Announcement

Collapse
No announcement yet.

Dr McDonald

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

  • #16
    Originally posted by bkilpatrick View Post
    I must have caught him at a good time. It was a few weeks ago. I had a response within a couple of hours.

    He even bore with me as I struggled to download the raw date properly. Had to change the settings on my MAC so it wouldn't unzip the file before sending! So nice and helpful.

    He was fairly brief and to the point when it came to the results, and I didn't get a percentage breakdown for the various European components, as some people do. What I did get was even better: he identified this tantalizing bit of Native American (below the PF threshold for FTDNA to report, I guess) with an explanation that it could also reflect Finnish/Sammi. So that opened up a whole new avenue for me.

    Hope you get results soon. But it's worth the wait, if that is what it takes.
    Could you please tell me what settings you had to change on your MAC.
    I have one too and can't get my raw data downloaded
    Many thanks

    Comment


    • #17
      Firefox.

      Preferences

      General Preferences

      Download

      "Always ask where to download" option and that way you'll have a chance to say don't unzip, then hit ok. I'm sure there's an easier way but this will get you started.

      MD

      Comment


      • #18
        Hi,

        Problem, as Dr. McDonald explained, was that MAC might be automatically unzipping files before sending. So I had to hunt around to figure out if this was the case and how to change it.

        Here's what I did, with the browser I use, Safari:

        Go to Safari, open preferences, and then see if there is a check box before "open safe files after downloading."

        If there is, you need to uncheck it!

        After that, it worked just fine. (After sending the files, I re-checked it.)

        Hope this helps!

        Blair

        Comment


        • #19
          How far back can Dr. McDonald's analysis go to find suspected NA ancestry? I have a 5th great-grandmother who is supposedly NA, but my PF results are 100% Western European (Orcadian) with 0.01% margin of error. The rest of my ancestry is Scots, English and German. A previous test by AncestryByDNA gave me 89% Caucasian and 11% Sub-saharan African, which seemed pretty random, though my grandmother was adopted and there was some question about her father (I found the birth mother, but she had been separated from the man she put on the birth certificate as the father for nearly a year before my grandmother was born). Could Dr. McDonald give me a breakdown between the Scots, English and German sub-populations of the Orcadian?

          Comment


          • #20
            Well, he found 0.8% Native American ancestry in my results, which he estimated to be from the time of Columbus (500 years old or so) so that's well beyond 5 generations.

            Comment


            • #21
              Originally posted by ahernandez View Post
              Well, he found 0.8% Native American ancestry in my results, which he estimated to be from the time of Columbus (500 years old or so) so that's well beyond 5 generations.
              I'm interested to hear how you interepret the Mozabite: do you think this is actual North African, or, say, something like a West African/East African/Near Eastern mix that averages out to be Mozabite?

              Comment


              • #22
                Originally posted by Javelin View Post
                I'm interested to hear how you interepret the Mozabite: do you think this is actual North African, or, say, something like a West African/East African/Near Eastern mix that averages out to be Mozabite?
                I interpret Mozabite as pretty much general North African (like the Native Population of the Magreb, the Amazigh people)and it's confirmed with Y and MT tests my extended family has done.

                For example, all my grandparents are Cuban of Spanish/Canary Islander descent (with a bit of NA thrown in apperantly) my maternal grandfather is E1b1b which is a North African/Berber Haplogroup. I think it's safe to say it's the result of the 700+ years of Moorish occupation of Iberia. My partnal side is from the Canary Islands and the Guanches (Canary Natives) were genetically related to the Berbers of Morocco and my paternal great grandmother's MT haplogroup is U6B, which is consistant with North Africa.

                It's funny you can almost see the history of invasion with the Y and MY haplogroups within my family. Those from the mainland: the women have European haps and the men have North African haps (North Africa invaded Iberia) with my Canary Islander family: the women have North African haps and the men have European haps (Europ invaded the CIs.)
                Last edited by ahernandez; 12 April 2011, 06:00 PM.

