Origin and Diffusion of mtDNA Haplogroup X

[Doug Welle]

http://www.dienekes.com/blog/archives/000402.htmlOrigin and
Diffusion of mtDNA Haplogroup X

Am. J. Hum. Genet., 73:000, 2003

Maere Reidla et al.

I think the article is still in copyright so I won't reproduce
it here - it suggests that Haplogroup X in Native Americans
comes from the Near East.

Doug
--
Doug Weller -- exorcise the demon to reply Doug & Helen's Dogs
http://www.dougandhelen.com Doug's Archaeology Site:
http://www.ramtops.co.uk

[Mb]

Doug Weller <dweller@ramtops.demon.co.uk> wrote:

> it suggests that Haplogroup X in Native Americans comes from
> the Near East.

..as do all X2 haplogroups of which the native american
halpogroup X2a is a subset.

Michael

[Gisele Hor]

On Mon, 03 Nov 2003 17:39:42 +0000, Doug Weller
<dweller@ramtops.demon.co.uk> wrote:

>http://www.dienekes.com/blog/archives/000402.htmlOrigin and
>Diffusion of mtDNA Haplogroup X
>
>Am. J. Hum. Genet., 73:000, 2003
>
>Maere Reidla et al.
>
>I think the article is still in copyright so I won't
>reproduce it here - it suggests that Haplogroup X in Native
>Americans comes from the Near East.

All of the sequences which begin with the letter 'H' in Figure
1 were from Herrnstadt et al's study in which, we were only
told, were derived from Europeans or Americans. I can not
think of a worse haplogroup to be uncertain about the
geographic sources of the samples.

It should also be mentioned that Reidla et al. believe
the Altaians only acquired haplogroup X within the last
6,700 years.

Gisele

[Doug Welle]

On Mon, 03 Nov 2003 20:48:34 GMT, in sci.archaeology, Gisele
Horvat wrote:

>On Mon, 03 Nov 2003 17:39:42 +0000, Doug Weller
><dweller@ramtops.demon.co.uk> wrote:
>
>>http://www.dienekes.com/blog/archives/000402.htmlOrigin and
>>Diffusion of mtDNA Haplogroup X
>>
>>Am. J. Hum. Genet., 73:000, 2003
>>
>>Maere Reidla et al.
>>
>>I think the article is still in copyright so I won't
>>reproduce it here - it suggests that Haplogroup X in Native
>>Americans comes from the Near East.
>
>All of the sequences which begin with the letter 'H' in
>Figure 1 were from Herrnstadt et al's study in which, we were
>only told, were derived from Europeans or Americans. I can
>not think of a worse haplogroup to be uncertain about the
>geographic sources of the samples.

I'm not quite clear what you mean. Do you mean we don't know
the actual sources of the samples?

>It should also be mentioned that Reidla et al. believe
>the Altaians only acquired haplogroup X within the last
>6,700 years.

What are the implications of this?

Do I gather that one shouldn't think that this article is the
last word?
:-)

Thanks.

Doug

--
Doug Weller -- exorcise the demon to reply Doug & Helen's Dogs
http://www.dougandhelen.com Doug's Archaeology Site:
http://www.ramtops.co.uk

[Philip Dei]

On Mon, 03 Nov 2003 21:37:09 +0000, Doug Weller
<dweller@ramtops.demon.co.uk> did some sarious thank'n and
scribbled:

>>It should also be mentioned that Reidla et al. believe
>>the Altaians only acquired haplogroup X within the last
>>6,700 years.
>
>What are the implications of this?
>
>Do I gather that one shouldn't think that this article is the
>last word?
>:-)

