PDA

View Full Version : What are the implications of DF13's age and early branching ?



Mikewww
05-02-2014, 12:37 PM
There is a particular phenomenon related to DF13 seems fantastic in its extensiveness.

We see blocks or string of SNPs for many ancient haplogroups. For instance, L21 has equivalents (so far) with Z245, Z290, Z260 and L459. I gather that these blocks of equivalents indicate bottlenecks of some sort where the other lineages that occurred and branched away but did not survive. This seems to be more of a rule than an exception for paternal lineages.

However, we have not found any equivalents to DF13. I think we've had a couple of dozen L21 guys test with FGC and about 300 with Big Y and we no equivalents are yet found. We are finding more and older subclades under DF13, but no equivalents.

Essentially, this means there are no SNP mutations between the L21/Z290/Z260/L459/Z245 MRCA and DF13 in the mappable regions of the Y.

The implication is that L21 MRCA and DF13 descendant were separated by just a few generations. DF13 could have easily have personally known of the L21 MRCA.

However, under DF13 there is no such bottleneck. The early expansion, or I should probably say the survivability of DF13's early expansion is quite extensive to say the least.

There are a number of subclades (think subclade not single SNP) underneath DF13. Many have blocks or strings of equivalents but we designate them with lead SNP names such as DF21, DF41, DF49, L513, L1335, Z251, Z253, Z255, FGC11134 (above CTS4466), along with CTS1751, CTS3386, Z16500, Z16502, S1026, S1051, S16264, MC14, FGC5496 etc. There are many more DF13* guys out there so I expect some consolidation but we'll probably also see new subclades identified.

What happened that gave DF13 such an expansive progeny that survived?

[[ Edit: added SNPs ]]

alan
05-02-2014, 03:05 PM
My first thought is that it had no rivals and was 'first in' in the isles. The relatively remote location of the isles, the need for maritime skills and the complete lack of any pre-beaker metallurgy in the isles makes it different from many area.

When you look at other areas, there are vast areas of Europe that did have a pre-beaker metal tradition including Iberia from shortly before 3000BC, southern France/Languedoc from just before that, Italy from at least 3500BC, the Alps from vary dates between 4000-3000BC, SE Europe long before that. Even corded ware groups in north-central Europe had metallurgy before beakers. Areas like Atlantic France and the Isles were not only completely lacking pre-beaker metallurgy but they were also areas with a lot of copper, tin and gold. So, you could say it was a perfect storm for beaker people being welcomed because they brought something entirely new to the table.

Metal seems likely to be the key. However, I think given the isles was the main area that DF13 really took off in a huge way we cannot ignore the fact that maritime skills were crucial. The combination of beaker metallurgy and maritime skills suitable for northern seas was probably something that land locked groups could not claim to possess. Its still a challenge to work out exactly who and where these skills came together but somehow metallurgical knowledge and maritime skills had to combine somewhere to make this possible. The lack of maritime skills and location is IMO probably the main reason U152 made little impact on the isles in comparison.

If I had to guess where metallurgical skills and maritime knowledge that allowed DF13 to do so well in the isles came together I would say somewhere between NW France and the Rhine. I tend to think NW France was crucial because it was the easiest way to get to the main early beaker source in Ross Island SW Ireland and also the Irish gold sources. It was a special source too as it was arsenical copper which was as hard as bronze - very rare in the isles. The distribution of isles-made beaker era metal products also points to NW France as part of its network. It seems to have been a crucial location where eastern and western influences met. However, its also clear that the Rhenish area was part of the network that linked to the isles and I tend to think middlemen in southern and eastern Britain linked the metal areas of the west to the Rhine area and culturally seemed more directly linked to the latter area than the actual metal ore areas in Ireland and the west. The best guess for parallels with the Irish beaker phenomenon point tentatively to NW France in terms of tomb types, pottery and distribution of beaker metalwork. Case even suggests some aspects of Irish beaker point to NW France. At the same time a lot of Irish beaker looks like British-Rhenish types so its complex and clearly there was some sort of network that went from the ore sources in Ireland through British middlemen who were in turn strongly connected to the Rhine.


There is a particular phenomenon related to DF13 seems fantastic in its extensiveness.

We see blocks or string of SNPs for many ancient haplogroups. For instance, L21 has equivalents (so far) with Z245, Z290, Z260 and L459. I gather that these blocks of equivalents indicate bottlenecks of some sort where the other lineages that occurred and branched away but did not survive. This seems to be more of a rule than an exception for paternal lineages.

However, we have not found any equivalents to DF13. I think we've had a couple of dozen L21 guys test with FGC and about 300 with Big Y and we no equivalents are yet found. We are finding more and older subclades under DF13, but no equivalents.

Essentially, this means there are no SNP mutations between the L21/Z290/Z260/L459/Z245 MRCA and DF13 in the mappable regions of the Y.

The implication is that L21 MRCA and DF13 descendant were separated by just a few generations. DF13 could have easily have personally known of the L21 MRCA.

However, under DF13 there is no such bottleneck. The early expansion, or I should probably say the survivability of DF13's early expansion is quite extensive to say the least.

There are a number of subclades (think subclade not single SNP) underneath DF13. Many have blocks or strings of equivalents but we designate them with lead SNP names such as DF21, DF41, DF49, L513, L1335, Z251, Z253, Z255, FGC11134 (above CTS4466), along with CTS1751, CTS3386, Z16500, Z16502, S1026, S1051, S16264, MC14, FGC5496, etc. There are many more DF13* guys out there so I expect some consolidation but we'll probably also see new subclades identified.

What happened that gave DF13 such an expansive progeny that survived?

alan
05-02-2014, 03:23 PM
To put it more simply there was a kind of a loop of beaker connections that went (with no certain starting point) Rhine-Britain-Ireland-NW France-Rhine-Britain-Ireland NW France etc. I would say the archaeological evidence best fits the direction of the network as

Rhine-NW France (contacts between two areas apparent even pre-beaker)

NW France-Ireland (similarity in beaker era tomb types and possibly pottery)

Ireland/Britain to NW France (distribution of Irish and British metalwork in NW France)

Britain to Rhine (strong Rhenish cultural influences despite use of Irish metal)

Britain to Ireland (much Irish beaker is of the British-Rhenish sort)

That IMO is why the way beaker manifests itself in the isles and adjacent continent is complicated with a varying mix of Atlantic and Rhenish influences. In classic archaeological vagueness its easy to see the network but hard to see the starting point.

Mikewww
05-02-2014, 03:25 PM
My first thought is that it had no rivals and was 'first in' in the isles. The relatively remote location of the isles, the need for maritime skills and the complete lack of any pre-beaker metallurgy in the isles makes it different from many area. ...

I agree the green fields "no rivals" type of situation is a viable alternative.

Another thing I think could be a bit of the reciprocal. That would be not that DF13 was so successful but that most of DF13's cousins and brothers suffered some ill fate somewhere, such as in France. The ill fate could be someone like Julius Caesar possibly to be later followed by folks coming across the Rhine from the east.

MJost
05-02-2014, 04:06 PM
Mike, I guess FGC5496 is the etc. ?

Late Bronze and Iron in NW Europe along with maritime trade ability appears to provide the best overall catalyst to DF13's massive expansion from 1500bc forward.

In an unpublished dissertation:

Re-evaluating Iron Age maritime societies: the North West Iberian Peninsula
by Rodrigo Pacheco Ruiz

https://www.academia.edu/437627/Re-evaluating_Iron_Age_maritime_societies_the_North_W est_Iberian_Peninsula


"There is no better way to approach the problem of our concern than by first considering Cunliffe’s work for the

Revista de Guimarãaes

entitled ‘Atlantic Sea-ways’. He argues that a key role the NW Iberian Peninsula is thought to have rendered was as a ‘stepping-stone’, in a world of social relationships from the Late Bronze Age to the Early Roman Period. Of course, one cannot begin to comprehend NW Iberia’s role, without first considering Cunliffe’s last idea in this epigraph: the ongoing uncertainty of assuming any kind of open sea connection with the rest of the Atlantic economic world. Cunliffe stresses both the importance of this region to the study of maritime archaeology and at the same time, the level of uncertainty that surrounds its exact role within an Atlantic maritime system.

...

2.1.1

Bronze Age metal trade and the Atlantic influence

The Bronze Age has been considered by several authors (Cunliffe, 1999) to be theclimax of maritime relationships between the Atlantic communities and the NW Iberian Peninsula. This viewpoint is clearly based largely on the evidence for involvement in the metal trade system. The concept of a ‘global economy’, directed Kristiansen’s (1994) research towards the Late Bronze Age metal production centres, such as southern England, Ireland, northwestern France and northwestern Iberia, together with Portugal, which he argues show some relation through a common metal tradition. Cunliffe (1999: 96-97) presents evidence for NW Iberian artefacts, such as the seventh to fifth centuries disc-footed fibulae and the antennae-hilted swords, which are rarely found outside this place of origin, actually being recovered in the south-west of Britain at sites such as Mount Batten and Harlyn Bay. This, he argues, points to a basic maritime relationship between these two areas, probably related to the tin industry, stressing the special bond that the Atlantic cultures had with NW Iberia.This proposed link could also be extended to the roundhouse building tradition and language similarities"

Overall good piece of information.

MJost

rms2
05-02-2014, 04:37 PM
I don't recall the earliest evidence of the manufacture and consumption of alcoholic beverages in the Isles and NW Europe in general, but I wonder what part the Beaker Folk played in that. The ability to summon the "spirits" must have been a tremendous advantage to those who possessed it.

So, I suggest the name of the first DF13+ man is actually known and celebrated. It was Charlie Mops (http://www.youtube.com/watch?v=DJVtFpZl7-Y).

:beerchug:

Mikewww
05-02-2014, 04:55 PM
Mike, I guess FGC5496 is the etc. ? Sorry, oversight... fixed.


Late Bronze and Iron in NW Europe along with maritime trade ability appears to provide the best overall catalyst to DF13's massive expansion from 1500bc forward.

In an unpublished dissertation:

Re-evaluating Iron Age maritime societies: the North West Iberian Peninsula
by Rodrigo Pacheco Ruiz

https://www.academia.edu/437627/Re-evaluating_Iron_Age_maritime_societies_the_North_W est_Iberian_Peninsula
....

In this scenario, would the surprisingly heavy dose of L21 in the Bologna, Italy area be due to an Atlantic back to the Mediterranean connection or due to a eastward migration from Transalpine Gaul over land?

I guess it would be important to know if the L21 in N. Italy was predominately DF13+ or not.

MJost
05-02-2014, 05:43 PM
Sorry, oversight... fixed.

In this scenario, would the surprisingly heavy dose of L21 in the Bologna, Italy area be due to an Atlantic back to the Mediterranean connection or due to a eastward migration from Transalpine Gaul over land?

I guess it would be important to know if the L21 in N. Italy was predominately DF13+ or not.
Good question. That would be a great new research project. Maybe they should log in and get opinions from posting community as to what SNPs to test for.

Thx for the fix... Maybe we need a size by nodes ranking.


MJost

MJost
05-02-2014, 07:01 PM
I don't recall the earliest evidence of the manufacture and consumption of alcoholic beverages in the Isles and NW Europe in general, but I wonder what part the Beaker Folk played in that. The ability to summon the "spirits" must have been a tremendous advantage to those who possessed it.

So, I suggest the name of the first DF13+ man is actually known and celebrated. It was Charlie Mops (http://www.youtube.com/watch?v=DJVtFpZl7-Y).

:beerchug:


You were posting several years ago that the aspect of "Maritime Bell Beaker predates the Atlantic Bronze Age Network, but what I meant was that network was of long standing before it reached its peak about 1300 BC - 700 BC."

Can you repost your thoughts updated with all the recent knowledge?

MJost

rms2
05-02-2014, 11:27 PM
You were posting several years ago that the aspect of "Maritime Bell Beaker predates the Atlantic Bronze Age Network, but what I meant was that network was of long standing before it reached its peak about 1300 BC - 700 BC."

Can you repost your thoughts updated with all the recent knowledge?

MJost

Whoa! That would be a long post!

Honestly, I don't even remember posting that.

alan
05-02-2014, 11:47 PM
The things is though when you look at the archaeology Atlantic Iberia actually was the final of the areas to join the Atlantic Bronze Age network and it seems more of a case of a much older northern Atlantic network expanding to include Portugal than the other way around. Personally I think that network in the late Bronze Age had virtually nothing to do with the clade pattern as there is a massive difference between the L21 dominated northern part of the network and the DF27 dominated Portuguese end of the network. I think the sharp division in clade pattern suggest those patterns date much earlier than the Atlantic Bronze Age and were pretty set in stone long before it came into existence. L21 seems to me a relic of the Isles-NW France-Lower Rhine triange of beaker period links, albeit that L21 has been badly truncated by later arrivals like U106 in the low countries and adjacent.

alan
05-02-2014, 11:57 PM
One other implication of DF13 IMO is that it really is very hard to see how it could be younger than 2400BC if its beaker linked, which seems very likely. Likewise, P312 really cannot be younger than say 2600BC and in all probability 2800BC and retain the beaker model for the pan-European nature of its spread. As there is absolutely nothing later than beaker which can adequately explain a pan-European spread I think that should be borne in mind in the debate on how to calibrate or formulate SNP counting dating. If P312 is not at least 4600-5000 years old then the beaker model falls. If it was even older then that is not such a problem as other options would arise for explanation but if it was younger that would be a huge problem.

alan
05-04-2014, 12:28 PM
Mike

Are you saying that U152, DF27 and U106 have no parallels for the frantic branching that started immediately after DF13? It does look like DF13 was the start of a sequence of continuous branching that kicked off immediately after that SNP. Timeframe wise DF13 is just the 2nd SNP after P312 on its line-only L21 intervenes - which is why I dont understand dates of 3500 years being put on L21 by some calculations. The SNP evidence suggests a constant branching started just two SNPs down from P312 so I am guessing this branching commenced about 200 years after P312.

One thing is clear to me is that despite the fact that DF13 has an entirely different geography from U152, DF13 exploded at a time when it and U152 and DF27 had only parted company two SNPs earlier - perhaps c. 200 years. That suggests to me that despite the differences in branching etc that DF13 had geographically moved into northern France and the isles only 200 years after P312.

The way I look at it is DF13 is just 2 SNPs below P312 so its very early in the story. The equivalent time period for U152 and DF27 would be the first SNP below them. I dont follow those clades closely but I am personally not aware of constant major branching happening with them.


There is a particular phenomenon related to DF13 seems fantastic in its extensiveness.

We see blocks or string of SNPs for many ancient haplogroups. For instance, L21 has equivalents (so far) with Z245, Z290, Z260 and L459. I gather that these blocks of equivalents indicate bottlenecks of some sort where the other lineages that occurred and branched away but did not survive. This seems to be more of a rule than an exception for paternal lineages.

However, we have not found any equivalents to DF13. I think we've had a couple of dozen L21 guys test with FGC and about 300 with Big Y and we no equivalents are yet found. We are finding more and older subclades under DF13, but no equivalents.

Essentially, this means there are no SNP mutations between the L21/Z290/Z260/L459/Z245 MRCA and DF13 in the mappable regions of the Y.

The implication is that L21 MRCA and DF13 descendant were separated by just a few generations. DF13 could have easily have personally known of the L21 MRCA.

However, under DF13 there is no such bottleneck. The early expansion, or I should probably say the survivability of DF13's early expansion is quite extensive to say the least.

There are a number of subclades (think subclade not single SNP) underneath DF13. Many have blocks or strings of equivalents but we designate them with lead SNP names such as DF21, DF41, DF49, L513, L1335, Z251, Z253, Z255, FGC11134 (above CTS4466), along with CTS1751, CTS3386, Z16500, Z16502, S1026, S1051, S16264, MC14, FGC5496 etc. There are many more DF13* guys out there so I expect some consolidation but we'll probably also see new subclades identified.

What happened that gave DF13 such an expansive progeny that survived?

[[ Edit: added SNPs ]]

R.Rocca
05-04-2014, 12:54 PM
Mike

Are you saying that U152, DF27 and U106 have no parallels for the frantic branching that started immediately after DF13? It does look like DF13 was the start of a sequence of continuous branching that kicked off immediately after that SNP. Timeframe wise DF13 is just the 2nd SNP after P312 on its line-only L21 intervenes - which is why I dont understand dates of 3500 years being put on L21 by some calculations. The SNP evidence suggests a constant branching started just two SNPs down from P312 so I am guessing this branching commenced about 200 years after P312.

