What do you think of this rough plan for storing and using haplogroups on WikiTree?

Question

WikiTreers, especially genetic genealogists and DNA Project members:

One thing we're often asked is why we don't associate maternal and paternal haplogroups reported from autosomal DNA tests (e.g. 23andMe) along the maternal and paternal lines.

The reason we don't do this is that our DNA test connection system was set up primarily with test comparison in mind. Haplogroups are just broad branches of trees. They're not useful for confirming relationships.

But, haplogroups are useful for rejecting relationships. If two people on a maternal or paternal line have conflicting haplogroups, it's likely that one of them has an error in their genealogy on WikiTree. I think there would be great value in displaying haplogroup information on profiles and reporting where there are conflicts.

Something that's kept us from moving forward on this in the past is the conflicting ways in which haplogroups are reported. I don't know much about this, but I know that the haplogroup tree has been reorganized to make room for recently-discovered groups, and this isn't definitively settled. Different companies use different taxonomies. And I think there are different levels of detail in the same taxonomy, because of the more advanced "Big Y"-type testing that's now available. Even though I don't fully understand all this, I would think that these are problems that can be solved. When a conflict is reported, we could just suggest that the members see if this is the reason for the conflict before they assume there's a genealogy problem.

Here's a rough outline of what I'm thinking.

We should start storing haplogroups in a new database table. Every individual could have multiple haplogroups associated with them. If they aren't all the same, we'll highlight this on the person's profile. We might someday do e-mail alerts about them. And/or maybe just have a special tool for reviewing the conflicts for people in your Watchlist.

Haplogroups will not be directly editable. They will automatically propagate from DNA tests. If you want to edit someone's haplogroup you'll need to edit what's entered for the test that's been associated with them, or edit a genealogical relationship that's leading to the connection.

We won't just auto-propagate from Y-chromosome and mitochondrial DNA tests. We'll do it from autosomal tests too, when haplogroups are reported from them. Even though they're from auDNA tests, they won't attach the way the basic DNA test connections for auDNA are made, i.e. out to seven or eight degrees in all directions. The haplogroups will propagate up and down maternal and paternal lines like the yDNA and mtDNA tests.

We could do pages that display everyone with a given haplogroup, and we could make it so that you can limit the display to those in your Watchlist.

This is just a rough sketch. If you've followed me this far, what do you think? Am I overlooking or misunderstanding something?

Thanks!

Chris

Comments

Another thought on the haplogroup naming problem, after rereading some comments from Peter Roberts elsewhere.

We don't need to depend on persistent URLs from ISOGG for explanations of haplogroups. We could have each haplogroup that's entered on WikiTree, however it's entered, link to a wiki page here, e.g. /wiki/Project:DNA/R-M269 or whatever. That page could link to relevant pages on the ISOGG wiki, explain that it's the same as R1b1-whatever and that R1b1-whatever shouldn't be considered a genealogical conflict. A volunteer here (Peter? others?) could head-off a lot of confusion.

---

If you have not done so already, please explore

http://www.wikitree.com/wiki/Space:Major_Y-DNA_and_mtDNA_Haplogroups

Most European males belong to Y haplogroup R which is at

http://www.wikitree.com/wiki/Space:Y-DNA_Haplogroup_R1b_%28aka_R-M343%29_Subhaplogroups

You can see the official hierarchy of those subhaplogroups at

http://www.isogg.org/tree/ISOGG_HapgrpR.html

That hierarchy changes (sometimes dramatically) as new SNPs are discovered.

One possibility is linking to the page for the major haplogroup on the ISSOG Y Tree (e.g. http://www.isogg.org/tree/ISOGG_HapgrpR.html) and then doing a keyword search for your SNP.  Anything equal to or higher than your SNP in the hierarchy could be a match.

Sincerely, Peter

---

Peter - the major haplogroup URL you referenced appears to have omitted E which certainly qualifies as one of the majors

---

E is under Y haplogroups.