This is probably diffusive backflow. Again this is
speculation based on HLA, but to me it looks as if when the
motion of genetic material began from WEA region into the
middle east it did appear to pick up genetic material from
the middle east and carry that material eastward. One of the
key HLA is the A28 (A*6801) which appears to have come
recently from africa. This HLA drops markedly in frequency
from Japan into SE asia and is completely absent in many
populations. It is substantially higher in gene frequency in
the Ainu versus the Japanese and pockets of this allele are
found in siberia at varying frequencies, also present in the
New world population in combination with New World specific
haplotypes. The path of this migration has been hard to
deduce, at present certain groups in pakistan would appear
to mark the edge of the migration per say and the caucasus
marks the other edge. Because to the rarity of the allele I
use other haps that are frequently found with this allele to
help track it. Going up and around into the biakal region
the tibetians mark another edge and the northern han to the
south, as one heads east this appears to break down as
streaks of the WEA/ME haps flow into se asia (confirmed by
other studies of mongolian genes). The oddity within the
data set is that the region between kazakhstan and
czechosolovakia on up into sweden is particularly devoid of
the haplotypes that make it appear that these peoples have a
certain degree of isolation relative to rather large
migrations across the whole of eurasia, this is particularly
strange given the recent history. If my estimation is
correct the ME component haps and alleles found in native
americans occurred 18 to 20 kya, and redistributions within
central north asia occurred after the last iceage. It is
understandable to me why the swedes and polish have lower
levels of these haplotypes, because swedes appear to have
been largely a more or less derivative population of the
islandic population of NW europe; therefore they would be on
the other side of the regional diffusion line that formed
before or as these groups left WEA; however, the absence in
the populations north of this region gives some possible
indications that during the LGM that peoples in europe
vacated along a southern route, possibly crossing over into
north africa as they headed eastward.

To give an example of the HLA and the connects.
1. I found strong association between belgium and armenian
(reason unknown, path unknown).
2. strong direct association between the basque and the czech
and the svanetians (reason unknown, path unknown).
3. Other studies have shown a connection between the
bulgarians, italians, turkic and kurdish, and jews. In many
instances it looked as gene flow was from west to east.
4. The omani looks as if they have haplotypes either directe
africa (west) or from western europe (not discounting
haplotypes shared with india)

Whereas the north asians group, discounting the affects of the
vikings, appear to have many haplotypes shared with middle
eastern and indian, and these have older and less easily
established connections with africa. These haplotypes have
some similarity with those found in europe at low and
scattered (undergone recombination) frequencies.

Therefore my conclusions is that humans settled into eurasia,
probably between india and the caucasus and moved bilaterally
both north and south (archaeologically suggest also west and
the A3 Cw7 B7 haplotype is in agreement. Subsequent waves from
africa and europe replaced the population in the middle
eastern region and one of these waves combined, and then
migrated NE into the biakal region heading west. Over time the
peoples who lived along this route diffused outward into
adjacent groups. If these people moved up 34 kya in accord
with the archeaology, the the prime period of african
readdition would be between 34 and 18 kya in the middle
eastern region (cautioning that prime does not exclude recent
addition, notable in the Basque with A29 Cw16 B44 and in the
greeks with a number of recent numbian allelotypes). Also, I
should add that there appears to be a genetic connect between
the sardinians and the romanians that reaches the finnic
population, more than likely a source of admixture. Sardianian
haplotypes do appear in europe, and may have been an early
arrival admixing with the central asian arrivals and later
pushed by western immigrants from africa back eastward. So
there are several wavefront potentials that need to be
considered. There are several papers now reporting as I
previously did that sardinian population appears to be older
and have a separate population history (relative to africa)
relative to the rest of europe. Thus there is tremendous
amount of complexity with the settling of europe. In fact
there is little in the way of easily understood zones in all
of eurasia. Of those the more easily understood are the
Irish/Cornish and the ryukuan in terms of places that were
settled and not later overrun with tremendous levels of gene
flow from another source. Potentially the saami and the lapps
might represent the edge of the uralic gradient and this would
create another anchor point.

Note also Gisele's critique, and also that Herrstats worked
has been strongly critiqued in the literature. Expect there
to be further reanalysis of these.