One thing is clear to me is that despite the fact that DF13 has an entirely different geography from U152, DF13 exploded at a time when it and U152 and DF27 had only parted company two SNPs earlier - perhaps c. 200 years. That suggests to me that despite the differences in branching etc that DF13 had geographically moved into northern France and the isles only 200 years after P312.

The way I look at it is DF13 is just 2 SNPs below P312 so its very early in the story. The equivalent time period for U152 and DF27 would be the first SNP below them. I dont follow those clades closely but I am personally not aware of constant major branching happening with them.

Just for reference, U152 is alone in its branch and L2, which makes up about 65% of all U152 frequency outside of Italy, is also alone on its own branch. L2 is standard WAMH and seems also to have started its branching early on in the P312 story. We are now up to 13 distinct subclades of L2.

Little bit
05-04-2014, 01:16 PM
Interesting topic, thanks for posting. My 2 guys are both DF13+. My son turned out to be L226+ from Rathkeale, Limerick Ireland and my grandpa is just DF13+ so far, from Wales. I'm beginning to wonder if my grandfather carries a particularly old version of DF13, perhaps basal, and I wonder if it could help with the research of DF13? I did a study to see if my grandpa would be a good candidate for FGC5496 but I think he's got too many off modal STR's (9), all of which are are included in my grandpa's rare or uncommon versions:

Off modal w/FGC5496:
439 10 - no others - fast mutating
458 16 - 1 other w/16 - fast mutating
448 18 - no others - medium mutating
460 10 - 1 other w/10 - fast mutating
442 11 - 4 others w/11 - medium mutating
511 9 - no others - medium mutating
534 14 - 1 other w/14 - fast mutating
444 11 - 3 others w/11 - medium mutating
481 23 - no others - fast mutating

These 4 are markers which my grandpa carries an uncommon version which seems more common in this group:

389ii 16 - medium mutating
576 17 - fast mutating
CDYb 39 - fast mutating
520 21 - medium mutating

I had calculated his 67 markers using a tool I got off the Wheaton Resources blog to figure out why my grandpa has so few matches at FTDNA. Just one at 12 markers, same surname, Griffith. Must be related but I haven't been successful in contacting him, I believe he passed away in 2010.

alan
05-04-2014, 02:57 PM
One thing that may explain why other P312 clades had a less spectacular take off might be indicated in the beaker maps. Beaker concentrations in continental Europe are in very specific pockets rather than widespread.

alan
05-04-2014, 03:00 PM
Is there a comparible tree for U152's early branching? I am just wondering how comparible its branching is to L21. If L2 is the biggie just one SNP below U152 and has a lot of early branches then L2 is comparable with DF13.

Also what about DF27 - is does it have a major cluster of branching just downstream?


Just for reference, U152 is alone in its branch and L2, which makes up about 65% of all U152 frequency outside of Italy, is also alone on its own branch. L2 is standard WAMH and seems also to have started its branching early on in the P312 story. We are now up to 13 distinct subclades of L2.

Webb
05-04-2014, 03:12 PM
Is there a comparible tree for U152's early branching? I am just wondering how comparible its branching is to L21. If L2 is the biggie just one SNP below U152 and has a lot of early branches then L2 is comparable with DF13.

Also what about DF27 - is does it have a major cluster of branching just downstream?

Yes it does. Z195/Z196 seems to take off quickly with multiple mutations just downstream.

Webb
05-04-2014, 03:26 PM
I've posted this before, but the Irish clan system may have been a catalyst for this type of mutation variance. Artificial bottle necking if you will.

alan
05-04-2014, 03:30 PM
That is quite reassuring. I would expect within a beaker model for there to be that sort of pattern among L2 as there was a pretty rich beaker culture in west-central Europe. I do tend to think U152 underwent its first major expansion in the central-east group of beakers and slowly expanded at the expense of the north-west European L21 group as that group morphed through the Unetice-Tumulous-Urnfield-Hallstatt sequence.


Just for reference, U152 is alone in its branch and L2, which makes up about 65% of all U152 frequency outside of Italy, is also alone on its own branch. L2 is standard WAMH and seems also to have started its branching early on in the P312 story. We are now up to 13 distinct subclades of L2.

seferhabahir
05-04-2014, 03:33 PM
Mike
Are you saying that U152, DF27 and U106 have no parallels for the frantic branching that started immediately after DF13? It does look like DF13 was the start of a sequence of continuous branching that kicked off immediately after that SNP. Timeframe wise DF13 is just the 2nd SNP after P312 on its line-only L21 intervenes - which is why I dont understand dates of 3500 years being put on L21 by some calculations. The SNP evidence suggests a constant branching started just two SNPs down from P312 so I am guessing this branching commenced about 200 years after P312.


This is an interesting thread, guys. I'm also having problems with the 3500 year calculation for L21 based on my FGC results where I seem to have up to 55-60 SNPs probably unique below Z251 (well, probably below Z251>S11566>S9294). Given that we might see a few more common SNPs below S9294 such as FGC11986, even if I use the faster 75 year (three generations) per SNP number, I get 55 x 75 = 4125 years for S92924, a SNP that is three down from DF13, putting L21 and P312 maybe at 4500-4800 years ago as Alan suggests to be consistent with the Beaker scenarios. Has there been anything found in FGC results to suggest that a 75-90 years per SNP estimate is wrong?

alan
05-04-2014, 03:46 PM
Can you link to the most up to date tree. How many SNPs down is Z195/6 from P312?


Yes it does. Z195/Z196 seems to take off quickly with multiple mutations just downstream.

alan
05-04-2014, 03:49 PM
I agree. How on earth can DF13 be that young when it is only 2 SNPs down from P312. It should only be what 160-300 years younger at most surely.


This is an interesting thread, guys. I'm also having problems with the 3500 year calculation for L21 based on my FGC results where I seem to have up to 55-60 SNPs probably unique below Z251 (well, probably below Z251>S11566>S9294). Given that we might see a few more common SNPs below S9294 such as FGC11986, even if I use the faster 75 year (three generations) per SNP number, I get 55 x 75 = 4125 years for S92924, a SNP that is three down from DF13, putting L21 and P312 maybe at 4500-4800 years ago as Alan suggests to be consistent with the Beaker scenarios. Has there been anything found in FGC results to suggest that a 75-90 years per SNP estimate is wrong?

seferhabahir
05-04-2014, 05:08 PM
Here an interesting read I found about some of the maritime events predating Western European Bell Beaker spread

http://www.minoanatlantis.com/Minoan_Spain.php

The author of this piece writes about Minoan culture and apparently runs a Minoan art gallery of some sort. Can't vouch for his credentials but the information sounds reasonable. Alan might be able to comment.

Gray Fox
05-04-2014, 05:30 PM
Can you link to the most up to date tree. How many SNPs down is Z195/6 from P312?

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

Z195/Z196 is the DF27 equivalent of DF13 for L21.

seferhabahir
05-04-2014, 06:13 PM
Here an interesting read I found about some of the maritime events predating Western European Bell Beaker spread

http://www.minoanatlantis.com/Minoan_Spain.php

The author of this piece writes about Minoan culture and apparently runs a Minoan art gallery of some sort. Can't vouch for his credentials but the information sounds reasonable. Alan might be able to comment.

OK, somewhat off-topic, but this person also has some things on the Sea Peoples on his web pages. Nobody knows what ethnic origin the Sea Peoples were. After their defeat by Ramesses III the Sea Peoples vanish from history, survivors of their last battle perhaps assimilating into Egyptian or other Levantine cultures. No records indicate where they came from and there are no accounts of them after 1178 BCE. I'm still looking around for what other North Mediterranean or southern European types might have taken a vacation (warlike or peaceful) into the Levant at some point and then decided to assimilate into the local community. Still want to find avenues that eventually lead to the Jewish Z251-11EE cluster and its almost unique value of DYS388=11 amongst all R1b haplogroups. Why is DYS388=11 not present anywhere in R1b except my poor little cluster? Might be a late mutation, but then it might have happened early in some population group that has been lost to later history.

Agamemnon
05-04-2014, 06:22 PM
I do tend to think U152 underwent its first major expansion in the central-east group of beakers and slowly expanded at the expense of the north-west European L21 group as that group morphed through the Unetice-Tumulous-Urnfield-Hallstatt sequence.

I also think this is the most plausible model to explain U152's expansion throughout western europe.

alan
05-05-2014, 11:49 AM
would this be an up to date tree for U152 that includes all the valid SNPs?

http://www.eupedia.com/images/design/R1b-S28-tree.png

R.Rocca
05-05-2014, 11:56 AM
would this be an up to date tree for U152 that includes all the valid SNPs?

http://www.eupedia.com/images/design/R1b-S28-tree.png

Nope...not even close ;)

http://r1b.org/imgs/U152_Tree_v003.png

alan
05-05-2014, 01:32 PM
cheers Rich. I was just digging about and was about to post another one but that is even better. What I am interested in is comparing the early generation branching patterns of the main P312 divisions and also U106. Basically a comparison of branching in terms of SNP generations which I understand are under debate in terms of length. Even if we dont know the exact time between SNPs it still interesting to see if there are any contrast between clades. I think this deserves a thread on its own.


Nope...not even close ;)

http://r1b.org/imgs/U152_Tree_v003.png

MJost
05-05-2014, 05:33 PM
I agree. How on earth can DF13 be that young when it is only 2 SNPs down from P312. It should only be what 160-300 years younger at most surely.

Lets say that P312>L21>DF13 were much old than 4k ybp, as you feel it must be, and we agree that DF13 initially had a very quick growth resulting in large numbers of branches, then a very sever bottle neck resulting in daughtering out many of these DF13 branches over the next 500+ years (20+ generations @25 years per) played a large roll affecting TMRCAs lowering its calculated ages.

MJost

alan
05-05-2014, 06:10 PM
To fit into a beaker model P312 could be no younger than about 4700 years old. Any less than that and the beaker model doesnt really work. In fact if the chronology of beaker pots perfectly mirror the spread of P312 or L11 in a literal way it would have to be a century of two older still.


Lets say that P312>L21>DF13 were much old than 4k ybp, as you feel it must be, and we agree that DF13 initially had a very quick growth resulting in large numbers of branches, then a very sever bottle neck resulting in daughtering out many of these DF13 branches over the next 500+ years (20+ generations @25 years per) played a large roll affecting TMRCAs lowering its calculated ages.

MJost

MJost
05-05-2014, 07:34 PM
So lets look at the most notable events in the 'weather history' across the British Isles. These major events could have stymied and caused branches to die off until there was later expansion to where the current HT variance numbers coalesce.

http://booty.org.uk/booty.weather/climate/4000_100BC.htm

MJost

Tolan
05-06-2014, 12:46 PM
Let me to give you an extact from the book "Protohistoire de la Bretagne", published in 1979, and the article by Jacques Biard, about the relationship between Brittany, British Isles and northern Germany, 4 000 years ago (old bronze).
I don't dare translate, I might make too many mistakes...

Au néolithique, les populations atlantiques avaient des relations
maritimes de l'Ibérie à l'Irlande, en passant par la Bretagne. Au
bronze ancien, l'activité se complète d'impulsions nordiques certainement,
le trafic de l'Ambre, venu probablement de la Baltique, joua un
grand rôle mais la connaissance du métal vient peut être en partie de
cette voie. Beaucoup de types d'armes,poignards triangulaires, haches
à petits rebords ont vu le jour semble-t-il dans ce creuset d'inventions
métallurgiques qu'est, vers 2000, la civilisation d'Unetice.
De là, vers le nord de l'Allemagne vont diffuser techniques et armes nouvelles.
Par la suite le mouvement semble redescendre au long des côtes de la Mer du Nord et de la Manche
pour atteindre les Iles Britanniques et la Bretagne.
Un certain nombre d'arguments plaident en ce sens. L'ambre, bien sûr, mais aussi
les nouvelles techniques de sépultures à structure en bois. C'est une nouveauté en Bretagne mais une chose courante
dans l'Allemagne du Nord ou le Danemark où les conditions de
conservation en milieu tourbeux ont préservé quelques-unes des cercueils faits de troncs de chêne
sciés en deux puis évidés à l'herminette.
D'autres relations sont certaines: parenté déjà évoquée des
épées et poignards de Gau-Bickelheim en Hesse rhénane avec les poignards bretons, parenté double puisqu'elle est à la
fois typologique (languette et 6 rivets) et technique (procédé de l'arseniure de
la surface des armes). Le jalonnement des influences nordiques se suit par le petit semis de tombes picardes ou normandes qui montrent
quelques parentés avec les ensembles bretons (Hervelinghen, Longues, La Hague).
La parenté la plus étroite est sans nul doute celle qui relie les tumulus bretons et ceux du Wessex à tel point que l'on a souvent
parlé de civilisation jumelle sinon de culture ou type britanno-armoricain. Toutefois sans faire preuve de
chauvinisme excessif, il faut souligner l'originalité de la civilisation armoricaine qui en bien des points
diffère de celle de Wessex. Wessex et Armorique ont en commun la présence des mêmes poignards à languette décorés des minuscules clous en or.
...
Chaque province a su garder son originalité à partir du même impact initial, combinant les nouvelles impulsions avec le substrat autochtone.
Dans les Îles Britanniques, il faut souligner l'influence beaucoup plus grande des campaniformes, et de leurs dérivés. Les "beakers", ces
campaniformes britanniques ont joué un grand rôle certainement dans la génèse de la civilisation du Wessex
...


About the St Jude's tumulus in Bourbriac (Côtes d'Armor)
..
Par ailleurs, c'est encore vers l'Allemagne du Nord que les comparaisons sont possibles puisque deux célèbres tumulus "princiers" de Saxe, ceux de
Leubingen et d'Helmsdorf montraient eux aussi des maisons funéraires composées de poutres arc-boutées recouvertes de chaume

MJost
05-06-2014, 01:33 PM
Bing translated
During the Neolithic, Atlantic populations had relations
Maritimes of Iberia to Ireland, passing through Brittany. To the
bronze, activity comes complete with Nordic impulses, certainly,
the amber traffic, came probably from the Baltic, played a
major role but the personal knowledge of the metal can be part of
This way. Many types of weapons, triangular daggers, axes
small edges have emerged it seems in this crucible of inventions
Metallurgical that is, around 2000, the Unetice civilization.
Thence, northward of the Germany will broadcast techniques and new weapons.
Subsequently the movement seems to back down to the long sides of the North Sea and the channel
to reach the British Isles and Brittany.
A number of arguments argue in this sense. Amber, of course, but also
new techniques of burials in wooden structure. This is a novelty in Brittany but one common thing
in the northern Germany or the Denmark where the conditions of
conservation in peat areas have preserved some of the coffins made of oak trunks
sawn in two and then seeded with an adze.
Other relationships are some: already mentioned parent of
swords and daggers of Gau-Bickelheim in Rheinhessen with Breton daggers, parent double since it is in the
typological times (tongue and 6 rivets) and technical (method of arsenide of)
the surface of the arms). Staking of the Nordic influences following by the small seedlings of Picardy or Norman graves that show
some relatives with Breton sets (Hervelinghen, long, the Hague).
The closer relationship is no doubt one that connects the Breton mounds and those of Wessex to the extent that there are often
spoken of twin civilization or culture or British-armorican type. However without evidence of
excessive chauvinism, it must be emphasized the originality of the armorican civilization in many respects
differs from that of Wessex. Wessex and Armorica have in common the presence of same tongue daggers decorated with tiny nails gold.
...
Each province has kept its originality from the initial impact, combining the new impulses with the native substrate.
In the British Isles, it should be noted the much larger influence of the campaniform, and their derivatives. "beakers", these
campaniform British played an important role definitely in the genesis of civilization of Wessex
...


About the St Jude's in tumulus Bourbriac (Côtes d'Armor)
..
Moreover, it even to the Germany of the North that comparisons are possible because two of "princely" famous mounds of Saxony, those of
Leubingen and Helmsdorf showed them also to funeral homes consisting of arc-boutees beams covered with thatched

Tolan
05-07-2014, 04:02 AM
Thanks, MJost! :)

alan
05-08-2014, 02:25 PM
I thought SNP counting was a way of getting around the problem of bottlenecks etc which we have with STRs? It seems there is a disagreement on which SNPs to count and how many average years per SNP etc but it does seem to be the way forward to work out the age of an SNP.

alan
05-08-2014, 02:29 PM
To me the keystone of all of this is the age of the P312 SNP. That is the common link of the pan-European spread and if beaker is P312 linked then it must have occurred before the very wide dispersal of beaker across Europe. I think getting SNP counting estimates for that SNP is vital and is more important to the beaker model than the age of its subclades which in theory could have arisen during or after dispersal. The age of P312 is the Rosetta Stone of the beaker model.


