I think part of the issue is that your definition of the "M222 group" deliberately includes not just people whose M222 status is unknown but people who are definitely M222- but have STR profiles similar to M222+ folks. I realize this has to do with the history of the North-West Irish group, but it is extremely confusing. To take an example from elsewhere in amateur Y-DNA studies, the modal values for L21 and its parent P312 are identical but we do not lump P312+ L21- men in with the "L21 group" even for convenience's sake. Rather than the "M222 group", perhaps "M222 and relatives" or "North-West Irish cluster" might clear some of this up. I appreciate that your genetic distance measure is independent of SNP information but surely it's a useful check on your data to know that the real TRMCA between a M222+ man and a M222- man must be greater than those for any two M222+ men. Just out of curiosity, have you done coalescence estimates on all men who are known to be M222+ from test results versus everyone with in the broader group defined by the STR signature? Steve On 8 July 2011 14:00, Bill Howard <weh8@verizon.net> wrote: > Hi, Paul, > Sorry about the "Pat" -- someone else used Pat along the way and I failed > to check it. My apologies. > I only wanted to make the point that the so-called plusses and minuses for > M222 are inextricably mixed so it doesn't matter if you use both in studies > related to the M222 group. > Indeed, the RCC time scale does not depend on M222 membership at all. It > will estimate the TMRCAs of any pair of haplotypes, and when it is looking > at a surname cluster, or any other cluster that has many members, the > estimate goes from about 180 years (SD) for a pair, downwards by a factor of > the square root of (n-1) where n is the number of haplogroups in the > cluster. The junction points on the phylogenetic tree are also pretty > well-defined in time using the same reasoning. > Whether or not it is a SNP is a red-herring. They only have to be > haplotypes. > - Bye from Bill Howard > > On Jul 8, 2011, at 12:52 PM, Paul Conroy wrote: > > > Bill, > > > > First of all my name is Paul Conroy, not "Pat" - he's the writer, I'm the > > hacker! > > > > Second, what I said was the R-L21 M222+ and R-L21 M222- are mostly not > > significantly different in STR values. So in areas where they both occur, > a > > NPE would likely go largely unnoticed based solely on STR values. However > > any calculations that date the TMRCA or possible Location of said > ancestor, > > based on Non-SNP tested data, are liable to be erroneous. That is as true > a > > statement now as ever before. > > > > However, I don't think anyone should be pressured into getting SNP tested > - > > although they should be encouraged to do so. But any investigator who is > > carrying out calculations based on Non-SNP tested data, should state that > > upfront, and add the caveat that results are thus likely to be erroneous > - > > simple as that. The alternative of course would be just to stick to SNP > > tested data, and avoid the confusion altogether?! > > > > I have no more to say on this - as IMO it should be obvious that you > can't > > predict dates for clusters defined by certain SNP's, when some of the > data > > don't actually have those SNP's. > > > > Cheers, > > Paul > > > > On Fri, Jul 8, 2011 at 12:34 PM, Bill Howard <weh8@verizon.net> wrote: > > > >> My primary interest in replying to Paul Conroy was to state that there > is > >> little if any difference between what he calls the M222+ and M222 - > groups. > >> I am glad to see that John McLaughlin and apparently David Wilson agree > with > >> me. This is crucial to what I am trying to do because the M222 group I > am > >> working on consists of both Conroy's groups and I find them > >> indistinguishable. I appreciate the remarks of John and others about the > >> history of what defines the M222 group. > >> > >> In John's note he cites a URL where the marker modals are found. It is > at: > >> <http://clanmaclochlainn.com/R1b1c7/M222repeat.htm> > >> > >> There MUST be a mistake of transcription there, because there are two > >> entries in the clanmaclochlainn.com file that disagree significantly > with > >> my modal list. Before we