How can we improve the use of X-DNA?

Question

Everyone who has taken an auDNA test (AncestryDNA, 23andMe, or Family Finder) also has X-DNA results.  Many of them don't know that.   WikiTree can tell us which ancestors could have contributed to our X-DNA:

http://www.wikitree.com/treewidget/Roberts-7085/89#X

and all of those ancestor's descendants who could have inherited their X-DNA:

http://www.wikitree.com/treewidget/Denmark-271/890#X

Using the above, and in a similar way that auDNA is currently displayed on profiles of our close relatives, may we please see a statement that "These X chromosome test-takers may share DNA with xyz."?

This will help inform auDNA test takers to take advantage of their X-DNA results. 

As an example of the usefulness of X-DNA, two male fifth cousins twice removed were able to confirm their ancestral lines (back 6 and 8 generations): http://www.wikitree.com/index.php?title=Special:Relationship&action=calculate&person1_name=Turner-12974&person2_name=Roberts-7085

Thanks and sincerely, Peter

http://www.wikitree.com/blog/wikitree-uncovers-xchromosome-ancestral-treasure/

Answer 1

Peter, I really have no idea what I am about to ask you so be patient with me.

I have my auDNA test indications on Wikitree.  Some of those profiles are carriers of my X Chromosomes and currently can only be discovered by the Wikitree X DNA tool.

Are you asking that the actual profile show an indicator of being a carrier of my X chromosome?

Are you asking that other auDNA test takers who share X DNA with the same ancestor be indicated on the profile?

Am I correct that some of these ancestors have my X Chromosome while other test takers indicated on the profile will not necessarily share X Chromosomes with me because of the way X Chromosomes are inherited?

How do we distinguish between those test takers that match and those who do not?

And now I need to learn what to do with that info once I understand it.

Comments

OK, I just read you blog post after the fact so I have some answers there.

---

Thanks Michael.  I'm proposing that WikiTree show a tester's relatives who could be an X-DNA match with that tester. These X-DNA relatives are a specific portion of the auDNA relatives which are currently displayed.

As it stands now, testers often don't bother to look for X-DNA matches.  I believe it is easier to work with X-DNA than auDNA (especially between males).
 

Answer 2

I like the idea Peter.  It would also serve as an educational tool about the special inheritance patterns.

Answer 3

Hi Peter,

I appreciate your persistence on this.

I don't know the right way forward. I'm uncomfortable treating all autosomal tests as essentially two separate tests and running connections for them separately. That would mean they appear on many profiles twice, and on the DNA surname index pages twice. It's a lot of clutter in the database and on pages. Especially since many members have taken multiple autosomal tests so they are already appearing multiple times.

We should probably store in a separate field in the connections database and display a little X by the autosomal test listings.

Part of the reason I haven't made this a priority is that the solution may connect with what we do with haplogroups. I still think we should be doing something different with haplogroups. Members who have them from autosomal tests always ask why the haplogroups don't attach up and down the maternal/paternal lines like mitochondrial and Y-DNA tests. The answer is that haplogroups can't be compared to confirm relationships. But they do have value. I know there is discussion right now on doing haplogroup categorization. That strikes me as too much manual labor and too hard to update. A database is the right way to do it.

One problem with moving forward on storing haplogroups (or automatically comparing them) is the inconsistency in naming them.

Maybe the haplogroup question should be put off and we should go ahead with something for X-DNA.

Chris

Comments

Thanks Chris,

AncestryDNA,  23andMe, and Family Finder each include two separate tests (auDNA and X-DNA).  X-DNA has a distinctively different inheritance pattern from auDNA.

It certainly causes additional clutter when a tester reports they tested (their auDNA and X-DNA) with multiple labs.  A possible solution is for the instruction/help page to recommend that auDNA testers only report one auDNA test and one GEDmatch ID.  In the note field for their auDNA test they could mention other labs they have tested with.

The chart at http://www.wikitree.com/treewidget/Roberts-7085/89 shows the little X that should be used.  My first thought was to have a separate list of X-DNA relatives  (in the way that Y, mt, and au are listed separately).  But I wonder (in order to avoid clutter) if it is possible to include the little X on the left side of the au for the X-DNA relatives of the profile that a user is looking at? Only some of our relatives who are likely to share some of our auDNA are able to share some of our X-DNA.