Lets say that P312>L21>DF13 were much old than 4k ybp, as you feel it must be, and we agree that DF13 initially had a very quick growth resulting in large numbers of branches, then a very sever bottle neck resulting in daughtering out many of these DF13 branches over the next 500+ years (20+ generations @25 years per) played a large roll affecting TMRCAs lowering its calculated ages.

MJost

MJost
05-08-2014, 04:39 PM
Ok, Lets look at my numbers from Full Genomes YDNA.

I have 304 total SNP's below DF13.

I have a total number of 1,495 private and public mutations under R1.

T. Karafet et al. estimated the age of R1, the parent of R1b, as 18,500 (12,500 - 25,700) years before present. (note this is a skewed Range slightly towards the longer)
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2336805/


18,500 minus the last 2000 that are presumed to be averaging 30 years per generations, equals 67. The previous 16,500 years at 25 years per generation would be 660 generations. Totaling a 727 Generations. Using 1,495 mutations over 727 generations, equates to 2 mutations per generation. Now using 304 mutations divided by 2 mutations per generations equals 152 generations. Taking 152 (67 generations back to 0AD, 85 Prior to 0AD) equals 2,000 + 2,125 total 4,125 years. This age is very close to my STR Founders calculations using variance.

Max at 27,500ybp: 67+ 1028 equals 1095 generations equates to 1.45 mutations per generation. 304 mutations below DF13 divided 1.45 equals 210 generations. 67 + 143 at 30 and 25 respectively totals 2000 + 3,575 totals 5,575ybp.

There is a slightly better probability that R1 could be bit higher than 18,500 than a lower age due to skewing.


MJost

alan
05-18-2014, 09:31 AM
I though Karafet's age of R1 had been discredited by the post-R* SNP string found in Mal'ta and people were now thinking R1 dated to about 25000 years ago.


Ok, Lets look at my numbers from Full Genomes YDNA.

I have 304 total SNP's below DF13.

I have a total number of 1,495 private and public mutations under R1.

T. Karafet et al. estimated the age of R1, the parent of R1b, as 18,500 (12,500 - 25,700) years before present. (note this is a skewed Range slightly towards the longer)
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2336805/


18,500 minus the last 2000 that are presumed to be averaging 30 years per generations, equals 67. The previous 16,500 years at 25 years per generation would be 660 generations. Totaling a 727 Generations. Using 1,495 mutations over 727 generations, equates to 2 mutations per generation. Now using 304 mutations divided by 2 mutations per generations equals 152 generations. Taking 152 (67 generations back to 0AD, 85 Prior to 0AD) equals 2,000 + 2,125 total 4,125 years. This age is very close to my STR Founders calculations using variance.

Max at 27,500ybp: 67+ 1028 equals 1095 generations equates to 1.45 mutations per generation. 304 mutations below DF13 divided 1.45 equals 210 generations. 67 + 143 at 30 and 25 respectively totals 2000 + 3,575 totals 5,575ybp.

There is a slightly better probability that R1 could be bit higher than 18,500 than a lower age due to skewing.


MJost

MJost
05-18-2014, 12:55 PM
T. Karafet et al. estimated the age of R1, the parent of R1b, as 18,500 (12,500 - 25,700) years before present, thus does have a statistical overlap. But not sure if YDNA was even extracted, only MtDNA, right? MJost



Raghavan, Maanasa,,et al
Upper Palaeolithic Siberian genome reveals dual ancestry of Native Americans
JA - Nature
2014/01/02/ - http://dx.doi.org/10.1038/nature12736

http://www.nature.com/nature/journal/v505/n7481/abs/nature12736.html#supplementary-information

"Here we sequence the draft genome of an approximately 24,000-year-old individual (MA-1), from Mal’ta in south-central Siberia (9) , to an average depth of 1×."

(9) 9.Gerasimov, M. M. in The Archaeology and Geomorphology of Northern Asia: Selected Works 5–32 (University of Toronto Press, 1964)

jdean
05-18-2014, 02:28 PM
I think Mal'ta boy is the oldest whole Y Chromosome yet analysed and was confirmed R x R1 & R2

1885

rms2
05-18-2014, 05:40 PM
Aren't SNP-based age estimates merely speculative until there are sufficient father-son full genome results, or, better yet, three-generation full genome results?

jdean
05-18-2014, 05:47 PM
I'd say so however the advantage with the Mal'ta results is they also come with radiocarbon dating.

Webb
05-18-2014, 06:01 PM
I'd say so however the advantage with the Mal'ta results is they also come with radiocarbon dating.

That is an often overlooked advantage of having ancient Ydna.

rms2
05-18-2014, 08:42 PM
I'd say so however the advantage with the Mal'ta results is they also come with radiocarbon dating.

I agree, but I meant SNP-based age estimates when applied to subclades with living representatives.

jdean
05-18-2014, 10:29 PM
I agree, but I meant SNP-based age estimates when applied to subclades with living representatives.

I haven't thought about this much but I'm guessing father son studies aren't going to be so usefull.

Possibly comparisons of people with well documented paper trails going back 3 or 4 hundred years to a common ancestor would be better ?

MJost
05-18-2014, 11:15 PM
The question to me is, the Mal'ta male had 138 are in the derived and 5 in the ancestral state, placing MA-1 as a lineage basal to hg R. How many years is that in downstream generations?

MJost

MJost
05-19-2014, 08:22 AM
With that many derived SNPs in Mal'ta boy and site carbon dating confirms 24 kya , then using 90 years per SNP, R is around 12.4kya older at 37.4kya. This age for 'R' is at the high end of Karafet 2008.

How many SNP are now found between R and R1? The R1 possible time of origin 12,500–25,700 years BP (Karafet 2008) would be pushed to the higher end closer to 25.7K.

Interesting to see this angle and leads me to tend to question if every TRMCA method needs corrected.

MJost

jdean
05-19-2014, 10:37 AM
With that many derived SNPs in Mal'ta boy and site carbon dating confirms 24 kya , then using 90 years per SNP, R is around 12.4kya older at 37.4kya. This age for 'R' is at the high end of Karafet 2008.

How many SNP are now found between R and R1? The R1 possible time of origin 12,500–25,700 years BP (Karafet 2008) would be pushed to the higher end closer to 25.7K.

Interesting to see this angle and leads me to tend to question if every TRMCA method needs corrected.

MJost

Certainly tweaking I'd say

Michał
05-19-2014, 11:05 AM
With that many derived SNPs in Mal'ta boy and site carbon dating confirms 24 kya , then using 90 years per SNP, R is around 12.4kya older at 37.4kya. This age for 'R' is at the high end of Karafet 2008.

How many SNP are now found between R and R1? The R1 possible time of origin 12,500–25,700 years BP (Karafet 2008) would be pushed to the higher end closer to 25.7K.

Interesting to see this angle and leads me to tend to question if every TRMCA method needs corrected.

MJost

All this has been discussed elsewhere some time ago:
http://www.anthrogenica.com/showthread.php?1507-Some-provisional-calculations-for-haplogroup-R1a-based-on-the-first-FGC-result/page4&p=20838#post20838

MJost
05-19-2014, 04:47 PM
Thanks, all my interest in R1 has been since my Full Genome results came in and I wasn't really following the subject prior.

From my Full Genome results, I have a total number of 1,495 private and public mutations under R1 and 304 private SNP's and InDel's mutations I have under DF13.


The R1 possible time of origin 12,500–25,700 years BP (Karafet 2008) would be pushed to the higher end closer to 25.7K, then what does that fact suggest?

MJost

Michał
05-19-2014, 06:15 PM
From my Full Genome results, I have a total number of 1,495 private and public mutations under R1 and 304 private SNP's and InDel's mutations I have under DF13.
Instead of counting all SNPs from your FGC report, you should try to select only those mutations that are considered as relatively reliable, let's say those marked as "over 99% genuine" or "over 95% genuine", though some additional verification of those results is usually required to refine the calculations. Also, the vast majority of INDELs reported by FGC have no phylogenetic value, and although some INDELs are equally stable (and reliable) as some high quality SNPs, it seems to be much more secure to rely on the reliable SNPs only when performing any NGS-based age calculations.



The R1 possible time of origin 12,500–25,700 years BP (Karafet 2008) would be pushed to the higher end closer to 25.7K, then what does that fact suggest?

This means that most STR-based TMRCAs calculated using the so-called genealogical mutation rates are simply underestimates. Also, its seems that the older a given clade is the mored underestimated its STR-based TMRCA value is.

MJost
05-19-2014, 08:02 PM
Of my Full Genome Y-DNA R1 and below SNPs at 95 or greater % (one and no asterisk ranked) and after removed all INDel's, I have 185 (68 under DF13). 185 SNPs at 27.5kya produces 149 years per SNP.

Of my 68 SNPs under DF13, I have Sanger sequenced validated 25, just over a third at 37%. if this 37% holds across the entire Y, and of my 185 would be 68 SNPs back to R1 would be a large 404 years per SNP. Even at 50% validated the 93 SNPs would age to 295 years per SNP.

I have tested my closest GD's as posted else where. A GD3/111 went positive for 23 of 25. My GD8/111 went 22 of 25 positive. Where as my GD13/111 went 17 of 25.

Michał
05-19-2014, 10:47 PM
Of my Full Genome Y-DNA R1 and below SNPs at 95 or greater % (one and no asterisk ranked) and after removed all INDel's, I have 185 (68 under DF13). 185 SNPs at 27.5kya produces 149 years per SNP.
I haven't seen too many FGC reports for R1b, but the ones I've seen (mostly the early ones) were lacking many well known SNPs between the M343 and L11 levels, which has never been seen for the corresponding part of the R1a tree in the FGC-tested R1a members. Therefore, I suspect that the number of your high quality SNPs downstream of R1 (or rather between the R1b-M343 and R1b-L11 levels), as reported by FGC, is significantly underestimated.

However, the above shouldn't apply for your FGC-tested SNPs downstream of DF13. I have roughly estimated that each such FGC-tested SNP corresponds to about 88 years (or between 80 and 100 years), so 68 SNPs under DF13 would correspond to about 5984 years, though I should warn you that only by taking an average number of SNPs for at least several unrelated sublineages of DF13 (and comparing it with similar results received for some brother clades of DF13) would make such calculations fairly reliable.




Of my 68 SNPs under DF13, I have Sanger sequenced validated 25, just over a third at 37%. if this 37% holds across the entire Y, and of my 185 would be 68 SNPs back to R1 would be a large 404 years per SNP. Even at 50% validated the 93 SNPs would age to 295 years per SNP.
The fact that you have validated 25 out of 68 SNPs downstream of DF13 does not mean that the remaining 43 SNPs are not reliable, so I wouldn't use this information when making any SNP-based age calculations.



I have tested my closest GD's as posted else where. A GD3/111 went positive for 23 of 25. My GD8/111 went 22 of 25 positive. Where as my GD13/111 went 17 of 25.
This is indeed a very nice correlation.

MJost
05-20-2014, 01:14 PM
In my FGC R1b Variant Results file does not include M207 but does M173 and it states:

#citations for reference data include:
# 1000 Genomes Project: An integrated map of genetic variation from 1,092 human genomes, McVean et al., Nature 491, 56-65 (01 November 2012) doi: 10.11038/nature11632
# Personal Genome Project: Ball, Madeleine P., et al. A public resource facilitating clinical use of genomes. Proceedings of the National Academy of Sciences 109.30 (2012): 11920-11927. http://www.pnas.org/content/109/30/11920.long
# A High-Coverage Genome Sequence from an Archaic Denisovan Individual, Meyer et al. Science 338, 222-226 doi:10.1126/science.1224344
# R. Drmanac, et. al. Science 327(5961), 78. [DOI: 10.1126/science.1181498]

Counting my positive SNPs only, it shows 1,495 total SNP and the non-private SNP (at or above DF13 back to R1) total 1,186 (corrected)

My Variant File shows:

SNPs shared with reference samples (1031 total variants)
INDELs shared with reference samples (155 total variants)

Private SNPs (285 total variants)
Private INDELs (24 total variants)

MJost
05-20-2014, 01:22 PM
However, the above shouldn't apply for your FGC-tested SNPs downstream of DF13. I have roughly estimated that each such FGC-tested SNP corresponds to about 88 years (or between 80 and 100 years), so 68 SNPs under DF13 would correspond to about 5984 years, though I should warn you that only by taking an average number of SNPs for at least several unrelated sublineages of DF13 (and comparing it with similar results received for some brother clades of DF13) would make such calculations fairly reliable.

The fact that you have validated 25 out of 68 SNPs downstream of DF13 does not mean that the remaining 43 SNPs are not reliable, so I wouldn't use this information when making any SNP-based age calculations.

This is indeed a very nice correlation.

Of my 68 SNPs, one can not assume that the non-Sanger Validated SNPs would match in every close cousin in my tree even if you could Sanger sequence each of this that have cross over events. Even INDel's are problematic in some respect.

MJost

Michał
05-20-2014, 03:01 PM
Of my 68 SNPs, one can not assume that the non-Sanger Validated SNPs would match in every close cousin in my tree

The same can be said about the Sanger-validated SNPs, so there is not much difference in this respect.


even if you could Sanger sequence each of this that have cross over events.

What do you mean by that? The FGC-derived SNPs are from the non-recombinant regions of chromosome Y, so they are not subject to crossing-over.



Even INDel's are problematic in some respect.

I would rather say that INDELs are much more problematic than SNPs (at least on average).

MJost
05-20-2014, 03:19 PM
Ok, assume that an X-crossover event means that within a 500bp range around a specific Y position it is so similar to the Chr X that almost all reads will be mapped to the X chromosome.

MJost

Michał
05-20-2014, 03:20 PM
In my FGC R1b Variant Results file does not include M207 but does M173 and it states:

Does your list of the FGC-tested positive SNPs include for example M269, L23, L51 and L11? Their reference alleles are the derived ones, so these mutations are not shown in the variantCompare file that is sent to the FGC-tested R1b members (at least in all those results that have been available to me).

Michał
05-20-2014, 03:30 PM
Ok, assume that an X-crossover event means that within a 500bp range around a specific Y position it is so similar to the Chr X that almost all reads will be mapped to the X chromosome.

Such SNPs are very rarely reported as high quality SNPs, or at least they shouldn't be reported as such. For example, I know that YFull rejects all such doubtful cases that are sometimes reported by FGC.

MJost
05-20-2014, 03:50 PM
Does your list of the FGC-tested positive SNPs include for example M269, L23, L51 and L11? Their reference alleles are the derived ones, so these mutations are not shown in the variantCompare file that is sent to the FGC-tested R1b members (at least in all those results that have been available to me).
I will suggest that FGC runs utilities that test each major SNP from the Root to AO-t to the main terminal SNP haplogroup match analysis based on FGC yKNOT utility. Mine shows 11 positive SNPs down from R to DF13 below. So just add seven SNPs to the total which are the main SNPs from P312 back up to L278+, M415+. I would say that Variant file are produced without these as they are assumed with all the R1 reference datasets used to matchup with my NGS SNPs. Here is my YKnot file data. MJost

|Matches: DF13+
|____R1b1a2a1a2c
|Matches: L21+, L459+
|____R1b1a2a1a2
|Matches: S116+
|____R1b1a2a1a
|Matches: L11+, L52+, L151+, S129+, S128+
|____R1b1a2a1
|Matches: L51+
|____R1b1a2a
|Matches: L23+, L150.1+
|*No-calls: S349?
|____R1b1a2
|Matches: L265+, M269+, S3+, S10+
|*No-calls: M520?, S13?, S17?
|____R1b1a
|Matches: P297+
|*No-calls: L320?
|____R1b1
|Matches: L278+, M415+
|*No-calls: P25_1?, P25_2?, P25_3?
|____R1b
|Matches: M343+
|____R1
|Matches: M173+, S1+, P225+, P231+, P233+, P234+, P236+, P238+, P242+, P245+, P286+, P294+
|____R

Michał
05-20-2014, 04:07 PM
I would say that Variant file are produced without these as they are assumed with all the R1 reference datasets used to matchup with my NGS SNPs. Here is my YKnot file data. MJost

The problem is that your list of positive SNPs lacks a huge number of FGC-tested SNPs that are positioned between the R1 and R1-P312 levels. For example, you have written in one of your recent posts:


Of my Full Genome Y-DNA R1 and below SNPs at 95 or greater % (one and no asterisk ranked) and after removed all INDel's, I have 185 (68 under DF13).
Please compare it with about 300 high quality SNPs (under R1) reported for each R1a member. This clearly suggest that more than 100 SNPs are simply missing on your list (and these are most likely those positioned between the R1 and R1b-P312 levels). Thus, it would be simply unwise to use the numbers from your posts to make any reliable TMRCA estimates.

