RootsChat.Com
General => Ancestral Family Tree DNA Testing => Topic started by: goldfinch99 on Sunday 15 December 24 10:22 GMT (UK)
-
I've noticed about three different lines in my matches that give abnormally high cM matches (up to about 25cM) in large numbers, dozens of matches on these lines, but the common ancestor is impossible to work out yet must be at least 400 to 500 years ago before the lines split into UK/N. American sections.
I am curious how likely it is that these are real lines vs something else? This Cannaday lineage, some light checking of trees in Ancestry suggests, descend from Kennedy noble lines in Scotland that had several generations of very close cousin marriages. I think this could be a valid reason for the DNA signals to remain strong for so long, and combine that with the thousands of descendants on these lines, it could mean that if just 1% of the descendants carry enough DNA and match to each other it could be real?
Has anyone else got this line in their DNA matches and researched it? Is it something that's known to genealogists?
Thank you if anyone can shine any light on this.
-
Hi,
Yes you are probably seeing pedigree collapse. It comes about when already related families, meet and have children.
This can be either because the families come from a relatively small physical area, or for religious or similar reasons. Quakers, Baptists and Jews are examples.
-
I've noticed about three different lines in my matches that give abnormally high cM matches (up to about 25cM) in large numbers, dozens of matches on these lines, but the common ancestor is impossible to work out yet must be at least 400 to 500 years ago before the lines split into UK/N. American sections.
I am curious how likely it is that these are real lines vs something else? This Cannaday lineage, some light checking of trees in Ancestry suggests, descend from Kennedy noble lines in Scotland that had several generations of very close cousin marriages. I think this could be a valid reason for the DNA signals to remain strong for so long, and combine that with the thousands of descendants on these lines, it could mean that if just 1% of the descendants carry enough DNA and match to each other it could be real?
Has anyone else got this line in their DNA matches and researched it? Is it something that's known to genealogists?
Thank you if anyone can shine any light on this.
You see this quite a lot, I call it sticky segments. In my case, I can see all the people have origins in a certain area, but working out what the common ancestor is may be very difficult if not impossible. ProTools can help as you can use to work out sub groups of closely related people. Then you have less overall lines to try and join up. In the case of American relatives, you may have had a relative who went to another rural area of the US, or became part of a relatively small religious group as David suggests, and their descendants consequently intermarried keeping that segment undivided.
-
Interesting, as the Cannaday/Kennedy line is supposed to descend from one of the Scottish kings!
Yes, the US lines do have continuing intermarriage with 2nd or third cousins and the same names appear in multiple places in peoples trees about 6 or 7 generations ago so that I can quickly see key names and know they're in this Cannaday grouping.
I can clearly see this without needing pro-tools. I would like to try pro-tools at some point though.
But I am in England, and I can't see how that line has combined with mine as it's happened so far back that I can't trace my lines that far back.
-
Interesting, as the Cannaday/Kennedy line is supposed to descend from one of the Scottish kings!
Yes, the US lines do have continuing intermarriage with 2nd or third cousins and the same names appear in multiple places in peoples trees about 6 or 7 generations ago so that I can quickly see key names and know they're in this Cannaday grouping.
I can clearly see this without needing pro-tools. I would like to try pro-tools at some point though.
But I am in England, and I can't see how that line has combined with mine as it's happened so far back that I can't trace my lines that far back.
Connecting US lines with any certainty is very difficult. Even if we are talking about closer relatives where British born people were still alive at the censuses, US censuses only give country of birth, not place. But sometimes it might be clearer because a surname is more unusual and you can see a clear pattern that matches with some other non US cousins on a line. Also sometimes, death certificates of some places, like NYC gave people's birth parents and at least countries of birth. Also if the line has been thoroughly researched on the American side tracing the origins in Britain or Ireland through means like wills, land deeds, local government records etc. (church records, except in large towns and cities rarely survive before the early 1800s) and published in various books you might also have a chance to work out the connection.
-
A couple of matches I found. One was someone who shared my 5 times great grandparents. When we compared our lines there were three occasions where brothers and sisters from a line married into our lines making more matches..
The other one really annoyed me. I found a match of 300 plus cM on an unknown line. Whacko I thought a clue. No! On one of her parents side my great grandparents were her grandparents and on the other her great grandparents. Grrrr
-
Sorry, I misread the previous post.
-
Interesting, as the Cannaday/Kennedy line is supposed to descend from one of the Scottish kings!
Yes, the US lines do have continuing intermarriage with 2nd or third cousins and the same names appear in multiple places in peoples trees about 6 or 7 generations ago so that I can quickly see key names and know they're in this Cannaday grouping.
