As usual excellent answer, Tim. Good question as well, it's not well explained by 23andMe but a very powerful tool and information that no other DNA testing service provides. Andreas > On 27 Nov 2017, at 09:47, Tim Janzen <[email protected]> wrote: > > Dear Eric, > The answer to your first question is "yes". > The reason that 23andMe includes the people in the "no" category is that > this information can be helpful if you are trying to organize clusters of > related people. Let's say that you are one of 4 people who have tested at > 23andMe and that these 4 people are all 4th cousins to each other on > different lines of descent from a shared ancestral couple. Let's call these > 3 cousins of yours A, B, and C to keep them straight. Let's say that you > share the same segment of DNA with A and B and thus the 3 of you form a > triangulated group. Let's say that C shares a different segment of DNA with > you than the one you share with A and B and that C also shares a DNA segment > with both A and B. It is helpful for you to know if A and/or B share DNA > with C. In such situations 23andMe will enter a "no" in the Shared DNA > column under C's name on the rows for A and B when you pull up C on your > match list in DNA Relatives. > > A secondary question has to do with the probability that a small percentage > in the column under C's name is genealogically significant when A and B > share an HIR with C. There are several possibilities here: > 1. The HIR that C shares with A or B is a false match. > 2. The HIR that C shares with A or B represents a true shared segment (IBD) > passed down through the same ancestral line that you share in common with A, > B, and C. > 3. The HIR that C shares with A or B represents a true shared segment (IBD) > passed down through a different ancestral line than the one you know you > share in common with A, B, and C. > > It is up to you to sort out which of the above the possibilities is the > correct one. Chromosome mapping and using phased data can help you sort > these issues out. I have often wondered what the probability is that > cousins such as A, B, and C will have either option #1 above occur or option > #3 occur. The probability of option #3 will of course depend to a > significant extent on the degree of endogamy among A, B, and C. The higher > the degree of endogamy, the higher the probability of option #3 occurring. > The probability of option #1 is probably reasonably high if the shared > percentage is 10% or less. So there is definitely some degree of "chaff" in > the "no" column. It is up to you to sort the wheat from the chaff. > > I would like it if 23andMe would allow us to sort the Shared DNA column and > the column next to it, but they haven't created that option yet. > Sincerely, > Tim Janzen > > > > -----Original Message----- > From: GENEALOGY-DNA [mailto:[email protected]] On Behalf Of > Eric S Johnson > Sent: Sunday, November 26, 2017 4:38 PM > To: [email protected] > Subject: [DNA] 23andMe TGs > > So I can't help but wonder whether I'm missing something: 23andMe's "new > experience" provides, at the bottom of a match page's left-hand pane, > "relatives in common." As I understand it, anyone for whom "yes" is marked > is part of a "triangulated group" consisting of "me," my 23andMe "DNA > cousin," and all the other "yes"-marked folks in this RiC list. Correct? > > So, why does this list include anyone marked as "no"? Aren't those "no"s > just chaff? In which case, why's 23andMe include them (thus forcing us to > click through page after page of RiCs in order to find the few "yes"s)? > > Best, > > Eric > > > ------------------------------- > To unsubscribe from the list, please send an email to [email protected] with the word 'unsubscribe' without the quotes in the subject and the body of the message