MJost
05-20-2014, 04:19 PM
The problem is that your list of positive SNPs lacks a huge number of FGC-tested SNPs that are positioned between the R1 and R1-P312 levels. For example, you have written in one of your recent posts:


Please compare it with about 300 high quality SNPs (under R1) reported for each R1a member. This clearly suggest that more than 100 SNPs are simply missing on your list (and these are most likely those positioned between the R1 and R1b-P312 levels). Thus, it would be simply unwise to use the numbers from your posts to make any reliable TMRCA estimates.
I am now assuming you are using your own FULLY filtering method to rank the R1a Full Genome BAM file results and not what was provided by FGC in their variant file?

MJost

Michał
05-20-2014, 04:38 PM
I am now assuming you are using your own FULLY filtering method to rank the R1a Full Genome BAM file results and not what was provided by FGC in their variant file?
Actually, I use both the original FGC reports and the YFull-filtered data. For example, the FGC reports for R1a are extremely useful for identifying most non-private SNPs under R1 (while the YFull reports include many unreliable non-private SNPs that are much more difficult to sort out, though I hope this will be improved very soon). On the other hand, I put much more trust in those private SNPs (and INDELs) that are reported by YFull (based on the BAM files produced by FGC), and it should be noted that the lists of private SNPs reported by FGC and YFull do not overlap in 100%.

MJost
05-20-2014, 05:06 PM
Using my BAM file with another Utility shows every SNP back to L138+ with 1,728 derived SNP positions associated with existing known SNP names. From another produced file with the same utility, I show 996 positions with ancestral Reference Genotype and my derived Genotype allele values without known SNP names which my novel and privates SNP position are found, and no INDel's positions appear to be listed.

The goal is to find NEW Terminal SNPs under my defined major subclade and work with those opportunities. I will continue to believe that FGC's 70-90 yps (years per SNP) estimate is acceptable at this time when comparing same source dataset results, as would be the prudent case for the BIGY yps. Building mixed source Trees may require a hybrid calculation with different yps numbers based on each NGS mutation rate using all or 'Gold' regions.

MJost

MJost
05-20-2014, 05:17 PM
Here is my closest YDNA STR kits showing their Kit numbers, 111 marker GD's and TIP calculations at 111 markers, compared to actual SGD's (SNP Genetic Distance) from me using tested my 25 Sanger Sequence validated SNPs.

have received all the SNP results, save one for 306/206731 which I assume he would be negative for, to create a SGD (SNP Genetic Distance) comparison using FtDNA's 111 Marker TIP and number of SNP differences. See attached PDF.

Watterson b. USA - 263/316063 is a SGD=2,

Codere b.Isle of Man - 306/206731 is a SGD=3 and

Ross - MDKA Caithness, Scotland - 113/155812 is a SGD=9.

The TIP yellow cell color show the probability percentages that are in the range of 50 to 68% (1-sigma) that I preferred to use in comparing Kits and it appears close when you follow the rows into the SGD column.

Ross has 17 SNPs out of 25 of mine.
Codere (Watterson) has 22 SNPs and
Watterson has 23 out of 25.

The comparison chart showing a nice relationship between STRs and SNPs. I am a predicted 3rd Cousin to 263/316063's father's sister* on 23andme as well.

A chart can be seen here:

https://drive.google.com/file/d/0By9Y3jb2fORNWkdIUlRjYTlKQWs/edit?usp=sharing

MJost

Mikewww
05-20-2014, 05:54 PM
Here is my closest YDNA STR kits showing their Kit numbers, 111 marker GD's and TIP calculations at 111 markers, compared to actual SGD's (SNP Genetic Distance) from me using tested my 25 Sanger Sequence validated SNPs.

have received all the SNP results, save one for 306/206731 which I assume he would be negative for, to create a SGD (SNP Genetic Distance) comparison using FtDNA's 111 Marker TIP and number of SNP differences. See attached PDF.

Watterson b. USA - 263/316063 is a SGD=2,

Codere b.Isle of Man - 306/206731 is a SGD=3 and

Ross - MDKA Caithness, Scotland - 113/155812 is a SGD=9.

The TIP yellow cell color show the probability percentages that are in the range of 50 to 68% (1-sigma) that I preferred to use in comparing Kits and it appears close when you follow the rows into the SGD column.

Ross has 17 SNPs out of 25 of mine.
Codere (Watterson) has 22 SNPs and
Watterson has 23 out of 25.

The comparison chart showing a nice relationship between STRs and SNPs. I am a predicted 3rd Cousin to 263/316063's father's Aunt on 23andme as well.

A chart can be seen here:

https://drive.google.com/file/d/0By9Y3jb2fORNWkdIUlRjYTlKQWs/edit?usp=sharing

MJost

Mark, you are doing great work. I somewhat miss the idea that there is less time for ancient haplogroup relationships and origin speculations, but I think more and more, we'll be doing the kind of true genetic genealogy research that you are doing.

BTW, I'm now counting 209 distinct SNP marked public branches under L21. I'm at 20 branches directly from DF13 with the vast, vast majority (almost all) of the 209 hanging below those 20. That 20 will have to become more numerous because there are many, many DF13* clusters that not done NGS testing. Even with some potential for consolidation among the smaller branches of DF13, I don't see how this won't grow.

In a sense, larger than L21, but all roads don't go back to Rome. They go to the WAMH MRCA ... must have been an L11 person.

MJost
05-20-2014, 09:47 PM
I will also add that the first male to produce children in the colonies born 1744 County Down, Ire. and arrived Virginia in 1764 is 10 generations back for me. Using aDNA predicted 4th cousin back to a common 10 generation ancestor who first arrived in the colonies in 1764 equates to four generations per SNP having two SGDs from me.

MJost

alan
05-20-2014, 09:56 PM
What are the average number of SNPs between today and P312? I mean in a linear sense not branching.

I think that is a crucial question as P312 is the last shared SNP of so much of R1b of western, central and northern Europe.

seferhabahir
05-20-2014, 10:14 PM
However, the above shouldn't apply for your FGC-tested SNPs downstream of DF13. I have roughly estimated that each such FGC-tested SNP corresponds to about 88 years (or between 80 and 100 years), so 68 SNPs under DF13 would correspond to about 5984 years, though I should warn you that only by taking an average number of SNPs for at least several unrelated sublineages of DF13 (and comparing it with similar results received for some brother clades of DF13) would make such calculations fairly reliable.


Mark's count of 68 SNPs under DF13 corresponds pretty well with my count of 68 SNPs total under DF13 (60 FGC private SNPs, 4 that appear to be recurrent but have not appeared in any other DF13 yet, one private INDEL, and then of course Z251, S11556, and S9294 below DF13). So there you go, another unrelated DF13 sublineage with about the same number. Perhaps not coincidence, and lends some support to the calculation of DF13 age, and therefore also P312 and L21.

MJost
05-20-2014, 10:26 PM
Mark, you are doing great work. I somewhat miss the idea that there is less time for ancient haplogroup relationships and origin speculations, but I think more and more, we'll truly entered in the kind of research of true genetic genealogy that you've entered.

BTW, I'm now counting 209 distinct SNP marked public branches under L21. I'm at 20 branches directly from DF13 with the vast, vast majority (almost all) of the 209 hanging below those 20. That 20 will have to become more numerous because there are many, many DF13* clusters that not done NGS testing. Even with some potential for consolidation among the smaller branches of DF13, I don't see how this won't grow.

In a sense, larger than L21, but all roads don't go back to Rome. They go to the WAMH MRCA ... must have been an L11 person.
As you see I am working backwards from my 25 SNPs starting with the 5496-A1A's and have completed this stage. Using STR GD as a guide we can now make decisions as to which group of the 25 SNPs to test into. With 98213 an -A1 now positive for FGC5496, he stated he is interested in testing into the SNPs above Ross with the FGC5496>FGC5521>FGC5539 called FGC5496-A1 panel.

Anyone who is in the 5495-A1 variety should test for FGC5496 and then, after going positive, test into the FGC5496-A1 panel. I have 40 plus guys in my private list, so the there is a solid base. There are others who are currently testing for FGC5496.

Trying to make some progress. Thanks.

MJost

seferhabahir
05-20-2014, 10:27 PM
What are the average number of SNPs between today and P312? I mean in a linear sense not branching.

I think that is a crucial question as P312 is the last shared SNP of so much of R1b of western, central and northern Europe.

Well, take the number 68 and then add in P312, L21, L459, and Z290 and we have something between 5,000 and 6,500 years ago (depending on whether you like 70 or 90 years per SNP). Works fine for your Beaker spread hypothesis.

MJost
05-20-2014, 10:55 PM
Mark's count of 68 SNPs under DF13 corresponds pretty well with my count of 68 SNPs total under DF13 (60 FGC private SNPs, 4 that appear to be recurrent but have not appeared in any other DF13 yet, one private INDEL, and then of course Z251, S11556, and S9294 below DF13). So there you go, another unrelated DF13 sublineage with about the same number. Perhaps not coincidence, and lends some support to the calculation of DF13 age, and therefore also P312 and L21.

Thanks, another confirmation.

I believe the 'no astrisk' (99%) ranking FGC SNP are mostly validating and the single asterisk ranked SNP are much lower confidence produced a lower number of validated 'Clean' SNPs but close to the average of the BigY's Gold location SNPs. We need to have a standard for trees that combine various sourced SNPs.

MJost

seferhabahir
05-22-2014, 12:29 AM
Chris McCown posted in another forum the following statistics:

FGC Duffy kit has 63 (*) or better better SNPs downstream of DF41 (including DF41 and equivalents).

FGC McCown kit has 61 (*) or better SNPs downstream of DF41(including DF41 and equivalents).

FGC Walker kit has 62 (*) or better SNPs downstream of DF41 (including DF41 and equivalents).

More evidence that there are multiple DF13 subclades having more than 60 downstream SNPs, assuming these are mostly single point mutations, and not INDELs.

alan
05-22-2014, 04:31 PM
I think that is a very believable date range for P312. Intriguing too as only at the youngest end of that sort of date range would P312 potentially being entirely or almost entirely chronologically within the beaker period. For most of the range of 5000-6500 years ago it would indicate that P312 had a pre-beaker existence as might its main early branches.


Well, take the number 68 and then add in P312, L21, L459, and Z290 and we have something between 5,000 and 6,500 years ago (depending on whether you like 70 or 90 years per SNP). Works fine for your Beaker spread hypothesis.

alan
05-22-2014, 04:42 PM
Full y testing of radiocarbon dated ancient remains is surely the best way we can refine y DNA dating. It seems to me we wouldnt need an awful lot of R1b guys with full y testing from across the prehistoric period to gain a lot of confidence in SNP dating.


I'd say so however the advantage with the Mal'ta results is they also come with radiocarbon dating.

jdean
05-22-2014, 06:59 PM
Full y testing of radiocarbon dated ancient remains is surely the best way we can refine y DNA dating. It seems to me we wouldnt need an awful lot of R1b guys with full y testing from across the prehistoric period to gain a lot of confidence in SNP dating.

A way of normalizing all the results would also need to be devised, but I don't see that would be insurmountable.

alan
05-22-2014, 07:01 PM
What are the main studies suggesting how many years between SNPs on the y chromosome? I understand that there is some variation in opinion. In layman's terms what are the various suggested average intervals?


Well, take the number 68 and then add in P312, L21, L459, and Z290 and we have something between 5,000 and 6,500 years ago (depending on whether you like 70 or 90 years per SNP). Works fine for your Beaker spread hypothesis.

seferhabahir
05-22-2014, 07:35 PM
What are the main studies suggesting how many years between SNPs on the y chromosome? I understand that there is some variation in opinion. In layman's terms what are the various suggested average intervals?

I'm just using Mark's recent quote "The Experts have deemed from studies that the range is 70-90 years per SNP mutation." I think Michal uses 88 years per SNP. Some would say it's three generations per SNP. Perhaps Mark or Michal or Warwick can elaborate on these ranges or point to specific studies. I'm assuming these reanges are for FGC high reliability SNPs, not Big Y or YFULL.

alan
05-22-2014, 08:06 PM
Thanks. That is kind of how I understand it and it seems to make sense but I have some vague memory that a paper by Wei or Zhao or someone estimated a lower amount of years per generation. I may be wrong but I vaguely remember something like that.


I'm just using Mark's recent quote "The Experts have deemed from studies that the range is 70-90 years per SNP mutation." I think Michal uses 88 years per SNP. Some would say it's three generations per SNP. Perhaps Mark or Michal or Warwick can elaborate on these ranges or point to specific studies. I'm assuming these reanges are for FGC high reliability SNPs, not Big Y or YFULL.

alan
05-22-2014, 08:10 PM
The big question I wonder is just how rare is the sort of preservation of ancient yDNA so a full y sort of testing can take place? Was a rare case and down to the incredibly cold and dry climate around Lake Baikal or is there hope this might be repeated frequently?


A way of normalizing all the results would also need to be devised, but I don't see that would be insurmountable.

seferhabahir
05-22-2014, 08:15 PM
Thanks. That is kind of how I understand it and it seems to make sense but I have some vague memory that a paper by Wei or Zhao or someone estimated a lower amount of years per generation. I may be wrong but I vaguely remember something like that.

There is a paper from 2009 that you might be remembering (four measured mutations in 13 generations)...

Human Y Chromosome Base-Substitution Mutation Rate Measured by Direct Sequencing in a Deep-Rooting Pedigree

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2748900/

alan
05-22-2014, 08:27 PM
Yes it does seem that it would be interesting to average the no of SNPs below P312 of all full y tested P312 derived modern people i.e U152, DF27, L21 etc people. I imagine that might be the best way to work out a more solid average figure for P312. I imagine people have posted some info on this but it would be scattered around a lot of different threads and forums and am struggling to find a summary.

alan
05-22-2014, 09:18 PM
Thanks again. I suppose that is pretty well not far off the same sort of ballpark depending on generation length.

Generation length figures are hard to work out. I personally think they varied as time went on and of course actual survival to breeding age may have meant that the first son frequently didnt leave genes and of course half of children born were female. So, it seems to me that we should not expect in any ancient society that the average father to son span commenced at the starting of his relations with women. I expect in very simple societies the length of male generations in a surviving line might have been low around 20 but I suspect it increased to mid 20s as farming and other advances took hold. However, it would be very hard to produce an all time average father to son span for generations for humanity. It probably slowly increased but its likely not as simple as that.

Thankfully at least when just looking at P312 there is general agreement it is post-dates the early Neolithic. That helps a bit. It means that little if any of the time related to hunter-gatherers and it is also likely that it post-dates the pioneer phases of the farming spread across most of Europe and SW Asia. That sort of sets it in a later Neolithic and copper age context in its early days. It sort of puts its early context into the sort of copper and Bronze Age societies.

I think by that period archaeology indicates and the very shape of P312 and downstream tends to confirm that a hierarchical society of varying levels had developed across most of Europe and there was an out of proportion impact of chiefs in terms of breeding and surviving children due to hogging of females and resources. Wealthier men who may have lived longer, bred a lot longer with serial wives as well as a lot of other breeding partners may have raised the average number of years between a father and any given surviving son as he may have had several born across a decade or two. I suspect that those sort of chiefdom societies may have dragged up the average amount of years between the father and the surviving sons quite a bit - maybe mid 20s or higher.

A lot of the P312 world slowly moved to something closer to a market economy and proto-urban living in the last two centuries BC and then was Romanised. This may have altered the social structure and breeding patterns. That could have brought an end to the Celtic mega breeding chiefs such as survived only in places Ireland, Scotland and Wales after the Romans. Its hard to generalise for the last 2000 years as different areas of the P312 world would have had different social structure that sometimes would promote young marriages and sometimes deter them and of course infant mortality levels would have an impact on the average age of a surviving son. It is something that definately could be subject to studies of historical sources, ancient cemeteries etc.

