Loading ...
Sorry, an error occurred while loading the content.

Mutation rates and tMRCA again

Expand Messages
  • craig
    I downloaded YSearch 37-marker matches to our Gaston haplotype and combined the 64 records with 21 records for our Ulster lineage. I then calculated the GDs to
    Message 1 of 17 , Jul 3, 2011
    • 0 Attachment
      I downloaded YSearch 37-marker matches to our Gaston haplotype and combined the 64 records with 21 records for our Ulster lineage. I then calculated the GDs to the Gaston modal haplotype for all records. These ranged from 0 to 4 for the Gastons and from 5 to 10 for the YSearch matches (assuming equal mutation weights for each marker). Since we know that the Gaston common ancestor lived in about 1650 I estimated 12 generations. The average GD of the Gastons is 1.8. This gives .0044 mutations per generation per marker, which falls about half way between the Chandler rate of .0033 and the McDonald rate of .0049

      Next I calculated the average GD (from the Gaston modal) for all 85 records, giving 6.1 (first having eliminated the double counting of 389-1 and 389-2). Multiplying this by the average number of generations per mutation derived from the Gaston records I get 41 generations for the group. This gives me a tMRCA of 1240 years (assuming 30 y/gen). The McDonald mutation rate would have given me 1000 years and the Chandler rate, 1500 years.

      Comments appreciated.

      Craig
    • Earl Beaty
      Craig, Your analysis is interesting. My impression is that the major problem is the criteria that you used to select the 64 records from Ysearch. If your
      Message 2 of 17 , Jul 5, 2011
      • 0 Attachment
        Craig,

        Your analysis is interesting. My impression is that the major problem is the
        criteria that you used to select the 64 records from Ysearch. If your
        criterion is too restrictive you will not get the family members with many
        mutations. That will cause your TMRCA to be too small. On the other hand if
        you have included haplotypes outside the family, your TMRCA will be too
        large.

        --Earl

        > -----Original Message-----
        > From: Beatty_Byrnes_DNA@yahoogroups.com
        > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
        > Sent: Sunday, July 03, 2011 9:16 AM
        > To: Beatty_Byrnes_DNA@yahoogroups.com
        > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
        >
        > I downloaded YSearch 37-marker matches to our Gaston haplotype and
        combined the
        > 64 records with 21 records for our Ulster lineage. I then calculated the
        GDs to the
        > Gaston modal haplotype for all records. These ranged from 0 to 4 for the
        Gastons and
        > from 5 to 10 for the YSearch matches (assuming equal mutation weights for
        each
        > marker). Since we know that the Gaston common ancestor lived in about 1650
        I
        > estimated 12 generations. The average GD of the Gastons is 1.8. This gives
        .0044
        > mutations per generation per marker, which falls about half way between
        the
        > Chandler rate of .0033 and the McDonald rate of .0049
        >
        > Next I calculated the average GD (from the Gaston modal) for all 85
        records, giving 6.1
        > (first having eliminated the double counting of 389-1 and 389-2).
        Multiplying this by
        > the average number of generations per mutation derived from the Gaston
        records I get
        > 41 generations for the group. This gives me a tMRCA of 1240 years
        (assuming 30
        > y/gen). The McDonald mutation rate would have given me 1000 years and the
        > Chandler rate, 1500 years.
        >
        > Comments appreciated.
        >
        > Craig
        >
        >
        >
        >
        > ------------------------------------
        >
        > Yahoo! Groups Links
        >
        >
        >
      • craig
        Good point, Earl. Might this apply to BBC as well, to some degree? I guess it depends on how well the 464x criterion corresponds to some presumed 5th century
        Message 3 of 17 , Jul 6, 2011
        • 0 Attachment
          Good point, Earl. Might this apply to BBC as well, to some degree? I guess it depends on how well the 464x criterion corresponds to some presumed 5th century Irish chieftain. We can never be sure if there are false positives and false negatives. However, at least by using that single criterion you allow all the other STRs to vary.

          But there really are no ideal criteria. The combination of a given surname and haplotype in the period since surnames became common is about the best one can do. Beyond that period one has to accept some unknown amount of uncertainty. Of course, a haplotype combined with a common SNP that originated at the appropriate time would be stronger evidence.

          I guess the most interesting thing about my experiment is that the average mutation rate based on our Gaston lineage fell mid-way between the bounds of the Chandler and McDonald estimates.

          Craig

          --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@...> wrote:
          >
          > Craig,
          >
          > Your analysis is interesting. My impression is that the major problem is the
          > criteria that you used to select the 64 records from Ysearch. If your
          > criterion is too restrictive you will not get the family members with many
          > mutations. That will cause your TMRCA to be too small. On the other hand if
          > you have included haplotypes outside the family, your TMRCA will be too
          > large.
          >
          > --Earl
          >
          > > -----Original Message-----
          > > From: Beatty_Byrnes_DNA@yahoogroups.com
          > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
          > > Sent: Sunday, July 03, 2011 9:16 AM
          > > To: Beatty_Byrnes_DNA@yahoogroups.com
          > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
          > >
          > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
          > combined the
          > > 64 records with 21 records for our Ulster lineage. I then calculated the
          > GDs to the
          > > Gaston modal haplotype for all records. These ranged from 0 to 4 for the
          > Gastons and
          > > from 5 to 10 for the YSearch matches (assuming equal mutation weights for
          > each
          > > marker). Since we know that the Gaston common ancestor lived in about 1650
          > I
          > > estimated 12 generations. The average GD of the Gastons is 1.8. This gives
          > .0044
          > > mutations per generation per marker, which falls about half way between
          > the
          > > Chandler rate of .0033 and the McDonald rate of .0049
          > >
          > > Next I calculated the average GD (from the Gaston modal) for all 85
          > records, giving 6.1
          > > (first having eliminated the double counting of 389-1 and 389-2).
          > Multiplying this by
          > > the average number of generations per mutation derived from the Gaston
          > records I get
          > > 41 generations for the group. This gives me a tMRCA of 1240 years
          > (assuming 30
          > > y/gen). The McDonald mutation rate would have given me 1000 years and the
          > > Chandler rate, 1500 years.
          > >
          > > Comments appreciated.
          > >
          > > Craig
          > >
          > >
          > >
          > >
          > > ------------------------------------
          > >
          > > Yahoo! Groups Links
          > >
          > >
          > >
          >
        • mikewww7
          For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt L159.2 under L21 was reliable enough to place on their versions of a Y DNA haplotree. I doubt
          Message 4 of 17 , Jul 6, 2011
          • 0 Attachment
            For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2 under L21" was reliable enough to place on their versions of a Y DNA haplotree.

            I doubt if DYS464X=2c2g is more stable but I don't know. These are really questions for genetic biochemists and population geneticists.

            Do keep in mind that L159.2 is possibly much older than BBC and O'Connor just might the be evidence of that. Perhaps his lineage survived some "bottleneck" as well as a more prolific BBC group.

            There are other subclades where there are large GD's for a few stragglers. Probably a few WTY type tests might uncover if there are any more SNP's in common with O'Connor.

            Mike



            --- In Beatty_Byrnes_DNA@yahoogroups.com, "craig" <craigpgaston@...> wrote:
            >
            > Good point, Earl. Might this apply to BBC as well, to some degree? I guess it depends on how well the 464x criterion corresponds to some presumed 5th century Irish chieftain. We can never be sure if there are false positives and false negatives. However, at least by using that single criterion you allow all the other STRs to vary.
            >
            > But there really are no ideal criteria. The combination of a given surname and haplotype in the period since surnames became common is about the best one can do. Beyond that period one has to accept some unknown amount of uncertainty. Of course, a haplotype combined with a common SNP that originated at the appropriate time would be stronger evidence.
            >
            > I guess the most interesting thing about my experiment is that the average mutation rate based on our Gaston lineage fell mid-way between the bounds of the Chandler and McDonald estimates.
            >
            > Craig
            >
            > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@> wrote:
            > >
            > > Craig,
            > >
            > > Your analysis is interesting. My impression is that the major problem is the
            > > criteria that you used to select the 64 records from Ysearch. If your
            > > criterion is too restrictive you will not get the family members with many
            > > mutations. That will cause your TMRCA to be too small. On the other hand if
            > > you have included haplotypes outside the family, your TMRCA will be too
            > > large.
            > >
            > > --Earl
            > >
            > > > -----Original Message-----
            > > > From: Beatty_Byrnes_DNA@yahoogroups.com
            > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
            > > > Sent: Sunday, July 03, 2011 9:16 AM
            > > > To: Beatty_Byrnes_DNA@yahoogroups.com
            > > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
            > > >
            > > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
            > > combined the
            > > > 64 records with 21 records for our Ulster lineage. I then calculated the
            > > GDs to the
            > > > Gaston modal haplotype for all records. These ranged from 0 to 4 for the
            > > Gastons and
            > > > from 5 to 10 for the YSearch matches (assuming equal mutation weights for
            > > each
            > > > marker). Since we know that the Gaston common ancestor lived in about 1650
            > > I
            > > > estimated 12 generations. The average GD of the Gastons is 1.8. This gives
            > > .0044
            > > > mutations per generation per marker, which falls about half way between
            > > the
            > > > Chandler rate of .0033 and the McDonald rate of .0049
            > > >
            > > > Next I calculated the average GD (from the Gaston modal) for all 85
            > > records, giving 6.1
            > > > (first having eliminated the double counting of 389-1 and 389-2).
            > > Multiplying this by
            > > > the average number of generations per mutation derived from the Gaston
            > > records I get
            > > > 41 generations for the group. This gives me a tMRCA of 1240 years
            > > (assuming 30
            > > > y/gen). The McDonald mutation rate would have given me 1000 years and the
            > > > Chandler rate, 1500 years.
            > > >
            > > > Comments appreciated.
            > > >
            > > > Craig
            > > >
            > > >
            > > >
            > > >
            > > > ------------------------------------
            > > >
            > > > Yahoo! Groups Links
            > > >
            > > >
            > > >
            > >
            >
          • Mr. Downing
            I agree that it is much more probable that L159.2 is older than we all currently surmise. Neal
            Message 5 of 17 , Jul 8, 2011
            • 0 Attachment
              I agree that it is much more probable that L159.2 is older than we all currently surmise.


