开云体育

开云体育

Greetings,

This was copied from a name project on FTDNA. I could only include 30 markers because one of the kits only had 30. I couldn’t compare haplogroups because one of them was R-M269, which doesn’t help. Is it possible to tell if these 2 guys are related?

DYS393 DYS390 DYS19 DYS385 DYS391 DYS385 DYS426 DYS388 DYS439 DYS389i DYS458 DYS459 DYS455 DYS454 DYS447 DYS437 DYS448 DYS449

DYS464 DYS460Y Y-GATA-H4 YCAII DYS456 DYS607 DYS576 DYS570 CDY DYS442 DYS438

?

Kit A:? 13 23 14 11 {12-14} 12 12 12 13 13 29 16 {9-10-10} 11 11 24 15 19 30 {15-15-17-17-17-17} 10 10 {19-23} 16 15 18 19 {37-38-39} 12 12

?

Kit B: 13 24 15 10 {11-15} 12 12 12 13 13 29 17 {9-10} 11 11 25 15 19 29 {15-15-17-18} 11 11 {19-23} 16 15 17 17 {37-38} 13 12

Thanks,
Mike

Hi Saul,

?

Determining origins of haplogroups is an exceptionally fraught business. This is largely down to trying to extract very precise information from extremely small amounts of data, when that data is incredibly biased in favour of certain countries (see my previous post #8025 and others). This is therefore not something you should expect a definitive answer on any time in the forseeable future. We can only guess.

?

I'll repeat the statistic I often quote, that we sample about 20 times fewer French families as British ones. What does that mean in practice? Let's imagine a large clan who are related back in about 700 AD. During the early Norman period following the 1066 conquest, half of the family move to England. There are 15 Englishmen from this family who test. How many Frenchmen should we expect to see? There's a 47% chance you'll see none, even if they still represent half the family today.

?

So yes, it could be that the Dutch tester at R-FT240903 is simply a back migration. Or it could be that they are sole representative of a much larger continental family that we simply can't identify because we haven't got enough testers from the parts of Europe where R-FT240903 is common.

?

In contrast, the relative youth of R-FGC12983 and R-BY60940 means they are likely to be English. There's a clear focus for these haplogroups in Somerset and Devon, so it appears that this branch of the family ended up here sometime in the Middle Ages. Whether that also applies to your more-distant branch of the family is much harder to tell.

?

As you reduce the testing quality from Big Y back to Y-STR-only testers, and ultimately to Y-12, you decrease the accuracy of the relationships you're testing. Some genuine matches fall off and many spurious matches appear. This first means that the relevance of your Y-67 match in Cornwall isn't entirely clear - if they are a spurious match from R-FGC12983 or R-BY60940, then they may not be representative of your ancestry in Great Britain / Ireland. (If they have taken a BigY test and share your "terminal" haplogroup, that's different, but the system isn't letting me check this right now.) This also means that you should not use Y-12 matches - they are only relevant in very specific contexts, and most people's Y-12 matches are related to them only thousands of years ago (see message #7996 for details and reasons).

?

Cheers,

?

Iain.

Hi Jason,

?

Without knowing precisely what you mean by "Big Y STR differences", it's hard to give a precise answer. However, what generally happens is that the Y-STR results from Big Y are delivered separately from the main (SNP-based) Big Y results themselves. This is because the technology used to isolate and measure the first 111 Y-STRs is fundamentally different from the rest of the test - the sample goes through a different machine.

?

If you are seeing new results at levels where you've previously tested, it could simply be that new testers have come along who match you.

?

If you are seeing new levels appearing (e.g. Y-67 and Y-111) that weren't there before, it's probably because the Y-STR portion of your test has finished going through this separate machine.

?

If you are seeing Big Y results in your Y-STR matches (i.e. numbers of out 700+ Y-STR matches alongside your existing Y-STR results) then that means the STR component of your BigY test has been validated and your main BigY results should be ready "soon". How soon that is can depend on many factors, but is likely to be within the next fortnight or so.

?

Cheers,

?

Iain.

Good afternoon,?

?

I had posed a similar question a few months ago, but have learnt a lot since then and have some thoughts and questions I was wondering if you guys could help me with.?

?

