 09:30 PM
Evidence is better than prediction. Congrats! :)

Tomenable
08-30-2016, 09:34 PM
Does it mean that Chisholms are likely also L617 (or at least DF27) ???

rms2
08-30-2016, 09:37 PM
I just got my R1b SNP Pack results. Very interesting - it says that I am R1b-L617:

https://www.yfull.com/tree/R-L617/

If this is correct then I am under R1b-DF27 - not under R1b-L21 - after all.

But I remember that such a possibility was also considered in this thread. :)

Congratulations!

Tomenable
08-30-2016, 09:53 PM
Apparently L617 can be found among about 2% of the Basques.

Are there any ethnic groups which have it at higher frequency?

TigerMW
08-30-2016, 09:59 PM
Just ordered SNP Pack and also Family Finder. ;)

How long will it take until I get the results back?

Congratulations.

Is this correct? You ordered the first week of August and you results are in the last week of August.

rms2
08-30-2016, 10:01 PM
Apparently L617 can be found among about 2% of the Basques.

Are there any ethnic groups which have it at higher frequency?

There is an L617 Project (https://www.familytreedna.com/public/R-L617?iframe=yresults). Have a look. You should join it.

Tomenable
08-30-2016, 10:36 PM
Is this correct? You ordered the first week of August and you results are in the last week of August.

Yes, very fast. :)

Results were "expected" after 21 September, but they always come sooner (I like it).


There is an L617 Project (https://www.familytreedna.com/public/R-L617?iframe=yresults). Have a look. You should join it.

Thanks, I've just sent a request to join.

Tomenable
08-30-2016, 10:52 PM
By the way:

My most distant known ancestor is from this area (https://en.wikipedia.org/wiki/Koźmin_Wielkopolski) in year 1832. I haven't been researching my genealogy very extensively, so I can probably easily get at least to the 1700s if I do some more research.

But I don't think that they lived in a very different area in the 1600s-1700s than in 1832.

And the surname is 100% Slavic Polish, and was like this already in the 1800s.

If I were descended from some Scottish immigrant from the 1600s, then we would expect a different surname (there are some Polonized surnames of Scottish origin). So this can be ruled out, I guess.

ArmandoR1b
08-31-2016, 02:07 AM
You should also take a look at the L617 portion of the BigTree (http://www.ytree.net/DisplayTree.php?blockID=618&star=false) but keep in mind that Europe outside of the British Isles is under-represented at FTDNA, YFull, and the BigTree.

Tomenable
08-31-2016, 06:00 AM
You should also take a look at the L617 portion of the BigTree (http://www.ytree.net/DisplayTree.php?blockID=618&star=false) but keep in mind that Europe outside of the British Isles is under-represented at FTDNA, YFull, and the BigTree.

Here is a map of distribution (my line is the one which goes back to year 1832 in Poland - in or around Koźmin Wlkp. (https://en.wikipedia.org/wiki/Ko%C5%BAmin_Wielkopolski); there is another guy whose line goes back to year 1855 in Lithuania - Šiauliai (https://en.wikipedia.org/wiki/%C5%A0iauliai); that guy had a different surname, but both surnames sound Polish/Slavic and both have "-ski" suffix). British customers are overrepresented, probably that's why most of people in L617 Project are from Britain - but within Britain itself, it seems to be concentrated in Southern England and also in Cornwall: https://s13.postimg.io/b2rmthmcn/L617_map.png

https://s13.postimg.io/b2rmthmcn/L617_map.png

Maybe L617 migrated from Eastern Europe to Western Europe in the Bronze Age, and "we" ("me" + "Sobolewski") stayed behind? Or maybe it came to Poland with Celts during the Iron Age? Anyway it is very interesting, I didn't expect such a rare subclade.

Tomenable
08-31-2016, 08:02 AM
1) Polish R1b according to Myres et al. 2010 (total: 18,35%):

http://www.nature.com/ejhg/journal/v19/n1/suppinfo/ejhg2010146s1.html

U106 - 5,94%

P312 - 5,47% - including:

P312>U152 - 3,47%
DF27 & P312* - 1,01%
P312>L21 - 0,99%

L51>L11* - 0,5%
L51(xL11) - 0,5%

L23(xL51) - 5,44%
M269(xL23) - 0,5%

2) Polish R1b according to Peter Gwozdz (total: 12,5%):

http://www.gwozdz.org/Results.html

U106 - 4,5% - including:

U106>L48>L47>"P Type" - 1,2%
U106>L48>L47>other - 0,4%
U106>L48>Z9 - 1,2%
U106>L48>other - 0,5%
U106(xL48) - 1,2%

