The geographical dimension of genetic diversity

In its natural framework, genetic information is embedded within a geographic context. Plants and animals are directly influenced by the specific characteristics of their surrounding environment. Therefore, spatial information is a potentially important element to be considered in trying to understand genetic resources. For many years, Geographical Information Science (GIScience) turned toward environmental modelling, generally to demonstrate how GIS basic features could be efficiently applied to fields related to the natural sciences. However, despite its current predominance in life sciences and its direct application to concerns of public society (health, food), genetics had heretofore remained outside the scope of research by the GIScience community while biologists were developing approaches based on GIS, which conducted to the elaboration of "landscape genetics". The present GIScience approach to linking genetics and geographics obviously places emphasis on geographic information, unlike most studies in this domain. This perspective is developed through an application to two case studies to assess the potential contribution of GIScience to conservation biology. The main one is provided by Econogene, a European research project aiming at promoting the sustainable conservation of genetic resources in sheep and goats. These small ruminants have considerable economic importance in marginal agrosystems in Europe. The surveying of their genetic resources makes it possible to highlight endangered breeds having high distinctiveness and priority for conservation. The second case study is complementary and supplies wild species data to demonstrate how genetic information is used to assess conservation measures applied to the endangered Scandinavian Brown Bear. Advanced molecular technologies make it possible to efficiently measure genetic information. In parallel, considerable advancements in computer science have led to the development of sophisticated GIS software and methods. The joining of molecular biology and GIScience enables novel and complementary methods of tackling some of the challenging issues related to evolutionary processes. This tentative application of diverse facets of GIScience to molecular biology addresses three distinct issues. Firstly, considering the vast quantity of information collected within the Econogene project, exploratory data analysis methods are applied to extract useful information from large spatially explicit genetic data sets. This category of GIS tools facilitates investigations to understand the geographic distribution of genetic diversity among sheep and goat breeds as well as its variation according to environmental parameters. Secondly, a review of population genetics literature having revealed weaknesses about the way spatial genetic data are generally represented on maps, the semiology of graphics is used to produce improved thematic maps representing patterns of genetic diversity. The recourse to high-performance cartography improves the interpretation of data, and facilitates the transmission of the results as well as their communication between researchers, or between researchers and the general public. Thirdly, this combination of GIScience with molecular biology mainly leads to the development of a spatial analysis method to detect signatures of natural selection within the genome. The method is adapted to a precise conception of environmental modelling, advocating the implementation of simple models in order to better grasp the functioning of natural processes, and recognizing an inescapable uncertainty. To find out these signatures, spatial analysis takes advantage of the evolution of genetics toward genomics and of the subsequent availability of large data sets generated by large genome scans. This permits to compute many simultaneous univariate logistic regression models and to identify specific regions of the genome which are selected by environmental parameters. These are identical to the ones detected by a standard approach developed in population genomics.