              Neal

              --- In Beatty_Byrnes_DNA@yahoogroups.com, "mikewww7" <mwwdna@...> wrote:
              >
              > For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2 under L21" was reliable enough to place on their versions of a Y DNA haplotree.
              >
              > I doubt if DYS464X=2c2g is more stable but I don't know. These are really questions for genetic biochemists and population geneticists.
              >
              > Do keep in mind that L159.2 is possibly much older than BBC and O'Connor just might the be evidence of that. Perhaps his lineage survived some "bottleneck" as well as a more prolific BBC group.
              >
              > There are other subclades where there are large GD's for a few stragglers. Probably a few WTY type tests might uncover if there are any more SNP's in common with O'Connor.
              >
              > Mike
              >
              >
              >
              > --- In Beatty_Byrnes_DNA@yahoogroups.com, "craig" <craigpgaston@> wrote:
              > >
              > > Good point, Earl. Might this apply to BBC as well, to some degree? I guess it depends on how well the 464x criterion corresponds to some presumed 5th century Irish chieftain. We can never be sure if there are false positives and false negatives. However, at least by using that single criterion you allow all the other STRs to vary.
              > >
              > > But there really are no ideal criteria. The combination of a given surname and haplotype in the period since surnames became common is about the best one can do. Beyond that period one has to accept some unknown amount of uncertainty. Of course, a haplotype combined with a common SNP that originated at the appropriate time would be stronger evidence.
              > >
              > > I guess the most interesting thing about my experiment is that the average mutation rate based on our Gaston lineage fell mid-way between the bounds of the Chandler and McDonald estimates.
              > >
              > > Craig
              > >
              > > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@> wrote:
              > > >
              > > > Craig,
              > > >
              > > > Your analysis is interesting. My impression is that the major problem is the
              > > > criteria that you used to select the 64 records from Ysearch. If your
              > > > criterion is too restrictive you will not get the family members with many
              > > > mutations. That will cause your TMRCA to be too small. On the other hand if
              > > > you have included haplotypes outside the family, your TMRCA will be too
              > > > large.
              > > >
              > > > --Earl
              > > >
              > > > > -----Original Message-----
              > > > > From: Beatty_Byrnes_DNA@yahoogroups.com
              > > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
              > > > > Sent: Sunday, July 03, 2011 9:16 AM
              > > > > To: Beatty_Byrnes_DNA@yahoogroups.com
              > > > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
              > > > >
              > > > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
              > > > combined the
              > > > > 64 records with 21 records for our Ulster lineage. I then calculated the
              > > > GDs to the
              > > > > Gaston modal haplotype for all records. These ranged from 0 to 4 for the
              > > > Gastons and
              > > > > from 5 to 10 for the YSearch matches (assuming equal mutation weights for
              > > > each
              > > > > marker). Since we know that the Gaston common ancestor lived in about 1650
              > > > I
              > > > > estimated 12 generations. The average GD of the Gastons is 1.8. This gives
              > > > .0044
              > > > > mutations per generation per marker, which falls about half way between
              > > > the
              > > > > Chandler rate of .0033 and the McDonald rate of .0049
              > > > >
              > > > > Next I calculated the average GD (from the Gaston modal) for all 85
              > > > records, giving 6.1
              > > > > (first having eliminated the double counting of 389-1 and 389-2).
              > > > Multiplying this by
              > > > > the average number of generations per mutation derived from the Gaston
              > > > records I get
              > > > > 41 generations for the group. This gives me a tMRCA of 1240 years
              > > > (assuming 30
              > > > > y/gen). The McDonald mutation rate would have given me 1000 years and the
              > > > > Chandler rate, 1500 years.
              > > > >
              > > > > Comments appreciated.
              > > > >
              > > > > Craig
              > > > >
              > > > >
              > > > >
              > > > >
              > > > > ------------------------------------
              > > > >
              > > > > Yahoo! Groups Links
              > > > >
              > > > >
              > > > >
              > > >
              > >
              >
            • Earl Beaty
              Craig, I think we haven t tried in the BBC to do a TMRCA calculation based on data derived from Ysearch. Our central assumption is that there was a guy who
              Message 6 of 17 , Jul 9, 2011
              • 0 Attachment
                Craig,

                I think we haven't tried in the BBC to do a TMRCA calculation based on data
                derived from Ysearch. Our central assumption is that there was a guy who
                lived something like 1000-2000 years ago who had the ccgg condition at 464X,
                and that we have identified a large collection of his descendants. I find it
                useful to give this guy a name, and the name I am preferring at the moment
                is Oog (Our old guy). This assumption is confirmed in many ways. We are
                lucky in that Oog had a very distinctive haplotype, which is present in the
                list of descendants we have developed. We have to be careful to not engage
                in circular reasoning (using our conclusions to justify our assumptions.) I
                think we are doing well on that front. This whole thing started with the
                Beatty family.

                Before the era of genetic genealogy a research sharing group of Beattys was
                assembled. When genetic testing became available we rather quickly ordered a
                bunch of tests and discovered that a large fraction had a common ancestor
                since the adoption of surnames. That group is now called Beatty Group 01. At
                the time an inside joke at Genealogy-DNA was that haplogroups were
                determined by "reading tea leaves". With time we learned that Group 01 is in
                haplogroup R1b1a2 (then called R1b1b2). We observed that the ancestral
                haplotype of our Group 01 was rather close to the ancestral haplotype of
                R1b1b2, with deviations only a 3 markers out of 37 and 5 out of 67. In
                addition we have elevated values at the 2 CDY markers. The 5 markers have
                rather low mutation rates and we surprisingly have modal values at the
                markers with relatively fast markers. This is a bit of good luck serving to
                define a somewhat small group. At some point the 464X test became available
                and we discovered that we are ccgg. We tested enough members of Group 01 to
                convince ourselves the all of Group 01 was ccgg. That meant that Oog
                preceded the adoption of the Beatty name. We didn't know any other
                descendants of Oog and for a while we regarded ccgg as a Beatty quality. We
                found several men with names other than Beatty who met the genetic
                conditions, including the ccgg condition. One of these was John Robert
                Burns.

                For a while we considered JRB a descendant of a Beatty man who lost his
                surname in an NPE. That lead to more Burns testing at 464X, and many more
                Byrnes were found with ccgg. The Byrnes had haplotypes which were similar to
                the Beattys, but with more diversity and with higher values at the CDY
                markers. This was followed by a testing of many men who had some relation
                with the men already found to be ccgg. This process produced the BBC. The
                Beatty and Byrnes are now the largest families in the project because of the
                way it developed. The Gaston family could easily turn out to be the largest
                when much more data are available.

                I believe that we have good reason to believe that Oog really is a man who
                is a male line ancestor for every (maybe except for a few) of the BBC. Our
                modal haplotype is almost certainly Oog's haplotype (except at perhaps 2 or
                3 markers).

                Working out when and where Oog lived is a difficult problem. I believe that
                the major problem is that we don't know how to deal with extinctions. We
                ought to be able to establish the order of the various mutational events,
                but relating them to history is hazardous.

