Go Back   Family Tree DNA Forums > Paternal Lineages (Y-DNA) > Paternal Lineage (Y-DNA STR) Advanced

Paternal Lineage (Y-DNA STR) Advanced Users can exchange information here on Y-DNA testing, results, and questions.

Reply
 
Thread Tools Display Modes
  #1  
Old 11th October 2017, 03:29 PM
TwiddlingThumbs TwiddlingThumbs is offline
FTDNA Customer
 
Join Date: Jan 2016
Posts: 57
Is TiP report garbage for non-matching surnames?

FTDNA guidelines on interpreting genetic distances assume the two men have the same surname. See, eg, Expected Relationships with Y-DNA STR Matches The implication is that, if the two men have different surnames (and there is no particular reason to believe there has been an NPE and that their genetic ancestors had the same surname), the probability of having a shared common ancestor within the genealogical time frame based on the genetic distances listed, are far less and the apparent "match" is most likely the result of convergent STR mutation rather than a common ancestor in the genealogical time frame.

As far as I can tell, however, the TiP reports do not suggest any need to distinguish between men with the same surnames. I think that is a problem. TMRCA calculations start with the assumption that two men have a common ancestor and then calculate the probabilities of how many generations it would take for one set of STRs to mutate to the other. With men with different surnames, however, that assumption is incorrect.

This relates to what I think may be the most common error in surname projects, which is to treat non-surname kits the same as surname kits. Many projects properly (in my view) exclude non-surname kits from matching groups unless there is reason to believe the non-surname kit had an NPE and is actually descended from someone with the surname. However, a significant number of projects seem to assume that every non-surname kit with a close genetic distance to a surname kit has that genetic distance due to a genetic common ancestor (and perhaps that an NPE occurred), whereas it is actually much more probable that the match was the result of convergent mutation.

Last edited by TwiddlingThumbs; 11th October 2017 at 03:37 PM.
Reply With Quote
  #2  
Old 11th October 2017, 10:13 PM
dna dna is offline
FTDNA Customer
 
Join Date: Aug 2014
Posts: 2,577
Yes, but...

Yes, but I think the real problem is that people tend to ignore probabilities that two individuals do not share a common ancestor within the last 1-24 generations.


Mr. W
Reply With Quote
  #3  
Old 12th October 2017, 09:45 PM
SurplusGadgets SurplusGadgets is offline
FTDNA Customer
 
Join Date: Mar 2016
Posts: 12
My experience has been different. The last thing I worry about is a surname when determining a match and grouping in a project. The first criteria (when available) is deep SNP test match; always. Next is STR match. Then we look at the surnames to classify as a variant, deviant or likely "NPE". People join our project either because of a surname match and/or an STR match to someone in our project.

I have one tightly knit group with over 10 surnames. And none are adopted. Just, for whatever reasons, these families decided to change to a mother's family name (more than once) over the last 300 years. They all descend from a common male ancestor after 1600. And they are all the only matches each other have (over 15 matches at y37; all patrilineal related). (We have shown that they were all living near each other in early, colonial Virginia with the families intermarrying.)

I have another group with similar surnames (variations, deviations) where with just a y12 result we can tell they are in the same group with the common ancestor from about 1500 or later. They simply have some very unique (rare) STR marker values in 3 of the y12 test suite markers. BigY has proven this out with the members.

Yet I have a few other groups that have 20+ y67 Matches that are not only not the same surname but known not to be the same patrilineal line. Their "real" patrilineal matches appear in the middle of the ordered STR match list. There are a few Haplogroups (DF27 in this case) where it appears this convergence occurs (in abundance). The STR test does not ever predict a very deep Haplogroup. Doing BigY or similar deeper SNP testing shows these testers are either in or not in the same patrilineal line. The ones in my surname study have "identical" matches to each other with BigY. Some of the other y67 STR matches who have done deeper SNP testing show they are down a different branch of P312 than DF27 (e.g. L20).

So, from experience, I do not think you should be making any statement of relatedness (or lack thereof) purely on given surnames. And not make any statement of likely convergence until you have deep SNP testing compared that shows they are different. Calculating and using GD and TiP without deep SNP testing is just guesswork before understanding more fully.

With all this said, in my experience, if they match STRs and also have the similar (variant, deviant) surname, then likely they match deep SNPs and a patrilineal relationship can be found if not known already.

I would finally add that the GD and TiP calculation is varied among these different groups as well. Some, with MRCA's of less than 200 years, have GD's of 3 or 4 at 67 markers. Others, with MRCA's over 300 years back, have GD's of 1 or 2 at 111 markers. So calculation of genetic distance and TiP from STR marker comparisons is still not a very exact science. Independent of Surnames matching but always only looked at if deep SNPs match. This large variance is even when taking into account different weighting factors for highly versus slowly varying markers. There is more going on than is maybe understood at this time. (I suspect related to the locations of markers in certain DNA areas prone to more major, connected changes.)

