This project is generated by computer and should not be manually edited

Randy,

Thanks for the update.

Is it possible to represent the hierarchy somehow on the index page?

Representing the hierarchy got tossed because it was thought to be problematic on narrow displays.

Another issue with all of this is that geni uses the haplogroup name as reported by the uploaded data. For example suppose a user who tested a couple of years ago uploads their test and it gets put in the J2 haplogroup using the older name scheme. Now suppose that the same person retested today and the testing company would probably classify them as J-M172 instead of J2. Both J2 and J-M172 are valid auto-populated projects on geni and both refer to the same DNA haplogroup. So we actually have just one more thing to confuse users.

Is it really confusing? The J2 haplogroup project page redirects people to the J-M172 project page after a brief explanation. I'm not sure it'd be *less* confusing to automatically send people who understand their haplogroup as "J2" to a project that doesn't have that name, without explanation.

Incidentally, Wikipedia's solution is that when people get redirected because of page renaming, a line "Redirected from <original page name>" gets inserted under the displayed page name.

Harald Tveit Alvestrand For a promising way of dealing with treating a fuller heiarchy listing of haplogroups I'd point you to this project https://www.geni.com/projects/H1-Mitochondrial-DNA/33536

Here Eric H Mercer includes text that explans the history of the haplogroup name, how it is used on geni, and an example tree.

Eric is exploring adding similar text to many of the auto-generated haplogroup projects to give better context on name changes and history.

Here are some additional examples:

https://www.geni.com/projects/A-Y-DNA/36549

https://www.geni.com/projects/J-Y-DNA/33929

https://www.geni.com/projects/J-CTS900-Y-DNA/38439

My thought is that the index just points to the individual project pages where a full treatment can be found on that haplogroup.

I'm R -> L48 -> L47 -> S20039 and a series of intermediate haplogroups that I always forget. I'd like to have hierarchy represented so that I can see that (for instance) R-BY3314 is also within L47, but not within S20039.

I think representing the hierarchy would require some multi-sectional thing like:

== Root ==
* A
** Subgroup level 1
*** Subgroup level 2
**** Subgroup level 3
***** <a href="subgroup4">Subgroup level 4</a>

== <a name="#subgroup4">Subgroup level 4</a> ==
* Subgroup level 5
** Subgroup level 6
**** Subgroup level 7
***** <a href="subgroup8">Subgroup level 8</a>

using max 4 levels of indent, and intra-page links to link between the sections of subgroups.

Harald, this would be more difficult than it seems, although not for reasons that are obvious to most people. Whenever I try to explain this to people I get verbally beaten up for it with "but I don't think it should be so hard" responses. Well it's hard, and I've put in a lot of frustrating time gaining expertise just to learn this. I'm expecting to regret speaking up (again). :-)

First problem is which naming authority to use. For mtDNA using PhyloTree is certain, but for Y-DNA the "obvious" choice of ISOGG is a problem because they are often very out of date, and ftDNA (which generates the most data) doesn't use them. At the J2-M172 project at ftDNA, where I'm a co-admin, two different people have tried to get ISOGG to update the J2-M172 section and subclades and given up because of frustrating organizational issues there. For J2-M172, ISOGG's tree is nearly useless. One person in our team is planning to try again soon, but he's dreading it.

At Family Tree DNA, they are using a trade-secret tree created by the Genographic Project, but it's not public nor is the underlying data easily available to anyone, and therefore it's a very poor choice for us. The YFull tree is usually the most up to date, but so much so that it can be quite experimental. Some major Y-haplogroup projects, like J2-M172 ( http://tree.j2-m172.info/ ) and R1b-P312 ( http://ytree.net/ ) keep their own trees. It takes someone familiar with a particular haplogroup to know which are the best sources of information currently, and to know what info is well established and what is speculative.

On top of that, none of the naming authorities make available proper serialized tree data formatted in either the older phylogeny file formats (e.g. Newick, Nexus) or newer ones (e.g. phyloXML, NeXML). For automated processing, a properly formatted data file is important, but the Y-DNA/mtDNA phylogeny community is not dealing with this issue. The J2-M172 and R1b-P312 groups are exploring this, because they have larger administrative teams than most, and have the needed sosftware development expertise, but they aren't really working together (yet).

So yes, it seems like it should be possible and not too hard, but the foundation work is not yet in place. Compared to the researchers working on gene networks or species phylogeny, which face similar issues, the Y-DNA/mt-DNA phylogeny data handling is years behind. So in the meantime, my approach is to provide small, high-level trees for the highest level haplogroups, then provide a more detailed but still stable tree at some deeper point in the hierarchy where an HTML version is manageable. All manually, slow, and a pain to do, but it will get us through a few years until the automation is ready.

(Glad to see that Geni has taken over the maintenance of this.)

Not involved with DNA myself, I occasionally would go in a random DNA project, just to see the list of profiles "with tree matches". I thought Geni could do more to advertise this little-known feature to users under each haplogroup, who can easily make the merges (with the imported trees from March 2016) that would increase the numbers, as well as making WFT more connected.

I also thought we would've seen more -- or any -- stories of which historical figures have got "confirmed" haplogroups (propagated from two or more different branches). I thought that would be the most exciting aspect of DNA on Geni, and it puzzles me why Geni isn't doing anything (I just searched the Geni blog). The staff probably doesn't know enough (of the scientific literature on genetics) to write a well-informed piece; perhaps someone here could contribute?

I thought I had said somewhere how I find DNA merges to do... so here's an improved version, especially for large haplogroups (with thousands of profiles).

- go to List of Profiles
- (show advanced control, if necessary)
- filter by "with tree matches"
- if too many matches with single profile trees (with HIDDEN name), make a filter by "deceased" or "public"
- if still too many and you'd want to work "near the bottom" of the tree, filter by date of birth "after 1880"

Many of the DNA trees imported back in March 2016 are only "ancestor tree", with no uncles and cousins. Working from the bottom is strongly advised. (Now that we have a link to "descendant comparison" on the bottom of any side-by-side merge comparison, it's easier to find the bottom.)

Again, I wish that Geni could pop up and say "Congratulations, you merged in ### many profiles to the haplogroup ***"

Hy Randy, I wonder if you could add my group to the index Y18103-Y-DNA. Thanks

You are I-M253, on your Geni Profile, so this project will only create Y18103 when you upload your results that will change your profile haplogroup.

Noted, automatically updated. Awesome stuff 👍