                --Earl

                > -----Original Message-----
                > From: Beatty_Byrnes_DNA@yahoogroups.com
                > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                > Sent: Wednesday, July 06, 2011 4:24 AM
                > To: Beatty_Byrnes_DNA@yahoogroups.com
                > Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again
                >
                > Good point, Earl. Might this apply to BBC as well, to some degree? I guess
                it depends
                > on how well the 464x criterion corresponds to some presumed 5th century
                Irish
                > chieftain. We can never be sure if there are false positives and false
                negatives.
                > However, at least by using that single criterion you allow all the other
                STRs to vary.
                >
                > But there really are no ideal criteria. The combination of a given surname
                and
                > haplotype in the period since surnames became common is about the best one
                can
                > do. Beyond that period one has to accept some unknown amount of
                uncertainty. Of
                > course, a haplotype combined with a common SNP that originated at the
                appropriate
                > time would be stronger evidence.
                >
                > I guess the most interesting thing about my experiment is that the average
                mutation
                > rate based on our Gaston lineage fell mid-way between the bounds of the
                Chandler
                > and McDonald estimates.
                >
                > Craig
                >
                > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@...>
                wrote:
                > >
                > > Craig,
                > >
                > > Your analysis is interesting. My impression is that the major problem is
                the
                > > criteria that you used to select the 64 records from Ysearch. If your
                > > criterion is too restrictive you will not get the family members with
                many
                > > mutations. That will cause your TMRCA to be too small. On the other hand
                if
                > > you have included haplotypes outside the family, your TMRCA will be too
                > > large.
                > >
                > > --Earl
                > >
                > > > -----Original Message-----
                > > > From: Beatty_Byrnes_DNA@yahoogroups.com
                > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                > > > Sent: Sunday, July 03, 2011 9:16 AM
                > > > To: Beatty_Byrnes_DNA@yahoogroups.com
                > > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
                > > >
                > > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
                > > combined the
                > > > 64 records with 21 records for our Ulster lineage. I then calculated
                the
                > > GDs to the
                > > > Gaston modal haplotype for all records. These ranged from 0 to 4 for
                the
                > > Gastons and
                > > > from 5 to 10 for the YSearch matches (assuming equal mutation weights
                for
                > > each
                > > > marker). Since we know that the Gaston common ancestor lived in about
                1650
                > > I
                > > > estimated 12 generations. The average GD of the Gastons is 1.8. This
                gives
                > > .0044
                > > > mutations per generation per marker, which falls about half way
                between
                > > the
                > > > Chandler rate of .0033 and the McDonald rate of .0049
                > > >
                > > > Next I calculated the average GD (from the Gaston modal) for all 85
                > > records, giving 6.1
                > > > (first having eliminated the double counting of 389-1 and 389-2).
                > > Multiplying this by
                > > > the average number of generations per mutation derived from the Gaston
                > > records I get
                > > > 41 generations for the group. This gives me a tMRCA of 1240 years
                > > (assuming 30
                > > > y/gen). The McDonald mutation rate would have given me 1000 years and
                the
                > > > Chandler rate, 1500 years.
                > > >
                > > > Comments appreciated.
                > > >
                > > > Craig
                > > >
                > > >
                > > >
                > > >
                > > > ------------------------------------
                > > >
                > > > Yahoo! Groups Links
                > > >
                > > >
                > > >
                > >
                >
                >
                >
                >
                > ------------------------------------
                >
                > Yahoo! Groups Links
                >
                >
                >
              • mikewww7
                Thanks for the summary, Earl. Is it Gaston Lineage 1 that appears to be of L159.2? http://www.worldfamilies.net/surnames/gaston/results
                Message 7 of 17 , Jul 9, 2011
                • 0 Attachment
                  Thanks for the summary, Earl.

                  Is it Gaston Lineage 1 that appears to be of L159.2?
                  http://www.worldfamilies.net/surnames/gaston/results

                  --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@...> wrote:
                  >
                  > Craig,
                  >
                  > I think we haven't tried in the BBC to do a TMRCA calculation based on data
                  > derived from Ysearch. Our central assumption is that there was a guy who
                  > lived something like 1000-2000 years ago who had the ccgg condition at 464X,
                  > and that we have identified a large collection of his descendants. I find it
                  > useful to give this guy a name, and the name I am preferring at the moment
                  > is Oog (Our old guy). This assumption is confirmed in many ways. We are
                  > lucky in that Oog had a very distinctive haplotype, which is present in the
                  > list of descendants we have developed. We have to be careful to not engage
                  > in circular reasoning (using our conclusions to justify our assumptions.) I
                  > think we are doing well on that front. This whole thing started with the
                  > Beatty family.
                  >
                  > Before the era of genetic genealogy a research sharing group of Beattys was
                  > assembled. When genetic testing became available we rather quickly ordered a
                  > bunch of tests and discovered that a large fraction had a common ancestor
                  > since the adoption of surnames. That group is now called Beatty Group 01. At
                  > the time an inside joke at Genealogy-DNA was that haplogroups were
                  > determined by "reading tea leaves". With time we learned that Group 01 is in
                  > haplogroup R1b1a2 (then called R1b1b2). We observed that the ancestral
                  > haplotype of our Group 01 was rather close to the ancestral haplotype of
                  > R1b1b2, with deviations only a 3 markers out of 37 and 5 out of 67. In
                  > addition we have elevated values at the 2 CDY markers. The 5 markers have
                  > rather low mutation rates and we surprisingly have modal values at the
                  > markers with relatively fast markers. This is a bit of good luck serving to
                  > define a somewhat small group. At some point the 464X test became available
                  > and we discovered that we are ccgg. We tested enough members of Group 01 to
                  > convince ourselves the all of Group 01 was ccgg. That meant that Oog
                  > preceded the adoption of the Beatty name. We didn't know any other
                  > descendants of Oog and for a while we regarded ccgg as a Beatty quality. We
                  > found several men with names other than Beatty who met the genetic
                  > conditions, including the ccgg condition. One of these was John Robert
                  > Burns.
                  >
                  > For a while we considered JRB a descendant of a Beatty man who lost his
                  > surname in an NPE. That lead to more Burns testing at 464X, and many more
                  > Byrnes were found with ccgg. The Byrnes had haplotypes which were similar to
                  > the Beattys, but with more diversity and with higher values at the CDY
                  > markers. This was followed by a testing of many men who had some relation
                  > with the men already found to be ccgg. This process produced the BBC. The
                  > Beatty and Byrnes are now the largest families in the project because of the
                  > way it developed. The Gaston family could easily turn out to be the largest
                  > when much more data are available.
                  >
                  > I believe that we have good reason to believe that Oog really is a man who
                  > is a male line ancestor for every (maybe except for a few) of the BBC. Our
                  > modal haplotype is almost certainly Oog's haplotype (except at perhaps 2 or
                  > 3 markers).
                  >
                  > Working out when and where Oog lived is a difficult problem. I believe that
                  > the major problem is that we don't know how to deal with extinctions. We
                  > ought to be able to establish the order of the various mutational events,
                  > but relating them to history is hazardous.
                  >
                  > --Earl
                  >
                  > > -----Original Message-----
                  > > From: Beatty_Byrnes_DNA@yahoogroups.com
                  > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                  > > Sent: Wednesday, July 06, 2011 4:24 AM
                  > > To: Beatty_Byrnes_DNA@yahoogroups.com
                  > > Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again
                  > >
                  > > Good point, Earl. Might this apply to BBC as well, to some degree? I guess
                  > it depends
                  > > on how well the 464x criterion corresponds to some presumed 5th century
                  > Irish
                  > > chieftain. We can never be sure if there are false positives and false
                  > negatives.
                  > > However, at least by using that single criterion you allow all the other
                  > STRs to vary.
                  > >
                  > > But there really are no ideal criteria. The combination of a given surname
                  > and
                  > > haplotype in the period since surnames became common is about the best one
                  > can
                  > > do. Beyond that period one has to accept some unknown amount of
                  > uncertainty. Of
                  > > course, a haplotype combined with a common SNP that originated at the
                  > appropriate
                  > > time would be stronger evidence.
                  > >
                  > > I guess the most interesting thing about my experiment is that the average
                  > mutation
                  > > rate based on our Gaston lineage fell mid-way between the bounds of the
                  > Chandler
                  > > and McDonald estimates.
                  > >
                  > > Craig
                  > >
                  > > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@>
                  > wrote:
                  > > >
                  > > > Craig,
                  > > >
                  > > > Your analysis is interesting. My impression is that the major problem is
                  > the
                  > > > criteria that you used to select the 64 records from Ysearch. If your
                  > > > criterion is too restrictive you will not get the family members with
                  > many
                  > > > mutations. That will cause your TMRCA to be too small. On the other hand
                  > if
                  > > > you have included haplotypes outside the family, your TMRCA will be too
                  > > > large.
                  > > >
                  > > > --Earl
                  > > >
                  > > > > -----Original Message-----
                  > > > > From: Beatty_Byrnes_DNA@yahoogroups.com
                  > > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                  > > > > Sent: Sunday, July 03, 2011 9:16 AM
                  > > > > To: Beatty_Byrnes_DNA@yahoogroups.com
                  > > > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
                  > > > >
                  > > > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
                  > > > combined the
                  > > > > 64 records with 21 records for our Ulster lineage. I then calculated
                  > the
                  > > > GDs to the
                  > > > > Gaston modal haplotype for all records. These ranged from 0 to 4 for
                  > the
                  > > > Gastons and
                  > > > > from 5 to 10 for the YSearch matches (assuming equal mutation weights
                  > for
                  > > > each
                  > > > > marker). Since we know that the Gaston common ancestor lived in about
                  > 1650
                  > > > I
                  > > > > estimated 12 generations. The average GD of the Gastons is 1.8. This
                  > gives
                  > > > .0044
                  > > > > mutations per generation per marker, which falls about half way
                  > between
                  > > > the
                  > > > > Chandler rate of .0033 and the McDonald rate of .0049
                  > > > >
                  > > > > Next I calculated the average GD (from the Gaston modal) for all 85
                  > > > records, giving 6.1
                  > > > > (first having eliminated the double counting of 389-1 and 389-2).
                  > > > Multiplying this by
                  > > > > the average number of generations per mutation derived from the Gaston
                  > > > records I get
                  > > > > 41 generations for the group. This gives me a tMRCA of 1240 years
                  > > > (assuming 30
                  > > > > y/gen). The McDonald mutation rate would have given me 1000 years and
                  > the
                  > > > > Chandler rate, 1500 years.
                  > > > >
                  > > > > Comments appreciated.
                  > > > >
                  > > > > Craig
                  > > > >
                  > > > >
                  > > > >
                  > > > >
                  > > > > ------------------------------------
                  > > > >
                  > > > > Yahoo! Groups Links
                  > > > >
                  > > > >
                  > > > >
                  > > >
                  > >
                  > >
                  > >
                  > >
                  > > ------------------------------------
                  > >
                  > > Yahoo! Groups Links
                  > >
                  > >
                  > >
                  >
                • John S Walden
                  RE . I then calculated the GDs to the Gaston modal haplotype for all records. These ranged from 0 to 4 for the Gastons and from 5 to 10 for the YSearch
                  Message 8 of 17 , Jul 9, 2011
                  • 0 Attachment
                    RE  . I then calculated the GDs to the Gaston modal haplotype for all records. These ranged from 0 to 4 for the Gastons and from 5 to 10 for the YSearch matches (assuming equal mutation weights for each marker).

