Genetic Structure of France

[Angela]

This promises to be very interesting. It was just presented at a conference, so all we have is the abstract (thanks to anthrogenica), but hopefully the paper will be out soon. It seems it's just about as predicted.

"Fine-Scale Human Genetic Structure in France

Aude Saint-Pierre, Céline Bellenguez, Luc Letenneur, Claudine Berr, Carole Dufouil, Philippe Amouyel,Emmanuelle Génin

Characterizing geographical population structure is critical to genetic studies of disease as it is an important cause of false positive results in genome wide association studies (GWASs). The genetic structure of several countries in Europe has been carefully studied but there is a lack of descriptive study of the French population. Indeed, apart from the work of Karakachoff et al (2015) that focused on the western part of France and detected interesting stratification, no study so far provided a comprehensive look at the French genetic landscape. Here we describe the genetic structure of the French population at a fine-scale using genetic data from the 3 Cities (3C) Study, a population cohort of French elderly individuals that served as controls in several GWAS conducted with French patients. From this cohort, we had access to 4,433 genotyped individuals sampled in three regions of France but born all over France.

We selected a subset of 770 individuals to cover evenly the different regions of France and applied methods that utilize haplotype information for detecting fine-scale population structure. The 770 individuals were partitioned into homogeneous clusters using CHROMOPAINTER and fineSTRUCTURE analysis (Lawson et al. 2012). Six clusters were identified that correlate well with the geographic origin of individuals. The coarsest level of genetic differentiation separates the samples from southwestern French from all the others. Subsequent splits reveal more subtle differentiation except for samples from western France which showed a relatively high degree of homogeneity.

For each cluster we used CHROMOPAINTER to estimate an “ancestry profile” which characterises the ancestry of the cluster as a mixture of the reference sample. Using the subsample of 770 individuals as a reference sample to assign the remaining 3C individuals, we found that the cluster assignment was coherent with the places of birth of individuals. The same procedure was applied using the five European samples from the 1000 Genomes Project as a reference sample. Contribution from European populations shows a cline roughly north-south, in ancestry profiles. Spain (IBS) is the largest contributor of the southwest and south clusters while the highest contribution of Great-Britain (GBR) population is observed in Brittany.

In conclusion, we provide evidence that there exist some levels of genetic stratification in France. The French population could roughly be divided into 6 genetic clusters that correlate well with geography. The knowledge of this stratification pattern will be useful to design robust and powerful association studies."

 

[LeBrok]

6 genetic clusters? It will be interesting to see how they are grouped.

 

[berun]

Let's see methodology applied but results seem OK with history: Vascones in Gascony and Britons in Brittany; there was also a cultural Iberian presence in 3 southern departments but the "south" is too big. "European populations" marking a cline north-south are the Franks?

 

[MOESAN]

North-South? it seems a bit simplistic at first look - but I wait to read it if I can -

 

[Maciamo]

6 genetic clusters? It will be interesting to see how they are grouped.

I don't have much autosomal data from France (about 15 samples), but based on the little I have, I'd say:

- Centre-North & Northeast (Franche-Comté, Alsace, Lorraine, Champagne, Picardy, Nord-Pas-de-Calais) => higher Germanic admixture than average
- Brittany => very Cornish & Welsh-like admixture
- Normandy => similar to Centre-North but with some British and Scandinavian admixture
- Centre and Centre-South (Centre, Limousin, Auvergne, Languedoc) => a bit higher Iberian and lower Germanic admixture than average
- Gascony => very high Iberian admixture
- Southeast (Rhône-Alps, Provence-Alpes-Côte d'Azur) => lower Germanic, higher Italian and Neolithic admixture

 

[mwauthy]

It will be interesting to see the results. French Canadians tended to come from the western and northern areas of France except for Brittany (La Rochelle and Dieppe major ports of departure). Phenotypically they kind of look Irish except with a Roman nose. Alpine features, pale skin, dark hair. I wonder if there is some Vasconic connection between northern Spain, western France, and ancient Britons and why this might account for the high Iberian and British Isles percentages that French Canadians get on their DNA tests.

 

[Angela]

​

Not really on topic, but for those interested in French Canadian genetics, the following might be interesting. As you can see, they're really not the same people anymore:

Wiki says the original population was from "Normandy, Perche, Beauce, Maine, Anjou, Touraine, Poitou, Aunis, Angoumois, Saintonge and Gascony.^[4]" but the "Annals " source doesn't seem to be on the internet.


See:
http://www.unz.com/pfrost/french-canadians-unexplained-genetic/



http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1003815


Also see:

http://blogs.discovermagazine.com/gnxp/2011/01/the-genomic-heritage-of-french-canadians/#.WPENeNIrKyI

"That paper found that nearly 70% of the immigrant founding stock in this data set came directly from France. For the period before 1700 that fraction exceeds 95%. Of the remainder, about 15% of the founding stock were Acadians, who themselves were presumably mostly of French origin. Because of the earlier migration of the French founding stock, they left a stronger impact on future generations:

 

[mwauthy]

Thanks for the info. I always thought that because French Canadians came from different regions of France their admixtures could represent an average for a decent portion of the country of France. However, I suppose they are too homogenous of a population though for this to be the case.