---

Absolutely!!  Pulled the trigger before looking at the title!  Sorry

---

I think that this is a fantastic idea, Chris. But perhaps one should limit the Y chromosome haplogroups to their stumps, for the longer versions of the haplogroup (forgive my non-scientific terminology here!) seem to be changing all the time. My haplogroup at FTDNA was R1b1A2; later it mysteriously changed into R-M269. At 23andMe my paternal haplogroup is R1b1b2a1a2d*. Perhaps the stump R1b could be used? Maternal haplogroups, on the other hand, seem to be more stabile: mine was and still is U5b2b3a at FTDNA and  U5b2b at 23andMe. Perhaps the stump U5b2b could be used at WikiTree?  

---

Albertus, R1b1A2 = R-M269.  FTDNA decided to start using the terminal SNP the defines the long-hand title of haplogroups a while ago. All anyone has to do is visit the ISOGG tree, especially for the R branch, to see why using the short title makes sense. Using the same system your 23andMe terminal SNP ISOGG title = R-M335 as far as I can determine.

---

I do not understand the "problem" about using haplogroups, particularly the Y haplogroups identified by 23andMe or FTDNA and which are listed on the ISOGG tree.

Both Peter and I have worked on building out the Y-I1 haplotree categories here on the WikiTree, and if you start from my profile you can find the Y-DNA Haplogroup I-S12289 category, currently my terminal SNP, and follow the tree up to the parent DNA category. You can also follow my paternal lineage and find that I've added the I-S12289 category to each father since we have enough evidence within the Beardsley/lee DNA project to be confident that the common ancestor of our surname belongs to this haplogroup.

Adding mtDNA haplogroups should be even less of a problem as they are much more stable, and conversely much less defining, than Y haplogroups. If you test and are given a mtDNA haplogroup it makes sens to add that category all the way up your maternal lineage. Until, or maybe better stated unless, you run into another haplogroup on a common ancestor it is for all intents and purposes certain that each member of your maternal line was and is from the same haplogroup. As far as mtDNA is concerned, ALL of your maternal lineage within the genealogical timeframe will have your exact haplogroup.

If I have misstated anything, please correct my understanding. If I'm confused about what the issue with using haplogroups concerns, please clarify the matter for me.

---

Thanks John.  

Just to be clear, I don't support categorizing haplogroups of direct paternal line ancestors of Y-DNA test takers or direct maternal line ancestors of mtDNA test takers.  This is because WikiTree already automatically associates those profiles with the DNA information of their descendants who have DNA tested.

Sincerely, Peter

---

I find categorizing helps with those many ancestors I find on FTDNA, but whose tested descendants have not added their tests to WikiTree.  Therefore there's no DNA info on the common ancestor's WikiTree profile.

Answer 1

Hi Chris,

Rather than attempting to link to an external tree, or make reference to constantly changing haplogroup labels, have you considered WikiTree becoming the haplogroup tree itself?  I am speaking here in reference to the Y-DNA tree, but a similar situation would hold for mtDNA.

It was only a few years ago that I would have agreed with you that haplogroups were broad branches useful only for rejecting relationships.  I don't think that's the case anymore and it certainly won't be the case in the future.  NGS testing has changed that.  There are men now who have a more refined haplogroup than their father, which is more refined than his father. Y-DNA haplogroups now reach out into the very leaves on the WikiTree.

I'm not sure how much longer we'll be able to maintain centrally managed Y-DNA trees such as the FTDNA tree, the YFull tree, or even the ISOGG tree. Every man is potentially his own branch on that tree.  I wonder if a distributed project like the WikiTree is the only possible solution for the future.  There is no sense in maintaining two parallel trees of humanity, when one WikiTree can serve both purposes.

Such a change will require a small change in how haplogroups are viewed.  Rather than seeing them as some label, tagging some diverse group of men, view them instead as a male line ancestor, an unknown number of generations back, whose name we may not know.  Each man can make reference to one of these unknown ancestors, and each of them to the unknown ancestor upstream of them.

Like the WikiTree, volunteers would have to maintain the relationships between these ancient ancestors.  This is already an activity of many FTDNA project admins, and would provide a central site for collaboration.  The DNA variations that define each of these ancient ancestors would also have to find a home on WikiTree.  They are what someone would compare their DNA results against in order to determine the most recent ancestor from which they descend.