                    Hmmm - did this lead to over counting the number of mutations ( and thus the GDs) because
                    a single mutation could show up in more than one record?

                    John W

                  • craig
                    I also thank you for the summary, Earl. I knew about the markers and all that but I didn t know how the BBC project got started. Yes, Mike, it is our lineage 1
                    Message 9 of 17 , Jul 9, 2011
                    • 0 Attachment
                      I also thank you for the summary, Earl. I knew about the markers and all that but I didn't know how the BBC project got started. Yes, Mike, it is our lineage 1 that is probably all 159.2. Two of us have been tested but we all have very similar haplotypes that lead us back to a mid-1600s patriarch in Co Antrim so I would be very surprised if any of us tested negative for this SNP.

                      You may be right, John, about over counting. Our main marker, 460, where there is an even split between 11 and 12, would be counted as a difference for about half of our group whereas there may be only a couple of mutations. This is really the only marker of the first 37 where this happens so the bias is not too great. There are also a couple in CDYa.

                      If I read between the lines, Earl, you are saying that the Gastons are an integral part of the BBC group so the most relevant tMRCA is the one that you calculate for membership based on the ccgg criterion. I don't disagree. I was only trying to get an intuitive understanding of how to do this based on the names that our members all see when they look at their FTDNA matches. Your and John's comments are helpful in this regard.

                      It seems to me that the problem of over counting is unavoidable. We will inevitably have an overrepresentation of apparent mutations because the recruitment to the BBC project is not random. Some lineages will be more represented than others and each lineage will have its own family markers. But there are other problems that can't easily be addressed, such as self-cancelling mutations, which bias the measurement in the opposite direction.

                      Craig

                      --- In Beatty_Byrnes_DNA@yahoogroups.com, "mikewww7" <mwwdna@...> wrote:
                      >
                      > Thanks for the summary, Earl.
                      >
                      > Is it Gaston Lineage 1 that appears to be of L159.2?
                      > http://www.worldfamilies.net/surnames/gaston/results
                      >
                      > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@> wrote:
                      > >
                      > > Craig,
                      > >
                      > > I think we haven't tried in the BBC to do a TMRCA calculation based on data
                      > > derived from Ysearch. Our central assumption is that there was a guy who
                      > > lived something like 1000-2000 years ago who had the ccgg condition at 464X,
                      > > and that we have identified a large collection of his descendants. I find it
                      > > useful to give this guy a name, and the name I am preferring at the moment
                      > > is Oog (Our old guy). This assumption is confirmed in many ways. We are
                      > > lucky in that Oog had a very distinctive haplotype, which is present in the
                      > > list of descendants we have developed. We have to be careful to not engage
                      > > in circular reasoning (using our conclusions to justify our assumptions.) I
                      > > think we are doing well on that front. This whole thing started with the
                      > > Beatty family.
                      > >
                      > > Before the era of genetic genealogy a research sharing group of Beattys was
                      > > assembled. When genetic testing became available we rather quickly ordered a
                      > > bunch of tests and discovered that a large fraction had a common ancestor
                      > > since the adoption of surnames. That group is now called Beatty Group 01. At
                      > > the time an inside joke at Genealogy-DNA was that haplogroups were
                      > > determined by "reading tea leaves". With time we learned that Group 01 is in
                      > > haplogroup R1b1a2 (then called R1b1b2). We observed that the ancestral
                      > > haplotype of our Group 01 was rather close to the ancestral haplotype of
                      > > R1b1b2, with deviations only a 3 markers out of 37 and 5 out of 67. In
                      > > addition we have elevated values at the 2 CDY markers. The 5 markers have
                      > > rather low mutation rates and we surprisingly have modal values at the
                      > > markers with relatively fast markers. This is a bit of good luck serving to
                      > > define a somewhat small group. At some point the 464X test became available
                      > > and we discovered that we are ccgg. We tested enough members of Group 01 to
                      > > convince ourselves the all of Group 01 was ccgg. That meant that Oog
                      > > preceded the adoption of the Beatty name. We didn't know any other
                      > > descendants of Oog and for a while we regarded ccgg as a Beatty quality. We
                      > > found several men with names other than Beatty who met the genetic
                      > > conditions, including the ccgg condition. One of these was John Robert
                      > > Burns.
                      > >
                      > > For a while we considered JRB a descendant of a Beatty man who lost his
                      > > surname in an NPE. That lead to more Burns testing at 464X, and many more
                      > > Byrnes were found with ccgg. The Byrnes had haplotypes which were similar to
                      > > the Beattys, but with more diversity and with higher values at the CDY
                      > > markers. This was followed by a testing of many men who had some relation
                      > > with the men already found to be ccgg. This process produced the BBC. The
                      > > Beatty and Byrnes are now the largest families in the project because of the
                      > > way it developed. The Gaston family could easily turn out to be the largest
                      > > when much more data are available.
                      > >
                      > > I believe that we have good reason to believe that Oog really is a man who
                      > > is a male line ancestor for every (maybe except for a few) of the BBC. Our
                      > > modal haplotype is almost certainly Oog's haplotype (except at perhaps 2 or
                      > > 3 markers).
                      > >
                      > > Working out when and where Oog lived is a difficult problem. I believe that
                      > > the major problem is that we don't know how to deal with extinctions. We
                      > > ought to be able to establish the order of the various mutational events,
                      > > but relating them to history is hazardous.
                      > >
                      > > --Earl
                      > >
                      > > > -----Original Message-----
                      > > > From: Beatty_Byrnes_DNA@yahoogroups.com
                      > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                      > > > Sent: Wednesday, July 06, 2011 4:24 AM
                      > > > To: Beatty_Byrnes_DNA@yahoogroups.com
                      > > > Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again
                      > > >
                      > > > Good point, Earl. Might this apply to BBC as well, to some degree? I guess
                      > > it depends
                      > > > on how well the 464x criterion corresponds to some presumed 5th century
                      > > Irish
                      > > > chieftain. We can never be sure if there are false positives and false
                      > > negatives.
                      > > > However, at least by using that single criterion you allow all the other
                      > > STRs to vary.
                      > > >
                      > > > But there really are no ideal criteria. The combination of a given surname
                      > > and
                      > > > haplotype in the period since surnames became common is about the best one
                      > > can
                      > > > do. Beyond that period one has to accept some unknown amount of
                      > > uncertainty. Of
                      > > > course, a haplotype combined with a common SNP that originated at the
                      > > appropriate
                      > > > time would be stronger evidence.
                      > > >
                      > > > I guess the most interesting thing about my experiment is that the average
                      > > mutation
                      > > > rate based on our Gaston lineage fell mid-way between the bounds of the
                      > > Chandler
                      > > > and McDonald estimates.
                      > > >
                      > > > Craig
                      > > >
                      > > > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@>
                      > > wrote:
                      > > > >
                      > > > > Craig,
                      > > > >
                      > > > > Your analysis is interesting. My impression is that the major problem is
                      > > the
                      > > > > criteria that you used to select the 64 records from Ysearch. If your
                      > > > > criterion is too restrictive you will not get the family members with
                      > > many
                      > > > > mutations. That will cause your TMRCA to be too small. On the other hand
                      > > if
                      > > > > you have included haplotypes outside the family, your TMRCA will be too
                      > > > > large.
                      > > > >
                      > > > > --Earl
                      > > > >
                      > > > > > -----Original Message-----
                      > > > > > From: Beatty_Byrnes_DNA@yahoogroups.com
                      > > > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                      > > > > > Sent: Sunday, July 03, 2011 9:16 AM
                      > > > > > To: Beatty_Byrnes_DNA@yahoogroups.com
                      > > > > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
                      > > > > >
                      > > > > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
                      > > > > combined the
                      > > > > > 64 records with 21 records for our Ulster lineage. I then calculated
                      > > the
                      > > > > GDs to the
                      > > > > > Gaston modal haplotype for all records. These ranged from 0 to 4 for
                      > > the
                      > > > > Gastons and
                      > > > > > from 5 to 10 for the YSearch matches (assuming equal mutation weights
                      > > for
                      > > > > each
                      > > > > > marker). Since we know that the Gaston common ancestor lived in about
                      > > 1650
                      > > > > I
                      > > > > > estimated 12 generations. The average GD of the Gastons is 1.8. This
                      > > gives
                      > > > > .0044
                      > > > > > mutations per generation per marker, which falls about half way
                      > > between
                      > > > > the
                      > > > > > Chandler rate of .0033 and the McDonald rate of .0049
                      > > > > >
                      > > > > > Next I calculated the average GD (from the Gaston modal) for all 85
                      > > > > records, giving 6.1
                      > > > > > (first having eliminated the double counting of 389-1 and 389-2).
                      > > > > Multiplying this by
                      > > > > > the average number of generations per mutation derived from the Gaston
                      > > > > records I get
                      > > > > > 41 generations for the group. This gives me a tMRCA of 1240 years
                      > > > > (assuming 30
                      > > > > > y/gen). The McDonald mutation rate would have given me 1000 years and
                      > > the
                      > > > > > Chandler rate, 1500 years.
                      > > > > >
                      > > > > > Comments appreciated.
                      > > > > >
                      > > > > > Craig
                      > > > > >
                      > > > > >
                      > > > > >
                      > > > > >
                      > > > > > ------------------------------------
                      > > > > >
                      > > > > > Yahoo! Groups Links
                      > > > > >
                      > > > > >
                      > > > > >
                      > > > >
                      > > >
                      > > >
                      > > >
                      > > >
                      > > > ------------------------------------
                      > > >
                      > > > Yahoo! Groups Links
                      > > >
                      > > >
                      > > >
                      > >
                      >
                    • mikewww7
                      Craig, Is there a direct FTDNA project screen for the Gaston project lineage 1? The World Families web site seems to only display 37 markers. I ll copy in any
                      Message 10 of 17 , Jul 9, 2011
                      • 0 Attachment
                        Craig,

                        Is there a direct FTDNA project screen for the Gaston project lineage 1? The World Families web site seems to only display 37 markers.

