View Full Version : Massive Y-DNA Haplogroup Projects at FTDNA

06-22-2020, 01:44 PM
Ever wonder about the utility of the truly ponderous y-haplogroup projects at FTDNA, like the R1b Project? Some of them are gargantuan.

I guess they might be good for data miners - maybe - but they seem too big to me. Who looks through all their pages? Anybody?

Perhaps it would be better if FTDNA had central hubs for all the major y-dna haplogroups. There one would find links to all the various smaller projects nested within the larger haplogroups.

Exactly how that would be done, I'm not sure. The hub for R1b could direct the curious to other hubs based on the major nodes of FTDNA's Haplotree for R1b, for example.

Just a thought.

Any other ideas?

Think the massive y-dna projects are a great idea?

J1 DYS388=13
06-22-2020, 01:49 PM
I can't load the data from the R1b and R1a projects, so they are useless to me.

06-22-2020, 02:09 PM
FTDNA has improved performance that now allows you to download these projects for the first time in a long time. Eventually - as these projects continue to grow, there will download issues in the future again.

If you belong to P312, U106 or L21, I now recommend testers leave the R1b project as it just bloats this largest project in FTDNA and will continually cause accessibility issues (as well as workload on FTDNA
servers which just drives up costs for minimal benefit). Also, if you belong to L21, there is no need to be a member of P312. There are several geographic projects that are getting very large as well - but these
make more sense unless accessibility issues arise in the future. For R-L226, I do recommend joining both Z253 and L21 (as backups in case there is a later issue with L226 or Z253). But as L21 grows too large,
this would be reduced to only Z253. U106 presents a future accessibility issue as well (over 6,000 members). It is OK for now - but it has no projects below it (predictable haplogroups that could be started now to address future growth of U106).

There is a huge need to create more haplogroup projects for those haplogroups that are predictable and chartable. L513 is only 1700 testers but L193 is a very large predictable haplogroup where a new haplogroup
project should be created. This would allow more focus on these large predictable haplogroups and would reduce the size of L513 over the long term. For these more manageable predictable haplogroups, YSNP prediction
should be done and promoted. Also, charting could be done as well which would better benefit the members of the predictable haplogroups. FTDNA did allow R1b to automatically join its members to lower projects
recently but it was not requested or rejected to remove these moves from R1b.

06-22-2020, 02:43 PM
I have been a member of the big J-M304 Project. We can easily detect specific branches in the J1 haplogroup even with 12 STR markers because they are quite ancient, several branches with more than 12000 yrs. I created our specific branch J-M365 and I always told J1 had a Northern Middle Eastern division. Of course J1 is a very geopolitical haplogroup full of ethnoreligious groups, if you belong to a specific cluster you belong to a specific religion and geography and I always told we were Northern Portuguese and Cristãos Velhos. When they created a new J1 Project we were not invited because we had our different identity and some people didn't like our ideas now proven correct. M365 was among the first discovered SNPs and just like in several other cases it is recurrent, but I also discovered our complete trail of SNPs when the full sequence tests were first available at FGC. I could pick some members to our group in the big J and J1 Projects because we can easily recognize the STR modal even with 12 markers, so the big Projects were quite useful for us.

06-22-2020, 04:14 PM
I like the large haplogroup projects personally and would hate to see them go away. On the data analysis side the consolidated data pool allows admins (if they care to do it) to create consolidated looks like allele frequency tables which in that particular example allows individual testers to see which STR mutations they have are particularly unique in their larger haplogroup and therefore more likely to indicate shared recent ancestry.

At the individual tester level the large haplogroups give STR testers a chance to work forward from ancient times to hopefully find a common subgroup closer to present day. Especially for testers with no (or few) matches on the match reports, haplogroup projects often represent the only chance for those testers to find whatever matches really are closest to them in the testing databases. And when your estimated haplogroup is something like "R-M269", the larger haplogroup projects are the only place where you can start.

Widespread SNP testing may eventually make those particular examples less important but with perhaps 750,000 STR-tested men and only perhaps 50.000-55,000 deep SNP tests (Big Y, WGS, etc but the number there is a very rough guess) I think we still have a need for regular STR analysis.

I think it's a shame that people are recommending that new testers not join the larger haplogroup projects just because they themselves have to wait longer to bring up the data. This is a field where very little progress can be made without comparing results with other people, and data privacy while a valid concern often already can make comparing results with other people difficult. We don't need another reason to isolate testers away from sources of comparative data that might help them.

06-22-2020, 04:40 PM
I believe that the suggestion was that people who already know that they are P312, U106, or L21 don’t join the R1b project. Not that new testers don’t join.

However, I do agree with much of what you said about benefits to people who are stuck at M269.

06-22-2020, 05:02 PM
Also, the slow load time for the data probably goes the other way as well.

If you only know that you are M269+. And have to sift through hundreds of pages that load extremely slowly (most of which are of no use to you) then you will probably be searching for quite a while.

Correction: it’s only 51 pages.

06-22-2020, 05:21 PM
When you don't upgrade your infrastructure for 15-20 years this is what happens.

06-22-2020, 05:23 PM
If you only know that you are M269+. And have to sift through hundreds of pages that load extremely slowly (most of which are of no use to you) then you will probably be searching for quite a while.

Some people are good in figuring out for themselves, most will likely benefit from asking admins to look at their results.

06-22-2020, 05:45 PM
Wouldn't a central hub with links to the various sub-projects serve the purpose while preserving the data and streamlining it?

There might still be an R1b Project for those stuck for the time being at R1b-M269.

06-22-2020, 06:21 PM
My solution to multiple pages is to display 9999 results per page (everything fits on 3 pages for R1b project) and then search by keyword through the list. It works on google chrome

06-22-2020, 06:24 PM
Useless, I once joined with a kit of mine , took me 10minutes to find my kit. Maybe ftdna should restrict theamount of members so administraators are forced to start a sub project, which also means they can give more attention.

Edward J
06-22-2020, 06:42 PM
I can't get some of them to load without crashing (the Scottish Group for example). Also, I have seen some Admins say that they are no longer manually grouping members due to performance issues.