I know this is a different approach than what you were suggesting.  From my perspective, Y-DNA and mtDNA haplogroups are not some other source that should be referenced on WikiTree.  Instead, they are the natural continuation of the tree into the past.

Thanks for all your great work Chris.

Alex

Comments

Hi Alex,

Thanks for your post here. I like the way your mind works.

We didn't move forward with the plan I described above. I realized I didn't understand enough. I still don't. But I never seem to have time to focus on it.

Your post helps me understand a little more. I think what you're recommending would lead to creating profiles for distant ancestors based just on haplogroups. Interesting. Not sure how this would work.

This isn't exactly the same, but reminds me of a conflict recently where a member added a DNA test for a distant ancestor based on what he knows from the tests of descendants. This is a hack and inappropriate but I can see there is a need for something along these lines. Maybe storing haplogroups for people in a separate database table is still the answer.

Unfortunately, I don't have time to spend on this today. I want to make sure Peter Roberts sees this. He's the member who has played the largest part in shaping our current DNA-related features and I'll be seeing him at RootsTech in a couple weeks, where I hope we will have lots of time to discuss.

I hope you'll stay involved and help shape how we move forward.

Thanks,

Chris

---

Hello Alex,

Thank you for your suggestions.

Do you see a way to track assorted paternal line cousins who have SNP tested at various levels and/or are reporting the same SNP with different names, or reporting equivalent SNPs (which currently define the same subhaplogroup branch)?

One feature which might be helpful to add (or reference) is a path or crumb trail such as what Brad McGuire reported earlier in this thread:

R-L513>S5668>Z16340>FGC9087>FGC9793>FGC9804>Z16337

So, if a distant cousin only SNP tested down to the level of FGC9793 then it would be clearer to see that he is likely the same as his direct paternal line cousin who has the Z16337 SNP.

I also want to paraphrase an observation by Thomas Krahn of YSEQ that next generation Y chromosome testing will not likely get much less expensive.  Instead it may not be too long before full genome sequencing becomes affordable for genetic genealogy.  However full genome sequencing may not cover as much of the Y as a dedicated next generation test of the Y chromosome.

I'm open to any corrections if I've misinterpreted Dr. Krahn.

Sincerely, Peter

---

Hi Chris,

What I'm proposing would involve creating profiles for the nodes on the Y haplogroup tree.  To use some information from Peter's response, you would actually create a profile for L513, S5668, Z16340, FGC9078, etc.  You would also need a somewhat more general concept of 'father', one that would allow for some arbitrary number of generations back in time.  It could mean grandfather or great-grandfather, or it could mean gggggggggggggggggggg-grandfather depending on the situation.  The number of generations could be estimated, but would never be exact.  Rather than specifying a haplogroup, men could indicate who there y-line ancestor is.  The meaning is exactly the same.

About 4000 years ago, somewhere in northern France or Germany (possibly Britain/Ireland) some man lived who first carried the L513 mutation.  Saying that your haplogroup is R-L513 is exactly the same thing as saying you are the ggg....g-grandson of this man who lived 4000 or so years ago.  This same man had a grandson, or great-grandson (we just don't know exactly), who first carried the S5668 mutation.  Indicating that your haplogroup is R-S5668 is the same as saying you're one of his male line descendants.  Clearly, you would still be considered a descendent of L513 as well.  The Y haplogroup tree is a genealogical tree of real men, who were our ancestors.  Including those men on the WikiTree is the most natural way of incorporating haplogroup information.

It was only 2 years ago that the progression of branches, would have ended at L513 for Brad - it was R-P312>L21>DF13>L513.  Now, for Brad McGuire, we've added a number of new descendants to his line, R-P312>L21>DF13>L513>S5668>Z16340>FGC9087>FGC9793>FGC9804>Z16337.  Things are progressing quickly, and the most recent branches are now hundreds of years old for some men, instead of thousands.  This is well into the genealogical time frame.

If haplogroup information is going to be useful on WikiTree, we need to be able to compare two men and determine if they belong to closely related haplogroups.  When haplogroups were thousands of years old, it would have been ok to say they must have the same haplogroup, if they are to be potentially y-line related.  Not any more.  For many branches, we now have surnames broken up into different haplogroups.  If we want to draw any conclusions about comparing men, we have to know the structure of the tree.  One way or another, this structure has to be recorded.  It might as well be part of the tree itself.

