Did you see the new one-to-one GEDmatch DNA test comparison links?

+60 votes
3.5k views

Hi WikiTreers,

I'm very excited to announce the following. It's something I've been wanting every time I see a distant cousin's DNA test connected to one of my ancestor's profiles (which is becoming more and more common -- 3.3 million of our profiles have them).

You will now see "[compare]" links next to GEDmatch kit IDs on profiles and DNA Ancestor Confirmation Aid pages such as Roberts-7085/899.

For your own Ancestor Confirmation Aid page, click here and then select the "DNA Confirmation" button.

If you select or enter two GEDmatch IDs you can go directly to a one-to-one comparison on GEDmatch with one click. (This also works for Ysearch and Mitosearch IDs. This capability isn't new, but we integrated the user interface for it.)

For more info, see Help:DNA_Comparison.

This is another step forward in our growing relationship with GEDmatch. And there's more on the way. John Olson, John Hayward, and Curtis Rogers are constantly improving GEDmatch's capabilities and we're working together to utilize their advanced genetic genealogy tools for our mission to grow an accurate single family tree.

In particular, GEDmatch is in the process of restructuring and improving their Multi-Kit analysis. And they're working on integrating the new Genesis kit IDs into their main database. When these changes are done, GEDmatch will enable WikiTree members to select more than just two kits at a time for comparison. This will be a major advance for triangulation, and for genetic genealogy in general, I'd say.

Onward and upward,

Chris

P.S. As with many of our DNA features, this would not have happened without Peter Roberts. It's to his credit, from start to finish. The last pieces fell into place at the FTDNA conference down in Houston. Peter, Mags Gaulden, and a whole contingent of other WikiTreers were evangelizing for the importance of the single family tree in genetic genealogy. Face-to-face interactions in the small community of advanced genetic genealogists really make a difference. Peter's discussions with John Olson there got the last pieces into place for these comparison links.

in The Tree House by Chris Whitten G2G Astronaut (1.1m points)

Looks nice

That sounds awesome, Magnus. It wouldn't be feasible on category pages. That would be too much to load from multiple databases. But I wonder how else it could be done. Maybe Ales or another independent developer would want to tackle it.
Hi Chris,

GREAT!!! job guys

If you ever get the GPS project for locations going, I'd love to help out. I've got 20 years experience building GUI SQL relational databases. I'm retired so I have a lot of free time. I think it could be a very easy fix and open up a lot of new options for Wikitree

Hi Richard,

Which GPS project? Now that we attempt to normalize location names using FamilySearch's Places database, it's becoming more feasible to map location names to GPS coordinates.

You should join the WT Apps Project and post here in G2G about your ideas. Maybe someone else would want to collaborate with you.

OK, will do. Is there place where I can read/learn about the data structure of Wikitree?

I've seen some talk on g2g about GPS locations. https://www.wikitree.com/g2g/508271/can-we-be-careful-about-local-location-names-please?show=508323#a508323 is a good one from Magnus.

and on data-doctors

https://apps.wikitree.com/apps/ is a good start I think....

Thanks for this exciting announcement, and special thanks to Peter Roberts for setting this up--I am super excited to give it a try!
Very cool. Thank u
I've been wishing for a easier and quicker way to compare. Thanks so much!
Thanks everyone,

We appreciate all your hard work!!!

Your should be awarded the max points for this!!!

Taylor
Yes! I've used them already. I finally got a cousin and a sister on gedmatch.com so have updated our profiles with this information. Would you mind taking a gander at my profile sources section to see if everything looks kosher?

3 Answers

+13 votes
Wow!  (It works better than I imagined it :-)
by Peter Roberts G2G6 Pilot (489k points)
edited by Peter Roberts
+7 votes
I tried it and got a match on one of the shared profiles. I added a header DNA Notes and copied the match information to the profile.

The other matches were not big enough to trigger a match with the criteria used. I didn't see the option of changing the criteria like you can do on Gedmatch.
by Sue Hall G2G6 Pilot (122k points)
  • I guess you played with Dunham-1259
    • GEDMATCH compare URL T350359 A510412

      Comparing Kit A510412 (Mary Hall) and T350359 (Millicent Miller Benbow)

      Minimum threshold size to be included in total = 500 SNPs
      Mismatch-bunching Limit = 250 SNPs
      Minimum segment cM to be included in total = 7.0 cM

 

Feels a Template should be created doing the same but you can have in the documentation in the bio section and explain your analyze of the result....

