U106 'the wander years'

[Mikewww]

"The variance of U106 in Ireland/Scotland/Wales" is a meaningless figure, as far as I'm concerned !!

Any statistic that is calculated with a low or not representative sample is probably meaningless. The problem in this recent set of numbers may not be related to the multiple expansions/exchanges but rather to the sampling. I'm just saying don't throw the baby out with the bath water, at least yet.

I'm sure these questions are solvable, but not in this way. The Genographic Project once showed diagrams on their web site describing M343 entering Europe. I haven't heard anybody describing things in those terms ever since. Now people are describing migrations in terms of U106. That's not going to work either. The granularity is wrong.

I don't think your example is applicable. We talk about the expansion of R1b into Europe all of the time and R1b is marked by M343. We are just refining the early levels of branching in our discussions looking at L23+ or L23-. There are clearly multiple subclades (some to be discovered) on both sides of the L23+ and L23-. That doesn't mean that branching isn't important. It may be critical, or it may be at L51+ and L51-, but still this is backup the phylogenetic tree not at the lowest level.

This is what I meant by no silver bullet. I doubt if there is any "right level" of granularity. All of this stuff has to be analyzed.  I'm in agreement that analysis has to be done at the lowest level possible, but other levels shouldn't be ignored, particularly the comparisons in the early (older) branching. I'm not saying our current statistics handle this. They don't, but progress is being made.  I suppose the next step would be a multi-interclade calculation with three or more clades. Well, that's way beyond me, but stuff like carbon-14 dating took awhile to gain precision. This is just science. It cranks away slowly to improve on concepts and methods and hopefully invent a few new ones.

I re-iterate that I am not saying to rely on any one thing, stand-alone. Look at everything in context.

[Peter M]

Peter, to your point and as I've mentioned before, some variance numbers are skewed because of a greater or lesser number of migrations into and out of a region that may or may not have occurred!

Well, in the case of the British Isles and U106, my guess is , they are skewed to the extent of being completely unuseable. $:-) At least that's my opinion.

To that point, Italy seems to not have been impacted by a major L48+ migration because unlike in most areas of Europe, L48 is much less common. We tested the following L48- kits a few months ago and all are now U106+ L48- Z156-   

179540   Giacomo Zeni, b. 1630 Verona, Italy
141915   Berardo Cesaroni, b.c. 1575, Cartoceto, Pesaro and Urbina, Italy
N12646   Veturio Cesaroni, b.1880, Terni, Umbria, Italy

One point in time, I thought one of the Cesaroni's might actually be Z18+ and on my suggestion he ordered a test. After all, there are strong links between southern Germany and northern Italy (more than both just liking fast cars). Unfortunately he turned out Z18-.

Italy is an interesting case that illustrates my point rather nicely. L48 is another U106 sub-clade but has a distribution that's different from the others. It's a good candidate for further serious research, I guess.

[Peter M]

Any statistic that is calculated with a low or not representative sample is probably meaningless. The problem in this recent set of numbers may not be related to the multiple expansions/exchanges but rather to the sampling. I'm just saying don't throw the baby out with the bath water, at least yet.

Well, about the use of variance in the study of the migration (of U106) to the British Isles we apparently see things differently. You say it's a matter of having more data and I think it's not possible in principle with any amount of data to say anything on the level of U106.

I'm sure these questions are solvable, but not in this way. The Genographic Project once showed diagrams on their web site describing M343 entering Europe. I haven't heard anybody describing things in those terms ever since. Now people are describing migrations in terms of U106. That's not going to work either. The granularity is wrong.

I don't think your example is applicable. We talk about the expansion of R1b into Europe all of the time and R1b is marked by M343.

My impression is, just as with U106 to some extent, there have been multiple cases of migration of (groups of people containing) R1b into Europe. Just as with U106, I therefore don't think it possible to study the "migration of R1b into Europe". But I must say, I have spent more time looking at U106 than at R1b. It's just an impression (!!) based on U106- (minus) finds.