go further, my modal file is identical when I > >> derive them from the O'Cathain group that John McLaughlin sent me AND > with > >> two larger files of M222, one of which contains 320 haplotypes and the > other > >> contains 683 haplotypes. My conclusion is that the > clanmaclochlainn.comfile MUST be wrong for the following DYS values: > >> > >> DYS My Value clanmaclochlainn.comValue > >> 426 12 13 > >> 454 11 12 > >> > >> On the website that gives modal values for M222, located at < > >> > http://www.littlescottishcluster.com/RL21/Files/R-L21%20Tree%2020110531.pdf > >, > >> the values for these markers agree with mine, not the posting on > >> clanmaclochlainn.com. > >> Let values on the latter site are wrong. They are crucial since they are > >> very slow mutators, hence are valuable for analysis of the phylogenetic > tree > >> at high values of RCC. I think someone needs to look into the details > of > >> how the clanmaclochlainn values were done. It surely needs correction > for > >> those two markers. > >> > >> - Bye from Bill > >> > >> > >> > >> On Jul 7, 2011, at 10:16 PM, Lochlan@aol.com wrote: > >> > >>> In a message dated 7/7/2011 7:00:26 A.M. Central Daylight Time, > >>> weh8@verizon.net writes: > >>> > >>> There has been considerable discussion both on- and off-line about how > >> the > >>> M222 SNP is defined. > >>> First, I understand that its early definition depended on the first 12 > >>> markers. > >>> Next, we have the deep clade test of FTDNA with a proprietary approach > >> we > >>> know little about. > >>> Next, there are discussions of how the markers agree or disagree with > >> the > >>> modal values of the deep clade test, but only with respect to the > first > >> 12 > >>> markers of the FTDNA string. > >>> > >>> Bill, a few others have chimed in on various aspects of these > >> assumptions > >>> but I have to add myself they are wrong. > >>> > >>> <First, I understand that its early definition depended on the first > 12 > >>> markers. > >>> > >>> 25 markers was the norm at FTDNA when David Wilson first announced his > >>> discovery of what he called the NW Irish cluster. He was soon followed > >> by the > >>> Trinity team with 17 markers who called the same cluster the Irish > Modal > >>> Haplotype (IMH). > >>> > >>> <Next, we have the deep clade test of FTDNA with a proprietary > approach > >> we > >>> know little about. > >>> > >>> > >>> This has already been covered. It's not proprietory and details are > >>> readily available. > >>> > >>> <Next, there are discussions of how the markers agree or disagree with > >> the > >>> modal values of the deep clade test, but only with respect to the first > >> 12 > >>> markers of the FTDNA string. > >>> > >>> This has never been true anywhere that I know of and never in the M222 > >>> project. David Wilson has had a full modal haplotype online at > Ysearch > >> as > >>> long as I can remember. It is true that FTDNA awards their Nial icon > >> based > >>> on an exact match at 12 markers. Perhaps this is what you're thinking > >> of. > >>> > >>> David's M222 modal is M5UKQ on Ysearch. > >>> > >>> < I include now a table that shows the percentage of M222 testees that > >>> have mutations at the various points in the haplotype. For example, > those > >> with > >>> 454 had a constant value of the modal for 454, and less than 50% of the > >>> testees had the modal for the two CDYs. > >>> > >>> That's already been done. > >>> > >>> _http://clanmaclochlainn.com/R1b1c7/M222repeat.htm_ > >>> (http://clanmaclochlainn.com/R1b1c7/M222repeat.htm) > >>> > >>> > >>> If this thread continues much longer I'll attempt to get David Wilson > >> back > >>> on board. He no longer is a member of the list but he surely will > have > >>> some strong opinions on some of the statements made lately. He began > >> the > >>> M222 project and was the one who made the decision that SNP testing was > >> not > >>> necessary for inclusion in the project. Why? Because membership in > >> the > >>> clade was easily detected via STRs alone. > >>> > >>> Pat Conroy threw out the Wilson sample as an example of a false M222 > >> based > >>> on STRS. I'm glad someone else mentioned that. I have five times in > >> the > >>> last month to no response. The