Sorting out DNA results, apparent "Clare" chromosome segment

Genealogy, Archaeology, History, Heritage & Folklore

Moderators: Clare Support, Clare Past Mod

Post Reply
smcarberry
Posts: 1281
Joined: Fri Mar 30, 2007 4:31 pm
Location: USA

Sorting out DNA results, apparent "Clare" chromosome segment

Post by smcarberry » Sun Jan 15, 2017 6:06 pm

My knowledge of DNA in all its scientific ramifications has not significantly increased since my posting on results of my test kit with 23andMe and my Carberry male cousin's test kit with FamilyTreeNDA. However, I continue to view whatever shows up on my company's "DNA Relatives" listings. This is quite varied, as even those who participate in "Open Sharing" often don't list surnames or useful U.S. locations. "Open Sharing" means that I can see a graphic presentation of how my gene segments compare with that of a cousin identified by the company. I am informed that 23andMe uses the number of shared gene segments and the length of any such segment, to determine the degree of relationship. I continue to have no close relatives appearing on 23andMe, whether in Open Sharing or not. In the top 10 such identified relatives, there is still a good number remaining anonymous.

However, for the cousins who are Nos. 7 and 9 on the list in order of relationship to me, there is a shared gene segment which seems to point to our common Clare heritage. Cousin No. 7 (named Peter) has a posted genealogy, with ancestors Falvey and Fitzgerald of East Clare (church records located). Another "Open Sharing" cousin, Patricia, (not in my top 10) with the same shared gene segment likewise has researched her family back to Clare; I have to contact her for more details but she has listed her U.S. location as the same county where my family settled. Cousin No. 9, another Peter, has not listed any surnames or U.S. locations although he states his maternal side is Irish. None of the paternal haplogroups match that of my Carberry cousin, but our Carberry line is not Clare in origin. It would be our old ancestor Carberry's wife who was born in Clare, and her surname appears to be Maloney from an 1816 baptism.

I am showing below edited versions of these cousins' Open Sharing gene graphics, so you can see that their gene segments, all on Chromosome No. 9 (as labelled by 23andMe) look remarkably alike. Due to that company' reliance on gene length in its assessments, I suppose that the segment length is carefully depicted, and I am hoping its position is also dependably shown. If so, then could I assume this particular location on this chromosome is a "marker"-type distinction for a Clare descendant ? Any opinion from a DNA savvy person would be appreciated.

On this working assumption, I will be contacting an adopted person identified as my 4th cousin (the same as Peter, Peter, and Patricia) whose 23andMe graphic likewise shows the same chromosome 9 segment position and length. She states she wants to learn about her biological relatives but has shown no clues to U.S. locations, which adoptees usually have in one form or another. Her current surname appears to be Polish.

This small set of conclusions has sifted out of hours of viewing profiles of those identified by 23andMe as my cousins. 23andMe provides no way to sort the mass of 1400-plus such cousins, except for surnames that some have listed in their profiles. I fortunately have a maternal side that is very distinct gene-wise from my paternal side. Thus in looking for other Clare descendants, I can fairly readily weed out the maternal-side cousins by their maternal haplogroups and U.S. locations.

Actually, I am surprised that I have arrived at this stage. The down side is identified cousins, even the ones who are active in family history, are not interested in going beyond what they already have. I would like to get at least the surname of a common ancestor. That seems possible, as the time period involved is 1800 to 1835, for which documents do exist if someone had property or a family tree already exists on file from the old Genealogical Office in Dublin.

This is just a memo for your general information. I am o.k. whether or not I get more results than currently. I feel bad for the adoptees who really need responses.

Sharon Carberry
USA
Peter & Peter with chromo 9 segment.jpg
Peter & Peter with chromo 9 segment.jpg (26.63 KiB) Viewed 5142 times
Clare desc. Peter & Patricia's chromo 9 segmnt position.jpg
Clare desc. Peter & Patricia's chromo 9 segmnt position.jpg (12.31 KiB) Viewed 5142 times
Adopted 4th cuz's chromo 9 segmnt.jpg
Adopted 4th cuz's chromo 9 segmnt.jpg (17.38 KiB) Viewed 5142 times

smcarberry
Posts: 1281
Joined: Fri Mar 30, 2007 4:31 pm
Location: USA

Re: Sorting out DNA results, apparent "Clare" chromosome seg

Post by smcarberry » Mon Feb 06, 2017 8:13 pm

New features at 23andMe are allowing more analysis of the “DNA Relatives” pool of subscribers derived from autosomal results of my test kit (covering both maternal and paternal lines). Up till now, each subscriber received a long list of people sharing any portion of any gene with you, without a way to know the exact relationships involved in the group as a whole. However, now it is possible to see how much each subscriber shares with you and your other pooled relatives. Still not a perfect system, as some subscribers don't show up in the comparison (perhaps blocked somehow), and a lot of people who do show up nonetheless don't provide ancestor locations and surnames. All eight of my great-grandparents came from different countries, so there is a lot of sifting to do to figure out which relatives in the pool are my Irish ones. The new feature allows me to confirm some of my prior estimations,

