Are matches at 25 markers meaningless if there is no match at 37 markers?

[gtkingston]

In terms of yDNA analysis, is there any significance to 25 marker matches if there is no reported match at 37 markers? My FTDNA results summary indicates that I have no 37 marker match and no exact 25 marker match. I do have two 25 marker matches at a genetic distance of 1, and both of these have the same surname (though different to my surname). I have seven 25 marker matches at a genetic distance of 2, and one of these seven again has that surname.

Although the FTDNA generated algorhythm suggests that there is a 98% chance I am related to the former two people, clearly the strength of this must be weakened by the absence of a match at 37 markers with these participants (all had 37 markers examined). But does the lack of a 37 marker match with these people mean that I should completely ignore the possibility of any distant relationship? Surely it is more than coincidence that three of my nine closest matches share a surname (an uncommon surname at that).

[fantome]

No 37 marker matches

Hi!
I'm in a similar situation and have the same question as you. I have 8 matches at 1 distance, 17 matches a 2 distance in the 25 marker group and no matches at 37. I had no exact or real close surname matches but most of the surnames were Jewish names.. My haploup results are R1a1, Ashkenazi-Levite. I always thought this was accurate as to a Jewish heritage. One respondent said a wild guess would be that the 25 marker matches implies an ancestor about 2,000 years ago. He said that if my patrilineal ancestry were Jewish, I would get much closer matches, simply because the Jewish community is very strongly represented in genetic testing. He further suggested the probability of a Belarusian or Ukrainian Catholic heritage? I know that DNA results do not prove a religious history, but now I'm not sure of which is an accurate intepretation of my DNA testing. Perhaps someone can clear this up.

best wishes,
fantome

[lgmayka]

To both of you:

1) If you have a Ysearch entry, and mention it, we may be able to help you better.

2) Many FTDNA customers set their User Preferences to match only against members of the projects to which they belong. Thus, you may get more matches if you join the project for your haplogroup--e.g.:

R1a Project:

http://www.ftdna.com/public/R1aY-Haplogroup/

R1b Project:

http://www.ftdna.com/public/r1b/

But the bottom line is that near-matches (1 or 2 off) at 25 markers may or may not be very significant. One cousin of mine was 24/25 with a known Scottish family; but he then found himself 34/37, and finally 56/67. The Scottish family belongs to a subclade whose members generally claim that they have occupied Scotland since the Ice Age. Needless to say, they deny-deny-deny any relationship with my cousin anytime within the past 10,000 years.

As my example shows, even 37-marker near-matches can be somewhat misleading. More and more, I urge my project members to "go for the gold [standard]": 67 markers. Even if your nearest matches have not yet gotten to 67 markers, you will be able to see your relationship among ethnic groups better, and your very presence at the 67-marker level will encourage your near-matches to upgrade too.

[gtkingston]

Thank you. My ysearch address is KT94U. I would be grateful for any advice I can get on this matter. I just find it difficult to accept that three of my nine closest matches (including my two closest matches) on the FTDNA database could have the same uncommon surname simply by chance.

[gtkingston]

It seems that none of my matches (if I am allowed to call them that) have loaded their results onto ysearch.

[Clochaire]

gtkingston,

I think I may have run into a similar situation to the one you describe. I had 7 "close matches" of 23 and 22 / 25 showing up on 'Recent Ancestral Origins' page, but was unable to find them on Ysearch.

It turned out that 4 of these individuals had indeed entered data on Ysearch. But they didn't show up on my Ysearch queries because our matches "fell apart" at 37 loci. That is, all of us had tested at least 37 loci in total, and the Ysearch query doesn't allow you to limit your search to the first 25 loci.

You can get around this by setting up a separate Ysearch ID which lists only your first 25 loci.

Of course, this allowed me to see some pulished genealogical info about the "close matches", and I made some inferences about the possible remote origins of my own people accordingly. But I know that the conclusions I draw have to be somewhat limited -- after all, they fell apart at 37 loci.

Hope this helps.

Jack

[cacio]

Another way to check for larger distances at many markers is to input the search parameters manually. When you do a search in ysearch, a URL will show up for the search result page with the various search options. You can change those manually and then load the page again.
For instance, you can set mismatches_max= whaterver number you want even if larger than 5. This will allow you to see, say, distances of 6 at 37 markers. It's not as good as creating another user, but it's a quick check.

cacio

[lgmayka]

My ysearch address is KT94U.

Here is your Ysearch neighbor list at 37 markers:

http://www.ysearch.org/search_results.asp?uid=KT94U&min_markers=35&mismatch_type=absolute&mismatches_max=10&mismatches_sliding_starting_marker=30&lastname_limiter=none&haplo=&region=

As you can see, 435GJ is considerably closer than anyone else.

