Final Report Summary - MOVE2ADAPT (Linking landscape genetic structure with local adaptation to changing environments) Project MOVE2ADAPT: Linking landscape genetic structure with local adaptation to changing environments*Please note that this project is ongoing, despite this being the final report, due to a 7-month early termination in mid-August 2015 (total funding period – April 2014-August 16 2015).SummaryTwo key processes determine a species’ genetic characteristics and vulnerability to habitat fragmentation and climate change: 1) the species’ ability to disperse and colonise new areas, and 2) the species’ ability to adapt to these new areas. This project’s aim was to investigate these processes in a range-expanding damselfly (Ischnura elegans) in Sweden (Figure 1), using current genomic sequencing technologies. The project has produced a genomic dataset of 432 individuals spanning 22 populations and a latitudinal gradient of about five degrees. So far, approximately a third of this data has been analysed, which we report on here*. However, numerous outcomes have been achieved already, providing novel insights and methodologies for understanding species range expansion and local adaptation.Key ResultsGenetic Structure: To date, we have analysed genomic data from nearly 200 I. elegans damselflies from 11 populations that span the ‘core’ and range expanding edge of the species distribution in Sweden. About 9500 genetic markers (single nucleotide polymorphisms) were obtained from the RAD sequencing data. The analysis showed strong genetic structuring among Ischnura elegans populations, where five genetic clusters were evident. This level of structure has not previously been reported in Sweden, which is likely due to the high resolution afforded by our genomic sequencing approach. The patterns show that there is high genetic structuring at the range limit, where habitat suitability is lower and populations are further apart. This indicates a reduction in gene flow between populations at the range edge compared to the southern ‘core’ region in Sweden. Loci under selection and environmental association analysis: Using the software Bayescan with all ~9500 loci, we identified 369 outlier loci (those under putative selection), which is approximately 4% of SNP markers. The whole suite of loci were analysed using what is called ‘Environmental Association Analysis’ or EAA, which uses regression and maximum likelihood approaches to statistically associate particular SNP markers with local environmental conditions. This analysis is used to identify genes and gene regions involved in local adaptation to environmental features (or even phenotypic traits).EAA analyses (also called genotype x environment correlation analysis) using two programs, LFMM and SAMBADA showed common significant associations between loci and 12 environmental variables (Figure 2). Most of the variables were derived from the WorldClim database and included various temperature variables, as well as soil type, ecoregion, altitude and tree cover. The majority of the significant locus associations were made up of the 4% outlier loci, therefore offering possible agents of selection on these pre-identified outlier loci (see Figure 2). Three environmental variables related to temperature 1) Mean Annual Temperature, 2) Mean Diurnal Temperature Variation, and 3) Temperature Seasonality, showed the highest number of genotype x environment associations (Figure 2). The overlap between the three variables indicate genes that may be more generally associated with temperature adaptation. The uniquely associating genes may be more specific to particular temperature variables, be artefacts, or represent genes that have linkages with temperature variables and other, unmeasured variables. Local adaptation at temperature-related genes across the sampling areaTo examine for patterns of local adaptation in relation to location at SNP loci, we conducted a bivariate local spatial autocorrelation analysis using the software GeoDa. As a preliminary investigation here, we present results from one SNP, whose association with temperature variables was extremely significant in both EAA analyses used above. Figure 3, shows whether the allele frequency of the chosen SNP increases, decreases or remains the same in relation to its local environmental variables, in this case, mean annual temperature. Figure 3 shows that the SNP locus in question shows increased frequency in the northern range limit populations, and decreased, lower allele frequencies in the southern core region. Two populations in between were found to neither statistically increase or decrease in frequency in relation to its local temperature. This analysis will be performed for many loci of interest (e.g. under selection) and eventually used to create an adaptive landscape for this species’ distribution in Sweden. This will provide insight in to the environmental processes driving local adaptation during range expansion. These results above are interesting in light of our recent published paper:Lancaster LT, Dudaniec RY, Hansson B, Svensson E. (2015) Latitudinal shift in thermal niche breadth results from thermal release during a climate-mediated range expansion. Journal of Biogeography. 42:1953-1963.The paper provides important information about thermal tolerance adaptation in I. elegans, and reflects more variation in phenotypic response at the range limit compared with the core, where upper thermal tolerance limits are under increased selection. Our current analysis will serve to possibly confirm this pattern using genetic signatures of local adaptive variation.ConclusionsThe results to date from this study are revolutionizing the way in which we use genomic information to understand local adaptation processes. The methods used are novel and implement recent techniques for associating genomic markers (mapped to a reference genome) with environmental variables. Developing these methods is key to getting at the roots of how species will respond to climate and land use change. The resolution of genetic structure we have already found along the sampled latitudinal gradient of I. elegans in Sweden is evidence of pockets of isolation and suggests location-specific adaptive processes, yet to be discovered.Impacts, use a socio-economic relevanceThe future of landscape and species management will need to incorporate the effects of climate and land use change. Informative, scientifically supported tools are required to do this effectively. This project is developing a methodology with extremely high potential to inform the future and current conservation of species under environmental change. Therefore, results may be used by land holders, conservation managers and planners. By characterizing the ‘genomic landscape’ of a species throughout its range, the outcomes of this project will prove useful for prioritizing management of landscapes to conserve habitat connectivity and adaptive variation. The techniques applied are not only of use to species of conservation relevance; they are also relevant for controlling the spread of pests and diseases in agriculture and in the wild by identifying possible containment routes or priority treatment areas. Thus, the socio-economic benefits of landscape genomics research are far-reaching.Project websites: http://www.biology.lu.se/rachael-dudaniecwww.dudanieclab.weebly.com final1-figure-3.png Related documents final1-figure_structure-.pdf final1-figure-2.pdf