Announcement

Collapse
No announcement yet.

Surprised by results

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

  • Surprised by results

    First reference population:British
    Second reference population: German

    Fair enough-this makes some sense.

    However Northern European 45 Mediterranean 36 South West Asian 17 Native American 2


    Family tree research so far indicates English/Scottish/Irish ancestry so rather surprised by the rather high(?) South West Asian and any Native American.

    Results seem at odds with family finder.


    Neanderthal 2.0 Denisovan 1.5

  • #2
    gatty congrats on your results. Thanks for sharing. I noticed in your PF signature you do have the Finnish, Russian results listed. I wonder if this was your clue to NA ancestry. Other threads have talked about the connection of NA ancestry and getting the Russian, Finnish results in PF.

    Comment


    • #3
      My brother Jon's results came in today.
      Northern European 42%
      Mediterranean 40%
      South West Asian 17%
      Neanderthal 2% Denisovan 1.5%

      His population finder results for here are:
      Europe (Western European) French, Orcadian, Spanish 92.58% ±7.51%
      Europe Tuscan, Sardinian 7.42% ±7.51%

      First reference population Danish
      Second reference population Tuscany

      Our known ancestry is English/Scottish/German back to 1600.

      Donna

      Comment


      • #4
        Here,cut and pasted is the reply i got from Dr McDonald which seems to refute the Native American component.

        N49571-autosomal-o-results.csv
        Most likely fit is 89.8% (+- 9.6%) Europe (all Western Europe)
        and 10.2% (+- 9.6%) Europe (various subcontinents)
        which is 100% total Europe

        The following are possible population sets and their fractions,
        most likely at the top
        Spain= 0.164 Irish= 0.836 or
        French= 0.292 Irish= 0.708 or
        Irish= 0.865 Tuscan= 0.135 or
        English= 0.875 Russian= 0.125 or
        Irish= 0.844 Italian= 0.156 or
        English= 0.867 Finland= 0.133 or
        French= 0.669 Finland= 0.331 or
        English= 0.932 Chuvash= 0.068 or
        Irish= 0.418 English= 0.582 or
        Irish= 0.928 Sardinia= 0.072

        Pop Finder (which is an early version of my program) got it right.

        It is essentially English, but with some Continental European compensating for
        a substantial Irish. Geno2 is over-interpreting the data. Yes, I see a small Native American
        on chromosome 11 if I set my program to "speculative", which I consider to be "in the noise".

        Comment


        • #5
          I am rather surprised to read multiple Geno 2 participants having such high
          South West Asia and South Asian percentages!

          Comment


          • #6
            Originally posted by Donald Locke View Post
            I am rather surprised to read multiple Geno 2 participants having such high
            South West Asia and South Asian percentages!


            I notice that McDonald make no mention of the 17% South West Asian component. Is this because it goes back beyond the period his analysis tests for or is this another over interpretation by Geno2?

            Comment


            • #7
              yayy I've been waiting for 2 months and a half and still nothing, I hope it'll be beter than yours

              Comment


              • #8
                Very prompt reply back.
                I notice that you make no mention of the 17% South West Asian
                component. Is this because it goes back beyond the period your analysis
                tests for or is this another
                over interpretation by Geno2?


                Neither, really. You will note the fit with Irish-Tuscan. You fit almost
                as well with such things as Irish plus 10% Jewish or Irish plus 5% Bedouin.
                My system takes into account geographic distances in certain cases,
                this being one, on the basis of probabilities. I also look at special tests
                for Jewish, Druze, Moroccan/Berber and Bedouin. You test definitely no
                Druze or BEdouin, and very very iffy for Jewish and Moroccan, so I discount the
                Mideast. Besides, I see no Mideast on the chromosomes even at speculative.

                So Geno2 is making a mistake that requires intelligence beyond that of a machine
                to prevent. I wrote FTDNA's Family finder ... it uses an earlier version of
                the "lists" I send ... and the computer tries very hard with numerous tests,
                but its not as good as a human.

                Doug McDonald

                Re Gedmatch highest frequency for West Asian is Dodecad v3 at 6.37. Others- West Asian 5.80(mdlp w -22)
                South West Asia 2.07(Eurogenes k-13) South West Asian 1.40(Harrapaworld) West Asian 4.00(Eurogenes eu)
                West Asian 3.43(Eurogenes j test)

                All significantly below 17%. If Dr McDonald is right(i'm not discounting possible bias towards his
                own system) then the Geno2 has done quite a poor job/his differing conclusions prove it's quite an inexact discipline.

                Comment


                • #9
                  Originally posted by Donald Locke View Post
                  I am rather surprised to read multiple Geno 2 participants having such high
                  South West Asia and South Asian percentages!
                  The Geno test takes a look at your global ancestry from a different time period than other whole genome tests. Most tests take a "snapshot" of an individual's ancestry about 500-1000 years ago. The Geno test attempts to look far deeper, more like 5,000-10,000 years ago or so.

                  Comment


                  • #10
                    Just remember all these companies use different populations and algorithms. Geno is supposed to be very deep ancestry. Most have not been impressed from what I have been able see on the forum in regards to the Geno test.

                    Dr. McDonald is good at finding NA, as he has samples that others don't. Most are satisfied by his analysis. I have read on another forum that he is not so great with distinguishing Northern Europeans. And I for one could have cared less for his analysis of me. I was basically told that my DNA could fit in any European country. His margin of error for me was higher than it is here.

                    I personally prefer the admixture tools on Gedmatch. Everyone's opinion on this will vary.

                    Comment

                    Working...
                    X