Go Back   Family Tree DNA Forums > Universal Lineage Testing (Autosomal DNA) > Family Finder Advanced Topics

Family Finder Advanced Topics Advanced discussion about Family Tree DNA's Family Finder Product.

Reply
 
Thread Tools Display Modes
  #1  
Old 25th June 2018, 05:52 PM
KATM KATM is online now
mtDNA: K1a3 | Y-DNA: R-L1308*
 
Join Date: Nov 2012
Location: Mid-Atlantic coast, U.S.A.
Posts: 1,111
Talking Riddle me this, Batman: FTDNA vs. GEDmatch results not the same

Did you ever have a match at FTDNA who is also at GEDmatch, but doesn’t match you the same way at GEDmatch as at FTDNA?

I had an email from a person, "Z," who matches an FTDNA kit I manage, "A." This person Z is an estimated 5th to Distant cousin to A. I found that Z also matches A's paternal 1st cousin "Q" (also as a 5th to Distant cousin), and two of A's children, "E" and "F". Using the Family Finder Chromosome Browser, I see that:
  • Z matches A, and A's children E and F, on chromosome #12 from 118775846 to 123922969, for 9 cM. That indicates the segment was passed on by A to E and F, completely intact.
  • Z matches Q on chromosome #12 from 118775846 to 124270507, for 9.87 cM. This indicates that Q received a bit larger segment from her parent, who was a full sibling to A's parent.
Since Z gave his GEDmatch number, I checked to see how A and Q matched him at GEDmatch, using the One-to-One comparison. Result: there is no match shown on chromosome #12 for either A or Q, compared with Z, and GEDmatch shows no segments shared at the default setting of 7 cM minimum segment size.

Usually I leave the GEDmatch setting for segment size at that default 7 cM. But I thought I'd see what showed if I lowered it to 3 cM, thinking perhaps chromosome #12 would show at least a little something. Alas, no:
  • Z matches A on chromosome #1 for 3.3 cM, and chromosome #15 for 3.8 cM.
  • Z matches Q on chromosome #1 for 3.5 cM (a different segment than A), on chromosome #4 for 4.3 cM, on chromosome #14 for 3.2 cM, and on chromosome #17 for 3.2 cM.
Note that these smaller matching segments shown at GEDmatch DO NOT match any small segments in the FTDNA Chromosome Browser for A or Q, when compared with Z.

I also checked to see if Z matched A's sibling, "B," but Z does not show up in B's match list at FTDNA.
At GEDmatch, Z matches B on two chromosomes, #15 and #18, for 4.5 cM each.

I realize that the two siblings and their cousin could inherit different segments, so may not match Z the same way, or at all, but wonder why is there the difference in matching for A or Q to Z between GEDmatch and FTDNA? Why does a 9 or 9.87 cM segment seen at FTDNA, or any portion of it, not show up at GEDmatch?

Last edited by KATM; 25th June 2018 at 05:54 PM.
Reply With Quote
  #2  
Old 25th June 2018, 06:39 PM
prairielad prairielad is offline
FTDNA Customer
 
Join Date: Feb 2011
Location: Canada
Posts: 2,022
Is any kit a transfer kit to ftdna or are both tested at Ftdna?

Are both kits on gedmatch ftdna kits?
Reply With Quote
  #3  
Old 25th June 2018, 07:19 PM
KATM KATM is online now
mtDNA: K1a3 | Y-DNA: R-L1308*
 
Join Date: Nov 2012
Location: Mid-Atlantic coast, U.S.A.
Posts: 1,111
Z's kit has the FTDNA "T" prefix, as does A's, but I know that A's kit was a transfer from 23andMe, several years ago (v. 3, so should have a good overlap, with almost as many SNPs tested). The "T" kit for A was from a file downloaded from FTDNA. There is a separate, research kit at GEDmatch for A, using the 23andMe downloaded file. That research kit also does not have any matching segments at the default 7 cM minimum size.

I just checked at GEDmatch to see how E and F would match Z, since they share the same segment with Z at FTDNA, in common with A. Both E and F did the Family Finder test at FTDNA. Results:
  • Z shares with E on chromosome #4 for 3.5 cM, and chromosome #8 for 3.6 cM.
  • Z shares with F on chromosome #13 for 4.1 cM, and chromosome #17 for 3.8 cM.
Neither of these match the small segments shown for A compared with Z at GEDmatch.

