Some years ago, I compiled a table of the number of individuals in each generation of your ancestry: http://dgmweb.net/Ancillary/OnE/NumberAncestors.html That was when I set my genealogical goal at getting all my lines back ten generations. It's almost ridiculous to be proud of an ancestor 20 or 30 generations back. At best, they've contributed a miniscule amount to your genetic endowment - possibly even none. Nothing's impossible, but I'm going to be very surprised to find my parents related within the reach of the FamilyFinder test. If these mutual matches I'm getting from both sides of my family are IBS, that's solves that problem, but then that raises the question as to whether I should have accepted the DAVIS in this table as our cousin: http://dgmweb.net/DNA/aDNA/FF-Descendants_Isaac_DAVIS.html#table Diana > -----Original Message----- > From: [email protected] [mailto:autosomal-dna- > [email protected]] On Behalf Of Jim Bartlett > Sent: Wednesday, February 15, 2012 8:10 AM > To: [email protected] > Subject: Re: [AUTOSOMAL-DNA] when is a match not a match? > > Diana > > Through intensive research, spurred on by atDNA testing and helped > by collaboration with my atMatches, I've determined my parents were > 8th cousins! They never knew it. From their common ancestor, Sion W > HILL 1654-1705 VA, their respective lines of descent followed very > different paths through VA until their marriage. > > When I learned this, it explained (to me) why I was finding > identical atDNA segments in atMatches with paper trails on both > sides. I have since learned that most of these small segments are > IBS flotsam and jetsam. > > In looking at my tree, I know all my ancestors out to 16/16. Then > it's 28/32, 48/64, 73/128, 108/256, 160/512, 192/1024, and 236/2048 > (roughly 10 percent). The point is that there is a high percentage > of unknown ancestors, where an atDNA Common Ancestor could be hiding > - including those that would make you your own cousin. > > Do the math at your 6th cousin level (# known ancestors/128) and > subtract from 1 (even an EE can do it) - this will give you a sense > of what you don't know about your ancestry. I picked 6th cousin > level, because cousins at that level would look like 4th or 5th. > > Jim - Sent from my iPhone - FaceTime! > > On Feb 15, 2012, at 2:58 AM, "Diana Gale Matthiesen" > <[email protected]> wrote: > > > I just noticed something that has me perplexed. I have now FF > tested > > three cousins on my father's side of the family and one on my > mother's > > side. I just noticed that that there are two people in common > with > > the me and with a cousin on each side of my family. What has me > > perplexed is that there is nothing in my parents' paper > genealogies to > > even suggest they have any recent ancestors in common. > > > > The relationships suggested here are 3rd and 4th cousin to 4th- > remote > > and 5th-remote. What is the probability that someone matching at > > these levels is *not* actually related in genealogical time? And > what > > is the probability of two people matching the same two people at > these > > levels and still not be related? > > > > Diana > > > > ______________________________ > For answers to Frequently Asked Questions about mailing lists, > please see: > http://dgmweb.net/MailingListFAQs.html > > > ------------------------------- > To unsubscribe from the list, please send an email to AUTOSOMAL-DNA- > [email protected] with the word 'unsubscribe' without the quotes > in the subject and the body of the message

I think I didn't frame my question correctly. I'm not asking for a yes/no answer to the question of whether or not these individuals are related - I know it's not that easy. I'm asking what the probability is or, rather, what everyone's empirical experience is. In other words, for the confirmed cousins you have tested (ones with a known paper connection), how many mutual matches did you get (matches "in common with")? In my case (one each): 1st cousin = 30 3rd cousin (1R) = 7 3rd cousin (2R) = 11 4th cousin = 6 4th cousin (1R) = 2 7th cousin = 2 I'm wondering what others have experienced along this line. What's bugging me, I think, is that, after nearly eight years of managing Y-DNA surname projects, I have a pretty good feel for what is and is not reasonable in terms of matching someone in genealogical time. In contrast, with FamilyFinder, I have far too little experience to even have a sense of what is or is not plausible. I'm just looking for ballpark figures here of the typical number of mutual matches at each level of cousin, so I know when to ignore something and when to take a closer look. Or has someone already compiled something along this line? Diana