As an interesting example, consider the haplogroup R-S781.  It originated in Scotland about 750 years old.  There are thousands of men who belong to this haplogroup.  To me, for WikiTree, simply indicating those men are R-S781 wouldn't do this situation justice.  In this case, we wouldn't even need to create a new profile for the man who founded this branch, he already has one.  He's Stewart-720.  There are just seven people on the tree who can claim him as their father, but potentially thousands of men who would also be his descendants but who can't yet trace their paper genealogy back to him.  Those men are his ggg...gg-grandsons, we're just uncertain on the number of greats.  There are also many more Stewart men who are related to him, but happen to be negative for S781.  This may seem like an extreme example, but it will become increasing common that men on the tree, are also the founders of haplogroups.

I look forward to discussing these options further.

Thanks,
Alex

---

Peter,

Ultimately, I think the haplogroup tree incorporated into WikiTree will look like a tree of phylogenetically equivalent blocks.  We should include all names and all equivalent mutations.  When a man has completed some yDNA single SNP tests, SNP packs or NGS tests, he'll compare the SNPs that he's positive for to what is on the tree.  He may not have results that indicate he is positive for all of the equivalents, but that doesn't matter.  He only needs to identify the most downstream block from which he descends.  I imagine some sort of interface where he can compare his own list of positive SNPs to the SNPs on the WikiTree, and can slowly work his way down the tree.

I have no idea of what pricing will look like in the future.  Prices have been dropping fairly quickly over the last decade, but that's not an indication of what the future will hold.  I have seen some Y-chromosome coverage information from whole genome NGS sequencing completed by FullGenomes.com and the results look promising to me, especially when completed at 20X to 30X.  It may even be the case that many men in the future will order WGS for the autosomal results, and the Y-chromosome data will just be an added bonus.  I may be wrong, but I think we'll have lots of yDNA data to work with going forward.

Alex

---

I really appreciated the ideas here, of both Alex and Peter, and I hope that someday those ideas will be implemented.  Since they require some development effort and prioritization though, I'd like to ask for something much simpler, for now, an interim method that's easy to implement, but practical, much better than we have at present.  It's basically Peter's suggestion of the "path or crumb trail", for the Y-DNA haplogroup box.

My yDNA haplogroup is currently R-Z17, and will shortly change again to something even less recognizable.  I think well over 90% of users would look at R-Z17 and compare it with R-M269 and say they don't match, and that's just wrong, as it's actually a subgroup of R-M269.  It's a sad thing here that the more you test your yDNA, the less useful your haplogroup becomes for most observers (and we certainly cannot require extensive training in yDNA haplogroup naming for everyone!).  And I can't at present enter anything more helpful in the yDNA haplogroup box.  I've put it in the DNA Notes box, but most won't ever see that.

If the Y-DNA box were expanded a little, and a few more characters allowed, then we could enter a simple and recognizable path, such as "R-M269->U106->Z18->Z17", something that is nearly intuitive even for the untrained.  I've only included notable points, which gets around the problem we have with constant new discoveries changing the path.  And new test results aren't likely to change anything, except add info at the end.  If I want, I could add the detailed SNP path in the DNA Notes.  If that untrained user compares this simple haplogroup path with R-M269, they can now draw the same conclusion as the trained person, that it could be a match but one is tested further than the other.  That's all we want them to conclude.  (And I'm not too worried about consistency or standardization in the pathing, anything even if inconsistent is better than now, plus no matter how different there should be enough in each path for easy eyeball comparisons.)

I hope you will consider this, it makes the haplogroup designation more useful for at-a-glance comparisons.

---

We could link whatever string is in the haplogroup field to a wiki page, e.g. https://www.wikitree.com/wiki/Help:R-M269 or https://www.wikitree.com/wiki/Category:R-M269

This would enable members to use the power of wiki editing, categorization, and redirects. The individual pages could give explanation and link to higher and lower-levels.

I don't know how far the DNA Project has gotten with https://www.wikitree.com/wiki/Category:Y-DNA_Haplogroups and how useful that has turned out to be, but it seems like this is a natural way to build the tree that Alex suggested. It's not a tree like our family tree. It's a hierarchical tree.