[Mark Jost]

My impression is, just as with U106 to some extent, there have been multiple cases of migration of (groups of people containing) R1b into Europe. Just as with U106, I therefore don't think it possible to study the "migration of R1b into Europe". But I must say, I have spent more time looking at U106 than at R1b. It's just an impression (!!) based on U106- (minus) finds.

Peter,

It would be interesting if maybe you or someone else can take all of the tested HTs in U106 and using the TMRCA estimator, that can slice and dice, and compile a list of Founders and the next higher Subclade Founder ages and then calculate the Interclade for the two and see what the results are and, where the majority of those are located. If the subclade has several regions, one can split those groups out. Even by Allele value if there is a nice split.

This is alot of work but would benefit everyone.

I, when I have some time, will do this from my DF13xSubclade Guys.

MJost

[stoneman]

The first U106 people arrived in Ireland 6000 ybp.

If an SNP was an infection where would most of the people be infected? Mikewww would probably think 8000 miles from the source. Do you think that one man with the defective gene known as M222  moved to the North West of Ireland and spead it to most of the population? I dont believe it.

[Peter M]

The first U106 people arrived in Ireland 6000 ybp.

If an SNP was an infection where would most of the people be infected? Mikewww would probably think 8000 miles from the source. Do you think that one man with the defective gene known as M222  moved to the North West of Ireland and spead it to most of the population? I dont believe it.

Possible, but as long as most Irish who are (or seem to be) U106, are not interested in any further investigation of their background, this statement will be hard to prove.

[Peter M]

My impression is, just as with U106 to some extent, there have been multiple cases of migration of (groups of people containing) R1b into Europe. Just as with U106, I therefore don't think it possible to study the "migration of R1b into Europe". But I must say, I have spent more time looking at U106 than at R1b. It's just an impression (!!) based on U106- (minus) finds.

Peter,

It would be interesting if maybe you or someone else can take all of the tested HTs in U106 and using the TMRCA estimator, that can slice and dice, and compile a list of Founders and the next higher Subclade Founder ages and then calculate the Interclade for the two and see what the results are and, where the majority of those are located. If the subclade has several regions, one can split those groups out. Even by Allele value if there is a nice split.

This is alot of work but would benefit everyone.

I, when I have some time, will do this from my DF13xSubclade Guys.

MJost

Mark, forgive my ignorance, but what is "the TMRCA estimator, that can slice and dice" ? Is this a certain program or algorithm or whatever ?? (don't get me wrong, this IS a serious question.)

Please keep in mind, I'm pretty sure of what I'm saying. I do NOT believe any results will be achieved by looking at U106 AS A WHOLE. I've tried to make a tree of a huge number of U106+ profiles using my own software in the past and that didn't work. The thing is too diverse to be handled as a whole. What needs to be done is to get far more people testing, preferably a few, levels of SNPs below U106, so that groups can be formed of much smaller subtrees for further investigation. Unfortunately, the SNPs below U106 tend to be a bit hard to predict; actual testing is needed. (Last week I made a Result Sheet of U106 after not having looked at tests there for a little under a year and was absolutely SHOCKED by the total lack of progress in that area; the undertested people that were there a year ago, still are there.)

[Peter M]

Somehow or another, a specific problem to resolve is how did U106's and P312's ancestral lineages come together in time and geography to the single L11* man that was their most recent common ancestor. That will line up with one more of these prehistoric cultures (or components of them) and will tell us today's majority paternal lineages populated Europe. We have enough data to say this probably happened relatively quickly.

This is a nice issue. How many reasonable L11 profiles do we have and do these profiles have a geographic origin specified with FT-DNA ??

[Mark Jost]

Mark, forgive my ignorance, but what is "the TMRCA estimator, that can slice and dice" ? Is this a certain program or algorithm or whatever ?? (don't get me wrong, this IS a serious question.)

