Proper protocol when auDNA reveals erroneous relationship?

Question

WikiTree’s relationship finder reveals two people are 1st cousins once removed.  However WikiTree’s autosomal DNA compare feature shows they share no matching autosomal DNA in GEDmatch.  What steps do you suggest to document and try remedy that situation (that there is an error somewhere in their relationship trail)? What would you say to each DNA tested WikiTree user? Do you agree that it should be a private message (and not posted in the comment section of their profiles)?  Do you recommend creating a category so this finding does not get overlooked?

Thanks and sincerely, Peter

Comments

Many thanks, J. People who suffer through and read my way-too-wordy posts deserve some sort of award. 

Al: Actually, the age disparity between two full-siblings makes no difference in the DNA distributions they received from their parents. An older sibling cannot, simply by dint of being older, reveal DNA matches to older ancestors. The parents certainly can, however, because they're one generation, one meiosis event, closer to the ancestral line. Ditto for grandparents. That's one important reason the standard advice for anyone embarking on autosomal testing for genealogy is to test the oldest family members first.

---

I agree 99.999999% but there are some oddities such as where one (or both) of the test takers has had a bone-marrow transplant just for one example.  Of course, why would one take a DNA test after such an operation?

---

And even then only in the case of an allogeneic transplant as opposed to an autologous transplant...in the world of "hematopoietic stem cell transplantation" that we used to be able to call simply "bone marrow transplants." I need a wearble Google device just to keep up with terminology...   

Oh, and even with an allogeneic transplant there's a reasonably good chance that a cheek-swab test would read correctly where a saliva test wouldn't. The cheek's epithelial cells would still contain the patient's pre-transplant DNA, not that of the donor.

---

All true, and that 's why I am squarely on your side on this viewpoint,  I was just trying, per PEter's request, to give an example where it a valid relationship MIGHT not show-up in a DNA result irrespective of the fact that the odds are infinitesimal.

---

Yeppers; gotcha. I just saw an opportunity to try to spell "hematopoietic" and ran with it....  

---

I have not yet seen this compare dna that this post is talking about.  Can someone advise me where I can find that?

---

Hello Janice,

Go to https://www.wikitree.com/wiki/Anderson-32105

On the right side of your profile, under DNA Tested, click on [compare] to the right of A812017

Then click on [compare] next to A988592.

Then in the pop up window click on the COMPARE button

 Then Login to GEDmatch

Sincerely, Peter

---

Thanks Peter.  So my next question is that it seems then the person you want to compare to has to be listed in the area of It is likely that these autosomal DNA test-takers will share DNA with Janice.  How does Wiki determine who is entered there?

---

At https://www.wikitree.com/wiki/Anderson-32105 following [compare], click on [test details]. The first paragraph says “This includes all blood relatives within eight degrees of separation — up to sixth great-grandparents and out to third cousins.”

---

Ok thanks.

Answer 1

This type of situation needs to be handled with discretion. Private messaging is more appropriate than posting on the comments section of the profiles. The Wikitree adoption angel team may be able to offer advice on the wording to use.

Comments

Edie, Thanks for the best answer star.

Answer 2

Lynda is correct. You have to be very discrete if there are any living family members. If all family concerned agree to document this, then it goes in the Biography. This is a very sensitive topic. If the non-paternal event occurred several generations ago it becomes less so. You would also need to determine if an adoption was involved. Definitely private messages and not comments.

Answer 3

I ask again so that my other questions don’t get overlooked.

Communication will be via private message and handled in a delicate manner. 

What steps do you suggest to document and try remedy that situation (that there is an error somewhere in their relationship trail)? What would you say to each DNA tested WikiTree user? Do you recommend creating a category so this finding does not get overlooked? Do you suggest something else?

Thank you and sincerely, Peter

Comments

I believe in facts, simple facts, presented without opinion and without conclusion, as simply and objectively as possible.  Present them in a way that lets them figure out the problem for themselves.  If the DNA result indicates zero matching DNA, don't say that the persons don't match, just say that this test did not find any matching DNA.  Period.  If they want a suggestion, all I would add is that if they are certain there should be matching DNA, then they may want another test.  That way, they will find the corroboration they need, and recognize for themselves the truth.  If they insist on your opinion, all I would add is that I have never heard of a close matching relationship without matching DNA.  And while it's not impossible the company made a significant mistake, I've never seen one happen so far.

