PDA

View Full Version : Comparing My Dad's MyOrigins Map to My Own



rms2
07-31-2014, 08:13 PM
My dad got his Family Finder results today. Here is his MyOrigins map beside my own.

2172 2173

Odd how my dad has 32% Scandinavian, and I don't have any of that. I have 53% Western and Central Europe, and my dad has none of that. I guess I need to order the Family Finder test for my mom now.

I might have received my red-head variant, Arg160Tryp, from my dad, since he has TC at rs1805008 like me (T is the risk allele). One of my dad's older sisters had red hair. My mom hasn't had the Family Finder test yet, however, so it's possible my dad gave me the C at rs1805008, if my mom has a T there. I doubt if that's the case, so more than likely I got the T red-head carrier trait from my dad.

My dad is also homozygous for lactase persistence at rs4988235 and rs182549 like me.

Of course, we show up in each other's Family Finder matches as Parent/Child with 3379.78 shared cm.

Mehrdad
07-31-2014, 08:56 PM
Very interesting, I wonder what happened to you having at least 4.5% Southern European and 2% Middle Eastern, was it covered up by something else?

Gray Fox
07-31-2014, 08:57 PM
Thanks for sharing. Aside from the Scandinavian input, I have similar percentages. I'm still scratching my head on that Eastern Middle East part. I like family finder for the simple fact that it gives you access to your raw data and provides fairly close matches with cousins. Myorigins seems awfully screwy and I could take it or leave it.. Probably leave it!

Are you finding the red-head variant and lactase persistence from 23&Me data? I seem to be a carrier for both myself. As I know I can have dairy and I have a copper/strawberry blonde color to my beard.

rms2
07-31-2014, 09:39 PM
Thanks for sharing. Aside from the Scandinavian input, I have similar percentages. I'm still scratching my head on that Eastern Middle East part. I like family finder for the simple fact that it gives you access to your raw data and provides fairly close matches with cousins. Myorigins seems awfully screwy and I could take it or leave it.. Probably leave it!

Are you finding the red-head variant and lactase persistence from 23&Me data? I seem to be a carrier for both myself. As I know I can have dairy and I have a copper/strawberry blonde color to my beard.

I never tested with 23andMe. I got the redhead variant and the lactase persistence data from my Family Finder raw data, which you can download and then use 7zip to open. I also did the Redhead test with BritainsDNA, which confirmed what I knew from Family Finder.

rs1805008 is one of the places on the MC1R gene for red hair. If you have a T there, you are a carrier of that red hair variant.

rs4988235 is one of the most common sites for lactase persistence. If you have at least one A there, you should be lactase persistent.

rs182549 is another location for lactase persistence. Having at least one T there is supposed to indicate lactase persistence.

Once you have downloaded your FF raw data to an Excel spreadsheet, you can use the Find & Select Function to run these things and find them.

rms2
07-31-2014, 09:43 PM
Very interesting, I wonder what happened to you having at least 4.5% Southern European and 2% Middle Eastern, was it covered up by something else?

I know my father contributed half of my chromosomes, but I guess he did not do it in a consistent fashion: you know, half his this and half his that. The differences are hard to account for though.

dp
07-31-2014, 09:51 PM
Dear Richard,
yeh, I don't know why immediate family members get such varied results.
thanks for the info on the trait alleles.
dp :-)

MitchellSince1893
07-31-2014, 11:04 PM
My dad has tested with both 23andme and FTDNA. I put a lot more stock in 23andme's ancestry composition than FTDNA's myorigins. 23andme's results are much more in line with the genealogical research.

While MyOrgins seems to misidentify or altogether miss ancestry.

My father has 10% German and Scandinavian ancestry, but MyOrigins says its 0%. 23andme says 7%.
He also has known South Asian ancestry (~2%). MyOrigins say 0%, and 23andme says 1.3%.
My father has no known Middle Eastern ancestry. MyOrigins says 7%, 23andme says 0%

Thus, I just don't have much confidence in MyOrigins.

Mehrdad
08-01-2014, 03:20 AM
I'm almost tempted to say to take anything below 10% on my origins with a grain of salt, but what would I know right :). But it would be interesting to run both your data and your dad's data through some of the calculators on Gedmatch and compare the differences and similarities.

Tolan
08-01-2014, 11:54 AM
There were others who were surprised with the results of their children ...

There are no genes specifically British, Scandinavian or Mediterranean, where I think, give fairly uncertains results.

The admixtures calulators are better in my opinion...
These are the only to tell me that I'm French!

rms2
08-01-2014, 12:14 PM
I'm almost tempted to say to take anything below 10% on my origins with a grain of salt, but what would I know right :). But it would be interesting to run both your data and your dad's data through some of the calculators on Gedmatch and compare the differences and similarities.

Good suggestion. I'll do that later today when and if I get the chance.

rms2
08-01-2014, 05:49 PM
I just finished uploading my dad's autosomal and x chromosome raw data to Gedmatch. I'm just waiting now until his data become available for comparisons.