---

Alex,
Thank you!
I would LOVE WikiTree to have a haplotree in it's heirarchy of SNPs.
Particularly if the "descendant" tree was expandable/collapsible so you could quickly see how the branches inter-relate.
I do this in a VERY small scale on my DNASurnames site where I try to keep track of the branches that interest me, along with the patriarch of the tested line.
eg http://dnasurnames.info/dchaplogroupr.htm

Anyone with a particular interest in a particular branch could add to the existing Category pages for that branch whereby patriarchs and testees if they wished, could simply add the category to the profiles to link them all up.
A pattern of associated surnames with each category SNP would be "interesting"
I would expect  any aliases for the SNP to be explained on an associated category page.

The power of WikiTree (for me) is its DNA friendliness, and this would be a great addition., the ultimate in yDNA/paternal ancestral line validation.

The current haplogroup categories don't seem to have any hierarchical structure to them (at least not the ones I checked) - and of course have a mixture of old and new terminology for the SNP names.
I then scrolled down and saw that there are indeed hierarchies in the haplo categories, woops.

---

Lorna, maybe you can make this your project, i.e. the tree of haplogroup categorization?

---

Chris, I think we should try to stick to your one tree philosophy.

Currently, FTDNA has their own tree, YFull.com has their own tree and ISOGG maintains a tree as well.  Although it may not be possible at the moment, I think it would be better to try and leverage one of the existing trees, then to maintain an independent, parallel tree.

It wasn't that long ago I advocated for WikiTree to have its own version of the Y-DNA and mtDNA haplogroup trees integrated into its own structure.  That may have been too ambitious.  Using something that is already maintained may be the more sensible and practical option.

---

way too big for one person Chris :)

What's your reaction to having an expandable/collapsible descendant chart? (Hoping that it's not already there and I've missed it)

You do it for one of the pedigree chart options.

Would be very useful for navigating around trees for "normal" people, and fantastic for "haplo" people if you were to permit haplogroups to become "people" in a hierarchy.

Answer 2

Chris, My Y-DNA Haplogroup is R1b with a deep clade positive for 312. The mtDNA is Haplogroup H. I don't see myself going the 23andMe route, though.

Comments

Hi Frank.

What do you mean by "going the 23andMe route"? I didn't mean to imply that people who have taken y and mt tests should take a 23andMe test to get their haplogroups from them, if that's what you thought I meant.

By the way, I notice that you don't have a public tree. We wouldn't be able to connect any haplogroups for you unless that was changed. At least, that's the rule we follow for y and mt DNA tests. It would reveal too much for people who have chosen a high-level of privacy.

Chris

---

Someone told me not to take the 23andMe test several years ago.

My recent post was to indicate what tests I have taken and it seemed more complete to indicate the one I am unlikely to take.

---

If by 312 you mean P312, then that SNP actually defines the branch currently labeled R1b1a2a1a2.  P312 is also called PF6547 and S116 by different labs.

---

Hello Peter,

I had it as R1b1a2a1a1b, has it been revised?

---

P312 is currently defines R1b1a2a1a2

in 2012 it was R1b1a2a1a1b

in 2010 it was R1b1b2a1a2

in 2008 it was R1b1b2a1b

R1b1a2a1a1b is defined by Z19 and Z18 (Z18 is also called S493 and YSC0000053).

---

It is much better to call yourself R-P312 than to say you are R1b1a2a1a2.

You will also have others who insist on saying they are R1b-P312 which adds some more inconsistency.

---

In any case, my understanding is that about half of the R1b's test positive for R-312. Does that sound about right to you?

---

Yes, your Y haplogroup R-P312 is very common but I've not been able to find a percentage.

Answer 3

Just to be clear, 23andMe is the only auDNA test which happens to also report Y-DNA and mtDNA haplogroups .  They are outdated labels from about 2009.  [Added later: Genographic 2.0 also reported haplogroups but its auDNA results are not compatible with 23andMe, AncestryDNA or Family Finder].

Many tester's mistakenly believe there is somehow a connection between haplogroups and autosomal DNA or X-DNA.  I hope WikiTree's feature does not add to that misperception.

