FTDNA flags goodbye to "longhand" haplogroup designations

[MMaddi]

If one's ancestry is western European and he is in y haplogroup R, then the distinction that matters isn't between R1a and R1b but between P312 and U106. That would be more pertinent information. But we don't need to clutter the shorthand with all that. Just look up the listed terminal SNP. If it's your own terminal SNP, surely you will do that much anyway. If it's somebody else's terminal SNP, then it depends on how much you care about it. But seeing an "R1a" or an "R1b" in the prefix isn't going to tell one much, not these days.

I'm not really interested in arguing about which is the better shorthand, R vs. R1a and R1b. Personally, I prefer R1a and R1b for reasons others have stated. If others, like Stevo, prefer R for its brevity, that's fine with me too.

I've bolded the first sentence in Stevo's quoted post above. I've done that because I do have a fundamental disagreement with it. Yes, the dominant divide in R among western Europeans is P312 vs. U106. But it is certainly informative to know if someone with British Isles ancestry is R1a vs. R1b.

R1a is far more common in eastern Europe than western Europe, which is why Stevo wrote what he did. But, precisely because R1a, say in the British Isles, is not nearly as common as P312 or U106 makes it important information to have. If you are R1a with British Isles ancestry you might infer, with some justification, that the deep ancestry of your paternal line is from eastern, not western, Europe. That would imply possible Scandinavian deep ancestry. That would be important information to consider.

[Stevo]

I think you missed my point. The argument against FTDNA's current shorthand is that it is not informative enough, that it needs to signify at a glance whether one is R1b or R1a, as if that were an absolutely essential distinction, one that must be conveyed in the shorthand system. That distinction isn't important enough, in my opinion, to justify the additional clutter in a shorthand system, since the purpose of a shorthand system is to be as brief as possible while conveying the essential information. I think FTDNA's current system does that just fine.

I never said it's not important for westerners to know whether or not they are R1a as opposed to R1b. Of course that's important. It's just not important or essential enough to include in the shorthand system, to the exclusion of other information that is just as, if not more, important, like whether one is P312+ versus U106+.

If a man opens his "myFTDNA" pages and sees that FTDNA's shorthand calls him R-Z284, is he going to stop there, completely baffled, or will he not try to find out what that means? Of course, he will try to find out what that means, and in the course of doing so discover that Z284 is well downstream of R1a, as well as other information that is just as important.

Contrariwise, would a man who opened his myFTDNA pages and saw that he was R1a-Z284 be satisfied with that? Would he also not have to find out what that means? Would he not have to consult the very same sources as the man in the example above, whose shorthand designation was R-Z284? He would. So, he is really not much better off with the three-letter prefix than he was with the single-letter prefix, and if he is someone brand new to genetic genealogy, he is no better off, since he doesn't carry the baggage that makes "R1a" or "R1b" significant, like that distinction seems to those of us who got into this when we didn't have much more to go on.

Here's another thing. As long as FTDNA is using the shorthand system it does, using something else, like "R1b" or "R1a" in the prefix, is a potential source of confusion for newbies: one more thing to have to explain, one more complication.

If the three-letter prefix suddenly prevails among the scientific community and becomes the standard, then I will be happy to switch to it. I know that in this thread it sounds like keeping the single-letter prefix is a big deal to me, but it really is not. I just think it is better than the three-letter prefix for a shorthand system, where the emphasis is on short, as in brief. One letter gets you where you need to go to find out what you need to find out. Three letters do the same thing but with less economy.

[fbirder]

the purpose of a shorthand system is to be as brief as possible while conveying the essential information. I think FTDNA's current system does that just fine.

Nope. To be as brief as possible they should scrap all the letter designations. After all, L1280 describes my haplogroup just as accurately as R-L1280.

[Lklundin]

Personally, I prefer

This is largely what this is about.

So the best solution for everybody would be if FTDNA would extend the 'myFTDNA - Account Settings' with a user preference for showing (everywhere!) the Y-DNA haplogroup using one of:
1) long notation (incl. version information of the tree),
2) short hand,
3) a combination of both 1 and 2,
4) extended short hand (e.g. R1b-SNP), and
5) which ever FTDNA thinks the users might like (e.g. x characters of the long notation dash the SNP)

This is what software usually does for other preferences such as metric/imperial or time-formats.

[Stevo]

Nope. To be as brief as possible they should scrap all the letter designations. After all, L1280 describes my haplogroup just as accurately as R-L1280.

Here is why you are wrong. The phylogenetic tree is massive and so is broken up into major branches called haplogroups. Those major branch haplogroups have letter designations. Currently ISOGG's phylogenetic trees follow those letter designations. As far as I know, ISOGG is the only game in town when it comes to an easily accessible y-dna phylogenetic tree and Y-DNA SNP Index.

Another thing is that SNPs are SNPs and not y haplogroups or subclades unto themselves. They characterize haplogroups and subclades, but they are not the same things. One of the reasons for this is that sometimes the very same SNP will arise in more than one y haplogroup and subclade. What makes the difference is the entire panoply of SNP descent. In other words, an R-U106 isn't R-U106 merely because he is U106+ but because he has the full pedigree of R SNPs leading up to and including U106, from M207 on down.

That is what makes single SNP testing risky in some instances. It's generally okay, however, if one's haplotype (STRs) supports the single SNP test result, but it's better to get at least a few SNPs tested that are upstream to confirm the pedigree.

So, the R prefix (or whatever single-letter prefix applies) is necessary because it puts one in the right phylogenetic tree, which is a y-dna pedigree. Think of R as the patriarch whose first name was "R" but whose surname is that long string of SNPs leading up to and including M207. The surname keeps getting longer and longer in each individual case but for us ultimately traces back to what it was at the M207 root.

