Announcement

Collapse
No announcement yet.

Maternal ancestral origins

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

  • Maternal ancestral origins

    Hi everyone

    I recently did a mtDNA+ test and got the attached results. The situation with my family tree is that I have managed to trace my maternal line back to a lady called Nelly Miller who was born in 1810 in Hobart Tasmania, Australia, just a couple of years after Hobart was settled.

    I am unable to trace back any further hence why I did the mtDNA test.

    I realise that I can upgrade my test to a full MtDNA test which will improve the resolution, however for the time being does the Ancestral origins listing tell me anything about which country or countries my maternal line may have come from?

    My Haplogroup is H by the way.

    Thanks Peter
    Attached Files

  • #2
    Hello again.

    I finally got the Advanced matches feature to work which has helped somewhat.

    I did a search on all people who match all of autosomnal as well at HVR1 and HVR2. This gave me a small number of positive results.

    At this point I would appreciate some advice regarding the above search. In order to trace back past my great great great grandmother I need to find matches with a common ancestor a little further back but not too far back. For example if I could find someone with a common ancestor who was Nelly Millers mother this would assist me.

    The shortlist of matches found where all of autsomnal, HVR1 and 2 all match, if I understand the theory correctly, should be people with a common maternal ancestor to me but reasonably recent common ancestor. That's because the autosomnal data should find only recent generations of people with matches of autosomnal dna and the HVR 1 and 2 matches should be people with a common maternal ancestor.

    I looked though the results of my search above and also their trees but got nothing of use. Is it worth doing the full mtDNA test then because whilst it should more precisely match me with people with the same maternal dna, those people should show up on the combined autosomnal/hvr1 and 2 search that I did. I would be unlikely to get any new matches in recent generations.

    Is this correct?

    cheers Peter







    Comment


    • #3
      For mtDNA Haplogroup H, it is a fairly common haplogroup in Europe.
      It has many subclade branches which would be more Genealogically relevant.
      In my opinion, especially for Haplogroup H, you will need to do Full Sequence if you wish to find matches/clues to your maternal mtDNA line.

      The main branch H (old parent branch of all H subclade branches) was formed roughly 20,000 to 25,000 years ago, that is alot of people you will match with on just the HVR1 and HVR2 level.
      Many matches on these levels will beyond a genealogical timeframe

      There is only a 50% confidence level that a HVR1 match is within 1300 years (52 generations), and a 50% confidence level that a HVR2 match is within 700 years (28 Generations)
      https://www.familytreedna.com/learn/...-relationship/

      https://www.eupedia.com/europe/Haplogroup_H_mtDNA.shtml

      Haplogroup H subclades (all which will test positive for SNP which classifies Haplogroup H )
      Scoll down until you hit the Haplogroup H branch and it subclades, you will see there are many possible branches that you may fall into
      https://www.phylotree.org/tree/R0.htm

      Comment


      • #4
        Thanks Prairielad

        I think you may have missed my point. Maybe if I explain it a different way.

        I have H haplogroup and basically you are saying that by getting a full test I can find the particular branch I fall into i.e. my subclade.

        Lets say for arguments sake I fall into subclade H1at1 and that my matches page on a full test identifies 10 people who we will call a,b,c,d,e,f,g,h,i and j. And lets say the first 7 are from over 500 years ago and the last 3 are from recent generations.

        My first question would be - will all of persons a to j appear in my current Hvr1 and 2 list or not? I think they will because my understanding is that my H1at1 matches are a subset of my H matches.

        Assuming I am correct then, as general proposition, is the following correct - autosomnal matches above a certain threshold are from recent generations eg last 300 years? I believe so because with longer periods the autosomnal dna gets spread amongst more descendents.

        So far so good. It would follow then that my more recent mtDNA matches (to my subclade even if not known) would also be autosomnal matches to me whereas my mtDNA matches with common ancestors further back would be less likely to be autosomnal matches of any significance.

        So if we know or presume that my match population from my H population (according to the combined HVR1 and 2 tests) includes my more recent maternal ancestors.

