Ralph, Thanks for the well thought-out response. I left out a lot of the detail you are asking for to make it easier to understand. I had posted the same question in greater detail on dna-forums.org, and got no responses. My suspicion was that people did not respond because it was too complicated. So it appears you can't win. We believe the man (John Francis) who was born in 1658 to have had 35 repeats at cdy-b because a) He had six sons who had children, and descendants of three (Robert,Joseph and Daniel) have been tested. The descendants of Robert and Daniel have cdy-b = 35. The man born in 1822(Theodore Francis), the 36 and 37s are all descendants of Joseph. b) Additionally, there is another family with a different surname (Rose) that is obviously related to ours. It lived in the same Connecticut town in the 1600s and has an identical modal haplotype at 37 markers (assuming ours is cdy-b=35). I believe there are 8 or 9 members of that family group who have been tested and all are cdy-b = 35. Based on this, we think it is probable that John Francis was cdy-b =35. Yes, he could have been 36, 37 or something else, but given a&b above, we think that cdy-b is, by far, the most likely scenario and that there were mutations down the line. So the question becomes, where did the mutations occur? Theodore Francis is the MRCA of the three individuals who were tested. While it is possible that he was a 35, that would imply parallel mutations to 36 in 3 different lines and further parallel mutations to 37 in two different lines. That seems wildly implausible to me. If Theodore Francis was a 36, that would mean the line had only undergone one mutation between the birth of Joseph Francis (1698) and the birth of Theodore in 1822. However, it would imply parallel mutations from 36 to 37 in two different lines. On the other hand, if Theodore was a 37, that would mean the line would have had to have gone from 35 to 37 between 1698 and 1822, but it would only imply one mutation after Theodore's time, and does not require parallel mutations. Personally, I think it is much more likely that Theodore was a 37 than a 36. Although both the 36 and 37 scenarios contain 3 mutations on cdy-b, I mistrust the parallel mutation scenario that is implied if Theodore were a 36. In a parallel mutation scenario, the exact same mutation must occur exactly the same way in two individuals in different lines. In other words, the mutations must occur on the exact same marker and in the same direction. If these mutations are truly random, a scenario including parallel mutations seems far less likely than one in which the same number of mutations occur in the same line. I hope this has cleared things up a bit. Please let me know if you think I've missed something. Re: [Y-DNA-projects] Y-DNA-PROJECTS Which mutation path is more likely? To: <y-dna-projects@rootsweb.com> Message-ID: <5154D39393C748EDBD0A14512F695DE4@Ralphs> Content-Type: text/plain; charset="us-ascii" Gregory wrote: "Please help settle a dispute. "1. A man born in 1658 has 35 repeats at cdy-b. His great-great-grandson born in 1822 has mutated up to 37 repeats at cdy-b. The second man's great-great-grandson born in 1965 has back-mutated to 36 repeats at cdy-b, while two other descendants (one through a different son of the man born in 1822, one through the same) still have 37. "2. The same man born in 1658 still has 35 repeats at cdy-b, but this time his great-great-grandson born in 1822 has only 36 repeats at cdy-b. Two of his descendants through different sons both mutate independently to 37, while a third remains at 36." Others will, no doubt, address the volatility, multi-copy & palindromic aspects of CDY (AKA, DSY724), as well as the reporting of CYDa & CDYb. I'll focus on the epistemology. It is important _HOW_ you "know" the posited values (for CMA b.1658 & GGS b.1822). The dispute may be an artifact of over-interpreting; both scenarios (and others) are plausible. The most likely scenario is not included above; it is that CMA b.1685, & GGS b.1822, had values found in descendants. The men born in 1658 & 1822 can not have been alive in the 21st century for actual Y-DNA testing. Without testing, we can not state their CDY values with any certainty. What can really be known here are the values for the men tested -- i.e., CDYb=37 for two & CDYb=36 for another. Assuming this is the only marker where mismatches are found, it is highly probable that the three men share a CMA within GTF and that that CMA had CDY values of either. As to "back-mutated", this term must be hypothetical or a conclusion. Perhaps, you've "triangulated" (with other descendants) to estimate CDYb values of b.1658 & b.1822. This method yields only probabilistic estimates, with confidence limits unknown to us readers. Conclusion: Neither mutation scenario is more likely than the other on the data provided -- the basis for assumed data is unstated. More likely than either is that CMA b.1658 & GGS b.1822 had CDYb=36 or CDYb=37. -rt_/) AKA, ralph

One thing to keep in mind: the most probable scenario isn't necessarily what actually happened. The only way to really know what happened is to test cousins until you've located the mutations. Diana > -----Original Message----- > From: y-dna-projects-bounces@rootsweb.com On Behalf Of Gregory Francis > Sent: Wednesday, June 23, 2010 1:40 PM > To: y-dna-projects@rootsweb.com > Subject: Re: [Y-DNA-projects] Which mutation path is more likely? > <snip> > > Personally, I think it is much more likely that Theodore was > a 37 than a 36. Although both the 36 and 37 scenarios contain > 3 mutations on cdy-b, I mistrust the parallel mutation > scenario that is implied if Theodore were a 36. In a parallel > mutation scenario, the exact same mutation must occur exactly > the same way in two individuals in different lines. In other > words, the mutations must occur on the exact same marker and > in the same direction. If these mutations are truly random, a > scenario including parallel mutations seems far less likely > than one in which the same number of mutations occur in the same line. > > I hope this has cleared things up a bit. Please let me know > if you think I've missed something. > >