PDA

View Full Version : Structure of Ydna December 2014



Sile
06-12-14, 12:00
new paper

http://mbe.oxfordjournals.org/content/early/2014/11/26/molbev.msu327.full.pdf+html

As for my T-Ydna ..the oldest basal for T found is in Bhutan :startled:



see page 26 for ages of Y

Angela
06-12-14, 19:54
Some conclusions from the paper:

*They give a TMRCA for the whole tree of 129 KYA , and a 46 KYA date for the DR Colonization of Eurasia node.

*"Hg E1b1a (here given a TMRCA of 6.9 KYA) has previously been associated with the expansion of Bantu languages, which spread widely from Central Africa ~3KYA together with farming and iron-working."

*"Haplogroups I and J divide at 31 KYA, and each then divides in two at similar times of 21– 23 KYA. Within hg I1 is a striking star-like genealogy dating to 3.5 KYA."

*As to y Dna "Q":
"Here as well as an example of the common native American Q,
native American Q-M3 lineage,we included rare examples of European Q chromosomes. One of the English chromosomes belongs to the deepest rooting
lineage within Q (Q M-378) and may reflect the Jewish diaspora (Hammer et al 2009); the other is distantly related, shares a deep node with the Mexican 0M3 chromosome and has an STR haplotype closely related to those of scarce Scandinavian hg Q chromosomes (unpublished data)."

*The TMRCA of hg R is19KYA, and within it bothhgs R1a and R1b compriseyoung, star-like expansionsdiscussed extensively elsewhere(Batini et al. submitted)"

*"The addition of Central Asian chromosomes here contributes a sequence to the deepest subclade of R1b-M269, while another, in a Bhutanese individual, forms an outgroup almost as old as the R1a/R1b split."


*"generally, the STRs perform poorly, giving a wide variety of TMRCAs for nodes with similar SNP-based dates, and correlation coefficient sconsistently below 0.6.

The evolutionary STR mutation rate consistently overestimates, and the pedigree rate underestimates, the TMRCAs of nodes (Figure 4a).

As expected, the pedigree mutation rate performs better for young nodes (<10 KYA; Table S6.while the evolutionary rate performs better for older nodes."

Also,

"This probably reflects the increasing importance of back-mutation in older clades.Despite the diminishingcost of NGS, it seems likely that researchers will wish to continue
to use STRs in dating; in order to provide a rational framework, careful analysis of large datasets comprising multiple STRs and MSY sequenceswill be needed. The citizen-scientist’ community, which now generates 111-locus STR haplotypes combined with ~10-Mb MSY NGS data, may be best placed to do this."

*"these findings support the idea that purifying selection is acting on single-copy MSY genes....
purifying selection is ongoing, and that past claims of terminal degeneration of the Y chromosome are exaggerated...while evidence is mounting that purifying selection is acting on MSY protein-coding genes, more work is required to understand their functional roles. Candidate genes are currently lacking for some established MSY-linked phenotypessuch as HIV-AIDS progression (Sezgin et al. 2009) and coronary artery disease susceptibility(Charchar et al. 2012), and there is a clear need to
understand the roles of non-coding RNA genes on the MSY, as well as the suite of protein-coding genes."

Ed. My OCD was acting up, so I cleaned up the format a bit.

Angela
06-12-14, 21:17
I haven't gone through the supplement, and probably won't have time today, but a few thoughts...

I'm sure all the gentlemen will be relieved to hear that the past claims of terminal degeneration of the Y chromosome are exaggerated...:smile: I am glad that there is continuing research into the relationship between the different y lineages and various disease susceptibilities. I have wondered more than once whether this might have something to do with the frequency of certain lineages in certain locales. While coronary artery disease has such a late onset that its effect on procreation would be negligible, susceptibility to certain infections might not. In a town near me, church records indicate that 90% of the population died at the height of the Black Death. Might there have been a difference in survival rates based on uniparental markers? I'd also be interested to know if they are looking into the possibility that certain male lineages might have a slightly higher rate of producing males instead of females.

I think we knew most of this with regard to yDNA STRs. I wonder if Gioiello is still interested in this topic? He was talking about back mutations in older clades years ago.

The very old R1b in Bhutan is certainly interesting, as is the data for "Q". Attaching "race" or even "ethnicity" to y Dna lineages should be one of those ideas that go straight to the trash.

The extreme "youth" of modern y Dna subclades world-wide, and the fact that it is tied to the massive expansions after the adoption of agriculture continues to astound.

As to the TMRCAs, I'm not sure. Maybe I'll have a clearer idea after I read the supplement.

