In a message dated 11/19/2011 6:28:23 A.M. Central Standard Time, ikennedy_msdn2@hotmail.com writes: Remind me how you define acceptance into this project, since without that key piece of information the data below doesn't seem to mean much, to me? And of course if it's not precisely defined, the stats would mean even less. That then runs the risk that you have let some in who are further away from the modal from ones you earlier rejected. Thanks Iain Iain, I don't think I answered this. I've been doing what David Wilson began, eyeballing STRs, seeing how well they matched up to the modal then admitting them into the project if they matched up well. By well I mean close or exact matches to the M222 modal at 37 markers, although mostly I just look at the first 25. If any are questionable or just have bits and pieces of the modal I suggest an SNP test and do not admit them. I have to admit this approach is more of an art than a science but in the case of M222 it seems to work well. I'm sure it would not work at all for any other haplogroup project in existence. David Wilson once said if you match two of three of the first modal markers you most likely are M222. That's 390= 25, 385b = 13 and 392 = 14. If anyone mismatches at two of three they aren't admitted. The other markers are also checked, especially 464. I re-ran the M222 modal markers using only M222+. You can see the results here: _http://clanmaclochlainn.com/R1b1c7/M222repeat.htm_ (http://clanmaclochlainn.com/R1b1c7/M222repeat.htm) Mostly the percentages are about the same. But we're dealing with about half as many samples. A few changed a fair amount like 385b. I'm not sure why this happened. I'll have to double-check the figures I ran on the entire project. Another one that changed a lot is 413a. I'll re-check that too. One thing that happens when you require SNP tests is you lose a lot of samples, especially from some of the larger surname groups. Lots of Dohertys, McLaughlins, Ewings and Cowans do not have SNP tests because they were advised it was unnecessary to test all by the experts. That probably changed some of the percentages. I think you said some of your Kennedys fall into that category too. But that's the reason I've so far resisted either requiring SNP tests or segregating the results into SNP tested and non-tested. I could do that easily though. I could also change the requirements for joining to include at least an M222+ result. I hesitate to do that though. I'm willing at this point to go along with what people in the project think is best. Would segregating the SNP tested and non tested into different groups satisfy the purists who demand SNP tests for every sample? Realistically everyone now in the project is not going to jump up and order an SNP test because the purists demand it. They're right of course. Sort of. You can't tell absolutely without a shadow of doubt if someone is M222 without an SNP test. But those who claim that one can't identify M222 on the basis of STRS also are full of it. I was just told that last week by someone in the L21 project. I don't have any idea what the certainty level based on STR analysis would be though. I'm sure it's very high. In the high 90s. A few ringers could sneak in. Bill Howard ran some statistical analyses of the project a while back. He mentioned in vague terms a few of them. Basically he found no difference in results from sets using 37 markers vs. 67 markers; and also found no significant differences between the SNP tested members and those who weren't. You'd have to ask him how he did this. I barely remember the conversations at this point. John

Thanks John. I honestly wasn't trying to restart the SNP/STR argument again, it just suddenly occurred to me that you hadn't described the STR definition. I guess for me at least, if not others, that now DF23 has surfaced there is great interest in the people who just fail entry. I've been following the dna forums discussions but they too seem to be following the 'finger in the air' test for likely DF23 candidates. cheers Iain ---------------------------------------- > From: Lochlan@aol.com > Date: Sun, 20 Nov 2011 00:10:38 -0500 > To: dna-r1b1c7@rootsweb.com > Subject: Re: [R-M222] M222 Modal Markers > > > > Iain > > Iain, I don't think I answered this. I've been doing what David Wilson > began, eyeballing STRs, seeing how well they matched up to the modal then > admitting them into the project if they matched up well. By well I mean close > or exact matches to the M222 modal at 37 markers, although mostly I just > look at the first 25. If any are questionable or just have bits and pieces > of the modal I suggest an SNP test and do not admit them. I have to admit > this approach is more of an art than a science but in the case of M222 it > seems to work well. I'm sure it would not work at all for any other > haplogroup project in existence. > > David Wilson once said if you match two of three of the first modal markers > you most likely are M222. That's 390= 25, 385b = 13 and 392 = 14. If > anyone mismatches at two of three they aren't admitted. The other markers > are also checked, especially 464. > > I re-ran the M222 modal markers using only M222+. You can see the results > here: > > _http://clanmaclochlainn.com/R1b1c7/M222repeat.htm_ > (http://clanmaclochlainn.com/R1b1c7/M222repeat.htm) > > > Mostly the percentages are about the same. But we're dealing with about > half as many samples. A few changed a fair amount like 385b. I'm not sure > why this happened. I'll have to double-check the figures I ran on the > entire project. Another one that changed a lot is 413a. I'll re-check > that too. > > One thing that happens when you require SNP tests is you lose a lot of > samples, especially from some of the larger surname groups. Lots of Dohertys, > McLaughlins, Ewings and Cowans do not have SNP tests because they were > advised it was unnecessary to test all by the experts. That probably changed > some of the percentages. I think you said some of your Kennedys fall into > that category too. But that's the reason I've so far resisted either > requiring SNP tests or segregating the results into SNP tested and non-tested. > I could do that easily though. I could also change the requirements for > joining to include at least an M222+ result. I hesitate to do that