My guess for the average father to son who survived to reproduce generation span since P312 occurred to today may be around 25 but its just a guess. If that is around the right figure and there is an SNP every 3.5 generations then that would be one every 87.5 years. If there are around 70 SNPs on average since P312 in modern full y testers then that would make P312 around 6125 years ago. The same sort of ballpark. I think getting as good an average for no of SNP since P312 in modern full y testers would probably be the most important variable to refine as I dont think the generations length thing can ever be fully resolved.


There is a paper from 2009 that you might be remembering (four measured mutations in 13 generations)...

Human Y Chromosome Base-Substitution Mutation Rate Measured by Direct Sequencing in a Deep-Rooting Pedigree

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2748900/

Heber
05-22-2014, 11:07 PM
Thanks. That is kind of how I understand it and it seems to make sense but I have some vague memory that a paper by Wei or Zhao or someone estimated a lower amount of years per generation. I may be wrong but I vaguely remember something like that.

A calibrated human Y-chromosomal phylogeny based on resequencing
Wei Wei, Qasim Ayub, [...], and Chris Tyler-Smith

In all five estimates (two GENETREE, one BEAST, two rho), calibration was achieved using the directly measured SNP mutation rate of 1.0 × 10−9 mutations per nucleotide per year or 3.0 × 10−8 mutations per nucleotide per generation (Xue et al. 2009). Where necessary, we converted estimates in generations to estimates in years using 30 yr per generation.

1888

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3561879/#!po=67.3077

Heber
05-22-2014, 11:34 PM
The shape of the three expansions under P312 (L2, DF27, DF13) can be seen here:

1891
1889
1892

The expansion of DF27 is the most extreme.

http://biorxiv.org/content/biorxiv/early/2013/11/22/000802.1.full.pdf

L2
http://www.pinterest.com/gerardcorcoran/r1b-u152-l2/

DF27
http://www.pinterest.com/gerardcorcoran/r1b-df27/

DF13
http://www.pinterest.com/gerardcorcoran/r1b-l21/

M269
http://www.pinterest.com/gerardcorcoran/r1b-m269/

The questions is did they happen in the sequence implied by the Phylogenetic Tree and why is the expansion of DF27 so extreme and what does it imply?

alan
05-23-2014, 12:03 AM
I did notice a post somewhere on this forum that seemed to be suggesting a fully y U152 guy had around 70 SNPs below U152. I think it seems increasingly likely that P312 cannot be younger than 5000 years even when lower calculations like 20 years per generation an SNP per 70 years are used. I personally would tend to think that both those would be too low so it does raise the interesting possibility that P312 had at least a few centuries of a pre-beaker existence somewhere and naturally that would mean L11 and L51 are some centuries older still. That is potentially significant as they are pretty well also known only from west and central Europe, suggestive that R1b had a pre-beaker but post-first farmers existence in that zone having arrived at some point between 6500 and perhaps 5200 years ago. Its strongly held by most people these days that the deeper roots of R1b above L51 almost certainly lay in SE Europe, eastern Europe, Anatolia or SW Asia. It does seem that there is something of a watershed between L51 and derived and L23xL51, M269. So, ideally we should be looking at L51 crossing that watershed at some point between 4500 and 3000BC.

Mikewww
05-23-2014, 03:07 AM
The shape of the three expansions under P312 (L2, DF27, DF13) can be seen here:
...
The expansion of DF27 is the most extreme.
....
The questions is did they happen in the sequence implied by the Phylogenetic Tree and why is the expansion of DF27 so extreme and what does it imply?

What do you mean by most "extreme"? Are you talking about geographic spread? or early branching? population growth? or something else?

MJost
05-23-2014, 04:21 AM
I'm just using Mark's recent quote "The Experts have deemed from studies that the range is 70-90 years per SNP mutation." I think Michal uses 88 years per SNP. Some would say it's three generations per SNP. Perhaps Mark or Michal or Warwick can elaborate on these ranges or point to specific studies. I'm assuming these reanges are for FGC high reliability SNPs, not Big Y or YFULL.
I already posted this but here it is again
https://drive.google.com/file/d/0By9Y3jb2fORNWkdIUlRjYTlKQWs/edit?usp=sharing

Out of my 25 SNPs, my GD3/111 had 23 positive, a SGD (SNP Genetic Distance) of 2. As a predict 3rd cousin on 23andme for his father's sister, has 10 generations back to the first male into the Virginia Colonies. My GD8/111 tested a SGD3 and has to be at least 11 generations but I suspect 12-13 generations back to a common ancestor. Then my GD13/111 is a SGD9. So depending on the number of years per generation, from b. 1744 to present (1950) at 10 Gens is 3.3 generations per mutation, that covered about 206 years which is about 20.5 year per gen. Just rough numbers.

MJost

Heber
05-23-2014, 08:06 AM
What do you mean by most "extreme"? Are you talking about geographic spread? or early branching? population growth? or something else?

Mike,

By looking at the trees above it is clear that the branching of DF27 and by extension population growth is more extreme than that of L2 or DF13 and this is the most rapid expansion within R1b and that R1b itself is the most rapid expansion within the entire Phylogenetic Tree as identified by Tyler Smith, Wei and Xue.

"Next-generation sequence data suggests a "rapid" and "extreme" expansion of R1b across Europe during the Neolithic."

"The tree was consistent with the established structure. It confirmed Hg E (Bantu), O (China) and R1b (Europe) expansions associated with the Neolithic transitions in different parts of the world, and revealed that the expansion in Europe was the most extreme."

What was identified by Tyler Smith, Wei and Xue as the extreme R1b expansion was is fact largely due to the subsequent expansion of L2, DF27 and Df13.

http://eurogenes.blogspot.ie/2012/09/next-generation-resequencing-data.html

This same expansion is identified by Patterson in his autosomal analysis.

"Ancient Admixture in Human History"

"We have detected here a signal of gene flow from northern Europe into Spain around 2000 B.C. We discuss a likely interpretation. At this time there was a characteristic pottery termed ‘bell-beakers’ believed to correspond to a population spread across Iberia and northern Europe. We hypothesize that we are seeing here a genetic signal of the ‘Bell-Beaker culture’ (HARRISON, 1980). Initial cultural flow of the Bell-Beakers appears to have been from South to North, but the full story may be complex. Indeed one hypothesis is that after an initial expansion from Iberia there was a reverse flow back to Iberia (CZEBRESZUK, 2003); this ‘reflux’ model is broadly concordant with our genetic results, and if this is the correct explanation it suggests that this reverse flow may have been accompanied by substantial population movement."

http://www.genetics.org/content/early/2012/09/06/genetics.112.145037.full.pdf

Again what Patterson detects as a Bell Beaker expansion from Iberia (and reflux) is supported by the rapid expansion of DF27 which has its highest frequency in Iberia.

alan
05-23-2014, 11:36 AM
If human genes are to be literally linked to beakers and an out of Iberia model for both P312 and the pots, the big challenge is to explain how P312 got INTO Iberia. There seems to be no doubt that R1b within the main body of Europe is a late arrival. Certainly ancient DNA has consistently failed to pick it up in pre-beaker times and much about R1b points to it being an east to west movement. SNP counting also seems to suggest that the west European forms of R1b wouldnt have existed until somewhere between 4500 and 3000BC. This is long after Cardial farming wave spread c 6500-5500BC all along the Med. and indeed the various Med. or near-Med. Neolithic and copper age people tested for ancient DNA of pre-beaker date are all not R1b.

So, it is pretty clear that R1b seems to have entered western Europe long after the early farming waves and so far there is no evidence of it at all pre-3000BC in western Europe. So, my conclusion is that it moved into the area at some point c. 4500-3000BC and that is a bigger mystery than a secondary movement in the beaker period. In order to spread with beakers R1b had to get into western Europe in pre-beaker times.

I have a personal theory on this that P312 entered Iberia with the rare all over corded AOC/all over ornamented AOO beakers which do seem early in Iberia but yet have a very different distribution compared to Maritime. These AOO/AOC beakers are distributed more around the coasts and much more eastern weighted than Maritime although they do partly overlap with the maritime beakers. http://repositorio-aberto.up.pt/bitstream/10216/20435/2/sojorgeallover000085668.pdf

My hunch is that the mistake has been to set a AOO Dutch origin vs an Iberian origin model fight for beaker. I think a better model would be that the roots of bell beakers stems from AOO/AOC beaker which was developed by people to the south-west of the corded ware zone through contact with corded ware through the west Alpine valleys to Liguria and the west Med c. 2800BC. The AOO/AOC beakers in Iberia have some early dates and a distribution that is strongest around the Pyrennes and lower Ebro etc in eastern Spain. They are also known in lesser numbers along the Iberia coasts and to a lesser degree inland and do overlap with maritime. I think if there is any evidence of an INTO Iberia root for beaker pot this is probably it and its highly unlikely that a rare type of pot like this in Iberia has been RC dated sufficiently to be conclusive. From memory I think Iberian AOO beaker has been linked to shafthole axe distribution too - another possible indicator of influences from he east. It does feel to me there is a hint of potential low level contacts between Iberia and the corded ware world c. 2800BC but that there is a geographical missing link in between. I suspect the missing link is somewhere around Liguria and the Rhone area where the people using Remedello symbolism's mines were running out and contacts with Iberia would have been desirable.


Mike,

By looking at the trees above it is clear that the branching of DF27 is more extreme than that of L2 or DF13 and this is the most rapid expansion within R1b and that R1b itself is the most rapid expansion within the entire Phylogenetic Tree as identified by Tyler Smith, Wei and Xue.

"Next-generation sequence data suggests a "rapid" and "extreme" expansion of R1b across Europe during the Neolithic."

"The tree was consistent with the established structure. It confirmed Hg E (Bantu), O (China) and R1b (Europe) expansions associated with the Neolithic transitions in different parts of the world, and revealed that the expansion in Europe was the most extreme."

What was identified by Tyler Smith, Wei and Xue as the extreme R1b expansion was is fact largely due to the subsequent expansion of L2, DF27 and Df13.

http://eurogenes.blogspot.ie/2012/09/next-generation-resequencing-data.html

This same expansion is identified by Patterson in his autosomal analysis.

"Ancient Admixture in Human History"

"We have detected here a signal of gene flow from northern Europe into Spain around 2000 B.C. We discuss a likely interpretation. At this time there was a characteristic pottery termed ‘bell-beakers’ believed to correspond to a population spread across Iberia and northern Europe. We hypothesize that we are seeing here a genetic signal of the ‘Bell-Beaker culture’ (HARRISON, 1980). Initial cultural flow of the Bell-Beakers appears to have been from South to North, but the full story may be complex. Indeed one hypothesis is that after an initial expansion from Iberia there was a reverse flow back to Iberia (CZEBRESZUK, 2003); this ‘reflux’ model is broadly concordant with our genetic results, and if this is the correct explanation it suggests that this reverse flow may have been accompanied by substantial population movement."

http://www.genetics.org/content/early/2012/09/06/genetics.112.145037.full.pdf

Again what Patterson detects as a Bell Beaker expansion from Iberia (and reflux) is supported by the rapid expansion of DF27 which has its highest frequency in Iberia.

George Chandler
05-23-2014, 08:24 PM
I already posted this but here it is again
https://drive.google.com/file/d/0By9Y3jb2fORNWkdIUlRjYTlKQWs/edit?usp=sharing

Out of my 25 SNPs, my GD3/111 had 23 positive, a SGD (SNP Genetic Distance) of 2. As a predict 3rd cousin on 23andme for his father's sister, has 10 generations back to the first male into the Virginia Colonies. My GD8/111 tested a SGD3 and has to be at least 11 generations but I suspect 12-13 generations back to a common ancestor. Then my GD13/111 is a SGD9. So depending on the number of years per generation, from b. 1744 to present (1950) at 10 Gens is 3.3 generations per mutation, that covered about 206 years which is about 20.5 year per gen. Just rough numbers.

MJost

The Sanger results have all been confirmed from my FGC testing for all 33 SNP's below DF13. The Hamilton kit had a GD18/111 and is SGD6 though it's likely the MRCA is from the Tower Ward of London about 1200 AD though that MRCA isn't certain unless it is somehow triangulated with others. The STR comparison fits and if that's the case it's 129 years per verified SNP from then to the time of my birth. I will be able to identify how many of the six SNP's are ancestral when the Chandler beta results are returned from the other cousins line with the MRCA being 12 generations back. If I use only Sanger verified SNP's it puts the age of DF13 at ~4,100 years. If you use all 58 FGC SNP's below DF13 in my kit that's 7,482 years for DF13. Obviously the exact minimum age won't be determined unless skeletal remains are tested and carbon dated.

George

Heber
05-23-2014, 10:21 PM
Mike,

By looking at the trees above it is clear that the branching of DF27 and by extension population growth is more extreme than that of L2 or DF13 and this is the most rapid expansion within R1b and that R1b itself is the most rapid expansion within the entire Phylogenetic Tree as identified by Tyler Smith, Wei and Xue.

"Next-generation sequence data suggests a "rapid" and "extreme" expansion of R1b across Europe during the Neolithic."

"The tree was consistent with the established structure. It confirmed Hg E (Bantu), O (China) and R1b (Europe) expansions associated with the Neolithic transitions in different parts of the world, and revealed that the expansion in Europe was the most extreme."

What was identified by Tyler Smith, Wei and Xue as the extreme R1b expansion was is fact largely due to the subsequent expansion of L2, DF27 and Df13.

http://eurogenes.blogspot.ie/2012/09/next-generation-resequencing-data.html

This same expansion is identified by Patterson in his autosomal analysis.

"Ancient Admixture in Human History"

"We have detected here a signal of gene flow from northern Europe into Spain around 2000 B.C. We discuss a likely interpretation. At this time there was a characteristic pottery termed ‘bell-beakers’ believed to correspond to a population spread across Iberia and northern Europe. We hypothesize that we are seeing here a genetic signal of the ‘Bell-Beaker culture’ (HARRISON, 1980). Initial cultural flow of the Bell-Beakers appears to have been from South to North, but the full story may be complex. Indeed one hypothesis is that after an initial expansion from Iberia there was a reverse flow back to Iberia (CZEBRESZUK, 2003); this ‘reflux’ model is broadly concordant with our genetic results, and if this is the correct explanation it suggests that this reverse flow may have been accompanied by substantial population movement."

http://www.genetics.org/content/early/2012/09/06/genetics.112.145037.full.pdf

Again what Patterson detects as a Bell Beaker expansion from Iberia (and reflux) is supported by the rapid expansion of DF27 which has its highest frequency in Iberia.

Andy Grierson gave an excellent talk at WDYTYA which addressed this topic.
Here are some of the slides from his presentation which is on YouTube.
What surprised me is that 28 Parallel Branches were identified under DF13 up from 11 established branches in the UK10K project. This indicates that we have a long way to go in identifying the full phylogeny of L2, DF27 and DF13.
The detailed paper from the long overdue Mayo and Asturias Genographic project should also help to flesh out Df13 and DF27.

1894
1895
1896
1897

http://youtu.be/rzgP_jDByh0

Mikewww
05-23-2014, 10:51 PM
The shape of the three expansions under P312 (L2, DF27, DF13) can be seen here:

1891
1889
1892

The expansion of DF27 is the most extreme.
....
The questions is did they happen in the sequence implied by the Phylogenetic Tree and why is the expansion of DF27 so extreme and what does it imply?

Gerard, what is the data specifically that you are using to propose that DF27 had the most rapid (I think what you are saying by "extreme") expansion?

I would think we'd need an SNP "depth" chart where blocks of phylogenetic equivalents are shown to demonstrate the speed of expansion at a particular level. I haven't really looked. Are our NGS results showing that there are no SNPs between DF27 and Z196 and its brothers or that just downstream of Z196 there is a wide flat diversity of branching? I'm not saying DF27 didn't have the most rapid early growth. I am just looking to see how we determine that.

Have we ever found equivalents to D27? I don't think so. How about for DF19? or U152? or Z196? It is my understanding no equivalents can be found for L2.