Gone are the days of people flat out saying there were
singular waves of migration from africa from a single
direction. We can only surmise why they said these types of
things to begin with, given the data of even their
publication cycle was in disaggrement and the coming times
of molecular evolution will be people, who UPON realizing
their work is under public scrutiny int the Public trenches,
take note sit down and SERIOUSLY consider their data in
light of everything else that has been published including
other peoples data and data from other genetic studies.
Gisele was one of the first to point out that mtDNAs had
been eliminated from study in new world because of
post-columbian contamination even though more similar haps
were found in asia and elsewhere outside of europe. The
whole issue of X has been undeniably critical of the way
native american genetic studies have been done, because it
turns out people willy-nilly excluded sequences they did not
beleive where native american even though genetic studies in
siberia and asia were not sufficient to rule out potential
ancestral-descendant relationships.

[Gisele Hor]

On Mon, 03 Nov 2003 21:37:09 +0000, Doug Weller
<dweller@ramtops.demon.co.uk> wrote:

>On Mon, 03 Nov 2003 20:48:34 GMT, in sci.archaeology, Gisele
>Horvat wrote:
>
>>On Mon, 03 Nov 2003 17:39:42 +0000, Doug Weller
>><dweller@ramtops.demon.co.uk> wrote:
>>
>>>http://www.dienekes.com/blog/archives/000402.htmlOrigin and
>>>Diffusion of mtDNA Haplogroup X
>>>
>>>Am. J. Hum. Genet., 73:000, 2003
>>>
>>>Maere Reidla et al.
>>>
>>>I think the article is still in copyright so I won't
>>>reproduce it here - it suggests that Haplogroup X in Native
>>>Americans comes from the Near East.
>>
>>All of the sequences which begin with the letter 'H' in
>>Figure 1 were from Herrnstadt et al's study in which, we
>>were only told, were derived from Europeans or Americans. I
>>can not think of a worse haplogroup to be uncertain about
>>the geographic sources of the samples.
>
>I'm not quite clear what you mean. Do you mean we don't know
>the actual sources of the samples?

Maybe Herrnstadt et al. knows but, apparently, they did
not think it necessary to make that information available
to others:

"Blood samples and frozen brain samples were collected from
maternally unrelated individuals (as determined from
family-history information) in the United States and the
United Kingdom after institutional review board approval and
informed consent. We have here included the 64 European and 2
African mtDNA sequences that were previously analyzed for
recombination (Elson et al. 2001). The U.S. and U.K. samples
were from the general population, and the individuals included
normal controls, patients with type 2 diabetes, and patients
with neurodegenerative disorders (e.g., Alzheimer disease [AD]
or Parkinson disease)." (Herrnstadt et al. 2002)

That's all we're told about the source of the 560 sequences
they obtained. This is unfortunate because this is the largest
study of greater portion of the complete mtDNA genome
performed thus far. I've scanned Reidla's article a couple
times now to find the part which states how it is known that
all of the haplogroup X sequences in Herrnstadt's study were
of European origin but can't find it. I suspect that since the
*two* Native American sequences which were compared clustered
together, the remainder were just assumed to be European.
However, all the New World haplogroup X sequences which are
currently presumed to be Native American are not homogeneous.

With regards to this statement:

"Indeed, if we assume that the two complete Native American X
sequences (from one Navajo and one Ojibwa) began to diverge
while their common ancestor was already in the Americas, we
obtain a coalescence time of 18,000 +- 6,800 YBP, implying an
arrival time not later than 11,000 YBP."

I wonder what coalescence time would have been determined if
the calculation had been based upon more than two sequences.

>>It should also be mentioned that Reidla et al. believe
>>the Altaians only acquired haplogroup X within the last
>>6,700 years.
>
>What are the implications of this?

Well, as far as theories are concerned, we are back to where
we were a couple years ago. There is good Old World haplogroup
X information now but good New World information is still
required to compare it to. Two complete sequences are hardly
sufficient. All the rest are partials (hypervariable)

>Do I gather that one shouldn't think that this article is the
>last word?
>:-)

For the determination of information about Near
Eastern/European haplogroup X sequences, this was a pretty
comprehensive study. For information concerning Native
American haplogroup X, I'll wait for the next one....

