Author Topic: DNA conundrum, can anyone help me make sense of this please?  (Read 998 times)

Offline Romilly

  • RootsChat Marquessate
  • *******
  • Posts: 5,431
    • View Profile
DNA conundrum, can anyone help me make sense of this please?
« on: Tuesday 29 March 22 10:29 BST (UK) »
I’m going round in circles trying to make sense of this, and would welcome some imput with it.

Because I’ve never been able to find my Paternal Grandfather anywhere before he married my Grandmother in 1893, I turned to DNA testing.

Thanks to my late Mother having tested with Ancestry, I can sort my matches into Paternal/Maternal sides in the main.

My most significant Paternal match on Ancestry is to someone at 133cMs across 4 segments. Ancestry says he’s my 2nd Cousin, once removed.

On 23andme, I have a DNA match to someone whose Grandfather is the 1st Cousin of this man. My match to her is 97cMs across 3 segments.

And so I think that the relationship between these 2 matches must be 1st Cousins, twice removed?

My problem is, how do I match to them?

23andme say that my match on that Site and I share a set of Great Great Grandparents. However, my Grandfather was born in 1860, and so I’m wondering if that skews the generations somewhat?

I’m feeling totally stuck with this  :o
Any help welcomed…

Romilly.
Any census information included in this post is Crown Copyright, from www.nationalarchives.gov.uk
Researching:
Wilson, Warren, Dulston, Hooper, Duffin, Petty, Rees, Davies, Williams, Newman, Dyer, Hamilton, Edmeads, Pattenden.

Offline brigidmac

  • RootsChat Marquessate
  • *******
  • Posts: 6,014
  • Computer incompetent but stiil trying
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #1 on: Tuesday 29 March 22 14:11 BST (UK) »
Romilly
Ancestry does not usually give you specific relationship based purely on DNA it will give you a range of relationships then a thru line may suggest that you relate to another person at second cousin once removed level based on their and your tree which may or may not be correct

Has your mum got any shared matches to this shared match

Have you used the DNA painter
I'm in the process of helping mrX work out patternity and am working thru the DNA amounts you may find that post useful

https://www.rootschat.com/forum/index.php?topic=860581.0

# tt may help to put your amounts into DNA painter and compare results as I've done
Does your match know that her first cousin is definitely full first cousin
Roberts,Fellman.Macdermid smith jones,Bloch,Irvine,Hallis Stevenson

Offline phil57

  • RootsChat Veteran
  • *****
  • Posts: 644
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #2 on: Tuesday 29 March 22 14:16 BST (UK) »
Ancestry cannot tell you what your relationship to DNA matches is. It can infer a best guess based on the most likely probability if it knows the length of the match between you. For instance, DNA Painter indicates that for a 133 cM match, there is a 50% probability that the relationship could be half 2C, 2C1R, half 1C2R or 1C3R. There are lower probabilities that the relationship could be different, but within the ranges given, they could all be valid for you and your match without supporting evidence.

For instance, I have a match at 122 cM, which is in the same ball park as your 133 cM match. He is actually my second cousin and that relationship is confirmed. But DNA Painter indicates that only about 12% of 122 cM matches will be between 2nd cousins. Nevertheless it is a valid match confirmed by other research, but Ancestry would suggest that we were most likely 2C1R or other relationships based on the highest probability.

Similarly, if Ancestry can match you and your 133 cM to people that can be found in family trees within its database (Thrulines) it can infer a relationship between you that may be a better guess. BUT it is based on the users tree information, so Ancestry still cannot say that it is accurate. It depends how correct (or otherwise) the trees it is using to infer the relationship happen to be - and there are a lot of incorrect trees on Ancestry and elsewhere.

So you really need to start with whatever information you have or can find on your match, and research their and your family trees until you discover a potential link via traditional research. If the suggested relationship is via Ancestry Thrulines, you have a head start if you can view the other users' trees, but with the proviso that you need to carry out your own research on the links that are being suggested, to confirm or disprove them to your satisfaction.

If or when you discover a potential link between you and your match, a tool such as DNA Painter can help you judge whether the suggested link is possible, and to what degree of probability.

In other words, DNA by itself cannot tell you anything other than that you and your match share a certain length of DNA. Tools such as DNA Painter can infer the probabilities of various relationships between two people sharing a given length of DNA, based on relationships at that level which have already been confirmed, but you really need to carry out the traditional research to try and discover your link and confirm whether it is possible, or probable.
Stokes - London and Essex
Hodges - Somerset
Murden - Notts
Humphries/Humphreys from Montgomeryshire

Offline Romilly

  • RootsChat Marquessate
  • *******
  • Posts: 5,431
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #3 on: Tuesday 29 March 22 14:26 BST (UK) »

Hi Both,

Many thanks for your replies.

I know that this DNA match is on my Paternal side, not maternal.

This is what Ancestry have put as the Match between myself and the match I'm looking at:

'You and ***** ****
2nd cousin 1x removed | Father's side'.

This isn't taken from ThruLines, it's from my list of DNA Matches.

I've tried the WATO tool and had a go at DNA Painter, but got nowhere.

This DNA Match has an extensive Public Tree on Ancestry, but my Grandfather would only fit if he changed his name, for some unknown reason...

I've been asking about him on here since I joined RC in 2005, but he continues to elude me  :o

Romilly.



Any census information included in this post is Crown Copyright, from www.nationalarchives.gov.uk
Researching:
Wilson, Warren, Dulston, Hooper, Duffin, Petty, Rees, Davies, Williams, Newman, Dyer, Hamilton, Edmeads, Pattenden.