P312 - 4,5% - including:

U152>L2 - 1,7%
U152(xL2) - 0,4%
DF27>Z196 - 1,2%
DF27(xZ196) - 0,1%
L21 & P312* - 1,1%

Z2103 - 2,2% - including:

Z2103>Y5587("EE Type")>BY593 - 1,5%
Z2103>Y5587("EE Type")>other - 0,5%
Z2103>all other subclades - 0,2%

M269>other - 1,1%
L754(xM269) - 0,2%

Tomenable
08-31-2016, 08:02 AM
So my subclade is less than 0,1% it seems. Because it is not under DF27>Z196, but DF27(xZ196).

MitchellSince1893
08-31-2016, 12:05 PM
Congrats!

razyn
08-31-2016, 12:41 PM
Congratulations -- I guess I know which kit you are, since I haven't logged in but one L617+ lately. You should also check out this thread: http://www.anthrogenica.com/showthread.php?4959-Z2552-(DF81-L617-YP4295-and-Z15001)

Looking at the current state of Alex's Big Tree for your parent group, of the three major subdivisions (currently identified) below Z2552, L617 looks less likely to be of Basque (or other Iberian) ancestry than the other two clades, YP4295 and DF81. http://www.ytree.net/DisplayTree.php?blockID=641&star=false

But it's early days, relatively speaking. In your case it may not be so much that a small percentage of the DF27 population is L617+, but more that a small percentage of the more eastern population from which you descend has been tested for anything. What we find is strongly affected by whom we test.

lgmayka
08-31-2016, 01:23 PM
I just got my R1b SNP Pack results. Very interesting - it says that I am R1b-L617:

https://www.yfull.com/tree/R-L617/
The other R-L617 in our project (https://www.familytreedna.com/public/polish?iframe=ysnp) is kit N17254. In his ancestry field, he lists Lithuanian geography but a Polish surname. He has taken the Big Y test. YFull lists him (https://yfull.com/tree/R-L617/) as R-L617*. In other words, his lineage diverged from the rest of R-L617 about 3800 years ago.

He is not on Ytree (http://ytree.net/DisplayTree.php?blockID=618&star=false).

Tomenable
08-31-2016, 01:32 PM
I will try to dig deeper in the family tree, at least into the 1700s. I got to 1832 only as a result of amateur research of parish records etc. which are available online. So I guess that a more professional kind of genealogical research could push my paper trail even far deeper into the past.

mouse
08-31-2016, 04:27 PM
Does it mean that Chisholms are likely also L617 (or at least DF27) ???

https://www.familytreedna.com/public/Chisholm/default.aspx?section=yresults

ADW_1981
08-31-2016, 04:27 PM
It looks like there is also a German L617 kit as well. If wonder if yourself, the German, and the Lithuanian Polish share a more recent ancestor than the other more western branches.

Tomenable
08-31-2016, 04:54 PM
My Y-chromosome ancestors were Roman Catholics (at least as far back as 1832):

Walery K. born in 1904 in Koźmin (https://en.wikipedia.org/wiki/Koźmin_Wielkopolski); married Jadwiga M. and they settled in nearby Dobrzyca (https://en.wikipedia.org/wiki/Dobrzyca)
parents: Franciscus K. and Ludovica B.

Franciscus K. born in 1871; married Ludovica B. in Roman Catholic parish of Koźmin in 1898
parents: Laurentius K. and Elisabeth G.

Laurentius K. born in 1832; married Elisabeth G. in Roman Catholic parish of Koźmin in 1857
parents: not listed by my online source

Now I have a few potential candidates for father of Laurentius, and I'm not yet 100% sure.

But I can easily determine this and then maybe I can get into the 1700s with my paper trail.

Tomenable
08-31-2016, 04:57 PM
As for the Sephardic Jewish hypothesis (from one of links posted by Razyn):

Sephardic Jews were expelled from Iberia around year 1500, IIRC. This hypothesis would require a Spanish Jew to come to Poland at some point between the 1500s and 1700s, convert to Roman Catholicism, and change his surname to a 100% Slavic Polish one without any signs of foreign origin - all before 1800.

I ordered "Family Finder" - if some Jewish admixture shows up, then I will consider this.

But I think that the presence of R-L617 in this region is older than 1500 AD.


https://www.familytreedna.com/public/Chisholm/default.aspx?section=yresults

I joined that Project too.

rms2
08-31-2016, 10:04 PM
Given the relative paucity of Poles and Lithuanians in FTDNA's database, I think it is probably significant that we already have two men of Lithuanian ancestry with Polish-looking surnames and one Polish man who are all R1b-L617. DF27 pops up in some odd places pretty far east, and I am the last guy who would resort to Iberia to account for it.

