Just a comment. I have been noticing that criteria used in autosomal DNA match searches mentioned in this long conversation have been at 5 or so cMs, which may very well thin out at 3 or 4 generations. I have also seen a useful result to 6 or 7 generations mentioned for 15 cM match searches; this was in a webinar by Lisa Louise Cooke. Don't know if this will help, but I am suggesting adjusting the criteria to see if the results extend further back. Donna B. -----Original Message----- From: lovelace-bounces@rootsweb.com [mailto:lovelace-bounces@rootsweb.com] On Behalf Of Brondak via Sent: Sunday, July 03, 2016 4:12 PM To: lovelace@rootsweb.com Subject: [LL] autosomal dna List and all you dna experts, I have to confess that I'm still a little bit at sea concerning the actual ancestral value of autosomal DNA. It seems that beyond the frist two or three generations, any matches are more random than definitive, However, that being said, I do have a question. In the cae of Abraham Lovelace whose line has daughtered out, therefore not available for Y-DNA, do you think it could help us put him with the Lovelace MD bunch (or noe) whichever the case may be? I personally think he belongs, but would love to find some proof more than trhe slim evidence we have. Lou Ann ------------------------------- To unsubscribe from the list, please send an email to LOVELACE-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message

Cuzzins, In answer to Donna about adjusting the parameters on autosomal matches... The centimorgan (cM) is, according to the ISOGG wiki (International Society of Genetic Genealogy), "... a unit of recombinant frequency which is used to measure genetic distance. It is often used to imply distance along a chromosome, and takes into account how often recombination occurs in a region." It is more or less interpreted as a measure of length of shared DNA segments. Again, from the ISOGG wiki, the major testing companies "use centiMorgans to denote the size of matching DNA segments in autosomal DNA tests. Segments which share a large number of centiMorgans in common are more likely to be of significance and to indicate a common ancestor within a genealogical timeframe." (http://isogg.org/wiki/Centimorgan) The way I interpret this is that as the "size" limit in cM is increased, the probability of a shared segment denoting a relationship is also increased. So "longer" shared segments between two people indicate a higher probability that the two share a common ancestor. ISOGG suggests adding together the cM for all shared segments of 5 cM or above to use in calculating the probable percentage of DNA shared with a match (http://isogg.org/wiki/Centimorgan#Converting_centiMorgans_into_percentages) . Gedmatch allows the user to specify the cM value to use when searching for matches, and they set their default at 7 cM. I tend to go with 5 cM per ISOGG. But others go even lower, sometimes down to 3 cM or even less. When you decrease the value of cM, you come up with more matching segments but they are much smaller, and the probability that they are random matches rather than matches due to relatedness is increased. Cuzzin Doug Lovelace and I are descendants of Ol' Barton the horse thief, Doug through Barton's son Asa and I through Barton's son Benjamin. Our relationship is calculated as 5th cousins 1 generation removed. ISOGG has a table of "Average autosomal DNA shared by pairs of relatives, in percentages and centiMorgans" here: http://isogg.org/wiki/Autosomal_DNA_statistics#Table According to that table of **averages**, Doug and I should share a total of 1.66 cM, or 0.0244% of our DNA. Using the Gedmatch default of 7 cM for the cutoff, Doug does not show as a match to me (7 cm >> 1.66 cM). The same thing for 5 cM and 3 cM. If I lower the threshold to 2 cM, we share a total of 4.9 cM (0.072%). If I lower it to the minimum, 1 cM, we share a total of 7 cM (0.103%). Using the ISOGG table of averages, we should be somewhere around fifth cousins (3.32 cM, 0.0488%) to fourth cousins once removed (6.64 cM, 0.0977%). So unless I really lower the matching threshold at Gedmatch, Doug and I don't match at all. Kim Philbeck Wood and I are both descendants of William Lovelace of Rutherford County. William is the grandson of Ol' Barton the horse thief. I descend from William's son George Logan Lovelace, and Kim descends from George's sister Eliza Ann (Lovelace) Philbeck. That makes us third cousins. According to the table at ISOGG, Kim and I should share **on average** a total of about 53 cM, or 0.781% of our DNA. If I use Gedmatch's default cutoff (7 cM), Kim and I share a total of 154 cM, about 3 times as much as expected on average! In the table, that would make us on the order of second cousins or second cousins once removed. What this demonstrates is that the predictions are just that, predictions, and that the actual relationships may vary around the **average** that is predicted. I just toss this out there as a demonstration of one way to use comparisons. Donna and I are a generation closer than Kim and I. Donna and I both descend from George Logan Lovelace, I through George's son Joseph Baxter, and Donna through his son James Dolphus. That makes us second cousins. Donna has submitted a sample to FTDNA, and I'm anxious to see how we match. I need to read up on triangulation. Lenny, do you have any good references to suggest? Peace, Part of the Tree, Greg -----Original Message----- From: lovelace-bounces@rootsweb.com [mailto:lovelace-bounces@rootsweb.com] On Behalf Of Donna Bockian via Sent: Monday, July 4, 2016 2:21 AM To: 'Brondak' <Brondak@aol.com>; lovelace@rootsweb.com Subject: Re: [LL] autosomal dna Just a comment. I have been noticing that criteria used in autosomal DNA match searches mentioned in this long conversation have been at 5 or so cMs, which may very well thin out at 3 or 4 generations. I have also seen a useful result to 6 or 7 generations mentioned for 15 cM match searches; this was in a webinar by Lisa Louise Cooke. Don't know if this will help, but I am suggesting adjusting the criteria to see if the results extend further back. Donna B.