Clark sample in the M222 negative > >> section > >>> is almost identical. David Wilson listed these two as possibly pre > >>> M222. There is one other similar sample (Hannan) listed in the M222 > >> negative > >>> section. He probably should go into the pre M222 section as well. > >>> > >>> The Wilson sample has an interesting history. David Wilson who was > >> also > >>> the admin. of the Wilson DNA project. He assumed this sample was > M222+ > >>> and was shocked to learn it was M222- by SNP test. At David's > urging, > >>> FTDNA retested the sample several times. It came back negative every > >> time. > >>> > >>> There are a few suspicious parts of both the Clark and Wilson samples. > >>> Both have 385 = 12-14 which is unusual in M222. The Clark sample at > 464 > >> has > >>> 15-15-16-17 at 464. That is highly unusual in M222. I see nothing > >> unusual > >>> in the Hannan sample at all. None of these samples stand out as > unusual > >>> in terms of genetic distance. Each has matchers within the project at > >>> 6,7-20 GDS which is no different from most other cluster members. > >> That's no > >>> different from the Conroy sample. > >>> > >>> If I were presented with any of these samples and had to chose whether > to > >>> admit them to the project based on STRS alone I probably would without > an > >>> SNP test. I think David Wilson would do the same. > >>> > >>> So I will be the first to admit there could be some samples in the > >>> project which appear to be M222 but are not. The question is how > many? > >>> > >>> What Bill has not mentioned to the list is a theory that the > bi-modality > >> he > >>> mentioned is caused by M222 project samples that are M222-. The twin > >>> humps his graph shows seems to be equally large on each side. That > >> would seem > >>> to indicate there are roughly as many M222- in the project as M222+. > >> Is > >>> that possible? > >>> > >>> Not long ago the count of project members with SNP tests was 334. > >> That's > >>> out of 668 or about half the project. I can't guess how many of the > >> non > >>> tested members might be M222-. Not too many is my guess. > >>> > >>> The way the project is currently organized shouldn't cause any > >> confusion. > >>> FTDNA clearly indicates those with SNP tests. Those who only want to > >>> include them in analysis can do so easily. Dropping half the project > >> without > >>> SNP tests would greatly reduce the surname diversity available. Is > that > >> a > >>> wise thing to do? > >>> > >>> I for one have no interest in getting an SNP test. I belong to a > >>> cluster of McLaughlins, four or five of whom have been SNP tested and > >> are M222+. > >>> As far as I'm concerned an SNP test for me would be a waste of money. > >> I > >>> won't spend the extra money just to keep the purists happy. > >>> > >>> On the other hand there could be a few samples in the project who are > not > >>> M222 positive. It would be interesting to find out for sure. Right > >> now > >>> we can only guess. > >>> > >>> > >>> That is a conundrum. > >>> > >>> > >>> 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 > >> > > 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 >

Steve, Since I cannot distinguish the two so-called groups of M222 from each other, and since all the haplotypes I include in the M222 database have at least 27 DYS modals that agree with the 37 DYS modal values, I do not deliberately include ones who are too far from the self-defined limits (above) of M222. I am doing nothing for convenience. Those haplotypes have to meet the 73% or more test. The answer to your other point(s) is that I do not find the two groups to be operationally distinguishable. - Bye from Bill Howard On Jul 8, 2011, at 2:30 PM, Stephen Forrest wrote: > I think part of the issue is that your definition of the "M222 group" > deliberately includes not just people whose M222 status is unknown but > people who are definitely M222- but have STR profiles similar to M222+ > folks. I realize this has to do with the history of the North-West Irish > group, but it is extremely confusing. > > To take an example from elsewhere in amateur Y-DNA studies, the modal values > for L21 and its parent P312 are identical but we do not lump