Gisele

[Philip Dei]

On Tue, 04 Nov 2003 03:46:31 GMT, Gisele Horvat
<g-horvat@shaw.ca> wrote:

>"Indeed, if we assume that the two complete Native American X
>sequences (from one Navajo and one Ojibwa) began to diverge
>while their common ancestor was already in the Americas, we
>obtain a coalescence time of 18,000 +- 6,800 YBP, implying an
>arrival time not later than 11,000 YBP."

Ouch! I assume that 6800 is 1 S.D. and the 11,200 to 24,800
= the 68% confidence interval. Real confidence is 4,400 to
31,600 years.

>I wonder what coalescence time would have been determined if
>the calculation had been based upon more than two sequences.

Actually the coalescence time they state as minimum is wrong
based on their own S.D. likewise the maximum. I can estimate
that if adding 2 more about the same mean the 96% confidence
interval would only retract by 31% about the mean (whatever
that mean is)

>>>It should also be mentioned that Reidla et al. believe
>>>the Altaians only acquired haplogroup X within the last
>>>6,700 years.
>>
>>What are the implications of this?
>
>Well, as far as theories are concerned, we are back to where
>we were a couple years ago. There is good Old World
>haplogroup X information now but good New World information
>is still required to compare it to. Two complete sequences
>are hardly sufficient. All the rest are partials
>(hypervariable)
>
>>Do I gather that one shouldn't think that this article is
>>the last word?
>>:-)
>
>For the determination of information about Near
>Eastern/European haplogroup X sequences, this was a pretty
>comprehensive study. For information concerning Native
>American haplogroup X, I'll wait for the next one....
>
>Gisele

[Zolota]

"Philip Deitiker" <Nopdeitik@att.net.spam> wrote in message
news:mmffqv8hc1igisq81ior6kq2narv2fammk@4ax.com...
> On Tue, 04 Nov 2003 03:46:31 GMT, Gisele Horvat
> <g-horvat@shaw.ca> wrote:
>
>
> >"Indeed, if we assume that the two complete Native American
> >X sequences (from one Navajo and one Ojibwa) began to
> >diverge while their common ancestor was already in the
> >Americas, we obtain a coalescence time of 18,000 +- 6,800
> >YBP, implying an arrival time not later than 11,000 YBP."
>
> Ouch! I assume that 6800 is 1 S.D. and the 11,200 to 24,800
> = the 68% confidence interval. Real confidence is 4,400 to
> 31,600 years.

Where is the logic that the divergence occured in North
America? The original sentence said "if we assume----common
ancestor was already in the Americas ". Nothing can prove that
the Navajo and Ojibwai diverged in north America.

Z

[Mb]

Gisele Horvat <g-horvat@shaw.ca> wrote:

> I wonder what coalescence time would have been determined if
> the calculation had been based upon more than two sequences.

Btw. I was just evaluating a program I recently revised and
let it trace haplogroups from a mtDNA collection.

At this occation I retrieved 'gi|29690980|Navajo_Na3X' which
is identical to what is labeled Nav125 in Figure 1 of the
paper. Since they didn't note the GenBank identifier I'd have
to dig through the literature in order to look up if the
sequences are actually the same. Something I'm to lazy to do
since this stuff doesn't interest me that much.

Case you eventually should have the GenBank ID for Nav125 you
may post it here. If the ID's are not identical then there
might at least exist three X2a seqs, although I don't think
they missed one.

Michael

[Gisele Hor]

On Wed, 5 Nov 2003 15:37:24 +0100, cai@pirin.ha (mb) wrote:

>Gisele Horvat <g-horvat@shaw.ca> wrote:
>
>> I wonder what coalescence time would have been determined
>> if the calculation had been based upon more than two
>> sequences.
>
>Btw. I was just evaluating a program I recently revised and
>let it trace haplogroups from a mtDNA collection.
>
>At this occation I retrieved 'gi|29690980|Navajo_Na3X' which
>is identical to what is labeled Nav125 in Figure 1 of the
>paper. Since they didn't note the GenBank identifier I'd have
>to dig through the literature in order to look up if the
>sequences are actually the same. Something I'm to lazy to do
>since this stuff doesn't interest me that much.
>
>Case you eventually should have the GenBank ID for Nav125 you
>may post it here. If the ID's are not identical then there
>might at least exist three X2a seqs, although I don't think
>they missed one.

