Hi Joe, Thanks for the detailed reply. Maybe I've been looking in the wrong places but I haven't been able to find answers to my question despite countless Google searches. I'm still puzzled about the criteria used to place testees into family groupings. Following is a set of 12 marker results for 6 fictitious individuals that might be considered as a single family group: 13 22 13 10 13 14 11 14 11 12 12 30 14 22 13 10 13 14 11 14 11 12 12 30 14 23 13 10 13 14 11 14 11 12 12 30 14 23 14 10 13 14 11 14 11 12 12 30 14 23 14 11 13 14 11 14 11 12 12 30 14 23 14 11 14 14 11 14 11 12 12 30 Each person matches on 11 of 12 markers with the person above and/or below on the list and there is a genetic distance of 1 on the marker that is not a match. Various websites explain the % chance of sharing a common ancestor within a certain number of generations where there is a genetic distance of 1. If they share a common surname that % should be the same for each of the 6 testees when comparing results with those immediately above or below but the % decreases as the genetic distance increases. There is a genetic distance of 5 between the first and last person so the chance of them sharing a common ancestor is far less. In isolation the first and last do not appear to belong to the same family group: 13 22 13 10 13 14 11 14 11 12 12 30 14 23 14 11 14 14 11 14 11 12 12 30 When viewed with the results of the other testees the links become obvious. When I compare my personal DNA results with others there are many that are similar but may have a genetic distance of 1, 2 , 3, 4 etc. Maybe I belong in one of their family groups but the people that would provide the necessary links just haven't been tested? Logically though this must mean I potentially belong to many family groups and it just depends on where the arbitrary lines are drawn between families. This is the reason why I doubt that family groupings are relevant except for exact matches. I think this just proves that we are probably all related if you look far enough back in time. Regards, Steve Joe Marshall wrote: > Hi , You won't find those answers on this group or the Marshall DNA > site, it is on other DNA sites and School sites, etc. > > my caps is sticking so sorry if I miss a few > > Our Marshall DNA site is 1999. Still no responses from FTDNA other than > "we will look into it". You would need to look around at other sources > to get the info. There are lots out there. I have a link below to one > page to look at. > > I had more than 12 markers done but he does not add them. I have > people write to me and say they never have been posted. There are a > bunch on there that should be in the Simon Granser Marshall line based > on the numbers > > on one other note first, we could actually find that the ancient > Marshall's did think they were related because of name. Or there could > have been some adoption in 1400 or something that created another DNA > line. It is only a method to see who is blood relative at some point in > time. It could be from 1200's when they first started using last names. > Or in some other case such as a totally different name, there could > have been some adoption in 800 for all we know. We will also all have > other name matches from back in time before the use of last names. > > Remember this, when there were not that many people, they did not use > last names. They were Sir John of Essex, for example. He was the only > one in that town with that name. You have to imagine a much smaller > neighborhood than we have today. > > > The help you get from it depends on when you are stuck in your tree. If > you are stuck 100 years ago, you would want numbers that are nearly > 100% identical. If you are stuck 500 years ago, the branches will have > changed a bit. > the closer you are on the tree, the closer your numbers usually are. > > to find the truth you must read and remember history and facts that > some are not taught > > your haplogroup came from one person long ago, it is a fact > (haplogroup is not even in the spell checker) > WHEN, is another question that can not be scientifically proven > > It is known though, that I1b is the Normans and R1b is the Irish > possibly "Picts" that were there longer than the Normans > > certain markers are prone to genetic mutation. others do not change at > all. it has to do with where you live ie: the sun rays hitting you/ > radiation. this means a higher mutation rate of those certain numbers. > If you live near the equator you will have more mutation. that is a > fact. when you go outside and feel the sun burning you, that is > radiation. You have to take in to account that some base mutation rates > are very low. some match exact because they are a father and son as > noted on the charts. You would probably have to have someone in your > exact branch to be an exact match with no mutation. Now if you get a > lot of Xrays or something, that will possibly do it. > > one mutation in Group F is the first digit 15 to 14 and the 12th digit > > > those certain numbers that do not change or not much, are probably the > ones you match with every one else. > > sometimes those numbers are in red on charts. the ones that change more > > > you can type in numbers here and experiment > > http://www.mymcgee.com/tools/yutility.html?mode=ftdna_mode > > I can not get this page to work on my site at the moment > > > > > > Joe Marshall > > gggg grandson Aaron Marshall b. abt 1750 > 10th Virginia Militia 1776 > ggggg grandson Col. John Marshel > > r[EVOL]ution calling > > http://www.MiscellaneaMarescalliana.org > > rootsweb: http://wc.rootsweb.com/cgi-bin/igm.cgi?db=aaronmarshel1740 > > DNA site: http://www.chipster.us/AaronMarshallDNAline/default.htm > > marshall DNA: > http://freepages.genealogy.rootsweb.com/~marshalldna/MarshallDNA1/ > > > -----Original Message----- > From: Stephen Marshall <smarshall@meridianfreight.com.au> > To: marshall@rootsweb.com > Sent: Tue, 29 Apr 2008 2:06 am > Subject: [MARSHALL] How are family groups determined in the Marshall > DNA Project? > > > > > > > > > > > Hi, > > Can anyone explain how family groups are determined for DNA test > subjects? > > I used to think that subjects were placed in groups where there was a > high level of marker matches, say 11 of 12, and where the non-matching > marker varied by only 1 or 2. Probably a common haplo group was also > important. I'm now unsure how scientific this process can be (I mean > the > arbitrary grouping of individuals not the DNA testing itself). > > By way of example, on the Marshall DNA Project website, Group A is > shown > as having 5 testees. As I read from testee 1 down to testee 5 I see the > similarities and assume that's why they've been grouped together. If > you > look, however, at 1 and 5 in isolation they match on only 8 of 12. If > results for the remaining testees were not available I presume 1 and 5 > would not be in the same group. > > Logically this then raises the question of whether many other testees > with perhaps only 8 of 12 matches may belong within a common family > group but proof is just not available (ie there may be "linking" > individuals who have not been tested). > > My next query relates to the increased use of 25, 37 and 67 marker > testing. In determining family groups are the first 12 markers given > more weight than remaining markers? Two individuals could match 100% on > the 1st 12 but vary considerably thereafter. Would they then belong in > the same family group? I have had a 67 marker test done for myself. I > match on more than 12 markers with a few other testees but not > necessarily the 1st 12 or even 12 consecutive markers. > > I find this really hard to understand. I'd love to read other people's > views on this. > > Regards, > > Steve Marshall > Sydney > Australia > > > ------------------------------- > To unsubscribe from the list, please send an email to > MARSHALL-request@rootsweb.com > with the word 'unsubscribe' without the quotes in the subject and the > body of > the message > > > > > > > ------------------------------- > To unsubscribe from the list, please send an email to MARSHALL-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message >