PDA

View Full Version : Big Y match?



MikeWhalen
10-26-2015, 11:39 PM
so I got an email from a fellow that said Big Y says we matched on 'SNP is 26,306.'

anyone have any clue what that one is, or could mean?

Evidently the fellow is African American who has run into the 'slave wall' when researching his genealogy in the US southern states

...I have no known or likely connection to any y lines from the southern states as I can trace my Y line directly back to Ireland, and from Ireland to Canada in the early 1800's

so, I don't know if this is some way ancient sort of snp or what, and the fellow is pretty new to this all....

any advice would be helpful

thanks

Mike

Petr
10-27-2015, 12:10 AM
FTDNA BigY matches have almost no value.

David Mc
10-27-2015, 12:15 AM
Hi Mike,

I believe I've been corresponding with the same person-- a very nice fellow he is, too. I've pointed him towards Mike Walsh and suggested he join the R-L21 and Subclades Project as he seems to be positive for R-L21 and DF13. He matches me on both of these but not on DF49, so our shared ancestor is very ancient indeed.

I don't know how closely matched he is to you, variant-wise, but if you do share a more recent ancestor it could have been a distant cousin who emigrated to the States from Ireland. It's possible that the link is a more ancient one as well (like mine). Either way, I hope he's able to make some kind of connection with someone, even if it's a regional one.

David Mc
10-27-2015, 12:19 AM
FTDNA BigY matches have almost no value.

I think they have value. The trick with Big-Y matches is filtering the results to include only your own subclade and then pairing the results with higher level STR matches. This isn't enough on its own, but it gives you a clear direction for further analysis.

MikeWhalen
10-27-2015, 12:32 AM
thanks David...I suspect its quite ancient but that is just a guess
Mike WWW sure is a great contact, I hope he makes it-that will give him a ton of info

Mike



Hi Mike,

I believe I've been corresponding with the same person-- a very nice fellow he is, too. I've pointed him towards Mike Walsh and suggested he join the R-L21 and Subclades Project as he seems to be positive for R-L21 and DF13. He matches me on both of these but not on DF49, so our shared ancestor is very ancient indeed.

I don't know how closely matched he is to you, variant-wise, but if you do share a more recent ancestor it could have been a distant cousin who emigrated to the States from Ireland. It's possible that the link is a more ancient one as well (like mine). Either way, I hope he's able to make some kind of connection with someone, even if it's a regional one.

Petr
10-27-2015, 12:34 AM
I agree that further filtering of the results may help.

razyn
10-27-2015, 02:18 AM
The top of the Shared Novel Variants list is useless, or largely so; but if you scroll to the bottom, and have a match with one to four people (or thereabouts), most of them are real; and if so, very useful indeed. As you go up the list (novel variants shared by ten, or forty, or lots of guys) look at the next column to the right and you'll notice that they also have SNPs marking some haplogroup that you know you aren't in. So those are probably highly recurrent -- or completely bogus, or shared with the mountain gorilla and all your mutual kin.

Like the 26,000 at the top of the list.

MikeWhalen
10-27-2015, 02:25 AM
rayzn-thanks for the heads up, I will check that out...but I think your conclusion is exactly correct, mountain gorilla's or sloths or toads or some such :)

Mike

David Mc
10-27-2015, 03:27 AM
The top of the Shared Novel Variants list is useless, or largely so; but if you scroll to the bottom, and have a match with one to four people (or thereabouts), most of them are real; and if so, very useful indeed. As you go up the list (novel variants shared by ten, or forty, or lots of guys) look at the next column to the right and you'll notice that they also have SNPs marking some haplogroup that you know you aren't in. So those are probably highly recurrent -- or completely bogus, or shared with the mountain gorilla and all your mutual kin.

Like the 26,000 at the top of the list.

In my case, when I've filtered out all non-DF49 matches I am left with a grand total of 34. Some of these are more closely related than others, looking at the PhyloTree, but they are fairy evenly spaced out from one another. But then there aren't a lot of DF49xDF23 who have tested yet, so it's all fairly (ahem) "relative."

Cofgene
10-27-2015, 11:57 AM
FTDNA has a new pop-up window which shows which SNPS which are now novel between your results and someone else on your Big-Y match list. More useful than before. I have not evaluated it to see how many crappy region calls are still left.

razyn
10-27-2015, 01:46 PM
One recent "issue" is that they are finally (though gradually) adding the SNPs discovered via BigY testing (over the past two years) to the "Known SNPs" list. In general, that's a good idea; but for the Shared Novel Variants matching function, it means they are no longer "novel" (now that they are "known"). The program isn't designed to put you in touch with people you match on known SNPs. Too bad, in a way; it was useful for finding people with BigY results who should be in one of the newer subgroups of the DF27 project -- but weren't, because they don't have any DF27 test result, as such.

MikeWhalen
10-27-2015, 03:26 PM
so I've done a bit of digging and have a better idea of what people meant.

the first thing is the fellow that contacted my mistook the number 26,306 for an actual SNP, rather than the actual number of matches we had out of the millions tested -its been so long since I looked at the big y stuff, I had forgotten that

we share 57 'novel variants' but that does not seem to be much use as none of them seem to have any meaning or value to them as yet-at least as best as I could tell

under the shared SNP's, it starts to get a bit confusing...after clicking on that tab, it says "Showing 1 to 18 of 510 entries (filtered from 36,287 total entries)", implying that we have 36,287 shared SNP's and yet on the original matching tab, it says we shared 26, 206...so which is it 36k or 26k?

I then checked to see if it says we share some of my key haplogroup defining SNP's, L21 and L513 and it said yes, he was derived for both of them
except!!
when on the main page after I type his surname in for comparison and it shows the basic info, under the 'Non Matching Known SNP's", it say there are 3 we don't share "CTS11903 CTS5396 L513"...so that's a major problem as it said we did share L513 on another page

-as a side interest, CTS5396 is an equivalent snp to L513, and I think, CTS11903 has something to do with Feline and Human imunoviruses
(see razyn we weren't far off in some of our speculation :) )

So the bottom line is, we seem to share the L21 snp but I have mixed results if we share the L513-one search says yes, one says no...
assuming I understood the Big Y site and tables and searches...which is always a roll of the dice with me

I hope the fellow does join the L21 group and ask for help, particularly with MikeWWW, who as you all know, is a super dedicated citizen scientist who has become the go to guy for L21 (amongst other ones)

anyway, thanks again for the help

Mike