I'm not so much concerned with the odd differences in the small segments shown at GEDmatch for any of these people (although I'd like to know why that's happening, too), but curious as to why a 9 cM segment would not show up there, when it is seen at FTDNA.

Last edited by KATM; 25th June 2018 at 07:32 PM.
Reply With Quote
  #4  
Old 25th June 2018, 07:37 PM
KATM KATM is online now
mtDNA: K1a3 | Y-DNA: R-L1308*
 
Join Date: Nov 2012
Location: Mid-Atlantic coast, U.S.A.
Posts: 1,111
I should add that Q (the 1st cousin to A) also took the Family Finder test at FTDNA, so no transfers for E, F or Q. I do not know if Z's kit was a transfer or not; he has not replied yet to my most recent email to him.

A's 23andMe research kit at GEDmatch gets the same results as his active "T" kit, when minimum segment size is set to 3 cM. I neglected to mention in my last post that I set the minimum to 3 cM when comparing E and F to Z at GEDmatch.
Reply With Quote
  #5  
Old 26th June 2018, 06:32 AM
Frederator Frederator is offline
FTDNA Customer
 
Join Date: Jul 2010
Posts: 819
Well, at under 10 cM, it's probably not worth the effort to verify this, but the most likely explanation is that there is a difference between the mismatch tolerance settings in the FTDNA and Gedmatch matching algorithms. Remember, the one-to-one utility has customizeable parameters called 'mismatch evaluation window' and 'mismatch bunching limit'.

We're talking about 700,000 or more SNPs in a typical commercial test kit. There are bound to be a certain number of call errors in there, hopefully very small in absolute terms. But if the evaluation space is less than 10 cM, even one could be the difference between registering as a match or not.

Frankly, with a segment this small, it could be even more trivial than that. It could be a freak interaction with any or all of those parameters and an incredibly minor difference in the Builds used by FTDNA and Gedmatch. If memory serves, Gedmatch uses an older Build version (i.e., set of assumptions about the cM value assigned to specific SNPs) than FTDNA. Not to mention the difference in the specific array of SNPs used by FTDNA and Ancestry.com that another poster mentioned earlier.
Reply With Quote
  #6  
Old 26th June 2018, 09:42 AM
KATM KATM is online now
mtDNA: K1a3 | Y-DNA: R-L1308*
 
Join Date: Nov 2012
Location: Mid-Atlantic coast, U.S.A.
Posts: 1,111
Thanks for your thoughts, Frederator. I may play with those settings at GEDmatch, and see if anything changes. It's true, such a segment may not be worth the effort. I only became interested in it when the match ("Z") contacted me. The ideas you put forth make sense.

Such are the vagaries of DNA matching! I think, in a way, this makes me even more wary of distant matches at FTDNA (and elsewhere). If such minor differences in builds, tolerances, etc. can end up showing a qualifying segment that, in actuality, may really not be a true match, how can you trust such matches?

Last edited by KATM; 26th June 2018 at 09:58 AM.
Reply With Quote
  #7  
Old 26th June 2018, 10:32 AM
Frederator Frederator is offline
FTDNA Customer
 
Join Date: Jul 2010
Posts: 819
True. But what can a person do? Most people undertake these tests to resolve some mystery or brick wall or some such, so it seems almost an imperative to wring the absolute most out of each match in order to put the whole into a wider context. I still obsess over these small matches myself.

I guess we just have to become more comfortable making our conclusions on such partial information tentative. And be prepared to be proven totally wrong, eventually.
Reply With Quote
Reply

Bookmarks


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Ran FTDNA results in GEDMatch. Native American DNA found that is confusing. mayita Family Finder Advanced Topics 3 31st May 2016 01:33 PM
Need Help Interpreting FTDNA vs GEDMatch Results ebenoodt DNA and Genealogy for Beginners 4 17th April 2015 10:25 PM
GedMatch vs FTDNA differences mabrams Family Finder Advanced Topics 15 25th October 2013 10:05 AM
FTDNA VS GEDMATCH RESULTS Mendomaid Family Finder Advanced Topics 0 11th May 2013 12:35 PM
Holy SNPS, Batman! Wing_Genealogis Paternal Lineage (Y-DNA STR) Advanced 20 16th December 2012 06:52 PM


All times are GMT -5. The time now is 10:36 AM.


Family Tree DNA - World Headquarters

1445 North Loop West, Suite 820
Houston, Texas 77008, USA

Phone: (713) 868-1438 | Fax: (832) 201-7147
Copyright 2001-2010 Genealogy by Genetics, Ltd.
Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.