Results 1 to 7 of 7

Thread: Methodologies for how "regions" are calculated

  1. #1

    Methodologies for how "regions" are calculated

    I'm wondering if anyone has info on how the various home-testing website calculate where DNA is "from", and what sort of timeframe they're supposed to represent?

    I did an Ancestry DNA test last year (which I've uploaded the results from to FTDNA and MyHeritage) and recently took a LivingDNA test (as I understand they have a strong British Isles focus) and about the only thing the tests seem to agree on is that I'm super-from the UK and Ireland (which I already knew from my inability to be out in the sun for more than five minutes without going a shade of lobster-red).

    However, the subregions they suggest vary wildly. I understand there's considerable overlap in a lot of regions (eg Ancestry's "Ireland/Scotland/Wales" and "Great Britain") but I get some wildly different estimations.

    Wall of text description of what I mean incoming:

    I'm British with an Irish grandmother, and know all four of her grandparents were also born in Ireland. I know you don't inherit exactly 25% of DNA from each grandparent but this seems like a good data point to look out for.

    Ancestry gives me 50% Great Britain, 33% Ireland/Scotland/Wales, with the remainder split between W. Europe and Scandinavia.

    MyHeritage, on the other hand, gives me 68.4% Irish/Scottish/Welsh, 29% English, and the remainder NW European.

    FTDNA gives me 54% W and C Europe, 44% British Isles and a trace remainder of Scandinavia.

    Living DNA gives me 96.4% UK and Ireland, 38% of which it labels as Southeast England with the remainder split between various other regions of the UK and Ireland (some of which I have no genealogical paper trail for, and only about 8% for Ireland itself).

    The genealogical paper trail suggests that basically all of my ancestors were born in the UK or Ireland as far back as we've got (200 years on average).

    As a Brit I was particularly interested in LivingDNA as I understood their data was based on extensive work in the British Isles and allowed you to "drill down" to almost the county level.

    I was surprised at the massive chunk of SE England-origin DNA in my Living DNA results, and how comparatively weak other regions are (Ireland in particular).

    This seemed to contrast quite a lot with what the paper-trail shows. My Irish grandmother's family were from around Dublin (though we've not got further back than mid-19th Century with them), my paternal grandfather was from Yorkshire (though his parents backgrounds are a mix of Northwest England, Yorkshire, Lincolnshire and a bit of Suffolk), my maternal grandmother's family were from Cheshire, Staffordshire and Shropshire, and my maternal grandfather's family were from Essex and East London (with some Cambridgeshire).

    Everything except Living DNA is based on data from my Ancestry test, and I know my Living DNA results haven't got mixed up as I've uploaded data from both tests to GEDMatch and am my own closest relative on there, as it were. All very confusing!

  2. #2
    Banned
    Posts
    6,336
    Sex
    Location
    Torun
    Ethnicity
    Central 75% + 25% Mazovia
    Nationality
    Pole
    Y-DNA (P)
    R1a > M198 > YP 1337

    Poland
    It depends on sample used in analysis.

  3. The Following User Says Thank You to lukaszM For This Useful Post:

     Mixed (01-07-2018)

  4. #3
    Registered Users
    Posts
    883
    Sex
    Omitted
    Ethnicity
    NW European
    Nationality
    Welsh
    Y-DNA (P)
    R-U106>DF96>S15663
    mtDNA (M)
    H13a1a
    mtDNA (P)
    H3a1

    European Union England Ireland Ireland Munster Wales
    Living DNA's Irish reference panel is only made up of 7 people at present. It will be updated in the future and your results will change. This is a known factor in the results of Irish people.