Michael,

Na3X (Mishmar's) is the Navajo sequence shown:

("...5 complete sequences (Lev2, MM, E18, Nav125, and Oj2)
were taken from the literature (Levin et al.1999; Maca-Meyer
et al.2001; Mishmar et al. 2003; Bandelt et al., in press)."

I have no idea why it was called Nav125 in Reidla's article.

The authors went to the trouble of obtaining 5 additional
complete Old World haplogroup X sequences and they surveyed
13,589 Old World mtDNAs but did not attempt to add, to the
genetic record, one iota of Native American haplogroup X
information:

"The data obtained from the analyses of complete mtDNA
sequences belonging to haplogroup X were then used to survey
13,589 mtDNAs (21,682 when mtDNA data from the literature are
included) from 66 populations of Eurasia and North Africa
(table 1).

This would be my main complaint. The article is not really
about haplogroup X as the title suggests but, rather, about
Old World haplogroup X. How can any group of researchers have
a good knowledge of the diversity of these sequences in the
Americas when they are only looking at two complete ones? I
don't want to complain too loudly, though, because I'm
greatful for all the RFLP information they obtained.

Did you notice that the Navajo sequence had the variant at
14560 like the X1 Druze example shown in Figure 1? Reidla et
al. failed to mention it and yet, it seems unlikely to me
that the same mutation occurred in both Old World and New
World haplogroup X sequences independently and what about the
recurrent 13708A variant in sub-groups X2d and X2b? This
variant was considered to be definitive of Eurasian
haplogroup J. Lev2 has two coding region mutations (13708A +
12612G) in common with haplogroup J... The authors did
mention that the 153G HVR II variant is found in *Native
American* haplogroup A sequences but did not really suggest a
connection. In the New World, A should be closest to X based
upon coding region variants.

A couple of Old World haplogroups (I & N1b) which should be
closer to X than most European/Near Eastern haplogroups
have the 1719A variant. Therefore, how can the absence of
it in certain sequences be indicative of the "root"
haplogroup X sub-group? There are many unanswered
questions... Certainly, as Doug Weller suggested, this is
not the last word on the subject.

Gisele

[Mb]

Gisele Horvat <g-horvat@shaw.ca> wrote:

> >Case you eventually should have the GenBank ID for Nav125
> >you may post it here. If the ID's are not identical then
> >there might at least exist three X2a seqs, although I don't
> >think they missed one.
>
> Michael,
>
> Na3X (Mishmar's) is the Navajo sequence shown:
>
> ("...5 complete sequences (Lev2, MM, E18, Nav125, and Oj2)
> were taken from the literature (Levin et al.1999; Maca-Meyer
> et al.2001; Mishmar et al. 2003; Bandelt et al., in press)."
>
> I have no idea why it was called Nav125 in Reidla's article.
>

Yes, Navajo_Na3X is submitted from Mishmar but I havn't
bothered with with Maca-Meyer and Levin and I can't know what
Bandelt is coming up with. Hence there is at least the
theoretical possibility that Nav125 do not refer to Mishmar,
though I think it actually points to the latter.

Well and relabelling samples while one is working on them
isn't unusual for practical reasons during a study but should
of course clearified if the work is going public.

> The authors went to the trouble of obtaining 5 additional
> complete Old World haplogroup X sequences...

I so fare merely glossed over the paper and, as I said, took
shorthand some of it's infos for rather heuristic reasons. And
since I'm going to leave for a couple of days soon, I've no
time to dig in deeper here, besides that this stuff is no
matter of interest for me. But maybe I'll pick up if I'm back.