Opinions and advice rarely work if they conflict with a person's made-up mind and opinion.  It just starts arguments, which may even harden their opinion.  Show the horse the water, and let it decide what it wants.  Ensure they have all the facts clearly, and leave them to figure it out.

Answer 4

Hi Peter. I too have used the WikiTree compare and many times I have looked like I do not share any DNA with the other WikiTree user. What you have to do is write down the GEDmatch numbers (from the WikiTree compare) and log into GEDmatch and then run the numbers on the site itself as you would with other matches you are investigating. Use the tool where you can compare both kits to each other. With the list that test generates you must use the 3D browser tool as the others do not show results for the X chromosome. It is possible to only share X DNA with another person (the WikiTree compare only provides info for chromosomes 1-22).

When you run the numbers yourself you should look at what chromosomes you share DNA on and where the segments are located to see if you have overlapping segments. If you and the disputed match and at least one other all share the same chromosome at the same location (all with each other) then you share a MRCA, If you share the same chromosome but at different locations on that chromosome then you are related to the disputed first cousin once removed but descend from different ancestors. Also the amount of cM's you share with another match and your relationship to that match is not set in stone.

There is a range of cM's for each type of relationship. I have a 8th cousin who I share enough DNA with that makes us look like we are 4th cousins for instance. You may share a small amount of DNA with the disputed first cousin once removed and therefore a match will not be picked up in a one to one comparison (which is what WikiTree compare is) because the amount of shared DNA is too small.

It is quite possible that you do not share DNA with the disputed first cousin once removed but do share matches in common with the disputed match in which case you would need to work out what the relationships are between you and the disputed match and those you share in common.

If there is a discrepancy I would contact the person privately. First cousins once removed means that you are one generation back from the generation who share a common grandparent (that means your relationship to the other person is quite close) and therefore the situation should be handled with care and sensitivity. If you are certain about other matches that are on GEDmatch in this same line you should also compare those known matches with the disputed first cousin once removed before you even think about contacting that person. I hope this helps and good luck.

Comments

Hi Peter, do you have the GEDmatch numbers that you used with the WikiTree comapre that led to the disputed first cousin once removed?

---

Hello Maree,

You said, “I too have used the WikiTree compare and many times I have looked like I do not share any DNA with the other WikiTree user. What you have to do is write down the GEDmatch numbers (from the WikiTree compare) and log into GEDmatch and then run the numbers on the site itself as you would with other matches you are investigating.”

WikiTree’s compare feature simply links to GEDmatch’s One-to-one utility.  That is why you have to login to GEDmatch to use it. The comparison is running on the GEDmatch site itself.

---

“It is quite possible that you do not share DNA with the disputed first cousin once removed ”.  I disagree. True first cousins once removed will share autosomal DNA with each other.  See https://isogg.org/wiki/Autosomal_DNA_statistics and https://isogg.org/wiki/Autosomal_DNA_statistics#/media/File:Shared_cM_version_3.jpg

Sincerely,

---

Hi Peter, that is true but what if your relationship is NOT quite first cousins once removed? That is what I meant by saying 'it is quite possible that you DO NOT share DNA with the DISPUTED first cousin once removed' NOT that first cousins once removed do not share DNA. You will only know which is the case by looking at the match in detail on GEDmatch.

And what if the DNA you share is on the X chromosome (chromosome 23)? These matches do not show up in a one to one comparison. Sometimes to solve a problem you have to be a bit more flexible, did you run the numbers on GEDmatch (comparing both kits to each other and using the 3D browser)?

If I stuck rigidly to the ISOOG table (which I believe is a valuable tool) I would be looking for some of the solutions to my DNA matches in the wrong place. As I said I have an 8th cousin who I share enough DNA with that on the ISOOG table that match would be my 4th cousin!