Mikewww
05-23-2014, 11:02 PM
If human genes are to be literally linked to beakers and an out of Iberia model for both P312 and the pots, the big challenge is to explain how P312 got INTO Iberia. There seems to be no doubt that R1b within the main body of Europe is a late arrival. Certainly ancient DNA has consistently failed to pick it up in pre-beaker times and much about R1b points to it being an east to west movement. SNP counting also seems to suggest that the west European forms of R1b wouldnt have existed until somewhere between 4500 and 3000BC. This is long after Cardial farming wave spread c 6500-5500BC all along the Med. and indeed the various Med. or near-Med. Neolithic and copper age people tested for ancient DNA of pre-beaker date are all not R1b.

So, it is pretty clear that R1b seems to have entered western Europe long after the early farming waves and so far there is no evidence of it at all pre-3000BC in western Europe. So, my conclusion is that it moved into the area at some point c. 4500-3000BC and that is a bigger mystery than a secondary movement in the beaker period. In order to spread with beakers R1b had to get into western Europe in pre-beaker times.

I have a personal theory on this that P312 entered Iberia with the rare all over corded AOC/all over ornamented AOO beakers which do seem early in Iberia but yet have a very different distribution compared to Maritime. These AOO/AOC beakers are distributed more around the coasts and much more eastern weighted than Maritime although they do partly overlap with the maritime beakers. http://repositorio-aberto.up.pt/bitstream/10216/20435/2/sojorgeallover000085668.pdf

My hunch is that the mistake has been to set a AOO Dutch origin vs an Iberian origin model fight for beaker. I think a better model would be that the roots of bell beakers stems from AOO/AOC beaker which was developed by people to the south-west of the corded ware zone through contact with corded ware through the west Alpine valleys to Liguria and the west Med c. 2800BC. The AOO/AOC beakers in Iberia have some early dates and a distribution that is strongest around the Pyrennes and lower Ebro etc in eastern Spain. They are also known in lesser numbers along the Iberia coasts and to a lesser degree inland and do overlap with maritime. I think if there is any evidence of an INTO Iberia root for beaker pot this is probably it and its highly unlikely that a rare type of pot like this in Iberia has been RC dated sufficiently to be conclusive. From memory I think Iberian AOO beaker has been linked to shafthole axe distribution too - another possible indicator of influences from he east. It does feel to me there is a hint of potential low level contacts between Iberia and the corded ware world c. 2800BC but that there is a geographical missing link in between. I suspect the missing link is somewhere around Liguria and the Rhone area where the people using Remedello symbolism's mines were running out and contacts with Iberia would have been desirable.

There is a new R1b project that is focused on DF100 CTS4528. I'm trying to get them highlight that it is for L11+(P310,P311) and P312- U106- people. This will be interesting to follow.

https://www.familytreedna.com/public/DF100-CTS4528-L11-P310-L151-P311/default.aspx?section=yresults

Heber
05-24-2014, 08:17 AM
Gerard, what is the data specifically that you are using to propose that DF27 had the most rapid (I think what you are saying by "extreme") expansion?

I would think we'd need an SNP "depth" chart where blocks of phylogenetic equivalents are shown to demonstrate the speed of expansion at a particular level. I haven't really looked. Are our NGS results showing that there are no SNPs between DF27 and Z196 and its brothers or that just downstream of Z196 there is a wide flat diversity of branching? I'm not saying DF27 didn't have the most rapid early growth. I am just looking to see how we determine that.

Have we ever found equivalents to D27? I don't think so. How about for DF19? or U152? or Z196? It is my understanding no equivalents can be found for L2.

Mike,

On Andy Grierson's chart we have
L2: 8
DF13: 11
DF27: 18
when the draft paper was originally published.
Visual inspection of their draft tree shows DF27 with the most rapid expansion.
I borrowed the qualifier extreme from the CTS, Wei, Xue paper.
I will revert to rapid at your suggestion.:).
Of course these numbers are changing as we make new discoveries with NGS.
For example the UK10K project predicts 28 branches under DF13.

1898
1900

The latest draft trees from L21 (Mike Walsh) and U152 (Rich Rocca), are:

1899

1901

and Alex Williamson's draft tree shows there is still a long way to go.

1902

http://www.littlescottishcluster.com/RL21/NGS/Tree.html

alan
05-24-2014, 10:29 AM
Going back to the main topic of the thread, I think it is probably fair to say that something of the relative expansion pattern can be deduced from the number of parallel SNP defined branches that share the same next upstream SNP with nothing intervening. I think this sort of approach probably will allow soon lead to some more confident understanding of the relative expansion histories of the branches.

It can be seem on the L21 line that there would seem to be a slight delay in its expansion as its not until the SNP level one below DF13 that there is a major take off. This kind of fits the NWern distribution of L21 and the c. 3-400 years delay between origin to maximum distribution and the NW was one of the last areas in Europe to receive beakers and copper.

Ireland for example probably received beakers and copper working not long before 2400BC and certainly after 2500BC. I am skeptical of the very earliest dates for beakers but I would be confident that the European beaker phenomenon had commenced by say 2800BC so beaker in Ireland was at least 300 years behind the beaker origin dates. So, I suppose you would think in SNP terms that local beaker expansion should be associated with SNP defined branches maybe 3-5 SNPs downstream from the earliest beaker dates if beaker had always been associated with P312. You could say that Irish L21 starts in the big generation just below DF13. Much of that generation of SNPs is shared with Britain and may not have originated in Ireland (as far as I understand the British beaker dates commence a little earlier than in Ireland although only by a couple of generations). So, its probably best to just treat the isles together.

So, if there is a close correspondence between beaker arriving in the isles c. 2500-2400BC and the big expansion of largely isles dominated L21 branches one SNP down from DF13 that might be one way of implying a date for that particular big branching generation. Its as close as I can think of an archaeologically derived date for a particular point in time for the L21 story.

The potential usefulness of this to the wider P312 story of such an archaeological dating is that by counting back SNPs from that generation to P312 it provides another possible way of arriving at a date for P312. As far as I understand that count would involve DF13, L21 and P312 - I am not clear on the other less well known SNPs within that range and which ones would be counted. In archaeological terms something between 3-500 years worth of SNPs would not be surprising between P312 and the first big branching below DF13 if the beaker model is correct.

jdean
05-24-2014, 01:35 PM
The potential usefulness of this to the wider P312 story of such an archaeological dating is that by counting back SNPs from that generation to P312 it provides another possible way of arriving at a date for P312. As far as I understand that count would involve DF13, L21 and P312 - I am not clear on the other less well known SNPs within that range and which ones would be counted. In archaeological terms something between 3-500 years worth of SNPs would not be surprising between P312 and the first big branching below DF13 if the beaker model is correct.

500 yrs makes sense, L459, Z245, Z260 & Z290 look to be phylogenetically equivalent to L21 and likewise Z2542 WRT DF13, though Z2542 doesn't play well with Big Y however the rejected results coming back certainly suggest this is the case.

That gives 7 SNPs after P312 to DF13.

alan
05-24-2014, 03:15 PM
My general feeling is 25 years between a father and surviving son who produced male progeny is the safest estimate for clades of the last 6000 years and the estimate of 3 generations per SNP seems popular. I tend therefore to favour around 75 years per SNP or more. I wouldnt be surprised from posts I have read if an average of 65 SNPs below DF13 in modern full y testers is in the right ballpark. That would come to 4875 years ago or around 2800BC. Given the amount of uncertains that is close enough to an age of around 2500BC I would feel would be the sort of date we should arrive at if DF13's explosion into so many one-down from DF13 branches which are isles dominated. As I image DF13 originated on the continent and it was the next step down which populated the isles with its many branches, the arrival in the isles may have been around a century after DF13 came into existence so that might suggest a date of around 2600BC may be a good guess for DF13 itself. That is pretty close to the that arrived at using SNP counting of modern DF13 people - just 200 younger which is no big deal given all the uncertainties and averaging involved.

I would certainly feel happiest from an archaeological point of view to see the massive branching one SNP below DF13 as associated with the beaker spreading through the isles c. 2500-2400BC. It to me seems to back up the sort of ages being guessed for SNP intervals of 75 years or more. The archaeological evidence is very much against the dating of DF13 to only 3500 years ago IMO. I just find it impossible to see DF13 as young as that given its wide distribution in the isles. So, I am firmly in the camp of the people dating SNP intervals in modern individuals to 75 or more years.

One warning I would give at using Mal'ta SNP count and radiocarbon date is that that person died a couple of thousand years into the LGM and had his entire ancestry in a very different kind of demographic context than that when looking at DF13. An LGM hunter-gatherer context would probably have had a horribly low average life expectancy from birth figure. There have been attempts to base this on modern hunter gatherers but I am not sure this can be applied to LGM Siberia and there are simply not enough Upper Palaeolithic burials to work this kind of thing out in a meaningful way. IMO most children would not have survived to reproductive age and it would appear that the population just remained at maintenance which surely would mean that only half of the population actually produced any sons who survived to reproduce. I cannot get have a handle on what typical father to son generation intervals would be but I suspect it was at least a little lower than in the copper age where hierarchy meant a small amount of men with access to both wealth and women probably reproduced more and over a longer span.

Certainly I think several generations just below DF13 must have involved men who were hogging both females and wealth and may have lived longer, bred over a longer span with several women and had a better chance of his children surviving. I tend to think the average span between father and surviving sons would have been around 25 in that sort of set up and could even be higher in the case of very resource hogging individuals.

The other complication of a hierarchical society is of course that a higher level of mutations may be present in those wealthy individuals who lived longer and had a sequence of wives and concubines. If those kind of likely longer lived elite individuals or chiefs tended to produce a very out of proportion amount of surviving male lineages then perhaps mutation rate would speed up due to him reproducing some of his children at an older age.

Then again, if you have a lot of children and this pattern of rapid branching continues then the high status will dilute and become less marked in your descendants as they can only maintain that status in further generations as long as there are new unexploited areas to expand into - after a pioneering phases like that this will eventually dry up and further power will become the source of competition then violence then lifespans may reduce again. I suppose the lesson of this is that a golden pioneering phase such as for example metal skilled people spreading out over an area where there was no competition is by definition limited and will eventually be curtailed by competition. Even in a clan system many sons wont become chief and there lines will begin a downward mobility unless they risk their necks in fighting over new areas.

So, I would still tend to feel that it might average out to 25 years between father and surviving son in the copper/Bronze and Iron Age.


500 yrs makes sense, L459, Z245, Z260 & Z290 look to be phylogenetically equivalent to L21 and likewise Z2542 WRT DF13, though Z2542 doesn't play well with Big Y however the rejected results coming back certainly suggest this is the case.

That gives 7 SNPs after P312 to DF13.

alan
05-24-2014, 03:36 PM
Another aspect of this is that DF27 seems to have branched profusely immediately under that SNP. If that is correct then it commenced some centuries before the big explosion below DF13. That actually would fit the early start beaker had in SW Europe compared to NW. Archaeology would suggest beaker started 3-500 years (radiocarbon and contextual issues keep this a big vague) earlier in SW Europe than the isles. So, my feeling is that DF27 probably represents the early beaker zone in SW Europe. DF13 probably represents beaker reaching the English Channel and then the isles. That appears to fit the branching patterns.

That leaves U152. In archaeological terms the date of beaker in central Europe is probably a couple of centuries younger than in SW Europe and a century younger than NW Europe. So it would be expected IMO for its heavy branching to be delayed compared to SW Europe but a little earlier than NW Europe.

All of this would of course depend on a very literal correlation between the radiocarbon dates of beaker pots by geography and the spread of P312. That is not necessarily a given. However, it is interesting that there is some correlation.

Michał
05-24-2014, 06:04 PM
There is a paper from 2009 that you might be remembering (four measured mutations in 13 generations)...

Human Y Chromosome Base-Substitution Mutation Rate Measured by Direct Sequencing in a Deep-Rooting Pedigree

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2748900/

I would rather warn you all from using this paper as the most reliable source the Y-DNA SNP mutation rate:
http://eng.molgen.org/viewtopic.php?f=77&t=1300&start=02

Michał
05-24-2014, 06:28 PM
It can be seem on the L21 line that there would seem to be a slight delay in its expansion as its not until the SNP level one below DF13 that there is a major take off. This kind of fits the NWern distribution of L21 and the c. 3-400 years delay between origin to maximum distribution and the NW was one of the last areas in Europe to receive beakers and copper.

It seems to me that it is rather the other way around, i.e. the SNP-calculated MRCA for a significantly expanding clade should predate the archaeologically attested expansion by about 500-1000 years, and I have already explained it in my previous posts:
http://www.anthrogenica.com/showthread.php?1716-R1b-and-IE-branching/page3&p=23493#post23493
http://www.anthrogenica.com/showthread.php?2420-SNP-based-TMRCAs-for-R1b-U106-and-subclades&p=37495#post37495

Alan, if you don't agree with my points, what is the basis for your assumption that the SNP split postdates migration instead of preceding it. Do you really believe that it was a single P312 man who initiated the Beaker expansion, and if not, what happened to all other members of his tribe/community (i.e. where are their descendants now)?

I am quite convinced that demographic expansion does not actually produce the genetic diversity but rather fixes it by preventing the genetic drift from negatively controlling the level off diversity in a relatively small and stable (non-expanding) population.

Michał
05-24-2014, 06:39 PM
I'm just using Mark's recent quote "The Experts have deemed from studies that the range is 70-90 years per SNP mutation." I think Michal uses 88 years per SNP. Some would say it's three generations per SNP. Perhaps Mark or Michal or Warwick can elaborate on these ranges or point to specific studies. I'm assuming these reanges are for FGC high reliability SNPs, not Big Y or YFULL.
This has been already discussed so many times that I would rather be unwilling to repeat all these arguments again and again. On the other hand, Alan is right that all this informations is dispersed in different threads, so it is difficult to get a general overview. Therefore, I am going to write a summary of all the data that I am currently using to make my own calculations, but I will rather post it in a more appropriate thread (STR Wars, GDs, TMRCA estimates, Variance, Mutation Rates & SNP counting).

alan
05-26-2014, 02:19 PM
While the oldest part of the sort of date range for P312 would put it within touching distance of the very late arriving first farmers in the north-west, it still seems unlikely to me on current data. I say this because when you look at the bigger picture of Europe like southern and central Europe even the older dates suggested by SNP counting for P312 are significantly too young to be related to first farmers. All the ancient DNA also refutes a connection with these farming waves and NW Europe's first farmers seem closest related to central European post-LBK middle Neolithic groups like Lengyel and Rossen which have a significant cultural inheritance from LBK which is R1b-free in ancient DNA results. The NW European Neolithic seems to owe nothing to the Cardial culture so its irrelevant. I would still feel that on balance the case for P312 linked to early farmers anywhere is poor when all the evidence available is looked at.

What the SNP counting data has raised is the possibility that P312 may have a pre-beaker history. I used to think any pre-beaker history would have to be related to L11*, L51* or L23xL51 but the new calculations suggest the possibility that P312 could have been around in say 3500BC before beaker. However there are other ways of looking at it like treating the generation of DF13 as coinciding with the beaker arrival in the isles c. 2500BC would perhaps place P312 just a century or so before 3000BC which is closer to fitting with a beaker model although still implying a pre-beaker origin of P312 although a shorter one.

I think on balance from the evidence and calculations I have seen I would tend to see P312 as having a copper age origin even if it is pre-beaker.

alan
05-26-2014, 03:06 PM
Michal

Actually I tend to agree with most of what you are suggesting. I was just suggesting that the most likely geographical correlation of the big branching just under DF13 would be the arrival of beaker in NW Europe c. 2500BC. From other posters that could be 7 or 8 SNPs down from P312 which could place P312 around 3200BC which is still pre-beaker.

I am somewhere in between on the issue of branching before or after migration - I actually tend to see it as simultaneous. The reason why I feel that branching did not happen to a massive degree pre-migration is that P312 is split into three groups with strongly contrasting overall geographies - something I would extend to L11 as a whole too. However, I am not a zealot on that opinion because of the possibility of founder effects.