I am answering this because my great-grandfather was adopted so I am kin to several people who may not even have him in there line. I did DNA to find out who he was related to. The DNA shows us related to alot of McDonalds. I know that people moved, people were born and died without records. I am not sure how to help just wanted to throw this out there. Virginia Parrr -----Original Message----- From: Paul Wright <[email protected]> To: autosomal-dna <[email protected]> Sent: Wed, Feb 15, 2012 8:41 am Subject: Re: [AUTOSOMAL-DNA] when is a match not a match? Still dealing with unphased data. One or both cousin could be related by ifferent ancestry than that you are related to them with. Not very ncommon when dealing with more rural populations with allot of ntermarrying. On Wed, Feb 15, 2012 at 1:58 AM, Diana Gale Matthiesen [email protected]>wrote: > I just noticed something that has me perplexed. I have now FF tested three cousins on my father's side of the family and one on my mother's side. I just noticed that that there are two people in common with the me and with a cousin on each side of my family. What has me perplexed is that there is nothing in my parents' paper genealogies to even suggest they have any recent ancestors in common. The relationships suggested here are 3rd and 4th cousin to 4th-remote and 5th-remote. What is the probability that someone matching at these levels is *not* actually related in genealogical time? And what is the probability of two people matching the same two people at these levels and still not be related? Diana ______________________________ For answers to Frequently Asked Questions about mailing lists, please see: http://dgmweb.net/MailingListFAQs.html ------------------------------- 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 _____________________________ or answers to Frequently Asked Questions about mailing lists, please see: ttp://dgmweb.net/MailingListFAQs.html ------------------------------ o unsubscribe from the list, please send an email to [email protected] ith the word 'unsubscribe' without the quotes in the subject and the body of he message

Still dealing with unphased data. One or both cousin could be related by different ancestry than that you are related to them with. Not very uncommon when dealing with more rural populations with allot of intermarrying. On Wed, Feb 15, 2012 at 1:58 AM, Diana Gale Matthiesen <[email protected]>wrote: > I just noticed something that has me perplexed. I have now FF tested > three cousins on my father's side of the family and one on my mother's > side. I just noticed that that there are two people in common with > the me and with a cousin on each side of my family. What has me > perplexed is that there is nothing in my parents' paper genealogies to > even suggest they have any recent ancestors in common. > > The relationships suggested here are 3rd and 4th cousin to 4th-remote > and 5th-remote. What is the probability that someone matching at > these levels is *not* actually related in genealogical time? And what > is the probability of two people matching the same two people at these > levels and still not be related? > > Diana > > > > ______________________________ > For answers to Frequently Asked Questions about mailing lists, please see: > http://dgmweb.net/MailingListFAQs.html > > > ------------------------------- > 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 >