On the one hand it will be great see those 23andMe haplogoups automatically associated with their respective direct paternal and direct maternal lines.  On the other hand it will be interesting to see what confusion (if any?) ensues.

Most sincerely,

Comments

Although I have not received my test results yet, I believe Genographic 2.0 is also testing mtDNA now.  Is there, or will there be a way to incorporate my results within WikiTree?  

Is the problem just nomenclature?  -NGP

---

The Genographic Project is now marketing "Geno 2.0 Next Generation." According to http://forums.ftdna.com/showthread.php?p=414447 Geno 2.0 had 150 thousand atDNA SNPs and Geno 2.0 Next Generation has 750 thousand auDNA SNPS.

I expect it will look at more Y-DNA and mtDNA SNPs. I've not seen any details about that.

Answer 4

Hey Chris!

That is a great idea to develop a tool on Wikitree that would allow people to test/confirm their paternal lines with Y haplogroup information and test/confirm their maternal lines with mT haplogroup information.  The people who study haplogroups are always looking for ways to make their research useful in genealogy, and this sounds like the type of tool that would have appeal to those who have done any haplogroup testing.

It would seem as if you would have to keep the haplogroup classifications rather broad at first (i.e. R-M269) simply because with the existing Y tests, the large majority of Y testers have only tested at 12, 25, 37 or (probably at most) at 67 markers.  As such, there would be relatively few testers that have done Big-Y or a full-genome to allow for deeper identification of their haplogroup.  I don't know as much about the mT-DNA database of tests, but my guess is that there are different levels of tests in that database as well.

In the long run, I think this tool could one day become really useful for us - especially when we reach the point where full genome tests are commonplace (thus allowing for a large database of deeply identified haplogroups).

I don't know if you have ever talked with Blaine Bettinger, but he is really into haplogroup testing and would probably have a lot of insight on how this tool could/should be developed.  If not, here is his blog for reference:

http://www.thegeneticgenealogist.com/

Answer 5

Chris, I like your idea. There has been a workaround for the 23andme problem, but your suggestion makes it automatic. Calling the 23andMe test an "autosomal test" has never been accurate.

Can the date of the haplogroup report also be included in the WikiTree database table? It is my understanding that a haplogroup report from three years ago may not be the same today, even if they are both from the same company.

Comments

Hi Rick. Interesting point about the date. That might be something that we should be asking for in the extra info section of the test editing page, along with GEDMatch ID, etc.

---

I'm not sure many people will know what date to enter.  Also, what do you do with the existing entries?

---

Good point, Ken. Probably not worth asking for a date. We do have the date the test was entered on WikiTree.

Answer 6

Hi Chris,

I can see the value of the 23andme haplogroups being used to help identify inconsistent relationships, but I would not plan on going beyond using them for more than just display.

I am not sure how accurate comparing the haplogroups will be between two DNA testers.  If one tester is FTDNA and the other is 23andme, it seems to me to be a monumental task to determine if they are not consistent.

23andme uses one version of ISOGG map and presumably will at some point change it.  There are already over 5000 users who have already indicated they have taken the 23andme test. Most seem to have entered some value for their Haplogroup(s).  At some point, if or when 23andme users will begin entering a different version of the ISOGG map.

I would like to know if someone along the paternal/maternal line has taken a 23andme test but I think I may have to visually compare the reported results to determine if they are not consistent. 

I don't think this paticular list is going to grow very quickly. Some comparisons will be obvious, and some may need some research such as matching FTDNA result of M-269 with 23andme's R1b1b2a1a2d3a.

Answer 7

This is a great idea, Chris.  On the WikiTree, Cooper, SmithConnections Northeastern, Overby and Arbuthnot DNA webpages that I manage, the full deep haplogroup is listed for each individual test, but they are grouped by the general haplogroup, such as B, E, G, I1, I2, J1, J2, Q, R1a, R1b, T and others.  This grouping reduces confusion for haplogroups R-BY15 that matches haplotypes in R-L21, R-L513, and R-M269 because they are all in an R1b haplogroup.