What I would say though is if there was a geographical spot or limited zone where there was a long period of existence and branching before any significant migration would it not still be detectable. today in the mix of clades and high variance in some zone? If so where is the likely spot. I would say its unlikely to have been at the extremes like the NW or Atlantic Iberia where L21 and DF27 respectively are very dominant and of course extremes of geography also simply seem unlikely from a geographical perspective. I would tend to see the most likely point of origin of P312 based on modern distribution as somewhere in the zone between Liguria, the Alps and the Rhine-Rhone axis. It also seems a good shout for the origin of L11* and L51* too. The fact that these lines could split off the larger lines at least shows that the ancestral forms passed through that area. Now, I presume that they being present in the Alps suggests branching was going on in that area before P312 so that would push back the date of presence in Alpine Europe even further than whatever date date we finally agree on for P312. Certainly few people would look at the distribution of L11* and L51* in that area but not in eastern Europe tends to suggest that at least a limited move west as far as the Alps commenced significantly earlier than the beaker period. It would on the surface seem to suggest that the idea of pre-beaker early copper age groups in that zone c. 3500-3000BC did include R1b even if the pre-beaker copper age ancient DNA has not picked it up yet - i.e. The Ice Man and the Languedoc samples.



It seems to me that it is rather the other way around, i.e. the SNP-calculated MRCA for a significantly expanding clade should predate the archaeologically attested expansion by about 500-1000 years, and I have already explained it in my previous posts:
http://www.anthrogenica.com/showthread.php?1716-R1b-and-IE-branching/page3&p=23493#post23493
http://www.anthrogenica.com/showthread.php?2420-SNP-based-TMRCAs-for-R1b-U106-and-subclades&p=37495#post37495

Alan, if you don't agree with my points, what is the basis for your assumption that the SNP split postdates migration instead of preceding it. Do you really believe that it was a single P312 man who initiated the Beaker expansion, and if not, what happened to all other members of his tribe/community (i.e. where are their descendants now)?

I am quite convinced that demographic expansion does not actually produce the genetic diversity but rather fixes it by preventing the genetic drift from negatively controlling the level off diversity in a relatively small and stable (non-expanding) population.

Dubhthach
05-26-2014, 03:15 PM
Just on side note saw this posted on Facebook, nice little article from Koch:

https://www.academia.edu/7142081/Celts_Britons_and_Gaels-Names_Peoples_and_Identities

rms2
05-26-2014, 05:08 PM
Just on side note saw this posted on Facebook, nice little article from Koch:

https://www.academia.edu/7142081/Celts_Britons_and_Gaels-Names_Peoples_and_Identities

Excellent article. I enjoyed it. I especially like the way it deals with the silliness of Celtoscepticism at the beginning.

alan
05-26-2014, 07:23 PM
Very good article. Koch has been pretty consistent and produced most of these ideas 20-odd years back in journals like Emania. I prefer Koch's contributions to his archaeological sidekick Barry Cunliffe in terms of the Atlantic Celts.

One correction I would make is that while the Ierne and Albion reference probably originates c. 550BC, long before the La Tene material 200-250 years later and does prove Celticity in pre-La Tene times, the idea that Dowris late Bronze Age culture survived that late to fill the apparent gap before La Tene material has been recently rejected in the recent book on Life and Death in Iron Age Ireland. Reviews of dating dating suggest Dowris metalwork, coarse pottery types, sites like hillforts etc ended not long after 700BC around the Hallsatt C phase of Europe (a small amount of material of which appears in Ireland) and there is no evidence whatsoever for a long twilight period of the Dowris Late Bronze Age material. Dowris elite metalwork simply came to an end and Iron but not La Tene or any other foreign cultural material probably did appear shortly after 700BC. The truth seems that the elite and system that worked on control of bronze and metals collapsed when iron (accessible to all in the form of bog iron ore and not subject to control like copper and tin) came in. Essentially Bronze elite metalwork disappeared probably along with the loss of power of the old elite and some sort of relatively mundane native culture using iron and no elite showy bronze work existed for some centuries c. 700-350BC. Extremely little if any of the iron objects survive in that period - possible because it survives very badly in Ireland - but iron slag and working sites do. At the moment we just dont know what form this pre-La Tene native Iron Age material culture looked like and whether the iron objects were copies of tools etc from Bronze Age types or Hallstatt types etc.


Just on side note saw this posted on Facebook, nice little article from Koch:

https://www.academia.edu/7142081/Celts_Britons_and_Gaels-Names_Peoples_and_Identities

Dubhthach
05-26-2014, 07:53 PM
A good example of Iron usage in this period is the Lackan spear head. Part of the shaft survived and was dated using C-14 to 800-675BC

http://www.100objects.ie/wp-content/uploads/2012/10/Iron-Spear-Head.jpg

Fintan O'Toole almost wet himself in Irish Times writing about it. (History of Ireland in 100 objects)



This iron spearhead is of a kind familiar enough from the Ireland of 500 CE. Andy Halpin of the National Museum of Ireland says that it “wouldn’t be out of place in the early medieval period”. The problem is that recent carbon-dating of the remains of its wooden shaft suggests that it may be more than 1,000 years older. If this is so, it explodes a myth about how the Iron Age came to Ireland.

The long-held belief was that the use of iron in Ireland was a result of the invasion of the Celts. These people are associated with the development of ironworking in Europe north of the Alps, and with a culture named after the Austrian town of Hallstatt. Greek writers refer to the existence of ‘Keltoi’ in central Europe in the sixth century BCE. The Hallstatt people seem to have been responsible for the westward spread of ironworking. It seemed logical that the appearance of iron in Ireland must be evidence of the arrival of these Celts. No one doubts that the influence of these central Europeans is evident in Irish artefacts from the sixth century BCE onwards. But the process was not one in which the Bronze Age suddenly ended and there is no reason why this date has to be regarded as wrong. It’s the combination of this early date and its superb quality that makes this spear so startling. “We are beginning,” says Halpin, “to get other evidence for ironworking technology at an earlier date than we thought. The idea that ironworking was happening here in maybe 600 or 700 BCE wouldn’t really be disputed any more. But the idea that they were producing something as nice as this at that period suggests not only that iron was being worked here, but also that it was being worked by very competent smiths much earlier than we think. Those smiths were not invading ‘Celts’. They were part of the same culture that was producing the dazzling gold and bronze objects we have already seen.


I see following quote from National Museum site


Recent radiocarbon dates from sites excavated in the Irish midlands suggest that knowledge of ironworking may have been known from as early as the eighth Century BCE. At Kinnegad 2, Co. Westmeath, charcoal found associated with iron slag and pottery of Late Bronze Age type, yielded a date range of 810 – 420 BCE. At Rossan 6, also in Co. Westmeath, charcoal associated with iron slag yielded a date range of 820 – 780 BCE. The transition to the widespread use of iron, in preference to bronze, appears to have happened slowly and over a long period, and it was not until around the third Century BCE that a distinctive Iron Age society, clearly recognisable in the archaeological record, emerged. What gives the period its distinctive character is the widespread use of ornament to decorate objects using an art style that was developed first in central Europe by Celtic peoples. Known as the La Tène style, the art occurs on early objects, such as two imported gold collars found in a bog at Ardnaglug, Co. Roscommon that were probably made in the Rhineland in the third Century BCE.
...



As you said it points to the use of Iron happening among the actual "indigenous population", you probably had some sort of elite collapse thence lack of "prestige items", could be connected with climatic changes evident in the Bog core records.

-Paul
(DF41+)

Michał
05-26-2014, 07:55 PM
I am somewhere in between on the issue of branching before or after migration - I actually tend to see it as simultaneous. The reason why I feel that branching did not happen to a massive degree pre-migration is that P312 is split into three groups with strongly contrasting overall geographies - something I would extend to L11 as a whole too. However, I am not a zealot on that opinion because of the possibility of founder effects.

Another possibility is that the migrating Bell Beaker groupings were different tribes or clans composed of some relatively closely related people, so it was much more likely that all/most members of the hypothetical "DF27 tribe" ended up in Iberia while most members of the hypothetical "L21 tribe" migrated to the Isles, even if their respective MRCAs lived a couple of centuries earlier in a completely different region of Europe.

alan
05-26-2014, 08:07 PM
I think Celtoscepticm is really based on old Victorian racial ideas, deliberate discouragement of the tend to separatism and devolution among the Irish, Scots and Welsh and frankly an irrational insistence on the importance of whether the word Celt was used by the isles Celts. As Koch points out you cannot expect native sources of c. 600AD to reflect identity or remember identity terms from 800 years earlier or whatever and identity terms are fluid and constantly changing. We have no idea what anyone called themselves in the iron age and we cannot trust classical sources on this. Classical sources have a bad habit of naming people or nations after one tribe ie Caledonia. Several of the Iberian Celtic tribes had names that included the word Celt and they may have just generalised that onto similar peoples. Alternatively the world Celt and Gaul were words they came across as self-describers from a small subset of Celtic speaking warriors who intruded on the Classical world - I suspect they were boast words from Celtic speakers that they clashed with and just meant something like 'we are the crazy hard men or warrior heros'. The word Celt was borrowed into German as hilder or something similar and meant hero. If so it may have just been a word used by migrating warrior groups. Also as Koch points out it early became a territorial terms with no specific meaning i.e. Gallia and Gallia Celtica and so would have just meant someone in or from those territories. Naturally once it had territorial meaning it wouldnt be applied to other territories.

Celto-skepticism is all pedantic rubbish and Anglo arrogance IMO. The Romans clearly state that the British, the culture and their language was a variant on that of Gaul and the Irish were variants of that in Britain. Linguists clearly confirm that they were closely related but its far from just that. Many of the cultural, religious (druids, bards etc), institutions (clientship), social structure, traditions, class divisions, societal, status, ritual, military aspects can be shown to exist across the Celtic speaking world and its rather more specific than just generic IE. The bottom line is that all that was shared and the only real differences variation in wealth, development and access to influences which depending on geography - some had access to certain trendy metalwork or Etruscan influences while others didnt, some were richer than others and had access to trade while others were less so, some had more advanced societies with coinages, oppida etc while some were rural and pastoral etc. However the core basic social structures and traditions are noted across the Celtic speaking world and strongly echoed in early Irish and Welsh literature.

Would the same 'skeptics' say a Germanic speaking tribe cannot be called Germanic because no Roman commentator in a toga records them self-describing as such? Do we have records of every Slavic speaking tribe describing themselves as Slavs? Does that make them less Slavic? IMO the Celts are being made to suffer simply because as the 'fathers of Europe' theirs was a very early dispersal and in time out of mind.

The truth is people exaggerate the power of oral history. There are virtually no peoples in Europe who accurately had knowledge of their ancestors once you move back millennia - the Romans, Germans etc simply traced themselves from gods like Thor or Romulus and Remis once it got much into the prehistoric period.


Excellent article. I enjoyed it. I especially like the way it deals with the silliness of Celtoscepticism at the beginning.

alan
05-26-2014, 09:55 PM
I would wonder how long a lineage could build up into a clan or small tribe before it would be obvious in the SNP vertical string it had a period like that. It has been suggested above that there could be 7 SNPs between P312 and DF13 derived lines that seem to have occupied NW Europe. That could represent quite a number of centuries - perhaps 500 years which is plenty time to build up a huge clan or tribe.

However, the problem is that most L21 in the isles share the same SNPs from P312 to DF13 suggesting they are from one lineage. This more seems to imply to me a short build up under DF13 before migration but ASAIK DF21, DF23 etc are just one SNP below so it only sounds like a maximum of a few generations of pre-dispersal build up before dispersal. Maybe if we look at it from a a more refined subclades point of view then things like P314.2 are a couple of SNPs below DF13 so there was a bit more time to form a larger clan before dispersal but my impression is probably that they moved about in small clans perhaps in the low dozens rather than large groups.

I suppose DF13 could have built up as DF13* for 3-5 generations without SNPs occurring to identify or allow reconstruction of this phases and therefore is essentially unreconstructable as it is below a 90 year interval SNP radar. I suppose a lot can happen in 90 years. It is possible too that a pioneering phase as metal controlling lineages in an area who wanted metals and had no knowledge or skills relating to it could lead to a situation of the skilled lineage producing a freakish amount of males from a common great grandfather although I would still suspect we are talking low hundreds even in an extreme situation and these spread across the isles. The SNPs directly below DF13 could represent the lineages a few generations later of beaker people in the isles continuing to grow at a big big rate - especially when considered that many lineages would have later died out in the 4000 plus years since then. For the large amount of lineages one SNP down from DF13 to have survived to today suggests there was a massive burst of growth.

In general I think the 'below the SNP radar' perhaps 90 years sort of period could translate into 4-5 generations of DF13* that could have built up before migration but this would still seem to me to be likely to have been no have amounted to the low hundreds at most. I would also tend to think that the nature of the apparent role of beaker people in introducing metals, mining etc across the isles would encourage fission and pioneering fanning out in small numbers into each area. After all tthere would be only so many such specialists and traders that any local pre-beaker tribe would want to absorb and support. Its most likely IMO that small subclans of a dozen or so people would have 'worked' a given area and provided metals, trade, metalworking etc skills and knowledge although they probably remained in contact with a wider network of distant cousins in other locations given that beaker culture and the metal network required this kind of networking.


Another possibility is that the migrating Bell Beaker groupings were different tribes or clans composed of some relatively closely related people, so it was much more likely that all/most members of the hypothetical "DF27 tribe" ended up in Iberia while most members of the hypothetical "L21 tribe" migrated to the Isles, even if their respective MRCAs lived a couple of centuries earlier in a completely different region of Europe.

Dubhthach
05-27-2014, 09:32 AM
With regards to Celto-sceptism and "Insular Celtic" languages. It's kinda par for the course with general revionististic orthodoxy that developed during the 1960's/70's. Some of which in Ireland anyways ties in with deep (almost atavistic) hatred that often boils to surface with regards to the Irish language in this country. Part of it I believe is to do with a "colonial mentality". For example you'll see it in comments on articles about Irish where you'll have people go on rants about "It was beaten into me in school.... it's a dead language ..... waste of time .... children should learn Chinese instead ...." (substitute Chinese for whatever is the "it" language for business at time).

In a framework of "there was never any migrations/movements of people/ The Irish were never Celts" you can see arguments like this:
"Well Irish isn't really our language, after all we would have spoken another language before hand, only adapted a "Celtic language" for trade etc., so if it dies out it's no great loss" (this ties in with concept that there was no migration since the stone age etc. and that cultural change was due to "trade"/"contact")

Leaving aside that I also think that part of why it was so popular form the 1960's onwards was part of a widespread "backlack" against what was seen as "nationalistic baggage" given context of what was happening in Northern Ireland.

alan
05-27-2014, 02:22 PM
Yeah I gather there is similar backlash against supporting Celtic FC in the Republic as its seen as part of an ultra nationalistic old fashioned thing wrapped up with Sinn Fein, the Irish language etc. OK in the north controversy is understandable given its a very divided population but in the south its a bit more surprising and seems a bit like self loathing.

Having said that I dont myself like politics and language/cultural aspects being too closely associated in a way that makes it look like a particular political view feels like ownership of something like the Irish language. It creates a barrier to taking up the language among many unionists - something that wasnt always the case and is a bit silly because a significant minority of their ancestors had spoken Gaelic in Galloway in SW Scotland and it had only started to die out just before the Ulster plantation. There are also a surprisingly significant amount of Ulster protestants with highland surnames despite it being mainly a lowland Scots and English plantation. I think they probably came in later in the 17th century than the plantation as the terms of the plantation banned highlands Scots from being planted in Ulster. So, its part of their heritage too even if that idea goes down like a fart in a spacesuit with many today.


With regards to Celto-sceptism and "Insular Celtic" languages. It's kinda par for the course with general revionististic orthodoxy that developed during the 1960's/70's. Some of which in Ireland anyways ties in with deep (almost atavistic) hatred that often boils to surface with regards to the Irish language in this country. Part of it I believe is to do with a "colonial mentality". For example you'll see it in comments on articles about Irish where you'll have people go on rants about "It was beaten into me in school.... it's a dead language ..... waste of time .... children should learn Chinese instead ...." (substitute Chinese for whatever is the "it" language for business at time).

In a framework of "there was never any migrations/movements of people/ The Irish were never Celts" you can see arguments like this:
"Well Irish isn't really our language, after all we would have spoken another language before hand, only adapted a "Celtic language" for trade etc., so if it dies out it's no great loss" (this ties in with concept that there was no migration since the stone age etc. and that cultural change was due to "trade"/"contact")

Leaving aside that I also think that part of why it was so popular form the 1960's onwards was part of a widespread "backlack" against what was seen as "nationalistic baggage" given context of what was happening in Northern Ireland.