  • {{DNAGEDCOM
    |T350359
    |A510412}}
  • Add third parameter could maybe be the changing the criteria... 

EDIT Feels gedmatch.com still has a referer check..... which makes it tough with direct URLs....

And the chromosome location. If you are trying to triangulate you are aiming for the same spot.

You have a template that some "Einstein" has destroyed and added red headlines with warning and no alternate solution I have never understood why people at WikiTree make creating Templates so complex... its just text substitution to avoid problems all other wikis understand this benefit and use them a lot.... as genealogy is using more and more external links it must be used to avoid having Aleš bot walking around and guess and change..... its a bad pattern that WikiTree first create all errors with imported GEDMATCHES external links with no templates and then have tools and competitions to fix them... would be easier do it the other way around....  

Below the match with the destroyed template

Chr Start Location End Location Centimorgans (cM) SNPs
13 90,166,047 96,324,226 8.1 995

 

{{auDNA Match 
| people = [[Hall-16256|Sue Hall]], [[Miller-28432|Millicent Benbow]]
| common = [[Dunham-1259|George Dunham 1785-1845]]
| chromo = 13
| start = 90,166,047
| stop = 96,324,226
| cM = 8.1
}}

This template should also have GEDMATCH ID and  creating the same URL as the new tool..... I guess its no rocket science to add a compare button in the same way we see upper right on profiles with DNA tests..... 

Adding this in a template then we in the future could have more tools t support if you find another person sharing this segment with you and give you a hint that a test at FTDNA also was for the same segment....

My understanding there are tools like dnapainter.com that use this kind of information to visualize shared segments. A template is a good start to get this information in a structured way video 


Big pic

Not to get too complicated, but would the template be set up to allow multiple people? Or would it be another template to aim at triangulation? common=A, desc= a,b,c,d chromo = 13, start=x,  cM=8

The  template above has named arguments ==> YES you can have more peoples etc....

The above template is rather simple as it has just "text" arguments with all the people involved

people = [[Hall-16256|Sue Hall]], [[Miller-28432|Millicent Benbow]] 

I guess DNA genealogy is developing fast so the requirements will change and what is needed in a template will change but having a template is a good start..... 

@ Peter Roberts 

Restricting to 7cM for a system with no restrictions feels odd....

  • I have a connection Bildtse-2 ==> third cousins once removed with a rather good paper documentation
    • DNA Genealogy
      • ==> 10% of the people with this relation has less than 7cM
      • Testing on FTDNA we dont see anything
      • Testing on GEDMATCH we can see 4 cM ==> its a weak evidence that the paper trail can be correct see Non-Confirmed_DNA_connections_2
If you and your match were both using phased auDNA then you could be confident at 4 cM.
phased auDNA?

I have matches on 3 different lines that match on the tree but are less than 7 cM. I documented the Pomeroy matches on a page that I created for the Pomeroy DNA
https://www.wikitree.com/wiki/Space:Pomeroy_DNA

There are matches over 7cM that are not close.
It would also be nice to have a template for Gedmatches where the other person is not on wiki tree but maybe ancestry or another site. Why not use these to demonstrate DNA matches?
Gedmatch ids could still be used. While it would be possible to add everyone to wikitree, it would be nice to show the match without making this mandatory.

There are over 200 Ancestry 4th cousins, one first cousin, one first cousin once removed.

  • The first cousin (no tree) shares 722 centimorgans shared across 34 DNA segments (Confidence: extremely high). We share 24 4th cousin matches). 
  • The first cousin once removed, Mirielle13, shares 380 centimorgans shared across 17 DNA segments (Confidence: Extremely High). We share 16 4th cousin matches. Note this cousin 1st cousin once removed is on Gedmatch.
  •  

>> If you and your match were both using phased auDNA then you could be confident at 4 cM

Yes maybe I can use Maijgren-1 F311279 to at least isolate what is on my fathers side.... to get a higher confidence that Bildtse-2 T786703 is a hit on my mothers side..... confident is a strong word in genealogy ;-) 

+5 votes
It is an absolutely wonderful feature and I use it all the time to past results to DNA Painter.

https://dnapainter.com
by J G G2G6 Mach 8 (84.4k points)

Related questions

+60 votes
8 answers
+25 votes
2 answers
+34 votes
3 answers
+26 votes
0 answers
+18 votes
1 answer
+25 votes
5 answers
+3 votes
1 answer
73 views asked Mar 16, 2018 in Genealogy Help by James Bonds G2G Crew (340 points)
+36 votes
3 answers

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright

...