                        I'll copy in any of the 67 length haplotypes I can from Gaston Lineage 1 into the R-L21 All haplotypes spreadsheet, but I can't find a Gaston FTDNA project public web site I can copy from. They are definitely "predicted" L159.2.

                        BTW, Gaston's run thick in my great-grandfather's hometown (very small) and one married my uncle. I think I have some of their blood in me too.

                        Regards,
                        Mike W

                        --- In Beatty_Byrnes_DNA@yahoogroups.com, "craig" <craigpgaston@...> wrote:
                        >
                        > I also thank you for the summary, Earl. I knew about the markers and all that but I didn't know how the BBC project got started. Yes, Mike, it is our lineage 1 that is probably all 159.2. Two of us have been tested but we all have very similar haplotypes that lead us back to a mid-1600s patriarch in Co Antrim so I would be very surprised if any of us tested negative for this SNP.
                        >
                        > You may be right, John, about over counting. Our main marker, 460, where there is an even split between 11 and 12, would be counted as a difference for about half of our group whereas there may be only a couple of mutations. This is really the only marker of the first 37 where this happens so the bias is not too great. There are also a couple in CDYa.
                        >
                        > If I read between the lines, Earl, you are saying that the Gastons are an integral part of the BBC group so the most relevant tMRCA is the one that you calculate for membership based on the ccgg criterion. I don't disagree. I was only trying to get an intuitive understanding of how to do this based on the names that our members all see when they look at their FTDNA matches. Your and John's comments are helpful in this regard.
                        >
                        > It seems to me that the problem of over counting is unavoidable. We will inevitably have an overrepresentation of apparent mutations because the recruitment to the BBC project is not random. Some lineages will be more represented than others and each lineage will have its own family markers. But there are other problems that can't easily be addressed, such as self-cancelling mutations, which bias the measurement in the opposite direction.
                        >
                        > Craig
                        >
                        > --- In Beatty_Byrnes_DNA@yahoogroups.com, "mikewww7" <mwwdna@> wrote:
                        > >
                        > > Thanks for the summary, Earl.
                        > >
                        > > Is it Gaston Lineage 1 that appears to be of L159.2?
                        > > http://www.worldfamilies.net/surnames/gaston/results
                        > >
                        > > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@> wrote:
                        > > >
                        > > > Craig,
                        > > >
                        > > > I think we haven't tried in the BBC to do a TMRCA calculation based on data
                        > > > derived from Ysearch. Our central assumption is that there was a guy who
                        > > > lived something like 1000-2000 years ago who had the ccgg condition at 464X,
                        > > > and that we have identified a large collection of his descendants. I find it
                        > > > useful to give this guy a name, and the name I am preferring at the moment
                        > > > is Oog (Our old guy). This assumption is confirmed in many ways. We are
                        > > > lucky in that Oog had a very distinctive haplotype, which is present in the
                        > > > list of descendants we have developed. We have to be careful to not engage
                        > > > in circular reasoning (using our conclusions to justify our assumptions.) I
                        > > > think we are doing well on that front. This whole thing started with the
                        > > > Beatty family.
                        > > >
                        > > > Before the era of genetic genealogy a research sharing group of Beattys was
                        > > > assembled. When genetic testing became available we rather quickly ordered a
                        > > > bunch of tests and discovered that a large fraction had a common ancestor
                        > > > since the adoption of surnames. That group is now called Beatty Group 01. At
                        > > > the time an inside joke at Genealogy-DNA was that haplogroups were
                        > > > determined by "reading tea leaves". With time we learned that Group 01 is in
                        > > > haplogroup R1b1a2 (then called R1b1b2). We observed that the ancestral
                        > > > haplotype of our Group 01 was rather close to the ancestral haplotype of
                        > > > R1b1b2, with deviations only a 3 markers out of 37 and 5 out of 67. In
                        > > > addition we have elevated values at the 2 CDY markers. The 5 markers have
                        > > > rather low mutation rates and we surprisingly have modal values at the
                        > > > markers with relatively fast markers. This is a bit of good luck serving to
                        > > > define a somewhat small group. At some point the 464X test became available
                        > > > and we discovered that we are ccgg. We tested enough members of Group 01 to
                        > > > convince ourselves the all of Group 01 was ccgg. That meant that Oog
                        > > > preceded the adoption of the Beatty name. We didn't know any other
                        > > > descendants of Oog and for a while we regarded ccgg as a Beatty quality. We
                        > > > found several men with names other than Beatty who met the genetic
                        > > > conditions, including the ccgg condition. One of these was John Robert
                        > > > Burns.
                        > > >
                        > > > For a while we considered JRB a descendant of a Beatty man who lost his
                        > > > surname in an NPE. That lead to more Burns testing at 464X, and many more
                        > > > Byrnes were found with ccgg. The Byrnes had haplotypes which were similar to
                        > > > the Beattys, but with more diversity and with higher values at the CDY
                        > > > markers. This was followed by a testing of many men who had some relation
                        > > > with the men already found to be ccgg. This process produced the BBC. The
                        > > > Beatty and Byrnes are now the largest families in the project because of the
                        > > > way it developed. The Gaston family could easily turn out to be the largest
                        > > > when much more data are available.
                        > > >
                        > > > I believe that we have good reason to believe that Oog really is a man who
                        > > > is a male line ancestor for every (maybe except for a few) of the BBC. Our
                        > > > modal haplotype is almost certainly Oog's haplotype (except at perhaps 2 or
                        > > > 3 markers).
                        > > >
                        > > > Working out when and where Oog lived is a difficult problem. I believe that
                        > > > the major problem is that we don't know how to deal with extinctions. We
                        > > > ought to be able to establish the order of the various mutational events,
                        > > > but relating them to history is hazardous.
                        > > >
                        > > > --Earl
                        > > >
                        > > > > -----Original Message-----
                        > > > > From: Beatty_Byrnes_DNA@yahoogroups.com
                        > > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                        > > > > Sent: Wednesday, July 06, 2011 4:24 AM
                        > > > > To: Beatty_Byrnes_DNA@yahoogroups.com
                        > > > > Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again
                        > > > >
                        > > > > Good point, Earl. Might this apply to BBC as well, to some degree? I guess
                        > > > it depends
                        > > > > on how well the 464x criterion corresponds to some presumed 5th century
                        > > > Irish
                        > > > > chieftain. We can never be sure if there are false positives and false
                        > > > negatives.
                        > > > > However, at least by using that single criterion you allow all the other
                        > > > STRs to vary.
                        > > > >
                        > > > > But there really are no ideal criteria. The combination of a given surname
                        > > > and
                        > > > > haplotype in the period since surnames became common is about the best one
                        > > > can
                        > > > > do. Beyond that period one has to accept some unknown amount of
                        > > > uncertainty. Of
                        > > > > course, a haplotype combined with a common SNP that originated at the
                        > > > appropriate
                        > > > > time would be stronger evidence.
                        > > > >
                        > > > > I guess the most interesting thing about my experiment is that the average
                        > > > mutation
                        > > > > rate based on our Gaston lineage fell mid-way between the bounds of the
                        > > > Chandler
                        > > > > and McDonald estimates.
                        > > > >
                        > > > > Craig
                        > > > >
                        > > > > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@>
                        > > > wrote:
                        > > > > >
                        > > > > > Craig,
                        > > > > >
                        > > > > > Your analysis is interesting. My impression is that the major problem is
                        > > > the
                        > > > > > criteria that you used to select the 64 records from Ysearch. If your
                        > > > > > criterion is too restrictive you will not get the family members with
                        > > > many
                        > > > > > mutations. That will cause your TMRCA to be too small. On the other hand
                        > > > if
                        > > > > > you have included haplotypes outside the family, your TMRCA will be too
                        > > > > > large.
                        > > > > >
                        > > > > > --Earl
                        > > > > >
                        > > > > > > -----Original Message-----
                        > > > > > > From: Beatty_Byrnes_DNA@yahoogroups.com
                        > > > > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                        > > > > > > Sent: Sunday, July 03, 2011 9:16 AM
                        > > > > > > To: Beatty_Byrnes_DNA@yahoogroups.com
                        > > > > > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
                        > > > > > >
                        > > > > > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
                        > > > > > combined the
                        > > > > > > 64 records with 21 records for our Ulster lineage. I then calculated
                        > > > the
                        > > > > > GDs to the
                        > > > > > > Gaston modal haplotype for all records. These ranged from 0 to 4 for
                        > > > the
                        > > > > > Gastons and
                        > > > > > > from 5 to 10 for the YSearch matches (assuming equal mutation weights
                        > > > for
                        > > > > > each
                        > > > > > > marker). Since we know that the Gaston common ancestor lived in about
                        > > > 1650
                        > > > > > I
                        > > > > > > estimated 12 generations. The average GD of the Gastons is 1.8. This
                        > > > gives
                        > > > > > .0044
                        > > > > > > mutations per generation per marker, which falls about half way
                        > > > between
                        > > > > > the
                        > > > > > > Chandler rate of .0033 and the McDonald rate of .0049
                        > > > > > >
                        > > > > > > Next I calculated the average GD (from the Gaston modal) for all 85
                        > > > > > records, giving 6.1
                        > > > > > > (first having eliminated the double counting of 389-1 and 389-2).
                        > > > > > Multiplying this by
                        > > > > > > the average number of generations per mutation derived from the Gaston
                        > > > > > records I get
                        > > > > > > 41 generations for the group. This gives me a tMRCA of 1240 years
                        > > > > > (assuming 30
                        > > > > > > y/gen). The McDonald mutation rate would have given me 1000 years and
                        > > > the
                        > > > > > > Chandler rate, 1500 years.
                        > > > > > >
                        > > > > > > Comments appreciated.
                        > > > > > >
                        > > > > > > Craig
                        > > > > > >
                        > > > > > >
                        > > > > > >
                        > > > > > >
                        > > > > > > ------------------------------------
                        > > > > > >
                        > > > > > > Yahoo! Groups Links
                        > > > > > >
                        > > > > > >
                        > > > > > >
                        > > > > >
                        > > > >
                        > > > >
                        > > > >
                        > > > >
                        > > > > ------------------------------------
                        > > > >
                        > > > > Yahoo! Groups Links
                        > > > >
                        > > > >
                        > > > >
                        > > >
                        > >
                        >
                      • craig
                        I just created a direct FTDNA project screen, Mike. http://www.familytreedna.com/public/Gaston There are about 5000 U.S. Gastons according to the census. My
                        Message 11 of 17 , Jul 10, 2011
                        • 0 Attachment
                          I just created a direct FTDNA project screen, Mike. http://www.familytreedna.com/public/Gaston