The various incarnations of the out-of-Iberia explanations for P312 and this or that subclade of P312 remind me of those Japanese beetle traps one sometimes sees in his neighbors' yards. You know the ones. They consist of an open bag on a little metal stand. There is some sort of sex lure in the bag that attracts the beetles, who fly in, get stuck, and cannot get out. The Iberian peninsula is like the bag and P312 the beetles. Some of us are like little children, who, upon seeing all the Japanese beetles in the bag, think that is where Japanese beetles come from.

11304

razyn
08-31-2016, 10:32 PM
Some of us are like little children, who, upon seeing all the Japanese beetles in the bag, think that is where Japanese beetles come from.
ROTFLMAO. And I rarely do that.

mouse
09-01-2016, 06:08 AM
As for the Sephardic Jewish hypothesis (from one of links posted by Razyn):

Sephardic Jews were expelled from Iberia around year 1500, IIRC. This hypothesis would require a Spanish Jew to come to Poland at some point between the 1500s and 1700s, convert to Roman Catholicism, and change his surname to a 100% Slavic Polish one without any signs of foreign origin - all before 1800.

I ordered "Family Finder" - if some Jewish admixture shows up, then I will consider this.

But I think that the presence of R-L617 in this region is older than 1500 AD.



I joined that Project too.

I don't think that you are related to any of the Chisolms. A lot of them are L21. Your haplotype is similar to theirs but you are DF27. You should have taken the Big Y test. If you want to make progress in future you will need take it and you spent money that could have went to your Big Y test. Do you think that you are descended from the Iberian Bell Beakers?

Tomenable
09-01-2016, 08:15 AM
Your haplotype is similar to theirs but you are DF27.

Actually it is puzzling because I have positive calls for P312 and L617, but not DF27.

My positive calls and DF27 call:

M343+, L278+, L389+, P297+, M269+, L23+, L51+, P310+, P311+, P312+, DF27 *, L617+

* No call or heterozygous call

However, DF27 is still listed in the SNP tree with golden font as "Presumed Positive".

mouse
09-01-2016, 09:03 AM
Actually it is puzzling because I have positive calls for P312 and L617, but not DF27.

My positive calls and DF27 call:

M343+, L278+, L389+, P297+, M269+, L23+, L51+, P310+, P311+, P312+, DF27 *, L617+

* No call or heterozygous call

However, DF27 is still listed in the SNP tree with golden font as "Presumed Positive".

If you are positive for L617 then you must be DF27+. The Big Y test would have given better resolution.

miiser
09-01-2016, 09:20 AM
Actually it is puzzling because I have positive calls for P312 and L617, but not DF27.

My positive calls and DF27 call:

M343+, L278+, L389+, P297+, M269+, L23+, L51+, P310+, P311+, P312+, DF27 *, L617+

* No call or heterozygous call

However, DF27 is still listed in the SNP tree with golden font as "Presumed Positive".

I wouldn't worry about it. There isn't always sufficient confidence to make a call for every tested SNP. This is just the nature of beast. Being positive for downstream SNPs makes it fairly certain that you are DF27.

lgmayka
09-01-2016, 01:37 PM
M343+, L278+, L389+, P297+, M269+, L23+, L51+, P310+, P311+, P312+, DF27 *, L617+

* No call or heterozygous call
DF27 is notoriously difficult to read correctly. At FTDNA, the individual DF27 SNP test costs $49 instead of $39 because of this extra difficulty/complexity.

My understanding is that if an R1b-M343 SNP Pack customer gets results that may indicate DF27*, FTDNA will run an individual DF27 SNP test if necessary to resolve the ambiguity. But that is not your case--your downstream positive SNP (L617+) makes clear that you indeed belong to the DF27 branch.

Tomenable
09-01-2016, 06:52 PM
It looks like there is also a German L617 kit as well. If wonder if yourself, the German, and the Lithuanian Polish share a more recent ancestor than the other more western branches.

