Likeliness of no 12 marker matches with R1B, and potentially meaningful mutations

[DrB]

Hi everyone,
I took a Y-37 test a while ago (2011) and had no matches at 37 markers, none at 25 and even none at 12! More than 6 years later.. still the same!
I know that it can happen with some rarer haplogroups, but I'm R1b! More than that, R-M269 (as far as the Y-37 test results go, I know this can be narrowed down).

How likely is such a thing to happen, to have absolutely zero 12 markers match with the most common haplogroup in European populations? Is there anything interesting or significant that can be derived from it, other than the fact that no one else even remotely patrilinearly related (or not even) has taken the test? Would it be useful to post the specific number of repeats/unusual mutations that have taken place (there are some unusual ones, of course, hence the utter lack of matches), can they actually tell something, be revealing somehow?

Thanks a lot in advance for your answers!

[prairielad]

Do you have an uncle or cousin who can test? I tested my Father (basic haplogroup R-M269) and he also had no exact 12 marker matches, even at 37 markers he only had 8 matches at a GD of 4
Tested his cousin and found out that my father has multiple recent mutations that happened in him, my Grandfather or both, one marker located in the 12 marker level. My Father to cousin are a GD of 4 at 37markers. Cousin has 23 surname matches at the 12 marker level that are a GD of 0.



Y-DNA12 Matches
11 out of 12 (A genetic distance of 1)*
12 out of 12 (A genetic distance of 0)
Y-DNA25 Matches
23 out of 25 (A genetic distance of 2)
24 out of 25 (A genetic distance of 1)
25 out of 25 (A genetic distance of 0)
Y-DNA37 Matches
33 out of 37 (A genetic distance of 4)
34 out of 37 (A genetic distance of 3)
35 out of 37 (A genetic distance of 2)
36 out of 37 (A genetic distance of 1)
37 out of 37 (A genetic distance of 0)
Y-DNA67 Matches
60 out of 67 (A genetic distance of 7)
61 out of 67 (A genetic distance of 6)
62 out of 67 (A genetic distance of 5)
63 out of 67 (A genetic distance of 4)
64 out of 67 (A genetic distance of 3)
65 out of 67 (A genetic distance of 2)
66 out of 67 (A genetic distance of 1)
67 out of 67 (A genetic distance of 0)
Y-DNA111 Matches
101 out of 111 (A genetic distance of 10)
102 out of 111 (A genetic distance of 9)
103 out of 111 (A genetic distance of 8)
104 out of 111 (A genetic distance of 7)
105 out of 111 (A genetic distance of 6)
106 out of 111 (A genetic distance of 5)
107 out of 111 (A genetic distance of 4)
108 out of 111 (A genetic distance of 3)
109 out of 111 (A genetic distance of 2)
110 out of 111 (A genetic distance of 1)
111 out of 111 (A genetic distance of 0)

* For Y-DNA12 matches, 11 out of 12 matches are only shown with both customers belong to the same group project.
However, to best serve our customers who are adopted, we provide at the Y-DNA12 level both 11/12 and 12/12 matching to the entire Family Tree DNA database to those in the Adoptee Project. This is because they cannot know the best Surname project to join in advance of testing.

[TwiddlingThumbs]

Dr. B - Could you please list your values for the first 12 markers? That way we can see what is causing your failure to match - eg lots of off-modal values or a single value that is extremely unusual. I suspect it may be the latter, in which case, you would want to see if you match up with anyone else closely on the other markers.

[DrB]

Thanks to both of you for your answers!

No, I do not have an uncle or a first cousin I could ask... I do have a 3rd or 4th degree patrilineal cousin... (but I've never even met him and I don't think he's very tech-savvy (I'm not sure he even has an email, I've only spoken to his wife a couple of times by email), it might seem a bit awkward..

Actually, interestingly, on ysearch, I match with 2 people, but who had other markers tested; so we have only like 8-9 markers which we both tasted... (those are exact matches, though, so the site gives us a genetic distance of 0.. but one was apparently adopted, and neither of them has replied after I contacted them, haha...)

Sure, here are the results, in the order and for the markers used by FTDNA:
13 24 14 11 10-14 12 9 12 13 13 29

[TwiddlingThumbs]

13 24 14 11 10-14 12 9 12 13 13 29

Here's the percentage of R1b people with the 12 marker values you posted:
DYS393 13 91%
DYS390 24 60%
DYS19 14 89%
DYS391 11 67%
DYS385a 10 03% (87% are 11)
DYS385b 14 66%
DYS426 12 98%
DYS388 9 00% (98% are 12, 1% are 13, <1% are 11 or 14)
DYS439 12 61%
DYS389i 13 80%
DYS392 13 86%
DYS389ii 29 63%
See: http://freepages.genealogy.rootsweb....logy/yfreq.htm

As you can see, all your values are the most common ones with two big exceptions. Your 10 on DYS385a is pretty rare, but since most R1b's have an 11, that would only give you a genetic distance of 1 from tons of people. More significantly, your 9 on DYS388 is so rare that it doesn't even show up in the frequency stats. Since 98% of R1b's have a DYS388 value of 12 and since FTDNA treats the difference between 12 and 9 as 3 separate mutations, that one marker alone is going to add 3 to your genetic distance from pretty much every R1b and make a match hard to find.

