Announcement

Collapse
No announcement yet.

Pseudoautosomal regions

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

  • Pseudoautosomal regions

    We usually talk about the Y chromosome as something very different from the X. But, if I have it right, there are 2 (or now 3) small pseudo-autosomal regions on the Y that are homologous to the X, and can bind together and crossover. I wonder how this is handled in the new X matching. There are probably more questions than these but I have these.

    1) Do the pseudo-autosomal regions of the Y get used in the X match? That might mean a male had 2 values of a SNP and would match people his mother did not because of that.
    2) I believe this means there are cases where a female will get an allele from their father's Y chromosome and a male can get an allele from their father's X chromosome. How does this affect matching, and our logic in using the matches?

  • #2
    Originally posted by JohnG View Post
    We usually talk about the Y chromosome as something very different from the X. But, if I have it right, there are 2 (or now 3) small pseudo-autosomal regions on the Y that are homologous to the X, and can bind together and crossover. I wonder how this is handled in the new X matching. There are probably more questions than these but I have these.

    1) Do the pseudo-autosomal regions of the Y get used in the X match? That might mean a male had 2 values of a SNP and would match people his mother did not because of that.
    2) I believe this means there are cases where a female will get an allele from their father's Y chromosome and a male can get an allele from their father's X chromosome. How does this affect matching, and our logic in using the matches?
    Great question John. Another poster on the forum has brought this up in the past, and has had me wondering about this from time to time. It seems to me like it might be possible to transfer this data on from say father to daughter. Really though I have no clue. I hope one of the experts will weigh in on this. Even if it is something that really has not been studied enough to give an exact answer. I for one would like to know where the science currently stands on this.
    Last edited by Táltos; 4th January 2014, 09:59 PM.

    Comment


    • #3
      I did the quick Google, Wikipedia and a list of papers look last night and it seems 30 years ago folks would have said there was no crossover, but now there are these 2, and in some people 3, areas that behave like autosomal DNA, not sex linked.

      I imagine FTDNA's X match has produced lots of interesting data, some of it getting reported by users as puzzles.

      Comment


      • #4
        Originally posted by Táltos View Post
        Great question John. Another poster on the forum has brought this up in the past, and has had me wondering about this from time to time. It seems to me like it might be possible to transfer this data on from say father to daughter. Really though I have no clue. I hope one of the experts will weigh in on this. Even if it is something that really has not been studied enough to give an exact answer. I for one would like to know where the science currently stands on this.
        Hi,

        I am consulting with Dr. Mittelman and a few others on this. I suspect we are going into territory --again-- where citizen science is about to be ahead of academics.

        Comment


        • #5
          I guess one simple question is if the pseudoautosomal regions on the Y are in the count for matching. If so, wouldn't that mean the equivalent regions on the Y are too?

          Comment


          • #6
            23andMe reports XY SNPs on the X chromosome, with SNPs covering the territory at the tips of the chromosome, from positions 1 to 2,697,868 and 154,939,018 to 155,236,747.

            It appears that FTDNA is not including XY SNPs in the X file, which starts at 2,700,157 and ends at 154,916,845. (The SNPs are in alphabetical order by rs number, so you need to sort by position.)

            AncestryDNA reports XY SNPs on chromosome "25", which may be the way the Illumina base calling software works. IF so, it's possible that FTDNA could report those, but they're small potatoes compared to the total size of the X.

            In any event, it doesn't sound like the pseudo-autosomal region is impacting the X matches here.

            Comment


            • #7
              • I am sorting through the X matches of known family members and may have found an example of this pseudo-autosomal Y crossover , or of insights the X match can provide, or just a distant mystery.
              • My first cousin once removed (EG) has an X match with her half 2nd cousin twice removed (ML) - one segment of 2.05 cM. On the autosomal matching, EG and ML match for a total of 42 cM with an 8 cM longest shared segment. ML also matches EG first cousin (AM) and EG's 1st cousin once removed PJ. ML does not match 3 others that are by paper records her half 3rd cousins once removed. So ML matches 50% of what are roughly 4th cousin to 5th cousin type relationships. OK for 4th, amazing for 5th.
              • The common ancestor is W. W had three wives. No children by the first. ML descends from W and his unknown by name 2nd wife. EG, AM and PJ descend from W and his third wife, through a child born after W died. So it was some reassurance to match ML, proving the parentage of that posthumous child.
              • Here is the question -
              • EG's line back to W is: Her mother, her maternal grandmother, and W is her maternal maternal great grandfather.
              • ML's line back to W is: Her mother, her maternal grandmother, her maternal maternal grandfather, her maternal maternal paternal grandfather and then W her maternal maternal paternal paternal great great great grandfather. So two generations of Y with no X.
              • One thought was that perhaps the unknown by name 2nd wife was a sister of the third wife. That would help account for three matches to descendants of the third wife. But that would still mean one generation of father to son transmission of this 2.05 cM segment.
              • The pseudoautosomal regions are called for short PAR1 and PAR2, and if you are in the 2% that have the third, PAR3. My expertise from Wikipedia, is that PAR1 is 2.6Mbp of the 155Mbp of the X and the 59Mbp of the Y. PAR2 is much smaller, 320kbp.
              • So the 2.05 cM shared between these two half second cousins twice removed is about right to be the PAR1. So one, hypothesis is this is the PAR1 passed down through some male generations (2 or 1, if the wives were related).
              • 2nd Hypothesis is that EG has dozens of other X matches with segments as long or longer than 2.05 cM, and this match with ML may just be like those, some currently unknown common ancestor many generations ago.