This is a German-American. I checked the distribution of his surname (using http://www.verwandt.de/karten/), and I think that he could be originally from the region just to the north of Frankfurt am Main. There are also three (not two)* L617 from Lithuania, but only one with exact location given. I hastily made a map of modern distribution of L617. Cornwall seems to be one of areas of high concentration - even within Britain. Other such areas include the Basque Land (4 samples of L617 in a sample of 229 Basques) and Cambridgeshire. But let's keep in mind that customers of British origin are overrepresented on FTDNA:

https://s4.postimg.io/79wnm2ucd/L617_distribution.png

https://s4.postimg.io/79wnm2ucd/L617_distribution.png

*Surname of one of these guys looks very Polish, but surnames of the other two guys - not so much. These two other surnames are similar to each other, but one ends with suffix "-as" (which is typically Lithuanian), while the other one ends with suffix "-sky/ski" (which looks Polish, but considering that "core parts" of both surnames are similar - perhaps the one with "-sky" is a Polonised version of the one with "-as").

Tomenable
09-02-2016, 01:29 AM
Here are the main branches (it seems that the Eastern/Lithuanian branch is the oldest one, L617*):

On the other hand, it seems that all of British samples belong to a younger branch L617>FGC14951:

https://s22.postimg.io/5xgcz72f5/L617_branches.png

https://s22.postimg.io/5xgcz72f5/L617_branches.png

Isidro
09-02-2016, 02:32 AM
Congratulations Tomenable.

It is great to find out a finer branch, Mine sits in age 3,900 years ago, in between yours; 4,300-3800 years old.

DF27-Z195-Z198-CTS4188-BY3270-S11121.

R.Rocca
09-02-2016, 03:50 AM
By the way:

My most distant known ancestor is from this area (https://en.wikipedia.org/wiki/Koźmin_Wielkopolski) in year 1832. I haven't been researching my genealogy very extensively, so I can probably easily get at least to the 1700s if I do some more research.

But I don't think that they lived in a very different area in the 1600s-1700s than in 1832.

And the surname is 100% Slavic Polish, and was like this already in the 1800s.

If I were descended from some Scottish immigrant from the 1600s, then we would expect a different surname (there are some Polonized surnames of Scottish origin). So this can be ruled out, I guess.

Congratulations Tomenable! Some recommended reading if you haven't seen them yet :D
The Construction of Social Structure: Bell Beakers and Trzciniec Complex in North-Eastern Part of Central Europe
https://www.academia.edu/2022466/The_Construction_of_Social_Structure_Bell_Beakers_ and_Trzciniec_Complex_in_North-Eastern_Part_of_Central_Europe

Northern and Southern Bell Beakers in Poland
https://www.academia.edu/2022469/Northern_and_Southern_Bell_Beakers_in_Poland

Gesellschaftliche Strukturen der Glockenbecherkultur im Gebiet zwischen Weichsel und Oder
https://www.academia.edu/2022462/Gesellschaftliche_Strukturen_der_Glockenbecherkult ur_im_Gebiet_zwischen_Weichsel_und_Oder

Tomenable
09-03-2016, 06:10 PM
This FTDNA page (https://www.familytreedna.com/learn/testing-process/return-kit-long-will-results-take/)gives typical test processing times:
Family Finder, 4-6 weeks
Big Y, 8-10 weeks
All others, 6-8 weeks

Of course, your mileage may vary. :)

So SNP Pack normally takes 2 weeks longer to process than Family Finder.

However, I'm still waiting for my Family Finder, even though I already got my SNP Pack.

Note that both tests were batched on exactly the same day (10.08.2016).

Tomenable
09-06-2016, 11:56 AM
When it comes to R1b-L617, maybe the following kits:

N17254 + 435024 + 360008 + 450077 + 271705 + 368640 + 307927

Belong to the same branch which is yet to be discovered?

Finn
09-14-2021, 03:08 PM
I agree only SNP testing can give definitive answer, but I have run your STRs in NevGen out of curiosity, since R1b level was recently upgraded and now supports 168 subclades. The most probable result was DF27>ZZ12> ZZ39 , though I must add probability of unsupported subclade was very high (29.85%), but this is because this level of prediction requires 67 or more STRs for optimal results. The number in brackets shows percentage if supported clades account for 100%.

1. R1b DF27>ZZ12> ZZ39 51,74 % (73,8%)
2. R1b L21>DF13> FGC5494 4,04 % (5,8%)
3. R1b P312>DF99 3,28 % (4,7%)
4. R1b L21>FGC11134>> S1121 2,35 % (3,3%)
5. R1b U106>Z381> Z156>DF96 1,97 % (2,8%)
6. R1b L21>DF13> ZZ10>Z253> BY60 1,64 % (2,3%)
7. R1b L21>DF13> Z39589>DF49> ZP20 1,04 % (1,5%)


Pribislav could you please run the result of my nephew? He has done the 37 marker, same problem as Tomenable some years ago.