Diana Through intensive research, spurred on by atDNA testing and helped by collaboration with my atMatches, I've determined my parents were 8th cousins! They never knew it. From their common ancestor, Sion W HILL 1654-1705 VA, their respective lines of descent followed very different paths through VA until their marriage. When I learned this, it explained (to me) why I was finding identical atDNA segments in atMatches with paper trails on both sides. I have since learned that most of these small segments are IBS flotsam and jetsam. In looking at my tree, I know all my ancestors out to 16/16. Then it's 28/32, 48/64, 73/128, 108/256, 160/512, 192/1024, and 236/2048 (roughly 10 percent). The point is that there is a high percentage of unknown ancestors, where an atDNA Common Ancestor could be hiding - including those that would make you your own cousin. Do the math at your 6th cousin level (# known ancestors/128) and subtract from 1 (even an EE can do it) - this will give you a sense of what you don't know about your ancestry. I picked 6th cousin level, because cousins at that level would look like 4th or 5th. Jim - Sent from my iPhone - FaceTime! On Feb 15, 2012, at 2:58 AM, "Diana Gale Matthiesen" <[email protected]> wrote: > I just noticed something that has me perplexed. I have now FF tested > three cousins on my father's side of the family and one on my mother's > side. I just noticed that that there are two people in common with > the me and with a cousin on each side of my family. What has me > perplexed is that there is nothing in my parents' paper genealogies to > even suggest they have any recent ancestors in common. > > The relationships suggested here are 3rd and 4th cousin to 4th-remote > and 5th-remote. What is the probability that someone matching at > these levels is *not* actually related in genealogical time? And what > is the probability of two people matching the same two people at these > levels and still not be related? > > Diana

I just noticed something that has me perplexed. I have now FF tested three cousins on my father's side of the family and one on my mother's side. I just noticed that that there are two people in common with the me and with a cousin on each side of my family. What has me perplexed is that there is nothing in my parents' paper genealogies to even suggest they have any recent ancestors in common. The relationships suggested here are 3rd and 4th cousin to 4th-remote and 5th-remote. What is the probability that someone matching at these levels is *not* actually related in genealogical time? And what is the probability of two people matching the same two people at these levels and still not be related? Diana

Thank you so much, Sam. I did have the FTDNA browser set to show the minimum size segements, but not the one at Gedmatch. I'll see how it looks now. Hopefully I won't feel as dense this time :) melissa On Tue, Feb 14, 2012 at 6:42 AM, Sam Eaton <[email protected]> wrote: > What you are seeing is likely the results of different settings. The FTDNA Chromosome browser has a few settings that are accessed by clicking on a drop down menu. GED match is almost infinitely adjustable by changing the settings on the page where you enter your kit numbers. > > As the math used is slightly different the start and stop values number of SNPs etc. may vary.  If both browsers are set to the same settings, like 1.0 cM 500 SNPs, the results should match very closely. > > Mostly the GEDmatch browser is much more powerful and complex to use. > > > ---------------------------------------------------------------------- > > > From: M Robards <[email protected]> > Subject: [AUTOSOMAL-DNA] Differing Info on Different Chromosome >    Browsers? > > > I need some help reading the results on FTDNA's AND GedMatch's > chromosome browsers for the same three people.  I'm getting different > readings from the two browsers.  Does this mean I've done something > horribly wrong in uploading their files from FTDNA?  For instance, > FTDNA says there is some matching segment(s) on Chromosome 16 between > A and B and C.  Yet GedMatch's browser shows no matching segments from > B and C on A's.  This is pretty confusing to me. > > melissa

What you are seeing is likely the results of different settings. The FTDNA Chromosome browser has a few settings that are accessed by clicking on a drop down menu. GED match is almost infinitely adjustable by changing the settings on the page where you enter your kit numbers. As the math used is slightly different the start and stop values number of SNPs etc. may vary. If both browsers are set to the same settings, like 1.0 cM 500 SNPs, the results should match very closely. Mostly the GEDmatch browser is much more powerful and complex to use. ---------------------------------------------------------------------- From: M Robards <[email protected]> Subject: [AUTOSOMAL-DNA] Differing Info on Different Chromosome Browsers? I need some help reading the results on FTDNA's AND GedMatch's chromosome browsers for the same three people. I'm getting different readings from the two browsers. Does this mean I've done something horribly wrong in uploading their files from FTDNA? For instance, FTDNA says there is some matching segment(s) on Chromosome 16 between A and B and C. Yet GedMatch's browser shows no matching segments from B and C on A's. This is pretty confusing to me. melissa springer robards