Here is the spreadsheet that I was alluding to.

https://docs.google.com/file/d/0By9Y3jb2fORNX2FjVV85bmVtWWs/edit

Just enter your haplotypes in CladeA and use the copy A to B button. This allows you to set up Interclade calculations.

MJost

[Peter M]

Mark, forgive my ignorance, but what is "the TMRCA estimator, that can slice and dice" ? Is this a certain program or algorithm or whatever ?? (don't get me wrong, this IS a serious question.)

Here is the spreadsheet that I was alluding to.

https://docs.google.com/file/d/0By9Y3jb2fORNX2FjVV85bmVtWWs/edit

Just enter your haplotypes in CladeA and use the copy A to B button. This allows you to set up Interclade calculations.

MJost

Thanks Mark. Essentially, this appears to be (based on) Ken's spreadsheet for doing interclade calculations. What does it do to, or how would you propose to, form U106 subgroups (using this spreadsheet) ??

BTW, apparently there are more versions of Ken's sheet. I have, together with Vince T., made an implementation in PHP (sort of "C") of generation5 that connects to a profile database of sorts, so as to overcome the problem of copying/pasting sets of profiles. I personally don't consider "just enter your haplotypes" as a simple operation, especially because of the problem of keeping things in sync with new test data coming in. This implementation is currently being used for the calculation of the TMRCA's in the results sheets on L257.org (a work in progress, there are a few issues !!!). I'll have a more close look at your version.

[Mikewww]

Somehow or another, a specific problem to resolve is how did U106's and P312's ancestral lineages come together in time and geography to the single L11* man that was their most recent common ancestor. That will line up with one more of these prehistoric cultures (or components of them) and will tell us today's majority paternal lineages populated Europe. We have enough data to say this probably happened relatively quickly.

This is a nice issue. How many reasonable L11 profiles do we have and do these profiles have a geographic origin specified with FT-DNA ??

I keep track of all I can find in public projects. I just wish everyone would join their major haplogroup projects.  Here is the count I currently have L11 family SNP confirmed folks.

4080 - L21
990 - U152
601 - DF27
958 - P312xL21xU152xDF27 (many of these could be DF27 with a few L21)
2148 - U106
61 - L11xP312xU106

About 80% of the above are 67 STRs in length.  Most have the Ysearch IDs, where I (or Vince T) could find them. On Old World MDKAs I cross-checked Ysearch and FTDNA and classified everyone as best I could to the County/Shire/Department level. We have almost 2000 111 STR haplotypes.

Mark J has a system of automatically copying these haplotypes out of the spreadsheets into a Ken Nordtvedt based TMRCA estimator so he can do estimates fairly quickly. Other than cleaning up/extracting surnames and geographical classifications I don't enter the data manually either.  I've got a copy/paste/transform system that helps detect STR upgrades and new SNP results. The haplogroups are generated (based on the Clade tables in other threads here) from the SNP results so I don't care that FTDNA's haplotree is way behind. I just need to see the Y DNA SNP reports turned on.

[Mark Jost]

Thanks Mark. Essentially, this appears to be (based on) Ken's spreadsheet for doing interclade calculations. What does it do to, or how would you propose to, form U106 subgroups (using this spreadsheet) ??

BTW, apparently there are more versions of Ken's sheet. I have, together with Vince T., made an implementation in PHP (sort of "C") of generation5 that connects to a profile database of sorts, so as to overcome the problem of copying/pasting sets of profiles. I personally don't consider "just enter your haplotypes" as a simple operation, especially because of the problem of keeping things in sync with new test data coming in. This implementation is currently being used for the calculation of the TMRCA's in the results sheets on L257.org (a work in progress, there are a few issues !!!). I'll have a more close look at your version.