My terminal clade is R-FT239204 and my earliest ancestor is from Kerry, Ireland. My surname however is English and I suspect that I have found documentation to support a Protestant > Catholic conversion around 1857, when records for this line stop (could indicate a migration into Ireland from England directly prior to this point).?

?

Anyway, having read a thread on here from a while back regarding the extremely distantly Kincades and Sinclairs and whatnot, which had mentioned haplogroup 995, I had assumed it to be post 1066. However, I have a match at y67 whose earliest ancestor seems to be from western Cornwall. I won’t reveal any information regarding them of course though. Therefore, my clade, in tandem to the two other subclades of R-FGC13546 (BY60940, and FGC12983) all cluster within Cornwall, with 940 and 980 being within just a few kilometers of each other. According to FTDNA, my clade with the others split off from 13546 around 700 CE.?

?

Would this not indicate that 13546 itself was from within Cornwall? Maybe even 995 was in England already? I know that there is a Dutch sample directly upstream of my clade, but this could be a back migration? I didn’t match with that individual and hence have no way of gathering their information.?

?

It also states that 940 and 983 both formed around the 1400s and 1500s, and so I am wondering as to if that would indicate a later split between those two clades and my haplogroup which formed around 750 CE, even though the TMRCA for all three states a common ancestor at the same time (700 CE).

I do have another question. I do not have a very deep understanding of strs and whatnot. I don’t really understand what it means. I have 1 y67 match who shares my terminal clade, and then 2 at y25. However, I have about 20 matches at y12 and here is where I am confused. Those at y12 mostly have vastly different clades to me. Some even have R1a and non R-U106. Despite this, they’re mostly at 1 step, I think I roughly understand what that means, and they are predicted to be related to me around the 1100s on average. Am I misinterpreting this? How can I, U106, be related to R1a or vastly distantly related R1b within 25 generations? Some are even located in Finland, Scandinavia and Poland, etc. I mean, for U106 that’s not irregular but to somehow be related to these individuals when my clade is clearly a different clade and completely separate from those regions around 1100 is strange.?

?

Thank you for all your help, :)?

?

- Saul

Hi Joe,

?

Please feel free to share.?

?

Cheers,

?

Iain.

Hi folks,

?

I've been keeping a track of the haplotree statistics for 4 1/2 years now, and saving copies of the haplotree for the last 1 1/2. The idea was that I can go back and see how the biases in the dataset are evolving, but this also lets us see some statistics of how these Family Finder tests are impacting our knowledge. Here, I'll be comparing the numbers to what they were six months ago, on 10th Nov., which is just before the FF results starting being released.

?

The total number of Y-DNA results in the haplotree has increased by 71% to 189,528. Given we're used to seeing an increase of around 6-8% each year, that's a huge increase in six months. Obviously, these aren't detailed haplogroups, but there are a surprising number of them.

The number of R-U106 has increased by 90% to 39,811. For comparison, R-P312 and R-M269 overall have both increased by 80%. I'm not certain why R-U106 has seen such a larger increase, but it may be down to tester geography. Some countries clearly prefer Family Finder over BigY tests (which isn't necessarily a good thing for us).

?

However, this increase is not all it seems. Most of these testers have not indicated their Y-DNA country of origin, or have a country of origin that is outside of Europe. This is maybe unsurprising as, if you've only taken an autosomal test, you're not going to be so interested in filling this information in. However, the total number of R-U106 testers in Europe has only increased by 30% to 11,702. It is this European statistic that informs us how much more information we can tell about the origins of R-U106 haplogroups in Europe. A 30% increase gives us roughly a 14% increase in geographic fidelity. Note that these Family Finder tests won't help us improve TMRCAs. This suggests we need to take action to engage with these new Family Finder testers and ensure they fill in their most-distant known ancestor information and include a country of origin.

?

In terms of where these increases have been happening, the approximate increases are:
British Isles: 24%
North-west Europe: 30%
North-central Europe: 35%
Scandinavia: 62%
Eastern Europe: 46%
South-east Europe / Balkans: 51%
Mediterranean: 36%

?