It still takes a lot of time, but now I can choose other subscribers to put into a comparison group, which speeds up obtaining the shared gene segment. Even better, 23andMe is now providing the exact segment measurements, plus a segment position is presented in numerical terms, as well as the exact number of SNPs involved (a unique identifier used by all the test kit companies) and also Genetic Distance expressed in cM units. Luckily, some of my many “anonymous” pooled relatives have segment positions showing up on the new feature although nothing has been showing on their individual profile pages. By shifting from screen to screen of pooled relatives, I now have found sets of very closely-related people, although I still lack many subscriber names. Previously I could put together a pair or group only if their surname, segment, and stated history all matched up. Now I have siblings, parents, and even twins. I also now believe that my closest matched relative at 23andMe, an “anonymous” who will not reply to me, is not a second cousin but rather a child of a first cousin whose whereabouts I determined a while ago (he also would not respond to me). That subscriber is referred to here still as a second cousin, to match the label placed on her by 23andMe.

In case you also have many “anonymous” relatives on a company website, this seems to be happening: one person is a leader in getting the test kits done, typically someone in a younger generation who can undertake the learning curve in handling results. That person provides a nice full profile page, and, at 23andMe, is allowed to manage the profiles of his/her relatives who have been persuaded to participate but have no desire to be active - the “anonymous” subscribers. I have several of those, at the 4th cousin level. When I know there are older folks in a group of subscribers, I will contact just the active one with a full profile. Now seeing their interrelationships within the group is quite valuable and lets me do better organization, using my time to get on to a new, different line to sift out of the huge relative pool.

The following illustrates my methods:

1. I compared the two Peters who seemed to share the same segment on Chromosome 9, known to be connected to Co. Clare, based on one Peter's documented family history. The new feature negated their direct relationship, so now they are in different lines. All I can learn from Peter M. is that his surnames include Clancy and Griffin; no locations provided. Nonetheless, the new feature has shown that Peter M. is related to my anonymous “second” cousin and myself, which means there is a Carberry or Clare connection.

2. Significantly, the new analysis confirms a prior suspicion about an Irish-derived segment on Chromosome 5. I started with my anonymous “second” cousin's own pool of DNA Relatives, a portion of which appears below (minus identifying names). In the 9 people shown, there are 5 who are exactly as close to that cousin as I am – very curious. Three were not worth following (one is a daughter of Peter with a known Clare tie; one is an adoptee; one is anonymous). The remaining 2 were very revealing. No. 2 on the list is Carla, who shares with me 2 segments on Chromo 5. No. 8 on the list is “KK” who turned out to be a first cousin of Peter M. (see KK's list below, where Peter is No. 2 on the list). KK's mother is listed as No. 4, one of those several anonymous 4th cousins of mine. Best of all, KK provided locations for all her ancestors: all in Ireland. I am working on learning more about prior exact locations. There are also 2 remotely-related cousins who have Conlon in their histories, plus one has Burke and Lynch.

3. No. 2 on my anonymous “second” cousin's pool list is Carla, who already provided enough background that I think I located her parents in the 1940 census, although the surname is not one in my known history (the location is right). Our long shared segment also links up to a subscriber named “Jim.” Our exact segment analysis is attached, with names deleted. You can see that Carla's long 5 segment starts exactly where Jim's does – very significant. I also show the color-coded version. Carla, Jim, and I all are 4th cousins to a subscriber surnamed Dyer, all in different degrees; unfortunately Dyer currently is a dead end for lack of information. I show Jim's DNA pool, with names deleted: my “second” cousin is No.1, Carla is No. 2, and Dyer is No. 4. Jim states in his profile that his surnames include O'Connor, a possibility in one of my Clare lines.

Sharon Carberry
Anon 2nd cuz's rellie pool, p.1.jpg
Anon 2nd cuz's rellie pool, p.1.jpg (21.95 KiB) Viewed 5023 times
Carla, Jim's exact sgmnt data.jpg
Carla, Jim's exact sgmnt data.jpg (11.46 KiB) Viewed 5023 times
Color version of Carla, Jim's segments.jpg
Color version of Carla, Jim's segments.jpg (21.47 KiB) Viewed 5023 times
Jim's ties (second column) compared to mine.jpg
Jim's ties (second column) compared to mine.jpg (8.37 KiB) Viewed 5023 times
KK's rellie pool.jpg
KK's rellie pool.jpg (15.79 KiB) Viewed 5023 times

Post Reply