I can clearly see this without needing pro-tools. I would like to try pro-tools at some point though.
But I am in England, and I can't see how that line has combined with mine as it's happened so far back that I can't trace my lines that far back.
Connecting US lines with any certainty is very difficult. Even if we are talking about closer relatives where British born people were still alive at the censuses, US censuses only give country of birth, not place. But sometimes it might be clearer because a surname is more unusual and you can see a clear pattern that matches with some other non US cousins on a line. Also sometimes, death certificates of some places, like NYC gave people's birth parents and at least countries of birth. Also if the line has been thoroughly researched on the American side tracing the origins in Britain or Ireland through means like wills, land deeds, local government records etc. (church records, except in large towns and cities rarely survive before the early 1800s) and published in various books you might also have a chance to work out the connection.
I think that's a bit too much for me to take on, but maybe in the next ten years as more people delve into DNA and records they will end up finding the info and someone will tie the connections together. This one is interesting as it comes from nobility and potentially royalty, so there should be a lot of researchers. I do have a lot of brick walls on that side of the tree that I think are currently impossible to break through, but that might change as more people develop their trees and do DNA tests.
-
Hi,
This article perhaps gives a better description than I managed. There are a lot of other articles out there on the web.
https://dna-explained.com/2021/07/23/whats-the-difference-between-pedigree-collapse-and-endogamy/
You may be pleasantly surprised how matches like this can break down brick walls, but to do this you need to make your tree as wide as possible. Cousins, great uncles and aunts, and their children. This will give Ancestry’s algorithms something to start generating matches against. It does take patience. You may also need to build trees for some of the matches. Pick some of the higher match ones who have a tree that goes back to 1900 or so. Not everyone has a full tree on Ancestry, but I have managed to push some of them back to where I can see the link.
Some of these go back to the 1700’s.
That being said I also have a good few matches both in the UK and USA who go back before records begin, but I can see either common names or areas.
Frustratingly Ancestry for some reason limit Thru lines to 5 great grandparents although I have proved some matches are actually 1 or 2 generations further back.
One last thing, it was by building a tree for a match that I broke one of my own brick walls.
This lead me to find research that other people had done on their line, that when combined with other documents (Wills, and Autobiography! And others) that allowed me to prove the link beyond doubt.
Happy Hunting
-
Thanks David. That also shows how the ProTools could help by using the chromosome browser on all these matches.
Colouring in a chromosome browser is on my 'to do' list.
-
Hi,
I don’t think that Pro Tools has a chromosome browser at present.
But you can do quite a lot of useful analysis by grouping DNA matches using the labels in Ancestry.
Using something similar to the Leeds Method or some other similar clustering method.
-
Oh, that's disappointing, hopefully A will bring out a chromosome browser soon. Still worth having a look at the tools, maybe a little Christmas present to myself.
I'm not familiar with the Leeds method or clustering. I will have to look them up.
-
I genuinely wouldn't worry about the Leeds method, it can be a lot of work trying to filter matches to four groups with each connecting to just one specific grandparent and is best tackled with matches between 90 and 400cM. Scanning through the posts it seems the matches fall below that amount in many cases. Additionally if there's a chance of pedigree collapse then the odds are people will be related to each other in multiple ways and Leeds Method grouping really doesn't work well when multiple relationship complications are added into the mix.
Ancestry have repeatedly cited privacy issues as the reason for not offering a chromo browser though it obviously hasn't stopped them introducing protools which I'd argue is potentially a bigger privacy issue.
-
Thanks Glen. Yes, these are going to be very, very distant confluence points, and the US lines are clear. On the UK/Ireland side, the only names I've seen are my dad's close relatives on my nan's side.
I think the higher US matches are about 20-25cM, with individuals on the Cannaday line taking up about 4 positions in about the 8th generation past. It's obvious looking at the US trees that go far enough back, or where a descendant on one of those lines has a name that stays in my memory as I've seen it so often that I know it comes from that lineage.
It's a thing of interest in my DNA. I saw either an article or YouTube video headlined something like "Think you're descended from Scottish kings? Think again" and that made me doubt that this could be real. So I just wanted knowledgeable opinions on how likely this was to be real.
It's cool, I think. It's like they say about so many people with Euro or Brit ancestry descend from certain key figures in history, but you're not going to find a DNA segment in your ancestry results connecting you to those figures. Except that it's possible that here's a rare example where it does show in the results for a small fraction of the descendants who have have trees with enough pedigree collapse or endogamy. wow!