Page 11 of 47 FirstFirst ... 91011121321 ... LastLast
Results 101 to 110 of 465

Thread: An nMonte and 4mix Guide for the Participants of the Basal-rich K7 and/or Global 10 T

  1. #101
    Registered Users
    Posts
    232
    Sex
    Location
    Netherlands
    Ethnicity
    South-Dutch
    Nationality
    Dutch
    Y-DNA
    I2a2a1b2-CTS1977
    mtDNA
    H13a1a1

    Netherlands Belgium
    Quote Originally Posted by Chad Rohlfsen View Post
    Not sure what to make of this, but I did include all possible sources of my Asian ancestry. Nothing about Saami makes sense though
    The minor admixtures feed on the mangled left-overs of the the major admixtures.
    Unless you have a good model of the major admixtures, modelling the minor admixtures is very unreliable.
    Personally I would anyway not recommend nMonte for hunting minor admixtures.

  2. #102
    Moderator
    Posts
    3,271
    Sex
    Location
    Normandy
    Ethnicity
    northwesterner
    Y-DNA
    U152>L2>Z367
    mtDNA
    H5a1

    Normandie Netherlands Friesland Finland
    Quote Originally Posted by Huijbregts View Post
    The minor admixtures feed on the mangled left-overs of the the major admixtures.
    Unless you have a good model of the major admixtures, modelling the minor admixtures is very unreliable.
    Personally I would anyway not recommend nMonte for hunting minor admixtures.
    I would not feel comfortable to have an argument with the Creator and Master of nMonte but, it depends on what you call a minor admixture. In the case of Chad, "Finnish" appears at 24% on a model that seems rather good, at least in my view:
    [1] "distance%=0.7223 / distance=0.007223"
    "French_East" 36.05
    "English_Kent" 33.75
    "Finnish" 24.8
    "Ashkenazi_Jew" 5.4

    This model is what he gets when removing "Saami" from the spreadsheet.
    Compare against what he gets with the Saami reference in the spreadsheet:

    Chad_Rohlfsen:Chad_Rohlfsen
    "English_Kent" 54.7
    "French_East" 34.15
    "Saami" 9.65
    "German" 1.5

    It looks like the finnish non saami components ( about 15%) have been in the second model integrated to the western populations ( about 85%).
    En North alom, de North venom
    En North fum naiz, en North manom

    (Roman de Rou, Wace, 1160-1170)

  3. #103
    Registered Users
    Posts
    232
    Sex
    Location
    Netherlands
    Ethnicity
    South-Dutch
    Nationality
    Dutch
    Y-DNA
    I2a2a1b2-CTS1977
    mtDNA
    H13a1a1

    Netherlands Belgium
    Quote Originally Posted by anglesqueville View Post
    It looks like the finnish non saami components ( about 15%) have been in the second model integrated to the western populations ( about 85%).
    That is a sensible comment on the relation between Saami and Finnish.
    But my argument was that, if the modelling of the major components is obscure, modelling of the minor components is futile.
    Compare the next two models, both in 7 dimensions:

    Chad_Rohlfsen
    "French_East" 64.8
    "Finnish" 28.6
    "English_Kent" 5.3
    "Ashkenazi_Jew" 1.3
    distance%=0.4827%

    Chad_Rohlfsen
    "Orcadian" 69.5
    "French_South" 27.3
    "Saami" 1.85
    "Nganasan" 1.35
    distance%=0.3747%

  4. #104
    Moderator
    Posts
    3,271
    Sex
    Location
    Normandy
    Ethnicity
    northwesterner
    Y-DNA
    U152>L2>Z367
    mtDNA
    H5a1

    Normandie Netherlands Friesland Finland
    Quote Originally Posted by Huijbregts View Post
    That is a sensible comment on the relation between Saami and Finnish.
    But my argument was that, if the modelling of the major components is obscure, modelling of the minor components is futile.
    Compare the next two models, both in 7 dimensions:

    Chad_Rohlfsen
    "French_East" 64.8
    "Finnish" 28.6
    "English_Kent" 5.3
    "Ashkenazi_Jew" 1.3
    distance%=0.4827%