 

[Pax Augusta]

A pre-print was avalaible two years ago, is it the same?

The fine-scale genetic structure of the French population (2015)

The existence of population stratification is a major problem in case-control association studies and there is a need for a better assessment of allele frequency variation within populations at all geographic scales. Such efforts have been conducted in different European countries where strong patterns of geographic variations were found. The genome-wide extent of variations in allele frequencies of common variants has however never been documented at the scale of France. In this study, we describe these patterns of variation using genome-wide SNP chip data from 4,433 individuals, recruited as part of the Three-City study and whose places of birth in France were available. We show that there is a strong correlation between the top three principal components extracted from the genetic data and the latitude and longitude of birth places. Using multiple linear regression models, we were able to determine the birth places within less than 197 km of the reported origin for 50% of the individuals. Using model-based clustering with seven main geographic regions, we found that individuals were assigned in majority to their true region of origin. However, we found that information on ancestry could not be retrieved by using a small panel of Ancestry-Informative Markers (AIMs).


Not working anymore
http://audesp.free.fr/Publis/FineScaleStructureFrance_StPierre.pdf

Thread on Eupedia opened by Moesan

http://www.eupedia.com/forum/threads/31546-Basque-and-Gascoyne-originality-new-surevey-about-France




And was this already discussed?

Fine-scale human genetic structure in Western France.

Karakachoff M1, Duforet-Frebourg N2, Simonet F3, Le Scouarnec S3, Pellen N4, Lecointe S3, Charpentier E3, Gros F3, Cauchi S5, Froguel P6, Copin N7; D.E.S.I.R. Study Group, Le Tourneau T8, Probst V8, Le Marec H8, Molinaro S9, Balkau B10, Redon R8, Schott JJ8, Blum MG2, Dina C8; D E S I R Study Group.

The difficulties arising from association analysis with rare variants underline the importance of suitable reference population cohorts, which integrate detailed spatial information. We analyzed a sample of 1684 individuals from Western France, who were genotyped at genome-wide level, from two cohorts D.E.S.I.R and CavsGen. We found that fine-scale population structure occurs at the scale of Western France, with distinct admixture proportions for individuals originating from the Brittany Region and the Vendée Department. Genetic differentiation increases with distance at a high rate in these two parts of Northwestern France and linkage disequilibrium is higher in Brittany suggesting a lower effective population size. When looking for genomic regions informative about Breton origin, we found two prominent associated regions that include the lactase region and the HLA complex. For both the lactase and the HLA regions, there is a low differentiation between Bretons and Irish, and this is also found at the genome-wide level. At a more refined scale, and within the Pays de la Loire Region, we also found evidence of fine-scale population structure, although principal component analysis showed that individuals from different departments cannot be confidently discriminated. Because of the evidence for fine-scale genetic structure in Western France, we anticipate that rare and geographically localized variants will be identified in future full-sequence analyses.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4795055/

 

[mwauthy]

What are the exact names for the 7 regions?
GE: Germanic?
GO: ?
IDF: ?
MED: Mediterranean?
NO: Nordic?
RA: ?
SO: Sud Ouest?

 

[berun]

Grand Est, Grand Ouest, Ille-de-France, Rhône-Alpes

 

[berun]

Not so good methodology:

Pour l’étude 3C, 9294 personnes âgées de 65 ans ou plus ont été recrutées dans les villes de Bordeaux, Dijon et Montpellier (d’où le nom Trois Cités), par tirage au sort sur les listes électorales. Quel avantage les personnes tirent-elles à participer à une étude de cohorte ? A priori aucun. Leur démarche, altruiste, s’apparente à un don, car elles ne bénéficieront vraisemblablement pas des retombées de la recherche.

the researchers got the samples from altruist city dwellers and were got by chance-series, and even if they selected regional samples, it seems that no sorting by deep local ancestry could be done. The results got with such data could be more blurred.

 

[MOESAN]

Not so good methodology:



the researchers got the samples from altruist city dwellers and were got by chance-series, and even if they selected regional samples, it seems that no sorting by deep local ancestry could be done. The results got with such data could be more blurred.

i agree: no serious sorting, and one more attempt to drown Brittany in a "Grand Ouest" when we know Brittany has some specificities (specific deportment in Jacobine France) ; they also drowned Germanic speaking areas, and Corsicans, and Basques into greater regions: always the same! SO their results has some worth, but they limited this worth by lack of method... the other study mentioned by Pax Augusta, limited to West, is very more instructive concerning past of France (I don't speak of future): I have the maps at home but I cannot attach them here to mu posts -