                Comment


                • #23
                  Originally posted by ahernandez View Post
                  Well, he found 0.8% Native American ancestry in my results, which he estimated to be from the time of Columbus (500 years old or so) so that's well beyond 5 generations.
                  Thats pretty amazing for 500 years.
                  Must be extreme luck.

                  Thats 20 Generations.

                  Mathematicaly
                  13 Generations (325 years) = 0,000122070312%

                  Sorry, anything longer ago than 13 generations wont show up as a normal number on my calculator ^^

                  I wonder how Dr McDonalds estaminates these 0,8% as "from the time of Columbus".

                  0,8% would be "Normal" for

                  1 GEneration (25 years)= 50%
                  2 Generations (50 years) = 25%
                  3 Generations (75 years) = 12%
                  4 Generations (100 years) = 6%
                  5 Generations (125 years) = 3%
                  6 Generations (150 years) = 1,5%
                  7 Generations (175 years)= 0.8%
                  8 Generations (200 years) = 0,4%

                  By the most simple formula of how aDNA is inheritated.

                  Comment


                  • #24
                    Hah, I don't know how he came up with the timeframe, maybe he was just throwing out a historical event for context, but your mathematical "norm" is off too for us, in 175 years there's only been 4 generations of my family not 7.

                    So it's all just estimates in the end.

                    Comment


                    • #25
                      Originally posted by ahernandez View Post
                      but your mathematical "norm" is off too for us, in 175 years there's only been 4 generations of my family not 7.

                      So it's all just estimates in the end.
                      Yeah. But then just look at the number of generations.
                      Theoretically the aDNA reduces itself to 50% in every generation.

                      I used 25 years for 1 generation, wich may be somewhat short.

                      Its just that I somewhere read that in earlier times, people had children with much younger years than today.

                      I read that in the late 19th century German women had their first child in the average age of 23.
                      In the 1960s this was 26, in the 1980 it was 27 and now its close to 28.

                      If I calculate yours, it means, your people do childs by the age of 44 in general, wich is for sure somewhat unusual too.

                      In the tribal age of the Germans (1. century for example) people married by the age of 12, had been rated as "fully adult" by 15 and rarely lived bejond 35.

                      Comment


                      • #26
                        Just got my results from Dr. McDonald, and I'm very grateful he had the time to review my data. This is what he sent me:

                        "You test best as (by hand)

                        32.7% Irish
                        44.4% Maya
                        8.3% Na-Dene
                        14.6% Ethiopian

                        or exactly as good

                        24.1% English
                        14.6% Jewish
                        42.1% Maya
                        11.3% Na-Dene
                        8% Mandenka"

                        As you can see, the 32.7% Irish is pretty close to the 24.1 + 14.6% English/Jewish percentages from the second set. My Illumina results seemed to absorb what Affy had reported as Middle East/North Africa.
                        I think there's that trend I had seen, matching results with family tradition -- Native American, European, Middle Eastern, African, and Asian. He didn't list South Asian, but that is my interpretation of the Na-Dene result, that it is Asian or Eurosian as I have no Na-Dene descent -- My NA ancestry is most likely from a tribe wholly unrelated to other tribes in Mexico and the US SW.

                        Comment


                        • #27
                          Daniel72 - Oh, I know. I agree with what you're saying. If I were going to do an average, I'd go with 25 years for generations too. I'm just saying McDonald probably meant it to be "very long ago" and not literally 1492.