though. > > I'm willing at this point to go along with what people in the project think > is best. Would segregating the SNP tested and non tested into different > groups satisfy the purists who demand SNP tests for every sample? > Realistically everyone now in the project is not going to jump up and order an SNP > test because the purists demand it. > > They're right of course. Sort of. You can't tell absolutely without a > shadow of doubt if someone is M222 without an SNP test. But those who claim > that one can't identify M222 on the basis of STRS also are full of it. I > was just told that last week by someone in the L21 project. I don't have > any idea what the certainty level based on STR analysis would be though. I'm > sure it's very high. In the high 90s. A few ringers could sneak in. > > Bill Howard ran some statistical analyses of the project a while back. He > mentioned in vague terms a few of them. Basically he found no difference > in results from sets using 37 markers vs. 67 markers; and also found no > significant differences between the SNP tested members and those who weren't. > You'd have to ask him how he did this. I barely remember the > conversations at this point. > > > > John > R1b1c7 Research and Links: > > http://clanmaclochlainn.com/R1b1c7/ > ------------------------------- > To unsubscribe from the list, please send an email to DNA-R1B1C7-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message

Iain It seems a McLea in the Ulster Heritage Mid Argyllshire group has ordered the DF23 test. If he should turn out to be DF23+, I think all of the McCain, Henry and Maguire M222- Ulster Heritage groups come into play with DYS481 = 24 DYS714 = 24 With luck, the results could be out by Wednesday. Sandy -----Original Message----- From: dna-r1b1c7-bounces@rootsweb.com [mailto:dna-r1b1c7-bounces@rootsweb.com] On Behalf Of Iain Kennedy Sent: 20 November 2011 11:36 To: dna-r1b1c7@rootsweb.com Subject: Re: [R-M222] M222 Modal Markers Thanks John. I honestly wasn't trying to restart the SNP/STR argument again, it just suddenly occurred to me that you hadn't described the STR definition. I guess for me at least, if not others, that now DF23 has surfaced there is great interest in the people who just fail entry. I've been following the dna forums discussions but they too seem to be following the 'finger in the air' test for likely DF23 candidates. cheers Iain ---------------------------------------- > From: Lochlan@aol.com > Date: Sun, 20 Nov 2011 00:10:38 -0500 > To: dna-r1b1c7@rootsweb.com > Subject: Re: [R-M222] M222 Modal Markers > > > > Iain > > Iain, I don't think I answered this. I've been doing what David Wilson > began, eyeballing STRs, seeing how well they matched up to the modal then > admitting them into the project if they matched up well. By well I mean close > or exact matches to the M222 modal at 37 markers, although mostly I just > look at the first 25. If any are questionable or just have bits and pieces > of the modal I suggest an SNP test and do not admit them. I have to admit > this approach is more of an art than a science but in the case of M222 it > seems to work well. I'm sure it would not work at all for any other > haplogroup project in existence. > > David Wilson once said if you match two of three of the first modal markers > you most likely are M222. That's 390= 25, 385b = 13 and 392 = 14. If > anyone mismatches at two of three they aren't admitted. The other markers > are also checked, especially 464. > > I re-ran the M222 modal markers using only M222+. You can see the results > here: > > _http://clanmaclochlainn.com/R1b1c7/M222repeat.htm_ > (http://clanmaclochlainn.com/R1b1c7/M222repeat.htm) > > > Mostly the percentages are about the same. But we're dealing with about > half as many samples. A few changed a fair amount like 385b. I'm not sure > why this happened. I'll have to double-check the figures I ran on the > entire project. Another one that changed a lot is 413a. I'll re-check > that too. > > One thing that happens when you require SNP tests is you lose a lot of > samples, especially from some of the larger surname groups. Lots of Dohertys, > McLaughlins, Ewings and Cowans do not have SNP tests because they were > advised it was unnecessary to test all by the experts. That probably changed > some of the percentages. I think you said some of your Kennedys fall into > that category too. But that's the reason I've so far resisted either > requiring SNP tests or segregating the results into SNP tested and non-tested. > I could do that easily though. I could also change the requirements for > joining to include at least an M222+ result. I hesitate to do that though. > > I'm willing at this point to go along with what people in the project think > is best. Would segregating the SNP tested and non tested into different > groups satisfy the purists who demand SNP tests for every sample? > Realistically everyone now in the project is not going to jump up and order an SNP > test because the purists demand it. > > They're right of course. Sort of. You can't tell absolutely without a > shadow of doubt if someone is M222 without an SNP test. But those who claim > that one can't identify M222 on the basis of STRS also are full of it. I > was just told that last week by someone in the L21 project. I don't have > any idea what the certainty level based on STR analysis would be though. I'm > sure it's very high. In the high 90s. A few ringers could sneak in. > > Bill Howard ran some statistical analyses of the project a while back. He > mentioned in vague terms a few of them. Basically he found no difference > in results from sets using 37 markers vs. 67 markers; and also found no > significant differences between the SNP tested members and those who weren't. > You'd have to ask him how he did this. I barely remember the > conversations at this point. > > > > John > R1b1c7 Research and Links: > > http://clanmaclochlainn.com/R1b1c7/ > ------------------------------- > To unsubscribe from the list, please send an email to DNA-R1B1C7-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message R1b1c7 Research and Links: http://clanmaclochlainn.com/R1b1c7/ ------------------------------- To unsubscribe from the list, please send an email to DNA-R1B1C7-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message