uncertain parents and DNA does not work well do we need a policy change?

Question

https://www.wikitree.com/wiki/Help:Uncertain_Parents

I believe parents should not be added if uncertain and marked uncertain.
Not only does this sometimes result in conflated profiles if you merge! But it messes up the DNA if it is not correct. By assigning DNA tests to all related profiles (per type of DNA testing) and creating multiple errors if not correct parents.

IF THE TREE IS WRONG THEN THE DNA IS TOO on those profiles that are connected because of the error

Comments

see https://voshart.medium.com/genealogy-and-uncertainty-c41fd106a3cf

which allowing uncertain parents to be added would then make the error rate even higher than the 25% claimed here which to me is not acceptable for a genealogy site or WikiTree. So it should be  about well sourced and documented trees not uncertain to make DNA useful.

1. Dry family trees are what paperwork says and is based on the nuclear family. For many, this is enough and most ancestry websites are based on dry family trees.
2. Wet family trees are DNA focused and require navigating Non-paternity events (NPE). Common NPEs include (but not limited to): illegitimacy outside marriage; infidelity within marriage; rape and adoption.
3. Social family trees vary depending on culture and explain someone’s lived reality. Depths of claims vary. If depth of claims are shallow the chances of dry and wet family trees aligning are higher than if claims go deeper. Some might denounce their dry or wet family trees. It is common for people to add social family through marriage, re-marriage, God-parents, ceremonial adoption, clan or totem systems, and informal adoption. Social claims need to be evaluated and re-evaluated with the social climate.

The error rates between dry and wet are much higher than people imagine. Contemporary NPE statistics range between 2–12%. Using an International median rate of 3.8%, which compounds each generation, suggests there is one in four chance of finding an error in your great-grandparent genealogy.

---

I am new to this platform.  This caught my eye because it may relate to a question I have and the way I work my DNA research in my own tree on ancestry.  It is NOT foolproof, but it is very handy in a directional way (finding which way to go);

I look for multiple DNA evidence in my 2nd great grand generation and older; in other words, I look for 5 or more DNA related people coming down from siblings of "my" line, before I mark a male as my ancestor and for the wife of the same male, I look for DNA evidence of the wife's parents the same way.  My dad's DNA goes back to his 5th generation quite nicely, and we have multiple DNA matches that cooborate this.  Is this how wiki tree works?  How do I indicate that here on wiki tree?  Should I upload my expansive tree here?  What is the best way to move forward.  Ty.

---

That's a good approach, Wikitree has help pages on adding DNA confirmation, for above the 3rd cousin level triangulation is required, that is a 3 way match between the descendants of 3 children of the most recent common ancestor, referred to as the 3 legged stool test (two legs are allowed to meet one generation below the MCRA).

 There is a app for generating the required DNA confirmation statement to be put on the profile (I put these under a '''DNA RESEACH NOTE''' header at the bottom of the profile (there's a stack of these confirmation statements on my profile which require a major tidy-up)

The app works well for triangulated relationships with one 3rd cousin and more distant relationships, eg 3C, 3C1R, 3C1R, but not for 3C, 3C, 3C1R, it also won't let you triangulate 3rd cousins.
 Relationships including and below 3C level don't require triangulation on Wikitree but still require a confirmation statement, which the app also generates unless there's a half 3rd cousin in the mix, that has to be manually generated.
 One day I intend to make a table for triangulations of my NZ Harris family 3rd cousins, I can get the full limit of a triangulation of myself and 6 3rd cousins, and have several triangulations.

 

DNA Confirmation Citation Maker app

Answer 1

During the tech meeting on Tuesday, we discussed adding an "uncertain" marker next to tests that have gone through an uncertain relationship.

But the policy of allowing parents to be marked "uncertain" isn't going to change.

Comments

Thanks for the answer

---

Did you discuss what to do if there is a DNA mismatch. Such as with Haplo groups that don't match and an obvious error in the tree proven by DNA?

---

Haplogroup mismatches are already reported in the weekly suggestions report.

---

Anyway thanks loads for all the work you techs do WikiTree has kept improving there is no way everyone is going to agree all the time. For me I am always against uncertain in regards to genealogy uncertain for me just belongs in the research notes when it comes to connecting parents wives and children

---

How does one see the weekly mismatch report? Does it show on an ancestor's WikiTree page?

---

It would be either Suggestion 162 https://www.wikitree.com/wiki/Space:DBE_162 or 167 https://www.wikitree.com/wiki/Space:DBE_167

Answer 2

If a father is marked as uncertain then it is possible to use Y chromosome matching of the patrilineal descendants of two of his sons as supporting evidence that he is the father.

Similarly, if a mother is marked as uncertain then it is possible to use mtDNA matching of the matrilineal descendants of two of her daughters as supporting evidence that she is the mother.

Also, if a mother is marked as uncertain then it is possible to use X chromosome matching of X chromosome descendants of her sons (or son AND daughter) as supporting evidence that she is the mother.

If uncertain parents are disconnected then it is exceeding difficult to use DNA matching to determine if the relationship is valid.

Comments

But until you can find the matches to have the proof that it is correct (or an error if they don't match) something you may never find -- until then if it happens You have spread that possible error to many profiles. Others then may think on other profiles they don't match when they do because of that error in the tree. The company you test and ged match etc is what you use to find the matches, not WikiTree?  When you do find the match, then you can add the parents based on those results?

---

If the DNA test results of descendants do not match, it is an indication of a problem with the genealogy. By examining the direct lines in WikiTree for Y or mtDNA, you can identify which relationship is marked as uncertain. In such cases, it is recommended to disconnect the parents and include a note explaining the disconnection based on the mismatch of the DNA test results of descendants.

If the possible parents are not connected then the DNA results of descendants of suspected siblings are of little help.

Answer 3

Full discloser, I have not taken a DNA test and have not tried to tie any test to profiles.

"IF THE TREE IS WRONG THEN THE DNA IS TOO"

That statement brings up the question of DNA relying on the Tree. If that is true what does the DNA add to the lineage? Seems it would be better if the DNA led us to correct parents. Not the other way around.

Just thinking outload.

Richard.

Comments

I work with Y DNA which is easier than autosomal as it is passed down male lines only with few changes. The Y DNA you do, if you are a male, is your line along your male ancestor line. If 2 male siblings or 2 males whose male lines whose ancestors are siblings back in time test and have a reliable DNA match. Then you know most of the Y DNA of their male ancestor lines and confirms your tree is correct. If one doesn't match the other then one has a problem with their biological male line tree compared to the paper tree. But which one you have to find the error on the tree. What happens when you add uncertain fathers if it is an error. That error spreads up and down all the male lines on WIKITREE  to all male lines directly related from the common ancestor. So it does not just make one error but multiple and they are often very difficult to correct as we have no idea where the error is along the line.

---

The “Unconfirmed” status is necessary in cases where the father could be either of two brothers. Y-DNA will confirm the same patrilineal connection regardless of which brother is the father. Autosomal is not useful if this case occurred >6 generations back. If the father is simply removed, rather than flagging the relationship as “Unconfirmed”, all descendants below that point will be disconnected from their WikiTree Y-DNA Connections. This exact scenario occurred to my line when someone decided to delete my “Unconfirmed” paternal ancestor who is one of two brothers.

---

I am referring to marking parents uncertain when unknown if correct or not. Confirming by DNA is whole other issue on Wikitree  which most people know nothing about and do not use for the most from what I have seen.