Author Topic: DNA ethnicity logic check / Unexpected result  (Read 6279 times)

Offline AndyH81

  • RootsChat Extra
  • **
  • Posts: 86
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #90 on: Monday 16 May 22 23:28 BST (UK) »
Just some food for thought. My half first cousins match me at 572 cM and 483 cM.  Whilst at the low end it is possible that the mother could be a half sibling.

I'd have to double-check, but I think her father died before I would have been conceived.

Also, I thought that half-siblings generally shared 25% or more DNA with another half-sibling (she and I "only" share 16%) and could be fairly accurately identified as half-siblings without too much guess work.

Offline Biggles50

  • RootsChat Extra
  • **
  • Posts: 86
  • Tongue in cheek avatar.
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #91 on: Tuesday 17 May 22 20:43 BST (UK) »
Also depends upon being tested with the same Company.

My Ancestry DNA matches have quite different cM figures on the comparison websites.

DNA Painter gives your 572cM as a 1/2 First Cousin at a 83% Probability and the 483cM the same at 89%

If you have not used their WATO tool, give it a go and see the vatios hypothesis’s that it presents

Online Nanna52

  • RootsChat Veteran
  • *****
  • Posts: 852
  • Edwin WB Vincent, my actor, (1881-1940)
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #92 on: Tuesday 17 May 22 23:35 BST (UK) »
Biggles the three of us tested with ancestry and all knew that there was a different grandfather.  Yes, dna painter is a great tool.
James -Victoria, Australia originally from Keynsham, Somerset.
Janes - Keynsham and Bristol area.
Heale/Hale - Keynsham, Somerset
Vincent - Illogan/Redruth, Cornwall.  Moved to Sculcoates, Yorkshire; Grass Valley, California; Timaru, New Zealand and Victoria, Australia.
Williams somewhere in Wales - he kept moving
Ellis - Anglesey

Gedmatch A327531

Offline AndyH81

  • RootsChat Extra
  • **
  • Posts: 86
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #93 on: Wednesday 18 May 22 00:07 BST (UK) »
Also depends upon being tested with the same Company.

My Ancestry DNA matches have quite different cM figures on the comparison websites.

DNA Painter gives your 572cM as a 1/2 First Cousin at a 83% Probability and the 483cM the same at 89%

If you have not used their WATO tool, give it a go and see the vatios hypothesis’s that it presents

I think the percentages the two of us are referring to might be two different things? Yours seem to be saying the percentage of probability of X relationship. The percentages I am talking about is the total percentage of DNA that we share.


Offline phil57

  • RootsChat Member
  • ***
  • Posts: 240
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #94 on: Wednesday 18 May 22 09:31 BST (UK) »


I think the percentages the two of us are referring to might be two different things? Yours seem to be saying the percentage of probability of X relationship. The percentages I am talking about is the total percentage of DNA that we share.

The only certainty in percentages of DNA shared is between parents and their children. All other relationships are variable, to the extents indicated by the DNA Painter tool and similar charts. You also have to beware of outliers - individuals who might share a substantially higher or lower percentage of DNA than the normal range for any given relationship.

DNA Painter expressed the probability of particular relationships for the percentage or amount of DNA shared between individuals. It is a handy tool, but it is important to recognise that just because a certain relationship has the highest probability, it doesn't follow that it is necessarily the correct relationship for your circumstances.

You have to do the traditional research, look at the probabilities of relationships between individuals you have researched being related in various ways, and compare those relationships with the probabilities suggested by tools such as DNA Painter. Nothing is certain, and trying to establish a specific relationship purely by the amount of DNA shared is fraught with difficulty.

I can give a couple of personal examples. My brother has a relatively low share of 59 cM with a 2nd cousin.

We have three 2C1Rs, all siblings born of the same parents. My shared matches with them range from 23 cM to 147 cM.
Stokes - London and Essex
Hodges - Somerset
Murden - Notts
Humphries/Humphreys from Montgomeryshire

Offline AndyH81

  • RootsChat Extra
  • **
  • Posts: 86
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #95 on: Wednesday 18 May 22 10:25 BST (UK) »
Ok. I thought based on everything I'd read that in order for someone to be a half-sibling you had to share at least 25% of your DNA or more to have that relationship and not that it could be lower.

Either way, that's not going to be the case as it's the women in the family I am related to, so the only way I could be the mother's half-brother is if her parents were too related, which to the best of my knowledge, they're not! Nobody was born with an extra head that I know of anyway.

Offline TonyV

  • RootsChat Senior
  • ****
  • Posts: 374
  • Expect the unexpected (I forgot to!)
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #96 on: Wednesday 18 May 22 11:02 BST (UK) »
Hi Andy

25% is a statistical average rather than an upper or lower limit. Because DNA gets shuffled each time a conception takes place it is possible for the figure to be higher or lower. The best example of the shuffle effect is that siblings get 50% of each parent's DNA but it is a different 50% each time, so they do not share the same DNA as other siblings. Nevertheless statistically the shared DNA is so high that it is highly probable that two people sharing somewhere around 2,600 cM are siblings.

As you get further from the parents the shuffle effect makes the average figures much less certain and it is possible for some more distant genetic ancestors not to pass on their genes to you at all while at the same time they might do to a 1st cousin of yours.

Offline Rosinish

  • RootsChat Marquessate
  • *******
  • Posts: 13,421
  • PASSED & PAST
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #97 on: Thursday 19 May 22 04:51 BST (UK) »
My half first cousins match me at 572 cM and 483 cM.  Whilst at the low end it is possible that the mother could be a half sibling.

Sorry to jump onboard but I've today found a new (unknown) match with...

1st – 2nd Cousin
8% shared DNA
578 cM across 21 segments

I've been in touch & will wait for a reply as to connections prior to jumping to conclusions  ;)

I can't recall if AndyH81 actually listed the % shared DNA & segments with his matches?

Annie
South Uist, Inverness-shire, Scotland:- Bowie, Campbell, Cumming, Currie

Ireland:- Cullen, Flannigan (Derry), Donahoe/Donaghue (variants) (Cork), McCrate (Tipperary), Mellon, Tol(l)and (Donegal & Tyrone)

Newcastle-on-Tyne/Durham (Northumberland):- Harrison, Jude, Kemp, Lunn, Mellon, Robson, Stirling

Kettering, Northampton:- MacKinnon

Canada:- Callaghan, Cumming, MacPhee

"OLD GENEALOGISTS NEVER DIE - THEY JUST LOSE THEIR CENSUS"

Offline AndyH81

  • RootsChat Extra
  • **
  • Posts: 86
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: DNA ethnicity logic check / Unexpected result
« Reply #98 on: Thursday 19 May 22 09:57 BST (UK) »
My half first cousins match me at 572 cM and 483 cM.  Whilst at the low end it is possible that the mother could be a half sibling.

Sorry to jump onboard but I've today found a new (unknown) match with...

1st – 2nd Cousin
8% shared DNA
578 cM across 21 segments

I've been in touch & will wait for a reply as to connections prior to jumping to conclusions  ;)

I can't recall if AndyH81 actually listed the % shared DNA & segments with his matches?

Annie

BC (the daughter) and I share: 7.7% of our DNA, 543.2 cM, 25 segments with the largest being 71.1 cM.

PW (the mother) and I share: 16% of our DNA, 1,130.7 cM, 37 segments with the largest being 88.1 cM.