Author Topic: Gedmatch - lowering parameters valid match or not?  (Read 229 times)

Offline sugarfizzle

  • RootsChat Veteran
  • *****
  • Posts: 896
    • View Profile
Gedmatch - lowering parameters valid match or not?
« on: Tuesday 04 December 18 18:39 GMT (UK) »
Asking for some thoughts. Sorry about the length of this post.

I mainly post about practical aspects of DNA tests, rather than the scientific. The scientific goes above my head sometimes, especially as I am getting on in years and my brain doesn't seem to understand things as easily as it used to.

I am a member of Steer mailing list, Steer was my maiden name. Quite a few of us descended from the same family have had autosomal DNA testing, but unfortunately do not match each other.

Those who have uploaded to Gedmatch match with me on a very small part of chromosome 6, found only by lowering the parameters, so not really valid matches in DNA terms.

I have access to the results of one list member, TDS, and she has a valid match at ancestry, who I have persuaded to upload to Gedmatch.

He matches her in the same place as the rest of us on chromosome 6, but for 10 cM.

I have painted them all into DNA painter, though documenting that this latest match is not with me, IYSWIM.

Does this imply that the lower matches are valid or not?

On paper, 4 of us are descended from one couple, my 4G Steer grandparents (TDS's 3G grandparents) Others are descended from the previous generation, my 5G Steer grandparents.

Trying to get my head around it.  ???

Regards Margaret

STEER, mainly Surrey, Kent; PINNOCKS/HAINES, Gosport, Hants; BARKER, mainly Broadwater, Sussex; Gosport, Hampshire; LAVERSUCH, Micheldever, Hampshire; WESTALL, London, Reading, Berks; HYDE, Croydon, Surrey; BRIGDEN, Hadlow, Kent and London; TUTHILL/STEPHENS, London
WILKINSON, Leeds, Yorkshire and Liverpool; WILLIAMSON, Liverpool; BEARE, Yeovil, Somerset; ALLEN, Kent and London; GORST, Liverpool; HOYLE, mainly Leeds, Yorkshire

Census Information is Crown Copyright, from www.nationalarchives.go

RootsChat is the busiest, largest free family history forum site in the country. It is completely free to use. Register now.
Also register instantly with Facebook or Twitter (and other social networks). Start your genealogy search now.


Offline davidft

  • RootsChat Marquessate
  • *******
  • Posts: 3,520
  • Census information Crown Copyright, from www.nationalarchives.gov.uk
    • View Profile
Re: Gedmatch - lowering parameters valid match or not?
« Reply #1 on: Tuesday 04 December 18 19:27 GMT (UK) »
My inclination is to stick to the default de minimums level that the various sites suggest e.g. 7cMs on DNA painter.

What you may be doing by lowering the threshold is hitting on false positives i.e. the lower the threshold the easier it is to find a match but that does not mean it is a genetic match just one out of coincidence.

The first person I got to reply to me on MyHeritage last week when discussing the potential match I had made agreed that the first segment was very likely a genetic match from a familial link however she pointed out that the second segment was likely a false positive as I matched more with the potential match than my father did, which of course should not be the case in a familial link. She suggested that this was what she called a "sticky" bit of ancient dna that many people might share with no familial connection (at least not on a recent timescale).

The way forward may be to see if you can triangulate the match with others, if you cannot it is not a "true" match. Also is there another generation in your direct line that can test (or has done already) to see how they match or not with this lower level potential match

So to summarise I would say not valid unless or until you can find stronger supporting data

I'll be interested to see what others suggest.

RootsChat is the busiest, largest free family history forum site in the country. It is completely free to use. Register now.
Also register instantly with Facebook or Twitter (and other social networks). Start your genealogy search now.


Offline familydar

  • RootsChat Veteran
  • *****
  • Posts: 767
    • View Profile
Re: Gedmatch - lowering parameters valid match or not?
« Reply #2 on: Tuesday 04 December 18 19:38 GMT (UK) »
To play devil's advocate, I feel the opposite to davidft.  I have something similar with one of my lines, no proper triangulation but a cluster of low matches in the same area of the same single chromosome.  TDS is closer to the mrca so is likely to have inherited more of their DNA than the rest of you.  Do any of you have any low grade matches with TDA on different segments that can't be attributed to other lines?

