What's going on with haplogroup A? What is the "everyman" SNP?

[GTC]

Okay, so I'll kick off with a post to this hitherto empty section ...

When I started out on this stuff a some years ago, the ISOGG tree showed M91 as the defining SNP for haplogroup A.

Given that all of men alive today are supposed to have a common male ancestor, I thought that M91 defined that link. Therefore it struck me as odd that my (V1 chip) 23andMe Y data didn't contain M91.

The current version of the ISOGG tree has moved M91 to BT. My 23andMe data for BT shows positives for M299, M42, M94 and SRY1532.1

Haplogroup A shows P305 as the defining SNP, together with a bunch of new V-series SNPs.

Unfortunately there is no "rs" number for P305 in the FTDNA Y Chromosome Browser so I can't see if it's in my 23andMe data.

So, what is the "everyman" SNP -- the one that every man alive today is positive for?

[David]

Neither the current nor the prior trees show any SNP that we are all derived for. In the previous trees, A and BT were peers, with the current tree showing A0 and A1 as peers and BT moved under A1. In neither case is a unifying SNP shown at the "Y" level above A/BT or A0/A1.

Deciding what is ancestral and what is derived at that level can be very difficult. The Cruciani (2011) paper that caused such an upheaval at the root of the tree touches on that to a fair extent. A sequence of 9 "T" nucleotides has always meant Hg BT, and a sequence of 8 "T" nucleotides has always meant YxBT, it is just the interpretation of which state is derived vs ancestral that has changed. Saying that Hg BT was ancestral for M91 worked when YxBT was simply Hg A, but with the additional complexity reported by Cruciani et al., that no longer worked and they realized that Hg BT was the derived state for M91, not the ancestral. I know there were a number of SNPs found during the WTY tests for a couple of Hg A samples that Thomas was having difficulty figuring out which state was ancestral, versus which state was derived.

If one looks at SNPs where essentially all humans share the same value, are we ancestral or derived at that location? If we also share the same value there with chimpanzees and bonobos, is it safe to say we are all ancestral? If there are differences, who is derived and who is ancestral? Interesting questions that I am not in the least qualified to comment on.

Dr Karafet (of the Hammer Lab) has been working on a massive update to the "YCC" haplotree for some time, and rumors were that it would be published sometime late this year. Whenver it comes out, I am going to be very interested as to what she does with the root of the tree.

Regards,
david

[GTC]

Neither the current nor the prior trees show any SNP that we are all derived for. In the previous trees, A and BT were peers, with the current tree showing A0 and A1 as peers and BT moved under A1. In neither case is a unifying SNP shown at the "Y" level above A/BT or A0/A1.

Yes, I guess I mixed up defining with unifying.

Deciding what is ancestral and what is derived at that level can be very difficult. The Cruciani (2011) paper that caused such an upheaval at the root of the tree touches on that to a fair extent. A sequence of 9 "T" nucleotides has always meant Hg BT, and a sequence of 8 "T" nucleotides has always meant YxBT, it is just the interpretation of which state is derived vs ancestral that has changed. Saying that Hg BT was ancestral for M91 worked when YxBT was simply Hg A, but with the additional complexity reported by Cruciani et al., that no longer worked and they realized that Hg BT was the derived state for M91, not the ancestral. I know there were a number of SNPs found during the WTY tests for a couple of Hg A samples that Thomas was having difficulty figuring out which state was ancestral, versus which state was derived.

I don't know what he uses as a reference. Is that done statistically? That is, if the majority of samples have (say) C in the subject position, does that make C ancestral by weight of numbers?

If one looks at SNPs where essentially all humans share the same value, are we ancestral or derived at that location? If we also share the same value there with chimpanzees and bonobos, is it safe to say we are all ancestral? If there are differences, who is derived and who is ancestral? Interesting questions that I am not in the least qualified to comment on.

It seems to me that, for hg A at least, the reference genome itself needs to be based on ancient samples. If we look at a modern chimpanzee, his mutations could be (comparatively) recent, could they not?

Dr Karafet (of the Hammer Lab) has been working on a massive update to the "YCC" haplotree for some time, and rumors were that it would be published sometime late this year. Whenver it comes out, I am going to be very interested as to what she does with the root of the tree.

Recently a number of us have been wondering what happened to the YCC. Previously it had a website -- albeit of a way out of date tree -- but that is now gone and the URL takes you to Hammer Labs home page. Googling around on YCC finds almost nothing much anymore. It's as if somebody has deliberately erased all traces of it.

From your mention of work being undertaken by Dr Karafet, it seems that YCC II may be in the works.

In association with Geno 2.0, a paper by Spencer Wells is expected soon. Perhaps Hammer, Karafet, et al, will be contributors to that?

We live in interesting times.

[David]

...
I don't know what he uses as a reference. Is that done statistically? That is, if the majority of samples have (say) C in the subject position, does that make C ancestral by weight of numbers?
...
It seems to me that, for hg A at least, the reference genome itself needs to be based on ancient samples. If we look at a modern chimpanzee, his mutations could be (comparatively) recent, could they not?

Both valid concerns, in my opinion. If weight of numbers is to be a valid means of selecting the ancestral value, then (to me) that implies a solid representative set of samples, and how does one be certain of achieving that?

I believe some researchers look beyond just chimpanzees for that reason; even looking at more than one set of primates, still seems kind of dicey to me.

Recently a number of us have been wondering what happened to the YCC. Previously it had a website -- albeit of a way out of date tree -- but that is now gone and the URL takes you to Hammer Labs home page. Googling around on YCC finds almost nothing much anymore. It's as if somebody has deliberately erased all traces of it.

From your mention of work being undertaken by Dr Karafet, it seems that YCC II may be in the works.

In association with Geno 2.0, a paper by Spencer Wells is expected soon. Perhaps Hammer, Karafet, et al, will be contributors to that?

We live in interesting times.

Yes, we do live in interesting times, that is perhaps the only certainty.

I personally think YCC is dead; it just is a nice shield to hide behind at times, as FTDNA did with their update in early 2011. No web site, no papers, zilch. The only real interest in standardization that I see seems to be in the hobbyist community.

--david

[GTC]

Also, thanks for the reference to the Cruciani paper. I recall it being discussed when it was published, but didn't have the time to read it ... and then promptly forgot about it. I have also been reading various references in it.

[Ann Turner]

Recently a number of us have been wondering what happened to the YCC. Previously it had a website -- albeit of a way out of date tree -- but that is now gone and the URL takes you to Hammer Labs home page. Googling around on YCC finds almost nothing much anymore. It's as if somebody has deliberately erased all traces of it.

Oh, no! It's been a long time since I referred to the website, but it should be kept for historical purposes. The original YCC effort was an amazing accomplishment, unifying research from many different labs, each with their idiosyncratic nomenclature. I even checked the WayBack machine and got this message:

"Page cannot be crawled or displayed due to robots.txt."

[GTC]

Oh, no! It's been a long time since I referred to the website, but it should be kept for historical purposes. The original YCC effort was an amazing accomplishment, unifying research from many different labs, each with their idiosyncratic nomenclature. I even checked the WayBack machine and got this message:

"Page cannot be crawled or displayed due to robots.txt."

Yep, I tried the Wayback Machine in my search for traces of the YCC.

About a that is left is their 2002 paper to which you refer, viz "A Nomenclature System for the Tree of Human Y-Chromosomal Binary Haplogroups".

Think I'll start a new thread on this subject.