In your scenario, I suggest that all the reported haplogroups be individually displayed on the profiles, but that the link to ISOGG go to the first letter level.  This will help explain that multiple tests for the same ancestor can have what appear to be different haplogroups, when they are actually deeper test results on the same haplogroup.  Does that make sense?  

This will be a great addition, Chris.  Thanks.  

Comments

Thanks, Kitty.

Note to self: If/when we do this, we should have a way to display by surname and sort by haplogroup, so, for example, you could see Smiths that begin with B, E, G, etc.

---

Good idea. But note that listing by surnames makes sense only for yDNA. It wouldn't do much good for mtDNA.

Answer 8

Good stuff Chris!  As an admin for FTDNA focused on Y DNA and given the recent surge in Next Generation Sequencing...your "tool" may become a fulltime job!!

R1b-L513>S5668>Z16340>FGC9087>FGC9793>FGC9804>Z16337 (aka>L69.5+)
R1b1a2a1a2c1b1a1a1a2

"Airghialla Mag Uidhir" Project Admin

I use several tools to assist in narrowing my project member's terminal branch. 
One of the best sites is Alex Williamson's Big Tree The Big Tree 

YFull is another great resource that I find helpful in finding genealogical branch points for various SNPs  YFull - Experimental YTree 

Just this week I've been able to confirm SNP branch points on our "known" Maguire pedigree or family tree. Essentially, this points to a tree that I pick up around 956AD and track up to about 1450 with subclades branching at known sept branches. I can email you an example if desired.

To the point of your query, I know within the L21>L513>S5668 crowd we have movement so rapid, that maintaining current nomenclature is a challenge.

My hat is off to your efforts!
Brad

Answer 9

I've created an experimental free space page for my I-Z59 Y haplogroup

http://www.wikitree.com/wiki/Space:Y_haplogroup_I-Z59

and a free space page for haplogroup I-Z58

http://www.wikitree.com/wiki/Space:Y_haplogroup_I-Z58

which is the ancestor of I-Z59.

Does this fit the plan for storing and using haplogroups in WikiTree?  Is something different being imagined?  Can this be improved?

Thanks and sincerely, Peter

Comments

I did something similar for R-L562 on the category page. [[http://www.wikitree.com/wiki/Category:Y-DNA_haplogroup_R-Z51/L562]]

---

Thanks James.  

Will this work better as a category project or as tagged free space pages (or a little of both)?

I am having trouble imagining how to capture a tree which is evolving and growing.  If each page for a haplogroup links to it's ancestor and links to each of its descendants (and links to each of its equivalents?) then the profile manager for that page only needs to be concerned with those links.

---

I do not understand how tags work for free space pages, so I cannot say which will work better, but as for links on the profile page either a category or free space page will work. There is a way to show what profiles are linked to a free space page.

Answer 10

Chris, I think this is an excellent ideal! An ideal that must be moved forward to stay fresh with the interest in having more DNA family information!

I have even seen DNA information on Facebook!

Thanks, Chris!

Answer 11

There may be another solution for doing comparisons of haplogroups when different haplogroup trees or labels are being used.

An automated process could point out when there is an exact match.  For example direct paternal line 5th cousins are both "R-M269".

There could also be an additional manual component or indicator (similar to the radio buttons for relationship status indicators) for when there is a "mismatch" which could actually be a match (e.g. R-M269 and R1b1a1a2, or R-M269 and R-CTS329, or R-M269 and R-P297, or R-M269 and R1b, or hundreds of other similar possibilities see http://isogg.org/tree/ISOGG_HapgrpR.html  ).  This manual selection would put the two haplogroups of descendants of the same direct paternal line in a possible match category.

The other indicator would be for when a mismatch can't be a match or is unlikely to be a match.  For example R-M269 and R-M73, or R1b1a1a2 and R1b1a1a1 or R-M269 and I-Z59, etc. This manual selection would put the two haplogroups of descendants of the same direct paternal line in a no match category.

Comments

A problem which remains is that two direct paternal line "5th cousins" could have a common surname like "Jones."  Because of their same surname, their direct paternal lines have been mistakenly joined together.  Those "5th cousins" both happen to belong to the common R-M269 haplogroup which would lead some to conclude their direct paternal line ancestries back to their most recent Jones ancestor is confirmed.  Only when you compare their haploTYPES does one see that they only match (for example) on 42 out of 67 markers which is too distant (not sufficient) for direct paternal line 5th cousins.