You might want to try exploring 2 avenues: 1 - join the R-M269 group and an applicable surname group and see who you match up closest with if you ignore DYS388. 2 - try to find info on DYS388 values of 9. I googled a bit and found a few references to that value: http://archiver.rootsweb.ancestry.co...-02/1203362619

Note that even though FTDNA treats the difference between 9 and 12 on DYS388 as 3 separate mutations, it could just as easily been the result of a single mutation and therefore you might actually be closely related to people with a 12 value.

[DrB]

Thank you so much! This is super interesting! No I have something to start from..
(when you say that I could "look at the matches regardless of the..." - but would that be relevant, to ignore such a change? Is it likely that I would be related to those people still?)

I will definitely look at information on these odd mutations.. 0% is quite nice, isn't it?

Regarding the projects.. I tried joining the R-M269 project yesterday, but somehow it didn't work/wouldn't let me.. as for my last name, it's an alsacian name which means short in german, and was giving to whoever was.. short, at that time.. so it's not even a real surname in the usual sense of the term...

[TwiddlingThumbs]

when you say that I could "look at the matches regardless of the..." - but would that be relevant, to ignore such a change? Is it likely that I would be related to those people still?

Yes. A change from 12 to 9 on DYS388 could have happened through a single mutation in one generation, so if you match someone reasonably closely based on the other marker values, you might well be related.

One thing I think you could do is go to ysearch.org and create a new user. Enter your values manually, but instead of entering 9 for DYS388, enter 11. (That way it will only be treated as a single mutation change.) Then see if you match up with anyone. (Note, however, that if you have other extremely rare values that are multiple steps from a common value, you might need to do the same things with those values as well.)

[DrB]

Hi,
Thanks! Ok, I tried.. It doesn't change much.. (I can find many genetic distance 1 matches on ysearch even with my real results.. but they are with less than 12 markers (different markers tested) - one does match with odd 9 one! But I don't think the contact feature works, every time I get an error message, which is especially frustrating in that particular case!

[Jim Barrett]

No, the contact feature hasn't worked in years.

[DrB]

Oh... So there's just no way to contact them? Well that's disappointing...

The best I can get, with 15 (only) markers tested in common with others (but the marker DYS 388 (and thus my are 9 result) is not counted in search) is one match with Yfilter on YHRD (also, DYS365 is missing from the list, as ftdna doesn't test it) it says:

Observed:
Found 1 match in 145,816 Haplotypes. This is approx. 1 match in 145,816 Haplotypes (95% CI: 26,172 - 5,759,425).

Expected:
DL (Yfiler) Discrete Laplace (Yfiler) could not be performed due to intermediate, duplicated or missing alleles within the Yfiler markerset.
n+1/N+1 Approx. 1 match in 72,909 Haplotypes (95% CI: 20,184 - 602,027)
Kappa Approx. 1 match in 188,259 Haplotypes

Does this bring any revelant information according to you?

By the way, I found another rare result at a particular marker (which isn't counted either in the YFiler search) thanks to your very useful link (allele frequenty databse): YCAIIb at 19(only 3% for R1B, with 23 for 81%... quite far!)

[Ric]

I have complained about this situation too. You are not alone, basically anybody with off-modal values can be disqualified to have valuable matches just because of one mutation. If this single mutation happened between your grand pa and your dad, you may not even match your own grandpa !!! And of course there is no way to know a priori when that mutation occurred.

There are several issues to consider here.

The 'off modal' values are not provided by your STR test, but they are visible in the groups for your relevant haplogroup, where the STR sequences of everybody in the group are actually displayed.
The Mode (the most common value as in statistics) for a particular STR position is indicated for everybody, so if your personal value is very far off this value/mode, you would know what this particular value 'should be' for you.
But to know that, you must do some SNP testing to know in which R1b subclade you belong.
Once you know what and where is the problematic STR, you can try to contact the people in the group who 'would match you' IF your off-modal STR wasn't off.

Problem: not everybody who has tested STR also join a group, this is a small subset of the testers and worse, you cannot contact them directly, you have to go through the admin of the group.

Beside joining your own haplogroup, I don't see anything else you can do, but FDNA certainly could do 2 things :
1) Relax the GD threshold so that the many people in this situation finally can have matches. Many matches will be irrelevant but a few may be true matches.
2) let the customer enters himself a particular vale for a str, typically that would be the Mode value that the customer retrieves from his haplogroup-group.
That's what Ysearch was supposed to do, but ysearch is dead now.