Aberdeen
06-12-14, 22:06
If the TMRCA for Y haplotype R is less than 20,000 years, does this mean the dating of Mal'ta Boy is off? As for those young star-like clusters of R1a and R1b expansions, how young are we talking about? We do have examples from Europe that are nearly 5000 years old and appear to be a long way from where R1a and R1b are assumed to have evolved. I don't really understand the science of estimating DNA dates but I think we should be concerned when estimates appear to conflict with known dates.

Sile
06-12-14, 22:27
If the TMRCA for Y haplotype R is less than 20,000 years, does this mean the dating of Mal'ta Boy is off? As for those young star-like clusters of R1a and R1b expansions, how young are we talking about? We do have examples from Europe that are nearly 5000 years old and appear to be a long way from where R1a and R1b are assumed to have evolved. I don't really understand the science of estimating DNA dates but I think we should be concerned when estimates appear to conflict with known dates.

Mal'ta boy is only R from a branch that split once R split from Q..........as I said before, he is not R1 or R2.
because he has no living "relative" he was ignored in this paper, but they knew about him.
They took the oldest living R which is R-M207.
So, the conclusion is that mal'ta split the same time as P1-M45 emerged which is ~24,300 years ago

bicicleur
07-12-14, 00:04
Mal'ta boy is only R from a branch that split once R split from Q..........as I said before, he is not R1 or R2.
because he has no living "relative" he was ignored in this paper, but they knew about him.
They took the oldest living R which is R-M207.
So, the conclusion is that mal'ta split the same time as P1-M45 emerged which is ~24,300 years ago

You may get away with that, but what about M/LT/NO/QR-M9 TRMCA 32.6 KYA?
Ust'-Ishim is 45 KYA and was K-M9

I estimate you should add some 50 % to the figures in the list. It would make much more sense.

Aberdeen
07-12-14, 00:38
Mal'ta boy is only R from a branch that split once R split from Q..........as I said before, he is not R1 or R2.
because he has no living "relative" he was ignored in this paper, but they knew about him.
They took the oldest living R which is R-M207.
So, the conclusion is that mal'ta split the same time as P1-M45 emerged which is ~24,300 years ago

No. The most common estimate for P1-M45 seems to be 35,000 years ago and for R-M207 24,000 - 35,000 years ago. Mal'ta Boy was R*. And of course he didn't have any living descendants - he died before reaching adolescence. But his first R-M207 ancestor did. Otherwise there would not have been any R1 or R2 people. But none of the oldest R-M207 people are still alive. If they were, they would be very old and would need mammoth ivory walkers in order to get around.

Sile
07-12-14, 01:10
No. The most common estimate for P1-M45 seems to be 35,000 years ago and for R-M207 24,000 - 35,000 years ago. Mal'ta Boy was R*. And of course he didn't have any living descendants - he died before reaching adolescence. But his first R-M207 ancestor did. Otherwise there would not have been any R1 or R2 people. But none of the oldest R-M207 people are still alive. If they were, they would be very old and would need mammoth ivory walkers in order to get around.

What are you talking about, mal'ta does not have M207

https://drive.google.com/folderview?id=0B7vzRsRM2aOQd3ZSVWdNOFB1WHM&usp=sharing#list

Sile
07-12-14, 01:17
You may get away with that, but what about M/LT/NO/QR-M9 TRMCA 32.6 KYA?
Ust'-Ishim is 45 KYA and was K-M9

I estimate you should add some 50 % to the figures in the list. It would make much more sense.

paper states
these are the oldest markers

(ii) Ancient population expansion: Within clade DR of the tree lies a deep Paleolithic lineage radiation, giving rise to haplogroups G, HF5, IJ, LT, NO and P, dating to between 23 and 33 KYA (we note that a single variant identified elsewhere (Poznik et al. 2013) resolves the polytomy of haplogroups G and H, with G branching earlier).


M ans S and R and Q come after P

you need to check the order that haplogroups are formed

Ust was K-M526

bicicleur
07-12-14, 11:10
paper states
these are the oldest markers

(ii) Ancient population expansion: Within clade DR of the tree lies a deep Paleolithic lineage radiation, giving rise to haplogroups G, HF5, IJ, LT, NO and P, dating to between 23 and 33 KYA (we note that a single variant identified elsewhere (Poznik et al. 2013) resolves the polytomy of haplogroups G and H, with G branching earlier).


M ans S and R and Q come after P

you need to check the order that haplogroups are formed

Ust was K-M526

Usht'-Ishim was K-M526 + 2 out of 7 SNPs for haplo X (pre-NO)
This was after split from MP and after split from LT
MP is ancestral to P and hence also to Q and R

Aberdeen
07-12-14, 16:14
What are you talking about, mal'ta does not have M207

https://drive.google.com/folderview?id=0B7vzRsRM2aOQd3ZSVWdNOFB1WHM&usp=sharing#list

Mal'ta Boy had Y haplotype R - see for example Dienekes' Blog of November 20, 2013. Elsewhere it's stated to be specifically R* - not R1 or R2 but still R.