Thanks again to Mehrdad for the great suggestion. I had nearly forgotten about Gedmatch, although my own data have been there for a long time now.

N21163
08-11-2014, 04:50 AM
Have just finished testing my parents with MyOrigins and were a little puzzled by the recombination for my results2365236623672366

Just waiting for raw data to be processed on gedmatch now

N21163
08-11-2014, 04:51 AM
Have just finished testing my parents with MyOrigins and were a little puzzled by the recombination for my results:

236523662367

Just waiting for raw data to be processed on gedmatch now

geebee
08-11-2014, 10:16 AM
Have just finished testing my parents with MyOrigins and were a little puzzled by the recombination for my results:

236523662367

Just waiting for raw data to be processed on gedmatch now

Yes, it's a little hard to see where your "Central/South Asian" part could have come from, since neither of your parents seems to have it.

We have sort of the opposite situation with my daughter. MyOrigins identifies her as "100% European", which is broken down (if you can call it that) as "97% Southern, Western, and Central Europe", and "3% Finland and Northern Siberia".

Both my wife and I have much more specific results. I have "94% European", which is "50% British Isles, 30% Southern Europe, 13% Scandinavia, and 1% Finland and Northern Siberia". I also have "3% Central/South Asian", "2% Middle Eastern", and "1% East Asian (Northeast Asia)". My wife has "93% European", which is "87% British Isles, 5% Southern Europe, and 1% Eastern Europe". She also has "5% Middle Eastern", and "2% Central/South Asian".

Now it's unsurprising that the smaller percentages don't show up with our daughter, but you'd think that she couldn't have ended up with less than "47% British Isles ancestry" from her mother, even if she inherited the least amount possible. Likewise, even if our daughter had inherited none of my British Isles ancestry, she'd have had to inherit everything else.

And, yes, with "97% Southern, Western, and Central Europe", the British Isles ancestry is surely in there somewhere. But why are my daughter's results so much less specific than mine or my wife's? And why is her "Finland and Northern Siberia" ancestry three times as much as that of the only parent shown as having that ancestry? Obviously, something not being counted that way for me (or possibly for my wife), is being counted that way for my daughter.

AJL
08-11-2014, 04:24 PM
Based on several comparisons of parents to children, it's clear the components cannot be taken literally. I ought to have a portion of Middle East–Asia Minor which disappears somehow from my father to me, yet I also somehow got several times more Scandinavia than both parents combined.

Goldschlager
08-12-2014, 12:06 AM
I did compare my results with my son and also his maternal grandparents.
Son: Ashkenazi Diaspora 91% Eastern Europe 9%
Me: Ashkenazi Diaspora 87% Eastern Europe 6% Asia Minor 7%
Gfather: Ashkenazi Diaspora 75% Eastern Europe 7% West Centr Europe 15% Asia Minor 3%
Gmother: Ashkenazi Diaspora 91% Asia Minor 6% Finland N Siberia 3%

It can be seen that Asia Minor did curiously disappeared. This seems to imply that whatever percentage is below, say, 10% may be unreliable.

AJL
08-12-2014, 01:32 AM
^ I suspect that's a higher percentage of Eastern Europe than fits in the typical range for Ashkenazim, have you taken 23andme and if so how do you read on East Euro there? Amongst my many Ashkenazi sharers the East Euro range is generally 0.1-5.0 % but I think I may have seen one or two as high as 10-12%.

N21163
08-12-2014, 05:05 AM
I wonder if a few things are happening here.
Analysis of autosomal DNA may produce seemingly inconsistent results, based on a "best fit" approach from algorithms and reference populations available, and some obscure populations could be seen as noise.
Or could we be seeing some DNA ancestry skip generations when DNA recombines?

While I appear to have a small percentage of Eastern European and Central Asian ancestry (that my parents appear not to) there is no Western and Central European component listed in my results.

I find this odd when my mother's Western and Central European results are listed as 39%

An article was posted some time ago in the FamilyTreeDNA forums entitled "Which Grandparent are you most related to?" http://www.slate.com/articles/health_and_science/human_genome/2013/10/analyze_your_child_s_dna_which_grandparents_are_mo st_genetically_related.single.html

Unfortunately I have no DNA samples from any of my grandparents to compare with my results.

Does this seem plausible to others or not?

AJL
08-12-2014, 03:35 PM
^ I haven't seen anything skip, as it were, but the best-fit approach seems right on the money.

One good thing about testing multiple generations is sometimes you have evidence for when components should show up but don't, sometimes you know when components are likely bogus, and sometimes components could best be described as borderline or marginal, but tending to the real. As an example my grandfather has a very slight Finnish/Yakut affinity across multiple tests, and sometimes I get that and sometimes I don't. I would call that borderline.

Incidentally, I am slightly less "related" to my grandfather than to my paternal uncle, to the point where the algorithm at Family Finder thought my grandfather was my uncle and vice versa. If you test enough people you may eventually start to identify which of your great-grandparents and great-great grandparents you most "take after."

MitchellSince1893
04-05-2017, 02:39 AM
disregard