Importantly, we have increases in France and Belgium of 29% and 30%, which helps us investigate these under-tested areas, as does the increase in eastern/south-eastern Europe. We have increases of 46% and 53% in the Czech and Slovak Republics, which may help us uncover the early phases of R-U106, particularly around the Unetice Culture. A 62% increase in Russian testers particularly helps us track the eastward expansion of R-U106 in later times.

?

A good indicator of how deeply people test is the R-Z2265/R-U106 ratio. Almost all R-U106 men are Z2265+, so this ratio tells us what fraction of people are stuck on R-U106 and not tested below. Globally, around 94.9% of R-U106 men are tested Z2265+, and roughly 93.6% of men with European origins. Regional statistics are:
British Isles: 94.4%
North-west Europe: 92.3%
North-central Europe: 93.2%
Scandinavia: 94.3%
Eastern Europe: 93.6%
South-east Europe / Balkans: 90.4%
Mediterranean: 85.8%
?

It's not clear why southern Europe should lack testing below R-U106 to 2-3 times the extent of northern Europe. These numbers have decreased with the latest data, so it is unlikely that they simply represent rare basal clades of R-U106. Possibilities are that there simply isn't the incentive for them to test deeply, or they have historically taken SNP pack tests more than BigY tests. Either way, these testers have a huge impact on our ability to define the origins of R-U106, so we need to prioritise their upgrades to BigY.

?

Looking at the major haplogroups within R-U106, the increase in European testers ranges from 21% in R-Z343 and 25% in R-L257, to 39% for R-L47 and 44% for R-FGC910. The other haplogroups all show a 26-36% increase. This is important, because there is a certain minimum size (about 400 testers) below which biases in testing mean we can't get an accurate picture of how a haplogroup is distributed around Europe. Without that picture, it is very difficult to work out where a haplogroup's origin is and how it has spread from there. This increase means we have more haplogroups where we can make these kinds of estimates.

?

The biases in haplogroup testing themselves have changed, and these upgrades have actually evened some of those biases out. My main measure for bias is the number of men tested in a country versus that country's estimated population in 1800. I've normalised this to the British Isles, so a number of 5 in country X means that we test roughly five times as many British/Irish families for every family from country X. For various reasons, this is an approximate measure. Nevertheless, the biases have moved from:
North-west Europe: 10.5 to 10.3
--- France ? ? ? ? 24.0 to 22.8
--- Germany ? ? ? ? 6.4 to 6.3
North-central Europe: 9.5 to 8.6
--- Poland ? ? ? ? ?6.0 to 5.5
Scandinavia: ? ? ? ?1.2 to 0.9
--- Norway ? ? ? ? ?1.2 to 1.0
--- Sweden ? ? ? ? ?1.4 to 1.0
Eastern Europe: ? ?16.1 to 15.1
--- Russia ? ? ? ? 20.3 to 20.0
--- Belarus ? ? ? ? 6.7 to 6.0
--- Ukraine ? ? ? ?12.0 to 10.4
South-east Europe / Balkans: 13.9 to 12.7
--- Hungary ? ? ? ? 8.8 to 8.3
Mediterranean: 12.9 to 12.6
--- Italy ? ? ? ? ?14.1 to 13.9
--- Spain ? ? ? ? ?11.9 to 11.7
?

So, overall, the Family Finder additions have really helped our European coverage, particularly in countries and regions that contain important trace elements of R-U106 but are grossly under-tested. Interestingly, parts of Scandinavia are now better tested than the British Isles! France is really the hinge point here: the more we can improve this number, the lower the minimum size a haplogroup needs to be before we can get a good idea of its origin. We really need to capitalise on these new upgrades, and encourage testers from these under-represented countries to upgrade to BigY.

?

Hopefully this provides a rough idea of where we are with the testing upgrades, how they affect us, and what we can do to make the most of them. I don't know how far FTDNA are through processing their Family Finder archives but, based on kit numbers alone, they're not even halfway yet.

?

There are things we probably want to do on a group level when this process completes. For now, I'd encourage you all to engage with any matches you have on Family Finder or your Y-STR tests that are coming back with R-U106 haplogroups - particularly if they are from European countries outside the British Isles and Scandinavia. Get them to enter their most-distant known ancestor information into Family Tree DNA and identify their male-line country of origin so that it shows up on the haplotree.

?

Cheers,

?

Iain.

开云体育