Sile
07-12-14, 18:54
Mal'ta Boy had Y haplotype R - see for example Dienekes' Blog of November 20, 2013. Elsewhere it's stated to be specifically R* - not R1 or R2 but still R.

R* = dead line

in genetics an asterix means no more SNP found, no more branches........dead end..............I stated this many many months ago. Dead line means no ydna relatives

Aberdeen
07-12-14, 20:06
R* = dead line

in genetics an asterix means no more SNP found, no more branches........dead end..............I stated this many many months ago. Dead line means no ydna relatives

The fact that a particular branch of R didn't produce any direct descendants doesn't make it not R. And, given how few people were around back then, the man who was the paternal ancestor of R1 and R2 was probably about a third cousin to Mal'ta Boy.

bicicleur
08-12-14, 11:16
You may get away with that, but what about M/LT/NO/QR-M9 TRMCA 32.6 KYA?
Ust'-Ishim is 45 KYA and was K-M9

I estimate you should add some 50 % to the figures in the list. It would make much more sense.

Here, we favoured a
rate (1.0 x 10-9/bp/year) estimated directly from NGS analysis of MSY sequences in a deeprooting
pedigree (Xue et al. 2009). Though the direct nature of the analysis and the proven
transmission of newly arising variants are positive features, the study’s major disadvantage is
that its mutation rate rests on only four observations. These numbers will improve as other
resequencing studies are published, but meanwhile other studies (Mendez et al. 2013;
Scozzari et al. 2014) have taken the genome-wide de novo mutation rate (based on a larger
number of observations) and scaled it to account for male-specific transmission, thus inferring
slower rates of 0.62 x 10-9 (Mendez et al. 2013) or 0.64 x 10-9 (Scozzari et al. 2014). Criticism
of this approach (Elhaik et al. 2014) has been based on its indirect nature, and the fact that the
resulting rates are at odds with phylogenetic mutation rate estimates (1.5 – 2.1 x 10-9 /bp/year
(Skaletsky et al. 2003; Kuroki et al. 2006)) based on human-chimpanzee MSY comparisons.
Calibration based on archaeological dates and assumptions about colonisation history (such as
the peopling of the Americas (Poznik et al. 2013) or of Sardinia (Francalacci et al. 2013))
has also been applied, although it introduces other sources of uncertainty. Further analysis of
deep-rooting pedigrees, combined with accumulating data on well-dated ancient DNA, should
help to give more reliable mutation rate estimates in the near future.

I would favour the slower rates of 0.62 or 0.64

sparkey
08-12-14, 18:53
What's wrong with this study's pedigree-based STR estimates? It's giving age estimates pretty consistently 2x too high. And given their Table 2 estimates being pretty consistent with estimates we've seen from other studies that use pedigree-based STR estimates, that seems to mean that their pedigree-based estimates are being scaled weirdly. Their figure 4a shows pretty tight fits around lines that show estimates 2x too high. Seems that they're putting a lot of stock into the 5 or so sub-10k YBP estimates that cluster to the right of the line. Are they using a particular scale?

Sile
08-12-14, 19:36
The fact that a particular branch of R didn't produce any direct descendants doesn't make it not R. And, given how few people were around back then, the man who was the paternal ancestor of R1 and R2 was probably about a third cousin to Mal'ta Boy.

I never said he was not R , I said he did not have R-M207 as you stated he did.

Sile
08-12-14, 19:41
What's wrong with this study's pedigree-based STR estimates? It's giving age estimates pretty consistently 2x too high. And given their Table 2 estimates being pretty consistent with estimates we've seen from other studies that use pedigree-based STR estimates, that seems to mean that their pedigree-based estimates are being scaled weirdly. Their figure 4a shows pretty tight fits around lines that show estimates 2x too high. Seems that they're putting a lot of stock into the 5 or so sub-10k YBP estimates that cluster to the right of the line. Are they using a particular scale?

They are not using dead lines, only using people who are living and they are using these as finding how old these tested peoples markers are

ages shown and compared with others in table 6

http://mbe.oxfordjournals.org/content/suppl/2014/11/26/msu327.DC1/Hallast_TablesS1-3_5-11Revised281014.xls

Aberdeen
08-12-14, 20:54
I never said he was not R , I said he did not have R-M207 as you stated he did.

