July 29, 2010

NordicDB paper on Finns, Danes, and Swedes

On the left is an MDS plot using ~45k SNPs. Some explanation on the datasets: CAPS are Swedish; SGENE and MS are Finnish (Helsinki region); Aneurysm is Finnish (Helsinki and Kupio).

A striking feature of the plot is the distinctiveness of the different Finish samples (light vs. dark brown points). This is not so difficult to explain if one considers that the light brown squares (DGI-FIN) are from Botnia. This parallels the results of Salmela et al. (2008) or Jakkula et al. (2008) in underscoring the internal structure of the population of Finland

The familiar V shape was also observed in the PCA produced by McEvoy et al. (2009) or Nelis et al. (2009). In my opinion, it is produced by the differential representation of the two main population elements of the Nordic countries, namely the Germanic and Finnic elements.

Here is the website of NordicDB.

European Journal of Human Genetics doi: 10.1038/ejhg.2010.112

NordicDB: a Nordic pool and portal for genome-wide control data

Monica Leu et al.

A cost-efficient way to increase power in a genetic association study is to pool controls from different sources. The genotyping effort can then be directed to large case series. The Nordic Control database, NordicDB, has been set up as a unique resource in the Nordic area and the data are available for authorized users through the web portal (http://www.nordicdb.org). The current version of NordicDB pools together high-density genome-wide SNP information from ~5000 controls originating from Finnish, Swedish and Danish studies and shows country-specific allele frequencies for SNP markers. The genetic homogeneity of the samples was investigated using multidimensional scaling (MDS) analysis and pairwise allele frequency differences between the studies. The plot of the first two MDS components showed excellent resemblance to the geographical placement of the samples, with a clear NW–SE gradient. We advise researchers to assess the impact of population structure when incorporating NordicDB controls in association studies. This harmonized Nordic database presents a unique genome-wide resource for future genetic association studies in the Nordic countries.

Link

11 comments:

onur said...

A clearer version of the same map can be found in the website of NordicDB with some information:

http://www.nordicdb.org/database/Map.html

onur said...

map = plot

Annie Mouse said...

To me what this map shows is a flow of genes between the UK and Scandinavia all the way up to Finland (pale orange).

The Scandinavian/Swedish green is also being pulled in another direction (Saami maybe? or perhaps another aboriginal).

There is also a separate flow between two different Finnish populations. One population is the one I mentioned above which is a the far end of the UK<->Scandinavia<->Finland flow (Atlantic coastal maybe?). The second separate population (dark brown) I would guess has a more eastern, perhaps asian influence (perhaps at the terminal of the inland flow between Asia and Europe).

aargiedude said...

Can someone detail the exact origin of all the samples in the graph?

Can someone with access to the study find out if it has genetic distances (FST)?

Is Botnia a precise historic region with agreed upon borders, or just a general term to refer to the region close to the Gulf of Bothnia?

What do you all think about the fact that Bothnia is closer to Sweden than any other place in Finland? Is this unexpected from the point of view of how Sweden interacted historically with Finland? Would it have been more expected that the southwest region, such as Helsinki, would be closer to Sweden?

onur said...

Argiedude, Helsinki samples probably consist of Finnish residents of Helsinki irrespective of whether they are orginally from there or recent migrants. So probably Helsinki samples are genetically an average of the whole Finnish population.

doug mcdonald said...

A comment about PCA plots, and the equivalent MDS ones:

Any PCA plot of a fairly continuous distribution of populations will have an L or U shaped distribution. It has to, mathematically. The second component cannot be a linear function of the first one, so its general nature has to be a quadratic function: a parabola. This is best seen in plots of purely Eastern Asian origin from the Yakuts or Mongolians in the north down to the Cambodians or Malays in the south.

But the usual European shape of the first two components is also from the same reason.
Doug McDonald

eurologist said...

In my opinion, it is produced by the differential representation of the two main population elements of the Nordic countries, namely the Germanic and Finnic elements.

Dienekes,

It looks to me that one can clearly distinguish a third component: the Baltic one - which is different from the other two. I believe there is a "Nordic" component that goes beyond Germanic and is common with many Swedes as well as NW Finns, and then the expected Baltic (regional paleolithic and early neolithic combined with SE finno-ugric - which could perhaps be separated at a higher level). That is, both the Nordic and Baltic deviate from central and northern central European.

It has to, mathematically. The second component cannot be a linear function of the first one, so its general nature has to be a quadratic function: a parabola.

Doug, could you explain this a bit more? That doesn't ring true to me. The second component is by definition orthogonal to the first, and plotted on a perpendicular axis. Populations harboring a larger value of this one with respect to others can be located anywhere along PC1: so you can have an L, a T,, a V, a U, a j - there is nothing quadratic about the relation to PC1, other than that it trivially cannot be linear.

Dienekes said...

I don't really see the Baltic component in Nordic countries.

If we are speaking of northern Europe more generally, it is quite clear that there is a distinction between Germanics and Balto-Slavs.

There are also clear clines within Germanics themselves (both towards the UK and towards C Europe), but it's not clear how these are formed. Potentially both differential admixture with pre-Germanic, pre-IE northern Europeans during the NW spread of IE languages, but also differentiatl admixture with non-Germanic languages (e.g. Celtic) during the S and W spread of Germanic languages may have played a role in creating such a cline.

eurologist said...

I meant Baltic as in ancient Baltic - not Slavo-Baltic. Of course, if you call this Finnish, it's just a matter of nomenclature - but I think the term Finnish is misleading.

This is the way I see it:

Finns are different from other (central) northern Europeans in three respects: they retain a larger fraction of the ancient northern European population, their admixture is with the most Nordic agriculturalists (northern Swedes), only, and and they have some recent Sami admixture (themselves highly related but isolated for a long time). Estonians are extremely close to Finns but lack the Sami admixture, and Latvians and Lithuanians are almost as close but have some recent Slavic admixture.

IMO, if you rotate the PC graph 45 degrees counter-clockwise, the vertical axis describes a Scandinavian-only cline between the most northern and northern-central European agriculturalists (Denmark = CEU!), while the horizontal describes the Baltic (but not Slavo-Baltic) gradient. So, as we know, some Finns are basically northern Swedes (but not Danish), whereas others fall more into the range of the ancient northern Baltic inhabitants (who continue to make up much of the genetic material in the Eastern Baltic, in general).

I agree it would be interesting to see how this would look with Estonian, Latvian, and Lithuanian populations included.

kenttsu said...

"I agree it would be interesting to see how this would look with Estonian, Latvian, and Lithuanian populations included."

I'm not sure if this is what you want but here's one with Sweden, Finland and Baltic countries.

http://blogs.discovermagazine.com/gnxp/2009/05/genetic-structure-of-eastern-european-populations/

Mundane man said...

Annie Mouse:"The second separate population (dark brown) I would guess has a more eastern, perhaps asian influence"

After all these years and all these studies people still insist there is an Asian influence?? It's not Asian, it's ancient Nordic/Baltic.