Michael

[Seppo Renf]

Doug Weller wrote:
>
> http://www.dienekes.com/blog/archives/000402.htmlOrigin and
> Diffusion of mtDNA Haplogroup X

Oh yeah... what a "great" URL... it produced the following
(till stopped): http://www.dienekes.com/blog/archives/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/b-
log/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/blog/s-
tyles-site.css

This is the correct URL:
http://www.dienekes.com/blog/archives/000402.html

>
> Am. J. Hum. Genet., 73:000, 2003
>
> Maere Reidla et al.
>
> I think the article is still in copyright so I won't
> reproduce it here -

Indeed, but then copyright lasts until 50 years AFTER the
author's death!

> it suggests that Haplogroup X in Native Americans comes from
> the Near East.

There are more than one kind of Haplogroup X and there is
still a lot of uncertainly about this as I understand it.

>

--
SIR - Philosopher unauthorised
--------------------------------------------------------
----------
The one who is educated from the wrong books is not educated,
he is misled.
--------------------------------------------------------
----------

[Mb]

Gisele Horvat <g-horvat@shaw.ca> wrote:

> ..
>
> Na3X (Mishmar's) is the Navajo sequence shown:
>
> ("...5 complete sequences (Lev2, MM, E18, Nav125, and Oj2)
> were taken from the literature (Levin et al.1999; Maca-Meyer
> et al.2001; Mishmar et al. 2003; Bandelt et al., in press)."
>

Being back, I traced for the 5 sequences in my mtDB. It
seems that MM is from Maca-Meyer, E18 likely is from
Mishmar as probably is Nav125. Lev2 and Oj2 were apparently
not submitted, so they are probably taken from Bandelts
upcoming paper.

OTH I found four additional mtDNA's from Finnilä which do fit
into haplogroup X, although two of them are identical. If I
see this correctly they would belong to X2c and X2b
respectively. Don't know why the left them out, but alas..

>
> ... The article is not really about haplogroup X as the
> title suggests but, rather, about Old World haplogroup X. ..

Well to me the title suggests that the paper tries to give a
clue about the origin of haplogroup X and it's initial
dispersal from there, but not necessarily wants to depict a
global history.

>
> Did you notice that the Navajo sequence had the variant at
> 14560 like the X1 Druze example shown in Figure 1? Reidla et
> al. failed to mention it and yet, it seems unlikely to me
> that the same mutation occurred in both Old World and New
> World haplogroup X sequences independently..

Maybe they oversaw this, or maybe the mutations are not the
same. 14560 belongs to a codon which stays for Valine, and
this codon is fourfold degenerated. Which mutation actually
happened isn't clear. One had to look up this Druze sequence.

But I must say that for me, as one who rather rarely involves
in this droughty haplostuff, the notations in figure 1 appear
sometimes ambiguous. While underlined positions shell indicate
that a mutation happens more than once, their kind is not
immediately clear.
e.g. 227 in MM is - referring to Anderson - a transition,
while 226 is a reversion of 226 in the root of X2b, IOW
here indicates indentity with the Anderson reference as
is the case for 13966.

> and what about the recurrent 13708A variant in sub-groups
> X2d and X2b? This variant was considered to be definitive of
> Eurasian haplogroup J.

According to table 2 of Herrnstadts 2002 paper this mutation
can be observed in a number of other haplogroups too, hence I
don't know in howfare this has to be considered as definitive
for a certain group.

> Lev2 has two coding region mutations (13708A + 12612G) in
> common with haplogroup J... ...

From fig.1 it is not clear for me that it is an A or a G
respectively. And even if so, so what?

> A couple of Old World haplogroups (I & N1b) which should be
> closer to X than most European/Near Eastern haplogroups have
> the 1719A variant. Therefore, how can the absence of it in
> certain sequences be indicative of the "root" haplogroup X
> sub-group?

Straightforward because all other X samples share 1719A, as
well as all
(e.h. X1 and X2) sequences share 6221:C 6371:T 14470:C
16189:C 16278:T?

Howevere, probably your approach to haplogroups is too static.
Haplogroups are defined by a unique set of mutations but not
by a set of unique mutations, which means that certain
mutations might show up in other haplogroups without
necessarily affecting the validity of one of this definitions.

Michael