PeterM,

Ok, I just loaded all the 67marker Hts from MikeW's latest U106, cleaned the Nuls and zero to empty cells, and loaded CladeA and B with the same set. Available to anyone.

https://docs.google.com/open?id=0By9Y3jb2fORNNF9ma2JfZEtKUlk

MJost

[Peter M]

I keep track of all I can find in public projects. I just wish everyone would join their major haplogroup projects.  Here is the count I currently have L11 family SNP confirmed folks.

4080 - L21
990 - U152
601 - DF27
958 - P312xL21xU152xDF27 (many of these could be DF27 with a few L21)
2148 - U106
61 - L11xP312xU106

About 80% of the above are 67 STRs in length.  Most have the Ysearch IDs, where I (or Vince T) could find them. On Old World MDKAs I cross-checked Ysearch and FTDNA and classified everyone as best I could to the County/Shire/Department level. We have almost 2000 111 STR haplotypes.

I guess, it would be nice to put all L11xP312xU106 on a geographical map to see what that would show. Let's start by restricting ourselves to (a) known and (b) non-British origins. How many tested L11* profile with a known origin do we have then ? That's those with coordinates specified PLUS those for which you would be able to specify/estimate a map location on the basis of "other" information.

BTW, I'm talking about Google Map location here; not just countries.

Mark J has a system of automatically copying these haplotypes out of the spreadsheets into a Ken Nordtvedt based TMRCA estimator so he can do estimates fairly quickly. Other than cleaning up/extracting surnames and geographical classifications I don't enter the data manually either.  I've got a copy/paste/transform system that helps detect STR upgrades and new SNP results. The haplogroups are generated (based on the Clade tables in other threads here) from the SNP results so I don't care that FTDNA's haplotree is way behind. I just need to see the Y DNA SNP reports turned on.

I always find it amazing to hear what things people have made for themselves. Think of the things we could have done if we could get some sort of co-operation going. ..... $:-)

[Peter M]

PeterM,

Ok, I just loaded all the 67marker Hts from MikeW's latest U106, cleaned the Nuls and zero to empty cells, and loaded CladeA and B with the same set. Available to anyone.

https://docs.google.com/open?id=0By9Y3jb2fORNNF9ma2JfZEtKUlk

MJost

Thanks, Mark, I will have a look.

[wing_genealogist]

...


Ok, I just loaded all the 67marker Hts from MikeW's latest U106, cleaned the Nuls and zero to empty cells, and loaded CladeA and B with the same set. Available to anyone.

https://docs.google.com/open?id=0By9Y3jb2fORNNF9ma2JfZEtKUlk

MJost

Thanks so VERY much for all of your continued work on this project. It is VERY much appreciated.

[alan trowel hands.]

The first U106 people arrived in Ireland 6000 ybp.

If an SNP was an infection where would most of the people be infected? Mikewww would probably think 8000 miles from the source. Do you think that one man with the defective gene known as M222  moved to the North West of Ireland and spead it to most of the population? I dont believe it.

Based on what? 

[Mikewww]

I guess, it would be nice to put all L11xP312xU106 on a geographical map to see what that would show. Let's start by restricting ourselves to (a) known and (b) non-British origins. How many tested L11* profile with a known origin do we have then ? That's those with coordinates specified PLUS those for which you would be able to specify/estimate a map location on the basis of "other" information.

BTW, I'm talking about Google Map location here; not just countries.

The best I can define things to is there in the spreadsheets. I don't have time to put them on Google maps, but if you know the geography its not hard to look at it.  We don't have that many L11* guys. I always say they are lightly scattered, but I guess that is just my characterization.