                          Another explenation is that it didn't dilute in a neat step-by-step downward path. When Spanish women were allow over the NA percentage diluted and then in the mid 19th century Cuba got a massive wave of Canary Islander immigrants that further added Euro genes to the mix. For a few generations the percents probably held, so you could have somthing like this:

                          1 Generation (25 years)= 50%
                          2 Generations (50 years) = 50%
                          3 Generations (75 years) = 25%
                          4 Generations (100 years) = 25%
                          5 Generations (125 years) = 25%
                          6 Generations (150 years) = 12.5%
                          7 Generations (175 years)= 6%
                          8 Generations (200 years) = 3%
                          etc

                          For example, my ME genes are from much further back than 5 generations, but after all these centuries it still holds at 20-some-percent because everyone in my family to some extent carries North African genes so it hasn't diluted as much.
                          Last edited by ahernandez; 12 April 2011, 07:15 PM.

                          Comment


                          • #28
                            Your SNPs reflect the populations and migrations of hundreds of years ago. The FTDNA FAQ is actually misleading. 500 years is very possible if that's what he told you.

                            Your recent ancestry is a reflection of its past history of the group you match. Your SNPs match the groups with supposedly no changes for around 500 years. That's what's supposed to be in the sample groups anyway. I doubt it always is.

                            SNP Segments last about 10 generations but SNPs are always there, constantly being mixed back and forth.

                            MD.
                            Last edited by mkdexter; 12 April 2011, 07:34 PM.

                            Comment


                            • #29
                              Your SNPs reflect the populations and migrations of hundreds of years ago. The FTDNA FAQ is actually misleading. 500 years is very possible if that's what he told you.
                              Well, of course.
                              But then, this is "normal" for almost the whole nation.

                              For example, Davidski calculated a possible 24% of Balto-Slavic admixture in an average German. And 10% Italian.
                              (was a try to check if aDNA tells a similiar story than Y-DNA)

                              Thats of course not caused by admixture from 5 generations ago but from 1500 or 2000 years ago.

                              But then again, its "normal" for a German to have 24% Balto-Slavic admixture and so it shouldnt show up as "admixture" at all, because hovering around 24% Balto-Slavic admixture is what is "German" atm. And what makes Germans different from say English.

                              Thats like Middle Eastern Admixture in Spanish or Italians.
                              Thats pretty ancient admixture that ALL of them possess. Thats not what we want to know.

                              And yes, I agree. If its "normal" for a whole nation to possess 1% of native american "blood", then this will last centuries or milenia around 1%.

                              Thats like my 1,5% North-East Asian, wich for sure is not from a full blooded Sibirian anchestor of 7 Generations ago but from my mysterious North_East European GGGMom I guess, who again may have an Sibirian anchestor from 4000 years ago.
                              Last edited by Daniel72; 12 April 2011, 08:33 PM.

                              Comment


                              • #30
                                For sure. A Roma person will see South Asian that is well over 500 years old, for example. It depends on may factors.

                                @ahernandez
                                That's pretty good, then, North African seems quite accurate. It's hard to get admixed people right but it seems to be working in your case.

                                @Daniel72
                                I see 30 years commonly given as average for generation. Let's say ahernandez is 30. That takes us back to 1981. The very brute calculation of probability would be something like:

                                1 Generation (1951)= 50%
                                2 Generations (1901) = 25%
                                3 Generations (1876) = 12.5%
                                4 Generations (1851) = 6.25%
                                5 Generations (1826) = 3.125%
                                6 Generations (1801) = 1.5625%
                                7 Generations (1776)= 0.78125%
                                8 Generations (~1750) = 0.390625%

                                But this assumes a single ancestor, exactly symmetrical inheritance all the way, nobody appearing more than once in a pedigree, etc. It's important to look at the possible range at each generation, not just brute probability. For example, he might have inherited 30-35% of his genes from one grandparent. With each generation the inheritance becomes more random.

                                I have traced two different blocks of about 12 centimorgans each to two different ancestors each born about 1745. These ancestors, several countries apart, happened to pass genes on to me. Many others did not -- most, at that level. It is completely credible that somebody could retain measurable Native American ancestry from 500 years ago at the levels mentioned.

                                Comment

                                Working...
                                X