Here is your 25-marker Ysearch neighbor list:

http://www.ysearch.org/search_results.asp?uid=KT94U&min_markers=25&mismatch_type=absolute&mismatches_max=3&mismatches_sliding_starting_marker=30&lastname_limiter=none&haplo=&region=

QCSYV is closer than anyone else on the list.

But I think you're saying that the near-matches you were talking about are not in Ysearch at all?

[Bill Harvey]

what is your interpretaion of 56/67 match?

To both of you:

snipped......................
But the bottom line is that near-matches (1 or 2 off) at 25 markers may or may not be very significant. One cousin of mine was 24/25 with a known Scottish family; but he then found himself 34/37, and finally 56/67. The Scottish family belongs to a subclade whose members generally claim that they have occupied Scotland since the Ice Age. Needless to say, they deny-deny-deny any relationship with my cousin anytime within the past 10,000 years.

As my example shows, even 37-marker near-matches can be somewhat misleading. More and more, I urge my project members to "go for the gold [standard]": 67 markers. Even if your nearest matches have not yet gotten to 67 markers, you will be able to see your relationship among ethnic groups better, and your very presence at the 67-marker level will encourage your near-matches to upgrade too.

If I am reading your post correctly you are saying that an initial 24/25 then a 34/37 "fairly close" indicated relationship really fell apart at the 67 marker level at 56/67?

Of course this won't hold true for all such 'matchings" across the board but it is certainly food for thought.

Bill

[R2-D2]

I received 12 e-mails today to tell me that I had one new 25 marker match. I logged in to check it out and the person was a GD of 2 at 25. He also had a 37 by his name but he was nowhere to be seen on my list of 37 marker matches. Does that mean that he is too far away at 37 markers to be a match or could his last panel still be in the oven? I have one 33/37 match who is also a 60/67 match with a different surname.

[lgmayka]

If I am reading your post correctly you are saying that an initial 24/25 then a 34/37 "fairly close" indicated relationship really fell apart at the 67 marker level at 56/67?

Yes, although it is still debatable as to the degree of falling-apart. The 56/67 is actually a total genetic distance of 12. I would claim that such a distance is small enough to indicate at least a deeper common ancestor--say, 2500 years ago. Others, as I mentioned, claim that the Scottish subclade has been there since the end of the Ice Age, and that any more recent relationship is impossible.

[josh w.]

fantome, implications of your 25 marker matches

fantome, I am not clear on the conclusions of your informant. Take your 23/25 matches. FTDNA analysis indicates that there is a 67% chance of a MRCA around 16 generations ago and close to a 90% chance of a MRCA 24 generations ago. Generations are assumed to last 25-30 years. Your 24/25 matches would be more recent than that.

This would be consistent with Jewish ancestry especially since R1a1 is common among Ashkenazim. The frequency of Ria1 in the Jewish population is likely the result of earlier conversion, probably in eastern Europe although there were some of that subclade in ancient Israel. On the other hand, a very significant portion of Jews became part of their faith as a result of conversion. As has been frequently mentioned it is possible that a majority of Ashkenazi Levites are R1a1.

[josh w.]

P.S. fantome, as has been mentioned you need to check out Y search to see if you have some matches at 37 markers. FTDNA only lists the closest matches (34/37 and beyond) but even more distant matches can be informative. This would especially be the case if your more distant matches appear to have Jewish surnames. Even if one has Jewish ancestry, the overrepresentation of Jews in the FTDNA sample hardly guarantees close 37 marker matches. The FTDNA sample also has an overrepresentation of people from the British Isles, but that does not mean that people from that region will find close matches.

[dcsmith]

I received 12 e-mails today to tell me that I had one new 25 marker match. I logged in to check it out and the person was a GD of 2 at 25. He also had a 37 by his name but he was nowhere to be seen on my list of 37 marker matches. Does that mean that he is too far away at 37 markers to be a match or could his last panel still be in the oven? I have one 33/37 match who is also a 60/67 match with a different surname.

I got the same thing today. 8 emails informing me of a 25 marker match only to find the man is 25 -2 and had tested for 37 markers.

I think I may call FTDNA tomorrow for an explanation and make a complaint.

I have a feeling their doing it to keep people from complaining that they're not seeing any matches, but it's starting to annoy me to see so many obvious mismatches on my page.

[fantome]

My ySearch ID

[QUOTE=lgmayka]To both of you:

1) If you have a Ysearch entry, and mention it, we may be able to help you better.
Greetings,
Pardon my delay in replying. My Ysearch I.D. is 7q9v9. after viewing, would you still suggest taking the 67 marker test?

Thanks,
fantome