I need some help reading the results on FTDNA's AND GedMatch's chromosome browsers for the same three people. I'm getting different readings from the two browsers. Does this mean I've done something horribly wrong in uploading their files from FTDNA? For instance, FTDNA says there is some matching segment(s) on Chromosome 16 between A and B and C. Yet GedMatch's browser shows no matching segments from B and C on A's. This is pretty confusing to me. melissa springer robards

It helps that I've been at this since 2004. It has taken all these years to find these three, and I had to pay for their testing to get them. Sometimes I feel like I'm trying to roll a dozen eggs down a football field with a toothpick. You make progress, first with this egg, then with that one, but it's sooooo slow! I will shortly have another group of four on my mother's side of the family (the ones I just did were on my father's side), but we're all first cousins (or 1st 1R), so no suspense here. But it will be interesting to see how many mutual matches we get. The most frustrating thing is how few people have uploaded GEDCOMs. Diana > From: Dwight Holmes > Sent: Sunday, February 12, 2012 10:31 AM > > Seems to me you're on the right track, Diana. > > I only have one such comparison, my 2nd cousin. > He & I only have 4 matches-in-common. A few > others here have much larger such family > databases. > > But I don't know how many people in the world > have "done this before", this exact thing you've > done. Very few of us are blessed with both the > energy, diligence, AND success rate that you've > got here in recruiting relatives to test. > Congratulations! (Sheesh, it took me 6 months > of obnoxiously persistent reminders and followups > with my own 1st cousins on my mother's side just > to get 1 of them to agree to do the test that I > sent to him...)

Seems to me you're on the right track, Diana. I only have one such comparison, my 2nd cousin. He & I only have 4 matches-in-common. A few others here have much larger such family databases. But I don't know how many people in the world have "done this before", this exact thing you've done. Very few of us are blessed with both the energy, diligence, AND success rate that you've got here in recruiting relatives to test. Congratulations! (Sheesh, it took me 6 months of obnoxiously persistent reminders and followups with my own 1st cousins on my mother's side just to get 1 of them to agree to do the test that I sent to him...) On Sun, Feb 12, 2012 at 8:47 AM, Diana Gale Matthiesen <[email protected]>wrote: > OK. I went to all four accounts, made lists of the names, then picked > out the matches that overlapped. It *seems* to me that the oddball is > related, but have I interpreted this correctly (scroll down the page > to see table)? > > http://dgmweb.net/DNA/aDNA/FF-Descendants_Isaac_DAVIS.html > > I've never done this before, so I don't really know what to expect. > > Diana > > > > From: Dwight Holmes > > Sent: Saturday, February 11, 2012 2:19 PM > > > > What I would do is make a list of the matches-in-common > > you have with the cousins whom you do match, then go > > into the oddball's complete list of matches and see if > > > > > > ______________________________ > For answers to Frequently Asked Questions about mailing lists, please see: > http://dgmweb.net/MailingListFAQs.html > > > ------------------------------- > 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 >

OK. I went to all four accounts, made lists of the names, then picked out the matches that overlapped. It *seems* to me that the oddball is related, but have I interpreted this correctly (scroll down the page to see table)? http://dgmweb.net/DNA/aDNA/FF-Descendants_Isaac_DAVIS.html I've never done this before, so I don't really know what to expect. Diana > From: Dwight Holmes > Sent: Saturday, February 11, 2012 2:19 PM > > What I would do is make a list of the matches-in-common > you have with the cousins whom you do match, then go > into the oddball's complete list of matches and see if

