Announcement

Collapse
No announcement yet.

Half Norwegian but 0% Scandinavian, WHY?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • spruithean
    replied
    Originally posted by jötunn View Post
    But some Eurogenes calculator give me things like "50% mediterranean" and some like "98% North European". I can't believe that mediterranean is that hard to distinguish from north Europe - any instructions on how to use it?
    K13 and K15 are generally the recommended Eurogenes calculators. K36 is decent when coupled with the correct tools outside of GEDmatch.

    Leave a comment:


  • jötunn
    replied
    But some Eurogenes calculator give me things like "50% mediterranean" and some like "98% North European". I can't believe that mediterranean is that hard to distinguish from north Europe - any instructions on how to use it?

    Leave a comment:


  • spruithean
    replied
    Originally posted by Lani Friend View Post
    Ah, that explains a lot, thx. Which of the GEDmatch calculators do you trust most for northwestern European? Eurogenes?
    Eurogenes is pretty decent for people of northern European descent. MDLP isn't bad either.

    Though in my case since my NW Euro ancestry is from all over I get a lot of strange results.

    Leave a comment:


  • Lani Friend
    replied
    Ah, that explains a lot, thx. Which of the GEDmatch calculators do you trust most for northwestern European? Eurogenes?

    Leave a comment:


  • spruithean
    replied
    The Orcadian people appear to be a pretty strong mixture of Scandinavian/Germanic ancestry and Celtic ancestry (British Isles). I would view it as a proxy for mixed Germanic and Celtic ancestry.

    Leave a comment:


  • Lani Friend
    replied
    Originally posted by spruithean View Post
    Each GEDmatch calculator has mildly different compositions per reference group. Hence the vast variety of calculators.
    I see, that makes sense. Would you have any idea why Orkney and Orcadian keep appearing in the initial position of many of my percentages in single and multiple pop comparisons? It comes before West Norwegian in one of the calculations. Don't think I have any Orcadian lines.

    Leave a comment:


  • spruithean
    replied
    Each GEDmatch calculator has mildly different compositions per reference group. Hence the vast variety of calculators.

    Leave a comment:


  • Lani Friend
    replied
    Originally posted by jötunn View Post
    It means that 5th generation ancestor gave you only 1/32 of your DNA - if all other 31/32 are 100% Scandinavian, you won't see the 1/32 of British isles, for example.

    Additionally, most of human DNA is shared between everyone. Which means that most of the DNA gave by such ancestor is likely to be shared by all your ancestors. So it leaves only a small portion of data to analyze. Since a lot of randomness appears in DNA, small portions of data can even be taken as some random mutation.

    Let's make an example

    Here is a simulation of your 5th generation ancestor's DNA:

    AAAAAAAAAAAAAAAABBBBBB

    which means 100% scandinavian

    He had a child with another 5th generation ancestor:

    AAAAAAAAAAAAAAAACCCCCC

    which means 100% british

    The kid was like that:

    AAAAAAAAAAAAAAAABBCCBB

    Inherited 2 C's and 9 A's from the British
    Inherited 4 B's and 7 A's from the Scandinavian ancestor

    Then the kid's partner is like:

    AAAAAAAAAAAAAAAACCBBCC

    So then, there is small chance that the result will be:

    AAAAAAAAAAAAAAAABBBBBB

    showing that you are 100% Scandinavian, and completly ignoring the 100% British ancestor.

    Since the real tested DNA code is much more than just 22 letters, the chances of completly losing the trace are much lower - so it will typically take 5 generations, not 2 like on my example.

    Still, there is a very, very low chance that if both of your parents are 50% scandinavians, you will be 100% scandinavian. It's completly random. It's like estimating the losses when buying 100000 lotto tickets.
    I think I understand what you're saying (reread it more), but if this is true, why does GEDmatch pick up on both Scandinavian and British (and more if you use several of the options not just K15)?

    Leave a comment:


  • Lani Friend
    replied
    Just found this entry from the Nature abstract in Nature volume 519, pages 309–314 (19 March 2015):

    "We use haplotype-based statistical methods to analyse genome-wide single nucleotide polymorphism (SNP) data from a carefully chosen geographically diverse sample of 2,039 individuals from the United Kingdom."

    So it is still haplotype-based. Also, I read from an earlier project tied in with this that they used ABO blood sample studies and an "antigen" study of some kind.

    Here is a good pdf of the findings:

    http://www.oxfordtoday.ox.ac.uk/site..._March2015.pdf

    Leave a comment:


  • Lani Friend
    replied
    I knew about the grandparents' proximity requirement. And that also made it pretty much rural which eliminated urban areas with unlimited diversity.

    I did not know that the original project used just YDNA and mtDNA and no autosomal.

    But the current project must have relied on just autosomal results to make it relevant to modern populations, right? The name of the project was not the "Haplogroup Map of Britain and Ireland", rather the "DNA Map etc."

    Please correct me if I'm wrong.