Last edited by SurplusGadgets; 12th October 2017 at 09:55 PM. Reason: Clarify some wording
Reply With Quote
  #4  
Old 13th October 2017, 04:05 PM
TwiddlingThumbs TwiddlingThumbs is offline
FTDNA Customer
 
Join Date: Jan 2016
Posts: 57
Good, detailed response SG. I still think you would agree with me that if all one knows is a gd or a TiP between two kits and the surnames, a close gd or TiP is much more likely to indicate a common male ancestor within the genealogical time frame if the surnames are the same than if they are not similar. (I think FTDNA does indicate that in the case of gd's and should also indicate that in the case of TiP.) I think your point is that we need to look at other available evidence besides STR matches to make a determination of of whether non-surname kits belong in matching surname groups.

I fully agree with that, but I think that in and of itself, suggested the need to treat non-surname kits different. If a kit with the project surname matches other kits in the project with the project surname, there is a very strong presumption that there is indeed a relationship there that is only overcome by strong countervailing evidence such as inconsistent SNP results. On the other hand, for a non-surname kit, the probabilities of the meaning of a match are different. In the case of kits with dissimilar surnames, I would say that the presumption should be that they are not related; however, there is no practical difference to just saying that other evidence must be examined before making a decision. Either way, I would say that if the non-surname kit (i) is not claiming male lineal descent from someone with the project surname (e.g., has not listed someone with the surname as his oldest known paternal ancestor), (ii) does not have its closest matches all with kits with the project surname, (iii) has not provided any genealogical evidence to suggest an NPE with a male with the project surname, and (iv) does not have SNP results that either prove or disprove a relationship, and thus the ONLY evidence in favor of a relationship is a low quality STR match, the non-surname kit should just be put in a non-surname group rather than in a group with the matching surname kits.

Last edited by TwiddlingThumbs; 13th October 2017 at 04:10 PM.
Reply With Quote
  #5  
Old 13th October 2017, 04:43 PM
clintonslayton76 clintonslayton76 is offline
FTDNA Customer
 
Join Date: Jun 2014
Posts: 152
I can only add an experience to this thread.
I have about 50 men in my overall project, but not all tested at FTDNA. We have YDNA evidence of two patrilines because about 15 are haplo I-P37.2 and 15 are I-M253. They cannot share a common Paternal in historic times, but all have used SLATON among their variants.
I was contacted by a male whose surname was not mine, but family stories suggested a surname very similar to ours, "Slater," for a man who killed a neighbor and took up a very common surname, Williams, managing to show in two later census in two states under that name. He later returned "home," whether surrendering or captured, and resumed his birth name of SLATON, serving a prison term under it. This was put together by good old detective work.
The male who contacted me (years ago) did his YDNA last year. His family were b as Williams, because the ancestor was using that name.
Had he simply looked for STR matches, and had not clues from the past, he would have found a very high match with a 37-marker male with his common surname, Williams. He would have found STR matches with our Slaton variants to be interesting but probably just a convergent coincidence (no SNP tests among our group except for me.) Why would he have even suspected a connection?
Interpretation of STR testing might well have caused years of fruitless research into WILLIAMS for this man. While I consider the surname to be an extremely unreliable indicator of relatedness, the relative guessing game of TiP and Gen Distances is just as questionable.
A Haplo admin grouped this man with WILLIAMS based solely on that STR match and the name in use, which was not unreasonable, but provably wrong. That is why surname admins must not rely solely on haplo admins, but must maintain good communications with them. They can help us if we help them.
The problem is bridging the huge gap between SNP and STR research, and I do not consider either to be "superior" to the other in genealogical research in all applications. The cost and complexity of
big SNP tests does not make it a good tool for those who are curious about their backgrounds in the most general sense.

Last edited by clintonslayton76; 13th October 2017 at 05:01 PM. Reason: clarity
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
how to interpret an Advanced Matching report? GoofusBroadway DNA and Genealogy for Beginners 3 7th January 2015 03:21 PM
Advanced Matching Report - Errors? Nite Owl Group Administrators - Advanced Chat 0 22nd April 2014 11:58 PM
No Matching Surnames MoberlyDrake Family Finder Advanced Topics 5 18th February 2012 12:28 AM
Things that work - matching, surnames, logic JohnG Family Finder Advanced Topics 0 25th August 2010 06:46 PM
Keeping Up With Matching Surnames MoberlyDrake Family Finder Advanced Topics 4 20th July 2010 06:10 AM


All times are GMT -5. The time now is 08:24 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 - 2017, Jelsoft Enterprises Ltd.