Maybe I'm not sure, either! ;-) What I would do is make a list of the matches-in-common you have with the cousins whom you do match, then go into the oddball's complete list of matches and see if any of them is there. On Sat, Feb 11, 2012 at 2:08 PM, Diana Gale Matthiesen <[email protected]>wrote: > I have access to all three accounts, in addition to my own, so I have > confirmed the relationship among the three of us. I can't confirm the > relationship between the oddball and the rest of us because he doesn't > show up as a match to any of us, at all, even distantly. > > I'm not certain I understand what you mean by "roundabout link." > > Diana > > > From: Dwight Holmes > > Sent: Saturday, February 11, 2012 1:40 PM > > > > Diana - another thing you can do is establish the FF "known > > relationship" link between all of you and your cousins > > wherever there is a FF match. Then you can easily see if > > there is a roundabout link between you and your cousin who > > doesn't match. It's perhaps unlikely, but worth a shot. > > > > > ______________________________ > For answers to Frequently Asked Questions about mailing lists, please see: > http://dgmweb.net/MailingListFAQs.html > > > ------------------------------- > 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 >

I have access to all three accounts, in addition to my own, so I have confirmed the relationship among the three of us. I can't confirm the relationship between the oddball and the rest of us because he doesn't show up as a match to any of us, at all, even distantly. I'm not certain I understand what you mean by "roundabout link." Diana > From: Dwight Holmes > Sent: Saturday, February 11, 2012 1:40 PM > > Diana - another thing you can do is establish the FF "known > relationship" link between all of you and your cousins > wherever there is a FF match. Then you can easily see if > there is a roundabout link between you and your cousin who > doesn't match. It's perhaps unlikely, but worth a shot.

> From: CeCe Moore > Sent: Saturday, February 11, 2012 3:58 AM > > Yes, Diana, you are right to wonder, but it can happen. Even with a > 3rd cousin 10 percent will not show a match. Since you are one > generation closer to the other cousins, it makes sense that you > match them, especially since the difference between them is on the > border of where autosomal inheritance falls off dramatically > (between 3rd and 4th cousin). The fact that no match shows at all > between any of the comparisons with this particular "cousin" is a > little suspect, but you are lucky that the others all do match > because a pair of full 4th cousins has only about a 50 percent > chance of matching. That's the rub. If the chance of a 4th cousin matching is 50%, then the chance that three would not match is getting close to being a probable non-match. One answer, of course, would be to test more cousins, and I've put another offer out, both for another THOMPSON and another DAVIS. > If you feel comfortable with Gedmatch, I would > upload everyone there and see if there are a number of matching > segments under FF's threshold, especially overlapping ones among you > all. I'm comfortable with GEDmatch, but I'm reluctant to ask three complete strangers to allow me to upload their results to a public database. I probably won't ask. > If not, I would start wondering about the possibility of a NPE > on their line that intersects with yours. That is what I'm wondering. I need to find a second DAVIS cousin to Y-DNA test and FF test. And I need to push the rest of my THOMPSON and DAVIS cousins to be tested - this time at *their* expense! Thanks for the input, Cece. Diana

I'll have to think about asking him for permission to do so. For a stranger, he's been very accommodating, but I don't want him to feel imposed upon because I'm asking too much of him. I wish FTDNA would give us these tools, so we don't have to go offsite for them. Diana > From: M. A. Farrell > Sent: Saturday, February 11, 2012 1:58 AM > > Put the Davis results into GedMatch, including his chromosome > browser upload and matches, and you might find matches in common > there, which includes 23Me people.. Can't hurt to look. > Mary Alice

Thank you for the links, Sam. I probably can't use methods that involve uploading their raw data somewhere else. Even though I paid for their testing, I think it would be over-stepping for me upload their data off site, and I'm uncomfortable even asking. These are not people I know well. All three had responded to my pleas on the web for cousins to test. So, Dr. Pike's utilities would appear to be the way to go. Thank you for bringing them to my attention. Diana > From: Sam Eaton > Sent: Saturday, February 11, 2012 10:18 AM > > Diana, > > Mary Alice is exactly correct. The match threshold for FTDNA or > 23andMe, while reasonable for screening out difficult matches to > find and real false positives, is not appropriate for identifying > matches between known, or strongly suspected relatives. > > There are alternatives like Dr Pike's utilities > > http://www.math.mun.ca/%7Edapike/FF23utils/ > > Dr Pike's utilities do require having a downloaded copy of the raw > data involved. But, the Raw Data is never uploaded. > > The beta phasing tool in this thread can also be used to examine the > Raw Data more closely > > https://www.23andme.com/you/community/thread/11466/#most_recent > > www.GEDmatch.com requires uploading the Raw Data > > Sam