        And I know or presume that my recent direct maternal line ancestors also appear in my autosomnal results.

        Then if I make a list of persons who appear in all HVR 1 and 2 and autosomnal match results, then the resulting list will include my more recent direct maternal ancestors (if they have been tested). It will also include false positives but the population will have been considerably narrowed down nevertheless.

        I think my logic is right but I am happy to be corrected.

        I would be grateful if someone who has done the full test could try the following. First get your list of subclade matches - we'll call that list A. Then using the Advanced matching tool, find all the people who match HVR1, HVR2 and autosomnal dna. We'll call that list B. Now according to my theory nearly all the people in List A should also appear in list B. Is that the case and if not what percentage did?

        Additionally you might ask - why do this?

        My logic is as follows. Matching precisely by one's subclade will narrow down the population considerably. But for two people to be precisely matched both must have done the full test. What about those people in my H population who haven't done the full test? I can match them generally but cannot be sure that the common ancestor is recent or thousands of years ago. By using my technique of HVR1/2/autosomnal matching I can be reasonably sure that the common ancestors of those matches will be relatively recent AND that they will be related to my maternal line.

        Is my logic correct or am I missing something ?

        Thanks Peter


        Comment


        • #5
          MtDna goes back "forever".

          But, autosomal, atDna, as advertised, only goes back ~ 5 generations; some opin slightly further back. Please remember atDna is ALL the dna from your mother AND your father, and not just lineal.

          I mtDna tested in 2012, and I have found no correlation to my autosomal testing. But, what I find interesting in my closest matches is that many are from Scandinavia. This is NOT reflected in my atDna.

          HVR 1 and 2 are meaningless in isolation. Even Full mtDna testing is limited. MtDna is good to rule in, or out, *Native American heritage on the strictly maternal line, matrilineal. * (NAs have their own haplogroups)

          These are just quick thoughts. Others at the Forum can add to this.
          Last edited by Biblioteque; 8th October 2019, 09:48 AM.

          Comment


          • #6
            Peter, I have FullSequence results and tried your suggestion. My mtDNA full sequence haplogroup is K1a3.

            List A: I have 25 matches total for my mtFullSequence (HVR1, HVR2, Coding Region). This includes zero matches at 0 genetic distance, two matches at GD 1, nine matches at GD 2, and fourteen matches at GD 3. As a matter of fact, I received my second GD 1 FullSequence match just recently - with no tree or any other information.

            List B: I then opened a window to do Advanced Matching: as you suggested, I selected HVR1, HVR2, and Family Finder. You didn't mention if you chose "Yes" or "No" to match all the selected tests. At first, I chose "Yes" for "Show only people I match in all selected tests." There were no matches that met all conditions. Then I tried it by selecting "No" for matching all tests. This time I got a list of 222 pages of those who met any of the conditions, so did a search by last name for all the 25 Full Sequence matches. Four of my Full Sequence matches did not take the FF test. Of the 25, two do not show up in the search for any of HVR1, HVR2, or Family Finder. 2/25, or 8%, do not match the Advanced Matching parameters. Those two had also taken the Family Finder test, so apparently they don't match me there, as well as not matching in HVR1 or HVR2.

            So - this would appear to be within your theory limits. But, I don't think this proves what you want it to prove: that you "can be reasonably sure that the common ancestors of those matches will be relatively recent AND that they will be related to my maternal line." My two matches (who do NOT match any of HVR1, HVR2, or Family Finder) both were at GD 3 at the Full Sequence level. This does not mean any common ancestor with them would necessarily be distant, as they could be recent, but have a heteroplasty or some other reason for not being a closer GD match. In order to be a Full Sequence match, you must also match at HVR1 AND HVR2.