    Leave a comment:


  • betadams
    replied
    Oxfordshire DNA matching service

    Background

    "This Geographical Dual DNA project provides for both Y-DNA and mtDNA for people with a direct paternal or maternal ancestral line originating in the English county of Oxfordshire. Participants can order either a Y-DNA (Y-chromosome) test to find out about their direct paternal line (your father, your father's father, etc.) or an mtDNA (mitochondrial DNA) test to learn about their direct maternal line (your mother, your mother's mother, etc.). The only requirement for joining the project is that participants must have a documented paper trail to Oxfordshire prior to 1900 on the direct paternal line for the Y-DNA test (preferably at least 19th Century or earlier) or on the direct maternal line for the mtDNA test (preferably three generations or more residency in Oxfordshire). If you have completed appropriate DNA testing with FTDNA (that is, you have an FTDNA kit number), you can join this project most simply by entering ‘Projects’ on your myFTDNA page and clicking ‘Join’. Members are asked to be as specific as possible with a geographical location for their Oxfordshire ancestors"

    The above were the criteria used for the original project between Oxford University and LivingDNA

    Leave a comment:


  • ltd-jean-pull
    replied
    Originally posted by Lani Friend View Post

    Question: If this is true, how come the brainiacs at Oxford Univ. were able to use atDNA to construct the phenomenal new "DNA Map of Britain and Ireland" published in journal Nature last year (I think):
    http://www.ox.ac.uk/news/2015-03-19-...-british-isles
    They had very precise criteria for selecting participants in the study. All their grandparents had to be born within so many miles of each other.

    Leave a comment:


  • Lani Friend
    replied
    jötunn:

    I very much appreciate your detailed response to my confusion about the FT staffer's reply. I have read your explanation several times, and I regret that it is far beyond my comprehension. Perhaps if you did a Youtube video about it..

    I admire your ability to master this concept, and you are very kind to try to explain it to me.

    I'm afraid I'm a bit mathematically-challenged. I once ended up in a basic math class (televised) where an eighty-year old instructor used ping pong balls to illustrate his examples and was always having to leave the screen for great lengths of time as he ran through the studio chasing them down

    However, if you are conversant with GEDmatch, I would really appreciate your reading my post about it on this page and answering my two very simple questions which should take someone with your brainpower about one minute to do...

    Sincerest thanks,
    Lani

    Leave a comment:


  • jötunn
    replied
    Originally posted by Lani Friend View Post
    It was the following quote from a FamilyTree staff member that confused me (this from someone who answers the questions sent in on the contact page):

    "Anyone who contributed DNA to your genome back farther than five generations will have a greatly diminished overall contribution to your genome that is most likely too insignificant to be detected or influence your ethnic percentage estimate."

    From this quote, it sounds like the results don't go back more than several generations, or they don't influence the MO percentages which from all the posts on this thread is clearly incorrect. I think now that his response just isn't correct, or I'm reading it wrong.
    It means that 5th generation ancestor gave you only 1/32 of your DNA - if all other 31/32 are 100% Scandinavian, you won't see the 1/32 of British isles, for example.

    Additionally, most of human DNA is shared between everyone. Which means that most of the DNA gave by such ancestor is likely to be shared by all your ancestors. So it leaves only a small portion of data to analyze. Since a lot of randomness appears in DNA, small portions of data can even be taken as some random mutation.

    Let's make an example

    Here is a simulation of your 5th generation ancestor's DNA:

    AAAAAAAAAAAAAAAABBBBBB

    which means 100% scandinavian

    He had a child with another 5th generation ancestor:

    AAAAAAAAAAAAAAAACCCCCC

    which means 100% british

    The kid was like that:

    AAAAAAAAAAAAAAAABBCCBB

    Inherited 2 C's and 9 A's from the British
    Inherited 4 B's and 7 A's from the Scandinavian ancestor

    Then the kid's partner is like:

    AAAAAAAAAAAAAAAACCBBCC

    So then, there is small chance that the result will be:

    AAAAAAAAAAAAAAAABBBBBB

    showing that you are 100% Scandinavian, and completly ignoring the 100% British ancestor.

    Since the real tested DNA code is much more than just 22 letters, the chances of completly losing the trace are much lower - so it will typically take 5 generations, not 2 like on my example.

    Still, there is a very, very low chance that if both of your parents are 50% scandinavians, you will be 100% scandinavian. It's completly random. It's like estimating the losses when buying 100000 lotto tickets.

    Leave a comment:


  • Lani Friend
    replied
    Yes, the GEDmatch data is amazing. I have been searching for my surname origin in Britain for years and finally narrowed it down to the areas of Kent or Devon based on frequencies of that name and other clues (Devon-born George Friend, the soccer star, is a deadringer for my Dad in his youth). GEDmatch puts it in the southwest, so I'm thrilled with that additional evidence!

    I see you are conversant with GEDmatch percentage breakdowns. Can you tell me why Orcadian is listed first in this list instead of West Norwegian when I don't have any relatives in Orkney?

    What do the numbers on the left side mean (1,2,3,4,etc)?

    Single Population Sharing
    # Population (source) Distance
    1 Orcadian 3.19
    2 North_Dutch 3.42
    3 West_Scottish 4.52
    4 Danish 4.52
    5 West_Norwegian 4.94

    Many thanks for any help!

    Leave a comment:

Working...
X