I want to avoid (put off) the haplogroup categorization issue.  I can only imagine it working if the ISOGG Y-Tree and phylotree had persistent links (urls) to the SNPs (which defined haplogroups) within those trees. That would allow one to know where their haplogroup fits in the hierarchy as the tree changes as new discoveries are added. 

I don't believe the solution for X-DNA or auDNA can connect with what is done with haplogroups because auDNA and X-DNA don't have haplogroups.  23andMe and the Genographic Project  report Y and mtDNA haplogroups but they have nothing to do with auDNA or X-DNA.

I believe the long term solution for auDNA and X-DNA is to automatically show the shared ancestry for people who all share the same auDNA (or X-DNA) segment with each other.  Then it will be much easier to confirm each of their ancestral lines back to their shared ancestor (or ancestral couple).  That is to say to have GEDmatch, or DNAgedcom, or DNA.Land allow a url which links to the tester's ancestry such as WikiTree's Relationship Finder or compact family tree.

---

Hi Peter,

Regarding ISOGG and persistent URLs, their tree is in their wiki, right? Persistent URLs are as simple as including redirects, something we do here. Old/alternate names could redirect to new names. I imagine Tim Janzen or whomever would be amenable to allowing the redirects if you explain the purpose and offer to do them.

I know au and xDNA don't have haplogroups. The reason I think we might want to think about them together is that including haplogroups apart from mt and yDNA tests will mean either a new DNA database or completely rethinking the current one.

Chris

---

Hi Chris,  The contact person who is helping with the ISOGG Y-Tree leadership transition told me that he will see if the programmer/others (at the ISOGG Y-DNA Haplogroup Tree) can take advantage having a persistent link to a SNP.

I hope we can look at making better use of X-DNA.

---

With regard to the 'clutter' that Peter mentions in reporting all auDNA tests,it seems we could eliminate some of that if Wikitree would have a single entry for auDNA tests.  Within that, there would be fields for the various sites & IDs (i.e. FTDNA kit #, Gedmatch ID #, ancestry.com member name, etc.).  When the person's name shows up on a profile as being an auDNA test-taker, if it would just show the person's name with a link to his DNA tests, it would greatly reduce the clutter, such as we have on this profile, where Landy has three separate entries due to having tested with all three testing companies:  http://www.wikitree.com/wiki/Gochenauer-7

---

Peter, the way you'd want to see the X connections, they would attach to every person up the tree in the X pattern, to the seventh(?) generation like auDNA tests, and then all the way the X pattern for each of those ancestors?

---

Hello Chris,

Every person up the tree in the X pattern, to the 7th generation and then all the way back down from each of those ancestors in the X pattern :-)

If the little X is on the left side of the au for the X-DNA relatives of the profile that a user is looking at, then there really is no additional clutter.  But there would be some who have an X and no au because we have the X going to the 7th generation (because it can useful out that far [and beyond]).

---

This discussion last week stirred up the haplogroup question in my mind. I'd really like to move forward with something, despite the haplogroup naming problem and link uncertainty.

I outlined my thoughts here: http://www.wikitree.com/g2g/192486/what-think-this-rough-plan-storing-using-haplogroups-wikitree

Answer 4

I have been using X-DNA for about a year now to find distant relatives.  I find a much higher hit ratio with people responding to me off of an X match than I do the regular autosomal match.     The beauty of the X is that it narrows down who in your chart the link may be coming from.  I am working with an adopted woman where we have an X match so that helps us eliminate entire parts of my tree in looking for a match with her... and because she has so little information to begin with, this is a big task.  The X has helped us narrow our focus.  I think just having the X on the DNA view is enough for most people to be able to quickly look at their trees and know where to target a search.

Comments

I also find working with X DNA matches to be easier and more productive than working with autosomal DNA matches.

This link shows who could have inherited Betsy's X chromosome:

https://www.wikitree.com/treewidget/Albury-14/890#X  and it reveals which of those descendants has had a autosomal DNA test.  Everyone with autosomal results also has X DNA results.