Is DNA confirmed appropriate for these circumstances?

Question

Confirmed with DNA is allowed if the testing company or GEDmatch predicts third cousin or closer.  Is that a WikiTree convention or a general genetic genealogy convention?

If GEDmatch predicts the MRCA is 4 generations ago, that's third cousin and confirmed with DNA is appropriate.  If GEDmatch predicts 4.5 generations, a confirmed by DNA is not allowed.  

If the one to many comparison on GEDmatch predicts 4.0 generations, but the one to one comparison predicts 4.1 generations, confirmed with DNA is not warranted.

If a third cousin (confirmed with paper trail) and I have our MRCA predicted at 4.5 generations, that is still a no go for confirmed by DNA.

Do I have that right?  Thanks!

Comments

Hey, Robert. Just a quick comment that, yes, the 3rd cousin or closer guideline is WikiTree's, not a general convention. There is actually no "confirmed with DNA" general convention at all. Given the way WikiTree is structured, it's reasonable to me that they would need to draw a line in the sand somewhere that is clear and simple to communicate. Third cousin is as good as place as any to draw it: even though it's possible for two 3Cs to share no measurable autosomal DNA, if they do it's likely to be in the range of 50 to 70 centiMorgans, a chunk large enough clearly to be evidence and not coincidence. At 3C1R, the theoretical amount of shared DNA is halved from 3C, and at 4C it's halved yet again; at 4C we're down to only an expected 13cM, so it's reasonable to want additional data to validate the DNA evidence for more distant cousins.

---

Thanks to everyone in helping me understand this topic of DNA confirmation.  I still have a lot of studying ahead of me!

Answer 1

No, Robert.  The amount of DNA shared can only predict the relationship.  This is why a solid and confirmed paper trail must accompany the DNA.  I have a confirmed 2C 2x removed who shows as a fourth cousin as far as how many centimorgans we share.  Did I use her to confirm the DNA on my relationship trail because she is less than a 3rd cousin? Yep. Because I have a solid paper trail with plenty of sources.

However, I very clearly stated in my confirmation statement that we were predicted at 4th cousins, but the paper trail confirmed we are 2C2R and the amount of centimorgans we shared was within an acceptable range for 2C2R based on ISOGG relationship charts.  

What any company predicts for a relationship (not just Gedmatch) is a ballpark estimate.  If I share 1700 centimorgans with someone, they could be my half sibling, aunt/uncle or grandparent.  This is why the paper trail is a must to prove which of the three it is.

Comments

Emma,

There is a discrepancy between what you wrote and what it says on the Help:DNA Confirmation page.  The criterion for GEDmatch says, "If GEDmatch's One-to-One utility predicts that your match could be a third cousin or closer and you know the person to be a third cousin or closer through traditional genealogy...".  There is similar language for other tests, "If the predicted relationship range in "Family Finder - Matches" says your match could be a third cousin or closer and you know the person to be a third cousin or closer through traditional genealogy..."  The help page specifies that the testing company must predict a 3rd cousin or closer match.

I personally agree with you that your 2C2R that is predicted as a 4th cousin, should be valid, but that is not how the help page reads.  Perhaps the language could be modified.  I'd support doing that.

Kerry

---

Thank you Kerry for bringing that to my attention.  This discussion has come up in the DNA project group.  I will mention it again and see if we can revise that section just a bit.  One of the problems is though, people skip over this part quite often:  

you know the person to be a third cousin or closer through traditional genealogy

P.S.  I am sending you a message on another topic...

---

Emma

Loved your answer on so many points, especially your flagrant disregard of the wikitree rules, which, albeit necessary, are rather arbitrary.

I would really like to get away from this concept of "Confirmed by DNA".

The most important thing is to get the paper trail right (and sometimes DNA matches can be great pointers as to where to look) and with that in place, the DNA matches will be consistent with the tree.

On my wish list would be to change the label from "Confirmed by" to "Consistent with". Perhaps you could mention this idea in the deliberations within the DNA group.

---

I never have a flagrant disregard for the wikiTree rules, Derrick. The help pages need to be updated to correctly reflect the intent behind the written word.  The intent is, a one to one comparison can be used for a third cousin or closer.  I have a proven 2C2R, thus I was able to use the one to one comparison with this match.

As to wanting to change the verbiage for the "confirmed by," you would need to start a new g2g thread for this because it is not up to the DNA Project to make system changes.  The collective voice of our members is what makes these kind of changes happen.

---

A 4th-cousin-level DNA match is well within the expected range for 2nd-cousin-2x-removed (equivalent to a 3rd cousin), but some skeptics would point out that the DNA match is equally likely to be a half-2nd-cousin-2x-removed. So the match you marked as "confirmed" is consistent with WikiTree guidance, but you would have more confidence in the DNA confirmation if another pair of testers in the family had a stronger match than you.

---

"flagrant disregard" was said tongue-in-cheek, my apologies if I touched a nerve. It is sometimes difficult to see the intent behind the written word.

The guidance notes manage to be specific and vague at the same time. Just what is "3rd cousin or closer"?

My simplistic mind works on the basis that a 2c1r is equivalent to a 3c on the easy to understand basis that 2+1=3. It would not occur to me to use the more complex argument that the average match for a 2c1r (123 cM - Blaine Bettinger's work) is larger than that for a 3c and therefor 2c1r is actually a closer relationship rather than being an equivalent one.

So I would have said that 2c2r is equivalent (in terms of genetic removal) to 4c, not 3c, and so falls outside the wikitree guidelines.

I'm guessing that your view is that as the average match for those two classes is the same then they are equivalent. There are several other relationships that fall in the same ballpark, including half 1c3r - I would struggle to say that half 1c3r is closer than 3c, but the figures suggest that it is.

My intent in writing this has nothing at all to do with an "I'm right, you're wrong" argument - that would be trite and a complete waste of time and energy. I actually agree with what you've done and believe that the guidance is poorly phrased, often ignored and framed around the wrong basic premise.

I'd appreciate your advice on how to raise the issue in G2G - as well as the obvious "DNA" what would be appropriate tags to add in order to bring the query to the attention of someone who is able to effect change?

I had assumed that you were exactly the right person to talk to in order to influence change, but hey, I got it wrong again.

---

The mathematics on percentage of DNA inheritance from common ancestry indicate that 2C2R is equivalent to 3C.

---

I actually interpreted "could be a third cousin or closer".  This means if the match has any possibility of being a 3rd cousins or closer. This means that just about any genetic distance could be valid. Could a genetic distance of 5 be a 3rd cousin? The answer is YES.