Wow, Geno 2.0 NG gave LESS haplogroup info for my dad than the old test!

[KATM]

If you ask me the SNPs you are looking for I can check.

Thanks for the offer, but right now the person in question has done no testing whatsoever, so we have no idea of his Y haplogroup, or any SNPs. I only speculate that he might be in one of the "I" groups, since they are prevalent in Scandinavia. It could be otherwise, of course.

I was just wondering if the "I" haplogroup SNPs were limited in the Geno 2.0 NextGen test.

[wombat]

I do see that they did add in one next test for L20 I think it was Z2191, I forget.

To correct my statement above the new sub-L20 test they added is for BY2171. I can't see to find what the ancestral value is. I assume my dad must have tested negative for this SNP change but since FTDNA seems a touch hit and miss in how they are tracking transferred SNPs I'm not 100% sure. They made it seem like he wasn't tested for it, but now going through the raw data I see that he was. (confusingly the old test did test for a CTS2171 on the Y, not sure if that is the same thing, maybe it already had it?) Actually I found the ancestral/derived for it on ISOGG YBrowse. But that would suggest he has positive for some of these. FTDNA transfer appears to suggest either not tested or negative. It could be a positive/negative strand issue. Or FTDNA transfer not working properly.

FTDNA is somewhat random in whether they say you should or shouldn't order a SNP because you have or haven't already been tested for it. They note some that were tested but fail to note others so that led to some confusion.

It seems I found another branch of L20 sub the old test tested for that the new one doesn't. So now it seems that is two subs for the old test and one for the new.

But it's complex since SNP names change.

I wish I knew what Geno/FTDNA considered to be the ancestral for each of the SNPs. I read that the chip sometimes uses different strand +/- for different SNPs in a few cases so how do you know when trying to parse it yourself?

Maybe L20 is not studied enough. Until I hear from them it's hard to know what is going on. Maybe the L20 handling is reasonable and maybe not. The V mtDNA is bad though for sure and that is what really got me going.

[wombat]

Hmm this is becoming confusing. For L20 using old or new I have what the ISOGG lookup calls derived which makes sense since am L20+. For Z291 and YSC0000193 branches on the old test (not tested on the new) have what it calls the ancestral values which makes sense since Z291- and YSC0000193-.

But for this new BY2171 that the new test has according to the ISOGG SNP lookup it would seem the derived value (AA) is showing up and yet FTDNA appears to be showing me on the chart as untested (which it might mean to be negative).

So does this mean FTDNA is messing up handling this SNP transfer, that a problem has been found with that probe so they are ignoring it on purpose or that this is one of the SNPs that they measured on the - strand instead of the + strand and you need to interpret ancestor and derived in opposite fashion?

If BY2171 actually is derived then maybe the new test has found a new sub-L20 branch that we are actually on. Perhaps in time they will update the terminal SNP on their site to show this when and if it appears as an official ISOGG in a year or two or when they get around to deciding how to handle it?

I'll have to contact them.

[wombat]

The Geno 2.0+ also has a problem with false positives and possibly false negatives. This is another reason why Geno 2.0+ is not a very good test.

Actually the thing where I said he had a - and + for something downstream of mixed negative results I don't think points to Geno 2.0 NG error at all. Looking over a bunch of BigY and such results it seems like many others have similar stuff going on so I think the problem is actually with the "needs more research tentative" item ISOGG has directly under their L20 branch, it seems that many get some of those negative and yet get L20 line and the line two below with some positive.

It could also be that I misinterpreted what they meant by the first line below L20. They might mean those were just a bunch of various potential branch points and that not all of them should be positive before being able to get to the next line below that one. So there may not even be any conflict at all possibly.

[Armando]

Armando, can you refer me to a website where I can see which Y haplogroup SNPs that Geno 2.0 NextGen does test? I gave a NextGen kit to a male who is likely to have an "I" haplogroup, if he ever uses the kit. He is of direct line male Scandinavian descent, back to about the mid-1700s. I was hoping to get at least a general idea (I1a, I2b, etc.), but now not sure if that will show, from what you are saying.

I don't know of any website and I don't know why they don't publicize the SNPs that are tested. I do think that you will be able to determine if he is I1a, I2b, etc because there are 179 I1 and I2 SNPs tested. Just be sure to have a haplogroup admin check the results for you once you get them transferred to FTDNA. You'll also be able to download the CSV file yourself and compare the results with what is found at http://www.isogg.org/tree/ISOGG_YDNA_SNP_Index.html and if you know how to use Vlookup or Index Match functions then you can automate the process.

[wombat]

Well I finally got an answer on the BY2171. And it turns out that is one of the ones they read strand reversed. So Geno 2.0 NG and FTDNA putting as down as negative for that is correct. The RAW data download from FTDNA is confusing though since the ones that are strand reversed are not labelled so off-hand there would be no way to know.

I guess that is sort of interesting. Between the old Geno 2.0 (which did actually test more R-L20 sub-branches than the new one!) and the new Geno 2.0 NG which did test a new and major BY2171 sub-branch that the old test did not it seems that we are negative all of the known R-L20 sub-branches then.

The question is, is that just nothing much and just an artifact that not enough R-L20 have tested anywhere yet or it is some clue that there is something unusual about our R-L20. We do know that R-L20 is found only very rarely in our main country of ancestry (I see only one other person on all of FTDNA reporting the same country.) so might this hint that R-L20 maybe could go way back in the Baltics and that there could be some unique, old, very rare branch there from way back when and it's not something that snuck in from some other region more recently? Hard to say yet. But interesting.