    Chad_Rohlfsen
    "Orcadian" 69.5
    "French_South" 27.3
    "Saami" 1.85
    "Nganasan" 1.35
    distance%=0.3747%
    if the modelling of the major components is obscure
    ah, yes, I'm understanding. In fact, I was thinking in the same way about my father's case. His finnish percentage ( strongly connected to other results, and overall to his relatives and to the examination of triangulated IBD segments) seems to be solid, but I don't pay too much attention to some "saami" or "nenets" percentages he got with nMonte. That said, when a percentage of Saami ( more than, say, 2%) appears in the results of an individual who has a not ridiculous finnish component ( say more than 5%) , I'm tempted not to see in it just an artefact.
    En North alom, de North venom
    En North fum naiz, en North manom

    (Roman de Rou, Wace, 1160-1170)

  5. #105
    Registered Users
    Posts
    1,960
    Sex
    Y-DNA
    R1b-DF27 FGC17112
    mtDNA
    H5a1

    Do i just delete the last three dimensions on my target and input files? I'll try some other mixes too.

  6. #106
    Registered Users
    Posts
    1,960
    Sex
    Y-DNA
    R1b-DF27 FGC17112
    mtDNA
    H5a1

    Quote Originally Posted by anglesqueville View Post
    Are you certain that you know all your tree back to 7 generations? If yes you are a lucky man. How many finnish relatives do you have on gedmatch ( rough estimation, on segments larger than 7 cM/700 snps)? If you have 3 or 4, OK, all northwesterners have a handfull of Finnish ( my mother has around 10 finnish relatives on gedmatch, none elsewhere, and she has very likely nothing to see with Finland). If this is the case for you, well, it's strange, because it seems that these new calcs of David, allied with nMonte, are really powerfull, and I have some difficulty to imagine how they can give you 24% finnish without any actual finnish relatedness.
    edit: Khana, I don't know if your question is to Chad or to myself. If it was to me ( in fact to my father) the first result was with the complete modern spreadsheet. ( and with Global 10, 7 PC)
    Are you talking about the one to many comparison?

  7. #107
    Registered Users
    Posts
    36
    Sex
    Y-DNA
    R1b>U152>L2>FGC14641
    mtDNA
    X2e2a

    Quote Originally Posted by Chad Rohlfsen View Post
    Do i just delete the last three dimensions on my target and input files?
    Yes, that is the way to go for modeling with 7 dimensions in Global 10.
    Hidden Content

    Hidden Content

    Y-DNA: R1b>L51>L151>P312>U152>L2>Z41150>DF90>FGC14641; Niğde, Turkey
    mtDNA: X2e2a; Drama, Greece
    Maternal Y-DNA: R1a>Z93>Z94>Y40>YP294>YP4866; Razgrad, Bulgaria
    Father's mtDNA: T2b; Niğde, Turkey
    Paternal grandfather's mtDNA: H2a1; Niğde, Turkey
    Maternal grandfather's mtDNA: H5; Razgrad, Bulgaria

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

     Chad Rohlfsen (12-17-2016)

  9. #108
    Moderator
    Posts
    3,271
    Sex
    Location
    Normandy
    Ethnicity
    northwesterner
    Y-DNA
    U152>L2>Z367
    mtDNA
    H5a1

    Normandie Netherlands Friesland Finland
    Quote Originally Posted by Chad Rohlfsen View Post
    Are you talking about the one to many comparison?
    Yes, just to give a rough idea. Something a little better would be to use the research by segments ( the first tier 1 tools on gedmatch), and the triangulation tool ( on tier 1 as well). But the 1toMany is a good starting point.
    En North alom, de North venom
    En North fum naiz, en North manom

    (Roman de Rou, Wace, 1160-1170)

  10. #109
    Registered Users
    Posts
    1,960
    Sex
    Y-DNA
    R1b-DF27 FGC17112
    mtDNA
    H5a1

    All non-African and Papuan pops at 10 dimensions.