Jim Cullen gives R1b S28 (25%) rest of the R1b 1%.

May be you can figure out more?
https://i.postimg.cc/6yWpndk0/Schermafbeelding-2021-09-14-om-17-07-24.png (https://postimg.cc/6yWpndk0)

Finn
09-14-2021, 04:30 PM
I agree only SNP testing can give definitive answer, but I have run your STRs in NevGen out of curiosity, since R1b level was recently upgraded and now supports 168 subclades. The most probable result was DF27>ZZ12> ZZ39 , though I must add probability of unsupported subclade was very high (29.85%), but this is because this level of prediction requires 67 or more STRs for optimal results. The number in brackets shows percentage if supported clades account for 100%.

1. R1b DF27>ZZ12> ZZ39 51,74 % (73,8%)
2. R1b L21>DF13> FGC5494 4,04 % (5,8%)
3. R1b P312>DF99 3,28 % (4,7%)
4. R1b L21>FGC11134>> S1121 2,35 % (3,3%)
5. R1b U106>Z381> Z156>DF96 1,97 % (2,8%)
6. R1b L21>DF13> ZZ10>Z253> BY60 1,64 % (2,3%)
7. R1b L21>DF13> Z39589>DF49> ZP20 1,04 % (1,5%)

On second thought, his four most close matches on FTDNA three matches from Germany, one from Scotland. The Scottish match gives R1b L51, two Germans DF96 and one DF98.

RobertCasey
09-14-2021, 04:51 PM
Y37 rarely works very well for YSNP prediction. It works only around 10 % of the time if you are lucky and have more than usual mutations in markers Y1 to Y37.

YSTR matches (genetic matches) are even more error prone. At Y37, it is very common to have 50 to 90 % error rates. You need to upgrade to Y111 (Y67 would do - but FTDNA removed that upgrade option). Y111 is rarely needed for YSNP prediction (less than 5 % of the time). Occasionally, there is some significant lack of divergence of mutations that lowers accuracy of YSNP prediction to 95 % to 98 %, Y111 usually puts the accuracy up to 99 to 100 %.

YSNP prediction and charting are far more accurate than using the FTDNA YSTR match lists as raw genetic distance is just a very poor methodology for relatedness. Here is a link to why genetic distance is so error prone:

https://www.youtube.com/watch?v=DHqqt0_wEgg&t=319s

Finn
09-14-2021, 07:22 PM
Y37 rarely works very well for YSNP prediction. It works only around 10 % of the time if you are lucky and have more than usual mutations in markers Y1 to Y37.

YSTR matches (genetic matches) are even more error prone. At Y37, it is very common to have 50 to 90 % error rates. You need to upgrade to Y111 (Y67 would do - but FTDNA removed that upgrade option). Y111 is rarely needed for YSNP prediction (less than 5 % of the time). Occasionally, there is some significant lack of divergence of mutations that lowers accuracy of YSNP prediction to 95 % to 98 %, Y111 usually puts the accuracy up to 99 to 100 %.

YSNP prediction and charting are far more accurate than using the FTDNA YSTR match lists as raw genetic distance is just a very poor methodology for relatedness. Here is a link to why genetic distance is so error prone:

https://www.youtube.com/watch?v=DHqqt0_wEgg&t=319s

Yes I regret that I didn't took a higher marker for him.

See from his matches and looking at the charts (on FTDNA with the ranges) this is (preliminary) pointing at: DF96

Which is in the Netherlands not odd (DF96 is quite high here).

But ok I guess I have to upgrade to be sure.

Tomenable
09-14-2021, 07:38 PM
This is my terminal SNP as of today:

https://www.yfull.com/live/tree/R-FGC31068/

https://i.imgur.com/JLa32E7.png

Helves
09-14-2021, 07:56 PM
This is my terminal SNP as of today:

https://www.yfull.com/live/tree/R-FGC31068/

https://i.imgur.com/JLa32E7.png

Where are these new samples without a flag from?

Tomenable
09-14-2021, 07:58 PM
Where are these new samples without a flag from?

Also from England AFAIK. Garimund knows them.

Finn
10-03-2021, 07:15 PM
What an oddity and what is DNA therefore interesting!

I just received the result of my nephew.

I expected some kind of R1b U106, TOTALLY wrong.

It's a subclade of L23, specific R-PF7562 and definitive:R-BY164569

That's in the Armenian kind of range:
https://i.postimg.cc/k4K95cfz/Schermafbeelding-2021-10-03-om-20-56-10.png (https://postimg.cc/bDy4CQX9)

My hunch is that it has a connection with (an old branch of!) Yamna/ Steppe ancestry....but any suggestion is welcome!

uintah106
10-03-2021, 07:21 PM
I think it has become more and more
obvious P 312 is omnipresent.