Remember, this is not a scientific sample and is probably biased towards American immigration sources.

fN9165   Manuchariants   R-L11*   Armenia, Yerevan (Armenian project)
f142633   Kiss   R-L11*   Hungary
fN39211   Klassen   R-L11*   Poland
f123914   Powers   R-L11*   Poland
fE4034   Nowak   R-L11*   Poland, Greater Poland Voivodeship, Poznań
f119582   Gronkiewicz   R-L11*   Poland, Subcarpathian Voivodeship, Jaslo Co., Jaslo
fN49180   D'Ambrosia   R-L11*   Italy, Campania, Salerno
f173307   Mülli   R-L11*   Switzerland, Zurich, Dielsdorf, Schöfflisdorf
fE8946   Escobar   R-L11*   France, Aquitaine, Pyrénées-Atlantiques, Biarritz
f67632   Demarcus   R-L11*   France
f71997   Marcotte   R-L11*   zzzUnkOrigin
f69898   Dick   R-L11*   Germany, Thuringia, Lerchenhügel
f110095   Ohlhaeuser   R-L11*   Germany, Baden-Württemberg
f151345   Merkel   R-L11*   Germany, Baden-Württemberg, Stuttgart, Schlierbach
f23858   Mayer   R-L11*   Germany, Bavaria, Middle Franconia, Dinkelsbühl
fN2197   Coontz   R-L11*   Germany
f103102   Jacob   R-L11*   Germany
f180962   Hulegaard   R-L11*   Denmark
fN70876   Madsen   R-L11*   Denmark, Viborg Amt, Mønsted
f159936   Edison   R-L11*   Netherlands
f128054   Van Der Meij   R-L11*   Netherlands
fN89073   Van De Putte   R-L11*   Netherlands, Zeeland, Retranchement
f142454   Jobse   R-L11*   Netherlands, Zuid-Holland, Ouddorp
f38215   Fox   R-L11*   England, East, Essex Co.
f120398   Swindell   R-L11*   England, East Midlands, Derbyshire, Derby
yUHFMG   Leigh   R-L11*   England, Nort West, Cheshire, Stockport, Cheadle
f154889   Fish   R-L11*   England, South East, Berkshire?, Windsor
f136502   Vawden   R-L11*   England, South West, Devonshire, Roborough
f169774   Adams   R-L11*   England, South West, Somersetshire, Barton St. David
f64311   Mason   R-L11*   England, West Midlands, Worcestershire, Pershore
fN14767   Baker   R-L11*   England
f41275   Bates   R-L11*   England
f135186   Blades   R-L11*   England
f206123   Diggs   R-L11*   England
f16296   Foster   R-L11*   England
f170071   Hilton   R-L11*   England
f187354   Medley   R-L11*   England
fN8721   zzzUnkName   R-L11*   England
f156511   Brown   R-L11*   Ireland, Ulster
f66955   Lyle   R-L11*   Ireland, Ulster, Co. Antrim, Larne
fN9273   Strathern   R-L11*   Ireland, Ulster, Co. Londonderry, Bellaghy
fN63307   Jackson   R-L11*   Ireland, Ulster, Co. Londonderry, Coleraine
f227066   Mitchell   R-L11*   Ireland
f197182   Shannon   R-L11*   Ireland
f117368   Smith   R-L11*   Ireland
fB1201   Forrester   R-L11*   Scotland
f91188   Meeks   R-L11*   Scotland
f6150   Turner   R-L11*   Scotland

For a possibly more representative set of data we might like at the Myres/Busby study data. If I get a chance I'll pull that up but Richard Rocca probably has it too.

[stoneman]

The farmers that built the Court Tombs in Ireland 6000 years ago came from North Eastern Europe.

The first U106 people arrived in Ireland 6000 ybp.

If an SNP was an infection where would most of the people be infected? Mikewww would probably think 8000 miles from the source. Do you think that one man with the defective gene known as M222  moved to the North West of Ireland and spead it to most of the population? I dont believe it.

Based on what? 

[Mikewww]

If an SNP was an infection where would most of the people be infected? Mikewww would probably think 8000 miles from the source. Do you think that one man with the defective gene known as M222  moved to the North West of Ireland and spead it to most of the population? I dont believe it.