    "Orcadian" 43.15
    "Spanish_Pais_Vasco" 36.5
    "Russian_Orel" 14.75
    "Khanty" 3.2
    "Avar" 1.25
    "Basque_French" 0.6
    "Scottish" 0.45
    "English_Kent" 0.1

    [1] "distance%=0.5178 / distance=0.005178"

  11. #110
    Registered Users
    Posts
    953
    Sex
    Location
    Pennsylvania
    Ethnicity
    West European
    Nationality
    USA
    Y-DNA
    R1b-U152-Z36+FGC6511
    mtDNA
    H11a2a

    United States of America Germany England France Scotland Ireland
    1) This Western European/Asian question has been a research topic of mine, too, through my father. On this PCA 10 with nMonte2, he resolves well with just the French East population, which does fit his ancestry:

    [1] "1. CLOSEST SINGLE ITEM DISTANCES"
    French_East
    0.007141428

    2) If I add just about any population with some East Asian in it, the fit improves (some better than others):

    [1] "distance%=0.3753 / distance=0.003753"

    Dad
    "French_East" 89.35
    "Saami" 10.65

    [1] "distance%=0.4399 / distance=0.004399"

    Dad
    "French_East" 82.25
    "Russian_Kargopol" 17.75

    [1] "distance%=0.4679 / distance=0.004679"

    Dad
    "French_East" 83.85
    "Finnish" 16.15

    [1] "distance%=0.6386 / distance=0.006386"

    Dad
    "French_East" 95.75
    "Uzbek" 4.25

    [1] "distance%=0.6849 / distance=0.006849"

    Dad
    "French_East" 96.95
    "Turkmen" 3.05

    3) If I test around the edges with various populations, the fit improves from the single population, alone (some better than others):

    [1] "distance%=0.4923 / distance=0.004923"

    Dad
    "French_East" 96.2
    "Nenets_Forest" 3.8

    [1] "distance%=0.5163 / distance=0.005163"


    Dad
    "French_East" 95.65
    "Karitiana" 4.35

    [1] "distance%=0.5208 / distance=0.005208"


    Dad
    "French_East" 95.6
    "Pima" 4.4

    [1] "distance%=0.541 / distance=0.00541"

    Dad
    "French_East" 97.35
    "Nganasan" 2.65

    [1] "distance%=0.6815 / distance=0.006815"


    Dad
    "French_East" 98.25
    "Japanese" 1.75

    [1] "distance%=0.6842 / distance=0.006842"


    Dad
    "French_East" 98.35
    "Korean" 1.65

    [1] "distance%=0.6893 / distance=0.006893"


    Dad
    "French_East" 98.4
    "Andamanese_Onge" 1.6

    [1] "distance%=0.7057 / distance=0.007057"

    Dad
    "French_East" 99.1
    "Han" 0.9

    4) If I add German, with Saami and all the edge populations, I get the best fit, though it doesn't need the Saami anymore:

    [1] "distance%=0.3449 / distance=0.003449"


    Dad
    "French_East" 65.8
    "German" 30.9
    "Nganasan" 2.45
    "Andamanese_Onge" 0.7
    "Nenets_Forest" 0.15
    "Karitiana" 0
    "Pima" 0
    "Japanese" 0
    "Korean" 0
    "Han" 0
    "Saami" 0

    I don't know the source of this, but the fortuitous (to me) recent upload of a second cousin to Gedmatch has allowed me to narrow the ancestral emigration point to Aschaffenburg, Bavaria or Obergimpern, Baden. She matches my father and I on some good-size, organized sub-segments of this admixture in the chromosomes.
    Last edited by randwulf; 12-17-2016 at 09:24 PM. Reason: Fixed some typos...

  12. The Following 3 Users Say Thank You to randwulf For This Useful Post:

     anglesqueville (12-17-2016), Huijbregts (12-17-2016), Radboud (12-17-2016)

Page 11 of 47 FirstFirst ... 91011121321 ... LastLast

Similar Threads

  1. Davidski's Basal-rick K7 + Global 10 Genetic Map Results
    By khanabadoshi in forum Autosomal (auDNA)
    Replies: 775
    Last Post: 09-16-2017, 11:31 AM
  2. Replies: 327
    Last Post: 03-29-2017, 07:24 PM
  3. Replies: 6
    Last Post: 03-02-2016, 01:00 PM
  4. An alternative for 4Mix
    By Huijbregts in forum Autosomal (auDNA)
    Replies: 95
    Last Post: 02-13-2016, 11:43 AM
  5. Replies: 8
    Last Post: 07-06-2014, 09:02 PM

Posting Permissions

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