R.Rocca
10-03-2021, 10:22 PM
What an oddity and what is DNA therefore interesting!

I just received the result of my nephew.

I expected some kind of R1b U106, TOTALLY wrong.

It's a subclade of L23, specific R-PF7562 and definitive:R-BY164569

That's in the Armenian kind of range:
https://i.postimg.cc/k4K95cfz/Schermafbeelding-2021-10-03-om-20-56-10.png (https://postimg.cc/bDy4CQX9)

My hunch is that it has a connection with (an old branch of!) Yamna/ Steppe ancestry....but any suggestion is welcome!

Finn, good to see he got his results, but please know that R-PF7562 is not below L23 and therefore has no known Yamnaya link (that I know of). In old studies, M269(xL23), which is likely R-PF7562, is most frequent in the modern day Balkans and somewhat rare elsewhere.

Finn
10-04-2021, 05:56 AM
Finn, good to see he got his results, but please know that R-PF7562 is not below L23 and therefore has no known Yamnaya link (that I know of). In old studies, M269(xL23), which is likely R-PF7562, is most frequent in the modern day Balkans and somewhat rare elsewhere.

Thanks for the comment Rocca, appreciated!

That's fine tuned! After research: it's indeed a sibling branch of L23 but it's not the Balkan line (that's below R-PF7563) it's (nowadays) a Caucasus line (that's below R-PF7562).

So I think it has still a connection with Yamna history....(to elaborate).

https://i.postimg.cc/hjkC6Yhf/Schermafbeelding-2021-10-04-om-07-43-58.png (https://postimg.cc/MMYbjd7S)

https://i.postimg.cc/k5kFgbCH/Schermafbeelding-2021-10-04-om-07-44-24.png (https://postimg.cc/DWPbCm8Q)

wiki:

Five individuals out of 110 tested in the Ararat Valley of Armenia belonged to R-M269(xL23) and 36 to R-L23*, with none belonging to known subclades of L23.[68]

Standardized Ape
10-04-2021, 07:13 AM
Thanks for the comment Rocca, appreciated!

That's fine tuned! After research: it's indeed a sibling branch of L23 but it's not the Balkan line (that's below R-PF7563) it's (noowadays) a Caucasus line (that's below R-PF7562).


I think an Alan link might be more likely here than old Yamnaya as it does not look 5000 years old. Alans at least in the Caucasus were quite rich in Caucasian ancestry and they certainly had more interaction with West Germanic speakers than most Caucasians.

Finn
10-04-2021, 08:21 AM
I think an Alan link might be more likely here than old Yamnaya as it does not look 5000 years old. Alans at least in the Caucasus were quite rich in Caucasian ancestry and they certainly had more interaction with West Germanic speakers than most Caucasians.

Thanks Standardizd Ape, can make some sense! "neo-steppe folk" ;)

https://en.wikipedia.org/wiki/Alans

Finn
10-04-2021, 08:56 AM
I think an Alan link might be more likely here than old Yamnaya as it does not look 5000 years old. Alans at least in the Caucasus were quite rich in Caucasian ancestry and they certainly had more interaction with West Germanic speakers than most Caucasians.

My mother reacted immediately that my great-grandfather sometimes went to a regional market at a distance from home, when he went home he stole a horse, rode home bareback, where he released the horse .... some traditions never disappear; )

CopperAxe
10-04-2021, 09:47 AM
Thanks for the comment Rocca, appreciated!

That's fine tuned! After research: it's indeed a sibling branch of L23 but it's not the Balkan line (that's below R-PF7563) it's (noowadays) a Caucasus line (that's below R-PF7562).

So I think it has still a connection with Yamna history....(to elaborate).

https://i.postimg.cc/hjkC6Yhf/Schermafbeelding-2021-10-04-om-07-43-58.png (https://postimg.cc/MMYbjd7S)

https://i.postimg.cc/k5kFgbCH/Schermafbeelding-2021-10-04-om-07-44-24.png (https://postimg.cc/DWPbCm8Q)

wiki:

If your nephew is up for it maybe consider contacting yfull. Having an estimation of the mrca between him and the Armenians with the same Y-dna allfather could be helpful in determining its origin and spread.

Finn
10-04-2021, 10:49 AM
If your nephew is up for it maybe consider contacting yfull. Having an estimation of the mrca between him and the Armenians with the same Y-dna allfather could be helpful in determining its origin and spread.

Yes of course, would mean taking a big Y I guess, and this would most probably change this tree somewhat. It will be the next step...

Pylsteen
10-04-2021, 07:01 PM
What an oddity and what is DNA therefore interesting!



Interesting and surprising; Is this a lineage from Drenthe?

Finn
10-04-2021, 08:06 PM
Interesting and surprising; Is this a lineage from Drenthe?

Yes indeed..pointing at the Alans. Although I'm still wondering.

And it's from Drenthe (my mothers' side). Unfortunately not traceable beyond 1820 when a boy was born in De Haar/ Sleen (near Emmen) with a NN father.

JonikW
10-04-2021, 09:12 PM
Yes indeed..pointing at the Alans. Although I'm still wondering.

And it's from Drenthe (my mothers' side). Unfortunately not traceable beyond 1820 when a boy was born in De Haar/ Sleen (near Emmen) with a NN father.

Fascinating stuff Finn. Beyond the U106, your family seems to have a talent for atypical Y results.;) I like the Alans theory. This is interesting from Wikipedia too, although I haven't verified its accuracy:

"Armenian and Dutch interactions are believed to have started in the 13th and 14th centuries, when Dutch merchants arrived in Cilicia and Armenian trading houses opened in the Low Countries. Armenians brought into the Low Countries carpets, dyes, cotton, and spices from Armenia and from around the world.

"Apart from the contemporary Armenian community spread out over the Netherlands, there had been an independent Armenian community concentrated in Amsterdam during the 17th and 18th centuries."

Finn
10-05-2021, 09:52 AM
Fascinating stuff Finn. Beyond the U106, your family seems to have a talent for atypical Y results.;) I like the Alans theory. This is interesting from Wikipedia too, although I haven't verified its accuracy:

"Armenian and Dutch interactions are believed to have started in the 13th and 14th centuries, when Dutch merchants arrived in Cilicia and Armenian trading houses opened in the Low Countries. Armenians brought into the Low Countries carpets, dyes, cotton, and spices from Armenia and from around the world.

"Apart from the contemporary Armenian community spread out over the Netherlands, there had been an independent Armenian community concentrated in Amsterdam during the 17th and 18th centuries."

Ai Aii:frusty: read better Finn....I confused downstream with a result......mmmm I'm going to shame somewhere in a corner.

(The endeffect somewhere in R1BU106??????:bump:

Finn
10-05-2021, 10:42 AM
And indeed first confirmation of Iain Mc Donald somewhere beneath FF96, more details follow.

anglesqueville
10-05-2021, 02:09 PM
Fascinating stuff Finn. Beyond the U106, your family seems to have a talent for atypical Y results.


I am jealous, my own Y is hopelessly boring. 46933

I await the results of my BigY 700 hoping for some improbable surprise, something that will add a little spice, a touch of a dream. I don't expect anything too crispy. In fact I'm afraid I broke my piggy bank just to be nice to an Anglo-American family.

Finn
10-05-2021, 08:11 PM
I am jealous, my own Y is hopelessly boring. 46933

I await the results of my BigY 700 hoping for some improbable surprise, something that will add a little spice, a touch of a dream. I don't expect anything too crispy. In fact I'm afraid I broke my piggy bank just to be nice to an Anglo-American family.

I just got the result from Iain DF96/R-S15585 level. Somewhere between a Finnish and a Schleswig family :behindsofa:

Bollox79
10-05-2021, 09:39 PM
I just got the result from Iain DF96/R-S15585 level. Somewhere between a Finnish and a Schleswig family :behindsofa:

So he's under DF96?

Finn
10-06-2021, 01:17 AM
So he's under DF96?

Yep definitely! And 'final' R-S15585 (1500-2500 ybp).

And pretty close to the Von Kroghs (Jürgen von Kroge Lübeck 1575). This subclade is most probably an offshoot of the Germanic migration times (fifth/sixth century).

It doesn't need much imagination to see the Von Kroghs (Adel) as heirs of the old "Anglo-Saxon" elite.

Schleswig-Holstein was the epicenter of a stream that not only had an influx to England but also had an impact on the North Dutch area.....