Having that R prefix insures the terminal SNP we're looking for is the one in the R family, with the right pedigree.

[Stevo]

This is largely what this is about.

So the best solution for everybody would be if FTDNA would extend the 'myFTDNA - Account Settings' with a user preference for showing (everywhere!) the Y-DNA haplogroup using one of:
1) long notation (incl. version information of the tree),
2) short hand,
3) a combination of both 1 and 2,
4) extended short hand (e.g. R1b-SNP), and
5) which ever FTDNA thinks the users might like (e.g. x characters of the long notation dash the SNP)

This is what software usually does for other preferences such as metric/imperial or time-formats.

No offense, and I appreciate having choices as much as anyone, but that strikes me as an invitation to chaos. Look at Ysearch, which is chock full of user errors when it comes to y haplogroup, subclade, SNP testing, map coordinates, you name it. It's a mine field.

FTDNA's current shorthand system is based on a scientific paper produced by the YCC (Y Chromosome Consortium) in 2002:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC155271/

I think we're better off sticking with it until the scientific community comes together and reaches some sort of consensus on a better system and FTDNA adopts that.

Like I said a few posts back, we've been using the current shorthand system at least since early 2008, and nobody made a big deal about it. It has only become a matter of minor (and thus far insignificant) controversy recently.

Take for example the R-L21 Plus Project, which I currently administer. Its title employs FTDNA's shorthand. Before anyone leaps to the conclusion that of course it does, because I named it, consider that I did not name that project; it was named by the guy who created it in late October of 2008. I inherited the project and did not alter the name of it. When P312 was discovered in early 2008, Rick Arnold and I created the R-P312 and Subclades Project. No one at that time criticized the name of the project or so much as raised a peep about it. There were absolutely no suggestions that we call it the "R1b-P312 and Subclades Project". None. Not one.

The idea that a three-character prefix is an absolute must represents a recent innovation.

[Wing_Genealogis]

One issue with simply using R-SNP is the fact Haplogroup R has gotten so large, there are a number of multiple SNP occurrences within it.

A perfect example is my clade, R-M157. It is found in both R1a as well as R1b. There is even a SNP (L217) which is found in two separate subclades of R1b-U106.

Granted, we can always use M157.2 (in my case) to signify this is the R1b SNP, rather than the R1a SNP, but it also points to the need to break down the ridiculously large R haplogroup (and not simply make R1b a new haplogroup, but move further down the tree).

[MMaddi]

Another thing is that SNPs are SNPs and not y haplogroups or subclades unto themselves. They characterize haplogroups and subclades, but they are not the same things. One of the reasons for this is that sometimes the very same SNP will arise in more than one y haplogroup and subclade. What makes the difference is the entire panoply of SNP descent. In other words, an R-U106 isn't R-U106 merely because he is U106+ but because he has the full pedigree of R SNPs leading up to and including U106, from M207 on down....

So, the R prefix (or whatever single-letter prefix applies) is necessary because it puts one in the right phylogenetic tree, which is a y-dna pedigree. Think of R as the patriarch whose first name was "R" but whose surname is that long string of SNPs leading up to and including M207. The surname keeps getting longer and longer in each individual case but for us ultimately traces back to what it was at the M207 root.

This is slightly off-topic, but I like to extend Stevo's analogy above and regard the haplogroup tree in the same way as a genealogist regards a family tree.

A genealogist will accumulate a paper trail of birth certificates to put together a family tree. The birth certificate will give the name of the father and mother and links one generation in the tree to the next.

This is the same thing a haplogroup tree provides - a line of descent. The difference in a haplogroup tree is that the haplogroup is uniparental - only one parent passes the SNPs to the next generation. Each SNP is like a birth certificate of the next subclade down, providing the name (haplogroup/subclade and SNP) of the "father."

Given, as Stevo explained above, that the same SNP can occur in different haplogroups, it's the package or descendancy of SNPs that gives you the context to know in which haplogroup the particular version of the SNP occurs.

[Stevo]

One issue with simply using R-SNP is the fact Haplogroup R has gotten so large, there are a number of multiple SNP occurrences within it.

A perfect example is my clade, R-M157. It is found in both R1a as well as R1b. There is even a SNP (L217) which is found in two separate subclades of R1b-U106.

Granted, we can always use M157.2 (in my case) to signify this is the R1b SNP, rather than the R1a SNP, but it also points to the need to break down the ridiculously large R haplogroup (and not simply make R1b a new haplogroup, but move further down the tree).

Right, and I believe there are other instances besides L217 of SNPs that have arisen in separate R1b subclades, so having a three-character prefix won't correct that problem. The only solution to such anomalies is separate names for the offending (pardon me for the expression) SNP, like M157.1, M157.2, etc.

[Stevo]

This is slightly off-topic, but I like to extend Stevo's analogy above and regard the haplogroup tree in the same way as a genealogist regards a family tree.

A genealogist will accumulate a paper trail of birth certificates to put together a family tree. The birth certificate will give the name of the father and mother and links one generation in the tree to the next.

This is the same thing a haplogroup tree provides - a line of descent. The difference in a haplogroup tree is that the haplogroup is uniparental - only one parent passes the SNPs to the next generation. Each SNP is like a birth certificate of the next subclade down, providing the name (haplogroup/subclade and SNP) of the "father."

Given, as Stevo explained above, that the same SNP can occur in different haplogroups, it's the package or descendancy of SNPs that gives you the context to know in which haplogroup the particular version of the SNP occurs.

Amen. Exactly right. That is why a y haplogroup is more than any single SNP. It's the whole blessed y pedigree leading up to and including that SNP.