---

Hi Peter, that is also true when you click compare you do have to log into GEDmatch. The thing is that that initial test is only the one to one comparison. You may not share enough cM's for it to show up as a match in the one to one comparison or you may share DNA on the X chromosome.

The one to one compare only analyses chromosome 1-22. That is why you need to do a compare both kits and use the 3D browser to be absolutely sure no match exists. I am not saying that the methods you use are bad only that I have had the same experience with the WikiTree compare and now I use it in a different way to be absolutely sure of my own results.

I just tried a WikiTree compare and there appears to be no way to get from the information that comes up (after you click compare on WikiTree and then log into GEDmatch) to GEDmatch's Home Page where you can use the full extent of tools available on their site. All I'm saying is that for more detailed and accurate results you need to log into GEDmatch youself and initiate the tests yourself.

You gave Kitty above some advice on how to use the WikiTree compare and then said that 'After you login to GEDmatch you will see that you share 12.4 cM with Peter Smith on chromosome 1'. Which is true, I did the compare just as you instructed and had the same result. But I also logged into GEDmatch myself and compared the two numbers with the 'People who match one or both of two kits' tool (and the 3D browser) and found that Kitty matches Peter Smith on chromosome 1 (12.6 cM's) and also matches Peter Smith on chromosome 22 (6.2 cM's) for a total of 18.8 cM's of DNA shared with Peter Smith.

Quite a difference don't you think!

---

Hello Maree,

The situation in the original question is first cousins once removed.  

You said “The one to one compare only analyses chromosome 1-22. That is why you need to do a compare both kits and use the 3D browser to be absolutely sure no match exists.“ 

Please provide two GEDmatch IDs (for 2nd cousins once removed or closer) that don’t match on One-to-one compare but do match on the 3D browser.

Most sincerely, Peter

---

Hello Maree, You said “I also logged into GEDmatch myself and compared the two numbers with the 'People who match one or both of two kits' tool (and the 3D browser) and found that Kitty matches Peter Smith on chromosome 1 (12.6 cM's) and also matches Peter Smith on chromosome 22 (6.2 cM's) for a total of 18.8 cM's of DNA shared with Peter Smith.”

What other GEDmatch ID did you include with Kitty and Peter Smith?  The 3D browser only lets you compare 3 or more people (not just two). I would be very surprised if the three of them all matched each other on that segment on chromosome 22.

Sincerely, Peter

---

Hi Peter, the situation in your original question was not a first cousin once removed as we both know it would be unlikely almost impossible for two people having that close a relationship not to share DNA. It is senseless of you to demand that I provide you with two GEDmatch IDs etc (as requested in your comment) as we both know that they would probably share DNA.

 I do not know what point you are trying to prove but why ask a question when you are not prepared to listen to the views of others?

I don't know what the validity of your statement is when you say 'I would be very surprised if the three of them all matched each other on that segment on chromosome 22'. Whether one or more people match the two kits being compared is irrelevant. It does not stop the fact that, in the case of Kitty and Peter, they share more DNA than a one to one comparison reveals.

I think we should agree to disagree as there is no gain in trying to change each others view not that I was trying to change your views. I just innocently happened along and answered a question from knowing my own experience with WikiTree compare and how I remedied it. I thought that I was helping but it appears I have just antagonised you and truly I don't know why. I wish you all the luck with your research and I think that should be the end of our conversation.

Answer 5

That's a challenging question.

Given a senario that a 1C1R on paper relationship has DNA tested with no matching result and they want Wikitree to publicly reveal this... what to do?

As just two DNA tests in a 1C1R on paper relationship does not reveal which of the parent-child relationships is an NPE, I would suggest that without a 3rd piece of evidence indicating where the NPE exists, that very little can be done to alter the wikitree tree structure.

A note could be made on each of the profiles about the situation.

Should any of the parent-child relationships be marked as "uncertain" when there is no evidence about any particular parent-child relationship being uncertain?   What is uncertain is the 1C1R relationship.

A Category set up to record such uncertainties won't reveal itself in a Relationship finder search.

So, I don't know the answer.

Regards

Veni