              Comment


              • #8
                OK, the match is not on the tips of the X, so not PAR1. Looks like Hypothesis 2 wins.

                Comment


                • #9
                  Originally posted by Rebekah Canada View Post
                  Hi,

                  I am consulting with Dr. Mittelman and a few others on this. I suspect we are going into territory --again-- where citizen science is about to be ahead of academics.
                  Rebekah and Ann Turner.. You are both awesome with your reports.. I am checking daily and sending others to look.. Thanks..

                  Comment


                  • #10
                    Originally posted by JohnG View Post
                    OK, the match is not on the tips of the X, so not PAR1. Looks like Hypothesis 2 wins.
                    Hi,

                    For such a small segment, I wonder if you are facing the proverbial possibility that the ancestor and all three of his wives shared common ancestry in pre-1790 Virginia.

                    Comment


                    • #11
                      Originally posted by JohnG View Post
                      I guess one simple question is if the pseudo-autosomal region are in the count for matching. If so, wouldn't that mean the equivalent regions on the Y are too?

                      JohnG and others.."pseudo-autosomal regions on the Y"?
                      Will something like that ever be informative for a female getting some information about paternal Y line?

                      Long story here but the short version is that Dad was an only child and his father from another country.. I tried to extract DNA from an object of Dad's at FTDNA some years ago.. Thomas Krahn helped with that ( it failed) and he once told me that SOME of the smaller X matches might some day be informative for that..

                      I tested the XSTRS ( both panels) for my Sister and I in 2007..had results in his database.. Just wondering..

                      Comment


                      • #12
                        Originally posted by Rebekah Canada View Post
                        Hi,

                        For such a small segment, I wonder if you are facing the proverbial possibility that the ancestor and all three of his wives shared common ancestry in pre-1790 Virginia.
                        Make that Massachusetts and it might be. Or New York.

                        I have another X match right now with a 6 cM common X segment between two 3rd cousins once removed who match on autosomal 65 cM total and 14 cM longest shared segment. But from the common ancestoral couple one is all daughters of daughters and the other is all sons of sons. So, the X match is not via that route. There is another common ancestor further back. I have another 3rd cousin once removed X match at 10 cM shared, so the 6 cM would likely represent several more generations back.

                        Comment


                        • #13
                          I have found a couple of cases where this might have happened (the crossing over from Y to X).

                          My paternal great grandfather, Josef Petermann, had a brother, Jakob Petermann. I had a granddaughter of Jakob tested for regular Family Finder purposes. Lo and behold, this granddaughter is on my X-match list. Since Jakob is on her mother's side, she could have inherited his X-chromosome without any crossing over.

                          I also had two great granddaughters of Josef Petermann tested, both my second cousins on different lines of descent, both descended from different sons of Josef. They also show up on the x-match list for the granddaughter of Jakob.

                          Makes me wonder if the Peterman Y-chromosome is more susceptible than normal to crossing over.

                          Timothy Peterman

                          Comment


                          • #14
                            Originally posted by T E Peterman View Post
                            I have found a couple of cases where this might have happened (the crossing over from Y to X).

                            My paternal great grandfather, Josef Petermann, had a brother, Jakob Petermann. I had a granddaughter of Jakob tested for regular Family Finder purposes. Lo and behold, this granddaughter is on my X-match list. Since Jakob is on her mother's side, she could have inherited his X-chromosome without any crossing over.

                            I also had two great granddaughters of Josef Petermann tested, both my second cousins on different lines of descent, both descended from different sons of Josef. They also show up on the x-match list for the granddaughter of Jakob.

                            Makes me wonder if the Peterman Y-chromosome is more susceptible than normal to crossing over.

                            Timothy Peterman
                            Hi,

                            How big are the segments and what was Jakob's ethnic background?

                            Comment


                            • #15
                              Originally posted by Ann Turner View Post
                              23andMe reports XY SNPs on the X chromosome, with SNPs covering the territory at the tips of the chromosome, from positions 1 to 2,697,868 and 154,939,018 to 155,236,747.

                              It appears that FTDNA is not including XY SNPs in the X file, which starts at 2,700,157 and ends at 154,916,845. (The SNPs are in alphabetical order by rs number, so you need to sort by position.)

                              AncestryDNA reports XY SNPs on chromosome "25", which may be the way the Illumina base calling software works. IF so, it's possible that FTDNA could report those, but they're small potatoes compared to the total size of the X.

                              In any event, it doesn't sound like the pseudo-autosomal region is impacting the X matches here.
                              I may need to backtrack on my statements here. I was basing my numbers on my son's transfer from 23andMe.

                              However, people with mother/child data at FTDNA see these numbers in the Chromosome Browser table

                              Segment boundaries: 1,370,495 to 154,570,039 (?build 36 numbers)
                              Number of SNPs 18,092
                              cM: 195.93

                              Could someone review a raw data download for the position of the 1st and last SNPs (build 36 and 37)?

                              The cM total appears to be based on female transmissions when I look at the Rutgers website. Rutgers reports sex-averaged rates for the autosomes, but not the X. Female rates are about 1.7 times higher for the autosomes and would be even more lopsided for the X, which stays intact every time it passes from a father to his daughter.

                              http://compgen.rutgers.edu/RutgersMap/MapBrowser.aspx

                              Comment

                              Working...
                              X