MJost
05-27-2014, 02:45 PM
I already posted this but here it is again
https://drive.google.com/file/d/0By9Y3jb2fORNWkdIUlRjYTlKQWs/edit?usp=sharing

Out of my 25 SNPs, my GD3/111 had 23 positive, a SGD (SNP Genetic Distance) of 2. As a predict 3rd cousin on 23andme for his father's sister, has 10 generations back to the first male into the Virginia Colonies. My GD8/111 tested a SGD3 and has to be at least 11 generations but I suspect 12-13 generations back to a common ancestor. Then my GD13/111 is a SGD9. So depending on the number of years per generation, from b. 1744 to present (1950) at 10 Gens is 3.3 generations per mutation, that covered about 206 years which is about 20.5 year per gen. Just rough numbers.

MJost

Let say that the above is correct, and there are about 75 SNPs back to P312, at 3.3 generations per SNP (75*3.3)x 21 years per generation calculates to 5,200ybp or 3,200BCsons just at the start of the Bronze Age period. With few SNPs under P312 down to DF13 , and considering a huge number of DF13 subclades now, and clearly to this point in time, little or no subclades being ancestral to another, this appears to be a very rapid expansion with many, many and/or few grandsons. Then by the existing STR dating, we see a much shorter TMRCA to the founder haplotypes. For these sons and grandsons, the lines then appears to have low or single son births, a long sustained bottle neck until the Iron age started in the 11th century BC, probably from the Caucasus, and slowly spread northwards and westwards over the succeeding 500 years, to around 600BC. In the STR side of TRMCA there are a number of DF13 subclades coalescing back to about this time frame, but a quite a few are later yet. I know my own subclade's each of the main two branches are pushing into the early AD period but when calculating the entire set of 67 marker HTs pushes it back 3200-3400 ybp with 19 Rho Mutations using a fixed 172 years between mutations. My 5496-A1* which would be FGC5521>FGC5539 guys had a 2300ybp (73+-26gen) TMRCA based on STR variances.

A long term bottle neck between DF13's initial huge number of sons and the later massive expansion after this bottle neck is the most pausable explanation for the difference amount time between STR variance and SNP counting methods.

In addition our two main subclades have two old continental branches back to 5-800BC-SNP or earlier time frame. Suggests a most probable continental origin of FGC5496.

MJost

Mikewww
05-27-2014, 04:08 PM
....
A long term bottle neck between DF13's initial huge number of sons and the later massive expansion after this bottle neck is the most pausable explanation for the difference amount time between STR variance and SNP counting methods.
...

I clearly agree there were multiple expansions, or fits and starts. The blocking of equivalent SNPs for some of the larger identifiable population subgroups is apparent and the result of some kind of pauses or bottlenecks. Here are some of the larger subgroups of DF13 that most of know by cluster name. Most noteworthy is M222 and the NW Irish.

DF13 > FGC1134 > CTS4466 and 11 equivalents (Irish II/South)

DF13 > DF49 > Z2976 and 1 equivalent > DF23 and 2 equivalents > Z2961 > M222 and 18 equivalents (NW Irish)

DF13 > L1335 and 7 equivalents (Scots + Wales II)

DF13 > L1335 and 7 equivalents > L1065 and 7 equivalents (Scots Modal)

DF13 > L513 and 3 equivalents > S5668 and 1 equivalent > A7 > L193 and 8 equivalents (A1 Scots-Irish)

DF13 > Z253 and 1 > Z2534 > Z2185 and 2 > L1066 and 2 equivalents (Irish IV/Continental)

DF13 > Z253 and 1 > Z2534 > Z2185 and 2 > L226 and 15 equivalents (Irish III/Dalcassian)

DF13 > Z255 and 11 equivalents (Irish Sea)

DF13 > DF21 > A50 and 17 equivalents (Airghelli I / Clan Colla / null 425)

The above are just the higher population examples I can think of. There are plenty of DF13* folks still out there and there are many smaller apparent direct subclades already identified.

Some how or some way, though, DF13 diversified very quickly to leave branching that could survive. In other words, its early expansion included many subclades (brothers that don't share SNPs other than DF13 and above) and put them in a situation to survive and in some cases thrive greatly. I don't see any other way than to conclude DF13 and his most immediate descendants were very mobile and successful colonizers.

I suspect that DF13 and his immediate male descendants had some advantage or advantages and diversified both geographically and genetically (inter-marriage) providing cultural, health and climatic protections. We forget that some diversifications (think colonies), similar to mutations, may not have worked but if you are prolific enough at diversifying some may get lucky and hit upon some good new combinations of skills, alliances and genes. If their networking remained well intact, the best new combinations could be shared.

alan
05-27-2014, 10:59 PM
I suppose the simplest answer to the quick big branching just under DF13 and then long periods of stasis in some lines is probably the old first-in or pioneering aspect where the opportunities for a lineage spreading into a land devoid of metal using groups were enormous for a century or so then possibly the opportunities dried up and it became very competitive. I would tend to picture beaker people as grabbing onto various nodes, trade networks, mines, friendly local tribes who wanted them to settle etc. However, if their role was primarily a specialist new one the best opportunities may have dried up within a century or so.

One archaeological way of supporting this for example in the isles is that in the first few centuries of beaker culture and copper use c. 2500-2200BC there only appears to have been one source of copper extraction used in the isles - Ross Island in SW Ireland. The reason this single mine was used was probably that prior to tin alloying copper was often rather soft and Ross Island was a rare, perhaps unique isles source of arsenical copper which is as hard as Bronze. A lot of the gold objects of isles origin in the same period may come from a single Irish source in the Mourne mountains. Other metal was likely imported from the continent.

This picture changed a few centuries later when other isles copper mines and sources opened up after than and probably Cornish tin led to Bronze. The latter meant that relatively pure soft copper could be exploited as it would be alloyed with tin to make Bronze.

Anyway, I am just outlining this to demonstrate that the opportunities for beaker copper workers to find niches may have been more limited than it may seem. Certainly in terms of copper extraction it was limited in the early beaker phase and copper and tin extraction would have continued in the later beaker phase in the isles to be limited in opportunities to certain areas. So there was a limiting factor in just how many people could have found those niches.

Of course extraction and primary production of copper was not the only role. What might have provided much more widespread opportunities for enclaves of beaker people would have been working the metal into objects. That may have become decentralised after a while as copper workers settled among tribes who desired those skills for local manufacture. Distribution was also a role that would have provided middlemen niches and arguably they seem to have had a role that brought a lot of wealth judging by rich beaker clusters in areas like Wessex which lay on the roots between the metal sources of the west and the metal poor but agriculturally wealthy lands in the east. There would also have had to be sailors etc involved.

It would be interesting to consider how much of a stranglehold in production of beaker period copper and gold objects was kept in new hands and when this transformed into the distribution of just the raw metal and local manufacturing. That sort of shift may have seen a new expansion phase.


I clearly agree there were multiple expansions, or fits and starts. The blocking of equivalent SNPs for some of the larger identifiable population subgroups is apparent and the result of some kind of pauses or bottlenecks. Here are some of the larger subgroups of DF13 that most of know by cluster name. Most noteworthy is M222 and the NW Irish.

DF13 > FGC1134 > CTS4466 and 11 equivalents (Irish II/South)

DF13 > DF49 > Z2976 and 1 equivalent > DF23 and 2 equivalents > Z2961 > M222 and 18 equivalents (NW Irish)

DF13 > L1335 and 7 equivalents (Scots + Wales II)

DF13 > L1335 and 7 equivalents > L1065 and 7 equivalents (Scots Modal)

DF13 > L513 and 3 equivalents > S5668 and 1 equivalent > A7 > L193 and 8 equivalents (A1 Scots-Irish)

DF13 > Z253 and 1 > Z2534 > Z2185 and 2 > L1066 and 2 equivalents (Irish IV/Continental)

DF13 > Z253 and 1 > Z2534 > Z2185 and 2 > L226 and 15 equivalents (Irish III/Dalcassian)

DF13 > Z255 and 11 equivalents (Irish Sea)

DF13 > DF21 > A50 and 17 equivalents (Airghelli I / Clan Colla / null 425)

The above are just the higher population examples I can think of. There are plenty of DF13* folks still out there and there are many smaller apparent direct subclades already identified.

Some how or some way, though, DF13 diversified very quickly to leave branching that could survive. In other words, its early expansion included many subclades (brothers that don't share SNPs other than DF13 and above) and put them in a situation to survive and in some cases thrive greatly. I don't see any other way than to conclude DF13 and his most immediate descendants were very mobile and successful colonizers.

I suspect that DF13 and his immediate male descendants had some advantage or advantages and diversified both geographically and genetically (inter-marriage) providing cultural, health and climatic protections. We forget that some diversifications (think colonies), similar to mutations, may not have worked but if you are prolific enough at diversifying some may get lucky and hit upon some good new combinations of skills, alliances and genes. If their networking remained well intact, the best new combinations could be shared.

Dubhthach
05-28-2014, 08:08 AM
I clearly agree there were multiple expansions, or fits and starts. The blocking of equivalent SNPs for some of the larger identifiable population subgroups is apparent and the result of some kind of pauses or bottlenecks. Here are some of the larger subgroups of DF13 that most of know by cluster name. Most noteworthy is M222 and the NW Irish.

DF13 > FGC1134 > CTS4466 and 11 equivalents (Irish II/South)

DF13 > DF49 > Z2976 and 1 equivalent > DF23 and 2 equivalents > Z2961 > M222 and 18 equivalents (NW Irish)

DF13 > L1335 and 7 equivalents (Scots + Wales II)

DF13 > L1335 and 7 equivalents > L1065 and 7 equivalents (Scots Modal)

DF13 > L513 and 3 equivalents > S5668 and 1 equivalent > A7 > L193 and 8 equivalents (A1 Scots-Irish)

DF13 > Z253 and 1 > Z2534 > Z2185 and 2 > L1066 and 2 equivalents (Irish IV/Continental)

DF13 > Z253 and 1 > Z2534 > Z2185 and 2 > L226 and 15 equivalents (Irish III/Dalcassian)

DF13 > Z255 and 11 equivalents (Irish Sea)

DF13 > DF21 > A50 and 17 equivalents (Airghelli I / Clan Colla / null 425)

The above are just the higher population examples I can think of. There are plenty of DF13* folks still out there and there are many smaller apparent direct subclades already identified.

Some how or some way, though, DF13 diversified very quickly to leave branching that could survive. In other words, its early expansion included many subclades (brothers that don't share SNPs other than DF13 and above) and put them in a situation to survive and in some cases thrive greatly. I don't see any other way than to conclude DF13 and his most immediate descendants were very mobile and successful colonizers.

I suspect that DF13 and his immediate male descendants had some advantage or advantages and diversified both geographically and genetically (inter-marriage) providing cultural, health and climatic protections. We forget that some diversifications (think colonies), similar to mutations, may not have worked but if you are prolific enough at diversifying some may get lucky and hit upon some good new combinations of skills, alliances and genes. If their networking remained well intact, the best new combinations could be shared.

I wonder if alot of this bottlenecking is actually during the Iron age. The prime example been M222, with all it's equivalents it's only after it found itself potentially in a position of power that we see diversivication of the SNP's under M222. Traditional TMRCA points towards end of Iron age as period where number of these "bottlenecked" lineages appear to expand.

http://www.kennedydna.com/M222_tree.png

If we look at Irish context and look at CTS4466, L226, M222 and A50 (and we assume that they appear connected to relevant dynasts) we see that these groups really only start to expanded in a historical sense in period after the rise of Christianity (400-500AD + --- for L226 it's even slightly later probably on order of late 9th/early 10th century if we believe it's connected to Dál gCais)

If that's the case then you'd obviously could have several hundred years before where the lineage was "restricted"/quite small, only then to undergo effectively a "dam burst" of growth over a short period of time.

-Paul
(DF41+)

jdean
05-28-2014, 08:44 AM
I wonder if alot of this bottlenecking is actually during the Iron age. The prime example been M222, with all it's equivalents it's only after it found itself potentially in a position of power that we see diversivication of the SNP's under M222. Traditional TMRCA points towards end of Iron age as period where number of these "bottlenecked" lineages appear to expand.

Any idea if an M222 x S7073 kit been identified yet ?

It'd be good to see FGC or Big Y for somebody on that branch.

Dubhthach
05-28-2014, 03:38 PM
Any idea if an M222 x S7073 kit been identified yet ?

It'd be good to see FGC or Big Y for somebody on that branch.

From what I've heard there is one anonymous M222+/S7073- known from Chromo2 testing, origin supposedly in Fife. Every other M222 tested in BigY/FGC or Chromo2 has come back as S7073+. S7073- is about as basal as you can get with M222 these days it would seem.

-Paul
(DF41+)

alan
05-28-2014, 04:46 PM
Certainly the archaeology would imply a really major downturn in Ireland soon after 700BC which was particularly intense around 700-350BC. This is not just down to metalwork but based now on radiocarbondating of random sites during the big road and pipeline projects during the boom years. In that period it seems a systems collapse meant the elite level become non-existent or very impoverished and probably power was very local. The improvement from say 350-0 is visible but not spectacular. According to pollen diagrams things only really take off in the later Roman era in Ireland c. 300AD or so and then an explosion in settlement sites after 500AD.

Its not often commented on but Britain too had a bit of a dark age c. 700-500BC and there is a massive drop off in elite indicators in what was the Hallstatt D and La Tene A period. A lot of Scotland and northern England seems to have been not dissimilar to Ireland in the lack of settlement sites, pottery, elite metalwork and burials in this period and even well after. I think Iron really killed off the elite who had their power in controlling copper and tin and it took a while to rebuild a new system where power was based more on land than control of metals. I know in England they are now thinking there was a return to feasting activities in hillforts around this time as an alternative way of bonding polities together instead of control and distribution of metals and exotica - kind of like a return to almost Neolithic practices. I believe the hengiform 'royal sites' like Tara, Dun Ailline, Navan and Cruachan some similar return to that sort of bonding through seasonal gatherings at festivals, feasting, games and ritual. Basically once the metal wealth had gone it becomes all about land agriculture again. It seems there was a period where large areas bonded a common identity that way.

I suspect it was a somewhat different set up from the predatory expanding lineage based societies of the Early Christian period which seem to have given rise to mega-lineages.


I wonder if alot of this bottlenecking is actually during the Iron age. The prime example been M222, with all it's equivalents it's only after it found itself potentially in a position of power that we see diversivication of the SNP's under M222. Traditional TMRCA points towards end of Iron age as period where number of these "bottlenecked" lineages appear to expand.

http://www.kennedydna.com/M222_tree.png

If we look at Irish context and look at CTS4466, L226, M222 and A50 (and we assume that they appear connected to relevant dynasts) we see that these groups really only start to expanded in a historical sense in period after the rise of Christianity (400-500AD + --- for L226 it's even slightly later probably on order of late 9th/early 10th century if we believe it's connected to Dál gCais)

If that's the case then you'd obviously could have several hundred years before where the lineage was "restricted"/quite small, only then to undergo effectively a "dam burst" of growth over a short period of time.

-Paul
(DF41+)

Heber
06-10-2014, 12:05 AM
"I believe the hengiform 'royal sites' like Tara, Dun Ailline, Navan and Cruachan some similar return to that sort of bonding through seasonal gatherings at festivals, feasting, games and ritual. Basically once the metal wealth had gone it becomes all about land agriculture again. It seems there was a period where large areas bonded a common identity that way.

I suspect it was a somewhat different set up from the predatory expanding lineage based societies of the Early Christian period which seem to have given rise to mega-lineages.

This is an interesting new find of a complete skeleton in a 3,000 year old setting. It would be interesting if useful DNA could be recovered.

"3,000-year-old remains of a baby have been found during inaugural archaeological works at a Meath site reputed to be the birthplace of Halloween.

The remains were found at the base of a 1.5 metre ditch at Tlachtga, near Athboy.

It’s believed the fully intact skeleton is of a baby between seven and 10 months old, but it is not thought the child was the victim of any human sacrifice on the ritualistic site.

The remains will now be taken to the School of Archaeology at University College Dublin for further examination.

Describing it as “an exciting find,” lead archaeologist on the site Dr Stephen Davis said: “We may never know what caused the death of the child. The skeleton probably dates back 3,000 years and was found on the bedrock at the base of a 1.5m ditch.”

The remains were found during a three-week excavation on Tlachtga — most commonly held to have been the first site to celebrate the feast of Samhain — Halloween."

http://www.irishexaminer.com/ireland/3000-year-old-remains-of-baby-found-in-meath-270998.html

http://tlachtga.wordpress.com/

1953