Diana - another thing you can do is establish the FF "known relationship" link between all of you and your cousins wherever there is a FF match. Then you can easily see if there is a roundabout link between you and your cousin who doesn't match. It's perhaps unlikely, but worth a shot. On Sat, Feb 11, 2012 at 1:33 PM, Diana Gale Matthiesen <[email protected]>wrote: > > From: Mary Guler > > Sent: Saturday, February 11, 2012 12:24 PM > > > > We have tested 4 siblings, a first cousin once removed, two second > > cousins and one third cousin from the same line. The last person > > tested was the third cousin. All of us match one another, except > > for the third cousin. She and I do not match each other. > > (I am the one who paid for her test....LOL!) This is > > the perfect example of the randomness of autosomal DNA inheritance > > :o) > > > > Mary Mitchell Guler > > Does this last third cousin not match just you? Or does she not match > *any* of the other seven? Are all of the seven mutually matching all > of the other seven? If not, how many of the seven does each one > match? > > I paid for all four of these tests, so I can look at the matching for > each of the other three, individually, from their perspective as well > as mine. The three of us are mutually matching (each of the three of > us matches the other two, from each individual perspective). The > fourth matches none of the other three. > > I'm not trying to be argumentative. I truly want to understand how to > view my situation compared to what others have experienced. I > understand that, at this distance, random variation can mean a > non-match. I guess what I'm asking is at what point does the number > of non-matches take us beyond the tails of the probability > distribution and into the realm of "there ain't no underlying > relationship here." > > Diana > > > > > > ______________________________ > For answers to Frequently Asked Questions about mailing lists, please see: > http://dgmweb.net/MailingListFAQs.html > > > ------------------------------- > 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 >

> From: Mary Guler > Sent: Saturday, February 11, 2012 12:24 PM > > We have tested  4 siblings, a first cousin once removed, two second > cousins and one third cousin from the same line.  The last person > tested was the third cousin.  All of us match one another, except > for the third cousin.  She and I do not match each other.  > (I am the one who paid for her test....LOL!)  This is > the perfect example of the randomness of autosomal DNA inheritance > :o) > > Mary Mitchell Guler Does this last third cousin not match just you? Or does she not match *any* of the other seven? Are all of the seven mutually matching all of the other seven? If not, how many of the seven does each one match? I paid for all four of these tests, so I can look at the matching for each of the other three, individually, from their perspective as well as mine. The three of us are mutually matching (each of the three of us matches the other two, from each individual perspective). The fourth matches none of the other three. I'm not trying to be argumentative. I truly want to understand how to view my situation compared to what others have experienced. I understand that, at this distance, random variation can mean a non-match. I guess what I'm asking is at what point does the number of non-matches take us beyond the tails of the probability distribution and into the realm of "there ain't no underlying relationship here." Diana

Yes, but I think that's the point. You *do* have a match through a third person. We now have four descendants of Isaac tested. Three match each other at the predicted 2nd-4th cousin level, but the fourth doesn't match *any* of the other three. It seems to me the probability that he is related is dropping... But I will give the other matching methods a try, to see if there's a match at a lower level. Diana > From: Marleen Van Horne > Sent: Saturday, February 11, 2012 1:03 AM > > Diana, > > I have a known cousin, her ggg-grandmother and my ggg-grandfather > are > brother and sister. She and I do not have an atDNA match, BUT we > both > have a match to a man who is a descendant, we think, of the brother > of > our gggg-grandfather. Go figure. > > Either situation is possible with your non-match. > > Marleen Van Horne