PDA

View Full Version : R_This and That



rms2
08-24-2019, 09:00 PM
Some years ago, probably at least ten, I got into a brief argument with Mike Walsh and Gary Corbett about y-haplogroup shorthand. I said that R-whatever was the best way to go. They said R1b-whatever was better.

Well, I have to confess they were absolutely right. I was wrong. Not only that, I was stupid about it.

The R1b prefix narrows things down significantly and removes a lot of doubt about what it is one is talking about. An "R" leaves a lot to doubt and forces one to look up the haplogroup.

Honestly, back then I favored the "R" because I thought it was important to show how closely related all the R1 haplogroups are. I've since changed my mind.

Sorry guys. You were right right right. I was wrong.

spruithean
08-24-2019, 09:19 PM
Some years ago, probably at least ten, I got into a brief argument with Mike Walsh and Gary Corbett about y-haplogroup shorthand. I said that R-whatever was the best way to go. They said R1b-whatever was better.

Well, I have to confess they were absolutely right. I was wrong. Not only that, I was stupid about it.

The R1b prefix narrows things down significantly and removes a lot of doubt about what it is one is talking about. An "R" leaves a lot to doubt and forces one to look up the haplogroup.

Honestly, back then I favored the "R" because I thought it was important to show how closely related all the R1 haplogroups are. I've since changed my mind.

Sorry guys. You were right right right. I was wrong.

I think this approach may be the best way to go about it. For those who are not familiar with R phylogeny would find it easier to deal with when we see R1b-whatever and R1a-whatever. I know for I-M170 it's nice to see I1-whatever, with perhaps some and I1a-whatever for clarity, so we don't mistake I1b or I1c SNPs for something under DF29.


However the simple R-snp method is still a hell of a lot better than that damn alphabet soup nonsense we still see in these papers. I find myself reading those papers and spending more time looking at the ISOGG tree trying to figure out what's going on. :lol:

TigerMW
08-30-2019, 08:14 PM
Wow, Richard. I forgot this argument but I see your memory is very good.

It grates on me a little that R1b doesn't get its own tab on these trees. Look at the number of branches for the R tree at FTDNA.

R-M207 10,797 (R all)

R-M420 1,404 (R1a)

R-M479 63 (R2)

R-M343 9,326 (R1b)

R1b is the vast majority.


Now here is something you can say I told you so on. There were those who thought L21 was a small subclade. You didn't and established the L21 project among other things.

R-L21 4,091

Admittedly there is a British Isles descendant bias among genetic genealogy testers but L21 has not been shy.

https://www.familytreedna.com/public/y-dna-haplotree/R

Actually, to their credit, YFull has an R1b tab.

Osiris
08-30-2019, 09:08 PM
The curse of naming something before you've even seen a small part of it.

rms2
08-30-2019, 09:17 PM
The curse of naming something before you've even seen a small part of it.

Well, one often has to decide a naming protocol up front, before much is known, and obviously Family Tree DNA has gone with R-whatever.

Like I said, I now prefer R1b- or R1a-whatever. That just makes things easier.

One thing I definitely don't like is the longhand. It's confusing, and it changes every time you turn around.

JMcB
08-30-2019, 09:23 PM
Well, one often has to decide a naming protocol up front, before much is known, and obviously Family Tree DNA has gone with R-whatever.

Like I said, I now prefer R1b- or R1a-whatever. That just makes things easier.

One thing I definitely don't like is the longhand. It's confusing, and it changes every time you turn around.

Yeah, they should drop that like a bad habit! The sooner the better!

TigerMW
08-30-2019, 09:59 PM
.. One thing I definitely don't like is the longhand. It's confusing, and it changes every time you turn around.
I agree with everything you say here but one thing. We still need the long hand. That doesn't mean it needs to be displayed on project Y classic screens or much of anywhere. I think it should be an option, to include that column.

Here is why. We need it for logical sorting of haplotypes. The way it is now within subgroups (or spreadsheets you download) you can't logically sort. The project Y reports sort within subgroup by STR1, STR2, STR3 etc. This puts 12-23-13-... then 12-23-14, 12-23-15, 12-24-13, etc.

If you had the longhand they could sort by the proper branch. I end up downloading the data, checking the haplogroup label agains a previously downloaded version of the haplotree with calculate branch sequence numbers. Then I can sort by branch sequence number.

rms2
08-30-2019, 10:11 PM
I agree with everything you say here but one thing. We still need the long hand . . .

I might have agreed a few years ago, but now it's so long and changes so often, I think it's just confusing, especially when the terminal SNP is omitted.

The longhand is worse than useless now.

dsm
08-30-2019, 11:02 PM
I might have agreed a few years ago, but now it's so long and changes so often, I think it's just confusing, especially when the terminal SNP is omitted.

The longhand is worse than useless now.

The real points re the longhand, in addition to how splits kept changing it, is that different labs/orgs would use their own minor variations.

I agree completely that longhand became so untrustworthy that it was rendered useless. We stopped using it at the S1194 project. The trustworthiness factor was the killer.

Wing Genealogist
08-31-2019, 12:15 AM
By and large everyone agrees NGS/WGS testing (and the resulting SNP tsunami) has basically broken the longhand naming system (as it is currently construed). IMHO, creating a new major clade somewhere along the line of R1b-L23 to R1b-L11/L151 would reduce (but certainly not eliminate) changes made to the longhand name (and basically stabilize the upper levels of both P312 & U106).

I believe it would be easier to convince the academia community to create a new Y-clade than to get them to abandon the longhand clade names. All too many SNPs have a variety of "lab" names (ie U106, S21, M405) so even switching to a SNP based tree is fraught with some level of dis-unity.

Osiris
08-31-2019, 01:27 AM
Well, one often has to decide a naming protocol up front, before much is known, and obviously Family Tree DNA has gone with R-whatever.

Like I said, I now prefer R1b- or R1a-whatever. That just makes things easier.

One thing I definitely don't like is the longhand. It's confusing, and it changes every time you turn around.
I was actually going all the way to when they assigned the 20 letters, we've learned so much since that time. It just seemed so random and could have wound up being almost any mutation if it was lucky enough to be found early on. But true, you have to build off of something and once it's decided it was assuredly built into the very core of the code that businesses like FTDNA use. I've often wondered how things would have changed if they were say fluid as say the taxonomy and the name of species.

And most definitely the sooner longhand is thrown in the dustbin the better.

JoeyP37
08-31-2019, 01:36 AM
I use R1a-YP445 for my subclade; it's just easier, and given my ancestry many people would think I was R1b so I don't want anyone to make an honest mistake, but we have accomplished a lot, sons of Grampa R1, his sons are the Athens and Sparta of the 3rd millennium BC, R1b Bell Beaker traders and explorers as Athens, and R1a steppe warriors as Sparta.

MitchellSince1893
08-31-2019, 02:52 AM
2019 ISOGG stops one SNP block above my current terminal SNP block. I don't think I could ever remember R1b1a1b1a1a2b1c1b3a1 Hell, it take me 7 seconds just to say it at normal talking speed!

And even that mouth full of 20 alphanumerics only gets you to ~1100 BC

rms2
08-31-2019, 12:07 PM
I use R1a-YP445 for my subclade; it's just easier, and given my ancestry many people would think I was R1b so I don't want anyone to make an honest mistake, but we have accomplished a lot, sons of Grampa R1, his sons are the Athens and Sparta of the 3rd millennium BC, R1b Bell Beaker traders and explorers as Athens, and R1a steppe warriors as Sparta.

What?

MikeWhalen
08-31-2019, 03:00 PM
rms2
you gotta scarey memory chum!

M