Author Topic: Strange match  (Read 1316 times)

Offline Raybistre

  • RootsChat Senior
  • ****
  • Posts: 422
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Strange match
« on: Sunday 20 July 25 09:29 BST (UK) »
Hi All, Did my DNA on Ancestry and have since uploaded to MyHeritage. On MyHeritage I have a match, lets call it XF, as follows: Shared DNA: 1.5% (102.9‎ cM); Shared segments: 6; Largest segment: 35‎ cM. Now I have been successful in linking all my matches at this level to myself except for this one. On the paper trails for XF and myself I have not found a link. I understand the paper trails can be wrong. I thought initially of half sibling relationships, adoption, or a NPE. What I find strange is that my daughter who has done a MH DNA test matches as expected to me and her mother but does not match to XF. From my limited understanding this is feasible possible but unlikely, but I stand to be corrected.
Any help?
Ray

Offline SouthseaSteel

  • RootsChat Member
  • ***
  • Posts: 127
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: Strange match
« Reply #1 on: Sunday 20 July 25 13:09 BST (UK) »

My rule of thumb of "converting" MH Matches to "equivalent" Ancestry levels, based on Ancestry's various algorithms addressing identity by descent (IBD) levels etc. is 50% for low levels like these.

I would look at this Match being more like 50 - 60 cM, which could be marginal.  And furthermore, nothing surprises me with MH Matches!!

Offline Biggles50

  • RootsChat Aristocrat
  • ******
  • Posts: 1,491
    • View Profile
Re: Strange match
« Reply #2 on: Sunday 20 July 25 14:03 BST (UK) »
Nothing on My Heritage surprises me.

It would seem that false positives can be presented in a large range of cM.

Take a look at the DNA tools on MyH, in particular the chromosome browser to see where you supposedly share segments.  Take and save a screenshot of the chromosome browser results for future reference.

I suggest uploading your Ancestry DNA to Gedmatch just to see if XF is there, if they are not then park them and see what happens.

Yes, it can happen that specific sections of DNA is not in following generations. 

My Wife has a 1C2R that she shares 187cM with (they share DNA along both his paternal & maternal lines), and in the shared matches there is one who shares 32cM with my Wife and the is another who shares 39cM with her and he is a close relative to the 32cM match.  The 1C2R is not a reportable DNA match to the 39cM match that my Wife has despite him being a closer relation to the 32cM match than my Wife is.

Good luck with your research.

Offline Glen in Tinsel Kni

  • RootsChat Aristocrat
  • ******
  • Posts: 1,461
  • Scottish Borders
    • View Profile
Re: Strange match
« Reply #3 on: Yesterday at 19:40 »
I'm not convinced there is a rule of thumb or conversion from one site to the other due to the testing and interpretation differences. 
I have matches on both that are within 2cM of each other yet others can be 5-60% higher on MH, the smallest in the group of 10 I just checked is 190cM with their MH figure being 230.

A father and son duo where the largest segment with the father is 45cM (28 segments in all), but the largest with the son is a standalone 65cM segment and just 21 segments overall.

Perhaps the most surprising is a half sibling around 1780 on Ancestry and over 400 higher on MH at a fraction under 2,200. Not the highest percentage difference by any means but enough to impact the relationship probabilities quite markedly. 


Offline rsel

  • RootsChat Senior
  • ****
  • Posts: 446
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: Strange match
« Reply #4 on: Today at 15:36 »
Ray,
   Have a look at the Chromosome Browser, and add both XF and you daughter, and do the compare.  What you may find is that in fact your daughter does match XF in the same places, but perhaps as lots of smaller chunks that are not quite big enough to trigger the match to actually be reported. 
   The reason i suggest this is from what you say whilst your total match size to XF is reasonable, its split over multiple segment's, and if 35cM is the largest its possible you daughter may not fully match you in those positions (say has 2 or 3 blocks).

Richard
Sellens - Sussex
Newham - Surrey
Wellington - Dagenham, Essex
Camp - South Essex
Wren - Essex
Livermore - Essex
Wane - Essex
Fisk - Essex / Suffolk
Bailey/Bayley - Sussex
Newton - Sussex
Funnell - Sussex
Streeter - Sussex
Coates - Sussex
Maisey - Surrey