P312+ L21- men > in with the "L21 group" even for convenience's sake. Rather than the "M222 > group", perhaps "M222 and relatives" or "North-West Irish cluster" might > clear some of this up. > > I appreciate that your genetic distance measure is independent of SNP > information but surely it's a useful check on your data to know that the > real TRMCA between a M222+ man and a M222- man must be greater than those > for any two M222+ men. > > Just out of curiosity, have you done coalescence estimates on all men who > are known to be M222+ from test results versus everyone with in the broader > group defined by the STR signature? > > Steve > > On 8 July 2011 14:00, Bill Howard <weh8@verizon.net> wrote: > >> Hi, Paul, >> Sorry about the "Pat" -- someone else used Pat along the way and I failed >> to check it. My apologies. >> I only wanted to make the point that the so-called plusses and minuses for >> M222 are inextricably mixed so it doesn't matter if you use both in studies >> related to the M222 group. >> Indeed, the RCC time scale does not depend on M222 membership at all. It >> will estimate the TMRCAs of any pair of haplotypes, and when it is looking >> at a surname cluster, or any other cluster that has many members, the >> estimate goes from about 180 years (SD) for a pair, downwards by a factor of >> the square root of (n-1) where n is the number of haplogroups in the >> cluster. The junction points on the phylogenetic tree are also pretty >> well-defined in time using the same reasoning. >> Whether or not it is a SNP is a red-herring. They only have to be >> haplotypes. >> - Bye from Bill Howard >> >> On Jul 8, 2011, at 12:52 PM, Paul Conroy wrote: >> >>> Bill, >>> >>> First of all my name is Paul Conroy, not "Pat" - he's the writer, I'm the >>> hacker! >>> >>> Second, what I said was the R-L21 M222+ and R-L21 M222- are mostly not >>> significantly different in STR values. So in areas where they both occur, >> a >>> NPE would likely go largely unnoticed based solely on STR values. However >>> any calculations that date the TMRCA or possible Location of said >> ancestor, >>> based on Non-SNP tested data, are liable to be erroneous. That is as true >> a >>> statement now as ever before. >>> >>> However, I don't think anyone should be pressured into getting SNP tested >> - >>> although they should be encouraged to do so. But any investigator who is >>> carrying out calculations based on Non-SNP tested data, should state that >>> upfront, and add the caveat that results are thus likely to be erroneous >> - >>> simple as that. The alternative of course would be just to stick to SNP >>> tested data, and avoid the confusion altogether?! >>> >>> I have no more to say on this - as IMO it should be obvious that you >> can't >>> predict dates for clusters defined by certain SNP's, when some of the >> data >>> don't actually have those SNP's. >>> >>> Cheers, >>> Paul >>> >>> On Fri, Jul 8, 2011 at 12:34 PM, Bill Howard <weh8@verizon.net> wrote: >>> >>>> My primary interest in replying to Paul Conroy was to state that there >> is >>>> little if any difference between what he calls the M222+ and M222 - >> groups. >>>> I am glad to see that John McLaughlin and apparently David Wilson agree >> with >>>> me. This is crucial to what I am trying to do because the M222 group I >> am >>>> working on consists of both Conroy's groups and I find them >>>> indistinguishable. I appreciate the remarks of John and others about the >>>> history of what defines the M222 group. >>>> >>>> In John's note he cites a URL where the marker modals are found. It is >> at: >>>> <http://clanmaclochlainn.com/R1b1c7/M222repeat.htm> >>>> >>>> There MUST be a mistake of transcription there, because there are two >>>> entries in the clanmaclochlainn.com file that disagree significantly >> with >>>> my modal list. Before we go further, my modal file is identical when I >>>> derive them from the O'Cathain group that John McLaughlin sent me AND >> with >>>> two larger files of M222, one of which contains 320 haplotypes and the >> other >>>> contains 683 haplotypes. My conclusion is that the >> clanmaclochlainn.comfile MUST be wrong for the following DYS values: >>>> >>>> DYS My Value clanmaclochlainn.comValue >>>> 426 12 13 >>>> 454 11 12 >>>> >>>> On the website that gives modal values for M222, located at < >>>> >> http://www.littlescottishcluster.com/RL21/Files/R-L21%20Tree%2020110531.pdf >>> , >>>> the values for these markers agree with mine, not the posting on >>>> clanmaclochlainn.com. >>>> Let values on the latter site are wrong. They are crucial since they are >>>> very slow mutators, hence are valuable for analysis of the phylogenetic >> tree >>>> at high values of RCC. I think someone needs to look into the details >> of >>>> how the clanmaclochlainn values were done. It surely needs correction >> for >>>> those two markers. >>>> >>>> - Bye from Bill >>>> >>>> >>>> >>>> On Jul 7, 2011, at 10:16 PM, Lochlan@aol.com wrote: >>>> >>>>> In a message dated 7/7/2011 7:00:26 A.M. Central Daylight Time, >>>>> weh8@verizon.net writes: >>>>> >>>>> There has been considerable discussion both on- and off-line about how >>>> the >>>>> M222 SNP is defined. >>>>> First, I understand that its early definition depended on the first 12 >>>>> markers. >>>>> Next, we have the deep clade test of FTDNA with a proprietary approach >>>> we >>>>> know little about. >>>>> Next, there are discussions of how the markers agree or disagree with >>>> the >>>>> modal values of the deep clade test, but only with respect to the >> first >>>> 12 >>>>> markers of the FTDNA string. >>>>> >>>>> Bill, a few others have chimed in on various aspects of these >>>> assumptions >>>>> but I have to add myself they are wrong. >>>>> >>>>> <First, I understand that its early definition depended on the first >> 12 >>>>> markers. >>>>> >>>>> 25 markers was the norm at FTDNA when David Wilson first announced his >>>>> discovery of what he called the NW Irish cluster. He was soon followed >>>> by the >>>>> Trinity team with 17 markers who called the same cluster the Irish >> Modal >>>>> Haplotype (IMH). >>>>> >>>>> <Next, we have the deep clade test of FTDNA with a proprietary >> approach >>>> we >>>>> know little about. >>>>> >>>>> >>>>> This has already been covered. It's not proprietory and details are >>>>> readily available. >>>>> >>>>> <Next, there are discussions of how the markers agree or disagree with >>>> the >>>>> modal values of the deep clade test, but only with respect to the first >>>> 12 >>>>> markers of the FTDNA string. >>>>> >>>>> This has never been true anywhere that I know of and never in the M222 >>>>> project. David Wilson has had a full modal haplotype online at >> Ysearch >>>> as >>>>> long as I can remember. It is true that FTDNA awards their Nial icon >>>> based >>>>> on an exact match at 12 markers. Perhaps this is what you're thinking >>>> of. >>>>> >>>>> David's M222 modal is M5UKQ on Ysearch. >>>>> >>>>> < I include now a table that shows the percentage of M222 testees that >>>>> have mutations at the various points in the haplotype. For example, >> those >>>> with >>>>> 454 had a constant value of the modal for 454, and less than 50% of the >>>>> testees had the modal for the two CDYs. >>>>> >>>>> That's already been done. >>>>> >>>>> _http://clanmaclochlainn.com/R1b1c7/M222repeat.htm_ >>>>> (http://clanmaclochlainn.com/R1b1c7/M222repeat.htm) >>>>> >>>>> >>>>> If this thread continues much longer I'll attempt to get David Wilson >>>> back >>>>> on board. He no longer is a member of the list but he surely will >> have >>>>> some strong opinions on some of the statements made lately. He began >>>> the >>>>> M222 project and was the one who made the decision that SNP testing was >>>> not >>>>> necessary for inclusion in the project. Why? Because membership in >>>> the >>>>> clade was easily detected via STRs alone. >>>>> >>>>> Pat Conroy threw out the Wilson sample as an example of a false M222 >>>> based >>>>> on STRS. I'm glad someone else mentioned that. I have five times in >>>> the >>>>> last month to no response. The Clark sample in the M222 negative >>>> section >>>>> is almost identical. David Wilson listed these two as possibly pre >>>>> M222. There is one other similar sample (Hannan) listed in the M222 >>>> negative >>>>> section. He probably should go into the pre M222 section as well. >>>>> >>>>> The Wilson sample has an interesting history. David Wilson who was >>>> also >>>>> the admin. of the Wilson DNA project. He assumed this sample was >> M222+ >>>>> and was shocked to learn it was M222- by SNP test. At David's >> urging, >>>>> FTDNA retested the sample several times. It came back negative every >>>> time. >>>>> >>>>> There are a few suspicious parts of both the Clark and Wilson samples. >>>>> Both have 385 = 12-14 which is unusual in M222. The Clark sample at >> 464 >>>> has >>>>> 15-15-16-17 at 464. That is highly unusual in M222. I see nothing >>>> unusual >>>>> in the Hannan sample at all. None of these samples stand out as >> unusual >>>>> in terms of genetic distance. Each has matchers within the project at >>>>> 6,7-20 GDS which is no different from most other cluster members. >>>> That's no >>>>> different from the Conroy sample. >>>>> >>>>> If I were presented with any of these samples and had to chose whether >> to >>>>> admit them to the project based on STRS alone I probably would without >> an >>>>> SNP test. I think David Wilson would do the same. >>>>> >>>>> So I will be the first to admit there could be some samples in the >>>>> project which appear to be M222 but are not. The question is how >> many? >>>>> >>>>> What Bill has not mentioned to the list is a theory that the >> bi-modality >>>> he >>>>> mentioned is caused by M222 project samples that are M222-. The twin >>>>> humps his graph shows seems to be equally large on each side. That >>>> would seem >>>>> to indicate there are roughly as many M222- in the project as M222+. >>>> Is >>>>> that possible? >>>>> >>>>> Not long ago the count of project members with SNP tests was 334. >>>> That's >>>>> out of 668 or about half the project. I can't guess how many of the >>>> non >>>>> tested members might be M222-. Not too many is my guess. >>>>> >>>>> The way the project is currently organized shouldn't cause any >>>> confusion. >>>>> FTDNA clearly indicates those with SNP tests. Those who only want to >>>>> include them in analysis can do so easily. Dropping half the project >>>> without >>>>> SNP tests would greatly reduce the surname diversity available. Is >> that >>>> a >>>>> wise thing to do? >>>>> >>>>> I for one have no interest in getting an SNP test. I belong to a >>>>> cluster of McLaughlins, four or five of whom have been SNP tested and >>>> are M222+. >>>>> As far as I'm concerned an SNP test for me would be a waste of money. >>>> I >>>>> won't spend the extra money just to keep the purists happy. >>>>> >>>>> On the other hand there could be a few samples in the project who are >> not >>>>> M222 positive. It would be interesting to find out for sure. Right >>>> now >>>>> we can only guess. >>>>> >>>>> >>>>> That is a conundrum. >>>>> >>>>> >>>>> 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 >>>> >>> 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 >> > 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

Bill, perhaps I was being unclear. I'm not proposing trying to distinguish M222+ and M222- groups on any other basis than a SNP test, and I fully accept they may not be "operationally distinguishable" on the basis of STRs. That is exactly why we do SNP testing. Surely we have real SNP results for some subset of the group in question, and that SNP data will be valuable in clarifying long-distance relationships for the reason I stated before: that any two men who are M222+ must have a common male-line ancestor more recently than any pair men who differ on the M222 SNP. (This is based on the theory of M222 as a unique event polymorphism, which we have no reason to doubt.) My use of the phrase "for convenience" was not pejorative and was only referring to your choice of nomenclature in using the phrase "the M222 group" to refer everyone with the North-West Irish STR signature, regardless of their status on the M222 SNP. I personally think it would be very interesting to see what your models predict for a TMRCA of all men who are provably M222+ (by a SNP test) versus a TMRCA for all men with the North-West Irish STR signature. The second number should be bigger than the first, but by how much? Steve On 8 July 2011 15:03, Bill Howard <weh8@verizon.net> wrote: > Steve, > Since I cannot distinguish the two so-called groups of M222 from each > other, and since all the haplotypes I include in the M222 database have at > least 27 DYS modals that agree with the 37 DYS modal values, I do not > deliberately include ones who are too far from the self-defined limits > (above) of M222. I am doing nothing for convenience. Those haplotypes have > to meet the 73% or more test. The answer to your other point(s) is that I do > not find the two groups to be operationally distinguishable. > - Bye from Bill Howard