I don't know a lot about genetics so you'll have to explain that to me. Does ISOGG have it wrong?

www.isogg.org/tree/ISOGG_HapgrpR.html

Sile
08-12-14, 21:56
I don't know a lot about genetics so you'll have to explain that to me. Does ISOGG have it wrong?

www.isogg.org/tree/ISOGG_HapgrpR.html (http://www.isogg.org/tree/ISOGG_HapgrpR.html)

Mal'ta is R* , which means an extinct line..........he did not have M207 ..........is it hard to understand that his line is non existant :bored:

he fits closest to current R haplogroup........most likely they did not want to asscoiate a new alpha marker to him

Aberdeen
08-12-14, 23:19
Mal'ta is R* , which means an extinct line..........he did not have M207 ..........is it hard to understand that his line is non existant :bored:

he fits closest to current R haplogroup........most likely they did not want to asscoiate a new alpha marker to him

Oh, so ISOGG isn't wrong, just really sloppy in showing R* under R-M207. But you haven't explained why you think that. And please - no more nonsense about R* not being R because R* is an extinct line, because that idea is just sile. If you think Mal'ta Boy was actually somewhere between P and R, please explain the science of it. I'd really like to understand why you think ISOGG was either mistaken or sloppy in saying R=M207 and that R* is a subset of M207. If there's a reason why you don't agree, you should be able to explain

Sile
09-12-14, 02:38
Oh, so ISOGG isn't wrong, just really sloppy in showing R* under R-M207. But you haven't explained why you think that. And please - no more nonsense about R* not being R because R* is an extinct line, because that idea is just sile. If you think Mal'ta Boy was actually somewhere between P and R, please explain the science of it. I'd really like to understand why you think ISOGG was either mistaken or sloppy in saying R=M207 and that R* is a subset of M207. If there's a reason why you don't agree, you should be able to explain

read maciano paper in this forum

http://dienekes.blogspot.be/2013/11/a-priori-y-chromosome-phylogeny-from.html

mal'ta is reclassified by some from R* to MA-1

oriental
09-12-14, 03:05
I am sure as Genetics is at its infancy there will be modifications to classes of Haplogroups as vanished groups are hypothesized and placed in between or ahead of the "known and present Haplogroups". There are gaps in the migrations and many groups vanished when the seas rose 300 feet. Of course, it was the closing of the North and South American continents that created the Gulf Stream which is probably the main cause of ending the Ice Age. The rise of the seas was gradual taking 23,000 years as was the continental movements 2 inches a year for a continent.

Aberdeen
09-12-14, 05:36
read maciano paper in this forum

http://dienekes.blogspot.be/2013/11/a-priori-y-chromosome-phylogeny-from.html

mal'ta is reclassified by some from R* to MA-1

I'd gladly read Maciamo's paper if you tell me where to find it - no doubt that would finally provide some clarity. But the article by Dienekes that you linked to says that scientists have discovered that there was a Y haplotype MP that lead to M and P (which led to Q and R) and it says that Mal'ta Boy was R. Scientists refer to him as MA-1 rather than Mal'ta Boy and MA-2 refers to the other set of remains found in the area that was less well preserved. These are not references to haplotypes. Is that what you've been on about all this time? LOL!

Sile
03-01-15, 19:52
Usht'-Ishim was K-M526 + 2 out of 7 SNPs for haplo X (pre-NO)
This was after split from MP and after split from LT
MP is ancestral to P and hence also to Q and R

LT split off , from K , way before MPS was created........it was not the same time.

Sile
03-01-15, 19:57
Here, we favoured a
rate (1.0 x 10-9/bp/year) estimated directly from NGS analysis of MSY sequences in a deeprooting
pedigree (Xue et al. 2009). Though the direct nature of the analysis and the proven
transmission of newly arising variants are positive features, the study’s major disadvantage is
that its mutation rate rests on only four observations. These numbers will improve as other
resequencing studies are published, but meanwhile other studies (Mendez et al. 2013;
Scozzari et al. 2014) have taken the genome-wide de novo mutation rate (based on a larger
number of observations) and scaled it to account for male-specific transmission, thus inferring
slower rates of 0.62 x 10-9 (Mendez et al. 2013) or 0.64 x 10-9 (Scozzari et al. 2014). Criticism
of this approach (Elhaik et al. 2014) has been based on its indirect nature, and the fact that the
resulting rates are at odds with phylogenetic mutation rate estimates (1.5 – 2.1 x 10-9 /bp/year
(Skaletsky et al. 2003; Kuroki et al. 2006)) based on human-chimpanzee MSY comparisons.
Calibration based on archaeological dates and assumptions about colonisation history (such as
the peopling of the Americas (Poznik et al. 2013) or of Sardinia (Francalacci et al. 2013))
has also been applied, although it introduces other sources of uncertainty. Further analysis of
deep-rooting pedigrees, combined with accumulating data on well-dated ancient DNA, should
help to give more reliable mutation rate estimates in the near future.

I would favour the slower rates of 0.62 or 0.64

when KLT split apart, LT became P326 and K became known as M526 ..................they split of the same time.

the LT branch was reclassified as K1 and the K-M526 was reclassified as K2