If you have a point that's fine, but why make an off-handed 8000 mile remark about me? There is no need to be personal.

Perhaps you are talking about the concepts of this paper "Surfing during population expansions promotes genetic revolutions and structuration" by Laurent Excoffier and Nicolas Ray
http://www.cell.com/trends/ecology-evolution/abstract/S0169-5347%2808%2900167-5

"The Fate of Mutations Surfing on the Wave of a Range Expansion" by Klopfstein et al
http://mbe.oxfordjournals.org/content/23/3/482.abstract

I don't come up with this stuff. There are scientists who think this can happen. No one, including me, says it always happen. Wave surfing is just something to consider.

[stoneman]

I dont know why you write so much about U106 when you are not in this group.Maybe I thought that you were having a go at me.

[Mark Jost]

...


Ok, I just loaded all the 67marker Hts from MikeW's latest U106, cleaned the Nuls and zero to empty cells, and loaded CladeA and B with the same set. Available to anyone.

https://docs.google.com/open?id=0By9Y3jb2fORNNF9ma2JfZEtKUlk

MJost

Thanks so VERY much for all of your continued work on this project. It is VERY much appreciated.

Your welcome. I was so interested in getting a mostly simple method to calculate alot of things but it was MikeW's Dashboard system in his Gen7.1 age Estimator and my desire to upgrade to the new 111 marker system of Nordtvedt's Engine that I decieded to upgrade things.

I hope you U106 guys produce a deep intraclade table between all your subclades and Geo-locations that Mike has worked so hard on identifying.

MJost

[Mikewww]

Message deleted

[stoneman]

Message deleted

[Jarman]

I am R-U106 but Jarman is not the only line I research. I sponsored DNA testing for a male in another lineage I am also researching - the STR matches were very informative and this line is R-L23 - I can converse and share info about that haplogroup too when I choose and am not limited to discussing U106.

[Peter M]

The best I can define things to is there in the spreadsheets. I don't have time to put them on Google maps, but if you know the geography its not hard to look at it.  We don't have that many L11* guys. I always say they are lightly scattered, but I guess that is just my characterization.

Remember, this is not a scientific sample and is probably biased towards American immigration sources.

fN39211   Klassen        R-L11*   Poland
f159936   Edison         R-L11*   Netherlands
yUHFMG    Leigh          R-L11*   England, Nort West, Cheshire, Stockport, Cheadle
f6150     Turner         R-L11*   Scotland

For a possibly more representative set of data we might like at the Myres/Busby study data. If I get a chance I'll pull that up but Richard Rocca probably has it too.

No, this is definitely not a scientific sample. It is heavily biased torwards British Isles and a few origins are remarkable: e.g. I've never come across the surname Edison as referring to a citizen from the Netherlands. $:-)

I've found a few more L11* in my database and added them to the list. Klassen is U106+, I have no SNP data for Leigh, but he's British Isles, so I don't see him as particularly critical and Turner appears to be L11 (has NOT tested U106 and P312; I must have a look if he's tested one of the sub-clades).

All in all some 25 L11* on the European Continent, but a few have not specified a geographic position; most have specified a country though.

I've created a Google Map using the V2 API and will publish that, if possible together with the rest of the data. The main problem here is that I'm rebuilding the L257.org web site at the moment (hackers apparently from Poland), and therefore a little improvisation is needed for a few days.

The biggest question: are there people who would be willing to sort out the geographic coordinates (latitude + longitude) for the people who haven't done so ? Please step forward.

EDIT: the map is here:http://www.groenebeverbv.nl/L11.jpg, but is a temporary thing on a temporary location to give everybody an idea what are we talking about; it's GM V2 (looks a little "old fashion") and static (you cannot click on pins). Future directions might be: (1) adding "old" P312 and/or U106; (2) adding M269+/L11- profiles; and (3) making the whole thing dynamic, so all Google Maps facilities are available. This will require volunteers to get the data together !!!