need a little help with ysearch results

[expatmutt]

I am trying solve the mystery of why my DNA and my dad's DNA do not match anyone with his surname on any database! Meanwhile, he matches with a lot of people I have no paper trail for. There is apparently an NPE somewhere in the line, I think maybe 2-3 generations back.

So, I have submitted my dad's Y results to ysearch.org to narrow things down. He did a 29 marker test with ancestry. After browsing the forums here, I submitted a search for matches on ysearch with a max distance of 1 and matching at least 20 markers. (There are no matches at 0 distance).

I got five results, with four surnames, three of which have been popping up quite frequently on my FTDNA matches.

I guess my questions are:

How closely are these people likely related to my dad?

What would you do next? It seems that I have been pointed in the directions of some surnames, but I am not quite sure where to go from there.

Should I spring for him to do a more comprehensive test at FTDNA? If so, which package should I run on him?

Thanks for any advice. I have a lot of experience with paper genealogy, but have only recently added DNA to the mix (to my consternation, in this case!)

[thormalen]

I would start with the FTDNA Y 67 marker. It seems evident that you have DNA that does not correspond to your last name. If you really want to know then just jump in and test some more.
I started with the same type problem and have ended up with just about every test FTDNA can do...and using a combination of my 67 marker and my Family Finder I have the answer I was looking for. Good luck.

[Dave123]

Part of the problem is the small percentage of European derived men who have been tested, and most people using these companies are Americans of European descent. I forgot to look at FTDNA's latest Y statistic, but I think it's like 150,000 men have had their Y tested at FTDNA. Not sure about other companies, but the total might be 300,000 or so. There must be about 300 to 400 million males worldwide of European descent, so only about 1/1000th of the entire Euro descended male pool has been tested. So it's almost like a lottery ticket, a few people luck out, but probably most don't.

My paternal name is Schroeder, but at the Schroeder Surname DNA Project only 25 or so people have posted their results. Most of them are in the R1b haplogroup, with only two in my E1b1b1 haplogroup. Niether of them are even close to my sequence at 12 markers. So far my closest matches have entirely different surnames than I have. One of them has paternal ancestry from a village only 6 miles from my paternal line ancestral village in northern Germany. We exactly match on 36 of 37 markers. DYS582, the second fastest mutating marker, differs by 2 steps. It could have been an NPE. In a few weeks his 67 marker results should be in, so we'll find out if our tight matching continues to hold.

[MMaddi]

The best thing to do is give us your ysearch ID, so that we can look at your results and those of your matches. That would help us better answer your question.

One factor is that there are a few markers where there is a conversion factor for results from Ancestry.com - see http://www.ysearch.org/conversion_page.asp#RG It' possible that you've not used the conversion fact for one or more markers and that would throw off who you're matching.

[expatmutt]

Sure, the ID is GKPTU. My father's surname is Bembry, and as far as I know, there is only one Bembry surname group in the U.S. He is white (Rb1) but there are probably more black Bembrys than white ones, and we are all most likely related--assuming my dad is actually a Bembry.

I am matching up with surnames in his family going back through his grandfather's generation on FTDNA and 23andme.com (my DNA) and ancestry (his DNA). However, before his grandfather's generation, I have yet to have one surname match from his paternal line. There are plenty from the maternal lines.

I did see the information about the ancestry glitch, but I don't think it's a problem. I checked on that before I uploaded his results.

I am going to see if I can get some kind of deal on an expanded test now...

[jbarry6899]

need a little help with ysearch results

You asked how close these matches might be to your father.

I looked at your matches on Ysearch for 20 markers or more. There is only about a 50% probability that the best match and you had a common ancestor within the last 500 years, so it's possible (even likely) that the matches you are seeing reflect common ancestors prior to the time that surnames were in use. Thus the differing family histories and locations could explain the different surnames.

Upgrading to 67 markers may be helpful, but if your surname is rare, you may not have much luck matching with men born more recently who share it. It might help you determine, however, whether an NPE is more likely.

You can also try charting the different surnames that you find and looking at their origins to try to pinpoint the general location of your common ancestor.

You might find FF useful to see if you find any distant or maternal relatives who might have more information.

IMHO, what you have found does not necessarily indicate an NPE, but it is a research challenge.

Good luck!

Jim

[MMaddi]

I'm looking at your ysearch account right now. You didn't put in values for DYS19, DYS459a-b, DYS437 or DYS442. I believe that Ancestry.com does test those markers. If you have values for those markers, add them to your account. That will give you more markers to compare with others in ysearch.

Using a combination of genetic distance and looking for shared off-modal markers, I would say you have matches worth looking into. Those are (genetic distance in parentheses): XMHA3 (25/26), 9TCNK (24/26), RJKHB (23/26) and 6GQ4S (25/28). The first and last are German paternal lines and the middle two are Grants from Scotland.

They all share with you these off-modal markers:
DYS388=13
DYS447=26
H4=12

All have 37 markers tested, except the last one who's tested 67. Matches at the level of 25 markers, just short of what you have to match, are not very reliable for finding a common ancestor within the last several hundred years. You do need to test more markers at FTDNA, either 37 or 67 markers. That's the level at which you can more reliably judge matches. Since only one of them already has 67 markers, it may be better for you to test 37. Or you can go for 67 and see if you come up with matches that don't show up at the 26 markers you presently have to compare.

[expatmutt]

This is great info, thanks. Ancestry didn't actually give me those markers you mentioned, except for DYS19. I wasn't sure where to put it before, so I put it under DYS19a and it didn't change the results.

Looks like we'll be doing a bit more testing!