Another problem is people making too many mistakes when manually selecting "haplogroup match" or "haplogroup mis-match"

---

Absolutely Peter. My brother is one of the ubiquitous R-M269, or R-P312, and now is R-Z279. It will be very difficult for people to determine if his match is valid or not w/o comparing the markers at at least the Y67 level.

---

Peter, as the advanced tests, Big Y etc., become more prevalent aren't SNP results more accurate than STR haplotypes for the Y DNA lineages? Not for close relatives (5 generations or less) but at least for mid -  distant levels?

In my Wolcott lineage SNP testing suggested by Big Y results has definitively proven a split in the branches just 8 generations back, while analysis of the STR haplotypes could only suggest the probability of the division.

One thing for sure, family DNA projects need administrator(s) who keep(s) up with the changes that are going on in the science.

---

Hello John,  In an earlier thread I paraphrased "...an observation by Thomas Krahn of YSEQ that next generation Y chromosome testing will not likely get much less expensive.  Instead it may not be too long before full genome sequencing becomes affordable for genetic genealogy.  However full genome sequencing may not cover as much of the Y as a dedicated next generation test of the Y chromosome."

So next generation Y testing may not become economical for most genealogists.

---

I agree Peter, but that is where having competent DNA project administrators are essential. It does not make sense for every member of a family to take full sequence tests, it is far more economical for identified members to do that test based on known (or suspected) branches in their trees. Ideally the cost of that test could be split among members of the project Then other members who are close surname cousins can test the specific SNPs which are believed to define those branches, at much less expense for all concerned.

If SNP tests are emphasized, STR testing above the 37 marker level seems to be an unnecessary expense, while testing less than 37 markers does not provide enough information. It is only necessary to test enough STR markers to confirm your haplogroup, then test targeted SNPs to confirm your branch.

This only applies to Y DNA of course, and it all probably gets thrown out the window if NPEs are uncovered though.

---

Good point about sharing the cost of next generation Y-SNP testing and then having suspected direct paternal line cousins doing specific SNP testing.  Thank you.

I hope we can encourage more Y-DNA (surname) project admins to use WikiTree.  

I like to view it as Y-SNP and Y-STR testing rather than Y-SNP instead of Y-STR testing.

---

Absolutely STR and SNP, not one or the other!

With FTDNA you must do a STR test before they allow SNP purchases to the best of my knowledge, certainly before they allow Big Y purchases. And that is a good policy. Now that they are the only game in town for Y DNA tests (well, except for YSEQ) it makes it easier for project admins to direct folks to the proper tests. YSEQ is best used after initial testing with FTDNA for specific SNPs that FTDNA doesn't offer. Their SNP panels are superior to FTDNA's, and Thomas Krahn is very responsive to requests to develop new individual SNP or specialized SNP panels.

 

Answer 12

HI Chris,

To give some feedback: I've uploaded screenshots of the groups my tested males are in at FT's various "public results" pages. These are results anyone can see so it seems helpful to simply identify the matches/grouping  that corresponds with a particular surname and that line's ancestors and descendants.  A photo does the job.

Basic haplogroups are going to the same for each member within a group, but in some instances participants have tested specific markers, that is always helpful. In a matching group you can work together on testing specific markers and be fairly certain the other group members would also test positive for the specific marker tested. And, those specific designations that flow into the basic haplogroup could throw a system based purely on haplogroup definition off, it I'm reading your proposal correctly.

For example: if you follow my Mitchell-17863 male line you will see our group as placed within surname Mitchell project's public results at FT. Most of this group are tested to 67 markers, maybe one to 111, the rest of us need not bother knowing we will match at 111.  For my brother's kit we have only added testing for L-21 and since we were positive can safely predict we would also be positive for additional markers others may test.

So I have placed screenshots on profiles of males who would connect (unless an NPE) in the event that someone who connects to these various surnames piles in and decides to test. Identifies our groups.

And the Autosomal for Mom and Me flows quite nicely with all listed surnames, that has worked quite well!

Sherrie Mitchell-17863