Announcement

Collapse
No announcement yet.

Population Size of Deepest Subclade

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

  • Population Size of Deepest Subclade

    Is anyone estimating the likely number of men in the deepest subclades? For example, the R1b subclades are now extending down to about ten more levels, with added levels being added frequently. Each lower level subclade will include fewer than the number in its parent subclade. At some point a person in the deepest subclade will share all the same SNPs with only a small number of people. Where are we so far in that progression of events?

  • #2
    How about this estimate?

    I've given this a little more thought after raising the question. For lack of real numbers, let me make some rough estimates:

    1. The FTDNA website says they have more than 220,000 Y-DNA records in the database.
    2. Assume that the R1b1s represent 50% of the database, or 110,000 men.
    3. One of the deepest R1b1 subclades in the 2012 Y-haplotree is R1b1a2a1a1b4h2a1a (L627), which is 13 levels down from R1b1.
    4. Assume that each lower level subclade reduces the number of people in the group by half (this is a crucial figure). So thirteen lower levels of subclade (down from R1b1) would reduce the number by (1/2)exp13, or 1/8182---say, about 1/8200.
    5. So of the 110,000 R1b1 Y-DNA records at FTDNA, we might expect only about 110,000/8200, or maybe a dozen men in subclade R1b1a2a1a1b4h2a1a (L627). That would seem to be a small enough group for a "family" reunion.
    6. The dozen men in that subclade would represent 12/220,000, or about 0.005% of the database. So if the FTDNA database is representative (which I doubt), we might expect a worldwide population of 0.005% of 3 Billion men, or 150,000 in that subclade.

    That's getting to be an interesting number now. We should be able right now to get our arms around all the men in the FTDNA database who share the same lowest-level subclade. In a few years maybe we can get our arms around the worldwide group that will have been narrowed down to a manageable size.

    Comment


    • #3
      6. The dozen men in that subclade would represent 12/220,000, or about 0.005% of the database. So if the FTDNA database is representative (which I doubt), we might expect a worldwide population of 0.005% of 3 Billion men, or 150,000 in that subclade.
      I think your logic here goes a little to far.
      Your sample set is American men who enjoy DNA testing which is probably an extremely small set of men when compared with the world population .

      Comment


      • #4
        Is the logic faulty?

        I think the primary uncertainty arises in my assumption that each clade has two subclades of equal size. If that is a reasonable assumption, then it doesn't matter how many men are in the highest level clade. The subclades down the line continue to be cut in half for each level. I used the R1b1 subclades because I just spotted some with thirteen levels beyond the R1b1. If you start with 8182 people and cut out half, and do that thirteen times, you will end up with one. The rest of the argument doesn't matter. Pretty soon the subclades will be small.

        Comment


        • #5
          Originally posted by Allen Blancett View Post
          my assumption that each clade has two subclades of equal size.
          Not at all. Look at the root of the tree. there are essentially two mega-clades, yDNA A, and BT -- that is, everyone else, including almost all non-Africans and the majority of Africans. The same applies right down the tree. Some subclades at the same level are much more populous than others -- and some better or more poorly represented in the FTDNA database.

          Comment


          • #6
            Allan is discussing something that I, too, have pondered. By the end of the decade, we could have an enormous y-DNA tree that terminates in many thousands of tiny subclades, all defined by an SNP, many with a MRCA in the last 500 or 1,000 years.

            Once things get this refined & y-DNA participants have all tested to this level, we can simply say that if two men share the same terminal SNP, they have a MRCA within the last whatever (500 or 1,000 years) & if they don't share the same terminal SNP, their MRCA must be a lot earlier.

            This will take the science of genealogical comparisons to a level that has never been seen before.

            Timothy Peterman

            Comment


            • #7
              We are already almost at the point where terminal SNPs are 1,000 years old.

              In fact, with the entire Y sequenced, most men in the world might have a private SNP.

              At that point STRs will become almost useless because SNPs will be able to chart all humanity so that haplogroups will no longer be so much "R-Z999" or "R-L999" as "R-Mr. George Fitzsimmons Smith of Lexington, Kentucky, and not his brother Frederic W. Smith."

              Comment

              Working...
              X