Hi, Steve, Mathematica can only operate on sets of haplotypes it is given. They can be in any order. Mathematica has no way of knowing who's a SNP member or not, but if it is presented with two groups of haplogroups, one in the group and one not, then we can compare the trees. Of course, mutations will still affect the positions on the tree, as will the different members of haplotypes presented to Mathematica. The list of M222 haplotypes I received from John McLaughlin consisted of 320 testees who apparently were in haplogroup R1b1a2a1a1b4b and they were all labeled as such. As I recall John said that all had not been tested (and in his and Wilson's opinion did not need to be). I had a larger group reported to be M222 but which were not represented as being tested. Both sets had similar statistics, as I wrote earlier. Now, if someone would send me a set of haplotypes that are bonafide as having been SNP-tested and put into the M222+ bin, I will run a tree separately on them and do an analysis on them. One of my efforts is to try to determine the time at which M222 originated. If membership in the SNP group is highly correlated with the marker strings of members in M222, then I will get a defendable result. It does indeed appear to be highly correlated so I feel quite safe with my approach. - Bye from Bill Howard On Jul 8, 2011, at 3:34 PM, Stephen Forrest wrote: > Bill, perhaps I was being unclear. I'm not proposing trying to distinguish > M222+ and M222- groups on any other basis than a SNP test, and I fully > accept they may not be "operationally distinguishable" on the basis of STRs. > That is exactly why we do SNP testing. > > Surely we have real SNP results for some subset of the group in question, > and that SNP data will be valuable in clarifying long-distance relationships > for the reason I stated before: that any two men who are M222+ must have a > common male-line ancestor more recently than any pair men who differ on the > M222 SNP. (This is based on the theory of M222 as a unique event > polymorphism, which we have no reason to doubt.) > > My use of the phrase "for convenience" was not pejorative and was only > referring to your choice of nomenclature in using the phrase "the M222 > group" to refer everyone with the North-West Irish STR signature, regardless > of their status on the M222 SNP. > > I personally think it would be very interesting to see what your models > predict for a TMRCA of all men who are provably M222+ (by a SNP test) versus > a TMRCA for all men with the North-West Irish STR signature. The second > number should be bigger than the first, but by how much? > > Steve > > On 8 July 2011 15:03, Bill Howard <weh8@verizon.net> wrote: > >> Steve, >> Since I cannot distinguish the two so-called groups of M222 from each >> other, and since all the haplotypes I include in the M222 database have at >> least 27 DYS modals that agree with the 37 DYS modal values, I do not >> deliberately include ones who are too far from the self-defined limits >> (above) of M222. I am doing nothing for convenience. Those haplotypes have >> to meet the 73% or more test. The answer to your other point(s) is that I do >> not find the two groups to be operationally distinguishable. >> - Bye from Bill Howard > 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

So R-M222 group is in reality the NWI modal group? M222+ is a (maybe large) subset of the NWI modal (the STR values) and previous emails are discussing the age of the NWI modal (not M222 directly). Now whether the age of NWI reflects then M222 occurred depends on what portion of NWI is also M222+? The same question answers the worth of SNP testing. So do we know the mix of M222+/M222- of those within the NWI bucket? And do we have have a good NWI bucket to sample from?