Offline phil57

  • RootsChat Veteran
  • *****
  • Posts: 644
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #4 on: Tuesday 29 March 22 14:39 BST (UK) »
This is what Ancestry have put as the Match between myself and the match I'm looking at:

'You and ***** ****
2nd cousin 1x removed | Father's side'.

This isn't taken from ThruLines, it's from my list of DNA Matches.

I'm pretty sure that Ancestry is only ever that definitive about a relationship if you have already updated the match to confirm how you are related, excepting suggestions via Thrulines.

Normally it will indicate something like "2nd - 3rd Cousin" and if you click on the link it will open a box indicating the range of possible relationships that could match, and their probabilities.

But there is no way to confirm an exact relationship by DNA alone. It can confirm a genetic link between two people (subject to the length of any given match, as matches at lower lengths may have probabilities of not actually being genetic matches at all) but for proving an exact relationship it is only a corroborative tool which needs to be assessed against other evidence to form a likely or definitive conclusion.
Stokes - London and Essex
Hodges - Somerset
Murden - Notts
Humphries/Humphreys from Montgomeryshire

Offline Romilly

  • RootsChat Marquessate
  • *******
  • Posts: 5,431
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #5 on: Tuesday 29 March 22 15:01 BST (UK) »

Thanks for your thoughts phil57.

Yes, I agree with much of what you are saying, but I've tried everything that I can think of to work out this match, and just wanted another opinion.

Myself, my sister, daughter and two paternal 1st Cousins all match to this man, anf his nephew and grandson.

I've linked my DNA Test to my Tree on Ancestry and if I put my Grandfather on it under the name he used from 1893 to his death in 1937, no DNA matches turn up on the paternal side, (apart from my 1st Cousins). However, if I add my Grandfather as 'Richard Kean' lots of DNA matches to this family turn up...

I've spent years trying to establish if my Grandfather William James Wilson, born Manchester 1860 and Richard Kean, born Manchester 1860 were one and the same man, but it's impossible to prove definitively. It is curious though that all of my Paternal DNA matches come back to the same Kean family.

Romilly.
Any census information included in this post is Crown Copyright, from www.nationalarchives.gov.uk
Researching:
Wilson, Warren, Dulston, Hooper, Duffin, Petty, Rees, Davies, Williams, Newman, Dyer, Hamilton, Edmeads, Pattenden.

Offline brigidmac

  • RootsChat Marquessate
  • *******
  • Posts: 6,014
  • Computer incompetent but stiil trying
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #6 on: Tuesday 29 March 22 15:59 BST (UK) »
The relationship is probably something you put in at an earlier time I'm in process of changing some of my relationships from second cousin to half second cousin
(because found out about an earlier marriage of common ancestor)

 .the  half second cousins have lower match amounts
To change relationships back to unknown s
!) you go in your match comparison
edit relationship / keep father's side ./press next  remove the relationship that is highlighted in blue? put "I'm not sure "or change to whatever your new now known relationship is
(....don't forget to confirm ...or it goes back as it was ...I've done that at least 6 times today .)

Punctuation and words added to make clearer

I mentioned about DNA biological generations not always corresponding with age
In my topic. Link added in reply one

Yes things do get squiffed ..paper trails say one thing but if a mother took on her daughter's child as her own or an aunt informally adopts the lines do have results which seem to be incorrect or "impossible"
Like you I'm lucky to have tests from older generation but the more family who test the better

Roberts,Fellman.Macdermid smith jones,Bloch,Irvine,Hallis Stevenson

Offline brigidmac

  • RootsChat Marquessate
  • *******
  • Posts: 6,014
  • Computer incompetent but stiil trying
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #7 on: Tuesday 29 March 22 16:03 BST (UK) »
Can someone help you set up a chart ?

Is it not possible that Mr Kean was your true grandfather or great uncle but had a child with the same lady as Mr Williams or they have 1 parent in common ?

There are all sorts of combinations ; what happens if you put them as brothers


Roberts,Fellman.Macdermid smith jones,Bloch,Irvine,Hallis Stevenson

Offline phil57

  • RootsChat Veteran
  • *****
  • Posts: 644
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA conundrum, can anyone help me make sense of this please?
« Reply #8 on: Tuesday 29 March 22 16:22 BST (UK) »
It can be frustrating at times. I have a match on Ancestry at almost twice the length of the one I mentioned to my 2nd cousin. He has not responded to my messages, so I have to assume he is completely disinterested, which is very frustrating for me.

He shares matches with several other matches on my father's side, although none of them are in my direct line of descent, but is not a match on Ancestry at least, to my 2nd cousin.

I have spent many months and quite a few pounds researching his tree, and have only found one possible way in which we could be related, which involves his grandmother being illegitimately fathered by my GG uncle. His great grandmother was a widow at the time of his grandmother's birth. That is circumstantial, but my GG uncle fathered two illegitimate children with my match's GG aunt in the same period of time, and his occupation was the same as that given for the fictitious father named on my match's grandmother's birth certificate. They all lived in the same town.

If correct, that would make him my 3rd cousin. It is possible - DNA Painter gives a 3% probability of that relationship at the length of our match. It sounds good, but it also means a 97% probability that the relationship could be closer if the numbers are turned around.

I live in hope that another close match to both of us will pop up in Ancestry sooner or later, but for now I am just left with a niggling doubt, unless or until I can find additional corroboration.
Stokes - London and Essex
Hodges - Somerset
Murden - Notts
Humphries/Humphreys from Montgomeryshire