                          There are about 5000 U.S. Gastons according to the census. My line remained in Antrim until the early 1900s and my father was born there. I live in Canada. We have convinced a few Gastons from Antrim to join our project. Unfortunately, they don't know any more than we do about the early history of the family.

                          Craig
                          >
                          > Craig,
                          >
                          > Is there a direct FTDNA project screen for the Gaston project lineage 1? The World Families web site seems to only display 37 markers.
                          >
                          > I'll copy in any of the 67 length haplotypes I can from Gaston Lineage 1 into the R-L21 All haplotypes spreadsheet, but I can't find a Gaston FTDNA project public web site I can copy from. They are definitely "predicted" L159.2.
                          >
                          > BTW, Gaston's run thick in my great-grandfather's hometown (very small) and one married my uncle. I think I have some of their blood in me too.
                          >
                          > Regards,
                          > Mike W
                          >
                        • Earl Beaty
                          Mike, I agree that the L159.2 data are telling us that the L159.2+ mutation preceded our ccgg event. The only problem is the Mystery Group. We have quite a few
                          Message 12 of 17 , Jul 10, 2011
                          • 0 Attachment
                            Mike,

                            I agree that the L159.2 data are telling us that the L159.2+ mutation
                            preceded our ccgg event. The only problem is the Mystery Group. We have
                            quite a few members who are ccgg and L159.2- . This seemed good evidence
                            that ccgg came first. The Mystery Group is aptly named.

                            This suggests a second ccgg event of perhaps a second L159.2 mutation.

                            --Earl

                            > -----Original Message-----
                            > From: Beatty_Byrnes_DNA@yahoogroups.com
                            > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of mikewww7
                            > Sent: Wednesday, July 06, 2011 6:33 AM
                            > To: Beatty_Byrnes_DNA@yahoogroups.com
                            > Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again
                            >
                            > For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2
                            under L21"
                            > was reliable enough to place on their versions of a Y DNA haplotree.
                            >
                            > I doubt if DYS464X=2c2g is more stable but I don't know. These are really
                            questions
                            > for genetic biochemists and population geneticists.
                            >
                            > Do keep in mind that L159.2 is possibly much older than BBC and O'Connor
                            just
                            > might the be evidence of that. Perhaps his lineage survived some
                            "bottleneck" as well
                            > as a more prolific BBC group.
                            >
                            > There are other subclades where there are large GD's for a few stragglers.
                            Probably a
                            > few WTY type tests might uncover if there are any more SNP's in common
                            with
                            > O'Connor.
                            >
                            > Mike
                            >
                            >
                            >
                            > --- In Beatty_Byrnes_DNA@yahoogroups.com, "craig" <craigpgaston@...>
                            wrote:
                            > >
                            > > Good point, Earl. Might this apply to BBC as well, to some degree? I
                            guess it
                            > depends on how well the 464x criterion corresponds to some presumed 5th
                            century
                            > Irish chieftain. We can never be sure if there are false positives and
                            false negatives.
                            > However, at least by using that single criterion you allow all the other
                            STRs to vary.
                            > >
                            > > But there really are no ideal criteria. The combination of a given
                            surname and
                            > haplotype in the period since surnames became common is about the best one
                            can
                            > do. Beyond that period one has to accept some unknown amount of
                            uncertainty. Of
                            > course, a haplotype combined with a common SNP that originated at the
                            appropriate
                            > time would be stronger evidence.
                            > >
                            > > I guess the most interesting thing about my experiment is that the
                            average
                            > mutation rate based on our Gaston lineage fell mid-way between the bounds
                            of the
                            > Chandler and McDonald estimates.
                            > >
                            > > Craig
                            > >
                            > > --- In Beatty_Byrnes_DNA@yahoogroups.com, "Earl Beaty" <ecbeaty@> wrote:
                            > > >
                            > > > Craig,
                            > > >
                            > > > Your analysis is interesting. My impression is that the major problem
                            is the
                            > > > criteria that you used to select the 64 records from Ysearch. If your
                            > > > criterion is too restrictive you will not get the family members with
                            many
                            > > > mutations. That will cause your TMRCA to be too small. On the other
                            hand if
                            > > > you have included haplotypes outside the family, your TMRCA will be
                            too
                            > > > large.
                            > > >
                            > > > --Earl
                            > > >
                            > > > > -----Original Message-----
                            > > > > From: Beatty_Byrnes_DNA@yahoogroups.com
                            > > > > [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of craig
                            > > > > Sent: Sunday, July 03, 2011 9:16 AM
                            > > > > To: Beatty_Byrnes_DNA@yahoogroups.com
                            > > > > Subject: [Beatty_Byrnes_DNA] Mutation rates and tMRCA again
                            > > > >
                            > > > > I downloaded YSearch 37-marker matches to our Gaston haplotype and
                            > > > combined the
                            > > > > 64 records with 21 records for our Ulster lineage. I then calculated
                            the
                            > > > GDs to the
                            > > > > Gaston modal haplotype for all records. These ranged from 0 to 4 for
                            the
                            > > > Gastons and
                            > > > > from 5 to 10 for the YSearch matches (assuming equal mutation
                            weights for
                            > > > each
                            > > > > marker). Since we know that the Gaston common ancestor lived in
                            about 1650
                            > > > I
                            > > > > estimated 12 generations. The average GD of the Gastons is 1.8. This
                            gives
                            > > > .0044
                            > > > > mutations per generation per marker, which falls about half way
                            between
                            > > > the
                            > > > > Chandler rate of .0033 and the McDonald rate of .0049
                            > > > >
                            > > > > Next I calculated the average GD (from the Gaston modal) for all 85
                            > > > records, giving 6.1
                            > > > > (first having eliminated the double counting of 389-1 and 389-2).
                            > > > Multiplying this by
                            > > > > the average number of generations per mutation derived from the
                            Gaston
                            > > > records I get
                            > > > > 41 generations for the group. This gives me a tMRCA of 1240 years
                            > > > (assuming 30
                            > > > > y/gen). The McDonald mutation rate would have given me 1000 years
                            and the
                            > > > > Chandler rate, 1500 years.
                            > > > >
                            > > > > Comments appreciated.
                            > > > >
                            > > > > Craig
                            > > > >
                            > > > >
                            > > > >
                            > > > >
                            > > > > ------------------------------------
                            > > > >
                            > > > > Yahoo! Groups Links
                            > > > >
                            > > > >
                            > > > >
                            > > >
                            > >
                            >
                            >
                            >
                            >
                            > ------------------------------------
                            >
                            > Yahoo! Groups Links
                            >
                            >
                            >
                          • mikewww7
                            I am trying to take a look at L159.2 test results vis a vis the Irish Sea modal haplotypes. I thought I was on to something with 126655 Mobley as an L159.2-
                            Message 13 of 17 , Jul 14, 2011
                            • 0 Attachment
                              I am trying to take a look at L159.2 test results vis a vis the Irish Sea modal haplotypes. I thought I was on to something with 126655 Mobley as an L159.2- guy who fits with another group in R-L21* as well, but I can see that other group could be called Irish Sea as well even though I hadn't.

                              To go further, after looking at the McHale's, I can see this problem is intractable. Probably not new news.

                              I'll look at the GD's of different combinations, although I don't know what that will determine.

