A novel approach for detecting polygenic adaptation applied to FORest tree candidate GEne NETworks

Informations projet

FORGENET

N° de convention de subvention: 705972

DOI

10.3030/705972

Projet clôturé

Date de signature de la CE 24 Février 2016

Date de début 1 Septembre 2017

Date de fin 1 Octobre 2019

Financé au titre de

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Coût total

€ 175 419,60

Contribution de l’UE

€ 175 419,60

Coordonné par

UNIVERSITAT ZURICH
Switzerland

Periodic Reporting for period 1 - FORGENET (A novel approach for detecting polygenic adaptation applied to FORest tree candidate GEne NETworks)

Période du rapport: 2017-09-01 au 2019-08-31

Résumé du contexte et des objectifs généraux du projet

Unravelling the mechanisms of selection that allow species to adapt to changing environments has been a perpetual goal of evolutionary genetics. A current paradigm shift seems to radically change our view regarding the major mechanisms of evolutionary change. Ample experimental data indicates that most traits that potentially play a role in adaptations are controlled by many genes and that most adaptation events in natural populations are reached via so-called polygenic adaptation, involving small changes at multiple genes at a time. FORGENET explored the statistical performance of several tests to detect the genomic signature of polygenic adaptation in simulated data sets. We addressed the question how gene flow and recombination between loci can influence the power and the false positive rate of single locus and pairwise tests. Single locus tests are based on the idea that there are major genes underlying ecological traits. Pairwise tests are based on detecting non-random associations between loci among populations, thereby allow to reveal the signature of selection on gene interactions. We found that pairwise tests are more powerful in most scenarios than single locus tests, notably when selection is recent, when there is a high gene flow, and when there are epistatic interactions between genes. The methods we developed and tested are implemented in software tools. Significant progress have been made to better understand drought adaptation in a European conifer, Abies alba. We illustrate the method to detect selection on gene interactions using the example of adaptation to the photoperiod in Arabidopsis thaliana and Populus tremula.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

Software implementation and simulation study

Quantitative traits with epistatic genetic architectures have been implemented in the simulation tool, Nemo. In particular, we implemented the multi-linear model that allows for additive by additive interactions among all pairs of loci that code for a quantitative trait. In a simulation study, we explored the effect of epistasis on adaptation. A wide range of demographic and selection scenarios were considered by varying several key parameters in a factorial design varying the degree of gene flow, the mutation rate, the recombination rate, the selection variance, and the strength of epistatic interactions between loci. Briefly, we found that non-directional epistasis can foster adaptation, when the epistatic effects are the same order of magnitude as the individual allelic effects. In comparison to a purely additive case, populations can adapt faster and reach the phenotypic optimum more closely.

FORGENET proposed to characterize the statistical properties of Ohta's linkage disequilibrium (LD) statistics for detecting the signature of selection under different demographic and evolutionary conditions. Ohta statistics were implemented in Nemo and in an R package, and the statistical properties of Ohta statistics were contrasted with the most commonly used methods for detecting selection, FST outlier tests. Briefly, we found that the performance of FST outlier tests can be diminished by several factors, notably by the presence of epistasis. In contrast, we identified several advantageous properties of Ohta's statistics.

Applications

Drought tolerance in conifers

Climate change increasingly causes drought stress in temperate forests. FORGENET originally foresaw identifying genes related to drought tolerance. However, the genomic resources for conifers are still in their infancy to allow for such an analysis. Instead, we assessed drought tolerance of an ecologically and economically important European conifer, silver fir (Abies alba), using an evolutionary quantitative genetic approach. In particular, we asked the question which silver fir populations could persevere better drought episodes, and identified particularly drought tolerant populations across the Western Alps. Most interestingly, populations originating from areas of low soil water holding capacity were more drought tolerant than others. However, these populations also had a show growth rate, which could be in conflict with foresters' interests. The results of this study can now be used to select populations for obtaining population genomic data, and attempting to identify gene networks that play a role in adaptation to drought stress response.

Application to illustrate the potential of Ohta's LD statistics

In collaboration with Prof. Olivier Francois (TIMC-IMAG, Université Grenoble Alpes) and with Prof. Pär Ingvarsson (Swedish University of Agricultural Sciences), we study adaptation to photoperiod in Arabidopsis thaliana and European aspen (Populus tremula). Data has been collected from populations from the same regions of Sweden for the two species. Photoperiod is supposedly the main selective driver of the timing of bud set between northern and southern Swedish aspen populations. The data analysis is in progress.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

Kick-off events

Summer School in Polygenic adaptation, 28 Aug – 30 Aug 2017
The school was aimed at providing a deeper knowledge of the methods that map genetic to phenotypic or environmental variation and their limitations, and to learn new approaches to analyze genomic data to detect the signature of selection on polygenic traits.

Symposium on “Detecting the Genomic Signal of Polygenic Adaptation and the Role of Epistasis in Evolution” 31 August - 1 September 2017
Reseach symposium organized along four main topics:
1. Theory of polygenic adaptation and the role of epistasis in adaptation
2. Evidence for polygenic adaptation in model systems
3. Lessons from long-term breeding experiments and experimental evolution
4. Statistical methods to detect the genomic signal of polygenic adaptation

Peer reviewed publication: K Csilléry, A Rodríguez-Verdugo, C Rellstab, F Guillaume. 2018 Detecting the Genomic Signal of Polygenic Adaptation and the Role of Epistasis in Evolution. Molecular Ecology 27 (3): 603-825


Software implementation and simulation study

Presentation of the main results at the ESEB meeting in Turku, Finnland 2019 (https://eseb2019.fi/).

A manuscript is in preparation for Genetics.


Exploitation and dissemination - Applications

Quantitative genetic studies to identify locally adapted populations for drought tolerance:

Presentation of the main results at the SMBE meeting in Yokohama, Japan, 2018.

Peer reviewed publications:
K Csilléry, O Ovaskainen, C Sperisen, N Buchmann, A Widmer, F Gugerli. 2020 Adaptation to local climate in a multi-trait space: evidence from silver fir (Abies alba Mill.) populations across a heterogeneous environment. Heredity, 124: 77–92 (Green Open Access, https://www.dora.lib4ri.ch/wsl/islandora/object/wsl:21386)

K Csilléry, N Buchmann, B Fady 2019 Growth and phenology are independent strategies to cope with drought stress in marginal silver fir (Abies alba Mill.) populations. Under revision in Evolutionary Applications EVA-2019-227-OA1 (Golden Open Access)

Application of NetAdapt in Arabidopsis and Populus

A manuscript is in preparation for Genetics.

Group photo of the Summer School in Polygenic Adaptation