            You can't really use mtDNA matches the way you use autosomal matches (Family Finder at FTDNA). With FF matches, you are "fishing" for cousins; the more DNA they share with you, the closer the relationship. But with mtDNA, even your close genetic distance matches, 0 or 1, might possibly be related long ago. "Fishing" is not generally productive. The best way to use mtDNA is to solve a particular question involving two females, to prove they could be related. A good example can be seen in the YouTube video, "DNA Stories: A Tale of Two Sisters featuring Bill Hurst" (Mr. Hurst is the administrator of the mtDNA K haplogroup project at FTDNA).

            FTDNA says on their "mtDNA (Maternal Lineage) Tests" page (my bolding):
            Your mtDNA may help you find genetic cousins along your direct maternal line. When we look at your mtDNA results, we look for differences in your mtDNA caused by small changes — copying errors — that are called mutations* or polymorphisms. We compare those differences to the differences of other people in our database. The range of possible generations before you share a common ancestor with a match is wide. Your mtDNA exact matches may be recent, but they may also be hundreds or thousands of years in the past.
            and
            The wide range in the test results does not prevent your results from being useful. You can use this clear maternal line to provide evidence to support a relationship. You first trace two or more female lineage descendants of a single woman utilizing traditional genealogy research. The descendants then test their mtDNA. If they are exact matches, it is evidence that supports the relationship. Not matching usually disproves the relationship.

            Planned comparisons are the best choice. However, you can still find your common ancestor with a match. To do so, use your known maternal genealogy. For each match, look for common geographic locations on the direct maternal line. Work through each of your ancestors on this line as well as their daughters, their daughters’ daughters, and so forth.

            Comparing genealogical records is vital when using mtDNA matching to help you in your research. You need to enter all that you know about your direct maternal line in your myFTDNA account.
            You may want to read Roberta Estes' blog post, "Concepts: Genetic Distance"; she covers genetic distance for both Y-DNA and mtDNA in the post, so scroll down to read the mtDNA section. A key quote from her post:
            Unfortunately, there is no “rule of thumb” for mitochondrial DNA genetic distance relative to years and generations distant.
            Last edited by KATM; 8th October 2019, 03:17 PM.

            Comment


            • #7
              Thanks for the replies.

              KATM when I do an advanced search and select HVR1, HVR2 and autosomnal then tick the box Yes for 'show people who match for all selected tests' I get 30 results.(Amazingly several of the matches are my closest autosomnal matches and are known to be second cousins on my fathers side of the family!) Go figure.

              Do take note though that the Advanced Matching search tool found under mtDNA doesn't appear to work properly whereas the one listed under Family Finder works fine.

              cheers Peter

              Comment


              • #8
                Peter, the Advanced Matching tool is the same page at the same URL address, whether you click on the link under Family Finder or mtDNA results. It should not work differently depending upon which link you click, so that is a puzzle. I just tried each, and both worked, although there was a passing glitch with one. You can try reloading and see if that helps. Also, one note: there is no "n" in "autosomal."

                As far as your Advanced Matching results, I think everyone's results are unique. It all depends on your particular ethnicity mix, haplogroup(s), and the factor of whether or not people who would match your DNA have tested yet, or at which company they may have tested. Thus our "matching mileage" may vary depending upon those factors. You got 30 results when ticking the "Yes" box, and I got none. Others may get some amount higher or in-between our results using the same choices.

                Comment


                • #9
                  Originally posted by KATM View Post
                  Peter, the Advanced Matching tool is the same page at the same URL address, whether you click on the link under Family Finder or mtDNA results. It should not work differently depending upon which link you click, so that is a puzzle. I just tried each, and both worked, although there was a passing glitch with one. You can try reloading and see if that helps. Also, one note: there is no "n" in "autosomal."

                  As far as your Advanced Matching results, I think everyone's results are unique. It all depends on your particular ethnicity mix, haplogroup(s), and the factor of whether or not people who would match your DNA have tested yet, or at which company they may have tested. Thus our "matching mileage" may vary depending upon those factors. You got 30 results when ticking the "Yes" box, and I got none. Others may get some amount higher or in-between our results using the same choices.
                  Thanks Katm

                  I just tested it and its working now. Maybe just needed the refresh.

                  cheers Peter

                  Comment

                  Working...
                  X