                              I do want to ask a basic question that is probably just an unknown. Is there any chance that a mutation to DYS464X could impact multiple other STR's, i.e. 448, 557, etc. at the same time... something like a recLOH event?

                              Mike W

                              --- In Beatty_Byrnes_DNA@yahoogroups.com, "mikewww7" <mwwdna@...> wrote:
                              >
                              > For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2 under L21" was reliable enough to place on their versions of a Y DNA haplotree.
                              >
                              > I doubt if DYS464X=2c2g is more stable but I don't know. These are really questions for genetic biochemists and population geneticists.
                              >
                              > Do keep in mind that L159.2 is possibly much older than BBC and O'Connor just might the be evidence of that. Perhaps his lineage survived some "bottleneck" as well as a more prolific BBC group.
                              >
                              > There are other subclades where there are large GD's for a few stragglers. Probably a few WTY type tests might uncover if there are any more SNP's in common with O'Connor.
                              >
                              > Mike
                            • dnalister@comcast.net
                              I have also wondered about a basic question of whether there can be mutation events affecting DYS464X test results and other results at the same time, and I
                              Message 14 of 17 , Jul 14, 2011
                              • 0 Attachment
                                I have also wondered about a basic question of whether there can be mutation events affecting DYS464X test results and other results at the same time, and I think that DYS448 may be on the same palindrome as DYS464, so at least DYS464X and DYS448 results may be affected by the same mutation events. That's assuming that I am remembering correctly that DYS448 is the singleton marker on that palindrome. Earl once wrote to me after a conference about something Thomas Krahn was explaining to him about DYS448 and some possible relationship to the stability of the DNA on the palindrome. At least, I think that's what Earl explained to me. Ever since then I have thought that I would like to talk to Thomas about this subject, but we are both busy and I haven't asked him about it so far. Maybe it's time to do that.

                                Kirsten


                                From: "mikewww7" <mwwdna@...>
                                To: "Beatty Byrnes DNA" <Beatty_Byrnes_DNA@yahoogroups.com>
                                Sent: Thursday, July 14, 2011 3:06:29 PM
                                Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again

                                 

                                I am trying to take a look at L159.2 test results vis a vis the Irish Sea modal haplotypes. I thought I was on to something with 126655 Mobley as an L159.2- guy who fits with another group in R-L21* as well, but I can see that other group could be called Irish Sea as well even though I hadn't.

                                To go further, after looking at the McHale's, I can see this problem is intractable. Probably not new news.

                                I'll look at the GD's of different combinations, although I don't know what that will determine.

                                I do want to ask a basic question that is probably just an unknown. Is there any chance that a mutation to DYS464X could impact multiple other STR's, i.e. 448, 557, etc. at the same time... something like a recLOH event?

                                Mike W

                                --- In Beatty_Byrnes_DNA@yahoogroups.com, "mikewww7" <mwwdna@...> wrote:
                                >
                                > For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2 under L21" was reliable enough to place on their versions of a Y DNA haplotree.
                                >
                                > I doubt if DYS464X=2c2g is more stable but I don't know. These are really questions for genetic biochemists and population geneticists.
                                >
                                > Do keep in mind that L159.2 is possibly much older than BBC and O'Connor just might the be evidence of that. Perhaps his lineage survived some "bottleneck" as well as a more prolific BBC group.
                                >
                                > There are other subclades where there are large GD's for a few stragglers. Probably a few WTY type tests might uncover if there are any more SNP's in common with O'Connor.
                                >
                                > Mike

                              • mikewww7
                                What s the opinion of the group on L159.2 testing? Given the problem on how the 464X and L159.2 testing seems to contradict itself in the mystery groups, I
                                Message 15 of 17 , Jul 15, 2011
                                • 0 Attachment
                                  What's the opinion of the group on L159.2 testing?

                                  Given the problem on how the 464X and L159.2 testing seems to contradict itself in the "mystery" groups, I think it would be wise for everyone in the "mystery" groups to test for L159.2.

                                  Actually, I think everyone with the Irish Sea modal type haplotype should test for L159.2. If we find that subgroups are also split on this maybe we would conclude that L159.2 is unstable?

                                  For instance if a McHale comes up L159.2+ then we need to ask FTDNA to relook at the L159.2- result from the other McHale. If both are confirmed I'd conclude L159.2 is unstable.

                                  Another example would be if any Beattie or Byrnes people in the main BBC groups come up L159.2-, then I'd also conclude L159.2 is unstable.

                                  We know that FTDNA and ISOGG think L159.2 is stable enough to be "officially" on the Y DNA tree. We can't really prove them right, but we can come close to proving them wrong.

                                  If we can't prove them wrong, my conclusion would be that the "mystery" groups are not really related to the Irish Sea common ancestor. They are just a crossing or converging branches from somewhere else on the R-L21 tree. This does happen... I've seen it a couple of times in the 11-13 (L513+) project.... of course, I'm trusting FTNDA and ISOGG in those cases as well.

                                  Mike


                                  --- In Beatty_Byrnes_DNA@yahoogroups.com, dnalister@... wrote:
                                  >
                                  > I have also wondered about a basic question of whether there can be mutation events affecting DYS464X test results and other results at the same time, and I think that DYS448 may be on the same palindrome as DYS464, so at least DYS464X and DYS448 results may be affected by the same mutation events. That's assuming that I am remembering correctly that DYS448 is the singleton marker on that palindrome. Earl once wrote to me after a conference about something Thomas Krahn was explaining to him about DYS448 and some possible relationship to the stability of the DNA on the palindrome. At least, I think that's what Earl explained to me. Ever since then I have thought that I would like to talk to Thomas about this subject, but we are both busy and I haven't asked him about it so far. Maybe it's time to do that.
                                  >
                                  > Kirsten
                                  >
                                  > ----- Original Message -----
                                  > From: "mikewww7" <mwwdna@...>
                                  > To: "Beatty Byrnes DNA" <Beatty_Byrnes_DNA@yahoogroups.com>
                                  > Sent: Thursday, July 14, 2011 3:06:29 PM
                                  > Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again
                                  >
                                  >
                                  >
                                  >
                                  >
                                  >
                                  > I am trying to take a look at L159.2 test results vis a vis the Irish Sea modal haplotypes. I thought I was on to something with 126655 Mobley as an L159.2- guy who fits with another group in R-L21* as well, but I can see that other group could be called Irish Sea as well even though I hadn't.
                                  >
                                  > To go further, after looking at the McHale's, I can see this problem is intractable. Probably not new news.
                                  >
                                  > I'll look at the GD's of different combinations, although I don't know what that will determine.
                                  >
                                  > I do want to ask a basic question that is probably just an unknown. Is there any chance that a mutation to DYS464X could impact multiple other STR's, i.e. 448, 557, etc. at the same time... something like a recLOH event?
                                  >
                                  > Mike W
                                  >
                                  > --- In Beatty_Byrnes_DNA@yahoogroups.com , "mikewww7" <mwwdna@> wrote:
                                  > >
                                  > > For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2 under L21" was reliable enough to place on their versions of a Y DNA haplotree.
                                  > >
                                  > > I doubt if DYS464X=2c2g is more stable but I don't know. These are really questions for genetic biochemists and population geneticists.
                                  > >
                                  > > Do keep in mind that L159.2 is possibly much older than BBC and O'Connor just might the be evidence of that. Perhaps his lineage survived some "bottleneck" as well as a more prolific BBC group.
                                  > >
                                  > > There are other subclades where there are large GD's for a few stragglers. Probably a few WTY type tests might uncover if there are any more SNP's in common with O'Connor.
                                  > >
                                  > > Mike
                                  >
                                • dnalister@comcast.net
                                  A goal that I have had for a while is to get at least one person from each family group in the BBC to do L159.2 testing. We are fairly close to meeting this
                                  Message 16 of 17 , Jul 15, 2011
                                  • 0 Attachment
                                    A goal that I have had for a while is to get at least one person from each family group in the BBC to do L159.2 testing. We are fairly close to meeting this goal for the families that have at least one member who has tested 2c2g on the DYS464X test. Of course whenever we get representatives of additional families to join the project and do DYS464X testing, we need to encourage L159.2 testing as well.

                                    Whatever happens with L159.2 testing, I have decided that it is phylogenetically useful in the sense that it has helped us distinguish the mystery groups. Of course Mystery groups 1 and 2 do have some STR differences from the L159.2+ population that I consider the core of the BBC. Mystery groups 1 and 2 share the similarity of having DYS448=19, making me think that they are in some way connected. The McHales are different in that respect, and it's harder to guess how they related to Mystery 1, Mystery 2, and the core. Even with Mystery 1 and Mystery 2, the number of key values to distinguish those subgroups is small enough that it's reasonably possible that there are men from those groups who would appear to be part of the core unless they had a result for L159.2, or that there might be a member of the core who appears to be a member of one of the Mystery groups unless tested for L159.2.

                                    I think it is possible that the McHales are pre-L159.2. They are not from Leinster, and their haplotypes seem a little different. I think it is plausible that they separated from the core BBC over 2,000 years ago, and while I think that the BBC could be over 2,000 years old, I don't think it is very much older than 2,000 years old.

                                    It seems to me that the determination of which SNPs are phylogenetically useful is in some ways an academic exercise. As more and more SNPs are placed on the tree, SNPs that are known to have mutated many times more than twice may be found useful. I think it has been stated SNPs known to have mutated more than once were too unstable to be placed on the tree. It's nice to have rules of thumb, but I think that if that has been one of the rules, it's not a rule that will stand the test of time.