Like you Margaret I struggle to get my head round this sort of thing of late - it used to come so easily!

Jane :-)
ALLEN
BARR, BARRATT, BERRY, BRADLEY,BRAMLEY,BRISTOW,BROWN,BUGBIRD,BUTLER
CAIN,CARR,CHAPMAN,CHARLES,CH*LTON,CHESTER,COCKETT
COLLASON,COLLYER,CORKERY
DARLING, DENYER,DICKERSON,DOLLING,DURBAN
FARMER,FURNELL
GIBSON,GILES,GROOMBRIDGE
HALL,HAMBIDGE,HARMES,HART,HICKS,HILL,HOLLOWAY
JACKSON
K*AT*S
LANCASTER,LINTON
MCDONALD,MCFADEN,MEARS,MILLARD
NICOLAS,NOAK,NORTH
PARFIT,PORTER
RIPPINGALE,ROBINS
SEARLE,SPENCER,STEDHAM
TYLER,TILLY,TUCKWELL
WADE,WAGER,WALKER,WATSON,WEBB,WITHRINGTON,WOOD

Offline sugarfizzle

  • RootsChat Veteran
  • *****
  • Posts: 896
    • View Profile
Re: Gedmatch - lowering parameters valid match or not?
« Reply #3 on: Wednesday 05 December 18 00:26 GMT (UK) »
Thank you both for your replies.

I realise that conventional wisdom is that lower matches are insignificant. However, logic dictates that if, say, my father inherited c 8 cMs from this Steer couple, then I could inherit c 4 cMs.

Unfortunately, we are all of about the same age group, nobody from the previous generation to test. TDS is in fact younger than most of us - youngest child of youngest child etc, plus ancestor having children late in life.

They all match with me at approx the same place on chromosome 6, TDS and this match spread across the rest of us for 10 cMs. None of them match with me anywhere else, even lowering the threshold at Gedmatch to about 3 cMs. Below 3 cMs there are several miniscule random matches.

I have 2 valid matches at ancestry from this line, unfortunately I cannot persuade them to upload elsewhere.

Thank you both again for your input, there will hopefully be more matches in the future on this line.

Regards Margaret
STEER, mainly Surrey, Kent; PINNOCKS/HAINES, Gosport, Hants; BARKER, mainly Broadwater, Sussex; Gosport, Hampshire; LAVERSUCH, Micheldever, Hampshire; WESTALL, London, Reading, Berks; HYDE, Croydon, Surrey; BRIGDEN, Hadlow, Kent and London; TUTHILL/STEPHENS, London
WILKINSON, Leeds, Yorkshire and Liverpool; WILLIAMSON, Liverpool; BEARE, Yeovil, Somerset; ALLEN, Kent and London; GORST, Liverpool; HOYLE, mainly Leeds, Yorkshire

Census Information is Crown Copyright, from www.nationalarchives.go

Offline JaneyH_104

  • RootsChat Extra
  • **
  • Posts: 93
  • My great-grandmother, Elizabeth Dyson
    • View Profile
Re: Gedmatch - lowering parameters valid match or not?
« Reply #4 on: Thursday 06 December 18 14:33 GMT (UK) »
I’d agree with davidft wholeheartedly.

My scientific knowledge is limited but growing, but there are good reasons why so many people urge caution over lowering the thresholds on GEDmatch and DNA painter. Essentially the risk of false positives becomes real and small segments can be identical by chance rather than identical by descent. Without other corroborating evidence I’d leave this on the back burner as a possible.

Related to this, if you’re working (like me) with Ancestry and have people who don’t want to upload elsewhere, have you seen the new website app called “Genetic Affairs”? You can upload data from Ancestry, FTDNA and 23andMe (so far). It runs a cross tabulation of all your matches against each other, but then groups them into clusters for you, automatically. It’s free for the first few runs and then you can either subscribe or pay for periodic updates. There are a few blog articles about it, although it’s only been out a week or so.
BOWDLER - Forest of Dean & Devon, DYSON, ENTWISTLE & TOWNEND - Huddersfield, CLARKE - Dorset, SCOBLE - Devon, HOUGH, COPE & WHITTAKER - Cheshire, BRACHER - Wiltshire, DENNISS - Herts/Hunts, SQUIRE - Hunts/Beds, BROWN - Herts/Beds