Finn
10-09-2021, 06:44 PM
Fascinating stuff Finn. Beyond the U106, your family seems to have a talent for atypical Y results.;) I like the Alans theory. This is interesting from Wikipedia too, although I haven't verified its accuracy:

"Armenian and Dutch interactions are believed to have started in the 13th and 14th centuries, when Dutch merchants arrived in Cilicia and Armenian trading houses opened in the Low Countries. Armenians brought intBy comparison to our other testers, I can securely put you into R-S15585, but I can't securely place you any further down the tree than that. You share some additional mutations with the R-S15585>FT419563>FGC21153 family (von Krogh) but not all the ones I would expect, so either you have some convergent mutations and are only related to them at the R-S15585 level, or you are in a new haplogroup between R-FT419563 and R-FGC21153, and have some back mutations that hide your true nature.
o the Low Countries carpets, dyes, cotton, and spices from Armenia and from around the world.

"Apart from the contemporary Armenian community spread out over the Netherlands, there had been an independent Armenian community concentrated in Amsterdam during the 17th and 18th centuries."

In the 'end' ;)

Of course I consulted the "R1b U106 crack" Iain Mc Donald. With many thanks to this man!

This is what he told me:

By comparison to our other testers, I can securely put you into R-S15585, but I can't securely place you any further down the tree than that. You share some additional mutations with the R-S15585>FT419563>FGC21153 family (von Krogh) but not all the ones I would expect, so either you have some convergent mutations and are only related to them at the R-S15585 level, or you are in a new haplogroup between R-FT419563 and R-FGC21153, and have some back mutations that hide your true nature.

R-S15585 is probably between 1500 and 2500 years old, and R-FT419563 is only a few generations younger. R-FGC21153 is private to the von Krogh family, and dates to some time after their stated common ancestor in 1580 AD.


The grandfather's line from my mother's side is called Grelling and comes from Drenthe in the Northern Netherlands. The bottleneck of this family is in the 19th century. The ancestor is Ernst Grelder born on August 30, 1820. He passed away on January 21, 1893, his name was so far 'Saxonized' to Grelling (-ing is son of). However, his father is NN!

The Dutch archive is now well digitized. We come across 14 hits with the family name Grelder throughout the Netherlands. 13 of these can be related to Ernst Grelder or his descendants.

All but one, namely Pieter Grelder. In Amsterdam on 30-9-1774 a Pieter Grelder married Grietje Rijnoldus. He is from Bremen and she from Groningen. Furthermore, the surname Grelder appears on a list of surnames in the former Swedish part of Vorpommern (Northeast Germany) in Stralsund of the 17th century. The Von Krogh family related via Y-DNA came from Lübeck. All three cities of Lübeck, Stralsund and Bremen are among the oldest Hanseatic cities, which were closely allied to each other. And meant also a circulation of people. Since the Grelders share the same ancestor with the Von Kroghs, they probably come from the original Hanseatic circuit. It is plausible that the NN father of Ernst Grelder comes from Bremen or Stralsund.

An alternative scenario is an Anglo-Saxon link. The Von Kroghs are a noble family of Lübeck origin. This is also the area from which the Anglo-Saxon spread, including to the Northern Netherlands in the fifth and sixth centuries.

However, I consider this scenario less likely because the Grelder family was clearly not indigenous. And the Anglo-Saxon genes were distributed in the Northern Netherlands about 1500 years before the birth of Ernst Grelder.

Ernst Grelder was born near the garrison town of Coevorden, so a link with a soldier, probably from a German Hanseatic city, is therefore plausible.

Ultimately proof is difficult, I think it will remain an assumption, an anonymous father never leaves much to 'paper' (no notarial deeds, etc.).

Nevertheless, this Y-DNA result has already created a little more clarity! (Still need the big Y).

For a 'broader' analysis see this:
https://anthrogenica.com/showthread.php?24951-Z156-(DF96-98)-the-missing-Bronze-Age-Link!&p=807029#post807029

Dewsloth
10-09-2021, 08:12 PM
R-S15585 is probably between 1500 and 2500 years old, and R-FT419563 is only a few generations younger

How many SNPs identified (public and/or private) below S15585?

Finn
10-10-2021, 12:53 AM
How many SNPs identified (public and/or private) below S15585?

No idea yet, may be I have to take the Big Y for that.

Dewsloth
10-10-2021, 08:26 PM
R-S15585 is probably between 1500 and 2500 years old, and R-FT419563 is only a few generations younger

Blocktree has it at about 28 SNPs, so the above is using ~53 to 89 years per SNP. FT419563 is about 19 SNPs.

Edit:

Or, going off the bottom of the block that includes S15585 (so more TMRCA than formation) it's 20 SNPs vs the 19 SNPs for FT419563 (which would be consistent with the "few generations younger."

But that makes the range for the estimates above 75 to 125 years per SNP.

FWIW, Yfull has the S15585 block formed 2100 ybp, TMRCA 1200 ybp.