                                    For study of the BBC, it would help a lot to know about other SNPs found within the group. Two members of the BBC core applied to participate in the
                                    WTY program during the WTY sale. Since we already had a member tested on about 100,000 base pairs early in the program, we would be hoping that testing of these two members would either turn up a SNP or SNPs in those same 100,000 base pairs that was missed due to poor quality sequencing or because it is not found in members of the BBC and/or to find a SNP or SNPs in the 100,000+ additional base pairs that WTY sequencing efforts can involve. Even if L159.2 is "unreliable" within R-L21 in the sense that there have been at least two mutations involving its location, it may well have been reliable within particular subclades of R-L21 associated with SNPs that remain unmapped.

                                    If the Mystery groups are less closely related to the BBC than some other non-BBC lines, then WTY sequencing for one or more of those Mystery groups might help us determine that faster than a single-minded focus on the core BBC. If another WTY sale comes along  before we resolve this question, then I think we should think carefully about WTY for at least one Mystery man. With the amount of sequencing and SNP discovery going on these days, I wonder if we might get lucky and find that some academic researcher has discovered a SNP or two that helps us answer some of our questions first. Even if that happens, I think that we will want to continue to work toward additional sequencing for members of the BBC. There must be more than one, two, three or four unmapped SNPs relevant to the BBC, and by finding more of them, we will be able to learn more about the history of the group.

                                    Kirsten


                                    From: "mikewww7" <mwwdna@...>
                                    To: "Beatty Byrnes DNA" <Beatty_Byrnes_DNA@yahoogroups.com>
                                    Sent: Friday, July 15, 2011 12:35:39 PM
                                    Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again

                                     

                                    What's the opinion of the group on L159.2 testing?

                                    Given the problem on how the 464X and L159.2 testing seems to contradict itself in the "mystery" groups, I think it would be wise for everyone in the "mystery" groups to test for L159.2.

                                    Actually, I think everyone with the Irish Sea modal type haplotype should test for L159.2. If we find that subgroups are also split on this maybe we would conclude that L159.2 is unstable?

                                    For instance if a McHale comes up L159.2+ then we need to ask FTDNA to relook at the L159.2- result from the other McHale. If both are confirmed I'd conclude L159.2 is unstable.

                                    Another example would be if any Beattie or Byrnes people in the main BBC groups come up L159.2-, then I'd also conclude L159.2 is unstable.

                                    We know that FTDNA and ISOGG think L159.2 is stable enough to be "officially" on the Y DNA tree. We can't really prove them right, but we can come close to proving them wrong.

                                    If we can't prove them wrong, my conclusion would be that the "mystery" groups are not really related to the Irish Sea common ancestor. They are just a crossing or converging branches from somewhere else on the R-L21 tree. This does happen... I've seen it a couple of times in the 11-13 (L513+) project.... of course, I'm trusting FTNDA and ISOGG in those cases as well.

                                    Mike

                                    --- In Beatty_Byrnes_DNA@yahoogroups.com, dnalister@... wrote:
                                    >
                                    > I have also wondered about a basic question of whether there can be mutation events affecting DYS464X test results and other results at the same time, and I think that DYS448 may be on the same palindrome as DYS464, so at least DYS464X and DYS448 results may be affected by the same mutation events. That's assuming that I am remembering correctly that DYS448 is the singleton marker on that palindrome. Earl once wrote to me after a conference about something Thomas Krahn was explaining to him about DYS448 and some possible relationship to the stability of the DNA on the palindrome. At least, I think that's what Earl explained to me. Ever since then I have thought that I would like to talk to Thomas about this subject, but we are both busy and I haven't asked him about it so far. Maybe it's time to do that.
                                    >
                                    > Kirsten
                                    >
                                    > ----- Original Message -----
                                    > From: "mikewww7" <mwwdna@...>
                                    > To: "Beatty Byrnes DNA" <Beatty_Byrnes_DNA@yahoogroups.com>
                                    > Sent: Thursday, July 14, 2011 3:06:29 PM
                                    > Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again
                                    >
                                    >
                                    >
                                    >
                                    >
                                    >
                                    > I am trying to take a look at L159.2 test results vis a vis the Irish Sea modal haplotypes. I thought I was on to something with 126655 Mobley as an L159.2- guy who fits with another group in R-L21* as well, but I can see that other group could be called Irish Sea as well even though I hadn't.
                                    >
                                    > To go further, after looking at the McHale's, I can see this problem is intractable. Probably not new news.
                                    >
                                    > I'll look at the GD's of different combinations, although I don't know what that will determine.
                                    >
                                    > I do want to ask a basic question that is probably just an unknown. Is there any chance that a mutation to DYS464X could impact multiple other STR's, i.e. 448, 557, etc. at the same time... something like a recLOH event?
                                    >
                                    > Mike W
                                    >
                                    > --- In Beatty_Byrnes_DNA@yahoogroups.com , "mikewww7" <mwwdna@> wrote:
                                    > >
                                    > > For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2 under L21" was reliable enough to place on their versions of a Y DNA haplotree.
                                    > >
                                    > > I doubt if DYS464X=2c2g is more stable but I don't know. These are really questions for genetic biochemists and population geneticists.
                                    > >
                                    > > Do keep in mind that L159.2 is possibly much older than BBC and O'Connor just might the be evidence of that. Perhaps his lineage survived some "bottleneck" as well as a more prolific BBC group.
                                    > >
                                    > > There are other subclades where there are large GD's for a few stragglers. Probably a few WTY type tests might uncover if there are any more SNP's in common with O'Connor.
                                    > >
                                    > > Mike
                                    >

                                  • Earl Beaty
                                    I will try to recall my lunchtime conversation with Thomas. A bit of background: In the Beatty Project all of Group 01 have DYS448
                                    Message 17 of 17 , Jul 17, 2011
                                    • 0 Attachment

                                      I will try to recall my lunchtime conversation with Thomas.

                                       

                                      A bit of background: In the Beatty Project all of Group 01 have DYS448<=18 and all the other Beattys tested have DYS448>18. The breaking out of Group 01 does not require 448. A question to Thomas was: Why is this marker so fixed in this family? His response was something like: Maybe it is locked. We didn’t discuss how such a lock might be implemented—it was just chatter over lunch. I came away with the impression that there are many “errors” in the copying process which happen often enough to be seen sometimes, but happen seldom enough to make a study difficult. STRs and SNPs aren’t the only things which mutate. A further impression was that during a palindrome repair there is a substantial variety of unexpected things that can happen.

                                       

                                      --Earl

                                       

                                       

                                      From: Beatty_Byrnes_DNA@yahoogroups.com [mailto:Beatty_Byrnes_DNA@yahoogroups.com] On Behalf Of dnalister@...
                                      Sent: Thursday, July 14, 2011 1:39 PM
                                      To: Beatty Byrnes DNA
                                      Subject: Re: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again

                                       




                                      I have also wondered about a basic question of whether there can be mutation events affecting DYS464X test results and other results at the same time, and I think that DYS448 may be on the same palindrome as DYS464, so at least DYS464X and DYS448 results may be affected by the same mutation events. That's assuming that I am remembering correctly that DYS448 is the singleton marker on that palindrome. Earl once wrote to me after a conference about something Thomas Krahn was explaining to him about DYS448 and some possible relationship to the stability of the DNA on the palindrome. At least, I think that's what Earl explained to me. Ever since then I have thought that I would like to talk to Thomas about this subject, but we are both busy and I haven't asked him about it so far. Maybe it's time to do that.

                                      Kirsten


                                      From: "mikewww7" <mwwdna@...>
                                      To: "Beatty Byrnes DNA" <Beatty_Byrnes_DNA@yahoogroups.com>
                                      Sent: Thursday, July 14, 2011 3:06:29 PM
                                      Subject: [Beatty_Byrnes_DNA] Re: Mutation rates and tMRCA again

                                       

                                      I am trying to take a look at L159.2 test results vis a vis the Irish Sea modal haplotypes. I thought I was on to something with 126655 Mobley as an L159.2- guy who fits with another group in R-L21* as well, but I can see that other group could be called Irish Sea as well even though I hadn't.

                                      To go further, after looking at the McHale's, I can see this problem is intractable. Probably not new news.

                                      I'll look at the GD's of different combinations, although I don't know what that will determine.

                                      I do want to ask a basic question that is probably just an unknown. Is there any chance that a mutation to DYS464X could impact multiple other STR's, i.e. 448, 557, etc. at the same time... something like a recLOH event?

                                      Mike W

                                      --- In Beatty_Byrnes_DNA@yahoogroups.com, "mikewww7" <mwwdna@...> wrote:
                                      >
                                      > For whatever reasons, FTDNA, EthnoAncestry and ISOGG all felt "L159.2 under L21" was reliable enough to place on their versions of a Y DNA haplotree.
                                      >
                                      > I doubt if DYS464X=2c2g is more stable but I don't know. These are really questions for genetic biochemists and population geneticists.
                                      >
                                      > Do keep in mind that L159.2 is possibly much older than BBC and O'Connor just might the be evidence of that. Perhaps his lineage survived some "bottleneck" as well as a more prolific BBC group.
                                      >
                                      > There are other subclades where there are large GD's for a few stragglers. Probably a few WTY type tests might uncover if there are any more SNP's in common with O'Connor.
                                      >
                                      > Mike




                                    Your message has been successfully submitted and would be delivered to recipients shortly.