  5. #4
    Banned
    Posts
    6,336
    Sex
    Location
    Torun
    Ethnicity
    Central 75% + 25% Mazovia
    Nationality
    Pole
    Y-DNA (P)
    R1a > M198 > YP 1337

    Poland
    Quote Originally Posted by Capitalis View Post
    Living DNA's Irish reference panel is only made up of 7 people at present. It will be updated in the future and your results will change. This is a known factor in the results of Irish people.
    What? I think I have those 7 "academic" Irish samples also. So LivingDNA uses such tiny references? Ubelievable and people pay for it.

  6. The Following 3 Users Say Thank You to lukaszM For This Useful Post:

     timberwolf (01-09-2018),  Tomenable (01-01-2018),  tomz (01-30-2018)

  7. #5
    Registered Users
    Posts
    5,344
    Sex
    Location
    Poland
    Ethnicity
    Polish
    Y-DNA (P)
    R1b-Z2552
    mtDNA (M)
    W6a

    Poland European Union Poland Pomerania
    A new study about Irish DNA has just been published this December, it shows that Ireland is more genetically diverse than previously thought:

    http://www.rcsi.ie/index.jsp?n=110&p=100&a=11226

    https://static-content.springer.com/...MOESM1_ESM.pdf


  8. The Following User Says Thank You to Tomenable For This Useful Post:

     tomz (01-30-2018)

  9. #6
    Registered Users
    Posts
    883
    Sex
    Omitted
    Ethnicity
    NW European
    Nationality
    Welsh
    Y-DNA (P)
    R-U106>DF96>S15663
    mtDNA (M)
    H13a1a
    mtDNA (P)
    H3a1

    European Union England Ireland Ireland Munster Wales
    Quote Originally Posted by lukaszM View Post
    What? I think I have those 7 "academic" Irish samples also. So LivingDNA uses such tiny references? Ubelievable and people pay for it.
    I don't work for Living DNA so I just relay what I read elsewhere. It would explain why so many Irish people don't match the reference panels.

    I've collected far more than 7 Irish GEDmatch kits just by checking my matches GEDCOM files for people with 100% Irish ancestry. I collect them because I wanted to know the variance in the Irish population and GEDmatch spreadsheets just quote averages.

    It's not hard to do, as you know.

  10. The Following User Says Thank You to Capitalis For This Useful Post:

     tomz (01-30-2018)

  11. #7
    Banned
    Posts
    6,336
    Sex
    Location
    Torun
    Ethnicity
    Central 75% + 25% Mazovia
    Nationality
    Pole
    Y-DNA (P)
    R1a > M198 > YP 1337

    Poland
    Quote Originally Posted by Capitalis View Post
    I don't work for Living DNA so I just relay what I read elsewhere. It would explain why so many Irish people don't match the reference panels.

    I've collected far more than 7 Irish GEDmatch kits just by checking my matches GEDCOM files for people with 100% Irish ancestry. I collect them because I wanted to know the variance in the Irish population and GEDmatch spreadsheets just quote averages.

    It's not hard to do, as you know.
    I had them also collected on Gedmatch. But those Irish academic references are very representative and better than most of Irish from Gedmatch which were mistaken many times for Welsh for example in my K36 Oracle. Now people from this region generallly are happy from the results.

    Problem is there are more numerous academic Irish samples but in databases which are avaialbale only for scientists. This is why I was suprised that such company can't access them...

  12. The Following User Says Thank You to lukaszM For This Useful Post:

     tomz (01-30-2018)

Similar Threads

  1. Replies: 7
    Last Post: 11-25-2019, 09:02 AM
  2. "Roman DNA" "Anglo-Saxon DNA" "Viking DNA"
    By RandomUsernameGuy in forum General
    Replies: 53
    Last Post: 12-23-2018, 07:25 PM
  3. Replies: 4
    Last Post: 10-18-2018, 07:44 AM
  4. Replies: 15
    Last Post: 01-30-2017, 09:51 PM
  5. "Indus Valley" and "Kalash" admixtures in Khvalynsk?
    By Tomenable in forum Ancient (aDNA)
    Replies: 12
    Last Post: 08-12-2016, 10:46 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •