PDA

View Full Version : Mapping admixtures



Sile
14-02-14, 10:15
interesting concept


http://admixturemap.paintmychromosomes.com/




http://dienekes.blogspot.com.au/2014/02/human-admixture-common-in-human-history.html (http://admixturemap.paintmychromosomes.com/)

ElHorsto
14-02-14, 12:13
interesting concept


http://admixturemap.paintmychromosomes.com/




(http://admixturemap.paintmychromosomes.com/)http://dienekes.blogspot.com.au/2014/02/human-admixture-common-in-human-history.html





Some aspects are interesting, but as Dienekes said, it contradicts the archeological evidence.

For instance 'Scottish' has overwhelming 'Irish' ancestry (no surprise), but 'Welsh' have zero 'Irish'.
The 'Welsh' are about half 'Italian', but 'Scottish' are zero 'Italian'.
That makes no sense. Or it shows only very recent admixtures.
Interesting though again the 'Baloch' and 'Lezgin' link for 'Scottish'.

Wilhelm
14-02-14, 13:27
Some aspects are interesting, but as Dienekes said, it contradicts the archeological evidence.

For instance 'Scottish' has overwhelming 'Irish' ancestry (no surprise), but 'Welsh' have zero 'Irish'.
The 'Welsh' are about half 'Italian', but 'Scottish' are zero 'Italian'.
That makes no sense. Or it shows only very recent admixtures.
Interesting though again the 'Baloch' and 'Lezgin' link for 'Scottish'.
The Welsh show 26% Italian, not half italian.

ElHorsto
14-02-14, 13:34
The Welsh show 26% Italian, not half italian.

You are right, thanks for correction. Yet my basic point remains.

martiko
14-02-14, 14:16
it has has sense, when they writes you Basque are in 96 % Spanish; purpose Spanish are in 12 % Basque? Especially when they know that basques are has big part of ancient Gaulishes of France.

Jackson
14-02-14, 16:11
Hmm this makes no sense to me, i did notice some of the samples are really low.

Angela
14-02-14, 18:42
I haven't plowed through the supplement yet, but I did look at the companion site quickly to check the results for northern Italians and Tuscans, as those are my two ancestral groups.

I find the ratio of "Cypriot" like, to "Welsh" like for the northern Italians, and "Cypriot" like and "French" like for the Tuscans a little surprising:
Northern Italians: 33/63
Tuscans: 45/55

I'm assuming that the "Cypriot" like population is the pre-Gallic "native" population? So, does that mean that the Italics, including the Romans, were "Cypriot" like, or just that the Italics were so low in number that they didn't change the prior EEF population very much, to use a Lazaridis term?

I suppose you could say that the 48% specifically English and Welsh for the northern Italians could be correlated with the Gallic invasions. The paper's date for the admixture, which is approximately in the middle of the first millennium B. C. would fit. But if the 8.6% German is a stand in for the Longobard admixture, that would have arrived much later in the AD/CE period.

Things are even more confusing for the Tuscans. It makes sense that their English and Welsh would be less, about 27% versus 48% for the northern Italians, but why do Tuscans have almost double the amount of German as the northern Italians? Everything I know says that the Langobard presence was stronger in the north. And why is the admixture dated so late in the Common Era for them, much later than for the northern Italians? The lowest edge of the range barely overlaps with the Lombard invasion, but what about the "British" admixture? Unless, as these dating programs seem to do, it's only picking up the last admixture date and folding in the older ones. That still doesn't explain why the specifically German one is higher than in northern Italians, however.

There's also the conflict with the Ralph and Coop paper to consider. While they did pick up the Gallic admixture in the middle of the first millenium B.C., they are very clear that they saw very minimal admixture in Italy after that time. So, which approach is more accurate? Perhaps Ralph and Coop as they are looking at specific, very small "bites" of the genome?

Well, if this paper is correct, it would explain why earlier work showed that while modern Tuscans are very similar to early medieval Tuscans, they're not similar at all, at least in terms of mtDNA, to the Etruscans. So, we full or half Tuscans have the Renaissance, but not the Etruscans, at least genetically? What a disappointment if that's true.

Also, it would prove how unreliable phenotypes can be...I constantly see people who look as if they are Etruscan statues come to life. The power of dominant genes perhaps?

It also occurs to me to wonder how this correlates to the yDNA signatures in Italy, especially R1b U-152. Is it Gallic then? If it is, what does that say about the millions of words expended trying to explain the arrival of the Indo-European languages like Italic into Europe with steppe herders in the second millennium B.C.? And what haplogroup would have carried the "Germanic" input? In northern Italy, you might say U-106, but what about in Toscana?

And how does this correlate to Lazaridis et al proportions for all the European populations?

Tons of questions, I know, but I wanted to put them out there...

MOESAN
14-02-14, 19:01
Hello Angela!
I did not read these papers, I will and shall do
concerning physical aspect don't forget genuine Etruscans of eastern origin seem having been only a narrow elite population - it seems confirmed by surveys about mt DNA showing the links between central Italy and Anatolia Near-East for mt DNA are OLD links and not recent history ones for the most - more neolithical - and mt DNA says nothing about global phenotype - historical Etruscans (as depicted and not depicted) were more an autochtonal population + some I-E italics (Umbrians) than a true etruscan one
more genrally I have some defiance against these comparisons with local populations like -'welsh', 'irish', 'german', 'basque'... no sense for me: only components have some value (and yet...) not global modern state or districts populations

Angela
14-02-14, 19:44
Hello Angela!
I did not read these papers, I will and shall do
concerning physical aspect don't forget genuine Etruscans of eastern origin seem having been only a narrow elite population - it seems confirmed by surveys about mt DNA showing the links between central Italy and Anatolia Near-East for mt DNA are OLD links and not recent history ones for the most - more neolithical - and mt DNA says nothing about global phenotype - historical Etruscans (as depicted and not depicted) were more an autochtonal population + some I-E italics (Umbrians) than a true etruscan one
more genrally I have some defiance against these comparisons with local populations like -'welsh', 'irish', 'german', 'basque'... no sense for me: only components have some value (and yet...) not global modern state or districts populations

Good points, Moesan. Let us know what you think after you read the paper.

I suppose part of my problem is that I've identified as purely Mediterranean my whole life. I should know better, of course...there's no "pure" anything. Still, these "northern" percentages seem really high. If I average the two groups, I come out about 39/61. My poor father, God rest his soul, must be turning over in his grave!

Ed. I don't mean any offense...absolutely nothing wrong with having northern ancestry, and much of which to be proud. It's just that my father was convinced and had me convinced when I was younger that we were the "pure" descendents of the Etruscans and the Romans. So much for that! :)

Sile
14-02-14, 20:34
I haven't plowed through the supplement yet, but I did look at the companion site quickly to check the results for northern Italians and Tuscans, as those are my two ancestral groups.

I find the ratio of "Cypriot" like, to "Welsh" like for the northern Italians, and "Cypriot" like and "French" like for the Tuscans a little surprising:
Northern Italians: 33/63
Tuscans: 45/55

I'm assuming that the "Cypriot" like population is the pre-Gallic "native" population? So, does that mean that the Italics, including the Romans, were "Cypriot" like, or just that the Italics were so low in number that they didn't change the prior EEF population very much, to use a Lazaridis term?

I suppose you could say that the 48% specifically English and Welsh for the northern Italians could be correlated with the Gallic invasions. The paper's date for the admixture, which is approximately in the middle of the first millennium B. C. would fit. But if the 8.6% German is a stand in for the Longobard admixture, that would have arrived much later in the AD/CE period.

Things are even more confusing for the Tuscans. It makes sense that their English and Welsh would be less, about 27% versus 48% for the northern Italians, but why do Tuscans have almost double the amount of German as the northern Italians? Everything I know says that the Langobard presence was stronger in the north. And why is the admixture dated so late in the Common Era for them, much later than for the northern Italians? The lowest edge of the range barely overlaps with the Lombard invasion, but what about the "British" admixture? Unless, as these dating programs seem to do, it's only picking up the last admixture date and folding in the older ones. That still doesn't explain why the specifically German one is higher than in northern Italians, however.

There's also the conflict with the Ralph and Coop paper to consider. While they did pick up the Gallic admixture in the middle of the first millenium B.C., they are very clear that they saw very minimal admixture in Italy after that time. So, which approach is more accurate? Perhaps Ralph and Coop as they are looking at specific, very small "bites" of the genome?

Well, if this paper is correct, it would explain why earlier work showed that while modern Tuscans are very similar to early medieval Tuscans, they're not similar at all, at least in terms of mtDNA, to the Etruscans. So, we full or half Tuscans have the Renaissance, but not the Etruscans, at least genetically? What a disappointment if that's true.

Also, it would prove how unreliable phenotypes can be...I constantly see people who look as if they are Etruscan statues come to life. The power of dominant genes perhaps?

It also occurs to me to wonder how this correlates to the yDNA signatures in Italy, especially R1b U-152. Is it Gallic then? If it is, what does that say about the millions of words expended trying to explain the arrival of the Indo-European languages like Italic into Europe with steppe herders in the second millennium B.C.? And what haplogroup would have carried the "Germanic" input? In northern Italy, you might say U-106, but what about in Toscana?

And how does this correlate to Lazaridis et al proportions for all the European populations?

Tons of questions, I know, but I wanted to put them out there...

I think people need to look deeper into what the "buttons" refer to.......like, the north-Italian button is the Bergamo admixture one, which refers to also tyrolese and eastern swiss ( as well as north italian ) people.
The hungarian would incorporate slovaks, ancient pannonia area etc etc

Sile
14-02-14, 20:41
Good points, Moesan. Let us know what you think after you read the paper.

I suppose part of my problem is that I've identified as purely Mediterranean my whole life. I should know better, of course...there's no "pure" anything. Still, these "northern" percentages seem really high. If I average the two groups, I come out about 39/61. My poor father, God rest his soul, must be turning over in his grave!

Ed. I don't mean any offense...absolutely nothing wrong with having northern ancestry, and much of which to be proud. It's just that my father was convinced and had me convinced when I was younger that we were the "pure" descendents of the Etruscans and the Romans. So much for that! :)

I read these mixture reading as going both ways. how are you reading it?

as an example. my ftdna haplogroup origns ( a very powerful indicator as mentioned by experts) under ydna has only 2 names, italy and ireland, to me it refers to irish have my italian mix and my italian has irish mix.
These buttons on this admixture map refer to the same system ( going both ways)

Angela
14-02-14, 21:18
I read these mixture reading as going both ways. how are you reading it?

as an example. my ftdna haplogroup origns ( a very powerful indicator as mentioned by experts) under ydna has only 2 names, italy and ireland, to me it refers to irish have my italian mix and my italian has irish mix.
These buttons on this admixture map refer to the same system ( going both ways)

But the admixture doesn't go both ways in this analysis unless I missed something. Northern Italians are 48% English and Welsh, but those groups are not 48% Italian. By all means check the other nationalities...maybe I got something wrong. I did it late last night.

As for yDNA haplogroups, they're far less informative in my opinion than autosomal DNA. It's only one marker. Someone could share the exact same subclade as you do, and because it is unusual in his country, he could share very little recent autosomal admixture.

Sile
14-02-14, 21:23
But the admixture doesn't go both ways in this analysis unless I missed something. Northern Italians are 48% English and Welsh, but those groups are not 48% Italian. By all means check the other nationalities...maybe I got something wrong. I did it late last night.

As for yDNA haplogroups, they're far less informative in my opinion than autosomal DNA. It's only one marker. Someone could share the exact same subclade as you do, and because it is unusual in his country, he could share very little recent autosomal admixture.

?
I just checked some samples
welsh to north-italian = 12.1%
north-italian to welsh = 11.6%

ok its not 100% , but .5 of a percent is close enough for me

Angela
14-02-14, 21:42
?
I just checked some samples
welsh to north-italian = 12.1%
north-italian to welsh = 11.6%

ok its not 100% , but .5 of a percent is close enough for me

O.K., we're comparing apples and oranges. The 48% which I jotted down and posted about is the total for English (36.2) and Welsh (12.1). I see the correspondence for the Welsh, but the English aren't listed as having any Italian admixture at all.

The English do have Welsh admixture according to this (41%), but even if you take 41% of the Welsh southern and northern Italian total (about 25%) you don't get anywhere near 36%.

Sile
14-02-14, 23:07
O.K., we're comparing apples and oranges. The 48% which I jotted down and posted about is the total for English (36.2) and Welsh (12.1). I see the correspondence for the Welsh, but the English aren't listed as having any Italian admixture at all.

The English do have Welsh admixture according to this (41%), but even if you take 41% of the Welsh southern and northern Italian total (about 25%) you don't get anywhere near 36%.

we differ on what we read
The more you delve into it the more you get out of it, especially the eastern european markers with the different split you can get ( top right corner)

tjlowery87
15-02-14, 02:13
I can not get the links to work on my computer .What does it say about england?thanks

Sile
15-02-14, 02:42
I can not get the links to work on my computer .What does it say about england?thanks

sometimes its takes 10 seconds to load and sometimes 2 minutes

Alan
15-02-14, 14:48
Some aspects are interesting, but as Dienekes said, it contradicts the archeological evidence.

For instance 'Scottish' has overwhelming 'Irish' ancestry (no surprise), but 'Welsh' have zero 'Irish'.
The 'Welsh' are about half 'Italian', but 'Scottish' are zero 'Italian'.
That makes no sense. Or it shows only very recent admixtures.
Interesting though again the 'Baloch' and 'Lezgin' link for 'Scottish'.


Thats what I thought too. It explains well the relation of populations by modern "admixture" but it completely ignores how these admixture portions are made up themselves. I mean, I see ~20% Scottish in Kalash, yet no Kalash in Scots. Why especially Scottish? And I doubt that any Scott moved down the South-Central Asia.

MOESAN
16-02-14, 18:00
so English people had a lot of Welsh but French people has a lot of English without Welsh.....................................
I'm too old!...
my poor brain is heating too much when I try to understand these "admixtures" calculations - as others here I have the impression carrots, potatoes and bananas have been mixed -
perhaps the new basic elements of comparisons are banarrots, potananas, carrotoes, potarrotonas...??? I'm going to take my pills and go to bed without supper nor even a drink!
I'm destroyed!

Angela
17-02-14, 06:56
so English people had a lot of Welsh but French people has a lot of English without Welsh.....................................
I'm too old!...
my poor brain is heating too much when I try to understand these "admixtures" calculations - as others here I have the impression carrots, potatoes and bananas have been mixed -
perhaps the new basic elements of comparisons are banarrots, potananas, carrotoes, potarrotonas...??? I'm going to take my pills and go to bed without supper nor even a drink!
I'm destroyed!

I have my issues with the conclusions drawn in this paper, but even with papers that make sense to me, like Lazaridis et al or Ralph and Coop, while I try to blunder through the math, at a certain point I just have to take it on faith...that or spend all my time on this. I thought I was pretty good at mathematics, but either I've forgotten a lot, or they've gone beyond me, lol.

LeBrok
17-02-14, 09:35
so English people had a lot of Welsh but French people has a lot of English without Welsh.....................................
I'm too old!...
my poor brain is heating too much when I try to understand these "admixtures" calculations - as others here I have the impression carrots, potatoes and bananas have been mixed -
perhaps the new basic elements of comparisons are banarrots, potananas, carrotoes, potarrotonas...??? I'm going to take my pills and go to bed without supper nor even a drink!
I'm destroyed!

I have my issues with the conclusions drawn in this paper, but even with papers that make sense to me, like Lazaridis et al or Ralph and Coop, while I try to blunder through the math, at a certain point I just have to take it on faith...that or spend all my time on this. I thought I was pretty good at mathematics, but either I've forgotten a lot, or they've gone beyond me, lol.
Lol, it sounds so human and you make my heart soft. This surely takes extreme brain gymnastics.
Thanks for opening up.

Sile
17-02-14, 12:01
I have my issues with the conclusions drawn in this paper, but even with papers that make sense to me, like Lazaridis et al or Ralph and Coop, while I try to blunder through the math, at a certain point I just have to take it on faith...that or spend all my time on this. I thought I was pretty good at mathematics, but either I've forgotten a lot, or they've gone beyond me, lol.

What maths do you need?

The arrows go in one direction..........example some of Tuscany ( arrow going to tuscany )
English (18.6%)GermanyAustria (15.0%)Welsh (8.1%)NorthItalian (6.5%)French (4.4%

You don't expect 15.0% of tuscans to go back to Germany..........do you.? the 15% germans would be replaced by someone else. That shows how migrations work

Basically each marker is separate and not aligned with each other...makes sense to me

Vallicanus
17-02-14, 18:41
Why trust a genetic atlas with samples of SIX or EIGHT!

ebAmerican
17-02-14, 20:36
For instance 'Scottish' has overwhelming 'Irish' ancestry (no surprise), but 'Welsh' have zero 'Irish'.
The 'Welsh' are about half 'Italian', but 'Scottish' are zero 'Italian'.
That makes no sense. Or it shows only very recent admixtures.
Interesting though again the 'Baloch' and 'Lezgin' link for 'Scottish'.

The Welsh dilemma is kind of a stumper, seeing that 10% of Brits have an Irish grandparent. The sample is realy small only 4 individuals. Going through the different nationalities most samples are small. I don't see how 50 samples could give you an accurate picture of an entire countries autosomal picture. It's a start and will be interesting to periodically take a look and see if their sample base increases.

tjlowery87
17-02-14, 22:13
they didnt add the dutch,danes or belguim

Angela
21-02-14, 02:04
What maths do you need?

The arrows go in one direction..........example some of Tuscany ( arrow going to tuscany )
English (18.6%)GermanyAustria (15.0%)Welsh (8.1%)NorthItalian (6.5%)French (4.4%

You don't expect 15.0% of tuscans to go back to Germany..........do you.? the 15% germans would be replaced by someone else. That shows how migrations work

Basically each marker is separate and not aligned with each other...makes sense to me


I don't blindly accept a paper's conclusions. I always at least attempt to check the assumptions, the methodology, the programs, and yes, the mathematics.

As for the direction of flow, I believe it was you who said the flow went both ways. I don't think I ever made any such claim.

I have a few issues with the conclusions drawn by the authors of this paper as regards Italy. For one, let's take the postulated date for the admixture (in the 900's) in Toscana. There is no large migration into the area in that century. The closest might be the Lombards, but the Duchy of Tuscia was constituted by them in the year 570, centered around their capital in Lucca. Now, as the Lombards initially instituted various laws against intermarriage between themselves and the "Romans", perhaps one could say the intermarriage only happened later? Still, the sources with which I'm familiar hold that the amalgamation of Langobard and Roman was well underway within a hundred years of the invasion. Or, maybe it was the arrival of the Frankish elite which led to the amalgamation of the Romans and the Lombards. Still, even that occurred in the 700's.

This online book provides quite a bit of good detail about the period:
http://books.google.com/books?id=lAAf3TPFta8C&pg=PA65&dq=How+many+Lombards+invaded+Italy&hl=en&sa=X&ei=ZLcDU86rKMqU0QHsmICoBA&ved=0CD4Q6AEwBA#v=onepage&q=How%20many%20Lombards%20invaded%20Italy&f=false

As does this book by Patrick Geary:
http://books.google.com/books?id=A26s-v2eEwAC&pg=PA125&lpg=PA125&dq=Population+of+Italy+at+the+time+of+the+invasion +by+the+Lombards&source=bl&ots=yJHx6hEUgY&sig=0H85jJXmf0rMQydf7vf3J8nu0KA&hl=en&sa=X&ei=_NEDU6ypD-jS0wGe-4HIDA&ved=0CFkQ6AEwCDgK#v=onepage&q=Population%20of%20Italy%20at%20the%20time%20of%2 0the%20invasion%20by%20the%20Lombards&f=false

Also, what is one to make of the fact that while 15% of the Tuscan genome is attributed to "Germans", 18.6% is English, and 8.1% is "Welsh". (Let's ignore the French, Basque etc. and other minor components for now.) Assuming for the moment that this almost 27% of the genome(English and Welsh) is attributable to general central European "Celtic"/"Germanic" alleles, are we supposed to assume that these elements were part of the greater Langobard genome that arrived in Italy in the sixth century? I know that the Langobards picked up other tribes in their migrations, but we've had no indication that they were such a large percentage of the general Langobard force.

There's also the fact that given the numbers posited for the Langobards and assorted barbarian tribes involved in the sixth century invasion, you would have to assume an almost complete depopulation of the Italian peninsula for that number of people to have this kind of genomic effect. There is indeed controversy as to the the numbers of people that remained in the peninsula after the horrors of the Gothic Wars and the attendant famine and disease, but even the most pessimistic accounts put the number in the millions. I think we can take some of Paul the Deacon's account as hyperbole.

A discussion of the small numbers involved in the Lombard invasion can be found here:
http://rbedrosian.com/Ref/Bury/ieb14.htm

So, while the authors of this paper claim that their dating method is far superior to the results available from the Roll-Off program, I'm not convinced.

The results also would seem to conflict with the IBD analysis by Ralph and Coop, who see no admixture in these areas of Italy after the arrival of the Gauls or Celts in the first millennium B.C. Then, there's the conflict with the findings of Lazaridis et al. How could the northern Italians, who are about 70% (?) EEF have 63% of their ancestry from north of the Alps?

While in the end they may be proven to be correct, as things stand now, some things don't quite add up.

Sile
21-02-14, 06:55
I don't blindly accept a paper's conclusions. I always at least attempt to check the assumptions, the methodology, the programs, and yes, the mathematics.

As for the direction of flow, I believe it was you who said the flow went both ways. I don't think I ever made any such claim.

I have a few issues with the conclusions drawn by the authors of this paper as regards Italy. For one, let's take the postulated date for the admixture (in the 900's) in Toscana. There is no large migration into the area in that century. The closest might be the Lombards, but the Duchy of Tuscia was constituted by them in the year 570, centered around their capital in Lucca. Now, as the Lombards initially instituted various laws against intermarriage between themselves and the "Romans", perhaps one could say the intermarriage only happened later? Still, the sources with which I'm familiar hold that the amalgamation of Langobard and Roman was well underway within a hundred years of the invasion. Or, maybe it was the arrival of the Frankish elite which led to the amalgamation of the Romans and the Lombards. Still, even that occurred in the 700's.

This online book provides quite a bit of good detail about the period:
http://books.google.com/books?id=lAAf3TPFta8C&pg=PA65&dq=How+many+Lombards+invaded+Italy&hl=en&sa=X&ei=ZLcDU86rKMqU0QHsmICoBA&ved=0CD4Q6AEwBA#v=onepage&q=How%20many%20Lombards%20invaded%20Italy&f=false

As does this book by Patrick Geary:
http://books.google.com/books?id=A26s-v2eEwAC&pg=PA125&lpg=PA125&dq=Population+of+Italy+at+the+time+of+the+invasion +by+the+Lombards&source=bl&ots=yJHx6hEUgY&sig=0H85jJXmf0rMQydf7vf3J8nu0KA&hl=en&sa=X&ei=_NEDU6ypD-jS0wGe-4HIDA&ved=0CFkQ6AEwCDgK#v=onepage&q=Population%20of%20Italy%20at%20the%20time%20of%2 0the%20invasion%20by%20the%20Lombards&f=false

Also, what is one to make of the fact that while 15% of the Tuscan genome is attributed to "Germans", 18.6% is English, and 8.1% is "Welsh". (Let's ignore the French, Basque etc. and other minor components for now.) Assuming for the moment that this almost 27% of the genome(English and Welsh) is attributable to general central European "Celtic"/"Germanic" alleles, are we supposed to assume that these elements were part of the greater Langobard genome that arrived in Italy in the sixth century? I know that the Langobards picked up other tribes in their migrations, but we've had no indication that they were such a large percentage of the general Langobard force.

There's also the fact that given the numbers posited for the Langobards and assorted barbarian tribes involved in the sixth century invasion, you would have to assume an almost complete depopulation of the Italian peninsula for that number of people to have this kind of genomic effect. There is indeed controversy as to the the numbers of people that remained in the peninsula after the horrors of the Gothic Wars and the attendant famine and disease, but even the most pessimistic accounts put the number in the millions. I think we can take some of Paul the Deacon's account as hyperbole.

A discussion of the small numbers involved in the Lombard invasion can be found here:
http://rbedrosian.com/Ref/Bury/ieb14.htm

So, while the authors of this paper claim that their dating method is far superior to the results available from the Roll-Off program, I'm not convinced.

The results also would seem to conflict with the IBD analysis by Ralph and Coop, who see no admixture in these areas of Italy after the arrival of the Gauls or Celts in the first millennium B.C. Then, there's the conflict with the findings of Lazaridis et al. How could the northern Italians, who are about 70% (?) EEF have 63% of their ancestry from north of the Alps?

While in the end they may be proven to be correct, as things stand now, some things don't quite add up.

I don't blindly accept a paper's conclusions. I always at least attempt to check the assumptions, the methodology, the programs, and yes, the mathematics.

As for the direction of flow, I believe it was you who said the flow went both ways. I don't think I ever made any such claim.

I have a few issues with the conclusions drawn by the authors of this paper as regards Italy. For one, let's take the postulated date for the admixture (in the 900's) in Toscana. There is no large migration into the area in that century. The closest might be the Lombards, but the Duchy of Tuscia was constituted by them in the year 570, centered around their capital in Lucca. Now, as the Lombards initially instituted various laws against intermarriage between themselves and the "Romans", perhaps one could say the intermarriage only happened later? Still, the sources with which I'm familiar hold that the amalgamation of Langobard and Roman was well underway within a hundred years of the invasion. Or, maybe it was the arrival of the Frankish elite which led to the amalgamation of the Romans and the Lombards. Still, even that occurred in the 700's.

This online book provides quite a bit of good detail about the period:
http://books.google.com/books?id=lAAf3TPFta8C&pg=PA65&dq=How+many+Lombards+invaded+Italy&hl=en&sa=X&ei=ZLcDU86rKMqU0QHsmICoBA&ved=0CD4Q6AEwBA#v=onepage&q=How%20many%20Lombards%20invaded%20Italy&f=false
[/QUOTE]

Yes, it was I in error, but after re-reading it , it is only way way traffic.

As per your tuscan time frame, it states between 400AD to 1200AD. The north-italian component with german included in it was most likely be Lombard with Rugii and Heruli with it ( lombards -milan area, Rugii-east lake garda area and heruli = Concordia-friuli area), the others would be the remains of the ostro-goths. Do you know how many samples?


As does this book by Patrick Geary:
http://books.google.com/books?id=A26s-v2eEwAC&pg=PA125&lpg=PA125&dq=Population+of+Italy+at+the+time+of+the+invasion +by+the+Lombards&source=bl&ots=yJHx6hEUgY&sig=0H85jJXmf0rMQydf7vf3J8nu0KA&hl=en&sa=X&ei=_NEDU6ypD-jS0wGe-4HIDA&ved=0CFkQ6AEwCDgK#v=onepage&q=Population%20of%20Italy%20at%20the%20time%20of%2 0the%20invasion%20by%20the%20Lombards&f=false


Also, what is one to make of the fact that while 15% of the Tuscan genome is attributed to "Germans", 18.6% is English, and 8.1% is "Welsh". (Let's ignore the French, Basque etc. and other minor components for now.) Assuming for the moment that this almost 27% of the genome(English and Welsh) is attributable to general central European "Celtic"/"Germanic" alleles, are we supposed to assume that these elements were part of the greater Langobard genome that arrived in Italy in the sixth century? I know that the Langobards picked up other tribes in their migrations, but we've had no indication that they were such a large percentage of the general Langobard force.


you have burgundian, frankish ( swabia and alemanni ), the welsh is celt and the english..unsure


There's also the fact that given the numbers posited for the Langobards and assorted barbarian tribes involved in the sixth century invasion, you would have to assume an almost complete depopulation of the Italian peninsula for that number of people to have this kind of genomic effect. There is indeed controversy as to the the numbers of people that remained in the peninsula after the horrors of the Gothic Wars and the attendant famine and disease, but even the most pessimistic accounts put the number in the millions. I think we can take some of Paul the Deacon's account as hyperbole.

there was only about 150,000 lombards, the rest where others.


A discussion of the small numbers involved in the Lombard invasion can be found here:
http://rbedrosian.com/Ref/Bury/ieb14.htm

So, while the authors of this paper claim that their dating method is far superior to the results available from the Roll-Off program, I'm not convinced.

The results also would seem to conflict with the IBD analysis by Ralph and Coop, who see no admixture in these areas of Italy after the arrival of the Gauls or Celts in the first millennium B.C. Then, there's the conflict with the findings of Lazaridis et al. How could the northern Italians, who are about 70% (?) EEF have 63% of their ancestry from north of the Alps?

While in the end they may be proven to be correct, as things stand now, some things don't quite add up.[/QUOTE]

I imagine , the more samples they use over time the more accurate it will be

Sile
21-02-14, 07:08
granted there are not many samples and as yet I cannot get full access to the paper.

some access I can see

http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE53626

http://www.admixturemap.paintmychromosomes.com/

http://www.mpg.de/7910586/human-genetic-history-map (http://www.admixturemap.paintmychromosomes.com/)

Greying Wanderer
21-02-14, 12:25
@Angela

The three (non-expert) random thoughts i had about this were

1) What effect does speed have on the timing? A Roman foundation myth type admixture event would be very fast but what if populations A / B are adjacent plains / mountain people and only inter-marry very slowly over many centuries?

2) Similar to the above what effect might selection in place have? Example, one population conquers another but the female part of the conquered population has some beneficial trait so over time although the male proportion of the conquering population increases as you might expect the female side of the conquered population increases in frequency also.

3) What happens if you have multiple unfinished admixture events like point (1) in the same region e.g. your Tuscan example, assume for the sake of argument the "Welsh" component is a Ligurian substrate, the "English" is a Gallo-Celt substrate and the "German" is the Lombards all from different time periods. How might that jumble up the admixture date? Might they all be timed to the last event?

Angela
21-02-14, 17:59
@Angela

The three (non-expert) random thoughts i had about this were

1) What effect does speed have on the timing? A Roman foundation myth type admixture event would be very fast but what if populations A / B are adjacent plains / mountain people and only inter-marry very slowly over many centuries?

2) Similar to the above what effect might selection in place have? Example, one population conquers another but the female part of the conquered population has some beneficial trait so over time although the male proportion of the conquering population increases as you might expect the female side of the conquered population increases in frequency also.

3) What happens if you have multiple unfinished admixture events like point (1) in the same region e.g. your Tuscan example, assume for the sake of argument the "Welsh" component is a Ligurian substrate, the "English" is a Gallo-Celt substrate and the "German" is the Lombards all from different time periods. How might that jumble up the admixture date? Might they all be timed to the last event?

I agree with all three of your points. One of my main issues is with their dating. Roll-off, imo, produces a date which corresponds only to the last date of admixture. Despite the protestations of the authors that their dating methodology is superior to Roll-off, I think it suffers from the same problem. Another example of the issue I have with the dating is the fact that the northern Italian admixture event is dated to to the time of the Gallic migration, which makes sense, but the admixture event is so much later for the Tuscans. Also, such an early admixture date for the northern Italians leaves the Lombards out altogether.

Sticking with the Tuscan example, the Welsh and some of the English alleles might have come to Tuscany with the Italic peoples who spoke a related language to Celtic, or, as you say, from the Ligures, who were at least partly Indo European according to the latest analysis of their language. That gene flow would have been enriched by the Gallic invasions. Then, I can make sense of the about 15% Germanic by attributing it to the Lombards, although, as I said, Ralph and Coop found no admixture after the Gauls, and the Lombard influence has always been held to have been the greatest in Friuli and the Veneto, and to a lesser extent, Lombardia.

I'm still rather surprised by this 55% number for northern admixture in Tuscans, and 63% in northern Italians, but it's possible, I suppose, although as I said, that seems to contradict the EEF number for northern Italians and Tuscans as per Lazaridis et al.

Perhaps when they have some more samples, and they fix up their dating methodology, this will make more sense.

Oh, as to your point number 2, I don't know that it would apply to northern Italy and Toscana, at least as applies to the Gallic and Lombard invasions. Both were more in the nature of folk movements. The Ostrogoths, Huns, etc., were more in the nature of small bands of warriors. I doubt they had much of a genomic impact at all. The Indo-Europeans, I don't know. I think that whole migration needs to be fleshed out by a lot of ancient DNA.

MOESAN
21-02-14, 18:14
What maths do you need?

The arrows go in one direction..........example some of Tuscany ( arrow going to tuscany )
English (18.6%)GermanyAustria (15.0%)Welsh (8.1%)NorthItalian (6.5%)French (4.4%

You don't expect 15.0% of tuscans to go back to Germany..........do you.? the 15% germans would be replaced by someone else. That shows how migrations work

Basically each marker is separate and not aligned with each other...makes sense to me

sorry but making maths with not well identified elements (or worst: varying elements) cannot send us very far ... or too far!) the 'english' elements in Italy are surely only SOME of the elements that produced "english" population and reached Italy WITHOUT BEING PASSED THROUGH England at any time, for the most: what worth for History of populations moves???... Or I'm an idiot (possible too: it will never distroy my appetite for life)

Sile
21-02-14, 22:31
maybe this link below is a better guide

http://www.nature.com/ejhg/journal/vaop/ncurrent/full/ejhg20141a.html



here are where samples and fst placements are from
Suppl. Table 1. Populations are shown with their geographic centres, obtained from [17], and the latitudes and longitudes of those centres [18]


Population
City
Latitude
Longitude


Czech Republic
Cihost


49.74


15.34



Finland
Kajaani


64.23


27.73


France
Paris


48.52


2.2


Germany
Thuringen


51.01


10.85


Greece
Delphi


38.48


22.49


North Italy
Verona


45.44


10.99


South Italy
Naples


41.54


12.29


Netherlands
Utrecht


52.09


5.12


Norway
Ogndalsfjella


64.02


11.53


Poland
Piatek


52.07


19.48


Spain
Madrid


40.24


3.41


Sweden
Flataklocken


62.38


16.3


UK
Slaidburn


53.98
-2.45

Angela
22-02-14, 01:16
Sile;426903]maybe this link below is a better guide

http://www.nature.com/ejhg/journal/vaop/ncurrent/full/ejhg20141a.html



Thanks for the link to this paper. I'll read it this week-end.

Meanwhile, I did look at a couple of the figures. Could you check the legend on the PCA's? I'm slightly color blind so I might be getting this wrong, but it seems as if they have the Northern Italy and Southern Italy labels switched.

Sile
22-02-14, 04:00
[QUOTE=


Thanks for the link to this paper. I'll read it this week-end.

Meanwhile, I did look at a couple of the figures. Could you check the legend on the PCA's? I'm slightly color blind so I might be getting this wrong, but it seems as if they have the Northern Italy and Southern Italy labels switched.

looks fine

But only 4 Germans where tested, so this portion is exaggerated and also means others that replaced German are exaggerated .

only 8 tuscans tested

12 north -italians

28 french .................indicates not much migration from there

Greying Wanderer
22-02-14, 08:08
@Angela

"I'm still rather surprised by this 55% number for northern admixture in Tuscans, and 63% in northern Italians, but it's possible, I suppose, although as I said, that seems to contradict the EEF number for northern Italians and Tuscans as per Lazaridis et al."

Isn't EEF 50% basal and 50% WHG (or WHG-like)?

.

"Oh, as to your point number 2, I don't know that it would apply to northern Italy and Toscana"

Yeah I was thinking more central Europe. If LBK was over-run but x% of the female side remained and if the farmers had lots of useful genes then it seems plausible that the farmer proportion might increase over time through the female line. This might not effect the admixture calcs but I wondered if an increasing mixture - say from 20% to 40% - through a process of selection in place might show up as admixture with a skewed date - if you see what I mean.

Angela
22-02-14, 20:30
@Angela

"I'm still rather surprised by this 55% number for northern admixture in Tuscans, and 63% in northern Italians, but it's possible, I suppose, although as I said, that seems to contradict the EEF number for northern Italians and Tuscans as per Lazaridis et al."

Isn't EEF 50% basal and 50% WHG (or WHG-like)?

.

"Oh, as to your point number 2, I don't know that it would apply to northern Italy and Toscana"

Yeah I was thinking more central Europe. If LBK was over-run but x% of the female side remained and if the farmers had lots of useful genes then it seems plausible that the farmer proportion might increase over time through the female line. This might not effect the admixture calcs but I wondered if an increasing mixture - say from 20% to 40% - through a process of selection in place might show up as admixture with a skewed date - if you see what I mean.



EEF, which is represented by the Stuttgart woman, is approximately 44% Basal Eurasian, and the rest is West Eurasian on their charts. Their analysis indicated that some proportion of that West Eurasian is probably attributable to WHG, but they were unable to specifically quantify the amount, because we don't yet have a genome from a West Asian farmer. Their range for that WHG component is from 2 to 32%. In some parts of the paper they use an average amount of about 18%. If we use that figure, then EEF is 44% Basal Eurasian, 18% WHG, and 38% ancestral West Eurasian.

What I was looking at is the EEF score for Northern Italians currently, which is about 70%. If I remember correctly, the EEF score for the Germans and the English hovers around the 50% mark.
If I accept the conclusion of this paper we are currently discussing, there was a major admixture event in the middle of the first millennium B.C. in northern Italy, which would correlate with the Gallic invasions. Let's assume for a moment that the population of northern Italy at that time was 100% EEF, or close to it. Now, I think we could use the current 50% figure for the EEF of the invaders, as there has been no migration from the south into the north since that time. Therefore, one of my questions is...would the admixture of a 50% EEF population with a 100% EEF population result in a population that is currently 70% EEF? I think, generally, that isn't a bad figure.

However, wouldn't a population in which 63% of the alleles can be attributed to populations north of the Alps be less EEF? Somewhere around 55% perhaps? That's why that 63% "northern" figure seems high to me.

The only way it would make sense is if their dating program is off, and the northern alleles also arrived with the Indo-European Italics, and therefore the northern Italians already had some amount of northern alleles by the time of the Gallic invasions. This would all apply to the Tuscans as well, just at a slightly lower level, since they show up in this paper as a 55% generally "northern" people.

There, I've now convinced myself that the admixture levels are reasonable so long as the dating program is incorrect. :)

If I've gone wrong somewhere, I hope someone will post and point out the errors.


As to your point about the LBK, I think I understand what you're getting at, but if you're speaking about the late Neolithic/early metal ages intrusions supposedly attributable to the Indo-Europeans, it seems to me that the mtDNA evidence shows that they didn't bring all that much new mtDNA into Europe. See:
http://dienekes.blogspot.com/2013/10/ancient-central-european-mtdna-across.html

In fact, I'm leaning toward the view that the "Indo-Europeans" probably already had an EEF component, as well as ANE and probably WHG. The exact proportions will of course have to wait for genome sequencing of some of them. And I hope they sequence samples not only from the steppe and the line of kurgans into Hungary, but from the population(s) that moved on from there, if they can find traces of them. As these people moved west, they probably absorbed more and more of the EEF from the people they encountered. After all, how else would Germans wind up about 50% EEF?

I also don't think that admixture works in the way that you envisage. The alleles attributable to the ability to digest lactose, or for light skin, or other such advantageous adaptations for a neolithic diet in central and northern Europe account for only a very small percentage of the autosomes.

Angela
22-02-14, 21:36
[QUOTE=Angela;426909]

looks fine

But only 4 Germans where tested, so this portion is exaggerated and also means others that replaced German are exaggerated .

only 8 tuscans tested

12 north -italians

28 french .................indicates not much migration from there

Sile, they definitely screwed it up. Hard to believe, I know, but look at the page of fst results:
http://www.nature.com/ejhg/journal/vaop/ncurrent/fig_tab/ejhg20141t3.html#figure-title

It's the northern Italians who should have an FST distance to the Spanish, the French, and the Greeks of .001, and to the Germans of .002.

And it's the southern Italians who should have an FST of .001 to the Greeks, .002 to the Spanish, .003 to the French, and .004 to the Germans.

Otherwise, everything we know about Italian genetics and European genetics would be wrong.

Similarly, on the PCA plot, it's the northern Italians who should be overlapping with the Spanish and the French, not the southern Italians. The colors and the labels would indicate that the southerners plot further north than the northern Italians.
http://www.nature.com/ejhg/journal/vaop/ncurrent/fig_tab/ejhg20141f2.html#figure-title

It's disgraceful when you think about it. This isn't even a preview...it's gone through peer review, supposedly.

Anyway, I noticed a few interesting things about the graphic:
l. The range in the southern Italian samples is interesting, larger than I thought it would be...a lot of variation there.

2. I also found it interesting that in this analysis the British seem to overlap much more with the Germans than with the French.

3. The Finns are even further away from other European populations in this analysis than in others I've seen.

Of course, PCA's are limited in their usefulness because they only show the first two principal components. Also, I always find the intra-European ones rather misleading. If you threw in east Asian and SSA populations and made it a global PCA, the positioning would change.

Greying Wanderer
22-02-14, 22:00
@Angela

"I also don't think that admixture works in the way that you envisage. The alleles attributable to the ability to digest lactose, or for light skin, or other such advantageous adaptations for a neolithic diet in central and northern Europe account for only a very small percentage of the autosomes."

Sure but you can't just marry 1% of a woman to get the SLC genes. You have to marry the whole woman so the kids will be 50/50. If you have admixture between two groups and a gene that was initially only carried by one of the populations has been strongly selected for then it seems to me their proportion of the autosomal DNA must have gone up dramatically from its initial level.

Example population A over-runs population B with the end result 100 A males, 100 A females and 20 B females. If the A females have an average of 1.9 surviving kids and the B females an average of 2.1 then over time the autosomal B DNA will go up from c. 10% to c. 50% (hmm?) Anyway, whatever the equilibrium level is it will be substantially above the start level because the selected for genes come in a full package.

ElHorsto
22-02-14, 22:18
As to your point about the LBK, I think I understand what you're getting at, but if you're speaking about the late Neolithic/early metal ages intrusions supposedly attributable to the Indo-Europeans, it seems to me that the mtDNA evidence shows that they didn't bring all that much new mtDNA into Europe. See:
http://dienekes.blogspot.com/2013/10/ancient-central-european-mtdna-across.html


Interesting the map were it can be seen than Britain is the only place in west Europe which was affected by mtDNA of all three metal-working cultures (CWC,BBC,UC), which supports Maciamo's and my humble opinion that Britain was particularly attractive for metal workers. Thanks for the link, it is often enlightening to take a look at past data and maps repeatedly.



In fact, I'm leaning toward the view that the "Indo-Europeans" probably already had an EEF component, as well as ANE and probably WHG. The exact proportions will of course have to wait for genome sequencing of some of them. And I hope they sequence samples not only from the steppe and the line of kurgans into Hungary, but from the population(s) that moved on from there, if they can find traces of them. As these people moved west, they probably absorbed more and more of the EEF from the people they encountered. After all, how else would Germans wind up about 50% EEF?


And the English too ;-)
I also think that EEF was important in at least part of IEans, yet I also think that IEans could have been diverse enough to have been much more ANE on one end (see ANE peak in Scotland), mostly ANE+WHG on another (Lithuanian language is very ancient IEan) and more EEF on another end (see the low WHG and perhaps too low ANE in several south european countries). But as you say, this diversity could have emerged by diverse subsequent admixtures into the 'original' IEans while moving west, depending on the route.

Sile
22-02-14, 22:54
[QUOTE=Sile;426919]

Sile, they definitely screwed it up. Hard to believe, I know, but look at the page of fst results:
http://www.nature.com/ejhg/journal/vaop/ncurrent/fig_tab/ejhg20141t3.html#figure-title

It's the northern Italians who should have an FST distance to the Spanish, the French, and the Greeks of .001, and to the Germans of .002.

And it's the southern Italians who should have an FST of .001 to the Greeks, .002 to the Spanish, .003 to the French, and .004 to the Germans.

Otherwise, everything we know about Italian genetics and European genetics would be wrong.

Similarly, on the PCA plot, it's the northern Italians who should be overlapping with the Spanish and the French, not the southern Italians. The colors and the labels would indicate that the southerners plot further north than the northern Italians.
http://www.nature.com/ejhg/journal/vaop/ncurrent/fig_tab/ejhg20141f2.html#figure-title

It's disgraceful when you think about it. This isn't even a preview...it's gone through peer review, supposedly.

Anyway, I noticed a few interesting things about the graphic:
l. The range in the southern Italian samples is interesting, larger than I thought it would be...a lot of variation there.

2. I also found it interesting that in this analysis the British seem to overlap much more with the Germans than with the French.

3. The Finns are even further away from other European populations in this analysis than in others I've seen.

Of course, PCA's are limited in their usefulness because they only show the first two principal components. Also, I always find the intra-European ones rather misleading. If you threw in east Asian and SSA populations and made it a global PCA, the positioning would change.

ignore my comments, i thought you where still talking about the first link and not the second link I provided. there are 203 north italians samples in the second link........but distance is from verona and not bergamo

MOESAN
11-04-14, 11:44
perhaps outdated observations of mine? it is about today components, without comparisons to ancient DNA

The Eurogenes-15 pools (type populations on the autosomals ground) seems to me better than the Dodecad (12B? The ones on Eupedia maps) ones concerning Europe, as a whole: principally concerning Western Europe where they seem having broken down the 'northwest european' block into a 'western' one and a 'northern' or 'north-sea' one:



in some way the Dodecad 'north-west-european' seems the cumul of the 'west-european' and 'north-sea'; the Eurogenes 'north-european'-'north-sea' has the merit to better check the classical 'nordic' collective phenotype, spite a slight predominence in West when comparing southern countries between them – and this in fact 'north-sea' component is not absent in the East of Europe, nor even in the Steppes -


by component now:
'mediterranean': Eurogenes is centered on Sardinia like Dodecad but as awhole more occidental (less in Italy, South specially, less in Greece, Balkans and Anatolia, less too in North Europe – more «local» in some way -

'northwest-european': broken down into 'north-european/»north-sea»' and 'west-european' by Eurogenes – the center of 'west-european' seems being in basque country at first and then on western coasts of Europe and in central to southern Spain, - strong enough yet in northern Europe and west-central Europe - it could seem linked to y-R1b but this is not so evident in eastern Europe (+ Balkans and Greece!) which shows a bit more density than Italy (only Russia and Ukraina have less or not more) - for 'north-european' the center is Scandinavia but also N-W Germany, the Netherlands and the British Isles with a gradient when coming southward or Eastward – I see some difference with the global 'northwest-european' of Dodecad in Bela-Russ and Romania where the cumul of the two pools does not show a so important lack as show the Dodecad unique pooling – (but I find the Dodecad drawing better cheks the Y-DNA markers distributions...

'east-european-baltic': Eurogenes not too different from the Dodecad map, seeming more precise for the center of density but not as precise for the less concerned regions – I find better the Dodecad map concerning eastern France, Flanders and Scotland, by instance - the two maps show the same slight impact in Greece (Slavs? Or before them?) -
-west-asian': Eurogenes close to Dodecad again here, but less precise again too – and here again, the pushing of people came through South is not so evident in Romania (and Italy) in the Eurogenes map (Starcevo-Cucuteni-Tripolje places) – also: the weight of this component is very less marked in N-E Africa compared to Dodecad – advantage of the Eurogenes map: it shows the South Asia regions for this component: it shows a push southwards from the Steppes North the Hindu Kush and into northern Afghanistan, reducing the weight of 'west-asian' there -

-'southwest-asian': no too big differences even if in Egypte the weight of this component seems decreasing in Eurogenes when we go southward, and at the contrary seems stronger in Italy and Anatolia, as opposed to the Dodecad map – the «hole» of Dodecad in Basque country and Catalonha does not appears in Eurogenes map -


my conclusions about the maps: Dodecad (+ Eupedia!) puts more distinction in the weak percentages regions, what is not without advantage – it seems too more precise in regions partitions (more data? - more local surveys?) - but the true progress in Eurogenes is the differenciation between 'west-european' and 'north-european' -



Now, what to put in the 'west-european' bag?: when? - from where? - what types? - it seems linked to an old population akin to a part of the basques ancestors, maybe 'cro-magnoid' types – not absent from north-eastern Europe but not dominent there where 'baltic' is the typical element; I 'm not sure these 'cromagnoids' are the only element present in this 'west-european'... – if I had the courage I could bet on a more 'brünnoid' (partly brachycephalized in a 'borrebylike' element) in the Baltic component, even if things are not so simple – these people could have been arrived in Europe after the LGM, with north Solutrean, coming from eastern far Steppes (the believing of some old scholars) – they could have met 'cromagnoids' came northeastward after LGM too, coming from West and N-Iberia – a ready-to-use thinking because we can imagine too some of the two origins could have met before that, even during the LGM, in central-north-central Europe (at Brussel latitude by instance), even if the 'Brünn' «cousin» 'Combe-capelle' appeared in West only about the 9000/7000 BC...

MOESAN
11-04-14, 14:31
[QUOTE=Sile;426919]

Sile, they definitely screwed it up. Hard to believe, I know, but look at the page of fst results:
http://www.nature.com/ejhg/journal/vaop/ncurrent/fig_tab/ejhg20141t3.html#figure-title

It's the northern Italians who should have an FST distance to the Spanish, the French, and the Greeks of .001, and to the Germans of .002.

And it's the southern Italians who should have an FST of .001 to the Greeks, .002 to the Spanish, .003 to the French, and .004 to the Germans.

Otherwise, everything we know about Italian genetics and European genetics would be wrong.

Similarly, on the PCA plot, it's the northern Italians who should be overlapping with the Spanish and the French, not the southern Italians. The colors and the labels would indicate that the southerners plot further north than the northern Italians.
http://www.nature.com/ejhg/journal/vaop/ncurrent/fig_tab/ejhg20141f2.html#figure-title

It's disgraceful when you think about it. This isn't even a preview...it's gone through peer review, supposedly.

Anyway, I noticed a few interesting things about the graphic:
l. The range in the southern Italian samples is interesting, larger than I thought it would be...a lot of variation there.

2. I also found it interesting that in this analysis the British seem to overlap much more with the Germans than with the French.

3. The Finns are even further away from other European populations in this analysis than in others I've seen.

Of course, PCA's are limited in their usefulness because they only show the first two principal components. Also, I always find the intra-European ones rather misleading. If you threw in east Asian and SSA populations and made it a global PCA, the positioning would change.

sorry cause I'm a bit l

MOESAN
11-04-14, 14:32
continue
Sorry cause I'm a bit late in the fight here: but as you Anglea, I think there has been an inversion between Southern and Northern Italy (technical paging error?)

joeyc
11-04-14, 19:21
Some of the conclusions of this study are very funny. Moreover the don't give any information about the samples they have used.