PDA

View Full Version : Nat Geno Pop differences



Claxon
07-19-2013, 12:17 PM
Thanks Scarlet... Solothurn and Rich R....
Previously I have mentioned that my Geno tests show me as 1 British, and 2, Romanian.
The British was 15 differences from my own, while Romanian was 5 differences.

Rich R mentioned he thought German was closer, but that was 8 differences. Sounded a bit strange

I have just looked at the populations list, including British, Germans and Romanian, and find the separate list is different than the pop graph put up, next to our own graph.

In the LISTED graph, German is 8 differences, while Romanian is 11, the British being 15.

So, apparently we are BOTH right, depending on the pop graph looked at.

I wonder how Nat Geno would explain that, if they ever choose to answer e mails. So, apparently we can choose our strongest population match, depending on what graph we look at.
And Nat Geno was supposed to be the "Be all, end all" of finding our origins ? They can't even get their own pop graphs to match a population.
Rich C

riya
08-01-2013, 07:34 AM
I still don't know, how to get access to my results without installing that software.

Scarlet Ibis
08-11-2013, 12:01 AM
I think the same thing is going on with my dad, in terms of his averages being closer to a population that isn't listed on his Who Am I? page.

Nat Geno says his top reference populations are: 1) British, and 2) Danish, which is accurate for his known ancestry. However, he seems to be closest to the German average on the populations list.

German: Northern Euro: 46% Mediterranean: 36% Southwest Asian: 17%
My Dad: Northern Euro: 45% Mediterranean: 36% Southwest Asian: 17%

Sein
10-01-2013, 04:12 AM
What they're trying to do is place you on a cline. They try to find two reference populations that can be averaged to produce an individual with statistically similar results to you. It is similar to the "two population approximation" you find at GEDMatch for the various Admixture calculators. Also, I think the first reference population they report is supposed to be closer to you than the second (relatively speaking). My first match is the Northern Caucasian reference population, while my second is the West Indian reference population. I'm not particularly similar in terms of percentages, but if you average both of these reference populations (but with a distinct shift towards the Northern Caucasians), you do get results very similar to my own. The averaged location on the map for these two populations is in Eastern Iran, and I guess that's close enough to home. My closest population in terms of percentages is probably the Iranian reference population, followed by the Pamiri Tajiks.

CelticGerman
10-03-2013, 01:05 PM
My results: 45% Northern European, 37% Mediterranean, 17% Southwest Asian.

My first reference population is British and the second German. In my opinion it should be other way round (?) ...

Solothurn
10-03-2013, 05:21 PM
I got:

1st Ref Pop = German
2nd Ref Pop = Greek

Shows how similar British and Germans really are :)


My results: 45% Northern European, 37% Mediterranean, 17% Southwest Asian.

My first reference population is British and the second German. In my opinion it should be other way round (?) ...

Scarlet Ibis
10-03-2013, 05:27 PM
It's become clear to me that their reference population calculations are "ok," but not even what I'd consider "good" compared to the consumer-driven tools we've had for a few years now, many of which are on gedmatch now. But, of course, most of us got the Geno 2.0 test for the Y-DNA results, and not for the population reference results.

CelticGerman
10-03-2013, 05:59 PM
I got:

1st Ref Pop = German
2nd Ref Pop = Greek

Shows how similar British and Germans really are :)

No doubt. At least Northern Germans and parts of Brits are really close
genetic cousins.

AJL
10-03-2013, 11:21 PM
I think by now there are enough doubts about the model for statistical distance that Geno 2 uses, that you are better off comparing your results to the reference populations yourself!

wombatofthenorth
04-08-2015, 09:04 PM
My results: 45% Northern European, 37% Mediterranean, 17% Southwest Asian.

My first reference population is British and the second German. In my opinion it should be other way round (?) ...


I don't believe the ordering means anything. I think they just list two matches in alphabetical order. I have not once seen a single person have match #1 begin with a letter that did not come earlier in the alphabet than their match #2. I have seen many people, my mom included, who are clearly a closer match to match #2. So I'm 99.9% convinced that the ordering of the matches has zero meaning. They just list them in alphabetical order.

wombatofthenorth
04-08-2015, 09:11 PM
Also you are not supposed to take the countries listed too seriously. They have very few countries. They show show what country in their base that your DNA's ancient ancestry mix is closest too. You might likely have zero ancestry from that particular country. Even if they have a country that a lot of your ancestry came from it might not come up since the average you got from your parents might be a closer match to some other country/region entirely. Your mix of ancient components could be much more like some country that you have not had any remotely recent, if ever, ancestors from.

It's shows a weird mix of recent 6 generations and ancient. It's hard to tell, from the raw result itself certain specifics as to whether it applies to anything recent or not. If you believe you had say mostly German background but the Mediterranean component came up much higher for Mediterranean and the Norther European much lower than it might imply that one recent relative perhaps came from Italy or something then those two components would likely be signalling that there was a very recent input from outside of Germany. Or if you had an unusual 2-3% result show up, that would imply likely a 3 or 4th great grand parent mixed with a completely different region, so again something in recent times. OTOH if you basically just got their population distribution for Germany than the Mediterranean and Southwest Asian components wouldn't imply at all that any recent ancestor was from either of those two regions and it would just all be the old signal from the spread of farming thousands of years ago.