News

GENESCALE - at which spatial scale does natural selection operate?

© 2013 EPFL

© 2013 EPFL

LASIG - together with the WSL group of Ecological Genetics, the Laboratory of Evolutionary Botany at the University of Neuchâtel and the G2C Institute at the University of Applied Sciences Western Switzerland in Yverdon-les-Bains (HEIG-VD) - was recently granted a SNSF interdisciplinary project to investigate the contribution of Very High Resolution (VHR) Digital Elevation Models (DEMs) acquired in particular by means of Unmanned Aerial Vehicle (SenseFly drones) for multiscale analysis in landscape genomics.

An emerging objective in molecular ecology is to identify the appropriate spatial scale at which to study adaptation in plants. An answer to this may come from landscape genomics, which amalgamates population genetics and evolutionary theory with landscape ecology, i.e. the conformation of landscape elements and local environmental conditions. To date, the establishment of links between genetic polymorphisms and local environmental conditions largely relies on limited numbers of molecular markers and on data from coarse interpolation of long-term climatic conditions. While deep genotyping of entire genomes has become feasible in recent years owing to frantic technological progress, the use of remote sensing for describing landscape and microsite conditions has remained underexploited. DEMs have great potential to produce environmental variables (primary and secondary topographic attributes) that may help to identify genomic regions possibly involved in adaptive processes. It has been recently shown that VHR DEMs can be ideally generalized using wavelet transform to produce environmental variables at different nested scales, resulting in a more continuous representation of the landscape, as it exists in nature. The processing of association models between environmental variables extracted from these DEMs and genome-wide polymorphisms are likely to provide important insights on the impact of scale on the significance of these associations.

Here, we propose to deduce environmental conditions across four regions in the north-western Swiss Alps using DEMs acquired by unmanned aerial vehicle (UAV, eBee, SenseFly technology), and by light detection and ranging (LIDAR) data. We will confront these high-resolution environmental descriptions with whole-genome polymorphisms, including variation in the distribution of transposable elements (TEs) obtained from next generation re-quencing-based genome characterization of individual plants. The goal of GENESCALE is to answer the following questions: (i) at what spatial scale, and in response to which environmental factors, can is it possible to identify signals of local adaptation, and (ii) to what degree do genic vs. non-genic fractions of the genome contribute to the genome-wide signatures of adaptation?

As a target species, we will use Arabis alpina, a widespread Brassicaceae with divergent ecological requirements and likely to become a model plant for ecological genomics.